Download Business Confidence and Depression Prevention: A Mesoeconomic

American Economic Association
Business Confidence and Depression Prevention: A Mesoeconomic Perspective
Author(s): Yew-Kwang Ng
Source: The American Economic Review, Vol. 82, No. 2, Papers and Proceedings of the
Hundred and Fourth Annual Meeting of the American Economic Association (May, 1992), pp.
365-371
Published by: American Economic Association
Stable URL: http://www.jstor.org/stable/2117429 .
Accessed: 30/01/2015 02:56
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
.
American Economic Association is collaborating with JSTOR to digitize, preserve and extend access to The
American Economic Review.
http://www.jstor.org
This content downloaded from 155.69.24.171 on Fri, 30 Jan 2015 02:56:00 AM
All use subject to JSTOR Terms and Conditions
(A MICRO-MACROANALYSISWITH
MESOECONOMICS
AND ITSAPPLICATIONS)t
NONPERFECTCOMPETITION
Business Confidence and Depression Prevention:
A Mesoeconomic Perspective
By YEW-KWANGNG*
Can a collapse in business confidence
(such as might be triggered by a sharp fall in
share prices) lead to a deep depression? If
so, how may it be prevented? Without attempting to address these important issues
in all relevant aspects, this paper provides
some insights, using a micro-macroeconomic analysis.
A share-market crash may trigger a depression mainly by reducing aggregate demand (through wealth effects and possibly
money contraction effects as bankers become more cautious in lending money) or
by causing a collapse in business confidence.
In a simple model of perfect competition
with the resulting classical dichotomy between the real sector and the financial sector, a reduction in (nominal) aggregate demand should only reduce the price level
without affecting real variables. Retaining
all simple features (comparative-static analysis, no time lags, misinformation, or any
other rigidities), just the relaxation of perfect competition is sufficient to break the
classical dichotomy (Section I), and make
self-fulfilling collapse in confidence possible
(Section II).
Elsewhere (Ng, 1977, 1980, 1982a, 1986) I
developed a micro-macroeconomic analysis
of a representative firm that is not necessarily perfectly competitive to analyze the effects of economy-wide changes on aggregate
output and the price level (dubbed "mesoeconomics" for brevity). It takes account of
the profit-maximization calculation at the
firm level as well as interfirm interactions
and repercussions through aggregate variables, including the effects of aggregate demand, aggregate output, and the price level
on the firm's demand and cost curves.
Though hailed by Robin Marris (1991) as
the modern pioneer of imperfect competition microfoundation of macroeconomics,
the analysis has not received widespread
attention. In this paper, the method is used
to analyze the effects of a change in business confidence in the form of a fall (or
increase) in expected real aggregate demand. Due to space limitation, mathematical derivations of results are not presented
here but are available in Ng (1992).
As shown below, a fall in expected real
aggregate demand will lead to a fall in aggregate output (and hence employment) by
the full extent whether nominal aggregate
demand falls with real aggregate demand or
is maintained or even increased by monetary and fiscal policies, if costs (mainly
wages) do not fall (as output falls) by a
sufficient amount to more than offset any
decrease in demand elasticity as may occur
when the number of firms decreases. While
the maintenance of nominal aggregate demand may also be important if complications such as time lags are introduced, cost
tDiscussant: Ian MacDonald, University of Melbourne, Australia.
*Professor of Economics, Monash University, Clayton, Melbourne, Victoria 3168, Australia. The final
draft was done while I was visiting the Faculty of
Business Administration, National University of Singapore.
365
This content downloaded from 155.69.24.171 on Fri, 30 Jan 2015 02:56:00 AM
All use subject to JSTOR Terms and Conditions
366
AEA PAPERS AND PROCEEDINGS
adjustment is of paramount importance in
the comparative-static analysis.
I. The Method of Mesoeconomic Analysis
Pricing and output decisions are made by
firms subject to demand and cost conditions. Concentrating on the decisions of the
firm is an adequate analysis provided factors affecting its demand and cost functions
are appropriately taken into account. The
concentration on a single representative firm
makes the analysis inappropriate for analyzing interfirm changes. However, for economy-wide changes (or for industry-wide
changes), it has been shown in a fully general-equilibrium analysis that (i) a representative firm exists that exactly represents the
response of the economy in average price
(i.e., the price level) and aggregate output
to any given economy-wide change in demand or cost, and (ii) a representative firm
defined by a simple method (weighted average) can be used as a good approximation
for all changes not involving big changes in
relative prices (Ng, 1986 appendix 31).
Essentially, the firm maximizes profit
given that (i) aggregate demand and the
average price of all other firms enter its
demand function and (ii) aggregate output
(related to employment) and average price
enter its cost function. This allows analysis
of the interaction between the individual
(typically imperfectly competitive) firm and
the rest of the economy without taking in
the full complication of a fully disaggregated general-equilibrium analysis. It can be
shown that the method is consistent with
consumer utility maximization and the introduction of an explicit labor-supply function. Since firms are the decision-makers
deciding on price and output levels, any
change must work through the demand and
cost functions of firms. The general functional forms for these demand and cost
functions are consistent with virtually all
reasonable specifications of a general-equilibrium model except that distributional
changes are ignored, as in all aggregate
analyses. The aggregate demand function is
also general, with various specific functions
as special cases.
MAY 1992
A. Breaking the Classical Dichotomy
A most simple model of a perfectly competitive economy consists of the following
four equations:
Y = F(L)
FL = W/P
W/P= ql(L)
kY= M/P.
The first three equations specify that real
output Y is a function of the only variable
input L whose marginal product FL must
equal its real wage rate W/P, which must
lie on its (inverse) supply function if; and
the last equation is the simple classical case
that demand for money equals supply. The
first three equations completely determine
the three real variables of the whole model:
Y, L, and W/P, leaving the last equation to
determine only the nominal variable P, the
price level. This classical dichotomy is broken simply by the introduction of nonperfect competition which necessitates replacing the second equation by AFL = W where
,u marginal revenue. As shown in Ng
(1980, 1982b), a change in (nominal) aggregate demand may change ,u at given real
output level by changing the elasticity of
demand for the product of the representative firm at given output level. A change in
money supply may change the output level
to a new equilibrium level. (For other mechanisms in which money may affect real output, see Olivier Jean Blanchard [1990]. In
particular, the significance of imperfect
competition has received close attention [see
e.g., Takashi Negishi, 1979; Oliver D. Hart,
1982; Dennis J. Snower, 1983; Robert M.
Solow, 1986; Blanchard and Nobuhiro Kiyotaki, 1987; Huw Dixon, 1987; Ian McDonald, 1987]. The belief that the result of
possible nonneutrality of money is inconsistent with some recent results [e.g., Blanchard and Kiyotaki, 1987; Jean-Pascal Benassy, 1987; Dixon, 1987] on the neutrality
of money in an imperfectly competitive
economy is based on the assumption of a
This content downloaded from 155.69.24.171 on Fri, 30 Jan 2015 02:56:00 AM
All use subject to JSTOR Terms and Conditions
VOL. 82 NO. 2
MESOECONOMICS (A MICRO-MACROANALYSIS)
367
unique equilibrium. In my model, a continuum of real equilibria is possible, whereby a
change in money supply or business confidence may trigger a shift from one equilibrium to another.)
B. Monetarist and KeynesianResults as
Special Cases
I have shown that an increase in nominal
aggregate demand (caused by an increase in
money supply or other factors) may leave
output unchanged and increase the price
level proportionately (the monetarist result)
or may increase output in the same direction without increasing the price level (the
Keynesian result). How demand elasticity of
the representative firm changes with respect
to real aggregate demand and how costs
respond to output are important determining factors.
C. Explaining Large Effects on Prices of
Cost Changes
An exogenous increase in prices of certain inputs (e.g., oil) increases the costs of
the representative firm only slightly. However, this also increases the average price
which feeds again into costs. The full impact on the price level could be magnified
many times.
II. ModelingBusinessConfidence
Changes in business confidence is modeled by changes in the expected value of
real aggregate demand. If an expected xpercent fall in real aggregate demand makes
it profit-maximizing for firms to reduce output by x percent or more, the collapse in
confidence may be self-fulfilling or even cumulative. This may be the case even if nominal aggregate demand is maintained intact
or increased. Moreover, the depressed output level could be an equilibrium position.
A. Self-Fulfilling Collapse in Confidence
Consider Figure 1: a decrease in real
aggregate demand (as will result if nominal
demand decreases with the price level un-
d~~~~~
0~~~~~~~~
0
~~~~~~~~~~~~
FIGURE 1. EFFECTS OF THE PRICE LEVEL
AND AGGREGATE DEMAND ON THE
DEMAND CURVE
changed) shifts the demand curve for the
productof a representativefirm from d to
d' in the absence of an elasticity change
(which may go either way). From the position of d', consideran increasein the price
level with real aggregate demand unchanged (nominal aggregate demand and
the price level increaseby the same proportion). This should shift the demand curve
verticallyupwardsince if this firm also increases its price by the same proportion,
quantity demanded should remain unchanged, as only nominal variables have
changed (homogeneity of degree zero).
Combiningthe above two changes, the demand curvewill be twisted from d to d" if
only the price level increaseswith nomiiial
aggregate demand unchanged (real aggregate demanddecreases).
If a collapse in business confidenceleads
to an expected decrease in real aggregate
demand while the nominal aggregate demand is unchanged, the demand curve of
the representativefirm is expected to twist
from d to d". As shown in Figure 2, this
also twists its marginal-revenuecurve to
MR".If marginalcost, MC, is not responsive to outputbut is proportionatelyresponsive to the price level, the new profit-maximizingpoint involves a higher price and a
lower outputby exactlythe same proportion
as the (originallyexpected) changes in the
price level and real aggregatedemand,mak-
This content downloaded from 155.69.24.171 on Fri, 30 Jan 2015 02:56:00 AM
All use subject to JSTOR Terms and Conditions
368
AEA PAPERS AND PROCEEDINGS
money wage rates did not respond fully to
the fall in the price level. It can be shown
mathematically that this tends to offset the
negative response in wage rates to unemployment, making self-fulfilling collapses still
possible. It might be thought that the case
of no negative response in wage rates to
unemployment illustrated in Figure 2 is not
possible in practice. However, it has been
shown (e.g., McDonald and Solow, 1981;
Ng, 1986 Ch. 12) that unchanged wage rates
as unemployment changes may be quite
consistent with unions' utility maximization.
p
p
Mc"
d
\
MR"
q"
FIGURE
TO
2. A
\R
MAY 1992
d#
B. Consideringthe Entry /Exit of Firms
q
TWIST OF THE DEMAND
CURVE
FROM
d
d" TWISTS
CURVE TO
THE MARGINAL-REVENUE
MARGINAL COST IS RESPONSIVE TO PRICE
LEVEL BUT NOT TO OUTPUT, MC SHIFTS TO MC"
MR"; IF
ing the collapse in business confidence selffulfilling.
The result illustrated in Figure 2 shows
that, under certain conditions (e.g., MC is
not responsive to output but is proportionately responsive to the price level; or
changes in MC are offset by changes in MR
due to a change in the demand elasticity), if
real aggregate demand is expected to decrease, the maintenance of nominal aggregate demand may not be sufficient to avoid
a decrease in real aggregate demand and
output. If real aggregate demand is firmly
expected to decrease, the maintenance of
nominal aggregate demand may just lead to
an expected (and eventually an actual) increase in the price level. Then, on top of a
decrease in real output, we have an increase
in the price level, as illustrated in Figure 2.
To prevent the depression illustrated in
Figure 2, costs have to be decreased
(through decreasing real wage rates) as output decreases (unemployment increases). In
the post-1929 depression, some negative responses of marginal costs (wages) to increases in unemployment probably applied.
However, nominal aggregate demand and
the price level fell in that depression, and
A depression is not just characterized by
output and employment curtailment by
firms, but also by the closing down of some
firms. This free entry/exit of firms is modeled by imposing the customary zero-profit
condition. Equilibrium then requires the
representative firm to operate at a point
involving not only MR = MC but also total
revenue = total cost. It may be thought that
the zero-profit condition should only apply
to the marginal firms, not the representative
firm, which should earn some positive profits. However, the ability of inframarginal
firms to earn positive profits must be due to
the possession of some superior factors (e.g.,
resources, position, good management, etc.).
Assuming competition for these factors, the
supernormal profits should be transformed
into higher prices for these superior inputs.
Thus, the zero-profit requirement is reasonable for modeling the effects of free
entry/exit, not to mention the issue of contestability.
In the long run, it is more likely that a
reduction in aggregate output will lower the
cost curves (through reduced input prices).
In this respect, a depressed equilibrium is
less likely than in the short run. However,
the exit effect may work to offset this. As a
depression sets in, some firms exit, decreasing the number of firms and hence lowering
the degree of competition, leading to a less
elastic demand curve. A collapse in business
confidence (real aggregate demand expected to decrease) may still be self-fulfilling. Figure 3 shows a case in which the
This content downloaded from 155.69.24.171 on Fri, 30 Jan 2015 02:56:00 AM
All use subject to JSTOR Terms and Conditions
VOL. 82 NO. 2
MESOECONOMICS (A MICRO-MACROANALYSIS)
369
AC AC
p
A
d
AC'
_____
p
~~~~~~AC
d
MCI
Mc
J,M
Mc
q'
q
FIGURE 3. A CASE IN WHICH A SELF-FULFILLING
COLLAPSE IN BUSINESS CONFIDENCE IS
ASSOCIATED WITH A DECREASE IN
NOMINAL AGGREGATE DEMAND
collapse is associated with a decrease in
nominal aggregate demand and Figure 4
shows a case in which the nominal aggregate demand is maintained unchanged. In
both figures, I allow for the full response of
costs to prices (not significant for Fig. 3 in
which prices remain unchanged).
An expected decrease in real aggregate
demand shifts the demand curve from d to
d'. The absolute demand elasticity decreases at given p due to the decreased
competition effect. Cost curves shift downward in Figure 3 due to the depressing
effects of a reduction in aggregate output
on input prices. Despite this, the new equilibrium point involves a lower output at an
unchanged price level, making the original
decrease in expected real aggregate demand
self-fulfilling.
As aggregate output falls, firms exit, and
the representative firm moves from A to B,
costs actually fall. Why do firms not reenter
to take advantage of the lower costs? The
answer is that firms are not making positive
profits; an individual firm considering entry
would see itself making losses as it forces
the demand curve for each firm in that
industry to move leftward. While if firms in
all industries reenter and expand, the econ-
MR
qI
q
FIGURE 4. A CASE IN WHICH A SELF-FULFILLING
COLLAPSE IN BUSINESS CONFIDENCE OCCURS
WITH No CHANGE IN NOMINAL
AGGREGATE DEMAND
omy can move back to the original equilibrium A, each firm sees it as unprofitable to
do so. In such a situation, an interfirm
macroeconomic externality exists, as analyzed in Ng (1986 Ch. 3).
In Figure 4, the price level is expected to
increase as real aggregate demand is expected to fall and nominal aggregate demand is expected to remain unchanged. This
shifts the demand curve d' vertically upward in comparison to the d' in Figure 3.
Cost curves move upward but proportionately less than the upward increase in price,
since a depressing effect through aggregate
output is allowed. In both figures, the depressed position remains an equilibrium
point, which confirms the original collapse
in business confidence.
The above analysis shows that, when the
exit effect is taken into account, the maintenance of nominal aggregate demand plus
the lowering of wages as aggregate output
falls, and the full response of wage rates to
a lower price level (which in combination is
sufficient to prevent a depression in the
short-run model where the number of firms
is given) may not be sufficient to prevent a
depression in the face of a collapse in business expectations in the presence of the exit
This content downloaded from 155.69.24.171 on Fri, 30 Jan 2015 02:56:00 AM
All use subject to JSTOR Terms and Conditions
370
AEA PAPERS AND PROCEEDINGS
effect. It may be necessary to lower wage
rates (as unemployment increases) by a
marginthat is more than sufficientto offset
the exit effect.
Both for the short-runand the long-run
cases, the required reduction in costs
(mainly throughwage rates) is only transitional. If the required reduction has been
achieved,the economyshould move back to
the high equilibriumoutput q in Figure 3.
At this point, costs (and wage rates) are
back to the originallevels (correspondingto
J). In contrast,if unions refuse to have the
wage rates loweredby the requiredamount,
the economy may be stuck at the low equilibrium q' with lower real wage rates (corresponding to J' instead of J) for a long
time, untilthe prolongeddepressionchanges
union militancy.
It may be thought that the exit effect is
unlikelyto occur since firmsare unlikelyto
go bankruptjust on the weight of a collapse
in businessconfidencebefore real aggregate
demand actually decreases substantially.
However, in a growingand changingeconomy, entry and exit take place as a matter
of routine. A collapse in confidence will
almostcertainlyspeed up exits and delay or
stifle planned entries. Thus, the numberof
firms may be substantiallycut as soon as
confidencecollapses. The entry/exit effect,
though classified as a long-run effect for
analytical purposes, may take place very
quickly,especially if the expected decrease
in real aggregatedemand is believed to be
prolonged.
My analysis showing the possibility of
self-fulfillingcollapse in businessconfidence
and the insufficiencyof the maintenanceof
(nominal)aggregatedemandin preventinga
depression under such conditions does not
mean that the real economyis actuallycharacterizedby these conditions.However,due
to the importance of preventing a great
depression, there should be adequate understanding of the mechanisms involved,
even if these apply to rare specific cases
only. The question of whether the required
conditions apply in practice is beyond the
scope of this paper. Nevertheless,the persistence of a high unemploymentrate in the
United Kingdom over many years in the
MAY 1992
1980's may partly be due to similar conditions makinga depressed situationpossible
as a long-runequilibrium.
REFERENCES
Benassy, Jean-Pascal, "Imperfect Competition, Unemployment and Policy," European Economic Review, February/March
1987, 31, 417-26.
Blanchard,Olivier Jean, "Why Does Money
Affect Output? A Survey," in Benjamin
M. Friedman and Frank H. Hahn, eds.,
Handbook of MonetaryEconomics, Vol. 2,
Amsterdam: North-Holland; 1990, pp.
779-835.
and Kiyotaki,Nobuhiro,"Monopolistic Competition and the Effects of Aggregate Demand," American Economic Review, September 1987, 77, 647-66.
Dixon, Huw, "A Simple Model of Imperfect
Competition with Walrasian Features,"
Oxford Economic Papers, March 1987, 39,
134-60.
Hart,OliverD., "A Model of Imperfect Competition with Keynesian Features," QuarterlyJournal of Economics, February 1982,
97, 109-38.
McDonald, Ian, "Customer Markets, Trade
Unions and Stagflation," Economica, May
1987, 54, 139-54.
and Solow, Robert M., "Wage Bargaining and Employment," American
Economic Review, December 1981, 71,
896-908.
Marris,Robin, ReconstructingKeynesian Economics with Imperfect Competition, London: Edward Elgar, 1991.
Negishi,Takashi, Microeconomic Foundations
of Keynesian Macroeconomics, Amsterdam: North-Holland, 1979.
Ng, Yew-Kwang,"Aggregate Demand, Business Expectation, and Economic Recovery without Aggravating Inflation," Australian Economic Papers, June 1977, 16,
130-40.
, "Macroeconomics with Non-perfect
Competition," Economic Journal, September 1980, 90, 598-610.
, (1982a) "A Micro-Macroeconomic
Analysis Based on a Representative
Firm," Economica, May 1982, 49, 121-39.
This content downloaded from 155.69.24.171 on Fri, 30 Jan 2015 02:56:00 AM
All use subject to JSTOR Terms and Conditions
VOL. 82 NO. 2
MESOECONOMICS (A MICRO-MACROANALYSIS)
, (1982b) "Macroeconomics with
Non-perfect Competition: A Reply,"
Economic Journal, September 1982, 92,
706-7.
, Mesoeconomics: A Micro-Macro
Analysis, Brighton, U.K.: Wheatsheaf,
1986.
, "Business Confidence and Depression Prevention: A Micro-Macroeconomic Analysis," Mathematical Social Science, 1992 (forthcoming).
371
Snower, Dennis J., "Imperfect Competition,
Underemployment and Crowding Out,"
Oxford Economic Papers, Supplement,
November 1983, 35, 245-70.
Solow,RobertM., "Monopolistic Competition
and the Multiplier," in Walter P. Heller,
Ross M. Starr, and David A. Starrett,
eds., Equilibrium Analysis: Essays in
Honor of Kenneth J. Arrow, Vol. 2, Cambridge: Cambridge University Press, 1986,
pp. 301-15.
This content downloaded from 155.69.24.171 on Fri, 30 Jan 2015 02:56:00 AM
All use subject to JSTOR Terms and Conditions