- Mohsin Khan, and Morris Goldstein
- Published Date:
- August 1982
For the past hundred years the rate of growth of output in the developing world has depended on the rate of growth of output in the developed world. When the developed grow fast the developing grow fast, and when the developed slow down, the developing slow down. Is this linkage inevitable?1
One of the more important yet puzzling aspects of the recent global stagflation has been the rather surprising resiliency of growth rates of real income in non-oil developing countries during the 1973–80 period in the face of the marked slowdown of corresponding growth rates in the industrial world.2 The primary purpose of this paper is to shed some light on this phenomenon by examining the relationship between the rate of economic growth in the non-oil developing countries and that in the industrial countries over the past decade or so.3 While this paper does not deal with projections or scenarios for growth rates in non-oil developing countries,4 the discussion identifies some of the key relationships and issues that need to be taken into account in any such exercises.
Table 1 presents growth rates (in real terms) for gross national product (GNP) and for certain trade aggregates in both the industrial countries and the non-oil developing countries during 1968–80. Four aspects of Table 1 deserve special mention as a background to the succeeding analysis.
|Growth in real terms of|
|GNP of industrial countries||4.5||3.1|
|Total imports of industrial countries||9.9||4.2|
|Imports of industrial countries from non-oil developing countries||7.7||5.6|
|Total exports of non-oil developing countries||9.3||7.2|
|GNP of all non-oil developing countries (excluding China)||5.8||5.2|
|By analytical group|
|Net oil exporters||6.0||6.5|
|Major exporters of manufactures||8.1||5.8|
|Other net oil importers||5.5||4.5|
|Terms of trade of non-oil developing countries||1.3||–1.2|
|Purchasing power of exports of non-oil developing countries||7.9||6.0|
|Total imports of non-oil developing countries||7.8||6.5|
To begin with, there is the pronounced decline in the average growth rate of real GNP in the industrial countries as one moves from the 1968–72 period (4.5 per cent) to the 1973–80 period (3.1 per cent).5 Also relevant, but not shown in Table 1, is the prospect of slow real output growth in these countries in both the short and medium term. After increasing at 1.2 per cent in 1981, it is envisaged that real GNP in the industrial countries will increase by 0.8 per cent in 1982 and by 2.5 per cent for 1983.6 Similarly, recent growth “scenarios” prepared by the International Monetary Fund7 and the World Bank8 for the 1980–85 period, point to annual average real GNP growth rates in industrial countries of 1.8 and 2.6 per cent, respectively, in the pessimistic “low growth” case, and perhaps more disappointingly, of only 2.2 to 3.3 per cent in the “high growth” case. Put simply, if, following the Lewis (1980) analogy, industrial countries are regarded as the “engine” of economic growth, then the engine has been slowing down since 1973, almost ground to a halt in 1981, and is considered unlikely to regain during the 1980s the average speed of the 1960s.
A second observation emerging from Table 1—and one that runs perhaps counter to the conventional view—is the absence of any rigid link between the rates of economic growth in the non-oil developing countries and the industrial countries. Essentially this link is expected to operate in the following way. A decline in growth in the industrial countries lowers their growth of import demand from all sources, including from the non-oil developing countries. The resulting fall in export growth from the non-oil developing countries in turn reduces their growth rates as well. However, as illustrated in Table 1: (i) the growth rate of real GNP in the industrial countries can be exceeded by the growth rate of their total imports; (ii) industrial countries’ overall import growth rate may be exceeded by the growth rate of their imports from non-oil developing countries; (iii) the growth rate of industrial country imports from non-oil developing countries can be exceeded by the growth rate of total exports of the non-oil developing countries; and (iv) this export growth rate may, in turn, be quite different from the real output growth rate of the non-oil developing countries. It should be emphasized that saying that non-oil developing countries’ growth is not tied rigidly to growth in industrial countries is not the same as saying either that the former could not have grown considerably faster since 1973 if the latter had grown faster over this period, or that non-oil developing countries can achieve reasonable growth rates indefinitely, regardless of the growth rate of industrial countries, or for that matter, that the post–1972 actual growth rates recorded by the non-oil developing countries are satisfactory. Rather, the point is simply that growth in the industrial countries is not the sole, nor even the overriding, determinant of growth in the non-oil developing countries. Indeed, as we argue later in this paper, a number of factors ranging from trade policies in the industrial countries to weather in the non-oil developing countries can alter the rate of the latter’s growth for any given rate of growth in the industrial countries. Further, by no means are all of these factors exogenous to the non-oil developing countries. In particular, the entries in Table 1 for industrial countries’ imports from non-oil developing countries and for the latter’s total exports and GNP are obviously sensitive to the economic policies adopted in the non-oil developing countries.
A third observation from Table 1 is the tremendous diversity of growth experiences within the non-oil developing country group—a diversity present in both periods. Whether classified by analytical group or by geographic regions, it is clear that different subgroups of non-oil developing countries have had widely varying degrees of success in coping with the post–1972 slowdown in industrial country growth. For example, whereas major exporters of manufactures saw their average growth rate drop from 8.1 per cent per annum in 1968–72 to 5.8 per cent per annum in 1973–80, the average growth rate in the low-income countries was actually marginally higher in the later period. Similarly, while non-oil developing countries in the Western Hemisphere recorded a large decline in average growth rates (7.2 per cent in 1968–72 versus 5.8 per cent in 1973–80) those in Asia grew faster on average in the second period.
The fourth observation—or set of observations—is in the nature of a caveat about the implications that can be drawn from aggregate average growth rates of real GNP alone. To start, because the growth figures in Table 1 are weighted averages of individual countries’ growth rates (where the weights represent individual countries’ shares in group GNP), it is possible for the group and subgroup averages to be dominated by the performances of a few large countries. In such a case, the group—and particularly the subgroup—growth figures may not be representative of the growth performance of a “typical” non-oil developing country over a given period. A graphic illustration of such a phenomenon occurred in 1979 when negative growth in India contributed to a 0.1 per cent average growth rate for the subgroup of low-income developing countries. In contrast, the median growth rate for this subgroup was 4.2 per cent. More generally, however, median growth rates have been within 0.5 to 1.0 percentage points of the weighted average growth rates (see Table 2), with the discrepancy being smaller in 1973–80 than in 1968–72. Nevertheless, Table 2 does indicate that the typical non-oil developing country was growing somewhat slower in both 1968–72 and 1973–80 than the weighted average figures in Table 1 would suggest.
|Weighted average||Median||Weighted average||Median|
|All non-oil developing countries1||5.8||4.9||5.2||4.7|
|Net oil exporters||6.0||5.4||6.5||6.3|
|Net oil importers1||5.8||4.8||4.9||4.5|
|Major exporters of manufactures||8.1||8.5||5.8||6.5|
|Other net oil importers||5.5||4.9||4.5||4.6|
Attention also needs to be paid to other characteristics of the distribution of growth rates in the non-oil developing countries which cannot be inferred from measures of central tendency (i.e., means and medians). Even when the mean or median growth rate has not changed much over time, there could well be quite significant differences in the dispersion of these growth rates across countries. Table 3 provides some evidence on this point by comparing various alternative measures of dispersion for real GNP growth rates of net oil-importing developing countries during 1968–72 and 1973–80.9 In short, these measures show that while average real GNP growth was only moderately lower in the latter period (4.9 per cent versus 5.8 per cent), the dispersion of these growth rates was considerably higher in the later period. Further, a greater proportion of countries had unsatisfactory growth records in 1973–80. For example, 17 per cent of all net oil-importing developing countries had negative average growth rates (versus 12 per cent in the earlier period); nearly one third had growth rates that implied constant or falling per capita real income (versus about one quarter in the earlier period); and for that 10 per cent of the sample that had the lowest growth rates, recorded average growth rates were less than half those of the earlier period.
|Time Period||Weighted Average Growth Rate (In per cent)||Coefficient of Variation||Average Growth Rate (In per cent)||Per Cent of Countries with|
|Lowest decile||Highest decile||Negative growth rates||Growth less than or equal to population growth|
A final caution concerns the dangers associated with drawing conclusions about changes in the relative wel-fare of groups of countries from real GNP growth rates alone. Two points are worth noting in this connection. One is that population growth rates are typically not independent of per capita income levels. Rather, empirical research reveals that population growth rates are usually negatively related to per capita income levels.10 The second point is that existing differences in levels of per capita income among country groups are so large that faster growth rates of real GNP in low-income non-oil developing countries are still consistent with an increasing absolute gap in per capita income levels between industrial and developing countries, or among different subgroups of developing countries themselves.11
Because the topic addressed in this paper is such a far-reaching one, it is necessary to place certain restrictions on the scope of the paper. The first such restriction is that since the focus of this study is on economic developments in non-oil developing countries, attention is focused on the transmission of economic activity from the industrial countries to the non-oil developing countries and any transmission effects in the opposite direction are ignored. While it is clear that this places emphasis on the dominant direction of causation, it does need to be acknowledged that the relationship between economic growth in the two regions runs in both directions and that the non-oil developing countries have in fact constituted a relatively buoyant market for exports from the industrial countries during the 1973–80 period.12
A second restriction is that the analysis is aggregate in character and does not deal with the effects of real income growth in the industrial countries on individual non-oil developing countries. Instead, wherever possible, account has been taken of the heterogeneous nature of the non-oil developing countries by giving separate attention to the four analytical subgroups of these countries commonly used by the International Monetary Fund, namely, net oil exporters, major exporters of manufactures, low-income countries, and other net oil importers.
The third restriction relates to the data used in this paper, and here reliance has been placed on the data available in the Fund and in the World Bank. One problem, however, is that the Fund’s country group classifications differ from those in the World Bank.13 Despite these differences, it is useful to sometimes use World Bank data to draw broad inferences about the non-oil developing countries where comparable Fund data are not available (e.g., the commodity composition of trade flows). The other data-related restriction concerns the time period of analysis. This analysis uses 1967 as a starting point because Fund data prior to that year are limited.14 The 1973–80 period is emphasized, however, in order to differentiate between the more recent, post oil-shock years in which the industrial countries exhibited slower growth and the earlier period.
The plan of the remainder of the paper is as follows: Section II considers the lines of influence between income growth in the industrial countries and the growth of exports from the non-oil developing countries. Section III carries the chain of argument further by examining the relationship between the growth of exports from the non-oil developing countries and their growth of real output. In brief, Sections II and III seek to identify the factors that can lead to divergencies between the growth rates of GNP in industrial countries and those of their total imports, their imports from non-oil developing countries, non-oil developing countries’ total exports, and GNP of non-oil developing countries. Finally, Section IV presents the paper’s main conclusions and some of the implications of the analysis.