Public Disclosure Authorized

Public Disclosure Authorized

Public Disclosure Authorized
Public Disclosure Authorized

Policy,Planning,and ReseLrch

WORKING PAPERS
Macroeconomic
Adjustment
andGrowth
CountryEconomicsDepartment
The WorldBank
December1989
WPS339

PrivateInvestment
and Macroeconomic
Adjustment
An Overview
Luis Serven
and
AndresSolimano

This paper reviews current investment theories, recent models

linking macroeconomicpolicies and private investment, and the
effect of uncertainty and credibility on irreversible investment
decisions. Empirical studies on the subject are also reviewed,
and the general implications of this literature for the design of
growth-oriented adjustment programs are discussed.
The Policy, Planning, and Research Complex distiibuies PPR Working Papers to disseminate the findings of work in progress and to
encourage the exchange of ideas among Bank staff and all others interested in development issues. These papers carry the names of
the authors, refnectinly their views, and should be used and cited accordingly Thefindings, interprtations, and conclusions are the
authors'own. They shotld not be attnbuted tothe World Bank, itsBoard of Directors, itsmanagement, orany of its mcmber countries.

Plc,Planning,and Research

Maroeconomic
Adjustment
andGrowth

Serven and Solimano review the literature on the

macroeconomic determinants of investment,
paying particular attention to the transmission
mechanisms and likely effects of different macro
policies on private investment.

credit also influenec how macroeconomic

policies affect investment.

What are the links between adjustment,

investment, and growth? Correcting riacro
imbalances and achieving macroeconomic
stability are prerequisites for achieving sustained
growth. A strong response from private investors to incentives introduced as part of an
adjustment program is crucial if the stabilization
effort is to be followed by sustained growth.

* To improve the design of macroeconomic

stabilization policies consistent with the res 7il)tion of growth, more research is needed on thc.
implications for private investment of (a) fiscal
adjustment (especially cuts in public investmenl)
(b) changes in the exchange rate, and (c) monctary restraint under altemative rin,-.ncialmarkec
arrangements.

Through what transmission mechanisms do

macroeconomic policies affect private investment? Serven and Solimano discuss how
different macro policies affect the variablesthe real interest rat<, the market price of installed
capital, and the price of new capital goodsthat influenec the profitability of capital. Demand conditions and the availability of real

* A high research priority should be theidCvelopment of models suitable for the empirical
study of irreversible investment under uncertainty, a relevant issue for understanding the
lack of investment response to incentives undec
unstable macroeconomic conditions. More work
is also needed on the investment consequences
of policy credibility, and the policy implicatinns
of the link bctwecn credibility and sustainabilily.

Serven and Solimano recommend more

research in two areas:

This paper is a product of the Macroeconomic Adjusunent and Growth Division,

Country Economics Depanment. Copies are available free from the World Bank,
1818 H Street NW, Washington DC 20433. Please contact Emily Khine, room NI I055, extension 61763 (42 pagcs with charts).

The PPR Working Paper Series disseminates the findings of work under way in thc Bank's Policy, Planning, and Rcsea;rch
Complex. An objectivc of the series is to get thesc fmdings out quickly, even if prcsentations are Iess than fully polishie(l.
The findings, interpretations, and conclusions in these papers do not necessarily reprcsent official policy of the Bank.
Produced at the PPR DisseminationCenter

Table of Contents

.. . . . . . . . . . . . . . . . . . . . . . . . .

1.

Introduction.

2.

Investment Theory: A Brief Review

3.

Macroeconomic Policies and Private Investment . . . . . . . . . . .

Monetary and fiscal policy and private investment . . . . . .

Exchange rate policy and private investment . . . . . . . . .

10
13

3.1.
3.2.
4.

The

incentive structure and investment response:

..
uncertaintyand irreversibility

5.

..

4s1

Irreversibility, uncertainty, and investment

4.2

.
The role of credibility

4.3

Empirical applications

Concluding Remarks..

REFERENCES

credibility,

.21
. . . . . . . .

22

27

. . . . . . . . . . . . . . . . . . . . 30

...............

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

* We thank Bela Balassa, Bill Branson, Ricardo Caballero, Vittorio Corbo

and Robert Pindyck for helpful comments and discussions.

36
38

1.

Introduction
The correctionof

external imbalancesin many developingcountries

during the eightieshas taken the form of major cuts in investmentrates rather
than increasesin domesticsavings.This investmentdecline,which mirrors the
decline in the externalresourcetransfersince 1982, has been especiallysharp
in the highly indebted countries,and has been accompaniedby a slowdown in
growth in all LDC's.

Both public and private investmentratas have fallen,

although the latter more drastically than the former. If this trend is
maintained, it will lead to a slowdown in medium term growth possibilitiesin
these economiesand will reduce the levels of long run per capita consumption
and income, endangeringthe sustainabilityof the adjustmenteffort.
The observedreductionin investmentin LDC 's seems to be the result
of several factors. First, the lower availabilityof foveign savings has not
been matched by a correspondingincrease in domestic savings. Second, the
deteriorationof fiscal conditionsdue to the cut in foreign lending.to the
rise in domestic interestrates, and to the accelerationin inflationforced a
contraction

ir

public

investment. Third,

the

increase

in

macroeconomic

instabilityassociatedwith the externalshocksand the difficultiesof domestic

governmentsto stabilizethe economy has hampered private investment.Fourth,
the debt overhanghas also discouragedinvestment,through its implied tax on
future output and the ensuing credit constraints in internationalcapital
markets.
In this paper we review current investmenttheories, recent models
of investment,
behaviorand empiricalstudieson the subject in order to examine
the linkages between macroeconomicadjustment and private investment. The

-2review serves two purposes: on the one hand, to get a furtherunderstandingof

the behaviorof investmentin LDC's during the eighties. On the other hand, we
seek to identify researchareas relevant for the design of policies that can
bring about adjustmentwith growth.
The paper is organized as follows. First we review in Section 2
different theories of investment, starting from Keynes and covering the
Accelerator, Neoclassical, Tobin's

Q,

Disequilibrium, Two-Gap

and

Irreversibilitytheoriesof investment. In Section3, we discussthe literature

on macroeconomicpolicies and private investment, examining the effect of
monetary, fiscal and exchange rate policy on private investment, paying
attentionto some economicor institutional
featuresspecificto LDCs (e.g.,the
degree of interventionin financial markets, the possible complementarities
between public and private investment,or the high relianceon importedcapital
goods) that may affect the transmissionmechanismsthroughwhich some standard
macropolicymeasures influenceinvestment.In the fourth sectionwe review the
recent literatureon credibility,uncertaintyand irreversibility
in investment
decisions,which is very useful in order to understandthe responseof private
investment to the change in economic iricentivesthat comes along with an
adjustmentprogram.Because investmentis at least partiallyirreversible,and
because it is guided by the uncertain future profitabilityof capital, it is
also extremely sensitive to cconomic and/or political instability.We discuss
how such factorscontributeto determinethe investmentresponseto a given set
of economic incentives,which is a key mechanism for stabilizationto be
followedby a resumptionof growth. Finally,Section5 presentssome concluding
remarks.

2.

InvestmentTheory: A Brief Review

Keynes was perhaps the first economist to call attention to the

existenceof an independentinvestmentfunctionin the economy in departurefrom

the prevailingnotion (i.e., the Wicksellian loan market) that all available
saving is automaticallyinvestedprovided an appropriateinterest rate exists
in the economy. Keynes' (1936)basic insightwas that investmentdependson the
prospective marginal efficiency of capital relative to some interest rate
reflectingthe opportunitycost of the investedfunds. In addition,he pointed
out the intrinsicvolatility of private investment,due to the fact that any
forecastof the returnsof investmentaccruingin the futurewill be necessarily
incompleteand uncertain. Accordingto Keynes,in such an environmentinvestors
would be left to their "animal spirits' in making their investmentdecisions
rather than to a rationalcalculationof an inherentlyuncertaindistantfuture.
After Keynes, the evolutionof investmenttheory was linked to simple
growthmodels in the Harrod-Domartradition. This gave rise to the accelerator
theory,popular in the fifties and early sixties and widely used even today in
practical growth exercises. The acceleratortheory makes investmenta linear
proportionof changes in output, as derived from a fixed proportionsproduction
technology. This extreme simplicityexplains the popularityof the approach:
given an incrementalcapital-outputratio (ICOR), it is easy to compute the
investmentrequirementsneeded to achievea given output growth target. In this
model, profitability,expectationsand cost of capital considerationsplay no
role in the determinationof investment.

These overly restrictiveassumptionsled Jorgenson (1967) and HallJorgenson (1971), among others, to formulate the Neoclassical approach to
investment. This approachintroducesfactor substitutionin the derivationof
the demand for capital from the firm's cost minimization (or profit
maximization)problem. The desired capital stock is shown to depend on the
rental cost of capital (which in turn dependson the price of capital goods, the
real interestrate and the depreciationrate) and the levelof output. Decision
and delivery lags (or implicitlyadjustmentcosts) create a gap between the
current and desired capital stocks, giving rise to an investment equation,
namely an equation for the change in the capital stock.
This approach, in turn, has been subject to several criticisms
regarding the consistency, and plausibility,of its assumptions: (i) the
assumptions of

perfect competition and

exogenously given output are

inconsistent;(ii) the assumptionof staticexpectationsis inappropriate,since

investmentis essentiallya forward looking process; (iii) delivery lags are
introducedin an ad hoc manner.
An alternative formulation of the investment function is the *QO
theory of investmentassociatedwith Tobin (1969). In this approachthe ratio
of the market value of the existingcapital stock to its replacementcost (the
Q ratio) is the main force driving investment. Tobin providedtwo reasonswhy
Q may differ from unity: delivery lags, and increasing marginal costs of
investment. Abel (1981) and Hayashi (1982) reconcile the neoclassicaland Q
approaches,by showing that the latter followsfrom the firm's optimal capital
accumulationproblem under (convex)adjustmentcosts.

In this setting what

matters for investmentis marginal Q, i.e., the ratio between the increasein

- 5the value ox the firm due to the installationof an additionalunit of capital,

and its replacementcost. However,marginalQ is not observed;moreover,it will
generallydiffer from the observedaverageQ (-fhich
is just the market value of
existing capital in terms of new capital),except under conditionsof perfect
competitionand constantreturnsto scale (see Hayashi (1982)). They will also
differ if firms face quantityconstraintsin real or financialmarkets. In that
case, average Q will not provide all the relevantinformationfor investment
decisions;the latter will also depend on the relevantquantityconstraints.
The basic assumption of convex installation costs is highly
questionable. While such an assumption is necessary to bound the rate of
investment(so that a meaningfulinvestmentdemand functioncan be defined),it
can be argued that the cost of additionsto an individualfirm's capital stock
is likely to be linear (or even concave) in investment,due to the 'lumpy'
nature of many investmentprojects. More importantly,disinvestment,if at all
possible,is much more costly than positiveinvestment:capital goods often are
firm-specific,and have a low resale value. An extreme but useful view of this
asymmetryis to considerinvestmentcompletelyirreversible. In this case, the
adjustment cost function is asymmetric with infinite adjustment costs for
negative investmentrates. The notion of irreversibleinvestmentwas first
introducedby Arrow (1968), who characterizedthe dynamics of irreversible
investment under conditions of certainty. He showed that irreversibility
creates a wedge between the cost of capital and its marginal contributionto
profits. However, it is Ander conditionsof uncertaintywhen irreversibility
can have importantimplicationsfor investmentdecisions:as a recent literature
(e.g., Bernanke (1983), McDonald and Siegel (1986), Pindyck (1988b,1989),

Bertola (1989))has emphasized,irreversibleinvestmentcan be very negativsly

affected by risk factors.

The intuitive reason is that if the future is

uncertain any addition to productivecapacity today increasesthe probability

that the firm may find itself tomorrowwith 'toomuch' capital,which cannot be
(costlessly)eliminated due to the irreversiblenature of investment;hence
firms will be extremelycautiousin their capacityexpansiondecisions. As we
shall discuss below, this suggests that uncertaintymay be more relevant for
investmentdecisions than other conventionalvariables such as interestrates
or taxes.
In the disequilibriumapproachto investment(Halinvaud(1980,1982),
Sneesens (1987)), investment is a function of both profitabilityand output
demand considerations.In Halinvaud (1982),investmentdecisionsare separated
in two btages: the decisionto expand the level of productivecapacity.and the
decision about the capital intensityof that additionalcapacity. This last
decision depends on profitabilityvariables like the relativecost of capital
(includingthe real interestrate) and labor. On the other hand, the capacity
decision depends on the degree of capacity utilization in the economy as an
indicator of demand conditions. The distinction between both decisions is
meaningful due to the assumptionof a putty-claytechnology, so that factor
proportionsare flexibleex-ante but rigid ex-post. In Sneessens (1987),net
investment is positively related to the gap between actual and long run
equilibriumcapacities. This in turn is a reflectionof differencesbetween
actual and equilibrium rates of capacity utilizationand between actual and
equilibriummarkup rates. Therefore investmentdepends both on profitability
(discrepanciesbetween actual and equilibriummark-up rates) and on sales

constraints (discrepanciesin rates of capacityutilization). The investment

decision, in turn, takes place in a setting in which some firms may be facing
current and expectedfuture sales constraints,an importantdepdrturnboth from
the Neoclassical(Jorgenson)and the Q models.
Disequilibriummodels have often been criticizeddue to the simplicity
of their expectationalassumptions.However,market disequilibriumand rational
expectationsare not necessarilyinconsistent. Neary and Stiglitz (1983) have
shown that rational expectationsand excess supply in the goods and labor
markets can coexist, in a context of forward-lookingagents that anticipate
future sales constraint in a world of wage and price rigidities (see also
Precious (1985)). This is particularly relevant for investment since the
outcomesof decisionsmade todaywill be observedin the future,so expectations
play a crucial role. On the other hand, importantproblems of macroeconomic
adjustment,like deviationsof output from full capacity in the face of demand
shocks,are associatedwith (transitory)disequilibriumin the goods and labor
markets. In such conditions,a combinationof expectationsand disequilibrium
may be needed for a an adequateunderstandingof investmentbehavior.
In the developingcountriescontextinvestmentmay be subjectto other
constraintsbesides that of sales. Rama (1987) has formulatedand estimat^d
investment equations in terms of profitability and sales and financing
constraints. At the aggregate level, savings availabilitymay be limited
because of a lack of foreign savings in economieswith a significantstock of
outstanding foreign debt. Large fiscal deficits also reduce the volume of
domestic savings availableto financeprivate investment. At the microilevel
firms may face binding financialconstraintsif quantity adjustmentsrule in

domestic capital markets. This may be the case because of the existence of
controlledinterestrates and also becausecredit rationingmay be a featureof
the equilibrium in the loan market, as demonstratedby Stiglitz an,dWeiss
(1981). Asymmetric information,adverse selection and incentive effects may
make interestrite changesan inefficientdevice to sort out good borrowersfrom
bad borrowers.

Under those conditions, credit rationing and quantitative

constraintsmay become a preferredtool for lendingallocationby the creditors.

There is a growing literatureon the effectsof financialconstraints
on investment(see Fazzari Hubbard and Petersen (1988a,1988b),Calomiris and
Hubbard (1989),Mayer (1989),Mackie-MRson(1989)). Its main contentionis that
internalfinance (retainedprofits)and externalfinance (bonds,equity or bank
credit) are not perfect substitutes. The discrepancyin the cost of different
sources of financing is due to asymmetric information:lenders in capital
markets cannot evaluate accurately the

quality of

firms, investment

opportunities,thus making the cost of new debt and equity differ substantially
from the opportunitycost of internalfinance generatedthrough cash flow and
retainedearnings. According to this view, investmentwill be very sensitive
to financialfactors such as the availabilityof internalfinanceor the access
to capital markets. This new strand is clearly a departure from the perfect
capitalmarket approachwere the financialstructureof the firm is irrelevant
for investmentdecisions;in this new setting :iemarket value of a firm is not
independentof its financialstructure.
Empiricalresearchalong these lines has been undertakenfor the U.S.
by Fazzari,Hubbard and Petersen (1988a). They test the role of the financial
structure of the firm in the Q,

neoclassical and accelerator models of

-9investmentdiscriminatingby firm size. The general finding is that financial

effects are important for inves,mentin all firms, but also that consistent
differences exist across firms regarding the sensitivity of investment to
balan e sheet variables that measure liquidity,dependingupon their retained
earningspolicies. An importantmacroeconomicdimension of these findingsis
that, provided fluctuationsin firms cash flows and liquidity are correlated
with movements in aggregate economic activity and the business cycle,
n,acroeconomic
instabilitymay affect investmentalso throughfinancialchannels,
mainly for firms relyingheavily on internal (and external)finance .
Another relevant feature of investmentin LDCs is the high import
content of capital goods. This raises an importantpoint emphasizedin two-gap
models (Chenery and Bruno, 1962 and Bacha, 1982), namely that the lack of
foreign exchange may constitutea major constraint to sustain high rates of
investmentand growth in LDCs. In fact, in economieswere domesticand foreign
capital goods are highly complementarythe lack of foreign resourcesto import
machinery and equipment will be an impediment to growth (in the medium run
import substitutionof capitalgoods and export promotionwould ease the foreign
exchange constraint). The foreign exchange constraint also has important
implications (discussed below) for the impact of exchange rate policy on
investmentdemand.

3.

MacroeconomicPoliciesand Private Investment

In this sectionwe examine the effects of macroeconomicpolicieson

private investment. In particularwe are interestedin studyingthe impact on

investmentof differenttools of monetary,fiscaland exchangerate policy aimed

10

at correctingunsustainablemacroeconomicimbalances. The traditionalmacro

package includesrestrictivefiscal and monetarypolicies supplementedwith a
real devaluationof the exchangerate. We review the most relevantliterature
on the macroeconomicdeterminantsof investment,paying particularattentionto
the transmissionmechanismsand likely effects of differentmacro policies on
private investment.
A summary of the linkagesbetween adjustment,investmentand growth
appears in Chart 1

The basic notion here is that the correction of macro

imbalancesand the achievementof macroeconomicstabilityis a prerequisitefor

achieving sustainedgrowth. In turn, a strong responseof private investment
to the set of incentivesput in place by an adjustment program is a basic
element for the stabilizationeffort be followedby sustainedgrowth. Chart 2
offers a

schematic view of the transmission mechanisms through which

macroeconomicpolicies affect private investment.The first three columns show

the variables that influence the profitabilityof capital (the real interest
rate, the market price of installedcapital,and the price of new capitalgoods)
and how they are affectedby the differentmacro policies.The fourth and fifth
columns single out demand conditions and real credit availabilityas other
determinantsof investmentthat may be affected by macroeconomicpolicies.We
now turn to a more detaileddiscussionof these effects.
3.1.

Monetary and fiscal policy and private investment

Restrictivemonetary or credit policiesaimed at reducing inflation

and!or the current account deficit may affect investmentthrough two "price'
channels. One is the rise in the real cost of bank credit, a major source of
investmentfinancingin LDC's. The second is the increasein the opportunity

11

cost of retainedearnings,also an importantsource of investmentfinancingin

most developingcountries,due to higher real interestrates. Both effects lead
to an impl'citor explicit (in the case of organizedequity markets) reduction
in the market value of existingcapital relativeto its replacementcost (the
Q ratio is expectedto fall with a monetarycontraction),and thus to a decline
of investment. In repressedfinancialmarkets,credit policy affectsinvestment
directly through the stock of credit available to firms with access to
preferential interest rates and through interest rates for firms operating
through the unofficialmoney market (for models of credit policy and growth in
financially repressed economies see Van Wijnbergen (1983a and 1983b)). The
institutionalset-up of the financial markets in developing countries is
certainly an important feature determining the impact and transmission
mechanismsof monetary and credit policy on investment(an empiricalanalysis
of monetary stabilizationpolicies for Korea with endogenousdzterminationof
investmentis providedby Van Wijnbergen, (1982)).
High fiscaldeficitsalso push up interestratesand crowd-outprivate
investment. However, the way a fiscal deficit is correctedalso matters from
the viewpoint of investment. Differentmixes of tax increasesand/or spending
reductionscan be expectedto have differenteffectson private investment. In
particular, due to institutionaland political rigidities in the ability of
governmentsto reduce currentpublic expenditure,fiscaladjustmentoften takes
tne form of reduced public investment, some of whose components may be
complementarywith private investment. In fact, the empirical evidence from
data on developingcountriesanalyzedby Blejer and Kahn (1984) indicatesthat
public investmentin infrastructureis complementarywith private investment

12

(and other types of public investmentare not). Similarly,Musalem (1989) finds

evidence of complementarilybetween private and public investmentin a timeseries study of investmentin Mexico.
However, Balassa (1988) reports cross section statistical results
showingthat public and private investmentare negativelycorrelated,with a one
per cent increase in public investmer.t
being associatedwith a 0.55 percent
decline in private investment.Furthermore he finds a negative correlation
between the share of public investmentin total investmentand the size of
incrementaloutput-capitalratios,arguing for a

lower efficiency of public

investmentrelativeto private investment.

The general issue is how monetaryand fiscal policiesaffect total and
private investmentand what are the more relevanttransmissionmechanismsat
work. A plausiblemechanism for restrictivedemand policiesto affect private
investment is through the market value of capital. As recent econometric
evidenceshows (see Solimano,1989)aggregateinvestmentprofitabiLityis L-Lghly
procyclical. Tobin's Q increases in upturns and falls in downturns- so we
should expect the market value of capital to fall in the short run in response
to a slowdown in economic activity following restrictive demand policies.
Another relevanttopic for researchin this area is the sensitivityof private
investmentto cyclical changes in activity levels. Econometricestimates of
investment functions show, in general, a strong response of investment to
changes in output. This is a puzzling finding since a non-negligiblepart of
output fluctuationsappear to be transitory(thereforethey should not affect
investment),and it is costly to install capital (so adjusting to transitory
shocks is also costly). Then this excessive output- related variability of

- 13

investmentin the cycle remains largelyunexplained(see Blanchard'sdiscussion

of Shapiro,(1986)).However,myopic expectationsand short investmenthorizons
may be consistentwith the observedlarge fluctuationsof investmentassociated
with output changes.
The initial downturn in economic activity often associated with
macroeconomic adjustment may also affect investment through its effect on
expectations. In fact, a

current recession could form the basis for

wpessimistic"expectations,leading investorsto postpone investmentuntil the

recovery arrives; this, in turn, may prevent a take-off of investment
(particularlyof projects with short gestation lags) and delay the recovery
itself,and the economy may get stuck in a low activityequilibriumbecause of
insufficientinvestmentarising from self-fulfilling
pessimism on the part of
investors. How to avoid such an outcome is an importantconsiderationin the
design of restrictivedemand policies that minimize the potentiallyadverse
impact on investmentand growth.

3.2.

Exchange rate policy and private investment

A key elementof almost any adjustmentand stabilizationplan seeking

a reduction in the size of the current account deficit is a combinationof

expenditure reducing with expenditure switching policies. The latter refers
basically to a real devaluation. A real depreciationmay affect investment
through several channels:
i)

The profitabilityof investment: a devaluation may affect the

profitabilityof investmentth_ough its impact on the relativeprice of capital

in the economy.In fact, Buffie (1986)and Branson (1986) show that if capital

- 14 -

goods have an import content then a devaluation raises the supply (or
reposition)price of capital in terms of home goods; ceteris paribus, this
effect terndsto depress investmentin the home goods sector. An empirical
confirmationof the presumptionthat a real depreciationreduce investment(in
the short run)

is provided by Husalem (1989) for the case of aggregate

investmentin Mexico.
In these models, investmentis treated as a compositegood produced
by combining domestic (i.e., construction or infrastructure)and foreign
components (i.e., machinery and equipment).

In this setting, a

teal

depreciation of the exchange rate acts as an adverse supply shock in the

"production'of investmentgoods.

Branson (1986) explicitly calculates the

impact of a devaluationon Tobin's Q in the home goods sector,concludingthat

profits fall (and along with them the market value of capital)and the cost of
capital (and its repositionprice) rises followinga real depreciation.Solimano
(1989) finds a negative effect of real devaluations on investment in his
empirical simultaneousequation model for Chile; his results show that the
economy-wideTobin's Q falls when the real exchange rate rises because of a
dominantrepositionprice effect followinga real dep-eciation(inprinciplethe
market value of capital rises for the traded goods sector after a devaluation,
but this effect may be Loo small relativeto the repositionprice effect and the
effect of devaluationon the market value of capital in the home goods sector).
The issue is also reviewed,conceptually,by Lizondo and Montiel (1988),who
distinguishbetween investmentin the traded and non-tradedgoods sectors in a
model in which capital is sector-specific. They decompose the effect of
devaluationon profitabilityinto three components: a) its impact on the cost

15 -

of capital, b) its effect on the productwage in both sectors (also examined

by Van Wijnbergen (1986) and R'sager (1984)),and c) its impact on the cost of
importedintermediateinputs. Their conclusionis that the net effect of a real
depreciationis generallyambiguous,since it tends to increaseinvestmentin
the traded goods sector and reduce it in the home good; sector.
Anotherchannelthroughwhich devaluationmay affect the profitability
of investment is the real interest rate.

Consider first the case of an

unanticipateddevaluation(we dlecuss below the anticipateddevaluationcase),

and assume that inte-estrates are determinedin domesticassets markets (i.e.,
in the money market). In this case a devaluationwill increasethe price level
through its impact on the cost of importedintermediateinputs and wages under
indexation;if monetary policy does not fully accommodatethe increasein the
price level, real money balanceswill fall pushing up the real interest rate
for a given rate of (expected)inflation. Thus devaluationwill depress the
market value of capital exertingan adverse effect on investment.On the other
hand, if devaluation was anticipated and if it succeeds in eliminating
devaluationexpectations,then it may result in an investmentexpansion,since
the requiredreturn on capitalwould tend to fall reflectingthe reductionin
the anticipatedrate of depreciation(whetherthis will be so depends on the
degree of capital mobilityand the import content of investment;see below).

ii) Financial effects of devaluation: the debt crisis of the eighties,

and the adjustmentpolicies adopted thereafter,has brought renewed attention
to the effects of devaluationon the real value of liabilitiesdenominatedin
dollars held by domesticfirms, banks and financialintermediaries.

- 16 -

In the case of foreign-indebted

firmsdevaluationautomaticallyraises
the burden of debt, hence reducingtheirnet worth. If domesticcredit markets
are imperfect (as it is often the case in LDCs) these firms may subsequently
have to face credit constraints,or will have to bear higher costs of outside
financingas creditorsraise their lendingrate to compensatefor the increased
default risk.

These financial pressures will lead directly to reduced

investmentfor those highly indebtedfirms in risk of bankruptcy. The increase

in the real value of firms' foreigndebt also affects investmentindirectlydue
to its adverse impact on the financialsystem. As the net worth of indebted
firms falls so does the quality of domesticcreditors'portfolios (i.e.,banks
and financial intermediaries). In fact, they may be forced to reduce their
exposure by cutting their loans --

or may simply go bankrupt. The ensuing

tighteningof creditmarketsmay result in a reducedsupply of credit (or higher

interestrates) even for firms that had no foreigncurrency liabilities. This
tighteningof credit conditions,in turn, discouragesinvestmentas financing
becomes more scarce or more expensive.
The financialeffects of an unanticipateddevaluationare sometimes
so significantthat firms and/or financialintermediarieshave been bailed out
by the public sector to avoid an epidemicof bankruptciesthat could result in
a major economiccrisis and lead to the failureof the adjustmentpackage. The
financingof the bailout, however,may lead in the future to a domestic debt
overhang, as the treasury has to issue bonds to cever the foreign exchange
losses of commercialbanks and/or firms indebtedin dollars terms. The ensuing
higher stock of public debt associatedwith the rescue of indebtedfirms puts
an upward pressureon interestrates, crowdingout private investment. It is

- 17 -

interestingto note the implicittrade-offbetween supportinginvestmenttoday

(via subsidizationof indebtedfirms) versus investmenttomorrow (arisingfrom
the crowding-outeffect of public debt issued in previousperiods).
Empirical studies of the financialeffects of devaluationand its
Jmpact on investmentare scarce;exceptionsare Easterly (1989) and Rosensweig
and Taylor (1989). Easterly (1989) sets up a ComputableGeneral Equilibrium
model (CGE) for Mexico, extendedto include financialflows in order to trace
the impact of a currencydevaluationon investment(whichis assumedto be selffinancedand/or face credit constraints). In this model a devaluationis shown
to result in a fall in both GDP and private investment,but with the latter
contracting substantiallymore than the former. The main cut in investment
comes from corporations,and is due to a sharp increasein their real foreign
indebtedness. Easterly reports that the cash flow of corporationsdeclines
substantiallyin the simulationsas a result of capital losses on dollar debt,
while the replacementcost of capital rises sharply. Rosensweig and Taylor
(1989) arrive to mixed results using a CGE model for Thailandwith endogeious
portfolio choice.

In their simulations GDP increases following a ieal

depreciation,under the assumption of a strong export response to relatLve

prices incentivesand no capital losses from devaluation. In turn, higher net
worth provides more deposits to banks, credit supply rises, and the interest
rate falls.

The result is an increase in investment. However, in their

simulationsincludingcapital losses on foreign liabilitiesassociatedwith a

devaluation,domesticcapitalformationcan be crowdedout, and the expansionary
net exports effect may be offset.

18

iii) Devaluation,activitylevels and investment: A third channel through

which devaluationmay affect investmentis providedby its effect on aggregate
demand. This may be especiallyimportantwhen firms face sales constraints,so
that the degree of capacityutilizationor other variable representingdemand
considerationshas a strong systematiceffect on investment (sucn effect is
often found empirically; see e.L. Musalem (1989) and Solimano (1989)). If
devaluationreducesaggregatedemand ex-ante,then ex-post investmentwill fall.
Moreover, if investmenthas a significantimport content,then output expansion
is likely to be a necessary (but not sufficient)condition for investmentnot
to fall ex-post. The literature on contractionarydevaluation (Krugman and
Taylor,1978; Van Wijnbergen,1986; Edwards,1987; Serven,1986; Solimano,1986;
Lizondo and Montiel, 1989) emphasizesthe slow working of substitutioneffects
arising from devaluation; hence in the short run the impact of a real
devaluationon aggregatedemand is dominatedby its adverse incomeeffects.The
latter operate through two main channels: one arises form the likely initial
trade imbalance,which results in a real income transfer to the rest of the
world (even at given terms of trade); the other from the negative impact on
consumptionof real income redistributionfrom wages to profits. On the supply
side, three transmissionmechanismsmay contributeto output contraction: the
increased real cost (in terms of domesticgoods) of importedinputs, the rise
of working capital costs, and real wage resistance. If the net effect of a
currency devaluation is contractionary,i.e., GDP falls, then the slump in
economicactivity is likely to form the basis for investorsto cut investment
spending-- unless they clearly perceivethe slump to be transitory. However,
with sufficiently strong substitution effects (e.g., a large impact of

19

devaluationon exports)an expansionaryoutcomewill result,and so devaluation

may raise real income and stimulate investment spending as the degree of
capacityutilizationincreases.
The discussionuntilnow has focusedon devaluationwithoutmaking any
explicit distinction between anticipatedand unanticipateddevaluation. An
anticipateddevaluationcan affect investmentthrough two additionalchannels:
the real interestrate and the import content of capital goods.
iv)

The real interest rate channel:

The effect of an anticipated

devaluation on interest rates depends crucially or. the degree of capital

mobility (that is,

the

costs of

portfolio adjustment) and

on

the

substitut'bilitybetween domestic and foreign assets. Let us consider the

general case of imperfect capital mobility and imperfect substitutability
between domesticand foreignassets. In this context,asset market equilibrium
makes the domestic real interest an increasingfunction of the foreign real
interestrate plus the expectedrate of depreciationof the real exchangerate.
Hence the perceptionby the public that the real exchangerate is overvaluedand
a real depreciationis imminentwill lead to higher real interest rates and
reduced investment;in addition,this effect will be more importantthe higher
the degree of substitutability(and also of capitalmobility) between domestic
equity and foreign assets. However, under conditions of imperfect asset
substitutabilityor restricted capital mobility, it is also possible that
investors will shift their portfolios towards imported capital goods in the
expectation of a devaluation.

v)

Let us explore this case now.

The speculative hoarding effect of imported capital Aoods:

The

anticipationof a real devaluationmay also have a positiveeffect on investment

- 20 -

demand, when capital goods have a significant import content, before a

devaluation actually takes place.

The mechanism that could produce this

outburstof investmentis the speculativehoardingeffect,which would increase

the purchasesof importedcapitalgoods in anticipationof a future devaluation
that would raise the replacementcost of investment. As argued by Dornbusch
(1984),the more plausibledynamicsis the following: firms and importerswill
attempt to increasetheir purchasesof foreigncapitalgoodswhen a devaluation
is expected in order to collect the anticipatedcapital gain; then, when the
devaluation actually occurs and the implicit subsidy embodied in the
overvaluation is eliminated, a sharp cut in investment may follow.

The

speculative hoarding of foreign capital goods may give way to a period of

depressed investment after the devaluation, as the over-accumulationis
reversed. (A similartime patternwould emerge in the case of transitorytrade
liberalization,when the latter includesa temporary reductionof tariffs on
importedcapital goods.) Obviously,a crucial assumptionfor this pattern to
emerge is that of imperfectcapitalmobility and/or

imperfectsubstitutability

between domestic equity and foreign assets, a requirementwhich in principle

seems quite realisticfor most LDCs. A close study of the observeddynamicsof
imports of investment goods during devaluation episodes is worth to be
undertaken,if we want to learn more on the dynamics of investment (and the
current account)during an adjustmentprogram.

21 -

4.

The incentive structure and investment response:

uncertaintyand irreversibility

credibility,

A key ingredient of most macroeconomicadjustment packages is a

change in economic incentivesthat switches spending towards domestic goods
(offsettingthe deflationarybias of the usual monetary and fiscal restraint)
and raises profitabilityin the tradablesector. This change in incentivesis
expected to lead to an outburst of investmentin the tradable goods sector,
increasing productive capacity and enhancing economic growth --

and thus

ensuringthe sustainabilityof the adjustmenteffort.

In practice, however, the investmentresponseoften is unexpectedly
weak, and involves long delays.

This poses major difficulties for the

adjustmenteffort, since in the absence of an investmentexpansion the shortrun deflationaryconsequencesof the expenditure-restraining
measures may be
magnified,leading to a persistentreductionin growth. In this way, the lack
of an adequate investment response in the tradable sector to the change in
economicincentivesincreasesthe cost of the adjustmentin terms of employment
and growth; ultimately, it may render the stabilization effort socially
unacceptableand thus unsustainable.
Thereforeit is essentialto improveour understaniingof the reasons
that underliethis slow reactionof investment,in order to improveour ability
to design sustainable adjustmentpolicies. In the theoretical framework of
symmetric convex adjustment costs to investment, this inertia could be
explainedby a combinationof high adjustmentcosts with sluggishexpectations
on the part of investors. However, the assumption that firms face rapidly
increasing marginal costs to capacity expansion appears questionable on

- 22

empirical grounds, and there is also no clear justificationfor a myopic

expectationalbehavior by investors. A more satisfactoryexplanationcan be
offered by emphasizingthe importanceof risk factors in investmentdecisions,
which would make investorsreluctantto undertake fixed investmentprojectsin
a context of high uncertaintyabout the future economic environmentand, in
particular,about the future incentivestructure.
Chart 3 provides a schematic illustrationof the implicationsof
uncertaintyfor asset decisions.When there is uncertaintyabout the economic
environment or about the permanence of economic incentives, irreversible
decisionswill be delayed to avoid long lastingmistakes.In particular,fixed
investment decisions will be postponed, with the corresponding negative
consequencesfor growth, in favor of more flexiblepositions in liquid assets.
Among these, capital flightwill be a preferredoption wheneverthere are major
doubts about the sufficiencyor the sustainabilityof the adjustmenteffort.

4.1

1
Irreversibility,
uncertainty,and investment

As an emerging literaturehas emphasized (see Pindyck, 1989) for

references), the key role of uncertainty in investment decisions follows
directly from the irreversiblenature of most investmentexpenditures. These
can be viewed as sunk costs, because capital, once installed, is firm- or
industry-specific
and cannot be put to productiveuse in a differentactivity
(at least without incurringa substantialcost). The decisionto undertakean
irreversibleinvestmentin an uncertainenvironmentcan be viewed as involving
the exercising of an option --

the option to wait for new informationthat

I The material in this section is largelybased on Pindyck (1989).

- 23

might affect the desirabilityor timing of the investment. Thus, the lost
value of this option must be consideredas part of the opportunitycost of
investment--

an issue which is overlookedin the conventionalnet present

value calculations(which would therefore underestimatethe opportunitycost

and overestimateinvestment). As recent studieshave shown, this opportunity
cost can be substantial,and is also very sensitiveto the prevailingdegree of
uncertai,aty
about the economicconditionsthat determinethe future returnsto
the investment. As a consequence,changes in uncertaintycan have a very
strong impact on aggregateinvestment;irom a policy perspective,the stability
and predictabilityof the incentive structure and the macroeconomicpolicy
environmentmay be much more importantthan tax incentivesor interestrates.
In other words, if uncertaintyover the economicenvironmentis high, tax and
related incentivesmay have to be very large to have any significantimpact on
investment.
is completely
It is importantto note that this effect of unicertainty
independentof investors' risk preferencesor of the extent to which their
risks may be diversifiable. Investorsmay be risk-neutral(as assumed by most
of the irreversibilityliterature)and their risks completelydiversifiable;
yet investmentwould continue to depend negativelyon the perceived degree of
uncertainty. The latter becomes important here simply because the fixed
investmentdecisioncannot be 'undone'(at least at zero cost) if future events
turn out to be unfavorable. In general,therewill be a value to waiting (i.e.
an opportunitycost to investingtoday rather than waiting for informationto
arrive) whenever the investment is irreversibleand its net payoff evolves
stochasticallyover time.

- 24

From a macroeconomic perspective, there are different forms of

uncertainty which may be relevant for investment decisions. Consider for
example the investmentdecisionof a firm facing uncertainfuture demand,which
has been analyzed by Pindyck (1988b) and Bertola (1989). If investment is
irreversible,then some of the firms' installedcapacitymay go unutilizedif
demand turns out to be low. Ex-ante,this will make firms want to hold less
capacity than they would under conditionsof reversibility. Mtoreover,
Pindyck
and Bertola also show that increaseddemand volatilitywill generally lead to
reduced investment,by worsening the 'worst case' scenarioin which the firm
regrets the irreversiblecapacity expansion (it also makes the 'high demand'
scenariobetter,but this can be taken care of by installingadditionalcapital
if needed, i.e. the adjustment cost function is asymmetric). The firm's
optimal investmentrule equates the expecteddiscountedvalue of profits from
the marginal unit of capital to the installationcost plus the 'value of
waiting' lost by undertakingthe capacityexpansion.
The case of uncertain real exchangerates has been studied by Dixit
(1987), Krugman (1988), and Krugman and Baldwin (1987), who consider the
behavior of a firm who must decide whether to enter (or exit) the foreign
market. They show that sunk entry costs combinedwith uncertain future real
exchange rates will cause firms not to enter the market even though favorable
exchange rate movements would seem to make entry profitable.

Similarly,

Caballero and Corbo (1988) show that uncertaintyover future real exchange
ra.es can depress exports. Dornbusch (1988) examines the related issue of
capital flight reversalfollowinga real depreciation;he argues that in order
to attract the previouslyevaded capital to irreversiblefixed investment,an

25

over-depreciationof the exchange rate may be needed, to compensate the

uncertaintyfaced by investorswith a frontloadingof the returnsto investing
in the domesticcountry.
Ingersoll and Ross (1988) examine the ro'e of interest rate
uncertainty in a context of irreversibleinvestmentwhere future returns are
known with certainty (see also Tornell,l988). As with uncertaintyover future
cash flows, this creates an opportunitycost for investing. They concludethat
the effect of interestrate variabilityon the optimal timing of investmentmay
be quite sizeable;moreover, they show that a fall in expectedfuture interest
rates need not lead to increasedinvestment. The reason is that such a change
also lovers the cost of waiting, and thus can have ambiguous effects on
investment. In other words, interestrate volatilitymay be more importantfor
investmentthan interestrate levels.
The relevanceof these results for macroeconomicpolicy, especially
in developingcountries,cannot be overemphasized.Consider,for example, the
problem of relativeprice volatility. Many developingcountries suffer from
high and unpredictableinflation,which is usually matched by high relative
price variability. The irreversibilityapproach suggests that this would
reduce the effectivenessof relativeprice zhanges in stimulatinginvestment.
Specifically,a history of frequentrelativeprice swings would make investors
extremely cautious in reacting to a

policy-induced change in sectoral

incentives;substantialtime may elapse before investorsbecome convincedthat

the change is permanent -- and before they are willing to give up their option
to postpone investment. Notice also that the implementationof an adjustment
packagemay well increaseuncertaintyin the short run, as private agents start

- 26 -

receivingmixed incentive signals -- some associatedwith the previous policy

rules, some with the stabilizationpackage, and some with the structural
reformsaimed at restoringmedium term growth. An exampiealong these lines is
provided by van Wijnbergen (1985), who shows that a trade reform which is
suspectedto be only temporarycan in fact lead to a fall in investment-- as
economicagents postpone investmentin both the home and traded goods sectors
in order to receive additionalinformation.
The debt overhang faced by many high-indebtedcountries creates a
similarproblem,which has been emphasizedby Sachr (1988). It arises from the
need to carry out an external transfer to the country's creditors, and
representsanother source of instabilityof the macroeconomicenvironment:in
a context of uncertainty,the level of the real exchangerate and/or the demand
management policies consistent with the required transfer also become
uncertain;the size of the transferitself is not known with certainty,as it
depends on uncontrollablefactors such as the future level of world interest
rates and the terms of trade. Carrying out the transfermay require future
real exchange rate changes, fiscal contraction,or both. Thus investorsmust
face the risk of large swings in relative prices, taxation, or aggregate
demand; as we saw above, each of them would lead to reduced investment.
In practice, this effect may be hard to identify,since foreign debt
may affect investmentadverselythrough two additionalchannels (emphasizedby
Borenzstein (1989)). First, the debt overhang,which acts as an anticipated
foreign tax on current and future income: since part of the future return on
any investmentwill accrue to the creditorsas bigger debt servicepayments,it
discourages capital accumulation and promotes capital flights. Second, the

- 27

credit rationing effect: a highly indebtedcountry is likely to face c.redit

constraints in internationalcapital markets, which is equivalentto facing
higher real interestrates, and this will also discourageinvestment.

4.2

The role of credibility

From a

policy perspective, an extremely important source of

uncertainty is the imperfect credibility of policy reforms. The latter is

related to the public'sperceptionsabout both the internalconsistencyof the
adjustmentprogram and the government'swillingness to carry out the program
despite its impliedsocial costs. Unless investorsview the adjustmentprogram
as fully credible in both senses, the possibilityof a future policy reversal
will become a key determinant of the investment response. As argued by
Dornbusch (1988), any adjustmentprogram can be undone by revertingeconomic
policies-- while investorscannot undo their fixed capital decisions. In such
conditions, the value of waiting arises from the losses (the 'irreversible
mistake', in Bernanke's (1983) terminology)that investors would incur if
policy were in fact reversedin the future. Clearly,the larger the perceived
probabilityof a future policy reversal,the less willing investorswill be to
undertake fixed investmentprojects -- or the larger the current return they
will require in order to compensate for the possibilityof an irreversible
mistake.
This implies that any given set of policy measures can have widely
different effects on investment depanding on the prevailing degree of
'confidence'of the public.

In particular, stabilizationmay entail large

social and economic costs if credibility is low --

since the investment

- 28

response will be insufficientto offset the deflationarybias of the usual

fiscal and moDnetaryrestraint measures; thus, a persistent recession may
develop before investorsbecome confidentenough that the adjustmentmeasures
will be maintained. This may be particularlyrelevantin economieswith a past
history of frequent policy swings or failed stabilizationattempts -features shared by many highly indebted countries --

two

in which the private

sectorhas learned to view adjustmentprogramswith considerableskepticism.

Hence

setting the

right economic incentives is

required

preconditionfor investmentand growth, but it does not guarantee that they

will in fact take place (Dornbusch,(1989)).Obviously,high credibilitywould
help speed up the investmentresponseand reduce the costs of the adjustment.
However,the questionof how can credibilitybe affectedby governmentactions
remainslargelyunresolved.Specifically,an importantissue here is the choice
between gradual and abrupt stabilization. The former would set initially
modest objectives,which can be achievedwith near certainty,in order to build
up the government'sreputation. The latter would start with an overadjustment
(e.g., an over-depreciationof the exchange rate) to frontload the incentives
to resourcereallocation(but also the costs of the adjustment). As argued by
Edwards (1988),the choice may largelydepend on the specificsof each country;
the social distributionof adjustmentcosts implicitin the program, together
with past policy experience,are likely to be importantissues here.
It is importantto emphasize that policy reversal is an endogenous
outcome in this framework,since current private sector decisions affect the
opportunity set of

future policy actions and ultimately determine the

sustainabilityof the adjustmentpolicy. As an example, consider again the

29 -

case of a large real depreciationthat due to low confidencefails to attract

investment to the tradable sector.

Its only visible effects will be a

deflationary real income cut and an income redistributionfrom labor to

capital, especially in the traded goods sector; however, because the
depreciationis not sufficientto compensatefor the lack of credibility,the
increased profits will be reflected in increased capital flight.

Social

pressureand balanceof paymentsproblemsmay eventuallyforce policy reversal,

thus confirmingthe initial skepticismof investors.
The alternativesituationstarts with high confidence,which allows
an investment boom and validates the adjustment program.

In both cases

expectationsmay be self-fulfilling,
which reflectsthe possibilityof multiple
equilibria in this framework --

an indeterminacythat also arises in the

literature on investment under monopolistic competition (see e.g. Kiyotaki

(1988),Shleifer and Vishny (1989)). In this case, it is due to the presence
of an externalitythat creates a wedge between the social and private returns
to investment:higher aggregate investmenthelps sustain the adjustmenteffort
and thereforeresults in higher returnsto investment,a mechanismthat will be
ignoredby the individualinvestor. Since the 'high confidence'equilibriumis
clearly better in a meaningful sense than its alternative,it is crucial to
investigate what

specific policy measures (e.g., additional temporary

investmentincentives)can lead the economy to this superioroutcome.As argued

by Dornbusch (1989), sufficientexternal support to the stabilizationeffort
may

play an

important role by

raising investors' confidence in

the

sustainabilityof the adjustment,thus giving way to the investmenttakeoff.

- 30 -

4.3

Empiricalapplications
The empirical literatureon uncertainty and irreversibilitystill

remains very scarce.

A simple nonstructuralapproach is used by Pindyck

(1986),who tests for the effectsof uncertaintyby introducingthe varianceof

stock returns as an explanatory variable in an otherwise conventional
investmentequation;his results with aggregateU.S. data indicate that the
variance of stock returns is an important factor for investment growth.
Solimano (1989) also investigatesthe effects of economic instability in an
empirical simultaneousequation model applied to Chile. He finds that the
volatilityof the real exchange rate and output exert a significantnegative
impact on private investment, and argues that the large swings in both
variables in Chile during the eighties may have resulted in a substantial
reduction in private investmentas compared to a counterfactualscenario of
lower relative price and output variability.Dailami and Walton (1989) also
argue that macroeconomicinstabilitymay be one major cause of low investment
in Zimbabwe.
A differentapproach is used by Bizer and Sichel (1988),who develop
a structuralmodel of capital accumulationwith asymmetriccosts of adjustment;
in their framework, irreversibilitywould imply higher downward adjustment
costs than upward ones. Their preliminaryresultsusing industrydata for the
U.S. manufacturing sector are somewhat mixed, perhaps due to aggregation
problems.More work along these lines, perhapsusing a cross-sectionof country
data, should be a researchpriority. However, it should be noted that sample
variances (or other sample measures of variability) represent imperfect
measuresof risk; in particular,they cannot c&pturethe 'peso-problem'
type of

31

uncertaintyassociatedwith the possibilityof a regime change (e.g.. a policy

reversal),which may be very hard to measure empirically. Moreover, the role
of irreversibilitymay be masked in aggregate data; as Bertola (1989) points
out, the irreversibilityconstraint is probably much more relevant at the
disaggregatedlevel.
Simulationmodels provide an alternativeway to assess the practical
importance of irreversibility. Rough numerical calculations reported by
several authors suggest that it may be very large indeed for 'reasonable'
parameter values. The developmentof a structuralsimulatior.
model should be
anotherpriority in the researchagenda. Such a model, parameterizedto fit a
particularcountry or industry,could be very useful to evaluatethe impact on
investmentof policy changes and of changes in specificforms of uncertainty.
In particular,it could be uted to analyze explicitlythe effects of perceived
possible shifts in the policy regime -- i.e., credibilitychanges.

32 -

Chart 1

Macroeconomic

Medium Term

Policies
_ *

initial
disequilibrium

Adjustment

~~*_

*______

Macro
Stability

Policies:
Exchange rate
Demand Management

Sustained
Growth

Incentivestructure
credibility

Private Investment

EconomicGrowth

- 33

Chart 2

Macroeconomic Policies and Private Investment

transmission
mechanism

I
real interest
rate

market value
of capital

policy

1. monetary
contraction

reposition
price of
capital

demand
levels

+/-

+/_

real credit
availability

2. fiscal
adjustment

i) spending

+/-

reduction

ii) tax

increase
3. devaluation

+/_

+l

34 -

Chart 3

Incentives and Investment Response

Liquid assets

Uncertainty

Cav'italFlight

Irreversible
capital

Postponement of
Investment

- 35 -

Chart 4
Effects on Private Investmentoft
(EmpiricalStudies)

Real depreciation
of the exchanSe rate
(short run)

Increasein
Increase in
capacityutilization public
investment

Blejer and
Khan (1984)

Positivefor
public investment in
infrastructure

Balassa
(1988)

Negative

Musalem
(1989)
Solimano
(1989)

Negative

Negative

Positive

Positive

Positive

- 36 5.

ConcludingRemarks
In this paper we have reviewed the linkages between macroeconomic

adjustmentand private investment.From the policy viewpoint,there are three

broad areas relevant for research. The first concerns the effects of
macroeconomic adjustment policies on private investment.

There are some

institutionalor economic features shared by many LDCs th&t may modify in a

substantialmannet the transmissionmechanismsthroughwhich fiscal,monetary,
and exchange rate adjustmentaffect investmentdecisions. To advance in the
design of macroeconomicstabilizationpoliciesthat minimizethe adverse shortterm impact on investment,we need to know more about the implicationsfor
private investment of fiscal adjustment (and especially public investment
reductions), of monetary restraint under alternative financial market
arrangements,and of exchangerate changes.
The second researcharea concernsthe implicationsof irreversibility
and uncertainty. A more completeunderstandingof their effects on investment
decisionsis a crucial prerequisitefor the design of adjustmentprograms that
;.t.xroduce
credible incentivesfor the expansion of investment,leading to a
resumption of growth and making the adjustment effort sustainable. The
development of models suitable for the empirical study of irreversible
investmentunder uncertaintyshould be a top priority in the researchagenda.
More work is also needed on the investmentconsequencesof policy credibility,
as well as on the policy implicationsof the investmentexternalityintroduced
by the credibility/sustainability
link.
The third research area is concerned with the links between the
reduction in the transfer of external resources and the drop in investment

- 37 -

observed in many LDCs. A furtherunderstandingof the transmissionmechariams

at work and the relatedpolicy implicationsare clearlyvery relevantareas for
policy oriented researchin the field of macroeconomicadjustmentand private
investment.

38

REFERENCES

Abel, A.(1980): "EmpiricalInvestmentEquations:AnIntegrativeapproach,"

Carnegie-RochesterConference Series on Public Policy on the state of
Macroeconomics.
and 0. Blanchard (1983): "Investmentand Sales: An Empirical Study,'
Working Paper, Harvard University.

Arrow, K. (1968): "OptimalCapital Policy with IrreversibleInvestment",in

Value, Capital,and Growtht Essays in Honor of Sir John Hicks, J. Wolfe,
ed. EdinburghUniversityPress.
Bacha, E. (1984): "Growth with Limited supplies of Foreign Exchange: A
Reappraisalof the Two Gaps Model."
Balassa, B. (1988) " Public Finance and Economic Development " PPR Working
Paper 1 31. The World Bank.
Bernanke,B. (1983): "Irreversibility,
Uncertainty,and CyclicalInvestment,"
QuarterlyJournal of Economics.
Bertola,G. (1989): "IrreversibleInvestment,"mimeo, PrincetonUniversity.
Bizer, D. and D. Sichel (1988): "Irreversible Investment: An Empirical
Investigation,"
mimeo.
Blejer, M. and M. Khan (1984): "GovernmentPolicy and Private Investment in
Developingcountries,"I.M.F. Staff Papers 31.
Borensztein,E. (1989) " Debt Overhang,Credit Rationing,and Investment"mimeo
I.M.F.
Branson,W. (1986):'Stabilization,
Stagflation,and InvestmentIncentives:The
Case of Kenya, 1979-1980," in S. Edwards and L.Ahamed eds. Economic
Adjustment and Exchange Rates in Developing Countries, N.B.E.R. The
Universityof Chicago Press.
Buffie, E. (1986): "Devaluation,Investmentand Growth in LDC's," Journal of
DevelopmentEconomcLs,Vol. 16. North-Holland.
Caballero,R. and V. Corbo (1988): "RealExchangeRate Uncertaintyand Exports:
Multi-country Empirical Evidence," Working Paper 414, Department of
Economics,ColumbiaUniversity.
Calomiris, C.H. and R.G. Hubbard (1989): "Price flexibility, credit
availability,and econoAicfluctuations:evidencefrom the United States,
1894-1909,"Quarterly .urnalof Economics,(Forthcoming).

- 39

Chenery,H. and M. Bruno (1962): 'DevelopmentAlternativesin an Open Economy:

The Case of Israel"EconomicJournal 72.
Dailami,M. and M. Walton (1989):"Private Investment,GovernmentPolicy
and ForeignCapital in Zimbabwe"PPR Working Paper# 248 The World Bank
Dixit, A. (1987): "Entry and Exit Decisions of Firms under FluctuatingReal
ExchangeRates,"mimeo, PrincetonUniversity.
Dornbusch, R. (1985): "Overborrowing:Three Case Studies," in International
Debt and the Developing Countries, G. Smith and J. Cuddington eds.
Washington,D.C.: World Bank.
(1987): "Open Economy Macroeconomics: New
Working Paper 2372.

Directions," N.B.E.R.

(1988): "Notes on Credibility and Stabilization," N.B.E.R. Working

_______

Paper 2790.
(1989): "From Stabilization to Growth", mimeo, MIT.
Easterly, W. (1989): "Portfolio Effects in a CGE Model: Devaluation in a
DollarizedEconomy" in L. Taylor (ED.), StructuralistComputableGeneral
EquilibriumModels: SociallyRelevantPolicy Analysis for The Third World
The MIT Press (Forthcoming).
Edwards, S. (1988): "Stabilization, Macroeconomic Policy and
Liberalization,"
mimeo.

Trade

Fazzari,S., R.G. Hubbard and B.C. Petersen(1988a):"Financingconstraintsand

Corporate Investment,"BrookingsPapers on EconomicActivity,Vol. 1.
(1988b): "Investment,Financing Decisions, and Tax Policy," American
EconomicReview.
___

Hall, R. (1977):"Investment,InterestRates, and the Effects of Stabilization

Policies,"Brookings Papers on Economic Activity,
Vol. 8, 7.
and D.W. Jorgenson (1971): "Application of the Theory of Optimum

Capital Accumuilation,"
in G. Frommm, ed. Tax Incentives and Capital
Spending,The BrookingsInstitution,Washington,D.C.
Hayashi, F. (1982): 'Tobin's Marginal q and Average q: A
Interpretation,"Econometrica,Vol. 50, 1.

Neoclassical

Ingersoll, J. and S. Ross (1988): "Waiting to Invest: Investment and

Uncertainty,"mimeo, Yale University.
Jorgenson,D.W. (1967): 'The Theory of InvestmentBehavior,"in R. Ferber ed.
Determinantsof InvestmentBehavior,N.B.E.R.

- 40 -

Kiyotaki, (1988): " MultipleExpectationalEquilibriaunder Monopolistic

Competition",QuarterlyJournal of Economics 103, p.696-713.
Krugman,P. (1988):ExchangeRate Instability.
and R. Baldwin (1987): "The Persister.ceof the U.S. Trade Deficit,"
Brookings Papers on EconomicActivity 1.

Devaluationin Developing
Lizondo,J.S. and P.Montiel (1989): "Contractionary
Countries:An AnalyticalOverview,"I.M.F. Staff Papers.
MacKie-Mason,J. (1989): "Do Firms Care Who Provides Their Financing'"mimeo
Universityof Michigan.
Majd, S. and R. Pindyck (1987): "Time to Build, Option Value, and Investment
Decisions,"Journal of FinancialEconomics 18, pp. 7-27.
Malinvaud, E. (1980): Profitabilityand Unemployment,Cambridge University
Press.
(1982): "Wages and Unemployment,"

Economic Journal 92.

Mayer, C. (1989): "Financial Systems, Corporate Finance and

Development,"mimeo, City UniversityBusinessSchool.

Economic

McDonald,R. and D. Siegel (1986): "The Value of Waiting to Invest",Quarterly

Journal of Economics101, pp. 707-28.
Musalem,A. (1989):
Neary P. and J. Stiglitz (1983): "Towards a Reconstruction of Keynesian
Economics:Expectationsand ConstrainedEquilibria,"QuarterlyJournal of
Economics.
Pindyck, R. (1986): "Capital Risk and Models of Investment Behavior," MIT
Working Paper, 1819-86.
(1988a): "Options,Flexibility and Investment Decisions," MIT Center
for Energy Policy ResearchWorking Paper: 88-018.
(1988b): "IrreversibleInvestment, Capacity Choice, and the Value of
the Firm," AmericanEconomicReview 78, pp. 969-85.
(1989): "Irreversibility,
Uncertainty, and Investment," mimeo.

Precious,M. (.985): wDemandConstraints,RationalExpectations,and Investment

Theory,"Oxford EconomicPapers.

41 -

Rama, M. (1987):'Inversi6nPrivada:Teoria Q y Enfoquede Desequilibrio,'7th

Latin American Meeting of Econometric Society: Abstracts and Papers,
Vol. 4, Sao-Paulo.
Risager, 0. (1984): 'Devaluation,Profitabilityand Investment:A Model with
AnticipatedWage Adjustment,'Seminar Working Paper 1287, Institute for
InternationalEconomicStudies,Universityof Stockholm.
Rosensweig,J. and L. Taylor (1989)s 'Devaluation,Capital Flows and Crowding
Out: A CGE Model with PortfolioChoice for ThaiYand,'in L. Taylor (ed.)
StructuralistComputable General Equilibrium Modelst Socially Relevant
Policiesfor the Third World. Cambridge:MIT Press.
Sachs, J. (1988): 'The Debt Overhang of Developing Countries,* in Debt,
Stabilization,and DevelopmentsEssays in Memorr of Carlos Diaz Alejandro.
Oxford: Basil Blackwell.
Schwartzmann,A. (1985): 'ExchangeRates and the Tax Structurein a 'q' Theory
of Investment,'unpublishedPh.D. dissertation,MIT.
Serven,L. (1986): 'The Effects of Devaluation,"mimeo, M.I.T.
Shapiro, M. (1986): 'Investment,Output and the Cost of Capital," Brookings
Papers on EconomicActivity, 1.
Shleifer, A. and R. Vishny (1989): *The Efficiency of Investment in the
Presenceof AggregateDemand Spillovers',Journalof PoliticalEconomy 96.
Stiglitz,J. and A. Weiss (1981):'Credit Rationing in Markets with Imperfect
Information
'
American Economic Review, 71 June
Sneessens,
H. (1987): 'Investment
and the Inflation-Unemployment
Trade-Off in
a Macroeconomic Rationing Model with MonopolisticCompetition,"European
Economic Review 31.
Solimano, A. (1989):
'Investment
Behavior and Macroeconomic Adjustment:
An
Empirical Analysis for Chile," PPR Working Paper (forthcoming),
World
Bank.
(1986): 'Contractionary
Devaluation in the Southern Cone: The case of
Chile," Journal of DevelopingEconomics23.
___

_______

(1988): 'La Deuda Externa y el problema de la Inversi6n en America

Latina: Situaci6nactual y opciones,'mimeo, Santiago-Chile.

Tornell,A. (1989): 'Real vs. Financial Investment:Towards an Explanationof

'Short-Termism',"
miamo, ColumbiaUniversity.

- 42 -

Tobin, J. (1969): 'A General EquilibriumApproachto MonetaryTheory,* Journal

of Money, Credit and Banking,Vol.l.
(1982): 'Honey and Finance in the MacroeconomicProcess,n Journal of
Money, Credit and Banking,Nobel Lecture,Vol. 14, 2.
___

Van Wijnbergen, S. (1985): 'Trade Reform,AggregateInvestmentand Capital

flight. On credibilityand the value of information,"Economic Letters
17.
(1983a): 'Credit Policy, Inflation and Growth in a
RepressedEconomy,"Journal of DevelopmentEconomics,13.

Financially

(1983b): 'Interest Rate Management in LDC's,' Journal of Monetary

Economics,12.
(1982): 'Stagflationary
Effects of Monetary Stabilization Policies:
A QuantitativeAnalysis'Journal of DevelopmentEconomics.

World Bank (1988):World DevelopmentReport 1988.

privinv.doc

PPR WorkingPaper Series

Contact
iwa

Author

Diu

for paper

WPS323 The Old and the New in Heterodox MiguelA. Kiguel

StabilizationPrograms: Lessons
NissanLiviatan
from the Sixtiesand the Eighties
WPS324 EthicalApproachesto Family
Planningin Africa

F.T. Sai
K Newman

December1989

S. Ainsworth
31091

WPS325 Manufacturers'Responsesto Infra- KyuSik Lee

structureDeficienciesin Nigeria
Alex Anas

December1989

L. Victorio
31015

WPS326 Do ExportersGainfrom VERs?

Jaimede Melo
L AlanWinters

WPS327 MakingNoisyDataSing: Estimating JamesR. Tybout

l,oduction Technologiesin
DevelopingCountries
WPS328 Europe,MiddleEast,and North
Africa (EMN)RegionPopulation
Projections,1939-90Edition

RodolfoA. Bulatao
EduardBos
PatienceW. Stephens
My T. Vu

November1989

S. Ainsworth
31091

WPS329 LatinAmericaand the Caribbean

(LAC) Region Population
Projections,1989-90Edition

RodolfoA. Bulatao
EduardBos
PatienceW. Stephens
My T. Vu

November1989

S. Ainsworth
31091

WPS330 Africa RegionPopulation

Projections,1989-90Edition

RodolfoA. Bulatao
EduardBos
PatienceW. Stephens
My T. Vu

November1989

S. Ainsworth
31091

WPS331 Asia Region PopulationProjections, RodolfoA. Bulatao

1989-90Edition
EduardBos
PatienceW. Stephens
My T. Vu

November1989

S. Ainsworth
31091

WPS332 India: EffectiveIncentivesin

Agriculture: Cotton, Groundnuts,
Wheat, and Rice

AshokGulhati

WPS333 SecondaryMarket Pricesunder

Stiin Claessens
AlternativeDebt Reduction
Swedervan Wijnbergen
Strategies: An Option Pricing
Approachwith an Applicationto Mexico
WPS334 Built-inTax Elasticityand the
JaberEhdaie
RevenueImpactof Discretionary
Tax Measures: An Econometric
EstimationMethod with Application
to Malawiand Mauritius

PPR WorkingPaper Series

iwl
Author

WPS335 MacroeconomicAdjustmentand
Labor MarketStructurein Four
Latin AmericanCountries:An
EconometricStudy

RamonE. Lopez
LuisA. Riveros

WPS336 ComplementaryApproachesto
FinancingHealthServicesin
Africa

GermanoMwabu

WPS337 ProjectingMortalityfor All

Countries

RodolfoA. Bulatao
EduardBos

WPS338 Supplyand Useof EssentialDrugs

in Sub-SaharanAfrica: Issuesand
PossibleSolutions

S. D. Foster

WPS339 Private Investmentand MacroeconomicAdjustment:An

Overview

Luis Serven
AndresSolimano

DbWror

Contact
pager

December1989

S. Ainsworth
31091

December1989

E. Khine
61763

WPS340 PrudentialRegulationand Banking VincentP. Polizatto

Supervision: Buildingan Institutional
Frameworkfor Banks

WDR Office
31393

WPS341 Cost-of-LivingDifferencesbetween MartinRavallion

Urbanand RuralAreasof Indonesia Dominiquevan de Walle

C. Spooner
30464

WPS342 HumanCapitaland Endogenous

PatricioArrau
Growthin a Large ScaleLife-Cycle
Model
WPS343 Policy Determinantsof Growth:
Surveyof Theoryand Evidence

WilliamR. Easterly
DeborahL. Wetzel

WPS344 Policy Distortions,Size of

Government,and Growth

WilliamEasterly

WPS345 Essayson PrivateTransfersand

Public Policy in Developing
Countries: A Surveyand Case
Study for Peru

Donald Cox
EmmanuelJimenez

WPS346 Indiia. The GrowingConflict

betweenTradeand Transport:
Issuesand Options

HansJurgen Peters

WPS347 HousingFinancein Developing

Countries: A TransactionCost
Approach

RobertM. Buckley

R. Luz
61588