Journal Issue

Guest article: The IMF’s role in developing countries: An exchange with the Editor

International Monetary Fund. External Relations Dept.
Published Date:
September 1984
  • ShareShare
Show Summary Details

An exchange between Tony Killick, Director of ODI, and the Editor of Finance & Development

Basic objectives of the IMF
Killick: I start with the objectives of the Fund, as stated in Article I. These include: “To facilitate the expansion and balanced growth of international trade, and to contribute thereby to the promotion and maintenance of high levels of employment and real income and to the development of the productive resources of all members as primary objectives of economic policy” and “To give confidence to members by making the general resources of the Fund temporarily available to them under adequate safeguards, thus providing them with opportunity to correct maladjustments in their balance of payments without resorting to measures destructive of national or international prosperity.”Editor: A number of international agencies were created in the postwar period with specific responsibilities and areas of competence: the World Bank in development matters, the GATT in trade matters, and so on. The Fund was given responsibility in international monetary affairs which, of course, includes the balance of payments situation of its members. The correction of balance of payments disequilibria is a central purpose of the Fund; it is not an end in itself to be sought to the exclusion of all else. Rather, the point is that a viable balance of payments is a critical ingredient in the attainment of the other objectives.
There is an immensely strong case in present-day circumstances for an organization such as the Fund to pursue these goals and the main question is about the Fund’s ability to achieve these desirable ends. Lest there be any confusion, I am not suggesting that the Fund be converted into another international aid agency for the financing of long-term development. Note, however, that Article I identifies multiple objectives, with employment, income, and economic development designated as primary objectives. The balance of payments is seen as a constraint on the achievement of these objectives, rather than an end in itself. However, over the years, the Fund’s emphasis on correcting disequilibria in the balance of payments has often meant that the primary objectives have been pushed into second place. I suggest that this pursuit of a single balance of payments objective, against the multifaceted goals of national governments, is an avoidable source of friction between the Fund and some developing countries.Article I indeed states that the purposes of the Fund are, inter alia, to facilitate the expansion and balanced growth of international trade, and to contribute thereby to the objectives of employment, income, and development of all members. The word “thereby” is crucial: it does not diminish the importance of the objectives but it clearly establishes how the Fund is to contribute to their achievement, namely, by facilitating the balanced growth of trade and, as Article I also states, through international cooperation, stability, and orderly exchange arrangements. It is not accurate to say that the primary objectives have been “pushed” into second place; they simply cannot be attained and sustained in the absence of a viable payments position.
Need for greater adaptability
Killick: I will return to that point when discussing the design of stabilization programs but there seem to me to be important reasons why the emphasis of Fund policies should change. One is the changing nature of the problems the Fund has to deal with and the constraints that inhibit it from adapting to them. For one thing, there have been major shifts in the type of country utilizing the Fund’s higher conditionality facilities. In recent years the great majority of stand-by and extended facility credits have gone to developing countries, and many to the poorer among them (credits to Argentina, Brazil, and Mexico in early 1983 were major exceptions). Thus of the stand-by and extended facility arrangements in effect in September 1983, 41 out of 45 such programs were to developing countries, of whom 34 are classified by the World Bank as low-income or lower-to-middle-income countries. When the Fund was created it was not at all envisaged that such would be the pattern of its lending nor was the design of its programs, as it emerged in the 1950s and 1960s, framed with such a pattern of resource use in mind.Editor: The Fund has changed and adapted its policies throughout its history: new facilities have been created (compensatory, buffer stock financing, oil, and extended facilities) and modified to meet the particular problems, especially of developing countries; lending policies have undergone review and change in response to changing circumstances, as was the case with the enlarged access policy. In the case of the three large debtor countries you mention, the Fund adapted its practices and took a coordinating role to help resolve their problems. As regards the type of country utilizing Fund resources, bear in mind that the Fund is a global institution. At various times in its history different groups of countries have required its assistance. The Fund has had programs with countries at various stages of development and differing economic and sociopolitical characteristics. I do not think that when the Fund was created a particular pattern of lending was envisaged, nor was this the case in the design of programs. The Fund follows certain general principles which must be applied uniformly, but nevertheless with due regard to the circumstances of each country.
Killick: The difficulty, as I see it, is that it is precisely for the poorer developing countries, with which it is now mainly dealing, that the Fund’s traditional approach to stabilization is not well suited. In our report we illustrate this with a case study. In attitudes toward economic policies, an appreciation of the gains to be had from maintaining an open and market-oriented economy, and the avoidance of the grosser forms of economic mismanagement, our case study (Kenya) appears to be precisely the type of country which could most easily do business with the Fund. Yet the relationship has often been a troubled one. All programs, except the most recent, have broken down. This illustrates the general difficulty of applying the Fund’s approach to stabilization to circumstances in the poorer developing countries. Indeed, there are those who say privately that the Fund is simply not the right kind of institution to assist such countries, which mainly need more long-term aid. Yet these countries also—in fact, particularly—suffer from disequilibria and need effective assistance from the Fund.Editor: Generalizations are often misleading. The relationship with member countries, as well as performance under Fund-supported programs, has varied. The less successful programs tend to receive most publicity. True, some programs have “broken down,” but the blanket explanation that this was due to the Fund’s approach is not borne out by the facts. Careful analysis shows that programs (including those in your case study) may break down for a number of reasons, such as a lack of commitment to their implementation, monitoring difficulties, political events, or a substantial change in external circumstances—matters that have little to do with the Fund’s approach. The difficult political and economic pressures make adjustment very hard in some cases and some programs do not succeed. There is no denying that the poorer countries are faced with especially difficult problems—mainly the basic problems of poverty and development. The Fund must do all it can, given its area of responsibility and expertise, to strengthen those countries’ development prospects by helping them ameliorate the balance of payments constraint. But if the need is for long-term aid, then this has to come from elsewhere, and the Fund’s contribution is to support conditions for its efficient use.
Killick: I will revert to the reasons for program breakdown shortly. But another reason for changes in the Fund’s approach is that the primary cause of many deficits has changed, just as the type of country experiencing disequilibria has changed. Deteriorating terms of trade and other exogenous factors have worsened the balance of payments of most oil importing developing countries during the past decade, difficulties that have sometimes been compounded by structural weaknesses of more domestic origin. At the same time, the control of domestic demand has remained the thrust of Fund programs, even though attempts to adjust the disequilibria by this means threaten to be “destructive of national prosperity” in terms of output, employment, and development. It is, moreover, universally accepted, not least by the Fund, that the structural adaptation required in such circumstances is necessarily a lengthy process. Yet the Fund continues to use conventional one-year stand-by programs that are not well designed to assist with this longer-term task, even if they are set within some medium-term framework.Editor: First, while the cause of deficits has shifted, adjustments must be made irrespective of the origin of balance of payments problems. Second, while control of demand is an important element of Fund-supported programs, it is not the thrust of programs to the exclusion of all else. The Fund’s emphasis on appropriate relative price policies has the aim of stimulating growth of output and employment, and also of shifting resources to the export sector—distinctly supply-oriented measures. Third, the length of Fund-supported programs. This must not be confused with Fund views about adjustment. There is no inconsistency between a medium-term view of adjustment and policy measures formulated for successive periods that are shorter. Programs can only cover periods for which governments are ready and able to offer meaningful policy commitments. It is better to have a succession of short programs to which there is reliable commitment rather than longer programs if these represent only intentions. Besides, shorter programs provide for more flexibility—something that you would approve of. Moreover, most governments are not prepared to make commitments for long periods for budgetary and other reasons (e.g., election timetables).
Killick: Yes, I certainly favor greater flexibility. Indeed, the Fund’s stated policy is to adapt conditionality to individual country conditions but to retain uniformity of treatment across countries. Our evidence suggested that, while there was not a stereotyped IMF package applied to all cases, there were fairly severe limits to the degree of flexibility which the Fund achieved in its programs in the 1970s. In particular, it usually (but not always) prescribed a program of demand management even in cases where excess demand was not found to be a principal cause of payments deficits.Editor: There are, of course, limits to flexibility imposed by the size of the necessary adjustment, resources available, and other factors. As regards your last point, a payments deficit implies a gap between supply and demand, that is a deficiency of supply or excess demand, or a combination of both. If, for the sake of argument, it is assumed that the deficit is caused by factors on the supply side alone, measures will clearly have to be taken to encourage supply. But this takes time, and while supply is recovering, demand will still have to be restrained (unless the country disposes of ample reserves or borrowing possibilities). In this simplified example, the extent to which adjustment takes place through demand management is determined by the responsiveness of supply—its extent and timing. In practice, of course, adjustment cannot be neatly compartmentalized and demand management, as well as supply management, must be considered part of an integrated policy package.
Killick: The Fund has sought to adapt to the new problems created by the oil shocks and the world recession in a number of ways: increasing its lending power; creating new facilities and enlarging existing ones; making increased use of the extended facility; and reducing insistence on rigorous conditionality. But it experienced difficulties in altering programs to accommodate the changed forms of adjustment required in response to exogenously generated payments deficits. In any case, the experiment with less rigorous conditionality, which started in the late 1970s, was put sharply into reverse in the latter half of 1981. It is hopefully not unfair to characterize the Fund’s 1982 policy conditionality as trying to tackle present-day problems with the policies of the 1960s?Editor: The perceived changes in conditionality in the mid-1970s have been addressed in the Managing Director’s interview with this journal (June 1982). To begin with, the original perception of a “loosening of conditionality” was not accurate. In the wake of the first oil shock, “Unconditional facilities, or facilities with light conditionality, were resorted to or designed in the expectation that the balance of payments problems might be reversible. It became very clear by 1978 that such an unconditional or ‘recycling-oriented’ approach was not warranted by the realities of the situation as we saw it evolve at that time. Balance of payments must be adjusted and the Fund must link its high credit tranche resources to the adoption of meaningful adjustment policies by member countries.” So what happened in the early 1980s was not a tightening of conditionality but rather a worsening of external conditions of the countries in question and the need for greater adjustment effort. One further point: it is not the origin of a deficit that determines the need for adjustment but its nature; in other words, whether it is temporary or permanent.
Killick: I can at this juncture refer back to an earlier point regarding the biases resulting from the “single objective” approach of Fund conditionality. As I mentioned, virtually without exception, programs are built around a balance of payments objective; any targets for such variables as economic growth and inflation are strictly subordinate to the balance of payments goal. Income distribution objectives do not feature at all. Yet there are potentially large economic costs involved in seeking to achieve economic restructuring largely by means of demand management (although such management certainly has an important role to play). I suggest that the Fund should be more willing than it has been in the past to accept development and the other “primary objectives” identified in Article I as imposing constraints upon the design of adjustment programs, and make a more conscious attempt to design programs which achieve adjustment in a least-cost manner.Editor: To recapitulate, the Fund is concerned with the balance of payments of members but not as an end in itself. A distinction must be drawn between the broader economic aspects of Fund-supported programs (which set forth the general direction of economic policy) and the performance criteria that apply to a few macroeconomic variables and help monitor performance. Matters other than the balance of payments do figure in Fund programs. Reducing inflation is certainly one. You also mention income distribution. This is a legitimate concern of economic policy. But it is not correct to say this factor does not figure in Fund-supported programs at all: indeed, the Fund encourages better income distribution by promoting realistic exchange rates and producer prices which benefit the poorer rural sector—generally the primary export sector in developing countries. But, in general, the Fund aims at staying clear of domestic priority areas. Note that the Fund is also taken to task for being already too involved in the economies of its members.
Impact of Fund Programs
Killick: In the course of our project, my colleagues and I reviewed the available published and unpublished evidence on the impact of Fund programs, and undertook tests of our own. We arrived at conclusions which strengthen the case for changes in Fund policies. Although the evidence is by no means uniform and there are large intrinsic difficulties in arriving at a satisfactory assessment, the studies indicated that Fund-supported programs in developing countries had limited effectiveness. To summarize: (1) while programs did tend to strengthen the basic or overall balance of payments, the known statistical significance of the results was low and the achievement often fell short of program targets, which tended to be overambitious; (2) only in some cases did Fund programs result in significantly larger inflows of capital from other sources, and there was no systematic association between programs and sustained liberalization of trade and payments; (3) programs did not generally have strong deflationary effects, although there were indications that such effects were becoming stronger in the more recent years. It was apparent that both stand-by and extended facility programs were subject to fairly frequent breakdowns, especially the latter. Extended facility programs probably brought smaller balance of payments benefits but did better with respect to growth. Overall, the evidence did not provide much support for the view that extended facility program results were markedly weaker than stand-by program results.Editor: The measurement of effectiveness of economic performance is a difficult area and one on which more work is necessary. There are more questions than answers: What is to be measured? Is performance under the program to be measured against performance before the program? Is performance of a country with a program to be compared with the performance of a country without a program? In my view the most reasonable measure is the actual situation compared with the situation that would have prevailed in the absence of a Fund-supported program (or with the situation under alternative prescriptions). In other words, it is insufficient to measure progress under a program exclusively against a set of economic targets; it is also important, difficult though it is, to measure against the alternatives of not adjusting or of disorderly adjustment. This necessarily involves estimates of likely developments and conjectures as to the policy measures that would have been implemented in the absence of a Fund program. How would your “results” stand up against this test? One more point: the success of a country’s association with the Fund is not limited to the program period alone. A broader view is necessary; often the Fund helps induce a reorientation of thinking and policy which goes beyond the performance under the program.
Killick: When we sought to explain the limited impact, we naturally started with the considerable difficulties which the Fund experienced in securing government compliance with key components of programs, but an only moderately strong connection was discovered between programs that were implemented and the achievement of desired results. The reasons for unachieved targets seemed rather to include the effects of the large dislocations that occurred in the world economy after 1973 and the ambivalence shown by some developing country governments toward stabilization and payments adjustment, partly because of the political sensitivity of effective policies. However, we also suggest that weaknesses in the design of Fund programs themselves contributed to the indifferent record of achievement—especially their short-term nature, the consequential tendency to overambition, and their relative neglect of measures directed at the quantity and composition of output.Editor: It is difficult to interpret terms such as “limited impact” or “indifferent record of achievement” as general judgments on the effectiveness of Fund programs. The experience has varied widely. With respect to “overambition,” it is not clear what the term means in practice. The Fund can, of course, be mistaken in some of its predictions. In many instances, though, “overambition” is used to argue in favor of lowering the adjustment effort that is required. A program must take into account many factors, including available finance, the authorities’ capacity to implement measures, and the external situation. As regards the relative neglect of measures concerning the quantity and composition of output, I have already explained that great emphasis is put in Fund programs on enhancing supply and shifting resources.
Directions of change
Killick: Let me turn now to how Fund conditionality policies could be changed. At the general level we urge the adoption of a richer mix of policies and of the principle that programs should be designed to resolve the causes of the problems. Second, we urge the conscious adoption by the Fund of a “cost-minimization” principle in designing its conditionality for any given country; namely, it should examine alternative courses of action on the balance of payments and select that which is most in harmony, or least in conflict, with other objectives of government policy. This implies giving greater weight to the “primary objectives” of economic policy, to the distribution of income, and to the political implications of the proposed measures. Third, while the Fund’s principle of uniformity of treatment among members should be upheld, we suggest that greater weight be placed on the different abilities of the authorities to formulate and execute integrated packages of measures, and on the varying speed of improvement it is realistic to anticipate. Taking countries’ differing “capacities to transform” more into account would move the Fund in the direction of longer-term programs in the poorer developing countries. A final general recommendation is that Fund conditionality should focus on achieving a consensus with the government about desirable policy measures, with a corresponding move away from the present importance placed upon quantified performance criteria relating to domestic credit and a few other macroeconomic variables.Editor: You raise several points. First, there is no disagreement that programs should address the adjustment needs within the constraints of each case. I assume by a “richer mix” you simply mean more conditions. Second, the Fund does examine, in discussions with member countries, “alternative courses” and it goes without saying that it follows a “cost-minimization” approach, namely, adjustment consistent with the exigencies of the situation, including growth potential. Fund-supported programs obviously can only contain measures that are based on a consensus with the authorities about desirable policy outcomes, taking into account social and political implications. But if you are implying that the Fund should assume as given a particular social objective, and build a program around it, this is a more difficult proposition since it may not be consistent with the objective of adjustment. Your third point is to a large extent already practiced, within the constraint of available financing resources. Finally, the purpose of quantified criteria is to monitor performance under the program; how does one monitor without some quantification? One way would be by a subjective and judgmental approach, and that could well create greater difficulties than do the present quantifiable targets.
Killick: I will return to that question in a moment. But let me point out that the principle that program design be related to the causes of the problem could have substantial implications. While in some cases the main thrust of Fund programs would be little affected, there is widespread agreement that the problems of many oil importing developing countries are long term and structural. For many there has, in the past decade, been an apparently fundamental worsening in their terms of trade and export prospects (largely as a result of higher real oil prices and slower growth in the industrial countries), so that they now need to adjust their structures of production and demand to these harsher realities. In such cases we urge the adoption of “real economy” programs. The precise nature of these would need to be related to specific country circumstances, and we provide a concrete illustration in our report. But I see them as possessing at least the following features:Editor: Is “real” used in the sense of physical quantities in contrast to “monetary” variables? Would, for example, realistic pricing policies be excluded from a “real economy” approach? In any case, labels are not important. Existing Fund-supported programs, and the alternative suggested, should be judged on the criterion of their effectiveness in attaining the objectives sought within the confines set by the particular situation faced by members.
(1) They are set in a cost-minimizing, growth-oriented framework and are also designed to be consonant with government objectives on poverty alleviation and income distribution. Cost-minimization would be pursued both by measures designed to increase the utilization of existing productive capacity and to protect productive investment and hence the expansion of capacity. If successful, they would permit equilibrium to be restored at a higher level of economic activity than under programs placing most weight on demand restraint.(1) The terms “cost-minimizing,” “growth-oriented,” and “government objectives on poverty alleviation and income distribution” cover attractive and laudable objectives. Again, however, we take issue with the notion that Fund programs place “most weight on demand restraint.” Demand management is important—and you yourself qualify it as being of “key importance”—but as indicated earlier it is not the focal point of Fund programs to the exclusion of all else. A further point: you single out the protection of investment. For any given degree of necessary adjustment, this implies that real consumption would have to be cut even more. It would be wrong to argue that real investment expenditures must never be cut in an adjustment program.
(2) Real economy programs would include measures to stimulate the production of exportable and import-substituting goods and services relative to nontradables, with these measures having at least the same status as other provisions of the program. A wide range of policy measures can be deployed to assist this type of structural change. Changes in relative prices in both the product and factor markets will, however, be of crucial importance and we agree with the stress placed by the Fund on the role of the exchange rate in influencing the pattern of relative prices.(2) This would appear to require the Fund to engage in a detailed regulation of economies, including decisions as to which sectors or industries should be promoted, the price that should be charged for specific factors, and so on. This may be neither feasible nor acceptable to member countries. The Fund cannot become the “super minister” of economic management. Moreover, some of the areas—such as investment decisions to effect changes in the structure of production—are the domain of the World Bank.
(3) They should be medium-term, designed to be executed over, say, five years. Policies of structural adaptation must be expected to require at least that period to be effective.(3) This point has already been discussed, but perhaps it is worth reiterating the difficulties governments often face in undertaking policy commitments over an extended period of time.
(4) The emphasis should be upon achieving a consensus on necessary policy changes, leading to a stronger government commitment to the programs. More widespread use of resident Fund representatives should help achieve such agreements; they could also help monitor the progress of programs in place.(4) Both of these already take place.
(5) Demand restraint would remain of key importance but would move somewhat from the center of stage. No matter what approach is taken, an improvement of the balance of payments necessitates a reduction in consumption and/or investment relative to income. However, our approach would increase the prospects of achieving this through increases in income. It would also recognize the positive role that domestic credit can play in permitting increased capacity utilization and investments in structural change.(5) Demand management is of key importance, although hitherto you have implied the Fund gives excessive attention to it. Does the last sentence mean that rather than agreeing with overall credit ceilings, the Fund should get into the business of rationing credit between sectors, industries, perhaps even individual entities?
(6) Quantified performance criteria should be replaced by a broader set of “review indicators.” Performance under these indicators—which would sometimes be quite similar to conventional criteria—would not govern eligibility for continued access to the credit, as in the case of existing performance criteria. But, like these criteria, they would trigger a review mission that would assess the overall progress of the program and make recommendations about continued access on that basis. A review mission could be dispatched at the initiative of either the government or the Fund.(6) Your suggestion is no different, in essence, from present practice because performance criteria also operate as “review indicators.” After all, if a performance criterion is breached, normally this triggers a review mission and new discussions.
(7) There should be an agreed timetable of execution for all, or a large proportion of, the program elements and explicit provision for the ways in which progress would be monitored. The experience of the World Bank with its structural adjustment loans—which have much influenced our own thinking—is that it is perfectly feasible to monitor such programs. It has also shown that they do not give rise to insuperable political objections, even though they are regarded as very demanding.(7) Two points: A recent Bank study on its structural adjustment loans concluded that “… conditions adequate to permit access to Fund resources are usually a prerequisite to a detailed program of structural change.” Second, structural adjustment programs include a monitorable set of specific actions—this is not that different from conditionality criteria.
(8) In addition to lending its own resources, the Fund would regularly initiate actions to attract additional supporting finance from other multilateral, bilateral, and, perhaps, commercial sources. It has, of course, recently done this in its well-known rescue operations for Brazil, Mexico, and other heavily indebted countries, but this was not previously a standard feature of the Fund’s work.(8) Indeed, this has recently been the case. But it is something that must be done on a case-by-case basis, rather than as a standard feature of the Fund’s work. The attraction of development aid, on the other hand, is a function of the World Bank through its consultative groups and donor groups.
In sum…
Killick: There is no doubt a good deal we could still argue about in the specifics of these proposals and gaps to be filled. But one point is unquestionable: they do not offer deficit countries a soft option, a way of obtaining a lot more money without taking tough policy measures. The programs would not usually be easy or popular, but they would have the merit of being seen as more relevant. The above suggestions are not radical. To a large extent we are urging the Fund—and the country representatives who determine its policies—to have the courage of their own convictions when they introduced the extended facility in 1975 and when they moved to adapt conditionality after the second oil shock in 1979-80. It is true that if it were to move far in the direction we propose, the Fund would need larger resources and our report discusses how these might be mobilized. I also believe that there is need for policy changes in many developing countries. It should be remembered, too, that any success that oil importing developing countries might have in reducing their deficits depends on the willingness of other countries to see their payments positions weaken. For as long as the international environment remains characterized by protectionist anti-trade policies and deflationary macroeconomic policies in industrial countries, the scope for adjustment through the domestic efforts of developing country governments will remain severely limited.Editor: A few general comments in response: First, as indicated, many of the points involve proposals that are already carried out by the Fund, or points that put a different emphasis on aspects of the Fund approach. Second, some of the suggestions would be “radical” in the sense that they would change the traditional role and activities of the Fund and transform it into a much more interventionist agent in the economic management of its members. Third, you appreciate, although you do not emphasize, that these proposals would require larger resources than the Fund currently possesses. Indeed, given the difficulties the Fund has experienced in securing the last quota increase and the difficulties in borrowing from official sources, it is surprising that the need for greater Fund resources, and how to obtain them, does not form a major focus of your attention. Finally, your statement on the injurious effects of protectionism is well taken.
Killick: While I have concentrated here on areas in which we believe Fund policies might be strengthened this does not mean I fail to recognize the often acute difficulties with which it has to wrestle. The Fund has made major contributions, not the least in the constructive role it has played in averting the incipient banking crisis of the past 18 months. But I am concerned that the constraints upon its attempts to adapt to changing circumstances threaten both to weaken it and to leave developing countries short of the assistance they need in dealing with acute foreign exchange shortages and the consequential costs to human well-being.Editor: Open-minded studies of the Fund and constructive suggestions are welcome. Many of your concerns are valid and have been, or are being, addressed. Others imply essentially a different emphasis. Others still, I think, are perhaps based on mistaken premises of what the Fund is, what it does, and what it can do. To be sure, there are a large number of constraints on the ability of the Fund to assist its members. But I do not believe that merely relaxing conditionality, or going to five-year (or longer) programs, or adapting a real economy approach addresses these constraints. Given its resources, objectives, and the circumstances of developing countries, I think the Fund performs a most useful role and provides a needed service for all its members, and thus for the international economy.

Other Resources Citing This Publication