Off Print:

Finance-led Capitalism?
Macroeconomic Effects of Changes in the Financial Sector

Edited by

Eckhard Hein, Torsten Niechoj, Peter Spahn, Achim Truger

Metropolis-Verlag
Marburg 2008
 Financial systems in developing
countries and economic development

Hansjrg Herr

1. Introduction
The wave of globalisation over the past decades finds its centre in the in-
ternational financial system. The deregulation of financial markets has
led to an enormous amount of new innovations and an explosion of inter-
national capital flows. The deregulation of financial systems started in
the developed world in the 1970s. However, after a time lapse most de-
veloping countries and later the countries of the former Soviet block libe-
ralised their domestic financial systems (e.g. by switching to market-
determined interest rates and allowing foreign currency deposits in the
domestic banking system) and at the same time reduced or eliminated
capital controls (e.g. allowing domestic firms to take credits abroad and
households to keep bank accounts, debt securities or shares abroad). The
question is whether these changes have increased the chance for develop-
ing countries to catch up quickly or whether development is becoming
more difficult under present conditions.
Believing in the neutrality of money, Robert Lucas (1988) argued that
the relation between financial and economic development is over-
stressed. More surprisingly, Joan Robinson (1952) pointed out that fi-
nancial development follows economic growth passively. Although she
supported a Keynesian tradition she underestimated the role of money.
As Jan Kregel (1985) puts it, Keynesian economics without money is like
Hamlet without the prince. In this essay money and the financial system
are considered of key importance for economic development in both de-
veloped and developing countries. The World Bank (2001) also strongly
supports the belief that financial systems are of paramount importance

3:32 , 04.11.2008, 07 Herr.doc

124 Hansjrg Herr

for economic development. Empirically, a positive development of fi-

nancial markets is a good predictor for future economic growth (King/
Levine 1993). The World Bank stresses the supply side and the produc-
tivity effects of financial systems. Therefore, the contribution of finance
to long-term growth is to improve the economys total factor productivity
rather than the quantity of capital. (World Bank 2001, 41). According to
the World Bank (2001) a more developed financial system leads to a bet-
ter allocation of financial funds, allows for the accumulation of small
amounts of savings, involves broader risk dispersion, improves gover-
nance structures in enterprises, and offers more venture capital, etc. Al-
though many of the points mentioned above can be supported, the role of
finance cannot be captured adequately if it is considered merely as an
improvement of total factor productivity or as quasi-technical progress.
In the following an interpretation of financial systems and develop-
ment in a monetary-oriented Keynesian tradition is given. In the second
part the importance of finance for development is stressed. In the third
part, this is applied in an analysis of developing countries. It is found that
in many developing countries the domestic financial system is not able to
create sufficient credit for dynamic development. A short case study on
China shows that under certain conditions the domestic financial system
can facilitate or even promote development. Part four deals with the pos-
sibility of tapping foreign financial markets to finance development. The
risks of such a strategy, notably boom-bust cycles and Dutch disease, are
also discussed. The final part offers concluding remarks.

2. A Schumpeterian-Keynesian approach to finance and development

It was Schumpeter (1911) who argued that development is only possible
if innovative entrepreneurs receive credit created ad hoc (in the German
original out of nothing) to invest:

Our second thesis now runs: in so far as credit cannot be given out of
the results of past enterprise or in general out of reservoirs of purchas-
ing power created by past development, it can only consist of credit by
means of payment created ad hoc, which can be backed neither by
money in the strict sense nor by products already existing. It can in-
deed be covered by other assets than products, that is by any kind of
property which the entrepreneur may happen to own. But this is in the

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 125

first place not necessary and in the second place does not alter the na-
ture of the process, which consists in creating a new demand for, with-
out simultaneously creating, a new supply of goods. (Schumpeter
1911, 106).

Keynes also argued along these lines:

The ex-ante saver has no cash, but it is cash which the ex-ante inves-
tor requires. >@ If there is no change in the liquidity position, the pub-
lic can save ex-ante and ex-post and ex-any-thing-else until they are
blue in the face, without alleviating the problem in the least. (Keynes
1937a, 665 f.).1

What is necessary for economic development is new credit created first

and foremost by the banking system. With such an approach savings will
be created out of new income stimulated by investment. In Figure 1 such
a process is shown. It starts when the banking system gives new and ad-
ditional credit to firms to invest. Banks do not need additional financial
funds or even savings to give additional credit. With the help of the cen-
tral bank, commercial banks can create additional credit out of nothing.
The act of credit creation leads automatically to the creation of additional
monetary wealth since a banking system simply gives credits by increas-
ing its deposits. The vital point is that additional credit leads to additional
investment. Investment is the link between the asset market and the
goods market. Investment creates demand, additional capacity, produc-
tion and employment, and also additional income in the form of wages
and profits. What we see here is a credit-investment-income-saving me-
chanism. The process is started by credit and investment; income and
savings are the result. Some of the created deposits flow to households
via income. Savings are kept by the population as cash, bank deposits or
deposits given to entrepreneurs via the capital market. Banks can start
giving additional credit to let the circuit of investment and income start
again. Usually banks are needed as circuit starters for investment and
production (Bossone/Sarr 2002), however, the central bank has to ac-
commodate an expansion process by money creation because part of this
central bank money will trickle away in additional cash holdings of the
public and additional reserve holdings of banks.

1
Also cp. Keynes (1937b):

3:32 , 04.11.2008, 07 Herr.doc

126 Hansjrg Herr

Figure 1: The credit-investment-income process

refinancing from
additional bank central bank
deposits

additional
debt-securities
and shares commercial
banking system

savings of
households,
firms, gov-
ernment credit for
investment in
working
productive capital
capital

income-
creation
process

Of course, finance is not the only factor needed to create development.

For example, entrepreneurs must also exist to invest in productive activi-
ties. Economic expansion can be inflationary if demand becomes too
high and a demand-driven inflation triggers a wage-price spiral. Under
conditions of full capacity utilisation and low unemployment, inflation
can become a problem. As inflationary processes are inherently unstable
and lead to the danger of cumulative developments, central banks must
fight against inflationary processes. In such a situation the central bank
does not become a circuit starter, it becomes a circuit stopper as restric-
tive monetary policy will lead to a slowdown of demand, to lower credit
and money expansion and to a reduction in GDP growth and employment
(Heine/Herr 2003).

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 127

3. Finance and development in developing countries

3.1 The hard macroeconomic monetary budget constraint
Empirically there is a close link between the development of the financial
system and the level of per capita income. A common indicator for the
development of financial systems is the relation between domestic credit
(given in domestic and foreign currency) to GDP. In Table 1 this relation
is given for countries with different levels of GDP per capita. In 2006 in
low-income countries domestic lending by banks reached only about
55% of GDP; in middle-income countries it was about 77%, and in high-
income countries nearly 200%.2 In some low-income or even middle-
income countries domestic credit to GDP is extremely low. In these
countries a domestic based credit-investment-income mechanism is quite
obviously not functioning. Indeed, in many countries the domestic finan-
cial system is fundamentally distorted and not able to deliver a sufficient
quantity of credit. It is noteworthy that China, a low-income country, has
a ratio of domestic credit to GDP of 138% and realises a considerably
higher percentage than one would expect. One of the secrets of the Chi-
nese development can be found in this extraordinary development of do-
mestic credit (see below).
Why is domestic credit to GDP so low in developing countries? A first
hint can be found in looking at dollarisation (including euroisation).3
Dollarisation means that part of domestic monetary wealth is kept in for-
eign currency.
To understand dollarisation we can follow Keyness analysis of mon-
ey. According to him the advantage of keeping money can be expressed
in a premium.

The possession of actual money lulls our disquietude; and the pre-
mium which we require to make us part with money is the measure of
the degree of our disquietude. (Keynes 1937c, 216).

2
Other indicators issued debt-securities or stock market capitalisation in % of GDP,
would show the same relationship.
3
Other currencies do not play an important role. In March 2006 45% of all domes-
tic foreign currency deposits were kept in U.S. dollar, 26% in euro, 4% in yen, and
3% in Swiss franc (BIS 2006).

3:32 , 04.11.2008, 07 Herr.doc

128 Hansjrg Herr

Table 1: Domestic Bank Credit in % of GDP in 2006

(credits in domestic and foreign currency)

Domestic bank
credit in % of GDP
Low-income countries 55
Middle-income countries 77
High-income countries 195
China 138
Armenia 8
Bangladesh 58
Brazil 82
Cambodia 6
India 64
Korea, Rep. 107
Malaysia 125
Mexico 40
Mongolia 25
Peru 15
Russian Federation 21

Source: World Development Indicators (2008).

The liquidity premium is a non-pecuniary subjectively given rate of re-

turn obtained from the holding of a certain stock of liquidity. The mar-
ginal liquidity premium falls with the quantity of liquidity held, as with a
higher stock of liquidity the disquietude is reduced (Keynes 1936, chap-
ter 17). Wealth owners look for the best money which satisfies their de-
mand to lull their disquietude in an uncertain world. Different national
monies fulfil these demands to a different extent as they have different
qualities.4

4
It can also be argued that different monies have different reputations or different
brand names.

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 129

Equation 1 gives the well-known interest-rate parity between two cur-

rencies. The gross rate of return for domestic interest bearing titles (the
left hand side of the equation) must have the same value as the gross rate
of return for foreign interest bearing titles (the right hand side of the equ-
ation) in equilibrium.

ee
(1+ iD ) = (1+ iF ) (1)
e
iD: domestic interest rate, iF : foreign interest rate, e: spot exchange rate (in-
e
crease means nominal depreciation), e : future or expected exchange rate (in-
crease means nominal depreciation)

Keyness idea of a liquidity premium can be transferred to the quality of

different currencies. In this case each currency in the world earns a spe-
cific non-pecuniary rate of return, a currency premium. The currency
premium can be interpreted as a country specific liquidity premium,
which expresses the specific quality of the currency.5 A wealth owner
may have a certain exchange rate expectation. However, in a world with
uncertainty he or she knows that not all future events can be known
and/or it is not possible to give all known future events certain probabili-
ties. This leads us to the conclusion that there exists at one point in time
an exchange rate expectation, but in addition there is a currency premium
which measures the subjective trust given to ones own expectation.6
Typically, low-quality currencies suffer from more known unknowns
and unknown unknowns than high-quality currencies and have a rela-
tively low currency premium. First of all, the currency premium depends
on the present and even more the expected internal and external stability
of a currency. Factors like the size of the currency area, the size, liquidity
and sophistication of the financial market in the currency or economic
policies in favour of wealth owners are also important. In addition, the
currency premium includes factors beyond narrow economic considera-
tions; it also expresses the political stability and the international role and
economic and military power of the country issuing the currency, etc.
This means that even if exchange rates were expected to be stable, it

5
Country risk premia are sometimes used to express a similar idea. A risk premium
reduces the rate of return of an asset; a currency premium increases the rate of re-
turn of an asset.
6
This is a method used by Keynes (1921).

3:32 , 04.11.2008, 07 Herr.doc

130 Hansjrg Herr

must be assumed that currencies will have substantially different curren-

cy premia. Equation (2) gives the equilibrium condition for the rate of re-
turn of monetary wealth in two currencies including currency premia.

ee
(1  iD )(1  lD ) = (1  iF )(1  lF ) (2)
e
lF: foreign marginal currency premium, lD: domestic marginal currency premium

In Figure 2 two hypothetical marginal currency premium functions of a

wealth owner for the Russian rouble and the U.S. dollar are drawn. Let us
assume the same expected pecuniary returns (interest rate differentials
plus exchange rate expectations) in Russian roubles and U.S. dollars.
Under this condition our wealth owner will restructure his or her portfo-
lio until the marginal currency premia of the two currencies have the
same value, let us say l1. Then, in our example, expressed in a common
currency the quantity WRouble is held in roubles and the quantity WDollar in
U.S. dollars. The stock of wealth kept in U.S. dollars is higher than the
stock of wealth kept in roubles. The wealth owner obviously believes that
the U.S. dollar is a better currency and more trustworthy than the rouble.
Even in countries with currencies of poor quality, wealth holders
usually keep some monetary wealth in domestic currency for transaction
purposes. However, much more important is the demand for monetary
wealth beyond the needs for daily transactions. This explains why in de-
veloping countries a small part of monetary wealth is kept in domestic
monetary wealth and the bigger part in foreign monetary wealth. Holding
a small quantity of monetary wealth in a low-quality currency may lead
to a relatively high marginal currency premium. However, in comparison
to high-quality currencies the marginal currency premium drops quickly
and substantially when the stock of wealth held in the currency increases.
Given the actual exchange rates and theoretically assuming that a
wealth owner holds the same quantity of monetary wealth in all curren-
cies of the world, we can construct a currency hierarchy according to the
subjectively calculated hierarchy of the marginal currency premia.
In Figure 2, wealth of WDollar held in U.S. dollars creates a marginal
currency premium of l1. If the same quantity of wealth is held in roubles
the marginal currency premium for wealth in roubles is only l0. Thus, in
our illustration the Russian rouble has a lower quality than the U.S. dollar
and ranks lower in the currency hierarchy than the U.S. dollar.

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 131

Figure 2: Marginal currency premia of two countries*

Marginal currency
premia (1)

11

U.S. dollar
10
Rouble

Quantity of wealth held in a

common currency
WRouble WDollar

Note: * The same pecuniary return in U.S. dollars and roubles is assumed.

Developing countries are usually unable to reduce domestic credit expan-

sion in domestic currency and thus domestic monetary wealth to such
small quantities that the national marginal currency premia reach the lev-
el of marginal currency premia in developed countries. This would lead
to the strangulation of the domestic economies. Usually they offer higher
interest rates to compensate for lower currency premia. Under this condi-
tion domestic credit expansion can be higher; however, very high interest
rates also lead to the strangulation of the economy. High interest rates are
also unfavourable as they lead to a low wage share as a percentage of na-
tional income and at the same time to the suppression of a credit-
investment-income-creating process. Thus developing countries have to
optimise between two evils, low credit expansion and high interest rates.
They are typically trapped in a situation of relatively high interest rates
and low domestic credit expansion in domestic currency.
Let us look at the empirical situation. In many countries in the world
domestic currencies play a subordinated role. In the case of deposit dolla-
risation the public keeps foreign currency deposits in the domestic bank-

3:32 , 04.11.2008, 07 Herr.doc

132 Hansjrg Herr

ing system. Table 2 shows that deposit dollarisation is high and exceeds
50% or more of domestic deposits in many countries. Domestic banks
use foreign currency deposits to give foreign currency loans inside the
country and/or channel the collected foreign currency deposits abroad if
there are no good domestic debtors. Table 2 reveals only the peak of for-
eign currency holdings. There are two other ways to hold wealth in for-
eign currency which are statistically difficult to detect. Firstly, foreign
wealth can be held inside the country in foreign banknotes and is part of
dollarisation. Especially in countries with unstable banking systems
and/or the danger of confiscation of monetary wealth by the government,
cash holdings are high (De Nicol et al. 2003). Secondly, the wealth held
abroad is quantitatively more important. Dollarisation can be considered
the capital flight of small wealth owners as it is costly and needs informa-
tion to keep wealth in a foreign country. Keeping wealth outside the
country, capital flight in the original sense, is typical for the rich in de-
veloping countries. Taking into account the high degree of dollarisation
in many developing countries and adding cash holdings in foreign cur-
rencies and capital flight to foreign countries, it becomes clear that finan-
cial systems in most developing countries are highly penetrated and na-
tional currencies only partly take over domestic monetary functions.
Central banks in developed countries, as mentioned above, will ac-
commodate a circuit of domestic credit expansion, investment and
growth as long as there is no inflationary danger. The key point is that
developing countries are confronted with a systematically tighter ma-
croeconomic monetary budget constraint than developed countries.7

7
Kornai (1980) spoke about monetary budget constraints to distinguish market
economies (hard budget constraints) from planned economies (soft budget con-
straints).

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 133

Table 2: Evolution of average foreign currency deposits

to total deposits (in %)

Number of
Regions countries 1996 1997 1998 1999 2000 2001
South America 8 45.8 46.1 49.4 53.2 54.0 55.9
Transition Econo- 26 37.3 38.9 43.5 44.3 46.9 47.7
mies
Middle East 7 36.5 37.2 37.7 37.5 38.2 41.9
Africa 14 27.9 27.3 27.8 28.9 32.7 33.2
Asia 13 24.9 28.0 26.8 28.8 28.7 28.2
Central America and 7 20.6 20.8 22.0 22.1 22.5 24.7
Mexico
Caribbean 10 6.3 7.6 6.8 6.7 6.1 6.2
Developed Countries 14 7.4 7.5 7.5 6.7 7 6.6

Source: De Nicol et al. (2003).

Countries with low-quality monies can afford only a relatively small cre-
dit expansion in domestic currency. Let us assume 100 units of credit and
monetary wealth in domestic currency are created in such a country. Ac-
cording to the preference of wealth owners a certain percentage of the
newly created wealth will be exchanged in hard currencies. If wealth
holders want to keep 50% of their monetary wealth in hard currency in-
side and/or outside the country a very conservative assumption for
most developing countries 50 units of domestic wealth in domestic cur-
rency will be exchanged in hard currency. Everything unchanged, this
leads to a depreciation of the domestic currency which can be accepted
only to a limited extent by a developing country. Firstly, a depreciation
may trigger inflationary processes, especially as in developing countries
prices and wages are sometimes pegged to the exchange rate. Secondly,
as in developing countries foreign debt is nearly completely denominated
in foreign currency, another consequence of the poor quality of their mo-
nies, any real depreciation leads to an increase of the real debt burden of
debtors in foreign currency. Thirdly, the export and import elasticity may
be very low, a case in which very large real exchange rate movements are
needed to improve the trade balance or a depreciation even increases a
current account deficit. And finally, a real depreciation may lead to such

3:32 , 04.11.2008, 07 Herr.doc

134 Hansjrg Herr

a reduction of real income that a (larger) part of the population falls into
poverty. It follows that countries with low-quality currencies have to stop
domestic credit expansion in domestic currency very quickly and are not
able to initiate a Keynesian-Schumpeterian credit-investment-income-
creating process which is the backbone of economic development. The
macroeconomic monetary budget constraint in many developing coun-
tries is too hard and does not allow for development.

3.2 Credit expansion and development in China

In China domestic credit in % of GDP is extraordinarily high for a devel-
oping country (Table 1). This relates to the overall very dynamic eco-
nomic development in China (Herr/Priewe 2005). Since economic re-
forms were introduced in 1978, China has been experiencing 30 years of
rapid GDP growth of around 9% per year (Table 3). Taking one U.S. dol-
lar a day as a measure, China is the country with the biggest absolute re-
duction in poverty in the world in spite of the alarming and increasing
unequal income distribution. It did not follow a big-bang strategy; a gra-
dual strategy of reform was chosen with far-reaching government inter-
ventions. For example, in the sequence of reforms privatisation was put
to the end of transition, property rights remained rather unclear and pric-
es were liberalised gradually as well as foreign trade. China grew slowly
out of the plan as quantity planning was relaxed slowly and for a long
time substituted by politically influenced credit allocation (Naugthon
2007, 85). Development in the enterprise sector has been based on three
pillars. As the first pillar, state-owned enterprises stimulated growth via
high investment and were used for industrial policy which combined in-
fant industry protection and support for certain industries. A second pillar
consisted of small- and medium-sized market based enterprises. After
1978 in this sector a Manchester type of capitalism with waves of enter-
prise foundations and bankruptcies developed. This is the sector with
partly very poor working conditions, no safety net and a general low lev-
el of legal regulation. Joint venture enterprises with foreign companies
and for a few years an increasing number of wholly-owned foreign enter-
prises constitute a third pillar. Over the years the share of state-owned en-
terprises has became smaller but in industrial production it is still around
30%.

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 135

In searching for the growth drivers in China it becomes clear that

growth has been stimulated by high investment demand and the external
sector. Growth rates in gross capital formation as well as the proportion
of gross capital formation in % of GDP have been high. In addition, Chi-
na has followed a policy of avoiding deficits in the current account. Es-
pecially since the second half of the 1990s high growth rates of exports
and high current account surpluses have characterised Chinese develop-
ment (Table 3). The Chinese central bank, the Peoples Bank of China
(PBoC), has been steadily intervening in the foreign exchange market to
prevent an appreciation of the renminbi (RMB). Thus the PBoC is the
cause of the high Chinese current account surpluses and not low Chinese
wages. China has accumulated the highest foreign exchange reserves in
the world amounting to about 2 trillion U.S. dollars in 2008.
Overall, China has managed to defend macroeconomic stability. Al-
though inflation rates started to get out of control in the early 1990s, they
could be reduced in the years that followed. In 1994 the RMB was
pegged to the U.S. dollar; since 2005 when China switched to a currency
basket peg, the RMB has been appreciating moderately vis--vis the U.S.
dollar. The overall good performance as far as inflation rates, exchange
rate stability, GDP growth and the current account are concerned led to a
relatively high confidence in the domestic currency and the domestic fi-
nancial system compared with other developing countries. Since 1978
China has had no currency crises and has also never suffered from the
erosion of the domestic monetary system. Until today, China has had a
comprehensive system of capital controls which has been relaxed a bit in
the past few years (Herr 2008). The logic of the system is simple: all
types of capital flows are controls with the exception of foreign direct in-
vestment (FDI) inflows. For example, capital outflows are strictly con-
trolled. Of course, there were some legal and also illegal ways of export-
ing and importing capital, however; these have never threatened the
overall development in China.

3:32 , 04.11.2008, 07 Herr.doc

136 Hansjrg Herr

Table 3: Basic macroeconomic data for China 19902006

expenditure,
Gross capi-

services, %

Current ac-
FDI net in-
flows in %
Final Con-

ance, % of
Exports of
% of GDP

goods and

count bal-
tal forma-
tion, % of

rate (CPI)
sumption
GDP per

Inflation
growth*

growth*

of GDP
of GDP
capita
GDP

GDP

GDP
Year

1978 11.7 10.2 38.2 62.1 6.6 0.7 0.0 n/a

1979 7.6 6.1 36.1 64.4 8.6 n/a 0.0 n/a
1980 7.8 6.5 34.8 65.5 10.7 7.5 0.0 0.1
1981 5.2 3.9 32.5 67.1 12.7 2.4 0.1 0.8
1982 9.1 7.5 31.9 66.5 12.3 1.9 0.2 2.0
1983 10.9 9.3 32.8 66.4 10.9 1.5 0.3 1.4
1984 15.2 13.7 34.2 65.8 11.3 2.8 0.5 0.6
1985 13.5 11.9 38.1 66.0 10.0 9.3 0.5 -3.8
1986 8.8 7.2 37.5 64.9 11.8 6.5 0.6 -2.4
1987 11.6 9.8 36.3 63.6 16.4 7.3 0.9 0.1
1988 11.3 9.5 37.0 63.9 17.1 18.8 1.0 -0.9
1989 4.1 2.5 36.6 64.5 16.7 18.0 1.0 -1.0
1990 3.8 2.3 34.9 62.5 19.2 3.1 1.0 3.1
1991 9.2 7.7 34.8 62.4 21.0 3.4 1.2 3.3
1992 14.2 12.8 36.6 62.4 22.5 6.4 2.7 1.3
1993 14.0 12.7 42.6 59.3 23.3 14.7 6.2 -1.9
1994 13.1 11.8 40.5 58.2 24.6 24.1 6.0 1.4
1995 10.9 9.7 40.3 58.1 23.1 17.1 4.9 0.2
1996 10.0 8.9 38.8 59.2 20.1 8.3 4.7 0.8
1997 9.3 8.2 36.7 59.0 21.8 2.8 4.6 3.9
1998 7.8 6.8 36.2 59.6 20.3 -0.8 4.3 3.1
1999 7.6 6.7 36.2 61.1 20.4 -1.4 3.6 1.4
2000 8.4 7.6 35.3 62.3 23.3 0.4 3.2 1.7
2001 8.3 7.5 36.5 61.4 22.6 0.7 3.3 1.3
2002 9.1 8.4 37.9 59.6 25.1 -0.8 3.4 2.4
2003 10.0 9.3 41.0 56.8 29.6 1.2 2.9 2.8
2004 10.1 9.4 43.2 54.3 34.0 3.9 2.8 3.6
2005 10.4 9.8 42.7 51.8 37.3 1.8 3.5 7.2
2006 11.1 10.5 42.5 49.9 40.1 1.5 3.0 9.4

* Calculated in constant prices, change to previous year.

Source: China Statistic Yearbook (2007), International Monetary Fund, World Eco-
nomic Outlook Database (April 2008), World Development Indicators (2008).

10:06, 18.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 137

Without permission Chinese banks, firms and households are not allowed
to take foreign credit. As a result, foreign debt in China compared with
foreign assets is low. For FDI inflows China is one of the most open
countries in the world. Prasad and Wei (2005, 19) from the IMF are cor-
rect when they state:

In the context of the discussion on the benefits and potential risks of

financial globalisation, the dominance of FDI in Chinas capital in-
flows implies that it has been able to control the risks and get more of
the promised benefits of financial integration than many other emerg-
ing markets that have taken a less cautious approach to capital account
liberalisation.

Actually, for a few years China has been suffering from too high and
partly speculative capital inflows especially because of high FDI in-
flows and other financial flows connected with FDI like credits and trade
flows between foreign parent companies and subsidiaries in China.
Without the PBoCs interventions in the foreign exchange market China
would most likely be pushed into huge current account deficits.
Compared with other developing countries dollarisation in China is
relatively low. Domestic foreign currency deposit in 2006 were below
4% of total domestic deposits; domestic foreign currency credits in % of
total domestic credits were below 6% (PBoC 2008). It is important here
to recognise that capital export controls and relatively low dollarisation
prevented the early curtailing of domestic credit expansion which is the
explanation of the low ratio of domestic bank credit to GDP in so many
developing countries.
Development in China would not have been possible without the fi-
nancial system supporting and pushing investment. The backbone of the
financial system are the four big state-owned commercial banks which
today cover over 60% of deposit collection and bank credits given. Near-
ly all other banks, for example city banks or so-called policy banks, are
also state owned. Until the late 1990s there had been a credit plan with
credit ceilings. A large part of credits was policy driven. At least for two
decades after the start of reform in 1978 the PBoC, or better the State
Council, decided on the credit volume including the regional distribution
of credits. On a provincial level credit allocation was negotiated between
local government, the local branches of the banks, the local branch of the
PBoC and state-owned firms. This implied that on a macroeconomic lev-

3:32 , 04.11.2008, 07 Herr.doc

138 Hansjrg Herr

el the monetary budget constraint was hard as in a developed market

economy, whereas on a microeconomic level it was partly soft.8 Credits
were given for investment, however, also loss-making firms were fi-
nanced to prevent higher unemployment and soften the problem of the
non-existence of a pension system and health insurance beyond the firm
level. As of today (end 2008) the PBoC continues to set deposits and
lending rates and uses moral persuasion to control credit volumes. The
Chinese financial system is not that much different from the ones in the
successful East Asian countries after World War II. Japan and the Asian
tigers followed a policy of low interest rates, credit rationing, policy dri-
ven credit expansion and allocation, industrial policy, capital controls and
export promotion (Stiglitz/Uy 1996). These successful models collapsed
after the liberalisation of the domestic and external financial system.
The Chinese financial system was not developed without costs. Since
the second half of the 1990s state-owned enterprises have come under in-
creasing pressure from the growing Chinese private sector which has not
had to pay for a firm-based welfare state system and the growing foreign
investment sector which has usually used superior imported technology.
These developments, together with a partly soft budget constraint, led to
the accumulation of non-performing loans. The resulting financial pres-
sure triggered the privatisation of small- and medium-sized state-owned
enterprises which were the least productive in the state-owned enterprise
sector (Lau 2000). Official Chinese data reported that in 2002 non-
performing loans in the banking system amounted to 19.0% of GDP; the
percentage went down to 6.7% in 2005 (Naughton 2007, 462).9 These
figures may not give the full extent of the problem; however, they show
that the problem is under control. And we should not forget that non-
performing loans present to a large extent quasi fiscal deficits as banks
took over government functions to finance the social safety net. And last
but not least non-performing loans are denominated in domestic curren-

8
Defending macroeconomic stability was not always easy for the PBoC. Especially
in the early 1990s commercial banks found ways to circumvent credit ceilings by
creating non-bank financial institutions. The banking reform in China in 1994 es-
tablished macroeconomic control of the PBoC again.
9
In 2004 the relation of domestic credit to GDP in China was 166.9%. The drastic
reduction of the ratio to around 120% in 2005 reflects that balance sheets were
cleaned up from non-performing loans (World Bank 2008).

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 139

cy. This stands in sharp contrast to many over-indebted developing coun-

tries which are indebted in foreign currency.
The financial system and especially the interaction between state-
owned banks and state-owned enterprises has been the demand-side
backbone of the Chinese economy. It has stimulated high investment
with all of its multiplier effects for the other sectors of the economy per-
manently. High investment was partly policy-driven the high invest-
ment of state-owned firms and partly driven by the expanding private
sector. The latter had nearly no access to credits from state-owned banks.
It financed investment from own profits and via the large grey and in-
formal credit market. FDI supported the dynamic of the Chinese econo-
my. Its main effect has to be seen in the transfer of technology and man-
agement skills and the opening of export channels. However, it would be
misleading to understand FDI as the main growth engine. In the 1980s
FDI inflows were very low and GDP nevertheless grew at the same rate
as later when FDI inflows exploded.
Thereafter, China managed to create a Schumpeterian-Keynesian cre-
dit-investment-income-creation process which led to economic prosperi-
ty. This process was driven by political credit expansion and allocation
and by a dynamic private sector including foreign enterprises. The ex-
pansion process was externally protected by strict capital controls and a
policy to avoid current account deficits.

4. Capital inflows and development

Countries with distorted domestic financial systems with a lack of credit
can tap foreign financial markets. Potential credit volumes are unre-
stricted and interest rates in New York, London or Frankfurt are lower
than in the domestic markets of many developing countries. In this sec-
tion it will be argued that foreign credit can become a substitute for a
lack of domestic credit, although tapping foreign credit markets involves
high risks of failure. Firstly, foreign credit in developing countries is
nearly always denominated in foreign currency. Under the condition of
high credit denominated in foreign currency, any sharp depreciation in-
creases the real debt burden of domestic debtors and leads to a simulta-
neous currency and banking crisis (twin crises). Eichengreen et al. (2003)
called this fundamental disadvantage for developing countries, for exam-

3:32 , 04.11.2008, 07 Herr.doc

140 Hansjrg Herr

ple compared with the highly foreign indebted USA, as original sin of
low-quality currencies. Secondly, even without the danger of currency
crises, capital inflows may erode the competitiveness of developing
countries and lead to long-run negative effects (Dutch disease).

4.1 Boom-bust cycles and twin crises

Since the deregulation of international capital flows in the 1970s the
world economy has experienced several global boom-bust cycles of capi-
tal flowing into developing countries and then suddenly flowing back,
creating twin crises (Williamson 2005).10 During the first boom phase in
the second half of the 1970s (cp. Figure 3) capital flowed mainly to Latin
America as other parts of the developing world still used capital controls.
During the bust phase in the early 1980s inflows halved as investors were
shocked by the Latin American debt crisis which started in 1982 in Mex-
ico and spread to nearly all countries in Central and South America.
About one decade later, in the early 1990s, a new huge wave of capital
flooded the group of developing countries. During this time most of the
capital swept into Asian and transition countries which had opened their
capital accounts. The so-called Tequila crisis in Mexico in 1994 brought
only a short interruption of this boom phase. This boom phase came to a
violent end in 1997. Almost unnoticed it started in the former Czechoslo-
vak Republic in spring 1997 and escalated later in the year in Thailand,
South Korea, Malaysia and Indonesia. The Russian crisis followed in
1998. After a long struggle, Argentina, a sweetheart of the IMF during
the 1990s, fell into crisis in 2001, as did Turkey. During this bust phase
capital flows to developing countries halved again as in the decade be-
fore. In 2003 a new wave of increasing capital flows to developing coun-
tries started. In 2008 a new bust phase started. All these crises, and only a
few significant ones are mentioned here, led to disastrous twin crises and
in some cases nearly wiped out the entire domestic financial system.
Any boom-bust cycle is unique and depends on many specific histori-
cal factors. However, a typical pattern can be given. Boom-bust cycles
are expectation driven. The boom phase takes place on the basis of high

10
The poorest developing countries are largely excluded from such cycles as they
rely on donors and cannot attract substantial private capital inflows.

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 141

confidence of foreign and domestic economic agents in the economic de-

velopment in the developing country. FDI and portfolio inflows increase
as foreign investors expect high returns. Domestic banks and other finan-
cial institutions go abroad to get additional financial means for expanding
credit. In addition, at least bigger firms can go abroad directly to take
bank credits or issue debt-securities. In such a situation foreign creditors,
usually foreign banks, do not hesitate to give credit to developing coun-
tries as they also believe in the positive development of the debtor coun-
try. Investors can become too optimistic and in retrospect even irrational.
Indeed in many cases the wave of capital inflows started shortly after
countries liberalised their capital accounts. For some years capital import
countries show a good economic performance with high credit expan-
sion, high investment, high GDP growth and positive employment ef-
fects. Some of the foreign investors, especially those giving credits and
buying bonds, may speculate that some governments, the IMF or other
international institutions will bail out over-indebted countries. Other fac-
tors also come into play. Waves of capital inflows add to a typically al-
ready existing asset price bubble.
The wave of capital inflows creates a greater and greater economic
fragility. Firstly, net capital inflows lead to current account deficits. For
countries with fixed exchange rates increasing current account deficits
are easy to finance. Capital imports may also lead to an appreciation and
a cumulative increase of current account deficits. Secondly, foreign debt
in relation to GDP increases as part of the current account deficits is fi-
nanced by credits. It should be kept in mind that some of these capital in-
flows finance capital outflows and it is the gross debt in foreign currency
including credit dollarisation which is important for the real debt burden
in the case of a later depreciation. The higher share of net FDI inflow and
net portfolio equity inflow in the last boom cycle (cp. Figure 3) is an im-
provement of the structure of inflows for developing countries. It would,
however, be an illusion to assume that a future bust phase can avoid twin
crises. Gross foreign debt and credit dollarisation is still alarmingly high
in developing countries. Thirdly, asset prices increase to irrational levels
make a later destructive asset price deflation more and more likely. If a
central bank starts to increase interest rates to fight against inflationary
developments a dilemma is created: In such a situation the harder a cen-
tral bank increases interest rates or tries to reduce domestic credit expansion

3:32 , 04.11.2008, 07 Herr.doc

142 Hansjrg Herr

by administrative means, the higher the incentive of domestic agents to

circumvent restrictive monetary policy by taking credit abroad.

Figure 3: Total net capital inflows, net FDI flows and net portfolio equity
flows in developing countries 19702005 (in million U.S. dollars)*

350.000,00

300.000,00

250.000,00

200.000,00

150.000,00

100.000,00

50.000,00

0,00

50.000,00
1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003

Total net inflow

Net FDI inflow
Net portfolio equity inflows

Note: * Total private and official flows except IMF credits.

Source: Williamson (2005, 40).

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 143

The pitcher keeps going to the well until it is broken at last. Even small
changes in expectations can lead to portfolio shifts which are difficult to
manage for the country. There may be a political shock, as in Mexico in
1994, triggering a currency crisis; there may be the fear by foreign in-
vestment funds that asset markets and the external value of a currency
will collapse as in Thailand 1997; there may be contagion effects as in
Malaysia or South Korea in 1997; and there may be no good explanation
at all for the change in sentiments. In all currency crises foreign and do-
mestic wealth owners want to secure their wealth and prefer to keep it in
high-quality currencies. A change in expectations leads to a negative
conventional judgement which implies depreciation expectations and a
collapse of the currency premium. Economically costly twin crises which
can lead to long-term stagnation and a change in the long-run growth
path characterise the bust phase.11

4.2 Capital inflows and Dutch disease

In the 1960s the Netherlands discovered large offshore natural gas depo-
sits, resulting in negative effects for its manufacturing sector. The export
of gas led to a real appreciation of the Dutch guilder and a shrinking of
manufacturing. This economic effect was called Dutch disease. The
same argument applies to development aid which also can crowd out the
manufacturing export sector:

In sum then, the bad news is that even if delivered with the best inten-
tions and used carefully by responsible recipient governments, there

11
Williamson (2005, 15 ff.) calculated the costs of the Asian crisis in 1997. The
cumulative loss of one years GDP from 1997 to 2000 was 82% in Indonesia, 27%
in South Korea, 39% in Malaysia und 57% in Thailand. According to IMF estima-
tions costs of bank restructuring in % of GDP were 32.5% in Indonesia, 19.5% in
Korea, 19.3% in Malaysia und 25.0% in Thailand. In all four countries in 2003 un-
employment rates were still higher than in 1997. Also, the population living under
the national or international poverty line increased after the Asian crisis. The Asian
crisis in 1997 also shows that economic development is path dependent. South Ko-
rea, for example, was able to overcome the crisis relatively quickly. In Indonesia
the crisis was much deeper and longer and pushed the country on a medium-term
growth path which is lower than before the Asian crisis.

1:13 , 05.11.2008, 07 Herr.doc

144 Hansjrg Herr

are side effects like adverse impacts on competitiveness, which can

offset aids beneficial effect on growth. (Rajan/Subramanian 2005, 8).

The argument is that aid inflows may be spent on domestic goods

building schools or giving money to the poor who buy domestic goods ;
in this case the price level of non-tradable goods goes up; and the result-
ing real appreciation leads to a flow of resources from the export sector
to the non-tradable goods sector and a shrinking export sector. For exam-
ple, skilled workers move from the export sector to increase production
in the non-tradable goods sector. As the export sector is exposed to inter-
national competitiveness it is usually considered to be the sector with the
highest productivity increases. Thus, any reduction of the export sector is
harmful for the long term development of developing countries. Empiri-
cal studies of the aid related Dutch disease effect do not come to clear
conclusions. However, there is no doubt about the potentially high nega-
tive side effects of development aid (Adam/OConnell 2004, Adam 2006,
Rajan/Subramanian 2005, Hoffmann/Zattler 2006).
This argument can be applied to private international capital flows as
any type of net capital inflow can potentially lead to appreciation and
Dutch disease effects. In the following different typical cases of net capi-
tal inflows are analysed. In many analyses of Dutch disease Says law is
considered to hold. This is not accepted here. Effective demand becomes
a decisive factor to determine output and employment and also for Dutch
disease effects.
a) In the first case increases in net capital inflows are neutralised by
central bank interventions which create artificial compensating cap-
ital exports. Under the assumption of successful sterilisation, the no-
minal and real exchange rates do not change and there is no negative
demand effect and no Dutch disease. One positive aspect is that cen-
tral banks reserves increase.
b) In the second case increases in net capital flows lead to additional
imports of goods which cannot be produced domestically and at the
same time do not crowd out domestic demand and production. Let us
say that capital imports are used to import consumption goods which
were not available in the country before and are consumed additional-
ly. There is no change in the exchange rate, no change in domestic
demand and no change in domestic production. Or let us assume

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 145

capital imports are used to import capital goods which were not
available before and do not crowd out domestic investment. In this
case domestic capacities increase. However, without additional do-
mestic demand there will be no additional output. Imported capital
goods may lead to higher domestic productivity, which is positive,
but higher productivity does not necessarily increase domestic de-
mand and production automatically.
c) The third case is a special version of the second case. Let us assume a
situation of excess domestic demand and physical bottlenecks to in-
crease production. Under such a condition net capital inflows and a
resulting inflow of needed capital or intermediate goods increase do-
mestic production. This special case fits to the traditional World Bank
model (Chenery/Strout 1966) and to the belief of many other interna-
tional institutions and development ministries. It is based on the as-
sumption that developing countries have a lack of domestic savings
and of physical capital, and that foreign savings and a deficit in the
current account should augment domestic saving and increase the
domestic capital stock. China, which in 1978 definitely was not a
country with a good and abundant capital stock, offers no support for
this idea. It developed without current account deficits and imme-
diately followed an export-led development. Easterly (1999) tested
the saving gap model. From the 138 developing countries he tested
there was only one country (Tunisia) that supports the saving gap
model. In all other cases sometimes even very high net capital in-
flows were unable to trigger development.
d) In the fourth case net capital imports are used to finance domestic
demand. There are again two sub cases. In the first sub case firms,
public households, etc. switch from domestic finance to foreign
finance without increasing domestic demand. Here we have a clear
reduction in domestic demand as the capital import leads to an ap-
preciation and a reduction in exports and/or an increase in imports
exactly equal to the additional capital import. In the second sub case
capital imports are used to increase domestic demand. Here we have
an appreciation and thus a reduction of demand equal to the net capi-
tal import. However, this loss of demand is compensated by addition-
al domestic demand. The overall effect is zero.

3:32 , 04.11.2008, 07 Herr.doc

146 Hansjrg Herr

The conclusion of this part is that under very special conditions insuffi-
cient physical domestic capital stock to produce more and sufficient do-
mestic demand capital inflows can increase domestic production. In
many other cases net capital inflows do not lead to an increase of domes-
tic production or even to damaging Dutch disease effects.

5. Conclusion
World financial markets are more integrated than ever before. However,
the markets are divided into different currency segments. At the top of
the currency hierarchy are currencies like the U.S. dollar, the euro, the
Japanese yen or the British pound which dominate international financial
markets. At the bottom of the hierarchy are the currencies of developing
countries. In many cases these currencies only partly take over domestic
currency functions. They are usually used for domestic transaction pur-
poses but do not serve as a store of wealth to be transferred into the fu-
ture. Compared with top currencies these currencies earn a low currency
premium and are confronted with dollarisation as well as capital flight in
hard currencies. The relatively low quality of these currencies is one of
the most important hurdles for development probably much more than
the conventionally emphasised lack of physical capital, technology, skills
of workers, etc. It prevents a domestically financed Schumpeterian-
Keynesian credit-investment-income process as a large part of domestic
monetary wealth during such a process is exchanged in foreign currency,
triggers depreciation and leads to an extremely tight macroeconomic
monetary budget constraint which suppresses development.
Strategies to increase the quality of national currencies are of para-
mount importance for economic development. There are several condi-
tions which must be fulfilled to achieve this goal. Firstly, macroeconomic
monetary stability must be realised; especially inflation rates must be low
and depreciations limited. Secondly, a very successful policy to increase
the quality of domestic currencies is to prevent current account deficits
and high foreign debt. Increasing exports together with current account
surpluses do no only stimulate domestic demand and trigger export-led
growth, they also reduce the likelihood of twin crises and thus stabilise
expectations in the stability of the domestic financial system and the do-
mestic currency. The group of developed countries should accept current

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 147

account deficits to allow the group of developing countries to have cur-

rent account surpluses. Such a policy would improve the conditions for
development of developing countries and would be much more efficient
than development aid as many types of developing aid stimulate domes-
tic exports of developed countries and can lead to Dutch disease effects
in developing countries. Thirdly, measures to reduce dollarisation are
needed. Fourthly, regulations can support and protect a domestically fi-
nanced development process. Traditional capital export controls can be
used here as well as, for example, the restriction of domestic banks,
pension funds, etc. to invest abroad.
Substituting a domestic accumulation process in domestic money with
a foreign financed accumulation process is theoretically possible, albeit
only under certain conditions. The most important ones are the creation
of sufficient domestic demand in spite of the demand-reducing current
account deficits, long-term stable capital inflows and successful measures
against Dutch disease effects. In too many cases the attempt to create sus-
tainable development via net capital inflows and current account deficits
have failed and ended in twin crises and a perpetuation of underdevelop-
ment.
Against this background it is no surprise that there seems to be no em-
pirical relationship between capital account liberalisation and economic
performance. It seems clear that the quick and comprehensive liberalisa-
tion of capital accounts add fundamentally to instability and crises and
not to stable development (Stiglitz 2000, 10751086, and 2004, 5771,
Rodrik 1998 and 2001, Diaz-Alejandro 1985). This argument does not
imply that a country should stay completely isolated from international
capital markets. Certain types of FDI and other capital flows can support
economic development. A superior strategy seems to be a combination of
some selected FDI inflows and restrictions of other types of capital in-
flows with current account surpluses. Central bank interventions are an
important and legitimate strategy for developing countries to prevent cur-
rent account deficits. Such interventions and other controlled capital ex-
ports can compensate FDI inflows and stabilise current account surplus-
es.

3:32 , 04.11.2008, 07 Herr.doc

148 Hansjrg Herr

References

Adam, C. (2006): Exogenous inflows and real exchange rates. Theoretical quirk
or empirical reality, in: Isard, P., Lipschitz, L., Mourmouras, A., Yont-
cheva, B. (eds.): The macroeconomic management of foreign aid. Oppor-
tunities and pitfalls, Washington D.C.. International Monetary Fund.
171193.
Adam, C., OConnell, S. (2004): Aid versus trade revisited: Donor and reci-
pients policies in the presence of learning-by-doing, in: Economic Jour-
nal, 114, 150173.
BIS [Bank for International Settlement] (2006): Quarterly Review, September,
Basle.
Bossone, B., Sarr, A. (2002): A new financial system for poverty reduction and
growth, International Monetary Fund, Working Paper, No. 02/178.
Chenery, H.B., Strout, A.M. (1966): Foreign assistance and economic develop-
ment, in: American Economic Review, 56, 679733.
De Nicol, G., Honohan, P., Ize, A. (2003): Dollarization of the banking sys-
tem: Good or bad?, International Monetary Fund Working, Paper, No.
03/146.
Diaz-Alejandro, C.F. (1985): Goodbye financial repression, hello financial
crash, in: Journal of Development Economics, 19, 125.
Easterly, W. (1999): The ghost of the financial gap. Testing the growth model
used in the international financial institutions, in: Journal of Development
Economics, 60, 423438.
Eichengreen, B., Hausmann, R., Panizza, U. (2003): Currency mismatches, debt
intolerance and original sin. Why they are not the same and why it mat-
ters, National Bureau of Economic Research, Working Paper, No. 10036.
Heine, M., Herr, H. (2003): Volkswirtschaftslehre. Paradigmenorientierte Ein-
fhrung in die Mikro- und Makrokonomie, 3rd ed. (Mnchen: Olden-
bourg).
Herr, H. (2008): Capital controls and economic development in China, in: Ares-
tis, P., De Paule, L.F. (eds.): Financial liberalisation and economic per-
formance in emerging markets (Cheltenham: Edward Elgar), 142172.
Herr, H., Priewe, J. (2005): The macroeconomics of development and poverty
reduction. Strategies beyond the Washington Consensus (Baden-Baden:
Nomos).
Hoffmann, M., Zattler, J. (2006): Mitigating the risks of increased ODA flows,
in: Internationale Politik und Gesellschaft, 2/2006, 2742.
Keynes, J.M. (1921): A treatise on probability, Vol. 8 of the Collected Writings
of John Maynard Keynes (London: Macmillan).

3:32 , 04.11.2008, 07 Herr.doc

 Financial systems in developing countries and economic development 149

Keynes, J.M. (1936): The general theory of employment, interest and money,
Vol. 10 of the Collected Writings of John Maynard Keynes (London:
Macmillan).
Keynes, J.M. (1937a): The ex-ante theory of interest, in: The Economic Jour-
nal, 47, 663669.
Keynes, J.M. (1937b): Alternative theories of the rate of interest, in: The Eco-
nomic Journal, 47, 241256.
Keynes, J.M. (1937c): The general theory of employment, in: The Quarterly
Journal of Economics, 51, 209223.
King, R.G., Levine, R. (1993): Finance and growth: Schumpeter might be right,
in: Quarterly Journal of Economics, 108, 717 737.
Kornai, J. (1980): The economics of shortage (North-Holland: Amsterdam).
Kregel, J.A. (1985): Hamlet without the prince: Cambridge macroeconomics
without money, in: American Economic Review, 75, 133139.
Peoples Bank of China [PBoC] (2008), URL: www.pbc.gov.cn.
Lau, R.W.K. (2000): Privatisierung in China, in: PROKLA. Zeitschrift fr Kri-
tische Sozialwissenschaft, 30, 240265.
Lucas, R.E. Jr. (1988): On the mechanics of economic development, in: Journal
of Monetary Economics, 22, 342.
Naughton, B. (2007): The Chinese economy. Transition and growth (Cam-
bridge, MA: MIT Press).
Prasad, E., Wei S.J. (2005): The Chinese approach to capital inflows, Interna-
tional Monetary Fund, Research Department, Washington, D.C.
Rajan, R.G., Subramanian, A. (2005): What undermines aids impact on
growth. International Monetary Fund, Working Paper, No. 05/126.
Robinson, J. (1952): On generalisation of the General Theory, in: The Rate of
interest and other essays (London: Macmillan).
Rodrik, D. (1998): Who needs capital-account convertibility?, Essays in inter-
national finance, No. 207.
Rodrik, D. (2001): Development strategies for the next century, in: Pleskovic,
B., Stern, N. (eds.): Annual World Bank conference on development eco-
nomics 2000 (Washington D.C.): World Bank
Schumpeter, J. (1911): The theory of economic development, English transla-
tion (1934) (Cambridge, MA: Harvard University Press).
Stiglitz, J.E., Uy, M. (1996): Financial markets, public policy, and the East
Asian miracle, in: The World Bank Observer, 11, 249276.
Stiglitz, J.E. (2000): Capital market liberalization, economic growth and insta-
bility, in: World Development, 28, 10751086.

3:32 , 04.11.2008, 07 Herr.doc

150 Hansjrg Herr

Stiglitz, J.E. (2004): Capital-market liberalisation, globalisation, and the IMF,

in: Oxford Review of Economic Policy, 20, 5771.
Williamson, J. (2005): Curbing the boom-bust cycle: Stabilizing capital flows to
emerging markets. Institute for International Economics, Washington
D.C.
World Bank (2001): Finance for growth (Oxford: Oxford University Press).
World Bank (2007): World development report (Oxford: Oxford University
Press).
World Bank (2008): World Development Indicators, Washington D.C.

3:32 , 04.11.2008, 07 Herr.doc