The macroeconomic data discussed in Section II suggest that saving, investment, and growth rates are positively linked in the developing countries. This section examines in more detail the relationship among these variables, starting with the relationship between saving and investment. It then considers the evidence on the effects of saving and investment on growth and ends with a brief discussion of the direction of causation among these variables.
The Saving-Investment Relationship
As discussed in Section IV of Part Two, if capital were perfectly mobile among countries, changes in domestic investment would be independent of changes in national saving. Thus, the importance of national saving for investment depends on the degree of international capital mobility.32
Table 4 suggests a close association between national saving and domestic investment in most groups of developing countries—the five regions and six groups classified according to problems related to debt servicing, inflation, and income level. The correlation between levels of national saving and domestic investment is quite close in the periods before and after the debt crisis for practically all the country groupings. In the majority of the groups, the correlation, as expected, turns out to be stronger in the post-debt-crisis period. It weakens substantially after 1982 only among country groups that have retained relatively favorable access to international capital markets—that is, countries in Asia and countries that have avoided debt-servicing problems.
Saving-Investment Correlations in Developing Countries
Saving-Investment Correlations in Developing Countries
| 1976–88 | 1976–81 | 1982–88 | |
|---|---|---|---|
| Developing countries | 0.68 | 0.60 | 0.70 |
| Countries with debt-servicing problems | 0.73 | 0.72 | 0.74 |
| Countries without debt-servicing problems | 0.88 | 0.84 | 0.71 |
| High-inflation countries | 0.80 | 0.78 | 0.76 |
| Low-inflation countries | 0.63 | 0.51 | 0.65 |
| High-income countries | 0.55 | 0.44 | 0.61 |
| Low-income countries | 0.78 | 0.69 | 0.76 |
| Africa | 0.74 | 0.66 | 0.75 |
| Asia | 0.84 | 0.91 | 0.57 |
| Europe | 0.88 | 0.80 | 0.94 |
| Middle East | 0.31 | 0.11 | 0.44 |
| Western Hemisphere | 0.63 | 0.71 | 0.65 |
Saving-Investment Correlations in Developing Countries
| 1976–88 | 1976–81 | 1982–88 | |
|---|---|---|---|
| Developing countries | 0.68 | 0.60 | 0.70 |
| Countries with debt-servicing problems | 0.73 | 0.72 | 0.74 |
| Countries without debt-servicing problems | 0.88 | 0.84 | 0.71 |
| High-inflation countries | 0.80 | 0.78 | 0.76 |
| Low-inflation countries | 0.63 | 0.51 | 0.65 |
| High-income countries | 0.55 | 0.44 | 0.61 |
| Low-income countries | 0.78 | 0.69 | 0.76 |
| Africa | 0.74 | 0.66 | 0.75 |
| Asia | 0.84 | 0.91 | 0.57 |
| Europe | 0.88 | 0.80 | 0.94 |
| Middle East | 0.31 | 0.11 | 0.44 |
| Western Hemisphere | 0.63 | 0.71 | 0.65 |
Effects of Saving and Investment on Growth
The downward trends in saving and investment have been accompanied by a slowdown of economic growth in developing countries. This slowdown has primarily reflected sharply lower growth in the countries that encountered debt-servicing difficulties; the countries that did not encounter such difficulties actually experienced some acceleration in output growth (see Table 2).
A Fund staff study has examined the relationship between investment and growth for 125 capital importing developing countries.33 The study found this relationship to be positive and statistically significant for all country categories, with a 10 percentage point increase in the ratio of investment to GDP, raising the growth rate of output by 1½ percentage points on average—a result that is broadly confirmed by other studies. A separate study (see International Monetary Fund, 1988a, Chapter IV) found that the contribution of investment to growth appears to have varied widely across regions (see Table 5). Overall, research on the growth process in developing countries suggests that, while investment has indeed made a significant contribution to growth, other factors such as the accumulation of human capital as well as improvements in the quality of resources and the efficiency with which they are used, also have been very important; see Otani and Villanueva (1988 and 1989) as well as Psacharopoulos (1988). The importance of total factor productivity and of human capital formation in empirical studies of the growth process in developing countries suggests that the endogenous growth literature discussed in Section III of Part Two may be as relevant to developing countries as to industrial countries.
Capital Importing Developing Countries: Contribution of Capital to Growth, 1974–87
(In percent)
Capital Importing Developing Countries: Contribution of Capital to Growth, 1974–87
(In percent)
| Contribution of Capital | |||
|---|---|---|---|
| 1974–81 | 1982–87 | ||
| All capital importing countries | 53 | 56 | |
| By region | |||
| Africa | 50 | 45 | |
| Asia | 57 | 50 | |
| Europe | 61 | 90 | |
| Non-oil Middle East | 33 | 38 | |
| Western Hemisphere | 88 | 83 | |
| By financial criteria | |||
| Countries with recent debt-servicing problems | 86 | 89 | |
| Countries without recent debt-servicing problems | 41 | 44 | |
Capital Importing Developing Countries: Contribution of Capital to Growth, 1974–87
(In percent)
| Contribution of Capital | |||
|---|---|---|---|
| 1974–81 | 1982–87 | ||
| All capital importing countries | 53 | 56 | |
| By region | |||
| Africa | 50 | 45 | |
| Asia | 57 | 50 | |
| Europe | 61 | 90 | |
| Non-oil Middle East | 33 | 38 | |
| Western Hemisphere | 88 | 83 | |
| By financial criteria | |||
| Countries with recent debt-servicing problems | 86 | 89 | |
| Countries without recent debt-servicing problems | 41 | 44 | |
Since growth depends critically on investment (broadly defined to include human capital), and resources for investment in developing countries are derived primarily from national saving, the latter is often seen as a key determinant of economic growth. Table 6 indicates that the saving rate and investment in human capital are indeed closely linked to economic growth.34 The findings for the three country groups classified by per capita income level provide insights into the contribution of saving and human capital formation to economic growth. For example, the saving rate was found to have contributed to growth significantly more in the high-income and middle-income groups than in the low-income group. Investment in human capital contributed to growth significantly more in low-income and middle-income countries.
Factors Determining Growth in Fifty-Five Developing Countries, 1970–851
(In percent)
Figures are expressed as annual averages.
Ratio of national saving to GNP.
Ratio of government educational expenditure to GNP.
In real terms.
Nominal interest rate on external debt (denominated in U.S. dollars) less the percentage change in export prices (in U.S. dollars).
Factors Determining Growth in Fifty-Five Developing Countries, 1970–851
(In percent)
| Portion Contributed by | |||||||
|---|---|---|---|---|---|---|---|
| Growth Rate of Per Capita Real GNP | Domestic saving rate2 | Rate of investment in human capital3 | Growth rate of exports4 | Growth rate of population | Real interest rate on external debt5 | Other factors | |
| All sample countries | 2.7 | 2.3 | 0.7 | 2.6 | –2.0 | 0.1 | –1.0 |
| High-income countries | 4.0 | 3.7 | 0.3 | 3.5 | –1.3 | 0.1 | –2.3 |
| Middle-income countries | 2.0 | 7.5 | 2.3 | 0.6 | –3.6 | 0.5 | –5.3 |
| Low-income countries | 1.4 | — | 1.8 | 2.6 | –3.1 | 0.4 | –0.3 |
Figures are expressed as annual averages.
Ratio of national saving to GNP.
Ratio of government educational expenditure to GNP.
In real terms.
Nominal interest rate on external debt (denominated in U.S. dollars) less the percentage change in export prices (in U.S. dollars).
Factors Determining Growth in Fifty-Five Developing Countries, 1970–851
(In percent)
| Portion Contributed by | |||||||
|---|---|---|---|---|---|---|---|
| Growth Rate of Per Capita Real GNP | Domestic saving rate2 | Rate of investment in human capital3 | Growth rate of exports4 | Growth rate of population | Real interest rate on external debt5 | Other factors | |
| All sample countries | 2.7 | 2.3 | 0.7 | 2.6 | –2.0 | 0.1 | –1.0 |
| High-income countries | 4.0 | 3.7 | 0.3 | 3.5 | –1.3 | 0.1 | –2.3 |
| Middle-income countries | 2.0 | 7.5 | 2.3 | 0.6 | –3.6 | 0.5 | –5.3 |
| Low-income countries | 1.4 | — | 1.8 | 2.6 | –3.1 | 0.4 | –0.3 |
Figures are expressed as annual averages.
Ratio of national saving to GNP.
Ratio of government educational expenditure to GNP.
In real terms.
Nominal interest rate on external debt (denominated in U.S. dollars) less the percentage change in export prices (in U.S. dollars).
Direction of Causation Among Saving, Investment, and Growth
The previous two subsections have provided evidence of a close correlation among changes in saving, investment, and growth rates in developing countries. The direction of causation among the recent changes in these variables, however, is not revealed by such correlations, since movements in any one of these variables may have caused the corresponding changes in the others, or all three may have been driven by the same exogenous factor. More concretely, however, several macroeconomic hypotheses may be consistent with the recent data:
● The reduction in growth may have been caused by a decrease in domestic investment prompted by a shortfall in national saving. This saving shortfall may itself have arisen because of an attempt to sustain domestic consumption in the face of an exogenous increase in external interest payments, or because of an exogenous increase in public consumption, or because of inappropriate domestic financial policies that resulted in negative domestic real interest rates. Together with reduced availability of foreign saving after 1982, the lower national saving rates would have contributed to a reduction in investment and thus in economic growth.
● Alternatively, a contraction in investment may have been the driving factor. Domestic residents may have switched their portfolio toward the acquisition of foreign assets instead of domestic capital, owing to a higher perceived risk of capital losses on domestic investment—brought about by the debt overhang, by adverse external developments regarding interest rates and the terms of trade, or by inadequate domestic macroeconomic policies. The resulting capital flight and lower private investment may have, in turn, reduced lending by external creditors. Public investment may also have fallen in order to finance higher external interest payments without curtailing public consumption or raising additional revenues. The result of reduced investment would be lower aggregate demand and slower expansion of capacity, both of which result in lower output growth. If this situation were to be perceived as temporary by the private sector, private consumption would tend to be sustained, thus reducing private saving. Meanwhile, if public consumption continued at previous levels, public saving would fall, since government revenues would decline with reduced levels of economic activity.
● Finally, lower output growth may have triggered reductions in saving and investment rates. Faced with severe foreign exchange constraints resulting from reserve depletion as well as adverse external developments, countries may have adopted restrictive demand management policies and exchange rate adjustments, both of which may have had contractionary effects on output in the short run. In many cases this effect would have been aggravated by import restrictions and foreign exchange rationing, creating supply bottlenecks. The contraction of demand, together with the supply restrictions, would have adversely affected domestic investment incentives, thus reducing medium-term growth through slower expansion of capacity. Private saving would also have fallen if depressed output levels were expected to recover within a reasonable period of time.
These alternative hypotheses are by no means mutually exclusive, and their relative roles may have been quite different in different countries. Assessing their respective historical roles would, therefore, require detailed country-specific analysis. The existence of alternative interpretations of past experience serves as a reminder that, in a general equilibrium setting, the interactions among saving, investment, and growth are quite complex. Nevertheless, the formulation of policies to promote economic growth in the context of adjustment programs requires taking these general equilibrium interactions into account.