This article was downloaded by: [Cornell University]

On: 28 June 2012, At: 22:02

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Review of Political Economy

Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/crpe20

Cycles and Growth: A Source of

Demand-Driven Endogenous Growth
a

Pierangelo Garegnani & Attilio Trezzini

Dipartimento di Economia, Universit degli Studi di Roma Tre,

Rome, Italy
Version of record first published: 16 Dec 2009

To cite this article: Pierangelo Garegnani & Attilio Trezzini (2010): Cycles and Growth: A Source of
Demand-Driven Endogenous Growth, Review of Political Economy, 22:1, 119-125
To link to this article: http://dx.doi.org/10.1080/09538250903392119

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-andconditions
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation
that the contents will be complete or accurate or up to date. The accuracy of any
instructions, formulae, and drug doses should be independently verified with primary
sources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.

Review of Political Economy,

Volume 22, Number 1, 119 125, January 2010

Cycles and Growth: A Source of

Demand-Driven Endogenous Growth
PIERANGELO GAREGNANI & ATTILIO TREZZINI

Downloaded by [Cornell University] at 22:02 28 June 2012

Dipartimento di Economia, Universita` degli Studi di Roma Tre, Rome, Italy

ABSTRACT This paper moves in a theoretical context in which the level of economic
activity is dependent on aggregate demand in both the long and the short period. It
shows that given two simple hypotheses, the economy will exhibit a tendency to grow
independently of any increase in the average level of ongoing investment (or any other
type of autonomous demand) over time. The two hypotheses are (a) that investment
oscillates over time and (b) that the communitys marginal propensity to consume is
lower when income contracts in slumps than when it increases in booms. This points to
a source of growth that is as endogenous to the system, as trade cycles are.

1. Introduction
The general theoretical background of our argument is one in which the level of
economic activity is understood to depend on aggregate demand both in the long
period, where productive capacity can change, and in the short period, where it
is a given. While we have discussed this background elsewhere (see Garegnani,
197879, 1992; Garegnani & Palumbo, 1998; Trezzini, 1995, 1998), it is a characteristic of the present paper that the question of the dependence of output on
long-term aggregate demand is not approached in the usual terms of a dichotomy
between autonomous and induced expenditure. Attention will instead be focused
on an endogenous mechanism of growth of aggregate demand, and hence of the
social product,1 based on the cyclical fluctuations of investment. This growth
would thus be as endogenous as are the cyclical fluctuations of investment.

Correspondence Address: Pierangelo Garegnani, Universita` degli Studi di Roma Tre, Centro Studi e
Documentazione Piero Sraffa, Dipartimento di Economia, Via Silvio DAmico 77, 00145 Rome,
Italy. Email: sraffa@uniroma3.it
1

The growth discussed here is in the first place the growth of aggregate demand. However,
in the above-mentioned theoretical framework of demand-driven growth, the normality of
a considerable degree of long-period elasticity of output makes it generally possible to
associate the suggested increase in aggregate demand with a corresponding increase in
social product. It may be recalled here how the argument for a long-period elasticity of
output with respect to aggregate demand rests importantly on the compound-rate nature
of the potential long-period growth of productive capacity (Garegnani, 1992).
ISSN 0953-8259 print/ISSN 1465-3982 online/10/010119 7 # 2010 Taylor & Francis
DOI: 10.1080/09538250903392119

Downloaded by [Cornell University] at 22:02 28 June 2012

120 P. Garegnani & A. Trezzini

Economists have long been aware of the existence of a cyclical asymmetry of
the marginal propensity to consume, understood as the comparative rigidity of
social consumption during a decrease rather than an increase in the social
product (see Samuelson, 1943; Duesenberry, 1948). Insufficient attention
appears to have been focused, however, on how this can generate growth endogenously once a long-period role of aggregate demand is admitted. As we shall see, a
marginal propensity to consume that is lower when the social product falls than
when it rises entails a progressive upward shift in the propensity to consume. Continuing cycle after cycle, this results in a tendency of the economy to grow even if
the average level of investment (or autonomous expenditure in general) remains
the same.
The above contrasts with the fact that cyclical fluctuations have generally
been considered irrelevant in determining the long-run trend of the social
product, not only in the obvious case of neoclassical theory, where the tendency
to full utilisation of resources would only allow oscillations around a trend determined by supply-side circumstances, but also whenever a long-period role of
aggregate demand is analysed under the assumption of steady growth. In our
earlier work, cited above, we noted that this assumption is essentially inconsistent
with an independent role of aggregate demand in the growth process. It also leads,
however, to a belief that short-period fluctuations in autonomous demand can only
cause the level of aggregate output to oscillate around a trend determined by
altogether different circumstances and thus to overlooking important phenomena,
such as the one presented here.

2. General Assumptions
The aim of the paper is to present the above mechanism engendering an endogenous demand-driven tendency of the economy to grow in its simplest form. Differences in the marginal propensities to consume in the different phases of the cycle
are therefore assumed without attempting any theoretical or empirical justification.2 The same circumscribed aim explains why technical progress will be
ignored and why constant returns to scale will be assumed with no scarce
natural resources and with a limitless supply of labour.
Investment, savings and the social product will be taken as gross. Since it will
be necessary to operate with value aggregates, the ground is cleared for this by
assuming a given real wage and the corresponding system of competitive
normal relative prices. Growth will accordingly leave relative prices unaffected.
As usual at this level of abstraction the effects of government and international
economic relations will be ignored.
It should be stressed that, while all the above assumptions will help to make
the argument clear and simple, the broad conclusions drawn from it are independent of them, except for the availability of labour.

One of us has addressed this question in a companion paper (Trezzini, 2005).

Cycles and Growth

121

Downloaded by [Cornell University] at 22:02 28 June 2012

3. An Endogenous Increase in Aggregate Demand with Constant

Ongoing Investment
Let us start by assuming that investment remains constant cycle after cycle in its
average level calculated over booms and slumps. While this is an unrealistic
assumption for what will then be shown to be a growing economy, it will help
to clarify that on the demand side this endogenous growth has nothing whatsoever
to do with any increases in the average level of investment. As already implied in a
demand-driven analysis of growth, however, absolutely no problem would arise if
we allowed for increasing investment; it would in general simply mean a further
increase in aggregate demand. The point of our assumption is that our focus here is
rather on the opposite side, namely on the fact that the increase in aggregate
demand discussed is independent of any increase in investment. If this is to be
shown in the clearest possible light, it must be in isolation from any changes in
investment; in other words, as it operates when the amount of ongoing investment
is constant.
To simplify, it will be assumed that the cycles are all of the same duration and
that the distribution of the constant amount of investment over the various phases
of each cycle is the same in all the cycles.3 In particular, the amount of investment
in the trough year of each cycle is assumed to be always the same, a constant
amount indicated by It. The same constancy is assumed for the amount of investment in the peak years, indicated by Ip.
In Figure 1, as in the usual Keynesian cross diagram with its 458 line, the
horizontal axis measures the social product in terms of the commodity chosen
as the numeraire. If the social product in period t is indicated as Yt, the vertical
axis will measure the corresponding aggregate demand in its component
elements, namely consumption, Ct, and investment, It. The previous assumption
about investment in trough years being the same cycle after cycle then suggests
that a straight line TT0 can be drawn parallel to the 458 bisector and at a vertical
distance of tt0 from the bisector equal to the trough investment It; i.e. tt0 It.
Consumption must be on that line in each trough year, at the level of income
corresponding to the trough in question. Similarly, a second line PP0 can be
drawn parallel to the 458 bisector and below TT0 . The vertical distance of this
PP0 line from the bisector, measured by the segment pp0 , is equal to the given
amount of peak investment Ip; i.e. pp0 Ip. Consumption must be on that line
in each peak year. The lines TT0 and PP0 will therefore provide the limits

It can be assumed, for example, that investment both rises and falls, from troughs to
peaks and vice versa, in accordance with a linear relationship. In this case, the
average level of investment over the cycle will be (Ip It)/2 regardless of the relative
or absolute lengths of the two phases. Any other assumption about the distribution of
investment over the cycle would, however, lead to the same conclusions for the
growth of the economy. As we shall see in the following section, the way in which
investment increases or decreases and hence the average level of investment is, in
fact, irrelevant as regards the expansive effect, which depends exclusively on the
boom and trough levels of investment.

Downloaded by [Cornell University] at 22:02 28 June 2012

122 P. Garegnani & A. Trezzini

Figure 1. The trajectory C0, C1, C2, C3, C4, . . . indicates the path of aggregate consumption when
investment fluctuates around a constant average level

within which aggregate consumption will be found to oscillate as income

increases cycle after cycle.
Let us now consider the path of aggregate consumption and the social product
more closely through, for example, two cycles, under the above hypotheses and
initially by means of an elementary numerical example. It is assumed that the marginal propensity to consume is mpcs 0.1 during slump years and at the higher
level of mpcb 0.5 in boom years.
Let us start from a year t 0 with a gross social product of Y0 100 in terms
of the commodity numeraire and then assume that, cycle after cycle, the investment of trough years, measured by the segment tt0 in Figure 1, is It 5 and
the investment of peak years, measured by pp0 is Ip 10. On the assumption
that t 0 is a trough year, investment will be I0 5 and consumption (on the
TT0 line for the point Y0 100) will be C0 95. A boom will then begin with
investment rising progressively to reach I1 10 pp0 in the peak year t 1.
Given the boom marginal propensity to consume mpcb 0.5, income will need
to have increased by the corresponding multiplier, which can be indicated as
the boom multiplier Mb 1=1  mpcb 1=0:5 2. The application of Mb to
the increase in investment DI 5 will therefore give an increase in income of
DY 10 and the income at t 1 will be Y1 110. With I1 10, consumption
will be at the corresponding point of the straight line PP0 giving C1 100.
A slump will follow, with investment falling gradually back to I2 It 5 in
the ensuing trough. Given the slump marginal propensity to consume mpcs 0.1,
application of the slump multiplier Ms 1=1  mpcs 1=0:9 1.11 to DI 5
will determine an income falling by DY 5.56 to Y2 104.44. Consumption,
C2, will accordingly be C2 Y2 It 104.44 5 99.44.
Comparison of the social products of the two subsequent troughs, Y0 100
and Y2 104.44, already shows the expansionary effect on the whole cycle.
The same effect is of course confirmed on proceeding to the next boom, where

Cycles and Growth

123

investment again increases by DI 5, which, with the boom multiplier Mb 2,

will give an increase in income of DY 10, thus pushing income to Y3
114.4. This peak level is higher than the Y1 110 of the previous peak also by
4.44. The same effect is finally confirmed with respect to the next trough when
its income of Y4 108.88 is compared with the Y2 104.44 and Y0 100 of
the previous two troughs with an increase of 4.44 over each cycle.

Downloaded by [Cornell University] at 22:02 28 June 2012

4. The Algebra
The segments C0C1, C1C2, C2C3, . . . thus show the irreversible path of consumption in the economy. In other words, they are segments of a consumption
function that changes at each transition from slump to boom within each cycle,
and then of course also changes from cycle to cycle. It is clear that these
increases in aggregate demand and social product are jointly engendered by
(i) the difference between the marginal propensity to consume in booms and
slumps, i.e. the steeper slope of C0C1 with respect to C1C2 or of C2C3
with respect to C3C4,4 and (ii) the oscillations of investment from It to Ip
and back.
On our assumptions, the overall increase in social product in the course of
the cycle, which can be labelled DYc, is in fact simply the difference between
the increase in output during the boomgiven by the boom multiplier Mb
1/(1 mpcb) applied to the difference between peaks and trough investment
(Ip It)and the decrease in output during the slumpgiven by the slump
multiplier Ms 1/(1 mpcs) applied to the same investment difference (which
in this case measures the change in investment during the slump):5
DYc Ip  It Mb  Ip  It Ms
or
DYc Ip  It Mb  Ms

(1)

This is the increase in social product over a cycle. The resulting yearly increase in

It should be noted that if the order of magnitude of the two marginal propensities to
consume were reversed, the oscillation of a constant amount of investment would have
a symmetrically depressive effect. This can be seen in Figure 1, where the arrows
would have to be reversed so that the steeper segments of the path would represent the
slumps and the flatter ones the booms. The economy would move over time in this case
from peak Y3 to peak Y1 and from trough Y4 to trough Y0.
5
We assume throughout this paper that the marginal propensities to consume remain constant within each different phases of the cycle, so that we have one multiplier for the
booms, Mb, and one for the slumps, Ms. Allowing the propensities to vary within each
phase of the cycle leads to no significant change in the relation defining the dimension
of growth. It would be sufficient to take the two multipliers as averages of the different
multipliers weighted by the fraction of the total change in investment to which they
apply, together of course with the possibility of identifying two successive phases in
which these averages give mpcs , mpcb.

124 P. Garegnani & A. Trezzini

aggregate demand and the social product will then of course depend also on the
duration of each cycle.6

Downloaded by [Cornell University] at 22:02 28 June 2012

5. A Further Result
Expression (1) has an interesting implication: the tendency of the economy to
grow is independent not only of any increase in the amount of ongoing investment,
as shown by our assumption of the constancy of that investment, but also to a large
extent of the size of that constant investment. Given the marginal propensities to
consume and therefore the two multipliers, the growth of aggregate demand
depends exclusively on the difference between peak and trough investment. The
same expansionary effect can therefore be obtained by any average amount of
investment included between a minimum determined by the trough investment
It, which can be approached by lengthening the slumps to occupy almost the
entire duration of the cycle and keeping investment in it close to its trough
level, and a maximum determined by the peak level Ip, which can be approached
in a symmetrical way.

6. Conclusions
What is outlined above is essentially a process whereby, due to the cyclical asymmetry of propensities to consume, the oscillations of investment in the course of the
trade cycle trigger successive upward shifts of the economys propensity to
consume from C0C1 to C2C3 or from C1C2 to C3C4 and so on, thus tending to
raise the social product along with them. The idea is simple and the assumptions
about consumption behaviour during the cycle are hardly controversial. As
indicated at the beginning, however, the emergence of this effect in the theory
requires two conditions that are rarely fulfilled together in the literature. The
first is a rejection of the notion of a long-period tendency of the competitive
economy towards the full utilisation of productive resources. The second condition
instead is that the analysis should not be conducted in terms of steady growth, but
rather, as has been done here, in terms of the traditional normal positions of the
economy which are compatible with any pattern in the evolution of outputs.
What is presented here is not, however, a model of growth. Its purpose is
simply to draw attention to an element capable of playing a role, perhaps an important role, in the tendency of the aggregate demand in a market economy to grow.
We doubt that formalised overall explanations of the growth of an economy are
likely to prove fruitful. Once the view that growth can be explained in terms of
autonomous changes in factor endowments, technical knowledge or tastes is abandoned, the central importance of historical and institutional circumstances in
the growth process clearly emerges, as it did within the analyses of the classical
economists from Adam Smith to Marx. And such circumstances are too
6

A total increase of 4% over a cycle of 4 years corresponds to an average yearly rate of

almost 1%. A lower average yearly rate of just under 0.5% would apply if the same
total increase took place over a cycle of 8 years.

Cycles and Growth

125

complex and variable from country to country, and period to period in the same
country, to permit any useful deductive quantitative treatment of the process as
a whole, as distinct from treatment of particular aspects of it, like the one
considered here.

Downloaded by [Cornell University] at 22:02 28 June 2012

References
Duesenberry, J.S. (1948) Income-consumption relations and their implications, in: L.A. Metzler
(Ed.) Income, Employment and Public Policy: Essays in Honor of Alvin H. Hansen
(New York: W.W. Norton).
Garegnani, P. (197879) Notes on consumption, investment and effective demand, parts I & II,
Cambridge Journal of Economics, 2 & 3, pp. 335353, 6382.
Garegnani, P. (1992) Some notes for an analysis of accumulation, in: J. Halevi, D. Laibman & E.J.
Nell (Eds), Beyond the Steady-State (Basingstoke & London: Macmillan).
Garegnani, P. & Palumbo, A. (1998) Accumulation of capital, in: H.D. Kurz & N. Salvadori (Eds),
The Elgar Companion to Classical Economics (Aldershot: Edward Elgar).
Samuelson, P.A. (1943) Full employment after the war, in: S.E. Harris (Ed.) Postwar Economic
Problems (New York: McGraw-Hill).
Trezzini, A. (1995) Capacity utilisation in the long run and the autonomous components of aggregate
demand, Contributions to Political Economy, 14, pp. 3366.
Trezzini, A. (1998) Capacity utilisation in the long run: some further considerations, Contributions
to Political Economy, 17, pp. 5367.
Trezzini, A. (2005) The irreversibility of consumption and economic growth, Quaderno di Ricerca
n. 6, Roma, Centro Ricerche e Documentazione Piero Sraffa.