Interest in and concern about the global effects of Fund-supported adjustment programs (hereafter referred to as Fund programs) has increased considerably in the 1980s. To some extent, this is a natural reflection of the number of member countries undertaking Fund programs. In 1980–83, for example, in the face of severe external payments deficits, an average of 23 countries a year had a Fund program. If countries borrowing under the compensatory financing facility are also included, the number of program countries rises to 31 a year. The corresponding figure for 1971–73 was only eight countries a year. (Program periods do not coincide exactly with calendar years; a country was assigned to the program group in any year in which it had a Fund program for at least seven months.)
The misgivings about the global effects of Fund programs may be classified into four areas. First, that simultaneous demand restraint policies in many program countries could impart a pro-cyclical deflationary bias to the world economy, with adverse consequences for real output and employment in non-program and program countries alike. Second, that the consistency of Fund prescriptions across countries was questionable—that, for example, prescriptions for appropriate monetary and fiscal policies in industrial countries might imply a lower demand for imports than the export objectives of Fund programs in non-oil developing countries. Third, that because one program country’s imports are another’s exports, trade links among program countries may frustrate the trade objectives of individual countries. Finally, that simultaneous exchange rate devaluation by many program countries, some of which export mainly primary commodities, would result mostly in a lower world price for their exports, with little beneficial effect on their export earnings. The common thread of these positions is that what might be feasible and desirable for a single program country acting alone will not be so for many program countries acting simultaneously.
This article is based on a much longer study of the same title published by the Fund as Occasional Paper 42.
The study on which this article is based examined the strengths and weaknesses of alternative ways of defining and measuring the effects of programs; identified the channels by which policies in program countries might be expected to affect both non-program and other program countries; reviewed the empirical evidence on the likely size of such “aggregation” or “interdependence” effects of Fund programs; and considered the ways in which the Fund takes global effects into account, both in the design of programs and, more broadly, in the advice it gives to member countries. The scope of the study did not allow an appraisal of individual Fund programs.
Definition and measurement
Before assessing the global effects of Fund programs, it is important to have a clear idea of how program effects should be defined and measured. Five separate standards for the evaluation of Fund programs were examined. They were:
• Factual standard: the difference between macroeconomic performance under the Fund program and performance prior to the program—the before-after approach;
• Normative measure: the difference between performance under the program and the performance specified in its targets—the actual-versus-target approach;
• Conjectural standard—I: the difference between performance under the program and performance had no program been implemented—the actual-versus-in-the-absence-of approach;
• Conjectural standard—II: the difference between performance under the program and the “optimal” performance—the actual-versus-optimal-policy approach; and
• Conjectural standard—III: the difference between hypothetical performance under Fund program-type policies and hypothetical performance under some other policies—the comparison-of-policies approach.
The main conclusion, simply put, was that not only the size but also the direction of program effects is likely to be quite sensitive to alternative definitions and estimating methodologies. The review of five possible interpretations of program effects showed that the measured effects of the same programs can vary substantially, depending, inter alia, on (1) whether changes in non-program factors between the pre-program and program periods are accounted for; (2) whether program targets make allowance for unexpected developments in the global environment; (3) whether program countries differ systematically from non-program countries prior to the program period in ways that matter for subsequent performance; (4) whether non-program countries are themselves affected by Fund programs; (5) whether the medium-and long-run as well as the initial effects of programs are considered; (6) whether, because of confidence and credibility factors, the implementation of a given policy has different effects within the context of a Fund program than it has without it; and, perhaps most important, (7) whether the most relevant comparison for the actual effect of a Fund program is what would have happened without it, or what could have happened under some hypothetical and optimal set of policies.
A good way of illustrating how the alternative definitions of program effects can color the evaluation of programs is to use the different methods to assess the much discussed recent import compression experienced by countries that had Fund programs in 1983. On a weighted average basis, the volume of imports by program countries fell by almost 8 percent in 1983. What role should be assigned to Fund programs in this decline?
Because the before-after approach cannot distinguish program from non-program determinants of outcomes, all changes are attributed to the program, and the interpretation would be that Fund programs “caused” the fall in import volumes. Since a lower demand for imports by program countries implies, ceteris paribus, lower exports for the rest of the world, this would imply, in turn, that Fund programs had a deflationary effect on global economic activity, depending of course on their weight in the world economy.
The actual-versus-in-the-absence-of approach means comparing the actual decline in imports with the change that would have occurred without Fund programs. In this connection, three points are relevant. First, the Fund’s lending in 1983 exceeded SDR 12 billion and helped to secure over SDR 20 billion in new bank lending to non-oil developing countries. Second, without the direct and “catalytic” effects of Fund lending, the flow of financing to 1983 program countries would have been much smaller. Finally, based on past empirical work, foreign exchange receipts are the main determinant of the demand for imports in developing countries. These factors suggest that as a result of Fund programs the decline in import volume was smaller than it would otherwise have been; Fund programs can then be viewed as having had an expansionary effect on global economic activity. In addition, based on preliminary trade data for 1984, the same group of 1983 program countries exhibited an average increase of 10 percent in their import volumes in 1984—which supports the proposition that the medium-term effects of programs are probably quite different from their initial impact.
Yet a third, more mixed, verdict might well emerge from the actual-versus-target or actual-versus-optimal-policy approaches. If, for example, import volumes fell more than targeted, the verdict might be that the external adjustment achieved under 1983 programs was both unavoidable and better managed than it would have been without programs, but still that the compression of imports went further than would be optimal or desirable from the perspective of longer-term growth. Under these methods, the decline of imports could be attributed to overachievements of fiscal targets, or to greater than anticipated adjustment pressures linked to higher than expected world real interest rates, or even to the application of restrictive trade controls by program countries that ran counter to program intentions. In any case, the conclusion from this perspective could be that the effect of programs on imports was expansionary but not as expansionary as would be desirable or optimal given the operating environment.
The effects of programs can, therefore, mean different things to different people. This is not all bad because none of the separate definitions of program effects is free of shortcomings. Still, unless these different definitions or interpretations of program effects are explicitly recognized, the danger exists that different views of the global effects of Fund programs will be due in large part to the application of different yardsticks to the same evidence.
It is clear that the global impact of Fund programs will be strongly influenced by the structural and behavioral characteristics of the program countries themselves. In the study, four of these characteristics were examined: (1) the share of program countries in world trade; (2) the degree of trade interdependence among them; (3) the share of program countries in international capital flows; and (4) the typical size of changes in import volumes, export prices, and real exchange rates in program countries that could have significant transmission effects. One might expect that, other things being equal, the larger these four parameters or disturbances, the greater would be the transmission of program effects to the rest of the world or to other program countries, that is, the larger would be the global effects of programs.
A detailed examination of the trade data for program countries showed that, first and most important, even though the combined share of these countries in world trade has risen steadily over the past decade, it is still quite modest, accounting in 1983 for about 7–8 percent of global trade and roughly 40 percent of the trade of all non-oil developing countries. To place these figures in perspective, in 1983 the seven largest industrial countries took 49 percent, and the United States alone accounted for over 15 percent, of world imports. Thus, even with nearly 40 countries implementing Fund programs in 1984, the potential for shifts in import demand in these countries to affect economic activity in the rest of the world would seem to be quite limited, especially compared with the potential leverage of the industrial countries. (The fact that program countries as a group typically have rather a modest share of world imports does not mean, though, either that individual countries or even industries could not be seriously affected by changes in the import behavior of program countries, or that these induced effects on exports would be roughly similar across countries and industries.)
A second point that emerged is that the share of world trade attributable to program countries varies considerably over time with changes in the numbers and types of program countries. The program countries’ share of world imports for 1983, for example, was ten times larger than their average share in 1973–75, and more than four times larger than their share in 1982. The main reason for the increase in their share of world imports in 1983 is that several large trading countries (Argentina, Brazil, Chile, Hungary, Korea, Mexico, and Turkey) were added to the program country group in that year. (Likewise, the program country share of world trade hit a peak of over 12 percent of world imports in 1977 because the United Kingdom and Italy had programs then.) Perhaps the main implication of this temporal instability in shares of world trade is that one should not expect any transmission effects from program countries to be stable from year to year. This of course complicates the estimation of the global effects of programs.
The greater the trade interdependence among program countries, the higher, ceteris paribus, would be the risk that any program-induced changes in the demand for imports would be mutually reinforcing—perhaps with larger multiplier effects on aggregate demand than desired or anticipated.
An analysis of the data on interdependence suggested two conclusions. First, the average degree of trade interdependence among program countries is rather low. For the 37 program countries examined, the (unweighted) average share of imports from and exports to other program countries was 9 percent and 8 percent, respectively. This rather low average level of trade interdependence among program countries reflects the more general fact that most program countries are non-oil developing countries and that these countries trade mostly with industrial countries. (In 1983, for example, industrial countries accounted for 59 percent of the total imports of non-oil developing countries and for 57 percent of their exports.) Trade among non-oil developing countries represented 20 percent of their total imports and 24 percent of their total exports in 1983.
Second, although average interdependence is low, there clearly are some program countries where intra-program country trade is significant. Out of the 37 program countries examined, 11 have more than 10 percent of their trade with other program countries; for four of them that average was above 25 percent.
For the majority of program countries, therefore, trade with other program countries accounts for only a small share of total exports or imports. Such trade has, however, been increasing, and there are some program countries where it is unmistakably important. Finally, it appears that trade among program countries is more capital intensive than the trade of program countries with industrial countries.
International capital flows
Policies in program countries could affect other countries through trade in financial assets as well as trade in goods and services. In addition, because the availability and terms of financing strongly influence the speed of external adjustment, and because current account deficits create a need for financing, the effects of capital flows on programs can often not be divorced from the effects of trade flows.
As with trade flows, one would expect the global effects of programs to be larger, the larger the weight of program countries in international capital flows. An evaluation of the size and structure of the external liabilities of all non-oil developing countries in both 1973 and 1983 shows that: first, non-oil developing countries have been more attractive to international lenders than to foreign investors; second, private creditors have become much more important than official creditors as a source of external lending to these countries over the past decade; and third, financial institutions, primarily commercial banks, have been at the forefront of this “privatization” of lending to the developing world, increasing their share in long-term debt from 15 percent in 1973 to 36 percent in 1983.
The share of program countries in the external liabilities of the non-oil developing countries is now much more important than it used to be. The piecemeal data available on foreign direct investment suggest: (1) that program countries accounted for roughly 55 percent of the total foreign direct investment in non-oil developing countries in 1983; (2) that only three countries of the program country group (Brazil, Mexico, and South Africa) accounted for about 70 percent of the total in 1983; and (3) that in earlier years, program countries seem to have had only a small share (less than 10 percent) of total foreign direct investment in non-oil developing countries.
The same pattern seems to have prevailed with external debt. Until 1983, program countries had only a modest share of total debt of all non-oil developing countries, ranging from about 8 percent in 1982 to roughly 23 percent in 1980. However, with the addition in 1983 of 11 major borrowing developing countries to the program country group, the situation changed dramatically: program countries then accounted for 56 percent of the total outstanding debt of non-oil developing countries, 79 percent of their short-term debt, and 67 percent of the long-term debt owed to private financial institutions. This sudden change reflects the concentration of bank lending to developing countries in a relatively small number of major borrowers, the serious debt-servicing difficulties of these borrowers in 1982–83 in response to a harsh external environment and inappropriate past domestic policies, and the implementation of Fund programs by these same countries. Lest the figures on the program country share of bank lending to non-oil developing countries be misinterpreted, it is important to recognize that even in the peak year of 1983, program countries probably accounted for only 3–4 percent of banks’ total (domestic and international) claims.
Finally, when considering the global effects of Fund programs that operate via international capital flows, it is crucial to account for the distinction between the influence of Fund programs and that of program countries. In recent years a number of important Fund programs have involved an understanding not only between the Fund and the program countries, but also between the Fund and various private financial institutions—and this precisely out of concern for the global or “systemic” effects that might follow if lenders and borrowers in international capital markets took too narrow a view of their own self-interest.
Effects of import changes
The broad characteristics of program countries that were reviewed earlier are helpful as an indicator of their potential to influence macroeconomic developments in the rest of the world. To gain a more precise view of the global effects of Fund programs, these effects were studied within the more formal framework of econometric global trade models. (Such models have certain advantages over rough calculations of trade shares: they can provide a better estimate of induced, “later round” effects; they are better at estimating the timing of effects; and they can better calculate marginal, as against average, trade propensities, and these are more relevant for assessing the effects of Fund programs.) The three models used for the simulations were the OECD Interlink model, the IMF World Trade Model, and the LINK model.
The simulation experiments conducted within the framework of these models showed that changes in imports by program countries do affect economic activity in the rest of the world, and in the expected direction. But, just as important, they strongly suggest that the size of such global transmission effects is small. Specifically, even the 7 percent (or $10 billion) fall in the value of imports by program countries that occurred in 1983 appears to have been associated with only a 0.1–0.2 percent fall in real GNP in industrial countries. This is not the stuff of which global recessions are made, or ended.
The same simulation exercises also indicate: (1) that the lion’s share of these trade and output transmission effects takes place within one year of the import change; (2) that the full or final effect on real GNP in industrial countries, albeit small, is considerably larger (say, two to three times) than the initial effect; and (3) that even among the seven largest industrial countries, these induced effects on output differ because of intercountry differences in both the share of total exports going to non-oil developing countries and the share of exports in GNP. Finally, the simulation results imply that the effects of changes in imports of program countries are likely to be much greater on their own real income and growth rates than on those of their partners.
Exchange rate changes
No other topic seems to have led to so much discussion of aggregation effects as simultaneous exchange rate action by primary producing countries. The World Bank and the Fund have been criticized for taking too “piecemeal” an approach to exchange rate policy and it has been suggested that developing countries should collectively devalue against the currencies of the developed countries. On the other hand, as noted earlier, some observers have taken precisely the opposite view—warning that exchange rate-induced increases in production and in exports of primary commodities, if implemented simultaneously by many program countries, will merely depress the world price of these commodities and unfavorably affect the instigator’s terms of trade—and this for little benefit since the demand for these goods is quite price inelastic.
In analyzing this aspect of the study, the aggregate or global effects of multilateral exchange rate changes by a group of program countries were examined. The analysis showed that the proposition that simultaneous exchange rate action by program countries could have serious aggregate effects on the prices of program country exports was applicable mainly to primary commodities. Moreover, the potential for significant aggregate price consequences depended mainly on the ability of program countries to affect world supply. While this potential is clearly much higher for some commodities (e.g., cocoa, coffee) than for others (e.g., wheat, citrus fruits), and in the long run rather than the short run, the risks are reduced in practice. This is because primary commodities now represent a significantly smaller share of exports by non-oil developing countries than they did two decades ago, and the share of non-oil developing countries in world exports of primary commodities is now considerably smaller than even a decade ago. Further, not all program countries change their exchange rates at the same time and those that do usually do not export the same products. Still, such interdependence and aggregation effects associated with multilateral exchange rate action need to be closely monitored, and it is possible to identify the individual non-oil developing countries and primary commodities where “market power” seems to be relatively high. A case has also been made for the view that exchange rate adjustment can be useful to protect the profitability of exporting even for those non-oil developing countries that face fixed external terms of trade. Last but not least, the important role played by industrial country policies in influencing the effects of exchange rate action by program countries must be recognized.
Aggregation and interdependence
The discussion thus far is relevant for assessing the likely size and direction of any global effects of Fund programs. Attention must, however, also be turned toward an equally important and closely related subject, namely, if and how the Fund accounts for aggregation and interdependence effects in both the advice it gives to member countries and in the design of Fund programs themselves. The latter is important because even if Fund programs did have strong potential global effects, such effects could in principle be offset for both in the design of programs and in the advice given by the Fund to non-program countries. For example, if the process of achieving greater fiscal responsibility in program countries had significant multiplier effects on aggregate demand in non-program countries, and if these spillover effects were larger than desired and their size and timing were known, then an adjustment could be made to the design of programs to reduce such spillovers. Similarly, if the international adjustment process is working smoothly, any reduction in spending in countries with balance of payments deficits should be offset by an increase in spending in surplus countries, leaving global aggregate demand little affected. In other words, trouble arises only if the global effects of programs are significant and if these effects are ignored in the design of policy in program countries and in the Fund’s advice to non-program countries.
By its very nature the Fund must be concerned about the global or systemic effects of policies of its individual members. Indeed, the raison d’être of the Fund and most of its activities is precisely the principle that the effects on other countries of the policies of individual members can be significant, and that an institution is needed to ensure that countries with balance of payments problems do not take measures that have large and unsatisfactory international repercussions. Both the Fund’s lending activities and its surveillance functions are largely directed toward meeting that objective. Also, some of the Fund’s most visible activities during the past few years have been motivated by such global concerns. Two of the best examples are the Fund’s efforts to deal with the debt problems of members that had borrowed beyond their debt-servicing capacity and the Fund’s recent policy advice to the United States concerning its fiscal policy. The relevant question is not whether the Fund ought to consider the global and systemic effects of its advice and programs but rather how it can best do so.
The main mechanism within the Fund for appraising the global effects of country policies, as well as the consistency of policies across country groups, is the World Economic Outlook exercise that takes place twice a year. Because of the sequential nature of Fund programs and because of the distribution to outgoing missions of information on both previous Fund programs and foreign demand and price developments, it is also possible for program design to incorporate effects from other programs. In a similar vein, the provisions for waivers and modifications in Fund programs represent a well-established mechanism for dealing with departures from performance criteria, including those attributable to unforeseen aggregation and interdependence effects.
As the analysis presented in this article has shown, while the concerns regarding the global effects of Fund programs cannot be dismissed, the situation is neither as straightforward nor as alarming as some would suggest. First, it must be recognized that alternative definitions and measurement standards for program effects can yield markedly different results. Second, it is important to view the impact of Fund programs in their broader sense—including, for example, the “catalytic” effect on enhancing the availability of financial resources—as well as to consider the situation that would have developed in the absence of a Fund program. Third, the global effects of Fund programs on trade and economic activity, during the 1980s, were limited, inter alia, by the small share of program countries in world trade and by the relatively low degree of trade interdependence among program countries. Fourth, the risk of adverse aggregate price effects of simultaneous exchange rate depreciations was reduced because, inter alia, of the lesser importance of primary commodities in the trade of the non-oil developing countries, and the relatively low share of these countries in world exports of non-energy primary commodities. However, the Fund monitors closely the global effects of exchange rate changes by program countries and must continue to urge industrial countries to improve access of developing countries to their large markets. Fifth, the global or systemic effects of policies of individual countries do matter. The Fund, in offering advice to all countries, is concerned about these effects and has evolved a number of procedures to consider the repercussions of Fund programs on other countries and the world economy as a whole.