Chapter

V Private Investment

Author(s):
Kalpana Kochhar, Erik Offerdal, Louis Dicks-Mireaux, Mauro Mecagni, Jianping Zhou, Balázs Horváth, David Goldsbrough, and Sharmini Coorey
Published Date:
August 1996
Share
  • ShareShare
Show Summary Details

The cross-country evidence discussed in Section III confirmed that the absence of a stable environment for economic decision making is an important impediment to private investment. Yet progress toward macroeconomic stability has sometimes proved insufficient to generate a rapid recovery of investment. Lags, at times long and protracted, in the response of private investment have raised concerns on possible links between the design of adjustment policies and the duration of these lags. The rate of domestic saving may also have influenced the pace of the recovery in investment, especially when access to international capital markets was limited.35

The experience of the eight countries reveals marked differences in the response of private investment to adjustment policies—in terms of the steepness of the initial decline, whether there was an identifiable period of “pause” before recovery began,36 and the magnitude of the observed recovery (see Table 4 and Chart 7). The record of falling private investment rates in the early years of adjustment ranged from a precipitous drop in Chile during the 1982-83 crisis (following an unsustainable boom fueled by expectations of a devaluation) to smaller, but still substantial, declines observed in India, Mexico, Morocco, and Thailand. Comparison of program targets with actual outcomes suggests that, on average, targets under IMF-supported programs for the eight countries were generally too sanguine about the prospects for an early rebound in private investment (Box 2).

A marked investment pause—lasting three to four years—occurred in two countries: Mexico during the first adjustment period and Morocco in 1985-88. In Ghana, private investment rose when the Economic Recovery Program was introduced, although the initial rebound suffered a brief hiccup in 1986. Private investment in Bangladesh and Senegal was persistently flat with minor fluctuations around low rates in the range of 5 percent to 10 percent of GDP. For these two countries, the record points to a protracted stagnation, rather than a simple postadjustment pause of investment, suggesting that structural impediments were especially important (see Section VIII).37

As for the strength of the subsequent recovery, when it existed three groups can be identified: (1) a weak, uneven recovery in Ghana and Morocco; (2) a delayed, but significant recovery in Mexico to investment rates around 15 percent of GDP; and (3) a stronger, more rapid recovery to investment rates in excess of 20 percent of GDP (in Chile) or 30 percent of GDP (in Thailand).

With this background, three questions are addressed in this section. First, does the gradual response of investment observed in many of the countries come as a surprise or should a lag be expected on the basis of rational economic behavior? Second, what is the empirical evidence on the role of public policies in influencing the observed pattern of private investment? Third, how can the lag in the investment response be minimized and the recovery accelerated?

Factors Affecting the Lagged Response of Private Investment

In general, a certain lag in the response of private investment during or after a period of adjustment should be expected based on the theory of investment behavior. Such a lag could reflect the rational response of private investors to adjustment costs or to uncertainty about the permanence of changes in macroeconomic and structural policies, as well as the more conventional effects stemming from demand management policies.

The considerations emerging from investment studies (Box 3) suggest that the perceived credibility of public policies is important in orienting private investors’ decisions. Indeed, if investors perceive that relative prices are highly uncertain, and policy measures poorly coordinated or likely to be reversed, they may delay committing resources until a more positive assessment is possible. Also, since firms have incomplete information and may in practice gauge investment prospects partly by observing competitors’ and partners’ behavior, the economy may get locked into a low-investment and low-growth equilibrium if a sufficient number of firms postpone investment. In contrast, policies perceived as consistent and unlikely to be reversed may turn expectations around and induce a positive response of private investors, thereby reducing the costs of adjustment.

Box 2.Investment: Program Targets and Outcomes

IMF-supported programs for the eight countries have, on average, experienced only small shortfalls between targeted and actual investment rates, with most of the shortfall occurring in the early program years.1 However, targets for private investment in the initial program year appear on average to have been too optimistic.

In the initial program year, the ratio of total investment to GDP was targeted, on average, to rise slightly, whereas the actual outcome was a small decline; the average shortfall was 0.4 percentage point of GDP (see chart). The shortfall for private investment was almost 1 percentage point of GDP. On average, programs targeted a pickup in private investment in the initial year, whereas actual investment changed little, suggesting that programs for the eight countries typically took insufficient account of the tendency for private investment to respond to adjustment with a lag. The shortfall was partly offset by public investment, which declined by less than targeted.2 Differences between targets and actuals for the second and third program years were generally smaller, with the main shortfalls occurring in public investment.

Change in Investment Rates: Program Targets and Outcomes

(Ratio to GDP, in percentage points; t denotes first program year)

Source: IMF staff estimates.

1 The sample covers initial targets in programs supported by stand-by arrangements under the upper credit tranches, the structural adjustment facility, and the enhanced structural adjustment facility. In contrast, for the 36 countries covered in the last review of IMF conditionality (Schadler and others (1995)) there was, on average, no shortfall in investment for programs covered by that review.2 Since not all programs contain a breakdown between targets for public and private investment, differences in sample coverage mean the components do not add up to the total.

Determinants of Investment and the Role of Public Policies

In order to examine the relative importance of these effects in the eight countries, econometric estimates of the determinants of investment combined traditional neoclassical influences with factors specific to developing countries and indicators of uncertainty and macrostability.38 The results are summarized in Table 8 and presented in more detail in Appendix IV. The impact of individual factors on the decline, pause, or recovery of private investment varies. Table 9 provides a summary of the salient influences behind major investment episodes in each country, drawing upon both the econometric and other evidence in the country case studies. Several broad conclusions emerge.

Table 8.Selected Factors Influencing Private Investment1
Explanatory Variables:
Financial variablesMeasures of uncertainty/ macroeconomic instability
CountrySample PeriodEndogenous Variable: Share of Investment in GDPAggregate activity (Lagged growth of real GDP)Real interest rate/ relative priceReal growth Of credit to private sectorPublic Investment (Share in GDP)Inflation rate/parallel market premium/ real effective exchange rateExternal debt/debt serviceSample volatility measures (or other)External shocks
Bangladesh1979/80-1993/94Private, constant pricesNS−−−+NS−−−−−−
Chile1975–93Nongovernment, constant prices+++−−−NS−−−NS−−−NSNS
Ghana1971–93Private, current prices+++−−NS−−−−−−NSNS++
India1973/74-1993/94Private, constant prices+++2_+−−−−−−
Mexico1972–93Private, current prices+++++−−−NSNSNS
1981–93Private, constant prices++−−−++−−−NS
Morocco1972–93Private, constant prices+−−++++−−−NSNS++34
Senegal1978–93Nongovernment, constant prices+NSNS−−−−−−−−5Š
Thailand1970–93Private, constant prices+++6Š−−−−−−NSNSNS
Source: Appendix IV, Table 18.

Sign of the estimated coefficient: +,– = not statistically significant at 5 percent level, but significant at a marginally higher level; ++,–– = statistically significant at 5 percent; +++,––– = statistically significant at 1 percent; NS indicates not statistically significant and not retained in final preferred specification; blank cells indicate variables omitted in individual countries.

Lagged growth in industrial production.

Export price shock (in percent of GDP).

Import price shock (in percent of GDP).

Government domestic borrowing requirement including involuntary accumulation of arrears (in percent of GDP).

Real GDP growth in partner countries.

Source: Appendix IV, Table 18.

Sign of the estimated coefficient: +,– = not statistically significant at 5 percent level, but significant at a marginally higher level; ++,–– = statistically significant at 5 percent; +++,––– = statistically significant at 1 percent; NS indicates not statistically significant and not retained in final preferred specification; blank cells indicate variables omitted in individual countries.

Lagged growth in industrial production.

Export price shock (in percent of GDP).

Import price shock (in percent of GDP).

Government domestic borrowing requirement including involuntary accumulation of arrears (in percent of GDP).

Real GDP growth in partner countries.

Table 9.Summary of Main Factors Explaining Major Episodes of Private Investment Performance1
CountryInvestment Performance (Corresponding period)Main Reasons
BangladeshProlonged stagnation and weak recovery (1980–93)Weak financial sector and large nonperforming loans leading to high interest rate spreads
Administrative and regulatory impediments
Macroeconomic policy uncertainty in early 1980s
Chile2Sharp decline (1981–83)Severe recession associated with financial crisis and external shocks
Weak financial sector balance sheets
Strong recovery (1983–93)Sustained recovery in economic activity
Substantial improvements in external debt-service profile
Successful structural reforms
GhanaUneven, weak recovery (1983–91)Persistent macroeconomic policy uncertainty
Weak financial sector balance sheets
Crowding-out effects of public investment
IndiaOverall flat trend followed by decline (1990/91–1992/93)Cost and availability of credit
Crowding-out effects of public investment prior to 1990/91
Slowdown of industrial production beginning in 1990/91
Uncertainty associated with greater volatility of inflation beginning in 1990/91
Continued regulatory impediments despite substantial reform
MexicoDecline (1981–83)Recession associated with financial crisis and external shocks
Severe erosion of private sector confidence in public policies
Sharp contraction of credit to the private sector
Pause (1983–87)Persistent macroeconomic instability (high inflation)
Stagnating economic activity
Uncertainty due to unresolved external debt situation and weak private sector confidence
Moderate recovery (1987–93)Progress in macroeconomic stability and disinflation
Progress in private sector access to bank credit through financial reform
Confidence effects from resolution of external debt problem (acting through interest rates, although these remained high for part of the period)
MoroccoDecline (1982–85)Reduction of complementary public investment
Rising real interest rates and higher cost of imported capital goods
Pause (1985–88) and weak, uneven recovery (1989–93)Reduction of complementary public investment Rising real interest rates
Slow growth of economic activity (supply shocks)
Higher costs of imported capital goods
Senegal2Flat trendWeak external competitiveness
Rising external debt in early 1980s
Crowding-out effects of public investment
Persistence of structural distortions
ThailandStrong recovery (1986–93)Decline in relative unit labor costs: and strong demand in partner countries
Macroeconomic stability
Pro-business orientation of policies
Source: IMF staff estimates.

Based on econometric estimates and additional evidence from country case studies.

Nongovernment investment.

Source: IMF staff estimates.

Based on econometric estimates and additional evidence from country case studies.

Nongovernment investment.

Box 3.What Does the Literature Say About Lags in Investment?

Studies of private investment behavior emphasize two complementary factors—the existence of costs to adjusting the capital stock and the role of uncertainty and irreversibility—in order to explain the lagged response of investment to changes in policies and incentives. These factors suggest caution in assuming a strong response of private capital formation in the early stages of an adjustment program.

If costs related to the purchase and installation of capital increase with the investment rate, they create incentives for a gradual transition from actual to desired levels of capital.1 However, these costs can be influenced by policies. In particular, the transition period may be unduly protracted by credit-rationing constraints and structural problems of financial intermediaries; by the large weight of the public sector, often hampering private initiative with distortions and regulations; or by constraints on access to foreign exchange or financing for imported capital goods.

The timing of private investment decisions may also be affected by uncertainty and irreversibility effects.2 An increase in uncertainty causes risk-averse firms to reallocate resources away from risky activities, thereby lowering the desired stock of capital.3 Also, if the outcomes of irreversible investments turn out to be worse than expected, risk-neutral investors may get stuck with low returns, whereas if prospects improve new entrants may compete away part of the gains. Consequently, downside risk may increase without corresponding upside gain, so that waiting may have value as it gives firms the opportunity to process new information before committing resources.4 The role of these factors however is hard to test empirically because the available indicators—such as measures of volatility of inflation or other relevant variables—may only partially capture the impact of policy uncertainty.

1 Costs related to the purchase, delivery, and installation of new capital goods play a central role in the dynamic response of investment in both the neoclassical theory and in Tobin’s q theory of investment.2 Irreversibility arises from sunk costs related to firm-specific investment and the large discount in valuing secondhand equipment.3 Inflation uncertainty may also lower investment by obscuring price signals, and credit rationing may intensify as uncertainty increases.4 See Dixit (1992)), Dixit and Pindyck (1994), and Pindyck (1991).

Effects from lagged economic activity played an important role in most countries except Bangladesh. In Chile, Ghana, India, Mexico, and Thailand39 these effects were strongly significant, less so in Morocco and Senegal. This implies that any contractionary effects of adjustment policies on output would have a short-run impact on private investment. However, this variable may also act as a channel of transmission for confidence effects and private sector expectations, which can be influenced, for instance, by protracted financial imbalances and measures affecting perceptions of the pro-business orientation of policies. Such influences appear to have been especially important in Mexico and Chile prior to and during the debt crisis.

High real interest rates and constraints on the growth of credit to the private sector contributed in several instances to a decline or slowed the recovery of investment (Bangladesh, India, Mexico, and Thailand).40 As discussed in the preceding section, both the effects of the mix of monetary and fiscal policies, as well as the impact of structural problems in the financial sector, were contributing factors in several cases.41 The results suggest that greater efforts to improve competition and efficiency of financial intermediaries and to ensure an adequate flow of credit to the private sector could have yielded a stronger investment response. There were, however, episodes where sharply increasing real rates did not prevent investment from rising. For example, in Chile the private sector spending spree on imported consumption and investment goods prior to the 1982 crisis occurred in the context of a sharp rise in real lending rates.42

Uncertainty and macroeconomic instability—captured by a variety of indicators in individual country estimates—exerted adverse and strongly significant effects on private capital formation. These factors appear to have been especially important in Mexico, where persistent, high inflation was an important deterrent to real private investment during the 1983-87 pause; in Ghana, where parallel market premiums captured the lingering impact of foreign exchange controls and slippages in financial policies induced the 1992-93 investment decline; and in Senegal, where inflation and rising external debt contributed to hold back nongovernment investment in the early 1980s.

However, there are also indications that broader confidence factors associated with internal and external financial crises had a major influence on the turning points of the investment cycle. In Mexico, the depth and speed of the investment decline in the aftermath of the 1982 crisis were sharper than predicted by the estimated investment equation and probably reflected the erosion of confidence in public policies due to the partial default on domestic debt, the imposition of exchange controls, and the nationalization of the banking system. In Chile, the sharp decline of nongovernment investment in 1981-83 was also linked to a severe financial crisis. In this case, major problems were the large external debt burden, the perceived inconsistency between exchange rate and wage policies, the associated private sector overspending, and the precarious condition of the financial system.

As for the role of public investment, evidence for the eight countries provides little support for the view that private and aggregate public investment are complementary. In six countries, the effect of public investment was found to be negative and strongly significant, implying that public investment competed with and crowded out private sector opportunities. The exceptions were Bangladesh, where no effect was detected, and Morocco where the two were largely complementary.43 In countries where there was evidence of a negative overall relationship between public and private investment, efforts to rationalize and better select public investment programs and associated privatization reforms appear to have been instrumental in promoting, rather than hampering, private investment. However, this is an area where generalizations are difficult, as shown by the different results reported in the empirical literature.44 Also, several qualifications are in order. First, lack of data prevented addressing the important distinction between infrastructural and noninfrastructural public investment.45 While Chile, Mexico, and Thailand made progress in enhancing the role of the private sector in utilities and infrastructural projects, in Thailand the cautious public investment policies resulted in infrastructural bottlenecks by the early 1990s. Second, while the rationalization of public investment programs created opportunities for the private sector, the associated reduction of demand affected private investment indirectly. Third, the efficiency of public investment, and hence its full impact on growth, cannot be deduced from the aggregate relationships discussed here.46

How Can Policies Accelerate the Response of Private Investment?

Experience in the eight countries suggests a number of conclusions:

• Achievement of greater macroeconomic stability was conducive to private investment. Progress in reducing inflation (Mexico and Senegal) and the parallel market premium (Ghana), in keeping the external debt-servicing burden under check (Chile, Bangladesh, and Senegal), in avoiding excessive real exchange rate appreciations and maintaining real wage flexibility (Thailand and Morocco) all helped promote conditions conducive to higher private investment rates. However, the magnitude of each effect varied across countries and over time.

• Uncertainty about the course of policies appears to have penalized private investment. Therefore, avoidance of measures adversely affecting private sector confidence in public policies, clear announcement signals, and sustained implementation of mutually consistent policies are likely to reinforce credibility and reduce lags in the response of investors.

• Except for Morocco, there is no evidence that the reduction of public investment had negative effects on private investment. Rather, the estimates indicate that, in terms of direct contemporaneous links, substitutability effects dominated in most of the countries, suggesting that efforts to rationalize public investment programs create opportunities for private investment. This does not mean specific public investments, especially in infrastructure or human capital development, are not essential to support higher private investment and growth. Nor does it say how squeezing public investments during fiscal adjustment would affect growth; indeed, it is difficult to answer such a question without detailed cost-benefit analyses. However, it does question the wisdom of a general, untargeted increase in public capital spending as a way to elicit a faster private investment response.

• The findings suggest that ensuring an adequate flow of credit to the private sector, including through the removal of credit constraints when they are in place, is likely to promote a faster recovery of investment. Both an inappropriate mix of monetary and fiscal policies and structural problems in the financial sector can reduce private investment. The experience of Bangladesh also suggests that easing financial repression and lifting administrative restrictions may not by themselves be sufficient to ease credit constraints if the latter stem from an inherited stock of nonperforming assets. Furthermore, increasing competition in the banking system and addressing macroeconomic and structural problems that contribute to poorly performing portfolios may reduce the incidence of excessive bank spreads and high real lending rates. The importance of removing structural distortions affecting financial intermediation and the allocation of resources is underscored by the experience of countries with a relatively flat investment record, all of which showed lagging progress in these areas.

    Other Resources Citing This Publication