A quantitative analysis of long-term growth performance, with policy implications for improving human capital, mobilizing domestic savings, and promoting exports
Compared with the record of the post-war period up to 1970, the growth performance of developing countries in the 1970s and the 1980s has been rather disappointing. Investment rates in most of these countries have declined, but they have still been higher than domestic saving rates, making necessary substantial foreign borrowing. As a result, external debt has increased rapidly and external payments positions have become unsustainable. More recently, as an essential element of the adjustment strategy, developing countries have restrained aggregate demand, mainly by reducing public sector expenditures, including those on both physical and human capital. The adverse implications of compressed investment activity have drawn attention from economic and business quarters, national authorities, and international organizations (see, for example, the report of the international Group of Twenty- Four on International Monetary Affairs, “The Role of the IMF in Adjustment with Growth,” IMF Survey, August 10, 1987).
These developments have engendered lively discussions on the appropriate growth- oriented adjustment policies for developing countries, particularly those countries with structural rigidities and heavy external debts. At the same time, calls have been made for the clarification of the relationships among savings, investment, and growth, as well as the identification of other factors contributing to growth. This article discusses some of these issues and presents the results of a new study on long-term growth performance and its major determinants in developing countries (see box).
Framework for analysis
In explaining the process of long-term growth, standard neoclassical growth analysis attaches considerable importance to capital accumulation, domestic savings, and exports. Typically, a higher ratio of capital to labor is associated with a higher rate of domestic savings, and the latter is a major constraint on the former in developing countries. As to the role of exports, it is argued that the export sector serves as a vehicle for technology transfer, through the import of advanced capital goods by that sector. Export expansion also provides a spur to the development of efficient and internationally competitive management and workers, the effects of which reach beyond the export sector. By earning foreign exchange that raises the country’s capacity to service external debt and thus improves its creditworthiness, the expansion of the export sector induces larger inflows of foreign credits that make higher rates of investment possible. Export growth also creates forward and backward linkages to producers of goods and services in the rest of the economy.
Standard growth analysis assumes that the labor force grows at a constant rate, unaffected by economic variables. In the short and medium term, an increase in domestic savings or in exports would finance an expansion of the capital stock and thus lead to a one-time increase in output. Over time, however, the fixed rate of growth in the labor supply would become a bottleneck to further output growth.
Recent theoretical studies using the “new” growth analysis have emphasized the powerful role of the accumulation of human capital. In this modified analysis, the growth rate of output is accounted for by increases in factor inputs and improvements in their efficiency or productivity. These improvements depend largely on society’s outlays on health and various forms of education, to build up human capital. Accordingly, as long as society devotes a portion of its resources to improving the productivity of its labor force, per capita output growth would rise with the increase in labor productivity.
The analytical framework adopted in this article incorporates the roles of investment in human capital, exports, and external debt into the standard neoclassical macroeconomics framework. The approach highlights the relationship between output growth and the availability of financing for investment in the long term. It posits that the long-term growth of per capita real GNP depends on structural parameters of the economy, such as the domestic saving rate, the rate of investment in human capital (approximated as the ratio of government educational expenditure to GNP), the growth rate of exports (in real terms) that is essential to finance necessary imports of capital goods, the real interest rate on external debt, and the rate of population growth. One would expect the first three of these factors to help finance the expansion of the capital stock (both physical and human) and thus to raise the growth of per capita real GNP, whereas the latter two are likely to contribute negatively to per capita real GNP growth. While many of these parameters may be influenced by long-term economic growth, it is beyond the scope of this article to investigate such influence.
This article draws on “Determinants of Long-Term Growth Performance in Developing Countries,” IMF Working Paper (WP/88/97), available from the authors, forthcoming in World Development vol. 18 (1990), Oxford, Pergamon Press. Also see “Theoretical Aspects of Growth in Developing Countries: External Debt Dynamics and the Role of Human Capital,” IMF Staff Papers (June 1989), by the same authors, and references cited therein.
|Portion contributed by|
|All sample countries||2.7||2.3||0.7||2.6||–2.0||0.1||–1.0|
To quantify the contributions of these structural factors to the long-term growth of per capita real GNP in different countries, the study used annual average data for 1970–85 for 55 developing countries. (By and large, tests showed that the equations used were able to capture cross-country variations in the growth of per capita real GNP, with the coefficient of determination being about 0.7.) The strength of the explanatory parameters used for the analysis, coupled with the actual annual averages of the values of these parameters, provides the basis for quantifying the determinants of per capita real GNP growth for the whole sample of countries and for three subgroups of countries at different levels of per capita income (see table).
For the sample as a whole, per capita real GNP grew, on average, at an annual rate of 2.7 percent over the period 1970–85. The most important contributing factor was the expansion of exports; this accounted for more than 95 percent of the average growth rate of per capita real GNP, that is, 2.6 percentage points out of the total per capita real GNP growth. The next most important determinants of GNP growth were the domestic saving rate and the rate of investment in human capital. The real interest rate on external debt, being negative on average over the period studied, contributed slightly positively to the growth of per capita real GNP. The growth rate of population, as expected, contributed negatively. Residual elements, representing factors excluded from the equation, also contributed negatively to economic growth. These elements include, for example, changes in the acreage and productivity of land and changes in working hours.
The finding that exports have a dominant influence on economic growth is not surprising, given the importance of the export sector as outlined above. For an average economy in the sample, the estimated growth effects of the export sector were larger than might have been expected from both the actual rate of export growth (about 7 percent a year) and the sector’s share in total output (about 25 percent).
The domestic saving rate was found to be almost as significant as exports in its positive impact on economic growth. On average, developing countries saved about a fifth of their output, providing a significant portion of the financing for investment, which in turn contributed to the growth of per capita output. The evidence in the table suggests that a once-and-for-all increase of 10 percentage points in the saving rate would raise the average growth rate of per capita real GNP by 1.2 percentage points a year.
Investment in human capital, measured as the ratio of government educational expenditure to GNP, amounted to about 4 percent a year. Its effect on economic growth was less than those of the rate of export growth or the saving rate, but was nevertheless noticeable. If one also took into account private-sector investments in education and the entire economy’s investments in health, nutritional intake, and on-the-job training, the financing of investment in human capital probably would have been a much more significant contributing factor than indicated in the table.
The negative contribution of population growth to the growth of per capita GNP—about 2 percent a year—was somewhat less than the rate of population growth itself. This suggests that, in a typical developing country, a reduction in the capital/labor ratio brought about by an increase in population growth led to only a small decline in per capita real GNP growth. Hence, the marginal productivity of capital was low.
The above aggregate results mask the diversity of experience among countries at different levels of development. The findings for the three income groups provide insights into the factors affecting economic growth in countries with different income levels. For example, the saving rate was found to have contributed to economic growth significantly more in the high-income and middle-income groups than in the low-income group. These findings also suggest that a once-and-for-all increase in the domestic saving rate of 10 percentage points would raise the long-term growth rate of per capita GNP by 1 or 2 percentage points annually in many of the high-income countries, and by as much as 3 to 4 percentage points in many middle-income countries. The saving rate seems to have contributed little to growth in the low-income countries. In many of these countries, the financial system is still rudimentary, so that the saving-investment-growth linkage in the period studied may have been tenuous.
Expenditures on the improvement of human capital contributed noticeably to economic growth in the middle-income and the low-income countries, but hardly at all in the high-income countries. This suggests that the rate of return on such expenditures may be higher in countries which have not entered the high-income or advanced industrial stages. This tendency is well argued by Rostow (see W.W. Rostow, The Stages of Economic Growth: A Non-Communist Manifesto, Cambridge University Press, 1960) and substantiated by empirical studies on the role of education in economic growth (see, for example, George Psacharopoulos, “Education and Development: A Review,” World Bank Research Observer 3, No. 1, January 1988).
Export growth contributed to economic growth substantially more in low-income and high-income countries than in middle-income countries. This may be because, in the early stages of development, growth depends largely on the expansion of primary commodity exports. As low-income countries develop, entering the “take-off stage” defined by Rostow, the development strategy often changes and the pattern of production shifts away from primary products toward more consumer-oriented goods and import-substituting products. Economic growth then tends to depend more on the expansion of production for the domestic market, and the dependence on exports lessens. As the economy nears the later phase of the take-off stage, the domestic market becomes saturated and the need to increase exports becomes imperative, as many newly industrialized economies and some other Asian countries have found. Thus export growth again plays an important role in economic growth in high-income countries.
What do these findings mean for growth- oriented adjustment policies? Economic growth is a complex process involving social, cultural, political, and economic factors, and the choice of growth-oriented adjustment policies obviously must be based on the particular circumstances of individual countries. Nonetheless, some general implications can be drawn from the analysis. First, the development of human resources plays an important role in economic growth. Expenditures on improving human capital appear to have a substantial effect on output growth when both public and private spending on the development of human resources (education, health care, and on-the-job training) are taken into account. Any planning for adjustment policies should consider the adverse consequences for economic growth of cutting such spending.
Second, a sustained increase in the domestic saving rate is crucial to growth-oriented adjustment efforts, because it is essential to raising the investment rate. In countries where government consumption expenditures nearly exhaust current revenues, it may be essential to combine tax reforms aimed at raising the revenue/income ratio with measures to reduce the budgetary share of consumption spending. In other countries, broadening the choice of financial instruments and institutions, pursued through an integration of unofficial with official financial markets, would encourage more households to increase their savings in financial assets and even to raise their saving rate in the long run.
Third, a steady expansion of the export sector appears to have powerful effects on growth. In some countries, the encouragement of exports may require adjustments in the exchange rate and trade systems, while in other countries the monetary and fiscal incentive systems may have to be improved in order to remove disincentives to exports. Such measures would have to be complemented by market-opening policies in industrial countries and in the developing countries themselves.
Fourth, the growth of population certainly provides a source of economic growth, but the effect on per capita income growth is negative. If this negative impact is to be mitigated, population growth needs to be accompanied by an increased rate of investment in human capital.
Fifth, the effect on output growth of the real interest rate on external debt is negative. This implies that general reductions in the world market real interest rate and in the risk premium would reduce the cost of external borrowing, help ease the burden of external debt, and help countries to realize their potential for growth.
FINANCE & DEVELOPMENT is available on microfilm from University Microfilms, P.O. Box 1346, Ann Arbor, Ml 48106, USA, and on microfiche (English only) from Microphoto Division, Bell and Howell Company, Old Mansfield Road, Wooster, OH 44691, USA
Ichiro Otani and Delano Villanueva