This chapter examines why the liberalization of currency transactions, and also (to some extent) of trade, was less effective outside the developed world, or why the idea of “one world” and of a single world economy was challenged. The IMF worked out a technical and analytical device for the achievement of stabilization, the so-called Polak model, but its operation was limited by what was at that time usually described as a failure or absence of political will in member countries. The resulting difficulties and failures obliged the IMF and the World Bank to begin to rethink issues concerned with the design of the international order and their own role within it.
The conference of Bretton Woods, held in the last phase of a great world war, had—perhaps not surprisingly—concerned itself only rather peripherally with development issues. Although Article I of the Fund’s Articles of Agreement stated an objective of “the development of the productive resources of all members,” development finance had not been a major concern in the circumstances of 1944. The conference had avoided any attempt to distinguish between groups of members, and its participants ruled out an additional phrase suggested by the Indian delegation as part of the second “purpose of the Fund” set out in that Article: “to assist in the fuller utilisation of the resources of economically under-developed countries.”1
Indeed it only became clear later that there was a specific problem about “development.” It was only in the course of the first postwar decades that the “developing world” began to define itself politically and economically. Politically, the new definition came about as a result of the breakup of colonial regimes and the advent of independence. Within a short space of time, international institutions gained a large number of new members. From 1960 to 1963, the membership of the IMF rose from 68 to 101. Economically, states now regarded themselves as “developing,” or in the more usual expressions of the 1950s and 1960s as “underdeveloped” or “less developed.” Development would be an important part of the assertion of national independence: it meant evolving a material basis that would eliminate economic dependence, and also satisfy the new demands and expectations of citizens. The new states searched for a particular economic strategy appropriate to their particular political condition. Their answer often lay in a partial disengagement from the world economy, and a reluctance to accept the IMF Article VIII goal of currency convertibility, or to see any substantial advantages in trade liberalization. The great movement of trade and currency liberalization, which reached a climax in the early 1960s in the developed world, thus left many poorer countries untouched.
Alvaro—“No, don’t touch me!
I can do it myself.”
Explaining why is not easy; and indeed the reasons clearly varied considerably from country to country. In general, some mixture of the following arguments led states to look for a separate path of development.
(1) The most obvious reason belongs to the realm of ideas and was the consequence of the application of what proved to be an inappropriate theory of development. According to a widely prevalent approach, development required a radical relative reduction of the agricultural and commodity producing sectors, which could only be achieved through the manipulation of prices and the creation of a different price structure to that prevailing on the world market. Thus a development strategy required the fostering of substitutes for previously imported manufactured products. The apparent success of the Soviet model of development reinforced this theoretical preference.2 Most development strategies envisaged a powerful role for the state in overcoming barriers to development and in planning growth.
(2) Another explanation is concerned with the domestic sociology of politics. The bias against agriculture, which was almost always presented as a necessary kick start of development and as an essential part of any successful modernization strategy, was initiated by urban elites, and as it proceeded built up its own urban vested interest, with a substantial political leverage. Again, the result was to emphasize the state’s role in development.
(3) A final explanation concerns the international order itself. Some features of the international system posed a deterrent to would-be members: for instance, the highly restricted international availability of private capital in the 1950s and 1960s, or sharply fluctuating commodity prices (especially in the first half of the 1950s), or more politically, fear of external influence and dependence.
In most cases, these causes overlapped and reinforced one another. Ideas (in this case about development) not only affect the way interest groups put their demands forward but also actually may move such organizations to form in the first place. Ideas, however, are not born in a vacuum. The existence of problems in the international order encourages people to formulate theories to explain why.
How easy would it have been to reverse these attitudes and policies? A rethinking would only be possible as a consequence of the perception that something had gone wrong. Frequently this only arises in the case of an acute crisis. But by that time the problem may be so deeply entrenched that it is impossible easily to solve. In some cases, where a large internal market and a dynamic entrepreneurial culture existed, import-substitution growth could be sustained at very high rates for a relatively long time period. The most striking example is to be found in the “Brazilian miracle” of the 1930s or again in the late 1960s and 1970s. In such cases, little rethinking even appeared to be necessary.
Could these problems have been tackled by a more forthcoming or better organized international system? The World Bank, the General Agreement on Tariffs and Trade (GATT), and the IMF tried to make access to the system easier, but their efforts were not always successful.
The World Bank tried to make up through its loans for the very sluggish private capital market of the early postwar years, or to spur the private sector into international lending by giving guarantees; but it would never be a perfect substitute for a private market, with its larger capacities and its ability to make a wide range of choices and assessments. For the moment, and for most countries, there was little external private sector investment available. In addition, in regard to its own operations, the World Bank was limited to lending for public sector projects.
The GATT initially was restricted largely to industrial countries. As it became more universal in the course of the 1960s (by the beginning of 1971 it had 77 members), it also attempted to adapt more to the particular needs of developing countries. Article 18 of the original charter allowed developing countries to withdraw concessions on customs duties, and also to give subsidies, if these measures were needed in order to establish new industries that would increase production and raise living standards; they could also take exceptional measures to protect their balance of payments. For their part, industrial countries under Article 37 had committed themselves to reduce or eliminate barriers to the exports of less developed countries. But when it came to putting these principles into practice, the European countries that were members of the European Economic Community (EEC) were unable to agree to a plan to remove restrictions contrary to the GATT; and the GATT instead formulated in 1965 a much vaguer and less substantive Part IV of the Agreement (which became Articles 37 to 39 and diluted the previous commitment).3 In all the most important areas of trade—in agriculture, steel, and textiles—industrial countries in fact evolved restrictive and sometimes discriminatory trade practices. As a result, some commentators began to speak of a “‘conspiracy‘ of noncompliance.”4
The IMF’s task was defined as the provision of short-term balance of payments support. This assumed a great significance in connecting economies with the world system. In the absence of available capital imports, balance of payments difficulties could prove an insuperable obstacle to integration. More fundamentally, analyzing the origins of balance of payments problems could provide a tool for diagnosing more wide-ranging economic difficulties. The balance of payments acted as a fever thermometer. Deficits might indicate the presence of wrong exchange rates or disincentives to export. The practical difficulty involved in responding to the diagnostic tool with short-term support that, to continue the analogy, might only provisionally lower the temperature without curing the sickness.
The Clash of Ideas
The first reason for the separation of many economies from the international system derived from the widespread conviction that an alternative set of economic rules or even a different logic applied to developing countries.5 The differences concerned the appropriate degree of exposure to the international economy and the desirability of domestic financial stability. (These issues were linked, in that the adoption of convertibility and fixed parities would necessarily rule out fiscal experimentation or inflationism.) A difference in understanding about the operation of the international economy, and the associated conviction that the other side was acting out of a fundamentally political logic, profoundly handicapped the IMF in its relations with many developing country members. In order to become more influential, it would have had to demonstrate the validity of an economic approach as much as provide financial resources. Many influential analysts, however, believed that poorer countries would be damaged by exposure to the international system, that emerging manufactures would be destroyed, and that the export of a limited number of commodities would create an intolerable dependence. Access to capital flows would be difficult or impossible.
The tyranny of the past played a part in creating these beliefs. They appeared as lessons of history, and especially of the prewar era. There was, it was thought, a general need for protection. In the environment of the 1930s, the adoption of tariffs and quotas had made sense as an insulation against the spread of deflation.6 And before that, the relatively open international financial system had transmitted financial shocks too easily. This was a preventative argument for intervention in trade, but there was also, it appeared, a powerful developmental case. In this view, trade measures could be used as a way of engineering a transformation of the economic structure. Countries interpreted the interwar experience as showing that, after an initial postwar boom, commodity prices were likely to weaken. In general, a commitment to agricultural production was thought to mean an obstacle to development, and the best hope for advance lay in moving underemployed people and resources out of agriculture as rapidly as possible. Multiple exchange rates had been an aspect of Nazi economic and trade policy in the 1930s that the United States had found deeply objectionable. But they had also been seen, particularly in Latin America, as a way of managing trade so as to promote import substitution and thus promote industrialization, as well as of raising revenue to pay off foreign debt and thus avoiding default. In short, lessons drawn about how to adapt to the dismal world of the 1930s were frequently much more compelling than the high hopes of Bretton Woods that there might be created a better world.
The practical experiences of the interwar period were systematized as an economic theory that explained why primary producing countries would inevitably experience a secular decline in their terms of trade. The historical record (see Figure 5-1), as well as assumptions about the propensity of developed countries to buy foodstuffs and raw materials as incomes rose, led Hans Singer and Raúl Prebisch to claim that the demand for manufactured goods would rise more quickly than for primary products and that differences in demand would be reflected in the evolution of relative prices.7 The policy conclusion derived from this economic reasoning was that only hy shifting substantial resources into the production of manufactured goods could developing countries escape a permanent poverty trap. The theory of the long-run decline of commodity terms of trade provided the intellectual underpinning for the politics of import substitution and for “export pessimism,” a view frequently associated with the United Nations Economic Commission for Latin America. Exports, according to the doctrine, could rarely be an adequate motor for development. Integration into an international economy would damage an initially uncompetitive industrial sector, which however held out the only hope for breaking out of the traps of poverty and underdevelopment.
Price Indices of World Trade
(Logarithm)
Sources: League of Nations; United Nations.Some analysts went further than Singer and Prebisch and asserted that the entire process of development was a political struggle. According to this view, developed countries, in order to be able to appropriate for themselves a greater share of the world’s resources, had used the theory of neoclassical economics as an ideological instrument. Their insistence on comparative advantage and the mutuality of gains from trade constituted a duplicitous and hypocritical masking of their own exploitative interests. Import substitution strategies, on the other hand, might provide an economic basis for the assertion of sovereignty and political independence, as well as self-enrich-ment, by developing countries. If this analysis were valid, the demand for speedy balance of payments adjustment looked like another weapon of developed countries to hold down developing countries below their optimum growth path.
The “structural approach” derived further support from the problem of commodity price volatility. Economists looked at a historical record in which these prices moved much more than those of manufactured products. Sudden and unanticipated drops in the prices of major exports would induce both immediate balance of payments problems and rising fiscal deficits as governments tried to compensate the losses of politically powerful producers. Adjustment programs in these circumstances would force subsidized prices to rise, marginal industries would suffer from the application of credit restrictions, and the exchange rate would fall: all of these responses would injure powerful groups who would challenge the goal of monetary stabilization. In these circumstances, the adjustment program could not affect the basic problem of commodity price instability, but would only make more damaging the consequences and repercussions. Serious remedial action would need to deal with the international issues of commodity prices and commodity trading.
The policy implication involved the conclusion that devaluations should be avoided and that the external economy should be managed through exchange control often with the use of multiple exchange rates (different exchange rates for different categories of product in a licensed trading system). These were a common phenomenon of the 1950s, in every Latin American economy, but also in countries as diverse as France, the Benelux countries, Israel, the Philippines, and Yugoslavia (which with about 200 different exchange rates had probably the world’s most complicated system). In 1958, 36 out of 58 members of the IMF operated such multiple rates. Such exchange control allowed countries to run higher rates of inflation without increasing imports. A substantial further difference in views about economic management concerned the appropriate level of inflation and an assessment of the economic consequences of inflation. The advocates of a protected national development saw fewer dangers in inflation.
In the experience of Europe, the United States, and some parts of Asia (especially India and Japan), inflation had been a legacy of wartime finance and was associated with major price distortions and in consequence irrational and misguided investment undertakings. A return to balanced economic development, in which price signals could be used as a basis for investment decisions, required a reduction of inflation rates through fiscal discipline and appropriate monetary policy. But Latin America had not been so directly affected by wartime finance and as a result developed a much more relaxed attitude to inflation. Investment often occurred, especially since the 1930s, under state direction. When a boost in investment was required in order to set the economy on the appropriate course of development, state expenditure would necessarily expand.
Inflation was thus seen as a product of “structural” political and economic forces that led to budget deficits and credit expansion. It would be misguided to intervene in this network of social pressures, and the cost would be economic and political collapse. Inflation acted not as an incentive to misallocation, but rather as a stimulus to development, which might remove obstacles such as labor immobility, disguised unemployment, or production bottlenecks.8 Nor were Latin American economists alone in proposing such an interpretation. They derived a substantial amount of support from economists in Western countries who told them that “a case could be made for making inflation an instrument of policy, rather than the control of inflation an object of policy.”9 Adjustment programs thus were seen as bearing no long-run benefits at all but only imposing the short-run costs of production lost and opportunities forgone. The result was a theory—and a political practice—that has been termed “economic populism”: “an approach to economics that emphasizes growth and income distribution and deemphasizes the risks of inflation and deficit finance, external constraints, and the reaction of economic agents to aggressive nonmarket policies.”10
In 1959 the Secretary-General of the United Nations, Dag Hammarskjöld, appeared to endorse this populist and proinflationary view when he asked thetorically: “Are we not, perhaps, rather inclined to solve the conflict between stability and growth too exclusively in favor of stability, to the detriment of the vigor and dynamism so characteristic of the world economy during the first post-war decade?… The issue cannot be sidestepped by simple resort to global measures of fiscal and monetary control, without serious risk that the economy may lose all forward momentum.”11
By contrast, the IMF in particular seemed to hostile observers (who were numerous) intent on applying the narrowly anti-inflationary and fiscal view of the 1920s League of Nations when it came to the discussion of currency stabilization. In the 1959 Annual Report the IMF stated its approach in the following way: “Notwithstanding the realization that is now fairly general that sound economic development is not compatible with the distortions that rapid or chronic inflation always creates, a number of the less developed countries have had great difficulty in abating or slowing down the rate of inflation.… The temporary deterioration of the standard of living, which in such circumstances [a stabilization program following a period of low saving] is inevitable, may be interpreted by some sections of the public as an indication of the failure of the program, and give rise to claims for prompt upward adjustments in wages and salaries and for more liberal credit terms, which, if granted, will again generate inflationary pressures.”12 The 1962 Annual Report concluded: “Experience shows that, if prices rise fast enough to initiate a wage-price spiral or to undermine confidence in the real value of assets denominated in money (e.g., savings deposits), economic growth will be discouraged.”13
Inflation in this picture encouraged the misallocation of investment. According to the IMF view, inventories would become irrationally large, as money could no longer be relied on as a store of liquidity. Funds would be channeled into luxury housing, and, as the exchange value fell, into foreign assets, despite attempts that governments might make to control capital flight. Inflation would deter foreign investment; and the fall of the exchange rate would stimulate excessive import substitution.14 In the light of these arguments, the best approach to stabilization lay in a shock therapy, since a rapid action stood a greater chance of breaking inflationary expectations, while “an attempt to slow down an inflation will take a long time to be effective and its final result will be uncertain.”15
This was precisely the claim that at the time most development economists found objectionable. They saw inflation as a desirable source of investment funds. Fighting it would reduce investment and the capacity for growth. If inflationary options could only be followed away from the international economic system—then the sacrifice was not a great one. In face of the argument that stabilization would create a more rational investment climate and attract foreign inflows, they did not need to be complete cynics to see that those resources might not in reality be so easily available. As Richard Cooper pointed out in 1968, “it is a perverse characteristic of international capital that it fails to move when it is most needed.”16
The Sociology of Politics
As the alternative and mercantilistic ideas about development came to be applied in practice, they evolved their own political momentum. Once programs for import substitution were under way, they created social groups with a vested interest in their continuation, who saw greater openness as only bringing losses for them. Industrialization created urban centers with apparently higher wages and conditions of work than prevailed in the countryside. The towns then exercised a powerful pull on the rural labor force. The new businessmen and workers were well placed to press governments for further concessions to their particular interests.
The initial impetus toward import substitution and the distortion of the domestic terms of trade against agriculture (the traditional beneficiary of state policy in many colonial states as well as in Latin America) created a powerful and well-paid urban working class and a strong interest in import-substitution industrialization. An overvalued exchange rate (which made agricultural exports less profitable and allowed cheap imports for specifically targeted groups) was frequently used to transfer income away from traditional agricultural producers.17 The exchange rate became a vehicle for forcing social change. The penalization of agriculture, and a de facto subsidization of urban consumers, led as an unintended consequence to the creation of greater rural poverty and to increased disparities of income and wealth. Those countries that applied import substitution most systematically as a result developed far more inegalitarian societies. In general, there was a remarkable contrast in this respect between import-substituting Latin American countries, and East Asian societies, which largely abandoned this strategy as early as the beginning of the 1960s (see Chapter 14).
Social inequality helped to produce increased political instability, strained democratic systems, and increased the likelihood of coups. Political instability in turn had economic effects: it encouraged the constant formulation of “new” or different approaches. In this way, it played a part in the process that produced rapidly alternating policies and economic performance and that helped to reduce the overall long-term rate of growth.
Perhaps the most dramatic example of this process was Argentina, where the case for the import-substituting strategy had been clearly enunciated by Prebisch, and a practical application had been implemented already earlier, in the 1930s, and then more dramatically immediately after the Second World War by the Peronist regime. After 1958, governments took an opposite tack and attempted to liberalize, decontrol, and integrate into the world economy with the support of IMF programs. The German currency reform and liberalization of 1948 and the Spanish reforms of 1959 were used as an explicit model; and the new course was supported through an IMF stand-by arrangement. “We must achieve,” Arturo Frondizi explained, “an Argentine miracle; we must show that we too are capable of transforming despair into hope and the ruins of defeat into great victories.” Alvaro Carlos Alsogaray, who in June 1959 became Minister of Labor and the Economy, was an explicit admirer of Ludwig Erhard. Alsogaray believed that German-style liberalization with a single exchange rate and a control of the money supply exercised only through changes in banks’ reserve requirements would produce the Argentine Wirtschaftswunder. In the particular circumstances of Argentina, such reforms would attract foreign capital. Roberto Alemann, who in 1959 was Under-Secretary in the Finance Ministry, later made this argument into a general theory of development, in which domestic liberalization provided new incentives for international capital inflows: “International cooperation leading to the elimination of restrictions on trade and on international long-term lending, as well as to the amelioration of the stimulus given by developed countries to investments in their own economies, should be explored, in order to reestablish a vigorous flow of genuine long-term private capital.”18 In Argentina, however, the attempted stabilization quickly collapsed, and the country began a cycle of reform, failure, and the reassertion of traditional populist policy priorities. The consequent extreme instability produced very sharp economic fluctuations and a very poor overall growth performance.
The Politics of the International System
In the 1950s, countries described at the time as “underdeveloped” grew quickly, but their growth rates were similar to, rather than markedly higher than, those of industrial countries. They appeared to have missed the process of “catching up” implicit in the notion of development. The widespread character of development did nothing to discourage the thought that the potential for growth might be greater outside the world economic system.
In addition, decadal or other long-term average figures on growth miss one of the most important features of 1950s and 1960s development: its jerky character in most developing countries. At the same time as some economists in developed countries were speaking of the demise of the business cycle, outside the industrial world there could be no doubt that growth rates were very volatile. The responsibility, however, lay, according to most observers, not in endogenous cyclical factors but rather in constraints imposed by the international system. Intensive growth spurts led to the emergence of balance of payments problems and the need to adjust by slowing down. The dislocations caused by joining the world economic system were most directly reflected in the external balance.
The IMF, as an institution immediately concerned with the management of payments, found itself gradually drawn into a much wider discussion of development issues. As a result, it began to function as a kind of doorkeeper to the international financial system, and quickly came to find itself dealing with issues much broader than simple short-term balance of payments problems. In particular, it necessarily faced the following questions posed by the theorists of development:
(1) Does development necessarily result in current account deficits, as a consequence of the need to import raw materials and capital goods?
(2) Does development necessarily result in fiscal deficits, as the state supplies saving that cannot be provided by an inadequate private sector? The first two hypotheses, as expounded by Hollis Chenery, became known as theory of the “two gaps” (the payments and savings gaps).19
(3) How should the gaps be financed? In particular, are private inflows of investment capital appropriate, or does the long time frame implied by the process of development rule out any involvement of private capital except on conditions in which it is given a substantial measure of control? And would that control be politically unacceptable by bringing a limitation of national sovereignty, which had just been obtained in many countries with great sacrifice and was passionately defended?
(4) If the capital flows were to be regulated by international government action, equally troubling questions would arise. How could recipient countries avoid the intrusion of political calculations or inappropriate economic conditions (such as donors’ insistence that aid be tied to the purchases of their own national products)?
(5) All these issues were logically connected in a highly politicized web. Supposing that private flows only come if they are shown a green light by official capital flows and that official flows are made conditional on the establishment of policies under the supervision of international institutions to encourage long-term and stable inflows of private funds? Or supposing that the answers given by multilateral institutions to short-term balance of payments problems are associated with recommendations designed to produce long-term solutions in which private and official flows play an important part? In these cases, dependence seemed to be reimposed, as a result of economic interconnectedness, and the promise and goal of political independence might be undermined.
Such linkages create some of the most painful political problems arising out of the development process. Discussion of the balance of payments as a constraint on development exposed some fundamental problems. If advice from the outside comes differently from different sources, it looks inconsistent. On the other hand, if it emphasizes the linkages between different elements of a reform strategy, it looks too well coordinated and too intrusive. If the same sorts of conditions are set for assistance by multilateral institutions, industrial countries, and private lenders, they create the suspicion that there is some sinister “external interest” aimed at the subversion of national priorities. If advice from the outside is to be effective, or even if it is to be implemented at all, it needs to be accepted by national governments and civil servants. It is most useless when imposed by external authority and most useful when it comes as the result of a common search for a solution to a problem that represents a general concern. The likelihood of a long-term success is greatest when a common framework for analysis is established, and in this way the suspicions that attend on external advice are dispelled.
From the beginning, one of the major functions of the IMF had been concerned with the transmission of ideas. An international institution can play a major part in national decision making not so much by intervention from the outside or even through financial support (though this was frequently crucial in the implementation of new economic strategies) but by bolstering the position of reformers in the bureaucratic structure, usually in the finance ministry or the central bank. The best way of doing this was to supply accurate information about world economic developments and helpful ideas. This side of the IMF’s activities acquired some new institutional dimensions in the 1960s. In 1964, the IMF Institute was created as a training institution; also in 1964, the Central Banking Service was launched, offering advice mostly to African countries. In 1969, the Bureau of Statistics began to provide technical assistance to help member countries to improve data collection and establish or extend central bank bulletins.
Already earlier, in the 1950s, IMF missions had developed into one of the most obvious ways in which the institution expressed its views to the outside world. The mission was usually composed of a relatively small team (four to six people) with representatives of the IMF area departments involved (these had been created in 1953), as well as of departments with a particular concern (Exchange Restrictions, Central Banking, or Fiscal Affairs). This team negotiated on the basis of an internal briefing paper agreed in advance through consultation and discussion in the Fund, and on its return produced a report for the Fund management that would then be the subject of an Executive Board discussion. This was both a way of investigating the external economic problems and difficulties of member countries, of establishing conditions on which financial assistance could be given, and of transmitting views and suggestions between the member countries and the Fund.
Even an agreement and a shared intellectual analysis is sometimes not enough. Often high short-term costs of adjustment are required in order to achieve the long-term benefits of the confidence established by the common outlook and vision. But when these costs are too great, they also can exercise a lasting and damaging effect on confidence and on the credibility of the political and administrative leadership. In order to be credible, policies require a certain commitment on the part of governments that there will be no sudden and dramatic reversal. Where growth produces instability, and political change leads to international upheavals, it is hard to establish commitment and credibility. This in turn makes the formulation of an appropriate policy more difficult and leads to an orientation toward the short term.
What are the chances of establishing a common, universal framework that will increase the chances of stability? Every country has its own individual problems, complex and peculiar. International orders depend on the creation of universal rules and principles; such was the great truth of Bretton Woods. But treating countries according to general rules may look harsh and inflexible in the light of the remorseless individuality of the suffering and sorrows of mankind. Such questions became acute as the general principles established at Bretton Woods began to shape not just the re-establishment of European currency convertibility, but in the late 1950s also the discussion of liberalization in the rest of the world economy. Providing answers to the dilemma of how to reconcile a universal institutional framework with the need to deal with the specificity of particular problems in individual countries provided a continual challenge to the system established at Bretton Woods.
Almost every non-Western country had a deep-seated and quite justifiable suspicion of imperialism and its legacy. In some cases, anti-imperialism strengthened the initial appeal of international multilateral economic institutions. Particularly in the early 1960s, as many African states became independent, they found working with the IMF and the World Bank an effective way of shaking off the patronizing advice of finance ministries and central banks from the former imperial powers. The Bank of England, for instance, systematically resisted attempts to set up new central banks and preferred to advise on the operation of currency boards or more limited types of monetary authority.20 The creation of central banks required, as a result, the technical assistance of the IMF.
In other cases, however, anti-imperialism soured the relations of poorer states toward the international financial system and international financial institutions. Many states believed, rightly, that there was a close involvement of the United States in many Fund and World Bank programs. Where American political power was believed to be a problem and a threat and where that perception generated a nationalist response, the role of the Bretton Woods institutions almost inevitably became highly problematic. A great deal of the IMF’s activity relating to non industrial countries took place in the western hemisphere; but this is where the attractions and the suspicions of U.S. power were at their greatest. The U.S. government, which historically had been very closely involved in Latin American affairs, felt vulnerable to the criticism that much of its aid to the hemisphere was being wasted or frittered away. It began to make some aid programs dependent on reaching an agreement with the IMF, which would guarantee the pursuit of sound policy, and might encourage also the involvement of private capital. Thus Peru concluded in February 1954 a $12.5 million stand-by arrangement with the IMF, borrowed from the U.S. government, and simultaneously established a substantial credit line with the Chase National Bank of New York. But such linkages created suspicions in many Latin Americans about the politics surrounding IMF involvement and recommendations. The influential Brazilian economist, Celso Furtado, for instance, claimed that the IMF was a “premature” international creation, which “operated primarily as a U.S.-serving control instrument over the economic and financial policies of other countries, especially the so-called under-developed countries.”21
At the same time, a growing international consensus insisted that growth was a natural, desirable, and steady long-term process. Academic observers insisted that the “tricks of growth are not all that difficult; they may seem so, at moments of frustration and confusion in transitional soci ties.”22
International unease about the limited success of development strategies resulted in new initiatives to remove “obstacles,” establish “preconditions,” and encourage faster “growth.” The United Nations declared the 1960s to be the “development decade.” The World Bank had already successfully used its relatively modest capital as a basis for a great expansion of lending using borrowed funds. Eugene R. Black (President 1949–62), a former Vice President of the Chase National Bank, had a very solid reputation in the financial community and set about using the guaranteed capital of the World Bank as a basis for financial intermediation. The Bank with its government-guaranteed capital could borrow from private sources on very favorable terms and raised money in the 1950s substantially on the abundant U.S. capital markets. In the 1960s, as the U.S. payments position deteriorated, it then turned to European (chiefly German) and then Japanese and Saudi Arabian markets for funds. The Bank, in effect, stepped in where markets were nervous because of inadequate information, and educated lenders as well as helping its borrowers. It played in this way a major role in the development of international capital markets, whose operation had been curtailed and whose confidence had been profoundly shaken during the interwar world economic crisis. The World Bank now encouraged markets to respond to the need for world financial flows, an encouragement particularly needed in circumstances when banks were nervous and hesitant about their foreign exposure.
These operations were conducted at market rates and allowed borrowers access to funds that they would have otherwise been unable to obtain. But there was also a need for funding for projects where the return would not be as immediately visible, at least in financial terms. Particularly in the case of very poor countries, borrowing on commercial terms was out of question. After a long debate, the World Bank in 1960 created the International Development Association (IDA) as a “soft loan” institution, with “a more benign personality than the Bank.”23 It was financed through subscriptions (called “replenishments”) from the richer member countries and gave credits with a substantial grant element for more “social” objects than the Bank, with its insistence on commercial returns, could provide.
In practice, the World Bank in the 1950s and 1960s had developed into a major development agency. It also encouraged emulation. In 1958, the European Development Fund was set up, for development aid, most of which went to francophone Africa, as well as the European Investment Bank, which supported projects for regional development within the EEC. Much later, in 1990, another European institution was added, the European Bank for Reconstruction and Development, to assist economic restructuring in Central and Eastern Europe. In 1959, an Inter-American Development Bank was established. The African Development Bank began operations in 1964, to provide a “concrete contribution to the human drive against backwardness,” but soon complained bitterly that the World Bank was taking away the better projects.24 The more successful Asian Development Bank, which started in 1966, was the outcome of a number of Japanese initiatives going back to the 1950s, and was modeled very directly on the World Bank.25 One of the principal attractions of the newly created regional banks was to attract more resources than could be made available by the Bretton Woods institutions. They were both “a tribute and a rebuke” to the World Bank: a tribute in that their operating principle was based on the same model, but a rebuke in that they reflected a feeling that the World Bank loans were insufficient and that the Bank was not sufficiently sympathetic to ambitions for regional integration.26
Approaching the International System
At Bretton Woods, few had anticipated that poorer countries might borrow, In the optimistic environment of the late 1950s, some began to assume that countries could solve their balance of payments difficulties relatively easily and that as a result the door to the international financial system would readily swing open, as capital markets became wider and more experienced. Only a limited amount of initial assistance would be required. The reality was much more problematic. The existence of balance of payments problems in poor countries involved the Fund in a whole range of development problems, led to an overlap and potential clashes with the World Bank, and to a discussion of relationship of the Bank to private capital flows. As long as the Fund had been largely inactive the ambiguities inherent in the delimitations of spheres of activity in Bretton Woods remained unproblemati-cal;27 with greater Fund activity it was bound to become clear that short-term balance of payments problems did not constitute a logically distinct problem and could not be treated as such.
In the aftermath of the 1956 political crisis of Suez and the resulting need of the United Kingdom and France for balance of payments assistance, IMF activities expanded quite dramatically with regard also to other countries. Total drawings rose substantially even apart from the major resources devoted to France or the United Kingdom: from $27.5 million in 1955 to $131.5 million in 1956 ($692.6 million if the United Kingdom is included) and $977.1 million in 1957, but then falling again to $337.9 million in 1958 and $179.8 million in 1959. In fact, measured as a share of world trade, drawings on the IMF were at their highest ever in the decade that followed 1956 (Figure 5-2). Most of the assistance involved developing countries.
Drawings on the IMF as a Share of World Imports
(In percent)
Source: International Monetary Fund.The new expansion of activity frightened many observers, and especially many Americans. The Fund appeared to hostile observers as being sucked into long-term lending, or into areas that commercial banks would not touch. The IMF’s new Managing Director, Per Jacobsson, told Jean Monnet that the United States was concerned that the exhaustion of quotas would be followed by a demand for a quota increase and that almost every one except the Americans (who could veto it because of the Articles’ requirement of special majorities) was in favor of such an increase.28 The conservative University of Chicago economist Melchior Palyi took the opportunity to restate the bitter accusations he had already made during the debate over the ratification of the Bretton Woods Agreements. Now he said that he feared a sudden burst of profligacy on the part of the Bretton Woods institutions: “The ineffectuality of both … and the very limited aid they provide have caused so much criticism among debtor countries as to make their survival questionable—or the deterioration of their credit standards a virtual certainty (which is what is happening since the Suez crisis).”29
The work of the IMF in this initial period of activism was based on the hope that developing countries would soon participate in the general movement that was so evident in Europe away from inflation, exchange controls, and multiple rates and toward liberalization and the growth of international trade. Per Jacobsson in particular embraced this activity with a missionary’s enthusiasm and energy. The campaign against multiple exchange rates produced major successes, and after 1956 a movement toward unified, but often floating, rates began, with Bolivia, Chile, and Iran leading the way. In 1957 an IMF Executive Board memorandum encouraged the process of simplification of exchange rate regimes. But financial resources were also required in the transition. Jacobsson realized that it would need a substantial financial commitment. In 1957, for instance, when India encountered an acute balance of payments crisis, Jacobsson became enmeshed in a struggle for a very large Indian stand-by arrangement. Eventually, the amount of the support was scaled down from the $200 million proposed by Jacobsson to $72.5 million at the insistence of the United States.30
Fund programs, it was believed, would create a new incentive structure to encourage stabilization, and stabilization would encourage long-term capital inflows. The purpose of systematic intervention by the IMF was to compensate for temporary falloffs in the long-term movement of capital. Per Jacobsson set out this view most explicitly during the Annual Meetings of 1961: “Insofar as such development is financed by long-term foreign capital, the actual use of these resources will normally be reflected in a deficit in the current account of the balance of payments. Provided that the other elements of the balance of payments are in equilibrium, this current account deficit would be matched by the external financing, and there would then be no over-all deficit.” Where this was not the case, Fund financing over a three-to five-year period would “provide the time needed for the appropriate measures to take effect, and thus to relate the long-term development to the available long-term finance.”31 Maintaining the five-year limit on the availability of Fund resources was a crucial element in the Fund’s mission to deal with short-term balance of payments problems by the provision of revolving credits. Longer-term credit would be supplied for specific projects by the World Bank or by private capital. The Fund should not, however, become too closely involved with the activities of the private banks, and international institutions should not try directly to give information to the markets. This issue arose particularly dramatically in the case of Argentina, when after an apparently successful stabilization attempt, foreign banks requested that the IMF take a direct part in debt rescheduling and restructuring.
European banks in particular were quite nervous about any new overseas lending. They had had little recent experience with overseas lending except in financing some commercial activity. Argentina had only just reached an agreement providing for the multilateraliation of payments and nondiscrimination in trade with Europe, as well as for the consolidation and repayment of debts over a ten-year period (the Act of Paris, May 30, 1956, which created an initially loose institutional mechanism, known as the Paris Club, for renegotiating official debt). In 1959 some European banks concluded a syndicated $70 million Argentine credit. But before they would be willing to go farther, they sought the involvement of international institutions in order to supply an additional measure of security. First they appealed to the IMF to act as a trustee for loans to developing countries. One European banker explained to Per Jacobsson that the banks wanted “a certain amount of moral support, which, under certain circumstances, might become very important, particularly if there should be a change of government in Argentina.” European banks did not have the extensive machinery of the U.S. government to follow the economic and monetary policies of non-European states and also did not have the experience of New York banks in Latin America. Jacobsson refused to allow the IMF to be used in this way: “Where what 1 would call ‘parallel’ credit arrangements … have been made by the Fund, and other international organizations, other countries, or private banks, each party considering the extension of credit should make its own decision on whether or not to make its funds available, and on what terms.” The Fund needed to retain the confidence of its members that all its decisions on drawings were “impartially arrived at on the basis of its own practices and policies, and influenced only by the appropriate technical considerations.”32 In order to be effective, the Fund had to dissociate itself from both commercial and political calculations, and establish its identity as a reservoir of economic skill.
Apart from institutional and financial assistance, countries required an appropriate intellectual and theoretical framework with which to interpret and analyze their economies, including the international aspects. In the course of the 1950s, the IMF developed an approach that linked balance of payments issues to monetary and fiscal behavior. The major breakthrough came as a result of a mixture of theoretical and practical work by J.J. Polak. Traditionally Keynesian analysis had neglected the external sector. In an article written with William H. White in 1955, Polak took a traditional Keynesian model but widened it so as to examine the money supply as influenced by the foreign balance.33 In the same year, Polak headed an IMF mission to Mexico, which worked very closely with the Bank of Mexico to invent a technique for ensuring external stability and avoiding a new devaluation of the peso. The result was a practical formula: the money supply should expand at a rate not faster than growth of real GNP. In Polak’s report, he set out a way of estimating the amount of money that could “safely” be created over a four-year period, based on estimates of output and of the increase of foreign exchange reserves and loans to the government. On the basis of these estimates, the amount of credit available to the private sector could be calculated and compared to the credit that would be created by the application of different marginal reserve requirements for banks.34
This model evolved as a practical matter of consensus between the Fund mission, the central bank, and private bankers in Mexico. Its most fundamental policy implication concerned credit creation, or the interaction of the central bank with the banking system. Bankers were willing to accept the high marginal reserve requirements imposed by the Bank of Mexico because they too wished to avoid the uncertainty that would result from a new balance of payments crisis and a new devaluation. One year later, another Fund mission reported as a personal impression that: “The Mexicans value very highly the technical report of our 1955 mission to Mexico, although the report did not contribute very much to the change of policy in Mexico. The Mexican authorities had previously embarked, long before the mission went to Mexico, upon the policies which are described in the mission report. They acknowledge, however, that the report made a significant technical contribution to the full analysis of the Mexican problem.”35
In 1957 Polak gave a formal and systematic statement of the new approach, which started with the observation of the “embarrassing inability” of analysts to make forecasts of the effect of monetary expansion on the balance of payments.36 Polak’s model provided, on the basis of estimates of capital movements, exports, and domestic credit creation (“the ultimate variable”) a basis for predicting the development of income, money, and imports. The paper was concerned with two levels of analysis. Theoretically, it showed how the behavior of the Keynesian multiplier was affected by an open economy, in which exports offered the possibility of income gains. Second, the discussion consistently paid attention to practical issues and problems, and in particular to the puzzle of why a country with increased exports does not continue to show payments surpluses in subsequent years. The critical concept was domestic credit expansion.
Polak came to the conclusion that in the absence of domestic credit expansion an increase in exports would lead to an equivalent increase in national income, an increase in the rate of imports, and an increase in the quantity of money. He then investigated the effects on the model of a long-run increase in credit expansion. This would produce the same increase in money income and money stock as would have been produced by an equivalent increase in exports, and also would lead to an increase in the rate of reserves equivalent to the increase in credit expansion. It would, however, have a very different effect on reserve behavior: the rate of reserve loss would approach the rate of credit expansion, and the loss of reserves would be equivalent to the total credit expansion that had taken place minus the increase in the quantity of money.
The argument contained important implications for monetary and trade policies. In particular, Polak had demonstrated that an increase in exports would only lead to a short-term improvement in the balance of payments and that import restrictions could not correct a balance of payments deficit in the long term. (The initially favorable effect would disappear once the income effect was realized, and increased incomes led to increased demands for imports.) Countries could control the behavior of their reserves through monetary policy, and the behavior of domestic credit expansion. This was also a rather practical model, in that it required for its application a knowledge of trade behavior and of domestic credit expansion, but did not require national income calculations (which would have been much harder to make in countries with only rudimentary statistical services). The model was later extended to encompass capital flows and to be applied to the problems of industrial countries.37 It was a crucial step in the development of the theory of international finance and of the integration of Keynesian and monetary economics.
Without such a framework for analysis, Jacobsson’s insistence on monetary stability as a fundamental part of a liberalization strategy would have been purely thetorical and practically ineffective.
Too often, unfortunately, the incentives provided by international institutions for the integration of their members in the international system and the economic analysis produced results that were disappointing in practice. A study of 36 trade reforms carried out in the 1950s and 1960s in developing countries showed that only 15 of the reforms were fully sustained.38 In many countries the economic pendulum constantly swung periodically and quite violently between enthusiastic acceptance of programs, slippage, negotiated modification, rejection, rapid growth following abandonment of the program, increased inflation and balance of payments difficulties, stagnation, and then eventually the elaboration of a newly agreed package. The attractions of an alternative theory of development to that involved in the Fund’s analysis provided a constant siren song, an inducement to break with rather than cooperate in the international system.
More Than Balance of Payments
By the mid-1960s a substantial and vociferous criticism had emerged in member countries, in newspapers, and among academics of the IMF’s approach to stabilization. The London Economist had written in 1961 that the Fund’s stabilization programs in Latin America had been so harsh that Per Jacobsson might be described as “Mr. Khrushchev’s secret weapon.” “There is little sense of shared purpose between the fund and the governments it is helping; instead, Latin American economists see the strict orthodoxy of the fund’s tenets as a challenge to them to find ways of outwitting the lawgiver.” The article concluded that “it is feared that if restrictive monetary policies are adopted without social safeguards, the countries run the risk of serious social eruption.”39
In November 1962, in the middle of protracted and difficult negotiations for an IMF stand-by arrangement, Brazil introduced in the United Nations General Assembly a resolution calling for a thorough empirical and analytical study of inflation and its effects in “under-developed countries,” and which implicitly condemned the IMF for ignoring the structural factors that had led to sustained inflation. In its original form, the resolution claimed that “inflation in under-developed countries is often aggravated by the process of urbanization and by the incorporation into the market economy of populations previously living in a subsistence economy, thus requiring difficult structural adjustments and creating additional pressures on the available supply of goods and services.” The resolution gained the additional support of Bolivia, Hungary, and Tunisia (though with the terminologically interesting difference that it now referred to the problems of “developing countries”).40
In 1963 an internal IMF staff memorandum concluded that it was “not too strong to say that the Colombian case tends to support many of the recent criticisms of [Fund] stabilization programs.”41 In 1966 the World Bank prepared a draft paper that argued that the balance of payments difficulties caused by development constituted a long-term and not simply a transitory phenomenon, with the result that short-term stabilizations necessarily imposed an excessive cost. The paper included bitterly critical passages arguing that long-term confidence in developing countries had often been undermined by frequent exchange rate changes, that the IMF’s approach had had the effect of discouraging savings, and that “a certain amount of inflation seems to be unavoidable in the development process, in which hitherto unemployed or underemployed resources have to be more fully utilized.” In conclusion, the authors recommended an urgent reconsideration of the IMF’s approach to the performance of developing countries.42
In some cases, most notably India in 1965–66, the World Bank seemed to be taking over some of the functions of the Fund by default, as a result of the inadequate financial capacity of the Fund in dealing with large-scale problems. In India, 1965 and 1966 were years of catastrophic rice and wheat harvests following monsoon failures, as well as massive dislocation as a result of the 1965 war with Pakistan and the temporary suspension of work by the consortium established in the 1950s with World Bank and donor countries’ participation. These crises indicated a deep-seated problem, but also required the much more conventional and limited solution of an IMF stand-by arrangement (of $200 million), a devaluation of 36 percent (largely at the suggestion of the World Bank), and $900 million provided by the consortium after pressure from the Bank. The 1966 World Bank India package, though dressed up as project lending, in practice amounted to support in a balance of payments crisis. During the discussion in the Bank’s Executive Board, one Director complained that “we are doing here the job of the Monetary Fund, which has the responsibility of financing short-term balance-of-payments deficits. The Bank has no such responsibility.”43
The increasing overlap between the activities of the Bretton Woods twins in the course of the early 1960s required some institutional or bureaucratic response. After a lengthy discussion between the Bank and the Fund, the two eventually reached an agreement in 1966 about procedures and cooperation, intended to reduce the frictions between the two institutions. According to this concordat, the Fund had primary responsibility for exchange rates and restrictive systems, adjustment of temporary balance of payments disequilibria, and financial stabilization. The Bank would deal with development programs and the evaluation of projects. The concordat also recognized the reality of the overlap, although its attempt to deal with it remained rather on the level of pious sentiment. Neither institution would ignore any aspect of the “‘structure and progress’ of member countries.”44 In other words, monetary stabilization and growth-oriented structural policies should not be considered in isolation from each other: though in practice, the balancing of stability and growth remained a contentious issue, not just in developing countries.
The criticisms directed against Fund programs that had emerged in the course of the turf war with the Bank could not be dealt with simply by the piecemeal tinkering accompanied by high-level diplomacy that essentially had characterized in particular Per Jacobsson’s distinctive and initially very promising approach to creating a single world economy. The attention of analysts turned to the structure of the world economy as a whole. These debates were inevitably reflected in discussion of the role of the Bretton Woods institutions, as well as in internal debates. During the course of the 1960s, two major problems stood out: the volatility of export earnings and the problem of reserve inadequacy. Both these issues became the subject of discussion, investigation, and reform initiatives both within and outside the Fund.
Commodities
A major demand of commodity producers reflected the experience of the volatility of the 1950s: first the Korean war boom, and then the collapse of prices. One of the most obvious problems lay in the way balance of payments difficulties emerged out of large and unanticipated fluctuations in export earnings that resulted from dependence on commodities. An important element of the initial Keynes proposals for an international bank had been concerned with the problem of commodity prices. Could not the volatility of the commodity market be reduced? The most radical solution involved intervention in price-setting. The stabilization of commodity prices was a major goal when developing countries in 1964 created the United Nations Conference on Trade and Development (UNCTAD). But, in practice, for the first nine years of its existence—until a commodity boom in the early 1970s created very different economic and also political conditions—UNCTAD produced no major initiatives on commodity prices. A less dramatic approach involved creating a support mechanism for countries that suffered from the effects of fluctuations, so that the vagaries of prices would not disrupt the process of development. In 1963 the IMF established a facility for the compensatory financing of export fluctuations, which was extended in 1966. Under this initiative, funds were made available if exports over the previous 12-month period had fallen short of the estimated medium-term level of exports defined as a moving five-year average. Brazil in 1963, with a $60 million borrowing was the first country to make use of this side of the new approach.
In 1967, at the Annual Meetings of the World Bank and the IMF in Rio de Janeiro, France and 14 African members of the franc zone submitted a resolution calling for the stabilization of commodity prices. A report produced in 1969 as a result of the Rio meetings recommended temporary financing of buffer stocks, and the first use was made in 1971 to fund a stock of tin maintained under the Fourth International Tin Agreement.45 But this initiative was overshadowed by the currency instability of the early 1970s and the dramatic development of the oil price in 1973.
In general, the historical experience with buffer stocks has been highly disappointing: both in the interwar period, when some schemes had collapsed spectacularly because of insufficient funding for the buffers, or because, for political reasons, the prices were set unrealistically high; and in the postwar era, when for a long time the only major international agreement involving buffers concerned tin.46 (There were few commodities in which any workable international agreement had been achieved. The others—wheat, sugar, and coffee—depended on a mixture of price limits and export quotas.) However, tin prices were, with the exception of a brief period of stability between 1956 and 1960, slightly more volatile after the 1956 International Tin Agreement than before, and no less volatile than the unstabilized prices of other commodities. Tin, with optimal conditions for such a solution (relatively few producers, a relatively constant industrial demand, little deterioration over time, and lower storage costs), scarcely held out an attractive precedent for the use of buffer stocks in order to reduce price fluctuations.
Reserves
The most far-ranging initiatives concerned with development involved the creation of new reserve assets as part of an overall strategy. Balance of payments crises in developing countries had been particularly severe and especially disruptive. Would the most economically efficient solution not lie in a channeling of reserves created to deal with the international liquidity issue to developing countries? This discussion proceeded in parallel with an increased concern in the 1960s with the general issue of international liquidity (which is the theme of the next chapter). One vision of what eventually became the SDR (special drawing right) saw an important role for what came to be called the “link” with development, on the grounds that the effects of a liquidity shortage in dealing with short-term balance of payments problems were felt most acutely outside the developed world.
Some of the early practical and academic suggestions for liquidity creation contained a quite explicit link. In 1958 Sir Maxwell Stamp proposed a scheme under which the IMF would create liquidity by issuing certificates that would be given to an aid-coordinating agency and used to purchase products from industrial countries. In 1962, Stamp gave this idea a more explicitly British twist by adding that the spending of the developing countries should take place specifically in those developed count ies which, like Britain at that time, could be held to be suffering from an underemployment of resources. In 1960, Robert Triffin’s scheme for monetary reform included the issue of certificates by the IMF that would be used to purchase World Bank bonds or equivalent securities. In 1966, Tibor Scitovsky proposed that reserves should be created for the purchase of goods by developing countries from developed countries with balance of payments deficits and underemployed resources.47
In establishing the SDR, the IMF groped for a way of reconciling the interests of developed and developing countries. The many failures to sustain heroic efforts at domestic stabilization, the idealistic discussions of the “development decade” in the early 1960s, and the resentments and disappointments engendered in the immediate aftermath of colonial imperialism, all had created the impression that there existed a difference of interest between two groups of countries. In the late 1960s, the political complexities grew. Clearly articulated distinctions and differences based on conceptions of national interest emerged in the debates about changes in the international monetary system between developed countries. The behavior of the major industrial countries, and in particular of the two reserve centers, seemed to spur others to consider the international order as the playground of national and power interests.
When the SDR was eventually launched, and the first issue of SDRs made in 1969, there were only rather oblique references to the concept of the “link.” The SDR had been created as a response to a perception that there was an international shortage of liquidity. The major evidence that there was indeed such a shortage came from the prevalence of balance of payments problems, as well as the ever-increasing share of foreign exchange (largely U.S. dollars) in official reserve holdings, “The main indications of reserve inadequacy in these years [after 1965] lie in the increased reliance on restrictions on international transactions and the increased recourse to international financial assistance, bilateral and multilateral, for the purpose of meeting payments deficits and sustaining reserves.”48 The problems of development were used here to explain the SDR creation, but SDRs were kept within the universalistic framework provided in the original legal framework of the Articles of Agreement before the First (SDR) Amendment. There was as a result no reference to any theory of a “link.”
In retrospect, both the attempts to alter the system in the 1960s, by attempting to control commodity prices through such schemes as bu fer stocks, or by using international reserves as an instrument of development aid, appear misguided. One of the most important insights of the classical economic tradition—it was formulated by Adam Smith and Sir James Steuart—is that every country is a “developing” country in that its economic structure is constantly undergoing change. If this is true, then the problems of development should be treated less as peculiar difficulties with their own separate logic, than as general economic issues. It would be more promising to reduce the extent of dependence on particular exports through the development of a broader range of export products (including manufactures) than to engage in the construction of elaborate support schemes. It would be better to open more economies more rapidly to international capital flows than to make use of the very small amount of external assistance that might have been provided under any conceivable version of the “link.” But these changes would require a revolution both in economic thought and in international financial institutions and capital markets that by and large had not yet occurred. In the absence of such a revolution, and in the face of the pressing problems of poverty and inequality, it was hard to argue that even the palliatives should not be tried.
