• international problems of ECONOMIC DEVELOPMENT Address by IRVING S.FRIEDMAN The Economic Adviser to the President INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT to THE CANADIAN POLITICAL SCIENCE ASSOCIATION OTTAWA, CANADA JUNE 7, 1967 . INTERNATIONAL PROBLEMS OF ECONOMIC DEVELOPMENT have been connected with international institutions for most I of my professional career- namely, the Fund and the World Bank Group since their inception-and I thought it may be of some interest to you if I were to talk today about the economic issues and problems which arise at the international level and which constitute the bulk of my work. Let me sketch in broad outline three major features of the problems that confront economists dealing with problems of devel- opment at an international level of action and policy. We are engaged in the process whereby the advanced indus- trial countries have come to the assistance of the less developed countries (LDC's) in order to accelerate their development. This process, whereby one group of nations has recognized an obliga- tion to help accelerate the economic development of another group of nations, has become so familiar a feature of our daily newspaper reading that we sometimes forget what a profound new develop- ment this is in human affairs. Professional historians have noted this fact and have suggested that this development may well be the most distinguishing feature of our age. This development has, at the same time, raised a number of questions with regard to the nature of international economic re- lationships which go far beyond the confines of the existing theory of international trade and finance. For the present, let me pick out just two questions which are of very particular interest to us as economists in the World Bank Group. The first is the question of burden-sharing-on what principles is the quantum of aid to be divided among the aid-givers? At present a great deal of aid effort is organized bilaterally, and historically it has been dominated by a variety of political factors. A part of the aid is, however, through multilateral chan- nels, and of late an increasing part of bilateral aid has come to be organized on an international basis through various coordinating mechanisms such as consortia and consultative groups. We have been involved in this, since most of these coordinating mechanisms have been devised under the aegis of the World Bank. One of our concerns here has been to increase the role of economic calculus in the search for criteria by which the international aid effort can 1 be shared among the donor countries. This is not easy, and, as recent discussions have shown, there are even problems in measur- ing the burden to the donor countries of the assistance they pro- vide. This assistance is provided under such different terms and under so many constraints on its use by the recipient countries that a comparison of the nominal amounts is clearly not adequate as indicating the relative burden borne by these countries. Various approaches have been made to this problem of burden- sharing; there are, for example, various proposals to link the total aid provided by each individual donor country to some percentage of national income. Then there is the approach which has related the amount of official aid to the budgetary situation in each coun- try. Another important economic factor that has played a role in this question is the balance of payments effect of aid. Flows of aid, representing a transfer of resources overseas, enter the balance of payments of a donor country as an extraneous element not di- rectly related to the economic activities of consumption and pro- duction. A transfer of resources overseas without a corresponding matching in terms of aid-financed commodity exports may alter the relative balance of payments situation of countries. This kind of concern with the balance of payments has been an important reason-though by no means the only one-behind tying of bilateral aid to procurement in the country providing funds. But on the other side of the picture, the use of such tied aid creates some problems for the recipient countries. The inefficiencies asso- ciated with aid-tying, principally in terms of the higher prices the LDC's have to pay for their aid-financed purchases, have been widely discussed elsewhere by professional economists and by rep- resentatives of developing countries in various international fo- rums. Estimates indicate that tying results in about a 15 per cent increase in prices of commodities financed by aid; for some indi- vidual countries the percentage is even higher. The inefficiencies which tying introduces have weighed so heavily that the Bank and the International Development Association (IDA) in their opera- tions have insisted on international bidding in the use of funds. The economic question in this is how to reconcile the advantages of international competitive bidding from the viewpoint of developing countries with the consideration of the balance of payments effects of such a procedure on industrial countries in balance of payments difficulties. Or to put it another way, the question of economic in- terest is not how to take aid out of the balance of payments but 2 rather how to take the balance of payments out of aid. The answer for some is that the magnitude of development assistance should not be related to balance of payments positions; any disequilibrat- ing payments effects should be offset by available equilibrating measures. The problem should be thought of purely in resources terms. This issue of the relation between assistance and balance of payments is now of critical importance in current negotiations on various proposals in development finance. * * * While these are some of the problems on the side of burden- sharing by the rich nations, there are equally complex problems in the flow of aid to the developing countries. Among these prob- lems are the amounts of assistance to be given to particular devel- oping countries; the purposes for which such assistance is given- for specific projects, for general import-financing, for budgetary support, etc. ; the relative amounts of grants and loans; the terms and maturities of loans; the relation between debt-servicing and net flow of assistance, etc. Each of these problems raises a host of others. If I may, I will comment today only on the problems of aid allocation. The total volume of aid today is not adequate to provide all the funds which the developing countries could productively use for their development, hence the need for some criteria by which aid can be allocated. So far political considerations have dominated the allocation of bilateral aid as shown, for example, by the extent to which aid from particular countries is largely concentrated in par- ticular regions-French aid to the franc area, ·British aid to the former Empire, U.S. aid to Latin America-and this has brought to the forefront of aid discussions the question whether and what economic criteria are at all applicable in this field. The obvious criterion of relative poverty as a principle of allocation of aid is not entirely satisfactory from the point of view of ensuring that development assistance is put to the most productive use. To ensure the ·effective use of aid, therefore, there is today increasingly a search for some sort of performance criteria which would relate aid to economic performance by developing countries. There is as yet no clear agreement among economists on what criteria should be appropriately applied in judging performance-- nor indeed on the precise meaning of the concept itself. What inter- pretation should be given to this concept? Is "performance" the same thing as creditworthiness of a country in terms of its ability 3 to service more loans? Should it be judged in relation to the extent to which countries attain the targets of their development plans? Or is it its adherence to what might be considered "right" policies irrespective of the results obtained? Or should it be identified with the degree of success which a country has achieved in its economic affairs, irrespective of the amount of aid it has received, and ir- respective of the favorable or unfavorable accidents which have attended its domestic activities or its foreign trade and other relationships? There are no clear answers to the question of evaluating per- formance in developing countries, either in theory or in practice. Economic science has been traditionally concerned with questions of efficiency and welfare which one might have thought would pro- vide us with some criteria for evaluating performance. But we do not get much help here. On the one hand we have elegant models, which rarely get out of algebra into arithmetic using reliable num- bers; on the other hand we have the rule-of-thumb approach, i.e., some universal formula based on some simple relationships. For example, a formula which would measure performance in terms of the increase in government capital formation as a proportion of GNP over the past few years. Then, there is the notion of various "correct" policies that should be followed to maximize economic welfare, with which economic science has been grappling ever since its birth-the. use of the price mechanism and competition, the encouragement of investment and innovation, the promotion of for- eign trade, etc., but the problem of formulating them in such a way as to take account of the great variety of circumstances found in the developing countries still remains largely unsolved. In practice, it has been customary to consider the performance of developing countries by the rate of growth of their national income in the past as well as the rate of growth likely in the future. Judgments on likely future rates of growth are arrived at on the basis of judgments on whether a country is having a satisfactory amount of savings and investment, whether it is allocating its resources properly, whether it is having a proper monetary policy, whether it is maintaining the right rate, and so on. It has been a mixture of objective criteria and subjective judgments. The use of the national aggregates, especially the rate of growth of national income, as a guide to performance is not entirely satisfactory, per se, since the efforts which the poor countries have to make to promote their own development is not necessarily re- 4 • fleeted in the growth of their national incomes in the short period. These countries are interested in economic development because they want to achieve higher living standards ultimately, but the pattern of investment that they choose to achieve this may offer little immediate returns. For example it is now more and more widely recognized that the developing countries must first achieve a basic structural transformation of their economies, particularly in order to harness modern science and technology to their pro- duction systems. An essential aspect of such a transformation is the building up of sufficient amounts of physical and social infra- structure, even at times ahead of the demand for such facilities. The real justification for these investments is not that they by themselves would lead in the short run to growth of the national incomes of these countries, but that they will eventually increase the returns to directly productive investments. Let me illustrate the way in which national aggregates can hide the extent of progress made in the field of development with the example of India. Over India's first three plan periods covering the 15 years 1950 to 1965, the national output in India increased at a little more than 3 per cent per year and per capita output by slightly over 1 per cent per year. This is quite modest, and may even be disquieting, but at the same time between 1950 and 1963 in- stalled electricity capacity increased 230 per cent,. and 93 per cent of the towns and villages counted in the 1961 census were electrified by 1966; railway goods traffic increased by 142 per cent; railway passenger traffic by 33 per cent; the number of doctors by 39 per cent; the literacy rate increased from 16.6 per cent in 1951 to 24 per cent in 1961; the enrollment ratio for primary education in- creased from 43 per cent in 1951 to 67 per cent in 1961, and so on. Are these not the sort of changes that are necessary for a country's development, in whatever way development as a process of modern- ization may be defined? It is for this sort of reason that I have increasingly felt that in judging performance, we should try to get away from simplistic notions of national income growth. All major aspects of the econ- omy need to be considered. In judging performance we should also look to the efficient implementation of projects within an over-all development strategy, and the contribution a country itself makes to the financing of these projects in relation to its capacity. Within the Bank we are still debating the issues of how to judge perform- ance, as I am sure some of you are too. We have no dogma on the subject nor do we want to give undue weight to the past. Instead, 5 we have often used what might be called the "policy approach," whereby we agree with member developing countries on certain policies they should follow in the future to promote their best feasible development, e.g., monetary and fiscal policies designed to accelerate resource mobilization, incentive taxation, forward plan- ning of investments, etc. Agreement on these policies is arrived at through a continuous process of consultations, and as long as the country implements these policies, its performance is judged ade- quate for purposes of future assistance from the Bank Group. In some cases, even this approach is not feasible, and the problem of judging performance for the purpose of Bank lending becomes quite perplexing. One such case is that of a country which has no record of good past performance, and no con;imitment to future policy actions which would give adequate priority to development and which could be agreed internationally. At the same time this is a country with vast reserves of an important mineral resource with a high and increasing export demand, which it wishes to de- velop with the help of an international project loan. What should the Bank do? Should it accept the argument that a successful ex- ploitation of this mineral resource is vital enough for the future economic development of the country for it not to bother with the general governmental economic policy? Or should it wait for im- provement in government policies for development before agreeing to lend, with the expectation that the country's eagerness to obtain financing for such an attractive project will help induce the country to adopt better economic policies? Fortunately these situations are exceptional, but they do illustrate some of the knottier prob- lems which may arise in evaluating performance. The Bank's interest in the economic performance of the de- veloping countries in recent years goes far beyond its own imme- diate lending operational needs. One of the important functions that the Bank assumes in agreeing to organize a consultative group for any country is to make periodic comprehensive economic re- ports on the country's development possibilities, problems and per- formance as a basis for the Group's deliberations. There are now 11 developing countries which are covered by consultative groups or consortium arrangements chaired by the World Bank. In addi- tion, we undertake economic reviews for our own purposes, evaluat- ing the performance of our member developing countries ; last year we made such reviews for 45 countries, and since March 1965 for 73 countries. As a part of these responsibilities, we are con- tinuously in search of more explicit performance criteria which can be applied consistently among the less developed countries. 6 This whole question of performance brings me to the second group of problems of development, namely, the role of economics in the practical task which the developing countries face in pro- moting their own development with their own resources and with external assistance. The task ahead of the developing countries, as has been often pointed out, remains a long and arduous one. They still have to make an enormous effort to raise their economies to a level where they can exploit modern science and technology, their main hope for rapid progress in a world where the other countries are ad- vancing so fast already. For this purpose, the developing countries have to bear heavy sacrifices of present consumption, already at a pitifully low level, in order to make investments for the more or less distant future. A myriad of problems confronts their leaders in the painful choices they have to make, between consumption and investment, between quick returns and long-run goals, between social stability and economic growth, etc. They bring only limited experience to this unfamiliar task. There are inescapable political factors in the decisions actually made by their governments. In the face of all these difficulties, we have to find methods of evaluating the priority given to development, which would recognize the political element in decision-making, and yet be economic and ob- jective enough to escape narrow political bias. The process whereby policy decisions are translated into con- certed public action and private response is necessarily slow and it is not surprising, therefore, that some of us are at times dis- couraged by the pace of development in the developing countries. It is still a fashionable exercise to predict a gloomy future for them in terms of an ever-widening gap between them and the advanced countries by projecting rates of growth observed in the recent past. GNP per capita in developing countries (excluding Sino-Soviet countries) for which we have data has been growing at the rate of 1.8 per cent per year over the period 1960-65. The rate of growth of GNP per capita in the United States and Germany has been, on the other hand, 3.2 per cent and 3.5 per cent, respectively. If the per capita income of these countries were to be extrapolated on the basis of past trends, in 1980 the per capita income of LDC's, which now stands at around $150 per year, would still be only $195. For particular regions and countries, especially those of South Asia, prospects on the basis of a simple extrapolation would be even bleaker. As against this, per capita income in the United States would have increased to $5,060 per year from the present level of $3,160; and in Germany from $1,590 to $2,650. 7 I, for one, do not share in this pessimism. By its very nature, economic development is not a simple extrapolation of steady rates of growth; it is a deliberate attempt to accelerate and change ex- isting trends. There are increasing signs that the investments of the past decade and those of the next will make a decisive change from the past stagnation, which, in many cases, goes back to the 1920's or earlier. What is perhaps less 'apparent is the great im- provement that has taken place in the way the less developed countries are organizing their development efforts by the use of planning techniques. On a recent count, at least 50 countries have adopted partial or comprehensive development programs. Some of this planning has been faulty and, at times, even useless. But behind these weaknesses lies a truly remarkable phenomenon-the accept- ance of the planning technique. By planning technique in this con- text, I do not mean a detailed control and regulation of economic activity. Rather, it is the method by which governments make com- mitments to future actions and policies, thus extending the time horizons within which economic calculations based on objective criteria can play a greater part. The planning technique then be- comes an important instrument for coordinating development activity on a number of different fronts, and for maintaining some sort of continuity in the pace of development. Seen in such a broad perspective, the progress that has been made is considerable, and what is needed now is to push further the basic planning ingredient of commitment to future action on a coordinated basis, consistent with the possibilities of practical implementation. While the im- provements in the use of the planning technique promise to make the development efforts of less developed countries more successful in achieving their goal, it also provides a convenient framework by which the international community can make an over-all evaluation of the development program and priorities, and at the same time a focus toward which international development assistance can be directed. A third important feature of the economic problems that we have to deal with is to find ways in which we can assist the devel- oping countries in the mobilization of resources for their develop- ment. There are a number of ways in which they can raise and utilize more of their own domestic resources for this purpose, but a particularly stringent constraint for most of the developing coun- tries is in their foreign exchange earnings and receipts. To these countries, the means to pay for foreign goods and services are crucial because they must rely on imports for the capital goods 8 which they cannot produce themselves at their lower level of tech- nological development. However, the developing countries generally face serious structural difficulties in transforming their domestic resources into foreign exchange under present conditions of world trade. The difficulties which the developing countries face in com- modity trade has been the subject of extended discussions and negotiations between the advanced countries and the LDC's in the UNCTAD, GATT and elsewhere. The other source of foreign ex- change for the LDC's is the flows of official and private capital from advanced countries. This, of course, is of very immediate interest to us in the Bank, and there are many economic questions which arise in dealing with the appropriate terms, level and nature of such assistance, but I would mention only one of these, namely, the concept of external capital requirements of developing countries. The concept of capital requirements is an indication of the total amounts that less developed countries can productively use for their development and as such is one of the factors that the donor countries have to take into account in their decisions about the scale of their assistance to the developing countries. In recent years attempts have been made in the World Bank, the United Nations, the U.S. Government and elsewhere, to estimate the external fi- nance needs of the less developed countries. The methodologies by which these estimates are arrived at differ considerably. The ap- proach we took was based on individual country experience. Con- sidered judgments were made by our country economic staff regarding feasible growth in income, investment and savings, im- ports and exports for each country taken individually. Our analysis showed that the developing countries as a whole could effectively absorb over the next five years an additional capital inflow of $3-4 billion per year. Notwithstanding the informational gap in such estimates, these projections do indicate that even with relatively optimistic assumptions about the growth of exports and effective mobilization of savings as compared to historical trends, there is likely to be for some time to come an increasing requirement for net foreign exchange resources, if these cGuntries are to sustain rates of growth which are compatible with their capabilities. * * * All these three features that I have described, of coping with the economic problems of development at an international level, find a place in one of the exercises we began in 1965. This is the preparation and international consideration of the Study on Sup- plementary Financial Measures that we did in response to a request from the United Nations Conference on Trade and Development in 9 1964. This was a study aimed at devising a way of assisting the LDC's to meet the problem of unexpected export shortfalls which threatened to disrupt their development program, and which were of a nature and duration which could not be met by short or medium-term balance of payments support. We aimed at producing a viable scheme based on economic research and analysis. The first question we asked ourselves in undertaking the study was whether there was a problem and, if so, how big the problem was. The instability in the international trade in primary goods, which constitute the bulk of the exports of the LDC's, has been recognized for some time. The notion of instability is quite com- plex and cannot be completely captured in any simple statistical index. It is only too easy to devise a statistical measurement, call it an "index of instability," and then practically define instability as that which is measured by such an index. Several such indices have appeared in recent years, on the basis of which some authors have even questions whether there is a problem arising from export in- stability in the LDC's. Our own studies have persuaded us that export instability is particularly important in the case of primary goods and is a serious problem for the LDC's-much more serious than for the developed countries. Moreover, this question of in- stability had been studied in some detail by the IMF staff in pro- viding the intellectual basis for the IMF Compensatory Financing Scheme. This facility, as you know, came into existence in 1963 with the purpose of assisting "members, particularly primary exporters, encountering payment difficulties produced by temporary export shortfalls." The Fund has defined the "norm" as a moving average for five years centered on the current shortfall year so that there would be a rough balancing of shortfalls and surpluses over a rea- sonable period of time. The assistance provided under the facility is within the framework of normal Fund operations-to be repaid in three to five years, thus maintaining the revolving character of the Fund's resources. There is, however, another aspect of the problem of instability which is not dealt with by the Fund Compensatory Scheme nor covered by the usual discussion of instability, namely, deviation in export earnings from anticipated earnings on the basis of which a program for development over medium-term periods of about five years has been based. These deviations are not measured from extrapolations of past trends but rather from estimates of future expectations. A key aim of developil}ent is often to bring about by deliberate policy measures a change in past trends in export per- 10 - ~ -. ,.._- .. ·:.---~ formance. This kind of instability with the consequent uncertainty could not be measured by moving averages or similar measures nor dealt with by short-term balance of payments support. Having isolated the nature of the problem, to which the UNCTAD resolution addressed itself, we next tried to estimate the size of the problem. We wished to test empirically whether the "adverse movement in export proceeds" (and which was defined in the resolution as "a shortfall from reasonable expectations") had been a major problem. In keeping with the logic of the resolution, we defined reasonable expectations as estimated future export eventualities incorporated in an objective export projection within the context of a development program. Establishing the size of the problem involved relating "actuals" to "expectations," and we had to know not only what had happened in the past but also what people expected to happen at any particular point in the past. Fur- thermore, we had to be confident that what was done in the past was reasonably objective and could be used as a basis for judging the reliability of the export projections which had been made. It was impossible to have precise data to conduct the exercise, since our concept was a new one. Fortunately, however, we had an avail- able body of experience within the Bank which could be used. As part of the process of determining the creditworthiness of prospec- tive borrowers and their economic positions and prospects of devel- opment, the Bank in a considerable number of cases had projected exports of developing countries for a period of years ahead. Some of these projections had been made entirely by the Bank staff, some had been made originally by technicians in the member countries and modified by Bank economists or accepted by them after analy- sis, as reasonable. We decided to use these projections made in the past as raw materials for developing the "expectations" side of our analysis, and to compare with them the export performance of the same countries on the basis of available trade statistics. An exami- nation of Bank reports yielded 113 quantitative projections made by Bank economists for 37 countries during the years 1949-64, in- clusive, covering projections extending for various periods of time, but typically five years, into the future. Where these projections overlapped for any country, they were combined together into a "composite projection"; this was used as a measure of "reasonable expectations" for a comparison with actual export performance. We then measured each country's actual export earnings against the .projected values to derive a series of country-by-country, year- by-year shortfalls and what we called "overages"-earnings above expectations. This analysis demonstrated that shortfalls below ex- 11 pectations had been significant for a number of countries, though the number of countries affected and the intensity of shortfalls varied from year to year. However, this analysis had an important limitation in that the number of countries in the sample varied considerably from year to year. We therefore decided to extract from the list of countries for which composite_projections ·were possible two sample groups, one consisting of 14 countries for the seven-year period 1957- 63 and another of 18 countries for 1959-63. (The first sample comprised Brazil, Burma, Colombia, Costa Rica, Ecuador, El Sal- vador, Ethiopia, Guatemala, Mexico, Nicaragua, Pakistan, Peru, Thailand and Sudan. The second sample comprised all of these plus Argentina, Panama, India and Tanganyika [now Tanzania].) When the results obtained on the basis of the two samples were extrapo- lated to a global basis, this exercise yielded an estimate of $1.6 billion per year of total gross shortfalls, i.e., without subtracting the "overages" earned by the same countries. Export projections are now being done in the Bank for nearly all countries, unless statistically impossible. The next question that we tackled was the causes and effects of export shortfalls. We again took the empirical approach and made a number of special case studies of country experience. We recon- structed the economic experience of these countries as best we could. We had the help of economists with rich experience of these countries and our past economic reports based on field investiga- tions and country policy memoranda, as well as the published liter- ature. We found that, in a number of cases, an unexpected export shortfall had in fact resulted in slowing down, and even halting, planned investment, and had led to hastily mounted import restric- tion programs which impeded efficiency, particularly in the indus- trial sector, as well as emerg~ncy cutbacks in private and public investment through monetary and fiscal policies. The axe has usu- ally fallen on long-term-but basic-investments, resulting in a loss of momentum in the development effort. In a number of these cases, the export shortfall was due largely to unexpected move- ments in supply and demand conditions abroad, over which the affected country had no control. To illustrate, let me give you an example-that of Colombia in the early 1960's. In Colombia there has been a significant shortfall in exports below the plan target made in 1961, largely because of the fall in coffee exports in the 1960's. Though some decline in coffee prices had been expected, the extent of that decline could not have been anticipated on any rea- 12 I sonable basis. The most immediate effect of the shortfall in export earnings was in the slowing down of the rate of growth of indus- trial output. The rate of capital formation in Colombia is heavily dependent on the flow of imported machinery and equipment which increased by some 20 per cent in 1960 and 7 per cent in 1961, but declined by 13 per cent in 1962 and 19 per cent in 1963. Actual investment expenditures of the national government in 1963 were less than two-thirds of the target established under their plan which had been received by the World Bank staff. The major victim of the government retrenchment was transport investment. Thus, we concluded on the basis of past experience that the problem of unexpected export shortfalls was a real one for many developing countries and that this constituted a serious danger to the process of planned development in these countries. A new scheme was therefore needed to meet this problem. The scheme we proposed is now being discussed by an expert group of govern- mental representatives consisting of 14 developed and developing countries set up by the Trade and Development Board of the United Nations. The main ingredients of the scheme are as follows: A member country wishing to benefit from the existence of the scheme would enter into an understanding with the international agency admin- istering the scheme with regard to its export projection, the main elements in its development program and the related policies. As long as the administering agency was satisfied that the member country was doing its best to implement these agreed policies and programs, it would be presumed that any shortfall from the export projection was a shortfall outside the country's control and was likely to disrupt the country's development pro~ram. The scheme under these circumstances would operate to provide speedy relief to the country affected by an export shortfall. The amount of assistance provided by the scheme would depend on the amount of previously accumulated earnings above expectations (so-called "overages"), the amount of financing available from other sources of finance including the IMF compensatory facility and, finally, the feasible adjustment measures that the country itself could under- take to absorb part of the effects of an export shortfall without disrupting its development program. The extent to which the coun- try would be called upon to use other sources of finance, as well as the nature of the feasible domestic adjustment, would be agreed to in advance with the agency. 13 The scheme would be administered by an international agency, which might be one already existing or a new one established for the purpose. Supporting member governments would provide re- sources permitting operations during an initial five-year period. In our study we estimated that a level of $300 million to $400 million per year would probably suffice. Benefits of this scheme would be available to all developing countries prepared to meet the criteria of the scheme. Assistance from the scheme would be on the same terms as basic official development finance, usually quite concessional. I realize that there are many parts of the scheme on which I might comment. However, within the time available to me, I think it would be more useful if I were to confine myself only to certain aspects of the scheme. One of our fundamental assumptions in developing the scheme was that any assistance from the scheme should be related to ade- quate economic performance by developing countries. This princi- ple also helped in coping with the problem of how to judge whether shortfalls were to be considered to be due to causes beyond the country's control. The concept of relating assistance to perform- ance criteria enabled us to use a relationship of fundamental im- portance in basic development finance and also one that organically linked supplementary finance with normal or basic development finance, since criteria applicable to the latter could be used in whole or in part in providing supplementary finance. The novelty introduced by our proposed Supplementary Fi- nance Scheme is that the scheme is based on performance criteria to be worked out for all developing countries and in such a way that donor and recipient countries agree on what is feasible and desir- able, i.e., worthy of support by the international community. As a consequence, both donors and ·recipients have a stake in the success of the agreed development programs. The development effort thus becomes truly international, with all countries-donors and recip- ients-prepared to accept limitations on their freedom of action but within the framework of general policies and procedures which they adapt and change-as needed. The benefit is not only the more efficient use of available development finance, but an alternative way to cope with the difficult relationship problems which arise in this sensitive and important field. This approach evolved for the Supplementary Finance Scheme is clearly applicable to all develop- ment finance and has been increasingly so recognized. 14 The use of agreed performance criteria also helps to meet an- other important difficulty, namely, the need to make the availability of assistance from the scheme certain and speedy, but still subject to administrative discretion of the agency, i.e., not automatic. A time-consuming examination of the performance of a country after a shortfall is avoided in the scheme by stipulating that an under- standing be arrived at in advance, at the beginning of a planning period, between the international administering agency and the individual country concerning the development program and re- lated policies. Certainty of assistance is assured by the provision that, as long as the country adhered to this understanding, the country would qualify for assistance in the event of an export shortfall. The international agency would be in continuous contact with the country and the implementation of the agreed program and policies by the country would provide an a priori reason to believe that any shortfall was due to causes beyond the country's control. The ingredients of this "policy package," as we call it in our study, would, of course, vary from country to country. Stand- ardized criteria would be avoided; much reliance would be placed on the objectivity and competency of the international agency to agree with the country concerned on a policy package adapted to the conditions of the country, but designed to achieve the objectives of economic development. Moreover, to avoid unrealistic rigidity the international agency and the country could agree at any time to needed changes in policies implementing the development pro- gram, but all the time the international community would be assured that its pledged support was for an internationally agreed program and policies. Another principle on which we based the scheme was that the export projection from . which shortfalls would be calculated would be not only as objective and scientific as statistical techniques and available data permitted, but also agreed to between the agency and the country. This projection, extending over the entire period of the development program, would be arrived at in the context of an understanding on the program as a whole. It would be subject to review and revision only in agreement with the international agency and as part of an agreed general recasting of the program. In suggesting a five-year projection period to coincide with the usual length of a development plan, we were keenly aware of the limitations of available techniques of projection over that long a period. Projections are, however, built into development programs in two ways: the development program contains commitments to 15 certain policy actions on which the projections are based, especially in export policies ; the development program in turn is based on the export projection and is tested for financial feasibility on the basis of anticipated export earnings and other sources of development finance. Many countries are already accustomed to making projec- tions to provide a basis on which investment programs can be built. We felt that a five-year export projection, as against a much shorter time period, was necessary since investment calculations over the plan period must be based on export projections and the length of the latter should by its very nature be tied to the former, namely, investment calculations. In suggesting this we were encouraged as to its feasibility by our own experience in this field. We compared IBRD projections of comparable time lengths with actual historical experience in 37 cases involving 19 countries. We ran a number of statistical tests to judge the reliability of our projections, and on the basis of these came to the conclusion that five-year export pro- jections were feasible and reasonable. The quality of projections can, of course, be improved by sharpening analytical tools and strengthening the basis for technical judgments, and we are plan- ning to do so in the World Bank in any case. This is a field in which improvement in technique can make a big difference to the effec- tiveness of planning in developing countries. However, if I may state the obvious, the improvement in projection technique will not eliminate the problem of unexpected shortfalls because unpredict- able changes in the world economy will continue to take place. Having decided on a scheme that would provide timely assist- ance for countries experiencing export shortfalls, the questions of how much finance and on what terms had to be analyzed and an- swered. We assumed that assistance from the scheme should be adequate to perform the job, but not more than necessary to protect an agreed development program from disruption. This minimum amount should be provided by the donor countries; after analysis, as I said before, we concluded that $300 million to $400 million per year for five years was a reasonable figure for the initial period. However, the recipient LDC should also do what it can to meet a shortfall, without disrupting its development pr.ogram. Thus we felt that there should be provision in the scheme to apply against the shortfall any "overages" in export earnings aceumulated dur- ing the plan or projection period. There would also be an· under- standing on appropriate use of the country's reserves. If these were inadequate, the compensatory financing facility of the IMF might be drawn upon if the country qualified for such assistance under the IMF rules. Other action that the country could take would de- 16 pend on the circumstances of the country. For example, additional drawing rights in the IMF or other sources of international assist- ance might be available, or the government might have latitude for the restriction of some luxury imports or revision of tax rates, etc. The provision that a country should make all feasible domestic adjustments raises many questions. What is to be strictly regarded as the development plan? How about cuts in consumption-are they appropriate when the levels of consumption are already very low? Should provision be made for refinancing drawings from the Fund, if such drawings remain outstanding after five years? Are changes in fiscal policy to absorb part of the effect of an export shortfall to be considered appropriate when there is a possibility that the short- fall may be reversed? We have tried to approach these more de- tailed questions in the same way as the original broad questions, i.e., as economists trying to find practical solutions based on objec- tive analysis. There are a number of aspects of the scheme that I have not touched upon, but I hope this presentation helps to provide you with a fair idea of our approach to the questions and issues of develop- ment as they arise at an international level. The questions and issues which I have mentioned today do not, of course, exhaust the area of concern to development economists and policymakers in this field. I do hope, however, that they give some indication to those of you outside of governments and international institutions as to our common needs for further research, :;i.nalysis and theory in this field. * * * Many of our problems as economists and economic policy- makers are created by the existence of national boundaries, yet we have many more national boundaries than ever before. This gives special importance to international institutions since, for the time being at least, they represent an attempt to reconcile the nation- state and the underlying unity of economic behavior. The interna- tional financial institutions thus have a special responsibility; not only are we given a mandate to help solve some of the world's criti- cal problems, but we have been entrusted with large sums of monies by the international community. As economists in these institu- tions dealing with very difficult problems on the policy frontiers of economic science, we need all the help and guidance we can get in fulfilling our tasks. 17 INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT 1818 H Street, N.W., Washington, D.C. 20433 U.S.A. Telephone number: EXecutive 3-6360 Cable address: INTBAFRAD Office for Europe: 4, Avenue d'Iena, Paris 16e, France Telephone number: KLEber 25-10 Cable address: INTBAFRAD