Economic Reforms in Chile

Economic Reforms in Chile
From Dictatorship to Democracy
Second edition
Ricardo Ffrench-Davis
This page intentionally left blank
From Dictatorship to Democracy
Second edition
© Ricardo Ffrench-Davis 2010
All rights reserved. No reproduction, copy or transmission of this
publication may be made without written permission.
No portion of this publication may be reproduced, copied or transmitted
save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency,
Saffron House, 6–10 Kirby Street, London EC1N 8TS.
Any person who does any unauthorized act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The author has asserted his right to be identified as the author of this work
in accordance with the Copyright, Designs and Patents Act 1988.
First published 2002 by The University of Michigan Press
This edition published 2010 by
PALGRAVE MACMILLAN
Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,
registered in England, company number 785998, of Houndmills, Basingstoke,
Hampshire RG21 6XS.
Palgrave Macmillan in the US is a division of St Martin’s Press LLC,
175 Fifth Avenue, New York, NY 10010.
Palgrave Macmillan is the global academic imprint of the above companies
and has companies and representatives throughout the world.
Palgrave® and Macmillan® are registered trademarks in the United States,
the United Kingdom, Europe and other countries.
ISBN 978-1-349-36709-2 ISBN 978-0-230-28965-9 (eBook)
DOI 10.1057/9780230289659
This book is printed on paper suitable for recycling and made from fully managed
and sustained forest sources. Logging, pulping and manufacturing processes are
expected to conform to the environmental regulations of the country of origin.
A catalogue record for this book is available from the British Library.
Library of Congress Cataloging-in-Publication Data
Ffrench-Davis, Ricardo.
Economic reforms in Chile : from dictatorship to democracy / Ricardo
Ffrench-Davis. — 2nd ed.
p. cm.
Summary: “This book provides an depth analysis of neo-liberal and of
progressive economic reforms in Chile since the Pinochet dictatorship _”—
Provided by publisher.
Includes bibliographical references and index.
1. Chile—Economic policy. 2. Chile—Economic conditions—1970–1973.

3. Chile—Economic conditions—1973–1988. 4. Chile—Economic
conditions—1988– I. Title.
HC192.F443 2010
339.983—dc22 2010023772
10 9 8 7 6 5 4 3 2 1
19 18 17 16 15 14 13 12 11 10
Contents
List of Figures vii

List of Tables viii
List of Boxes xi
Preface to the Second Edition xii
First Part: An Account of Four Decades 1

I Economic development in Chile since the 1970s 3
1 A historical perspective on development strategies 5
2 The neoliberal strategy, 1973–89 9
3 Democracy, reforming the reforms and development,
1990–2009 24
Annex 43
Second Part: The Neoliberal Experiment in Chile, 1973–1982 51
II Import Liberalization in 1973–82 53
1 The import liberalization process 54
2 Import composition 61
3 Effects of trade liberalization on manufacturing 64
4 Towards an evaluation 68
5 Concluding remarks 75
III Domestic Financial Liberalization and External Debt
in the 1970s: Building a Major Crisis 77
1 Financial liberalization and foreign indebtedness 78
2 Financial capital flows and indebtedness: volume,
sources, and uses 83
3 Indebtedness and macroeconomic adjustment 88
4 Interest-rate differentials 97
5 External vulnerability and the dynamics of indebtedness 102
6 Some lessons from this experience 104
Third Part: Economic Recovery and the Heritage of the
Dictatorship, 1982–1989 107
IV Policy Rectifications and Recovery from the Debt Crisis,
1982–9 109
1 The debt crisis 110
2 Debt management in 1982–9 117
3 A more pragmatic economic policy 121
v
vi Contents
4 Strong economic recovery, weak average growth 124

5 Macroeconomic (im)balances by 1989 126
V Export Development during the 1980s 130
1 The role of exports in national development 131
2 Trade policy in the 1980s: a departure from orthodoxy 133
3 Export performance: dynamism and diversification 141
4 Dynamic exports and limited economic growth 144
Fourth Part: Development Challenges for Democracy 147

VI Export Dynamism and Economic Growth since the 1990s 149
1 Trade policies since 1990 150
2 Export performance 156
3 Exports and foreign investment 165
4 Exports and economic growth 167
VII Inequality and Poverty since the 1970s 174
1 Trends in income distribution and poverty 176
2 The role of neoliberal reforms 193
3 Reforms to fight poverty and income concentration 198
VIII Managing Capital Inflows in the 1990s and the “Encaje” 209
1 Capital inflows: magnitudes and composition 211
2 The policy counter-cyclical response 216
3 The effectiveness of macro-stabilizing measures 226
4 Savings, investment, and growth in the 1990s 232
5 Some policy lessons from this experience 234
IX Economic Policy after the 1999 Recession 236
1 The origins of the break 238
2 The impact of the crisis, 1998–9 240
3 The recessive phase, 1999–2003 243
4 Dynamic recovery in 2004–5 247
5 Deceleration from 2006 249
6 Two relevant issues: the structural fiscal balance and
the exchange rate 255
X Concluding Remarks and Challenges 268
1 An account of sixteen years of dictatorship versus
sixteen years of democracy 268
2 Four general conclusions 271
3 Ten specific challenges 274
References 278
Index 291
List of Figures
I.1 Actual and potential GDP, 1960–2009 (log scale, GDP*

1996 = 100) 15
A.1 Comparison of two estimates of the Output Gap, 1960–2008
(% of GDP*) 49
II.1 Average tariff and real exchange rate, 1973–82 59
III.1 Simple conventional framework of capital markets
liberalization 79
VII.1 Chile and developed countries, per capita GDP and income
distribution, 2007 (PPP US$) 176
VII.2 National unemployment rate, 1960–2009 (percentage of the
labor force) 183
VII.3 Income distribution in Greater Santiago, 1960–2009
(Q5/Q1 ratio, three-year moving average) 190
VIII.1 Evolution of the real exchange rate, 1986–2009 (1986 = 100) 225
IX.1 Pension fund outflows and RER, 1993–2000 (% of total funds,
1986 = 100) 241
IX.2 Real copper price, actual and trend, 1959–2009 (2009 US$/lb) 248
IX.3 Trend GDP: annual review by Committee of Experts,
2006–9 (annual growth rate, %) 262
IX.4 Panel A: evolution of exports and imports of goods and
services, 2000–9 (annual real growth rates, %). Panel B:
current account balances, 2000–9 (% of current GDP) 265
X.1 Per capita GDP growth, 1973–89 versus 1990–2009 (real
indexes, 1973 and 1989 = 100) 271
vii
List of Tables
I.1 Comparison of key macroeconomic variables, 1959–2009 8

I.2 Evolution of GDP and its composition, 1975–81 (annual
average rates of growth) 17
I.3 Macroeconomic variables during dictatorship, 1974–89
(annual averages) 22
I.4 Public and private budget balances, 1990–2007 (% of GDP
at current prices) 31
I.5 Gross savings ratios, 1982–2008 (% of GDP at current prices) 39
I.6 Per capita GDP growth, 1974–2009 (annual average growth
rates) 41
A.1 Potential GDP growth and output gap, 1960–2007 (annual
average growth rate, % of GDP*) 45
II.1 Tariff liberalization, 1973–9 (rates on cif value) 56
II.2 Cost per dollar of imports, 1973–82 58
II.3 Main imports of consumer goods and total imports, 1970,
1980, 1981 (1977 US$ millions) 62
II.4 Shares by income brackets of consumption of imported goods:
main non-traditional imports in 1978 (% of total consumption) 63
II.5 Manufacturing output: Chile and the world economy,
1974–82 (1973 = 100) 65
II.6 Balance of payments, 1973–82 (1977 US$ millions) 70
III.1 Deficit on the current account and capital flows, 1970–82 84
III.2 Total foreign debt and private financial creditors, 1974–82 85
III.3 Public and private net foreign debt, 1973–82 (US$ millions) 86
III.4 Gross annual flows of loans, by debtors, 1976–82 87
III.5 Domestic and external real interest rates in pesos, 1975–82
(annual percentage rates) 98
IV.1 Foreign debt and public guarantees, 1975–89 (US$ millions
and percentages of total) 111
IV.2 Production, consumption, investment, and external shocks per
capita, 1980–9 (in 1977 pesos as a share of 1981 actual GDP) 114
IV.3 Net transfers of funds abroad by creditor, 1983–9 (US$ millions) 118
viii
List of Tables ix
IV.4 Indexes of stock prices in pesos and converted to US dollars,

1980–9 (1980 = 100) 121
V.1 Average tariff and real exchange rate, 1973–89 138
V.2 Growth of export volume, 1961–89 (annual averages, %) 141
V.3 Composition of goods exports, 1970–89 (%) 142
V.4 Geographic distribution of Chilean exports, according to
degree of elaboration, 1970–89 143
VI.1 Growth of export volume, 1986–2009 (annual averages, %) 156
VI.2 Average tariff, collected tariff, and real exchange rate,
1984–2009 157
VI.3 Indicators of export diversification, 1990–2008 (number of
markets, firms, and products) 161
VI.4 Geographic distribution of Chilean exports, according to
technological content, 1983–2008 163
VI.5 Export participation in total output, 1961–2009 (% of GDP) 168
VI.6 Exports and economic growth, 1961–2009 (annual average
rates of growth, %) 171
VII.1 Wages, family allowances, and social expenditure,
1970–2009 (real indices, 1989 = 100) 181
VII.2 Distribution of household expenditure in Santiago, 1969,
1978, 1988 (%) 182
VII.3 Poverty evolution in Chile, 1987–2006 (%) 186
VII.4 Poverty dynamics in Chile, 1996–2001–2006 (% of
population) 187
VII.5 Distribution of household expenditure and income in
Santiago, with imputed rent for homeowners, 1988 and
1997 (%) 189
VIII.1 Composition of capital flows, 1980–2008 (2008 US$
as % of GDP) 212
VIII.2 Net capital inflows and deficit on the current account,
1980–2008 (% of GDP) 225
VIII.3 Implicit cost of reserve requirement on foreign borrowing,
1991–8 (annualized rates) 227
IX.1 Investment, savings and growth, 1974–2009 (annual growth
rates, % of GDP) 237
x List of Tables
IX.2 Gross investment and depreciation, 1995–2009 (% of GDP

at constant 2003 prices) 255
IX.3 Fiscal indicators, 2001–2009 258
X.1 GDP, GDP per capita and income distribution, 1974–2009
(annual growth rates, %) 270
List of Boxes
III.1 The “crawling” exchange rate: a pioneer reform in 1965–70 92

IV.1 Debt-swaps programs 119
V.1 Incentives to exports by the late 1980s 137
VI.1 Molybdenum in the national economy 160
VII.1 Wages and the poverty line, 2000 and 2005 (monthly
income in UF) 177
VIII.1 Regulations on capital flows by mid-1998 221
xi
Preface to the Second Edition
This book reflects my research and views on crucial developments in the

Chilean economy from the 1970s up to 2009.
I have organized the text chronologically, in four subperiods: the first half
of the Pinochet dictatorship, the second half of that regime, and then the
two decades since the return to democratically elected presidents in 1990.
In my view, under a restrictive definition, the four subperiods correspond
to different approaches to building a market economy, progressing from an
extremely orthodox neoliberalism, in the first half of the Pinochet regime
(1973–82), to some significant deviations (pragmatism with a regressive
bias) in the second half (1982–9), and then two variants of meaningful trials
intended to achieve growth with equity (pragmatism with a progressive bias,
but inconsistencies) with diverse inputs and outputs.
This book is a combination of newly prepared material and previously
published papers. In this second English edition, I have deleted three
chapters from the first edition, abridged several sections, and added some
50,000 words, including several new sections and two new chapters.
The older historical sections are built on papers written close to the times
of the events, on import liberalization (Chapter II), on the building of the
debt crisis of 1982 (Chapter III), and on the rectifications with recovery in
the 1980s (Chapter IV). These articles were written while I was working at
the Center for Economic Research on Latin America (CIEPLAN) between
1976 and 1990, no doubt the most inspiring period of my professional life.
I have edited them thoroughly, but the original tone and emphasis have
been retained. Happily, I continue to feel comfortable with them today.
Although the knowledge informing them has improved and deepened
(I hope), there is no discontinuity of approach, and I feel that the various
pieces are consistent. But the reader is to be the judge. Here and there a
line of thought has been expanded by incorporating paragraphs from other
essays or that are newly written. Naturally, duplications among chapters
have been reduced as deemed convenient, and I have deleted what I judged
to be details not relevant today.
Subsequently, for the first edition, I wrote the initial draft of an essay on
income distribution and poverty, and a chapter on exports. For this second
edition, the two chapters were carefully revised and expanded. I prepared
a new chapter on reforms and policies after the contagion of the Asian
crisis, as well as a chapter on exports and growth, focused on explaining the
sharp variability observed in the correlation between export dynamism and
GDP performance in different subperiods. A brief conclusion compares the
xii
Preface to the Second Edition xiii
overall performance during the dictatorship with that of the four demo-
cratic governments that have followed since 1990, and poses some issues
and challenges for the next decade.
These chapters, and several new sections, were written while I was at the
Economic Commission for Latin America and the Caribbean (ECLAC); the
main input of another chapter was coauthored in a highly fruitful associa-
tion with Manuel Agosin and with valuable input from workshops held at
ECLAC.
The entire typescript has been revised and updated at the Department of
Economics of the University of Chile during 2008 and 2009. I have made
a serious effort to harmonize statistical series and to make use of excellent
research by other colleagues on data harmonization and explanations of
events. They are cited through the book.
I have attempted to deal with each issue and period in a way that com-
bines pragmatism with a solid analytical base, and to consistently take
into account the objectives of growth with equity, while trying to keep my
distance from fashion, myths, and extremes.
In this book I stress relationships – reforms, policies, and outcomes – with
other Latin American nations, on issues that I discuss at length in my book
Reforming Latin America’s Economies after Market Fundamentalism, prepared
during my stay at ECLAC, a period which I recall gratefully. Eight years of
work at the central bank are reflected in some of the emphasis throughout
the book.
In the preparation of successive editions, in which statistical series were
updated and harmonized, figures checked, duplications eliminated, sections
of other writings incorporated, and language improved, I enjoyed the excel-
lent collaboration and efficiency of my research assistants Heriberto Tapia
and Rodrigo Heresi, both at ECLAC. In the final stage of this second edition
in English, I acknowledge the support of Rodrigo Heresi, David Coble, and
Felipe Labrín, at the Department of Economics of the University of Chile,
and of Julia Escobedo in the checking of files. Finally, I would like to express
my appreciation to CIEPLAN for its valuable support for this publication.
To all those mentioned, and to the many who contributed but have
been unjustly omitted here, my deepest gratitude. All responsibility for
the content of this book and any errors it may contain are, of course,
solely mine.
Santiago, July 2010
First Part
An Account of Four Decades
I
Economic development in
Chile since the 1970s
One main reason for the relevance of the Chilean experience for other
emerging economies, with the implementation of a free market economic
model, rests in the depth of transformations and the long period since
this process was started. The first intense reforms were launched in 1973,
under the dictatorship of general Pinochet. Usually, it has been accepted
that there is “one” successful Chilean model. The fact is that the more
than one-third of a century since 1973 includes several subperiods, with
different approaches, diverse external environments, and notably diverse
economic and social outcomes. This diversity is of great significance, since
Chile teaches about successes but also deep mistakes and failures to be
avoided.
The first stage of the reforms (1973–81) was characterized by the imple-
mentation of a neoliberal model in its purest ideological form. Deep trade
and financial liberalizations, and the adoption of “neutral” economic poli-
cies, were accompanied by massive privatizations. In general, by 1981, suc-
cess was achieved in reducing the inflation rate and eliminating the fiscal
deficit, but at the expense of the external balance and a low investment ratio.
The consequence was an economic and social collapse in 1982, with a GDP
drop of 14%, high unemployment exceeding 30% of the labor force, and a
significant increase in poverty with a worsening income distribution.
The second stage (1982–9) involved moves towards more pragmatic poli-
cies to overcome the effects of the deep crisis. It involved a series of foreign
debt renegotiations, several policy interventions aiming to balance the
external deficit – such as tariff increases and “selective” export incentives –
and the direct takeover of the collapsed financial system, followed by
reprivatization when balance sheets were in order thanks to heavy public
subsidies to banks and debtors. At the end of this period, the economy had
recovered, while income distribution had worsened even further than in the
seventies. During recovery, actual GDP grew vigorously, but after due con-
sideration of the 1982 recession, it emerges that average growth was under
3% in both halves of the Pinochet regime.
3
4 Economic Development in Chile
By 1990, in the return to democracy, the Chilean economy faced the

challenges of achieving a sustained high average GDP growth and of serv-
ing the great social debt accumulated in the years of dictatorship. Thus,
a third variant of the economic model began in 1990, under democracy.
We have named this stage that of the reforms to the reforms, since it started
from what had been inherited; the formal slogan of the Concertación
Democrática, a center-left coalition of socialists and Christian democrats,
was “change with stability” to achieve growth with equity in the socio-
economic dimension of the program of the new government. There were
significant reforms of the market model, strengthening the social compo-
nent and correcting severe failures of economic policies. These included
labor reforms (which restored several labor rights) and a tax reform (which
raised public revenue in order to improve social expenditure). In addition,
substantive counter-cyclical changes in fiscal, monetary, capital markets,
exchange rate, and regulation policies were implemented, aiming for a
stable and sustainable macroeconomic environment, functional for eco-
nomic development. It was in this context in which Chile expanded its
productive capacity in a sustainable manner in the nineties, growing at
annual rates exceeding 7%, improving at the same time the social indi-
cators; that is, it partly achieved the elusive combination of economic
growth with equity.
Nevertheless, gradually, after the mid-1990s, Chile weakened its counter-
cyclical macroeconomic approach and became, as a consequence, vulner-
able to the turbulences originating in the Asian crisis: the real exchange
rate appreciated significantly and the external deficit doubled. As a conse-
quence, the economy exhibited a stagnating actual output and a drop in
the growth of potential GDP in the quinquenium 1999–2003. After a partial
and hesitating recovery in 2004–8, and the arrival of the contagion of the
global crisis in 2009, today it should be seeking reforms to the model that
would allow the resumption of sustained growth after the weak economic
performance of 1999–2008. We analyze all these situations through the
text.
Section 1 of this chapter presents a brief summary of outstanding features
of the Chilean economy since the fifties, advancing quickly up to the period
on which this book focuses, which begins with the military coup of 1973.
Section 2 covers the Pinochet regime. Section 3 summarizes the four demo-
cratic governments of Presidents Patricio Aylwin, Eduardo Frei Ruiz-Tagle,
Ricardo Lagos, and Michelle Bachelet. Finally, a methodological annex digs
into the long-run trends of Chile’s economic growth during the past half-
century; we emphasize (i) the contrast between average figures recorded in
dictatorship and in democracy, and (ii) the respective instabilities exhibited
by the real economy (employment, investment, output) in diverse subperi-
ods; finally, the annex presents alternative measurements of potential GDP
or productive frontier.
Economic Development in Chile 5
1 A historical perspective on development strategies
At the time of the Great Depression, the Chilean economy was one of the
most developed of the region. The Great Depression affected the Chilean
economy severely. The decline in the terms of trade and the collapse of
exports were disastrous (see Muñoz, 1986). Following the worst years,
however, the Chilean economy enjoyed a significant recovery, with a rate
of industrial growth that mitigated the constraints imposed by the trade
slump. Principally, this was the result of economic policies that responded
actively to the crisis with new industrialization strategies.
During the 1950s, however, the development model began to encounter
new problems.1 On the one hand, indiscriminate protection given to import
substitutes reduced its productivity and discouraged the development of
new exports. The instability of traditional export prices was transmitted to
the domestic economy through recurrent balance of payments shocks. One
main warning signal about the intensity of failures in the Chilean economy
was the accelerating inflation of 1952–5, when the annual rate of increase
in the consumer price index (CPI) jumped from 12 to 86%. The govern-
ment of President Carlos Ibáñez, which had been elected by a large majority
in September 1952, with the support of independents and leftists, lost its
popularity, encountered growing social unrest, and ended up adopting an
orthodox stabilization program. Both the money supply and government
spending were sharply curtailed, and the complex system of regulations intro-
duced during the Depression and World War II was cut back. But soon, the
recessionary effects of these initiatives led to widespread social rejection.
In 1958 a rightist government was elected. Given the prevailing high infla-
tion, President Jorge Alessandri focused on achieving price stabilization. In his
view, that was a prerequisite for stimulating private investment. Stabilization
was to be achieved by eliminating the “inflationary financing” of the fiscal
deficit and pegging the nominal exchange rate. Consequently, a stabilization
program anchored to a pegged exchange rate and import liberalization was
designed, supported by abundant foreign loans to the government.
This program enjoyed temporary success. Inflation was indeed substan-
tially reduced in 1960–1, but the balance of payments deficits grew so large
that international reserves were soon depleted. The investment ratio and
output had risen, but the increase in exports was unable to offset the great
expansion in imports, which exceeded the available external financing. As
a result of a currency crisis, in 1962 the exchange rate was devalued, import
restrictions were reinstalled, and inflation climbed back.
In 1964 the Christian Democrat Eduardo Frei Montalva was elected
President. The government strategy was based on a three-point platform: a
1
The next three presidential periods – 1952–8, 1958–64 and 1964–70 – are analyzed
in detail in Ffrench-Davis (1973).
gradual, multi-anchored, non-recessionary stabilization program; an indus-

trial modernization program that reactivated the role of the State as the
generator of investment initiatives, introducing new, leading sectors (such
as telecommunications and the petrochemical industry), and developing
non-traditional exports; and a program of structural and social change that
included a significant agrarian reform, the first steps in the nationalization
of the nation’s large copper mines, and the development of community and
labor-based grassroots organizations (Ahumada, 1966; Molina, 1972).
In 1964, productive capacity was underutilized and real wages were depres-
sed, which made it possible to reconcile a boost in output, wage increases, and
a reduced inflation. This was facilitated by improved terms of trade in 1965–6.
Increased fiscal expenditures in 1965 were funded by a significant tax reform,
which raised revenue, diminished evasion, and improved tax equity. After
strong growth of gross domestic product (GDP) in 1965–6, underutilized pro-
ductive capacity gradually became exhausted, while investment continued at
moderate levels. Real wages also rose much faster than planned, with a sharp
surge in the organized sectors. This resulted in cost pressures on prices and
had a negative impact on the government budget and on inflationary expec-
tations. Strikes spread, and relations between the government and urban
labor unions deteriorated. Inflation, which had been dampened in 1965–7,
began to rekindle in the subsequent years.
The macroeconomic outcome was a gradual rise in the underutilization of
productive capacity after 1967. In fact, actual average growth was a vigorous
5.9% in the first two years and a modest 3.1% in the final four years of the
Frei presidency. After reaching a high rate of use of labor and capital in 1966,
the gradual underutilization of capacity had generated a significant output
gap by late 1970.2 Nevertheless, no disequilibria as traumatic as those of
1955 and 1962 in the past, or 1975 and 1982 in the future, were recorded.
A great disequilibrium that increased over the years took place in the socio-
political arena. Political antagonisms prevented the formation of a broad-
based coalition government. The scope of the new development strategy
was extremely ambitious. Evidently, without a broad and solid coalition the
chances of success were slim. Differences concerning the nature and degree
of change prevented the formation of a large center-left majority in favor of
progressive reforms. Only after the military coup would these two political
sectors reach consensus for the accomplishment of sustained economic and
2
After a drop in 1967, a sharp improvement of the terms of trade in 1968 and 1969
was recorded (equivalent to 6% of GDP), followed by a partial negative shock of 3%
in 1970. A significant share of the improved external balance was captured by the
Chilean government and saved as international reserves in the central bank, in order
to face future deterioration of copper prices. This was a pioneering approach to the
implementation of a counter-cyclical copper stabilization fund, seeking to stabilize
both fiscal expenditure and the foreign exchange market.
social reforms, winning all elections after 1988 and in charge of the four
presidential regimes from 1990 to 2010, until defeat in the latter year.
In brief, the Frei government implemented an ambitious program of pro-
gressive reforms. Productive capacity growth averaged only 4.3%, rather low
compared with the 5.6% average recorded by Latin America. But currency
crises were avoided and inflation was curbed (see Table I.1). A significant
reform of the tax system was accomplished and 51% domestic control
over the large copper mines was achieved, leading to the capture of a siz-
able share of the economic rent from that rich natural resource. The rural
sector – with the implementation of structural reforms, most notably
agrarian reform – and the industrial sector were modernized. Exports were
diversified, with a steady increase in non-mining items, and there was
a strengthening of Latin American regional integration, which was also
instrumental in trade diversification. The State apparatus was modern-
ized by providing it with better qualified human resources. A stable real
exchange rate policy was put in place and sustained, assisted by a sort of
pioneer copper stabilization fund, and strides were made in rationalizing the
import regime and export promotion.
In 1970, President Salvador Allende was elected by one-third of the
national electorate. The Unidad Popular (UP) government sought far-reaching
structural change, particularly with respect to property and without regard
for macroeconomic equilibrium. It proceeded without a social and political
majority.
President Allende inherited idle installed capacity and large international
reserves. This allowed an expansionary policy with rapidly rising wages
and social expenditures. Economic activity responded positively, with an
8% rise in GDP and no significant inflationary pressures in 1971. However,
the expansion was carried out with public revenue losses due to drops in
real (public) utility tariffs, an appreciating real exchange rate, and a rapidly
expanding money supply, which financed a rising fiscal deficit. The fiscal
deficit climbed to 12% of GDP in 1972–3, financed principally by central
bank money printing. The huge deficit was strongly influenced by repressing
price controls on the goods and services sold by public enterprises (Larraín,
1991), while public and private investment became depressed. Meanwhile,
structural changes, including completion of the nationalization of the large
copper mines and the nationalization of the banking system and several
other enterprises, were under way. In addition, many firms and farms were
taken over arbitrarily by workers or political groups.
Aggregate demand rose disproportionately with respect to the slowing
creation of new productive capacity, while external, fiscal, and monetary
balances worsened at an accelerating pace. This deterioration was intensified
by the persistent worsening of the terms of trade from 1970 to 1972, and
by the sharp cut-off of private and US government loans (see Bitar, 1979;
Dornbusch and Edwards, 1991).
8
Table I.1 Comparison of key macroeconomic variables, 1959–2009a

During the government of
Allesandri Frei M. Allende Pinochet Concertación Aylwin Frei R.-T. Lagos Bachelet
(1959–64) (1965–70) (1971–3) (1974–89) (1990–2009) (1990–3) (1994–9) (2000–5) (2006–8)
GDP growth (%) 3.7 4.0 1.2 2.9 5.0 7.7 5.4 4.3 2.8
Exports growth (%) 6.2 2.3 –4.1 10.7 7.2 9.6 9.7 5.9 2.4
Inflation rate (%)b 26.6 26.3 293.8 79.9 7.0 17.7 6.1 2.9 4.0
Unemployment rate (%)c 5.2 5.9 4.7 18 8.7 7.3 7.5 10.7 8.8
Real wages (1970 100) 62.2 84.2 89.7 81.8 129.7 99.8 123.4 140.0 153.5
Investment (% of GDP)
In 1977 pesos 20.7 19.5 16.1 15.7 25.0 20.7 25.1 24.2 30.2
In 2003 pesos 17.9 16.8 13.9 13.6 21.5 17.9 21.6 20.8 26.0
Government surplus (% of GDP) –4.7 –2.5 –11.5 0.3 1.9 2.0 1.2 0.7 4.3
Structural surplus (% of GDP)d n.a. n.a. n.a. n.a. 0.7 0.4 0.8 0.7 0.4
Income distribution (Q5/Q1)e 12.5 13.9 11.4 18.5 15.3 13.1 16.8 15.0 14.0
Population growth (%) 2.5 2.1 1.8 1.6 1.3 1.8 1.5 1.1 1.0
Source: Central Bank, Budget Office (DIPRES) and National Bureau of Statistics (INE).
a
Annual growth rates of GDP and exports; average of annual rates of inflation and unemployment. bFrom December to December. cWith emergency
employment correction; without correction it is 13.3% in 1974–89, 7.4% in 1994–9, 9.7% in 2000–5 and 8.1% in 2006–9. dOfficial estimates that use trend
GDP as structural tax base, underestimating potential GDP.
A change of the head of the economic policy team in the middle of

1972 was accompanied by a serious attempt to correct the course taken.
Nevertheless, the magnitude of the macroeconomic imbalances and the
lack of governance frustrated this and several later attempts at rectification.
During 1972–3, actual GDP declined by 4.1% because of sectoral imbalances
and bottlenecks resulting from import restrictions, countless strikes, distor-
tions in official relative prices, a growing black market, rising underutiliza-
tion of potential GDP, and accelerating inflation (Bitar, 1979).
Meanwhile, income distribution initially improved due to significant
expansion of employment and nominal wages, but it deteriorated later
with the climbing hyperinflation (an annualized 700% in the four months
before the September 1973 coup) and the underutilization of potential GDP
in 1972–3. In all, this period shows a distributive improvement but within a
deteriorated economy: a less concentrated distribution of a smaller cake.
Economic policy ended up hurting the government politically, provid-
ing ammunition to a tough opposition. The size of idle productive capac-
ity at the outset as well as the State’s inability to enforce price controls (at
levels sometimes as low as one-tenth of generalized “black” market prices)
and to sustain balance of payment imbalances were highly overestimated.
Macroeconomic imbalances were fully felt in the second year of the admin-
istration, and from there on the struggle for power absorbed most of the
energies of both government and opposition. Economic disequilibria, black
markets, hyperinflation, low governance and the growing inability to reach
political agreements finally gave way to the dramatic institutional break-up
and the prevalence of the coup supporters in September 11, 1973 (Bitar,
1979; Baño, 2003).
2 The neoliberal strategy, 1973–893
The period covered by the dictatorship of Augusto Pinochet is longer than

that of the three democratic governments that preceded it (Alessandri, Frei
Montalva, and Allende) and the three that followed it (Aylwin, Frei Ruiz-
Tagle, and Lagos). This long period can be divided into two halves.4 The
3
This section is based on an abridged and revised version of “Economic and political
instability in Chile: 1950–89,” co-authored with Oscar Muñoz, published in S. Teitel
(ed.), Towards a New Development Strategy for Latin America: Pathways from Hirschman’s
Thoughts, Inter-American Development Bank, 1992. I am indebted to Sergio Bitar,
Manuel Marfán, Patricio Meller, Oscar Muñoz, Dagmar Raczynski, Joseph Ramos,
John Sheahan, and Simón Teitel for their valuable comments.
4
The economic and social dimensions are examined in CIEPLAN (1982, 1983), Foxley
(1983), Ramos (1986), Edwards and Cox-Edwards (1987), Larraín (1987), Fontaine
(1989), Büchi (1993), Meller (1996a), and Larraín and Vergara (2000). Texts of wider
coverage, with abundant references, are Moulián (1997), Hunneus (2001), and Correa
et al. (2002).
first, between 1973 and 1981, is a case of orthodoxy or neoliberalism in its

purest or extreme form. The second, from 1982 until March 1990, within
the same general approach, introduced numerous heterodox interventions,
which gave a more pragmatic, down to earth, shade. They were stimulated
by the severe crisis resulting from the deep ideological policies of the first
half. This second period can be identified as pragmatism (which is positive
because it tries to adapt to the real economy); nevertheless, several of the
interventions had a strong regressive bias (evidently negative), favoring sec-
tors with high incomes at the expense of medium and low income sectors.
That explains the significant worsening of income distribution and poverty
during the 1980s.
(a) Pure neoliberalism, 1973–81

One distinctive feature of neoliberalism is its “globalism”: that is, its
neglect of the implications of sectoral imbalances; of the heterogeneity in
productive structures and among economic agents, and in access to voice
and power of different sectors; of the social and allocative implications of
market segmentations, and of the difficulty of transmitting transparently
information to all sorts of economic agents so that they can contribute to
fulfilling the expectations of policymakers. Ultimately, neoliberalism also
underestimates the frequent presence of destabilizing adjustment processes,
lags, and overshooting, and the incompleteness of markets and institutions
in developing nations. These elements represent severe obstacles that pre-
vent “neutral” and indirect global economic policies from being effective,
by themselves alone, in emerging economies that are in the process of deep
transformation like Chile was.
In parallel with the changes in the economic field, there were structural
reforms in social organizations. According to the rhetoric of the dictator-
ship, these were part of the project to create a competitive society of “free
men.” This involved reforms in the university system, in the organization
and dependence of elementary schools, in health services, professional asso-
ciations, and student and labor organizations.5
The initial concerns of Pinochet’s government lay with controlling the
macroeconomic disequilibria and especially the hyper-inflation inherited in
1973.6 Soon, the arguments shifted into the inefficiencies of the prevailing
economic system, according to the international speech that became fash-
ionable in the following years. Increasingly, structural changes deepened, as
an extreme neoliberal group extended its power until it dominated public
policymaking.
5
See Campero and Valenzuela (1981), Vergara (1981), and several articles in Revista
Mensaje, especially Ruiz-Tagle (1979, 1980) and Zañartu (1980).
6
Consumer prices rose by 600% in 1973; as said, in the last four months of the UP
government the annualized inflation rate reached 700%.
(i) The reforms

The main reforms were abolition of price controls; across-the-board import
liberalization; sharp deregulation of the domestic financial market, in terms
of both access by new financial institutions and interest rate and lending poli-
cies, followed at the end of the decade by a deregulation of capital inflows;
reduction of the public sector and restrictions on the activities of public enter-
prises; privatization of the pension system and part of the national health
service; the return of expropriated businesses and lands to their former own-
ers; privatization of many traditional public enterprises; suppression of most
current labor union rights; and a tax reform, which, along with eliminating
some distortions (e.g. the cumulative effects of sales taxes corrected by imple-
menting a value added tax), sharply reduced the share of direct and progres-
sive taxes. The traditional role of the State, as entrepreneur and promoter of
investment and industrialization, was to be curtailed as quickly as possible so
that those functions might be fulfilled exclusively by private agents in liberal-
ized open markets, under the “neutral” rules of a free-market economy.
Privatization was not limited to transferring businesses expropriated
during the regime of President Allende. It was also extended to enterprises
created during successive governments after the establishment of the public
Development Corporation (CORFO) in 1939. In 1970, CORFO controlled
the ownership of forty-six enterprises, a number that rose to around three
hundred in 1973.7 In 1980 there were only twenty-four enterprises left in
the hands of this institution, half of which were in the process of being sold.
There were also a dozen public enterprises dependent on other governmen-
tal departments. Among these were the Copper Corporation (CODELCO)
and the National Oil Enterprise (ENAP).
The sale of enterprises was largely conducted in periods of domestic reces-
sion and extremely high domestic interest rates. Hence, few agents were able
to participate as buyers. This was one of the causes of the resulting acute
concentration of wealth.8 It is interesting that there was weak participation
by multinational corporations, in contrast with the official expectations of
a vigorous flow of foreign direct investment (FDI). However, massive loans
from international commercial banks provided a substantial share of the
financing required by the domestic economic groups acquiring the enter-
prises being privatized.
In the agricultural sector, the transfer of ownership had dramatic signifi-
cance. The agrarian reform that had taken place during the governments of
7
This figure does not include about 220 enterprises subject to intervention in 1973.
See Vergara (1981). Bitar (1979, Chapter X) examines the social property area pro-
gram, its evolution, and resulting problems.
8
It is well documented that the transfers were made at prices significantly lower than
normal market values. See Dahse (1979), Foxley (1983), Marcel (1989), and Devlin and
Cominetti (1994).
Presidents Frei and Allende came to an abrupt end. After 1973, around one-
third of the expropriated land was returned to former owners and close to
another one-third was auctioned to non-rural dwellers. Barely one-third of
the area was allocated to peasants. Given the curtailment of State provision
of credit and technical support to peasants and cooperatives, these were some
of the principal victims of the restructuring of public expenditure. It is esti-
mated that, as early as 1979, about half of the peasants who had been assigned
land had been forced to sell or rent out their farms (see Ortega, 1987).9 At the
same time, a massive expulsion of peasants from the farms on which they
had been living before and during the agrarian reform took place.
In 1980, another major step in the process of privatization was taken in
the social security system. The pension regime, hitherto financed through a
pay-as-you-go system, was replaced with individual capitalization in private
social security societies (AFP) created by the new system.10 Existing pensions
and those of workers who would retire within five years continued to be
the responsibility of the public sector; the rest of the workers could choose
between remaining in the old system or transferring to an AFP. For merely
making the transfer, the worker benefited from an automatic take-home
increase of 11%.
One enterprise of great importance that was able to evade privatization
was CODELCO, the copper public firm. It underwent powerful onslaughts
from the economic team but succeeded in warding them off. Even so, it
suffered budgetary restrictions and systematic constraints on its expansion
imposed by the Ministry of Finance, despite the substantial profits it con-
tributed to the Treasury. It was only permitted to make investments that
maintained the production level reached in 1977. Within the contradictions
produced by the privatization dogma, the government encouraged, unsuc-
cessfully hitherto, the development of other copper deposits to be operated
by foreign companies.11
With regard to trade liberalization, practically all non-tariff restrictions
were removed. Tariffs were rapidly reduced from the high level predomi-
nant in 1973 (a simple mean rate of 94%) to a uniform tariff of 10% for all
goods from 1979. Likewise, trade reforms included the suppression of price
bands, anti-dumping devices, and public purchasing mechanisms designed
9
Two financial factors that contributed to the pressure on peasants to sell or lease
their allotted land were the high cost of credit in the domestic capital market and the
lack of prior relations between peasants and commercial banks. On the agriculture
and peasant situation see Ortega (1987).
10
The features of the pay as you go system and a comparative analysis with other
options are discussed in Uthoff (2001). The health reform is analyzed in Titelman
(2001).
11
The main foreign investment in the 1970s was performed by Exxon through the
acquisition of a deposit in exploitation. See Vignolo (1980) and Bande and Ffrench-
Davis (1989).
to attenuate the transmission of external instability into the domestic econ-

omy. In line with the objective of unilateral and across-the-board opening to
international trade, Chile withdrew from the Andean Pact in 1976. Further,
a quite liberal statute for FDI was launched early in 1974. Chapter II covers
in detail import and other trade policies.
In the financial field, a drastic reform of the domestic market was intro-
duced in 1975. The banks that had been nationalized under the President
Allende regime were privatized. Interest rates were left totally free, regula-
tions with respect to the terms and allocation of credit were eliminated,
and new financial local and foreign entities were authorized with few
restrictions. Finally, in relation to the capital account, there was a gradual
relaxation of restrictions on financial inflows. Chapter III analyzes financial
reforms, and their effective contribution to the acute recession that started
in 1982.
(ii) Achievements and failures

Utilization rates of installed productive capacity had recovered in the first
twelve months following the military coup. Labor discipline (imposed
through the repression of unions and persecution of leaders), the liberali-
zation of prices, exchange rate devaluation, increased public works invest-
ment, and high copper prices had removed bottlenecks and favored higher
use of potential GDP. Rising copper prices in 1973–4 more than offset
increased spending on oil imports, with a net improvement in the terms of
trade equivalent to almost 5% of GDP in 1974 compared with 1972.
A novel feature for Chile was the strength of the increase in the export
volume. There were four causes for this outcome: (i) a very sharp real devalu-
ation; (ii) export capacity installed in earlier years; (iii) the removal of bot-
tlenecks in the sector and the liberalization of imports of inputs; and (iv) a
sharp reduction in domestic demand (see Ffrench-Davis, 1979).
The implementation of the neoliberal strategy was disturbed by two devel-
opments that negatively affected the Chilean economy during the 1970s:
(i) an extremely high domestic inflation rate, which the monetarist stabiliza-
tion policy had great difficulty in controlling; and (ii) the first international
oil shock, which, coupled with the sharp decline of copper prices in 1975,
created a severe balance of payments crisis.
Until 1976, anti-inflationary action was based solely on monetary policy.
From May to August of 1973, as noted above, the annualized inflation rate
had skyrocketed to 700%.12 A few days after the coup, most of the control-
led prices were freed within a context of high uncertainty. The foreseeable
result was a dramatic upsurge of inflation (88% in October 1973). As the
12
All the inflation figures used here refer to the consumer price index as corrected
in Cortázar and Marshall (1980). The official index significantly underestimated the
actual rise in prices, mainly in 1973 and in 1976–8.
fiscal situation was being brought under control, monetary policy became
effectively restrictive in the course of 1974. The official line was that the new
price fixers, the private entrepreneurs, had to take money supply behavior
into account in order to define the price of their products. The concrete fact
is that the information on money supply became widely available with a lag
of some months and with various divergent indicators, and given the high
inflation, prices were often adjusted more frequently than once a month.
Under these circumstances, the main point of reference for each economic
agent became the actual behavior of entrepreneurs as a whole, measured
through changes in the official monthly CPI, the most easily available
and up-to-date indicator. The consequence was that annual inflation rates
exceeding 300% still persisted by the third year of implementation of the
model, despite monetary restrictions, a fiscal budget already under control
in 1975, and a large recessive output gap (20% in 1975–6).
Additionally, the price of copper dropped sharply by the second half of
1974, while the oil shock persisted, with a net negative terms of trade effect
amounting to the equivalent of 6.4% of GDP in 1975 as compared to 1972.
This sharp negative shock, coupled with persistent inflation, prompted the
government to introduce a tougher adjustment program in 1975, based on
a restriction of aggregate demand, led by fiscal and monetary contraction
and significant exchange rate devaluation.
The monetary restrictions, rather than influencing principally the price
level, had a greater impact on economic activity: during 1975 industrial pro-
duction fell by 28%, GDP declined by 17% (see Figure I.1),13 and open unem-
ployment (including emergency programs) peaked at 20% (see Jadresic, 1986).
The predominance of sharp demand-reducing policies over a weaker set of
switching policies (those affecting the composition of demand and supply)
explains the significant underutilization of productive capacity or output gap.
This generated high unemployment, numerous bankruptcies, and depressed
capital formation. The “price,” which was in fact adjusted swiftly downward,
was that of labor: by 1975 wages had lost about 40% of their purchasing
power owing to the drastic repression of unions, huge unemployment, and
the legal readjustment based on an underestimated CPI.14
The monetarist recipe for controlling inflation multiplied, in the domestic
economy, by three the depressive effects deriving from the negative external
shocks and involved a notably high cost, both socially and in terms of eco-
nomic activity (see Foxley, 1983; Ramos, 1986). By mid-1976, the economic
13
Naturally, the direct impact of the deterioration in the terms of trade observed in
1975 is not included in the figure of GDP decline. GDP measures real output; the
terms of trade affect the real purchasing power resulting from a given GDP.
14
As Cortázar and Marshall (1980) document, in that period there was a systematic,
month after month, underestimation of the CPI; the official CPI was used as reference
for wage and pension readjustments. Nonetheless, even the corrected CPI shows a
gradual drop in the rate of inflation.
106
Alessandri Frei Montalva Allende Pinochet Aylwin Frei R.T. Lagos Bachelet
104
GDP growth and output gap
102 (annual average growth rates, % of GDP)
1974–89 1990–2008
100 Actual GDP (%) 2.9 5.3

5.4
Potential GDP (%) 2.5 5.6
Output Gap (%) 10.4 2.5
98
96
94
92
90
Actual GDP Potential GDP (ICOR)
88
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
Figure I.1 Actual and potential GDP, 1960–2009 (log scale, GDP* 1996 100).
Source: Author’s calculations detailed in the annex of this chapter.
team recognized – implicitly – that monetary control was proving to be unable

to restrain inflation on its own. Then the exchange rate anchor was incorpo-
rated into the anti-inflationary policy. Thus began a long process in which
this rate was used to slow down inflation by reducing the cost of imported
goods and attempting to influence inflationary expectations: analytically,
it was a transition from closed to open monetarism. In June 1976 and March
1977, exchange rate revaluations were made (a reduction in the number
of pesos per US dollar), and these were accompanied by a systematic mass
media campaign.15 The measure had a significant effect, since inflation rap-
idly fell to levels below 100% annually after the first revaluation and below
60% after the second. There was a belated understanding that inflation was
not being generated by an excess of demand and monetary expansion. The
belated realization was, however, incomplete, since it implied an excessive
conditioning of the exchange rate to the anti-inflationary policy, thereby
sacrificing external equilibrium and the production of tradables. The severe
crisis of 1982 would show that had been a wrong bet.
The evolution of anti-inflationary policy ended in 1979 with the freez-
ing of the exchange rate. As a consequence, a new stage of automatic
macroeconomic policy was introduced when the government fully adopted
the “monetary approach to the balance of payments” then in fashion in
15
After the publicized revaluations, daily mini-devaluations were applied. Exchange
rate policy is analyzed, abridgedly, in Chapter II, and at length in Ffrench-Davis
(1981).
several academic and orthodox financial circles. The new official version
was that, with a fixed exchange rate, in an economy with free imports,
domestic prices could not rise more rapidly than international inflation.
This policy was supported by heavy foreign lending, which more than cov-
ered, until 1981, an expanding external deficit (see Chapter III).
When the exchange rate was pegged, domestic annual inflation exceeded
30%, while international inflation neared 12%. Subsequently, a convergence
between the two rates occurred, but only gradually; as a consequence, for a
couple of years, domestic inflation was markedly higher than the external
rate, so that the exchange rate lost purchasing power.16 Hence, the regime of
free imports and an appreciated exchange rate caused a flood on the domes-
tic market and an unsustainable disequilibrium in the current account
during 1981. To face the external deficit, the official policy relied on an
“automatic adjustment” in the style of the gold standard: it claimed that
the real exchange rate would automatically adjust with the contraction of
monetary liquidity associated with the current loss of international reserves
in the central bank. This contraction was expected to have provoked a
sharp drop in domestic prices and nominal wages. However, the fact that
the exchange rate between 1979 and 1981 had accumulated, not by 2 or
3% but by 30% (besides the still lagged effect of import liberalization, as
analyzed in Chapter II) was overlooked. The outcome was a drastic fall in
sales, output and employment, and a progressive strangulation of business
firms through increasing indebtedness at extremely high real interest rates
(see Chapter III).
Despite numerous tough restrictions on labor union activities and a real
wage that remained below the average level of 1970, the authorities blamed
salaries for their failure to achieve a fluent and rapid automatic adjustment.17
In mid-1982, they tried to establish a general reduction of nominal wages,
but the economic team was unable to impose such a measure. Consequently,
they turned to exchange rate devaluation, but with unsustainable distur-
bances in the productive and financial systems: between June and October of
1982, the nominal exchange rate was devalued by over 70% amidst a general
crisis. The corresponding inflationary impact implied that the CPI increase
16
It is relevant that external inflation (IEP,measured in US dollars) also decreased due
to the sharp appreciation of the US dollar with respect to the currencies of the other
industrialized countries: in the twelve months before June 1982, the IEP reached an
annualized average of –2%.
17
The lower limit for the adjustment in wages was determined by the official CPI of
the preceding period. Various authors blame this rule for the costly recessive adjust-
ment. Nonetheless, there were no international precedents of cases of massive defla-
tion that had resulted from restriction of the money supply, which had operated
fluently, at the required intensity, and without causing severe problems for debtors
or economic activity.
in the following twelve months jumped to 32%, in contrast with the 4%

recorded in the previous period.
In brief, the exchange anchor was effective in curbing inflation, which
by early 1982 stood at the international level, even below zero in some
months.18 But once again the severity of the macroeconomic imbalances
generated was underestimated. Efforts had been concentrated on bringing
down inflation, while external equilibrium and investment in human and
physical capital were overlooked. As said, since 1979, the real exchange
rate had lost a third of its purchasing power, the external debt doubled,
the export boom faltered in 1981–2, and the current account deficit had
climbed to 21% of GDP in 1981.19
(iii) Vulnerability and spurious growth

The weaknesses of the model are highlighted when the composition of GDP
is disaggregated (see Table I.2). First, there was a rise in external indebted-
ness and in its cost. Around one-fifth of per capita GDP “growth” recorded
between 1974 and 1981 corresponded to interest and profit payments
Table I.2 Evolution of GDP and its composition, 1975–81 (annual average rates of
growth)
Total Per capita
1975–80 1975–81 1975–80 1975–81

(1) (2) (3) (4)
1 GDP 3.8 4.0 2.3 2.5

2 GNP 3.4 3.5 1.9 2.0
3 Value added
(a) Marketing of imports 15.5 16.2 13.8 14.5
(b) Financial services 14.6 14.2 12.9 12.5
4 GNP excluding value added in 3 1.9 1.8 0.4 0.3
Source: Calculations based on official figures from National Accounts, 1960–81, in 1977 pesos.
Revised figures by Marcel and Mellor (1986) give GDP growth of 2.6% for 1975–81, and not the
official 4.0%; the main correction was in the industrial sector (here included in row 4). There are
no disaggregated corrected figures from which to construct a revised table.
18
Several indicators of this episode in Chile are similar, in sign and size, to those in
the currency crisis of Argentina that exploded in 2002: negative inflation preceding
the crisis, a sharp GDP drop, huge unemployment, and a dramatic rise in poverty
with the crisis.
19
Figures were calculated on the basis of the average exchange rate in 1976–8. With
the very appreciated exchange rate of 1981, the deficit is 14.5% of GDP. Notice that
GDP measured in current US dollars was US$15 billion in 1978, US$33 billion in 1981
and US$16 billion in 1985. Given the enormous volatility in these years, it is wise to
“normalize” the exchange rate used, in order to make inter-temporal comparisons of
GDP expressed in foreign currency.
accrued to foreigners, which meant that the rate of expansion of the

national product was lower than that of GDP. Second, the value-added in
(i) the marketing of imported products and (ii) financial services contributed
a high share to GDP “dynamism”; these two sectors exhibited a dramatic
cumulative expansion of 13% annually.
The first sector expanded as a result of an unsustainably rapid growth of
imports of consumer and intermediate goods. As discussed above, most of
these imports were financed not by higher exports but by an increase in pri-
vate foreign debt. This source of “dynamism” – value-added in the marketing
of imports – was unsustainable in an economy that lacked vigorous real pro-
ductive support and that was experiencing a notably high external deficit. The
second source of “dynamism” was associated with the financial reform and
responded principally to the spread between the rates of interest on deposits
and loans and to the transfer of foreign credits in Chile. Thus, these two sources
of “dynamism” rested on outlier values of imports and financial spreads, both
of which were prejudicial to productive activity and investment.
There can be no doubt that, owing to the distortion implied by both sec-
tors in the Chilean economy, the overall performance contained a sizable
share of spuriousness. Hence, it is very significant that the rest of per capita
value-added, which in 1974 amounted to 91% of gross national product,
remained virtually stationary, as can be seen in column 4 of Table I.2.20 Over
and above this poor performance, GDP fell by 14% in 1982–3, while Latin
American GDP diminished only 3.2% in the same period.
The link with the future relates to national savings and investment.21
The supporters of the model claimed that it would achieve a substan-
tial increase in savings, investment, and efficiency, all crucial sources of
economic growth. The foregoing analysis has shown that the results were
negative with respect to production. However, this could be consistent with
a vigorous process of slow maturing investment. Unfortunately, in each of
20
As mentioned earlier, the production of exportables also grew at a significant pace.
Therefore, a contraction in the rest of the economy – the non-exporting sectors –
occurred. Within this subset, the industry, which accounted for the major share of
economic activity, suffered the negative pulls of the recession in 1975, trade liberali-
zation, and then exchange rate appreciation. See Chapter II and Marcel and Meller
(1986).
21
By investment, is meant what in the national accounts is called gross capital forma-
tion or gross investment in fixed capital, or what I also call productive investment, in
contrast to purely financial or speculative investment. Besides investment, there are
many other connections to the future that are not considered here. These include:
the impact that the model may have had on the capacity for technological absorption
and adaptation; the degree of creativity of the technical and university education sys-
tems; national cultural development; the channels of participation and social capital
that could serve for development strategies, based on a national consensus; and the
dynamism and efficiency of the state as a development leader.
the years between 1974 and 1980, the gross fixed investment ratio was lower
than that of each year in the 1960s; only 1981 recorded a ratio comparable
with that decade. In 1974–81, the average investment ratio was 15.7%, in
contrast with 20.2% in the sixties. In parallel, savings financed a lower
share of investment; in 1970, around 90% was covered by national savings,
whereas in 1978–81 scarcely one-half came from this source.
The high rate of external borrowing, followed by a sudden stop in 1982,
led the country into a new recessionary crisis as severe as the 1975 one.
For the second time in a decade, the Chilean economy underwent a sharp
recession, the deepeest in Latin America in 1982, with GDP declining
14%, followed by a widespread bank crisis and massive unemployment in
1983.
Those disequilibria had been induced by excessive domestic spending on
the part of the private sector, spurred by financial liberalization, huge capital
inflows, and the so-called monetary approach to the balance of payments.
Underlying these disequilibria there was a severe diagnosis error. The gov-
ernment assumed that, since it had achieved a fiscal surplus and external
borrowing was being decided by private agents, a foreign exchange crisis
would never occur. The explicit and strong support of the International
Monetary Fund (IMF) (see Robichek, 1981) reassured the government in that
wrong assumption. It failed to realize that an unsustainable medium-term
deficit could be generated in the private sector (Marfán, 2005). With the
crisis, the producers of GDP faced massive bankruptcies. Political discontent
against the iron-fisted dictatorship spread, while demonstrations of opposi-
tion proliferated even among many regime supporters.
National accounts, both official and corrected (Marcel and Meller, 1986),
show high “growth” between 1976 and 1981. However, in the first place, the
model was not initially put in motion in 1976 but (in partial form) in 1973.
Second, in 1975 there was a sharp recession, which multiplied threefold
the depressive effect of external shocks, with a fall of 17% in GDP. Hence,
to measure economic evolution from the low 1975 point implies measur-
ing as “growth” what in fact was simply a recovery to the 1974 level. This
gave rise to the mistaken impression that Chile was growing vigorously and
would continue to do so at rates in the order of 8% per year,22 irrespective
of what might happen abroad. Whereas 1976–81 gives an annual per capita
GDP increase of 4.7%, the period 1974–81 averages only 1.4%. It is obvious
that the greater the recession of 1975, the greater could have been the sub-
sequent recovery. Thus, the greater the loss of production as a result of the
recession (a true social and private cost) the higher appears the “growth” if
the measurement begins at the lowest recessed point. This is an extremely
gross technical error, but it is quite frequent. This misleading implication
would emerge again in the second half of the dictatorship.
22
See, for example, the illustrative statements quoted in Foxley (1983).
(b) The introduction of pragmatism, with a regressive bias in 1982–9

The impacts of the deep crisis of 1982–3 and the subsequent recovery were
the dominant features of this subperiod. Renegotiations of foreign debt with
bank creditors, the IMF, the World Bank and the US government played
an outstanding part. As the government’s power weakened with the 1982
recession, it was compelled to revise its strategy in several respects, includ-
ing the change of the head of the economic team and the “nationalization”
of the private debt. The government, which by 1981 held merely one-third of
total debt, by 1987 held 86% of it. The determinant of the level of economic
activity, in a very explicit way, was the availability of foreign currency; thus
the amount required to service the debt and the loans from international
financial institutions was crucial for the speed of recovery. With respect to
the policy approach, the crisis revealed the extreme vulnerability to exter-
nal shocks created by neoliberal policies and the passivity of the State. The
apparent economic success of the late 1970s had been based on a shaky
policy of external and domestic borrowing with minimal regulations. The
economic collapse and the climate of discontent made possible the reconsti-
tution of some social movements that had been dismantled, especially labor
unions and political parties of the center and left.
In the economic arena, several adjustments were made by the government
over the next four years, including successive devaluations, the increase of
import tariffs, band prices for the main agricultural products, subsidies to
non-traditional exports, stringent regulation of the financial system, implicit
nationalization of private debt and massive financial aid to the private sec-
tor (see Chapter IV). All these changed policies, previously sharply dismissed
by the neoliberal approach, were signals of “pragmatism,” in order to reduce
the “scarcity of foreign currency” (the binding external constraint) and to
recover economic activity. This implied a substantial change in fiscal policy,
which after exhibiting a strong surplus until 1981 went into deficit, averag-
ing 3.4 % of GDP in 1983–5. However, this greater pragmatism was biased in
favor of upper-income sectors, including generous subsidies, while a tough
position was maintained towards labor and grassroots organizations. The
consequence was a further deterioration in income distribution; indeed, the
year 1987 recorded the worst income distribution since statistics have been
available (see Chapter VII).
A strong and sustained recovery of economic activity and domestic output
began in 1986. Exports were quite vigorous, encouraged by a spectacular
depreciation of the real exchange rate (130% between 1982 and 1988). In
1986–7, a gradual recovery of domestic activity proceeded in a sustainable
macroeconomic framework, under a binding external restriction, but gener-
ously alleviated by loans from international financial institutions (IDB, IMF
and World Bank). In the next two years, the situation changed, with the dis-
appearance of the restriction on external financing, due to a spectacular rise
in the copper price. As a consequence, the expansion of aggregate demand
and economic activity accelerated, culminating in the overheating of the

economy in 1989, when actual GDP increased by 10%. The jump in aggre-
gate demand resulted from expansion of the money supply, tax reductions,
and import liberalization with some exchange rate appreciation, which
made imports cheaper. The installed capacity still available by 1987 and the
sizable improvement in the terms of trade (copper prices) in 1988–9 made a
sharp transitory jump in economic activity feasible.
The last biennium of the Pinochet regime exhibited that high growth
of actual GDP, in contrast with a quite moderate rise in potential GDP.
Paradoxically, then, as had happened in the late 1970s, the domestic reces-
sion of 1982–7 was useful for the “marketing” of the model. The model
was able to show “growth” for several years, a circumstance that attracted
extensive publicity in the national and foreign mass media. The prior sharp
recession was underestimated or ignored by the media under a strong dic-
tatorial regime.
In brief, the 1980s ended with the Chilean economy recording a high rate
of capacity utilization and growth of actual GDP. However, average growth
in the whole 1982–9 period was low, merely 2.9% annually, while per capita
GDP only in 1988 reached the level it had exhibited in 1981. Additionally,
the economy did show substantial disequilibria. During 1988–9, a number
of macroeconomic variables exhibited inconsistent trends over the medium
term. Real aggregate demand had grown swiftly – by 22% – in the biennium,
and GDP had risen by 18%. Exports grew vigorously in the two-year period,
but imports rose even faster. The large gap between changes in expenditure
and production was compensated by an improvement in the terms of trade
equivalent to 6 and 7% of GDP in 1988 and 1989, respectively, as compared to
1986. Output, in turn, rose sharply due to the progressive exhaustion of idle
capacity. However, productive capacity was expanding by a total 8% in the
biennium. The contrast with the 18% increase in actual GDP led to full utili-
zation of installed capacity and overheating of the economy. This manifested
itself in accelerating inflation and a deteriorating external sector. By January
1990, inflation in twelve months reached 23%, twice the 1988 rate.23
(c) A summary of outcomes

Notwithstanding that some key policies were significantly different, eco-
nomic growth recorded similar annual averages in both halves of Pinochet’s
government: 3 and 2.9%, respectively (see Table I.3). How can these similar
results be explained? The low average of both periods was associated with
the severe and abrupt crises of 1975 and 1982–3, and with the gradual recov-
ery processes, which in both cases involved high rates of underutilization of
23
This is the inflation rate for the twelve months ending in January 1990. Between
August 1989 and January 1990 the annualized rate of increase in the consumer price
index (CPI) was 31%.
Table I.3 Macroeconomic variables during dictatorship, 1974–89a (annual averages)

1974–81 1982–9 1974–89
GDP growth (%) 3.0 2.9 2.9

Exports growth (%) 13.6 7.8 10.7
Inflation rate (%)b 138.9 20.8 79.9
Total unemployment rate (%)c 16.9 19.2 18.0
Official unemployment rate (%) 13.0 13.6 13.3
Real wages (1970 100) 75.7 88.0 81.8
Investment (% of GDP)
In 1977 pesos 15.9 15.6 15.7
In 1996 pesos 15.4 15.8 15.9
Government surplus (% of GDP) 1.6 –1.1 0.3
Source: Same sources as Table I.1.

a
Annual growth rates for GDP and exports; average annual rates for inflation and unemployment.
b
From December to December. cWith emergency employment correction.
productive capacity for several years. This persistent gap between actual and
potential GDP, in turn, was the main factor discouraging capital formation
(Servén and Solimano, 1993; Agosin, 1998), which was under 16% of GDP
in each subperiod; that is, around four points (one-third in net investment)
lower than in the 1960s.
The reforms had significantly different effects on the productive structure.
Trade liberalization contributed to the significant expansion of exports.
However, applied simultaneously with a tough monetary stabilization pol-
icy, they induced a depression that featured a 26% drop in industrial output
in 1975. Nonetheless, in parallel with numerous bankruptcies, the sector
achieved a rise in productivity among the surviving companies and by 1978
manufactured exports had rosen by an average of 15% (Vergara, 1980), with
greatly increased heterogeneity in the sector. It must always be recalled that
what is relevant for the nation as a whole is the evolution of productivity,
including all people, not only those with a job!
The high rate of business bankruptcies cannot necessarily be attributed
to inefficiency resulting from protection under the earlier development
strategy. In fact, after 1973, the long recession, real annual interest rates at
an average of 38%, and the accelerated import liberalization-cum-exchange
rate revaluation can be identified as the determinant factors leading to high
business mortality. Manufacturing lost a significant share of GDP. Exports,
on the other hand, achieved great dynamism, particularly of non-traditional
goods. In fact, between 1974 and 1980, the share of non-traditional exports
(including manufactures) in total sales of goods abroad rose from 9 to 19%.
These exports exhibited a falling share in 1981–5 (averaging 16%), and then
renewed an upward trend to 22% in 1989. In all, in 1974–89, the volume
of non-copper exports averaged an annual growth of 15%, undoubtedly a
significant figure (see Chapter V).
The modernized business sector featured a surge of new economic

groups and managers exhibiting vigorous innovative competitiveness. This
modernization went in parallel with that of the Internal Revenue Service
and of the Budget Office. Together with the export drive, they contributed
to, but did not determine, the economic success achieved later in the 1990s.
Furthermore, many classical conditions for development arose, among them
the “correction” of some prices (especially exchange rate depreciation in the
1980s and reduced costs of imported inputs), market deregulation, and guar-
anteed property rights. But it was also the consequence of the substantial
fall in real wages and in taxes on capital, and the elimination of many labor
rights, while modernization still eluded the large majority of firms.
At the end, there was notable progress for some people, with the exclu-
sion of many. An extreme neoliberal interpretation reports vigorous growth,
because only the results obtained after 1986 are considered. But, it is not
justified to assess a set of reforms by counting only the years with positive
results and ignoring the negative ones. That leads to severe errors in diagno-
sis and action, and encourages other nations to repeat the experience, which
happened to a large extent among Latin American economies in the 1990s
(see Ffrench-Davis, 2006, Chapter I). The concrete fact is that average GDP
growth was mediocre in both subperiods.
Until 1982, “price corrections” had been very contradictory. In fact, the
neoliberal orthodoxy did not consider that financial liberalization could
lead to high outlier interest rates or that trade liberalization could be accom-
panied by continuous exchange rate appreciation, as was the case between
1979 and 1982. Neither was it foreseen that the private sector would be
“promoted” amidst a sharp restriction of aggregate demand, as in 1975–6
and 1982–3. In turn, the debt crisis dismantled the financial system, imply-
ing that the government allocated the equivalent of 35% of annual GDP
to the rescue of some affected sectors, diverting these resources from the
generation of economic and social development. This combination of
events helps to explain why modernization in some sectors coexisted with
a mediocre economic growth averaging 2.9% between 1974 and 1989, why
the average investment ratio was notoriously below the 1960s level, and
why income distribution worsened so much (see Chapter VII).
As on many other occasions in Chilean history, economic policy was
strongly influenced by a notable transitory improvement in the copper
price.24 It is undeniable that 1988–9 would have been quite different had
there been a “normal” copper market in those years. Based on that abnor-
mally high price, the Pinochet regime could finally boast an economy with
impressive export values and actual GDP growth figures in 1988–9. But evi-
dent macroeconomic imbalances had to be corrected in 1990. This explains
the strong policy adjustment carried out in January 1990 – in full transition
24
In fact, as was to be expected, copper prices declined from late 1989.
from dictatorship to democracy – in order to stop an overheating of the

economy led by excessive aggregate demand. The adjustment was imple-
mented, in January 1990, by the new independent central bank, launched
by Pinochet a few days before the presidential election of 1989. The bank
acknowledged the severity of the disequilibria and the risk in waiting until
the inauguration of the democratic government in March 1990.
Income inequality was notably greater than two decades earlier, as a con-
sequence of strong worsening in the seventies and eighties. In fact, the share
of the poorest quintile in total expenditure had been reduced from 7.6%
in 1969, to 5.2% in 1978 and 4.4% in 1988 (see Chapter VII, table VII.2).
A determining variable of that outcome was the deterioration of the labor
market, with average and minimum wages that in 1989 were below those in
1981 and 1970. The major swings in distributive worsening were associated
with the 1975 and 1982 crises.
In political terms, a salient development was the effective organization
of social movements and political parties, which were able to compel the
democratization of the system, even under the rules unilaterally imposed by
the dictatorship. Following the triumph of the opposition, led by a center-
left coalition, the Concertación Democrática, in the October 1988 plebi-
scite and the December 1989 presidential election, a democratic president,
Patricio Aylwin, took office in March 1990.
3 Democracy, reforming the reforms and development,

1990–200925
From 1990 the political arena was dominated by the administration of the
Concertación de Partidos por la Democracia. The Democratic Concertación,
a center-left coalition of Christian democrats and social democrats, took
over power with the successive elections of Patricio Aylwin (1990–4),
Eduardo Frei Ruiz-Tagle (1994–2000), Ricardo Lagos (2000–6), and Michelle
Bachelet (2006–10). The first two administrations gave rise to a period of the
greatest prosperity in Chilean economic history, with a sustained average
7% GDP growth rate between 1990 and 1998. It marked a clear break with
the historical trend (see Table I.1). High capital formation and a generalized
atmosphere of stability prevailed until 1998, when the Asian crisis hit Latin
American countries.
The vigorous growth was led by a 10% annual expansion of exports, a figure
similar to the average recorded in the seventies and eighties (see Chapters V
and VI). Nevertheless, GDP growth was radically different: 7% in the nine-
25
Collections of studies with diverse approaches and authors can be found in
Bosworth et al. (1994), Pizarro et al. (1996), Cortázar and Vial (1998), Larraín and
Vergara (2000), Ffrench-Davis and Stallings (2001), Larraín (2005), Meller (2005),
Muñoz (2007), and Vega (2008).
ties and 2.9% in the two previous decades.26 Given the shared good export
performance, evidently, the main factor explaining the success in the nine-
ties was the strong expansion of the rest of GDP (non-tradables plus import
substitutes), which averaged 6.5% per year in 1990–8 (see Chapter VI, Table
VI.6). As in the three decades of intensive development of Korea and Taiwan
(1965–95), key to success were the links between the export sector and the rest
of the economy, and the persistence of a comprehensive real macroeconomic
balance.
The end of President Frei’s administration (1999) and the first years of
President Lagos’s (2000–3) were characterized by a depressed economic
environment. The sharp and sustained fall in economic activity was focused
in the non-exported GDP, which represented around 70% of the whole
economy. A drop in export dynamism – undoubtedly very significant –
explains only 1.0 out of 4.5 points of lower growth in 1999–2003 with
respect to 1990–8 (see Chapter VI).
In all, GDP growth averaged 5.3% in the nineteen years between 1990
and 2008 (5% if the recession of 2009 is included). In turn, per capita GDP
expanded by 4% annually in the same period, in contrast with 1.3% in
1974–89. This notable difference, which is associated with improvements in
the quality of macroeconomic policies since 1990 and some efforts to “com-
plete” factor markets, explains the substantive economic and social progress
achieved. Nevertheless, the brake in 1998 is significantly associated with
setbacks in the quality of macroeconomic management (see Chapter IX).
(a) From the reform of the reforms to the Asian contagion, 1990–8
The new administration of Patricio Aylwin focused its efforts on achieving
stronger, stabler, and more equitable GDP growth. This required stabiliz-
ing the economy after the 1988–9 overheating generated by the Pinochet
regime, and, among other things, an increasing investment ratio, the imple-
mentation of macroeconomic policies achieving sustained equilibrium in
financial markets and in the real economy, diminishing vulnerability to
external shocks, and progress in solving the most urgent social demands by
enabling larger segments of the population to benefit from the moderniza-
tion of the economy. The aim was to reconcile macroeconomic and mac-
rosocial equilibria, and implement a style of economic policy that would
become legitimate within the new democratic framework.
The new government had decided to avoid a radical change in exist-
ing economic policy, seeking “change-with-continuity” and thus breaking
with the rehashing tradition of several previous governments. In order
26
Notice that in the three cases we “end the decades” in peak years (1981, 1989,
and 1998) in order to be “fair” in the comparison. From peak to peak, actual growth
reflects the creation of productive capacity better than using mechanically calendar
decades.
to accomplish this goal, the government of President Aylwin had to gain

the support of trade unions and incorporate their organizations into the
macrosocial decision-making process. The new administration had to cope
with the potential conflict between macroeconomic stability and demand
for more resources to be allocated to lower income groups. On the one hand,
it changed the composition of public expenditure, increasing the share of
social spending in the budget, and, on the other, it rapidly presented to
the parliament a tax reform to increase fiscal income. Tax reform included
a reintroduction of profit taxes abolished in 1988, and an increase of
2 percentage points in value-added tax.27
Likewise, in 1990 the government proposed a reform of the labor code
to Congress. It was aimed at balancing the bargaining power of employers
and workers and sought to endow current labor legislation with greater
legitimacy. To get this law passed, an agreement was reached between
the government, labor and employer organizations, and most political
parties. However, the reforms agreed (including the tax reform) were always
less comprehensive than those originally proposed by the government
(Cortázar, 1996; Marfán, 1998). A determining fact was the group of senators
that had been appointed under the Constitution designed by Pinochet, who
more than compensated for the majority achieved by candidates of the new
government in the 1989 and 1993 parliamentarian elections.
In 1990, a tripartite agreement was also reached between the government
and the representatives of unionized workers and employers; this agree-
ment provided for an increase of 28% in the real minimum wage between
1989 and 1993. In the early 1990s, significant progress in income distribu-
tion and poverty reduction was achieved in this constructive climate. From
1994, poverty continued to decline, although more slowly. While 45% of
the population lived in poverty in 1987, by 1994 this had been reduced to
27.5%, and by 1998 to 22% (see Chapter VII).
It should be pointed out that greater social effort was attained with nota-
ble fiscal responsibility. As a result of the 1990 tax reform, the expansion of
economic activity and imports, a higher than expected copper price (cap-
tured by Chile thanks to the fact that the large copper mines had remained
nationalized; now they were grouped under Corporación Nacional del
Cobre, CODELCO), and a decline in tax evasion, the fiscal burden rose by
3% of GDP (to 18% of GDP). This allowed the government to increase pub-
lic spending, in particular social expenditure, and at the same time expand
27
It can be argued that the increase in value-added taxes included in the reform
would tend to impose a regressive effect as low income families consume a larger part
of their income. Nevertheless, a coherent comparison also has to consider that most
resources are transferred to lower income families through an increase in social spend-
ing. The net effect is evidently progressive.
non-financial public sector savings from 2% in the 1980s to nearly 5% of

GDP in the 1990s (see Table I.5, below).28 Higher savings not only financed
public investment but also generated an average fiscal surplus of 1.8% of
GDP in 1990–7; the surplus was used to reduce the high stock of public debt
accumulated during the dramatic financial crisis of 1982–3.
A new political agreement in 1993 enabled the approval of several previ-
ously transitory modifications on a more permanent basis. Subsequent evi-
dence rejected the prediction of critics of the 1990s tax reform, who claimed
that it would have a negative impact on investment. After a decline of the
investment ratio in 1991 – associated with the lagged effect of the 1990
adjustment – capital formation increased in 1992 and again in 1993, reach-
ing in the next five years levels never achieved in the previous three decades
(see Table I.1). This high productive investment was the main factor behind
the outstanding annual GDP growth, which rose from below 3% in 1974–89
to over 7% in 1990–8. As empirical studies show robustly, private invest-
ment, given its irreversibility, is positively correlated with real macroeco-
nomic equilibria whenever they appear to be sustainable and fulfill two key
conditions. First, for real equilibrium, effective demand has to be consistent
with the productive capacity being generated, and, second, key macroprices
(interest and exchange rates) must be right (see Servén and Solimano, 1993;
Agosin, 1998; Coeymans, 1999; Ffrench-Davis, 2006, Chapter II). This is
what we call real macroeconomic balances.
Given the macroeconomic disequilibria generated in 1988–9, a severe
adjustment through the increase in interest rates had been carried out in
order to control the expansion of aggregate demand and a new outbreak of
inflation. This adjustment was considerably complicated soon after by large
capital inflows, which, like other economies in the region, Chile had been
receiving since the early 1990s. The gap between domestic and international
interest rates had increased significantly. In the meantime, risk-rating agen-
cies had upgraded the Chilean economy, inducing a strong inflow of short-
term “hot money” and an appreciation in the exchange rate in the second
half of 1990 (with a drop from the depreciated ceiling to the appreciated
floor of its 10% crawling band of the price in pesos of the US dollar). The
28
Depreciation of capital goods in public firms is included in gross private savings.
Moreover, the fiscal sector generated financing to cover the deficit of the public social
security system. Under the social security reform of 1981, the public sector continued
paying already retired workers and financed part of the new pensions, while income
was shifted to the private system. The fiscal figures do not consider the quasi-fiscal
deficit of the central bank – which was initially caused by the government’s interven-
tion to prevent a massive bankruptcy of the domestic financial system in 1983; it was
enlarged, subsequently, with the operational losses in monetary sterilization in the
1990s (on this latter issue, see Chapter VIII).
central bank was forced to buy large amounts of foreign currency to defend
the band’s floor.
Particularly, economic authorities faced the need to differentiate between
permanent appreciation pressures, resulting from Chile’s net improvement
in productivity and from having surmounted the debt crisis, and transitory
pressures. When the former had been identified, an attempt was made to
avoid the latter in order to maintain the competitiveness of tradables. The
strong external supply of both short-term and portfolio capital threatened
to considerably diminish the capacity of the authorities to conduct mon-
etary policy and to avoid excessive fluctuations in both the real exchange
rate and aggregate demand.
Faced with a massive capital inflow, the Chilean authorities sought to
reconcile these two objectives – an interest rate suited for keeping domestic
balances and an exchange rate consistent with external balances – by apply-
ing several counter-cyclical policies. Among these were active exchange rate
policy and monetary sterilization; selective liberalization of capital outflows;
establishing a reserve requirement (encaje) on foreign loans and liquid
inflows, which increased their costs, in order to avoid what was considered an
excess supply; and the extension of a tax, which had previously applied only
to domestic currency loans, to include foreign currency loans.
Empirical research documents that these policies were successful in reduc-
ing short-term and volatile inflows and provided space for monetary policy
and avoiding destabilizing exchange rate appreciation (see Chapter VIII).
But FDI – risk capital was exempted from the reserve requirement – became
increasingly large. FDI was stimulated by the attractive features of the Chilean
economy: rich natural resources and the almost tax free transfer of the eco-
nomic rent abroad (a loophole inherited from the dictatorship that required
correction),29 high quality macroeconomic policies, and the positive percep-
tion of the democratization process. Therefore, a large surplus in the capital
account, with effective long-term financing, higher than the moderate deficit
in the current account, was generated. The central bank responded with active
purchases of foreign currency and open market monetary sterilization.
The set of policies, especially those affecting short-term capital inflows,
contributed to keeping the current account deficit within sustainable levels
(2.3% of GDP in 1990–5) and preventing an excessive increase in more vola-
tile external liabilities. In so doing, Chilean economic authorities contrib-
uted significantly to macroeconomic stability, to the improvement of the
social indicators, to the export strategy, and to overall growth. This became
evident when Chile showed nearly complete immunity during the Mexican
crisis of 1994–5 (see Ffrench-Davis, 2006, Chapter IX).
In 1990–5, GDP growth peaked at 7.8%. If the dynamism achieved in this
period is compared with that of other episodes of high GDP growth in the
29
In 2006 it began to operate a tax that plays a similar role to a royalty.
past four decades, in contrast to previous occasions, this time a comprehen-

sive set of positive features were fulfilled: (i) GDP growth, both actual and
potential, was sustained for several rather than one or two years; (ii) growth
occurred in a context of fast rising productive investment and national sav-
ings; (iii) growth occurred without significant inflationary or external account
pressures; (iv) an orderly fiscal balance was maintained, and (v) no major
macroeconomic disequilibria were built, thanks to several mini-adjustments,
thus avoiding the common need for traumatic maxi-adjustments.
After each of these years of unsustainable macroeconomics in the past
four decades, an adjustment program with heavy welfare costs had to be
implemented. These significant changes in the macroeconomic environ-
ment are associated, on the external front, with the instability of the terms
of trade and financing. On the domestic front, they reflect the high sensitiv-
ity of aggregate demand to external shocks, especially when the economy is
operating under pro-cyclical policies.
The impact of the adjustment program in 1990 on other economic vari-
ables was less severe and quickly reversed. As mentioned earlier, investment
ratios recovered in 1992 and reached record levels from 1993. One main
merit of policies in 1990–5 rests in that they successfully resisted the tempta-
tion of achieving a faster disinflation with an increased domestic absorption
of capital inflows and at the expense of exchange rate appreciation and a
larger external deficit.
Up to 1995 the authorities implemented an effective monitoring of the
counter-cyclical regulation on capital inflows. The strength of regulations
(the cost or implicit fee and the coverage of the encaje) was adjusted continu-
ously to the strength of the external supply of funding. In the six year period
1990–5, actual and potential GDP rose at similar rates, with the economy
working close to the production frontier (that is, with a minor output gap
between potential and actual GDP).
However, policies lost their strength after 1995, thus allowing a real
appreciation of the peso and imbalances in the external accounts in 1996–7.
Chile did bend, partially, to the powerful international fashion promoting
the liberalization of capital accounts. That fashion was in general force,
pressed by the US government, the World Bank and the IMF, the OECD, and
the academic world in the most influential spheres. It had been reinforced
under the belief that the management of the tequila crisis had shown that
the world had learnt to control financial crisis; such overoptimism was
absorbed domestically by business leaders and some public authorities.
Recall that the IMF was on the road to imposing on all countries the obliga-
tion to open capital accounts. That extreme position experienced a sudden
stop with the arrival of the Asian crisis in 1997. Fortunately, Chile did step
back, from 1995, to a middle-of-the-road position. It did not dismantle
regulations but allowed a weakening of their effects, as discussed in Chapter
VIII (see also Le Fort and Lehmann, 2003) Consequently, Chile entered
“vulnerability zones” – particularly with an appreciated real exchange rate

and a large external deficit – where it was caught by the Asian crisis.
Additional factors, beyond the acute international fashion, can explain
the policy change. First, the strength shown by the Chilean economy in the
face of the Mexican crisis in 1995 created a misleading sense of invulner-
ability. It led policymakers to disregard the fact that immunity had been the
result of a policy approach that had prevented (i) excessive exchange rate
appreciation, (ii) a high current account deficit, and (iii) a significant stock
of liquid external liabilities; moreover, (iv) most inflows had been directed
to capital formation, thus strengthening the capacity to respond to critical
situations and to absorb inflows efficiently. Second, after 1995 a change in
central bank priorities could be observed, with the prevalence of an anti-
inflationary bias. Third, the outstanding performance of Chile transformed
it into a preferred destination for foreign investors, in a context in which
huge amounts of capital were being supplied to emerging economies.
Despite this larger capital surge, most counter-cyclical regulations were kept
unchanged instead of being strengthened. Evidently, counter-cyclical poli-
cies must be adapted to the strength of the cycle, which did not happen.
Therefore, when the Asian contagion reached Chile in 1998, with a strong
negative terms of trade shock, the economy had accumulated significant
imbalances: the real exchange rate had appreciated by 16% between 1995
and October 1997, and the current account deficit had jumped to 4.8% of
GDP in 1996–7, as compared to 2.3% in 1990–5, which worsened further
with a sharp negative terms of trade shock in 1998.
The fiscal policy of these two years (1996–7) deserves special considera-
tion, since it has been an object of a misinformed debate. The fiscal policy
of the nineties harmonized a significant expansion of social expenses with
an increase in both public savings and investments. As said, a responsible
fiscal management financed increases in expenses with the tax reform and
the reduction of evasion. Real macroeconomic balances placed the economy
in the productive frontier or potential GDP, with a corresponding high fiscal
revenue. In the particular case of 1996–7, fiscal responsibility was kept. The
actual fiscal surplus was 2.1% of GDP, even exceeding that of the previous
years. Consequently, the assertion that fiscal performance was responsible
for the macroeconomic imbalance of this period has no support. The exter-
nal deficit was evidently located in the private sector and was financed and
encouraged by the weaker regulation of the capital account (see Table I.4).
It is true that fiscal expenditure registered a rise in those two years. But
it must be said that it was an increase in items unanimously agreed by all
political parties, related to education, health, and pensions; additionally,
recall that it was financed with taxes and not with indebtedness. In the
meantime, GDP and productive investment kept growing up to 1998. All
evidence shows that it was necessary to act strongly on the cause of the real
macroeconomic imbalance, that is, the excess of capital inflows in 1996–7.
Table I.4 Public and private budget balances, 1990–2007

(% of GDP at current prices)a
Total Private sector Public sector
1990–5 –2.3 –4.4 2.0

1996–7 –4.2 –6.4 2.1
1998 –4.9 –5.3 0.4
1999–2003 –0.9 0.1 –1.0
2004–7 3.2 –2.7 5.8
2008 –1.5 –6.8 5.3
Source: Author’s calculations based on Balance of Payments data from

the central bank and DIPRES for the public sector.
a
Total balance is the current account deficit or use of external
saving. The public sector corresponds to central government (does
not include public enterprises). The private sector is constructed by
difference.
Because of that excess, the Asian crisis found the Chilean economy vulner-
able, with a too cheap dollar price and a high external deficit.
(b) Recessive adjustment, 1999–2003

As mentioned above, the final year of President Frei’s administration (1999)
and the first four years of President Lagos’s (2000–3) were characterized by a
depressed economic environment that frustrated the expectations of a spon-
taneous rapid recovery. The sharp and sustained fall in economic activity
was focused in the non-exported GDP, which represented around 70% of the
economy. A drop in the volume of exports – undoubtedly very significant –
explains only 1.0 out of 4.5 points of lower growth in 1999–2003 with
respect to 1990–8.
Contagion from the Asian crisis arrived through two channels. On the
one hand, there was a sharp deterioration of the terms of the trade (equiva-
lent to 3% of GDP). On the other hand, a widespread reduction of capital
flows towards emerging economies took place. Thus, from late 1997 there
arose strong expectations of depreciation, which the central bank resisted
during 1998, due to the fear of inflationary overheating and the explicit
concern of allowing domestic economic groups to reduce their debt denomi-
nated in dollars. First, massive sales of foreign currency were carried out by
the central bank with an artificially low market price. Then, by mid-1998,
the bank drastically reduced the crawling band width, intending to give a
signal of nominal exchange rate stability, in combination with a rise in the
real interest rate, which reached 14.5%. In this critical context, not only was
there a drop of financial inflows, but also a massive residents’ capital flight.
In fact, from early 1998, there were voluminous capital outflows, principally
from the private pension funds – speculating against the domestic currency,
in the context of successive capital account liberalizations – totaling almost
5% of GDP in eighteen months (see Ffrench-Davis and Tapia, 2001; Zahler,

2006). It naturally had a strong contractionary impact on domestic liquidity
and aggregate demand.
Consequently, once again, a macroeconomic imbalance led by excessive
capital inflows in 1996–7 was followed by an adjustment that was costly
for both economic growth and equity. Beginning by mid-1998, aggregate
demand fell sharply (with a 6% drop in 1999, while GDP decreased by
0.8%). Meanwhile, productive capacity kept rising, resting on the still high
investment ratio of 1998. Consequently, with the fall in actual GDP and
the increase in potential GDP, a large recessive gap between both output
measures emerged.
It is useful to examine in some detail the size and evolution of the output
gap. It is not obsolete physical capacity. It is economically productive capacity
that was in use in 1997.30 New capacity was being created at a speed of about
7% per year, which, as said, continued in 1998 and 1999, determined by the
high investment ratios up to 1998. The downward adjustment in economic
activity started by mid-1998, with actual GDP rising by only 3.2% in 1998
and falling 0.8% in 1999. All added, a sizable recessive gap of about 6% of
potential GDP was generated in 1999 and was still persisting in 2003. That
was the determinant of the sharp drop in the investment ratio in 1999–2003.
As a consequence, the plateau of potential GDP growth had dropped from
7 to 4% per year, but actual growth fell even further, to 2.6%.
As has been shown repeatedly, a significant gap between actual GDP
and the production frontier is usually followed by a drop in productive
investment. As in Mexico in 1995, Argentina in 1995 and 1999–2001, and
Korea in 1998, in Chile the investment ratio diminished substantially (by
18% in 1999) and remained depressed until 2003; the output gap plus the
drop in investment had a deepening impact on employment too.
The presidential campaign of 1999 was marked by the target of a fast return
to 6–7% growth rates. Both Concertación Democrática and the opposition
based their programs on expectations that ex-post turned out to be overop-
timistic. This mismatch between expectations of an imminent recovery and
a depressed economy, with a stagnated aggregate demand, determined an
imbalance between the resources needed to fulfill the government program
and the depressed tax revenue. To face this context, the government imple-
mented a new fiscal rule that works with the concept of fiscal structural
balance (see Marcel et al., 2001; Tapia, 2003). The rule consists of sustaining
a level of expenses consistent with structural fiscal income; that is, for each
budget year the fiscal income is estimated as if the economy were fully using
the “potential or trend GDP” and were facing the expected medium-term
30
In order to estimate that capacity, the actual GDP of 1997 was adjusted downward
by an estimate of the output identified as non-sustainable in that year (see Annex).
copper price.31 Consequently, when the economy is overheated the Treasury

accumulates savings; and under a recession it uses those funds (or borrow) to
cover the foregone wastage of productivity and employment associated with
an economic activity below potential GDP. This new public policy seems to
me to show great conceptual progress in fiscal and macroeconomic manage-
ment, given the volatility of international trade and financial markets.
This innovative fiscal rule was accompanied by two features that are not
intrinsic to the rule. One was defining as a target a structural surplus of 1%;
after some time, inevitably, that goal leads to a net creditor state, a strange
and undesirable situation in a developing economy.32 The other one was
to define as potential GDP what had been the trend GDP of the Chilean
economy, which evidently includes the intense recessions that it suffered.
Obviously, trend GDP moves well below potential GDP or the productive
frontier (see Annex and Chapter IX).
The features of the Chilean rule implied a neutral fiscal policy with respect
to the economic cycle, which involves progress with respect to the pro-
cyclical traditional norm of balancing actual fiscal budget period after period.
Nevertheless, initially it did not move decidedly into a counter-cyclical rule,
a shortcoming that became evident in 2001–3. Beyond the highly laudable
progress achieved in facing international volatility, the evidence provided by
1999–2003 shows that there is a need for effective counter-cyclical policies to
stabilize economic activity, and to become able to recover fast in recessions.
As I stress throughout this book, that capacity is an essential ingredient of
real macroeconomic balances.
The central bank foreign exchange policy evolved from the mid-nineties,
when the monetary authorities began to lose confidence in the instruments
that had been so successful in the first half of that decade. In September
1999, the exchange rate was fully liberalized, leaving behind the crawling
band. During 2000–1, in what was considered to be a policy consistent with
the new floating regime, the majority of the remaining controls on financial
flows were eliminated. One of the effects has been intense financial activ-
ity, with voluminous inflows and outflows. Associated with the recessive
environment, and helping to keep it like that, in 2002–3 a net outflow of
portfolio investments equivalent to 3% of GDP took place. Thus, national
investors joined international markets in their pro-cyclical behavior.
Notwithstanding the persistent output gap, the government continued
developing social reforms. In 2001 there was a second reform enhancing
31
Two independent consulting committees, set up by the Ministry of Finance, pro-
vide each year an estimate of the trend price of copper (used to determine withdrawals
and deposits to a copper stabilization fund) and inputs to estimate “trend GDP” for
the next budget year.
32
By 2008 the Treasury was a heavy creditor, particularly in foreign currency.
Correspondingly, in a belated decision, the surplus target was reduced to 0.5% of GDP
in 2008 and to 0% in 2009.
labor rights. Also, temporary public employment programs were intensified,

covering nearly 2% of jobs in 2002–3. It continued with educational reforms,
all-day schooling, infrastructure improvements, and some modernization
of educational programs. Public health was significantly improved with the
new “AUGE” health program, which includes universal access to treatment
for a progressively growing number of pathologies. The “Chile Barrio” and
“Chile Solidario” programs were initiated, in order to eradicate shanty towns
and to incorporate indigents into the social network supported by the State
(see Galasso, 2006).
In 2002, an unemployment insurance scheme was started that is financed
by contributions of both private workers and employers (which go to an
individual account of the worker) and government contributions (which go
to a “solidarity fund”); in September 2009, after seven years in force there
were 3.2 million active contributors (about 85% of AFP private employers).
The high coverage in terms of number of workers affiliated has been
accompanied by very modest amounts of benefits and a minimal use of
the solidarity fund (see Chapter VII). In brief, this insurance was progress,
but it needed great strengthening in terms of its magnitude, expeditious
access, expansion of the solidarity fund, and its connection with effective
labor training programs. In 2009 the parliament approved a project of the
government that enhanced the extent of and access to the solidarity fund,
including workers with short-term contracts, introduced counter-cyclical
features in the benefits, and offered job search intermediation labor training
for the unemployed.
All this progress, very novel and promising, has had limited financing
because of the output gap. But as significant is the regressive impact that
the gap has had on the labor market. In the period between mid-1998 and
2003, the number of workers with a job (including people in special pro-
grams financed by the government) grew in total by scarcely 3.3%, while
the eighteen-year-old or over population increased by around 9%. The rate
of labor participation, which had been rising during the 1990s but was still
low by 1998 (54%), diminished by one point, and open unemployment rose
by four points (exceeding10% of the labor force). The main determinant
factor of this labor market deterioration was the real macroeconomic imbal-
ance, as we define it here: the high gap between actual and potential GDP,
which implied underutilization of labor and capital, discouraging produc-
tive investment. In fact, the investment ratio (at 2003 constant prices) in
1999–2003 was three points below the average recorded in 1995–8.
Chile could have implemented a vigorous domestic positive shock, tak-
ing advantage of all the strengths accumulated by the Chilean economy:
large international reserves, a Treasury and a central bank with low exter-
nal liabilities, outstanding fiscal discipline, among other attributes, and,
a main issue, the great recessive output gap prevailing in all the period
1999–2003.
(c) Recovery led by a positive external shock in 2004–8 and

the contagion in 2009
During 2003, there was a sharp rise in the international prices of com-
modities, which lasted until the contagion of the international crisis in
2008.33 The commodity boom implied a strong positive shock for several
Latin American countries (LACs).34 In the case of Chile, the terms of trade
improved by the equivalent of 10% of GDP between the recessed 1999–2003
period and 2004–5; the hike in the price of copper and other commodities
exported by Chile was notably stronger than that for the imports of oil.
This exogenous shock contributed to a significant jump in GDP growth,
from 2.6 to 5.8% in the same subperiods. Given that potential output was
rising merely about 4% annually, this implied a noticeable reduction of the
output gap. Generalized improved terms of trade and volume of exports
enhanced, directly, the spending capacity of the private sector, and expecta-
tions returned to broad optimism. After a usual lag, productive investment
started to rise.
Naturally, the outstanding merits accumulated by the Chilean economy
were a factor supporting the sharp recovery. But the leading force was the
positive external shock. This reveals a macroeconomic weakness, since the
merits of the Chilean economy were already present in the recessive period.
Thus, the conditions for a domestic reactivating shock were at hand from
the moment the excessive external deficit as well as the over-appreciated
exchange rate of 1998 had been corrected in 1999.35
Given a large output gap by the beginning of the positive external shock,
domestic supply was able to respond with a rising GDP and low inflation
pressures (within a target band of 2–4% set by the autonomous central
bank). In the meantime, once again, the bank allowed the real exchange
rate to appreciate by 20% between March 2003 and December 2005, con-
tributing to anchor inflation and to further increase the purchasing power.
Nevertheless, the partial recovery lost force in 2006, opening a period in
which actual GDP grew below the expansion of potential GDP. In contrast
with the experience of the 1990s, the Chilean economy performed below
33
Surprisingly, in 2009, amidst the intense global recession, prices relevant for the
Chilean economy – such as of copper and oil – resumed comparatively high levels.
A similar outcome has been recorded in stock markets: high prices in a depressed
world economy could be interpreted as risky bubbles, as stressed on several occasions
by the Nobel prizewinner Paul Krugman.
34
It is noteworthy that actual GDP growth also jumped in Latin America, from 1.3 to
5.4%, between those two periods. The driving force was the same one for Chile. See a
discussion in Ffrench-Davis (2006, Chapters I and VII).
35
For instance, a domestic positive shock, implemented by Korea and Malaysia in
1999, was highly successful, achieving a sharp recovery in one year (see Mahani et al.,
2006). In my view, that option was also available for Chile.
the average speed of Latin America, thus losing the lead that it had enjoyed
in the region since the late 1980s. It is true that, by 2008 it exhibited a much
better record from 1990, with an average 5.3% vis-à-vis the 3.2% of all LACs,
but in the margin (2004–8) it had lost ground, with a 4.9% average, lower
than the 5.3% recorded by Latin America.
A mix of factors directly explains that weakening. The direct origin is
located in 2006. The central bank overshot the rise of the interest rate and
left the exchange rate to revalue further; moreover, the Ministry of Finance
sterilized the positive impact of the copper price rise, and allowed most of
the negative impact of the increasing oil price to penetrate the domestic
economy. A great virtue – the copper stabilization scheme – overshot, steri-
lizing in excess. The moderated increase in aggregate demand was strongly
biased towards imports. Again, a severe inconsistency between the current
exchange rate policy and the consensus objective of fostering exports with
higher value-added prevailed, while small and medium-sized enterprises
(SMEs) faced external competition with reduced import tariffs and an appre-
ciated exchange rate (see Chapter VI).
In brief, there was a combination of excessive priority for inflation, a slid-
ing of economic policy towards more neutral policies, and the belief that
actual GDP was already too close to potential GDP.36 The fact is that in 2006
actual GDP grew somewhat less than the potential GDP, thus resulting in an
enlarged instead of a diminished recessive output gap. That implied forgone
employment, wages, and profits. Additionally, uncertainty was reintroduced
in entrepreneurs’ minds, particularly in exporters and small entrepreneurs,
and optimism was weakened.
During the next two years, effective demand experienced several ups and
downs, affecting economic activity; the significant effects of demand on
the response of real supply (that is, actual GDP) reflected the fact that the
Chilean economy was operating persistently below the production frontier.
Exchange rate instability contributed to the weak performance. Actually,
imports rose systematically nearly twice as fast as the volume of exports;
high prices hided that persistent fact. Evidently, the exchange rate was an
outlier, and quite unstable. That is, the exchange rate regime combined
two bads for the contribution of exports, and more broadly for tradables in
general, to development.
The international scenario played a relevant role, naturally enhanced
by the openness of the Chilean economy. There were positive features
for Chile, such as the spectacular price of copper and other large exports,
which allowed the Treasury, very responsibly, to accumulate sizable funds
for eventual bad years (or awful years such as 2009); fiscal surpluses jumped
further to an annual average of over 7% of GDP in 2006–8. Further, world
36
This belief was associated with the biased methodology that, working with a trend
that includes recessive periods, leads to underestimating potential GDP.
trade was dynamic up to the arrival of the international financial crisis. On

the contrary, international prices of food were climbing. From mid-2007 to
mid-2008, the price of food in the Chilean CPI increased by 22%, which
generated significant inflationary pressures, explaining about half of the
nearly 10% annual inflation recorded at the peak of the commodities boom
(by the third quarter of 2008; see Ramos, 2008). It was, mainly, an imported
inflation. The central bank consistently expressed its bias for the inflation
target, at the expense of growth. By late 2008, when Chile was already
exhibiting negative monthly inflation, the monetary policy interest rate
(8.25%) exceeded by over seven points that of the USA.
When the contagion of the global crisis arrived in 2008, the economy moved
into open recession. In 2009 it experienced a drop in GDP of 1.5%, similar to
the Latin American average. Now the government adopted a decided counter-
cyclical approach, making use of the stabilization fund, which provided broad
space for counter-cyclical fiscal policy. Notwithstanding a drop in tax revenue
resulting from a depressed demand, some taxes were reduced transitorily
(on fuels, loans, and SMEs). In parallel, fiscal expenditure increased by 18%,
which implied a 4.5% actual deficit. The strong counter-cyclical fiscal policy
was the main force compensating for the negative shocks, principally that
suffered by the volume of exports. In fact, the volume of Chilean exports,
which had risen 7.9% in the previous two decades, dived 5.6% in 2009. Fiscal
policy softened to a significant degree the multiplication to the domestic
market of the shock on exports. The domestic economy only contracted by a
negligle amount (see Chapter VI, Table VI.6). For the first time since the open-
ing of the economy, starting in 1973, the recessive adjustment, in the face
of external shocks, was stronger on exports than on the domestic economy.
Counter-cyclicality of fiscal policy was effective and efficient in 2009.
In all, as said, growth was moderate, but low compared with the poten-
tiality of the domestic economy and the positive shocks from abroad. In
2004–8, the average 4.9% rise of GDP was determined by a 6.6% expansion
of the volume of exports and 4.2% of the rest of GDP. This latter figure
contrasts sharply with the 6.5% dynamic growth of non-exports in 1990–8.
Systemic competitivity was failing. The failure was associated with the
second variable behind the weak economic performance: a hesitant devel-
opment agenda, whose shortcomings rank from unlucky shortages in the
supply of energy, to a faulty exchange rate policy, and to lags and lacks in
the incomplete links of capital markets with productive sectors (particularly
SMEs), labor training, and incentives to innovation. There have been rel-
evant announcements of the design of a more powerful productive develop-
ment policy focused on SMEs; but, notwithstanding some concrete progress,
these are still mostly only announcements.
Chile made ambitious social reforms in this period; pensions and health
were substantially improved, as detailed in Chapter VII. They contributed to
an improvement in income distribution, which today is notably less unequal
than in the 1980s. Nonetheless, the Chilean economy is still quite regressive.
This is related to the fact that the strong social agenda was not well matched
by the economic agenda. Fiscal responsibility was outstanding but derailed
from a development concern. Structural reforms (as well as macroeconomic
management, as shown) were weak and somewhat contradictory to the
introduction of equitability in market behavior. The latter required deep
reforms in the capital market, moving away from the priority on the overnight
markets and decidedly towards enhancing the long-term market segment;
developing segments for SMEs, and for entrepreneurs without wealth or his-
tory. Further, incentives for innovation were weak and have been taking a
relevant shape only recently. Labor training for untrained workers has been
improving but too mildly. The sharp increase in the number of years of educa-
tion has become associated with lowered quality, which demands even more
effective labor training. In brief, Chile missed, to a significant degree, what we
call taking the road that leads from financierism to productivism.
(d) A summary of outcomes

Notwithstanding the recessive gap in 1998–2003, actual GDP rose by 5.3%
in 1990–2008, in contrast with the 2.9% recorded in the 1970s and 1980s.
The leading force behind that outstanding performance was the vigorous
investment ratio achieved in the 1990s. The average ratio (21.4% in 1990–
2008) was seven points larger than during the neoliberal experiment (13.6%
in 1974–89; all in 2003 prices). Even in the recessive period of 1999–2003,
capital formation was significantly higher than that in the seventies and
eighties, and sustained a potential GDP growth rate of 4%.
It is important to stress that though FDI exhibited a very significant boom,
82% of the generation of productive capacity in the nineties was carried out
by domestic agents (see Ffrench-Davis, 2003). After the recessive adjust-
ment from 1998, greenfield FDI also contracted, temporarily, but the major
reduction focused on private national investment, demonstrating the strong
sensitivity of domestic investment ratios under a persistent output gap.
On the other hand, in 1990–2008 the national savings ratio averaged 22%
(at current prices), the highest in recent decades (see Table I.5). The higher
savings ratio was associated with the stimulating macroeconomic environ-
ment faced by firms, with rather small gaps between actual and potential
GDP, leading to greater use of installed capacity, higher profit margins, and
larger reinvestment of profits (Agosin, 1998).37
37
As pointed out above, the convergence between the productive frontier and effec-
tive demand is an essential ingredient for an efficient macroeconomic policy. The
absence or weakness of this fundamental macroeconomic equilibrium has been
characteristic of Latin American economies since the 1980s. See Ffrench-Davis (2006,
Chapter II).
Table I.5 Gross savings ratios, 1982–2008 (% of GDP at current prices)

Gross saving ratio
External National Public sector Stabilization funds Other

(1) (2) ⴝ (3) ⴙ (4) ⴙ (5) (3) (4) (5)
1982–4 8.4 3.1 –1.8 – 4.9

1985–9 4.9 16.5 3.3 1.6 11.7
1990–5 2.5 22.1 4.6 0.8 16.8
1996–8 4.7 22.7 4.8 0.0 17.9
1999–2003 0.9 20.6 2.3 – 0.5 18.7
2004–8 –2.2 23.8 5.1 3.8 14.8
Source: Author’s calculations based on National Accounts data from the central bank and
DIPRES.
(3) The public sector includes cash profits of public firms, principally of CODELCO (the public
copper producer), collected by the Treasury, excluding stabilization funds. (4) Corresponds to the
Copper Buffer Fund until 2006, and the Economic and Social Stabilization Fund and Pension
Reserve Fund since then. (5) Includes net private savings plus the central bank balance, profits
of public firms not transferred to the Treasury and capitalized by these firms, and depreciation
reserves of all public and private firms.
The savings capacity is strongly affected by the terms of trade, which

continue to be extremely unstable for Chile, as documented by the several
mentions above of fluctuations of the terms of trade effect as measured by
the national accounts. The main determinant of these fluctuations has been
the price of copper, with a smaller influence of oil imports. For instance,
in 1989 the high copper price implied additional foreign currency inflows
equivalent to 3.7% of GDP into the copper buffer fund (CBF), which is one
source of public and domestic savings. To the contrary, in 1999 the fund
decumulated 0.8% of GDP. The change represents a net difference of 4.5
percentage points, which ought to be taken into account when analyzing
the evolution of public and national gross saving figures in Table I.5; that is
required in order to measure actual savings effort in each year. However, that
is only part of the story. The CBF covered only the proceeds of the large pub-
lic firm CODELCO. But price fluctuations of copper have an effect on the
taxes paid by the private producers.38 With the replacement of the CBF with
the Social and Economic Stabilization Fund in 2006, the stabilization force
of the fund was enhanced by including the fluctuations of tax proceeds of
private copper producers.
38
All the profits of CODELCO are transferred to the Treasury. The issue was a matter
of discussion, particularly in recent years, when the fiscal budget had a huge surplus
with funds provided by the profits of CODELCO, and this firm had to borrow abroad
to finance its investment. In 2008 the Ministry of Finance allowed CODELCO to
retain a fraction of its profits after taxes, raised to a significant US$1 billion (0.6% of
GDP) for 2009.
In all, the copper stabilization fund has represented an outstanding factor

of macroeconomic stabilization. In general, the authorities have operated
efficiently and responsibly with it. The CBF contributed to the implemen-
tation of the structural budget, thus allowing a move from pro-cyclical to
neutral fiscal policies during the recessive years. But that valuable “credibil-
ity asset” was underutilized during 1999–2003, principally, as made explicit
by the authorities, because of the fear of being criticized and “punished”
by financial agents if they had adopted a counter-cyclical macroeconomic
policy. Exports were the driving forces behind economic growth, increasing
the external links of the Chilean economy and its potential for sustainable
growth. It is interesting to note that the rate of export growth was relatively
similar in the past three decades, contrasting with highly volatile GDP
growth. In this context, it must be underlined that GDP growth in the 1990s
performed notably better than in the Pinochet regime, because the rest of
GDP (non-exports) also grew dynamically, reflecting broader systemic com-
petitiveness and the positive impact of sustainable macroeconomic equi-
libria. In fact, non-exports rose anually by 6.5% in 1990–8, by merely 1.7%
in 1999–2003, and then recovered to 4.2% in 2004–08. The sharp swings are
mostly associated with domestic macroeconomic changes.
Income distribution continues to be very unequal in Chile. Nevertheless, it
must be recognized, considering the diverse available data (see Chapter VII),
that in 1990–8 an improvement in income distribution took place. Both the
decennial Household Budget Survey and the annual employment survey of
the University of Chile for Santiago show a diminishing ratio between the
incomes of the richest and the poorest quintiles of population. For example,
the latter survey indicates that the ratio was 13in the sixties, 15 in 1974–81,
20 in 1982–9 and 14 in 1992–5 (see Figure VII.3). The regressive impact
of recessive output gaps is evidenced in subsequent years, particularly in
1999–2002. With the recovery of economic activity and strengthened social
policies, distributive progress was recorded in 2004–9. Nevertheless, there is
still an enormous pending debt of the economy to most Chileans.
The labor situation is a main determining factor of income distribution.
It is a fact that the improvement of the labor market recorded between
1990 and 1998 contributed to the distributive progress in those years, and
that its worsening was a key factor explaining the partial deterioration in
1999–2002. With regard to the unemployment rate, although it was lower
than one-half the average rate under the Pinochet regime, it did not recover
to the level of the 1960s in a sustainable pattern. Still more, after the long
recessive imbalance from mid-1998, unemployment and income inequality
worsened, and labor market informality arose as one of the greater chal-
lenges to restarting the path of growth with equity. In recent years the posi-
tive features have been a significant increase in the participation of women
in the labor market, with its share rising to 43%, and the gradual increase
in the share of the labor force making social security contributions, which
means progress towards formality and workers having the protection that
contracts likely grant. The share of members of the labor force contributing
to the private system had risen from 41% in 1989 to 54% in 2008.
In summary, the Concertación administration compares favorably with all
regimes since the 1950s in terms of (actual, potential, and per capita) GDP
growth, productive investment ratio, inflation, real wages, social expend-
iture, and fiscal surplus (see Table I.1). This good average performance
implied that, in the years of democracy, Chile shortened its distance from
the developed world, as documented in Table I.6. Nevertheless, this per-
formance was not sustained or continuous. As is known, the first half of
the period (1990–8) involved vigorous growth (three times the speed of the
USA’s), with a strong convergence with the developed countries and a reduc-
tion in income inequality. In the second half (1999–2008), the distance
continued to be shortened, but the per capita GDP growth trend halved,
and the strong development convergence exhibited in 1990–8, for example,
with the United States was weakened, as were previous improvements in
income distribution.
What explains this remarkable change in the development trend? A mix
of factors. Here, we mention four that are, naturally, interrelated. First,
doubtless, the Asian crisis constituted a significant negative external shock;
however, we show that its direct influence via slowing export growth only
explains a minor share of the slower GDP growth (see Chapter VI).
Second, there is a structural element: dynamism in exports was strongly
influenced by natural resource exploitation (as in copper and forestry) and
the development of public services (as in energy and telecommunications)
in mega-projects that could be hardly replicated, at the same level of ris-
ing productivity, in the 2000s (Moguillansky, 1999). Consequently, a large
number of smaller projects in sectors more demanding of the still pending
systemic competitiveness became necessary in this new framework.
Third, the strong social agenda was not well matched by the economic
agenda. Evidently, there has been a lot of progress. But a rather poor economic
outcome in the second decade needs an explanation. There was missing a
more comprehensive effort to complete long-term capital markets; encourage
Table I.6 Per capita GDP growth, 1974–2009 (annual

average growth rates)
Chile Latin America USA
1974–81 1.5 2.0 1.5

1982–9 1.2 –0.7 2.6
1990–8 5.4 1.4 1.7
1999–2008 2.6 2.0 1.6
2009 –2.5 –2.8 –3.3
Source: Central Bank of Chile, ECLAC and the IMF.

diffusion, assimilation, and adaptation of technology; broaden labor train-

ing; and open external markets for Chilean non-traditional products (all with
special emphasis on SME development). All these are essential ingredients of
a renewed national agenda for growth with equity. My view is that these short-
comings had a relevant influence on the defeat of Concertación Democrática
in the recent presidential (January 2010) and parliamentary elections. But
political analysis is beyond the reach of this book.
Fourth, there is consensus that fiscal responsibility was outstanding, but
some structural progressive reforms were weak and some changes were
contradictory with the introduction of equitability in the market behavior.
The latter required deep reforms in the capital markets, away from priority
being given to overnight markets and decidedly towards enhancing the long-
term market segment; developing segments for SMEs, and for entrepreneurs
without wealth or history.39 Further, incentives for innovation were weak,
even though they have been taking shape recently.40 Labor training for
untrained workers has been improving but too mildly. The sharp increase in
the number of years of education have been associated with lowered quality,
demanding even more effective labor training. There appears to be a grow-
ing shortage of more trained workers and entrepreneurs, in an economy that
has doubled GDP per capita, and whose requirements for growth are now
more demanding. In brief, Chile missed, to a significant degree after the
promising start in the early 1990s, what we call taking the road that leads
from financierism to productivism.
A last, but not least, factor refers to the macroeconomic environment.
As detailed in Chapter IX, policies applied in the second half of the 1990s
gradually lost coherence and the ability to control the vulnerability of the
Chilean economy in the face of external shocks. Consequently, when the
Asian crisis exploded, a climate of instability returned to Chile once again,
opening a significant gap between actual and potential GDP from mid-1998.
This gap, as demonstrated throughout the text, was the main cause of the
sharp drop in the investment ratio in 1999–2003 and the loss of entrepre-
neurship pull.
It is essential to find the way back to real sustainable macroeconomic
equilibria, after some confusing swings between the neoliberal and growth-
with-equity approaches. Thus, along with deep microeconomic development-
friendly reforms (pro-SMEs and pro-employment) – decidedly enhanced,
39
There have been several reforms of the capital markets, which have improved the
access to financing for SMEs and microcredit. However, the market is still intensive
in short-term and liquid dealings and remains quite limited for SMEs (see Consejo de
Trabajo y Equidad, 2008).
40
In 2008 it was decided to focus the allocation of the proceeds of a royalty recently
established on mining to a selected group of clusters. It represented an encouraging
deviation from allocative neutrality (see Consejo Nacional para la Innovación, 2007).
located indeed as a priority of public policy – Chile can recover sustained high
rates of growth of GDP and a progressive reduction of inequality. For both
growth and equity it is necessary to reach sustainable real macroeconomic
balances. Beyond low inflation and fiscal responsibility, foreign exchange and
interest rates approaches functional for productive development are required,
as well as an active management of aggregate demand at levels consistent
with productive capacity.41 Recent performance has been deficient in this
matter.
Annex
Long-term trends and fluctuations in economic growth:
estimating potential GDP
This annex reports estimates of the evolution of potential GDP (GDP*) and
the gaps registered between GDP* and actual GDP. Recall that output gaps
have enormous effects on investment ratios, employment and the evolution
of total productivity (TFP).
The expansion of productive capacity – economic growth – is not a given
immutable figure, but the result of public action and the behavior of social,
political, and economic agents. Naturally, it also depends on the external
environment and the ideas in fashion that influence the behavior of differ-
ent agents; for example, the relative weight of short term “financieristic”
visions focused on windfall gains versus “productivistic” visions focused on
productive development.
The evolution of the potential productive capacity depends on the com-
ponents of the production function: gross capital formation, the labor force
and its quality (human capital), technological change and productivity. As
well, the actual rate of use of potential capacity is associated with socio-
political and macroeconomic environments. I have argued that macroeco-
nomic policies intended to ensure a high use of productive capacity are a
key ingredient of an efficient public policy. They not only contribute to sus-
tained economic growth, but also, especially, they fortify equity through its
positive impact on SMEs and employment. In order to grow with equity – as
shown throughout this book– the improvement of macroeconomic policies
is a core requirement. For a better analysis of this issue, there is need to exam-
ine the evolution of GDP* (productive frontier (PF) or installed capacity).
Counting with a credible estimate of GDP* contributes to understanding
41
It is fashionable to repeat that with the adoption of a floating exchange rate the
national economy was immunized from external shocks. The truth is that exchange
rate crises are eliminated, but at the expense of transferring great instability to the
real economy, especially to the allocation of resources between tradables and non-
tradables, and to aggregate demand.
economic history and to guiding future macroeconomic policies: that is,

monetary, fiscal, and exchange rate policies, and prudential counter-cyclical
regulation of the capital account.
There are many sophisticated methodologies to estimate potential output,
but they usually do not control comprehensively in the estimates for the
effects of strong recessions. Therefore, these estimates tend to take account of
the trend or historical actual GDP average, including the sharp cycles expe-
rienced by the domestic economy, which implies a downward bias of poten-
tial GDP.42 The use of these estimates affects the quality of both historical
analyses and macroeconomic policy, because it tends to reproduce past reces-
sions in the future and hide the crucial variable: the sustainable economic
frontier – of full utilization of capital and labor and efficiency in economic
management – at which the battery of macroeconomic policies should be
aimed.
This annex presents the methodological options that were used to esti-
mate GDP*, avoiding the problem mentioned above. With those estimations
of GDP* the differences from actual GDP are reported; that is, the output
or recessive gap. This gap reflects the evolution of aggregate demand, its
relationship with effective demand (which is that located on output by
domestic resources), and the match between the structures of demand and
supply of goods and factors markets. That is an interaction between macro
and microeconomics. GDP* expanded in a fairly stable fashion in the 1960s,
with relatively high capacity utilization, compared to the large output gaps
that prevailed in the 1970s and 1980s. In fact, capital formation and the
growth of productive capacity were more unstable in these two decades
(see Figure I.1 above, and Table A.1). The low investment ratio and slow
expansion of the productive frontier are associated with the low average
capacity utilization and real macroeconomic disequilibria of these two
decades.
The instability of economic activity has negative repercussions in two
respects. First, underutilization of capacity tends to reduce the actual social
and market profitability of capital; it diminishes the availability of invest-
ment funds and worsens the financial condition of enterprises. In addition
to discouraging investment, it reduces actual productivity. On the other
hand, this instability depresses productive employment, destroys part of
capacity as a result of massive bankruptcies, and negatively affects the
sustainable future income level. That is another link between macro and
microeconomics.
Massive bankruptcies caused by macroeconomic imbalances implied a
destruction of capital (a lot of which would have been productive under
42
For example, the popular Hodrick–Prescott technique and estimations of produc-
tion functions that do not duly control for cyclical changes in the rate of capital utili-
zation or the labor supply usually give erratic estimates of total factor productivity.
Table A.1 Potential GDP growth and output gap, 1960–2007 (annual average growth
rate, % of GDP*)
Potential gap Output gap
ICOR Production function ICOR Production function

(1) (2) (3) (4)
1960–71 4.6 3.8 1.9 –0.5

1971–3 1.2 3.0 5.0 2.4
1974–81 1.9 2.4 8.7 8.7
1982–9 3.0 3.5 12.1 13.2
1990–8 7.2 6.5 –0.5 –0.7
1999–2003 4.0 4.2 6.5 6.5
2004–8 4.2 4.2 3.7 3.8
1974–89 2.5 2.9 10.4 11.0

1990–2008 5.6 5.3 2.5 2.3
Source: Author’s calculations detailed in this annex.

Columns (1) and (2) are annual average growth rates of potential GDP or productive frontier
(PF). Columns (3) and (4) report the average output gap of different subperiods, as a percentage
of potential GDP. From 2002 on, the growth of potential GDP corresponds to the estimates made
by the Ministry of Finance in 2009.
normal demand and right prices). Losses resulting from sharp import liber-
alization and large exchange rate appreciation were very significant in the
1970s (see Chapter II). The situation improved in the 1980s, partly as a result
of better aligned macroprices (the exchange and interest rates), but it was
restricted by the recessive effects of the binding foreign currency scarcity,
and the distortion and the uncertainty generated during the debt crisis.
In the 1960s, excessive protection and administrative obstacles fostered
inefficiencies, but less real macroeconomic instability helped to improve
the efficiency of all productive factors and to keep more enterprises afloat.
It allowed the concentration of energies on creating enterprises rather than
on transferring existing assets, and provided more predictable patterns of
demand and stabler relative prices (which stimulated productive investment,
given its irreversibility). As a consequence, development was also more inclu-
sive or integrated, which offered greater productive opportunities for broader
sectors of society. This environment, despite numerous distorting inefficien-
cies, explains the better performance of productivity in the 1960s vis-à-vis the
1950s and the near match with potential productivity in the 1980s (recall that
actual GDP growth was 2.9% in 1982–9 and 4.8% in 1961–71).
The years of great underutilization of capacity have been associated with
deliberate or involuntary recessive adjustments following expansions, with
fiscal, monetary, or balance of payments disequilibria. Underutilization also
intensified when stabilizing policies rested on only one policy tool (generally
the exchange rate) instead of using multiple anchors. Major gaps occurred in
1954–6, 1959, 1973, 1975–9, 1982–7, and most recently 1999–2003.
In the 1990s, once most idle capacity became utilized in 1989, the produc-
tive frontier started to increase vigorously at annual rates of around 7% in
response to an increase in the investment ratio by 7 points of GDP between
1982–9 and 1990–8 (see table A.1). Prevailing domestic stability throughout
almost the whole decade, achieved by means of prudent counter-cyclical
policies like selective regulation of short-term or volatile capital inflows,
determined the framework for a virtuous cycle of higher utilization of the
existing capital stock, thus generating higher investment flows, and gener-
ally a more efficient use of productive resources; naturally, this reflected in
significant actual productivity growth. The Asian crisis implied a sudden
stop for the virtuous circle of 1990–8, enforcing a fall of the growth rate of
actual GDP to below 3% in 1999–2003. The recessive gap between actual
and potential GDP re-emerged, implying a significant slowdown of invest-
ment dynamism; consequently, a sharp diminution of the growth rate of
potential capacity took place, from a 7% to a 4% plateau.
Estimating potential GDP or the productive frontier
GDP* is the maximum aggregate supply of goods and services that can be
achieved in any period, given the imperfections prevailing in the produc-
tion process and factor quality. The determinant variable for achieving that
maximum is an actual demand consistent with potential supply.
There is a significant asymmetry in the actual performance of the
economy. Actual GDP can be placed much below the PF (or GDP*), while
it cannot stay above that productive capacity in a sustained way. The PF
can be surpassed only temporarily, but with the exhaustion of inventories,
or growing inflationary pressures, or non-financeable external deficits, or
under transitory terms of trade positive shocks.
In this Annex GDP* and the output gap are estimated through two meth-
odologies, which take into account the essential issues of our interest. The
first one is quite old in the literature and applied economics; it is based on
the ex-ante election and correction of peaks of use of the installed capacity,
then the calculation of a relation (incremental capital-output ratio, ICOR)
between the increase of output and the increase of the capital stock. The
second uses an aggregate production function approach. Both methods, in
this specific case, report consistent results (see Table A.1).
(a) ICOR methodology, with corrected peaks

This is a very simple method with limitations. However, for economies with
great fluctuations of economic activity it tends to give results better fitted
to reality than sophisticated conventional methods that omit the control by
those fluctuations or asymmetries.
With available information, we select the peaks of use of capacity in the

period studied. In this sense, there is wide consensus among economists
on when peaks have been reached: in the past three decades, the Chilean
economic peaks were located in 1971, 1974, 1981, 1989, and 1997. I assume
that in those years potential and actual GDP were rather similar. Consequently,
a first estimation of GDP* is the level of actual GDP in those years. However,
for greater precision and based on complementary information, I search for
the main biases in each of the annual peaks identified in getting a figure
for GDP*.
I have corrected for the two main sources of bias. In one the PF is under-
estimated by actual GDP; this happens when the peak is not achieved
throughout the year (for instance, in 1974, with a recession starting in the
third quarter). The other overestimates the PF, in the case of an excess of
imports without sustainable external financing; the excess allows a large
value-added in the marketing of imports, which implies that actual GDP
overestimates GDP* (overestimates GDP* moderately in 1997; overestimates
GDP* greatly in 1981, partly compensated by the underutilization in pro-
ducing tradables, due to a highly appreciated exchange rate).
Then the net increase in output was calculated (GDP corrected by depre-
ciation of capital) between two peaks, and the net increase in the stock
of capital (the sum of the net investment of each year lagged one year).
The ratio between the two is usually called the ICOR (or reciprocal of the
standard ICOR) or coefficient of gross marginal productivity of capital.43
Subsequently, this ratio and annual investment were used to estimate poten-
tial GDP between peaks.44
This estimation of GDP* is based (i) on the assumption that capital is the
only (or main) exogenous factor in the production function, and (ii) on
the ex-ante selection of peaks, when actual GDP is located on the produc-
tion function and not “below” it. Both assumptions are reasonable as a first
approach. On the one hand, in countries like Chile capital is the scarce
factor and, therefore, is the binding element of the productive frontier;
employment as well as the absorption of technology have a positive asso-
ciation with vigorous investment. On the other hand, peaks were selected
by means of the revision of diverse indicators that allow a good degree of
certainty on the proximity between actual GDP and GDP*. However, it is
43
The reciprocal of the ICOR was used associating increases in output with net fixed
capital formation. However, apart from investment in human capital, there have been
changes in the use of other factors, such as technology and natural resources (Marfán
and Bosworth, 1994). In fact, the use of natural resources intensified. Numerous
empirical studies examine the quantity and quality of labor and capital.
44
Estimates by Marfán (1992) for 1960–88 yield annual rates of change in GDP*
similar to those obtained by the two methods used in this chapter. The similarity
of results, within the time frame common to both studies, is accounted for by the
choice of the peaks method and the agreement in the identification of peaks and of
relevant to ask how much estimates change when relaxing both assump-
tions, including labor as an independent factor in an econometric estimate
of a production function.
(b) A production function with labor and capital

By definition, the production function represents an efficient technical
relation between the output level and the factors required to produce it.
Therefore, those points of underutilization of productive factors – that is,
points with idle capacity and productive inefficiency – are not located on
the production function.
In order to deal with this relation empirically, normally the production
function would be estimated using as dependent variable the actual GDP
and as independent the productive factors actually used. The potential GDP
would be obtained, then, when applying the parameters obtained to the fac-
tors in its “full” employment level. The problem is that, given the absence
of direct data, it is difficult to get the level of capital actually used and, on
the other hand, it is also difficult to establish the full employment level of
labor; for that reason, the estimates available usually have significant biases
using this this concept.45
An alternative approach, which I use here, makes use of valuable infor-
mation by estimating the production function with a corrected sample of
observations that excludes the points with evident inefficiencies (since we
know that the points where actual GDP is significantly below GDP* are not
on the production function). For it, first the following equation in loga-
rithms by ordinary least squares (OLS) was estimated:
ln(Yt ) a0 a1 ln(Kt ) a2 ln(Lt ) ut (1)
where Y is actual GDP (constant prices of 1996), K is the capital stock

(constant prices of 1996), L represents the labor factor (we use total hours
worked corrected by education), u is a random error term.46 Then, I build a
periods when the economy is placed sharply below the production frontier (that is,
out of the production function). Given the huge cycles experienced by the Chilean
economy, empirical studies that do not take the level of activity into account can be
significantly biased in their results concerning total factor productivity. For instance,
Roldós (1997) gives a change of total factor productivity (TFP) per annum of –3.8% in
1981–5 and 0.9% for 1986–90, evidently distorted by the 1982 recession. The interest-
ing paper of Beyer and Vergara (2002) estimates –2.2 points of contribution of the TFP
in 1981–85 and 2.3 points in 1986–90. Given the small output gap of 1981 and 1990,
the average approximates to real structural TFP.
45
See diverse estimates for the Chilean case in Morandé and Vergara (1997), Roldós
(1997), Coeymans (1999), Hofman (1999), Marcel et al. (2001), Beyer and Vergara
(2002), Contreras and García (2002), and Gallego and Loayza (2002).
46
The series of capital stock and hours worked constructed by the Ministry of Finance
was used.
25.0
ICOR PF
20.0
15.0
10.0
5.0
0.0
5.0
10.0
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
Figure A.1 Comparison of two estimates of the Output Gap, 1960–2008 (% of GDP*).
Source: Author’s calculations detailed in this annex: columns (3) and (4) of Table A.1.
corrected sample that does not contain the observations of years with more
significant inefficiencies. In order to approximate to it I excluded the obser-
vations in which the error term is negative. With that exclusion I correct for
the bias of the evidently asymmetrical fluctuations of actual GDP around
GDP*. But, beyond econometrics, one must check for the consistency of the
resulting selection with other data. The excluded years are 1975–8, 1982–8
and 1999–2003. This selection of years when actual GDP has deviated
sizably from the PF is consistent with other empirical studies like Coeymans
(1999) and Marfán (1992), among others.
With the remaining observations, (1) was re-estimated. Then the new
parameters obtained plus the actual value of K and estimates of “full
employment” L were used (in logarithms) to forecast the dependent vari-
able through the full sample, obtaining thereby the GDP* series. Figure A.1
compares the resulting output gap in both methods. Method 1 renders a
higher estimate of the output gap because it considers only years close to
the frontier, while method 2 retains years clearly below the PF, such as 1979
and 2004. Beyond this evident source of divergences, the similarity is clear,
in spite of the differences in the assumptions, and the fact that the produc-
tion function estimates GDP* with fixed parameters; meanwhile the ICOR
methodology allows for changes in productivity from peak to peak.
Second Part
The Neoliberal Experiment in
Chile, 1973–1982
II
Import Liberalization in 1973–82*
The distinctive characteristic of trade policies implemented in 1973 was the

sharp unilateral import liberalization, which was undertaken to a degree then
unprecedented in modern economic history in other semi-industrialized
or emerging economies. By establishing a uniform tariff of 10% for nearly all
imports, and deleting non-tariff restrictions, this liberalization suppressed
all selective criteria for regulating them. Trade liberalization was accom-
panied by a reduction in restrictions on foreign exchange dealings and on
capital flows.
The implementation of free trade raises four questions. First, did the
policy allow for a more efficient use of available resources or did it pro-
duce significant new divergences between market and social “efficiency”?
Second, what degree of dynamism characterized the process, compared to
its historical behavior, and how did the intensity and timing of liberaliza-
tion affect employment, investment, and consumption? Third, how much
competition or economic concentration did the policy reform produce?
Last, how effectively “neutral” were these “non-discriminatory” economic
policies? Policies, seemingly “neutral,” were applied to different segments
of the economy that coexisted in a framework marked by inequalities and
sharp heterogeneity in productivity among workers and among firms;
the heterogeneity of productivities is an outstanding feature of develop-
ing economies, which deters economic growth and equity. It implies that
“neutral” policies may produce asymmetrical effects, evidently non-neutral,
among different productive and social groups. There are also different effects
on the production of exportables that experience positive pulls and the pro-
duction of importables that receive negative pulls. Additionally, there are
* Based on “Import liberalization: the Chilean experience, 1973–82,” in Samuel and

Arturo Valenzuela (eds), Military Rule in Chile: Dictatorship and Oppositions, The Johns
Hopkins University Press, Baltimore, 1986. I appreciate the comments of Augusto
Aninat, Vittorio Corbo, René Cortázar, Jaime Estévez, Alejandro Foxley, Dominique
Hachette, and Pilar Vergara.
53
54 Import Liberalization
likely transfers of externalities to the rest of the domestic economy gener-

ated by the exposure to foreign competition (Tybout et al., 1991; Herzer and
Nowak-Lehmann, 2004).1
It is evident that the effects of a deep process of reforms need a long time
for completion. Nevertheless, given a positive time preference rate, it should
be kept in mind that a real peso today will be worth more than twice as
much as one earned ten years later. On the other hand, many other policy
changes were implemented together with trade reform, making it difficult
to distinguish between the effects of each of them.
At the outset, it must be discarded the simplistic dilemma that the only
alternative to the specific implementation of sharp import liberalization by
the Pinochet government would have been to maintain the chaotic situa-
tion of 1973 must be discarded. Actually, in 1967–70 a very pragmatic policy
(in my view) rationalizing trade policies was implemented (see Ffrench-
Davis, 1973). Why, for instance, could not that pragmatic approach to trade
reform be adopted again? Its main features were gradual implementation,
direct complementarity with export promotion, reduction of the vari-
ance and average effective protection to import substitutes, and an active
crawling-peg exchange rate policy consistent with the tariff reform.
Section 1 shows the path followed by the reduction of import restrictions,
focusing on tariff liberalization and the evolution of the real exchange rate
in order to determine to what degree this macroeconomic price played the
compensatory role assumed by policymakers. Section 2 analyzes the behav-
ior of the principal categories of imports, and examines some distributive
effects. Section 3 studies the overall impact of import liberalization on
manufacturing and employment, examining disaggregated information on
the performance by industrial branches. Section 4 presents the main conclu-
sions of this chapter.
1 The import liberalization process
In late 1973, before the introduction of reforms, Chilean foreign trade

was subject to a great deal of government interventionism:2 a simple tariff
average of 94%, with sharp dispersion, countless non-tariff restrictions and
multiple exchange rates. A remarkable reduction in the extremely high bar-
riers that protected import substitutes constituted the core of the new trade
policy. The rapid liberalization process, which began in 1973 and ended in
1
In this chapter, I focus on import policies. Export policies, their effects, and inter-
relationships between both dimensions are analyzed in Chapters V and VI.
2
Notice that this was the situation in 1973. As said, in the second half of the 1960s a
reform was in progress that included the gradual rationalization of the import regime,
the improvement of an export promotion scheme, and the systematic implementation
of a crawling-peg exchange rate policy (see Ffrench-Davis, 1973, Chapters III and IV).
Import Liberalization 55
June 1979, induced a drastic change in market comparative advantages by

modifying both the profile of effective protection and its average.3
During the period of liberalization, the target of the process underwent
significant changes. What initially appeared to be a moderate reform, with
maximum tariff rates of 60%, in the end became a drastic revamping of the
tariff structure with a final uniform tariff of 10%.
The first steps in the process consisted of suppressing the main non-tariff
barriers and dropping to 220% all tariffs that were above that level. Most
import prohibitions and deposits were eliminated. These import deposits,
with a rate of 10,000%, which were charged on more than half of imports,
represented a discretionary control on imports in 1973. The deposits were
waived on the condition that importers fixed their volumes within the quotas
recommended by the government. Since this mechanism was applied along
with an undervalued exchange rate to several thousand products, it generated
countless supply bottlenecks and speculative windfall gains for importers.
The “normalization” of the foreign exchange market (see Chapter III), which
took place in October 1973, and a significant increase in the price of copper
during the second half of that year facilitated the rapid removal of quantita-
tive restrictions and initial tariff reductions.
In early 1974, the government announced a tariff liberalization scheme
that would take place gradually over a three-year period and indicated that it
would proceed pari passu with real exchange rate depreciation. The finance
minister asserted that it would generate export-led growth and that work-
ers would benefit, “as it would create more jobs in expanding sectors than
the number of jobs that could disappear in some highly inefficient sectors.”
It was argued that “greater growth perspectives are based in the opening to
trade, in the development of export industries and in the intensification of
Latin American integration.”4 The announcement gave no indication of the
levels that tariffs were projected to reach.
The first such indication came in May 1974, when the government
declared that “in 1977, no tariff will be higher than 60%. This clearly defines
the tariff policy that will be followed in the future, so that domestic indus-
tries can make whatever adjustments are necessary and prepare themselves
so that they will be in good shape to meet foreign competition.”5
Despite these announcements, strategic differences emerged between
proponents of a fully orthodox policy and other government officials with
a more pragmatic bent. Towards late 1974, reserved government documents
mentioned a range of 25–35%, within which the majority of tariff rates
3
In Ffrench-Davis (2006, Chapter IV) a comparative analysis is made with the trade
reforms implemented by many LACs since the mid-1980s and in the 1990s.
4
See DIPRES (1978, pp. 35 and 61).
5
Comments of the finance minister, October 1974, reprinted in DIPRES (1978,
p. 107). The minister reconfirmed this goal on April 24, 1975 (p. 172).
would fall.6 Later, in 1975, the government announced that the tariff range
would be placed between 10 and 35% and that it would be reached during
the first half of 1978, by gradual reductions.
Although step-by-step the initial, more pragmatic, official policy gave way
to the free trade approach, the policy as it was presented by 1975 contained
two important heterodox elements. On one hand, it contemplated non-
uniform tariff rates (from 10 to 35%) according to the degree of elaboration
of different categories of products. On the other hand, it also contemplated
maintaining the tariff preferences agreed under the Andean Pact, which
meant that tariff reductions for several products stopped at the minimum
common external tariff in place in 1972–6, according to the Cartagena
Agreement.
The gradual tariff reductions were made approximately once every six
months, as can be seen in Table II.1. Nevertheless, the final reductions,
scheduled for the first half of 1978, were anticipated in August 1977, when
99.6% of all tariffs reached the 10–35% range, with a 20% simple average.
Table II.1 Tariff liberalization, 1973–9 (rates on cif value)

Maximum tariff Modal tariff
Date Rate (%) % of all Rate (%) % of all Simple No.

items items average of
tariff (%) items
(1) (2) (3) (4) (5) (6)
12/31/73 220 8.0 90 12.4 94.0 5,125

3/1/74 200 8.2 80 12.4 90.0 5,125
3/27/74 160 17.1 70 13.0 80.0 5,125
6/05/74 140 14.4 60 13.0 67.0 5,125
1/16/75 120 8.2 55 13.0 52.0 5,125
8/13/75 90 1.6 40 20.3 44.0 4,958
2/09/76 80 0.5 35 24.0 38.0 4,952
6/7/76 65 0.5 30 21.2 33.0 4,956
12/23/76 65 0.5 20 26.2 27.0 4,959
1/8/77 55 0.5 20 24.7 24.0 4,981
5/2/77 45 0.6 20 25.8 22.4 4,984
8/29/77 35 1.6 20 26.3 19.8 4,985
3/12/77 25 22.9 15 37.0 15.7 4,993
6/78 20 21.6 10 51.6 13.9 4,301
6/79 10 99.5 10 99.5 10.1 4,301
Source: Central Bank of Chile.

Note: Dates refer to the official decrees on general changes of custom tariff rates between
December 1973 and December 1977. On the latter date the government issued a decree of
monthly adjustments that lasted until June 1979.
6
De la Cuadra and Hachette (1992) and Hachette (2000) detail the trade reforms of
the 1970s.
Repeated official statements implied that the reduction of protection for

import-substituting firms had finished in August 1977. Three months later,
however, the finance minister announced a target tariff of a uniform 10% to
be reached in mid-1979. The additional tariff liberalization was carried out
in monthly steps between December 1977 and June 1979. By the later date,
a uniform tariff of 10% was being charged on nearly all imports into Chile.7
Non-tariff restrictions, which were generally inefficient and generated capi-
tal gains, had been mostly eliminated; and several hundreds of tariff exemp-
tions in favor of firms, regions, or individuals had been suppressed. But price
bands softening the transmission of external instability (such as on wheat,
oil seeds, and sugar) were also eliminated. Imports of capital goods, which
had been generally exempt from tariffs, now became subject to the uniform
tariff, although they could be paid in installments. Evidently, relative prices
changed against imports of capital goods and in favor of consumer goods,
as well as reducing the cost of exportables.
Thus the process of discussion finished in the government with a clear
predominance of the most orthodox approach. In fact, the 10% flat rate
was unusually low for developing countries in the 1970s, and even in devel-
oped countries a uniform rate was exceptional. Some comparative informa-
tion illustrates this. In a newly industrialized country such as South Korea,
even after a decade of implementation of its successful economic reforms,
tariffs still ranged between 0 and 150%, with many locally produced items
enjoying nominal rates of protection between 30 and 60%.8 Korea as well as
Taiwan and other Asian countries had carried out a profound trade reform
led by export promotion rather than by import liberalization (see Ffrench-
Davis, 2006, Chapter IV).
Developed countries, despite their position at the forefront of world
industry, usually maintained selective (non-uniform) rates of protection,
with levels notably higher than 10% for significant groups of items. Effective
tariff protection in the United States, Japan, and the European Economic
Community was relatively high for product categories such as textiles and
clothing, processed food, and light manufactures. For instance, textiles and
clothing enjoyed effective tariff protection of about 40% in these countries
(Roningen and Yeats, 1976; Baldwin, 1981).
After 1973, public officials stated repeatedly that the real exchange rate
would depreciate as effective protection dropped (DIPRES, 1978, pp. 275
and 291). The government declarations implied a naive view of causal
7
With this ends the “first trade reform.” The “second,” which initially reintroduced
diverse forms of protection, began in 1983; see Chapter V.
8
See Balassa (1977, pp. 148–51) and Frank et al. (1975). For other developing countries
analyzed in a comprehensive project by the National Bureau of Economic Research
(NBER), see Bhagwati (1978). All these authors support the use of selective moderate
tariffs. See also Sachs (1987).
relationships in the economy, among others assuming no capital flows.

In practice, however, large financial inflows implied significant deviations
from this supposedly univocal causal relationship.
In fact, during periods in which the most sizable tariff reductions were
made, the real exchange rate appreciated in tandem. Table II.2 shows, for
selected dates, the evolution of the real exchange rate (col. 1), nominal rates
of protection for items subject to the maximum and average tariffs (cols 3
and 4), and the total cost per dollar of imports (col. 5). Dates were selected
according to the evolution of the cost of imports and the relation between
the two components that are considered here: the exchange rate and nomi-
nal tariffs.9 From this information, we can distinguish four phases in the
liberalization process (see Figure II.1).
The tariff reductions in phase I, which lasted until April 1975, were made
when exchange rates were highly depreciated. Moreover, a significant pro-
portion of the notably high tariffs was then redundant, which meant that the
initial reductions did not have substantial effects, as they represented mostly
Table II.2 Cost per dollar of imports, 1973–82

Phase Date Real Change of Nominal tariffs (%) Average Change of
exchange (1) in each total (5) in each
rate phase (%) Maximum Average exchange phase (%)
rate
(1) (2) (3) (4) (5) (6)
10/73 59.20 220 94 114.85

I 67.2 31.0
4/75 98.98 120 52 150.45
II –39.4 –51.4
7/77 59.96 45 22 73.15
III 14.2 2.9
6/79 68.46 10 10 75.31
IV –31.3 –31.3
6/82a 47.02 10 10 51.72
Source: Calculations based on Central Bank of Chile, Boletín Mensual, several issues, Ffrench-Davis
(1984), and Table II.1.
Col. (1) is the nominal exchange rate deflated by the corrected CPI (see Cortázar and Marshall,
1980) and multiplied by the index of external prices (IEP; see Ffrench-Davis, 1984), in 1986 prices.
Col. (5) is obtained by multiplying col. (1) by [1 (4)/100]. aMonthly average until the day before
devaluation.
9
The selected dates are somewhat arbitrary, since the real exchange rate levels are
sensitive to the price indexes used to calculate them. The average tariff can also be
calculated in very different ways. I present here the simple average used by different
government and independent sources. It is well known that the simple average is
highly sensible to the extent of disaggregation.
110
Phase I Phase II Phase III Phase IV
100
100
90 80
Real exchange rate
RER, 1986 = 100
Average tariff, %
80
60
70
60 40
Average
50 tariff
20
40
30 0
1973
1974
1974
1974
1975
1975
1975
1976
1976
1976
1977
1977
1977
1978
1978
1978
1979
1979
1979
1980
1980
1980
1981
1981
1981
1982
1982
1982
Figure II.1 Average tariff and real exchange rate, 1973–82.
Source: Tables I.1 and I.2.
“water” in nominal protection. Because of this, there were no significant

increases in “non-traditional” imports during this phase. During phase I,
the average cost of a dollar’s worth of imports increased by 31%.10 Hence, the
reform process was principally one of “rationalization” in which the large
dispersion of effective protection rates was diminished without causing a
substantial impact on the manufacturing of import substitutes. However, the
latter industries suffered the negative effects of the strong domestic recession
then in progress. Import liberalization and exchange rate depreciation did
have a positive effect on exportables; given that exports previously enjoyed
customs exemptions for imported components, it was the exchange rate
policy that had stronger effects on increasing exports (Ffrench-Davis, 1979).
The situation faced by importables changed substantially during phase II,
which lasted from April 1975 until mid-1977. During this period, reductions
in nominal protection were more effective than in phase I, dropping from
an average of 52 to 22%, and the maximum nominal protection decreased
from 120 to 45%. The exchange rate appreciated by 39%, strongly reinforc-
ing the effects of falling tariffs. Thus, for two years, while tariff liberaliza-
tion rapidly reduced the effective protection, the real exchange rate was
revalued in parallel. Consequently, the thirty-point reduction in the average
10
Obviously, the nominal tariff for several items, in particular for consumer goods,
dropped notably more than the average tariff. The corresponding reduction in effec-
tive protection was even sharper. Therefore, for many of these categories the real
exchange rate depreciation did not compensate for the effects of tariff liberalization.
Data on effective and nominal tariffs before tariff liberalization appear in Behrman
(1976, pp. 137–44).
nominal tariff meant a 51% fall in the average cost of each dollar of imports
(Table II.2, col. 6). There was little chance for the market to gradually
adjust to the strong impact of this rapid liberalization, unexpected because
it contradicted repeated official statements that the exchange rate would
“indissolubly” compensate for the dismantling of tariffs.11 The net result of
tariff reforms in phase II was a rapid increase in non-traditional imports,
particularly non-food consumer goods.
In phase III, which lasted until mid-1979, when the nominal price of the
dollar was frozen, the exchange rate was periodically adjusted to compen-
sate for the newer tariff reductions. Consequently, at the end of phase III
the average cost of the import dollar was about the same as at its beginning.
Naturally, products that were relatively more protected at the start of phase III
lost their privileged position as customs duties converged towards a uniform
10%. These changes occurred in an economy growingly more sensitive to
the evolution of the international economy, including a constantly rising
import coefficient. The average tariff had dropped from 94 to 10% during
these three phases (and non-tariff restrictions had disappeared), while the
real exchange rate at the end of the process was only 16% above the price
at the outset of the reform. Exports also benefited from the broadening of
the range of imported inputs, then either liberalized or subject to a uniform
tariff. Substitutors, on the other hand, had to compete with imports that on
average were 34% cheaper than in late 1973.
Finally, in phase IV the real exchange rate was revalued steadily. This
appreciation was a consequence of the fixed nominal rate and a domestic
inflation that, from above, converged only gradually to the international
rate during the three years in which the exchange rate remained frozen.
In the end, foreign currency finished 21% cheaper than at the beginning of
the reform, and the average cost per dollar of imports was 55% lower (see
Table II.2). In June 1982, this phase ended with an abrupt devaluation and
a deep recession in process.
In summary, import policy took shape with successive official announce-
ments, with each one of them presented as the final one. Thus, the policy
evolved from a moderate opening to trade, explicitly and officially declared
to be consistent with the process of Andean integration, to a sharp, across-
the-board trade liberalization. As for the assumed compensatory role of the
exchange rate, the facts show that in general this macro-variable did not
behave according to the assumptions of the economic model. Therefore, an
economic and social outcome that differed so much from that expected by
the government is not surprising.
11
In this phase, Chile left the Andean Pact. As a result, producers of importables
faced less Andean competition, while exporters lost their significant inroads in that
market.
2 Import composition
The drastic changes in the structure and level of protection had a significant
impact on the composition of imports. As could be foreseen, imports of con-
sumer goods (particularly non-food), which had previously been the most
restricted category, were the most favored by across-the-board liberalization.
Many variables other than trade policies affected the behavior of imports.
Among the most significant variables were the sharp contraction in aggregate
demand in 1975–6 and the recovery in 1977–81, the low investment ratio
during nearly the entire liberalization period, and the rise in oil prices.
Between 1970 and 1981, total imports increased by 127% in real terms
(104% excluding oil and lubricants), as shown in Table II.3.12 We must recall
two factors. First, until 1980 imports of equipment and machinery, as a
share of GDP, were significantly below the level of 1970 and insufficient to
raise productive investment and recover the 1960s growth rates. Second, per
capita GDP barely increased by 10% in 1970–81, while per capita imports
other than equipment and machinery grew by 115%; that is a gross income
elasticity of 11. Evidently, increased imports were not associated with an
income effect, but were predominantly due to import liberalization and
exogenous changes in the supply and demand of importables (e.g. oil prices
and changes in income distribution).
Different import categories behaved heterogeneously. The influence of
liberalization can be observed mainly in non-food consumer goods, where the
greatest number of “new” non-traditional imports is concentrated. Purchases
of non-food consumer goods increased by 534% between 1970 and 1981. The
share of machinery and equipment in total imports dropped from 21 to 11%,
and their participation in GDP fell by one-tenth between 1970 and 1981, with
an even more noticeably lower share in the intermediate years. This drop
reflects a depressed productive investment ratio in the period.
Due to a reversal of the import substitution of final products, a reduction
of demand for intermediate goods took place. Consequently, net expenditure
in foreign currency involved in these “new” imports was less than their gross
value. Additionally, relaxed or dismantled national integration requirements
for manufactures caused an increase in the share of imported components
in consumer goods in which substitution survived. This occurred in the car
industry, for instance. Likewise, gross industrial output figures tended to
overestimate the level of domestic activity due to the falling integration of
domestic inputs.13
12
Given that 1973 presents significant distortions, in Table III.3 I used figures for
1970, which was considered a relatively “normal” year. Real figures are nominal dol-
lars deflated by the index of external inflation faced by Chile.
13
This was the main source of overestimation of GDP, in the national accounts calcu-
lated with fixed coefficient matrices. See Marcel and Meller (1986).
Table II.3 Main imports of consumer goods and total imports, 1970, 1980, 1981
(1977 US$ millions)
1970 1980 1981 Change 1970–81
(%)
Confectionery items 0.2 8.2 10.5 5,150

Leather and fur manufactures 1.3 9.0 17.5 1,246
Alcoholic beverages and cigarettes 1.1 22.8 27.5 2,400
Carpets, clothing, knitwear and 24.8 171.9 271.6 995
fabrics
Photographic and cinematographic 8.0 17.4 25.2 215
products
Footwear, hats, and umbrellas 2.1 24.0 43.3 1,962
Musical and optical instruments 4.4 18.1 28.7 552
Toys and recreational goods 3.5 32.0 42.4 1,111
Processed foods 5.3 34.6 41.3 679
Perfumery and cosmetics 0.1 13.7 19.6 9,500
Television sets 0.7 49.0 66.2 9,357
Radios 4.7 46.0 45.8 874
Cars and motorcycles 19.5 144.4 263.9 1,249
I Total main non-traditional 75.7 591.1 902.6 1,093

consumer imports
II Wheat, maize, and sugar 43.6 309.9 262.1 757
III Fuels and lubricants 118.0 666.9 689.5 484
IV Other consumer and 1,155.5 1,561.2 1,714.3 48
intermediate goods
V Transport equipment 157.4 317.5 395.8 152
VI Other capital goods 408.6 376.6 480.6 18
Total imports 1,959 3,823 4,445 127
Source: Calculations based on National Customs Authority for 1970 and import applications of
the Central Bank of Chile for categories II, III, V, and VI in 1980 and 1981. Current dollars were
deflated by the IEP.
Table II.3 shows the groups of consumer items (food and non-food) that
increased most between 1970 and 1981. The thirteen groups disaggregated
in this table cover 50% of all consumer goods imports in 1981. During the
eleven-year period they grew by 1,093% (i.e. twice the increase in fuels),
while all other imports expanded by 62%. As can be seen, most of these
non-traditional imports were items that had traditionally been considered
“luxury” goods. In several cases, these new imported items were not locally
produced, even though they did replace similar domestically produced
items. There was, then, a significant diversification in the composition of
consumption.
The consumption of goods whose imports grew most significantly was
concentrated among high-income groups. Naturally, this phenomenon
was related to the worsening of income distribution, although this effect
was magnified by the fact that the distribution of consumption is more

concentrated for these importables than for total consumption.
Undoubtedly, the low-income sectors were able to buy new varieties of con-
sumer products, but these purchases were bound by their limited and declin-
ing earnings. According to household budget surveys conducted in Santiago
by the National Bureau of Statistics, families in the upper quintile increased
their share of total consumption from 44.5 to 51% between 1969 and 1978,
while families in the lowest quintile decreased their share from 7.6 to 5.2%
(see Chapter VII, Table VII.2). The 1978 survey also reflects the concentration
of consumption of non-traditional imports that year (see Table II.4).
In eleven of the thirteen groups of importables included in Table II.4, the
richer quintile consumed a higher percentage than its share in total expend-
iture (as said, 51% in 1978). The two groups in which this richer bracket
purchased proportionally less than its average include “inferior goods”
(negative income elasticity) for the higher-income levels. For example,
as income rose purchases of black-and-white television sets were replaced
with purchases of color televisions, and transistor radios were passed over
in favor of stereo sets and tabletop radios.
Table II.4 Shares by income brackets of consumption of imported goods: main

non-traditional imports in 1978 (% of total consumption)
Item Highest 20% Middle 60% Lowest 20%
of households of households of households
Color television sets 100.0 0.0 0.0

Automobiles 98.6 1.4 0.0
Imported whisky 94.0 6.0 0.0
Imported cigarettes 92.0 8.0 0.0
Cassettes 72.8 26.8 0.4
Tennis racquets 71.8 28.2 0.0
Electric blenders, mixers, 71.8 28.2 0.0
and food processors
Motorcycles 65.3 34.7 0.0
Watches 59.7 34.7 5.6
Toys 56.1 41.4 2.5
Stereo equipment, record 51.3 48.5 0.2
players, and tape recorders
Transistor radios 32.9 57.8 9.3
Black-and-white television 18.8 71.2 10.0
sets
Share of each bracket in 51.0 43.8 5.2
total consumption
Source: Instituto Nacional de Estadísticas, III Encuesta de Presupuestos Familiares, vol. 3, Santiago,
May 1979.
Note: Only includes those products that could be identified in the survey as “non-traditional”
imports. Unless otherwise specified, includes consumption of domestic and imported goods.
3 Effects of trade liberalization on manufacturing
Evaluating the effects of import liberalization is a complex task. First, the

effects were significantly different in each phase. Second, many other impor-
tant changes took place simultaneously. On the one hand, there was a drop
in aggregate demand, wages, and investment, and a rise in unemployment;
these factors strongly influenced the nature of the adjustment process. On
the other hand, export expansion, which began before the impact of liber-
alization was substantial (especially in 1974), contributed to the recovery of
economic activity and offered opportunities for investment in the produc-
tion of exportables.
In this section, first, I examine the global changes in the industrial sector.
In the second part, I try to progress to a disaggregated analysis by sectoral
branches.
(a) Macroeconomic effects

Manufacturing production was drastically affected by the economic reces-
sion of 1975. That year, industrial output dropped by 26%, while GDP fell
by 17%. As could be predicted, this meant that there was room for a subse-
quent intensive recovery. Actually, subsequently, the manufacturing sector
exhibited high rates of “growth” from 1977 to 1979. Still, this was only
a partial recovery. Moreover, it lost velocity in 1980 and fell off in 1981.
Evidently, the invasion of imported consumer goods in 1980–1 played a
crucial role in this outcome. By 1981, eight years after the neoliberal eco-
nomic policy started to be implemented, industrial value-added per capita
was 18.5% lower than in 1973. Table II.5 depicts the contrast between
the evolution of the manufacturing sector in Chile and in developed and
developing countries.
This performance resulted in a notorious drop in the share of manufac-
turing in GDP, from 26% in 1970 to 20% in 1981. Finally, this deterioration
was also revealed in industrial employment: since 1975 it remained mark-
edly below that in 1970.14 This was caused, partly, by the diminished impor-
tance of labor-intensive industrial activities such as textiles and garments.
There was also a drop in employment in areas where output increased but
where value-added per unit of gross output decreased, which naturally
resulted in a reduction in labor demand. Finally, without even having
14
Several studies assert that manufacturing was not affected negatively by trade
liberalization. The causes of that contradictory conclusion, in sharp contrast with
evidence, are: (i) the comparison is made with 1975, after the great fall of that year
or (ii) the comparison centers on the eighties, disregarding that liberalization took
place in the seventies and that tariff and exchange rate protection were reintroduced
(though mildly) in the eighties.
Table II.5 Manufacturing output: Chile and the world economy, 1974–82
(1973 100)
Total value-added (VA) Chile
Year Industrialized Developing Chile Value-added Industrial

countries countries per capita employment
(1) (2) (3) (4) (5)
1974 100.1 106.3 99.1 97.5 97.5

1975 91.8 108.1 73.0 70.6 88.8
1976 100.1 116.7 74.9 71.4 86.1
1977 103.7 125.3 79.9 75.1 87.1
1978 107.9 133.6 85.0 78.8 88.8
1979 113.3 139.7 91.0 83.1 88.2
1980 112.3 146.8 93.3 83.9 88.8
1981 112.8 147.0 92.1 81.5 87.3
1982 108.5 149.6 72.8 63.4 71.0
Source: For Chile, calculations based on central bank data, Jadresic (1986), and Marcel and Meller
(1986). For developing and industrialized countries, United Nations, Monthly Bulletin of Statistics,
May 1983.
recovered its 1973 or 1974 levels, the sector suffered another spectacular
drop of 21% in 1982.
Let us return to 1974.15 The actual level of industrial production in 1974
underestimates the actual productive capacity: a sharp contraction of aggre-
gate demand had already begun in the second half of 1974 and was nega-
tively affecting the annual level of output.16 The effect can be partially seen
by observing that between October 1973 and September 1974, that is, dur-
ing the first twelve-month period of the new government, industrial output
was 3.4% higher than during the 1974 calendar year.
From mid-1974 until 1981, the sector operated below its productive
capacity and significantly below its historical trend. The gaps between
actual and trend supply were caused by a series of events. Output losses in
1975 were due primarily to the contraction of aggregate demand, led by
15
The manufacturing value-added had expanded by 5.9% per year in the sixties.
Nevertheless, the assertion that the industry had been “destroyed” by 1973 is
disproved by the output performance immediately after the military coup. In general,
there was not “destruction,” but growth did stop in 1971–3. The latter is one of the
costs of macroeconomic imbalances and lack of discipline in those years.
16
There was a large drop in real wages in 1973 and 1974; the surveys conducted in
January, April, and July 1974 by INE show that the average decrease was 16% com-
pared to the same months in 1973. The decrease strongly influenced sectors that
produced goods intensively demanded by middle- and low-income consumers. An
outstanding example was that of the textile and clothing sectors.
recessive external shocks.17 During the 1977–81 period, however, production

responded principally to the changes in demand patterns induced by the
across-the-board import liberalization with exchange rate appreciation, the
concentration of income, and the drop in domestic investment as a result
of a recessive and unstable macroeconomic environment and high outlier
interest rates (Mizala, 1992; Agosin, 1998). In fact, negative pulls for imports
de-substitution were significantly stronger than the positive pulls for spe-
cialization in some importables and export dynamism.
It is undeniable that the specific way in which import liberalization was
implemented contributed strongly to the overall poor performance of the
industrial sector and of the entire Chilean economy between 1973 and
1981. The productive capacity of the sector was seriously damaged, and
many firms needlessly went bankrupt; survivors, naturally, tended to be
strong. Altogether, despite the evident need to reduce the average level and
dispersion of effective import protection in 1973, the net balance appears
to be overwhelmingly negative for manufacturing. Disaggregated data pre-
sented in the next subsection provide additional support for this assertion.
(b) Effects on the structure of manufactured output

The structure of industrial output changed significantly during the 1970s.
To identify the impact of import liberalization more precisely, I now exam-
ine the behavior of the different production branches.
The close relationship between the domestic industrial sector and for-
eign trade can be seen first in the global evolution of exports and imports
of manufactured goods. Exports grew significantly from 1974 until they
totaled 10% of the gross value of the sectoral output in 1981, while imports
rose to 35%. The respective figures for 1969–70 were 3 and 17%. It is well
known that in the real world the behavior of exports is not univocally
tied in a unique way to import policy (see Bhagwati, 1978); in fact, export
promotion can be fully consistent with a policy of selective import substi-
tution, as the East Asian countries have demonstrated beyond any doubt.
Consequently, the effects of import and export policies on domestic produc-
tion can be analyzed separately.
Information broken down into twenty-nine groups (ISIC, rev. 2, three
digits) compares the 1969–70 average to 1978; that is, a year before the
freezing of the exchange rate in 1979. The composition of consumption,
17
The contraction of aggregate demand was associated in part with a significant
worsening of terms of trade beginning in the second half of 1974 (see Chapter V).
The negative terms of trade shock was equivalent to 6.4% of GDP in 1975 compared
to 1972. It must not be forgotten that this worsening followed a strong 5% improve-
ment in 1973–4. In fact, during the first year in which the new economic policy was
put to work, the copper price was extremely high. Nevertheless, the transitory high
revenues were not saved but disbursed as they were received.
output, and trade changed substantially during the decade (see Vergara,
1980). First, trade was dynamic in the sense that exports increased in sixteen
groups while imports rose in eighteen; both imports and exports grew in ten
of these groups, implying that, at three-digit information, there was intra-
industrial specialization. There were drops in both output and consumption
in seventeen branches. This suggests that domestic demand decisively influ-
enced output in this period.
At this level of disaggregation, many of the groups include goods whose
production processes and marketing channels are quite diverse. Notwith-
standing this heterogeneity, the data allow us to advance some conclusions
(Vergara, 1980; Foxley, 1983, Chapter 3).
First, only two groups show output growth that is associated with exports
(wood and paper); in two other groups, exports play a significant role (food
and industrial chemicals), even though their role is not as dominant as in
the preceding groups.18 When the data are broken down still further, it can
be shown that a large share of export expansion was concentrated in only
five sectors: pulp and paper, wood, molybdenum oxide, fish meal, and semi-
manufactured copper. After a sizable diversification in the period 1974–6,
the share of these products fell in 1976 to 58% of industrial exports, but in
1978 it rose to 64% and in 1981 to 66%.
For imports, diversification was greater, as shown in section 2. This diver-
sification is also reflected in the large number of groups for which imports
became significant. Three branches were strongly affected by imports:
electrical machinery, transport equipment, and professional equipment. In
the first two groups, the rise in domestic demand ameliorated the negative
impact of imports on output, while in the third group a receding domestic
demand exacerbated the negative impact. In six groups, opening to trade
together with a significant reduction in domestic demand led to a decline
in output (textiles, clothing, leather, oil derivatives, pottery and china, and
non-electrical machinery). In four other groups (footwear, printing and pub-
lishing, non-metallic minerals, and iron and steel), the determinant vari-
able in declining output prior to 1978 seems to have been the decrease in
domestic demand. Data on the other groups are more difficult to interpret,
as the results depend largely on which years are compared and the method-
ology used to estimate the real change of each variable.
As shown, the behavior of domestic demand had a determinant effect
on the level of output. This effect makes it difficult to evaluate the impact
of import liberalization, while depressed domestic demand contributed to
increased exports of items in excess supply in the local market. Naturally,
to the extent that domestic demand recovered subsequently, the relative
weight of different variables changed; thus, after 1978 the effects of import
18
The main food products exported are fish meal and frozen seafood. The most
important exported chemical substance is molybdenum oxide.
68
liberalization gained importance vis-à-vis aggregate demand as an explana-

tory factor of the poor performance of manufacturing. It was clear that by
1978 the effects of import liberalization had not yet been fully felt. At that
stage, the growth of export quantum was losing speed, while imports, par-
ticularly of consumer goods, were rising quickly. This trend was at work in
1978 and 1979, before the nominal exchange rate was fixed. Subsequently,
the sharp real appreciation of the peso that followed reinforced the lagged
effects of import liberalization. The negative impact of non-traditional
imports increased as compared to changes in domestic demand and the pos-
itive effects of exports. Aggregate demand became more import-intensive,
the quantum of non-resource-based exports ceased growing in 1980, and
imports (especially of consumer durables, as shown in Table II.3) rose in
1980–1 notably faster than in the previous two years.
The sector had adjusted to foreign competition in three ways. Some went
into bankruptcy or closed down plants. In other cases, firms began to special-
ize within the industry in two ways: merging with other firms and suspending
lines of production within a firm. Finally, some firms also began to import
goods that they marketed in place of those they had previously produced.
Marketing imported products enabled firms affected by import liberaliza-
tion to capitalize on the relative advantage they had because of their sales
outlets and knowledge of demand. This adjustment mechanism had several
interesting effects. First, in this case producing and importing were not inde-
pendent functions but were managed by the same decision unit; thus, for a
while foreign competition would be operating in a more limited fashion than
is assumed by orthodox theory. Second, a larger share of firms leaned towards
commercial and financial activities rather than producing goods. The extent
of this bias is shown by changes in the composition of GDP recorded in the
national accounts (see Chapter I, Table I.2). Third, the resulting growing cur-
rent account deficit, which was created by the asymmetrical response of pro-
ducers in sectors hurt by new trade policies and in sectors favored by them,
was financed by an unsustainable increase in foreign debt. Fourth, although
some producers defended themselves by switching to marketing imports,
this change negatively influenced employment; in fact, productive employ-
ment decreased per unit of sales and even per unit of output.19
4 Towards an evaluation
In these concluding remarks, I look first, briefly, at the global effects of trade
policy on the balance of payments. Next some points are raised about the
19
This is a microeconomic increase in productivity. However, it has negative social and
economic consequences (i) when it implies a greater reduction in employment than
in production, instead of a greater increase in production than in employment, and
(ii) when it takes place within a framework of widespread unemployment and wors-
ened income distribution, as was the case.
effects on the efficiency, dynamism, and competitiveness of the Chilean

economy.
(a) Balance of payments and the current account

Practically all components of foreign trade expanded during the period under
study, especially non-traditional imports and exports. The export quantum
recorded a most dynamic behavior. After an abrupt jump and diversifica-
tion in 1974–6, exports moderated their increase until 1980. Undoubtedly,
exports became the dynamic productive sector of the Chilean economy.
Actually, they recorded a remarkable 13.6% annual expansion in 1974–81.
Exports even provided some dynamism to the highly depressed agricultural
sector. But the net effect was limited since, after “enforcing discipline” in
the rural sector in 1974, growth reached merely 2% between 1974 and 1981.
The dynamic agricultural exports, consisting mainly of fruit and forestry
products, shared the sector with traditional agriculture, which decreased
spectacularly (especially in grains, sugar beet, and oil seeds).
The expansion of imports was much larger than that of exports, leading
to a growing trade deficit that increased markedly between 1976 and 1981.
A component that deviated most from “normal” values was imports of
equipment and machinery, whose share of total imports lost ten percent-
age points; they fell to merely 11% of imports, notwithstanding that the
Chilean GDP growth was highly dependent on imported capital goods.
The trade deficit was affected by the copper price,20 which recorded a level
one-fifth lower than the “normal” or trend price.21 Nevertheless, foreign
exchange proceeds from copper exports increased as a result of two factors.
On the one hand, large investments made between 1967 and 1970, with
long maturing lags, made it possible for copper output to increase by 50%
immediately after “discipline” was imposed in 1973. On the other hand, the
1971 nationalization of the large mines allowed the government to capture
a greater share of the economic rent derived from the rich Chilean copper
deposits. These two positive effects were “permanent,” while the low price
could be assumed to be “transitory.”
A huge current account deficit in 1981 was covered with extraordinarily
and unsustainably large capital inflows. These inflows, which were primarily
directed to the private sector, not only financed the current account deficit
but allowed a balance of payments surplus and the respective accumulation
of international reserves (Table II.6). In the meantime, external debt was
20
A component that showed a notable change in the opposite direction was the price
of molybdenum, a copper byproduct whose real price increased sixfold during the
1970s. The higher value of this export was equivalent to 46% of the increased expen-
diture on oil imports in the same period. See in Chapter VI, Box VI.1 for the role of
molybdenum in the external accounts.
21
See an early methodological discussion on the “normal” or trend copper price in
Ffrench-Davis (1973, Chapter IV).
70
Table II.6 Balance of payments, 1973–82 (1977 US$ millions)

1973 1974 1975 1976 1977 1978 1979 1980 1981 1982
I Current account –441 –256 –536 159 –551 –960 –925 –1,361 –3,213 –1,600
1 Trade balance 32 434 76 693 34 –376 –276 –527 –1,817 43
Exports (fob) 1,968 2,612 1,734 2,282 2,185 2,171 2,984 3,250 2,605 2,573
Copper 1,577 1,971 947 1,330 1,161 1,076 1,469 1,467 1,180 1,170
Non-copper 391 641 787 952 1,024 1,096 1,515 1,782 1,425 1,403
Imports (fob) 1,936 2,179 1,657 1,588 2,151 2,547 3,260 3,777 4,422 2,529
2 Non-financial services –328 –478 –313 –212 –295 –214 –186 –269 –476 –385
3 Financial services –165 –225 –310 –352 –365 –432 –525 –642 –994 –1,334
4 Unrequited transfers 20 13 11 30 75 62 62 78 74 76
II Net capital inflowsa 507 94 305 143 669 1,589 1,740 2,220 3,259 791
III Balance of payments 65 –163 –230 303 118 628 815 859 45 –809
Source: Calculations based on data from the Central Bank of Chile, Balanza de Pagos.
Note: The figures were deflated by the IEP (see Ffrench-Davis, 1984). aIncludes errors and omissions.
climbing fast, mostly devoted to finance consumption of imports (see

Chapter III).
Given the evolution of the sector, the authorities held on to two valid
assertions, although with a faulty interpretation of their implications. It was
stated that if the exchange rate were to be liberalized by the Central Bank
the rate would strongly appreciate. This evidently would have happened,
confirming that foreign exchange flexibility under a capital surge tends
to exacerbate imbalances (see Chapter IX and Harberger, 1985). A lesser
evil had been the fixing of the exchange rate at $39, but the most efficient
policy would have been to moderate capital inflows and to manage a real
devaluation in order to enhance a positive restructuring of production. The
other weak point was the assumption that the excess of imports would slow
down spontaneously. Slowing did tend to occur but with a significant lag.
Additionally, since the exchange rate continued to appreciate until 1982,
it stimulated a fast-rising expenditure increasingly intensive on imports,
which crowded out domestic output and savings. These effects left indelible
footprints on the Chilean economy over a prolonged time span. As under-
lined in Chapter III, the deficit on the current account of 21% of GDP in 1981
revealed an impressive and absolutely non-sustainable disequilibrium.
(b) Efficiency, dynamism, and competition

The theoretical foundation of import liberalization asserted that the market,
free of government interference, would allocate resources according to a sort
of unmistakably identifiable “comparative advantage.” Actually, market com-
parative advantages depend on the level and stability of the exchange rate, on
the degree of activity in domestic and world markets, on international price
fluctuations, and on many other factors (such as the availability of long-term
funding, infrastructure, and trained labor). Additionally, market compara-
tive advantages tend to diverge from social comparative advantages because
of the disequilibria and distortions characteristic of developing economies;
particularly relevant are the presence of severe unemployment and the
obstacles to acquiring dynamic competitivity. The differences between the
two can be striking, and highly costly, in a country facing a radical change
in several economic policies, with high capital and labor unemployment,
and inconsistencies among reforms, such as import liberalization and real
exchange rate appreciation.
(i) The macroeconomic framework and efficiency

The efficiency of any economic policy depends on the context in which it is
applied. The generally recessive domestic economic situation of this period
was relatively conducive to export promotion, but on the other hand it
increased the transition costs of import liberalization. Consequently, that
macroeconomic and reforming environment would have required a trade
opening led by export promotion, as East Asia had done, instead of the
approach led by import liberalization, as Chile did in the 1970s and Latin
America did in the 1990s (see Ffrench-Davis, 2006, Chapter IV).
The expansion and diversification of exports provided an outlet for excess
production that otherwise would not have had a market. In fact, the sharp
contraction of domestic demand (in particular, during the period from mid-
1974 to 1976) left a significant share of industry with underutilized installed
capacity. The prevailing depreciated exchange rate, the access to the Andean
market, and the efforts of PROCHILE (the government export promotion
agency) supplied a market abroad for many firms with excess capacity. Later,
productive capacities were expanded, and non-traditional and traditional nat-
ural resources were exploited, enhanced by prior investment during the 1960s
that developed the bases for the forestry, fruit, and fishing industries, and the
accelerated liberalization of imported inputs. Dynamism was reversed in the
early 1980s, when an overwhelming exchange rate appreciation induced a
generalized drop of exports. In general, the expansion of non-copper exports
promoted efficiency in the allocation of resources, principally through an
increase in the rate of utilization of capital and labor, the exploitation of natu-
ral or acquired comparative advantages, and the generation of externalities.
For imports, the situation was the opposite. In fact, if a trade liberalization
process goes too far, is too rapid, or is undertaken at the wrong juncture,
it will provoke premature and unnecessary plant shutdowns, the underu-
tilization of capital and labor, and a decrease in investment. Therefore, to
evaluate the effects on the Chilean economy we must distinguish between
the different stages of the liberalization process and take into account the
macroeconomic context in which it was implemented.
In phase I, clearly redundant levels of protection were eliminated. But
the first tariff reductions served to limit national producers’ capacity to set
monopolistic prices. In a second stage (phases II and III), the additional tariff
reductions, which lowered the maximum nominal protection from 120 to
10%, had a much greater effect. The most painful part of import liberalization
was carried out rapidly, with misleading announcements, and its negative
effects were reinforced by exchange rate revaluations. This policy was imple-
mented during a time in which wages were deteriorated, domestic demand
was very depressed, and open unemployment was remarkably high.
First, wage repression acted as an artificial protection mechanism that,
although it was obviously regressive, compensated for reduced tariffs on
imports. In fact, in 1976 the ratio of indexes of wages/exchange rate was
less than half that of 1970, and in 1979 it had recovered to only two-thirds
(see Ffrench-Davis, 1981).
Second, given the depth of recession, recovery rates for demand and pro-
duction were bound to be high. Since tariff liberalization took place during
recovery, a naive examination of the data may lead to the wrong conclusion
that the liberalization process encouraged an output increase. As was
demonstrated earlier, however, the opposite occurred. The implementation of

free imports contributed to maintaining the recovery of domestic production
at a level notably lower than that of the recovery of aggregate demand;
output and aggregate demand became increasingly import intensive. The
recession itself negatively affected the efficiency of the process. The underu-
tilization of productive capacity and the extremely high real interest rates
(averaging 38% between 1975 and 1982; see Chapter III) tended to raise the
average cost of output to domestic producers, making it more difficult for
them to face foreign competition.
Third, high unemployment created a significant gap between market and
social comparative advantage. Thus, the opportunity cost of resources freed by
the sectors negatively affected by reform was lower than their market price.
(ii) Dynamism and efficiency

Here, the discussion will be confined to two interrelated points, concerning
the investment ratio and asymmetry of adjustments, and “dynamic com-
parative advantages.”
A remarkably low level of gross domestic investment contributed to an
asymmetrical adjustment: strong negative pulls and weaker positive pulls.
The widespread underutilization of installed capacity discouraged domes-
tic investment (see Solimano, 1990; Agosin, 1998; Ffrench-Davis, 2006,
Chapter III); the low investment ratio was also associated with the remark-
ably high real interest rates. It is obvious that resource reallocation is easier
in an economy with a high rate of capacity utilization and dynamic growth.
The stagnation exhibited by the domestic economy during this period made
it necessary for many of the hurt sectors to reduce absolute output in order
for the relative adjustment to take place. Limited sectoral and regional
mobility of resources and the reduced rate of investment were obstacles to
the effective reallocation of freed resources: it was predominantly the expan-
sion achieved in the export sector that compensated, but quite partially, for
the resulting lack of dynamism in the economy. As stated in Ffrench-Davis
(1979), a growing share of the scarce domestic investment was channeled
to the export sector. Investment was mainly concentrated in activities
intensive in natural resources. It was less significant in products intensive in
value-added and in “acquirable” comparative advantage.
A reliable hypothesis is that it became easier to identify comparative
advantages exhibiting a defined base of natural resources or already acquired.
The countless changes taking place in the Chilean economy – depressed
domestic demand, unstable and outlier interest and exchange rates, and
incomplete strategic markets – made it difficult to identify the whereabouts of
potentially acquirable comparative advantages. The diffuseness of compara-
tive advantages was, presumably, one additional factor explaining the low
actual investment ratio.
(iii) Competition and efficiency

One result the government expected from import liberalization was
increased “competitiveness” in the domestic market. This would be achieved
by means of the effective or virtual presence of foreign importables, which
would put an upper limit on domestic prices. It is undeniable that this did
happen to a significant degree. What also happened, however, was that
there were important deviations from the types of relationships that were
supposed to characterize a “competitive” economy.
First, a significant proportion of non-traditional imports belonged to cate-
gories in which product differentiation played a decisive role. Consequently,
competition among suppliers of these products was based on product
differentiation to a larger degree than on pricing. The segmentation of the
capital market (one example is the persistent gap between domestic and
foreign interest rates) also introduced competition based on the terms of
suppliers’ credit. These factors provoked effects quite distinct from those
that “competition” should have generated according to orthodox theory.
Second, marketing channels were not completely open to every competitor;
for instance, in a number of cases the producer of import substitutes became
an importer of “competitive” goods. Third, the abrupt-cum-appreciation
trade opening promoted greater concentration in the ownership of domestic
productive activities. This phenomenon was reinforced by the depression in
aggregate demand and by the manner in which the capital market operated.
These factors gave a significant advantage to economic groups that were
linked to financial activities and had access to foreign credit.
A crucial argument for free trade policy refers to the benefits that competi-
tion allegedly brings to consumers, with the availability of a wider variety
of goods and lower prices. Within the framework of orthodox consumer
theory, the opening to foreign trade is seen as positive because it allows
demanders to equalize their marginal utility with the marginal cost of
importing (which is assumed to be the same as the international price for
a “small country”). The diversification of consumption is seen as welfare
increasing, as it would increase the freedom of choice of consumers. Ceteris
paribus this is perfectly true. However, it is relevant to add two comments,
concerning the indirect effects on consumers in their role as producers and
on equity. First, the “de-substitution” of imports during the adjustment
process generated unemployment and delayed the recovery of overall eco-
nomic activity. These discouraged investment, which in turn had a negative
impact on the creation of new job opportunities. Therefore, low-income
consumers (who suffered the highest levels of unemployment), in their
roles as producers (workers), bore much higher costs than the contingent
benefits derived from the diversification of the basket now available in the
market to those with purchasing power. Depressed producers cannot perform
as dynamic consumers.
Second, the diversification of consumption enabled a small, high-income

sector to rapidly adopt the consumption patterns of the well-off in the rich-
est economies. The notorious increase in inequality that took place during
these years (see Chapter VII) manifested itself in a noticeable differentiation
of lifestyles rather than in higher savings rates and productive investment.
The collapse in national savings supports this hypothesis (it dropped to
9.6% of GDP in 1981 and to 3.4% in 1982, in current prices).
5 Concluding remarks
The Chilean experience left heterodox lessons. The national economy in

1973 had excessive effective protection for numerous import categories;
therefore, a significant trade liberalization was required. However, trade
liberalization was excessive and ill-timed; it was not coordinated with
exchange rate and capital account policies, and disregarded the challenge
of completing underdeveloped factor markets. A gradual procedure should
have been adopted and a deliberate search for dynamic complementa-
tion between import substitution and export promotion should have been
undertaken in the East Asian style. Positive export expansion would have
been essentially consistent with a more pragmatic tariff policy than the one
adopted. It could have involved a greater conversion of import substitutes
into exportables, and a more diversified export basket that would have been
more intensive in value-added.
Consequently, a drastic dismantling of tariff protection, instead of a grad-
ual and more comprehensive reform, including all ingredients for productive
development, should not have been undertaken in a macroeconomic envi-
ronment such as the one predominant in the Chilean economy. Supporters
of the policy being implemented argued that if the policy had not been
implemented so rapidly it would have been impossible to carry it out at
all. The answer to this argument is threefold. First, it would have been bet-
ter not to undertake tariff reductions as drastic as those performed from
1975 (covering phases II and III) rather than attempting to impose them
amidst a depression, with high unemployment and low investment. As a
consequence, the corresponding de-substitution of imports, predictably, was
higher than naturally necessary, and therefore was inefficient in many cases.
Neither excessive protection nor extreme liberalization was the appropriate
solution. It is evident that “corner” solutions were not the optimal choice.
Second, foreign exchange appreciation should not have been allowed dur-
ing tariff liberalization. This misstep flagrantly contradicted the govern-
ment’s repeated assertions. Third, developed or complete domestic markets
for long-term financing, labor training and innovation were missing. These
generated a real environment that differed substantially from the theoretical
framework on which the arguments in favor of free trade were built.
Finally, the conventional hypothesis that an unrestricted opening to

imports would promote the expansion of labor-intensive tradables and
result in a contraction of capital-intensive activities was contradicted by
the characteristics of changes that occurred in the productive structure.
This performance was directly linked to the richness and characteristics of
natural resources, but also to the macroeconomic framework in which
trade was liberalized, to the excessive intensity and lack of selectivity of the
reform, and to the absence of a comprehensive development strategy.
III
Domestic Financial Liberalization
and External Debt in the 1970s:
Building a Major Crisis*
The Chilean economy recorded great fluctuations in its economic activity

during the period 1974–81 (see Foxley, 1983; Ramos, 1986; Edwards and
Cox-Edwards, 1987). In all, GDP growth averaged 3% per year; a figure
notably lower than the 5.6% recorded by Latin America in the 1970s. Note
that in the year after the end of that period, 1982, the debt crisis brought
a generalized recession in the region; it was particularly acute in Chile. It is
extremely informative that the deepest recession in the entire region took
place in a Chilean economy that had eliminated the fiscal deficit, privatized
many public firms, sharply liberalized imports and the domestic capital
markets, and was praised by international financial markets and institutions
as a highly “successful” economy. The Chilean outcome shares causes and
results with those of Mexico in the “tequila” crisis of 1995, and with the
currency and debt crisis of Argentina in 1998–2002.1
The poor outcome had six relevant causes: (i) the rapid and indiscrimi-
nate import liberalization (especially of consumer goods); (ii) the exchange
rate appreciation in combination with sharp reductions in effective protec-
tion; (iii) the persistence of high real interest rates on the domestic capital
market; (iv) the overwhelming permissiveness resulting from the lack of
regulation of financial institutions and capital flows; (v) the composition of
inflows, with a minor share directly linked to productive investment, while
the increase in debt was principally directed to consumer imported goods;
and (vi) the difficulty of identifying comparative advantages in a macroeco-
nomic environment with a large output gap and “wrong” macro-prices.
* Partially based on “The external debt, financial liberalization, and crisis in Chile,”
in M. Wionczek (ed.), Politics and Economics of External Debt Crisis: The Latin American
Experience, Westview Press, Boulder and London, 1985. I gratefully acknowledge the
comments of researchers at CIEPLAN, and those of Robert Devlin, Rudi Dornbusch,
Jorge Marshall Silva, and Carlos Massad. I appreciate the assistance of José de
Gregorio.
1
In Ffrench-Davis (2006, Chapter VII) an analysis of the crises of several LACs since
the implementation of the “Washington Consensus” is made.
77
78 Building a Major Crisis
Finally, great vulnerability in the national economy was generated, which

resulted in the severe crisis of 1982. In fact, in view of the passive and
neutral domestic policies pursued, the authorities were left with weakened
policy tools to deal with external stocks. Furthermore, growing indebtedness
and the magnitude of the current account deficit obviously could not have
been sustained, even had the international financial crisis not taken place.
Consequently, the external sector was placed on a course that would inevita-
bly call for a traumatic adjustment process. The seriousness of this situation
was exacerbated, of course, by the fact that during the 1970s the domestic
productive base had been weakened given extremely low investment ratios.
In fact, the average gross investment ratio fell by almost one-fifth compared
to the 1960s and the GDP growth rate diminished proportionally more.
Section 1 starts with the official conceptual framework, then presents
the main features of the financial liberalization process and the policies
adopted. Section 2 presents the evolution of capital flows and external debt;
special attention is given to bank loans and the behavior of capital flows
received by private debtors. In section 3, the macroeconomic impact of
external indebtedness is examined, especially the effects on liquidity and
on exchange-rate policy, and the disequilibrating adjustment processes to
which it gave rise. Section 4 analyzes the evolution of interest rates, stress-
ing the persistent gap between domestic and foreign rates. In section 5, a
discussion of the various sources of external vulnerability to which Chile
was exposed as a result of its financial openness is made. I argue, contradict-
ing the neoliberal approach, that the problems were concentrated in the
private sector segment of the debt. I also attempt to explain why the mas-
sive financial inflows were accompanied by a drop in national savings and
a decline in the average investment ratio. Finally, section 6 summarizes the
lessons provided by the 1970s neoliberal experiment for capital account and
financial markets management. All are quite relevant for the early twenty-
first century.
1 Financial liberalization and foreign indebtedness2
A full liberalization of the domestic financial market and a rather broad

opening of the capital account were the official policy in Chile. This sec-
tion presents the conceptual framework on which the financial opening
was based and then examines the way in which the principles were imple-
mented. Finally, the volume and composition of capital flows and external
indebtedness are analyzed.
2
For more detailed analyses of domestic financial reforms in Chile, see Arellano
(1983), Eyzaguirre (1988), Zahler (1988), and Fontaine (1989). A comparative analysis
of Latin American cases in the 1970s and early 1980s is presented in Ffrench-Davis
(1983a).
Building a Major Crisis 79
(a) The analytical framework of the experiment

There are of course some very sophisticated neoliberal versions of how the
liberalization of capital markets should work and of its effects compared to
those of so-called financial repression. The essential aspects of the official
version can be conveyed by means of a very simple scheme, however.
The liberalization of the domestic financial market and the opening up
to capital flows sought to increase overall investment and to improve the
allocation of resources. It was expected that there would be an increase in
the volume of investment and its efficiency, which would provide the basis
for vigorous and sustained economic growth.
In simple terms, the conceptual framework may be described in Figure III.1.
Curves O and D represent the supply of and demand for funds in the domes-
tic market, which are identified with savings and investment, respectively.
To begin with, there prevails a situation of “financial repression” in the
sense that there are restrictions regarding the organization of new banks
and the operations they can carry out, and there is extensive rationing of
demand. Against this background, the authorities fix an interest rate (rc )
lower than the equilibrium rate for a closed economy (re ). This determines
a volume of savings, Vc, and demand is rationed at the same level (for
the sake of simplicity, it is assumed that initially there are no net capital
inflows). The volume of savings and investment (Vc ) is less than that in a
closed market situation with a free interest rate (Ve ). At the same time, some
unprofitable investments are made, since rationing means that not all of
Interest rate
r
D O
O1
re
rf Of
rc
Volume
Vc Ve V
Vf
Figure III.1 Simple conventional framework of capital markets liberalization.

the most efficient investments take place. Thus, some investments would
have returns equal to rc, whereas others, with a higher yield, would remain
without financing (in some of them, the return would exceed re, since total
investment would be Vc ).
Domestic financial liberalization would allow a rise of r to its local equi-
librium, and savers would face more options for investing their funds.
Given the broader variety of options, there would be a larger supply of
national savings (O1, to the right of O) replacing the investment in non-
productive assets. On the other hand, the financial opening would allow
capital inflows, which would complement the local funding for investment.3
A small economy, like Chile, would face a horizontal supply of external
funding (Of).4 With free capital inflows, an investment volume of Vf > Ve and
an equalization of the interest rate at rf would be reached.
The national savings rate would rise, despite the lower interest rate, if
the shift of supply were sufficiently strong, as in Figure III.1. This is a com-
mon assumption, explicit or implicit, in the neoliberal reforms of financial
markets.
The exchange rate policy is a determining factor in the final outcome.
To reach an equalization of interest rates, there must be no expectations of
variation in the real exchange rate (deflated by net inflation). Initially, this was
sought through a crawling-peg policy of mini-devaluations. Subsequently, a
fixed nominal rate was adopted in 1979. Official policy assumed that –
in a free trade regime, such as was already at work in 1979, with a fiscal
surplus and, supposedly, an exchange rate close to the “equilibrium level” –
freezing the nominal exchange rate would rapidly prevent domestic infla-
tion from exceeding external inflation. Consequently, the exchange rate
would become the anchor for stabilizing the level of domestic prices. The
authorities thus formally adopted the “monetary approach to the balance of
payments,” with its neutral monetary policy.
The official approach assumed that financial liberalization would work
efficiently since capital flows were managed by private agents and mostly
without official guarantees. Thus, capital inflows would take place only if
the borrower expected a net return for its use that would exceed interest
payments. It was said that the fast growth in the external debt that took
place reflected a healthy economy (De la Cuadra, 1981, p. 1025) and private
debt would not present any threat of insolvency. This problem would only
arise in the case of public debt (Robichek, 1981, p. 172).
3
The relevant figure for the contribution of foreign capital to domestic capital forma-
tion refers to the net transfer of funds (NTF) – that is, the net capital inflow minus the
rent of foreign capital. Domestic savings plus NTF national savings plus net capital
inflows gross domestic investment.
4
See Harberger (1985) for the case of an upward sloping supply, associated with the
“negative externality” resulting from rising liabilities of a small economy.
A naive vision of the working of private institutions prevailed. Statements

by the authorities that users knew how to distinguish between good and
bad financial institutions were frequent. The evident fact is that prudential
supervision was extremely lax in Chile.5
In all, the discrepancies between the official approach and the real world
were very marked. The economic team expected that liberalization of the
domestic financial market, in association with the gradual opening of the
capital account, would lead to an increase in national savings. In parallel,
the quality of investment would improve in response to the suppression
of former subsidies, free interest rates, and the removal of discrimination
among different borrowers.
The volume of national savings and capital formation was less than Vc ;
that is, instead of growing, ratios decreased sharply. As documented below,
the domestic interest rate stood at unexpectedly high real levels, was quite
unstable, and its average was spectacularly higher than international rates;
terms were extremely short, mostly of thirty-days loans, with a negligible
share of mid-term lending. Fixing the nominal exchange rate in 1979 led
to a significant loss of purchasing power of the dollar price and to a grow-
ing external deficit. Private external indebtedness grew notably, encouraged
by the risky assumption that its real cost would remain low and that there
would continue to be easy access to loans in the future. Finally, the exces-
sive indebtedness, with the correspondingly rising service, by financing
competitive imports, weakened the productive system and the payments
capacity instead of strengthening them.
(b) The new institutional framework

In this period, three subperiods may be distinguished with regard to the
implementation of financial policies. The first ran from 1973 to 1975, when
there were no substantial changes in the domestic financial system; banks
nationalized under the previous regime remained under state control. On
the other hand, varied measures undertaken in order to encourage foreign
investment and foreign loans carried negligible net inflows.
The second period began in 1975 with the drastic reform of the domestic
financial system and the privatization of almost all commercial banks.6 In
addition to freeing the interest rate, in 1975 the government eliminated
the norms relating to quantitative controls of credit in domestic currency
and the selectivity of bank reserve regulations, which were largely intended
to channel funds into production rather than consumption. Moreover,
5
Interestingly, something similar occurred in industrial countries with a deregulation
of loans to developing countries. See Valdés-Prieto (1989) on the relaxation of norms
to facilitate the recycling of oil surpluses.
6
The largest commercial bank, the Banco del Estado, founded in 1953, remained
public, but its market share fell from about half of the system by the early 1970s to
14% of outstanding loans in 1981.
restrictions relating to maturity terms were removed (except for a gen-

eral restriction of thirty days minimum). Next, a gradual uniformity was
imposed on the different financing institutions for both the operations
permitted and their relaxed regulations.
The tendency towards uniform treatment also included foreign banks.
Their activities had been restricted during the governments of presidents
Frei and Allende. In December 1974, restrictions on their operation in Chile
were lifted; in the same year, a permissive statute for foreign investment
(Decree 600) was enacted. By 1982, there were nineteen foreign banks,
though only covering 10% of outstanding loans. In respect of the capital
account, liberalization was slower. This gradualism contrasted with the
rather abrupt trade and domestic financial reforms. The gradual nature
of the liberalization of controls on the size and terms of capital flows was
meant to ensure control over money supply, which for several years played
the principal role in anti-inflation policy. It was considered that, in view of
the sizable difference between domestic and foreign interest rates, abrupt
liberalization of the capital account would attract inflows in such a volume
that it would endanger price stabilization.
To control the size of inflows, several tools were applied: quotas on borrow-
ing and guarantees that banks could grant were established, as well as quotas
on the volume of foreign loans that could be sold monthly in the domestic
foreign exchange market, and a reserve requirement on financial inflows (to
inflows via article 14). Indeed, with the passage of time, the capital account
was liberalized through releasing quantitative restrictions or replacing them
with more flexible controls. In 1981, when the net use of foreign savings
jumped to 21% of GDP,7 financial inflows (under article 14 of the law on
foreign exchange operations) were subject to a minimum stay in the country
of twenty-four months and the compulsory deposit of a percentage of the
credit (10 or 15%, depending on the term) in the case of financial opera-
tions for less than sixty-six months. There were no special restrictions on
the volume of credit that the banks could borrow abroad and lend in foreign
currency in Chile, but there was still a limit on the guarantees they could
grant.
In all, there were heavy capital inflows, which grew rapidly between 1977
and 1981. The determining factors were the abundant international supply
of funds, the low initial bank debt of Chile, the image of creditworthiness
that Chile had been achieving, and the weakness of prudential regulations.
The poor effectiveness of capital controls was due to numerous leakages or
insufficiencies, including: (i) a climbing share of financial flows was dealt
7
Current account deficit in 1977 constant prices, with GDP converted to US dollars
with the real RER of 1976–8. In current US dollars, the net use of foreign capital was
14.5% of GHDP, due to the abnormally increased value of GDP expressed in US dollars
as a result of an extremely appreciated (outlier) exchange rate in 1981.
away from banks (37% in 1976–81); (ii) only loans with terms under sixty-
six months were subject to the reserve requirement, in an international
environment in which the standard international financial loan exceeded
that term; (iii) in loans under sixty-six months, the cost of the reserve
requirement represented a negligible fraction of the huge spread between
the domestic interest rate and the cost of external loans (see Table III.5);
(iv) short-term trade credit was free to expand vigorously, duplicating their
participation in a total debt that grew rapidly towards 1981.
Finally, from late 1981 onward, despite the further liberalization of
capital inflows, the supply shrank drastically, as a result of both the emer-
gence of the international financial crisis and the late recognition by bank
creditors of the excessive indebtedness of Chile. When the Mexican debt
crisis officially exploded in August 1982, Chile already found itself amid
a deep crisis. In fact, from the second half of 1981 on, GDP had been
decreasing persistently (see Marcel and Meller, 1983). During 1982, the
minimum term of twenty-four months was eliminated, the compulsory
deposit was set at 5% of the foreign loan, which was a rate similar to the
reserve requirement for domestic bank deposits, and capital outflows were
liberalized.
2 Financial capital flows and indebtedness:

volume, sources, and uses
Capital flows grew rapidly from 1977 onward. Even though Chile recorded a
large and growing current account deficit in the following years, net capital
inflows were sufficient to allow a significant accumulation of international
reserves until during 1981. This capital surge took place in a context of
dynamic expansion of the external sector, especially of imports of consumer
goods. Inflows were overwhelmingly concentrated in lending to the private
sector with no state guarantee. FDI and loans to the public sector accounted
for less than one-fifth of inflows.
The government expected a vigorous inflow of FDI in response to the
“economic and political system” that the dictatorship offered and to the
new statute for foreign investment (Decree 600, of 1974). It was hoped
that FDI would arrive to exploit the “comparative advantages” previously
repressed, which the model was liberating. However, the response of FDI was
disappointing for the expectations of the economic team. Actual inflows
were rather low and concentrated in the last part of the period (see Lahera,
1981). On the other hand, a significant share of inflows corresponded to two
components not involving direct creation of productive capacity. One cor-
responds to risk capital contributed by transnational banking branches and
the other to the acquisition of firms or of packages of equity shares.
Table III.1 shows several indicators of annual capital flows, expressed in
constant prices. They increased significantly in the second half of the 1970s
(col. 2). Columns 3 and 4 show that capital flows rose sharply as a share
of gross domestic investment and GDP. This was partly a result of the rela-
tive stagnation of the latter two variables during the 1970s. Foreign savings
(deficit on the current account) and the capital service also grew, with some
ups and downs, in relation to exports (cols 5 and 6), in spite of the dynamic
growth of the latter in the early years of the neoliberal experiment. The total
debt service in 1982 amounted to 88% of exports of goods and services, that
is, three times the average coefficient recorded in 1970–4.
In short, the data show that from 1977 capital inflows captured a grow-
ing relative weight in the Chilean economy. The coefficients reflecting their
incidence exhibit a debt-servicing burden substantially greater than for
Latin America as a whole in the 1970s and early 1980s (Bacha and Díaz-
Alejandro, 1983; Ffrench-Davis, 1983a).
Table III.1 Deficit on the current account and capital flows, 1970–82
Year Balance Gross External External Export Debt-service
on current inflow of financing financing deficit (%) coefficient
account loans of invest- to GDP (%) (%)
(1977 US$ (1977 US$ ment (%)
millions) millions)
(1) (2) (3) (4) (5) (6)
1970 –166 941 6.8 1.4 6.5 27.0

1971 –367 772 15.5 2.8 16.9 38.4
1972 –690 1,302 36.4 5.4 40.0 27.9
1973 –441 1,106 24.7 3.6 20.7 25.4
1974 –256 1,064 12.0 2.1 8.9 35.1
1975 –536 1,109 32.5 5.0 29.3 55.6
1976 159 1,086 –11.4 –1.4 –6.5 52.7
1977 –551 1,390 34.1 4.6 22.5 52.8
1978 –960 2,559 50.9 7.4 38.5 58.7
1979 –925 2,691 42.2 6.6 27.3 50.8
1980 –1,361 3,270 51.2 9.0 36.8 47.7
1981 –3,214 4,640 108.1 20.7 103.2 70.8
1982 –1,600 2,238 87.2 12.2 53.4 88.5
Source: Calculations based on Central Bank of Chile, Balanza de Pagos, Deuda Externa de Chile, and
Cuentas Nacionales in 1977 pesos.
Note: All nominal figures were deflated by the Index of External Prices (IEP; see Ffrench-Davis,
1984), in order to convert them into figures at 1977 prices. Column (3) measures the ratio
between the deficit on the current account and gross fixed capital formation. Column (4) is the
deficit on the current account as a share of GDP. Column (5) is the ratio between the deficit on
current account and exports of goods and services. Column (6) entries are amortizations plus net
interest payments as a share of exports of goods and services. For the conversion into US dollars
of the figures for GDP and investment, which were originally in 1977 pesos, the average real
exchange rate for the three-year period 1976–8 was used, expressed in 1977 pesos per US dollar
of that year.
Debt activity (as reflected by the volume of gross inflows and amortization
payments) increased faster than net capital flows since the terms of the loans
became shorter. This was a direct consequence of the increased share of pri-
vate creditors in total debt and their shorter maturity terms. The magnitude
reached by capital movements is reflected by the fact that, in the two-year
period 1980–1, average gross inflows were equivalent to 24% of GDP.
During the period under analysis, significant changes took place with
regard to the agents (creditors and debtors) participating in capital flows.
Among creditors, 84% of the external debt in 1981 was with banks and
financial institutions, which had accounted for only 19% by 1974 (see
Table III.2); in parallel, the nominal amount of debt with official institutions
declined. This “privatization of creditors” was partly the result of greater use
of the supply of foreign banks, which before 1977 had been scarcely used by
Chile compared to other emerging economies. From then on, however, it
rapidly grew-up. By 1982, Chile’s per capita bank debt exceeded US$1,000,
compared to a regional average of US$600 and only about US$500 in the
case of Brazil. On the other hand, Chile’s bank debt increased by 57% per
year between 1977 and 1981, compared to an average of 28% for develop-
ing countries.
With regard to debtors, after 1975 the growing net inflows were received
mostly by the private sector, while the government moved towards a budget
surplus and amortized its external debt. This situation was in line with a
deliberate policy of reducing state participation. This was facilitated by the
Table III.2 Total foreign debt and private financial creditors, 1974–82
Financial institutions
Year Total debt US$ millions Share in

(US$ millions) total (%)
(1) (2) (3)
1974 4,776 923 19.3

1975 5,453 1,352 24.8
1976 5,392 1,506 27.9
1977 5,763 2,144 37.2
1978 7,153 3,723 52.0
1979 8,790 5,885 67.0
1980 11,325 8,579 75.8
1981 15,700 13,169 83.8
1982 17,263 14,986 86.8
Source: Central Bank of Chile, Deuda Externa de Chile, 1982, August 1983,
Tables 1, 3, and 11.
Note: Total debt refers to the disbursed outstanding stock at year-end. In addi-
tion to the traditional foreign debt, it includes liabilities in national currency,
liabilities with the IMF, and short-term debt contracted by sectors other than
the monetary system, with the exception of direct foreign-trade operations.
change that took place in international markets: the loss of weight on the
part of official financial institutions, which operated mostly with govern-
ments; and the vigorous emergence of private international capital markets,
which offered access to both public and private debtors.
Table III.3 shows the composition of total outstanding foreign debt of
public and private borrowers. Since international reserves grew steadily
until 1981, part of the debt was not absorbed in the economy. I distinguish
between total “gross debt” and “net debt” (the share used to finance the
deficit on the current account). This has implications for determining the
sectoral origin of the macroeconomic disequilibrium that Chile experienced
and the effects of external debt on domestic purchasing power.
An estimate of the net impact of capital flows is shown in columns 3 and 4
of Table III.3, which present the total debt minus the international reserves
of the respective sectors. As the accumulation of assets was concentrated in
the public sector (central bank), from 1975 up to 1981 this was reflected in
a substantial reduction in its net liabilities. In the case of the private sector,
in contrast, net indebtedness grew very rapidly, multiplying by thirteen
between 1974 and 1981. It should be noted that most of the private debt
was contracted without state guarantee. Thus, in 1981 almost two-thirds of
Chile’s total debt lacked an official guarantee (see Chapter IV, Table IV.1).
That high share could have constituted a decisive bargaining factor in the
renegotiations of the external debt.
Table III.3 Public and private net foreign debt, 1973–82 (US$ millions)
Gross debt Net debt Private sector
share in net debt
Year Public Private Public Private (%)
sector sector sector sector
(1) (2) (3) (4) (5)
1973 3,276 786 3,063 716 18.9

1974 3,896 879 3,709 773 7.2
1975 4,426 1,027 4,252 931 18.0
1976 4,252 1,140 3,718 1,016 21.5
1977 4,319 1,444 3,763 1,339 26.2
1978 4,858 2,295 3,648 2,147 37.0
1979 5,018 3,772 2,882 3,053 54.9
1980 4,905 6,426 1,569 5,986 79.2
1981 5,145 10,561 1,878 9,761 83.9
1982 5,892 11,371 3,866 10,586 73.2
Sources: Calculations based on IMF, International Financial Statistics Yearbook, 1982; Central Bank of
Chile, Boletín Mensual, and Deuda Externa de Chile.
Note: Column (1) excludes state-guaranteed private debt and debt contracted by the Banco del
Estado. Column (2) includes private debt, state-guaranteed debt, and debt contracted by the Banco
del Estado. Columns (3) and (4) represent the gross debt minus the international reserves of the
central bank and the financial system, respectively. For measuring reserves, holdings of gold were
valued at a constant real price of US$42,222 per ounce of fine gold, base 1977.
With regard to the intermediaries, up to 1977 a significant share of the

private sector debt had been borrowed directly abroad, because of quanti-
tative restrictions faced by domestic banks. Two qualifications should be
posed, however. On the one hand, the segment of the non-financial private
sector with most access to inflows was that with the closest connections to
national and foreign banking institutions. On the other hand, a substantial
share of these loans had bank guarantees. From 1978 onward, the domestic
financial sector gained greater importance in the direct intermediation of
external private financing. Table III.4 provides details on liabilities under the
main channel of financial inflows (the so-called article 14). Private borrow-
ers captured 97% of gross inflows in 1976–81.
The official figures for the real net outstanding debt showed a reduction
over the five-year period 1976–80: it decreased from US$5.3 billion in 1975
to US$4.3 billion in 1980 (in 1977 dollars). Two external factors explained
this decline in contrast with high annual inflows in current dollars. The
main factor was the rate of international inflation, which eroded the real
value of the debt stock; second, net debt also decreased in response to the
rise in the value of international reserves maintained in gold. These two fac-
tors explain a drop of US$2.7 billion in real net debt during the five years
(see Ffrench-Davis and Arellano, 1981). Thus, instead of having decreased by
18% in real terms, the debt would have grown by one-third and would have
shown a sharp acceleration toward the end of that period in the absence of
these two factors. Actually, in 1981 there was an additional increase in debt
that was spectacular in both nominal and real terms. It was therefore clear
Table III.4 Gross annual flows of loans, by debtors, 1976–82

Breakdown (%)
Private non-financial sector

Domestic
Year Total flows Public Non- financial
(US$ millions) sector guaranteed Guaranteed institutions
(1) (2) (3) (4) (5)
1976 263 13.3 86.3 0.4

1977 336 13.2 80.1 6.7
1978 780 4.2 31.0 57.0 26.0 38.8
1979 1,245 1.8 34.7 56.3 21.6 41.9
1980 2,504 3.1 14.5 32.1 17.6 64.8
1981 4,517 1.9 20.6 25.2 4.6 72.9
1982 1,771 24.4 24.7 30.8 6.1 44.8
1976–81 9,645 3.1 37.2 59.8
Source: Based on data from Central Bank of Chile, “Créditos Liquidados Artículo 14,” December
1980; and Boletín Mensual, No. 662, April 1983.
Note: Column (1) shows the gross annual flow of disbursed loans minus compulsory deposits. The
breakdown over cols (2) through (5) was estimated on the basis of that for Santiago.
that the rapid growth of external indebtedness was generating macroeco-

nomic vulnerability as well as was being harmful to national development,
as is shown later (see also Chapter II). Nevertheless, the government insisted
right up to the end that “borrowing abroad was good business” because the
real interest rate was very low or negative; furthermore, debtors were in the
private sector, which was subject to “free market laws” so that, in the official
view, there could be no doubt that borrowing was efficient.
A major share of foreign loans was used to finance consumption by the
private sector. In fact, about three-fourths of the net debt rise in 1977–82
was used to increase the import/GDP coefficient of the Chilean economy
(see Ffrench-Davis and De Gregorio, 1987). The massive net inflows, also,
encouraged the excessive exchange rate appreciation and contributed to
sustain it for several years. In fact, if the supply of inflows had been weaker,
the government would have been forced to moderate the tariff liberalization
and/or the exchange rate appreciation. Indeed, the loans available to Chile
were larger than could be absorbed productively. In contrast to other coun-
tries, which channeled external financing into investment, Chile, instead
of having a “debt-led economic growth” (as in Brazil), incurred a “debt-led
deficit on current account,” with a crowding-out of savings and domestic
output (particularly manufactures and investment in tradables) by climbing
imports of consumer goods. The growing and unsustainable external gap
made a future downward adjustment unavoidable.
It is interesting to highlight two contrasts, one positive and one nega-
tive, between Chile and other LACs. On the one hand, capital flight was
rather minor in Chile (Arellano and Ramos, 1987), while in Argentina and
Venezuela it was notably high (Frenkel, 1983); in fact, in these two countries,
in parallel with inflows, substantial outflows were recorded. On the other
hand, the investment ratio was low in Chile, while Brazil and Colombia
exhibited significant productive capital formation as a consequence of more
efficient absorption of foreign resources (see Bacha and Díaz-Alejandro,
1983; Perry and Junguito, 1983).
3 Indebtedness and macroeconomic adjustment
The massive process of external indebtedness between 1977 and 1981 had
significant effects in many areas of the national economy. The process pro-
foundly affected aggregate demand and its composition, contributed to the
spectacular concentration of wealth (see Chapter VII), considerably altered
the functioning of the savings/investment process, and conditioned to a
crucial extent the management of monetary and exchange rate policies.
The initial impact of external indebtedness involved an increase in the
availability of foreign exchange. This gave rise to two possibilities. One
was an increase in international reserves, which is usually accompanied
by a rise in the money supply; the other consisted of an expansion of the
current account deficit. In practice, up to 1981, the growing indebtedness

manifested itself in both ways, since net capital inflows were larger than the
capacity to absorb them, notwithstanding that the current account deficit
steadily increased by considerable amounts. In 1980, net use of foreign
capital was the equivalent of 9% of GDP, in contrast with an average of 5%
for Latin America. The difference between inflows received and those used
gave rise to the increase in international reserves registered, which in 1980
represented 68% of annual imports of goods. Additionally, in 1981, both the
debt stock and the deficit on current account experienced sharp jumps, with
a drop in the reserves/imports ratio.
The rapid building-up of reserves until 1980 had substantial effects on
monetary and foreign exchange policies. They contributed, evidently, to
the recovery of economic activity, thus reducing the large recessive output
gap that emerged in the crisis of 1975; however, in parallel, other macr-
oeconomic disequilibria were being built. Furthermore, the huge capital
inflows meant that a large share of the total credit available in the domestic
economy had originated from foreign sources.
(a) Monetary policy, inflows, and the crowding-out of

domestic credit
From 1975 onward, net purchases of foreign exchange by the central bank
represented the main source of expansion of the money supply: in the three-
year period 1978–80, these operations represented more than 100% of the
total high-powered money issued (Ffrench-Davis and Arellano, 1981, Table
13). As noted, the overwhelming share of net purchases of foreign exchange
by the central bank came from the private sector. Indeed, in some years
foreign exchange operations with the public sector even had a contractive
effect. This situation continued until 1981, when the severe macroeconomic
imbalances that had been building in the Chilean economy started to emerge
openly and the loss of international reserves began. Correspondingly, the
monetary effect of foreign exchange operations became markedly restrictive
in 1981.
As indicated in section 1, during certain periods direct restrictions on
capital inflows were imposed as an instrument of monetary control. This
limitation was enforced – with successive modifications in the maximum
amounts of foreign exchange operations authorized – in September 1977
and was held up to April 1980. These restrictions were not sufficient to keep
inflows to the private sector down to a volume consistent with the monetary
expansion desired by the economic authorities. Consequently, the latter took
action on the other sources of money issue and on the exchange rate. In
both cases, there was a crowding-out of the domestic economy by the sec-
tors associated with foreign financing. In fact, domestic credit to banks was
restricted in the face of the increase in high-powered money. The exchange
rate, for its part, was revalued (see point (b) of this section) in response to
the accumulation of reserves, and concerned with the stubbornness of infla-

tionary pressures. As a consequence, there was a significant crowding-out of
domestic producers of tradables.
Quotas in the expansion of liquidity remained in force as long as the closed
economy monetary approach predominated. Subsequently, with the adop-
tion of the open economy monetary approach, the nominal exchange rate
was frozen, and a “neutral” monetary policy was explicitly adopted. Thus,
changes in the international reserves were to be the determining factor of
liquidity of the domestic economy against the background of a balanced fiscal
budget (actually, a surplus) and stable low bank reserve requirements. Interest
rates fluctuated freely according to financial market forces.
The monetary approach to the balance of payments survived until mid-
1982. During the last year in which it was in force, a contractive “automatic
adjustment” did operate, with disastrous effects on employment and output
(see Chapter IV).
(b) Foreign exchange policy: instability and appreciation8

In October 1973, a second experience with a crawling-peg regime was
adopted (the first experience in 1965–70 is briefly described in Box III.1).
Both experiences involved frequent small adjustments of the nominal
price of the dollar, lasted more than five years, and were established during
periods of acute balance of payments difficulties and sizable inflation rates.
However, they differed with respect to the political environment in which
they were implemented and the role assigned to the exchange rate, the sta-
bility of the policy, and the criteria used to determine modifications of the
real exchange rate (RER).
In this second experience, initially, the authorities sought to return the ER
to the purchasing power parity of 1970, and to compensate for significant
import liberalization. But subsequently they were strongly influenced by
changes in gross reserves, and the ER became more a monetary policy tool
than a resource allocator. Finally, the ER was used as the main instrument
regulating inflationary expectations and as anchor for tradable prices, rather
than as a signal for resource allocation and determinant of sustainable cur-
rent account balances. The changing role of the ER – the changes in policy
priorities and in the analytical approach – meant large fluctuations in the
RER.9
8
A detailed analysis of foreign exchange policies between 1965 and 1982 can be
found in Ffrench-Davis (1981).
9
By RER, I mean the nominal exchange rate divided by an index of domestic
prices and multiplied by one of external inflation. Since Chilean trade was diver-
sifying and price relations and exchange rates among the country’s main trading
partners had changed notoriously, we built a weighted index of external inflation
faced by Chile that reflects these phenomena (see Ffrench-Davis, 1973, Appendix I;
Ffrench-Davis, 1984).
In a first stage, between October 1973 and 1976, the main rate, of the
then multiple rates, experienced a 38% real depreciation between the aver-
age of 1974 and January 1976. The devaluation was directly associated with
a depressed copper price and a worsened current account. Real devaluation
compensated approximately for the average fall in nominal restrictions on
imports during those years, and both factors combined to enhance the com-
petitiveness of exports.
By early 1976, a growing balance of payments surplus (owing to the exter-
nal credit boom experienced by Chile since 1976) and a rising monthly rate
of inflation (to levels between 11 and 15% per month) invited a change
in the trend of the RER. Thus, it began a second stage, with a substantial
modification in the role of the exchange rate, associated with its impact over
inflation. The main features of this second phase, which covered the period
up to February 1978, were the daily frequency of mini-devaluations, accord-
ing to schedules (tablas) announced monthly, and a series of abrupt exchange
rate revaluations. Immediately, the real rate started to appreciate rapidly, los-
ing 14% in five months up to June. In this same period, import liberalization
became more effective because tariff reductions already faced a clearly fading
redundancy. Now liberalization came together with RER appreciation, rather
than the reverse as the authorities had repeatedly announced.
Then an abrupt revaluation of 10% took place in June 1976, when the
policy change was made official. This was a notorious change in the policy,
and caught public attention not so much because of its intrinsic economic
effects but because of the conditions and arguments that surrounded it. The
minister argued that: (i) in 1975 the ER had increased faster than domestic
prices, producing an “unrealistic” ER that was not sustainable in the long
run; (ii) the new parity would be adjusted systematically because its new
level was realistic; (iii) the new system would allow the exporter to work
with certainty; (iv) the revaluation would reduce industrial costs and infla-
tion from July onward, thus, he asserted, increasing real wages; (v) imports
and economic activity would be fostered by revaluation; and (vi) money
expansion originating in net reserve purchases by the central bank would
be reduced.10
The decision to revalue can be traced to a combination of factors. First, the
monthly rate of inflation had been rising, complicated by a slow recovery
from the downswing in economic activity in 1975, and open unemployment
was at a very high level. It appeared that a spectacular event was needed to
ensure the survival of the drastic orthodox policies being enforced, since eco-
nomic policy was under heavy attack from government supporters; criticism
from the opposition was not allowed by the dictatorship. Second, the purely
10
Speech of the Minister of Finance, in DIPRES (1978, pp. 261–2). About 80% of
the changes in high-powered money had been originating in the net purchase of
reserves.
Box III.1 The “crawling” exchange rate: a pioneer

reform in 1965–70*
Chile was a pioneer in implementing an exchange rate (ER) policy

belonging to the family of crawling peg approaches. It happened
between April 1965 and July 1970, making Chile the first country to
implement a policy of this nature systematically. After Chile, Colombia
and Brazil implemented similar policies in 1967 and 1968, respectively
(see Williamson, 1981). The policy was conducted by the central bank in
coordination with the Treasury.
The policy selection was influenced by several factors. First, it was
believed that both the nominal peg and totally free rates – the two corner
solutions that were then in fashion in the literature – had a negative
impact on macroeconomic stability and on resource allocation. The pur-
pose was to have a reasonably stable real exchange rate (RER), without
large jumps, in order to provide better allocative signals and avoid specu-
lative capital gains and sharp shocks to the domestic economy. Second,
the authority intended to limit the transmission of copper price insta-
bility to the domestic economy. It was necessary to ensure that govern-
ment expenditure was guided by a “normal” copper price. This was done
through annual agreements between the central bank and the Treasury
on the use of the tax revenue from large copper producers; by using an
account in dollars in the Central Bank where the “excess” revenues were
deposited and sterilized abroad. This was the “informal” birth of the
Copper Buffer Fund, created two decades later. In spite of facing intense
appreciation pressures, the economic authority resisted them success-
fully. In fact, in the period 1965–70, a persistent stability of the RER was
achieved, with some real devaluation associated with a gradual program
of rationalization in import restrictions. The Frei Montalva government
wanted to pursue in a more efficient fashion the substitution of imports
and to expand and diversify exports, with stable incentives.
The main variable guiding the nominal ER adjustments was the
expected rate of domestic inflation. It was the purpose of the authorities
to achieve a real devaluation: first, because there was a desire to reduce
the upward trend of the previous decade in foreign indebtedness; and,
second, because the average level of protection for import substitutes was
to be reduced gradually, thus requiring a compensatory adjustment in
the ER. The real rate was devalued by about 25% during the more than
five years that the policy lasted.
Two episodes illustrate the quality of the exchange rate reform. In
1967, the authorities foresaw early a balance of payments imbalance, to
which they responded with a reduction in fiscal expenditure and a faster

devaluation (with more frequent adjustments). This conjuncture had
removed the ER from its former role as a main expectations destabilizer
during critical situations.
Second, because of a higher price for the main export of Chile (copper,
which represented about two-thirds of total exports) and an increase in
the share of the economic rent accruing to the government, the bal-
ance of payments showed sizable surpluses from 1968 onward. This also
induced an increase in short-term capital inflows into a “successful”
emerging economy. The inflows were fostered by the larger reserves, and
relatively tight domestic credit. Obviously, strong pressures were exerted
on the government in order to freeze the ER. But the economic authori-
ties managed to retain the real devaluation achieved in 1967–8 and to
enlarge it by continuing to crawl nearly pari passu with the rate of domes-
tic inflation; thus external inflation provided a positive differential.
In summary, given available policy tools, in 1965–70 public policy
searched for stability in the RER rather than in reserves, (i) because of its
direct allocative effect, particularly with regard to exports; (ii) because
of the fear of ratchet price effects if the ER started following transitory
changes in the terms of trade or capital flows; and (iii) because of the
belief that the actual price of copper was above its “normal” level and
would fall drastically at some time in the future, as it did by mid-1970.
*Based on R. Ffrench-Davis, “Exchange rate policies in Chile: the experi-
ence with the crawling peg”, in J. Williamson (ed.), The Crawling Peg: Past
Performance and Future Prospects, Macmillan, London, 1981.
monetarist approach to stabilization (à la Friedman) had been weakened by

the negative experience with the dynamics of inflation in previous months
(Foxley, 1983). Thus, the emphasis was placed on the anti-inflationary effect
that a widely announced revaluation could exert via expectations.11
The fact is that all the weight of the government-controlled mass media
was put into promoting the newly announced policy. The revaluation
11
The orthodox closed economy monetarist approach to inflation that had prevailed
until mid-1976 was questioned by the emergence of an open economy monetarist
approach, to which most official economists seemed to have been converted rapidly.
Nonetheless, the latter approach took the lead only in 1979 when the nominal ER was
pegged. It is curious that economists who in the previous period had stated that if the
money supply was held constant the ER would have no effect on the price level later
argued that the ER determines the price level. Trade liberalization did not justify such
an extreme change of approach. Thus, we witness, once again, the role of fashion in
economics at the expense of pragmatism.
actually did change expectations, reducing the monthly rate of inflation to

6% in August and September; but it stagnated at that level for two quarters.
The minister of finance announced in March 1977 a new revaluation of
10%, with the set of mini-devaluations fixed at 4 and 3% per month, for
March and April, respectively. This was to be followed, again, by adjust-
ments according to the rate of inflation in the preceding month. Most of the
arguments provided by the government were the same as those given for the
first revaluation, though with more emphasis on the assertion that exports
(except for “marginal” ones) would hardly be affected because the ER “was
20% above October 1973.”12
The rate of inflation was indeed reduced to 3 or 4% a month, but it again
stagnated at the new level. The drop of inflation, led by revaluations, implied
a sharp real appreciation. Most of the real devaluation recorded in 1973–5
was reverted in this period, notwithstanding the deep tariff liberalization.
The phase ended in February 1978 when the monthly schedule was
replaced with a daily schedule for the rest of the year. The formula used
in the previous biennium, based on a nominal ER dependent on previous
inflation, was replaced with a schedule, determined by the annual inflation
target. The policy change considered decreasing devaluation adjustments,
starting with 2.5% (roughly the official rate of inflation in the previous three
months) and ending with 0.75% for December. The Minister declared that
“given the excellent balance of payments situation, the opening to foreign
trade and the announcement of the exchange rate for 1978, competitive
imports would rapidly enter if prices of domestic products were increased
excessively,” and that “the government is providing the foundations, so that
1979 will mark the start of a period of unprecedented price stability.”13 On
this occasion, the content of the official announcement, and of conferences
and interviews of several government representatives, gave the impres-
sion that, more than the regulation of expectations, the crucial factor was
the belief that the economy was already so open that the law of one price
would apply.14
Domestic inflation, measured by the official CPI, fell from 64% in 1977 to
30% in 1978 (from 84 to 37%, according to the corrected CPI). This apparent
success led to the adoption of the same approach for the following year. In
December 1978, a schedule of daily adjustments was announced for 1979.
12
Speech of the Minister of Finance, March 1977, in DIPRES (1978, pp. 308–9). The
official figure clearly overestimated the RER. See Table II.2. Recall that the official CPI
systematically underestimated actual inflation.
13
Speech of the Minister of Finance, February 1978, in DIPRES (1978, pp. 369–71).
14
This law implies that domestic and external prices of tradables, corrected by tariffs
and ER, are equal. As a result, it is argued that in a small country domestic prices cannot
change except through tariffs or ER changes. According to the extreme view, prices of
non-tradables similarly have no autonomy because they are determined by their inter-
relation with tradables through goods and factors markets.
The compound annual rate of devaluation to be implemented was 14.7%,

similar to the rate of inflation that had been announced informally as the
rate expected for 1979. No account was taken of either external inflation
(which was over 14% per annum in 1978–9) or the declining gap between
effective and potential GDP. If it was to be true that the law of one price
applied, exchange rate policy would clearly lead to inflation above 15%.
It is possible that the economic authorities were still in transition from an
expectations theory to a law of one price approach. The fact is that during
the first half of 1979, instead of decreasing, inflation began to rise. In the
twelve-month period ending in June, inflation reached 35%.
This last phase ended abruptly in June 1979 with the freezing of the ER
at a level 5.7% above the price prevailing then; the new level anticipated
the schedule for the end of the year. By mid-1979, after almost six years of
several policy variants and oscillations of the RER, the rate was at a level
close to that established at the outset. However, during the process import
restrictions had been drastically reduced: most non-tariff obstacles disap-
peared, and tariffs were reduced from an (unweighted) average of 94% to a
flat 10%. Additionally, the nominal ER was fixed in the context of a 35%
inflation trend.
The ER remained pegged at 39 pesos per US dollar up to mid-1982. Annual
domestic inflation decreased from 35% to close to zero, but during the tran-
sition between both dates the RER had appreciated by one-third. As a result,
given the pegged nominal rate, in order to return to the June 1979 level of
RER, apparently estimated as one of equilibrium by authorities (McKinnon,
1981, p. 34), a domestic inflation 30 points lower than the external one was
required. That was evidently out of touch, given the severe problems that a
large deflation poses for financial markets.15
An initiative to accelerate the adjustment process, through a legal reduc-
tion of nominal wages, could not be imposed by the economic team. In fact,
after some months of the so-called automatic adjustment, the outcome was
disastrous. Unemployment increased sharply and a rapid fall in manufactur-
ing output occurred.16 The end of the fixed exchange rate came when an
abrupt devaluation of 18% was applied in June 1982, followed by a series of
successive policy changes.
In short, since the mid-1970s, huge financial inflows were crucial to allow
diverse exchange rate approaches other than those of efficient resource
15
The wholesale price index fell by 8% between May 1981 and May 1982, while the
CPI declined by 1% between February and May 1982. On the other hand, interna-
tional markets exhibited negative inflation as well: the external price index fell by 4%
in the first half of 1982 as a result of a US dollar revaluation.
16
Open unemployment (excluding public emergency programs) grew from 11 to
23% of the labor force between September 1981 and June 1982. The value-added by
manufacturing fell by 21% in 1982.
allocation; in the process, rising inflows over-financed a climbing deficit

on current account. The use of the exchange rate to guide expectations (in
1976–9) and/or to anchor domestic prices (1979–82) did indeed result in
lower inflation. Nevertheless, inflation persistently decreased more slowly
than expected by the government. The exchange rate was used as a variable
to repress inflation, appreciating the peso during the process. This, together
with import liberalization and the recovery in economic activity recorded
between 1977 and 1981, led to an unsustainable current account deficit.17
At the same time, the gradual real exchange rate appreciation reduced the
cost of external indebtedness vis-à-vis domestic loans. Consequently, the
flows were encouraged by the evolution of the real exchange rate, enhanc-
ing the growing current account deficit and the surplus on the capital
account, in a pro-cyclical behavior.
(c) Worsening of portfolios and prudential supervision

External indebtedness contributed to a domestic credit boom. Total avail-
ability of funding was expanded by non-guaranteed (col. 3 in Table III.4,
above) and guaranteed (col. 4) direct foreign loans, and external loans inter-
mediated by local banks (col. 5). The rest of lending by domestic banks was
based on the liquidity generated by foreign exchange operations with the
corresponding accumulation of international reserves by the central bank.
The credit expansion took place in an environment of lax prudential regula-
tion and supervision.18 As said, loans to related parties grew at an accelerated
pace; most of them without effective guarantees. To avoid the prevailing regu-
lations, debtors made use of cross credits and turned to dummy firms as well
as operating from offshore institutions. In the face of high domestic interest
rates, banks renewed the principal and granted new loans with the objective
17
The increase in the deficit was also associated with the rise in interest rates and the
deterioration of the copper price. With regard to this latter item, the smaller fiscal
income from this source in 1981, compared to the average for 1960–70, was equiva-
lent to 0.7% of the 1981 GDP. The current figures on fiscal income contributed by
the large copper mining industry have been deflated by the external price index. The
deterioration in the copper price was partly compensated for by the improvement in
the price of molybdenum, a rise in copper output, and the capture of the economic
rent of the copper deposits for Chile due to the nationalization of these activities in
1966, 1969, and 1971 (see Ffrench-Davis and Tironi, 1974).
18
The permissiveness of banking regulation and supervision, and the absence of
public guarantees to deposits, rested on the assumption that self-regulation would
work together with the capacity of bank depositors to distinguish between the qual-
ity of different banks; irresponsible and risky banks would be rejected by the public.
Consistently with this view, the authorities repeated systematically that no private
loses would be financed by the state. However, after the intervention of a bankrupt
bank, in 1977, a guarantee was established on deposits in domestic currency up to the
equivalent of nearly US$3,000 (Held and Jimenez, 2001).
of being paid interest commitments (or the capitalization of interest); between

1976 and 1981, the outstanding stock of bank lending rose 38% annually in
real terms. Non-performing loans seemed to be low, and the banking system
exhibited high real profitability on capital and reserves, on the order of 17%
in 1978–80. Seemingly, for the government and diverse analysts, the market
was working fluently and without menacing risks.
Meanwhile, provisioning for risky portfolios, both total and individual,
was low. In 1979, total provisioning was reduced from 2% of loans to 0.75%,
and it was stated that priority would be given to individual provisions but
without implementing the latter. When the banking crisis exploded in
January 1983, it appeared that 19% of outstanding loans in 1982 had been
made to related parties and represented 249% of the capital and reserves of
domestic private banks. The deep weaknesses of the financial system as a
result of the neoliberal reforms became evident in the following year, with
enormous economic and fiscal costs mounting to nearly one-third of an
annual GDP (Sanhueza, 1999; Held and Jiménez, 2001).
4 Interest-rate differentials
Policymakers expected that the market, once freed from public interven-
tion, would achieve equalization of domestic and external interest rates,
an integrated financial market, and increased investment and its efficiency.
As shown in Table III.5, despite the huge volume of inflows, the outcome
was different: (i) there were persistent gaps between domestic and external
rates of up to over twenty percentage points annually, even in 1980–1 when
huge capital inflows were recorded; (ii) in the domestic market, the spread
between lending interest rates and passive rates (deposits) was around fif-
teen points; (iii) the nominal and real rates were extremely unstable, as were
the mentioned spreads; (iv) consumer loans expanded, predominantly for
imported consumer goods; and (v) the high cost of credit, its instability,
and the short maturities (mainly thirty days) discouraged productive invest-
ment. What non-speculative investment could pay real interest rates with
annual averages of 38%?
It may be noted that the ex-post gap between the domestic and external
rates for loans never dropped below eighteen percentage points per year.
In this respect, the traditional explanation that the differential was due to
expectations of a higher devaluation than the effective evolution of the
official exchange rate does not appear to have been valid. For example,
between 1977 and 1982 the parallel or black market rate was very similar
to the official rate. The easy access to the foreign exchange market that
existed at that time and the spot nature of the parallel rate do not make it
a precise indicator with regard to expectations of devaluation over twenty-
four months, which was the minimum term for the entry of capital under
article 14, but they do reflect the prevailing atmosphere of a relaxed foreign
Table III.5 Domestic and external real interest rates in

pesos, 1975–82 (annual percentage rates)
Year Domestic External Differential
(1) (2) (3)
1975a 121.0 – –
1976 51.2 –21.1 72.3
1977 39.4 0.2 39.2
1978 35.1 3.8 31.3
1979 16.9 –0.9 17.8
1980 12.2 –8.0 20.2
1981 38.8 12.4 26.4
1982 35.2 45.0 –9.8
Source: Calculations based on data from the Central Bank of Chile,

Instituto Nacional de Estadísticas, Cortázar and Marshall (1980),
and Ffrench-Davis and Arellano (1981).
Note: In 1982, the “preferential” exchange rate fixed by the govern-
ment for debtors was used in calculating the external interest rate.
This rate reflects the dominant segment of the market, of loans
between 30 and 89 days. The international rates paid correspond
to the interest rate for bank credits through Article 14 plus the
cost of compulsory deposits and the financial spread, all converted
into their peso equivalent. a Second semester, after the interest rate
liberalization.
exchange market in which there were continuing sales of foreign exchange

by the public over bank counters.
It is quite possible that the market was not aware, then, of the need
to make an adjustment for external inflation when measuring the real
exchange rate. It is therefore probable that the expected “foreign interest
rate,” comparable to the real domestic rate, would be closer to its nominal
level in dollars than to the ex-post rate given in column 2 of Table III.5. This
nominal rate fluctuated between 14 and 23% in 1976–82. Consequently,
even using this hypothesis, there would still be a substantial gap with the
domestic lending rate.
In addition to the domestic–external gap, there were substantial dif-
ferentials between the domestic rates for deposits and loans. There were
various reasons for these high spreads, and their significance was changing
over time. This issue has been dealt with elsewhere.19 Here I summarize the
aspects that have the greatest implications for the focus of this chapter.
Traditional explanations are as follows: (i) high bank reserve requirements
are a determining factor in the gap between domestic interest rates on
deposits and loans (the financial spread); (ii) the fiscal deficit and the inelastic
19
See Ffrench-Davis and Arellano (1981), Arellano (1983), Harberger (1985), and
Ramos (1986, Chapter 8).
demand for credit by public enterprises are responsible for the high interest
rates on loans; and (iii) the restrictions on capital flows are responsible for the
differential between rf and re (Figure III.1). None of these possible causes was of
significant importance during the whole period, however. The first one was of
some importance only in 1975–6 because of the high requirements for non-
interest-bearing bank reserves and over 300% inflation per year. Nevertheless,
very large financial spreads, net of the costs of reserves, persisted during most
of the period from 1975 to 1982. Second, the fiscal deficit fell rapidly (reach-
ing equilibrium in 1975) and turned into a solid surplus from 1976 onward
(Larraín, 1991, Table 4.4). Finally, in spite of the persistence of restrictions
on capital movements, such flows were huge, as was shown in section 2.
Consequently, orthodox analysis is not able to explain why, with net capital
inflows equivalent to an average of 8% of GDP in 1978–80, the gap between
domestic and external interest rates stood at an average of twenty-three per-
centage points per year (see Table III.5). In this period, as noted, it was evident
that there were still no expectations of a massive devaluation.
Therefore, there are other significant factors that explain the behavior of
domestic interest rates and the financial spreads.
First, diverse data suggest that the banking system was subject to rising
operational costs after the reform. One of the reasons for this, prior to 1977,
was the underutilization of installed capacity. Furthermore, the fact that the
system operated with such short terms for both deposits and loans tended
to increase costs. This contributed to explain why in 1978 the operating
cost of the system averaged 8% of total loans, an abnormal figure in com-
parison with international standards. Nevertheless, even after discounting
the operating costs, the spread still remained at abnormal levels. In all, as
documented, bank profitability was very high.
Second, the short maturities of financing facilitated the prevalence of high
interest rates.20 Those who had no access to external credit faced a severe
domestic recession simultaneously with interest rate liberalization. Against
the background of heavy propaganda to the effect that the recession would
be brief, many businesses and individuals resorted to expensive short-term
credit instead of closing down their operations, expecting a rapid reactiva-
tion of domestic demand and employment. Under these circumstances,
debtors did not view themselves as borrowing, for instance, at a real interest
rate of 40% per year but rather at 2.5 or 3% for thirty days, with the expecta-
tion of renewing the loan for a few months with a decreasing cost.
Effective demand, however, remained generally depressed until 1981 (and
lagged behind aggregate demand), and interest rates continued to be high
and unstable. Remember that only in 1981 did effective GDP reached a level
close to the productive frontier (see Chapter I). Given the continuing delay
20
On various occasions, the cause of the rise was the falling rate of inflation and the
lagging adjustment of the nominal interest rate.
of the expected reactivation, a devaluation for producers of exportables,

and a reversal of the trade reform hoped for by producers of importables,
entrepreneurs engaged in successive renewals of their bank loans, with the
increasing riskiness involved.
Third, some opportunities for highly profitable “investments” did arise.
Numerous public enterprises were sold at prices significantly below normal
market prices (see Foxley, 1982; Marcel, 1989). Similar opportunities were
offered by investment in real estate and financial assets, whose prices rose
notably in real terms: For instance, the stock exchange price index multiplied
by in constant prices between 1975 and 1979 and almost doubled in 1980.
The elimination of regulations on the use of credit made its rapid redistribu-
tion towards these uses possible, creating extremely profitable investments
from the private point of view but no new productive capacity.
Fourth, a spectacular increase in consumer credit, especially for imported
durables, was observed. Here, too, the suppression of previous restrictions
on bank lending for consumers facilitated the switch in the composition
of expenditure. Furthermore, import liberalization promoted an expansion
of the demand for credit, to be used for marketing imported consumer
goods. Thus, on its transit through the financial system, the savings of some
nationals leaked towards the consumption of imported goods. This is one
explanation for the drop observed in the rate of national savings in contrast
to the sharp increase in financial savings, which multiplied by six between
1976 and 1981.21 In the four-year period 1979–82, which was the period
of alleged great success of the economic model, the national savings ratio
(measured as gross capital formation minus the utilization of foreign capital,
as a share of GDP), reached barely 9% of GDP.
Fifth, the gradual recovery of domestic economic activity and wages, from
the very depressed levels of 1975, together with massive official publicity
within the prevailing authoritarian framework, helped to create an image of
a dynamic and rapidly growing real economy.22 And as economic recovery
on the basis of the utilization of installed capacity come to an end, aggre-
gate demand was fed with the very large inflows that allowed exchange
rate appreciation and sustained the consumption boom until far into 1981.
The atmosphere of success thus created, together with the biased expecta-
tions of a growing permanent income, induced consumers and firms to
continue increasing their indebtedness, while it prompted the banks to
21
The stock of loans of the banking system in domestic and foreign currencies rose
from 9% of GDP in 1976 to 55% in 1981. Figures that include assets of the central
bank are 33 and 62%, respectively. See Arellano (1983).
22
Schmidt-Hebbel (1988, p. 178) argues that “the explosive rise in consumption of
tradables between 1976 and 1981 … can be explained, in one half by diminished
restrictions on credit or consumer liquidity, and the other half by expectations of
higher permanent income and personal wealth, to a large degree stimulated by official
euphoria and propaganda.”
renew and rapidly expand their loans: it was an inter-temporal destabiliz-

ing adjustment generated by an over-optimistic euphoria. It was not an
enhanced appetite for risk, but an assuming away of risk. For instance, by
1981 many borrowers were moving from debt in pesos to debt in dollars,
because of the lower interest rate they faced. What an enormous capital loss
they suffered in 1982. Within this over-optimistic framework, the banking
institutions competed with each other, partly, by reducing the guarantees
requested from their borrowers. Pari passu, there was a huge crowding-out of
domestic savings by foreign savings; dropping domestic savings contributed
to increased real interest rates.
Sixth, banks allotted a significant share of funds to related companies and
individuals. Thus, for example, in 1982 the main private bank, which was
controlled by the largest economic group in Chile, had 42% of its portfolio
lent to firms that were directly related to the same group (Harberger, 1985;
Held and Jiménez, 2001). Consequently, financial transactions became mere
internal group operations, weakening appraisal criteria and procedures for
recovering loans.
Seventh, in a framework of frequent renewals of short-term lending and
borrowing to pay interest, the high cost of loans in the domestic capital
market implied an increase in the volume demanded rather than reducing
it: The latter effect predominated over the pure price effect. The magnitude
of the effect of financial costs is illustrated by the fact that between 1976
and 1982 an average debtor paid “excess” interest, over a real “normal” rate
of 8% per year, amounting to the equivalent of 300% of the initial loan.
In other words, a borrower who effectively disbursed the 8% each year,
renewed the principal, and capitalized the interest commitments in excess
of 8% would be liable by the end of 1982 for a debt four times the original
amount in constant purchasing power.23 It should be noted that, in con-
trast, a debtor in foreign currency, also disbursing the 8% annually, in 1982,
just before the devaluation, would be liable for a real debt 44% below that
contracted in 1976. Consequently, even after a massive real devaluation of
80%, the debtor would have ended up with a debt just equal to the original
amount (and equivalent to only a quarter of the liabilities of a person with
a debt denominated in pesos). This calculation is, of course, sensitive to the
rate assumed as “normal” and the period taken as a starting point. Thus,
for example, “late” debtors who borrowed in foreign currency only in 1981
suffered a huge loss, taking into account the real devaluations recorded in
1982. This is in sharp contrast to the case of “early” debtors.
23
These calculations were made using the corrected consumer price index. If the offi-
cial consumer price index had been used, the amount of the outstanding “real” debt
would have been 5.3 times.
Finally, capital inflows played a role different from the one tradition-
ally assumed. It is usually supposed that these funds enter into an inte-
grated market characterized by great substitutability between domestic and
external funding, so that the two types of interest rate would tend to equal-
ize. It is true that the external credit did relax liquidity constraints and,
evidently, there were some borrowers who had simultaneous access to
domestic and external sources of funds in diverse proportions.24
External financing became available on a very large scale, representing as
much as 40% of total loans available in the Chilean financial system (includ-
ing foreign loans) in 1981. The interest rate differentials, therefore, had
substantial effects for both resource allocation and income distribution (see
Zahler, 1980). Small and medium-sized producers were mostly relegated, at
best, to the segment where high interest rates prevailed. In contrast, borrowers
related to financial institutions and the main economic groups had easy access
to external credit, either directly or through the intermediation of domestic
banks. This noteworthy and persistent market segmentation helped to explain
the spectacular concentration of income and wealth in these years.
In all, the eight points summarized above, therefore, pushed up the lending
domestic interest rate. This was the result of persistent market segmentation.
In fact, only some debtors could borrow directly abroad or gain access to
credit through the intermediation of local banking. As a result, the difference
between the domestic and external interest rates reached notably high levels.
As already stated, the explanation for this does not lie in the expectations of
devaluation, since until far into 1981 these were absent, but is to be found
in the significant segmentation and incompleteness that prevailed in the
financial market.
5 External vulnerability and the dynamics of indebtedness
The growing indebtedness generated great vulnerability for the Chilean

external sector and for macroeconomic sustainability. In fact, the form
assumed by the transfer of foreign resources and the incentives provided
by the economic model led to a crowding-out of domestic savings and a
decline in productive investment (see Tables I.1 and III.1). Paradoxically, it
also crowded out the production of tradables, which experienced a drop in
their share in GDP (the production of exportables increased less than the
reduction in the production of importables).
The ease with which new loans could be obtained by “credible debtors”
and the low real international interest rates led many countries to take a
24
There was also significant heterogeneity within the domestic market itself. For
example, the average publicly offered lending interest rates exceeded the weighted
annual average rate calculated by the central bank by five and ten percentage points
in 1979 and 1980, respectively (Mizala, 1992).
complacent attitude during the 1970s.25 This was backed up by the belief, in
domestic official circles and international financial institutions (IFIs), that
since indebtedness was predominantly private its use would naturally be
efficient (see Robichek, 1981, pp. 171–2).
The growing indebtedness made gradual exchange rate appreciation fea-
sible. This, in turn, made it still more attractive to resort to external loans;
thanks to the appreciation, the real cost of foreign loans was negative during
most of the period. At the same time, domestic asset prices rose vigorously.
The process was thus self-encouraging, exacerbating capital inflows, which
increased domestic demand and allowed the continued exchange rate appre-
ciation. This led to the rising accommodation of the national economy to
massive financial inflows.
In the productive sector, however, the opposite of what was expected by
neoliberal reformers was taking place. The savings and investment ratios
were notably below the levels reached in the 1960s: the rate of gross fixed
capital formation barely reached an average of 15.5% of GDP in 1974–82,
and in its best year (1981) it did not exceed the average of 20.2% of the
1960s. An increasing share of inflows was directed towards the consump-
tion of imported goods, crowding out spending on national products
and domestic savings. The most obvious “comparative advantages” were
located in the purchase in the domestic market of underpriced equity from
deeply indebted firms. Except for some sectors making intensive use of
natural resources – such as fresh fruit, forestry, and fisheries (which indeed
expanded vigorously), and luxury construction – investors ran into the
difficulty of identifying comparative advantage: exchange rate appreciation,
high interest rates, the cutback of public support for productive develop-
ment, the reduction of public investment, the sharp trade liberalization, and
a significant recessive gap, all combined to provide a discouraging environ-
ment for productive investment (see several articles in CIEPLAN, 1983).
Consequently, in spite of the climate of euphoria and the close linkages
between the government and economic groups, productive investment
languished. The “financieristic” bias imposed on the domestic economy
prevailed; recall that the two “dynamic sectors” in the generation of GDP,
between 1974 and 1981, corresponded to value-added by the financial sec-
tor and by the marketing of imports (see Chapter I, Table I.2).
Apart from the poor performance of the productive system, it was obvi-
ous that the dynamics of the external sector could not be sustained for
long, even if there were no changes in the international environment.
Nevertheless, the official view was that the process would be self-regulated.
It was believed that since (i) there was no fiscal deficit, (ii) money supply
25
Interestingly, while Chilean liabilities and the external deficit were increasing
rapidly, the spreads it faced (which supposedly should reflect a growing risk) were
diminishing.
was less than the value of international reserves, and (iii) monetary policy
was “neutral,” a currency crisis could not arise. Several authorities even
asserted that there were economic arguments in favor of a revaluation
(De la Cuadra, 1981, p. 1024). It was thought that imports of consumer
goods would rapidly reach saturation and that the adjustment capacity of
the economy had been strengthened by the reforms imposed since 1973.
In contrast to these beliefs, when the international financial problems were
about to emerge, in 1981, the trade deficit amounted to 12% of GDP and
the current account deficit stood at 21%. For a long time, there had been
an evident need to reduce the external imbalance and to devalue the outlier
exchange rate.
The Chilean difficulties in capturing large external financing by late 1981
coincided with a situation in which there was a pressing need for fresh
foreign currency in order to cover increasing interest and amortization
payments and finance the huge trade deficit. Regarding the first two items,
the composition of external debt showed three strongly negative features.
First, the prevalence of flexible interest rates amid significant rate increases
in the international markets, together with a climbing outstanding debt
led, between 1978 and 1982, to interest payments being multiplied by four
(to 7% of GDP). Second, short-term indebtedness rose from a “normal,” sus-
tainable amount of trade credit to 13% of GDP. Third, amortization commit-
ments of private debt increased at an accelerated pace and were projected to
triple between 1981 and 1985.
Furthermore, the government had dismantled most economic regula-
tions, productive capacity had been weakened, and firms were heavily
indebted. Thus, the effects of external shocks were multiplied in the domes-
tic economy, with a decline in GDP of 14% in 1982, concentrated in the
manufacturing and building sectors.
In short, the external shock found Chile in a highly vulnerable posi-
tion, and this multiplied its negative effects on the domestic economy. As
documented in Chapter I, if the value added in financial activities and the
trading of imported goods (both based on shaky grounds) is deducted, the
per capita national product in 1981, before the effects of the shock, stood
only at a level similar to 1974 (see Chapter I, Table I.2). Despite emerging
modern sectors in the domestic economy, this was an unmistakable sign of
stagnation compared to the growth of the region in the 1970s.
6 Some lessons from this experience
I would like to underline three lessons from this financial opening process.
Financial liberalization is not neutral concerning the allocation of resources,
especially during the transition from a closed to an open economy, because
of the pressures generated on variables such as the stability of aggregate
demand, the composition of money supply, and the level and stability of
the exchange rate; they all have allocative affects and have an incidence
on the supply of factors of production. The key questions that emerge are
how to ensure that inflows are effectively channeled towards investment,
and whether access to and costs of the supply of external financing will be
stable in the future. Financial markets work with very short-term horizons
and naturally do not take into account the effects on productive activity.
Consequently, it is essential that the opening process be regulated in a man-
ner consistent with macroeconomic sustainability (see Ffrench-Davis, 2006,
Chapter VI).
First, access to external credit is not homogeneous for different eco-
nomic agents. In practice, this type of funding is usually available mainly
to large firms and economic groups. The cost differential observed among
agents during the expansive stage affected not only allocative efficiency but
income distribution as well. In particular, the opening of the capital account
in Chile provided substantial profits for those who were able to obtain for-
eign loans (see Zahler, 1980). Also, during the recessive stage, the outcome
is regressive since SMEs and informal segments tend to be affected more
intensively by drops in domestic demand and the emergence of liquidity
constraints.
A second lesson is connected with the destabilizing macroeconomic
effects of capital flows. In developing countries, where domestic markets
are not fluid and integrated, the instability in the supply of funding can be
quite disturbing for domestic economic activity. A feature of great practical
significance is that international financial markets experience sharp fluctua-
tions, which are promptly transmitted to domestic markets unless there is
some form of counter-cyclical regulation, principally during the expanding
part of the cycle (e.g. in 1977–81). Particularly, the supply of funds is subject
to abrupt changes in response to fluctuations in the lenders’ expectations of
short-term profitability. In critical situations, the latter tends to affect part of
the total stock of the debt rather than solely the new inflows.
Specific taxes can reduce short-term speculative flows by making them
more expensive compared to longer term flows (see Chapter IX). Beyond
these, the existence of external instability in the supply of funds requires
more complex counter-cyclical macroeconomic mechanisms, geared to sta-
bilize the volume of net flows recorded in each period. This can be done,
for instance, through the introduction of borrowing limits for national
banks or currency mismatch regulations (mechanisms in widespread use in
developed economies), the auction of external debt quotas, ad valorem taxes
that fluctuate according to the intensity of the supply of external funds, and
non-remunerated compulsory deposits like the encaje implemented by Chile
in the 1990s (see Chapter VIII).
Third, except in the case of compensatory flows, the regulation of capital
flows should consider their being channeled towards complementing national
savings and investment. That happens automatically with greenfield FDI.
Insofar as inflows are directed towards the domestic financial market and to
real estate and stock market bubbles, they can easily filter through to consump-
tion. The experience of various developing countries suggests that the final use
of funds is determined to a significant extent by the source of funding and the
way in which capital inflows are regulated and channeled. Consequently, in
incomplete capital markets such as in emerging economies like Chile, there is
a need to push the market explicitly towards a term structure consistent with
productive investment; the effective channeling of external funds towards
productive investment represents a necessary condition if external savings are
to contribute to capital formation and domestic development.
Third Part
Economic Recovery and the
Heritage of the Dictatorship,
1982–1989
IV
Policy Rectifications and Recovery
from the Debt Crisis, 1982–9*
During the 1980s, most Latin American countries were facing a dramatic for-
eign debt crisis. The sharp deterioration in international markets recorded
from 1981affected the emerging developing nations with unusual severity;
the drop in export prices and worsening access to the markets of industrial-
ized countries, the rise in international interest rates, and the sharp reduc-
tion in capital inflows all contributed to the strongest negative external
shocks in the past half century.
In 1982, the external shocks that struck Chile – the cutoff of bank loans,
the rise of international interest rates, and falling terms of trade – and the
“automatic domestic adjustment” carried out implied a 14% plunge in GDP,
the largest drop among LACs in that year. The Chilean economy began to
recover in 1984. However, only during 1988, after seven years of recessive
adjustment, did GDP per capita surpass its pre-crisis peak, recorded in 1981.
In the early stages, even after the explosion of the crisis in August 1982,
the adjustment followed a rigorous orthodox approach in continuity with
previous years. Domestic demand was curtailed sharply through an auto-
matic adjustment associated with losses of international reserves; most
public policies were kept “neutral.” In 1982, domestic demand fell by 25%,
diminishing sharply imports, but also domestic output and employment. It
is interesting to underline that a widely privatized economy, with free trade
and a fiscal surplus, suffered that huge 14% drop in GDP, notably more
severe than in the rest of the “non-reformed” LACs. Indeed, there was a
causal relation between the nature of reforms and the severity of the crisis.
Subsequently, the severity of the crisis led the authorities to introduce
deviations from the extreme market fundamentalism of the neoliberal
* Partially based on “Debt and growth in Chile: trends and prospects,” in David
Felix (ed.), Debt and Transfiguration? Prospects for Latin America’s Economic Revival,
M.E. Sharpe, New York, 1990. I appreciate the comments of José Pablo Arellano,
Sebastián Edwards, David Felix, Stephany Griffith-Jones, Manuel Marfán, and Carlos
Massad, and the research support of Andrés Gómez-Lobo.
109
110 Policy Rectifications and Recovery from Debt
approach. Public policy moved towards more pragmatic, active and selective
policies. Among other important changes, (i) export subsidies were estab-
lished; (ii) import tariffs were raised, (iii) in combination with a strong real
exchange rate depreciation; (iv) private debt was – directly or indirectly –
“nationalized,” with the public sector assuming the responsibility for a
significant part of private debt, with its share rising from one-third in 1981
to 86% of total external debt in 1987.
In 1987–8, Chile experienced three positive external shocks – a sharp rise
of the copper price, an agreement with creditor banks to postpone half of
interest payments from 1988 to 1991–3, and other foreign exchange sav-
ings resulting from debt swaps. As a consequence, the economy benefited
from the full relaxation of the acute binding external restrictions (scarcity of
foreign currency) faced since 1982. The elimination of liquidity constraints
allowed a significant expansion of aggregate demand in 1988–9. The supply
of output was able to respond vigorously, but essentially through the use of
idle capacity. Therefore, actual GDP growth between the previous peak, in
1981, and the new peak in 1989 averaged only 2.9% per year.
Section 1 briefly summarizes the situation just before the explosion of the
crisis. Then it focuses on the large negative net transfers taking place in 1982–9.
Then the macroeconomic adjustments in 1982–9 are analyzed, assessing the
evolution through the decade of (i) external shocks, (ii) demand-reducing
policies, and (iii) supply- and demand-switching policies. In section 2 a dis-
cussion of debt management after the crisis follows. Section 3 reports several
of the more pragmatic changes in policy design forced by the economic crisis.
Section 4 focuses on the recovery of GDP growth, investment, and its sustain-
ability; the performance of investment is stressed, arguing that the macroeco-
nomic environment was determinant of a low average investment ratio as
well the low economic growth recorded between the peaks of 1981 and 1989.
Section 5 analyzes the macroeconomic (im)balances by the late eighties.
1 The debt crisis1
(a) Accumulation of imbalances by 1981

As documented in Chapter III, Chile accumulated a large foreign debt in
1977–81. Borrowing was principally by private Chilean firms from for-
eign private banks, without any guarantee from the Chilean government;
in parallel, the public sector exhibited a surplus balance, which relieved the
Treasury of the need to borrow. Thus, there was a major privatization of the
sources and uses of debt (see Table IV.1).
Until late 1981, net capital inflows exceeded the absorptive capacity of
the domestic economy. Therefore, international reserves accumulated and
1
The 1980s performance of Latin America is examined in Feinberg and Ffrench-Davis
(1988), Devlin (1989), Griffith-Jones and Rodríguez (1992), and reports by ECLAC,
IDB, IMF, and World Bank.
Table IV.1 Foreign debt and public guarantees, 1975–89 (US$ millions and percentages of total)
Year Total Private with Public and pub- Private without Capitalized Total includ-
guarantee licly guaranteed guarantee debt ing capitalized
debt (1) + (7)
Amount % Amount %
(1) (2) (3) (4) (5) (6) (7) (8)
1975 5,453 21 4,667 85.6 786 14.4 5,453

1976 5,392 30 4,434 82.2 958 17.8 5,392
1977 5,763 46 4,479 77.7 1,284 22.3 5,763
1978 7,153 48 5,198 72.7 1,955 27.3 7,153
1979 8,790 76 5,369 61.1 3,421 38.9 8,790
1980 11,325 72 5,304 46.8 6,021 53.2 11,325
1981 15,700 69 5,623 35.8 10,077 64.2 15,700
1982 17,263 62 6,770 39.2 10,493 60.8 17,263
1983 18,133 1,815 10,497 57.9 7,636 42.1 18,133
1984 19,746 2,130 13,212 66.9 6,534 33.1 11 19,757
1985 20,607 2,348 15,242 74.0 5,365 26.0 85 20,690
1986 20,898 3,408 17,160 82.1 3,738 17.9 355 21,236
1987 20,722 3,276 17,894 86.4 2,828 13.6 1,187 21,844
1988 19,012 2,829 16,083 84.6 2,929 15.4 2,112 20,965
1989 17,569 2,120 13,568 77.2 4,001 22.8 3,436 20,629
Source: Central Bank of Chile, External Debt of Chile, 1990. Figures refer to end-of-year disbursed outstanding debt.
Note: (1) Total including IMF and debt payable in domestic currency and excluding short-term trade credit to non-bank debtors; the
latter amounted to US$800 million in 1985 and US$1.1 billion in 1989. (2) Chilean private debt publicly guaranteed. (4) and (6) are
percentages of col. (1). (7) Face value of debt capitalized through D.L. 600 and debt-equity operations through Chapter XIX. (8) Column
(1) plus 100% of debt capitalized through D.L. 600, plus the redenomination value of capitalized debt through Chapter XIX: averages
of 93% in 1985–7, 89% in 1988, and 84% in 1989.
111
exerted pressure for real exchange rate appreciation and import expansion.
Since the growth of imports outran that of exports, the deficit in the trade
balance underwent persistent and substantial increases. The external deficit
was led by the huge financial inflows; most inflows were financial, with a
low share of greenfield FDI. Undoubtedly, the lack of prudential regulation
and supervision facilitated the excess of external indebtedness and then the
depth of the bank crisis in 1983.
The Chilean economy had become particularly vulnerable to eventual
negative external shocks because the government had done away with
most of the regulatory mechanisms that coped with external instability
(instead relying on the automatism of the dollar standard or currency board)
and because the private sector (consumers and firms) was overleveraged
(Arellano, 1983; Eyzaguirre, 1988).
(b) The shocks leading recession and recovery in the 1980s

Chile suffered three severe negative external shocks in the early 1980s, which
had an unusually large multiplier effect on the domestic economy. GDP fell
by 14% in 1982, with manufacturing output declining by 21%. Expenditure-
reducing policies dominated over expenditure-switching policies.
The strongest shock was related to gross capital flows; rather than a sharp
rise in outflows, the larger shock was the drop with respect to the volumi-
nous inflows (and huge external deficit) recorded in 1981. After climbing
to 19% of GDP in 1981, the use of foreign savings fell to one-half of that
figure in 1982 and to one-fourth in 1983 (see Table V.2).2 Clearly the fig-
ure recorded in 1981 reflected a domestic policy, which allowed excessive
indebtedness for several years. Since 1977, the Chilean economy had been
adjusting (in both the productive sector and expenditures) to an unsustain-
able level of capital inflows. Consequently, aggregate demand was exceeding
domestic output by unsustainable amounts, and even without an interna-
tional debt crisis a major readjustment would have been needed in the near
future. The fact that in the second half of 1981 – long before the Mexican
crisis of August 1982 – GDP and employment had begun to fall (Marcel and
Meller, 1983, Table 1) is robust evidence.
Adjustment was unavoidable, no matter how well Chile might have man-
aged relations with creditor banks after 1982: the deep economic problems
had been sown during the previous boom years. Given the vulnerabilities
already accumulated and the international scenario, the best bargaining
2
The figures used in this chapter are based on official national accounts, calculated in
1977 relative prices. Naturally, the external sector weight (and its associated shocks)
depends on the year base. For example, in 1981 prices, it decreases; in 1976–8 and in
1986 prices, it increases. In general, in this book, in order to compare peso figures of
GDP with dollar figures of the current account, I use the average RER of 1976–8. That
gives an external deficit of 21% of GDP.
Policy Rectifications and Recovery from Debt 113
outcome certainly would have been a zero net transfer of financing, and
even that would have implied a huge decline in the net capital inflows com-
pared to 1980–1. A second shock was the increase of interest payments that
had been associated with growing indebtedness and an impressive jump in
the financial cost of foreign debt since late 1979. The third shock was the
decline in the terms of trade, which was led by the copper price drop.
Subsequently, an “automatic adjustment,” set in motion by the monetary
effects of the corresponding loss of international reserves, exacerbated the
reduction in domestic liquidity and aggregate demand. This resulted in a
17.2% drop in GDP per capita in 1982–3. As a consequence, there soon arose
a large gap between potential and actual GDP, and a decrease in investment.
The expansion of the production frontier, in turn, lost speed because of the
lower investment ratio.
The ideal adjustment in a perfectly flexible economy would eliminate
excess aggregate demand without generating a gap between actual and
potential GDP. In an economy with initial underutilization of capacity in the
production of tradables, adjustment with an appropriate dose of switching
policies could even achieve an increase in resource utilization and output. But
in an economy with price inflexibility, imperfect factor mobility, and limited
or confusing information, neutral demand-reducing policies (i.e. policies that
are intended to affect uniformly all expenditure components) can cause a
significant drop in output because the demand for both tradables and non-
tradables diminishes. In the real world, adjustment processes usually involve
a drop in output. This causes a lower rate of utilization of installed capacity
and, subsequently, a fall in gross capital formation.
Selective policies that facilitate switches in the composition of output and
expenditure can dampen output-reducing effects. A good combination of
expenditure-reducing and switching policies would tend to allow an out-
come closer to a constant rate of utilization of potential GDP (see Ffrench-
Davis and Marfán, 1989). In fact, if an excess of expenditure prevails, as in
1981, then aggregate demand needs to be reduced urgently. However, this
can be done in combination with switching policies that are geared towards
reducing the demand for tradables and encouraging domestic investment
and the supply of exports (Ffrench-Davis, 2006, Chapter II).
There now follows a brief account of estimates of the external shocks and
the paths taken by GDP, aggregate demand, exports, and imports during
1980–9. The purpose is to provide rough estimates of two components of
the economic costs of recessive adjustment: underutilization of installed
capacity and slackened creation of new capacity.
All variables fluctuated widely in this period, as is shown in Table IV.2.
Economic activity peaked in 1981 (row 1a). The actual GDP of 1981 is taken
as a good indicator of productive capacity as of that year. However, there was
some idle installed capacity, principally in the agricultural and manufactur-
ing sectors, as a consequence of the sharp trade liberalization with a strong
114
Table IV.2 Production, consumption, investment, and external shocks per capita, 1980–9 (in 1977 pesos as a share of 1981 actual
GDP)
Average
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1983–7 1988–9
1 (a) Actual GDP 96.2 100.0 84.6 82.6 86.5 87.1 90.5 94.1 99.3 107.4 88.2 103.4
(b) Potential GDP 97.0 99.0 102.3 103.0 101.5 103.0 103.9 103.9 105.6 107.4 103.0 106.5
2 Domestic 102.7 112.9 84.4 79.2 84.5 81.6 84.5 89.2 95.5 105.3 83.8 100.4
expenditure
3 Consumption 79.7 85.3 75.0 71.6 71.3 69.5 70.9 72.4 77.6 81.9 71.1 79.7
4 Gross fixed 17.0 19.5 12.7 10.6 11.4 12.9 13.6 15.5 16.9 20.0 12.8 18.4
investment
5 National savings 13.1 9.6 2.9 4.6 7.8 10.5 12.4 15.7 15.4 19.4 10.2 17.4
6 Non-financial –6.5 –12.9 0.2 3.4 1.9 5.5 6.0 4.9 3.8 2.1 4.4 3.0
current account
(a) Exports 22.8 20.4 21.1 20.9 21.9 23.0 24.9 26.6 27.8 31.6 23.5 29.7
(b) Imports 29.2 33.3 20.9 17.4 20.0 17.5 18.9 21.7 23.9 29.5 19.1 26.7
7 Terms of trade 1.9 0.0 –2.1 –1.1 –2.8 –3.3 –3.0 –1.6 1.5 0.9 –2.3 1.2
effect
8 Net interest and –3.4 –5.1 –6.7 –6.5 –7.3 –7.2 –7.2 –6.0 –6.4 –6.1 –6.8 –6.3
profits paid
9 Current account 8.2 18.5 9.3 4.6 8.5 6.0 4.4 2.7 1.1 2.4 5.2 1.8
deficit0ab
(a) Capital flowsab 13.1 18.7 4.8 2.6 8.5 5.6 3.6 2.8 3.2 3.5 4.6 3.3
(b) Change in 4.9 0.3 –4.5 –2.1 0.1 –0.4 –0.8 0.1 2.0 1.1 –0.6 1.6
reservesb
Source: Author’s calculations based on official figures of the Central Bank of Chile and Chapter I.
a
Unrequited transfers included. bFigures based on balance of payments data, which can differ from those of the national accounts.
exchange rate appreciation in preceding years. We have estimated that this

underutilization was more than compensated by “overuse” in the non-
tradable sector, especially by the value-added in the domestic marketing of
imports (see Chapter I, section 2 and Table I.2). Consequently, principally
based on the “value-added on an estimate of non-sustainable volume of
imports”, I consider the 1981 actual GDP as a 1% overestimation of poten-
tial GDP in that year (rows 1a and 1b).
In Table IV.2, all figures are adjusted by population growth and as
percentages of per capita GDP in 1981. Comparing any figure in a given
row to its value in 1981 indicates the change with respect to a situation of
both constant GDP per capita and constant shares of all other variables in
the table. For instance, the 1987 figure for fixed capital formation indicates
a fall from 19.5 in 1981 to 15.5. As a share of the absolute level of GDP in
1987, the decline was from 19.5 to 16.4% (15.5/94.1). Actual GDP per cap-
ita, in 1983–7, averaged 88.2% of the 1981 GDP. Because in the meantime
potential output was growing somewhat (Chapter I), the gap between actual
and potential GDP was even wider: 14.8 points per annum (rows 1a and
1b). Only in 1989 did actual and potential GDP converge. The significant
underutilization shows that demand-reducing policies had strong negative
effects on output, while switching policies (exchange rate and trade policies,
the selective fiscal policy, etc.) were weak. This shortcoming was reinforced
by the inflexibility in the composition of aggregate demand and supply.
The three external shocks are measured in rows 7, 8, and 9. Row 9 shows
the net use of foreign capital (current account deficit). Initially, after the
supply became dry, Chile could use its international reserves. That explains
why the financial shock does not appear as intense in 1982 as the cutoff in
external financing. The central bank lost reserves of 5% of GDP to compen-
sate for the foreign capital shortage.
The reserve losses, with a fixed exchange rate and passive monetary policy
at the outset of the crisis, gave rise to the sharp reduction in domestic
expenditure (row 2) on both consumption and investment. They were trans-
mitted to the external sector (row 6), reducing imports (sharply) and increas-
ing exports (slightly).
The 11.8% drop in actual GDP in the years 1983–7 with respect to 1981 is
the “static” output-reducing effect of the combination of external shocks and
weak switching policies. In the “dynamic” dimension, by 1983 investment
per capita had diminished by 45% with respect to 1981 (row 4), reducing
both potential GDP growth and the capacity to restructure the composition
of supply and demand, which is directly associated with the level of new
capital formation. Empirical work shows that the drop in capital formation
was significantly associated with the large gap between potential and actual
GDP (Agosin, 1998; Ffrench-Davis, 2006, Chapter III). Meanwhile, the drop
in capital formation weakened the capacity to restructure the composition
of supply and demand, which is a positive function of output expansion.
As a consequence, the low investment made a constructive adjustment of

the economy more unlikely, which became a determinant factor of the poor
behavior of real wages in that period (see Chapter VII).
An “automatic adjustment” mechanism, as was used in 1982, relies heavily
on the shock effects of demand-reducing policies. After any one-shot endog-
enous downward adjustment of domestic expenditure, as time goes by, some
switching in demand and supply composition gradually takes place sponta-
neously. That reallocation was assisted additionally (after several months)
by an increase in tariffs, the introduction of export subsidies, and a series of
sizable exchange rate devaluations with a return to a crawling-peg scheme
(see Chapter V).
Subsequently, there was a gradual recovery of output and expenditure.
This was greatly accelerated by a sharp positive shock in the international
price of copper; in 1988–9, after half a decade of severe recession, the siz-
able improvement in the terms of trade allowed a strong recovery of eco-
nomic activity. However, by 1988, GDP per capita was merely reaching the
level achieved in 1981, while investment per capita still was 13% lower.3
This was, in addition, the end of a cycle, because a macroeconomic over-
heating in 1989 demanded a significant new adjustment in late 1989 and
1990.
In the external sector, the automatic adjustment of 1982 was stronger and
came sooner on imports, which are highly responsive to drops in domestic
expenditure. Again, subsequently, time allowed a change in the behavior
of these two variables. Encouraged by large exchange rate depreciations,
exports grew faster in the latter part of the period.
It is interesting to measure the changes in exports and GDP between the
peaks of economic activity in 1981 and 1989. Cumulative export growth per
capita was significant: 55% and nine points as a share of GDP (six points if
compared to 1980). However, the cumulative GDP growth scarcely averaged
2.9% per year. This implies that the rest of GDP (non-exported GDP) growth
was quite weak (as shown in Chapter V).
The behavior of imports is also interesting. Their strong recovery ensued
following the spectacular drop recorded between 1981 and 1983 and the
swings in 1984–6, which were linked to a mini-recession experienced in
1985. Per capita imports grew by 56% between 1986 and 1989, a twofold
increase over the per capita export growth in those years (see Table IV.2,
row 6a). In the recovery following an adjustment such as the one carried out
in 1982, it is normal for imports to grow faster than exports. Nevertheless,
3
Figures in 1977 pesos. Later national account figures, in 1986 pesos, give 1988 as
the year in which the GDP per capita of 1981 was reached again. Weights of differ-
ent components of GDP change with relative prices. For instance, between 1986
and 1989, the terms of trade improve by the equivalent of 4% of GDP in 1977 pesos
accounts, and 7% in 1986 pesos; the main explanation is that trade is weighted in
1986 with a RER nearly two-thirds more depreciated.
the difference in the rate of growth between both variables was quite large,
as shown below. The disequilibrium was not evident in the trade balance at
current prices due to a remarkable improvement in the terms of trade – led
by the copper price – between 1986 and 1989.
2 Debt management in 1982–9
In 1982, external debt was four times the value of exports; debt service cap-
tured 89% of export proceeds. In the following years, there were five negoti-
ation rounds with creditor banks framed by agreements with the IMF and
the World Bank (three Structural Adjustment Loans (SAL) agreements).4
Liabilities with international commercial banks represented, as shown ear-
lier, more than four-fifths of the country’s total debt. The negotiations with
banks were geared towards rescheduling maturities, maintaining short-term
trade credit, and obtaining new loans to finance part of interest payments.
Since Chile fully paid its interest commitments, the remaining interest
payments were financed with a trade surplus and net transfers from public
creditors. Net transfers to bank creditors implied a negative flow of approxi-
mately 5% of GDP per annum in 1985–7.5
On the other hand, official creditors played a crucial role as net suppliers
of funds. Multilateral institutions were the largest lenders, the World Bank
and IDB provided positive net transfers of US$300 million per annum
(Table IV.3). This was a relatively large figure, as the net transfers of funding
from these two institutions to all Latin America averaged only US$1 billion
annually in 1985–7 and turned negative in 1987 (World Bank, 1989).6 As a
consequence, private creditor banks reduced their participation to 65% of
the total debt of Chile, while official creditors increased their share from 3%
in 1982 to 30% in 1989. The small outstanding debt with IFIs climbed, with
an average growth rate of 32% per annum.
With regard to disaggregation by debtors, after 1982 the state assumed
responsibility for an overwhelming share of total debt. In 1987, the public
sector share peaked at 86%, if the publicly guaranteed debt is included. This
figure was 36% in 1981 (Table IV.1). The debt “nationalization” was carried
out in three ways. The main one was the growing indebtedness incurred by
4
The five rounds – of 1983, 1984, 1985, 1987, and 1988 – are discussed in Ffrench-
Davis (1992). An analysis of the Latin American rounds is provided in Devlin and
Ffrench-Davis (1995).
5
Additionally, creditor banks received prepayments through debt swaps, either in
cash or in shares of Chilean firms. See below.
6
The “preferential” treatment for Chile is discussed in Felix and Caskey (1990) and in
Ffrench-Davis (1992). In 1985–7, Chile received 30% of the net new loans of the IDB
and the World Bank to Latin America, whereas it produced only 3% of the regional
GDP (at 1986 exchange rates). It must be recalled that the multilateral Chilean debt
was negligible in 1982.
Table IV.3 Net transfers of funds abroad by creditor, 1983–9 (US$ millions)
1983 1984 1985 1986 1987 1988 1989
Direct foreign 51 –51 –56 –139 –57 –102 –5

investmenta
Multilateral 167 248 349 266 253 109 33
institutionsb
Bilateral –57 –101 –69 –60 –38 213 82
officialb
Suppliersb –241 –154 –140 –7 131 –30 –19
Banks MLTb –178 –589 –718 –990 –988 –442 –791
Other MLTc –21 –69 –24 –86 –100 –119 –104
Short termd –834 906 0 198 –64 –423 405
Total –1,113 190 –658 –818 –863 –399 –399
Source: Calculations based on data from the Central Bank of Chile.

a
Net flows of foreign investment are disbursements minus net profits remitted after taxes; they
exclude debt swaps, capitalization of credits, and reinvested profits. bNet transfers from multilat-
eral banks of medium- and long-term foreign net loans minus actual interest payments. cIncludes
net interest payments to the IMF and medium- and long-term flows other than foreign debt.
d
Includes credit lines, trade credit, interest receipts, and errors and omissions. It excludes the
counterpart of debt swaps.
the central bank and other public organizations in order to provide foreign
currency to cover interest commitments, for both public and private agents.
This took place in an environment in which voluntary loans to LDCs had
disappeared. Second, the state granted ex post the public guarantee to the
foreign debt of domestic banks, under pressure from creditors and some
governments of developed nations.
Third, a debt swap scheme was started in May 1985, directed to reduce
debt (see Ffrench-Davis, 1990). It included two tiers that allowed the pre-
payment of debt by residents with creditor banks and debt/equity swaps
performed by foreigners (see Box IV.1); in that time, at market prices, the
foreign debt of Chile represented 130% of a sharply diminished annual
GDP. Later, there were also prepayments and direct write-downs by the
government (Elórtegui, 1988; Larraín and Velasco, 1990). The scheme was
based on: (i) the debt promissory notes (pagarés de la deuda externa) held by
creditor banks, which were priced at an average discount of 40% of the face
value in the international secondary market; and (ii) the direct capitaliza-
tion or conversion of external loans into equities. By December 1989, US$9
billion had been swapped through the different channels of this program;
there were implied “economic rents” (the difference between the face value
and market price of the debt notes) of about US$3.4 billion, which were
captured by the agents involved in these operations.
The main benefit for the Chilean economy accruing from the equity swaps
was the reduction of the outstanding debt stock; consequently, interest
Box IV.1 Debt-swaps programs

Proposals for debt conversion have been around for a long time in the aca-
demic and applied literature. As a matter of history, in the 1930s many Latin
American countries resorted to a form of conversion based on repurchase of
their own foreign debt bonds at substantially reduced prices: Chile enjoyed
an average discount of 69%. In the course of the 1980s crisis, such proposals
were renewed in Latin America, either in the framework of official programs
or as isolated operations. A rapid rise in the volume of swaps was encouraged
by high discounts of debt notes in international markets.
The Chilean program that operated between 1985 and 1990 was one of the
earlier formal schemes in Latin America and – given the size of the Chilean
economy – was relatively the most significant. The official scheme was
composed of two main tiers. One associated with foreign investment was a
debt–equity swap program (under Chapter XIX of the Compendium of Rules
for International Exchange in Chile). It was available to residents abroad and
designed to convert medium- and long-term debt owed by Chilean residents
to foreign banks. The latter could use their notes representing loans directly
or sell them at a discount in the international secondary market. The investor
then exchanged the debt paper for the debtor’s equity capital or with other
debtors for cash or notes in Chilean currency. The latter could be sold in the
domestic secondary market and then used for equity investment or to pur-
chase productive assets, subject to remittances of profits and principal after a
minimum period of stay.
Between 1985 and 1989, Chilean debtors had an average discount of 39.5%
with respect to face value. The gap between face value and secondary market
value represented an “economic rent” or capital gain. In Chapter XIX, the
foreign debt notes were recognized in Chile at an average of 88% of their face
value (that is, Chilean debtors caught only 12 of 39.5 points). Thus, a signifi-
cant share of discount was captured by investors or buyers in the secondary
market (the difference between 60.5 and 88%). From the point of view of the
foreign exchange market, the outcome implied that investors were benefiting
from a price of the dollar significantly above the market exchange rate.
The other formal component of the scheme was not associated with FDI.
Its aim was to reduce foreign debt (via Chapter XVIII) without subsequent
access to the official foreign exchange market. The central bank periodically
auctioned to residents in Chile the right to purchase debt notes abroad.
The auction enabled the bank to capture in 1985–9 around 30% of the
discount of the Chilean debt notes, the remaining 70% being captured by
other domestic agents. The purchase of debt paper was paid for with foreign
currency obtained in the “informal” domestic market, capital that had previ-
ously “fled” abroad, or “laundered” money from other foreign sources.
It is interesting to note that the central bank also made direct transactions,
receiving a larger discount, of between 42 and 50%.
Source: Based on Ffrench-Davis (1990).
payments and amortization commitments were diminished. However, they

also generated a series of other benefits and costs, the specifics of which
depended on the particular features and management of the swaps. For
example, debt–equity swaps involve a change in the composition of external
accounts in debtor nations; interest payments are replaced by profit remit-
tances, and amortizations are replaced by depreciation reserves and capital
withdrawals. Then, the source of external imbalance moves from “credit
accounts” to “investment accounts.” It must be stressed that, in order to
compensate for the expected rise in accrued revenue by foreign risk capital,
the program established minimum periods of stay for converted capital.7
The drop in bank debt due to debt equity swaps was nearly offset by a rise
in debt with multilateral institutions. It was only in 1988 that large swaps
produced a net drop in the total debt. That year the Latin American debt
also experienced a decrease, though to a lesser extent: only 2% in the region
compared to 8% in Chile (ECLAC, 1988).
After the negotiations with creditor banks in 1987–8, which had addi-
tionally relaxed the debt service, the external environment faced by Chile
improved notably. The price of copper rose sharply, as did other export
prices (such as cellulose and fish meal). The terms of trade improvement
together with the postponement of debt service for 1991–58 implied that the
binding external constraint prevailing in 1982–7 had disappeared in 1988.
In fact, these positive shocks led to a spectacular increase in the foreign
exchange available (ECLAC, 1988), which allowed the full use of productive
capacity and the overheating of the economy in 1989.
The improvement in domestic financial conditions from the mid-1980s
was also reflected in the behavior of stock prices indexes (which tend to
be notably pro-cyclical). Table IV.4 shows that between the second half of
1980 and 1985, for instance, real prices in pesos dropped by 64%.9 Given
that the recession involved exchange rate devaluation as an adjustment
mechanism, that decline is strengthened when stock market prices are
expressed in US dollars, showing an even larger real price drop of 77%.
Therefore, when debt–equity swaps were initiated in 1985, the space for
7
A significant share of debt–equity conversions was associated with privatization of
public enterprises or firms temporarily under government administration as a result
of the economic crisis of 1982. Consistent with the government’s desire to privatize as
quickly as possible, with either local or foreign buyers, the rates of return on converted
capital were considerably higher than interest rates. The creditor banks, which made
some of the largest transactions directly and acquired 40% of the gross capitalization,
thus converted “bad” (risky) loans into “good” equity capital (Dornbusch, 1988).
8
The rescheduling implied that the amortizations of the stock of debt by 1989 would
multiply by 3.2 between the annual averages of 1987–8 and 1991–5.
9
Making comparisons of the index over time is troublesome. Aside from the divi-
dends policy in cash and in stock issues, the Chilean economy suffered dramatic
changes of firm ownership, firm boundaries, and asset and liability structures.
Table IV.4 Indexes of stock prices in pesos and converted to US dollars, 1980–9
(1980 100)
Year Real Stock price index
exchange rate
In pesos Converted
to US$ real
Nominal Real
(1) (2) (3) (4)
1980 I 104.2 87.5 93.1 89.4

II 95.8 112.5 105.6 110.2
1981 I 87.2 102.9 88.0 100.9
II 82.4 87.2 71.3 86.6
1982 I 81.4 79.9 64.4 79.2
II 111.5 78.3 56.5 50.7
1983 I 120.8 65.3 41.5 34.4
II 116.1 63.0 35.6 30.7
1984 I 119.6 78.7 41.5 34.7
II 127.0 77.4 36.8 29.0
1985 I 140.4 87.3 35.1 25.0
II 161.9 113.1 40.9 25.3
1986 I 164.7 164.0 54.1 32.8
II 168.1 229.1 70.6 42.0
1987 I 170.9 318.7 89.2 52.2
II 176.2 394.8 100.4 57.0
1988 I 186.5 407.8 97.2 52.1
II 183.5 461.9 104.5 56.9
1989 I 176.0 629.7 131.5 74.7
II 185.0 686.3 129.2 69.8
Source: Based on data from the Central Bank of Chile and the Santiago Stock Exchange; averages
per semester of monthly figures.
Notes: (1) Nominal index deflated by the CPI and inflated by the index of external prices faced by
Chile. (3) General stock price index deflated by the CPI. (4) Equal to (3) 100/(1).
recovery was enormous. In fact, by 1989 the real index in US dollars had
recovered by 187% (and by 243% in real pesos). It is interesting to underline
that the strong increase in stock market prices by 1989 involved a substan-
tial capital gain for “early” investors in debt–equity swaps.
3 A more pragmatic economic policy
After the first years of adjustment in response to the debt crisis, a more
pragmatic macroeconomic policy, departing from the straight “neutrality”
of the reforms of the 1970s, was put into practice. It aimed, with “selective”
policies, to expand non-traditional exports, increase savings, and strengthen
the corporate and financial sector. It implied recognition by the authorities
that there were failures in strategic areas such as the financial and export
markets that needed to be corrected (Larraín and Vergara, 2000).
When external funding dried up in 1982, an increase in national savings
(that is, domestic savings minus interest and profits accrued to foreign capi-
tal, which had collapsed in 1980–3), turned out to be essential for financ-
ing investment. In order to encourage increased private savings, in 1984 a
tax reform was implemented. Moreover, several measures were adopted to
cope with the weakening of public finances.10 These included setting wage
increases for the public sector below the inflation rate and reductions in
several categories of public spending, including pensions and other social
expenditure (see Table VII.1). As a result, the fiscal deficit improved from
3.5% of GDP to a balance in 1987 (Larraín, 1991). Given the way in which
this balancing was achieved, it had a significant regressive bias, as discussed
in Chapter VII.
With respect to private non-financial companies, from 1982 most of
them suffered the effects of the high interest rates, depressed demand, and
exchange rate devaluation on their foreign debt (particularly, in the case of
producers of non-tradables). This, together with opaque banking practices
and the permissive regulatory framework generated by the financial reforms
of the 1970s, laid the foundation for an enormous financial crisis, which
forced the government to intervene in several financial companies that were
experiencing solvency problems, including the two largest private banks.
The government reacted to the financial crisis by implementing an aid
program for local debtors and banks (with an estimated cumulative cost of
35% of annual GDP), which included, among other measures, a preferential
rate for dollar debts, loans at subsidized rates for the financial sector, and
the central bank’s purchase of the banks’ non-performing portfolios, with a
commitment from the latter to repurchase them (Sanhueza, 1999).
In the mid-1980s, a second round of privatization was carried out. This
affected forty-six companies (including financial institutions that had been
taken over in 1983, and traditional public firms, such as the large electric
power company, ENDESA), which were quickly transferred to the private
sector (Hachette and Lüders, 1992; Devlin and Cominetti, 1994).
The Chilean banking crisis, which followed on the heels of a mas-
sive capital surge in the late 1970s, left in its wake a number of valuable
lessons that were reflected in Chilean rules or “legislation.”11 Therefore,
a deep financial reform was carried out, which implemented a rigorous
prudential supervision and strengthening of the regulatory agency (the
Superintendencia de Bancos). The prudential regulation and supervision
10
Weakening caused by the heavy financial cost to the Treasury of the 1981 social
security reform (see Chapter VII), as well as of depressed economic activity.
11
See Valdés-Prieto (1992), Aninat and Larraín (1996), Reinstein and Rosende (2000),
and Held and Jimenez (2001).
included the continuous monitoring of the quality of bank assets; strict

limits on banks’ lending to related agents; automatic mechanisms to adjust
banks’ capital when its market value falls below thresholds set by regula-
tors; and the authority to freeze bank operations, prevent banks in trouble
from transferring funds to third parties, and restrict dividend payments by
institutions not complying with capital requirements.
Among other significant changes made after the debt crisis were reforms
to trade and foreign exchange policies. Some tariff protection was reintro-
duced, with an increase in the uniform rate from 10 to 35% between
1983 and 1984, and price bands for the three main agricultural products,
dumping rules against unfair competition, and a 10% drawback to minor
exports were established (see Chapter V). The reshuffling of exchange rate
policy was dramatic, with a return to a crawling peg, after having pegged
the nominal peg for three years (1979–82), and having jumped to a totally
flexible rate for a brief period. Between 1983 and 1989, the authorities used
a floating band of ±2% (widened to ±3% in 1988 and to ±5% in mid-1989).
The “official” rate was devalued daily, in line with the differential between
domestic inflation and an estimate of external inflation. On a number of
occasions, discrete nominal devaluations were added, which contributed to
achieve a remarkable degree of real depreciation following the 1982 crisis:
130% between 1982 and 1988. Policies were accommodated to the severe
shortage of foreign currency in Chile (initially, a sharper shortage than in
most of the region, as discussed in section 1). These devaluations allowed a
stronger response to external conditions by encouraging both exports and
a recovery in the production of importables and reducing the demand for
imports.
The monetary policy also experienced a significant change. The policy
of totally free interest rates was replaced by active open market operations
of the central bank, and “persuasion” that allowed for significant inter-
est rate reductions in bank loans. Consequently, the average real cost of
38% recorded between 1975 and 1982 diminished to an average of 13% in
1983–5 and to 8% in 1986–8.
Thus, a competitive exchange rate, the introduction of protection to
tradables output, reasonable interest rates for the real sector, and a set of
public actions of sectoral support were key allocative policies behind the
new exporting take off, the recovery of the output of importables and the
restructuring of non-tradable production, which allowed closing the gap
between actual and potential GDP.12 But the process took seven long years,
12
Inflation, which had reached one digit by 1981 (even recording negative figures by
early 1982), averaged 21% in 1982–9, exhibiting a decreasing trend up to the over-
heating in 1989. Even though the government was concerned about inflation, the
emphasis was placed on the recovery of economic activity; that is another relevant
difference between the 1970s and the 1980s.
with the Chilean economy working for most of the period under a large
recessive gap; that is one main variable explaining the low average invest-
ment ratio in 1982–9.
4 Strong economic recovery, weak average growth
The Chilean economy had been very unstable in past decades. In terms of
output (GDP), in the 1980s Chile exhibited the worst drop in 1982 and the
greatest expansion in 1989, compared to all other Latin American coun-
tries. Both changes (i.e. the fall and the expansion) were related to external
shocks: worsening in the terms of trade and foreign debt service, in the first
case, and a sharp terms of trade improvement in the second.
Apologetic accounts of the economic policy implemented by the Pinochet
regime emphasize the evolution of the Chilean economy after the fall, usu-
ally starting to count from 1986, focusing particularly on the 10% increase
in GDP in 1989. This is misleading for two reasons. In the first place, the
greater the fall in economic activity, the stronger the subsequent recovery
can be. Consequently, it cannot be ignored that there was a huge GDP drop
in 1982–3, which implies an artificially low base of comparison for perform-
ance only measured since 1986; additionally, by 1986 there was a sizable
gap between actual and potential GDP, which explains the high increase of
actual GDP in parallel with a low investment ratio. In assessing the results
of economic policies, therefore, it is necessary to weigh the overall economic
and social effects and not just one part of the results (i.e. solely the recovery)
because this provides a biased view, particularly in an economy that has had
so many ups and downs. Notice again that it took seven years for GDP per
capita to overtake the 1981 level. Furthermore, only in 1991 did the average
wage surpass that of 1981.
The second reason is that the great expansion of 1989 was not sustain-
able. It was based on transitory factors that allowed a sharp recovery of GDP
growth, and they could not avoid an overheating of the economy, causing
negative effects on macroeconomic and macrosocial balances, which actually
required a demand-reducing adjustment in 1990. In fact, the modest invest-
ment ratio allowed a limited sustainable GDP growth of about 4% in 1988–9.
Moreover, in the whole period covered by the Pinochet regime, the average
increase of GDP was a mediocre 2.9% per year, including drops of about 15%
in 1975 and 1982, and increases of 8% in 1977 and 1979 and 10% in 1989.
To the contrary, export performance was successful in achieving a vigor-
ous overall expansion, in both halves of the regime. The volume of exports
experienced a real increase of 11% per year between 1974 and 1989 (15% for
non-traditional goods). In per capita terms, exports explain all the modest
increase of GDP, while the average of the rest of the economy stagnated (see
Chapter VI).
The main reason why economic growth was so moderate from 1973 is that
investment had been low. Average investment in the sixteen-year period was
15.7% of GDP. This figure is much lower than the 20.2% ratio recorded in
the 1960s (both figures measured in the same constant prices). This is why,
notwithstanding changes in efficiency, productive capacity expanded con-
siderably more in the 1960s: the annual rate was 4.3%, in contrast with 2.9%
during the dictatorship.
Associated with this poor global performance, in the 1970s and 1980s, real
wages experienced a significant drop: in 1989, wages were 5% lower than
in 1981. Four factors can explain that outcome: (i) GDP per capita rose by
barely 1.2% per year in 1982–9; (ii) a significant share of production leaked
out abroad to service the heavy foreign debt accumulated between 1976 and
1982; (iii) the investment ratio per member of the labor force was 15% lower
in the eighties than in the sixties; and (iv) the Chilean economy became
much more regressive and repressive for workers (see Chapter VII).
Low levels of investment were related to the fact that productive capacity
was underutilized for many years; as was stressed in Chapter I, a signifi-
cant gap between actual and potential GDP prevailed during most of the
sixteen years of the Pinochet regime. It is worthwhile to repeat that when
capacity is underutilized, it tends to reduce the average return on invest-
ment and the liquidity of firms, and consequently new capacity creation is
deterred.
On the other hand, real interest rates remained very high until 1982 (38%
in 1975–82), which undermined the financial position of many enterprises
and discouraged new projects. However, after 1982, interest rates were
reduced sharply, with some intervention by the authorities, but an addi-
tional negative factor was added: the distressed financial situation of many
firms and the heavy foreign debt service. This implied that a significant
share of national savings was not invested domestically but was used for
interest payments to foreign creditors.
A last factor that discouraged productive investment was that entrepre-
neurs devoted a significant portion of their efforts to the purchase of existing
assets in a highly active process of property transfers. These were induced by
the recessive imbalances that prevailed for many years in the Chilean econ-
omy, by an intensive privatization of state-owned enterprises, and later by
the government takeover of private enterprises that had gone bankrupt in
the early 1980s and were subsequently subjected to reprivatization. But not
all capital formation was depressed. In fact, as already underlined, the policy
shift towards pragmatism included significant incentives for the private sec-
tor to invest in exportables and public utilities. As shown in Moguillansky
(1999), huge transfers took place, given the form of privatization or repriva-
tization that was conducted, by means of the rescue of agents caught in the
debt crisis, the economic rents involved in debt–equity swaps, the highly
efficient drawback to exporters, and, of course, the much needed dramatic

exchange rate devaluation.
What about productivity? Productivity was much higher after the open-
ing of the economy and the liberalization of markets in several firms, in the
successful surviving firms (Tybout et al., 1991). But that progress tended to
be offset by a higher mortality rate among enterprises and a low average rate
of utilization of productive capacity in a majority of producers.
Consequently, there was greater heterogeneity among the surviving
enterprises, including the new ones, than in previous years. Modernization
reached into many economic sectors, but did not reach the majority of
them. For example, there were peasant sectors, small businesses in trade
and services, industrial workshops, and self-employed individuals that
suffered severe capitalization difficulties and operated, in part, with lower
productivity and income levels than two decades before. They managed
to stay in business or find a job by scaling down their incomes. Recall the
strong fact that average GDP growth was low and income distribution
worsened significantly, data that fit the increased structural heterogeneity
of the actual productivity of the Chilean economy recorded through these
sixteen years.13
The foregoing coexisted with the benefits of having a growing segment
of highly productive and dynamic enterprises, both urban and rural, and
several of them becoming successful exporters. This was a significant
and positive legacy for the future. The export-oriented economic policies
and the establishment of a market for natural resources and land (which was
certainly aided by the previous land reform) provided the favorable breeding
ground that enabled entrepreneurial initiative, particularly among members
of the younger generation, to venture into risk taking and an exploration
of new forms of production. But, these positive features cannot overcome
the evident fact that global growth and equity failed in the first and second
halves of the dictatorship.
5 Macroeconomic (im)balances by 1989
There is a long history in Latin America verifying the importance of sup-

porting basic macroeconomic balances. These refer to the amount of money
issued, the fiscal budget, the gap between export earnings and import
expenditures, the aggregate demand level and several macro-prices, such as
the exchange rate and interest rates. The neoliberal approach to macroeco-
nomic balances has focused mostly on the control of inflation and fiscal dis-
cipline, what we have called the financieristic approach, based on those two
13
The degree of informality is reflected, partly, in the fact that, by 1989, only 41%
of the labor force was contributing to the privatized social security system created by
the dictatorship in 1981.
pillars (see Fisher, 1993; Ffrench-Davis, 2006, Chapter II). There is no doubt
that persistent large imbalances in these two areas have caused severe crises
and hyperinflation, and have ended in sharp recessions in the past; several
Latin American economies, including Chile in 1971–3, exhibited fiscal
imbalances leading to pervasive hyperinflation. Therefore, any consistent
policy must take care of those two essential components of macroeconomic
balances. However, that is insufficient; moreover, frequently inflation has
been repressed artificially, with exchange rate appreciation or fiscal balance
being achieved by repressed expenditure in education and public works
(see Easterly and Servén, 2003).
The serious caveat of the neoliberal approach is the underestimation of
other key elements of sustainable macroeconomic balances, in order to
become functional for development. Thus, there is a third balance, which
should be explicitly included in macroeconomic equilibria, consisting of the
relationship between aggregate demand and potential GDP and the behav-
ior of macroprices, such as the exchange rate. The magnitude of the gap
between actual and potential GDP affects very significantly the level of capi-
tal formation, and subsequently productive employment; the behavior of
the real exchange rate is determinant of the performance of the production
of tradables and their link with the rest of GDP, as shown in Chapter VI.
In general, the Pinochet government was very cautious with regard to
monetary expansion and maintained the fiscal budget either balanced or
in surplus. A positive consequence of these policies was the convergence
towards a moderate one digit inflation in 1981 (quite moderate by Latin
American standards). Nevertheless, there is evidence that such policies are a
necessary but insufficient condition for economic stability. In fact, the exter-
nal sector underwent a prolonged and rising deficit, which was financed
with external borrowing for a long period of time (from 1977 to 1981). This
led to the 1982 crisis, even before the explosion of the region’s debt crisis.
On the other hand, during most of the second half of the regime(i.e the
1980s) a large gap again prevailed between productive capacity and its utiliza-
tion. Following the severe economic recession of 1982 and a few subsequent
years marked by uncertainty and huge unemployment of labor and capital,
a strong and sustained recovery of economic activity began in 1986. Fiscal
policy moved from a large surplus in 1981 to a significant deficit in 1982–5. In
1986–7, a sustainable recovery of economic activity took place, with a
progressive decline in the capacity utilization gap, and a return of fiscal
policy to a significant surplus in 1987;14 as said, relevant pragmatic adjust-
ments to trade, monetary and exchange rate policies contributed to this
recovery. However, in the following two years (1988–9) the situation
changed: demand and economic activity expanded at great speed (at more
14
The surplus was achieved, partly, with a sharp drop in social expenditure (see
Chapter VII, Table VII.1) and investment in infrastructure.
than twice the speed of increase in capacity) and this culminated, in 1989,
in an overheated economy when the growth rate of GDP reached 10%.
The great expansion was led, principally as of late 1987, by an increase in
private aggregate demand as a result of a considerable monetary expansion,
a significant reduction in taxes, and real exchange rate revaluation, which
made imports cheaper.15
Real aggregate demand expanded by 22% in 1988–9; GDP increased by
18%. The gap between the change in expenditure and production was
covered by an unexpected improvement in the terms of trade (i.e. the price
of copper and other items). Productive capacity (potential GDP) expanded
only by a total of 7–8% in the biennium. The increase in GDP was based
on the existence of installed capacity, which became exhausted around
1989 (see Chapter I, Figure I.1), when the economy exhibited evident
signals of overheating. Actual GDP had reached the productive frontier, and
consequently any additional increase of output required, in contrast to the
preceding years, a concomitant increase of productive capacity, which was
expanding slowly and quite below the rapidly rising aggregate demand.
This imbalance emerged as evident with a notorious acceleration of
inflation and a strong rise in imports. In fact, on an annualized basis, infla-
tion reached 31% between September 1989 and January 1990, a threefold
increase over the 10% rate recorded by late 1988. The volume of exports
experienced a vigorous increase of 23% during the biennium, but imports
grew 41%. The gap was covered by the extraordinary inflow of funds gener-
ated by the previously mentioned improved terms of trade. This positive
shock was centered on the price of copper, which doubled between 1985–6
(US$0.63 per pound) and 1988–9 (US$1.24). Copper represented half of
Chile’s exports, was produced primarily by a state-owned enterprise, and
yielded high revenues from taxes and profits to the government.
Due to the intensity of maladjustments, the Pinochet government carried
out successive adjustments, which included a devaluation in June 1989,
increases in the interest rate in April and September 1989, and finally a
sharp one in January 1990.16 Thus, in March 1990, when the democratically
15
The supply of money (M1A) increased by 56% in the twelve-month period ending
in October 1988 (the same month in which the plebiscite that the Pinochet govern-
ment was forced to call was carried out). Taxation decreased by approximately 4% of
GDP in 1988–9, and the real exchange rate underwent a 12% revaluation between
January 1988 and June 1989, when the Pinochet government reversed this policy due
to the accelerated increase in imports. The reversal was effective, since it achieved a
devaluation from the floor (appreciated bottom) to the ceiling of the exchange rate
band, which worked fluently in those years.
16
The latter was carried out after the central bank had begun operating as an inde-
pendent institution, according to the tailor-made legal framework imposed by the
Pinochet regime, which was enforced as of December 9, 1989; that is, five days before
the presidential election and sixteen and a half years after the coup of 1973.
elected President Aylwin took up the presidency, interest rates were notably
higher, reaching real active levels over 16%.
In short, when the democratic government was inaugurated in 1990, there
was a macroeconomic imbalance, and an adjustment process was already
under way in order to correct it.
V
Export Development during the
1980s*
The successful growth of the export sector can be seen as an outstanding

outcome of the Chilean economic reforms since 1973. The debt crisis was
an additional incentive for export expansion, given the binding external
restriction that Latin America as a whole experienced in the 1980s. An
annual growth of the volume, averaging 9.1% in the over one-third of a
century elapsed up to 2008, places the export performance as an outstand-
ing quantitative achievement of the Chilean economy;1 in that same period,
WTO data show that the volume of world trade averaged a 4.8% annual
growth. Notwithstanding the Chilean exporting success, the pulling capac-
ity of its exports over the rest of the economy has been robust only in some
periods, without achieving a sustained positive association across time. For
instance, in the decade on which this chapter focuses (1982–9), exports of
goods and services grew at an annual average rate of 7.8%,2 whereas GDP
only expanded 2.9%. This poor pull effect has structural and macroeconomic
causes.
It is important to distinguish different stages in trade performance and
policies in the one-third of a century covered by this book. The first half of
the military regime, characterized by the early trade liberalization, high rates
of growth of exports until 1980, and the parallel recovery of GDP recorded
by the Chilean economy after the deep crisis of 1975, would seem to indicate
a successful trade reform. However, this first trade reform3 was characterized
* I appreciate the comments of Sebastián Sáez.

1
Several interesting studies analyze the export performance in Chile. See, for instance,
Meller and Sáez (1995), Meller (1996a), Agosin (1999, 2007), Sachs et al. (1999), Herzer
and Nowak-Lehmann (2004), Macario (2000), and DIRECON (2009).
2
These figures correspond to the gross value of exports of goods and services
in 1996 constant prices (or quantum of exports). In Chapter VI I report esti-
mates of exports net of imported components; that is, value-added to GDP by
exports.
3
The first trade reform, detailed in Chapter II, was implemented between late 1973
and June 1979.
130
Export Development 131
by an import liberalization bias more powerful than the export promotion

policies: a trade deficit of 12% of GDP in 1981 represents evidence of the
bias; the liberalization of imports reduced the production costs of exports,
but this pro-exports effect was much weaker than the overall pro-imports
effect.4 The imbalance caused by the import-led (instead of a direct export-
promoting) nature of the reform was reinforced by exchange rate apprecia-
tion in the late 1970s, building up a discouraging environment for exports
by the early 1980s (see Chapter II). Actually, by 1981 most exports were
decreasing in quantum, reinforcing the negative impact of worsening terms
of trade.
A second export takeoff emerged after the debt crisis of 1982, as a result
of improved conditions, such as a sharply depreciated real exchange rate
in the 1980s and somewhat active public policies; these included a rather
pragmatic second trade “reform” that reintroduced some protection to
importables and direct incentives to non-traditional exports.
This chapter covers the main features of the Chilean export model since
1973, with special emphasis on the export take-off during the eighties.
Section 1 briefly summarizes the channels by which exports can contribute
to national development, influencing the pace of economic growth. Section 2
analyzes the role of trade and exchange rate policies, both marked by the
effects of the debt crisis. Then, the strategy of the seventies, simply based
on unilateral tariff liberalization, was replaced by a policy that, maintaining
the uniform tariff, raised its level, and introduced some interventions on
imports and diverse direct export incentives. Section 3 reviews the export
sector performance, in terms of dynamism, the composition of its basket,
and diversification. Section 4 starts the evaluation of the links between
exports, recovery, and economic growth, issues that are deepened in
Chapter VI, which analyzes trade policy since 1990.
1 The role of exports in national development
There are different channels by which exports can boost a country’s rate of
economic growth: by (i) generating the foreign exchange that enables the
purchase of imports required for economic expansion, at a lower real cost
than that of producing import substitutes; (ii) exploiting economies of scale
and specialization, directed to the broader external markets; (iii) giving rise
to linkages with other local activities, which make it possible to exploit
4
Estimated with GDP in pesos converted into US dollars with the average RER of
1976–8, expressed in currency of 1981. In current prices, the trade deficit in 1981
was 8% of GDP. Given the sharp appreciation of the exchange rate, the current dollar
value of GDP appeared artificially high in 1981, which in turn determined external
deficit and debt ratios that implied an underestimation of the severity of rising exter-
nal imbalances. See Chapter III.
132 Export Development
underutilized capital and human resources and stimulate new investments

by domestic suppliers to exporters; (iv) bringing greater contact with inter-
national best practice, exposing export activities and their suppliers to the
demands of competitiveness, and generating externalities (knowledge and
spillovers, partly but not only via the direct linkages in iii) in the domestic
economy;5 and (v) exports also playing a macroeconomic role. In economies
under a binding external constraint (BEC), as in the 1980s, increased exports
contribute to raise the rate of use of resources thanks to the positive impact
of the increased supply of foreign currency on effective demand.
From the standpoint of generating foreign exchange, what is important is
not only its volume at present, but also its future rate of growth. Thus, it is
relevant to promote exports of goods and services that feature a growing and
vigorous external demand over time, particularly in markets in which the
country is not a small supplier. One of the pitfalls of a strong concentration
of traditional exports in natural resources-based products is that their long-
term world demand tends to grow slowly. In addition, many developing
countries put pressure on the supply, causing a fall in international prices. In
order to sustain a high growth of export quantum it is necessary to diversify
the export basket towards products with more dynamic demand.
As regards the spillover effects from trade on the rest of the economy, the
larger the number of firms and productive sectors associated with exports,
the greater tends to be the impact on the production structure. Similarly,
this impact increases along with the national capacity to absorb the knowl-
edge acquired by export firms. This is why it is important to deepen the link-
ages between export activities and the rest of the productive system, and to
improve domestic mechanisms for the transfer and dissemination of tech-
nology, as well as human resources training. This is, in general, the concept
of systemic competitivity (see Fajnzylber, 1990; Ocampo and Parra, 2007)
and its feedback with exporting quality: vigorous exports in a low productiv-
ity domestic economy do not involve a dynamic economic growth; likewise,
efforts aimed at systemic competitiveness in a closed small economy tend to
end in a deadlock quickly.
Thus, public policies to support exports are key ingredients to enhance
development. A strategy of international insertion must give priority to
growth and diversification of exports as a long-term goal. But, at the same
time, the strategy has to strengthen the linkages of exports with the rest of
5
The generation of externalities is associated with horizontal and vertical diversifica-
tion of exports. Agosin (2007) develops an interesting empirical analysis, comparing
the experiences of Latin America and Asia. He concludes that the diversification
(i) tends to reduce the volatility of exports value and with it that of GDP, and (ii)
stimulates clusters and the transmission of best practices and “learning by doing.”
Herzer and Nowak-Lehmann (2004) make an estimation of these effects for Chile,
concluding that they are empirically significant.
the economy (see Chapter VI, section 3), by coordinating export promot-
ing policies with a set of economic policies including measures to promote
productive development, via technological innovation and labor training –
all in a context of consistent exchange rate, fiscal, and financial policies
(see Bouzas and Keifman, 2003).
2 Trade policy in the 1980s: a departure from orthodoxy
In response to the deep recession of 1982, Chile had to adjust its policies to
face the severe external constraint and stimulate domestic recovery. In fact,
there were some significant changes from the orthodox model implemented
in the 1970s to a more pragmatic economic approach. This included some
significant stepbacks, from an orthodox perspective, with respect to the
first reforms (see Chapter IV and Moguillansky, 1999). In this new context,
the generation of a trade surplus as a means of serving the foreign debt
was given top priority. The strategy implied both the reduction of imports
and the promotion of exports, which was done through three actions:
an increase in the uniform import tariff, the use of a battery of instruments
to encourage exports such as an innovative drawback for non-traditional
items, and an active exchange rate policy to ensure external competitiveness
and enhance the production of foreign currency by the Chilean economy.
The strong financial external constraint stimulated substantial devaluations
in Chile as well as in the rest of Latin America.
(a) The export momentum in the 1970s

In late 1973, a trade policy reform was launched that covered the elimina-
tion of all non-tariff trade barriers, a sharp process of tariff reduction, and
the establishment of a single exchange rate. Although it was not one of the
initial goals, by June 1979 a uniform tariff of 10% had been established,
and the exchange rate was pegged. However, there was no systematic direct
effort for export promotion. Trade policies responded to the assumption
that market forces would reallocate (without net costs) resources expelled
by import liberalization towards the export industries in which the country
had comparative advantages; this would lead to both fast export and GDP
growth.
During the first trade liberalization program, sharp tariff reductions and
the dismantling of quantitative controls appear to have had a great impact,
favorable to export dynamism. The point of departure was one in which
there was enormous room for the reduction of the costs of producing
exportables by substituting imported inputs for domestic ones, and broad
opportunities for improving productivity. Additionally, in late 1973 there
was a significant underutilization of installed export capacity, due to the
huge distortions then prevailing in the Chilean economy. That is the more
significant reason for the spectacular increase and diversification of exports
during 1974–80, especially focused in 1974.6 Nevertheless, the tariff liber-

alization of 1974–9 was unnecessarily costly, because a sizable share of the
manufacturing installed capacity was destroyed, instead of being gradually
reallocated to export sectors (Agosin, 1999).
Just as import substitution has an “easy” stage, so there is an initial easy
stage in the promotion of exports in emerging economies. The increase
in non-traditional exports in the 1970s relates in general to this stage. In
fact, it relied on rich natural resources and underutilized installed capaci-
ties, intensified by the great depression in domestic demand in 1975–6
and its subsequent slow recovery. This situation initially enabled exports to
expand without major investments. The increase in exports was spurred by
four additional factors. First, a crawling-peg exchange rate policy of mini-
devaluations was applied, which, despite contradictory movements from
1976, in combination with the sharp drop in labor costs initially encour-
aged exports. Second, the presence of Chile in the Andean Pact until 1976
provided an enlarged market for more than one-third of the increase in
new exports (Ffrench-Davis, 1979). Third, there was a reduction in the cost
of imported inputs to those exporters who had not benefited from tariff
exemptions previously. Finally, together with the above factors, the privi-
leged position assigned to export promotion in the official rhetoric gave
a significant pull to the hitherto incipient export mentality of domestic
entrepreneurs.7 However, the dynamism of exports transmitted quite weakly
to the rest of the economy. That feature was replicated during the 1980s.
The fact remains that, because of the recessionary situation in which most
of the reform was implemented, its suddenness, and the subsequent trends
exhibited by the exchange rate (appreciation) and interest rates (notably
high), the dynamism from the strong export performance was weakly
transmitted to the rest of the economy; indeed, fixed investment was far
below its historical level and, as shown in Chapter II, the economy exhib-
ited a sharp de-industrialization during the 1970s.
(b) Export promotion and the second trade reform in the 1980s
The domestic and balance-of-payments crises that hit Chile in 1982, as a
result of a combination of errors in economic management and the three
6
The growth rate of the volume of exports of goods was spectacularly high in 1974
(38%); growth of copper exports experienced a huge jump based on the maturing of
large investment projects of the late 1960s; but the volume of non-copper exports also
expanded by a notable 34%. A factor that contributes to explain the strong increase in
the volume of non-traditional exports in 1974–6 was the dynamic demand of coun-
tries of the Andean Community.
7
The official policy included active promotion through a public institution
(PRO-CHILE). This involved a deviation with respect to orthodoxy, which claimed to
base the promotion of exports exclusively on the liberalization of imports and the
supposed compensatory exchange rate devaluation. As the orthodox approach gained
control in public action, PRO-CHILE rapidly lost importance.
negative external shocks, caused aggregate demand to fall by 28% and GDP
to shrink by 14%. In an effort to cope with the crisis, a number of discrete
devaluations were applied, and later a crawling peg was reintroduced. It was
an answer to the evident need to generate a trade surplus in order to be able
to serve the external debt, especially in the absence of new voluntary capital
inflows and with depressed terms of trade.
At the same time, the uniform tariff was raised in stages up to 35% in
September 1984 (with annual averages of 24% in 1984 and 26% in 1985).
As the severe shortage of foreign exchange began to ease, the tariff was
gradually lowered again, first to 30% in March 1985, then to 20% in June,
to 15% in 1988, and to 11% by mid-1991. It is important to mention that,
in spite of these successive reductions, in 1991 the uniform tariff was still
above the 10% reached in 1979. It is a serious mistake to present the eighties
as a trade liberalization decade with respect to the seventies. Import liberali-
zation took place between 1974 and 1979.
Following the 1982 crisis, trade policy became more selective in several
respects. The government began to make somewhat active use of anti-dumping
measures to protect the economy from unfair trade practices.8 To this end, the
total tariff (the uniform rate plus compensatory surcharges) was raised to a
maximum of 35% – the level to which Chile had committed itself under the
terms of the GATT in 1979 – on imports that Chile could prove were being
dumped. In addition, in order to reduce the transmission of external insta-
bility into the domestic economy, price bands (supposedly consistent with
international price medium-term trends) were set for three main agricultural
products (wheat, sugar, and oilseeds); given the then depressed international
prices, this had significant encouraging effects on traditional agriculture.
Evidently, this also constituted a departure from the uniform tariff.
The explicit promotion of exports was made through a battery of instru-
ments, of which the simplified tariff drawback created in 1985 stands out.
The objective was to reimburse exporting firms a compensatory amount
for the tariffs paid on imported inputs, used to produce non-traditional
exportables. The instrument allowed exporters to obtain a maximum refund
of 10% of the fob value of exports. The requirements to benefit from the
incentive were: (i) the firm had to export products with less than 50% of
imported inputs; and (ii) the total exports of the country in the respective
item should not exceed a given annual threshold. The threshold value of
exports that determined the loss of the benefit was raised annually to take
into account inflation. In 1987, it introduced two tiers to allow for a gradual
loss of the subsidy. During 1988 the sales of inputs to exporters qualifying
for the RS were added to the list of beneficiaries.
8
The domestic legislation to attack trade distortions was established in 1981, in the
presence of pressures fed by the sharp appreciation of the exchange rate recorded after
1979. See Sáez (2005).
Careful econometric research on the impact of the simplified drawback

shows that after the introduction of this instrument, the number and value
of manufactured exports grew more rapidly (see Agosin et al., 2009). In addi-
tion, the automatic extinction of the subsidy was an important feature; in
fact, as the respective item developed, and its volume exceeded the respec-
tive thresholds, the percentage of the refund was reduced automatically and,
finally, the product did not qualify for reimbursement (precisely because of
its success), liberating resources for newer exports.
The drawback scheme sought (with efficiency, despite some evasion and
fraud) to promote the entry of new products and new external markets,
enhancing productive and exporter diversification (with their respective
externalities). The policy contributed to increase the number of exporting
firms, especially SMEs (Macario, 2000). In 1986, the amount spent by the
government in this instrument was US$25 million, a figure that rose to
US$66 million in 1989. The scheme reached a peak in 1998 with an expend-
iture of US$200 million. Though the spirit of the instrument is a tariff
refund, in practice, it corresponded to a net incentive to non-traditional
exports; we return to the issue in Chapter VI. Box V.1 summarizes the main
export promotion mechanisms in force by the late eighties.
On the other hand, one of the outstanding initiatives of export promo-
tion was developed by the Chile Foundation, a semi-public institution. The
Foundation’s first projects, in the seventies, were directed mainly at providing
technical assistance to certain productive sectors. Few of these early projects
got beyond the pilot stage. In view of this problem, the Foundation decided
it needed to gain more experience by launching business projects of its own.
The idea was to determine which activities might benefit from new technolo-
gies, and then to acquire and adapt these technologies. Once a technology
was assimilated, the Foundation would take charge of commercial production
and marketing through a subsidiary. When the subsidiary became profitable,
it would be sold, thereby completing the technology-transfer process.
The salmon industry was an example of a very successful initiative. In
1981, the Foundation decided to carry out a pilot project using a cage-based
technique of freshwater salmon farming. Commercial production began in
the period 1986–7, and output doubled in the following period. The project
began to deliver profits in 1988, and the technology-transfer cycle was com-
pleted that same year when the Foundation sold the project to a Japanese
fish and shellfish company. This project gave a definite boost to overall
salmon output in Chile, which became one of the top Chilean exports and
the main non-traditional one in the 1990s (ECLAC, 1998).
In the eighties, two additional changes (which involved intervention
in the relationship between domestic and external markets) took place.
On the one hand, the application of the debt–equity swap program, sum-
marized in Chapter IV, did not have the neutrality and automatism with
Box V.1 Incentives to exports by the late 1980s
• Drawback. Recovery of tariffs paid when importing inputs used to

manufacture exports. It allows firms to have access to a wide range of
imported inputs. However, this scheme involves complex paperwork
requirements, and large companies have a greater ability to comply
with the requirements. By contrast, SMEs tended to prefer the simpli-
fied drawback scheme described below.
• Simplified drawback. Simplified tax rebates or drawback on minor or

non-traditional exports, in the form of refunds of 10 or 5% of the fob
value of exports. This scheme was set up in 1985.
• Value added tax refund. Exports are exempted from the value added
tax and benefit from the refund of taxes paid on inputs incorporated
into exports. This mechanism is designed to avoid double taxation
on final products and follows the principle of taxation in the place of
destination instead of production.
• Export warehouses. Allow firms manufacturing exports to store

imported inputs that will be used to produce exports, without paying
tariffs or value added tax.
• Scheme for importing capital goods. Deferred payment (up to seven

years and three installments) of customs duty on imports of capital
goods. This provision is of general coverage and is not confined to
exporters. If the capital goods imported under this policy are used
to manufacture exports, then it is possible for the company to avoid
paying the tariffs (the firm must have exported at least 10% of its
total sales in the previous two years – for the first installment – which
increases to 60% for the next two payments).
• Financing of collateral for non-traditional exports. The Fondo de Garantía

para Exportaciones No-Tradicionales, set up in 1987, provides export
firms with up to 50% of what banks require as collateral when grant-
ing loans to finance non-traditional export investments (with a maxi-
mum of US$200,000 per year for each exporter).
• Export investment financing. CORFO has a long-term borrowing facility

(at subsidized rates) to finance export firms’ investments, with annual
sales under US$30 million. Also, it provides assistance with export
insurance for firms with yearly sales under US$10 million.
Source: Ffrench-Davis et al. (1992) and Macario (2000).

respect to the policies implemented in the seventies. On the contrary, the

program implied a strong subsidy to foreign investment in projects that
were approved case by case, with priority given to new non-mining exports
(Ffrench-Davis, 1990). Thus, the authorities prompted an idea of industrial
policy (Agosin, 1999). During the years that it was operating (1985–91),
about 60% of investments carried out under this program were directed to
manufactures, forestry, wood pulp, and paper.
On the other hand, the government strengthened the export-promoting
public institution Pro-Chile. Given the dominant strategy in the seventies, its
role had been minimized. By contrast, by the mid-eighties it was rearticulated,
strengthening its role of information provider on and for external markets.
In summary, after 1983, Chile carried out a more pragmatic “second trade
reform,” with a mix of restrictions, tariff increases, and explicit export pro-
motion. While it is true that the basic characteristics of the country’s trade
policy – in terms of the dismissal of non-tariff barriers and the adoption of a
uniform tariff – had not changed since 1979, it must be taken into account
that the tariff had once again become relatively high by 1984 and was
accompanied by anti-dumping measures and the price bands mentioned
above. In fact, the tariff level averaged 20% in 1984–9, which was double the
average rate for 1979–82 (see Table V.1).9 The greatest difference, however,
Table V.1 Average tariff and real exchange rate, 1973–89

Average tariffa (%) Real exchange rateb
(1986 100)
1973 94.0c 65.1c

1980 10.1 60.9
1981 10.1 51.8
June 1982 10.1 47.0d
1982 10.1 60.1
1983 17.9 72.1
1984 24.2 74.3
1985 25.8 91.0
1986 20.1 100.0
1987 20.0 104.3
1988 15.1 111.2
1989 15.1 108.6
1980–2 10.1 57.6
1984–9 20.1 98.2
Source: Based on Central Bank of Chile. aAnnual average, excluding exemp-

tions and preferential arrangements negotiated with LACs. bAnnual average.
The nominal exchange rate was deflated by the Chilean Consumer Price
Index (CPI; duly corrected in 1973–8) and inflated by an external price index.
c
December 1973. dMonthly average, until the day before devaluation.
9
Nonetheless, it was notably lower than the 94% nominal average tariff recorded by
1973, and lower than usual average rates in other Latin American countries in those times.
was that during the first liberalization drive the exchange rate had appreci-
ated steadily in the second half of the 1970s, and in the early 1980s. During
the 1980s, on the other hand, the reduction of the tariff from its maximum
level of 35% in 1984 to 15% by 1988 was accompanied by a sharp real depre-
ciation, under the pressure of the debt crisis. This sent out powerful positive
incentives to exporters, while at the same time encouraging the production
of import-competing goods.10 Unlike in the first liberalization effort, the
increase in exports was also coupled with a strong upturn in the output of
import substitutes, especially between 1984 and the late 1980s. The modifi-
cation of the exchange rate policy, stimulated strongly by the sharp shortage
of external financing, played a central role.
(c) The evolution of exchange-rate policy

Exchange rate policy has experienced substantial change over time. In
the mid-sixties, a crawling-peg policy was introduced; then, the exchange rate
was actively used to stimulate the production of exportables and to reduce the
current account deficit. In 1971–3 the main exchange rate was pegged as an
anti-inflationary anchor. In 1973–5 it recovered the role of resource allocat-
ing, with a return to a crawling-peg policy. Starting in 1976, the nominal rate
began to be used again to fight inflation. This was because inflation stubbornly
refused to slow down in reaction to a deep economic recession, which also
caused a surplus on current account in 1976. The real revaluation that usually
results from using the exchange rate as an anchor was intensified in 1979,
when the rate was fixed in current pesos, at a nominal parity of 39 pesos that
was maintained until the currency and financial crisis of 1982; a significant
real appreciation took place during these three years. After the crisis, there fol-
lowed a period of experimentation with successive policy changes during a few
months. In fact, in June 1982, after an 18% devaluation, a mini-devaluation
program was announced; it appeared to be a return to the crawling-peg policy
in force before the rate freeze in 1979. But, given a strong loss of credibility
and the evident insufficiency of devaluation and the schedule announced, the
outflow of international reserves continued disproportionably high.
Then, after the ministry that had directed the neo-liberal reforms in
the seventies was replaced, the full liberalization of the foreign exchange
market was decreed. The authorities believed that given the liberalization,
the markets would calm down. The opposite happened, because given free
access, in a few weeks the remaining international reserves evaporated, in
the presence of massive demand. Thus, the government reintroduced con-
trol of the exchange rate and of access to the market.
A crawling-peg exchange rate was again adopted from 1983 onward.
Basically, the central bank fixed a benchmark price for the dollar on the
10
See quantitative estimates in Moguillansky and Titelman (1993) and Agosin
(1999).
official market (called the “agreed” exchange rate), with a floating band that
was gradually enlarged from time to time. The “official” rate was devalued
daily, in line with the differential between domestic inflation and an esti-
mate of external inflation. On a number of occasions, discrete real devalu-
ations were added, helping to achieve the spectacular real depreciation
following the 1982 crisis (130% between the monthly minimum in 1982
and the average of 1988; see Table V.1).11
During 1988, the real exchange rate peaked. In that year, revaluations
together with tax and tariff reductions – financed by a huge jump in the
copper price – were implemented to reconcile a reduction in inflation with a
sharp recovery in aggregate demand and economic activity (see Chapter IV,
Table IV.2). This preceded a significant political event: in October 1988, for
the first time since 1973, Chileans were able to vote, though with extreme
restrictions. A plebiscite asking Chilean people whether Pinochet should
continue in power for one or eight more years took place. The first option
prevailed.
Economic recovery was completed in 1989, with actual GDP climbing
up to the production frontier or potential GDP (see Chapter I, Figure I.1).
The tax and tariff reductions and exchange rate appreciation in 1988 were
directly associated with the additional income generated by the sharp jump
in the copper price in 1987–9; the improvement in the terms of trade in
1988 with respect to 1986 was equivalent to 6% of GDP (measured at 1986
constant prices). A sizable increase in imports and in the current account
deficit in 1989 (if the current account is recalculated using the “normalized”
price of the Copper Stabilization Fund), together with new inflationary pres-
sures, led the central bank to reverse earlier reductions in interest rates.
By mid-1989, the floating band of the dollar was widened further to ±5%.
The central bank’s action was accompanied by a shift in the foreign currency
market expectations, which led the market to move quickly to the ceiling of
the band (the depreciated corner). Thus, with efficiency and no great trauma,
a significant depreciation was achieved without modifying the “official” rate.
For about a year, which included the return to a democratic regime, presiden-
tial elections (in December 1989), and the inauguration of President Aylwin (in
March 1990), the observed exchange rate remained at the ceiling of the band.
This occurred despite the fact that, in January 1990, a sharp tightening of the
adjustment process was made in order to control a sizable jump in inflation
(which had reached an annualized rate of 31% in the five preceding months).
11
Because various foreign exchange controls remained in force (except for a few
weeks in 1982), an illegal (but openly tolerated) market was operating in parallel. This
was legalized as the “informal” foreign exchange market only in April 1990, under the
provisions of the Central Bank Autonomy Act issued in the last days of the Pinochet
government.
3 Export performance: dynamism and diversification
During the first half of the military regime, the Chilean economy experi-
enced a notable expansion of exports, which represented a clear take off
from previous historical performance. In the seven-year period 1974–80, it
was particularly vigorous, with the volume of non-copper exports climbing
over 20% per year. As shown in Chapter II, total and manufacturing exports
stagnated in the early 1980s because of the significant appreciation of the
real exchange rate and the slowdown of the world economy. In fact, in 1981
the total quantum of exports fell 3.3%, led by a 7% drop in non-copper
items (see Sáez, 1991). Subsequently, encouraged by the sharp exchange
devaluations taking place since 1982, non-copper exports recovered some
speed. In all, in 1974–85 they expanded by 16% annually (Table V.2).
Table V.2 Growth of export volume, 1961–89 (annual averages, %)

1961–70 1971–3 1974–85 1986–9
Copper 3.9 –2.2 5.0 3.3

Non-copper 7.8 –8.5 16.0 13.1
Total goods 4.9 –4.5 9.4 8.8
Total goods and services 3.6 –4.2 10.7 11.1
Source: Sáez (1991) for copper, non-copper, and total exports of goods in 1961–85; Central Bank
of Chile balance of payments for copper and non-copper in 1986–9; national accounts (in 1977
pesos) for total exports of goods and services.
A second cycle of high export growth rates began well after the 1982
crisis. In the second half of the 1980s (1986–9), non-copper export quantum
grew by 13% a year; 9% for traditional goods and a notable 22% for non-
traditional items (see Chapter VI, Table VI.1). As said in previous sections,
the latter goods benefited from a highly depreciated exchange rate and the
simplified drawback.
Chilean exports became increasingly diversified but remained strongly
natural resource-intensive. The share of copper decreased from an average
of 72% in 1970–2, to 47% in 1982–9 (see Table V.3).12 Non-copper tradi-
tional exports – composed of fresh fruit, iron, nitrate, wood and natural
12
The share of copper in total exports is notably affected by the international price.
The figures mentioned here are corrected by a trend copper price series; the trend
was estimated on the basis of real prices and the Hodrick–Prescott filter. Additionally,
keep in mind that during the eighties, substantial decreases in production costs were
registered in the main copper companies around the world. Productivity improve-
ment in mining treatment and management overcame the deterioration in the laws
of minerals. For this reason, in the following years, the expected “normal price”
for copper tended to be much lower than in the preceding decades (see Bande and
Ffrench-Davis, 1989). Cost decreases continued in the nineties.
Table V.3 Composition of goods exports, 1970–89 (%)

Copper Traditional Non-traditional
non-copper
1970–3 74.5 17.2 8.3

1974–81 53.8 30.4 15.8
1982–9 47.0 35.6 17.5
Source: Based on BADECEL. The figures are in current US dollars. The value of copper exports
was corrected by estimates of a trend of the nominal copper price, derived from estimates of
the trend real price with a Hodrick–Prescott filter (see Figure IX.2).
resource-based manufacturing, such as fishmeal, wood pulp and paper –

grew vigorously in the 1970s, jumping from 17% of total exports of goods
in 1970–3 to 30% in 1974–81. In the early 1980s their real value declined as
a result of the anti-export bias of the exchange rate appreciation; however,
from the mid-eighties their real value resumed a dynamic growth increasing
its participation to 36% in 1982–9.
Non-traditional exports – which include several hundreds of manufactured
items with higher value-added, and new resource-based products – initially
expanded during the first trade reform. Their share increased from 8% in
1970–3 to 18% in 1980–1; their participation weakened in the next few
years. As mentioned, from 1985, again this group exhibited a strong growth
and, as a consequence, its share reached 22% of total exports in 1989. On
the other hand, the number of products sold abroad increased from 200 in
1970 to 2,300 in 1990. The most outstanding new item was salmon.
Exports became more diversified not only in terms of items, but also with
respect to markets of destination. In fact, the number of countries covered
rose from 31 in 1970 to 129 in 1990.
Traditionally, the European Union (EU) was the most important destina-
tion for Chilean exports. In 1970–3, for instance, 54% of total exports went
to Europe, while the rest was principally distributed between Japan (17%),
Latin America (13%) and the United States (10%) (see Table V.4, row d).
After the first trade liberalization, the geographic distribution experienced
some changes, mainly due to the growing importance of Latin and North
American markets at the expense of the participation of the EU. The debt
crisis, and its strong impact on Latin America, however, implied a setback
in that trend, with a return in the share of the region to 12% in 1983, after
reaching a peak of 25% in 1980. In all, in 1982–9 Latin America absorbed
16% of total Chilean exports, the USA took 21%, while the EU decreased to
37% after having averaged 42% in 1974–81. Meanwhile, the Asian countries
(other than Japan, which shows decreasing importance from a high level)
and a group of numerous other countries had emerged as relevant partners,
increasing their participation from 6% in 1970–3 to 16% in 1982–9; this
trend became even stronger in the 1990s and 2000s (see Chapter VI).
Table V.4 Geographic distribution of Chilean exports, according to degree of

elaboration, 1970–89
Classification Composition Geographic distribution (%)

(% of total)
Latin EU USA Asia Japan Other
America 11
1970–3
(a) Non-processed 16.9 10.0 25.7 10.8 1.1 48.8 3.7
natural resources
(b) Processed natural 80.7 12.4 61.5 10.0 4.5 9.8 1.8
resources
(c) Manufactures 2.3 52.8 12.1 9.6 3.7 7.7 14.0
(d) Total 100.0 12.8 54.2 10.1 3.9 16.5 2.4
1974–81
(a) Non-processed 19.9 16.5 26.2 12.4 4.3 32.1 8.5
natural resources
resources
(c) Manufactures 5.2 59.3 11.8 11.8 4.0 1.8 11.2
(d) Total 100.0 24.0 41.5 11.4 4.4 12.2 6.5
1982–9
(a) Non-processed 29.1 10.9 26.8 24.4 6.8 23.4 7.7
natural resources
resources
(c) Manufactures 6.3 43.5 18.6 23.8 4.6 1.7 7.9
(d) Total 100.0 15.5 36.7 20.6 7.5 11.5 8.2
Source: Based on COMTRADE (ECLAC) figures, processed by Jaime Contador. Figures in current
US dollars, classified according to ECLAC (1992). Based on SITC rev. 1. The composition does not
always add to 100% because of unclassified transactions summing around 1% of total exports.
The composition of Chilean exports varies hugely according to their

geographical destination. Table V.4 shows that unprocessed and processed
natural resources represent an overwhelming share of total exports of goods;
a trend that is stronger in those directed to industrialized countries. On the
other hand, more than one-half of manufacturing exports were directed to
Latin American countries; in 1970–89, 52% of Chilean manufactures were
sold to countries in the region, much more than the 16% sold to the USA,
15% to the EU, and 3% to Japan and Asia, respectively; on the negative side,
sales to Latin American partners have been extremely sensitive to recessive
macroeconomic situations in regional markets, as was the case during the
debt crisis. The outstanding importance of the Latin American countries as
destination of manufactured products is crucial for the expansion of the
volume and quality of exports intensive in value-added. This issue is re-
examined in Chapter VI.
4 Dynamic exports and limited economic growth
Exports have frequently been called the “engine” of Chilean economic

growth. As has been shown, they have exhibited a vigorous expansion, grow-
ing systematically faster than GDP during the past third of a century. Actually,
they were one of the fastest rising components of GDP. Consequently, the
gross value of exports of goods and services raised its share of GDP from 10%
in the 1960s to 25% in 1989 (measured at constant prices of 2003). Evidently,
this has been a potential channel of transmission of externalities resulting
from exposure to foreign markets. Nonetheless, the dynamic growth of the
volume of exports of goods and services (10.7% annually in 1974–89) took
place in parallel with a GDP rise that scarcely averaged 2.9%.
As shown in Chapter IV, the external debt crisis and the subsequent auto-
matic adjustment carried out meant a generalized fall in domestic output in
1982–3 (recall that there was a drop of 14% in GDP), when demand-reducing
policies caused a high underutilization of productive capacity; naturally,
given a sizable devaluation, tradable production tended to increase subse-
quently. By contrast, underutilization was concentrated in non-exportables,
whose capital formation was discouraged as a consequence.13
The export-promoting measures of the 1980s worked as supply-switching
policies, reallocating available resources, against the background of a very
low overall investment ratio (and, as a consequence, low creation of new
capacity). Depressed domestic demand and the increased export profitabil-
ity, as a result of exchange rate depreciation, provided the conditions that
encouraged investment and output in exportables. Consequently, it was
not precisely a case of export-led growth, but rather of an export drive led
by the binding external constraint and a recessionary domestic adjustment.
Evidently, strong export expansion, which averaged 7.8% per annum in
1982–9, was a factor contributing to the recovery of economic activity
throughout this subperiod; however, it could not compensate for the near
stagnation in non-export GDP, which grew by only 1.7%, merely as much as
the increase in population. This share of GDP that was sold in the domestic
markets was generally working under macroeconomic disequilibria, in the
sense that demand was frequently below productive capacity.
Thus, exports were unable to transmit their dynamism to total GDP,
which expanded modestly by 2.9% per annum (see Chapter VI, Table VI.6).
Naturally, most probably, ceteris paribus, if exports had grown less, GDP would
have tended to be even smaller. But the straight message is that vigorous
export growth is not sufficient to generate a dynamic economic expansion.
It must be recalled that, frequently, there was a huge output gap between
potential and actual GDP; generally, that recessive gap was placed in the
13
In Ffrench-Davis (2006 Chapter II) I have documented the strong negative impact
of the gap between potential GDP and actual GDP on the investment ratio.
production of non-exports (an exception being, as said, 1981). This implies

that a strong depressing effect of the gap was countervailing the pulling
effect of rising exports. But, beyond this macroeconomic consideration,
there is the structural feature of an economy with incomplete capital mar-
kets, excessively intensive in short term channels, outlier high interest rates,
exchange rate instability, low capital formation and weak innovation forces
for most of the domestic economy. What an unfriendly environment for the
producers of wealth!
I fully share the approach that exports should lead overall economic
growth. However, notwithstanding the broad consensus around this view,
during the seventies and eighties, we observe a significant export drive
that was not accompanied by a respectable average economic growth. This
reveals that a dynamic expansion of exports is a necessary, but insufficient,
condition for development. I return to this relevant issue in Chapter VI.
Fourth Part
Development Challenges for
Democracy
VI
Export Dynamism and Economic
Growth since the 1990s*
The nineties marked a new stage for Chilean trade policy. Export dynamism
was encouraged by a more comprehensive policy that combined both the
principles of an open economy and free trade and integration agreements
with several countries or groups of them – at the beginning, particularly
in Latin America, and subsequently with diverse developed countries and
other emerging regions. This strategy was implemented in a new economic
environment characterized by high domestic investment and overall grow-
ing productivity, especially until 1998.
In all, in the nineteen-year period (1990–2008), a 7.9% average expansion
of exports was associated with a relatively vigorous GDP growth averaging
5.3%. These figures are in sharp contrast with the 10.7 and 2.9%, respec-
tively, recorded in 1974–89 in Chile; interestingly, the two latter figures
are not too different from the growth of exports (6.9%) and GDP (3.2%)
exhibited by Latin America in 1990–2008. The contrast is focused on the
advantage taken by Chile in 1990–8, and is associated with differences in
the macroeconomic environment, particularly the quality of exchange rate
and capital account policies.
In section 1 the trade policy approach since 1990 is examined, empha-
sizing the role of the series of free trade agreements and the evolution
of exchange rate policy. Section 2 describes the general performance of
Chilean exports, in terms of both the basket of products and markets of
destination and their diversification. The interaction between exports
and foreign investment is examined in section 3. The changing relation
recorded between the evolution of exports and of economic growth is
explored in section 4. Finally, some concluding remarks are presented in
section 5.
* I appreciate the valuable comments of Sebastián Sáez, Rodrigo Heresi, and Heriberto
Tapia.
149
150 Export Dynamism and Economic Growth
1 Trade policies since 1990
The economic team that took office in 1990 maintained the basic princi-
ples of the 1980s trade policy. Principally, the economy remained open to
trade, and there was continuity in terms of maintaining a uniform nominal
import tariff rate towards the rest of the world. The rate of 11% established
in 1991was held up to 1999, when a gradual unilateral reduction of 5 points
was initiated, reaching 6% from 2003 to date. However, at the same time,
numerous free trade agreements were launched, involving import liberali-
zations accompanied by increasing market access and tariff preferences for
exports to trade partners. Undoubtedly, the net balance for Chile of this
innovation differs from a unilateral liberalization: liberalization of imports
from partner countries came together with liberalization of the access of
Chilean exports to the markets of those partners.
On the other hand, macroeconomic policies were substantially modi-
fied, in a significant reform of the reforms in the early 1990s (see Chapters
I and VIII). As a consequence of the macroeconomic reform, especially in
1990–5, (i) a more stable real exchange rate stimulated exports with higher
value-added and (ii) an environment was generated in which the overall
economy was placed closer to the productive frontier. As a consequence, the
rest of GDP – what we call non-exported GDP – recorded a notably higher
growth rate than in previous periods. Then, an outstanding outcome was
that the rather vigorous growth of exports was achieved pari passu with a
vigorous growth of GDP until 1998.
During the 1990s there were significant changes in the international
environment faced by Chile. The entry into force of the World Trade
Organization (WTO, which has included China since 2001) in 1995 implied
changes in diverse trade policy instruments. Chile adjusted to the new rules
of the multilateral system; especially relevant was the dismantling of the
simplified drawback that had been benefiting, effectively, non-traditional
exports (see Agosin et al., 2009). At the same time, an explosion of bilat-
eral and regional trade agreements has taken place, which has fragmented
international trade. Particularly, in 1994, the North American Free Trade
Agreement (NAFTA, including Canada, Mexico and the USA) modified rela-
tions between the United States and Latin America, stimulating a wave of
agreements of this type in the region.
These bilateral or regional agreements have allowed Chile to rely on a
(imperfect but effective) stabilization of the application of the rules that
govern trade (see DIRECON, 2009). Many non-tariff restrictions have been
deleted, while limiting the capacity of partner countries to adopt restrictions
against Chilean exports in periods of crisis; they have provided mecha-
nisms to solve controversies in case a country acts in opposition to the
assumed commitments or the commercial interests of Chile. At the same
time, they have implied continuing the opening of international markets
Export Dynamism and Economic Growth 151
independently of the multilateral negotiations of the WTO; a costly coun-

terpart has been the implementation, forced by the Uruguay Round and
some agreements, of a reform agenda that limits the space for productive
development and counter-cyclical macroeconomic policies in developing
economies (see Rodrik, 2001b).
(a) Towards a “reciprocated” trade policy

An outstanding feature of Chilean trade policy in the 1990s was the search
for preferential agreements to expand access to export markets. The new
political conditions allowed a turning point in the Chilean approach, mov-
ing from an across-the-board indiscriminate unilateral opening, towards a
strategy that also included preferential free trade agreements (FTAs) subject
to reciprocity. Actually, today, the bulk of Chilean exports receive the pref-
erential treatment of a free trade agreement (DIRECON, 2009).
Since the Chilean economy already exhibited a wide trade openness, the
benefits from further unilateral liberalization were estimated to be small in
a world where trade areas and economic blocks were increasingly important.
Advancing toward economic integration with Latin American countries –
already returned to democratic regimes – was seen as a way to be an actor
in globalization, a process named open regionalism. In particular, it was con-
sidered that exports to the region would tend to be more intensive in non-
traditional products, acquiring new comparative advantages with a higher
share of value-added.
As a result of the active trade policy, economic complementarity agree-
ments were signed with countries of the Andean Community; that is,
Bolivia (1993), Venezuela (1993), Colombia (1994), Ecuador (1995), and
Peru (1998). An agreement was signed in 1996 with MERCOSUR (Mercado
Común del Sur) – the main Latin American market, which then included
Argentina, Brazil, Paraguay, and Uruguay. A broad agreement was reached
with Mexico in 1991, which was deepened in 1999. Likewise, in February
2002, there entered into force the FTA with Central America (including
Costa Rica, El Salvador, Honduras, Guatemala, and Nicaragua).
Intensification of the linkages with other regions and developed econo-
mies was another ingredient of foreign policy. In 1994, Chile entered the
Asia-Pacific Economic Cooperation Forum (APEC), enhancing its trade rela-
tions with this dynamic region. In 1997, a comprehensive free trade agree-
ment with Canada was reached; it must be stressed that, among its positive
features, it included a special clause to allow the use of capital controls by
Chile. As a consequence of the contacts developed in the nineties, a compre-
hensive agreement with the European Union came into force in 2003 (with
a much broader coverage than FTAs), and a FTA was made with the United
States in 2004. Since then, a series of agreements with many other countries
or groupings of them have taken place, including China, India, Japan, and
Korea (see DIRECON, 2009).
In all, restrictions and tariffs faced abroad by Chilean exporters have been
sharply reduced. The average nominal tariff paid was 0.4% in 2008, a drop
of 88% with respect to the “most favored nation tariff.” In the case of the
two larger world markets – the European Union and the United States –
the agreements included the immediate elimination of tariff barriers for
a large number of products. In fact, the percentage of exports facing zero
duty rose from 74% for the European Union, and 70% for the United States,
before the agreements, to about 85% at the beginning of each agreement,
and to 97% in 2008.1 For the remainder products a gradual reduction of
tariffs is contemplated and, in several cases of remaining duties, there are
quotas for sales free of duty. In the case of the European Union the maxi-
mum term for full liberalization is ten years, while in the case of the United
States it is twelve years.
These agreements are relevant for Chile since they are with the major
markets of the planet, which are relatively closed to the emergent econo-
mies due to their protectionist practices. Both the European Union and the
United States use tariff escalation – tariffs that increase as the value-added
of the items rises – thus discouraging the expansion of upgraded exports to
these markets.2
The natural counterparts of trade preferences benefiting producers of
exportables are the preferences granted to imports from partner countries.
The intensified liberalization via trade agreements has meant that the aver-
age nominal tariff actually paid by importers is notoriously lower than
the uniform 6% in force. By 2005, the tariff actually paid averaged 2.1%,
while in 2009 it was less than 1%. Naturally, the reduction of actual tariffs,
together with a real exchange rate appreciation in 2005–9, has discouraged
the production of import substitutes.3 But, at the same time, it has reduced
the costs of imported inputs and capital goods for the production of export-
ables. With the trade agreements new opportunities for Chilean exports are
opened, especially for non-traditional items with higher value-added. But
costs also arise (such as the loss of tariff revenue and the, at least transi-
tory, unemployment resulting from the crowding-out of local producers of
importables).
One of the more troubled points of the negotiations with the United
States refers to the room for Chile to apply capital controls. Finally, Chile
1
The percentages are basically the same for both agreements, according to estimates
based on the exporting basket of 1998–2000 in the case of the European Union, and
the basket of 2001 in the case of the United States. See Leiva (2004) on the negotia-
tions of the treaty signed with the European Union.
2
Average MFN tariffs of the USA are very low but escalated. For instance, in 2002,
textile imports, shoes, and metal products imports faced tariffs averaging 33, 38, and
25%, respectively (DIRECON, 2009).
3
Official SOFOFA reports repeatedly mention the exchange rate appreciation as a
cause of domestic production being crowded -out by imports.
preserved the right to impose controls on short-term financial inflows. The

agreement provides that Chile will incur no liability and will not be subject
to claims for damages arising from the imposition of restrictive measures
that were incurred within one year from the date on which the restrictions
were imposed. Consequently, Chile kept policy space, but in a reduced form.
In fact, it curtailed the degrees of freedom for Chile to handle, freely and in
a transparent way, its financial linkages to the rest of the world and, conse-
quently, its domestic real macroeconomic stability.
(b) Domestic incentives

At the domestic level, Chilean trade policies in the 1990s faced a quite
different context to that of the 1980s. GATT rounds and the subsequent
rules of the WTO restricted the use of subsidies to exports; this is par-
ticularly relevant for developing countries in the process of acquiring new
comparative advantages. Therefore, the use of both simplified tax rebates
(reintegro simplificado) and the deferred payments of customs duties on
imports of capital goods – instruments of great effectiveness and efficiency
in the 1980s and 1990s – were limited and scheduled to gradually disap-
pear, because they were declared non-compatible by the Uruguay Round
Agreements. Consequently, by 2004, fiscal expenditure under the simplified
rebate was already negligible (fading away to only US$2 million, while in
1998 it had reached a peak of US$200 million). Although the simplified
drawback is supposed to be a refund of tariffs paid by exporters on their
imported inputs, the instrument was in fact a subsidy on non-traditional
exports; the reimbursement received by exporting firms usually exceeded
the tariffs paid on their imported inputs.
The subsidy implied by the simplified drawback had several interesting
features, which made it an efficient tool for upgrading the quality of exports,
as tested by Agosin et al. (2009). In lieu of having to apply for a drawback
of duties paid on imported inputs, small exports started receiving a 10%
subsidy on the value of their exports. When the entire tariff line (defined at
the eight-digit SITC level) surpassed a certain threshold the subsidy disap-
peared. From 1991, instead of an abrupt elimination of the subsidy, three
rates began to be used: 10% for annual exports below US$10 million, 5% for
exports between US$10 and 15 million, and 3% for exports between US$15
and 18 million. As a follow-up to the Uruguay Round Code on Subsidies and
Countervailing Duties, in the mid-1990s, even though no complaints had
been received from trading partners, the government declared its existence
as a subsidy to the WTO and committed to dismantling it, which it finally
did in 2003.
In modern terms, the mechanism worked as an effective way of subsidiz-
ing self-discovery (Hausmann and Rodrik, 2003) and letting the market
select the sectors to benefit from the incentive. The fact that the continua-
tion of the subsidy depended not on exports from a specific firm but on all
firms exporting the same product discouraged rent-seeking and ensured that
it was in effect a subsidy to infant exports (Agosin et al., 2007). Additionally,
in contrast to the standard drawback, which encourages the use of imported
components, it encouraged domestic value-added; it was restricted to
exporters with less than 50% of imports, which paid taxes on their imports
and received a fixed percentage subsidy on the revenue returned to Chile.
The research available shows that the incentive was effective in diversifying
exports, and discovering new comparative advantages, particularly in the
early 1990s.4
During the 1990s the authorities developed other efforts to promote
exports by seeking to correct more directly some market failures. The
main programs, managed by the Chilean export promoting institution,
PROCHILE, provide information to exporters and support activities fos-
tering Chilean products in new markets, launching campaigns to create
a positive image of the country. Corporate groups are stimulated to form
associations in order to promote their products and to perform activities
that allow a better knowledge of markets.5 These policies have been success-
ful in terms of easing market access to participant firms (Álvarez and Crespi,
2000). The growth of exports quantum continued to be vigorous during
the 1990s – around 10% per year, similar to that of the Pinochet regime.
Nevertheless, it lost dynamism from the late 1990s, as documented in
section 2.
(c) Exchange rate policy and the new capital surges

During most of the 1990s, a crawling-band exchange rate regime was main-
tained, but with much intensified activism. In line with the experience of
other emerging economies, Chilean authorities had to deal with a sharp
private capital surge that heavily influenced the exchange rate in the oppo-
site direction to the 1980s’ depreciating pressures. Chile’s real exchange
rate tended to appreciate in response to a large supply of capital inflows to
LACs during the first half of the 1990s (see Chapter VIII). However, due to
the counter-cyclical policies implemented by Chile, the actual appreciation
was notably more moderate than in other LACs; measured in a comparable
way by ECLAC methodology, by late 1994, just before the tequila crises,
the Chilean RER had appreciated by only 4.1% with respect to the average
of 1987–90, while in Latin America (weighted average of sixteen countries)
4
As shown by Agosin et al. (2009), “the exports from sectors that did receive the
incentive grew more rapidly than sectors that did not. The effect is strongest during
the 1991–96 period, when, according to the different estimates used, exports from
supported sectors grew by 89 to 60 percent more than other exports.”
5
The financing of the activities abroad and the costs of administration of these
associations or Committees of Exporters are subsidized in a diminishing scale for a
maximum period of six years.
it had appreciated by 23%.6 Additionally, since Chile was coming out of

the severe debt crisis of the 1980s and its productivity was increasing more
rapidly than that of its trade partners, there was space for some relative
appreciation in a movement towards equilibrium, rather than the destabi-
lizing RER misalignments brought about by excessive appreciation in most
of the region.
As witnessed by the moderate current account deficit (2.3% of GDP in
1990–5), the appreciation recorded in Chile was actually equilibrating, con-
sistent with net increases in productivity and the improvement in the terms
of external debt. The behavior of the RER responded to a very active public
policy, coordinated between the central bank and the Ministry of Finance.
In fact, significant modifications were introduced to exchange rate policy to
resist the appreciating trend; among them, the official exchange rate became
pegged to a basket of currencies, rather than just the US dollar, in order to
deter speculative capital inflows predominantly in dollars, and the central
bank actively intervened in the foreign exchange market, including intra-
marginal intervention (within the band). Furthermore, taxes and unremuner-
ated reserve requirements were applied to foreign loans and deposits to soften
the external supply in the face of capital surges, in such a way that the share
of short-term capital inflows was actually reduced. An explicit goal of avoid-
ing an excessive RER appreciation and instability was protecting the export
model and its contribution to economic development (Ffrench-Davis et al.,
1995; Zahler, 1998).
However, prudential macroeconomic policies lost their strength in the
second half of the decade, when a larger capital surge seeking the Chilean
markets was not faced with a timely and proportional reaction from the
central bank. Therefore, the overflood by inflows led to a real appreciation
of 15% between 1995 and October 1997, whereas the current account deficit
jumped to 4.8% of GDP in 1996–7 (see Chapter VIII, Table VIII.2).
The Asian crisis meant a sizable capital outflow and subsequent depreciat-
ing pressures. The central bank resisted them by narrowing the band and
raising interest rates sharply up to September 1999, when the exchange rate
was left to float freely. This change of regime allowed the correction of the
over-appreciated real exchange rate, but it did imply introducing higher
6
See Ffrench-Davis (2006, Table VII.2). There are diverse estimates of the magnitude
of real appreciation. The central bank of Chile reports 15%; it must be recalled that
the bank uses for external inflation indices of wholesale prices, whereas ECLAC uses
CPI. See figures for RER by ECLAC and central bank in Table VI.2. In all, the diverse
estimates give a significantly smaller real effective appreciation in Chile than in other
countries like Argentina, Brazil, and Mexico. Additionally, in the nineties, the reduc-
tion of import tariffs in the region involved a drop of imports costs that was several
points higher than that of Chile. Consequently, depreciations instead of appreciations
were required in Latin America in relation to Chile.
volatility,7 harmful for the objective of maintaining real macroeconomic

stability and for enhancing the contribution of exports to a vigorous and
sustained economic growth. In the subsequent years, exchange rate insta-
bility prevailed, with sharp devaluations through 1999–2003 and sharp
appreciations in the course of 2004–7, followed by sharp ups and downs
during 2008 and 2009. As said, that is a pervasive feature for upgrading the
quality of exports.
2 Export performance
There are diverse export indicators that are relevant for understanding their
role on development. In this section I examine the evolution of export
value, volumes, and prices, as well as the evolution of the export composi-
tion (diversification) or “quality.”
(a) Quantum
Exports quantum of goods continued to grow strongly through the 1990s.
This expansion was mainly explained by non-traditional exports, with an
annual increase that averaged 14.1% in 1990–8, while copper and non-
copper traditional exports grew by 9.4 and 5.1%, respectively.
Export dynamism remained strong despite the real exchange rate appre-
ciation recorded during the second half of the decade (see Table VI.2, and
Chapter VIII, Figure VIII.1). Four factors contributed to this positive outcome
of exports. First, world trade was more dynamic during the 1990s. After the
Table VI.1 Growth of export volume, 1986–2009 (annual averages, %)

1986–9 1990–8 1999–2003 2004–7 1990–2007 2008–9
Copper 3.3 9.4 4.5 4.8 7.0 –1.7

Non-copper 13.1 9.8 7.0 8.2 8.7 –3.1
Traditional 9.0 5.5 5.7 6.7 5.8 –1.4
Non-traditional 21.7 14.1 7.9 9.3 11.3 –4.3
Total exports 8.8 9.6 6.1 7.1 8.1 –2.6
Total services 16.2 10.6 6.3 7.3 8.7 1.1
Source: Balance of Payments of the central bank for exports of goods. Traditional non-copper
exports, due to availability of information, include fresh fruit, fishmeal, cellulose, paper, iron,
nitrate, silver, gold, molybdenum, raw and sawn wood, and methanol. Exports of services are
from National Accounts of the central bank. Figures for 2009 are preliminary.
7
Caballero and Corbo (1989) and Moguillansky and Titelman (1993) show empirically
that RER volatility has strong negative effects on both the level and quality of exports.
Rodrik (2007a, b) and Williamson (2000) analyze the key role of the exchange rate
on exports and productive development. Dodd and Griffith-Jones (2006) examine the
performance of the Chilean derivatives markets and implications for RER stability.
Table VI.2 Average tariff, collected tariff, and real exchange rate, 1984–2009
Average tariff a Collected tariff b Real exchange ratec (1986 ⴝ 100)
(%) (%)
Central bank ECLAC
1984–9 20.1 20.1 98.2 96.4

1990–5 12.0 10.0 99.5 100.6
1996–2000 10.8 8.0 80.8 85.6
2000–4 6.8 3.6 99.0 100.3
2005–9 6.0 1.4 94.5 93.2
Source: DIPRES, Central Bank and ECLAC; data for 2009 are preliminary.
a
Simple averages of nominal tariffs, excluding exemptions and preferential arrangements in trade
agreements. bThe collected tariff is the ratio between the actual value of import taxes and total
imports of goods. cThe real exchange rate is the observed nominal exchange rate multiplied by
the ratio between external and domestic inflation indexes. The central bank uses WPI of trade
partners as external inflation, weighted by their relative importance in imports and exports –
excluding oil and copper, respectively. ECLAC uses CPI for external inflation and does not exclude
any item of the basket trade. All figures are annual averages.
stagnation of world trade in the early 1980s, from 1984 it recovered speed,
and during the 1990s the volume of world trade expanded 5.7% annually.
Second, Chile gained access to new markets under preferential access thanks
to the several trade agreements. Third, the record domestic investment ratio
of the period allowed a great increase in productivity and external competi-
tiveness. Fourth, it must be kept in mind, on the one hand, that a major part
of the moderate exchange rate appreciation of the first half of the 1990s was
an equilibrating movement after the currency drought of the 1980s, and, on
the other, that during those years there was a deliberate policy to protect the
export development model through prudential management of the capital
account with positive lagged effects on the second half of the 1990s.
In all, the 9.9% growth of the volume of exports of goods and services in
1990–8 was vigorous in comparison with the eighties (7.8% in 1982–9) and,
in particular, with respect to the world trade growth average in the decade
(5.7%). The trade boom was interrupted by the Asian crisis that hit the Latin
American economies in 1998–9. It implied a sharp negative terms of trade
shock, which strongly reduced the export value, especially in traditional
items. The value of non-traditional exports, on the other hand, exhibited
strength during the crisis; commodities were the main victims in this period.
The vigorous growth recorded by that group of exports, from their take-off
in 1984, explains why their set matched the value of copper exports in the
late 1990s.
However, after the exceptional growth of the quantum of non-traditional
exports from the 1970s, their dynamism has experienced a significant decel-
eration: annual growth averaged 21.7% in 1986–9, 14.1% in 1990–8, and
8.6% in 1999–2007 (see Table VI.1). The real exchange rate appreciation
during the second half of the 1990s and the persistent instability from
there on appear as the main cause of the declining competitivity of non-
traditional tradable sectors (Díaz and Ramos, 1998). Econometric estimates
(Moguillansky and Titelman, 1993) of the price elasticity of the Chilean
export supply with respect to the real exchange rate across industries sup-
port this hypothesis. The research indicates that movements in the RER
have differentiated effects, depending on the type of good exported, and
that a persistent real depreciation (instead of instability and appreciation)
tends to have a positive (negative) impact on the volume and diversification
towards higher value-added. Thus, it is likely that a depreciated and stable
RER constituted a strong incentive for investment in exporting sectors, for
basket diversification and the fortification of its pulling-up effect on the rest
of the economy; and vice versa for periods of overvaluation, due to a mix of
financial volatility and lack of counter-cyclical exchange rate policies.
Slowing of the diversification of exports towards non-traditional items
explains most of the drop in the overall rate of expansion of exports. In
fact, since the late 1990s, the volume of exports has shown a deceleration;
initially, directly, due to a drop in Asian demand, and an overall depressed
world demand with a terms of trade deterioration. These factors, together
with exchange rate volatility, seem to have weighed more than the significant
real depreciation recorded between 2000 and 2003; anyway, soon after there
came another stage, with deep appreciation from 2004 (see Table VI.2).8
The pervasive feature of RER instability prevailed, in that when optimism
reigns during capital surges or booms, the exchange rate appreciates, thus
discouraging the production of tradables, while in recessive periods the RER
is sharply depreciated but the macroeconomic environment and liquidity
constraints are discouraging for most productive investment.
Already for a decade, the quantum of exports has been moving on a
plateau that, even though it is “respectable,” is sharply below that of the
previous decade. Additionally, progress in the upgrading of exports has
weakened. Poor foreign currency policy bears the major responsibility, but
there also are structural factors behind the decaying export dynamism.
Despite the weakening recorded since the late 1990s, in the nineteen
years from 1990 to 2008 exports of goods and services expanded by 7.9%
annually. That is a significant record; however, its contribution to overall
productive development and wellbeing is not that significant, an issue that
needs to be addressed by substantial corrections in macroeconomic and
productive development policies.
8
Although during 2002 the Chilean RER was relatively depreciated with respect to
the set of all trade partners (Table VI.2), with respect to Latin America it experienced
an appreciation of 28% in that year, determined principally by the Argentine maxi-
devaluation. Chilean exports to MERCOSUR fell in 2002 by nearly equivalent to 1%
of GDP, beyond the already depressed level of 2001.
(b) Export prices and terms of trade

The real evolution of trade prices has three types of significant effect in the
domestic economy: (i) on private and social returns of each sector; (ii) on
disposable national income that a given volume of exports contributes; and
(iii) on the domestic macroeconomic environment, through exchange rate,
monetary, and fiscal effects.
Chilean exports are highly concentrated in traditional natural resource-
based products. A distinguishing characteristic of this kind of products is
the great volatility of the prices, not only of copper, but also of other items
such as cellulose, fishmeal, and molybdenum (see Box VI.1). Naturally, not
all prices always move synchronously; thus, although they have great indi-
vidual fluctuations, the average price of the set of exports is less unstable.
Nonetheless, in practice, there is a significant synchronization or contagion
of fluctuations, even with imports (for example, copper and oil prices in
recent years). This is an effect of the intensive globalization of volatility.
For developing economies, many export prices constitute a given param-
eter (this is the common assumption for “small open economies”); then,
price fluctuations are a consequence not of domestic policies but of inter-
national market forces. Notwithstanding this, several developing countries
have become large producers of some commodities (thus influencing inter-
national prices), as is the case with Chile for copper, molybdenum, and
salmon. As documented by Moguillansky (1999), Sachs et al. (1999), and
Agosin (1999), the impressive growth of copper output exhibited in 1995–9
(rising 15.1% annually, and more than doubling in the period) was not sus-
tainable, due to both investment constraints and the fact that markets were
becoming limited for Chile. In that sense, too high a growth in the produc-
tion of an item for which world demand grows slowly (in the case of copper,
then by about 3% per year) can negatively affect the average price received,
damaging the profitability of the activity and the economic rent captured by
Chile. Indeed, the copper price was remarkably depressed in the following
years; for instance, by late 2002 a pound of copper was selling at less than
US$0.60. After, as usual, a new cycle followed with significantly high prices of
commodities in 2005–8, averaging US$3.23 in 2007. It was followed by a
sharp (though very short, on this occasion) new recessive cycle that started
in the course of 2008 (see Chapter IX, Figure IX.3). For that reason, it is so rel-
evant to diversify the exporting basket towards greater value-added, as well as
keeping a stabilization fund of copper income, as the government has done.
(c) Diversification
Table VI.3 summarizes the overall diversification achieved by Chilean
exports. Between 1990 and 2008, the number of markets of destination
increased from 129 to 190 countries (nearly all the countries in the world).
In turn, the number of exporting firms increased from 4,100 to 8,240 in the
Box VI.1 Molybdenum in the national economy

Chilean mining exports reached a total value of US$43.1 billion in 2007, of which
US$3.8 billion corresponded to molybdenum sales abroad (US$37.6 billion were of
copper). Molybdenum, for CODELCO a significant by-product in the production
of copper, constituted the second largest export product of Chile, even exceeding
fruit, forestry, cellulose, and salmon.
A jump in the share of molybdenum is related to the particular dynamism
shown by its price since 2004. In fact, after recording levels in the range of
US$2–5 per pound (an average of US$3.6 per pound in 1985–2003), the nominal
price observed in 2005 was US$32. The price fluctuated around this high plateau
(US$30–35) up to well into 2008, when it collapsed to US$10, in November, some-
what after the sharp drop of the copper price. There are several reasons for the
high plateau. First, on the supply side, numerous molybdenum mines in China
were closed. Second, there was greater demand due to the recovery of the world-
wide steel industry, which intensively uses molybdenum. Third, there was a broad
speculative pull in commodities.
The increasing share of molybdenum in total exports and in fiscal revenue
became a significant issue from the point of view of public finance. There was
broad consensus that current prices were not sustainable in the long run, as
increasing incentives to invest in new projects of primary mines and to increase
the capacity in copper mines that produce molybdenum as by-product will arise.
Consequently, since 2006, a cyclical correction to fiscal revenue from molybde-
num sales by CODELCO has been incorporated in the estimate of the structural
fiscal balance, differentiating it from copper revenue. It is interesting to note that
CODELCO does not estimate the cost of producing molybdenum but subtracts
the value of its sales from the total costs of production to estimate the net costs
of producing copper.
The cyclical adjustment made by the Ministry of Finance is as follows: the
cyclical component is estimated as the physical molybdenum sales of CODELCO
multiplied by the difference between the actual price and the “expected long-term
reference price.” The fundamental difficulty is to determine the long-term price,
since market information is scarce. Through expert assessments, for the estimation
of the 2007 fiscal budget, a trend price of US$14.7 per pound was used, a figure
equivalent to the moving average of monthly prices in 2002–6.
45 45
40 Av. real price in 2008 40
35 Real price October 2008 35
30 Real price November 2008 30
25 25
20 20
15 15
10 10
5 5
0 0
1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008
Table VI.3 Indicators of export diversification, 1990–2008 (number of markets, firms,

and products)
1990 1994 1998 2002 2005 2008
Markets of 129 141 172 158 184 190

destination
Firms 4100 5844 5847 6118 6880 8240
exporting
Products 2300 3622 3828 5160 5303 5161
exporteda
HHI indexb 0.399 0.280 0.279 0.271 0.345 0.383
Corrected HHI 0.364 0.273 0.305 0.305 0.275 0.258
indexc
Source: ProChile and UNCTAD for HHI.

a
In 2002, a further disaggregation of the trade nomenclature was made. bThe Herfindahl–
Hirschman index, calculated in current US dollars, reflects the degree of concentration of the
export basket, and takes values between 0 and 1 (maximum concentration). In 2001–5, the simple
average for Latin American countries was 0.273; for Korea and Taiwan it was 0.156; for the G-7
nations it was 0.099. The indexes were calculated by the author, based on UNCTAD methodology.
c
Corresponds to a modified HHI, based on the UNCTAD methodology, but using a trend copper
price. This reduces the distortion created by the rise of the copper price in recent years.
same period;9 however, the degree of concentration is notable, as fifteen

firms captured nearly 60% of total sales in 2008. The number of products
sold abroad increased notably too; although, still, ten items (out of over
5,000 items exported) represented as much as 69% of the total value of
goods exported.10
The concentration of the Chilean exporting basket is also remarkable
in an international comparison. The Herfindahl–Hirschman Index (HHI),
a measure of exports concentration that takes values between 0 and 1
(where 1 represents maximum concentration), places Chile as far more
concentrated than other developing and developed nations. According to
UNCTAD figures, in 2000–5 the index averaged 0.289 for Chile, whereas
Latin America averaged 0.273. The index for developing economies like
Korea and Taiwan was 0.156, whereas the G-7 developed nations averaged
0.099. In Table VI.3 it can be observed that Chile appeared to be worsen-
ing further in 2005 (0.345) and 2008 (0.383), due to the increasing share of
copper in total exports, as a consequence of the high copper price. Table VI.3
presents a “corrected” HHI Index, re-estimated with a trend price of copper.
9
That figure represents about 1% of the number of firms registered by the internal
revenue service. However, among large firms (close to 10,000), one-quarter have
become exporters.
10
See ProChile (2007). In 2007, 58.3% of exporters (4,611) sold less than US$100,000
annually and covered just 0.1% of total exports.
Nevertheless, even normalizing with a trend copper price, the corrected

index shows that Chilean exports are comparatively concentrated, but after
the step back during the contagion of the Asian crisis, they continued to
improve, though slowly.
Table VI.4 depicts product diversification by markets of destination. The
first column shows that, despite diversifying in recent decades, Chilean
exports continue to rely heavily on natural resources. In fact, it can be
seen that manufacturing exports also include a high share of commodity-
intensive goods. Nonetheless, there is a trend towards more manufactured
goods and several non-traditional natural resources. The classification in
Table VI.4 signals that the percentage of natural resource-based exports
decreased from 95% in 1983–9 to 90% in 2004–8, but they still cover a
notably high share.11 This feature is not surprising because Chile has a rich
endowment of natural resources combined with geographic factors that pro-
vide strong static comparative advantages. Nevertheless, this dependence on
natural resources and the limited value-added of manufactures also under-
lies the price instability faced by Chilean exports and the low dynamism of
the demand for its exporting basket. Additionally, it verifies that there have
not been effective, systematic, and powerful efforts to change in a more
significant way the inertia of this composition.
The geographic distribution of Chilean exports exhibits a significantly
larger diversification than in previous decades. The European Union’s share
has been falling sharply from 37% in 1983–9 to 24% in 2004–8. The United
States’ share also diminished, from 21 to 14%, whereas Latin America
receives 18% of Chilean exports, associated with several trade agreements
covering the vast majority of the markets of the region.
Likewise, the Asian market increased strongly in importance in the first
half of the nineties, becoming as a region the principal destination for
Chilean exports. The group Asia (10) doubled its share in 2004–8 (to 10.8%)
with respect to the average in 1983–9, whereas Japan’s share remained at
11%. The sharpest change is with China, which jumped from 2 to 12%,
concentrated solely in natural resources. The share of Asian markets suf-
fered a setback in 1998 as a result of the crisis that originated in that region.
Nevertheless, the fast recovery of the East Asian nations and the strong
expansion of exports to China made this region by far the most important,
taking 34% of exports in 2004–8 (adding together China, Japan, and the
other Asia 10 countries).
The negative effects of the Asian crisis were multiplied in Latin America,
since it found several countries in a financially vulnerable situation.
Especially, by 1998 the international capital markets had “dried off,” pro-
voking severe crises in several countries, which translated into a “half lost
11
The share of natural resources is biased upwards since 2004 due to the sharp rise of
the price of copper, expected to be transitory.
Table VI.4 Geographic distribution of Chilean exports, according to technological content, 1983–2008
Classification Composition Geographic distribution
(%) of total
Latin EU (15) USA Asia (10) China Japan Other
America (33)
1983–9
(a) Primary goods 77.4 10.6 40.8 22.0 5.4 1.8 12.8 6.5
(b) Natural resource-based 17.4 26.4 26.4 13.5 8.5 4.7 8.2 12.1
manufactures
(c) Manufactures and others 5.2 44.4 11.0 21.0 1.7 2.1 1.8 17.9
Low technology 1.2 42.3 10.2 37.8 1.5 0.1 0.6 7.5
Medium technology 2.6 56.2 13.6 17.7 2.7 3.3 3.3 3.3
High technology 0.3 39.8 19.6 35.7 0.8 0.0 0.3 3.8
Other transactions 1.1 25.9 0.8 1.0 0.2 0.0 0.0 72.1
(d) Total 100.0 14.9 36.8 20.8 5.6 2.3 11.5 8.1
1990–7
(a) Primary goods 67.9 11.1 32.4 14.9 14.7 2.0 20.3 4.6
manufactures
Low technology 3.2 58.8 8.0 27.2 2.8 0.1 0.6 2.9
High technology 0.6 63.7 4.5 25.9 1.2 0.0 0.3 4.4
(d) Total 100.0 18.8 27.8 14.9 12.7 1.7 17.5 6.5
(Continued)
163
164
Table VI.4 Continued
Classification Composition Geographic distribution
(%) of total
Latin EU (15) USA Asia (10) China Japan Other
America (33)
1998–2003
(a) Primary goods 61.9 13.2 28.7 16.5 11.6 6.8 16.7 6.5
manufactures
Low technology 3.0 66.2 5.9 22.6 1.7 0.6 0.2 2.7
High technology 0.8 74.7 4.7 16.5 0.5 0.3 0.3 3.0
(d) Total 100.0 22.2 24.7 17.7 9.3 5.4 13.1 7.6
2004–8
(a) Primary goods 71.9 10.9 26.2 12.5 12.4 14.7 13.8 9.4
manufactures
Low technology 1.6 70.9 5.4 17.0 1.7 0.9 0.1 3.9
High technology 0.4 72.3 6.4 16.8 0.8 0.3 0.3 3.1
(d) Total 100.0 17.7 24.3 13.8 10.8 12.1 11.3 9.9
Source: Based on United Nations ComTrade database, in current US dollars, classified by technological content, based on the Standard International
Trade Classification Rev. 1.
decade” for the continent in 1998–2003, with a negative per capita growth
(–0.3% per year). The most dramatic case was that of Argentina, whose GDP
fell by 18% in 1999–2002. Evidently, the regional market became notably
restricted for Chilean exports, with a diminishing participation from 24%
in 1998 to 18% in 2004–8. Among Latin American destinations, Mexico
has been the most dynamic, with a growing share in both total exports and
manufactures.
Actually, as was analyzed in Chapter V, the composition according to
geographic markets indicates that manufactures (with higher value-added)
are sold principally on Latin American markets, whereas commodities are
exported principally to the developed countries; in 2004–8, nearly one-half
of exports to the region, including MERCOSUR, were manufactured, as
compared to 10% of total exports. Furthermore, Table VI.4 shows that the
higher the degree of manufactures elaboration the larger tends to be the
participation of LACs. On the other hand, only 11% of primary goods have
intra-regional trade as their outlet. Thus, Latin American countries play a
crucial role in the diversification of Chilean exports towards manufactured
products. This diversification has been closely associated with the signifi-
cant trade liberalization of the region, including numerous regional trade
arrangements.
The question remains, however, as to the sustainability of Chile’s manu-
facturing export growth. On the one hand, much of the demand for these
products continues to depend on the Latin American region’s economic
growth performance, macroeconomic sustainability, and exchange rates
favoring intra-regional trade. On the other, it also depends on the strength-
ening of productive development policies focused on building capacity
in goods and services more intensive in value-added and innovation, as
discussed below.
3 Exports and foreign investment
One desirable effect of FDI is its contribution to export development. Besides

the intrinsic advantages that the expansion of exports could have, there are
at least two reasons why it seems desirable to link them to FDI.
First, from the point of view of the balance of payments it is necessary to
take account of the long-term equilibrium between the supply and demand
of foreign currency. Though capital inflows generate an increased supply of
dollars in the short run, this situation may not necessarily be permanent;
it depends on how much of this supply is allocated to the production
of tradables; even more, the profits earned will be repatriated, and then
the capital will follow. That is, FDI inflows also become a liability for the
host economy and, therefore, it must generate the resources to pay itself.
Naturally, exports are the only sustainable source of foreign currency in the
long run. Consequently, for the foreign exchange market it makes a great
difference whether inflows are directed to creating new capacity instead of

acquiring existing firms, and, in the case of creation, whether they are allo-
cated to the production of tradables or to non-tradables. In large developed
economies with deep capital markets this might not be a relevant issue, but
in emerging and small economies it has been quite relevant, as the case of
Chile demonstrates.
There are two opposite effects of FDI on exports. On the one hand,
there is the direct and positive impact that foreign capital can have in the
development of sectors geared to external markets and, on the other hand,
there is the indirect effect on the foreign exchange market; that is, the
exchange rate appreciation resulting from capital inflows and its negative
impact on the competitiveness of exports. In the case of Chile, the trade
pattern has been strongly affected by FDI in its dual role. In fact, FDI has
significantly contributed to the growth of Chilean exports,12 particularly
in copper output.13 These larger inflows, in turn, were partially responsible
for the peso overvaluation in 1996–8. This process had not only negative
macroeconomic consequences (see Chapter VIII), but also productive effects
because of the Dutch disease affecting several tradable sectors, especially
non-traditional exports, which are more elastic to the level and stability of
the RER (ECLAC, 1998, Chapter IV).14
Second, the space for trade policy to stimulate export development has
experienced a great tightening in the past decades. The WTO agreements
have imposed narrow limits on the use of incentives to exports. Part of
this is positive and benefits Chile, which is a much open economy; but
there is also a negative effect since restrictions were imposed on the use of
mechanisms such as the simplified drawback, of great importance since the
mid-eighties for non-traditional exporters. Thus, some selectivity in FDI is
12
The importance of foreign capital in exports is revealed by the fact that fourteen
of the twenty largest exporting firms are foreign-owned, and nine of them are in
mining.
13
The value-added to the Chilean economy in this copper output is notably lower
than that of CODELCO, because privately produced copper is less processed and
actual taxes on profits are lower (Meller, 2002). Another important factor is that a
high share of FDI during the late 1990s and the subsequent recessive period included
acquisitions of existing assets that do not directly generate new productive capacity
(Ffrench-Davis, 2003). According to UNCTAD data, 58% of FDI gross inflows in 1990–7
corresponded to greenfield FDI, falling to 19% in the recessive 1998–2002 period. In
contrast, during 2003–6, greenfield FDI reached four-fifths of total FDI inflows.
14
In those years, there arose propositions to ration FDI in the production of copper
through auctions. The main objectives were (i) to reduce the volume of inflows (thus
moderating the exchange rate appreciation), (ii) to weaken the depressive impact on
the future copper price, and (iii) to capture part of the natural resource economic rent,
which was subject to a low taxation that was unlinked to the quality of the resource
(Agosin and Ffrench-Davis, 1998). A royalty that captures an additional (though
small) share of the copper economic rent began to be applied only in 2006.
outlined as one of the most important mechanisms available in the future to

upgrade the quality of exports (Agosin and Bravo-Ortega, 2007);15 recently,
the authorities have been moving in this direction.
These facts highlight the importance of a comprehensive set of poli-
cies to ensure a sound export development, more diversified, with higher
value-added and with a dynamic international demand. An active exchange
rate policy and coherent productive development policies are absolutely
necessary for a successful outcome. This approach has started to take form
in recent years, departing from the “policy neutrality” inherited from the
Pinochet dictatorship. CORFO has developed a program to attract FDI in
information and communications technology. More significantly, beyond
FDI, the government has selected seven areas on which to focus public sup-
port in order to build productive clusters around them. I return below to this
relevant progress in use of policy space.
4 Exports and economic growth
Exports have been frequently called the engine of Chilean economic

growth. The growing share of exports in GDP has been a transmission chan-
nel of externalities resulting from exposure to foreign markets. The direct
incidence, however, is not easy to estimate quantitatively and research on
this issue is limited.16 In this section I concentrate on two issues. In the
first place, I identify the actual weight of exports in the domestic economy,
because that is the main determinant of how much their evolution directly
affects GDP: for example, a 10% increase in a component with a weight
of 30% gives a direct 3% change in GDP. There follows an analysis of the
“microeconomic” interrelation between exports and the growth of produc-
tive capacity. Finally, a macroeconomic interrelation between exports and
the rest of GDP, which has been very influential on the rate of economic
growth, is examined.
(a) What is the real weight of exports?

It is commonly repeated that foreign trade represents around 60 or 70% of
the economy. That is based on adding exports and imports and dividing them
by GDP. In practice, exports are part of GDP, but imports are part of domestic
15
It would have been relevant that the WTO new round (Development Round)
would have made a clear distinction between productive and financial investment.
Regarding the productive dimension, it requires that developing countries could
apply functional selectivity to “complete” markets of technology, human capital, and
long-term financing. With respect to financial investment it is crucial that developing
countries strengthen their capacity to apply active counter-cyclical macroeconomic
policies; among others, it implies being able to regulate capital flows. See Ocampo
(2001, 2007).
16
See Macario (2000) for pioneering research on Brazil, Chile, Colombia, and Mexico.
expenditure and belong to the output of the rest of the world. Additionally,
it is necessary to differentiate between gross exports (which are the figures we
normally observe) and net exports. The former correspond to the total value
of exported output (reported in the trade balance and national accounts),
while the latter discount the imported content of exports goods and, there-
fore, reflect the actual direct value-added contributed to GDP.17
Alternative measures of the ratio between gross exports and GDP are pre-
sented in Table VI.5. In the period 2004–9, for example, they represented
43% of GDP measured at current prices, 48% in 1977 prices, and 39% in
2003 prices. The resulting ratios depend on the level of the RER and relative
prices of exports in the respective base year of national accounts; both vari-
ables have been extremely unstable.
In 2003 prices, the present base year of National Accounts, gross exports
of goods and services represented, as said, 39% of GDP in 2004–9, while net
exports – that is, the value added by exports to GDP, which is a sum of value
added by all sectors of the economy – only makes up less than 30% of GDP.
This is still a very significant weight, and has been rising vigorously from
9% in the 1960s.
(b) Bases for export-led growth

The pulling capacity of exports on the rest of the domestic economy is
associated with the linkages between exports and national development.
Table VI.5 Export participation in total output, 1961–2009 (% of GDP)

Gross exports
Current Constant prices of

prices
1977 1996 2003
1961–70 15.8 12.1 8.8 9.6

1971–3 11.7 10.1 7.3 8.0
1974–81 21.8 20.2 14.9 16.4
1982–9 28.1 27.9 20.6 22.5
1990–8 29.0 35.5 26.3 28.8
1999–2003 32.4 43.6 32.3 35.4
2004–9 43.0 47.6 35.3 39.0
Source: Based on official central bank figures and Marcel and Meller (1986). Exports include
goods and services. Annual averages. For exports net of imported imports see Table VI.6.
17
GDP, on the expenditure side, is the sum of value-added on goods and services
consumed and invested domestically, and of net exports of goods and services (gross
value minus imported content).
Undoubtedly, there was progress in the past decades, since together with
traditional natural competitive advantages, there was also a development of
dynamic advantages (many of them natural resources-based).
Nevertheless, there are other arguments for additional diversification
towards goods and services with greater degree of elaboration, beyond
purely natural resource-intensive products. To achieve progress on this issue
is an important challenge. First, the fact that a high share of Chilean exports
are still concentrated in primary goods means that the economy as a whole
becomes highly vulnerable to the intense commodity price fluctuations.
Second, the long-term dynamism of those products is limited, which in the
future could become a negative factor for potential economic growth (Sachs
et al., 1999); this argument assumes that the optimism in fashion today that
China will continue to pull up vigorously the demand for copper is not
sustainable, and that supply will not respond as much. Third, the production
of goods with a greater degree of elaboration involves positive externalities
for the rest of the economy, by means of the learning-by-doing processes
and the benefits caught in the acquisition of new dynamic competitive
advantages in related products.
There is some encouraging progress in economic activities developed
around some of the successful exports, such as forestry, wine, salmon,
and fresh fruit. According to some analysts, those successful experiences
could support a “Nordic” development strategy for Chile, based on higher
incorporation of value-added to natural resources (Díaz and Ramos, 1998).
However, the speed of emergence of new exports diminished by the late
1990s.18 The exchange rate policy became detrimental for non-traditional
exports and the simplified drawback for small exports disappeared. More
recently, in a significant step forward, the newly created National Council for
Innovation and Competitivity has identified seven existing exports on which
to focus public support of and cooperation with the private sector, in order
to foster the development of productive clusters around them. The selection
includes sectors such as copper mining, aquaculture, fruit production, proc-
essed foods, and tourism. This is relevant progress in policy space.
A strong positive link between exports and national development implies
achieving systemic competitivity; that is, succeeding in becoming produc-
tive in producing for exporting and for the domestic market. That would
require the correction of market failures in the capital market, labor train-
ing, and research and development. In order to capture the potential gains
from these corrections a reform in macroeconomic policies is crucial.
18
Export discoveries tend to bunch in the period 1991–7. In 1998–2003, export dis-
coveries fall off dramatically. The weakening of export discoveries has much to do
with the lagged impact of the real exchange rate (see Agosin and Bravo-Ortega, 2007;
Agosin et al., 2009).
(c) Macroeconomic environment and non-exported GDP

Here I concentrate on the influence of domestic macroeconomic policies on
the association between the evolution of exports and the rest of GDP.
In fact, there is a significant positive correlation between export growth
and actual GDP growth after the 1980s debt crisis, and in the nineties.
During such periods the exports were of great dynamism, whereas actual
GDP did recover in the late eighties (after its dramatic fall in 1982–3) and, in
the nineties, it entered the period of greatest vigorous and sustainable growth
in its history. Most econometric research on this matter has detected a clear
positive effect of exports (especially non-copper) on overall GDP growth
(Meller, 1996b; Coeymans, 1999; Agosin, 2007); however, the elasticity of
GDP with respect to exports is quite variable and notably less than one.
For exports to be the real engine of growth, their expansion must be
linked to the creation of new productive capacity in the rest of the economy,
a feature strongly present solely in the 1990s. In that decade, the capital
formation ratio rose to unprecedented levels, and this process was intense
not only in the export sector. Non-tradables exhibited notably higher rates
of expansion that in the seventies and eighties: for instance, infrastructure
works and telecommunications recorded annual growth of 18.3 and 32.4%,
respectively, in 1990–7.19
Therefore, in the 1990s, global actual GDP growth was led by both the
tradable and non-tradable sectors. Actually, (gross) exports and non-exported
GDP grew 9.9 and 6.5% per year, respectively, in 1990–8 (see Table VI.6).20
This, accompanied by an environment of full utilization of productive
capacity (see Chapter I, Figure I.1), with real macroeconomic stability and
prudential regulation, accounts for the high growth rate of the Chilean
economy during most of the decade.
In the second half of the 1990s, however, the real exchange rate over-
appreciated, led by excessive capital inflows (see Chapter IX), implying a
distortion in resource allocation and the trade balance. As a consequence, a
significant macroeconomic vulnerability was generated that led to a recession
in 1999, after the contagion from the negative external shocks from Asia.
During 1999–2003 the economy worked significantly below its produc-
tive frontier. Some share of the poor performance in those five years can
be explained by the falling growth rate of exports. Nevertheless, given that
19
These figures must be compared with an annual growth in infrastructure invest-
ment of 2.6% in 1982–9 (Moguillansky, 1999).
20
Most of Latin America had also embarked on an export-led strategy. Actually, the
volume exported by the region rose by 8.3% in 1990–7, up to the peak before the
arrival of the Asian contagion; but, in contrast with Chile, its GDP expanded merely
3.3%. In that sense, the region exhibited an export and GDP outcome rather similar
to that of Chile in the 1970s and 1980s, but quite worse than the good performance
of Chile from 1990.
Table VI.6 Exports and economic growth, 1961–2009 (annual average rates of
growth, %)
GDP Exportsa Non-exported
GDPb
1961–70 4.4 3.6 4.5

1971–3 1.2 –4.1 1.7
1974–81 3.0 13.6 1.5
1982–9 2.9 7.8 1.7
1990–8 7.1 9.9 6.5
1999–2003 2.6 5.5 1.7
2004–8 4.9 6.6 4.2
2009 –1.5 –5.6 –0.1
Chile 1990–2008 5.3 7.9 4.6
Latin America 3.2 6.9 2.7
1990–2008
World 1990–2008 3.4 5.9 –
Source: Based on Central Bank National Accounts and Marcel and Meller (1986). For Chile
exports include goods and services at constant 2003 prices. ECLAC and IMF for LACS and the
world, respectively.
a
Diverse sources and methodologies explain differences from figures in Table VI.1. bNon-
exported GDP is equal to GDP minus the domestic content or value-added in exports. Imported
inputs in exports were assumed to be equal to the share of non-consumer imports in GDP.
For Latin America, based on Ffrench-Davis (2006, Table V.2) and data processed by ECLAC for
nineteen LACs.
exports then represented only a quarter of the domestic value-added, that

fall only translated into around 1 point of lower GDP growth. The growth
rate of the volume of exports diminished by around 40% in 1999–2003
(from 9.9 to 5.5%), but that of GDP diminished by two-thirds (from 7.1 to
2.6%). That is to say, the fall from a 7.1% growth plateau to 2.6% is mainly
explained by the meager increase in the rest of GDP; that is, the output
directed to domestic markets. The non-exported GDP grew by hardly 1.7%
in 1999–2003, in contrast with 6.5% in the previous period.
In 2004–8, a moderate recovery of the volume of exports (to 6.6%) as
well as a significant recovery of non-exported GDP (from 1.7 to 4.2%) took
place, determining growth rates of total GDP averaging 4.9% in that period.
As shown, while the prices of exports are notably more unstable than their
volume, the volume or quantum of non-exported GDP – that is, GDP sold
in the domestic economy – has been quite unstable, reflecting failures in
macroeconomic policies: sharp changes in aggregate demand and the RER.
The exception was the period 1990–8.
It is interesting to compare the performance of Chile with that of Latin
America in the whole period 1990–2008. Table VI.6 shows that the sharp
difference between the 5.3% GDP growth of Chile and the meager 3.2% of
Latin America is not focused in the export behavior, which is rather similar
(7.9% versus 6.9%). Actually, the gap is located, principally, in the evolution
of the rest of the economy: a rise of 4.6% in Chile in contrast with 2.7% in
Latin America. This is quite illustrative for the coherency of development
strategies.
Finally, it is evident that the abrupt changes that have taken place in
the relation between exports and GDP, recorded from year to year, do not
respond to structural swings. They are associated, principally, with instabil-
ity in the domestic macroeconomic environment, which has affected the
production sold in the domestic market. This instability, naturally, addition-
ally, weakens the productive link between exports and GDP and conspires
against the export-led strategy.
In all, during the past three decades, Chile witnessed a period of significant
export growth. This outstanding performance was associated, during the
past decades, with some active heterodox policies aimed at sustaining a
competitive real exchange rate and generating export capacity, rather than
solely with the neo-liberal (orthodox) economic reforms of the 1970s.
Four reciprocally reinforcing groups of factors appear to be more signifi-
cant for explaining the behavior of exports. First, it is a well known fact that
the level and stability of the real exchange rate has a determinant impact
on the overall export performance. Second, the diversification toward
goods and services with value-added is crucial for enhancing the quality
of exports, in the sense of benefiting from (i) more dynamic demand and
better international prices and (ii) stronger linkages of exports with the
domestic economy; the stability of the real exchange rate, incentives like the
simplified drawback for non-traditional exports, and intraregional integra-
tion have been significant variables at hand. Third, to upgrade the quality of
exports there is a need for an intensive national effort to complete domestic
markets for technology, labor-training, and long-term segment of the capital
market; all these are crucial ingredients for achieving systemic competitivity.
Fourth, the export drive needs to be accompanied by a domestic macroeco-
nomic environment suited for economic growth, sustainable and with the
right macroprices, consistent with productive development, in order to con-
tribute to export-led growth. This was the determinant factor in explaining
the positive association between export dynamism and sustainable growth
of overall GDP during most of the 1990s.
The Chilean economy seems to be at a crucial point with respect to its
international trade insertion. The signing of free trade agreements with
large economic blocs represents opportunities for many productive sectors
to extend their potential markets, but at the same time it raises challenges.
The case of Mexico, whose performance has been disappointing since the
enforcement of NAFTA, shows that the benefits of these associations are not
automatic. On the other hand, there are potential costs derived from the
losses by the authorities of degrees of freedom to implement certain wise
interventions. In this new scenario, it is essential to maintain the principles
explored above to guarantee an export development that not only takes
advantage of static comparative advantages but also generates a dynamic
systemic competitiveness.
VII
Inequality and Poverty since
the 1970s*
Inequality has once again become a major public policy issue in Chile. Despite
effective efforts made from the 1990s to reverse the ramping inequality of
the 1970s and 1980s, poverty is still a fact of life for too many Chileans;
a sharp inequality in opportunities, wealth, and income still prevails. Lack of
equity is a marked feature of the Chilean economy and society.
Significant changes in poverty and income distribution have been
recorded over the past one-third of a century. Its evolution can be summa-
rized in the following four points:
1 In the 1970s and 1980s, during Pinochet’s dictatorship, income distribu-

tion deteriorated sharply, and the share of the population living under
the so-called poverty line rose to 45% in 1987; that situation, associated
with the crises of 1975 and 1982, was the result of the higher concentra-
tion of wealth, worsening of real wages, increased unemployment, and
the fall of per capita social expenditure.
2 Poverty was reduced substantially from 45% of the population in 1987 to
14% in 2006.
3 After the great deterioration recorded during the dictatorship, income
distribution exhibited an improvement in the first half of the 1990s, but
it stepped back partially with the Asian crisis contagion; subsequently,
partial progress has been recorded, due to some economic reactivation
and new quite significant social reforms, moving closer to the best level
reached in the first half of the 1990s.
4 Today income distribution is less unequal than in the 1980s, rather
similar to the 1970s, and more regressive than in the 1960s. The ratio
* I am grateful for the valuable comments of David Bravo, Juan Carlos Feres, Osvaldo
Larrañaga, Carlos Massad, Jaime Ruiz-Tagle P., Jaime Ruiz-Tagle V., and Humberto
Vega; the exchange of views with Harald Beyer, Leonardo Moreno, Dagmar Raczynski,
Joseph Ramos, Claudia Sanhueza, Claudio Santibañez, Daniel Titelman, Arístides
Torche, and Andras Uthoff; and the efficient assistance of Rodrigo Heresi, Heriberto
Tapia, David Coble, and Felipe Labrín.
174
Inequality and Poverty 175
rich/poor quintile (GINI), which was around 13 times (GINI 5 49) in the
1960s, worsened to 20 (GINI 5 57) in the 1980s, and in recent years has
averaged about 14 times (GINI 5 52).
Notwithstanding the significant social improvements achieved during the

two decades of democracy, the net balance of income distribution, compar-
ing the 1960s and recent years, indicates that Chile lost ground over the
past four decades rather than progressing towards greater distributive equity.
Equity and macrosocial balances (that is, employment, wages, social programs,
sense of partnership in the national society, with life conditions improving
in a sustainable manner) are essential ingredients of modernization.
In this chapter, I review the most outstanding distributive develop-
ments over recent decades. Several social policies can affect poverty and
distribution. They include, among others, pensions, family allowances, the
minimum wage, incentives to job search for women, health and housing
subsidies, and unemployment insurance. But, in a developing country, the
top priority is to improve the performance of the labor market, creating bet-
ter job opportunities for mid and low income workers; generating “decent
jobs,” with rising real incomes. Three aspects are stressed in this area. Two
of them are structural, the need (i) to improve the quality and quantity of
investment in people, or human capital, and (ii) to enhance productive
investment and its link to productive employment. Both factors contribute
to raise productivity across society, and thus to spread opportunities to
larger segments of the labor force. The third aspect is (iii) the attainment of
sustainable macroeconomic balances of the real economy.
The efficient approach for obtaining both growth and equity implies a
comprehensive definition of macroeconomic equilibria. This includes far
more than low inflation and a sustainable fiscal balance. It also requires
equilibrium of the real economy; that is, making full use of available produc-
tive capacity, avoiding excessively fluctuating and outlier real interest and
exchange rates, and securing a favorable macroeconomic environment for
productive investment. I show that the sharpest setbacks in income distribu-
tion and poverty have been caused by critical macroeconomic imbalances:
the hyperinflation of 1973, and the recessions of 1975 and 1982. To these
should be added the cases in which macroeconomic “balances” have been
achieved at the expense of other balances, such as the cases of macrosocial
disequilibrium with drops in social expenditure in 1985–7, and of external
imbalance in 1996–8, followed by a recessive environment and a distribu-
tive set-back in 1999–2003. The subsequent recovery of economic activ-
ity, in 2004–7, brought an improved distribution, encouraged by notable
improvements in social policies, but, in all, merely to regain the situation
achieved in the first years of the return to democracy in the 1990s. Real
macroeconomic imbalances generate structural, long lasting, regressive
effects.
176 Inequality and Poverty
1 Trends in income distribution and poverty
Despite vigorous economic growth, particularly in the nineties, Chile is still

a developing emerging economy. Its per capita income level (in purchas-
ing power parity, PPP) is less than one-third that of the most developed
economies, and income inequality is notably higher, as can be seen in
Figure VII.1. For instance, the ratio rich/poor quintile in the G-7 nations is
7/1, while in Chile the comparable measurement was 15/1.
(a) How is it quantified?

Good measurements are important because they provide information on the
effectiveness of socioeconomic policies aimed at reducing the inequalities
and poverty characteristic of underdevelopment. Nonetheless, the measure-
ment of poverty and inequality is beset with great difficulties.
The definition of poverty is a conventional one. The generally accepted
definition of the poor is that they are “those whose income per capita is lower
than the cost of two baskets of food and basic non-food needs” (ECLAC,
1997, Box 1). This is the dividing line between the poor and the non-poor
used in the CASEN survey in Chile, conducted by the Ministry of Planning
(MIDEPLAN). The poverty line has remained constant in real terms in the
nine surveys made between 1987 and 2006, based on the household budget
survey of the National Bureau of Statistics of 1987–8. The urban poverty line
is equivalent to about US$100 monthly per capita; US$400 for an average
poor family of four. Box VII.1 shows the urban and rural level of the poverty
line, and presents diverse income indicators for different groups in society
in 2000 and 2005.
The definition does not tell us anything about how far the many people
who ceased to be numbered among the poor between 1987 and 2006 have
risen above the poverty line or about the previous position of those who
Per capita GDP

40.000
20% richest v/s
35.000 20% poorest Per capita GDP
30.000 7 times 20% richest v/s
25.000 G-7 38,578 (USA 8 times)
20% poorest
20.000 15 times
15.000 (USA 44,765)
10.000 13,745
5.000
0
G-7 Chile
Figure VII.1 Chile and developed countries, per capita GDP and income distribution,
2007 (PPP US$).
Source: IMF, World Economic Outlook Database (2008), World Bank, World Development
Indicators (2008).
Box VII.1 Wages and the poverty line, 2000 and 2005
(monthly income in UF)
The Table shows the monthly amounts in Unidades de Fomento (UF, the price-
indexed unit of account, is a financial unit of account, adjusted daily by changes
in CPI; in 2005, one UF was equivalent to US$31) of monthly wages in Chile
according to the main sources available. It also shows the value of the (per capita)
poverty line (rural and urban). It must be noted that the sources do not all register
the same concept of wage. While some gather the liquid wage, others report the
assessable salary. If workers’ contributions to social security (of approximately
20%) are discounted from assessable salary, then the monthly average liquid wage
in 2005 was between 12.2 UF (according to unemployment insurance figures) and
17.4 UF (University of Chile); that is to say, between US$380 and 545 (2005
prices). The information from private health insurance (ISAPREs), which includes
the population with higher income, exhibits an average liquid wage of 26.5 UF
(US$830 of 2005).
2000 2005 Wage Coverage
Average wages
ACHSa 20.0 21.4 Assessable National
AFPa 18.3 19.3 Assessable National
Male 19.3 20.3 Assessable National
Female 16.7 17.7 Assessable National
INEa 15.5 15.7 Assessable National
ISAPREa 27.4 33.3 Assessable National
Unemployment insurancea – 15.4 Assessable National
CASENb 14.2 14.2c Liquid National
Male 15.6 15.4c Liquid National
Female 11.9 12.2c Liquid National
University of Chiled 16.6 17.4 Liquid Greater Santiago
Minimum wagea 6.4 7.1 Assessable National

5.1 5.7 Liquid National
Poverty line
Urbanb 2.6 2.6 National
Ruralb 1.7 1.7 National
a
December. bNovember. c Year 2003. dJune. The assessable wage is the gross salary before
health and pension fund contributions. The liquid wage is the net amount actually received.
Universes (2005). ACHS: Chilean Association for Labor Accidents includes 1,718,166 work-
ers in 35,421 enterprises. AFP: Pension Fund Administrators includes 3,321,793 contributors
(98% dependants, 63% male). ISAPRE: Private Health Insurance System includes 1,241,018
contributors and 2,673,409 beneficiaries. In 2005, the population was 16.3 million and the
labor force 6.8 million.
The average number of persons by household for quintiles I and II was somewhat larger than
four. This means that an urban household of four people, for example, needs a monthly
income of 10.4 UF to surpass the poverty line. This amount was somewhat less than twice the
liquid minimum wage of 2005.
have fallen into poverty. A complementary measurement is the estimation

of points of GDP needed to raise the income of all the poor up to the pov-
erty line: that objective could be achieved with funds equivalent to 2% of
the incomes of non-poor people (see Feres, 2001b).
Poverty is not static. In particular, the precariousness of the poor or close
to poor population is accentuated under recessive economic situations. In
MIDEPLAN (2008), as will be seen later, this volatility is quantified, show-
ing the household shifting away from and towards poverty between 1996
and 2006.
Inequality is much harder to measure. Here I focus, principally, on income
distribution. The inaccuracies correspond, especially, to the extremes of the
distribution, and particularly the high income brackets and the rent of capi-
tal. The measurement of the income of dependent workers is more accurate,
in particular of mid and low wages, due to its being formal and subject to
social security contributions.
In turn, even if the figures available were correct, there are still many
alternative ways of organizing the information – for example, by income
or by expenditure per household or by per household member – and the
differences are substantial among diverse indicators. Once the information
has been classified, there are also various ways of measuring distribution;
indicators range from the more cryptic (such as the traditional GINI indica-
tor) to simpler ones such as the ratio between the shares of the richest and
poorest quintiles or deciles. One drawback of the latter indicator is that it
does not take into account the 60% or 80% in the middle brackets and it
gives great weight to the richest income bracket, where measurements are
very defective. The evolution of the intermediate sectors is caught, among
others, implicitly by the GINI coefficient (see Table VII.5).
There are different sources of information on income and expenditure
distribution in Chile. Some distributive outcomes differ radically between
the various sources. The data source of longest standing is the employment
survey of the University of Chile, which has been collecting information
on income in the capital, Santiago (covering over 40% of the Chilean
population), annually since 1958.1 The CASEN national survey is available
for 1987, for every two years between 1990 and 2000, and for 2003 and
2006; the new 2009 survey is being processed. Both the coverage and the
survey itself were improved in the 1990s, so that comparability with 1987
is limited; nevertheless, its aim is the characterization and measurement of
poverty and the access of the poor to social transfers and services, and it is
weak in the measurement of income distribution, particularly in the higher
1
The distribution data refer to household per capita income according to the University
of Chile employment survey. See Larrañaga (2001) and Center of Micro Data (Centro
de Micro Datos), Department of Economics, University of Chile, January 2010.
quintile. Paradoxically, notwithstanding those weaknesses and the fact that

it has been available only since 1987, it contains the data most commonly
used by analysts and mass media to measure income distribution in Chile.
This is due, probably, to the fact that the CASEN data are easily available and
have national coverage. Reliance on it is a risky option that leads to wrong
conclusions, such as that distribution has been rather stable for decades.
It misses the sharp worsening in the 1970s and 1980s.
Once a decade or so, the National Bureau of Statistics (INE) carries out a
detailed survey of household budgets in Santiago, which is solid on expendi-
ture levels and their composition. Data are available for 1969, 1978, 1988,
and 1997; a newer survey, for 2007, is partially published at present, cover-
ing on this occasion most large towns. INE also collects income data in a
survey on employment.2
(b) Progress and setbacks during the Pinochet regime, 1973–89

Notwithstanding social improvements made in the 1950s and 1960s, and
the comparatively progressive position within Latin America, the distribu-
tion pattern that prevailed around 1970 was regarded as highly inequitable
(Ahumada, 1966). Consequently, a number of proposals for improving
the situation were put forward. Several of these were implemented during the
administrations of Presidents Frei and Allende. A significant land reform was
carried out between 1965 and 1973 (Ffrench-Davis, 1973; Bitar, 1979; Ortega,
1987). In 1971, in Allende’s presidency, wages and social spending (pen-
sions, family allowances, education and health budgets, etc.) were increased
massively, although in an evidently unsustainable manner. The inflationary
surge of 1972–3, with annual rates in excess of 200 and 600%, respectively,
and the fall in real wages testify to it.
Some social indicators continued to improve during the Pinochet regime,
while others went sharply into reverse. The illiteracy rate, already down to
20% in 1952, fell to 10% around 1973 and to 6% in 1989, while the number
of students enrolled in primary schools, as a percentage of the population
aged six to fourteen, rose from 65 to almost 100% in 1973. As for secondary
education, the proportion of fifteen- to eighteen-year-olds enrolled climbed
from 10% in 1952 to 51% in 1973 and 75% in 1989.3
Developments were very positive with regard to life expectancy and infant
mortality. In particular, infant mortality fell markedly. In the early 1960s
the infant mortality rate was 110‰ and by 1970 had improved to 82‰.
2
There are many other expressions of inequality, such as in relation to gender,
the environment, the political system, wealth, power, the social or ethnic origin,
transparency or corruption, participation and voices. See, for example, Sen (2000),
Garretón (2003), and UNDP (2004).
3
See ECLAC, Statistical Yearbook, based on official Chilean information.
Between 1973 and 1980 it progressed from 66 to 33‰. In the 1980s Chile
exhibited the lowest level in Latin America, along with Costa Rica, Cuba,
and the English-speaking Caribbean. This enhanced performance was the
result of public efforts to improve mother and child care, including nutri-
tion programs for nursing children; a decline in the number of births;
and irreversible factors such as improvement in the education of mothers
(Raczynski and Oyarzo, 1981; Monckeberg, 1998).
Nonetheless, the performance of other indicators was quite negative (see
Table VII.1). This was essentially a reflection of: (i) economic and labor
reforms since 1973 that were strongly biased against workers and their econo-
mic organizations; (ii) great real macroeconomic instability; and (iii) a low
ratio of investment per member of the labor force (generating a negative
impact on productivity per worker). As a result, average real wages in 1989
were 8% below their 1970 level. In other words, over nearly two decades,
average wages fell; the worsening was enhanced by an increase in wage
inequality (Ruiz-Tagle V, 2007). The minimum wage declined by a similar
percentage over the period, and its coverage narrowed considerably, with
even lower levels being applied to apprentices, workers under twenty-one
(later changed to those under eighteen), and those over sixty-five (Cortázar,
1983). Similarly, family allowances, which had played a progressive role,
growing continuously in importance until the early 1970s (Ffrench-Davis,
1973), went into a steady decline after 1974; a regressive trend continued
until 1989, when they were 72% below their 1970 level.
Per capita public social expenditure (including health, education, hous-
ing, and retirement pensions) also fell, declining 16% in 1989 in comparison
with 1970.4 Public spending on education and health dropped substantially,
by 37 and 30%, respectively. Only social security spending increased, owing
to a rising number of pensioners, which over-compensated a reduction in
the average pension. However, some social spending (though not the major
part) was targeted at the poorest members of society, which appears to have
partially offset the decline in labor income.5 Many of these indicators wors-
ened during the 1970s, exhibited a partial recovery in 1979–81, and deterio-
rated again with the long recession started in 1982; average and minimum
wages began to rise only in 1988, and family allowances in 1990 and public
social spending in 1991, under the return to democracy (see Table VII.1).
The declines in labor income and in monetary social expenditure, as well
as the regressive bias of tax reforms in the 1970s and 1980s, were reflected in
a worsening distribution of consumer spending. Household budget surveys
4
Public spending is financed by fiscal funds and the contributions of beneficiaries.
A large discrepancy between official and “corrected” figures for social spending in the
1970s is discussed in Marshall (1981).
5
Sometimes faulty targeting implied the crowding-out of the middle class or even
the poor; for example, exclusion from the university, after the rather free access was
suppressed without a comprehensive system of scholarships.
Table VII.1 Wages, family allowances, and social expenditure, 1970–2009 (real indices,
1989 100)
Per capita public social expenditure
Average Minimum Family GDP per Education Health Total

wage wage allowance member of
labor force
(1) (2) (3) (4) (5) (6) (7)
1970 109.2 108.9 352.1 98.8 159.7 143.7 119.2

1980 97.2 141.6 287.3 100.6 141.5 118.2 107.4
1981 105.7 147.8 284.8 103.8 147.1 107.3 116.2
1985 90.9 93.7 192.2 81.8 121.4 92.0 107.9
1986 92.4 89.5 160.9 83.3 114.2 89.8 103.6
1987 92.1 84.0 134.2 89.2 105.0 88.4 101.0
1988 98.1 89.6 117.0 91.8 102.6 101.1 102.6
1989 100.0 100.0 100.0 100.0 100.0 100.0 100.0
1990 101.8 106.8 118.7 103.1 94.9 94.0 97.4
1991 106.8 116.8 145.8 108.9 105.0 108.9 104.6
1992 111.7 122.2 149.3 121.0 118.7 125.1 113.7
1993 115.6 128.2 151.9 123.8 127.2 138.6 123.1
1994 121.0 132.9 154.6 124.1 137.1 150.9 129.1
1995 129.2 138.9 159.4 136.9 151.1 155.3 136.7
1996 134.5 144.9 166.0 147.2 168.5 166.7 148.0
1997 137.8 150.1 174.9 153.9 183.4 174.4 154.5
1998 141.4 162.5 182.9 155.7 201.8 186.4 164.1
1999 144.8 173.8 187.8 148.0 213.1 189.7 174.7
2000 146.8 186.1 189.1 151.2 228.8 203.0 183.2
2001 149.2 193.2 191.0 154.8 244.9 217.1 192.4
2002 152.2 198.9 194.6 153.6 260.5 225.1 196.1
2003 153.7 201.5 196.3 149.7 261.4 236.1 198.0
2004 156.5 207.2 199.3 148.8 278.7 250.9 206.7
2005 159.4 211.2 198.9 150.5 274.6 270.3 214.9
2006 162.5 216.6 200.6 145.6 288.9 284.4 226.1
2007 167.0 220.3 227.0 151.0 318.9 338.9 242.4
2008 166.7 219.4 244.7 158.4 364.6 359.2 261.7
2009 173.9 229.8 265.1 – – – –
Average annual growth

1982–9 –0.7 –4.8 –12.3 –0.5 –4.7 –0.9 –1.9
1990–7 4.1 5.2 7.2 5.5 7.9 7.2 5.6
1998– 2.0 3.6 3.5 0.3a 6.4a 6.8a 4.9a
2009
Source: National Bureau of Statistics (INE) and Jadresic (1990) for wages; Cortázar and Marshall (1980)
for corrected CPI; Cabezas (1988) and the Budget Office (DIPRES, since 1986) for social expenditure.
Column (1) is the general wage index until April 1993, and later the hourly wage index; the latter
increased by 6% more than the other index between April 1993 and 2005. Column (2) represents liquid
income. Column (3) is the family allowance of blue-collar workers in 1970, then the uniform allowance,
and later that for the low-income bracket. Column (4) is an estimate of net national product based on
national accounts and depreciation from Table IX.2. Column (7) includes expenditure on education,
health, housing, pensions, and others. All are average figures for each year. aIncludes only up to 2008.
(Encuestas de Presupuestos Familiares, EPF) conducted in Santiago in 1969,

1978, and 1988, on which the basket of the CPI has been successively built,
show a steady decline in household spending in the three lowest quintiles.6
Furthermore, the poorer the sector of the population, the proportionally
greater is the decline (see Table VII.2). For example, the share of expenditure
of the poorest 40% of households (the first and second quintiles) fell from
19.4% in 1969, to 14.5% in 1978, and to 12.6% in 1988; in other words, they
lost one-third of their share of total expenditure. By contrast, the relative
situation of the richest quintile improved consistently, with its share rising
from 44.5% in 1969, to 51.0% in 1978, and 54.9% in 1988. Furthermore,
this is the only quintile in which spending per household rose in real terms
between 1969 and 1988.
This information on the distribution of expenditure shows that the sec-
ond half of the Pinochet government (1982–9) was even more regressive
than the first half, so that the concentration of wealth and income observed
during the first stage, from 1974 to 1981, was accentuated in the 1980s.
Table VII.2 Distribution of household expenditure in

Santiago, 1969, 1978, 1988 (%)
Quintile 1969 1978 1988
I 7.6 5.2 4.4

II 11.8 9.3 8.2
III 15.6 13.6 12.6
IV 20.6 21.0 20.0
V 44.5 51.0 54.9
Total 100.0 100.0 100.0
QV/QI 5.9 9.8 12.5
Source: National Bureau of Statistics, Household Budget Surveys

for Greater Santiago. Quintiles ranked according to household
expenditure; excluding imputed rent for home-owners, which is
available only since the 1988 survey (see Table VII.5).
6
We have searched for a possible distributive bias in the household budget survey
(EPF) associated with the fact that Santiago includes only 40% of Chile’s population
and that it is essentially urban. After comparing the results of the five CASEN surveys
recorded during the 1990s, Feres (2001a) finds that the GINI coefficients are relatively
similar for Santiago (average of 57.2) and for Chile (average of 57.4). Preliminary data
of the new EPF, for 2006–7, that now covers all regional capitals, exhibit the same
similarity: the GINI is not significantly different between Santiago and the average of
the rest of capitals. The EPF is used by the INE to build the CPI, which from 2010 will
be national instead of only for Santiago.
Information from the University of Chile employment survey for Santiago

likewise shows deterioration in income distribution, though differing in
intensity and with large fluctuations from year to year. According to this
source, between 1975 and 1987, the situation worsened steadily, stabilizing
temporarily during the upswing of 1977–80 only to deteriorate again after-
wards. The worst indicators were reached in 1987, whether measured by the
GINI coefficient or the ratio between the first and fifth quintiles (Larrañaga,
2001).
In previous chapters it was examined the role that a number of reforms
and policies of the dictatorship played in exacerbating the severe crises that
Chile faced in the 1980s. The dogmatic approach, particularly during the
first half of its rule, increased the country’s vulnerability to external shocks.
Among the worst consequences of the resulting recessions was the persist-
ent domestic unemployment that prevailed (see Figure VII.2). By 1975, the
percentage of unemployed already stood at 15.7%, a figure that rises to
17.6% if those working in the PEM (minimum employment program) and
the POJH public programs (employment program for heads of households)
are included. By 1983, the number of unemployed stood at 19% of the labor
force; emergency job programs absorbed another 13% of the labor force,7
for a total of 31.3%. The problem was later alleviated by the recovery of
35
Excluding emergency programs
30
Including emergency programs
25
20
15
10
0
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
Figure VII.2 National unemployment rate, 1960–2009 (percentage of the labor force).
Source: National Bureau of Statistics and Jadresic (1986), and Budget Office for Emergency programs.
Annual averages.
7
Although the emergency jobs programs were originally designed for a working week
of only fifteen hours, in practice full-time work was required. The wage paid, mean-
while, eventually fell to less than one-third of the minimum wage, with no social
security coverage. Unemployment benefits were virtually non-existent, although the
PEM performed partly as a subsidy in exchange for work. Earnings from the POJH
varied between 1.6 and 2 times those for PEM.
economic activity, but only in 1989 was there a return to single-digit unem-
ployment rates, with open unemployment of 8%. In a situation in which
unemployment hit the lowest income groups the hardest, with a lack of
unemployment insurance and weakened public social networks, the wors-
ening in income distribution is easily explained.
(c) Income distribution and poverty after the return

to democracy, 1990–2009
Since 1990, four periods with different socioeconomic outcomes can be
distinguished. During the first years (1990–5), there were significant real
improvements in the average and minimum wages and social expenditure,
which meant a recovery from the depressed levels of the two prior decades.
In parallel, employment, coverage by social security, and labor participa-
tion increased persistently. The macroeconomic environment, with fac-
tors of production (labor and capital) operating close to full employment,
provided a significant progressive equity dividend. In the second period
(1996–8), progress in poverty reduction slowed and wages experienced
smaller increases, while other social indicators kept improving. In the third
period (since 1999), under a reappearing recessive output gap, the environ-
ment worsened dramatically, with a sharp rise in unemployment and a
noticeable moderation in poverty reduction (CASEN 2000, 2003). However,
the minimum wage – which is proposed annually by the government to
the Parliament – rose significantly through the establishment of a trien-
nial program for 1998–2000. Similarly, wages of teachers and workers in
the national health service as well as public pensions experienced special
raises. The associated higher fiscal expenditure was duly financed with tax
revenue.
Finally, the fourth period starts with the recovery of economic activity
in 2004 and sharp reinforcement of social policies, especially intensive in
2008–9, notwithstanding the contagion of the global crisis. The CASEN
2006 survey shows substantial progress in poverty reduction, of 5 percent-
age points with respect to the 2003 survey (18.7–13.7%). The moderation
in poverty reduction in 2000–3 as well as its acceleration in 2003–6 shows
how determining the macroeconomic environment is, through its impact
on the labor market. In parallel, employment has recovered and with greater
formality, and the minimum wage has continued to rise (Table VII.1 signals
that in 2009 it corresponds to an amount 2.3 times the real level inherited
from the dictatorship); meanwhile, social programs, such as Chile Solidario
and the solidarity pillar of the ambitious Social Security Reform of 2008,
were focused on poor and more vulnerable people.
In 2007, the authorities and public opinion acknowledged with force the
severity of the income inequality still prevailing in Chile, and the govern-
ment launched a Presidential Advisory Council for Labor and Equity.
(i) Poverty
In 1987, 45.1% of the Chilean population was living in poverty according
to the CASEN survey.8 Subsequent measurements by CASEN, always main-
taining the same methodology and keeping a constant real height of the
poverty line, are summarized in Table VII.3.9 It shows the steady progress
made in this area, the percentage of poor people having fallen to 21.7% of
population in 1998, and to 13.7% in 2006.10 The percentage of extreme poor
diminished from 17% in 1987 to 3% in 2006.
Poverty is a dynamic concept: some households may have risen above the
poverty line, but be close to it, and a minor income deterioration can throw
them under the line again; or may have been employed at the time of the
survey, which refers to one month only (not to the year average), though in
many other months they were unemployed, given the precariousness of labor
markets. Table VII.4 shows that a high number of people move around the
poverty line, even if only three (monthly) points of time are considered in a
very revealing panel survey: in 1996, 2001 and 2006. Of the 23.5% of poor
in 1996, 12.2% were not poor in 2001, but in this year 7.4% fell into poverty,
completing 18.7% of population in 2001. Considering the third point in
time, only 4.4% appeared as poor in the three instances (a figure close to that
of indigents), while 29.8% were in poverty once or twice. Consequently, in
at least one of the three surveys (concentrated in the months of November),
over 34% of the population was in this situation; evidently, many more suf-
fered poverty in the course of the decade between 1996 and 2006. Those that
never appeared as poor in the panel survey averaged 14.3 years of education,
8
Previous studies with a rather similar poverty indicator but methodologies that are
not fully comparable reveal that 20% of the population was poor in 1970 (Altimir,
1982; ECLAC, 1991). It is evident that the share of poor people in 1970 was lower
than in 1987, even if the 20% figure for 1970 were highly underestimated with respect
to a percentage resulting from the present criteria of measurement.
9
Expenditure on food was adjusted, in the official estimate, by the increase in the
CPI of food and then was multiplied by 2. In 1987 (a year of record concentration
of income in the richest quintile), the government chose quintile III as the reference
bracket for the basket of food of the poverty line. This quintile recorded in that year
a factor close to 2 between food and total expenditure. The overall CPI shows that
between 1987 and 2006, the price index of non-food increased 12.7% more than
that of food; if that percentage were representative of the poor, the coefficient 2
should have been raised to 2.13 (that is 6.3%) in order to keep constant the level of
the expenditure basket of the poor (instead of that of food). The Fundación para la
Superación de la Pobreza (FSP) estimated, based on CASEN 2003, that the coefficient
for the poor should be raised to 2.2.
10
Historically, the incidence of poverty in the rural sector has been greater than in
urban areas, in spite of a lower absolute poverty line. The rural poverty line is lower
on the assumption that they benefit from self-provision and lower food prices. That
situation has reverted in the 2006 survey, since poverty reached 12.3% of the rural
population, whereas for the total it was 13.7%.
186
Table VII.3 Poverty evolution in Chile, 1987–2006 (%)

1987 1990 1992 1994 1996 1998 2000 2003 2006
Population
Indigents 17.4 12.9 8.8 7.6 5.8 5.6 5.7 4.7 3.2
Other poor 27.7 25.7 23.8 19.9 17.4 16.1 14.9 14.1 10.5
Total poverty 45.1 38.6 32.6 27.5 23.3 21.7 20.6 18.7 13.7
Households
Indigents 13.5 10.6 7.2 6.2 4.9 4.7 4.6 3.9 2.7
Other poor 24.5 22.7 20.5 17.0 14.8 13.1 12.0 11.5 8.5
Total poverty 38.0 33.3 27.7 23.2 19.7 17.8 16.6 15.4 11.3
Source: MIDEPLAN, national data from the CASEN surveys.

Table VII.4 Poverty dynamics in Chile, 1996–2001–2006

(% of population)
1996 2001 2006
Poor Non-poor Total
Poor (11.3) Poor (11.3) 4.4 7.0 11.3

Poor (12.2) Non-poor 1.6 10.6 12.2
Non-poor Poor (7.4) 1.5 5.8 7.4
Non-poor Non-poor 3.2 65.9 69.1
Total 10.7 89.3 100.0
Source: Based on CASEN Panel Surveys 1996–2001–2006, covering

Santiago and three other regions that represent about 60% of the
total national population (MIDEPLAN, 2008). In the National CASEN
2006 survey, the poor population was 13.7% for all Chile, while in the
regions covered by the Panel Survey it was 13.3% as compared to the
10.7% recorded in the Panel Survey.
while the remaining 34.1% averaged only 10.3 years of education (well below
the threshold of 12 years.
There is another relevant issue in a country whose income per capita
has doubled since the definition of the poverty line in the 1980s. Poverty
is not an absolute concept.11 The “poverty line” should not be identical in
an economy with US$40,000 per capita and in one with US$1,000. Several
specialists recommend a relative concept of poverty; for instance, a moving
line that is a given percentage of the average wage or of national income.
Naturally, a shift in this direction converges with measures of income dis-
tribution and hides the positive effects of economic growth on social
welfare.
In the specific case of Chile, which has achieved such a substantive aver-
age income improvement in the past two decades, the policy target should
be adjusted in the fight against poverty. That implies raising the height of
the poverty line so that the basic food needs are satisfied with higher quality
goods and a larger share of non-foodstuff.12
In parallel, despite the progress obtained in poverty reduction, solving
the unacceptable inequality remains a pending challenge for the Chilean
economy.
11
The sensitivity of the number of poor is notably intensive with respect to the height
of the line. Estimates by Fundación para la Superación de la Pobreza (FSP, 2005) show
that increasing the line by 11.4% augments the number of poor from 18.7 to 22.8%
in the CASEN 2003.
12
The Ministry of Planning (MIDEPLAN) is in the process of revising the poverty line
for the CASEN 2009, probably ending with a new higher line.
(ii) What has happened with income distribution?

The evolution since the return to democracy shows an improvement;
according to several sources, progress was significant in the earlier 1990s,
with subsequent reversals in the recessive 1998–2003 period, and an
improvement in the late 2000s. In any case, while the information available
generally points to a significant improvement compared to the 1980s, it is
clear that distribution is still very regressive. A major national effort will
be required, then, to correct this sharp inequity.
The most comprehensive source of information on expenditure distribu-
tion is the EPF of INE, for Santiago; it is a survey of wide coverage, collected
systematically for twelve months, with annotations in a passbook and
checkups. The last fully published survey was carried out in 1996–7 (INE,
1999). Due to methodological differences, these results are not comparable
with the series of the three previous surveys reported in Table VII.2, in par-
ticular with respect to the treatment of the rent of housing.13
The omission of imputed rent in an evaluation of income or expenditure
distribution represented a severe bias for the estimation of progress in the
well-being of poor households, which were increasingly becoming home-
owners, as a consequence of intensive government programs. Estimates of
imputed rent are recorded from the 1988 survey on. The progress in hous-
ing programs achieved by Chile has had a significant effect on quintile I, by
raising its share of total expenditure by 1.3 percentage points in the 1997
survey and 0.5 points in 1988, as compared to data without imputed rent
(see Feres, 2001a).
Demography has also played an important role. Per capita data (it would
be even more accurate to work with “an equivalent adult”)14 turn out to be
more precise than figures per household, since a “representative household”
of a given quintile in 1988 is not comparable with one in 1997. In fact, the
average number of members of households fell in all quintiles between 1988
and 1997, with a 6% drop in the total average (from 4.09 to 3.84 persons).15
This change has been stronger in quintile I than in quintile V. It implies
that the fight against inequality presents, rightly so, a greater improvement
with quintiles ordered by per capita expenditure and income rather than
household income.
Table VII.5 exhibits the distribution of per capita expenditure (cols 1 and 2)
and income (cols 3 and 4) for “comparable” quintiles of households in 1988
and 1997. It shows an overall improvement in 1997, when the poorest quintile
13
These three surveys rank households by paid expenditure, excluding imputed hous-
ing rent. The 1997 survey records acquired expenditure and generated disaggregated
data on imputed housing rent.
14
It would seem advisable to adjust for the number of household members, age, and
economies of scale (see Contreras and Ruiz-Tagle V., 1997).
15
The average fell further to 3.55 members per household in the 2007 EPF.
Table VII.5 Distribution of household expenditure and income in

Santiago, with imputed rent for homeowners, 1988 and 1997 (%)
Expenditure distribution Income distribution
Quintile 1988 1997 1988 1997

(1) (2) (3) (4)
I 6.4 6.7 4.8 6.3

II 10.2 10.4 8.6 10.0
III 13.6 14.2 11.8 13.6
IV 19.8 20.4 18.6 19.5
V 50.0 48.2 56.2 50.6
Total 100.0 100.0 100.0 100.0
GINI 0.39 0.37 0.45 0.39

QV/QI 7.8 7.3 11.8 8.0
Source: Adapted from Feres (2001a), based on tabulations of the fourth and
fifth Household Budget Surveys for Greater Santiago. Data ordered by per
capita figures of the household.
raised its participation between 0.3 and 1.5 points of GDP, and the GINI
improved by between 3% and 13% with respect to the 1988 survey. It also
shows that income inequality is more intense than that in expenditure, but
precise differences are unavailable given the larger weakness of the quality
of income figures.
In brief, the information from the EPF for 1988 and 1997 shows, on the
one hand, that extreme care must be taken in analyzing the distribution,
since the same surveys processed with different criteria can show very differ-
ent results.16 On the other hand, the EPF shows a reduction in both expendi-
ture and income concentration in the nineties.17 These improvements are
reconfirmed by provisional figures from the 2007 EPF.
The results provided by income data from the University of Chile employ-
ment survey for Santiago are consistent with those from the INE household
expenditure survey. Figure VII.3 presents the ratio between quintiles I and V.
It shows a statistically significant improvement since 1990 (an average of
15.7/1 in 1990–2009) compared to the 1970s and 1980s (an average of
17.7/1 in 1974–89, with close to 20/1 in the 1980s). Nonetheless, despite
that improvement, the data indicates that today income inequality is still
16
For example, the GINI calculated according to different criteria of ordering and
disaggregation changes by as much as twelve points. Feres (2001a) presents twelve
alternative forms of organization of data.
17
The comparison of income with expenditure of the same survey of 1997 shows
indebtedness in deciles 1 to 7 and savings in deciles 8 to 10 (INE, 1999).
24
22 19.9
20
18 16.3
15.1
16 14.4
13.7
12.9
14 12.3
12
10
GINI = 48.9 46.6 51.9 56.6 50.9 53.7 51.8
8
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
Figure VII.3 Income distribution in Greater Santiago, 1960–2009 (Q5/Q1 ratio, three-
year moving average).
Source: Based on the University of Chile Employment Survey, processed by Larrañaga (2001), and
revisions and updates by the Center of Micro Datos, Department of Economics, University of
Chile, January 2010. The data are ordered by income per capita of households. The numbers on
the horizontal lines correspond to the average of the ratios richer quintile/poorer quintile in each
subperiod under each line. Figures at the bottom are the averages of the GINI coefficients in the
same subperiods.
worse than it was in the 1960s (12.9/1 in the 1960s).18 Similar conclusions
arise from GINI coefficients.
This survey is annual, so that it allows the identification of changes in
shorter periods than a full decade. After improvements in the early years of
the return to democracy (to a ratio of 13.7/1 in 1992–5), a trend reversal is
recorded in the University of Chile employment survey for Santiago, with
a partial worsening in the second half of the nineties (an average of 16.3/1
in 1996–2002, and recovering to 14.4/1 in 2003–9). Thus, progress in distri-
bution took place mainly in the early years, when the reforms of the reforms
were carried out; these injected a dose of equity into the regressive neolib-
eral inheritance. An expression of it was the sustainable increase in both
real average wages and minimum wages in the first period, as well as rising
employment. The average wage rose 4.2% in 1991–8 (4.8% in 1991–6). In
the second period, in 1999–2009 annual rises averaged only 1.9%. In par-
allel, unemployment figures were already showing some deterioration in
18
These figures are based on a classification by household income per capita
(Larrañaga, 2001). Ruiz-Tagle V. (1999) provides data for total family income and
income per capita adjusted by economies of scale and per adult equivalent (see also
World Bank, 1997). The ranking for the averages of the sub-periods into which we
have grouped the information is similar in the three categories. The signs of changes
are also similar if the GINI coefficient is used. Nonetheless, the observations (which
refer to June of each year) suffer from a great deal of “noise”; that is why I use moving
averages of three years in Figure IX.3.
1998, under the contagion of the Asian crisis, which intensified with the
recessive gap in 1999–2002, but recovered thereafter.19
The best known indicator, the CASEN survey, shows an increase in the
share of monetary income received by the poorest quintile of households
between 1987 and 1992 and a decrease in the share of the richest quintile.
From that time onward, however, the survey depicts ups and downs. But,
still in 1996, all available data, both original and adjusted, ranked both by
household and by per capita income, show an improvement in the poorest
40% of households (and, in most cases, in quintile I) and a reduced share
in the richest 20% of households. That year, the GINI index had improved
in all cases with respect to 1987 (see Feres, 2001a). In 1998, when the long-
lasting recessive adjustment began – it was already evident in November, the
month when the survey was applied – a worsening is observed with respect
to the 1996 survey, which persists in the 2000 survey (nonetheless, these
changes have low statistical significance). The 2003 CASEN survey shows
some distributive improvements, capturing the start of economic recovery;
that trend was accentuated during the following years. Consequently, in the
2006 CASEN, the ratio between the richest and the poorest quintile dimin-
ished to 13.1, from 14.4 in 2000 (the GINI improved from 0.58 to 0.54 in
the same years).20
Finally, there has been a significant increase in “investment in people”
since 1990. The effects take time to make it felt. Consequently, it is interest-
ing to observe what happens to distribution when the monetary transferences
and the free services provided by the state (whose benefits emerge in the long
term) are added to people’s monetary incomes.21 As can be seen above, in
Table VII.1, there has been a substantial rise in per capita public spending
on education and health care, aimed essentially at the poorest quintiles.
Consequently, when corrected for non-monetary items, income distribution
improves considerably, bringing down the gap between the richest and poor-
est quintile from 13.1to 7.1 for the CASEN survey of 2006 (considering the
19
As against this outcome, there were substantial improvements in pensions and the
minimum wage in 1998–2000, which reflected the public concern over social issues,
but also the frustrated expectation that GDP would keep growing at around the 7%
plateau.
20
Figures based on household per capita autonomous income; that is, income from
labor and capital.
21
The monetary subsidies are transfers mostly targeted to low-income people. The
main programs corresponded to the family allowance, which is paid to low-wage
earners according to number of dependents; the welfare pension (PASIS), paid to poor
senior people or invalids; and the unique family subsidy (SUF), which is a family
assignment for poor persons without social safety. Other monetary transfers include
the subsidy for consumption of drinkable water (SAP), a monetary payment to the ben-
eficiaries of Chile Solidario, and a public complement of the pensions paid by the AFP
to guarantee a minimum level.
monetary transferences such as pensions of non-contributory seniors, family

allowances, and unemployment insurance). As social spending rises, a key
variable is whether it actually leads to an increase in the volume and/or qual-
ity of services or whether the same services are merely provided at greater
cost to the State.22 The actual volume has evidently improved; in the case
of education, the number of years studied has increased sharply as reported
here; quality needs to be improved sharply in education and in the services
provided by the National Health Service. To ensure that spending produces
results, effective pressure needs to be applied to achieve higher productivity
and a better service to beneficiaries.
(iii) The social situation and public policies

The return to democracy brought with it greater attention on the part of the
state to equity and poverty issues. Consequently, in the 1990s the authori-
ties introduced reforms to the reforms. Public spending was restructured to
increase funding in social areas, in parallel with a fiscal reform that raised
progressive taxation and the VAT rate. In the labor market, reforms led to a
substantial improvement in the minimum wage and reduced the imbalance
of power between workers and employers. Furthermore, innovative targeted
social programs were carried out, such as priority for basic schools in poor
zones and young labor training in the early 1990s (Raczynski, 1996); and,
in the present decade, the ambitious, more comprehensive, Chile Solidario,
which directly benefited the poorest households, the gradual health reform
AUGE, and a notable reform of social security.
Additionally, significant reforms concerning macroeconomic manage-
ment in the first half of the nineties had major positive effects on productive
employment and wages. The introduction of the unremunerated reserve
requirement and other prudential mechanisms for regulating volatile capital
inflows played a key role in achieving sustainability of macroeconomic bal-
ances and the 7% average GDP growth (see Chapter VIII).
The result of this set of policies was a substantial growth in the real aver-
age wage, which in 1998 was 41% higher than in 1989, while the minimum
wage was 63% above that in 1989. The recessive adjustment initiated in
1999 weakened the growth rate of wages; nevertheless, in 2009 the average
real wage exceeded by 74% that of 1989, while the minimum wage had
grown 137% in the same period. The family allowance had multiplied by
2.6, which meant recovering nearly 90% of the fall recorded in the eighties
22
A substantial share of the increase in social spending in the 1990s was directed
to raise the wages of teachers and national health system personnel. In 1990 their
salaries were extremely out of line with the market and below the minimum required
for efficient functioning. Unfortunately, while the quality of services fell as a result
of the decline in wages and in the social status of these public servants in the 1980s,
improved incomes are not automatically followed by a recovery in quality.
(see Table VII.1). Additionally, as said, per capita social expenditure in

education and health multiplied by 3.6 in 1990–2008 (7% annual average
growth), strongly directed to recovering the wage levels notably reduced in
the eighties. The unemployment rate also improved considerably, averaging
8.7% in 1990–2009 compared to 18% in 1974–89 (see Table I.1 and Figure
VII.2). Naturally, the macroeconomic conjuncture has had a significant
impact on employment. This is revealed in the rate of 6.1% of open unem-
ployment during the boom year 1997 compared to 10.5% in the depressed
1999–2003 period, or the extremely high rate of 31.3% recorded in 1983
during the debt crisis compared to 7.9% at the 1989 peak.
There are two positive trends in the labor market that must be underlined.
One is the rising share of workers in formal segments, as reflected by the
number contributing to social security. Democracy inherited a low share of
workers contributing to the private social security scheme created by the
dictatorship in 1981; the majority of the market was informal, with only
41% of the labor force contributing monthly. A rising trend from 1990 led to
46% contributing by the mid-1990s; then, there was a standstill until 2003,
broken thereafter in parallel to the recovery of economic activity and a legal
reform enforcing some formalization of the labor market,23 with the share
climbing to 54% in 2008. The other outstanding trend is the increase in the
participation of women in the labor force, as documented below.
Social policies, together with more efficient macroeconomic management –
which in the early 1990s laid the foundations for faster growth and large-scale
job creation – enabled a drastic reduction of poverty and indigence, and some
modest improvements in the still unequal income distribution.
At the beginning of this chapter, I noted the dissatisfaction that was felt
regarding the macrosocial imbalances prevailing around 1970. Although
income distribution has improved during these two decades, inequality
continues to be a quite negative feature of the Chilean economy. More con-
sistent, vigorous, and effective efforts need to be made.
2 The role of neoliberal reforms
The economic reforms applied in Chile in the 1970s and 1980s have had
a significant incidence in the social field. There were large, direct, negative
effects on various social indicators owing to the enforcement of a model
that focused on the neutrality of policies. The failure to take account of
the structural heterogeneity of agents, combined with significant market
failures and segmentation, translated into costly adjustment processes and
severe recessions, generating a background of low productive investment
23
President Lagos enacted legislation demanding large firms to take responsibility for
the fulfilment of labor contracts and payment of social security by their sub-contractors
(Ley de Sub-contratación).
and high unemployment (see Chapter I). Thus, indirectly these policies had
a negative impact on an unprotected population, and underlie the sharp
worsening taking place between 1973 and 1989. Naturally, the regressive
effects were compounded by the debt crisis. But, as shown in Chapter III, the
severe social and economic effects of the crisis were not totally independent
of the model adopted by the dictatorship.
As far as the reforms are concerned, one of the greatest changes was in the
fiscal sphere. Reforms to the tax system, in 1975, included the abolition of
wealth and capital gains taxes and a substantial reduction in the burden on
profits. On the other hand, the adoption of a value added tax was completed
and existing exemptions for basic consumer goods were abolished.24 The offi-
cial objective of these changes was to reduce the tax burden and concentrate
it on taxes that were “neutral and efficient.”
In 1984 a second sizable tax reform took place (see Marfán, 1998). The
income tax was drastically changed. A 40% additional tax on profits was
removed, while a 10% general tax on profits was retained; additionally, the
latter tax paid by the firm became a credit to the personal progressive income
tax of shareholders. The marginal rates of the income tax were sharply
reduced and the income brackets were enlarged. Shortly after, extremely
generous tax privileges were created for purchases of stock of firms being
privatized. The progressiveness of the tax system was notably weakened
(or regressiveness increased). Finally, before the plebiscite of 1988 VAT was
reduced from 20 to 16%, taxes on several luxury goods were diminished or
suppressed, and in 1989 the 10% tax on profits was eliminated for reinvested
profits. Thus, in practice, taxation on profits was gone, since distributed prof-
its benefited from the full tax credit already mentioned. The tax burden fell
to 16.5% of GDP in 1989. An evidently transitory high price of copper, col-
lected by CODELCO (the copper-producing public firm), filled the vacuum
left by reduced tax revenue. The regressive tax reforms were accompanied by
a progressive effort to fight tax evasion: equal tax collected for equal income
is an evidently equitable principle.
Public spending, as a percentage of GDP, was reduced by more than a
quarter, in comparison with the late 1960s, after having spiraled out of
control in 1972–3. There was a dramatic fall in public investment, which
declined by more than one-half between 1970 and 1979 as a share of
GDP. Support for private productive activities, in the form of subsidies
and infrastructure, also fell. Per capita social spending in 1981 was lower
than it had been in 1970, exhibiting a further drop with the debt crisis. By
24
There is no doubt that the replacement of progressive taxes with the VAT was regres-
sive in itself, and it was accompanied by regressive exemptions from this tax (such
as to the building industry) and by drops in social spending. This should not lead to
overlook the efficiency and high yield of VAT or the fact that a rate rise directed to
increasing focused social spending is clearly progressive.
1990, the democratic regime had inherited a level 18% below that of 1970
(see Table VII.1).
The privatization of many of the means of production owned by the state
took place from the mid-1970s. The process was conducted against the back-
ground of a domestic recession and extremely high domestic interest rates.
For this reason, only a few private groups were able to buy the privatized
enterprises, mainly those groups that had greater access to external credit;
hence, further strengthening the concentration of wealth and power (see
Dahse, 1979; Devlin and Cominetti, 1994; Monckeberg, 2001). Interestingly,
in spite of a quite liberal treatment of FDI, it was rather absent in Chile dur-
ing the 1970s.
Chapter III analyzed the way in which the reforms in the financial market,
far from enhancing productive investment, were characterized by extremely
high real interest rates and a great deal of financial activity, but contributed
none or little to the generation of new productive capacity and were asso-
ciated with crowded-out national savings. Increasing financial inflows led
the economy through a path of unsustainable current expenditure, which
ended in the deep crisis of 1982. Then, the state had to intervene in the
financial system to avoid a deeper collapse; the high cost of subsidies to
the financial system and to borrowers contributed to a markedly regressive
redistribution of wealth, involving reductions in social spending and public
sector investment in the 1980s and the long-lasting recession in the years
following the crisis (see section 3d). Recall that only during 1988 was the
1981 level of GDP per capita restored.
Labor legislation also underwent major changes that had a negative impact
on workers: the level and coverage of the minimum wage were reduced, the
dismissal of workers was made easier, and labor tribunals were abolished
(being restored in 1986). The unions were suspended in September 1973;
later, in 1979, they were authorized but with limited powers that excluded
collective negotiation inter-firm, restricted the rights of union leaders,
and fomented the segmentation of labor unions (Cortázar, 1983; Mizala and
Romaguera, 2001). In combination with political repression and economic
depression, the legislation was effective in reducing the power of social
organizations and their ability to defend their rights (Tokman, 2004).
In the early 1980s, deep reforms in the architecture of social security were
introduced, affecting the health and pension systems. They had a major
impact on the fiscal budget, and distributive effects at the time the reforms
were implemented.
The structural changes to health care culminated in the creation of a dual
system, with a public component acting through a National Common Fund
(the Fondo Nacional de Salud, FONASA) and a private component consist-
ing of health insurance companies (ISAPRES). By then, to a large extent,
health care was being financed by a payroll tax. The reform initially meant
that 11% of the beneficiaries moved to the new ISAPRES system with 48%
of the revenue from health contributions (Titelman, 2001). Regardless of

the quality of the reform, this effect evidently constituted a regressive form
of “targeting,” and it contributed to deepen the crisis in the public health
system. Most people in the four poorest quintiles are covered by the public
health system, and only in the richest quintile do ISAPREs cover a higher
proportion of people than FONASA.25 In 2007, only 17% of the population
was covered by health insurance with ISAPREs.
Private pension fund management companies (the Administradoras de
Fondos Previsionales, AFPs) began to operate in 1981, in what marked the
transition from a pay-as-you-go system (which remained alive for a reduced
and decreasing number of people) to an individual capitalization one. There
are varied consequences for distribution. On the one hand, the higher pen-
sions became conditional on a larger present value of contributions, which
involved clearly progressive effects. On the other hand, there are three nega-
tive features we emphasize here. First, the reform led to a decline in public
revenue as revenue was transferred to the AFPs from 1981. As the public
sector was left with the responsibility for financing existing pensions and
the pensions of those due to retire in the coming years, the public social
security deficit rose from 2% of GDP in 1980 to 7% in 1983–6 (Uthoff,
2001). In a time of recession, this increased the strain on the fiscal accounts
and was one more factor behind the spectacular cuts in social spending,
particularly on education and health, between 1980 (before the social secu-
rity reform was implemented) and 1987.
Second, the reform did not succeed in including lower income informal
or self-employed workers. Whereas 60% of workers were contributing to the
social security system in 1974 (some 79% were affiliated), in 1988 just 55%
of the working people were contributing to the private and public systems;
in between, the rate had fallen, spectacularly, to a point as low as 40% in
1982 (Arellano, 1989). In turn, only 41% of the total labor force (including
employed and jobless) were contributing in the private system in 1989.
Third, another significant distributive implication is the enormous con-
centration of diverse expressions of power (see Molina, 2006). In this respect,
we have to mention the concentration of power in the hands of AFP owners,
who use the funds of workers to buy shares and thus acquire a strong say
in appointing directors in the boards of corporations, with no voice for the
actual owners of the funds.
Finally, the trade reforms introduced since the mid-1970s have proved to
be a key factor in explaining factorial income behavior. On the one hand,
25
Another important point, with regard to distribution, is the risk of discrimination in
the private system in the way in which the insurance premium is linked to the health
risk of the person concerned. It discriminates greatly against the elderly, children,
women of fertile age, and in general those who are most in need of health services. It
is noteworthy, for example, that for those aged sixty-five or over, even in the richest
quintile most people, some 56%, are treated by the public sector (Titelman, 2001).
they had a great impact on the production structure of the country (see
Chapter II), which led to a significant relative decline (and often an absolute
one) in employment in some sectors (this was particularly steep in manu-
facturing), accompanied by a positive but weaker dynamism in expanding
sectors (Valdés, 1992); with it, a clear negative net effect on employment,
productive investment, and activity level was recorded during the deep
adjustment to import liberalization. This net negative balance was strength-
ened by the exchange rate appreciation of 1979–82 and by the pro-cyclical
bias of macroeconomic policies.
Since the economy was opened up, there has been a marked increase in
the premium to higher education, with worsening wage income distribution.
The forces behind this change in relative prices are rooted in the relative
decline in the demand for unskilled labor accelerated by abrupt import lib-
eralization and low productive investment, acting in tandem with greater
human capital requirements in the economy following the reforms. But this
tendency towards a growing wage gap would appear to be, partly, the result
of an exogenous increase in demand for qualified workers (this trend is asso-
ciated with the direction of global technological change and is transmitted
via growing trade links with the rest of the world, where wage inequality is
also on the rise), vis-à-vis a rather inflexible composition of the supply of
human capital. At the same time, the productive structure in the 1980s was
dominated by a natural resource intensive export sector using little unskilled
labor, while the rest of the economy was mostly depressed, accentuating the
inequality of income distribution.26 In 1990–8, the vigorous export expan-
sion was accompanied by a 6.5% annual rate of growth of the rest of the
economy, thus pulling up the demand for labor, which contrasts with a
negligible 1.7% in 1982–9 (see Chapter VI).
The concentration of human capital investment opportunities was quite
intense: (i) only a minority of the labor force, rising but slowly, attained
more than twelve years of schooling (which was then the threshold denot-
ing a break in the yield curve to school years); (ii) there were significant
gaps in the quality of education between that for students from the richest
quintile and that for the rest; (iii) labor training efforts were extremely weak
and biased towards trained people;27 and (iv) the system of higher and pro-
fessional education became even more regressive when public funding for
the universities was cut back in the 1980s and public institutions yielded in
26
Like other countries in Latin America, Chile has comparative advantages in the pro-
duction of natural–resource–intensive goods. The use of capital-intensive processes
to produce them entails a worsening of wage and income distribution, in contrast to
what would happen in other developing countries that specialize in the production
of labor-intensive goods, such as the Asian ones (Fischer, 1999).
27
Labor training can contribute significantly to flexibilize the supply of labor. There
has been some mild progress in this field since the 1990s. The percentage of the labor
force trained under the tax-exempt National Training and Employment Service (SENCE)
importance to private ones. All these contributed to perpetuate historical

inequities across generations.28
3 Reforms to fight poverty and income concentration
The debate over which variables best determine poverty, distribution pat-
terns and development was reactivated in recent years (see MIDEPLAN,
2002a; FSP, 2005). The analysis of the preceding sections demonstrates that
there have been remarkable changes in income distribution and poverty. We
have stressed the significant role of socioeconomic reforms – corrections and
stepbacks – over the past third of a century.
Here I mention four strong variables determining poverty and income
distribution in which there have been progress in recent years, progress that
needs further enhancement. Then, challenges in two development policy
areas that are crucial for growth with equity are discussed.
(a) Factors underlying income distribution

Evidently, different factors are interrelated, and in turn, they are a result of
socioeconomic structures and of public policies. Two of these variables are of
structural character and can be improved only gradually. The three following
ones – unemployment, social expenditure and macroeconomic environment –
affect both the structure and the conjuncture and can experience sharp
changes in the short run.
First, income levels have a highly positive relationship with years of
schooling. In particular, there is a notable wage gap between people who
completed university and those who only completed secondary school.
A worker with a complete university education earns, on average, more
than three times per hour what another with full secondary education earns
(Mizala and Romaguera, 2004). Additionally, the returns to tertiary education
have been increasing over time. Though the coverage of tertiary education
grew from 16 to 37% (2.3 times) between 1990 and 2003, the differences by
quintiles continued to be notable; for the first quintile it increased from 4.4
to 14.5%, whereas for the fifth quintile it rose from 40.2 to 73.7%. In general,
scheme increased from 4% in 1990 to nearly 8% in 1998. Nonetheless, only 20% of

companies made full use of this benefit and the distribution of their spending was
highly regressive (Benavente and Crespi, 1998). The problem remains, as stated by the
Consejo Asesor Presidencial de Trabajo y Equidad (2008). In fact, 1% of firms captured
42% of the resources, with 80% of the public subsidy being received by large firms; addi-
tionally, nearly four as many workers in the highest quintile as in the poorest received
some training (10.8 and 2.9%, respectively) according to CASEN 2006. The public pro-
gram requires a sharp reshuffle, with a significant increase in the funds allotted.
28
Núñez and Gutiérrez (2004) develop deep research on the persistence in the cor-
relation between the economic level of parents and the probable income level of their
children.
as a result of the broader coverage in all educational levels, the average years
of schooling of the population aged eighteen years or more increased from
9 to 10.2 years between 1990 and 2006 (CASEN 1990, 2006).29
This relationship is subject to two qualifications, which have deep impli-
cations for public policy. On the one hand, high-quality education and the
matching of supply of and demand for skills are essential. This is illustrated
by the fact that, although the average worker had 3.5 years more schooling
in 1992 than in 1970 (Hofman, 1999), the average wage was similar, having
been depressed in the intermediate years. On the other hand, schooling is
measured on the basis of the number of years of traditional education, with-
out taking into account the training accrued during people’s working lives.
Training is essential as a way of enhancing the productivity of workers with
few years of education or schooling whose quality does not match the cur-
rent demand for labor. Labor training makes it possible to partly compensate
the international trend towards higher wage inequality, principally associated
with the bias of technological change and educational differences (Beyer,
1997; Bravo and Contreras, 1999; Larrañaga, 2006).
Second, increasing participation of women in the labor force is a key
factor for reducing the number of households in poverty. According to the
Census of 1992, 28% of total workers were women; this percentage rose to
38% in the 2002 Census, and to 43% in the 2006 CASEN. That is significant
progress, but still quite short of the 73% employment rate for men. The
gender inequality is particularly acute among poorer women. According
to CASEN 2003, only 25% of women in the poorest quintile were in the
labor force, a percentage that doubles for the richest quintile (52%), with an
almost linear progression in the intermediate brackets. CASEN 2006 reports
that among women head of households, 49% of those in the lower quintile
were in the labor force, rising to 73% in the richest quintile.
What has happened is that progress in the participation of women has
been located in the higher income brackets.30 In fact, in households with a
head and couple, in the lowest quintile, 12% were both employed in 2006
as compared with 5% in 1990, while in the highest quintile the figures were
61% and 43%, respectively; that is, an improvement of 18 points in the
richer people versus 7 points in the poorer. Improving opportunities and
facilities (such as day nurseries and pre-school for children, and part-time
jobs) for working women with lower incomes, as well as the elimination of
the wage gaps between men and women with similar qualifications, are key
factors for increasing gender and overall equity.
29
According to the CASEN surveys, the coverage of primary education increased from
96.8 to 99.1% between 1990 and 2003, whereas that of secondary education rose from
80.3 to 92.8%.
30
Paradoxically, labor income distribution can be worsened by the increase in wom-
en’s participation, if the increase is focused on women who are from high income
brackets, more educated, and with a lower fecundity rate (see Larrañaga, 2001).
Third, unemployment is another very influential factor. In the CASEN

surveys from 1987 to 2006, the unemployment rate averaged nine times as
high in the first quintile as in the fifth, and its sensitivity to the economic
cycle, and even to slight fluctuations such as those of the late 1990s, is very
intense. In the first quintile, unemployment fell from 24% in 1987 to 15%
in 1992, as a consequence of improving macroeconomic balances and eco-
nomic growth. With the arrival of the Asian crisis it worsened from 16% in
1996 to 26% in 2000, evolving again to 16% in 2006. Unemployment is also
substantially higher among young people and those with less schooling.
CASEN 2006 shows that in the lower quintile, with deficient or limited edu-
cation, close to 30% of youngsters (15 to 24 years of age) were not working,
studying, or searching for a job; the percentage of women was twice that
of men (see Jélvez and Alvarado, 2009). Consequently, policies intended to
strengthen the demand for labor, to reduce the gender gaps, and to make
the supply more flexible and better able to adapt to technological changes –
with sustainable macroeconomic stability, vigorous physical capital forma-
tion, and increasing investment in people – play a very significant role in
improving the distribution of opportunities.
A palliative to the unemployment effects is the increasing coverage of a
new insurance for private workers. Previously, there was only an unemploy-
ment subsidy of a negligible amount, and a severance payment of one month
per year of work (with an eleven-year limit) with the same employer and
under some conditions of dismissal. It was mostly applicable to organized
workers with stable jobs. The new insurance scheme was implemented in
October 2002, being operated by a society of all the AFPs. In very abridged
terms, it includes two separate funds. One is financed by workers’ and
employers’ contributions, which are credited to the individual account of
the worker; the contribution of the employer (1.6% of the wage, in the case
of contracts without time limit) is credited against the severance liability of
the employer. The other is a solidarity fund, which is financed by the govern-
ment (with a given annual lump sum) and contributions by the employer
(0.8% of the assessable wage). All wage earners who sign a new contract
(with some exceptions, such as public officials) are part of the new system,
and also those who apply voluntarily for the insurance (very few, actually).
In September 2009, after seven years in force, it had 6.2 million registered
members (afiliados, workers who at some point in time have passed through
the new system), and among them there were 3.2 million active contribu-
tors (about 85% of AFP active contributors); this implies that fewer than one
in five waged workers had not left or lost their job in the seven-year period.
The mechanism started to operate in a relatively high unemployment period,
mostly as self-insurance.
In an economy averaging about 600,000 unemployed workers in the over
seven years in which the insurance has been in force, a large number of
workers have made use of the scheme; 4.8 million benefit payments have
been delivered. The huge number of operations reflected the precariousness

or actual flexibility of the labor market. But the benefit received, covering
an average of 2.5 months per request, has been quite small and has been
almost completely self-financed by their own contributions while they were
employed. In turn, up to 2008, only 3% of the benefits paid included a com-
plement from the solidarity fund. In all, notwithstanding that the insurance
was created in a recessive period, the two funds had accumulated a stock of
US$2.1 billion. Actually, the scheme had behaved pro-cyclically, given that
it accumulated funds during recessive years.
This unemployment insurance, overall a relevant progress, needed
strengthening or deep corrections in: (i) the magnitude of benefits; (ii) the
access to benefits; (iii) the extent of the solidarity fund; (iv) its macroeco-
nomic impact; and (v) the connection with effective training programs, with
the solidarity benefit conditional to participating effectively in the training.
In 2009 the parliament approved a government project that enhances the
extent of and access to the solidarity fund, giving access to workers with
short-term contracts, introduces counter-cyclical features in the benefits,
and seeks to improve job search intermediation for the unemployed and to
provide access to labor training for them.
Fourth, social expenditure has had a progressive incidence, representing a
rising share of the access of the poor to goods and services since the 1990s.
A small share was made up of monetary transfers, while the great majority
corresponds to free or subsidized goods and services delivered to the poorest.
However, recently, there have been sizable programs in progress that imply
significant flows of monetary transfers. In fact, besides the improvements
made to the unemployment insurance, there has been a deep reform of the
pensions system since 2008. It changes the balance of the 1981 reform, from
principally a private self-insurance system to one with an ambitious solidar-
ity pillar. The reform launched a progressive process, starting with members
of households in the two poorer quintiles and rising gradually up to three
quintiles of households in 2011. It establishes a higher basic pension for
people over 64 years old since 2008; it will provide a complementary sub-
sidy, that decreases gradually up to a pension more or less equivalent to
1.5 times the minimum wage (255,000 pesos in 2011); it starts a subsidy
equivalent to the social security contributions for two years to workers up
to 35 years old, in formal jobs; it offers a subsidy for mothers when retiring
for each child they have had (including adopted children). Additionally, the
reform allows the creation, in parallel with the private AFPs, of a public AFP,
a proposal that was strongly disputed by the opposition; the government
had not made use yet of the “policy space” gained.
The reform improves significantly the welfare of senior people, with
a high financial cost to the government. The reform was designed and
enacted when the fiscal budget was in a huge surplus in 2006–7, led by the
high price of copper.
A very relevant program to fight extreme poverty is Chile Solidario.

This program seeks to improve the life conditions of the poorest 225,000
families through three specific objectives: (i) to make effective the access of
indigent families to social benefits and services available, which, because of
their marginalization, they were not benefiting from; (ii) to give psychoso-
cial support, promoting the development of productive potentialities; and
(iii) to generate minimum conditions for the most vulnerable members of
the families to improve their employment position (this last has been less
successful).31
(b) Structural factors

The regressive trends of the 1980s were not exclusive to Chile. In general,
income distribution worsened, real wages fell, and the level and quality of
employment declined throughout Latin America. Something similar hap-
pened in the United States and Great Britain in that decade, with the ratio
between the richest and poorest quintiles rising. In the United States, the
household income of the poor fell, while that of the richest 10% improved
substantially in the 1980s (Krugman, 1990).
Poverty and income distribution are determined essentially in the pro-
duction process itself.32 Consequently, it is of great relevance to ensure
that productive transformation is associated with equity, which requires a
consistent move from “neutral” policies (that actually use to favor the upper
segments) towards strongly biased policies that improve the productivity
and workability of middle- and low-income sectors of society. It is clear
that the choice is not between growth and equity. It is not simply a matter
of choosing growth, since to achieve this, and make it sustainable, is no
easy matter: Chile has succeeded only in exceptional periods, one of which
was 1990–8. Consequently, the key thing is to identify the determinants of
growth, which encompass, at the present stage of Chilean development,
crucial complementarities between the sources of growth and equity, and
between macroeconomic and macrosocial balances.
The generation of productive employment is the main channel through
which economic and social progress is transmitted. This depends, obviously,
on supply and demand, and both are affected by public policies. To create a
vigorous demand for labor, productive investment must be notably higher
than it was during the neoliberal regime. Larger capital formation makes
it possible to achieve higher growth with greater job creation and better
wages. The fact that average wages were still lower in 1989 than in 1970 is
closely associated with the low investment ratio recorded in the 1970s and
1980s. Similarly, the rising investment ratio between 1992 and 1998 helps
31
See Galasso (2006) for a preliminary evaluation of the program.
32
Bourguignon and Walton (2007) demonstrate why and how poverty is detrimental
for development.
to explain the sustained although still insufficient improvement in wages

that occurred in those years.
Enhanced capital formation does not take place in a vacuum. It works
together with technological innovation. Much of it is embodied in machin-
ery and equipment and in the capabilities of labor. This means that much
higher productive, physical and human, investment will be needed in order
to incorporate technological development and thus improve national total
factor productivity.
A high rate of technical progress requires a rising stock of capital and
flexible and increasingly highly qualified labor if excessive “technologi-
cal unemployment” is to be avoided. In particular, in order to countervail
the expanding gap between low and high wages, an increase in physical
investment and investment in people associated with SMEs is essential.
Crucial aspects are improving the skills of labor and providing better train-
ing for workers and small entrepreneurs during the course of their active
lives, whence the importance of reforming education, strengthening labor
training, and encouraging the development of SMEs. They all contribute to
increase average productivity and to reduce the inequality gaps.
There are two ways of looking at productivity. One does not control for
economic cycles or changes in the rate of utilization of capacity, issues
related to the quality of macroeconomic balances, to which we return later.
When in the course of an economic cycle output falls sharply, say by 14%, as
happened in Chile in 1982, what is really falling is the actual rate of resource
utilization, not the structural capacity of factors (TFP) or potential GDP.
That reduced actual productivity is restored once the recessionary stage of
the cycle has given way to the expansionary one, even when structural or
potential productivity (the second concept) remains unchanged. Actual
productivity is quite variable in economies with pro-cyclical macroeconom-
ics; those changes do not reflect structural changes in productivity. The
structural or potential productivity is concerned with efforts to innovate,
with new combinations of productive resources and improvements in their
quality. This second type of productivity is one of the determinants of
long-term growth. Frequently, research mixes up these two components,33
which obscures the interpretation of what has happened with the TFP in
Chile. In all, nonetheless, it appears to be evident that, after a jump in
the 1990s, it has slowed its speed and its contribution to reducing equity
gaps.
Naturally, physical investment and innovation are not enough. There is
also a need to increase and enhance the supply of human capital, given the
dynamics of innovation and technological progress. Investment in people,
one of the two components of social spending, is a factor of production.
33
See essays by several authors in Morandé and Vergara (1997) and Loayza et al.
(2005).
An even more important feature, though, is that investment in people –

particularly education, labor training, and healthcare from the first seconds
of life – prepares them to participate more effectively in the market and is
instrumental in bringing to a halt the perpetuation of poverty, whereby the
children of the poor are condemned to be poor themselves (see Núñez and
Gutiérrez, 2004). Better nutrition and education of higher quality are key
inputs for a more flexible labor supply, which contributes to an effective
adjustment to the demand requirements, within a globalized environment.
The other component is continuous redistributive spending, aimed at com-
pensating those who lose from modernization, those who cannot earn a better
living in the market, or those who have ended their working lives and have
insufficient pensions or none at all. The rights-based social policy approach
of the Bachelet government, that is a culmination of the progressive social
policies of the four presidential regimes of the Concertación Democratica, has
made significant progress in facing this human rights challenge.
(c) Macroeconomic stability, investment, and income distribution

The macroeconomic environment affects poverty and income distribu-
tion in the short term through its impact on the labor market, as stressed
throughout this book. It also affects structural evolution through the effects
on: (i) physical and human capital formation, vis-à-vis mergers and acqui-
sitions and purely financial investment; (ii) the SMEs’ situation; (iii) the
value-added in non-traditional exports; and (iv) the authorities’ capacity
to focus on the future, on the long term, instead of on short-term critical
emergencies.
Comprehensive stability – what we call real macroeconomics – is essential
to equity and economic growth. If we look at what happened to wages and
employment during recessive periods in the one-third of a century, two asym-
metries emerge: (i) the fall in labor income was disproportionately high and
informal working increased during the bust, which tends to be abrupt; and
(ii) the recovery is usually slow. With both asymmetries, given the instabil-
ity of economic activity, the level of employment and formality in the labor
markets deteriorate more intensively than GDP. Given the regressive effect
that adjustment processes normally have on low-income sectors and wage
earners, it is clear that efforts need to be made to remove the factors that
cause instability and uncertainty (Rodrik, 2001a).
The definition of stability is crucial. CPI stability is essential, but it is only
one ingredient of comprehensive stability, of which the stability of the real
economy is the most significant. That implies using productive capacity
(potential GDP) with right macroeconomic prices. This conclusion is even
more compelling when we observe that real instability also discourages pro-
ductive investment. When firms are producing below capacity and capital is
underutilized, it is obvious that there are fewer market incentives to invest
in the creation of new productive capacity. Empirical evidence shows that
one of the most common tendencies during recessive adjustment is a sharp

drop in private investment.34 Public investment is generally cut as well, and
this discourages private investment still further.
Economies with large fluctuations tend to discourage not only produc-
tive investment but also technological innovation, as instability leads both
to large losses and great opportunities for easy profit. In such periods, the
profits of some are frequently made at the expense of others (in a zero or
negative sum game). What happens, therefore, is that cyclical instability
processes tend to result in underscoring medium- and long-term structural
productivity, and to generate an environment that is prone to specula-
tive investment (which creates neither new productive employment nor
economic growth) rather than to technological innovation and productive
investment.
Environments of great real instability, like those recorded in Chile during
the 1970s and 1980s, tend to be accompanied by other phenomena whose
effects on society as a whole are regressive. In fact, in these situations, losses
tend to emerge in productive or financial sectors, with an inclination to
sustain them by public sector subsidies. One case at hand is that of the
Chilean banking system after the 1983 crisis, as discussed earlier, with a total
final cost equivalent to 35% of a year’s GDP (Sanhueza, 1999). Thus, over
the course of a few years, the equivalent of one-third of national output (or
public expenditure on education during a full decade) was transferred from
some sectors to others in order to cope with this banking crisis. Most of the
large transfers of wealth that took place in the 1970s and 1980s were only
possible in the framework of great real instability, reinforced by the regres-
sive and ideology-driven approach of the dictatorship.
It has been shown throughout this book that financial volatility has been
at the heart of real macro-instability. Most of the spectacular increase in
international capital flows up to the explosion of the present global crisis
were not tied to productive investment, but were of a speculative financial
nature. These movements are guided by short-term expectations about dif-
ferentials in interest rates, exchange rates, and stock market prices among
countries, which have brought about a dizzying increase in the speed at
which speculative capital can move from one country to another. This has
led to great instability in macroprices and economic activity. There are too
many agents devoted to the capture of capital gains rather than generating
productivity gains. This neo rent-seeking financierism appears to be one of
34
See Solimano (1990), Agosin (1998), and Ffrench-Davis (2006). Another conse-
quence has been underinvestment by domestic SMEs. Correcting this imbalance has
been slowed by downward adjustment processes, as increases in interest rates above
“normal” levels and domestic demand restriction affect such firms more severely
than large ones, which are more diversified and can obtain financing through diverse
channels.
the reasons for the weakness of productive investment and the inequality in
Latin America (see Ffrench-Davis, 2006, Chapter VI).
Equity is an outstanding feature of developed nations, which are remark-

ably more equitable than developing countries like Chile. Nevertheless, the
more successful emerging economies have improved their income distribu-
tions, conscious that poverty and inequality are negative for development.
There is a need for a close interrelation of progressive productive and social
agendas.
Undoubtedly, the net balance since the 1970s shows a highly skewed
income distribution in Chile. Nevertheless, it is a mistake to state that it
has been always as inequitable as today. There have been large changes over
time. In the sixties it was not satisfactory, but it worsened notably in the
1970s and 1980s. A significant improvement was recorded in the first half of
the nineties, owing to better economic and social policies in democracy than
in dictatorship. However, subsequently, a long-lasting recessive adjustment
took place across the economy from 1998, and despite government efforts
to increase social spending the least affluent sectors suffered most. The result
in 1999–2003 was an increase in unemployment, a slowing in the reduction
of poverty, and some worsening of income distribution. Then, the recovery
of economic activity, in 2004–7, and a noticeable increase in progressive
social reforms, partly countervailing the global crisis, brought an improved
distribution in 2004–9, but merely to the same level achieved in the first
years of return to democracy in the 1990s. In all, inequality continues to be
a demanding challenge for the Chilean democracy.
Three key variables that are behind these different outcomes are: (i) the
size and allocation of social expenditure, today restricted by a low 18% tax
burden;35 (ii) the macroeconomic environment in which investors, produc-
ers, and workers operate, i.e. the real sector of the economy; and (iii) how
account is taken of the enormous structural heterogeneity prevailing in
domestic markets and of the disadvantageous environment in which SMEs
operate. The environment and the signals received by the set of producers
have played a decisive role in the phases of increased inequality in this one-
third of a century. The deep and loud message for economic policy is that
macroeconomic responsibility is related not only to low inflation and fis-
cal discipline, but also to the recessive or expansive environment of the
domestic economy, and to key macroprices such as the exchange and inter-
est rates, which asymmetrically affect small and medium firms versus large
35
In recent years the restriction imposed by the low tax burden has been sharply
relaxed by the jump in the price of copper benefiting fiscal revenue.
corporations, traditional versus non-traditional exporters, workers with low

qualifications versus highly qualified workers. Indeed, the macroeconomic
environment makes a remarkable difference for both growth and equity. An
alternative approach to the neoliberal fashion is what we have called real
macroeconomics or macroeconomics for development versus the merely finan-
cieristic macroeconomics (excessively focused on inflation control and capital
account opening).
Equity and growth capacity also depend upon specific sectoral policies:
how trade, financial, and tax reforms are designed and executed. These three
reforms – implemented with a neoliberal approach – left regressive tracks,
which help us to understand why they were associated with low productive
investment ratios and worsened income distribution. The structural reforms
were carried out with a very regressive bias, in contrast with, for example,
those in Korea or Taiwan, where trade and financial reforms were success-
fully executed pragmatically, with greater harmony between growth and
social equity.
The challenge is to grow while distributing equity; that is, distributing
the capacity to operate efficiently in the markets, particularly for small and
medium firms and for workers with lower qualifications. It is in these sectors
that the large income and productivity gaps are located. There, also, lies the
source of the huge GDP per capita gap between Chile and richer economies.
It is in these sectors of the Chilean economy that there is broad room for
efficiency gains. Then, growth-with-equity would become attainable.
Securing structural improvements in the distribution of opportunities and
income is a long-term task. It has been addressed willingly since the return
to democracy, but with inconsistencies. Although considerable progress has
been made in reducing poverty and indigence, and there has been some
improvement in income distribution, there still is pending a major chal-
lenge for the nation. Among other macro- and mesoeconomic corrections
required, the approach that needs to be taken includes:36
1 Improving active counter-cyclical macroeconomic management to make

the economy less vulnerable to external shocks, the effects of which are
regressive; this includes rebuilding an active exchange rate policy in order
to provide greater predictability for non-traditional exports and import-
competing output.
2 Strengthening the fight against tax evasion and legal avoidance (regres-
sive loopholes), which are detrimental to fiscal equity.
3 Implementing a systematic educational reform, improving and standard-
izing educational quality, and upgrading programs and teaching staff.
4 A sharp increase in the quantity and quality of labor training, in order to
enhance the flexibility of the labor supply.
36
A more comprehensive and detailed list is developed in Chapter X.
5 Fostering the productive development of SMEs with significantly greater

access to long-term domestic financing, modern technology, entre-
preneurial and labor training, and more stable domestic markets; and
enhancing their ability to conquer foreign markets or to become suppli-
ers of large domestic exporters.
VIII
Managing Capital Inflows in the
1990s and the “Encaje”*
In the nineties, private capital inflows returned to Latin America (see Calvo
et al., 1993; Ffrench-Davis and Griffith-Jones, 1995). Undoubtedly, the
resumption of capital flows, which had been interrupted in the eighties
with dramatic recessive effects on economic activity, had positive short-run
effects. In fact, it implied relaxing the shortage of foreign currency (binding
external constraint, BEC) under which most countries had operated dur-
ing the debt crisis. However, both the large magnitude of the new capital
inflows and their composition (prone to volatility) caused a growing macr-
oeconomic disequilibria in the recipient countries.
Chile was one of the first Latin American countries to attract the renewed
flows of foreign capital and one that faced the largest supply in relation to
its economic size. One main feature in Latin America was that the huge capi-
tal inflows were directed to consumption and to the acquisition of already
existing assets; Argentina and Mexico were outstanding examples in 1991–4
of a crowding-out of their national savings. It will be argued that, in the
nineties, one reason for the greater degree of Chilean success in channeling
foreign capital to investment was the explicit policy set that included, as
an outstanding component, counter-cyclical regulations discouraging an
“excess” of short-term inflows; associated with this, a high participation
of greenfield foreign direct investment (FDI) in total capital inflows was
recorded. The Chilean experience does indeed suggest that, when capital
inflows take the form of FDI, there is a greater likelihood that flows will be
more permanent and complement national savings instead of crowding-out
them, in contrast with the case in which foreign capital is in the form of
liquid or short-term flows or acquisitions of existing capacity.
Policies during most of the 1990s represented a significant move towards
a pragmatic, counter-cyclical, pro-development approach to capital flows.
* This chapter is partially based on Agosin and Ffrench-Davis (2001). I appreciate the
authorization of Manuel Agosin, Oxford University Press and UN/WIDER to make use
of that material.
209
210 Managing Capital Inflows
In a nutshell, the policy response during the 1990s surges in the supply of
foreign capital can be described as a successful attempt to discourage short-
term inflows while maintaining openness to long-term flows. Particularly,
policies were directed towards increasing the cost of short-term inflows via
non-interest-bearing reserve requirements (el encaje); this is a price-based
regulatory policy tool intended to modify relative market costs or profits. Its
level and coverage were modified in a counter-cyclical way.
The authorities also resorted to intra-marginal intervention in order to
slow down RER appreciation (in the face of inflows that surpassed the barrier
of the reserve requirement) and sterilize the monetary effects of reserve accu-
mulation. In parallel, there was a fiscal budget surplus and prudential regula-
tion and supervision of the financial sector was enhanced. This set of policies
sought to protect a development strategy whose main elements were (i) the
growth and diversification of exports, and (ii) a stimulating and sustainable
macroeconomic environment for productive activities and employment.
Policies were effective in achieving their targets during most of the 1990s.
During this period, productive capacity expanded briskly and economic
activity was close to its output capacity up to 1998. This ensured a persist-
ently rising investment ratio and increased employment and wages, in a vir-
tuous circle. Crucial variables were macroeconomic prices, such as interest
and exchange rates, which stayed well aligned. For example, the latter had
a very moderate appreciation in Chile in comparison with the rest of Latin
America. As a consequence, the current account deficit averaged only 2.3%
of GDP in 1990–5; in that period, exports grew vigorously and diversified
strongly (see Chapter VI). In those years the regressive trend in income dis-
tribution of the seventies and eighties stopped, and social equity exhibited
an improvement (see Chapter VII).
However, in 1996–7 this policy mix and the strength with which it was
applied remained rather unchanged, in spite of a new vigorous surge in capi-
tal flows to most countries in the region; even more, the systematic monitor-
ing made up to 1995 in order to avoid leakages lost force. As a consequence
of the lack of stronger comprehensive counter-cyclical action on the capital
surge – such as a higher reserve requirement or other similarly restrictive
policy tools – and despite heavy intervention in foreign exchange markets,
the central bank allowed a sharp real exchange rate appreciation and a worri-
some rise of the deficit on current account during this biennium. In parallel,
broad exit channels of national capital were being opened (principally for
the privatized social security funds), without significant outflows as long as
there prevailed optimistic expectations and exchange rate appreciation was
expected. In this way, for both capital inflows and outflows a pro-cyclical
macroeconomic vulnerability was being created, which exploded in 1998–9.
Nonetheless, as discussed below, the benefits of the active regulation
implemented in previous years had left large net international reserves,
a rather low stock of foreign liabilities, and a small share of volatile inflows,
Managing Capital Inflows 211
which allowed a recessive adjustment in 1999 that was remarkably less trau-
matic than those in 1975 and 1982.
This chapter studies the phenomenon of massive capital inflows in Chile
in the 1990s, the policy approaches utilized to deal with it, and their effects
on the domestic economy. Section 1 describes the size and composition of
capital inflows. Section 2 discusses the policy approaches taken to deal with
capital surges. Section 3 analyzes macroeconomic impacts and the effective-
ness of counter-cyclical policies implemented. Section 4 briefly presents the
outcomes in investment, savings, and economic growth. Finally, section 5
concludes with a discussion of counter-cyclical policy lessons.
1 Capital inflows: magnitudes and composition
The return to democratic rule in 1990 coincided with a new capital surge
to emerging economies (EEs). By 1986 there had been an initial spurt of
private capital towards Chile, which was associated almost exclusively with
the debt–equity swap program started by the authorities in 1985. Owing to
the large exchange rate subsidy implicit in the swap scheme, the program
was successful in attracting significant amounts of foreign investment; it is
well known that generally they did not involve an actual currency inflow
to Chile, even though in a process of financial engineering they implied a
reduction of the debt with foreign banks (Ffrench-Davis, 1990). The swap
program was abandoned by foreign investors in 1991, mainly because the
rise in the international price of bonds (pagarés) representing Chilean debt,
at the same time that the new authorities reduced incentives, made it no
longer profitable to invest via debt swaps. However, FDI not associated with
swaps did gain strength. In fact, during the 1990s, greenfield FDI became a
major share of net capital inflows (see Table VIII.1). Acquisitions, following
a strong world trend, also rose in Chile, but up to 1998 remained the minor-
ity of FDI inflows; the composition would revert sharply in the recessive
environment of 1999.
The supply of short-term private inflows also figured prominently in the
capital surge towards EEs. For interest-arbitraging capital inflows to take
place, the domestic interest rate must exceed the international rate by a mar-
gin that is more than sufficient to compensate for the expected exchange
rate depreciation and the country risk premium. These conditions prevailed
in Chile from the late 1980s, supported by a combination of five circum-
stances. First, in 1992 and 1993 international dollar interest rates reached
a thirty-year low. Second, notwithstanding the record high investment
ratio of Chile in the 1990s, it still had a low stock of productive capital.
Obviously, that shortage of physical capital tends to provide a higher trend
rate of return. Therefore, the interest rate must tend to be higher than in a
developed economy. Thus, monetary policy, in order to be consistent with
sustainable macroeconomic balances, must hold real interest rates over the
212
Table VIII.1 Composition of capital flows, 1980–2008 (2008 US$ as % of GDP)

1980–81 1982–89 1990–95 1996–97 1998–99 2000–03 2004–07 2008
a 6.8 8.9
FDI net inflows 1.4 1.1 3.5 7.8 9.0 4.9
Greenfield FDI 1.4 1.0 2.5 4.8 2.3 1.8 5.1 6.4
M&A 0.0 0.1 1.0 3.0 6.7 3.0 1.7 2.5
FDI net outflowsb –0.2 0.0 –0.8 –1.7 –2.7 –2.3 –1.7 –4.7
Official gross loansc 0.4 2.1 0.9 0.2 0.1 –0.1 –0.1 0.0
Public amortizationsd –5.0 –1.5 –1.5 –2.7 –0.6 –1.1 –1.3 –1.4
Private amortizationsd –2.9 –1.7 –1.2 –2.4 –3.9 –7.3 –8.0 –6.7
Portfolioe 0.2 0.0 0.7 1.8 –3.8 –1.2 –6.0 –5.2
f
Other capital 25.6 5.5 4.0 4.3 2.5 8.1 7.1 14.3
Net Capital Inflows 19.5 5.4 5.8 7.4 0.6 0.9 –3.3 5.2
Sources: Balanza de Pagos data from the Central Bank; Mergers and Acquisitions (M&A) from UNCTAD. Balance of payments figures in current US
dollars were deflated by an index of external prices faced by the Chilean economy, obtained from Ffrench-Davis (1984) and Central Bank, scaled
to 2008 100. The GDP series in 1996 constant prices was scaled-up to the 2008 price level, and transformed to 2008 US dollars using the average
nominal exchange rate of that year.
a
FDI in Chile, net from repatriation of capital. Includes net loans associated with risk capital of FDI and excludes debt-equity swaps. bNet investment
by Chilean firms abroad. cIncludes loans from foreign official and multilateral institutions. dAmortizations, including pre-payments. eNet balance of
flows of investment funds, ADRs and bonds. fShort and mid-term credit lines of banks and others.
international ones. Third, in addition to the “structural gap” in interest

rates, in 1990 a downward adjustment in economic activity took place that
relied on a significant rise in domestic interest rates. In January 1990, as dis-
cussed in Chapter IV, before the assumption of President Aylwin in March of
that year, the central bank had carried out a sharp interest rate hike.1
Two other requirements for interest arbitrage were also favorable to
capital inflows. Chile experienced a RER depreciation of 130% in 1982–8, in
response to the sharp scarcity of foreign currency; together with improved
terms of trade it appeared that Chile was leaving the debt crisis behind.
Consequently, expectations regarding the RER turned from depreciation
to appreciation. Additionally, there was a dramatic reduction in country
risk expected in EEs by international rating agencies and investment funds.
Consequently, in the early 1990s expectations of exchange rate apprecia-
tion, the capital inflow itself, and an improved current account position
made short-term round-tripping in Chile appear very profitable. Short-term
private flows were very sizable until 1992, when they began to decrease as a
consequence of the new policy measures adopted in 1990 and 1991 to stem
them and their intensification in 1992 (see section 2).
Novel components of flows were portfolio inflows. They took two forms:
investments through mutual funds set up in the major international capi-
tal markets and the issue of American Depositary Receipts (ADRs) by large
Chilean corporations. The ADR is a mechanism by means of which foreign
corporations can issue new shares on the US stock markets. The original or
primary emission of ADRs is put under rigorous requirements; for example,
the issuing company must satisfy established minimums on capital and
exceed requirements on risk classification, and still more importantly, the
ADR must correspond to an increase in the capital of the issuing firm. The
“primary” issue of ADRs implies an expansion of the capital of large firms
at relatively low cost, since capital in US markets naturally tends to be less
expensive than in Chile. The relatively developed domestic stock market
and the low country risk made Chilean stocks a prime candidate for inves-
tors seeking new and more exotic financial vehicles.
There is also a “secondary” issue of ADRs through the purchase by non-
residents, in the Chilean market, of the rest of the stock of the Chilean firms
that have issued ADRs in the USA. Subsequently, this purchase can be con-
verted into ADRs, after which they become tradable in the US markets (see
Ffrench-Davis et al., 1995). This “secondary” operation does not constitute
an enlargement of the capital of the issuing company but only a change in
ownership from nationals to foreigners. These shifts in ownership (without
expanding risk capital) involve exposing the economy to an additional
degree of volatility, since when foreign investors’ mood changes they can
easily reverse the operation. These flows have played a destabilizing role.
1
In the first months of 1990, the real lending interest rate averaged 16% annually.
They inflated the stock exchanges in 1994 and 1997 and depressed them in
1995 and 1998, in a clear case of pro-cyclical behavior.
At the same time that private flows were increasing, public debt was
being reduced. This reduction corresponded to amortizations of liabilities
contracted during the debt crisis (see Table VIII.1); these amortizations,
including substantial prepayments, were undertaken to alleviate the large
accumulation of international reserves by the central bank, to reduce net
financial costs, and to relax appreciating pressures on foreign exchange
markets and improve the balance of the bank.
From 1991, several large Chilean corporations made direct investments
abroad (see FDI outflows in Table VIII.1). The destinations were mainly
neighboring countries. The largest investments, which increased through
the 1990s, were principally directed to electricity generation and distribution
(mostly in recently privatized companies, first in Argentina and then in other
Latin American countries), and to other sectors such as light manufacturing
and retailing (Calderón and Griffith-Jones, 1995). In 1996–7, after persistent
expansion since 1991, the net flow of investment abroad represented 1.7% of
GDP.2 In turn, the main institutional investors (private pension funds, AFPs),
in spite of the increasing facilities provided by policy to invest abroad, in
mid-1997 had only exported 0.5% of their total assets.
As is documented in section 3, the measures to discourage speculative
short-term inflows were effective in isolating the Chilean economy from
the contagious effects of the tequila crisis. Thus, while the economies of
Argentina and Mexico contracted dramatically in 1995, the GDP of Chile
expanded vigorously (10.6%), with an increasing investment ratio (22.2%
of GDP in 1995 versus 17.6% in 1990; for Argentina and Mexico the invest-
ment ratio in 1995 was 16.5 and 15%, respectively).
Only the stock market experienced a drop of foreign portfolio invest-
ment. However, it did not have recessive macroeconomic effects, due to its
small share in total inflows; particularly, the depreciation that took place
was within the margins of the exchange rate band. Again, the contrast with
Argentina and Mexico is remarkable: between September 1994 and March
1995, the stock price index of those countries deteriorated by 32 and 65%,
respectively, whereas in Chile an almost negligible drop of 4% took place.
The Mexican and East Asian crises are illustrative of these dangers of
financial destabilization. In the case of Mexico, as emphasized by Sachs
et al. (1996), domestic policy failures, particularly the large increase in credit
that resulted from a poorly regulated financial system, were important fac-
tors. In both crises, domestic credit booms, however, were triggered by large
2
Balance of payments data underestimate the size of these investments because a
large share of them is financed with funds raised on international capital markets that
never enter the country. A similar situation emerged regarding Korean investments in
its neighbors (see Mahani et al., 2006).
capital inflows, in a similar process to the one that took place in Chile in
the seventies. The herding behavior displayed by foreign portfolio investors
has been recognized as a critical element in the Mexican crisis (Calvo and
Mendoza, 1996); a similar pattern preceded the Asian crisis. Since assets of
firms from a particular developing country are normally a very small propor-
tion of international investors’ portfolios, it may not pay to go to the trouble of
obtaining costly information. Therefore, they tend to follow “signals.”
The pro-boom signal by the early 1990s was that Mexico was undertaking
decidedly market-oriented reforms (and was entering NAFTA and then the
OECD) that would, in the eyes of the international banks, raise returns on
Mexican corporate assets. By late 1994, the signal for a reversal of the finan-
cial capital inflow was the notion that current account deficits had become
“unsustainable” and the exchange rate had appreciated “excessively” in
Mexico; but that should have been a surprise to no one because it was an
ongoing process at least since 1992 (see Ffrench-Davis, 2006, Box VII.1).
Of course, (i) the large current account deficits and outlier macroeconomic
prices, particularly an appreciating RER, had principally been a consequence
of the exogenous (and collective) behavior of foreign investors in the first
place. And (ii) it had been going on for several years.
Paradoxically, a “successful” country can see its fundamentals – such as the
deficit on the current account, exchange rate, savings, and bank portfolio –
worsened by a large capital surge. From a theoretical point of view, what
we have is the possibility of multiple equilibria: an appreciated exchange
rate with large capital inflows, and a depreciated exchange rate with capital
outflows. Moreover, there are dynamics involved: capital inflows appreciate
the RER, and the latter, if it is gradual, encourages additional inflows. This
can proceed for several years, as happened in 1976–81 and 1990–4 in several
LACs. After a while, when the deficits on current account accumulate due to
increasing stocks of external (particularly liquid and short-term) liabilities,
the appreciation trend is replaced with expectations of depreciation, which
in turn tends to lead to a reversal in the flows direction (see Ffrench-Davis,
2006, Chapters VI and VII). This indicates that there is a need for policies
that reduce the more volatile components of capital inflows, and demon-
strates that the “fundamentals” are dependent on policies towards financial
flows. Moreover, some equilibria are more “desirable” than others, in terms
of their effects on economic growth and its sustainability.
Nevertheless, the new capital surge towards the region in 1996–7, espe-
cially to “successful” countries like Chile, which had weakened its counter-
cyclical strategy, generated new macroeconomic vulnerabilities.3 In 1998,
for the third time in sixteen years, a drastic generalized reversal of financial
3
Surprisingly, capital inflows as a percentage of GDP were only slightly larger in
1990–5 than during the debt crisis (5.9% versus 5.5% in 1982–9). Nevertheless, in
1997 total inflows peaked to 9% of GDP.
flows to Latin America took place. Then, the Chilean economy suffered the
counterpart of the financial inflows received in 1996–7: financial outflows
began by late 1997 and accelerated in 1998–9. Unlike what happened in
1982 – when the majority of creditors were banks, and loans were subject
to flexible interest rates – now most resources caught by Chile corresponded
to FDI, a more “friendly” component in periods of crisis (see Table VIII.1).
Furthermore, the “service” of greenfield FDI tends to be counter-cyclical
in recessive situations because of the reduction of profits under a domestic
recession.4 In fact, their net profits after taxes fell from US$1,850 million in
1997 to US$1,050 million in 1999.5 The recessive situation that emerged in
1998 and dominated the macroeconomic environment in 1999 was caused
by the reversal of financial inflows by foreigners and by a newcomer to
crises: outflows by domestic institutional investors.
2 The policy counter-cyclical response
In the 1990s, the Chilean monetary authorities deployed a wide range of mac-
roeconomic policies to regulate capital surges and their effects. On the one
hand, the central bank attempted to discourage short-term and speculative
inflows while maintaining open access for FDI. On the other hand, it sought
to moderate the impact of financial inflows on the domestic economy, by
intervening in foreign exchange markets so as to prevent an unduly appreci-
ated real exchange rate and sterilizing the excessive monetary effects of the
rapid accumulation of international reserves (see Ffrench-Davis et al., 1995).
Two other policy factors contributed to the success achieved in managing
capital inflows. The first was fiscal discipline. A higher level of social expendi-
ture was financed through new taxes. Chile ran a significant public sector
surplus during 1990–7, amounting to 2% of GDP.6 The responsible stance
on fiscal policy, including compliance with the rules of a copper stabiliza-
tion fund,7 eased the task for the monetary authorities in managing capital
4
Even though they have significantly different productive and macroeconomic
implications, acquisitions are customarily registered together with greenfield FDI
inflows. They have been disaggregated as M&A in Table VIII.1.
5
Interestingly, the behavior of profits is counter-cyclical, but that of the rate of
remittances tends to be pro-cyclical: the share of profits remitted rose from 59 to
75% between 1997 and 1999. However, the net effect was a drop in total profit remit-
tances; this compensating effect must be taken into account when examining the
implications for the domestic economy of terms of trade changes.
6
Only in 1999 was a deficit recorded, determined by a significant drop in fiscal
income. This effect, evidently not a cause, was associated with the severe adjustment
process that depressed aggregate demand.
7
Given the weight of copper in the balance of payments and fiscal accounts, and the
instability of its price, a Buffer Fund was created in the 1980s under a SAL with the
World Bank. The fund was active throughout the 1990s, playing a stabilizing role for
fiscal income and the foreign exchange market.
inflows and aggregate demand, while a counter-cyclical macroeconomic

approach was applied decisively and with consistency. Second, subsequently
to the 1982–3 banking crisis, prudential banking regulations were introduced
and have been perfected over the years. This, again, prevented capital inflows
from unleashing a lending spree by the commercial banks, which, in turn,
eased the task of keeping the current account and the exchange rate within
sustainable bounds up to 1996.
(a) Implementing a new macroeconomics and the reserve

requirement: 1990–2
In a context of a massive supply of capital inflows, the two main goals of
exchange rate and inflow management policies should be (i) to achieve
sustained macroeconomic balances, especially in an economy prone to
huge cycles (recall that in 1975 and 1982 Chile had experienced the sharp-
est recessions in all Latin America) and (ii) to protect the growth model
adopted by the authorities, which granted a leading role to the expansion
and diversification of exports. For the former, it is crucial to manage aggre-
gate demand at levels consistent with the evolution of potential GDP; for
the latter, it is crucial to manage the exchange rate at ranges consistent with
a sustainable balance of the current account.
For exports to act as an engine of growth for the Chilean economy, the
level and stability of the real exchange rate are crucial. Consequently, the
Chilean authorities opted to regulate the foreign exchange market in order
to prevent large misalignments in the RER relative to its long-term trend.
The option chosen to make long-term fundamentals prevail over short-term
factors, influencing the exchange rate and the aggregate demand, assumes
that there exists an asymmetry of information between the market and the
monetary authorities, because the latter have a more comprehensive set of
objectives and knowledge of the factors driving the balance of payments
and economic activity; and, principally, because they have a longer plan-
ning horizon than agents whose training and reward are associated with the
results that they get at the short-term end of the market.
Here I give an analytical account relating policy changes implemented in
the 1990s. The changes taking place in global markets, the increasing inter-
national approval of domestic economic policies, the vigorous economic
activity, high domestic interest rates in Chile, and a smooth transition to
democracy stimulated a growing capital inflow since 1990.
These events were quickly reflected in a RER appreciation. By July 1990,
the exchange rate was at the floor of the band managed by the central bank
(in Latin American terms, i.e. the appreciated extreme). Even during the
Iraq crisis in September 1990, the market rate stayed on the floor of the
band, despite the fact that Chile was then importing 85% of its oil needs;
Chile reacted to the shock brought by the oil price by drastically raising the
domestic price of fuel, which caused an inflationary shock in September
and October. The CPI, whose inertial component implied a rise of about
2% monthly at the time, jumped to 4.9 and 3.8%, respectively, in those
months. The speed and close coordination with which the central bank and
the government reacted to external shocks may explain why pressures in the
foreign exchange market continued to encourage appreciation and inflation
was quickly reduced.
In early 1991, the strict crawling-peg system that had been followed by
the monetary authorities was modified, still without intra-marginal inter-
vention. The adjustment was in order to introduce “exchange-rate noise,”
which would discourage short-term flows, while deeper reforms were being
designed. The rate was moderately revalued on three different occasions
and then, in compensation, devalued gradually. Thus, at the end of each of
these moves, the “official” rate returned to its initial real level; the devalua-
tions within each move made it more costly for short-term inflows and thus
served as an effective tool for temporarily stemming the excess supply of
foreign exchange. However, the measure could not be repeated too often,
since the market would then anticipate the revaluation and the policy
would lose its effectiveness, which actually happened in the third move.
Nevertheless, during almost six months the authorities gained time to
design a policy that would act efficiently in a more prolonged transition
period: that main policy reform was the imposition of an unremunerated
reserve requirement (URR) on financial inflows.
Undoubtedly, this policy reform advanced against the fashionable opin-
ion of multilateral institutions and financial agents, which stressed the
need for an across-the-board capital account opening. The reform adopted
by the new national authorities was based on the expectation that the large
supply of financing was not permanent, and short-term factors affecting
the current account – such as the high price of copper, the high domestic
interest rates, and the transitory depressed level of imports – would tend
to change in the near to medium term. It was recognized, however, that
part of the observed improvement in the current account – a considerably
improved non-financial services account, a more vigorous non-traditional
exports, and a reduction in the external debt burden – was more structural
or permanent.
In June 1991, in response to this combination of factors, and pari passu
with a 2% revaluation and an import tariff reduction from 15 to 11%,
a non-interest bearing reserve requirement (URR) of 20% was established on
foreign loans (covering the whole range of foreign credits, including those
associated with FDI to trade credit). The reserves had to be maintained with
the central bank for a minimum of ninety days and a maximum of one
year, according to the maturity of the operation. At the same time, a stamp
tax on domestic credit, at an annual rate of up to 1.2%, was extended to
foreign loans. In July, an alternative to the reserve requirement was allowed
for medium-term credits, which consisted of making a payment to the

central bank of an amount equivalent to the financial cost of the reserve
requirement. This cost was then calculated by applying the LIBOR rate plus
2.5% (at an annual rate) to the amount of the reserve requirement. The
URR, the option of paying its financial cost, and the tax on credits all had a
zero marginal cost for lending exceeding one year, and, as discussed below,
the first two were particularly onerous for flows with short maturities (see
Table VIII.3, below).
It should be noted that the stages of the business cycle in Chile and in
its “financial center” (the United States) coincided during most of 1991.
However, in the ensuing months, US interest rates continued to decline,
exerting pressure on the central bank. In parallel, the Chilean economy
was booming, and its GDP growth rate had risen well into two digits.
Consequently, for the sake of macroeconomic equilibrium, the central bank
wanted to raise rather than lower domestic interest rates. To avoid encour-
aging arbitrage and under a large supply of financial inflows, the system
of reserve requirements was tightened and extended to other international
financial transactions.
In fact, beginning in January 1992, it was extended to time deposits in
foreign currency; in May, the rate of the reserve requirement was raised to
30%, and the period during which the deposit had to be maintained was
extended to one year, regardless of the maturity of the inflow. The spread
charged over LIBOR in the option of paying the financial cost of the reserve
requirement was increased from the original 2.5 to 4%.8 Later, in July
1995, its coverage was extended to purchases of Chilean stocks (“secondary
ADRs”) by foreigners.9 In order to close a loophole through which the
reserve requirements were being evaded (since equity investment was
exempt), the authorities decided to screen FDI applications; permission to
enter the country as FDI exempted from the reserve requirement was denied
when it was determined that the inflow was disguised financial capital. In
such a case, foreign investors had to register their funds at the central bank
as financial investments subject to the reserve requirement.
With the Asian crisis, and the sudden sharp scarcity of financial inflows,
the reserve requirement rate was reduced to 10% and then to zero in 1998.
8
On diverse occasions, as a counter-cyclical regulation mechanism, the rate of refer-
ence and the spread considered in calculating the entrance fee were modified.
9
It is not difficult to impose reserve requirements on foreign portfolio investments.
If funds that will be used for the investment are deposited with a Chilean bank, the
foreign deposit is liable to reserve requirements. For those funds that do not use a
Chilean bank as intermediary, the reserve requirement can be imposed when the asset
is registered in the name of an agent with a foreign address. In order to be converted
into ADRs, such assets must also be registered with the central bank.
The authorities announced, however, that the policy tool would remain
available in case there were new capital surges (Massad, 2000), and it was
restated that the URR was a counter-cyclical macroeconomic tool.
A summary of policy actions taken to tackle the excess of foreign cur-
rency during the period of abundant external financing can be found in
Box VIII.1.
From 1991, an attempt was made to facilitate capital outflows as a way of
alleviating downward pressure on the exchange rate. In particular, gradu-
ally, pension funds (AFPs) were allowed to invest up to 12% of their total
assets abroad. With a similar goal, as well as the objective of enhancing the
productive development of Chilean firms, residents seeking to invest abroad
were granted access to the formal foreign exchange market. The policy was
effective in encouraging significant flows of FDI and purchases of foreign
firms by Chilean companies in neighboring countries (the so-called chapter
XII of foreign exchange regulations; see Calderón and Griffith-Jones, 1995).
However, higher rates of return on financial assets in Chile than abroad and
expectations of peso appreciation discouraged financial investments abroad
by Chilean pension and mutual funds. These investments had been rising
very slowly as the domestic firms became better informed about foreign
financial assets. By mid-1997, the AFPs had only US$200 million invested
abroad, which represented 0.5% of the pension funds. An immediate effect
of liberalizing outflows probably was to encourage additional inflows as
a result of a fall in the country risk perceived by international investors.
Consequently, as argued by Williamson (1993) and Labán and Larraín
(2000), the resulting effect tends to be the opposite of the desired one.
The market takes advantage of opportunities to place foreign currency
abroad when expectations of appreciation are replaced with expectations of
depreciation.10 In the face of such a change in expectations in 1998–9, there
were massive outflows in the channels opened (see Chapter IX).
(b) Reforms to exchange rate policy since 1992

The large capital inflows of the early 1990s put strong pressure on the real
exchange rate. In order to moderate this trend, the authorities operated
through exchange rate policy, as noted earlier. However, appreciating pres-
sures continued in the ensuing months. Many observers began to hold the
view that a modification of exchange rate policy with a significant revaluation
was unavoidable. Consequently, the official rate began to lose its allocative
10
Nationals of the countries concerned have been observed to behave in much the
same way as foreign portfolio investors if they are allowed to do so. Thus, the ultimate
cause of exchange rate and asset price volatility is associated with the openness of
the capital account and the ease of moving into and out of assets denominated in
foreign currency. For an analysis of Chilean institutional investors, see Ffrench-Davis
and Tapia (2001) and Zahler (2006).
Box VIII.1 Regulations on capital flows by mid-1998

• Foreign direct investment. The only restriction on FDI inflows is the require-
ment that investments remain in Chile for a one-year period. There are
no additional restrictions on outflows, except for a tax on profit remit-
tances. FDI must be financed with a maximum debt component of 30%
(70% equity). This limit was reduced from 50% in October 1997.
• Portfolio investment inflows through ADRs. The minimum amount of ADR

issue is US$25 million, reduced from US$50 million in September 1994. A
minimum risk rating of BBB is required for non-financial firms and BBB+
for banks issuing ADRs. A 30% reserve requirement on secondary ADRs
was established in July 1995.
• Other financial and portfolio inflows. Subject to the 30% reserve require-
ment for a one-year period. With the Asian crisis, this was reduced to 10%
in June 1998 and to 0% in September. The inflows include trade credits,
foreign currency deposits, foreign loans including those associated with
FDI, and bond issues. Bond issuers face the same quality-enhancing
restrictions as ADR issuers. From 1992 foreign loans faced a monthly tax
of 0.1%, with a maximum of 1.2% on credits of one year or more.
• Investment abroad by the Chilean non-financial private sector. Currency pur-

chase in the formal market for investment abroad is authorized. Investors
not wishing to have access to the official foreign exchange market need
only inform the central bank of their investments abroad. Those wishing
to have access to the official market need permission from the central
bank. This is not difficult to obtain. Generally, the formal and free market
exchange rates were similar.
• Foreign investment by Chilean institutional investors. Foreign investments by

pension funds, mutual funds, and life insurance companies are subject to
certain limits as to the amounts and types of foreign assets that they can
hold. Pension funds were allowed to hold abroad up to 12% of their total
assets (raised to 16% in 1999), and investment in stocks was limited to
one-half of total foreign holdings.
• Foreign investment by banks. Foreign financial investments by commercial

banks are limited to 25% of bank capital and reserves and are restricted
to fixed income securities issued or guaranteed by foreign governments
or central banks. Banks are authorized to use foreign currency depos-
its to finance trade among countries belonging to the Latin American
Integration Association (LAIA). Commercial banks may hold equity in
foreign banks provided that they have a capital adequacy index of at least
10%.
Source: Central Bank reports.

capacity between tradables and non-tradables. In January 1992, the official

exchange rate was revalued by 5% and the floating band in the formal mar-
ket was expanded to 10%.11 The observed rate abruptly appreciated by 9% in
the market; that is, nearly the sum of the appreciation of the official rate and
the lowering of the floor of the band. There followed an overwhelming wave
of expectations of more revaluations, which was fed by capital inflows in the
formal and informal markets. These flows were encouraged by the certainty
that the central bank, under its own rules, could not intervene within the
band. In fact, in a market persistently situated near the floor, it intervened
only by buying at the bottom price. The market’s expectation was that, if
something changed, the floor exchange rate would be revalued, as in fact it
had been in January 1992.
For a long time, a proposal circulated in the central bank that a “dirty”
or regulated float should be initiated within the band; proponents of this
view argued that the prevailing rules, with a pure band, an increasingly
active informal market, and a more porous formal market, would lead to
an observed exchange rate leaning towards either extreme of the band (on
the ceiling in 1989–90, on the floor later). The sudden revaluation of the
observed rate by nearly 10% between January and February 1992 contrib-
uted to the bank taking the decision to initiate the dirty float in March of
that year. The observed rate then fluctuated for several years within a range
of one to eight points above the floor (i.e. normally not on the floor itself),
with the bank continuing to make active purchases but also frequent sales
(though with a significant accumulation of reserves).
The widening of the band had apparently signaled that the central bank
had renounced attempts to deter revaluating pressures in defense of the
export strategy, allowing the market, dominated by the short-termist seg-
ment, to determine the observed rate within a very wide range. To the
contrary, the establishment of the dirty float gave back to the central bank a
greater management capacity, enabling it to strengthen long-term variables
in determining the exchange rate for producers of exportable and import-
able goods and services.
Finally, in July of the same year, the dollar peg of the official rate was
replaced with a peg to a basket of currencies (of which the dollar repre-
sented 50%, the deutschmark 30%, and the yen 20%) as the new benchmark
11
It must be noted that Chile was coming out of a profound debt crisis, which
was accompanied by a sharp exchange rate depreciation. Consequently, there was
space for some appreciation. However, as Chile was moving from a restricted to an
overabundant supply of external savings, the authorities wanted to avoid an over-
adjustment of the exchange rate. One specifically troublesome feature is that, as
the expectations of foreign agents change from pessimism to optimism, they seek a
higher desired stock of investment in the “emerging market” over a short period of
time. This implies excessively large inflows for a while. Obviously, these are transitory
rather than permanently higher levels of periodical inflows.
exchange rate.12 The purpose of these measures was to make the dollar–peso
arbitrage of interest rates less profitable by introducing greater short-term
exchange rate uncertainty, given the daily instability of international prices
among these three currencies. The replacement of a peg to the dollar with a
basket of currencies also tended to give greater average stability to the peso
values of proceeds from exports. Indeed, unlike financial operations, which
are largely dollar-denominated, trade is fairly diversified in geographic terms –
with the United States representing then only one-fifth of the total – and it
also operates with a more diversified basket of currencies.
As a result of the policy mix implemented in 1990–4 (plus some “good
luck,” the improvement in the terms of trade in 1994–5), Chile was enjoy-
ing a solid external sector with sustainable macroeconomic “fundamentals”
(a small deficit on the current account, a sustainable exchange rate, and a
limited amount of external short-term liabilities) when the tequila crisis
exploded in late 1994 and its contagion effect reached Argentina in 1995.
Therefore, the across-the-board cutoff in liquid resources for Latin America
did not dampen the Chilean economy. As said, the main (but minor) shock
was observed in the outflows of foreign capital from the stock market.
However, due to the financing of the external deficit with long-term funds,
this shock did not cause a recessive situation. In fact, 1995 saw an excellent
performance in terms of economic growth, productive investment, and
employment, in acute contrast with the depressed situation of Argentina
and Mexico in that year.
Towards mid-1995, speculative capital flows began to return to the region
and with special intensity to Chile. Given overwhelming expectations of
currency appreciation, after Chile had shown itself to be immune to the
Tequila shock, the large interest rate differential between the peso and
the dollar (together with good prospects for the Chilean economy) gave
foreign portfolio and short-term investors what amounted to a very profit-
able one-way bet; expected profits far exceeded the toll they had to pay
because of the reserve requirement for entering domestic financial markets.
This trend towards appreciation could have been softened by intensifying
price restrictions on inflows (i.e. by increasing the rate, term, and/or extent
of the reserve requirement; see Le Fort and Lehmann, 2003); the counter-
cyclicality of policy lost consistency. The generalized over-optimism that
financial crises had been left behind by the world and the risky temptation
to speed the reduction of domestic inflation with a significant exchange
12
The basket of currencies and its matrix of real exchange rates were already operat-
ing implicitly through an index of external inflation (see Ffrench-Davis, 1984). This
index was used to determine the official exchange rate month to month. The new
policy adjustment implied capturing daily those variations in the peso/dollar rate,
following the evolution of exchange rates between the rest of currencies of the basket
and the dollar.
rate appreciation weakened the previous highly successful macroeconomic

policy. In fact, the external deficit increased pari passu with exchange rate
appreciation and rapid aggregate demand growth, which were stimulated by
the capital inflows of 1996–7.
Exchange rate management encouraged speculative inflows after 1995. In
spite of its formal adherence to a crawling band, in 1996–7 the central bank
was in fact maintaining an almost fixed nominal exchange rate. There are
two factors contributing to the peso pressures since 1995: (i) the establish-
ment of an annual exchange rate appreciation of 2%, on the basis of the
assumption of greater productivity growth in Chile with respect to its main
trading partners, and (ii) overestimates of the external inflation used for
the adjustments of the exchange rate band. Moreover, in order to lower the
floor of the band, in 1997 the authorities tinkered with the weights assigned
to each currency in the basket, making less credible the peg to a currency
basket rather than to the dollar.13 Consequently, it was a failure not of the
policy tools (the band of a basket of currencies with intra-margin interven-
tions and the encaje) but of (i) their contradictory implementation and
(ii) the lack of a decision to strengthen the URR.
The effects of the Asian crisis, including worsened terms of trade in 1998
(thus, now combined with some “bad luck”) and a slower pace of exports
expansion, found Chile with an appreciated exchange rate (which had not
happened prior to mid-1995) and a deficit on the current account now twice
as large as the average for 1990–5 (see Figure VIII.1 and Table VIII.2).14
It took quite a long time for the bank to correct the outlier RER after the
contagion of the Asian crisis arrived. It argued that in an economy near to
the productive frontier and with a high external deficit, to allow a strong
devaluation would have substantial inflationary effects. Undoubtedly, the
pass-through effect on inflation is normally greater in an economy operat-
ing close to its productive frontier. However, this is also a powerful argu-
ment for counter-cyclical action during booms by avoiding the exchange
rate misalignment before the critical situation arises. Increasingly, from the
mid-1990s, the anti-inflationary objective had become an overwhelming
priority for the central bank, at the expense of the RER and, consequently,
at the expense of efficient resource allocation and real macroeconomic
13
In November 1994, the weight of the US dollar was reduced from 50 to 45%, reflect-
ing the falling incidence of that currency in Chilean trade. In January 1997, it was
arbitrarily raised to 80%. For a comparative analysis of bands in Chile, Israel, and
Mexico, see Helpman et al. (1994). For an excellent analysis of intermediate regimes,
with references to Chile, see Williamson (2000).
14
An enlarged deficit on current account (even after being adjusted by the trend terms
of trade) is a revealed proof of an overly appreciated exchange rate, which was moving
faster than net productivity improvements. I contend that in 1990–5 there was a stabi-
lizing appreciation, while in 1996–7 there was an outlier overvaluation that exceeded
by far actual net increases in TFP in Chile, thus destabilizing the external balance.
115
110 108*
106 104
105 12%
6% 99
100
95
19%
90 91 91*
85
80
80
75
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Figure VIII.1 Evolution of the real exchange rate, 1986–2009 (1986 100).
Source: The real exchange rate is the amount of real pesos per one unit of a real weighted basket of
currencies of trade partners. Index based on official figures from the central bank. The numbers
correspond to annual averages of sub-periods delimited by horizontal lines. The arrows show the
appreciation percentages between the respective peaks or sub-periods of annual averages. *The
values for 2008 and 2009 correspond to monthly peak and valley values of each year, respectively;
both values correspond to December of each year.
Table VIII.2 Net capital inflows and deficit on the

current account, 1980–2008 (% of GDP)
Net capital inflows Deficit on the
current account
1980–4 12.6 13.7

1985–9 4.6 4.1
1990–5 6.3 2.3
1996–7 7.8 4.8
1998–2003 0.8 1.6
2004–7 –3.2 –3.4
2008 5.7 1.6
Source: Balance of payments data from the central bank for capi-
tal inflows and current account, in US dollars. National Accounts
from the central bank and Marcel and Meller (1986) for GDP, in
pesos. The GDP figures in current prices were normalized with
a trend series for the nominal exchange rate estimated with a
Hodrick–Prescott filter in order to avoid the misleading bias
introduced by a volatile RER.
balances. In the end, it was a policy reversal at the expense of sustainable

growth as well as of equity.
(c) Strengthening banking regulation and supervision

As noted, tough bank regulation and supervision prevented the excess liquid-
ity of banks from fueling a consumption boom and deterioration in the qual-
ity of bank assets. This was a legacy of the banking crisis of 1981–6 in the
aftermath of the preceding foreign capital surge, which led to a virtual col-
lapse of the entire banking system (see Díaz-Alejandro, 1985). The pruden-
tial regulation and supervision adopted since then includes: (i) continuous
monitoring of the quality of bank assets; (ii) strict limits on lending by banks
to related firms; (iii) automatic mechanisms of bank risk capital adjustment
when its market value falls below the limits required by the regulators; and
(iv) faculties to freeze banking operations, forbid fund transfers outside of
troubled banks, and restrict the payment of dividends by institutions that
fail to comply with capital adequacy requirements. Chilean financial mar-
kets have also acquired a depth that allowed for the orderly infusion of new
funds, and for their withdrawal, without significantly affecting the quality of
bank portfolios (Aninat and Larraín, 1996; Held and Jiménez, 2001).
Capital adequacy ratios along the lines of the 1988 Basle Accord were
introduced into the new banking law approved by Congress in 1997. But
banks’ capital, in practice, was well above the Basle norm of 8%. In addi-
tion, the central bank imposed limits on banks’ open positions in foreign
exchange, although these were still fairly crude in that they did not differ-
entiate between loans made in foreign currency to firms that earn foreign
currency and to firms whose earnings are in domestic currency. Neither do
these limits differentiate between different currencies. Currency risk is only
one aspect of credit risk evaluation, which as a whole is quite good in Chile.
Therefore, this compensates for the weaknesses in the norms on open posi-
tions in foreign exchange.
3 The effectiveness of macro-stabilizing measures
Macroeconomic counter-cyclical, prudential, regulations also have microeco-

nomic effects. Both are associated with the financial costs imposed by the sys-
tem of reserve requirements and taxes on foreign lending. The implicit total
tax consists of the extra interest costs imposed by the reserve requirement
and the tax on foreign credits. The calculations can be seen in Table VIII.3.
As a result of the lengthening (in 1992) of the reserve requirement holding
period to a full year, and regardless of the maturity of the financial transac-
tion, the implicit annualized tax rate on foreign borrowing increased dramat-
ically as maturities shortened. Before the term of the reserve requirement was
extended, the implicit annualized tax rate was identical on transactions as
short as a quarter (the minimum holding period up to May 1992) or as long
as a year. These very large estimates of the implicit tax rate on short-term
operations suggest that, if the regulations were not evaded, they must have
implied strong discouragement of short-term and portfolio flows.
How effective has the reserve requirement (together with interest and
exchange rate management) been in deterring short-term flows and
preventing excessive external deficit and exchange rate appreciation? How
does it affect different kinds of agents and, therefore, what is the impact on
Table VIII.3 Implicit cost of reserve requirement on foreign borrowing, 1991–8

(annualized rates)
1991 II 1992 I 1992 II 1996 1997 1998 I 1998 Q3
Reserve requirement 20 20 30 30 30 30 10
(%)
Minimum holding 3 3 12 12 12 12 12
period (months)
LIBOR (%) 5.5 4.3 3.6 5.6 5.8 5.7 5.6
Implicit cost (%)
12 months 2.8 2.8 3.3 4.2 4.1 4.3 1.8
6 months 2.8 3.2 5.5 7.2 6.9 7.3 2.4
3 months 2.8 4.0 9.7 13.3 12.7 13.4 3.7
Source: Author’s calculations, based on information from the central bank. Implicit cost includes
the 0.1% monthly tax, with an annual ceiling of 1.2%.
equity? How does it alter the macroeconomic environment for investment

and productive development? All these questions have been in the theoreti-
cal, empirical, and political debate. Here, I deal with them.
(a) Arguments against capital account management

A traditional argument is that (i) the regulation does not have effects because
in the present economic world (and the one of the nineties) it stimulates
evasion, offsetting all impact from the start or, gradually, soon after. A vari-
ant states that (ii) the regulation affects the composition of flows, but this
does not have any economic impact. A third variant (iii) accepts that there
are positive effects but, at the same time, it worsens regressively the access
of small companies to financing.
Some observers have claimed that the efficacy of measures intended to
discourage capital inflows is only temporary, since they assume that pri-
vate operators usually find ways to fully evade them (see Valdés-Prieto and
Soto, 2000). In principle, this can be done through several mechanisms.
Here I mention five of them. One is the under-invoicing of imports or the
over-invoicing of exports. The second is to delay payment for imports or
accelerate export receipts. Third, funds can be brought through the informal
foreign exchange market. Fourth, short-term funds can be disguised as FDI.
Fifth, agents can arrange back-to-back operations in which, for example, an
agent pays for imports with a bank deposit in Chile rather than with foreign
exchange; at the same time, the exporter is paid in foreign exchange by a
bank in his or her country. All of these (and other forms of evasion as well)
are possible, but they are not costless, some have a one-shot impact and
only cause delay of the effects, and some may have undesirable effects on
the tax liabilities of the avoiders of regulation.
The authorities were conscious of the risks and the dynamics of the mar-
ket, which, by its nature, looks for evasion channels. For that reason, up
to 1995 they maintained a permanent monitoring, closing loopholes that

were appearing. Here I mention two. First, the financial flows that banks or
investment funds disguise as risk capital or productive investment; when
this was becoming an important loophole, the authorities moved to close it
by means of the case-by-case analysis of those flows, and by placing under
the reserve requirement those that corresponded to financial investments.
While some evasion is inevitable, there is no hard evidence that the meas-
ures targeted to discourage short-term capital inflows had been massively
evaded, as is verified below.
Second is the case of investment funds purchasing secondary ADRs. By
1993, the secondary issue of ADRs started to rise as a source of short-term
inflows with particularly volatile characteristics. Thus, the extension of
reserve requirements to these inflows in 1995 was an effective attempt
to deal with an incipient problem that was already causing difficulties in
policy management. However, after a temporary lull in 1995, they again
surged in 1996, though now paying the corresponding cost of the reserve
requirement. The evidence suggests that the entry fee came to be perceived
as cheap in the face of positive fundamentals, optimistic expectations, and
a strong likelihood of further RER appreciation. I repeat that, evidently,
the policy answer then should have been a strengthening of the restrictive
power of the reserve requirement, in order to adapt to the increased liquid-
ity of international financial markets.
In contrast to several studies that show a significant effect over short-term
inflows, a common line of attack against the use of disincentives has been
to claim that, with regard to their behavior, it is impossible to distinguish
between capital inflows such as FDI or long-term lending, on the one hand,
and short-term or liquid flows on the other. In an influential study (widely
promoted by the supporters of the “Washington Consensus”) Claessens et al.
(1995) claimed that balance of payments categories have nothing to do with
the stability of flows themselves, long-term flows being just as likely to be
unstable as short-term flows.15
It is interesting to note that after many empirical works that verified the
lack of consistency of those critics, more recently another one arose. This
one accepts that the regulations had desirable effects on the composition of
flows in the case of Chile, but that they were regressive. In that sense, Forbes
15
Part of the explanation of the finding that FDI is as likely to be volatile as short-
term flows may stem from the fact that, for the countries that they selected, FDI
flows were a small percentage of total foreign financing, as reported by IMF statistics.
Fluctuations in small numbers tend to be greater than fluctuations in large ones. On
the other hand, the period covered excludes the tequila crisis, when portfolio flows
played a significant destabilizing role. It is evident that instability must be tested in
critical situations rather than during booms. Finally, we repeat the relevance of distin-
guishing between FDI and mergers and acquisitions. FDI is mainly carried out in fixed
assets (it is “irreversible” investment), whereas M&A involves “liquid” inflows.
(2003) finds that the reserve requirement affected “small” firms more inten-
sively by imposing financial constraints on them.16 Gallego and Hernández
(2003) conclude that the reserve requirement affected the financial structures
of Chilean firms, reducing their leverage, increasing their reliance on self-
generated funds (retained earnings), and increasing the maturity profile of
their debt. Both microeconomic works use as their sample a group of listed
companies in stock markets.17
It is evident that any tax imposes some monetary cost to taxpayers and,
in doing so, changes relative prices. The crucial point is what the net effect
of capital controls is on overall welfare, after a double test: (i) to contrast
their eventual microeconomic costs and their macroeconomic benefits; and
(ii) to assess the effects throughout the economic cycle; that is, not only
during the boom but also during the fall that usually follows. There is strong
evidence that a bust has a significant bias against SMEs, as shown by the
regressive impact of the recession of 1999 against SMEs and labor of lower
qualification.
(b) Arguments in favor of the regulation of inflows and

empirical tests
There are two kinds of evidence that one can use to assess the net effect of
regulations. The first is qualitative. There is broad consensus that, in the first
half of the nineties, Chile faced a larger supply of external finance (relative
to its GDP) than other countries in the region, because of its better economic
performance and greater political stability. However, during those years the
exchange rate appreciation and the current account deficit (as a share of
GDP) were smaller than in other countries in the region that were major
recipients of foreign capital (see Ffrench-Davis, 2006, Chapter VII). In addi-
tion, FDI represented a much larger share of inflows in Chile than in other
countries;18 in those years of re-establishment of democracy, besides restrict-
ing capital inflows, Chile restored labor rights and some taxes on profits. We
can assume that the reason why greenfield FDI responded positively was not
these variables but other policies, such as the better macroeconomic envi-
ronment that counter-cyclical policies generated. These provided a better
16
Meanwhile, the positive macroeconomic effects of the reserve requirement had
been acknowledged by academic circles and authorities of institutions such as the BIS,
the IMF, and the World Bank.
17
Most companies listed on Chilean stock markets are among the biggest in the
domestic economy. Therefore, evidently, conclusions from these works cannot apply
directly to the overwhelming majority of SMEs, which are not listed on the stock
market and have limited access to capital markets.
18
It should be noted that the loans associated with FDI were subject to the reserve
requirement. Since the average maturity of these loans averaged about seven years,
the incidence of the restriction merely for their first year was low. However, this
avoided the danger of short-term credit being disguised as long-term credit.
environment for productive investment, which attracted FDI and, what is

about “four times as relevant,” also spurred domestic investment.19
The second is quantitative. Significant econometric evidence shows that
policies directed towards the capital account did work rather well for about
five full years. These studies indicate that the combination of disincentives
to short-term inflows together with the other reforms in the macroeco-
nomic approach, at least up to 1995, had been able to reduce short-term
and liquid inflows and to avoid an unsustainable deficit on current account
(Agosin, 1998; Larraín et al., 2000). As discussed below, the policy approach
of Chile changed markedly in the following years in the face of a new
generalized capital surge towards emerging economies, when restrictions
on inflows were, paradoxically, left unchanged by the autonomous central
bank rather than being increased in response to the huge capital surge of
1996–7. Consequently, the destabilizing effects in that biennium cannot be
attributed to the counter-cyclical policies of the first half of the nineties.
In fact, there is robust evidence that the capital controls applied in Chile
modified the maturity structure of inflows, reducing the share of the short
term component.20 Indeed, actual short-term and liquid flows were limited
and a significant share of them entered paying the reserve requirement. The
magnitude of the financial revenue from the toll reached around 1% of GDP
during all the years it was in force: a positive by-product.
In order to check the hypothesis that the composition of the flows does
not make a difference, a series of econometric tests were run to determine
the degree of persistence of different types of private flows (see Agosin and
Ffrench-Davis, 2001). These tests lead to the conclusion that FDI is consider-
ably less volatile than short-term borrowing and portfolio flows, and that
it is advisable to target policies of prudential macroeconomic management
(such as the reserve requirement) on short-term or liquid inflows. Moreover,
persistent flows tend to be associated with productive investment and not
consumption. This is what the Chilean authorities successfully reached
when they took the decision to apply a comprehensive counter-cyclical
policy. Undoubtedly, during the first half of the 1990s, short-term and port-
folio inflows would have been much larger in the absence of the reserve
requirement. Together with sterilized intervention in foreign exchange and
monetary markets, the policy approach adopted prevented undue exchange
rate appreciation and a consumption boom, thus keeping the current
account deficit within reasonable bounds up to the mid-1990s.
The policy mix also had financial costs for the authorities during the boom
period. The accumulation of large volumes of foreign exchange reserves
19
The ratio of investment by nationals to FDI has been around four to one.
20
See Edwards (1999), De Gregorio et al. (2000), Gallego et al. (2002), and Le Fort and
Lehmann (2003). See also the comments about the reserve requirement’s significant
effectiveness in Williamson (2003).
imposed a financial cost, since the returns on these assets abroad were lower
than the interest payments on the central bank liabilities issued to steri-
lize the monetary effects of reserve accumulation; the losses for the central
bank were estimated at about 0.5% of GDP per annum. That is the cost
of “insurance” for economic stability; the initial accumulation implied a
movement towards “equilibrium” since in 1990 net liquid international
reserves were notably low (actually, negative if measured comprehensively),
but probably became unnecessarily large by 1994 (though most welcome
during the sudden stop of inflows of 1995). Undoubtedly, a more flexible
and restrictive management of the reserve requirement and other prudential
macroeconomic policy tools by the authorities would have moderated that
cost.
From the point of view of investment and growth, the impressive growth
performance of the 1990s supports the hypothesis that the positive effect
of the whole macroeconomic approach (including the management of capi-
tal inflows) was much stronger than any associated microeconomic costs.
Actually, the investment ratio of Chile in the 1990s was the highest recorded
in its past history. In this sense, “financial constraints” as defined and
reported by Forbes (2003) were not an impediment to expanding produc-
tive capacity.21 Moreover, the microeconomic switch from debt to retained
earnings in the financial structure, as well as the shift towards longer-term
liabilities of “small” firms recorded by Gallego and Hernández (2003), can
be considered as a positive by-product of Chilean capital controls. Indeed,
the main sources of private savings are non-distributed profits and deprecia-
tion reserves of firms; the improved macroeconomic environment implied
a higher rate of use of installed capacity, higher effective productivity, larger
profits, better expectations, and higher reinvestment of profits. That is not a
negative outcome, but a highly positive one. This is consistent with a severe
incompleteness in the Chilean capital markets with respect to long-term
financing for SMEs and non-wealthy new entrepreneurs.
In the macroeconomic dimension, there is robust evidence that the access
to financing and spreads of SMEs are more intensively affected than those
of large firms during crises. Avoiding crises by discouraging capital inflows
during the boom stage naturally involves SMEs paying higher interest rates
than otherwise during the boom. However, in the bust phase of the cycle,
the policy contributes to avoiding both sharp increases in costs during the
avoided bust and the corresponding binding financial constraints that they
usually face during recessions; these restrictions could be avoided in 1995 and
attenuated as in 1999, thanks to the prior prudential regulations on inflows.
21
Forbes (2003) defines “financially constrained” firms as those that depend on their
own sources of financing to invest. In fact, most SMEs do not have access to the for-
mal financial market, and therefore the macroeconomic environment they faced is a
key variable determining its liquidity.
But, though I must point out that it is my interested view as a participant

in the design and implementation of this set of policies, the overall balance
provides a case of very effective and efficient counter-cyclical policies.
In summary, (i) there is strong evidence that the capital controls applied in
Chile modified the maturity structure of inflows, reducing the share of short-
term components. This conclusion emphasizes a very positive characteristic
of this policy tool, because the probability and severity of crisis appear to be
closely associated with a greater liquidity of external liabilities. Empirical stud-
ies also recognize that the URR (ii) allowed a variable gap between domestic
and external interest rates to be maintained, thus providing space for an active
monetary policy. This was of great importance for the sustained growth proc-
ess recorded during almost all the decade; in fact, frequent mini-adjustments
from the central bank allowed the avoidance of maxi-adjustments, result-
ing in an economy working persistently on or near the production frontier.
On the other hand, (iii) the effect of the URR on both total flows and real
exchange rate appreciation has been found (see Le Fort and Lehmann, 2003;
Ffrench-Davis and Tapia, 2005). In all, (iv) the reserve requirement contrib-
uted to moderate the stock of liabilities and to improve its profile (from both
micro and macro perspectives). According to most international evidence
these two factors strongly determine both the probability of crises and its
associated costs.
There is another dimension that is dynamic, linking with the future: (v) an
economy with a high rate of use of productive capacity and stable long-
term flows, usually exhibits higher rates of productive investment. This
generates two positive effects that increase the efficient absorption capacity
of external savings: first, gross capital formation is intensive in imported
inputs, which is the reason why a given amount of capital inflows causes
less exchange rate pressure when it is channeled to investment than when
it goes to consumption; second, a high investment ratio generates a larger
output in the future, with its respective demand for imports, creating an effi-
cient and sustainable higher absorption capacity, without a misalignment of
the exchange rate.22
4 Savings, investment, and growth in the 1990s
The remarkable increase in the productive investment ratio, and the fact
that the ratio kept rising up to 1998, is of great importance. In most of the
period 1991–8, there was no recessive gap (gap between actual GDP and
potential GDP). It has been shown that the recessive gap is a significant
explanatory variable of the investment ratio (see Agosin, 1998; Ffrench-
22
This relevant effect on absorptive capacity, associated with the composition of
inflows, is usually omitted from econometric research.
Davis, 2006, Chapter III). It also happens that a larger investment ratio, in
response to a negligible recessive gap, tends to be associated with stronger
SMEs and overall better employment.
The gap remained negligible until 1998, encouraging productive invest-
ment, even though other macroeconomic disequilibria were being built: a
rising external deficit and an appreciated exchange rate. These two explain
why Chile was hit by the Asian crisis from 1998 and it did not remain
immune as it had been in the Mexican crisis of 1994–5. For that matter, it
is important to make a sharp distinction between the period in which the
counter-cyclical approach was implemented in a coherent way (1990–5),
and the period of increasing abandonment of the active counter-cyclical
strategy (1996–7). In the first period, the Chilean economy became one of
the less vulnerable in a region facing financial and exchange rate crises,
escaping from the contagion of the Mexican crisis. In the case of the Asian
crisis, the negative effect was rather moderated and, as discussed, was mostly
linked to policy failures like allowing the accumulation of macroeconomic
imbalances and the careless liberalization of outflows by residents during
the boom phase in 1996–7. When the Asian contagion arrived and a signifi-
cant recessive gap emerged, the investment ratio fell sharply in 1999.
Undoubtedly, the 1990s marks a clear-cut improvement in the growth of
productive capacity, in comparison with 1974–89. The gross fixed invest-
ment to GDP ratio rose steadily: from 13.6% in 1974–89 to 20.2% in 1990–8
(in constant 2003 prices; see Chapter IX, Table IX.1). This increased ratio
allowed Chile to sustain a GDP growth averaging over 7% per annum in
that same period.
In those nine years, actual and potential GDP grew at similar and sustain-
able rates, with the economy operating regularly close to the productive
frontier. That is one of the main conditions that must be fulfilled by an
efficient macroeconomic policy. As already stressed, that positive feature
of a macroeconomics-for-development was determinant of the high rate of
productive investment recorded in the nineties, and therefore of the increas-
ing potential GDP growth. In fact, this is strongly associated with the
correction of the macroeconomic environment, with high rates of use of
productive capacity (and well aligned macroprices for the sake of sustain-
ability). Naturally, this high investment ratio contributed to the generation
of productive jobs and to the adoption of technological advances imbedded
in imported capital goods.
The increase in the gross savings ratio was also strong, rising from 12%
during the Pinochet regime to 22% in 1990–8 (in current prices). This reveals
that, in the nineties, national and foreign savings worked as complements, as
opposed to the substitution that took place in Argentina and Mexico before
1995 and in Chile before 1982 (see Uthoff and Titelman, 1998). As shown in
section 1, capital inflows averaged 5.8% of GDP in 1990–5, whereas the use
of foreign savings was reduced to 2.3%, with the difference being accumu-
lated in the international reserves. This reveals that the sterilization policies
of capital inflows, by preventing excessive exchange rate appreciation and
avoiding a high external debt, allowed the economy to absorb less foreign
capital than the amount offered, regulating it to amounts consistent with an
efficient absorption capacity. This was enhanced by the fact that most flows
were associated directly with productive investment.
The Chilean policies directed towards restraining capital surges and mod-
erating exchange rate appreciation can be credited with a significant share
of the success achieved with regard to investment, savings, and economic
growth. On the one hand, the management of inflows has had a positive
impact on real macroeconomic stability, and has contributed to keeping
effective demand close to productive capacity, which is essential for invest-
ment expenditure to rise. On the other hand, when capital arrives in surges
rather than trends, and takes the form of volatile financial flows rather than
FDI or financing of imports of capital goods, it tends to crowd-out national
savings. Foreign savings stimulate consumption through their effects on
domestic liquidity, the exchange rate, and asset prices. Thus, success in mod-
erating capital surges and modifying its composition contributed to a sharp
increase in national savings ratios (Solimano, 1990; Agosin, 2001).
5 Some policy lessons from this experience
The Chilean experience of prudential macroeconomic management of capi-

tal inflows provides several relevant lessons. For developing countries, the
swings in capital flows can be of extraordinary magnitude relative to the
size of their economies. Totally passive policy stances will inevitably gener-
ate external vulnerabilities, resulting in enormous volatility in key domestic
macroprices (exchange and interest rates) and economic aggregates (aggre-
gate demand, output gap and external balance). By depressing investment,
these fluctuations have adverse effects on long-term growth, productive
employment and social equity.
Contrary to the neoliberal view, it is possible to discriminate between
(i) flows that are stable, of a long-term nature, and contribute to the coun-
try’s growth (such as greenfield FDI that directly create new capacity) and
(ii) those that are basically speculative and lead to excessive domestic
volatility. In the Chilean case, the market-based discouragement applied
to speculative flows had no adverse effects on FDI, which, on the contrary,
reached unprecedented levels during the decade.
Some evasion is inevitable: any system of discouragement makes it attrac-
tive for some operators to attempt to circumvent it. In the Chilean case, it
was necessary to close loopholes when it became obvious that agents were
creating them. In fact, circumvention can be kept to a minimum with a well
designed and transparent system such as the reserve requirement on capital
inflows (encaje), and continued monitoring by authorities.
The objective of sustaining economic growth in the face of volatile capital

flows (or volatile export prices, as in Chile) requires the use of a battery of
counter-cyclical policy instruments. In the Chilean case, the combination of
tax-like instruments meant to deter speculative inflows, a crawling band with
intra-marginal intervention (which in my view, was utilized too sparingly),
and sterilization of the monetary effects of capital inflows worked well for
over half a decade. It contributed to the creation of a friendly macroeco-
nomic environment for productive investment, job creation, and economic
growth, keeping under control the external sources of vulnerabilities that
caused severe crises in Argentina and Mexico in 1995.
IX
Economic Policy after the 1999
Recession
In 1998 the Chilean economy ended a period of exceptional (actual and

potential) GDP growth. A high ratio of productive investment gave support
to that dynamism, which allowed improvement in the quality of employ-
ment, and contributed to partially reverting the regressive outcomes of the
economic reforms of the 1970s and 1980s. However, since 1998, the eco-
nomic dynamism and some key features of economic policy have shown
significant changes. Consequently, from the perspective of the economic
performance of the Concertación governments, there is a need to distin-
guish subperiods with different emphases and priorities. While per capita
GDP climbed by 5.4% per year in 1990–8, the rate of increase dropped to
2.6% in 1999–2008; actually, it was a sharp slowdown, even though it was
still above the 2% average of LACs and the 1.6% of the United States.
The first half of the nineties (1990–5) had especially positive features.
On the domestic front, actual GDP grew at an outstanding annual rate of
7.8% (similar to the potential GDP expansion), led by growing investment
ratios, and sustaining significant real increases in wages and employment.
Additionally, the worsened income distribution trend inherited from the
dictatorship not only stopped, but recorded some improvement. On the
external front, in spite of a massive supply of capital inflows, the govern-
ment was successful in maintaining a sustainable external deficit with mod-
erate real exchange rate appreciation in a way consistent with economic
fundamentals (see Chapter VIII and Balassa, 1981).
In the 1996–7 biennium, both output and investment stayed high; nev-
ertheless, there was a weakening of distributive improvements (with some
worsening in the inequality indicators), at the same time that risky external
imbalances emerged, which, soon, implied high economic and social costs
associated with the contagion of external negative shocks derived from the
Asian crisis in 1998–9. As a result, in 1999–2003, the economy remained
depressed, with a sizable gap between actual and potential GDP; only in
2004–5 did it experience a significant recovery, averaging 5.8% actual GDP
growth, now led by a quite positive external shock via the terms of trade.
236
Economic Policy after the 1999 Recession 237
However, notwithstanding that the positive external shock persisted well

into 2008, the speed of GDP growth diminished in the next triennium.1
This performance implied a significant average output gap as compared
with the negligible gap outstanding in 1990–8, with the economy now
generally operating significantly below its potential capacity. This real mac-
roeconomic imbalance was present during most of the period 1999–2008,
and naturally aggravated in 2009. Actual GDP and potential GDP did not
resume a sustained dynamism, evolving far below the record achieved in the
1990s (see Table IX.1).
It is relevant to stress that, in spite of the persistent recessive output gap,
the government continued to develop social reforms of great significance:
for instance, the Chile Solidario and Chile Barrio programs, the new AUGE
health program, and a deep reform of retirement pensions. In order to face
the mismatch between the required resources to fulfill the social program
and the depressed tax revenues in recessive situations, a structural fiscal pol-
icy was implemented, which consists of maintaining an expenditure level
consistent with (estimates of) permanent or medium-term sustainable fiscal
revenue. This instrument not only represents a great formal advance to
guide fiscal policy, but also has desirable macroeconomic properties, in as
much as it allows the authorities to avoid public expenditure curtailments in
depressed periods and to avoid increases when the economy is overheated
and tax revenue exceeds its normal or structural level. The present global
crisis finds Chile with a Treasury that is a net creditor and with a notably
Table IX.1 Investment, savings and growth, 1974–2009 (annual growth rates, % of
GDP)
GDP Potential Output gap Gross fixed Current Gross
growth GDP investment account national
growth deficit savings
(1) (2) (3) (4) (5) (6)
1974–89 2.9 2.5 10.4 13.6 6.9 12.0

1990–5 7.8 7.6 –0.3 18.9 2.3 22.1
1996–8 5.7 6.4 –0.9 22.9 5.0 22.7
1999–2003 2.6 4.0 6.5 19.7 0.8 20.6
2004–5 5.8 3.9 4.3 22.7 –1.7 22.8
2006–9 2.8 4.2 4.9 26.2 –2.8 23.7
Source: Based on Chapter I and on National Accounts from the central bank.
Columns (1)–(4) are in 2003 constant prices. Columns (5) and (6) are based on current prices, with
the same adjustment made in Table VIII.2. Data for 2006–9 in columns (2) and (3) were taken from
potential GDP growth, estimated by the Ministry of Finance, with inputs from the Trend GDP
Committee. Column (6) includes fiscal savings, particularly large in 2004–8.
1
While GDP rose by 28% between 2003 and 2008, gross national income jumped
43%, pulled up by the sharp improvement in the terms of trade.
238 Economic Policy after the 1999 Recession
high liquid stock in stabilization funds, crucial positive features in this criti-
cal situation for international financial and trade markets.
Here I examine the origin of the break in the growth trend located in
1996–7 (section 1), the breaking point in 1998–9 (section 2), the depressed
years 1999–2003 (section 3), the recovery of 2004–5 (section 4), and its
weakening in 2006–9 (section 5). Section 6 analyzes some key issues for
macroeconomic policy: the structural fiscal balance and its relation with
trend GDP, and the exchange rate and counter-cyclical policies. Section 7
presents some conclusions.
1 The origins of the break
In general, up to 1995 the stabilizing macroeconomic policies established

with the return to democracy in 1990 had discouraged the most volatile
components of capital inflows and prevented an excessive real exchange
rate appreciation. But, even though a larger supply of external financing
to successful emerging economies was recorded after 1995, the authorities
maintained the intensity of the policy instruments being used. The obvious
consequence was that, in the following years, the capital inflows surpassed
the capacity of domestic markets to absorb them efficiently and without
generating future unsustainable imbalances. Indeed, the central bank was
unable to prevent a sharp real appreciation of the peso; in fact, the RER
average of 1996–8 was 19% below the average of 1990–5 (see Chapter VIII,
Figure VIII.1). This appreciation, as well as high liquidity, strongly stimu-
lated aggregate demand; that stimulus, given the appreciation, was biased in
favor of the demand for tradables, which contributed to expand the current
account deficit that peaked at 4.8% of GDP in 1996–7.
Evident signals of key macroprice misalignments emerged, and it became
necessary to fortify the instruments in order to face the excessively abun-
dant supply of financial inflows; but the central bank did not show suf-
ficient commitment to the defense of a competitive real exchange rate. In
fact, the monetary authority increasingly privileged the anti-inflationary
target (obviously facilitated by real appreciation), underestimating the
increasing external vulnerability that was being generated. Recall that then
there was a contagion of extreme optimism in international financial mar-
kets with respect to the emerging economies. The opinion prevailed that
after the effective control of the spread of the contagion of the tequila crisis,
the probability of financial crisis was vanishing, being replaced by the great
contagion of overoptimism in financial quarters.
Additionally, a series of measures by the central bank liberalizing out-
flows, specifically related to pension and mutual funds, accentuated the vul-
nerabilities, as strongly verified in 1998–9. However, it must be pointed out
that the bank had not yet dismantled its policies with respect to the capital
account, in contrast to several other LACs and Asian countries that had
already done so prematurely; but it lacked the decisiveness to strengthen

the comprehensive macroeconomic approach prevailing in the first half of
the decade.
A controversial issue has been the fiscal responsibility for the excess of
aggregate demand of 1996–7. Neoliberal advocates attribute the later fall of
GDP to the fiscal performance before the crisis, a hypothesis that is contra-
dicted by robust data. It is true that a rise in fiscal expenditure was recorded
in that biennium (nearly 7% annually). Nevertheless, (i) the fiscal expendi-
ture that had a macroeconomic impact represented less than one-fifth of the
economy; therefore, a fiscal contribution to moderate the excess of global
absorption would have been evidently insufficient by itself. (ii) Second, the
main components explaining the growth in public expenditure were educa-
tion, justice, and infrastructure: all fields where transformations were force-
fully demanded by society, with a political consensus on a greater increase
in their budget, pari passu with tax increases to finance them. (iii) Third,
the fiscal budget showed an increasing surplus (2.1% of GDP in 1996–7),
whereas the private sector exhibited an increasing deficit (6.4% of GDP; see
Chapter I, Table I.3); consequently, fiscal policy was not the leading source
of the imbalance. Moreover, the government had restrained its access to
credits of the World Bank and the IADB and had prepaid debt. Thus, once
again as in the 1970s, the imbalances originated in external shocks and
were located, predominantly, in liabilities of the private sector (see Marfán,
2005).
The responsibility of the government, at most, was the lack of greater
efforts to enforce coordination with the autonomous central bank.2 These
problems, related to the independence of the central bank, were taboo
subjects. The fear of debate ignores the fact that there is no unique form of
autonomy in the world. In fact, many alternative variants exist.
Although the renunciation of active handling of the reserve requirement
and other prudential measures attracted the imbalances, the intermediate
option to maintain the status quo, instead of the extreme option of full capi-
tal account liberalization, significantly contributed to moderate the macr-
oeconomic disequilibria being built. On the one hand, the external deficit
was limited until 1995, thanks to which the stock of external liabilities had
a moderate level in that year and subsequent imbalances were limited to the
rather brief period until the arrival of the Asian crisis. Furthermore, the URR
had some restrictive effect in 1996–7. Therefore, if those policies had not
been applied, the vulnerability of the Chilean economy and the resulting
costs of the Asian crisis would have been much greater.
2
The lack of coordination between the government and the central bank was evident.
As said, the monetary authorities did not show that they cared about the imbalances
in the external sector; however, on several occasions, the government made explicit
the need to strengthen the URR in order to moderate exchange rate appreciation.
2 The impact of the crisis, 1998–9
The Asian crisis interrupted the period of the greatest economic boom in
the history of Chile. The economy had to adjust in response to the two
negative external shocks faced from 1998 – the deterioration of the terms of
trade and a drop in capital inflows. As said, the contagion from Asia found
the economy in a vulnerable position, with a sharp real appreciation, a
dangerous climbing in the current account deficit, and facilities for capital
outflows by residents.
The excessive external deficit implied, by definition, an excess of aggre-
gate demand over potential GDP. Moreover, actual GDP exceeded to some
degree potential GDP, with economic activity operating at a speed some-
what faster than the sustainable one. Therefore, it became necessary to
reduce the rate of expansion of aggregate demand and, to a more moderate
degree, economic activity, in order to diminish the external deficit by about
two points of GDP and to close the “positive” output gap (excess of actual
GDP over potential GDP, which naturally is not sustainable). Even without
the contagion of the Asian crisis, some adjustment in the speed of those
variables was needed.
The symptoms of the crisis emerged in Chile by mid 1998, and they were
felt with force during 1999. The contagion occurred through two channels.
On the one hand, there was a trade shock with worsening terms of trade
and falling rates of export expansion. On the other hand, the large inflows
of 1996–7 gave way to outflows of both domestic and external funds. As a
consequence, and given an outlier exchange rate, strong depreciation expec-
tations arose; the central bank decisively resisted them during 1998, due to
its fear of inflationary pressures in a still overheated economy, and with the
explicit purpose of facilitating the amortization of the dollar-denominated
debt of domestic economic groups. In order to resist devaluation pressures,
the bank resorted to drastic rises in real interest rates.
Nevertheless, it was evident that the exchange rate was too appreciated.
In that context, a rise in the interest rate contributes to restraining aggre-
gate demand, but it hardly restrains the demand for dollars. For example,
if we assume an expected 20% devaluation, then economic agents can
borrow even at a high annual interest rate of 20%, buy dollars, and obtain a
positive return, even if the devaluation is delayed for nearly twelve months.
Undoubtedly, in 1998, both the devaluation horizon and the interest rate
were below those assumed in this simple exercise.
Since the strong rise of the interest rate was insufficient, the bank was
forced to make massive foreign currency sales at the artificially low market
price. Soon, the bank drastically reduced the width of the exchange rate
band, together with an intensification of the domestic adjustment led by
increases in the policy real interest rate to 14.5%; this was done to stress that
the central bank would not give in to market devaluation pressures. After
expansion of the band width by the end of 1998, its use was suspended in
September 1999, starting a free floating regime, now in a context of very
depressed domestic expenditure and strong peso depreciation.3
The recessive environment prevailing from late 1998 performed as an
effective control on inflationary expectations; it allowed sizable depreciation
without threatening the inflation target. On the other hand, since then the
exchange rate has exhibited significant volatility; for example, successively
guided by news on the economies of Argentina (the economic and political
collapse in the early 2000s) and Brazil (the election of President Lula, a leftist
leader, in 2002), the price of copper, and global uncertainties (the attacks on
New York on September 11, 2001, and the wars in Afghanistan and Iraq).
In this critical context, not only was there a sudden stop of foreign loans,
but also massive capital outflows by residents took place. In fact, from the
outset of 1998 a voluminous exit of flows was recorded, mainly from pen-
sion funds (AFPs), which could speculate against the peso when expecta-
tions shifted from appreciation to depreciation. The previous liberalizations
and the liquidity of their funds facilitated that pro-cyclical action (see
Figure IX.1). AFP outflows between January 1998 and June 1999 climbed
to the equivalent of 4.8% of GDP and to 12% of their total funds.4 That,
18 100
Appreciation zone Depreciation zone
16
95
Outflows (% of total funds)
14
RER (1986 = 100)

12 90
RER
10
Investment 85
8 abroad
limit
6 80
Investment
4
abroad 75
2
0 70
Dec-93
Mar-94
Jun-94
Sep-94
Dec-94
Mar-95
Jun-95
Sep-95
Dec-95
Mar-96
Jun-96
Sep-96
Dec-96
Mar-97
Jun-97
Sep-97
Dec-97
Mar-98
Jun-98
Sep-98
Dec-98
Mar-99
Jun-99
Sep-99
Dec-99
Mar-00
Figure IX.1 Pension fund outflows and RER, 1993–2000 (% of total funds, 1986 100).
Source: Based on central bank and Pension Fund Superintendence figures.
3
The real exchange rate depreciated by 22% between September 1999 and the annual
average of 2003.
4
A premature financial liberalization runs the risk of leaving too many open doors
for outflows, which usually become massive in the case of financial distress and
increasing depreciation expectations (see Ffrench-Davis et al., 1995). Indiscriminate
liberalization of the capital account usually tends to bring exchange rate and
macroeconomic instability; it makes more painful the international financial crises.
naturally, had an intense recessive impact on monetary liquidity and aggre-

gate demand.
The economic and social costs were significant. Domestic output adjusted
downwards from the sustained 7% annual growth trend to an actual GDP
drop of 0.8% in 1999. The investment ratio fell strongly (by 18% in 1999),
and open unemployment climbed to 10% of the labor force.
Thus, once again, after the macroeconomic imbalance led by excessive
capital flows in 1996–7, a recessive adjustment, costly in terms of growth
and equity, took place. During a decade, the economic agents had observed
annual real increases of aggregate demand around 7–9%, and suddenly they
were facing a 6% drop in 1999, with output decreasing by the mentioned
0.8%. In the meantime, productive capacity continued expanding, due to
the still high investment of 1998. Consequently, given the fall of actual GDP
and the rise of potential GDP, a strong recessive output gap emerged. After
that, a significant output gap prevailed for many years, reflecting a severe
flaw in macroeconomic policies. The neoliberal two-pillar macroeconom-
ics, based on the control of inflation and a fiscal surplus, was fully at work,
but the third pillar, of a sustainable exchange rate and an effective demand
consistent with productive capacity, failed.5 In 1996–7 that capacity was
exceeded moderately; to the contrary, there was a significantly depressed
effective demand in 1999–2003, revealing a standard asymmetry of effects,
with a net recessive balance even persisting in the subsequent years (see the
output gap in Table IX.1).
It is relevant to emphasize that, after 1998, the domestic financial system
did not radically worsen its situation: the non-performing portfolio of banks
as a percentage of total lending increased from 0.97% in December 1997
to 1.8% at its worse moment in April 1999, which is similar to the level
recorded in 1992, a year without crisis as reported by the Superintendence
of Financial Institutions (SBIF). This is remarkable, in the context of a 6%
drop in aggregate demand, and reflects the stability dividend of the rather
strict prudential regulation and supervision in force from 1986; it shows
how beneficial has been the capacity of the authorities to resist the forceful
and repeated demands to relax regulations.
However, the Chilean experience shows that prudential counter-cyclical
regulation is essential, but not enough to avoid the recessive impact of
external shocks. First, there is usually a sudden stop of access to credit for
In 1998–9, pension funds and residents investing abroad (the so-called chapter XII)
were the main determinants of international reserve losses, and of the sharp recessive
adjustment of domestic demand (Ffrench-Davis and Tapia, 2001). See Zahler (2006)
for an excellent analysis of pension funds’ macroeconomic implications.
5
Effective demand is understood as that part of aggregate or domestic demand that
falls on domestic output plus the demand for domestic output by foreigners (the
value-added to GDP by exports).
small and medium enterprises; second, a growing share of resources are

channeled outside the formal financial system; third, beyond the quality
of prudential supervision, macroeconomic imbalances that abruptly end in
massive devaluations, the emergence of very high interest rates, and bub-
bles in stock markets and real estate can suddenly deteriorate the banking
portfolio. Sustainable macroeconomic balances are an unavoidable partner
of sustainable (counter-cyclical) prudential regulation and supervision.
In the particular analysis of the 1999 recession, there is a series of variables
that explain the benign evolution of non-performing portfolios in domes-
tic banks, beyond the new regulation and supervision established after the
1982–3 crisis. Additionally, the management of capital flows had contrib-
uted to moderate indebtedness of banks and firms up to the mid-1990s. The
effect of excessive inflows in 1996–7 was compensated by the central bank’s
delay in correcting the exchange rate, because it gave time to large debtors
to reduce their dollar liabilities with a cheap dollar; naturally, at the expense
of the central bank balance. SMEs (which operate mainly in national cur-
rency) suffered a drier financial market, being proportionally more affected
by the 1999 recessive gap; the negative impact on SMEs explains part of the
weaker GDP performance from 1999.
3 The recessive phase, 1999–2003
During 1998 a long recessive adjustment began. Although from a short-run

perspective the adjustment was very expensive, in a historical context the
fall of 0.8% in 1999 was remarkably modest in comparison with 17 and 14%
GDP drops in 1975 and 1982, respectively. Therefore, given the domestic
and external conjuncture, what we have to explain is not why GDP fell in
1999, but why it decreased so mildly.
The solid position of the domestic financial system has already been
stressed. Despite the real macroeconom ic worsening recorded in 1996–8,
the benefits of the active capital account regulation implemented in previ-
ous years had left high international reserves, reduced external liabilities,
and a less volatile composition; moreover, the strict fiscal discipline implied
a decreasing national debt and a strong fiscal surplus (2.1% of GDP) in the
biennium before the Asian contagion. Consequently, now Chile was in a
comparatively better situation to face the contagion of the Asian crisis.
It can be argued, forcefully, that the sound fiscal performance helped to
moderate the bust at the outset of the recession. Undoubtedly, that fiscal
position contributes to moderate any financial crisis. Nevertheless, in 1981
the Treasury also had an exceptionally positive financial situation, and GDP
fell by 14% in 1982. It follows that the depth of the recession, in both cases,
was strongly associated with the macroeconomic environment and the
accumulated imbalance in the private sector (see Marfán, 2005).
Even though Chile underwent a comparatively smooth recession on this

occasion, the economic and social costs were significant, with output losses
around US$7 billion in 1999 (the gap between actual GDP and the produc-
tive frontier) and a persistent output gap during 1999–2003. Two conse-
quences were (i) a high open unemployment rate (around 10% of the labor
force) and (ii) a drop in the gross investment ratio averaging 3.2 points of
GDP throughout those years with respect to 1996–8. Why was that moder-
ate fall in 1999, in an economy with really sound fundamentals, followed
by such a long recessive situation?
Contrary to more fashionable views, a significant share of the explana-
tion rests on step-backs in the quality of the macroeconomic approach. In
fact, macroeconomic policies experienced significant reforms in subsequent
years. By the early 2000s, the government and the authorities of the central
bank decided on the conformation of a new macroeconomic framework
including (i) a fiscal policy guided by a structural surplus rule (see section 6),
which reinforced the responsible behavior of fiscal policy in the 1990s and
a stable trend for public expenditure; in the opposite direction, the three
other components shared pro-cyclical implications for economic activity:
(ii) a monetary policy focused exclusively onn fulfilling the inflationary
target (a range from 2 to 4% annually, though in a “long” horizon of 24
months), eliminating price indexing in monetary policy;6 (iii) full exchange
rate flexibility; and (iv) full capital account liberalization. Consequently,
most of the remaining capital account regulations were eliminated. In April
2001, the use of the URR (then with a 0% rate) was suspended, declaring
that the full financial opening of the economy had been achieved.7 It is rel-
evant to emphasize that a large recessive output gap then prevailed, result-
ing from a binding external constraint.
Additionally, the regulations on pension fund outflows were increas-
ingly relaxed. After the ceiling on investment abroad was raised, from 12 to
16% of total funds in January 1999 (in the middle of the crisis; see above
Figure IX.1), it was augmented further to 20% in February 2002, to 25% in
May 2003, and, in several other steps, had jumped to 60% by late 2009.
Authorities stated that raising the ceiling would significantly increase the
return for contributing workers. There was no concern, not even in 2008,
about the disequilibria that were being built in international financial
markets; neither was it considered that, if the channels were created in the
6
The bank had been using as monetary policy tool a rate indexed to the CPI (actu-
ally, to the Unidad de Fomento, UF). In 1999 it was replaced by a nominal interest
rate, in the search for further reductions in inflation. In my view, the UF had been
an extremely effective tool for generating some segments of long-term savings and
lending. See Shiller (2008) for a strong similar view.
7
It must be noted that the policy tool of the reserve requirement is still available in
the face of future capital surges if the authorities decide to make use of this counter-
cyclical regulation.
domestic capital market, those savings could generate higher returns in

Chile than in more capital-intensive mature economies.
In parallel with these policy changes, a large output gap still persisted in
2004–8.8 In 2000–3 potential GDP increased by about 4% annually, whereas
actual GDP growth averaged 3.5%. Evidently, on every day that actual out-
put grew less than 4%, it accumulated a larger gap; the opposite to what the
policy sought, as real macroeconomic equilibrium mandated. Consequently,
there arose two negative market adjustments: (i) part of the underutilized
productive capacity became progressively destroyed and (ii) the generation
of new capacity was weakened by depressed investment and innovation; in
both adjustments, the SMEs and less trained workers were the greatest losers.
As I have shown repeatedly, a significant gap between actual GDP and the
production frontier is usually followed by a drop in productive investment.
As in Mexico and Argentina in 1995, and Korea in 1998, in Chile the invest-
ment ratio diminished substantially, and in 2003 it was still three points
below the 23% recorded in 1998. That was a consequence of an economy
that, in spite of facing historically low real interest rates in 1999–2003, was
working with a high output gap, averaging some 6% of GDP in 1999–2003
(see Table IX.1). That gap reflects a real imbalance, because the economy
underutilized a significant share of the available productive resources.
As shown, the larger the recessive gap, the larger the drop experienced
by investment ratios. Further, a lower investment ratio affects the future
growth of potential GDP and the quality of the labor market.
In fact, the output gap together with the depressed investment ratio also
had a deepening impact on labor markets. In the five and a half years in the
period from mid-1998 to 2003, employment (including special programs
financed by the government) grew by barely 3.3%, while the population
aged eighteen or over increased by 9%. A main determining variable of
the labor market worsening was the macroeconomic imbalance as I have
defined it here: the high gap between actual and potential GDP, which
implied underutilization of labor and capital in that recessive period, and
discouragement of additions to the stock of capital. These factors moved the
Chilean economy away from the 7% annual rate of expansion in the nine-
ties (for both actual and potential GDP), downgrading it to the 4% plateau
during the first half of the 2000s.
Nevertheless, the sound macroeconomic foundations of previous years
allowed the authorities, with political decisiveness, to intensify advances in
8
Vulnerability to external shocks remains a severe problem for emerging economies.
The drop in 1998–9 as well as the ephemeral recovery of 2000 and the recession in
2001–3 involved the majority of LACs population. As is well known, there has been a
significant convergence of policy approaches in Latin America towards the macroeco-
nomics of “two pillars,” successful in obtaining low inflation and fiscal responsibility,
but failing with respect to growth and equity.
social expenditure during the recessive years. A comprehensive program to

support the poorest households (Chile Solidario) was created, and an ambi-
tious health reform (AUGE) started up. This was possible, partly, thanks to
the new scheme of structural fiscal budget, which avoided pro-cyclical fiscal
reactions in those recessive years. In this way, the government could imple-
ment new focused social programs that allowed a continuation of poverty
reduction (the number of poor people fell to 19% of the population in 2003)
in spite of the recessive environment.
Since in 1999 it became increasingly evident that the macroeconomic
pull from abroad, expected by the authorities, would be delayed, I argued
that a domestic shock should be implemented, taking advantage of all the
accumulated strengths of the Chilean economy: (i) a fiscal rule of structural
balance, which implies restraining expenditure increases in boom periods,
and should be able to accelerate them during busts; (ii) a moderate total
external debt, with a low share of short-term liabilities; (iii) a central bank
without external liabilities and with high international reserves (free of
mortgages); (iv) a government with a minimum debt and ordered fiscal
accounts, with an actual surplus of 1.9% of GDP (average 1990–8); (v) a low
actual and underlying inflation rate, persistently below the center of the
inflation band;9 and (vi) a private sector able to produce around 5–7% more
than it was producing in those years.
Thus, there were enough conditions to generate a reactivating domes-
tic shock in the private sector, since, additionally, the real exchange rate
appreciation and the excessive external deficit of 1998 had already been
corrected in 1999. In fact, it was already known that a positive domestic
shock, spurred by the government through monetary and fiscal policies, had
been highly successful in Korea and Malaysia, under a similar recessive situ-
ation and rather similar macroeconomic features (see Mahani et al., 2006):
after the abrupt GDP fall in 1998, both Asian nations experienced a sharp
recovery in 1999, led by their decision to correct the real macroeconomic
imbalance.
In contrast, national economic authorities rejected the proposal of a reac-
tivating domestic shock, arguing that financial markets would negatively
evaluate that action and, therefore, would worsen risk ratings and raise the
spreads charged to Chile. The negative impact of this, they argued, would
be greater than the positive effect of the reactivating domestic shock. Most
business press and opposition politicians supported and praised the official
approach. Moreover, they criticized, though mildly, the actual fiscal deficit
resulting from the implementation of the structural 1% surplus. Fortunately,
the government was able to sustain the structural principle adopted formally
in 2001, and thus progressed from a pro-cyclical fiscal policy to a neutral
9
Annual CPI inflation averaged 2.2% in 2001–4, below the 3% center of the 2–4%
band.
one, by maintaining the trend of expenditure notwithstanding drops in

revenue during recessive situations. It failed to move beyond this, towards
a strong counter-cyclical increase in expenditure.10 While it was estimated
by a committee of experts that potential GDP was increasing by around
4% (see section 6), actual GDP rose by just 2.6% during the quinquenium
1999–2003; a poor performance, but better than the 1.3% recorded by Latin
America overall.
Unfortunately, the fact is that recovery only arrived in 2004, led by a posi-
tive external shock via export prices. The evident lesson was that the step-
back experienced in the quality of macroeconomic policy must be corrected
to achieve sustained growth with equity. The uncorrected policy failures
once again exerted negative effects from 2006 (see section 5).
4 Dynamic recovery in 2004–5
Through 2003, international markets provided a significant improve-

ment in commodities prices and trade volume, which implied a strong
positive external shock in economies intensive in the production of natural
resources. In fact, Chile benefited from a remarkable increase in its terms of
trade, equivalent to roughly 10% of GDP between the recessive 1999–2003
period and the dynamic 2004–5; a significant share was retained abroad
as profits of FDI exporters. But, in all, there was a sizable improvement for
local exporters and the public sector. This positive exogenous shock raised
aggregate demand and, with it, the average growth rate of actual GDP from
2.6% in 1999–2003 to 5.8% in 2004–5. Since potential GDP was moving
around a 4% plateau in these years, with that 5.8% actual growth a signifi-
cant reduction of the output gap was recorded (the gap fell by around 3–4
points of GDP).
The fact that the 2004–5 actual rate of growth doubled the average of
1999–2003, without any change in structural variables, was a robust proof
that the output gap prevailing since 1999 was mainly a macroeconomic
failure, which was relaxed with the positive external shock. As shown, the
domestic failure was not in fiscal policy, but was placed, principally, in the
wrong signals provided by the central bank to the private sector. Chile had
in those years, after having corrected the excessive external deficit and the
appreciated exchange rate, all the objective conditions to carry out by itself
the positive shocks required to offset negative external shocks suffered from
1998. That would also have changed the negative expectations.
Naturally, the intensity of recovery was also based on the merits accumu-
lated by the Chilean economy in the preceding years, but the predominant
10
There was a relevant expansive action through the emergency employment pro-
grams. This and some other counter-cyclical expenditures are reflected in the fact that
the structural fiscal surplus averaged 0.65% of GDP in 2002–3, below the 1% target.
force was the positive external shock.11 This reveals a macroeconomic

weakness (in domestic policies as well as in the international financial
architecture), since those merits were already present during the 1999–2003
recessive situation. Along with economic recovery, though with a lag, the
investment ratio started to rise vigorously.
The fiscal structural rule continued in force. The actual deficit of 0.7%
in 2001–3 moved to a huge 4.6% surplus in 2005. The main determi-
nant of such an outcome was a notable jump in the price of copper (see
Figure IX.2). The outstanding increase of the actual price implied increas-
ing profits for CODELCO (the public copper firm) and taxes paid by private
copper producers, both enlarging the revenue collected by the Ministry of
Finance. The Treasury continued to keep in action a Copper Stabilization
Fund of proceeds from CODELCO, which was a natural counterpart of the
structural fiscal rule; in 2005, the fund accumulated the equivalent of the
gap between the actual average price of 167 cents per pound and the esti-
mated trend price of 93 cents of sales by CODELCO.12 The Treasury was,
rapidly, passing from debtor to creditor of the nation and of the world.
500
Actual copper price Trend copper price Committee
450
400
350
300
250
200
150
100
50
0
1959
1961
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
Figure IX.2 Real copper price, actual and trend, 1959–2009 (2009 US$/lb).
Source: Calculations based on Ffrench-Davis and Tironi (1974), central bank and Budget Office figures.
The long-term trend series is estimated with a Hodrick–Prescott filter (lambda 100) until 2000; in
order to avoid the end-of-sample bias of the filter, from 2001 to 2009 the annual average growth rate
(4.4%) of the real price between the actual price in 2000 (2009 US$1.30 per pound) and the trend price
reported by the Committee for 2009 (2009 US$1.99) is used.
11
It must be emphasized that similar positive external shocks allowed the average GDP
growth rate of Latin America to jump from 1.3% (1999–2003) to 5.6% (2004–7).
12
Estimate in 2004 for the structural budget of 2005, made by the Copper Price
Committee, an independent pluralistic counsel of the Minister of Finance. The esti-
mate in 2006 for the budget of 2007 was 121 cents per pound of copper. The structural
Naturally, generalized improved terms of trade and fast rising volumes of

exports enhanced, directly, the spending capacity of the private sector, and
expectations returned to optimism. Given a large output gap by the start
of the positive external shock, domestic supply was able to respond with a
rising GDP and low inflation pressures (within a target band of 2–4% set by
the autonomous central bank). In the meantime, the bank allowed the real
exchange rate to appreciate by 20% between March 2003 and December
2005, reinforcing an increase in aggregate demand, despite the gradual rise
in the interest rate and the oil price. But, given a significant appreciation,
aggregate demand became increasingly import-intensive.
5 Deceleration from 2006
By late 2005, a generalized optimism was reflected in expectations of con-

tinued growth of actual GDP around the 6% plateau. Estimates of potential
GDP dynamism were rising, supported by a significant increase in the
investment ratio. Improved expectations of businesses, after the custom-
ary lag, were responding to the relevant reduction that had been exhibited
by the output gap, and the prospect that the dynamism of economic activity
was to be sustained.
However, after the recovery experienced in 2004–5 came a period in
which actual GDP has persistently grown below the expansion of potential
GDP, even placing the Chilean economy below the average speed of Latin
America, thus losing the lead that it had held in the region since the late
1980s. It is true that by 2008 Chile still exhibited a much better record since
1990, with the average 5.3% vis-à-vis the 3.2% of LACs, but in the margin
(2004–8) it was losing ground as documented below.
Different explanations have emerged, such as (i) the arrival at maturity,
(ii) the increased obstacles to growth set by the government and the lack of
more neoliberal reforms, (iii) a lack of heterodox reforms, and (iv) failures
in the macroeconomic approach.
By late 2005, consensus expectations for growth in 2006 were perfectly
feasible.13 However, after a strong start, economic activity lost momentum.
In all, instead of the output gap being reduced by about 1% of GDP it was
increased by a similar figure. That implied employment, wages, and profits
that were lost forever; additionally, some uncertainty was reintroduced to
entrepreneurs’ minds, particularly exporters and small entrepreneurs, and
optimism was weakened.
or trend estimate in 2008 for the 2009 budget (at the 2009 price level) was 199 cents.
See Table IX.3.
13
In July 2005, within the framework of the trend GDP estimation for the structural
fiscal balance, an output gap of 1.7% was reported by the Ministry of Finance for that
year. If the gap is calculated under the concept of potential GDP, the recessive gap
rises to around 4–5% (see Chapter I, Annex).
It cannot be argued that the slowing could be explained by having reached

maturity. Indeed, the Chilean per capita GDP (at PPP), in spite of the progress
achieved in the past two decades, reached merely one-third of the level of the
USA. A weak performance in recent years – especially given the spectacular
terms of trade that Chile was facing – was the result of a complex combina-
tion of macroeconomic and microeconomic features, and a lack of long-term
perspective.
The conventional interpretation of most of the opposition was that the
government had put a brake upon the economy with red-tape bureaucratism
and a lack of additional privatizations; particularly, it was stressed, those of
CODELCO and of Banco Estado (the commercial public bank). With respect to
the privatizations requested by the opposition, CODELCO has provided huge
income to the Treasury, especially in the period to be explained, while Banco
Estado played, among domestic financial institutions, the main counter-
cyclical role during the 2009 recession, particularly lending to SMEs. With
respect to bureaucratism, obstacles to entrepreneurship, and so on are a real
challenge on the road to the development of Chile. But (i) Chile fares reason-
ably well in international comparisons of developing economies, and (ii) the
structural and bureaucratic environment for entrepreneurship and innova-
tion was not notably different to 2004–5 when growth was significant, and
even, I would say, rather similar to that of the golden period of the 1990s.
Something rather similar or somewhat worse cannot explain a sharp worsen-
ing of the dependent variable.
The available information shows that the main causes of this lost dyna-
mism since 2006 (beyond particular situations in the mining and energy
sectors)14 are not a lot of new micro-restrictions or “bottlenecks” imposed
by the government, but mainly failures in two strategic areas, and a set of
specific situations to which I turn later. One area refers to variables affect-
ing the macroeconomic environment and depressing the expectations and
dynamism of both private and public economic agents; actual total factor
productivity (TFP) suffers with underutilization of installed capacity. The
second strategic area relates to the economic development agenda, which,
notwithstanding recent progress in innovation policies, has been weak and
hesitant; innovation and, consequently, structural TFP suffers.
(a) The macroeconomic environment

I have argued that macroeconomic failures have been at the heart of the
worsened performance of the economy after the golden years. The case is
illustrated with an analysis of 2006. I believe that 2006 was a crucial year for
consolidating creative expectations. I summarize the depressing role of four
14
Such as a strike in the copper mine La Escondida, a landslide in the main mine of
CODELCO (Chuquicamata), and shortages in natural gas supply from Argentina.
policy variables that contribute to explain why the impulse was weakened
instead of strengthened:
1 A premature upward adjustment of the interest rate by the central bank,

mainly during 2005 and 2006. Inflation had risen from a 1% average in
2004 (below the 2–4% band set by the central bank) to 4% in 2005 (in the
upper border of the band). The bank overreacted, bringing down infla-
tion, by late 2006, below the center of the band. Inflation (both total and
core inflation) was restrained at the expense of economic recovery, with
a strong imbalance between economic policy objectives.15
2 An excessive exchange rate appreciation and its negative impact on trad-
able production, as is demonstrated by a sizable increase in imports and
a decreased demand for domestic output. As Figure IX.4 exhibits, the
increase in the quantum of imports had been notably larger than that
of exports since 2004. Exchange rate appreciation and the reduction of
the actual tariff paid by importers explain why domestic demand became
much more import-intensive in 2006;16 I re-examine this issue below.
In fact, in that year domestic demand expanded by 6.4%, but since the
import quantum grew by 10.5%, the local demand for domestic output
only involved a 3.6% rise.17 Again, there was a severe inconsistency
between the current exchange rate policy and the consensus objective of
fostering exports with higher value-added.
3 The authorities sterilized a substantial share of the expansive effect of the
high copper price; meanwhile most of the recessive impact of a higher
oil price was allowed to operate. Naturally, the higher oil price, which
averaged US$56 and 66 per barrel in 2005 and 2006, respectively, caught
an increasing share of the Chilean budgets. Part of the copper revenues
should have been used to compensate for that effect; for instance, by
15
The core CPI, which excludes oil and perishable food, in December 2006 exhibited
an annual inflation of 2.6%. That is, below the center of the 2–4% target. In the
course of 2006, nevertheless, the core CPI moved close to the ceiling of the band, but
during 1999–2004 it was usually below the center.
16
In 2002, imports from Korea, Europe, and the United States faced a uniform tariff
of 7%. The actual average tariff paid by all imports was reduced from 3.8% in January
2003 to 1.6% in 2006. The positive counterpart for national production is that the
lower tariff reduces the production costs of exports, and tariff preferences with partner
countries facilitate sales abroad. Exports continued leading GDP growth with a 7.6%
rate in 2004–6. Nevertheless, a decreasing trend is observed (13.3% in 2004; 4.3%
in 2005; 5.1% in 2006), because of some particular problems and likely due to the
cumulative exchange rate appreciation and the elimination of the subsidy to non-
traditional exports (see Chapter VI).
17
Notice that, as discussed in Chapter VI, part of imports is used to produce exports,
so that they are re-exported. It must be kept in mind that GDP is the sum of the
value-added of exports (which of course are sold abroad), and the rest of GDP (which
is sold in the domestic market).
subsidies to the public transportation of lower income population, by

making an existing Oil Stabilization Fund work more effectively, or in a
simple stabilizing adjustment (a transitory, not a permanent reduction)
of the oil-specific tax.18
4 The insufficient power of the interesting structural fiscal balance policy,
which advanced from pro-cyclicality to nearly neutrality, but without
reaching strong counter-cyclicality (as argued in section 6). The stabili-
zation of fiscal expenditure constitutes great progress, but is still insuf-
ficient as an input of the sector for macroeconomic stability. In order
to improve the macroeconomic performance, Chile needs to carry out
additional corrections with a more comprehensive approach. An evident
correction would have been to revise the 2006 public budget, readjusting
social and productive investment; either with a drop in the target of a 1%
structural surplus or by re-estimating structural tax proceeds with a cop-
per price of 121 cents (the revised estimate, in 2006, of the trend price).
The economy was operating in 2006 as if the trend copper price were
only 99 cents. This set of facts contributes to explain why the domestic
economy grew at a slower pace during 2006, notwithstanding the posi-
tive external situation and the domestic output gap.
During the next biennium, effective demand experienced several ups and
downs, affecting the stability of economic activity; the significant effects of
demand on the response of real supply (that is, actual GDP) reflected the fact
that the Chilean economy was operating persistently below the production
frontier.
Evidently, in the years before the global crisis there was a complex interna-
tional scenario. There were positive features for Chile, such as the spectacular
price of copper and other large exports, which allowed the Treasury, with
great responsibility, to accumulate sizable funds for eventual bad years (or
awful years such as 2009); fiscal surpluses jumped further to an annual aver-
age exceeding 7% of GDP in 2006–8. Moreover, the world trade volume was
dynamic until the arrival of the international financial crisis. On the negative
side for Chile, international prices of oil and food were climbing. From mid-
2007 to mid-2008, the price of food in the Chilean CPI increased by 22%,
which generated significant inflationary pressures, explaining about half of the
nearly 10% annual inflation recorded at the peak of the commodities boom
(by the third quarter of 2008; see Ramos, 2008). It was, mainly, an imported
inflation. Of course, Chile was not the only importer, since it became a public
bad available to all open economies. The central bank consistently expressed
18
The oil tax is clearly a progressive one from a distributive perspective, and is envi-
ronmentally friendly. I strongly support heavy taxes on fuels for environmental and
distributive reasons. From a macroeconomic perspective, the negative impact of the
increased price of oil should have been compensated for with additional fiscal expen-
diture financed with a fraction of the copper proceeds.
its bias for the inflation target, at the expense of growth. By late 2008, when
Chile was already exhibiting negative monthly inflation, the monetary policy
interest rate (8.25%) exceeded by over 7 points that of the USA.
When the contagion of the global crisis arrived, the government made
intensive use of the tools at hand, principally the stabilization fund, which
provides ample room for counter-cyclical fiscal policy. Notwithstanding a
drop in tax revenue resulting from a depressed demand, some taxes were
reduced transitorily (on fuels, credits, SMEs); tax revenue would have fallen
23% in 2009. In parallel, expenditure increased 18%, which implied a 4.5%
actual fiscal deficit. The strong counter-cyclical fiscal policy, based on the
official structural policy (see section 6a), was the main force compensating
for the negative shocks, principally that on the volume of exports. In fact,
the volume of Chilean exports, which had risen by 7.9% in the previous two
decades, dived 5.6% in 2009. Fiscal policy softened to a significant degree
the multiplication to the domestic market of the shock on exports. The
domestic economy only contracted by 1.1% (see Chapter VI, Table VI.6). For
the first time since the opening of the economy, which started in 1973, the
recessive adjustment in the face of external shocks was stronger on exports
than on the domestic economy. Counter-cyclicality of fiscal policy was
effective and efficient in 2009.
But returning to 2008, before the arrival of the global crisis, in all, as said,
during the previous boom period growth was moderate, not bad, but came
after five years of significant recession (average growth of 2.6%); then, natu-
rally, there was a recessive gap waiting to go into production, which allowed
“an easy stage for recovery.” Thus, the record is low compared with the
potentiality of the domestic economy and the huge positive external shocks.
Additionally, it is noticeable that the Chilean economy grew by less than the
average of Latin America during the long international boom from 2004 to
2008. The economic growth of Latin America averaged 5.3% in those years,
while Chile recorded a 4.9% growth. This rise of GDP was determined by a
6.6% expansion of exports and 4.2% of the rest of GDP. This moderate figure
contrasts sharply with the 6.5% dynamic growth of non-exports in 1990–8
(see Chapter VI, Table VI.6). Systemic competitivity was failing. The failure
was associated with the second variable underlying the weak economic per-
formance: a weak and hesitant development agenda.
The arrival of the global crisis, with its new recessive impact, reinforces
the need to review past experiences and to introduce corrections to the
development strategy. The 1.5% drop of economic activity in 2009, again
provides room for “an easy stage for recovery” during 2010, and room for
introducing pro-development reforms.
(b) The development agenda

Economic growth has slowed, but the social agenda has been particularly
innovative and has contributed to a more inclusive development.
Chile made ambitious social reforms in this period. As detailed in Chapter

VII, pensions and health were substantially improved, unemployment insur-
ance, initially mostly a self-insurance, was strengthened with a meaningful
solidarity pillar, care of children has expanded powerfully, the extreme poor
are better taken care of with Chile Solidario, the quality of housing for poor
and medium-low segments has risen and the subsidies are more generous, and
subsidies for the hiring of young workers have been established. The social
agenda has been quite powerful, well designed, and transparent.
But the strong social agenda was not well matched by the economic
agenda. Evidently, there has been progress in several areas. But a rather poor
economic outcome in the second decade is what needs an explanation and
not the success in the first decade. The poor outcome is mostly explained,
I believe, by significant shortcomings. My view is that these shortcomings
had a relevant influence in the defeat of Concertación Democrática in the
recent presidential (January 2010) and parliamentarian elections. But politi-
cal analysis is beyond the reach of this book.
There is consensus that fiscal responsibility was outstanding, but structural
progressive reforms (as well as macroeconomic management, as shown)
were weak and somewhat contradictory to the introduction of equitability
in the market behavior. The latter required deep reforms in the capital mar-
kets, away from the priority for the overnight markets and moving decidedly
towards enhancing the long-term market segment; developing segments for
SMEs and for entrepreneurs without wealth or history.19 Further, incentives
to innovation were weak, even though they have been taking relevant shape
only recently.20 Labor training for untrained workers has been improving
but too mildly. The sharp increase in the number of years of education have
been associated with lowered quality, which demands even more effective
labor training for the workers that have suffered that faltering quality.
There appears to be a growing shortage of more trained workers and
entrepreneurs, in an economy that has doubled GDP per capita, and whose
requirements for growth are now more demanding. In brief, Chile missed, to
a significant degree after the promising start in the early 1990s, what I call
taking the road that leads from financierism to productivism.
19
There have been several reforms of the capital markets, which have improved access
to financing for SMEs and microcredit. However, the market is still intensive in short-
term and liquid dealings and remains quite limited for SMEs. The Counsel on Equity
designated by President Bachelet (Consejo Trabajo y Equidad, 2008) stressed its con-
cern for the incompleteness of capital markets and their regressive biases.
20
In 2008 it was decided to focus the allocation of the proceeds of a royalty recently
established on mining to a selected group of clusters. It represented a sharp, and
encouraging, deviation from allocative neutrality. See Benavente (2005) and Consejo
Nacional para la Innovación (2007) for substantive analyses of the challenges faced
by Chile and alternatives for action.
Table IX.2 Gross investment and depreciation, 1995–2009 (% of GDP at

constant 2003 prices)
Gross fixed investment Depreciation Net capital formation
(1) (2) (3) ⴝ (1) ⴚ (2)
1995–8 22.8 7.2 15.6

2006–9 26.0 10.6 15.4
Sources: Based on National Accounts of the central bank and Data Base for the Trend GDP
Committee, Ministry of Finance, August 2009.
(c) Some specific situations

It is interesting that gross capital formation, as measured in constant prices
by the National Accounts, had risen substantially. National accounts iden-
tified a sizable drop in the price in pesos of capital goods that are mostly
imported; exchange rate appreciation plus drops in international prices
underlie that performance. Consequently, a given level of domestic savings
could purchase a sizably larger volume of capital goods. The gross invest-
ment ratio reached record levels in 2005–9.
Why was the higher ratio generating lower potential GDP growth? There
are some robust and some tentative explanations. First, it is customary to
look at gross investment figures, while what are relevant are the net figures.
Table IX.2 presents data on gross fixed investment, depreciation of capi-
tal, and net fixed investment, in constant prices, as a share of GDP. Two
subperiods with peak gross ratios are compared: 1995–8 and 2006–9, with
a rise in the gross ratio of over 3 points of GDP. However, depreciation of
capital also experienced a sharp rise, from 7.2% of GDP to 10.6%. Thus, in
net terms, the rise of the ratio becomes negligible. But, still, why is a similar
ratio associated with a lower GDP growth? Additional explanatory variables
are the sudden problems experienced in the supply of energy that affected
the productivity of some energy-intensive installed capacity (see Central
Bank, 2009), an increase in “green” environmentally friendly investment
that tends to increase the cost of projects, and investment in protection
from personal insecurity (an unfortunate world trend related to inequality
and reduced space for social integration).
6 Two relevant issues: the structural fiscal balance and

the exchange rate
(a) Improving the structural fiscal balance21

As part of a counter-cyclical policy package, the concept of structural fiscal
balance (SFB) is an outstanding fiscal component. There are different varieties
21
In Marcel et al. (2001) the basic features of the Chilean rule are exposed. Tapia
(2003) develops an analysis of the macroeconomic implications and proposes a
of SFB, but the essential component is the measurement of the balance

across the business cycle, estimating at each point of time what would be
the public income and spending in a framework of sustainable full employ-
ment of human and physical capital; that is, in simple terms, an economy
making full use of potential GDP (GDP*). If terms of trade fluctuations are
relevant for fiscal proceeds – via profits of public companies or taxes paid
by private exporters – the purchasing power of GDP* should be estimated at
the trend terms of trade. Given a tax burden, those two trends must guide
the sustainable path of public expenditure.
The fiscal instruments used to implement counter-cyclical policies must be
pragmatically chosen (see Stiglitz et al., 2006). During booms, for example, a
reduction in public expenditure will probably be insufficient to compensate
for an excess of of the private sector expenditure led by capital inflows. An
increase in taxes, instead, can directly affect agents with a higher propensity
to spend and spread its effects across the different sectors of the economy.
During an economic downturn, a tax relief may be ineffective under a
depressed macroeconomic environment and a private sector reluctant to
consume and invest. Public expenditures in non-exportables can, in this lat-
ter case, be a more effective instrument – for instance, by accelerating public
works if unemployment in the sector is high.
Fiscal policy ought to be part of a flexible policy package. Given that
emerging economies are especially vulnerable to global economic down-
turns, overreliance on monetary policy may bring poorer macro results, as
compared to a more balanced framework of counter-cyclical fiscal, exchange
rate, and monetary policy, as well as prudential regulation of capital flows.
The use of counter-cyclical fiscal policy needs as a convenient precondition
to be on a path of solvent and sustainable fiscal accounts.
The government of President Lagos implemented, in 2001, a formal fiscal
rule that operates within the concept of structural balance. Conceptually, the
structural balance methodology isolates the impact of the business cycle on
public finances, providing a long-term picture of the fiscal situation in terms
of both income and spending. It consists of maintaining in each annual
budget a level of fiscal expenditure consistent with revenues collected as if
the economy were making “normal” use of the productive capacity (an esti-
mate of a “trend” GDP) and the copper price were at its long-term level.
Therefore, when the economy is overheated, the government naturally
collects a larger than “normal” (to be defined below) tax revenue, but it
does not increase expenditure, thus accumulating savings; and in periods
of bust, the government uses those savings (or borrows) to cover depressed
series of adjustments to improve its effective counter-cyclicality. Recent important

adjustments are presented in DIPRES (2007) and Velasco et al. (2010). The issue is
further discussed in Ffrench-Davis (2010); the research support of Rodrigo Heresi was
crucial in this publication.
tax revenue resulting from the slower economic activity; thus it maintains
the trend of expenditure. This rule implies relevant conceptual progress in
fiscal and macroeconomic management, with respect to the standard policy
recipe in which expenditure follows economic cycles. It is a significant
improvement with respect to the pro-cyclical traditional norm that seeks
to balance the actual fiscal budget each year. Given the strong volatility of
international trade and financial markets, the standard neoliberal recipe is
highly pro-cyclical.
The use of the new fiscal framework requires estimating several key struc-
tural variables, among them (i) the trend GDP growth rate and (ii) the long-
term price of copper. The estimation of both parameters is made with inputs
provided by two independent committees of experts on an annual basis.
The definitions of the key parameters and assumptions have been made
increasingly transparent, the disclosure of information to the public has
been improved, and the methodology has been refined. Table IX.3 shows
the evolution of key fiscal indicators.
Evidently, fiscal responsibility was not a novelty in Chile when this frame-
work was started in 2001. The nearly 2% average surplus in the 1990s, fol-
lowing the return to democracy in 1990, testify to it. The concept of trend
copper price was also already operative, with a copper stabilization fund at
work for over a decade, and a significant precedent established successfully
in the 1960s, though dismissed since the 1970s. But, in this case, the formal
launching of the stabilization approach was quite useful for the quality of
policy, including interesting features, such as the creation of committees of
independent experts to estimate the trend price of copper and inputs for
trend GDP, when Chile was facing a recessive gap from 1999. In fact, the for-
mal adoption of the structural balance took place in a context of a depressed
economy, the macroeconomic need for an actual fiscal deficit, and the
assumption of a new president from a leftist party. The formal and well
advertised launching was an opportune and efficient step from an economic
and political perspective. In August 2006, the structural rule and fiscal policy
that previously depended exclusively on administrative decisions and politi-
cal will were institutionalized. The new law, the Fiscal Responsibility Law,
reinforced both the credibility and the transparency of fiscal policy.
This positive fiscal rule was accompanied by certain features that are not
intrinsic to it, but options in its implementation. Here, I mention three:
(i) the level at which the structural balance is targeted; (ii) the degree of
counter-cyclicality; and (iii) the definition of potential GDP (GDP*) or trend
GDP.
First, a key feature was the level at which the structural balance is targeted.
During the first years, a structural surplus target equivalent to 1% of GDP
was set with the aim of ensuring the accumulation of assets with which to
reduce the liabilities inherited from the debt crisis in the 1980s, and meet
future public sector commitments, including particularly the contingent
258
Table IX.3 Fiscal indicators, 2001–9

2001 2002 2003 2004 2005 2006 2007 2008 2009
GDP growth (%) 3.4 2.2 3.9 6.0 5.6 4.6 4.6 3.7 –1.5
Trend GDP growth (%) 4.1 3.9 4.0 4.2 4.9 5.0 5.3 5.0 4.9
Current copper price (US$/Lb) 71.5 70.7 80.7 130.0 166.9 304.9 322.9 315.5 232.4
Trend copper price (US$/Lb) 90.3 91.2 88.0 88.0 93.0 99.0 121.0 137.0 199.0
Fiscal balance (% of GDP) current –0.5 –1.2 –0.4 2.1 4.6 7.7 8.8 5.3 –4.5
Structural fiscal balance (% of GDP) current 1.0 0.6 0.7 1.0 1.0 1.0 1.0 0.7 –0.9
Fiscal income growth (%) constant 2008a 4.1 0.8 5.3 20.0 19.1 23.1 12.3 –8.3 –23.2
Fiscal expenditure growth (%) constant 2008a 3.6 4.2 1.6 6.1 6.6 6.9 8.7 8.0 17.8
Fiscal income (% of GDP) current 21.7 21.1 20.7 22.0 23.8 25.9 27.5 26.5 20.5
Fiscal income (% of trend GDP) current 21.2 20.2 19.8 21.5 23.4 25.3 26.7 25.3 18.2
Fiscal expenditure (% of GDP) current 22.2 22.3 21.2 19.9 19.3 18.2 18.7 21.2 25.1
Fiscal expenditure (% of trend GDP) current 21.7 21.4 20.3 19.4 18.9 17.8 18.2 20.3 22.2
Sources: DIPRES ( January 29, 2010) and Velasco et al. (2010). The estimates in each year of the trend variables for the next budget year have been
used here. For 2005 the official estimate of the output gap was 1.7% of trend GDP. Potential GDP implied an output gap of 4.6%
a
Deflated by the CPI which, sometimes, notably differs from the implicit deflator of GDP.
liabilities generated by the public guarantee for a minimum pension. In

addition, another argument for maintaining a structural surplus was the
structural operating deficit of the central bank of Chile, as a result of losses
arising from the bailout of the private banking system during the 1980s.
After some years of strong fiscal savings owing to the persistent boom
in commodity prices, the Treasury became a net creditor to the rest of the
world, with growing stabilization funds. By late 2008, the Economic and
Social Stabilization Fund (the successor of the Copper Buffer Fund) and
the Pension Reserve Fund had accumulated the equivalent of 18% of GDP,
while fiscal liabilities were negligible, after the significant amortizations
made with the previous surpluses of the fiscal balance. The socially profit-
able allocation for that 1% of GDP was to finance social investments and
productive development, such as better quality education, training for work-
ers and small entrepreneurs, innovation support, regional infrastructure
improvements, and incentives for long-term financing of SMEs and new
entrepreneurs. All these items were needed to secure faster development
convergence with rich nations.
Thus, it did not made sense any more to continue accumulating money
year after year in a country with social needs that were still high.22
Consistently, the structural surplus target for 2008 was reduced to 0.5% of
GDP. The contagion of the global crisis led to further reduce the balance
targeted to 0% in 2009. In parallel, Chile moved sharply from the rather
cyclically neutral approach to a strong counter-cyclical one. In 2009, it
reached a 0.9% structural fiscal deficit and a 4.5% measured deficit, with an
18% rise in expenditure.
Second, the structural rule allowed the maintenance of a level of expendi-
ture consistent with medium-term trends (estimated or, as to be shown,
rather underestimated) of GDP. This was very positive in the context of the
recessive gap in 2001–3 and 2009, under depressed fiscal revenue. A struc-
tural surplus was consistent with an actual deficit in those years. In contrast
to the neoliberal recipe, it implied sustaining fiscal expenditure in spite of
a depressed GDP, but it did not consider increasing expenditure to offset
recession. Consequently, it implied a fiscal policy that was rather neutral
with respect to the economic cycle, but fell short of being properly counter-
cyclical.
Nonetheless, some counter-cyclical spending was carried out, principally
in the form of emergency employment programs. The Anti-unemployment
Contingency Program, which previously required annual approval under the
22
Given that terms of trade shocks have been strong, the stabilization fund should
take care of stabilizing fiscal expenditure as well as the supply of foreign currency. To
achieve the twin counter-cyclical targets, the effective coordination of the Ministry of
Finance and the central bank becomes essential. While the stability of fiscal expendi-
ture has been exemplary, that of the exchange rate has failed.
budget law, became permanent, with the Fiscal Responsibility Law establish-
ing its objective, conditions and financing. The instrument is available at
national, regional, or local levels. The program can be activated whenever
the conditions established by the law are met: in this case, when the national
three-month rolling average unemployment rate, measured by the National
Statistics Institute (INE), exceeds its average for the previous five months, or
when it reaches at least 10% of the labor force (DIPRES, 2007).
In order to progress towards a more efficient macroeconomic policy, it
is necessary to go further beyond neutrality. It must advance decisively
towards a counter-cyclical approach, what has been done in 2009, under
the challenge of the global crisis in process (in fact, the estimated structural
balance for 2009 is a deficit of 0.9% of GDP; see DIPRES, 2010).
An effective counter-cyclical fiscal approach would involve anticipating
public expenses in recessive situations such as that in 1999–2003, and tran-
sitorily reducing some taxes, such as VAT or social security contributions.
And vice versa, in overheated situations such as in 1989. This would imply a
move towards effectively counter-cyclical taxes and expenditure. The doses
of each component should be associated with the expected effectiveness
of tax changes vis-à-vis expenditure changes, and perceived shortages or
excesses in each of them. Given the low tax burden in Chile (18% of GDP)
and the insufficiency of investment in public works, education, and inno-
vation, an asymmetrical treatment would be recommended: increase public
investment in recessions and raise taxes in booms.23
Third, the definition of the concept of “potential GDP” is essential for
implementing an approach of structural fiscal balance. It can be defined as
a production frontier or GDP* or as trend GDP (GDPt). The former is defined
as the maximum sustainable level of production by the economy; in volatile
economies there usually is a significant average output gap between GDP*
and actual GDP, where the latter is usually below GDP*. In fact, in unstable
economies, actual GDP can be sharply below GDP*, while only exception-
ally can it be above it; this is quite a relevant asymmetry for macroeconomic
policies. GDPt can be defined as the level of production consistent with a
“normal” use (in statistical terms) of productive inputs; that is, the trend
value or permanent component of actual GDP, which implies symmetrical
positive and negative gaps.
Then, the central question is what concept is the relevant one for calculat-
ing the structural fiscal accounts: maximum attainable or trend output? The
methodology of the Ministry of Finance made an explicit option in favor
of the trend GDP concept. The arguments are, first, that GDPt is the most
23
A related subject is that of the tax burden and its composition. On this matter, two
high-priority problems, presently, are the need to dismantle regressive tax exemptions
and combat tax evasion and avoidance; this requires clear messages to taxpayers that
there is a growingly transparent and efficient use of tax revenue.
standard methodology used by “the profession,” and the feature of sym-

metry between positive and negative gaps is assumed to maximize transpar-
ency and diminish risks of lack of credibility.24 Estimates of GDPt evidently,
usually, include the sharp recessions suffered by the Chilean economy. As a
consequence, it is obvious that, in this economy, average GDPt moves well
below GDP* or productive frontier, thus providing biased information for
counter-cyclical macroeconomic policies. Consequently, the main disadvan-
tage of using the trend measure is the weakening of counter-cyclical policies.
In practice, economic cycles are not symmetrical from the point of view of
the duration and intensity of booms and busts; moreover, as said, they are
not symmetrical with respect to the production frontier. Thus, although
statistically a trend of recent history can be obtained, it frequently does not
reflect a state of “normality.” Actually, if macroeconomic policy seeks to
reach GDPt and to stop there, that involves maintaining an output gap with
respect to GDP* that discourages capital formation.
On the other hand, if the economy falls below the estimated trend for a
relatively long period, the trend growth rate itself would tend to fall further,
implying a spurious reduction in the output gap (a “lowering the bar” effect),
and, therefore, a weakening of overall counter-cyclical policies. Thus, guiding
monetary and fiscal policies according to GDPt instead of GDP* results in a “self-
fulfilled prophecy” that depresses the future productive frontier; Figure IX.3
shows clearly the downward adjustment in the trend GDP growth rate made
by the committee, even before the global crisis, maintaining an output gap
with respect to GDP* that discourages capital formation. But, naturally, it
contributes to controlling inflation with greater force. Those policies should
be reformed, in order to place economic activity closer to the level of potential
GDP. Interestingly, year after year the authorities have made changes in the
methodology used which have reduced partly the differences with GDP*.
In summary, the Chilean fiscal policy has evolved in the past two decades
to combine discipline, transparency, and macroeconomic management.
Since 2001 there has been a rule that, in spite of several shortcomings (such
as the insufficient intensity of counter-cyclical effects throughout its life),
has served to avoid the standard pro-cyclical bias and give stability to public
expenditure. As the concept of structural budget has gained credibility, it
has been easier to introduce improvements and windows of discretion; for
example, allowing for an unprecedented expansive, counter-cyclical, reac-
tion to the 2009 crisis in a context of fiscal sustainability.
Overall the Chilean experience shows the importance of the introduction of
structural budgeting as a principle, and the value of learning in policymaking,
24
It is interesting that modal methodologies, with standard filters, like that of
Hodrick–Prescott, use estimates for several future years in order to avoid the bias
introduced by the “final years” of the historical series. Naturally, the future years are
not responding to “purely objective” estimates.
5.4
5.2
2006
5.0 2007
4.8
2008
4.6
4.4
2009
4.2
4.0
3.8
3.6
2006 2007 2008 2009
3.4
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Figure IX.3 Trend GDP: annual review by Committee of Experts, 2006–9 (annual
growth rate, %).
Source: Based on Ministry of Finance data. Each year, with the new historical data and the forecasts
for the following years by the Committee, the Ministry re-estimates the trend GDP for past and
future years.
paying attention to local structural specificities. Key challenges for the future
are a greater understanding and guiding principles to deal with the macroeco-
nomic effect of fiscal policy on economic activity, prices, and the exchange
rate determination. A particular problem is how to achieve a management of
public savings that efficiently serves both short-term macroeconomic policy
and long-term economic development.
Progress in fiscal policy must be matched by enhanced counter-cyclical
capacity in the management of aggregate demand and the exchange rate,
which in recent years have become quite unstable in response to pro-cyclical
capital flows. Chile had, in the 1990s, an outstanding and successful experi-
ence with counter-cyclical regulation of financial inflows and achievement
of comprehensive real macroeconomic balances. It is time to consider its
reinstallation for the sake of stronger sustained growth-with-equity.
(b) Correcting the exchange rate policy

The exchange rate is a macroprice whose evolution is extremely relevant for
the efficient allocation of resources. The exchange rate regime, particularly
after trade liberalization, has become an increasingly influential variable in
emerging economies, for both trade and finance. It is subject to two conflict-
ing demands. The first demand comes from trade: with the dismantling of
traditional trade policies (tariff and non-tariff restrictions), the real exchange
rate has become a key determinant of international competitiveness and a
crucial variable for an efficient allocation of resources into tradables.
Indeed, a “competitive” and stable real exchange rate is an input for sound
trade development. A rather depreciated real exchange rate improves export
competitiveness, and its stability favors productive investment in tradables
and higher value-added activities. The positive connection between a com-
petitive level of the real exchange rate and success in economic growth of
EEs has been documented by Williamson (2000) and Rodrik (2007). It is
noteworthy how the two extreme proposals in fashion (corner solutions)
disregard these relevant analyses based on robust facts.
The second demand is derived from an open capital account. Volatility in
international financial and trade markets generates a demand for large fluc-
tuations of the exchange rate so as to equilibrate the balance of payments
in response to positive and negative shocks generated during the cycle. This
objective frequently cannot be reconciled with the trade-related goals of
exchange rate policy, because the equilibrium of the balance of payments
uses to take place at the expense of disequilibrium of the current account.
It becomes particularly detrimental for a growth strategy based on export
expansion and diversification. Intermediate regimes of managed exchange
rate flexibility – such as crawling pegs and bands, and dirty floating – attempt
to reconcile these conflicting demands (see Williamson, 2000; Ffrench-Davis
and Ocampo, 2001). The kind of flexibility of the exchange rate that we do
get becomes crucial.
From 1999, the central bank adopted a free exchange rate regime. The
RER has fluctuated widely since then, with a pro-cyclical pattern detrimen-
tal for the efficient production of tradables, particularly for those intensive
in value-added. Figure IX.I depicted the behavior of the RER in the 1990s
(see also Chapter VIII, Figure VIII.1). Here the focus is on the more recent
period.
Notwithstanding the sharp sterilizing effect of the stabilization funds, the
series of jumps in the price of copper (up from 60 cents per pound at a point
in 2002 to 400 cents in 2008) generated a foreign exchange bubble for an
artificially, outlier, strong peso. Between 2003 and early 2008 the exchange
rate appreciated 24%; in parallel tariffs were approaching zero and the sub-
sidy for non-traditional exports had disappeared. It was such an extreme
appreciation that the central bank, even though it was strongly committed
to a free rate, announced a program of purchases in the foreign exchange
market that were effective in stopping the destabilizing trend and reverting
it, soon assisted by the arrival of the contagion of the international crisis.
The bank did not argue that the rate was an outlier and needed correction,
but stated that it was just seeking to fortify international reserves in the face
of rising uncertainty. After a significant depreciation up to late 2008, appre-
ciation was resumed in 2009.
By December 2009 the real exchange rate had appreciated by 16% with
respect to late 2008. It was close to 8% below the RER in the golden period
(1990–5), with average nominal tariffs reduced to 1% from 10% as well as
the elimination of the 10% subsidy to non-traditional exports. In the 1990s

there was room for stabilizing appreciation, but in the 2000s that room
has elapsed with a much stagnated productivity changes (TFP). As com-
pensation, the trend price of copper is today much higher, which allowed
the current account to be in balance in 2009 (see Figures IX.2 and IX.3).
Instability has been the dominant feature in a country that supposedly has
an export strategy. If the balance is achieved by an increased price of a com-
modity such as cooper, the country would face the risk of a “Dutch decease”;
even worse, the present estimate of the trend could be a misleading signal.
I recall that, by the mid-2000s, the trend price was estimated to less than
one-half the present estimate. The fact is that the copper market is notably
cyclical and estimates are strongly influenced by the short run. There is a
high risk that today we are on the overoptimistic side of expectations. An
efficient insurance against that risk is to bet for a systematic diversification
of exports, which requires reform of the exchange rate policy.
Figure IX.4 is quite illustrative of the instability generated by an unstable
RER. Since 2003, the volume of imports has been growing, frequently, several
times faster than the volume of exports (see panel A). A surplus in the current
account in 2001–4 (adjusted by the trend price of copper, in panel B), has
been replaced by a fast rising deficit, peaking at 8% of GDP in 2008. In paral-
lel, exports were losing speed (see Chapter VI). The exchange rate regime is
in sharp conflict with the export strategy. This type of wide inconsistency is
costly for both economic growth and equity.
The characteristic volatility of freely floating exchange rate regimes does
not pose significant inefficiencies when market fluctuations are short-lived
random walks; in such a case they are easily faced with derivatives (see Dodd
and Griffith-Jones, 2006). But fluctuations become a major concern when
there are longer waves – medium-term waves – as has been typical of the
access of EEs to capital markets in recent decades. In this case, persistent
appreciation of the exchange rate during capital surges tends to generate
perverse effects on irreversible resource allocation. Moreover, under freely
floating regimes with open capital accounts, counter-cyclical monetary
policy exacerbates pro-cyclical exchange rate fluctuations, with significant
costs derived from an inefficient resource allocation and under-utilization
of GDP*.
The ability of a flexible exchange rate regime to smooth the effects of exter-
nal cycles depends on the capacity to effectively manage a counter-cyclical
monetary and credit policy without enhancing pro-cyclical exchange rate
patterns. This is only possible under intermediate exchange rate regimes
along with capital account regulations. That was, clearly, the successful case
of Chile in the first half of the 1990s (see Chapter VIII).
Managed flexibility – including diverse types of bands or other inter-
ventions by economic authorities – strengthens real exchange rate stabil-
ity, keeping the ability to partially absorb the effects of moderate shocks.
A
20 18.4
17.2
16 14.5
13.3
12.2
12 10.1 10.6
9.7
7.2 7.6
8 6.5
5.1 5.1
4.1 4.3
3.1
4 2.3
1.6
4
5.6
8
12
14.3
16
Exports Imports
20
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
B
6
4.9
5 4.5
4
3 2.6
2.0 2.2
2 1.2 1.2
0.8
1
0
0.3
1 0.9
0.6 0.5
1.2 1.1
2 1.6
1.5
2.1
3 2.9
2.6
4
5
6
7
8 Current account Current account corrected by trend copper price
7.9
9
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Figure IX.4 Panel A: evolution of exports and imports of goods and services, 2000–9
(annual real growth rates, %). Panel B: current account balances, 2000–9 (% of current
GDP).
Source: Based on central bank figures. Panel A shows series from National Accounts in prices of
2003. Panel B shows the ratio between the current account from the Balance of Payments and GDP
in current prices. The nominal trend copper price, here, is derived from the real trend price in
Figure IX.2.
Consequently, the exchange rate fulfils more efficiently its allocative role
between tradables and non-tradables. If we pretend to give priority to the
exchange rate as a resource allocator in the trade strategy embraced by Chile –
multiple trade associations with different blocks and competitive nations –
it is imperative to give signals of certain medium-term real stability to both
investors and producers. There is a need not only for present producers of
tradables to remain competitive and increasingly linked with the rest of the
domestic economy, but also for many other producers to be born in the
future. The dynamic role of SMEs in this process must be stressed.
Obviously, intermediate regimes may also exhibit shortcomings (see
Ffrench-Davis et al., 2003). First, intermediate regimes are subject to specula-
tive pressures if they do not achieve credibility in markets; in critical conjunc-
tures, particularly after the exchange rate has become an outlier price, the
costs of defending a given rate are huge. Then, it may be wise to temporarily
move to full flexibility. Second, reserve accumulation during long booms
may become financially costly. Lastly, the capital account regulations needed
to manage intermediate regimes efficiently reduce those costs, but are only
partially effective. However, all things considered, intermediate regimes offer
a healthy alternative to certainly costly outlier macroprices frequently result-
ing from the corner solutions.
In the neoliberal approach a statement has emerged that the economic
authorities are unable to affect the “market exchange rate,” because that
would imply the intention of defeating fundamental market forces. However,
the relevant point is that the intermediate approach is intended not
to move “against the market” of producers of tradables, but to avoid short-
termist agents or transitory terms of trade fluctuations dominating the for-
eign exchange market. When one or the two latter dominate, this tends to
lead to rates that are not sustainable in the mid-term, with huge costs to the
agents that really are at the core of the fundamentals; that is, the producers
of tradables. This pervasive outcome has been extremely frequent, in recent
decades, in several emerging economies. Thus, the issue is which role to be
assigned to the exchange rate, and my view is that it must be the allocation
of irreversible resources between tradables and non-tradables.
In brief, a totally free exchange rate policy implies accepting an inability
to achieve a sustainable real macroeconomic policy, as I have defined it in
Ffrench-Davis (2006, Chapter I), and an export strategy with strong pull
effects on the rest of the economy.
In a context of massive and volatile capital flows and erratic terms of
trade, a totally free rate fluctuates remarkably. In fact, since the late 1990s,
the exchange rate has shown great volatility. That volatility discourages
(i) the generation of value-added on natural resources, (ii) new exports by
SMEs, (iii) subsistence of the national industry competing with imports and
the employment it generates, (iv) productive investment in tradables sec-
tors, and (v) exploiting new opportunities presented by trade agreements – it
also enhances their risks and costs. In order to renew a growth-with-equity

path, Chile needs to recover its ability to maintain real macroeconomics
balances, as an essential condition. The central bank would have to return,
explicitly or implicitly, to an intermediate policy between both extremes in
fashion.
In an open economy like the Chilean one, high growth depends on the
world economic situation, but also on “sound” domestic macroeconomic
management and microeconomic reforms. This implies an active macroeco-
nomic policy, focused on the real economy (“productivism”), involving an
economy working near to the productive frontier. Lengthy recessive cycles
generate structural damage for labor generation, depressed fiscal revenue,
and losses and bankruptcies of enterprises.
In summary, the great GDP instability reflects a failure of macroeconomic
policies, as illustrated by the recessive gap of 1999–2003 and the rather
slow growth of 2006–8. In order to avoid or to moderate the intensity of
these situations, an improvement is required towards a more comprehen-
sive counter-cyclical approach, (i) so that the authorities can focus, free of
urgency, on the design of a development strategy that involves complex
reforms and microeconomic policies, and (ii) so that producers can focus
on investment and productivity improvements, with right macroprices and
a potential GDP consistent with effective demand.
Once the macroeconomic framework is corrected, the road towards
pending deep microeconomic reforms is cleared. These reforms are associ-
ated with labor and entrepreneurial training, innovation support, capital
market reforms with priority for long-term financing, and a comprehensive
SMEs program. In all these issues there has been progress, but insufficient,
sometimes hesitant, and with contradictions. A growth-with-equity strategy
requires coherent action and to be placed systematically at the core of the
national agenda.
X
Concluding Remarks and Challenges
The Concertación Democrática completed two decades in command of

the Chilean economy, through four presidential regimes. The first three
democratic regimes covered a period of time as long as the sixteen years of
dictatorship, when neoliberal reforms were implemented. In this closing
chapter, first, a brief summary of the sixteen years of dictatorship versus the
first sixteen years of democracy is made, focused on economic growth and
income distribution outcomes; summary outcomes for the full two decades
of Concertación regimes are also presented. Then, four general conclusions
that I consider must be present in the analyses of Chilean development are
reported. Finally, I raise ten challenges, in a non-exhaustive listing, that
I have selected as of high-priority action for Chile at the opening of the
2010s.
1 An account of sixteen years of dictatorship versus

sixteen years of democracy
When I was starting the preparation of this edition, sixteen years had
elapsed from the return to democracy; a period as long as the prolonged
dictatorship. Thus, it was an opportune time for making a global assessment
of both experiences, at a time when Chile was to start its third century as
an independent nation.
Notwithstanding the development gaps in both GDP and equity that re-
emerged from 1999, great social and economic progress had characterized
the sixteen years of democratic development. There are evident, huge, con-
trasts with the sixteen years of dictatorship.
Economic growth in the neoliberal regime of Pinochet, between 1973 and
1989, averaged only 2.9%, and income distribution deteriorated remarkably.
That outcome was associated with the deep fundamentalism of reforms and
reformers, implying numerous policy failures that severely affected eco-
nomic growth and social welfare.
268
Concluding Remarks and Challenges 269
During the dictatorship of Pinochet diverse modernization took place in

Chile. Undoubtedly, several of them have constituted a permanent basis for
democratic development strategies, but others constitute a sharp negative
burden for economic development. The four Concertación governments,
a center-left coalition of Christian Democrats and Social Democrats, in a
strategy of continuity with change, did seek reforms to the reforms, in order
to correct the inherited model. The objective was to introduce pragmatism
and progressiveness. In particular, they tried to diminish the vulnerability
to external shocks in a context of increasing volatility in international mar-
kets, and to advance in correcting the inequality featured by the economic
model. The outcome of the set of changes was that during the 1990s there
was a vigorous expansion of productive capacity without precedent in Chile
(averaging 7.1% annually in 1990–8), along with significant reduction of
the number of people in poverty (from 45 to 21% of the population in the
same period) and some improvement in income distribution. Nevertheless,
a recessive output gap in 1999–2003 revealed flaws and contradictions, and
the lack of deeper reforms to the reforms.
With the Asian crisis contagion, the final year of President Frei’s govern-
ment (1999) and the first part of President Lagos’s (2000–3) evolved in a
recessive economic environment. The sharp fall of 1999, and the subse-
quent stagnation of economic activity, was concentrated in non-tradable
sectors, which represented around 70% of GDP; that stagnation involved a
negative impact, mainly on SMEs and employment. A slower expansion of
the volume of exports was accompanied by a terms of trade deterioration.
However, its impact was multiplied in the rest of the economy through the
capital account and the monetary and exchange rate policies. That recessive
impact prevailed until 2003, constituting a substantive failure of Chilean
macroeconomic policies.
If Chile had had an indebted economy, short of international reserves,
without access to external credit, and a fiscal deficit, it could have justified
a domestic multiplication of the external shocks as extended as it was in
1999–2003. But no such conditions occurred in Chile in these years. The
result of such a long and severe recessive period was a policy option of
not facing the negative external shock with a positive domestic reactivat-
ing shock. The external shocks were multiplied due to the adoption of a
“financieristic” macroeconomics, at the expense of sustained growth and
equity. I have stressed this issue throughout the book because of its high
relevancy for the future.
The arrival of the present world crisis reinforces my criticism and the
direction of my proposals. The contrast between 7.1% growth in 1990–8 and
3.7% in 1999–2008 reflects significant weaknesses in the design of economic
policy.
In all, in the sixteen years between 1990 and 2005, GDP growth averaged
5.5%, in contrast to 2.9% in 1974–89, and per capita GDP expanded by 4.1%
270 Concluding Remarks and Challenges
in comparison with 1.3% in 1974–89 (see Table X.1).1 As said, the economic
performance of the Concertación Democrática regimes was weaker in its
second half. The presidential election of 2009, in which the center-left coa-
lition was defeated, took place when the contagion of the financial crisis
was at work and the weakening of economic policy was reflected in poorer
economic growth and a worsened trend in the labor market. Nonetheless,
even including this latter period, the overall annual average of the four
democratic governments is notably superior to that of the dictatorship: per
capita GDP growth was 3.6 and 1.3%, respectively. Additionally, by 2009
social policy had acquired a solidly stronger power as shown in chapter VII,
which is reflected in the improved distributive figures by 2009 as compared
to 1999–2005, in spite of the weaknesses of the labor market and of eco-
nomic growth.
The differences in the annual averages imply a notorious cummulative
gap between the two as time goes by. Figure X.1 is illustrative. By the end
of its sixteen years the Pinochet regime had achieved (in 1989) a GDP per
capita merely 23.6% higher than in 1973; the Concertación Democrática
recorded an 88.8% increase between 1989 and 2005. That remarkable dif-
ference has been associated with the improvement in the quality of macr-
oeconomic policies since 1990 (in particular, I repeat it, mostly in the first
Table X.1 GDP, GDP per capita and income distribution, 1974–2009 (annual growth
rates, %)
GDP growth (%) GDP per capita Q5/Q1 ratio GINI
growth (%) coefficient
(1) (2) (3) (4)
1974–1981 3.0 1.5 15.1 51.9

1982–1989 2.9 1.2 20.2 56.7
1990–1998 7.1 5.4 15.5 52.9
1999–2009 3.3 2.1 15.2 52.6
1974–1989 2.9 1.3 17.7 54.3
1990–2005 5.5 4.1 15.7 53.1
1990–2009 5.0 3.6 15.3 52.7
Source: Author’s calculations based on Chapter I and National Accounts from the central bank.
Columns (1) and (2) are based on 2003 constant prices. Columns (3) and (4) are based on the
University of Chile Employment Survey for Santiago.
1
If the comparison includes the twenty years of democracy up to 2009, the record of
GDP growth is slightly lower: 5.0%. The comparison underestimates the performance
of the Bachelet presidency because it ends in a year of recession, 2009, which implies
that potential GDP was significantly above actual GDP. Actual GDP is the data used
in the comparison, but the higher potential GDP implies that the economy, in sub-
sequent years, can grow much faster than the generation of new capacity. It is a posi-
tive inheritance, in contrast with that of 1989, which made unavoidable a downward
adjustment in 1990.
220.0
Per capita GDP 1973–1989 Per capita GDP 1989–2009
200.0
188.8
180.0
160.0
140.0
123.6
120.0
100.0
80.0
1973–1989
1974–1990
1975–1991
1976–1992
1977–1993
1978–1994
1979–1995
1980–1996
1981–1997
1982–1998
1983–1999
1984–2000
1985–2001
1986–2002
1987–2003
1988–2004
1989–2005
2006
2007
2008
2009
Figure X.1 Per capita GDP growth, 1973–89 versus 1990–2009 (real indexes, 1973 and
1989 100).
Source: Based on data from Marcel and Meller (1986) and central bank for GDP, and National
Bureau of Statistics for population.
years of democracy), with the sustained increase in social expenditure that

has provided better opportunities for more people, and with efforts aimed
at correcting micro- and mesoeconomic market and policy failures. These
last efforts, in fact quite insufficient, consisted of trying to free themselves
from neoliberal fundamentalism, and to progress, too mildly, towards
“completing” long-term capital markets, labor training, and technological
innovation.
2 Four general conclusions
Classifications and categories tend to be arbitrary, but if they are coherent

they provide a guiding framework for analysis and policy actions. Four
groups of economic lessons or conclusions follow. They are drawn from the
recent decades of Chilean economic history.
(a) Equity is crucial for sustained growth

The challenge of making growth consistent with distributive equity is ever
more important, especially in light of the sharp drop in the standard of liv-
ing of a wide segment of the population in the 1970s and 1980s, and the fact
that democratic governments have only partially met the expectations they
aroused in the 1990s. It is clear that substantial political consensus among
the main social and political groups will be required to meet the challenge.
Only in this way can the country implement action over time that will
distribute the costs and benefits of economic growth in a more equitable
fashion, with progressive improvement in the distribution of opportunities,
productivity, and income. Democracy requires that growth and equity

progress together.
(b) The need to recover a macroeconomics for development

There is overwhelming evidence that macroeconomic balances are crucial to
the success of any development strategy. One component that tends to be
omitted from the catalog of macroeconomic factors, and should always be
in the foreground, is the relationship between new productive capacity crea-
tion and actual increases in production (or utilization of capacity). As has
been documented, there were deep imbalances in the relationship between
the two variables during the Pinochet regime. Underutilization was notably
high in both 1975–9 and 1982–7. In both periods average capital forma-
tion plunged, reducing sharply the creation of new capacity, productive
employment, and equity. In fact, the consequences of that macroeconomic
disequilibrium were long-lasting. A similar sort of disequilibrium emerged in
1999–2003, though quite moderately. Consequently, the effects were of the
same negative sign – lower capital formation, higher unemployment, more
inequality – but were notably milder than in the two older recessions.
It must be stressed that the most significant macroeconomic balance is the
sustainable use of economically productive capacity. An economy working
on the production frontier, with correct or right macroprices (particularly
exchange rates), achieves higher productivity, encourages larger capital for-
mation, and fosters the demand for labor. This is what I have called real
macroeconomics or productivistic macroeconomics, in contrast to the financieris-
tic macroeconomics (with an unbalanced focus on inflation and disregard for
variables such as the stability of the exchange rate and domestic demand).
The cost of real macroeconomic imbalances is high indeed. Beside the
fact that such a situation leads to a reversal of the initial success that may
have been achieved in growth and redistribution, experience shows that
it entails costly political losses for governments that succumb to populist
temptations, whether from the left (as with unfunded social spending) or
the right (e.g. with tax cuts that make it unfeasible to finance the invest-
ment in human capital that is essential for growth-with-equity, or with the
excessive financial deregulation and capital account opening that paves the
way for macroeconomic imbalances).
The ways of achieving macroeconomic balances can be quite diverse.
They may have progressive or regressive effects; they may be pro-cyclical or
counter-cyclical depending, among other things, on the relative weight
given, by policymakers, to variables such as inflation, employment,
economic activity, the composition of public spending and revenue, the
nature and regulation of financial institutions, and public actions that con-
tribute to building the capacity and organization of low-income segments
of population. In all, they may be biased towards fostering financierism or
towards productivism; the former tends to be regressive and volatile, while the
latter provides the bases for growth-with-equity. Sustainable comprehensive

macroeconomic balances are essential, and they must be designed consist-
ently with macrosocial balances.
(c) Inheritances, positive and negative, make a significant difference

In the 1970s and 1980s there were several reforms directed to modernize
the country’s economic organization. Undoubtedly, many constituted per-
manent and valuable bases for future development strategies. The changes
included significant growth and diversification of exports, the achievement
of a more efficient fiscal budgeting process and organization of the internal
revenue service, and the development of a new generation of entrepreneurs
who were more dynamic and modern than the traditional business class.
This progress contributed considerably to the outstanding performance of
the 1990s. However, there is a serious lack of pluralism and pragmatism in
the sector holding the highly concentrated economic power, mass media,
and wealth. This lack of pragmatism, or deep neoliberalism, has been a heavy
handicap to implementing policies effectively accelerating growth-with-
equity.
In the other corner of society, the Concertación Democrática had to work
in a social environment with weak social organizations, few and divided
unions, and an extreme imbalance of voices made heard in Chilean society.
There have been corrections, but also significant contradictions (see Garretón,
2003). A rebalancing of voices and power is crucial for achieving the reforms
geared towards comprehensive social and economic development.
(d) A deficit of development-friendly reforms

Structural reforms suffered from various failures that had severe repercus-
sions for potential economic growth and the population’s welfare. The
1990s saw significant reforms to the reforms of previous decades, with the
specific objective of introducing a progressive pragmatism. These reforms
reflected a great concern to reduce the vulnerability of the Chilean economy
to the international environment, with its increased volatility, while stress-
ing policies contributing to greater equity in the distribution of opportuni-
ties, productivity, and income. The results of this change of focus, despite a
number of contradictions, have been notable. The recession of 1999–2003,
nevertheless, highlighted failures and insufficiencies, the lack of greater
reforms of the reforms, and a step-back in macroeconomic policies.
There have been relevant social reforms launched recently, such as Chile
Solidario for the extreme poor, the AUGE program for health, unemploy-
ment insurance and its new reform, and the ambitious social security
reform. Nonetheless, I emphasize again that in an economy that is not rich
and exhibits shameful inequality, outcomes are principally determined in
the mal-working of markets. There is an absolute need to make sharp correc-
tions in the working of markets: corrections pro-labor and jobs, pro-SMEs,
and pro-real macroeconomic stability. This is what leaders of the official

coalition have called deepening reforms of the model.2
From an optimistic perspective, the positive side of the recessive situation
has been a growing recognition of the problems and the search for ways to
solve them. Public opinion has been expressing its discomfort, despite the
progress achieved; this is partly reflected in the fast growing number of peo-
ple not participating in political elections. Even people, paradoxically, can
react to reward the regressive forces. Chile has all the conditions needed for
moving forward, with strengthened dynamism, in the direction of sustain-
able development. With this book, I hope to contribute to the thoughtful
dialogue needed if we are to achieve growth with equity – a dialogue for
constructive and progressive action.
3 Ten specific challenges
How can vigorous growth be achieved while advancing towards equity?

First, there must be an understanding of how this can be achieved and what
the country’s real record is so far. Development occurs when growth is sus-
tainable over long periods of time. If progress in Chile is measured by the
decade, counting both good and bad years, only the 1990s recorded a figure
of 7% for (actual and potential) GDP growth. In the 1960s, GDP growth
averaged 4.3% and in the 1970s and 1980s less than 3%. Some outstanding
priorities that need to be kept up-front in order to replicate the exceptional
record of the 1990s and grow with more equity are grouped here under the
ten points that follow.
(1) Chile must regain a macroeconomic environment of sustainable devel-
opment with three features. The first is to achieve, as in most of the 1990s,
a level of effective demand close to the productive capacity of the Chilean
economy – always a key objective for efficient macroeconomic balances
and vigorous productive investment. This implies avoiding unsustainable
booms and applying vigorous and consistent counter-cyclical actions dur-
ing recessions. Second, it will be necessary to maintain a competitive and
stable real exchange rate, which is not feasible with corner solutions (neither
a fully free rate nor dollarization). Third, interest rates must be kept within
a moderate range. These three objectives require a more flexible fiscal policy,
with effective anti-cyclical measures (in the fiscal budget, monetary policy,
and banking supervision), and an increased ability to discourage excesses in
speculative capital inflows. The great danger is a “populist” attitude that fails
to weigh the broad negative effects on the economy of excessive capital flows
that are speculative, short-term, merely financial, or aimed at purchasing
2
Acentuar las correcciones del modelo. See Manifiesto de los Partidos de la Concertación
(2005), prepared during the presidential campaign of President Michelle Bachelet
(2006–10).
what already exists without creating new productive capacity. In fact, the
main cause of the recession of 1999 was the huge capital inflow recorded
in 1996–7, as shown in Chapter IX. The long recessive gap up to 2003 was
associated with the explicit dependency from the voices of short-termist
agents who specialized, in simple terms, on overnight finance. What a misuse
of the potentiality of the Chilean economy this was.
(2) Export dynamism is a determining factor for the ability to grow. Chile
cannot afford to continue simply doing more of what it has always done.
Markets become saturated and resources are exhausted or become less pro-
ductive. Non-traditional exports must be developed, without losing what
has been gained with traditional items. This means (i) moving forward
strongly to add value and technology to the traditional natural resources
and developing the production of intermediate goods and services as well
as capital goods linked to the productive process in those traditional areas
(the recent decision to move ahead with selected clusters is encouraging);
(ii) developing non-traditional natural resources; and (iii) finding niches
for the ability and experience of national entrepreneurs and technicians in
order to acquire competitive advantages in services and manufactures. This
will require an intense national effort to complete markets for technology,
labor training, and long-term capital.
A dynamic export drive cannot be achieved with the present wide
exchange rate instability. Notwithstanding the counter-cyclical stabilization
funds of the government, the foreign exchange market has been dominated,
since the mid-1990s, by volatile financial flows and terms of trade swings.
There is a need to recover the exchange rate as a powerful and efficient
resource allocator.
(3) With respect to export markets, Chile has signed free trade agreements
with most of the world, including, among others, the USA, the European
Union, Japan, the Republic of Korea and China in this decade. Chile
needs to better use the broad opportunities offered by its partners; as seen,
exports expanded more slowly in the 2000s than in the 1990s. As far as
non-traditional exports are concerned, negotiating access to Latin American
markets, taking advantage of geographic proximity, will be crucial.3 New
obstacles will probably emerge – among them, macroeconomic instability
in some neighbors and historical or political conflicts – but they must be
confronted constructively. A well designed regional integration is one way
to make, rather than take, globalization.
(4) Development always leaves some productive sectors behind. It is
better to anticipate malaise than to ignore it and have to deal with a mori-
bund patient later. Comprehensive macroeconomic stability can make a
3
Geography still counts for a lot, as testified by the North Americans, who sell half of
their exports to their two neighbor partners in NAFTA, or the Europeans, for whom
the intra-regional share is close to two-thirds.
great contribution to development and can help to minimize the number

of losers, but a pragmatic, forward-looking sectoral and regional policy is
also essential. Significant sectors, such as in agriculture, are urgently in
need of further thought and timely action for efficient, pro-development,
reconversion.
(5) Growth is not possible without physical investment. The Chilean
economy needs more, not less, productive investment. First, a dangerous
rent-seeking bias, intensified in recent years, must be avoided: there are too
many efforts and capabilities devoted to mergers and transfers of existing
assets. It is necessary to strengthen creative skills; upgrading the quality
of macroeconomic policies and enhancing actions to complete markets of
factors and exports are essential. Second, domestic capital formation is an
overwhelming proportion of investment everywhere in the world. Only one
of every ten dollars invested in the world comes from abroad. Domestic sav-
ings and investment are determinant. Policy must be designed to assure that
Chile’s long-term savings – such as those of the private pension funds – are
invested in domestic capital markets, and allocated to long-term lending for
development. It is a gross mistake to believe that those funds are too large
for Chile; actually, they are not enough, but the channels for their proper
allocation are missing.
The prevailing tendency in Chile to transfer the funds of the private pen-
sion system abroad presents an unfortunate contradiction with the objec-
tive of raising the domestic savings rate so as to increase capital formation
in Chile. Priority should be given to improving the transmission channels
between those long-term funds and productive investment – naturally,
always with guarantees and prudential regulations to protect the savings of
workers. This is consistent with the fact that average profitability is usually
higher in Chile (still a capital-scarce country) than in developed countries.
(6) Investment in human capital is a key input for stability and growth-
with-equity. It must be given the priority it deserves, always keeping in
mind that education is dependent on teachers; social trust and recognition
of their importance must be rebuilt. Present public funding is evidently
insufficient for attacking the deeply unfair inequality it holds at present.
Education, however, operates over the long term. Most of the labor force
of this new decade will consist of Chileans who have already finished their
formal education. Chile, then, faces the great challenge of enhancing the
flexibility of the supply of labor with far-sighted and well focused national
labor-training programs; small and micro entrepreneurs also need effective
training programs.
(7) The environment has become a leading issue in modernization.
Chile must make up for its lag in addressing environmental issues and in
integrating them in the design of public policies. A highly pragmatic and
participatory approach is needed to reconcile the development of today’s
productive capacity with preventive action, sustainability over time, and
the present welfare of citizens. Growth-with-equity for present generations

must become sustainable.
(8) Income distribution is highly inequitable. Chile must intensify its
efforts to improve the quality and coverage of social spending, make an
effective impact on hard core poverty, and eliminate the possibility that
a young, intelligent person will not receive a good education because of
a poor social origin. Chile Solidario, the AUGE program, and the Social
Security reform are significant steps forward. Strengthening the fight against
tax evasion and the regressive loopholes that still exist in the tax system
would provide financing and contribute to tax equity. Once again, global
consistency of policies and their interrelations is crucial. The quality of
macroeconomic policies strongly affects poverty and income distribution,
through its impact on productive employment.
(9) The state must be reformed. The main issues here are professional-
izing public administration, raising productivity systematically, improv-
ing the quality and friendliness of public service, making space for the
participation of people in identifying priorities and failures (in education,
health, public works, labor training, etc.), better using the public purchas-
ing power to develop SMEs, and heightening transparency in government.
Simultaneously, salaries must be gradually raised to levels consistent with
those prevailing in the private sector.
(10) Paradoxically, since the return to democracy the quality and intensity
of thinking about the future of Chilean people has eroded. It has become
too focused on the short term and concentrated on what “the market
finances.” Too much confidence has been placed in what other countries
think and do (some specialists even believe that other countries should drive
Chile’s macroeconomic policy, e.g. through the capital account opening),
and too little attention has been paid to the importance of each country’s
or region’s specific characteristics and targets. For many, globalization has
taken the shape of mythic immutability. However, globalization is a highly
heterogeneous process, human-made, incomplete, and rather unbalanced
with respect to social and economic factors. It is urgent that Chile rethink
the role of globalization in its economic life, in order to understand how to
move more effectively than in recent years towards sustained growth-with-
equity and to position itself to act cogently during this new decade.
References
Agosin, M. R. (1998) “Capital inflows and investment performance: Chile in the

1990s,” in R. Ffrench-Davis and H. Reisen (eds), Capital Flows and Investment
Performance: Lessons from Latin America, OECD Development Centre/ECLAC, Paris.
Agosin, M. R. (1999) “Trade and growth in Chile,” CEPAL Review 68, August.
Agosin, M. R. (2007) “Trade and growth: Why Asia grows faster than Latin America,”
in R. Ffrench-Davis and J. L. Machinea (eds), Economic Growth with Equity: Challenges
for Latin America, Palgrave Macmillan, Basingstoke.
Agosin, M. R. and Bravo-Ortega, C. (2007) “The emergence of new successful export
activities in Latin America: the case of Chile,” Latin American Research Network,
IDB. Published as Working Paper No. 236, Departamento de Economía, Universidad
de Chile, April.
Agosin, M. R. and Ffrench-Davis, R. (1998) “La inserción externa de Chile: experi-
encias recientes y desafíos,” in R. Cortázar and J. Vial (eds), Construyendo opciones:
propuestas económicas y sociales para el cambio de siglo, Dolmen Ediciones, Santiago.
Agosin, M. R. and Ffrench-Davis, R. (2001) “Managing capital inflows in Chile”, in
S. Griffith-Jones, M. F. Montes and A. Nasution (eds), Short-Term Capital Flows and
Economic Crises, Oxford University Press/WIDER, London and New York.
Agosin, M. R., Larraín, Ch. and Grau, N. (2009) “Industrial policy in Chile,”
Department of Economics, Universidad de Chile.
Agostini, C. (2008) “Los impuestos importan,” Observatorio Económico, Universidad
Alberto Hurtado, Santiago, August.
Ahumada, J. (1966) La crisis integral en Chile, Editorial Universitaria, Santiago.
Altimir, O. (1982) “Extent of poverty in Latin America,” World Bank Staff Working
Paper 522, World Bank, Washington, DC.
Alvarez, R. and Crespi, G. (2000) “Impacto de las políticas de fomento sobre el dina-
mismo exportador chileno,” Working Paper 3, Economics Department, University
of Chile, April.
Angell, A. (1993) “Chile since 1958,” in L. Bethell (ed.), Chile since Independence,
Cambridge History of Latin America, Cambridge.
Aninat, E. and Larraín, Ch. (1996) “Capital flows: lessons from the Chilean experi-
ence,” CEPAL Review 60, December.
Arellano, J. P. (1983) “De la liberalización a la intervención: el mercado de capitales
en Chile, 1974–83,” Colección Estudios CIEPLAN 11, December.
Arellano, J. P. (1989) “La seguridad social en Chile en los años 90,” Colección Estudios
CIEPLAN 27, December.
Arellano, J. P. and Cortázar, R. (1982) “Del milagro a la crisis: algunas reflexiones sobre
el momento económico,” Colección Estudios CIEPLAN 8, July.
Arellano, J. P. and Ramos, J. (1987) “Capital flight in Chile,” in D. Lessard and
J. Williamson (eds), Capital Flight and Third World Debt, Institute for International
Economics, Washington, DC.
Bacha, E. and Díaz-Alejandro, C. (1983) “Mercados financieros: una visión desde la
semiperiferia,” in R. Ffrench-Davis (comp.), Relaciones financieras externas y desarrollo
nacional en América Latina, Fondo de Cultura Económica, México.
Balassa, B. (1977) Policy Reforms in Developing Countries, Pergamon Press, New York.
278
References 279
Balassa, B. (1981) “Tipo de cambio y paridad del poder adquisitivo: una reevaluación,”
in R. Ffrench-Davis (ed.), Lecturas sobre intercambio y desarrollo, vol. I, Fondo de
Cultura Económica, México.
Baldwin, R. (1981) “Restricciones no arancelarias del comercio internacional,” in
R. Ffrench-Davis (ed.), Lecturas sobre intercambio y desarrollo, Fondo de Cultura
Económica, México.
Bande, J. and Ffrench-Davis, R. (1989) “Copper policies and the Chilean economy:
1973–88,” Notas Técnicas CIEPLAN 131, June; and Institute for Developing
Economies, Tokyo.
Baño, R. (ed.) (2003) La Unidad Popular treinta años después, LOM Ediciones/
Universidad de Chile, Santiago.
Behrman, J. (1976) Foreign Trade Regimes and Economic Development: Chile, NBER, New
York.
Benavente, J. M. (2005) “Innovación tecnológica en Chile, dónde estamos y qué se
puede hacer,” Economía Chilena 8, Banco Central de Chile, April.
Benavente, J. M. and Crespi, G. (1998) “Sesgos y debilidades del SNI en Chile,” in
M. Agosin and N. Saavedra (eds), Sistemas nacionales de innovación. ¿Qué puede
América Latina aprender de Japón?, Dolmen Ediciones, Santiago.
Beyer, H. (1997) “Distribución del ingreso: antecedentes para la discusión,” Estudios
Públicos 65, Santiago, summer.
Beyer, H. and Vergara, R. (2002) “Productivity and economic growth: the case of
Chile,” in N. Loayza and R. Soto (eds), Economic Growth: Sources, Trends and Cycles,
Central Bank of Chile, Santiago.
Bhagwati, J. (1978) Foreign Trade Regimes and Economic Development: Anatomy and
Consequences of Exchange Control Regimes, NBER, New York.
Bitar, S. (1979) Transición, socialismo y democracia: la experiencia chilena, Siglo XXI,
México.
Bosworth, B., Dornbusch, R. and Labán, R. (eds) (1994) The Chilean Economy: Policy
Lessons and Challenges, The Brookings Institution, Washington, DC.
Bourguignon, F. and Walton, M. (2007) “Is greater equity necessary for higher long-term
growth in Latin America?” in R. Ffrench-Davis and J. L. Machinea (eds), Economic
Growth with Equity: Challenges for Latin America, Palgrave Macmillan, Basingstoke.
Bouzas, R. and Ffrench-Davis, R. (eds) (1990) Conversión de deuda y financiación del
desarrollo en América Latina, Grupo Editor Latinoamericano, Buenos Aires.
Bouzas, R. and Keifman, S. (2003) “Making trade liberalization work,” in
P. P. Kuczynski and J. Williamson (eds), After the Washington Consensus: Restarting
Growth and Reform in Latin America, Institute for International Economics,
Washington, DC.
Bravo, D. and Contreras, D. (1999) “La distribución del ingreso en Chile, 1990–96:
análisis del mercado del trabajo y las políticas sociales,” Department of Economics,
University of Chile.
Brunner, J. J. (1981) La cultura autoritaria en Chile, FLACSO, Santiago.
Büchi (1993) La transformación económica de Chile, Grupo Editorial Norma, Bogotá.
Caballero, R. and Corbo, V. (1989) “The effect of real exchange rate uncertainty on
exports: empirical evidence,” The World Bank Economic Review, vol. 3(2), May.
Cabezas, M. (1988) “Revisión metodológica y estadística del gasto social en Chile:
1970–86,” Notas Técnicas CIEPLAN 114, May.
Calderón, A. and Griffith-Jones, S. (1995) “Los nuevos flujos de capital extranjero
en la economía chilena: acceso renovado y nuevos usos,” Serie Desarrollo Productivo,
24, ECLAC, Santiago.
280 References
Calvo, G., Leiderman, L. and Reinhart, C. (1993) “Capital inflows and real exchange
appreciation in Latin America: the role of external factors,” IMF Staff Papers, 40(1),
March.
Calvo, G. and Mendoza, E. (1996) “Petty crime and cruel punishment: lessons from
the Mexican debacle,” American Economic Review, 86(2), March.
Campero, G. and Valenzuela, J. (1981) “El movimiento sindical chileno en el
capitalismo autoritario,” ILT-Academia de Humanismo Cristiano/IPET, Santiago,
December.
Central Bank of Chile, Balanza de Pagos, Boletín Mensual, Deuda Externa de Chile,
Estudios Monetarios, Informe de Estabilidad Financiera, Informe de Política Monetaria,
y Memoria Anual, several issues, Santiago.
Centro de Microdatos, several Reports, Department of Economics, University of
Chile.
CIEPLAN (1982) Modelo económico chileno: trayectoria de una crítica, Editorial Aconcagua
(confiscated by the dictatorship), Santiago.
CIEPLAN (1983) Reconstrucción económica para la democracia, Editorial Aconcagua,
Santiago.
Círculo de Economía (1980) Seminario sobre políticas económicas para el desarrollo y la
igualdad, Academia de Humanismo Cristiano, Santiago, June.
Claessens, S., Dooley, M. P. and Warner, A. (1995) “Portfolio capital flows: hot or
cold?” The World Bank Economic Review 9(1), January.
Coeymans, J. (1999) “Determinantes de la productividad en Chile: 1961–97,”
Cuadernos de Economía 107, April.
Consejo Asesor Presidencial Trabajo y Equidad (2008) Hacia un Chile más justo: trabajo,
salario, competitividad y equidad social, Final Report, Santiago, August.
Consejo Nacional de Innovación (2007) Hacia una Estrategia Nacional de Innovación
para la Competitividad, Tomo I, Santiago.
Contreras, D. and Ruiz-Tagle, J. (1997) “¿Cómo medir la distribución de ingresos en
Chile?” Estudios Públicos 65, Santiago, summer.
Contreras, G. and García, P. (2002). “Estimating gaps and trends for the Chilean
economy,” in N. Loayza and R. Soto (eds), Economic Growth: Sources, Trends and
Cycles, Central Bank of Chile, Santiago.
Corbo, V. (1985) “Reforms and macroeconomic adjustments in Chile during
1974–84,” in V. Corbo and J. de Melo (eds), Liberalization with Stabilization in the
Southern Cone of Latin America, World Development 13(8), August.
Correa, S., Figueroa, C., Jocelyn-Holt, A., Rolle, L. and Vicuña, M. (2002) Historia del
siglo XX chileno, Tercera Edición, Editorial Sudamericana, Santiago.
Cortázar, R. (1983) “Derechos laborales y desarrollo: desafíos y tensiones,” in CIEPLAN,
Reconstrucción económica para la democracia, Editorial Aconcagua, Santiago.
Cortázar, R. (1996) “A labor policy for a new reality,” in C. Pizarro, D. Raczynski and
J. Vial (eds), Social and Economic Policies in Chile’s Transition to Democracy, CIEPLAN/
UNICEF, Santiago.
Cortázar, R. and Marshall, J. (1980) “Indice de precios al consumidor en Chile:
1970–78,” Colección Estudios CIEPLAN 4, November.
Cortázar, R. and Vial, J. (eds) (1998) Construyendo opciones: propuestas económicas y
sociales para el cambio de siglo, Dolmen Ediciones, Santiago.
Dahse, F. (1979) Mapa de la extrema riqueza. Los grupos económicos y el proceso de concen-
tración de capitales, Editorial Aconcagua, Santiago.
De Gregorio, J., Edwards, S. and Valdés, R. (2000) “Controls on capital inflows: do
they work?,” Journal of Development Economics, 63(1).
References 281
De la Cuadra, S. (1981) “Política cambiaria y deuda externa,” Boletín Mensual 639,

Central Bank, May.
De la Cuadra, S. and Hachette, D. (1992) Apertura comercial: experiencia chilena, Facultad
de Ciencias Económicas y Administrativas, Universidad Católica, Santiago.
Devlin, R. (1989) Debt and Crisis in Latin America: The Supply Side of the Story, Princeton
University Press, Princeton, NJ.
Devlin, R. and Cominetti, R. (1994) “La crisis de la empresa pública, las privatizacio-
nes y la equidad social,” Serie Reformas de Políticas Públicas 26, ECLAC, Santiago.
Devlin, R. and Ffrench-Davis, R. (1995) “The great Latin American debt crisis: a decade
of asymmetric adjustment,” in G. Helleiner (ed.), Poverty, Prosperity and the World
Economy, Macmillan, London.
Díaz, A. and Ramos, J. (1998) “Apertura y competitividad,” in R. Cortázar and J. Vial
(eds), Construyendo opciones: propuestas económicas y sociales para el cambio de siglo,
Dolmen Ediciones, Santiago.
Díaz-Alejandro, C. F. (1985) “Goodbye financial repression, hello financial crash,”
Journal of Development Economics 19(1/2), September/October.
DIPRES (1978) Somos realmente independientes gracias al esfuerzo de todos los chilenos,
Ministerio de Hacienda, Santiago.
DIPRES, Estadísticas de las finanzas públicas, Ministerio de Hacienda, several issues,
Santiago.
DIRECON (2009) Chile: 20 años de negociaciones comerciales, Ministry of Foreign
Affairs, Santiago.
Dodd, R. and Griffith-Jones, S. (2006) “Report on derivatives markets: stabilizing
or speculative impact on Chile and a comparison with Brazil,” in Working Paper,
Projects of CEPAL, October.
Dornbusch, R. (1988) “Our LDC debts,” in M. Feldstein (ed.), The United States and the
World Economy, University of Chicago Press/NBER, Chicago.
Dornbusch, R. and Edwards, S. (eds) (1991) The Macroeconomics of Populism in Latin
America, University of Chicago Press, Chicago.
ECLAC (1991) “Una estimación de la magnitud de la pobreza en Chile, 1987,”
Colección Estudios CIEPLAN 31, March, Santiago.
ECLAC (1997) “Evolución reciente de la pobreza en Chile,” (LC/R.1773) Santiago,
December.
ECLAC (1998) Policies to Improve Linkages with the Global Economy, 1995; second
updated edition published in Spanish by Fondo de Cultura Económica, Santiago.
Edwards, S. (1999) “How effective are capital controls?” Journal of Economic Perspectives
Autumn.
Edwards, S. and Cox-Edwards, A. (1987) Monetarism and Liberalization: The Chilean
Experiment, Ballinger, Cambridge, MA.
Elórtegui, C. (1988) Endeudamiento externo y conversión de deuda en la economía chilena,
Ediciones Universitarias de Valparaíso, Universidad Católica de Valparaíso.
Engel, E., Galetovic, A. and Raddatz, C. (1999) “Taxes and income distribution in Chile:
some unpleasant redistributive arithmetic,” Journal of Development Economics 59.
Eyzaguirre, N. (1988) “La deuda interna chilena, 1975–85,” in C. Massad and R. Zahler
(eds), Deuda interna y estabilidad financiera, vol. II, Grupo Editor Latinoamericano,
Buenos Aires.
Fajnzylber, F. (1990) “Industrialization in Latin America: from the “black box” to the
empty box,” in Cuadernos de la CEPAL 60, Santiago, and in M. Torres (ed.) Fernando
Fajnzylber – una visión renovadora del desarrollo en América Latina, Libros de la
CEPAL, 92, Santiago, 2006.
282 References
Fazio, H. (1997) El mapa actual de la extrema riqueza en Chile, LOM Ediciones,

Santiago.
Feinberg, R. and Ffrench-Davis, R. (eds) (1988) Development and External Debt in Latin
America, University of Notre Dame Press, South Bend.
Felix, D. and Caskey, J. (1990) “The road to default: an assessment of debt crisis man-
agement in Latin America,” in D. Felix (ed.), Debt and Transfiguration? Prospects for
Latin America’s Revival, M. E. Sharpe, New York.
Feres, J. C. (2001a) “Evidencia empírica en torno a la medición de la desigualdad:
algunas advertencias metodológicas,” mimeo, ECLAC.
Feres, J. C. (2001b) “La pobreza en Chile en el año 2000,” in Serie Estudios Estadísticos
y Prospectivos 14, ECLAC, July.
Ffrench-Davis, R. (1973) Políticas económicas en Chile: 1952–70, Ediciones Nueva
Universidad, Santiago.
Ffrench-Davis, R. (1979) “Exports and industrialization in an orthodox model: Chile,
1973–78,” CEPAL Review 9, December.
Ffrench-Davis, R. (1981) “Exchange rate policies in Chile: the experience with the
crawling peg,” in J. Williamson (ed.), The Crawling-Peg: Past Performance and Future
Prospects, Macmillan, London.
Ffrench-Davis, R. (comp.) (1983a) Relaciones financieras externas y desarrollo nacional en
América Latina, Fondo de Cultura Económica, México.
Ffrench-Davis, R. (1983b) “Una estrategia de apertura externa selectiva,” in
CIEPLAN, Reconstrucción económica para la democracia, Editorial Aconcagua,
Santiago).
Ffrench-Davis, R. (1984) “Índice de precios externos: un indicador para Chile de la
inflación internacional, 1952–83,” Colección Estudios CIEPLAN 13, June.
Ffrench-Davis, R. (1990) “Debt–equity swaps in Chile,” Cambridge Journal of Economics
14(1), March.
Ffrench-Davis, R. (1992) “Adjustment and conditionality in Chile, 1982–88,” in
S. Griffith-Jones and E. Rodríguez (eds), Cross-Conditionality, Banking Regulation and
Third-World Debt, Macmillan Press, London.
Ffrench-Davis, R. (2003) “La inversión extranjera directa en Chile,” in O. Muñoz (ed.),
Hacia un Chile competitivo, Editorial Universitaria/FLACSO, Santiago.
Ffrench-Davis, R. (2006) Reforms for Latin America’s Economies: After Market
Fundamentalism, Palgrave Macmillan, London and New York.
Ffrench-Davis, R. (2010) “The structural fiscal balance in Chile,” Journal of Globalization
and Development, 1, January.
Ffrench-Davis, R., Agosin, M. R. and Uthoff, A. (1995) “Capital movements, export
strategy and macroeconomic stability in Chile,” in R. Ffrench-Davis and S. Griffith-
Jones (eds), Coping with Capital Surges. The Return of Finance to Latin America, Lynne
Rienner Publishers, Boulder, CO.
Ffrench-Davis, R. and Arellano, J. P. (1981) “Apertura financiera externa: la experien-
cia chilena en 1973–80,” Colección Estudios CIEPLAN 5, July; and in Ffrench-Davis
(1983a).
Ffrench-Davis, R. and De Gregorio, J. (1987) “Orígenes y efectos del endeudamiento
externo en Chile: antes y después de la crisis,” El Trimestre Económico 213 and 214,
México, January–March.
Ffrench-Davis, R. and Griffith-Jones, S. (eds) (1995) Coping with Capital Surges. The
Return of Finance to Latin America, Lynne Rienner Publishers, Boulder, CO.
Ffrench-Davis, R., Leiva, P. and Madrid, R. (1992) “Trade liberalization in Chile: expe-
riences and prospects,” Trade Policy Series 1, UNCTAD, Geneva.
References 283
Ffrench-Davis, R. and Marfán, M. (1989) “Selective policies under a structural foreign

exchange shortage,” Journal of Development Economics 29.
Ffrench-Davis, R. and Stallings, B. (eds) (2001) Reformas, crecimiento y políticas sociales
en Chile desde 1973, LOM Ediciones/CEPAL, Santiago.
Ffrench-Davis, R. and Tapia, H. (2001) “Three varieties of capital surge management
in Chile,” in R. Ffrench-Davis (ed.), Financial Crises in Successful Emerging Economies,
Brookings Institution, Washington, DC.
Ffrench-Davis, R. and Tapia, H. (2005) “The Chilean style of capital controls: an
empirical assessment,” Working Paper, CEPAL.
Ffrench-Davis, R., Tapia, H. and Larraín, G. (2003) “How optimal are the extremes?
Latin American exchange rate policies during the Asian crisis?” in R. Ffrench-
Davis and S. Griffith-Jones (eds), From Capital Surges to Drought: Seeking Stability for
Emerging Markets, Palgrave Macmillan, London.
Ffrench-Davis, R. and Tironi, E. (eds) (1974) El cobre en el desarrollo nacional, Ediciones
Nueva Universidad, Santiago.
Fischer, R. (1999) “Income distribution and trade liberalization,” Documento de Trabajo
67, Departamento de Ingeniería Industrial, Universidad de Chile, Santiago.
Fontaine, J. A. (1989) “The Chilean economy in the 1980s: adjustment and recovery,”
in S. Edwards and F. Larraín (eds), Debt, Adjustment and Recovery: Latin America’s
Prospects for Growth and Development, Basil Blackwell, Oxford.
Forbes, K. (2003) “One cost of the Chilean capital controls: increased financial con-
straints for smaller traded firms,” in NBER Working Paper 9777, June.
Foxley, A. (1983) Latin American Experiments in Neoconservative Economics, University
of California Press, Berkeley.
Foxley, A. (2005) “Lessons from Chile’s development in the 1990s,” in T. Besley and
R. Zagha (eds), Development Challenges in the 1990s: Leading Policymakers Speak from
Experience, Oxford University Press/World Bank, New York.
Frank, Ch., Kim, K. and Westphal, L. (1975) Foreign Trade Regimes and Economic
Development: South Korea, NBER, New York.
Frenkel, R. (1983) “La apertura financiera externa: el caso argentino,” in R. Ffrench-
Davis (comp.), Relaciones financieras externas y desarrollo nacional en América Latina,
Fondo de Cultura Económica, México.
FSP (2005) “Umbrales sociales 2006: propuesta para la futura política social,”
Fundación para la Superación de la Pobreza, Santiago, August.
Galasso, E. (2006) “With their effort and one opportunity: alleviating extreme poverty
in Chile,” Working Paper, World Bank, Washington, DC, March.
Gallego, F. and Hernández, L. (2003) “Microeconomic effects of capital controls: the
Chilean experience during the nineties,” in Working Paper 203, Central Bank of
Chile, February.
Gallego, F. and Loayza, N. (2002) “The golden period for growth in Chile. Explanations
and forecasts,” in N. Loayza and R. Soto (eds), Economic Growth: Sources, Trends and
Cycles, Central Bank of Chile, Santiago.
Garretón, M. A. (2001) Cambios sociales, actores y acción colectiva en América Latina,
CEPAL, Santiago.
Garretón, M. A. (2003) The Incomplete Democracy in Latin America and Chile, North
Carolina University Press, Chapel Hill.
Griffith-Jones, S. and Rodríguez, E. (eds) (1992) Cross Conditionality, Banking Regulation
and Third World Debt, Macmillan, London.
Hachette, D. (2000) “La reforma comercial,” in F. Larraín and R. Vergara (eds), La
transformación económica de Chile, Centro de Estudios Públicos, Santiago.
284 References
Hachette, D. and Lüders, R. (1992) La privatización en Chile, CINDE, Santiago.

Harberger, A. (1985) “Observations on the Chilean economy, 1973–83,” Economic
Development and Cultural Change 33, April.
Hausmann, R. and Rodrik, D. (2003) “Economic development as self-discovery,”
Journal of Development Economics 72(2).
Held, G. and Jiménez, F. (2001) “Liberalización, crisis y reforma del sistema financiero
chileno,” in R. Ffrench-Davis and B. Stallings (eds), Reformas, crecimiento y políticas
sociales en Chile desde 1973, LOM Ediciones/CEPAL, Santiago.
Helpman, E., Leiderman, L. and Bufman, G. (1994) “A new breed of exchange rate
bands: Chile, Israel and Mexico,” Economic Policy 19, October.
Herzer, D. and Nowak-Lehmann, F. (2004) “Export diversification, externalities and
growth,” in Discussion Papers 099, Ibero America Institute for Economic Research (IAI).
Hofman, A. (1999) The Economic Development of Latin America in the Twentieth Century,
ECLAC, Edward Elgar, Aldershot.
Huneeus, C. (2001) El régimen de Pinochet, Editorial Sudamericana, Santiago.
INE, “Encuesta de Presupuestos Familiares,” several issues, Santiago.
Jadresic, E. (1986) “Evolución del empleo y desempleo en Chile: 1970–85, series
anuales y trimestrales,” Colección Estudios CIEPLAN 20, Santiago, December.
Jadresic, E. (1990) “Salarios reales en Chile: 1960–88,” Notas Técnicas CIEPLAN 134,
Santiago, September.
Jélvez, M. (ed.) (2007) Perspectivas económicas para el Chile del Bicentenario: desafíos y
oportunidades, Centro de Estudios del Desarrollo (CED) y PNUD, Santiago.
Jélvez, M. and Alvarado, M. (2009) “Cómo continuar avanzando hacia el desarrollo:
propuestas para una política nacional de empleo,” ILO, Santiago.
Krugman, P. (1990) The Age of Diminished Expectations, MIT Press, Cambridge, MA.
Kuczynski, P. P. and Williamson, J. (eds) (2003) After the Washington Consensus:
Restarting Growth and Reform in Latin America, Institute for International Economics,
Washington, DC.
Labán, R. and Larraín, F. (1997) “Can a liberalization of capital outflows increase net
capital inflows?” Journal of International Money and Finance 16(3).
Lahera, E. (1981) “The transnational corporation in the Chilean economy,” CEPAL
Review, 14, August.
Larraín, F. (ed.) (1987) Desarrollo económico en democracia: proposiciones para una socie-
dad libre y solidaria, Ediciones Universidad Católica de Chile, Santiago.
Larraín, F. (1991) “Public sector behavior in a highly indebted country: the contrast-
ing Chilean experience, 1970–85,” in F. Larraín and M. Selowsky (eds), The Public
Sector and the Latin American Crisis, ICS Press, San Francisco.
Larraín, F., Labán, R. and Chumacero, R. (2000) “What determines capital inflows? An
empirical analysis for Chile,” in F. Larraín (ed.), Capital Flows, Capital Controls, and
Currency Crises: Latin America in the 1990s, University of Michigan Press, Ann Arbor.
Larraín, F. and Velasco, A. (1990) “Can swaps solve the debt crisis? Lessons from the
Chilean experience,” Princeton Studies in International Finance.
Larraín, F. and Vergara, R. (eds) (2000) La transformación económica de Chile, Centro de
Estudios Públicos, Santiago.
Larraín, G. (2005) Chile fertil provincial: hacia un estado liberador y un mercado revolucio-
nario, Ramdon House Mondadori, Santiago.
Larrañaga, O. (2001) “Distribución de ingresos: 1958–2001,” in R. Ffrench-Davis and
B. Stallings (eds), Reformas, crecimiento y políticas sociales en Chile desde 1973, LOM
Ediciones/CEPAL, Santiago.
Larrañaga, O. (2006) “¿Qué puede esperarse de la política social en Chile?” Universidad
de Chile, mimeo.
References 285
Le Fort, G. and Lehmann, S. (2003) “ El encaje y la entrada neta de capitales: Chile en

el decenio de 1990,” Revista de la CEPAL 81, December.
Leiva, P. (2004) The Strategic Association Chile–European Union, CELARE, Santiago.
Loayza, N., Fajnzylber, P. and Calderón, C. (2005) “Economic growth in Latin America
and the Caribbean: stylized facts, explanations and forecasts,” Documento de Trabajo
265, Banco Central de Chile, Santiago.
Lüders, R. (1998) “The comparative economic performance of Chile 1810–1995,”
Estudios de Economía 25(2).
Macario, C. (2000) “Chile: in search of a new export drive,” in C. Macario (ed.), Export
Growth in Latin America, Lynne Rienner, Boulder, CO.
McKinnon, R. (1981) “Foreign exchange policy and economic liberalization in LDCs,”
Alternativas de políticas financieras en economías pequeñas y abiertas al exterior, Estudios
Monetarios VII, Banco Central de Chile, Santiago.
Mahani, Z. A., Shin, K. and Wang, Y. (2006) “Macroeconomic adjustments and the
real economy in Korea and Malaysia since 1997,” in R. Ffrench-Davis (ed.), Seeking
Growth under Financial Volatility, Palgrave Macmillan/CEPAL, New York.
Marcel, M. (1989) “Privatización y finanzas públicas: el caso de Chile, 1985–88,”
Colección Estudios CIEPLAN 26, Santiago, June.
Marcel, M. and Meller, P. (1983) “Indicadores líderes de recesión y expansión
económica,” Colección Estudios CIEPLAN 11, Santiago, December.
Marcel, M. and Meller, P. (1986) “Empalme de las cuentas nacionales de Chile 1960–85.
Métodos alternativos y resultados,” Colección Estudios CIEPLAN 20, Santiago,
December.
Marcel, M., Tokman, M., Valdés, R. and Benavides, P. (2001) “Balance estructural del
Gobierno Central, metodología y estimaciones para Chile: 1987–2000,” Estudios de
Finanzas Públicas 1, September.
Marfán, M. (1992) “Re-estimación del PGB potencial en Chile: implicancias para el
crecimiento,” Cuadernos de Economía 87, año 29, August.
Marfán, M. (1998) “El financiamiento fiscal en los años 90,” in R. Cortázar and J. Vial
(eds), Construyendo opciones: propuestas económicas y sociales para el cambio de siglo,
Dolmen Ediciones, Santiago.
Marfán, M. (2005) “La eficacia de la política fiscal y los déficit privados: un enfoque
macroeconómico,” in J. A. Ocampo (ed.) Más allá de las reformas: dinámica estruc-
tural y vulnerabilidad macroeconómica, CEPAL/Alfaomega, Bogotá.
Marshall, J. (1981) “El gasto público en Chile: 1969–1979,” Colección Estudios CIEPLAN
5, July.
Massad, C. (2000) “Capital flows in Chile: changes and policies in the nineties,” in
The Globalization of Financial Markets and the Emerging Economies, ECLAC/Jacques
Maritain Institute, Santiago.
Massad, C. and Zahler, R. (eds) (1988) Deuda interna y estabilidad financiera, vol. II,
Grupo Editor Latinoamericano, Buenos Aires.
Meller, P. (1996a) Un siglo de economía política chilena, 1980–1990, Andrés Bello,
Santiago.
Meller, P. (ed.) (1996b) El modelo exportador chileno: crecimiento y equidad, CIEPLAN,
Santiago.
Meller, P. (2002) “El cobre Chileno y la política minera,” in P. Meller (ed.), Dilemas y
debates en torno al cobre, Santiago.
Meller, P. (ed.) (2005) La paradoja aparente. Equidad y eficiencia: resolviendo el dilema,
Taurus, Santiago.
Meller, P. and Sáez, R. E. (eds) (1995) Auge exportador chileno: lecciones y desafíos,
Dolmen Ediciones/CIEPLAN, Santiago.
286 References
MIDEPLAN, “Resultados de la encuesta de caracterización macroeconómica nacional

(CASEN),” several issues.
MIDEPLAN (2002a) “Sistema Chile solidario: protección social integrada a las 225.000
familias más pobres del país,” Santiago, May.
MIDEPLAN (2002b) “Dinámica de la pobreza: resultados de la encuesta panel
1996–2001,” Santiago, December.
MIDEPLAN (2008) “La encuesta Panel CASEN 1996, 2001, 2006,” Fundación para la
Superación de la Pobreza, Observatorio Social de la Universidad Alberto Hurtado y
MIDEPLAN, Santiago.
Ministry of Finance, Exposición sobre el estado de la Hacienda Pública, several issues.
Ministry of Finance (2009) “Resultados del Comité Consultivo del PIB Tendencial,
2009,” August.
Ministry of Finance (2010) Fondos Soberanos 2009, Annual Report, Santiago.
Mizala, A. (1992) “Las reformas económicas de los años setenta y la industria manu-
facturera,” Colección Estudios CIEPLAN 35, September.
Mizala, A. and Romaguera, P. (2001) “La legislación laboral y el mercado del trabajo:
1975–2000,” in R. Ffrench-Davis and B. Stallings (eds), Reformas, crecimiento y políti-
cas sociales en Chile desde 1973, LOM Ediciones/CEPAL, Santiago.
Mizala, A. and Romaguera, P. (2004) “Remuneraciones y tasas de retorno de los
profesionales chilenos,” in J. J. Brunner and P. Meller (eds), Oferta y Demanda de
Profesionales y Técnicos en Chile, RIL editores, Santiago.
Moguillansky, G. (1999) La inversión en Chile: ¿el fin de un ciclo en expansión?, Fondo
de Cultura Económica/ECLAC, Santiago.
Moguillansky, G. and Titelman, D. (1993) “Análisis empírico del comportamiento
de las exportaciones no cobre en Chile: 1963–90,” Serie Documento de Trabajo 17,
ECLAC.
Molina, S. (1972) El proceso de cambio en Chile, Editorial Universitaria, Santiago.
Molina, S. (2006) Es el tiempo de la equidad, Academia Chilena de Ciencias Sociales,
Políticas y Morales, published by Banco del Desarrollo, Santiago.
Monckeberg, F. (1998) Jaque al Subdesarrollo Ahora, Dolmen Ediciones, Santiago.
Monckeberg, M. O. (2001) El saqueo de los grupos económicos al Estado chileno, Ediciones
B, Santiago.
Morandé, F. and Vergara, R. (eds) (1997) Análisis empírico del crecimiento en Chile,
Centro de Estudios Públicos/ILADES, Santiago.
Moulian, T. (1997) Chile actual. Anatomía de un mito, LOM Ediciones, Santiago.
Moulian, T. and Vergara, P. (1980) “Estado, ideología y políticas económicas en Chile:
1973–78,” Colección Estudios CIEPLAN 3, June.
Muñoz, O. (1986) Chile y su industrialización: pasado, crisis y opciones, Ediciones
CIEPLAN, Santiago.
Muñoz, O. (2007) El modelo económico de la Concertación: 1990–2005, Catalonia/
FLACSO, Santiago.
Núñez, J. and Gutiérrez, R. (2004) “Classism, discrimination and meritocracy in the
labor market: the case of Chile,” Documento de Trabajo 208, Departamento de
Economía, Universidad e Chile.
Ocampo, J. A. (2001) “A new look at the development agenda,” in CEPAL Review 74,
Santiago, August.
Ocampo, J. A. (2007) “The macroeconomics of the Latin American boom,” CEPAL
Review 93, December.
Ocampo, J. A. and Parra, A. (2007) “The dual divergence: growth success and collapses
in the developing world since 1980,” in R. Ffrench-Davis and J. L. Machinea (eds),
References 287
Economic Growth with Equity: Challenges for Latin America, Palgrave Macmillan,
New York.
OECD (2005) Economic Survey of Chile, OECD, Paris.
OECD (2007) OECD Review of Chile’s Innovation policy, OECD, Paris.
Ortega, E. (1987) Transformaciones agrarias y campesinado: de la participación a la
exclusión, CIEPLAN, Santiago.
Perry, G. and Junguito, R. (1983) “Política económica y endeudamiento externo en
Colombia en la década de los setenta,” in R. Ffrench-Davis (comp.), Relaciones finan-
cieras externas y desarrollo nacional en América Latina, Fondo de Cultura Económica,
México.
Pinto, A. (1973) Chile, un caso de desarrollo frustrado, Editorial Universitaria, Santiago.
Pinto, A. (1981) “Chile: el modelo ortodoxo y el desarrollo nacional,” in El Trimestre
Económico 192, México, October–December.
Pizarro, C., Raczynski, D. and Vial, J. (eds) (1996) Social and Economic Policies in Chile’s
Transition to Democracy, CIEPLAN/UNICEF, Santiago.
Política y Espíritu (1975) “Informe económico,” Política y Espíritu 357, Santiago.
Raczynski, D. (1996) “Targeting social programs: lessons from the Chilean experi-
ence,” in C. Pizarro, D. Raczynski and J. Vial (eds), Social and Economic Policies in
Chile’s Transition to Democracy, CIEPLAN/UNICEF, Santiago.
Raczynski, D. and Oyarzo, C. (1981) “¿Por qué cae la tasa de mortalidad infantil en
Chile?” Colección Estudios CIEPLAN 6, December.
Ramos, J. (1986) Neoconservative Economics in the Southern Cone of Latin America,
1973–83, Johns Hopkins University Press, Baltimore.
Ramos, J. (1993) “Macroeconomic equilibria and development,” in O. Sunkel (ed.),
Development from Within: Toward a Neostructuralist Approach for Latin America, Lynne
Rienner Publishers, Boulder, CO.
Ramos, J. (2008) “La economía chilena actual: adios al milagro, bienvenido el blind-
aje,” El Mostrador, electronic media, Santiago, October.
Reinstein, A. and Rosende, F. (2000) “Reforma financiera en Chile,” in F. Larraín and
R. Vergara (eds), La transformación económica de Chile, Centro de Estudios Públicos,
Santiago.
Robichek, W. (1981) “Some reflections about external public debt management,”
Alternativas de políticas financieras en economías pequeñas y abiertas al exterior, Estudios
Monetarios VII, Banco Central de Chile, Santiago.
Rodrik, D. (2001a) “Why is there so much economic insecurity in Latin America,”
CEPAL Review 73, April.
Rodrik, D. (2001b) The Global Governance of Trade as if Development Really Mattered,
UNDP, New York.
Rodrik, D. (2007a) “What produces economic success,” in R. Ffrench-Davis and
J. L. Machinea (eds), Economic Growth with Equity: Challenges for Latin America,
Palgrave Macmillan, Basingstoke.
Rodrik, D. (2007b) “The real exchange rate and economic growth: theory and evi-
dence,” mimeo, Harvard University.
Roldós, J. (1997) “El crecimiento del producto potencial en mercados emergentes:
el caso de Chile,” in F. Morandé and R. Vergara (eds), Análisis empírico del crecimiento
en Chile, Centro de Estudios Públicos/ILADES, Santiago.
Roningen, V. and A. Yeats (1976) “Non-tariff distortions of international trade: some
preliminary empirical evidence,” Weltwirtschaftliches Archiv 112.
Ruiz-Tagle, J. (1979) “Seis años de política social: una revolución en marcha,” Revista
Mensaje 282.
288 References
Ruiz-Tagle, J. (1980) “Desarticulación de la educación,” Revista Mensaje 291.

Ruiz-Tagle, V. J. (1999) “Chile: 40 años de desigualdad de ingresos,” Documento de
trabajo 165, Department of Economics, Universidad de Chile, Santiago.
Ruiz-Tagle, V. J. (2007) “Forecasting wage inequality,” in Estudios de Economia 34(2),
December.
Sachs, J. (1987) “Trade and exchange-rate policies in growth-oriented adjustment
programs,” in V. Corbo, M. Goldstein and M. Khan (eds), Growth-oriented Adjustment
Programs, IMF and World Bank, Washington, DC.
Sachs, J., Larraín, F. and Warner, A. (1999) “A structural analysis of Chile’s long-
term growth: history, prospects and policy implications,” report requested by the
Ministry of Finance.
Sachs, J., A. Tornell, and A. Velasco (1996) “Financial crises in emerging markets: the
lessons from 1995,” Brookings Papers on Economic Activity 1.
Sáez, S. (1991) “Indicadores para las exportaciones chilenas: 1950–89,” Notas Técnicas
138, CIEPLAN, January.
Sáez, S. (2005) “Keeping animal spirits asleep: the case of Chile,” Documento de Trabajo
3615, World Bank, Washington DC, May.
Sáez, S. and Soto, J. (2007) “Tariff and non-tariff barriers to Chilean exports,” Working
Paper, ECLAC, Santiago.
Sanhueza, G. (1999) “La crisis financiera en los años ochenta en Chile: análisis de sus
soluciones y su costo,” Economía Chilena 2(1), Banco Central de Chile, April.
Schmidt-Hebbel, K. (1988) “Consumo e inversión en Chile (1974–1982): una inter-
pretación ‘real’ del boom,” in F. Morandé and K. Schmidt-Hebbel (eds), Del auge a
la crisis de 1982, ILADES, Santiago, June.
Sen, A. K. (2000) Development as Freedom, Anchor Books, New York.
Servén, L. and Solimano, A. (1993) “Economic adjustment and investment perfor-
mance in developing countries: the experience of the 1980s,” in L. Servén and
A. Solimano (eds) Striving for Growth after Adjustment. The Role of Capital Formation,
World Bank, Washington, DC.
Shiller, R. J. (2008) The Subprime Solution, Princeton University Press, Princeton, NJ.
Solimano, A. (1990) “Inversión privada y ajuste macroeconómico. La experiencia
chilena en la década de los 80,” in Colección Estudios CIEPLAN 28, March.
Solimano, A. (2006) Vanishing Growth in Latin America, Edward Elgar, MA.
Solimano, A. and Pollack, M. (2006) La mesa coja: prosperidad y desigualdad en el Chile
democrático, Colección GIGLOB, Santiago.
Stiglitz, J. (1998) “The role of the financial system in development,” Fourth Annual
World Bank Conference on Development in Latin America and the Caribbean, San
Salvador, June.
Sunkel O. (ed.) (1993) Development from Within: Toward a Neostructuralist Approach for
Latin America, Lynne Rienner Publishers, Boulder, CO.
Tapia, H. (2003) “Balance estructural del gobierno central de Chile: análisis y propu-
estas,” in Serie Macroeconomía del Desarrollo 25, CEPAL, August.
Titelman, D. (2001) “Reformas al financiamiento del sistema de salud en Chile,” in
R. Ffrench-Davis and B. Stallings (eds), Reformas, crecimiento y políticas sociales en
Chile desde 1973, LOM Ediciones/CEPAL, Santiago.
Tokman, V. (2004) Una voz en el camino. Empleo y equidad en América Latina: 40 años de
búsqueda, Fondo de Cultura Económica, Santiago.
Torche, A. (1987) “Distribuir el ingreso para satisfacer las necesidades básicas,” in
F. Larraín (ed.), Desarrollo económico en democracia: proposiciones para una sociedad
libre y solidaria, Ediciones Universidad Católica de Chile, Santiago.
References 289
Tybout, J., de Melo, J. and Corbo, V. (1991) “The effects of trade reforms on scale and
technical efficiency: new evidence from Chile,” Policy Research and External Affairs
Working Papers 481, World Bank, Washington, DC.
UNDP (2004) Desarrollo humano en Chile. El poder: ¿para qué y para quién?, Programa de
las Naciones Unidas para el Desarrollo, Santiago.
Uthoff, A. (2001) “La reforma del sistema de pensiones y su impacto en el mercado
de capitales,” in Ffrench-Davis and B. Stallings (eds), Reformas, crecimiento y políticas
sociales en Chile desde 1973, LOM Ediciones/CEPAL, Santiago.
Uthoff, A. and D. Titelman (1998) “The relationship between foreign and national sav-
ings under financial liberalization,” in R. Ffrench-Davis and H. Reisen (eds), Capital
Flows and Investment Performance: Lessons from Latin America, OECD Development
Centre/ECLAC, Paris.
Valdés, R. (1992) “Cuantificación de la reestructuración sectorial generada por la liber-
alización comercial chilena,” Colección Estudios CIEPLAN 35, September.
Valdés-Prieto, S. (1989) “Orígenes de la crisis de la deuda. ¿Nos sobreendeudamos o
nos prestaron en exceso?” Estudios Públicos 33, Summer.
Valdés-Prieto, S. (1992) “Financial liberalization and the capital account: Chile,
1974–84,” in G. Caprio and J. Hanson (eds), Financial Reform: Theory and Experience,
Cambridge University Press, Cambridge.
Valdés-Prieto, S. and Soto, M. (2000) “New selective capital controls in Chile: are they
effective?” Empírica 25(2).
Valenzuela, A. and Valenzuela, S. (eds) (1986) Military Rule in Chile: Dictatorship and
Oppositions, Johns Hopkins University Press, Baltimore.
Vega, H. (2008) En vez de la injusticia, 2nd edn, Random House Mondadori, Santiago.
Velasco, A., A. Arenas, J. Rodrigruez, M. Jorrat and C. Gamboni (2010), “El enfoque de
Balance Estructural en la Política Fiscal en Chile,” in Estudios de Finanzas Públicas,
Ministry of Finance, February.
Vergara, P. (1980) “Apertura externa y desarrollo industrial en Chile: 1974–78,”
Colección Estudios CIEPLAN 4, November.
Vergara, P. (1981) “Las transformaciones de las funciones económicas del Estado de
Chile bajo el régimen militar,” Colección Estudios CIEPLAN 5, July.
Vignolo, C. (1980) “Inversión extranjera en Chile, 1974–79,” Revista Mensaje 286.
Vignolo, C. (1982) “El cobre en el desarrollo nacional: evolución reciente, perspectivas
y estrategias,” Centro de Estudios del Desarrollo/CED, Santiago, August.
Williamson, J. (1981) “The crawling-peg in historical perspective,” in J. Williamson
(ed.), The Crawling-Peg: Past Performance and Future Prospects, Macmillan, London.
Williamson, J. (1996) The Crawling Band as an Exchange Rate Regime: Lessons from Israel,
Chile, and Colombia, Institute for International Economics, Washington, DC.
Williamson, J. (2000) “Exchange rate regimes for emerging markets: reviving the
intermediate option,” Policy Analyses in International Economics 60, Institute for
International Economics, Washington, DC.
Williamson, J. (2003) “Overview: an agenda for restarting growth and reform,” in
P. P. Kuczynski and J. Williamson (eds), After the Washington Consensus: Restarting Growth
and Reform in Latin America, Institute for International Economics, Washington, DC.
World Bank (1997) “Chile: poverty and income distribution in a high-growth econ-
omy: 1987–1995,” Report 16377-CH, November.
Zahler, R. (1980) “Monetary and real repercusions of financial opening-up to the
exterior: the case of Chile, 1975–78,” CEPAL Review 10, April.
Zahler, R. (1988) “Estrategias financieras latinoamericanas: la experiencia del Cono
Sur,” Colección Estudios CIEPLAN 23, Santiago, March.
290 References
Zahler, R. (1998) “The central bank and Chilean macroeconomic policy on the
1990s,” CEPAL Review 64, April.
Zahler, R. (2006) “Macroeconomic stability and investment allocation by domestic
pension funds in emerging economies: the case of Chile,” in R. Ffrench-Davis (ed.),
Seeking Growth Under Financial Volatility, Palgrave Macmillan, New York.
Zañartu, M. (1980) “Gasto social en los pobres,” Revista Mensaje 290, Santiago.
Index
acquisitions 211, 216 Chilean exports to 142, 143, 163–5

ADRs (American Depositary see also specific countries
Receipts) 213, 219, 221, 228 Asian crisis 4, 24, 29–31, 41, 42, 46,
Afghanistan war 241 269
AFPs (Administradoras de Fondos capital flows 155, 214, 219, 221,
Previsionales) 12, 196, 220, 224, 233
241 exports 157, 165
unemployment insurance inequality and poverty 174, 191,
scheme 34, 200 200
agricultural sector 6, 7, 11–12 recession (1999) and recovery 236,
challenges 276 239–40, 243
debt crisis (1982) and recovery 113, Asia-Pacific Economic Cooperation
123 Forum (APEC) 151
exports 69 AUGE health program 34, 192, 237,
price bands 135 246, 273, 277
aid program, debt crisis (1982) and automobile industry 61
recovery 122 Aylwin, Patricio, and regime 24, 25–9
Alessandri, Jorge 5, 8 exchange rate 140
Allende, Salvador, and regime 7–9 interest rates 129
agrarian reform 7, 12 macroeconomic variables 8
foreign banks 82
inequality and poverty 179 Bachelet, Michelle, and regime 24, 34,
macroeconomic variables 8 35–8
nationalization program 7, 11, 13 Council on Equity 254
American Depositary Receipts GDP 270
(ADRs) 213, 219, 221, 228 macroeconomic variables 8
amortization 104, 120, 213, 240, 259 social policy 204
Andean Pact/Community 13, 56, 60, balance of payments
134, 151 Alessandri regime 5
anti-dumping measures 12, 135, 138 Allende regime 9
Anti-unemployment Contingency capital flows 214, 217
Program 259–60 exchange rate policy 91, 263
Argentina exports 134, 166
capital flows 88, 209, 214, 223 external indebtedness 88
currency and debt crisis (1998–2002) financial liberalization 80
17, 77, 158, 165, 241 Ibáñez regime 5
economic complementarity import liberalization 69–71
agreement 151 Pinochet regime 13, 15–16, 19
investment ratio 32 balance of trade
savings 233 debt crisis (1982) and recovery 112,
tequila crisis 223, 245 117
article 14 87, 97 export development 133, 135, 170
Asian countries external indebtedness 104
capital account 238–9 Banco del Estado 81, 250
291
292 Index
banking/financial sector financial liberalization 78, 79, 80,

Allende regime 7, 13 82–8, 105–6
capital flows 210, 216, 221, 225–6 Frei Ruiz-Tagle regime 31–2
debt crisis (1982) and recovery 19, interest rates 99, 102
122, 123 Latin American countries 88, 154,
external indebtedness 96–7 209, 214, 216
financial liberalization 81–3, 85, 87 macroeconomic effects 105
inequality and poverty 205 macrostabilizing measures,
interest rates 97–8, 99, 101–2 effectiveness of 226–32
nationalization 7, 13 magnitude and composition 211–16
Pinochet regime 11, 13, 18 policy counter-cyclical response
privatization 13, 81 216–26
regulation and supervision 81, 83, policy lessons 234–5
90, 96–100, 225–6 recession (1999) and recovery 236,
reserves see reserves 238, 240, 241, 256
bankruptcies 14, 19, 22, 44–5 savings, investment, and growth in
import liberalization 66, 68 the 1990s 232–4
Basle Accord 226 structural fiscal balance 256
best practice, and export capital formation
development 132 capital flows 232
binding external constraint (BEC) 132, challenges 276
209, 244 debt crisis (1982) and recovery 113,
black market 9 115, 125–6, 127
Bolivia 151 democracy 27, 30, 38, 170
Brazil 85, 88, 90, 151, 241 dictatorship 22, 144
Budget Office 23 exports 144, 170
building sector 104 external indebtedness 103
bureaucracy 250 financial liberalization 81
inequality and poverty 202–3
Canada 151 macroeconomics for
capital account development 272
Asian countries 238–9 recession (1999) and recovery 255
Aylwin regime 28, 29, 30 capital market segmentation 74
capital flows 218, 220, 226–9 car industry 61
exchange rate policy 263, 264, 266 Cartagena Agreement 56
exports 157 CASEN surveys 178–9, 185, 191
external indebtedness 96 Central America, free trade
financial liberalization 81, 82, 105 agreement 151
Frei Ruiz-Tagle regime 29, 30 central bank
Latin American countries 238–9 capital flows 210, 212–13, 216–21,
recession (1999) and recovery 238–9, 222–6, 230, 231, 232
241, 243, 244 debt crisis (1982) and recovery 120,
capital adequacy ratios 225, 226 122
capital flows 209–11 democracy 28, 30, 31, 33, 35,
challenges 274–5 36, 37
debt crisis (1982) and recovery 77, dictatorship 24
112, 114 exchange rate policy 91, 92–3,
exchange rate policy 266 139–40, 155, 263
exports 155, 170 external indebtedness 96
external indebtedness 103 import liberalization 71
Index 293
recession (1999) and recovery recessive adjustment (1999–2003)

238–41, 243, 246, 247, 249, 31–4
251–3, 259 recovery (2004–9) 35–8
structural fiscal balance 259 reform of the reforms (1990–8)
Central Bank Autonomy Act 1990 140 25–31
challenges for the future 274–7 slogan 4
children 254 social policy 204
Chile Barrio program 34, 237 Consejo Trabajo y Equidad 254
Chile Foundation 136 Constitution 26
Chile Solidario program 34, 184, 191, construction sector 104
192, 202, 237, 246, 253, 273, 277 consumer goods, and import
China 150, 151, 160, 162, 163–4 liberalization 61, 62–3, 64, 68
Christian Democrats 5 consumer price index (CPI)
CODELCO (Corporación Nacional del Bachelet regime 37
Cobre) capital flows 217
democracy 26, 39, 160, 166 Ibáñez regime 5
dictatorship 11, 12 Pinochet regime 14, 16–17, 21
exports 160, 166 recession (1999) 244, 246, 251, 252
molybdenum 160 Copper Buffer Fund (CBF) 39, 40, 92,
recession (1999) and recovery 248, 216, 259
250 copper industry
tax reform 194 Allende regime 7
Colombia 88, 92, 151 capital flows 218
commodity boom 35, 37 debt crisis (1982) and recovery 110,
comparative advantages 113, 116–17, 120, 128
export development 133, 151, 153, democracy 26, 33, 35, 36, 39–40
154, 162, 173 exports 156, 157, 159, 160–1,
external indebtedness 103 166–7, 169
financial liberalization 77, 83 dictatorship 12, 13, 14, 20–1, 23
import liberalization 71, 72, 73 exports 134, 141, 142
Compendium of Rules for International exchange rate policy 90, 91, 95, 140,
Exchange in Chile, Chapter XIX 263, 264
119 exports
competitivity democracy 156, 157, 159, 160,
Bachelet regime 37 166–7, 169
challenges 275 dictatorship 134, 141, 142
competition policy 123 Frei Montalva regime 6, 7
exchange rate policy 262 globalization of volatility 159
exports 132, 133, 172–3 import liberalization 55, 66, 69
import liberalization 71, 74–5 inequality and poverty 201
Lagos regime 37 and molybdenum 159, 160
recession (1999) and recovery 253 recession (1999) 241, 248–9, 251–2
Concertación Democrática (CD) 24–5, structural fiscal balance 256, 257
268 tax reforms 194
comparison with dictatorship see also CODELCO
268–71 Copper Price Committee 248
defeat in 2010 elections 254 Copper Stabilization Fund 6, 7, 248,
inheritances, impact of 273 257
macroeconomic variables 8 CORFO 11, 137, 167
outcomes 38–43 Costa Rica 151, 180
294 Index
Council on Equity 254 capital flows 211

crawling-peg exchange rate 90–3 export development 136–7
Aylwin regime 27 demand 43
capital flows 217–18, 224 Allende regime 7
debt crisis (1982) and recovery 116, Aylwin regime 26, 27, 28, 29
123 Bachelet regime 36
exports 134, 135, 139–40, 154 capital flows 216, 217, 234
external indebtedness 90 challenges 274
financial liberalization 80 debt crisis (1982) and recovery
Frei Montalva regime 97 109–10, 112–13, 116, 121–2, 124,
Frei Ruiz-Tagle regime 31, 33 127, 128
Pinochet regime 54, 134, 135, export development 132, 134, 135,
139–40 144
credit external indebtedness 88, 103
capital flows 218 financial liberalization 79, 104, 105
external indebtedness 88–92, 96–7, Frei Ruiz-Tagle regime 32
102 import composition 61
financial liberalization 81, 82, 83, import liberalization 64, 65–6, 67–8,
105 72, 73
interest rates 97–9, 100–2 interest rates 100
Mexican and East Asian crises 214 Pinochet regime 13, 14, 15, 20–1,
microcredit 254 23, 24
Pinochet regime 13 recession (1999) and recovery
recession (1999) and recovery 242–3 238–40, 242, 247, 249, 251–2, 262
creditworthiness, Chile’s reputation democracy see Concertación
for 82 Democrática; specific regimes
Cuba 180 demographic changes 188, 245
current account development policy
Aylwin regime 28 Bachelet regime 37
capital flows 210, 212, 215–18, challenges 275–6
224–5, 229, 230 deficit of development-friendly
debt crisis (1982) and recovery 78, reforms 273–4
114, 115 historical perspective 5–9
exchange rate policy 139, 155, 263, macroeconomics 272–3, 274–5
264 recession (1999) and recovery 253–4
external indebtedness 89, 96, 104 dictatorship see Pinochet, Augusto and
financial liberalization 82, 83, 84, regime
86, 88 dynamism, and import
Frei Ruiz-Tagle regime 30 liberalization 72, 73
import liberalization 68, 69–71
Mexican 214 Economic and Social Stabilization
Pinochet regime 16, 17 Fund 259
recession (1999) and recovery 237, economic complementarity
238, 240 agreements 151
economies of scale and
debt crisis see under recessions and specialization 131
financial crises Ecuador 151
debt paper, purchase of 119–20 education and training
debt swaps 110, 117, 118–20, 121, challenges 276, 277
125 democracy 34, 38, 42
Index 295
dictatorship 10, 179, 180 exports

export development 132 democracy 150, 152, 154–6, 166,
inequality and poverty 185–7, 167, 169, 171, 172
191–3, 197–201, 203–4, 207 dictatorship 131, 133, 134, 135,
recession (1999) and recovery 239, 138–40, 141
254 external indebtedness 88, 89–96,
structural fiscal balance 259 97–8, 103, 104
efficiency and import liberalization financial liberalization 80, 81, 88,
71–5 105
El Salvador 151 Frei Montalva regime 7
employment see labor market import liberalization 55, 57–60, 71,
ENAP 11 72
encaje 28, 29, 105, 210, 224, 234 manufacturing sector 66, 68
ENDESA 122 inequality and poverty 197, 206–7
energy sector 255 Mexico 214
entrepreneurs potential GDP 46
bank loans 100 recession (1999) and recovery 236,
capital flows 231 238–44, 246–7, 249, 251, 256,
challenges 275, 276 262–7
debt crisis (1982) and recovery 125, structural fiscal balance 256, 259,
126 262
democracy 36, 38, 42 export investment financing 137
export development 134 exports
inequality and poverty 203 1950s 5
recession (1999) 249, 250, 254 acceleration of receipts 227
structural fiscal balance 259 Alessandri regime 5
environmental issues 255, 276–7 Asian countries 57, 71, 75
equity, as crucial to sustained capital flows 210, 217, 224
growth 271–2 challenges 275
European Union (EU)/European debt crisis (1982) and recovery 110,
Economic Community 114–17, 121–2, 125–6, 128
Chilean exports to 142, 143, 152, democracy 24–5, 40, 41, 149–73
162–4 Aylwin regime 28
exports from 251 Bachelet regime 36, 37
tariffs 57 Frei Ruiz-Tagle regime 31
trade agreement with Chile 151 Lagos regime 31, 35
exchange rates 43 dictatorship 13, 17, 18, 20, 21, 22,
Alessandri regime 5 130–1
Allende regime 7 dynamism and diversification
capital flows 210–11, 215–18, 220–5, 141–3
229, 232–4 economic growth 144–5
challenges 274, 275 national development 131–3
debt crisis (1982) and recovery 77, trade policy 133–40
110, 112, 115–16, 121–2, 126–8 exchange rate policy 90–3, 263, 264,
democracy 27–8, 30, 31, 33, 35, 36 265, 266
exports 150, 152, 154–6, 157–8, Frei Montalva regime 6, 7
166, 167, 169, 171, 172 and import liberalization 57, 59, 60,
dictatorship 13, 14, 15–17, 20–1, 22 69, 71–2, 73, 75
exports 131, 133, 134, 135, manufacturing sector 64, 66–7
138–40, 141 inequality and poverty 197
296 Index
exports – continued Aylwin regime 26–7

non-traditional 137, 141–2, 151–2, Bachelet regime 37, 38
156–8, 161 Frei Ruiz-Tagle regime 30–1, 32–3
over-invoicing of 227 Lagos regime 33
recession (1999) 240, 247, 249, 251, dictatorship 14, 20
253 external indebtedness 90
subsidies see subsidies inequality and poverty 192
export warehouses 137 recession (1999) 237–9, 243, 244,
external debt see foreign indebtedness 246–7, 248, 252–3, 255–62
external inflation 16 see also fiscal balance
externalities, export development 132 Fiscal Responsibility Law 257, 260
external vulnerability 102–4 Fondo de Garantía para Exportaciones
Exxon 12 No-Tradicionales 137
Fondo Nacional de Salud
family allowances 180, 181, 191, 192 (FONASA) 195–6
family subsidy 191 food prices 37, 252
Finance, Ministry of foreign banks 82, 85
Bachelet regime 36 foreign currency 28, 31, 165, 209, 212,
exchange rate policy 155 240
molybdenum 160 foreign direct investment (FDI)
Pinochet regime 12 capital flows 209, 211, 213–16, 219,
recession (1999) and recovery 248 221, 228–30, 234
structural fiscal balance 259, 260 debt crisis (1982) and recovery 118
financial crises see recessions and democracy 28, 30, 38, 166–7
financial crises dictatorship 11, 13
financial liberalization 104–6 exports 166–7
debt crisis (1982) and recovery 78–83 financial liberalization 81, 82, 83,
inequality and poverty 195 105
financial sector see banking/financial inequality and poverty 195
sector recession (1999) and recovery 247
financierism 38, 42, 43, 103, 127, 254, short-term funds disguised as 227,
269, 272 228
inequality and poverty 205–6, 207 foreign exchange 43
fiscal balance capital flows 213, 216, 217–18, 221,
Alessandri regime 5 226, 227, 230
Allende regime 7 chapter XII of regulations 220
Aylwin regime 27 debt crisis (1982) and recovery 110,
Bachelet regime 36 120, 123
capital flows 210 exchange rate policy 263
debt crisis (1982) and recovery 121–2 exports 131, 132, 135, 165
Frei Ruiz-Tagle regime 30 external indebtedness 88–96, 97–8
interest rates 99 Frei Montalva regime 6
structural (SFB) 33, 255–62 Frei Ruiz-Tagle regime 33
see also fiscal policy Pinochet regime 19
fiscal policy 43 exports 131, 132, 135
capital flows 216 import liberalization 55, 69–71, 75
challenges 274 foreign indebtedness 78–8, 104–6
debt crisis (1982) and recovery capital flows 218
127–8 debt crisis (1982) and recovery 111,
democracy 42 112–13, 117–18, 121–2, 125
Index 297
dynamics 102–4 inequality and poverty 192, 194

export development 133, 135 infrastructure investment 170, 239
import liberalization 68 Pinochet regime 12, 13
interest-rate differentials 97–102 recession (1999) and recovery 239,
macroeconomic adjustment 88–97 244, 246, 256–7
Pinochet regime 18, 20 structural fiscal balance 256–7, 260
foreign loans see also social expenditure
Alessandri regime 5 Great Britain 202
Allende regime 7 Great Depression 5
capital flows 218 gross domestic product (GDP)
external indebtedness 103 Allende regime 7, 9
financial liberalization 81, 88, 105 capital flows 217, 218, 233
Pinochet regime 16 debt crisis (1982) and recovery
recession (1999) and recovery 241 109–10, 112–16, 124–8
foreign reserves see reserves democracy 24–5, 38, 40, 41, 149–56,
free trade agreements (FTAs) 150–1, 167–73
173, 275 Aylwin regime 25, 27, 28–9
Frei Montalva, Eduardo and regime Bachelet regime 35–6, 37
5–7 Frei Ruiz-Tagle regime 30, 31,
agrarian reform 6, 7, 12 32, 33
crawling exchange rate 92 Lagos regime 31, 32, 35, 37
foreign banks 82 dictatorship 13–15, 17–19, 21–3, 41
inequality and poverty 179 dictatorship and democracy
macroeconomic variables 8 compared 268, 269–71
trade reform 54 estimating potential GDP 43–9
Frei Ruiz-Tagle, Eduardo and export development 130, 133, 135,
regime 24, 25, 29–33, 269 144–5
macroeconomic variables 8 financial liberalization 78, 84
Frei Montalva regime 6
General Agreement on Tariffs and Trade recession (1975) 64
(GATT) 135, 153 recession (1999) and recovery 236–7,
GINI coefficient 175, 178, 189, 190, 240, 242–3, 245, 247, 249, 252–3,
191 255, 257, 267
dictatorship and democracy structural fiscal balance 257, 260–1,
compared 270 262
global financial crisis 37, 237–8, 252, trend versus potential 33
253, 269 see also productivity
exchange rate policy 263 gross national product (GNP) 17, 18,
structural fiscal balance 259, 260 104
globalization/globalism Guatemala 151
challenges 275, 277
neoliberalism 10 health sector
open regionalism 151 AUGE health program 34, 192, 237,
government spending 246, 273, 277
Allende regime 7 inequality and poverty 180, 191–2,
Aylwin regime 26 193, 195–6
Bachelet regime 37 Lagos regime 34
Frei Montalva regime 6 Pinochet regime 10, 11, 180
Frei Ruiz-Tagle regime 30 recession (1999) and recovery 237,
Ibáñez regime 5 254
298 Index
Herfindahl–Hirschman Index democracy 26, 37–8, 40–1

(HHI) 161, 162 dictatorship 11, 20, 23, 24
Honduras 151 dictatorship and democracy
housing 253 compared 268–9, 270
hyperinflation 9, 10, 127 external indebtedness 88
financial liberalization 105
Ibáñez, Carlos and regime 5 import composition 61, 62–3, 66, 75
illiteracy rate 179 interest rate differentials 102
import deposits 55 recession (1999) and recovery 236
import liberalization reforms 193–206
Alessandri regime 5 trends 176–93
democracy 150 income tax 194
dictatorship 11, 13, 21, 22, 53–4, 75–6 incremental capital-output ratio
composition of imports 61–3 (ICOR) 46–8, 49
evaluation 68–75 India 151
manufacturing sector 64–8 industrial modernization program,
process 54–60 Frei Montalva regime 6, 7
trade development 131, 135 industrial relations 6, 195
external indebtedness 93, 95, 96 see also labor unions and
financial liberalization 77, 88 organizations
inequality and poverty 197 inequality see income distribution
interest rates 97, 100 infant mortality rate 180
imports inflation 43
Alessandri regime 5 Alessandri regime 5
Allende regime 9 Allende regime 7, 9
Bachelet regime 36 Aylwin regime 27
capital flows 218 Bachelet regime 36, 37
debt crisis (1982) and recovery capital flows 217, 225
109–10, 112, 114–17, 123, 128 debt crisis (1982) and recovery 124,
delayed payment for 227 127, 128
exchange rate policy 264, 265, 266 exchange rate policy 88–9, 90–2,
and export development 131, 133, 93–6, 98, 139, 140
137 external 16
external indebtedness 103, 104 financial liberalization 82
Frei Montalva regime 7 Frei Montalva regime 6, 7
liberalization see import liberalization Frei Ruiz-Tagle regime 30, 31
marketing 68, 74, 103, 115 hyperinflation 9, 10, 127
Pinochet regime 15–16, 18, 21 Ibáñez regime 5
recession (1999) and recovery 249, import liberalization 60
251 interest rates 99
substitution see import substitution Lagos regime 35
under-invoicing of 227 Pinochet regime 13–17, 21
import substitution 5, 61 recession (1999) and recovery 240,
and export development 133, 139, 241, 244, 246, 249, 251–3
152 structural fiscal balance 261
income distribution 174–5, 206–8 informal labor market 196
Allende regime 9 infrastructure
capital flows 210 capital formation 170
challenges 277 public spending on 170, 239
debt crisis (1982) and recovery 126 structural fiscal balance 259
Index 299
inheritances, impact of 273 democracy 38

innovation Aylwin regime 25, 27, 29
democracy 38, 42 exports 157, 167
inequality and poverty 203, 205 Frei Ruiz-Tagle regime 30, 32
recession (1999) and recovery 250, Lagos regime 32, 35
254 dictatorship 18–19, 23
structural fiscal balance 259 exports 132, 134, 144
Inter-American Development Bank exchange rate policy 263
(IADB) 239 exports
interest rates 43 democracy 157, 167
Aylwin regime 27, 28, 129 dictatorship 132, 134, 144
Bachelet regime 36, 37 export investment financing 137
capital flows 210, 211–12, 217, 218, external indebtedness 88, 102, 103
222–3, 231, 232 FDI see foreign direct investment
challenges 274 financial liberalization 79–80, 81,
debt crisis (1982) and recovery 77, 105
122, 123, 125, 128–9 capital flows 84, 88, 105–6
export development 134, 140, Frei Montalva regime 6
155 import composition 61
external indebtedness 90, 96–102, import liberalization 64, 66, 73
103, 104 inequality and poverty 202–3,
financial liberalization 79–83, 88 204–6
Frei Ruiz-Tagle regime 31 interest rates 100
import liberalization 73 potential GDP 44
inequality and poverty 206–7 recession (1999) and recovery 236–7,
Pinochet regime 11, 13, 16, 17, 22 242, 244–5, 248–9, 255
exports 134, 140 Iraq wars 217, 241
recession (1999) and recovery 240, ISAPRES 195–6
243, 249, 251, 253
intermediate goods 61 Japan
Internal Revenue Service 23 Chilean exports to 142, 143, 163–4,
International Development Bank 165
(IDB) 20, 117 tariffs 57
international financial crisis 37, 237–8, trade agreement with Chile 151
252, 253, 269 justice, public expenditure on 239
exchange rate policy 263
structural fiscal balance 259, 260 Korea
international financial institutions agreement with Chile 151
(IFIs) 103, 117–18 capital flows 214
International Monetary Fund (IMF) 19, domestic positive shock 35, 246
20, 29, 117 exports 161, 251
international reserves see reserves intensive development period 25
investment investment ratio 32, 245
Alessandri regime 5 macroeconomic imbalance 246
Allende regime 7 tariffs 57
capital flows 210, 213, 220–1, 229, trade and financial reforms 57, 207
231, 232–4
challenges 276 labor force participation
debt crisis (1982) and recovery 78, rate 34
113, 114, 116, 122, 124–6 women 40, 193, 199
300 Index
labor market GDP 36, 41, 149, 170–2, 236, 247–9,

Allende regime 9 253
capital flows 210, 232, 233 hyperinflation 127
challenges 276 import liberalization 72
debt crisis (1982) and recovery 109, infant mortality rate 180
112, 127 international financial
democracy 184, 190–3, 195–8, institutions 117
200–1, 203–4 investment ratio 88
exchange rate policy 266 open regionalism 151
import liberalization 64, 65, 68, 74 per capita bank debt 85
inequality and poverty 40–1, 175, regional integration 7, 275
202, 204 regressive trends (1980s) 202
informal 196 tariffs 138
Lagos regime 34 trade liberalization 165
output gap 34 Latin American Integration Association
recession (1999) and recovery 245, (LAIA) 220
247, 254 law of one price 94, 95
training see education and training life expectancy 180
unemployment see unemployment life insurance companies 221
labor relations 6, 195 literacy rate 179
see also labor unions and Lula da Silva, Luiz Inacio 241
organizations
labor rights 23, 26, 34, 195, 229 macroeconomic adjustment to external
labor unions and organizations indebtedness 88–97, 102
Aylwin regime 26 macroeconomic balance 42–3, 272–3
Frei Montalva regime 6 capital flows 216–17
inequality and poverty 195 debt crisis (1982) and recovery
Pinochet regime 10, 11, 13, 14, 126–9
16, 20 exchange rate policy 267
Lagos, Ricardo and regime 24, 25, 31, export development 144
32, 33–4, 35, 37, 269 inequality and poverty 175, 204–6,
fiscal policy 256 207
macroeconomic variables 8 labor market 34
social security system 193 private investment 27
land reform and ownership recession (1999) and recovery 243,
debt crisis (1982) and recovery 126 245
Frei Montalva and Allende macroeconomic effects
regimes 179 of capital flows 105
Pinochet regime 11, 12 of import liberalization 64–6
Latin American countries (LACs) macroeconomic framework and import
capital account 238–9 liberalization 71–3
capital flows 88, 154, 209, 213, 215 macroeconomics for development
Chilean exports to 142, 143, 151, 272–3, 274–5
162–4 Malaysia 35, 246
commodity boom 35 manufacturing sector
debt conversion 119 debt crisis (1982) and recovery 112,
exchange rates 154–5 113
exports 149, 161, 170, 171–2 dictatorship 22, 136, 142, 143
foreign capital, net use of 89 effect of trade liberalization on 64–8
foreign debt crises 109, 120 exchange rate policy 95
Index 301
exports NAFTA 150, 173, 215

democracy 161–2, 165 National Bureau of Statistics (INE) 179,
dictatorship 136, 142, 143 188
external indebtedness 104 National Common Fund 195–6
inequality and poverty 197 National Council for Innovation and
marketing imported products 68, 74, Competitivity 170
103, 115 nationalization programs 6, 7, 11, 13
market segmentation 102 National Training and Employment
MERCOSUR 151, 158, 165 Service (SENCE) 197–8
mergers 68 neoliberalism 3, 9–10
Mexico domestic financial liberalization and
capital flows 209, 214–15 external debt 77–106
Chilean exports to 165 import liberalization 53–76
economic complementarity inequality and poverty 193–8, 207
agreement 151 outcomes 21–4
investment ratio 32 pragmatic, with regressive bias
NAFTA 173, 215 (1982–9) 10, 20–1, 109–29
savings 233 pure (1973–81) 10–19
tequila crisis 28, 30, 77, 83, 214–15, net transfer of funds (NTF) 80
223, 228, 233, 245 Nicaragua 151
microcredit 254 non-tariff barriers 55, 57, 60, 133, 150
military coup 6, 9 North American Free Trade Agreement
minimum wage 180, 181, 184, 192, (NAFTA) 150, 173, 215
195
molybdenum 69, 96, 160 oil industry
monetarism 93 democracy 35, 36, 39
monetary policy dictatorship 13
Aylwin regime 28 globalization of volatility 159
capital flows 211–12, 232 and import composition 61
debt crisis (1982) and recovery 115, Iraq crisis 217
123, 128 recession (1999) and recovery 249,
exchange rate policy 264 251–2
external indebtedness 88, 89–90, 92, oil shocks 13, 14
104 one price, law of 94, 95
financial liberalization 80 open regionalism 151
Pinochet regime 13–16, 22 Organization for Economic Cooperation
recession (1999) and recovery 244, and Development (OECD) 29,
246, 253, 256 215
structural fiscal balance 256, 261 output see gross domestic product; gross
monetary sterilization 28 national product; productivity
money supply
Allende regime 7 Paraguay 151
debt crisis (1982) and recovery 128 parliamentary elections 26, 42, 254
external indebtedness 88, 89, Pension Reserve Fund 259
103–4 pensions
financial liberalization 82, 104 debt crisis (1982) and recovery 122
Ibáñez regime 5 democracy 184, 191, 196, 201
Pinochet regime 14, 16, 21 dictatorship 11, 12, 180
multinational corporations 11 private funds see private pension
mutual funds 213, 221, 238 funds
302 Index
pensions – continued debt crisis (1982) and recovery 123

recession (1999) and recovery 237, export development 135, 138
254, 259 price liberalization 11, 12, 13, 57
Peru 151 private debt, “nationalization” of 20,
Pinochet, Augusto and regime 3, 9–10 110, 117–18
Central Bank Autonomy Act 1990 private pension funds 196
140 capital flows 220, 221
comparison between dictatorship and challenges 276
democracy 268–71 Frei Ruiz-Tagle regime 31
Constitution 26 recession (1999) and recovery 238,
109–29 privatization
import liberalization 11, 13, 21, 22, debt crisis (1982) and recovery 110,
53–4, 75–6 120, 122, 125–6
composition of imports 61–3 inequality and poverty 195
evaluation 68–75 interest rates 100
manufacturing sector 64–8 Pinochet regime 11, 12, 13
process 54–60 recession (1999) and recovery 250
trade development 131, 135 PRO-CHILE 72, 134, 138, 154
inequality and poverty 174, 179–84 product differentiation 74
macroeconomic variables 8, 22 production function 48–9
outcome of regime 21–4 productive frontier (PF) 33, 38
plebiscite (1988) 140 capital flows 224, 232, 233
pragmatic neoliberalism, with debt crisis (1982) and recovery 113,
regressive bias (1982–9) 10, 20–1 128
pure neoliberalism (1973–81) 10–19 democracy 29, 30, 32, 36
savings 18–19, 233 estimation 43, 44, 46–9
social security reform 12, 27 exports 171
political parties 20, 24 macroeconomics for
portfolio investment 213, 214, 220, development 272
243 recession (1999) 252
poverty 174–5, 206–8 structural fiscal balance 260, 261
Aylwin regime 26 productivism 38, 42, 43, 254, 267,
challenges 277 272–3
Chile Solidario 34, 184, 191, 192, productivity
202, 237, 246, 253, 273, 277 Allende regime 9
definition 176 Aylwin regime 28
recession (1999) and recovery 246, capital flows 210
254 challenges 274
reforms 193–206 debt crisis (1982) and recovery
trends 176–93 109–10, 112–16, 124–6
poverty line 176–8, 185–7 exchange rate policy 95
Presidential Advisory Council for Labor external indebtedness 104
and Equity 184 Frei Montalva regime 6, 7
presidential campaigns and elections Frei Ruiz-Tagle regime 32
1989 24 import liberalization 64–8, 73
1999 32 inequality and poverty 203
2010 42, 254, 270 macroeconomics for
price controls and bands development 272
Allende regime 9 Pinochet regime 13, 14, 21–2
Index 303
recession (1999) 236–7, 242, 244, external indebtedness 88–90, 96,

251 98–9
see also gross domestic product; gross financial liberalization 81, 83, 86
national product Frei Montalva regime 6
professional associations 10 import liberalization 69
public expenditure see government Pinochet regime 16
spending recession (1999) and recovery 239,
public purchasing mechanisms 12–13 243, 246
resource allocation
quotas, and financial liberalization exchange rate policy 262, 264, 266,
82, 90 275
exports 171
real estate 243 financial liberalization 79, 104
recessions and financial crises import liberalization 72
1975 18, 19, 21, 24, 64 interest rate differentials 102
1982 debt crisis 3, 13, 19, 20, 21, 24 risk rating of Chilean economy 27,
accumulation of imbalances 213, 246
110–12
capital flows 213 sales taxes 11
causes 77–106 salmon industry 136, 142, 159–60
export development 131, 133 SALs (Structural Adjustment
public debt 27 Loans) 117, 216
recovery 109–10, 117–29 savings
shocks 112–17 capital flows 84, 105–6, 209, 231,
1999 236–8, 267 232–4
decelerated recovery challenges 276
(from 2006) 249–55 debt crisis (1982) and recovery 112,
dynamic recovery (2004–5) 247–9 114, 122, 125
exchange rate policy democracy 27, 33, 38–9
corrections 262–7 dictatorship 18–19, 233
impact 240–3 external indebtedness 88, 102, 103
origins 238–9 financial liberalization 79–80, 81, 84,
recessive phase (1999–2003) 243–7 105–6
structural fiscal balance 255–62 import liberalization 75
2009 37 interest rates 101–2
Ibáñez regime 5 recession (1999) and recovery 237,
structural fiscal balance 33, 255–62 255, 262
reform of the reforms 4, 25–31, 150, structural fiscal balance 262
269, 273 self-employment 196
inequality and poverty 190, 192 SENCE (National Training and
regional integration 7, 275 Employment Service) 197–8
regulatory framework 5 seniors, inequality and poverty 201
reserves 81, 83, 90, 97–9 September 11, 2001 attacks 241
Alessandri regime 5 shanty towns, eradication
Allende regime 7 programme 34
capital flows 86, 210, 213, 216, small and medium-sized enterprises
218–19, 224, 226–33 (SMEs)
debt crisis (1982) and recovery 110, capital flows 229, 231, 232
113, 114, 115 democracy 36, 37, 38, 42
exchange rate policy 139 exchange rate policy 266
304 Index
SMEs – continued subsidies 110, 116

export development 137 democracy 153–4
financial liberalization 105 dictatorship 20
inequality and poverty 203, 205, GATT/WTO restrictions 153
206–7, 208 inequality and poverty 195
recession (1999) and recovery 243, Supintendencia de Bancos 123, 242
245, 250, 254
structural fiscal balance 259 Taiwan 25, 57, 161, 207
social expenditure tariffs
Allende regime 7, 179 Bachelet regime 36
Aylwin regime 26 capital flows 218
capital flows 216 countries other than Chile 57
challenges 277 debt crisis (1982) and recovery 116,
democracy and dictatorship 123
compared 271 drawback scheme 135–6, 153–4
inequality and poverty 206 exchange rate policy 263
democracy 184, 191–2, 193, exports
194–5, 201 democracy 150, 152, 157
dictatorship 180, 182 dictatorship 133, 134, 135–6, 138–9
Pinochet regime 180, 182 import liberalization 53, 54, 55–60,
trends 181 72, 75
see also government spending Pinochet regime 12, 133, 134, 135–6,
social movements 24 140
social policy 273 recession (1999) and recovery 251
dictatorship and democracy taxation
compared 270 capital flows 105, 216, 218, 226, 229
inequality and poverty 204 challenges 277
recession (1999) and recovery 237, debt crisis (1982) and recovery 122,
246, 253–4 128
social security system democracy 26, 27, 28, 30, 37, 39
capital flows 210 inequality and poverty 184, 192,
debt crisis (1982) and recovery 122, 194, 206, 207
126 dictatorship 11, 21, 23, 182
democracy 40–1 Frei Montalva regime 6, 7
dictatorship 12, 27, 180 income tax 194
inequality and poverty 193, 195 rebates 153
reform (2008) 184, 192, 273, 277 recession (1999) and recovery 237,
South Korea see Korea 239, 248, 253, 256–7
steel industry 160 sales taxes 11
stock markets structural fiscal balance 256–7, 260
capital flows 213, 214, 223, 229 value-added tax (VAT) 11, 26, 137,
120–2 technology
interest rates 100 capital flows 233
recession (1999) and recovery 243 challenges 275
strikes 6, 9 export development 132, 137
Structural Adjustment Loans inequality and poverty 203, 205
(SALs) 117, 216 terms of trade
structural fiscal balance (SFB) 33, Allende regime 7
255–62 capital flows 212, 223, 224
Index 305
debt crisis (1982) and recovery 113, dictatorship 14, 19, 183–4
114, 116–17, 120, 124, 128 exchange rate policy 95
democracy 29, 30, 31, 35, 39 and import liberalization 64, 71, 72,
exports 157, 158, 159–61 73, 74
dictatorship 13, 14, 21, 131, 135 insurance 34, 200–1, 206, 254, 273
exchange rate policy 140, 266 recession (1999) and recovery 242,
exports 244, 256
democracy 157, 158, 159–61 structural fiscal balance 256
dictatorship 131, 135 trade policy 152
Frei Montalva regime 6 Unidad de Fomento (UF) 244
import liberalization 66 Unidad Popular (UP) 7
recession (1999) and recovery 236, United Kingdom 202
237, 240, 247, 249, 250, 256 United States of America
structural fiscal balance 256, 259 ADRs 213, 219, 221, 228
time deposits 219 capital account liberalization 29
total factor productivity (TFP) 43, 44, capital flows 218, 222–3
48, 203, 250, 264 Chilean exports to 142, 143, 152–3,
trade deficit see balance of trade 162–4
trade liberalization exports 251
debt crisis (1982) and recovery 113 free trade agreements 150, 151
effects on manufacturing 64–8 GDP 41, 236
export development 133 loans to Chile 7, 20
external indebtedness 103 NAFTA 150
Frei Montalva regime 54 regressive trends (1980s) 202
Latin American countries 165 tariffs 57
Pinochet regime 12–13, 22, 53 unremunerated reserve requirement
see also import liberalization (URR) 218–19, 224, 232, 239,
trade policy 244
debt crisis (1982) and recovery 123, Uruguay 151
128 utility tariffs 7
democracy 149–56, 166, 196–7
exchange rates 262–3, 266 value added
export development 130–1, 132–40 debt crisis (1982) and recovery 115
inequality and poverty 196–7 exchange rate policy 95, 266
trade surplus see balance of trade exports
trade unions see labor unions and democracy 152, 154, 158, 161–2,
organizations 165, 167–9, 172
training see education and training dictatorship 142
Treasury external indebtedness 103
crawling exchange rate 90, 92–3 import liberalization 64, 65, 73, 75
debt crisis (1982) and recovery 122 Pinochet regime 17, 18
global financial crisis 237–8 tax (VAT) 11, 26, 137, 192, 194
recession (1999) 243, 252, 259 Venezuela 88, 151
unemployment wage levels

Anti-unemployment Contingency Allende regime 7, 9, 179
Program 259–60 Aylwin regime 26
debt crisis (1982) and recovery 127 capital flows 210
democracy 34, 40, 184, 190–1, 193, debt crisis (1982) and recovery 116,
200–1 122, 124, 125
306 Index
wage levels – continued poverty line 177–8

exchange rate policy 93, 95 trends 181
Frei Montalva regime 6 water subsidy 191
import liberalization 64, 65, 72 women
inequality and poverty 177–8, 184, labor force participation 40, 193, 199
190, 192–3, 197, 198–9, 204 unemployment 200
minimum wage 180, 181, 184, 192, World Bank 20, 29, 117, 216, 239
195 world trade 156–7, 252
Pinochet regime 14, 16, 23, 24, 180, World Trade Organization (WTO) 150,
202–3 151, 153, 167

Economic Reforms in Chile - From Dictatorship To Democracy

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Economic Reforms in Chile - From Dictatorship To Democracy

Transféré par

Droits d'auteur :

Formats disponibles

From Dictatorship to Democracy

1. Chile—Economic policy. 2. Chile—Economic conditions—1970–1973.

List of Figures vii

First Part: An Account of Four Decades 1

4 Strong economic recovery, weak average growth 124

Fourth Part: Development Challenges for Democracy 147

I.1 Actual and potential GDP, 1960–2009 (log scale, GDP*

I.1 Comparison of key macroeconomic variables, 1959–2009 8

IV.4 Indexes of stock prices in pesos and converted to US dollars,

IX.2 Gross investment and depreciation, 1995–2009 (% of GDP

III.1 The “crawling” exchange rate: a pioneer reform in 1965–70 92

This book reflects my research and views on crucial developments in the

By 1990, in the return to democracy, the Chilean economy faced the

1 A historical perspective on development strategies

gradual, multi-anchored, non-recessionary stabilization program; an indus-

Table I.1 Comparison of key macroeconomic variables, 1959–2009a

A change of the head of the economic policy team in the middle of

2 The neoliberal strategy, 1973–893

The period covered by the dictatorship of Augusto Pinochet is longer than

first, between 1973 and 1981, is a case of orthodoxy or neoliberalism in its

(a) Pure neoliberalism, 1973–81

(i) The reforms

to attenuate the transmission of external instability into the domestic econ-

(ii) Achievements and failures

100 Actual GDP (%) 2.9 5.3

team recognized – implicitly – that monetary control was proving to be unable

in the following twelve months jumped to 32%, in contrast with the 4%

(iii) Vulnerability and spurious growth

1975–80 1975–81 1975–80 1975–81

1 GDP 3.8 4.0 2.3 2.5

accrued to foreigners, which meant that the rate of expansion of the

(b) The introduction of pragmatism, with a regressive bias in 1982–9

and economic activity accelerated, culminating in the overheating of the

(c) A summary of outcomes

Table I.3 Macroeconomic variables during dictatorship, 1974–89a (annual averages)

GDP growth (%) 3.0 2.9 2.9

Source: Same sources as Table I.1.

The modernized business sector featured a surge of new economic

from dictatorship to democracy – in order to stop an overheating of the

3 Democracy, reforming the reforms and development,

to accomplish this goal, the government of President Aylwin had to gain

non-financial public sector savings from 2% in the 1980s to nearly 5% of

past four decades, in contrast to previous occasions, this time a comprehen-

“vulnerability zones” – particularly with an appreciated real exchange rate

Table I.4 Public and private budget balances, 1990–2007

1990–5 –2.3 –4.4 2.0

Source: Author’s calculations based on Balance of Payments data from

(b) Recessive adjustment, 1999–2003

5% of GDP in eighteen months (see Ffrench-Davis and Tapia, 2001; Zahler,

copper price.31 Consequently, when the economy is overheated the Treasury

labor rights. Also, temporary public employment programs were intensified,

(c) Recovery led by a positive external shock in 2004–8 and

trade was dynamic up to the arrival of the international financial crisis. On

(d) A summary of outcomes

Table I.5 Gross savings ratios, 1982–2008 (% of GDP at current prices)

External National Public sector Stabilization funds Other

1982–4 8.4 3.1 –1.8 – 4.9

The savings capacity is strongly affected by the terms of trade, which

In all, the copper stabilization fund has represented an outstanding factor

Table I.6 Per capita GDP growth, 1974–2009 (annual

1974–81 1.5 2.0 1.5

Source: Central Bank of Chile, ECLAC and the IMF.