FOREIGN AID AND MEETING PROJECT-RELATED RECURRENT COSTS IN LESS DEVELOPED COUNTRIES by Gregory Bernard Bogut A professional paper submitted in partial fulfillment of the requirements for the degree of Master of Public Administration MONTANA STATE UNIVERSITY Bozeman, Montana March 1988 ii APPROVAL of a professional paper submitted by Gregory Bernard Bogut This professional paper has been read by each member of the committee and has been found to be satisfactory regarding content, English usage, format, citations, bibliographic style, and consistency, and College of Graduate Studies. 2-r Up- 8-y Date is ready for submission to the Chairperson, Graduate Committee Approved for the Major^ Department ft i Date Headf Major Department Approved for the College of *2. - Date Graduate Dean iii STATEMENT OF PERMISSION TO USE In presenting this paper in partial fulfillment of the requirements for a master's degree at Montana State Uni¬ versity, I agree that the Library shall make it available to borrowers under the rules of the Library. Brief quota¬ tions from this paper are allowable without special permis¬ sion, provided that accurate acknowledgement of the source is made. Permission for extensive quotation from or reproduc¬ tion of this paper may be granted by my major professor, or in his absence, by the Director of Libraries when, in the opinion of either, the proposed use of the material is for scholarly purposes. Any copying or use of the material in this paper for financial gain shall not be allowed without my written permission. iv TABLE OF CONTENTS Page LIST OF TABLES vi ABSTRACT viii 1. INTRODUCTION . 1 2. SPENDING PRESSURES IN DEVELOPING COUNTRIES 13 Population and Modernization Effects 13 Population Effects 14 Modernization Effects 15 Summary 18 3. FINANCING INVESTMENT NEEDS IN LDCs 21 The Public Sector Savings-Investment Gap 21 Financing Investment by Public Sector Saving 22 Financing Investment by Private Sector Saving 26 Structural Restrictions 26 Financial Sector Restrictions.... 28 Financing Investment by Money Creation 30 Summary 34 4. OFFICIAL DEVELOPMENT ASSISTANCE AND RECURRENT COSTS .. 39 Expenditure Effects 40 Increased Investment 41 Changes in the Content of Investment 42 Donor Characteristics and External Incursions 43 The Impact of External Incursions 46 Adaptive Responses to Incursions 48 Revenue Effects 56 Taxation and Borrowing and Leakage from the Recurrent Budget 56 Summary 58 V TABLE OF CONTENTS—Continued 5. MEETING RECURRENT COSTS? SHRINKING FISCAL RESOURCES 62 Revenue Constraints 65 Government Policies and the Tax Base.. 65 The Preemption of Domestic Credit 66 The Undervaluation of Capital 66 The Undervaluation of Foreign Exchange 68 Expenditure Inducers 70 Government Policies and Committed Expenditures 70 Civil Service Wages and Salaries 71 Social Insurance Obligations 73 Interest Payments on the Public Debt.... 75 Subsidization of State-Owned Enterprises 76 Revenue Constraints and Expenditure Inducers: The Uncommitted Revenue Squeeze 79 Summary 83 6. LIMITATIONS TOWARD FINANCING RECURRENT COSTS IN WEST AFRICA AND THE SAHEL 86 Why West Africa and the Sahel? 87 The Public Sector Savings-Investment Gap 96 Revenue Constraints 97 Poverty and Low Private Saving 98 Existing Tax Systems 99 Expenditure Inducers 110 The Gap 112 Other Domestic Financing Options: Private Sector Savings and Money Creation 115 Official Development Assistance as a Means of Finance 118 The Fiscally-Induced Uncommitted Revenue Squeeze. 122 Uncommitted Revenue as a Support Mechanism for Recurrent Costs 125 Summary 138 7. CONCLUSION.- 145 REFERENCES CITED 156 APPENDIX 159 vi LIST OF TABLES Table Page 1. Public sector wages and salaries as a percentage of total current expenditure, nineteen nation world sample, 1975, 1983. 73 2. Public sector social insurance as a percentage of total current expenditure, eleven nation world sample, 1975, 1983 74 3. Public sector debt service as a percentage of total current expenditure, nineteen nation world sample, 1975, 1983 76 4. Public sector subsidies as a percentage of total current expenditure, nineteen nation world sample, 1975, 1983 78 5. Other goods and services as a percentage of total current expenditure, nineteen nation world sample, 1975, 1983 81 6. Summary of the shares of committed and uncommit¬ ted expenditure in total current expenditure, nineteen nation world sample, 1975, 1983 82 7. Basic economic and demographic indicators for the seven-nation composite sample and other select economic or geographic groups 89 8. Health and education-related indicators and phy¬ sical quality of life indices for the seven- nation composite sample and other select economic or geographic groups 92 9. Trade-related indicators for the seven-nation composite sample and other select economic and geographic groups 102 10. Select principle export commodities of the seven- nation composite sample: change in market share and price. 104 11. Nominal and real interest rates for the five- nation Sahelian subset of the seven-nation composite sample, 1974-1979 107 vii LIST OF TABLES—Continued 12. Net foreign assets of the seven nation composite sample, 1970-1983... 108 13. Absolute nominal increase of committed public sector expenditures for five nations of the seven-nation composite sample, 1977-1982 Ill 14. Key budget indicators and the public sector savings-investment gap for the seven-nation composite sample, 1976 113 15. Key budget indicators and the public sector savings-investment gap for the seven-nation composite sample, 1980 114 16. Financing from the non-bank public as a percen¬ tage of total deficit financing for the seven- nation composite sample, various years 115 17. Financing of deficit by deposit money banks and monetary authorities for the seven nation composite sample, 1982 118 18. Per capita official development assistance (ODA), ODA as a percent of gross domestic investment for the seven-nation composite sample 120 19. Total loans from external sources as a percent of total deficit financing for the seven-nation composite sample, 1982 121 20. Uncommitted expenditure as a percent of current expenditure, average annual change of the propor¬ tion of uncommitted revenue in current expenditure for the seven-nation composite sample, various years 126 21. Nominal change in committed expenditure vs. nom¬ inal change in uncommitted expenditure for the seven-nation composite sample, various years 130 22. Real change in committed expenditure vs. real change in uncommitted expenditure for the seven-nation composite sample, various years - 135 viii ABSTRACT Throughout the developing world the productivity of public investments that are already in place has been seriously jeopardized by the failure of govern¬ ment officials to adequately provide for their oper¬ ation and maintenance over time. While this outcome is due, in part, to economic constraints, whether externally or internally imposed, which limit over¬ all revenue, unmet project-related recurrent costs can largely be attributed to deliberate fiscal poli¬ cies and practices which effectively limit fiscal resources which could potentially support such costs. This process is intensified by the influx of large amounts of official development assistance which allows an overall increase in the number of installed investments, particularly those with high recurrent costs, without emphasizing an accompanying need for fiscal resources to support their operation and maintenance. This paper describes many of the processes, foreign or domestic, economic or fiscal, which both increase expenditures and/or limit funds for project operation and maintenance and demon¬ strates that a definitive preference in favor of expenditure categories supportive of objects-of- expenditure other than those related to development projects is operative among seven West African and Sahelian nations. 1 CHAPTER 1 INTRODUCTION The leadership of Less Developed Countries (LDCs), in response to the emergence of strong domestic opinions fa¬ voring enhanced public services and encouraged by the ac¬ tive development efforts of the community of donor nations, have increasingly resorted to investment in public sector projects and programs as a primary means for economic growth and/or social change. Such investments, evaluated as socially profitable, are thereafter assumed to realize their full productivity, thus contributing to national economic welfare. Desired outcomes of this sort, however, appear'inconsistent with numerous descriptions of public sector projects and programs as undermaintained, undersup¬ plied, and subsequently, underproductive. In fact, many LDC economies have been characterized by disinvestment as productive equipment falls into disuse, not because it is obsolete or no longer viable, but because of lack of main¬ tenance and spare parts. Outcomes such as these have become more prevalent, even for relatively new projects, since domestic fiscal resources, which are often scarce, are increasingly being used for such essential govern¬ ment consumption needs as wages and salaries, social 2 insurance obligations, and debt service, and not for the recurrent costs associated with the repair and maintenance of installed investments. As LDCs and donors place more and more emphasis on new programs and projects to provide for social infrastructure, the problem of unmet project- related recurrent costs, and the loss of productivity which stems from it, is likely to intensify. Specifically, since the productivity of any given project is reliant upon future outlays for its operation and maintenance, any in¬ crease in the number of investments will require the need for subsequent increases in recurrent expenditure. These increases, in turn, may be beyond the normal capacities of the budget. Subsequently, project-related recurrent costs may go unmet. From the perspective of influences which increase project-related operation and maintenance expenditure, unmet recurrent costs associated with public sector in¬ vestment can be essentially interpreted as a by-product of the active development efforts of the last several decades. In this regard, LDCs, either through their own efforts or assisted by aid from outside sources, have steadily in¬ creased public investments' share in their gross national products. Depending upon the prevailing political climate, this process can reflect either an active approach toward specific ideological or social goals, a response to the demands of national elites for consumption standards 3 reflecting those of the developed market economies, or a means whereby national rulers can secure approval or main¬ tain power or prestige. To the extent where the general population is enfranchised, the process may also reflect a response to electoral calls for an expansion in projects to meet basic human needs. The process can also be represen¬ tative of a general willingness by donors and recipients to provide and to accept increasing amounts of development assistance for its political, social, or economic value. While the idea of development through public sector investment offered new hope to the nations of the develop¬ ing world, the process was sure to be inhibited by the limited capacity of LDCs to generate a surplus of revenue for investment. At the time of independence, many LDCs were almost fully dependent on their former colonial stew¬ ards for recurrent budget support while funding for capital budgets commensurate with investment needs was almost non- 1 existent. Given their limited fiscal capacity and in view of the fact that either the parochial or the popular call for investment in social investment projects was becoming increasingly prominent, LDCs had to look more and more to the outside world for help. Foreign economic assistance, in the shape of support for investment programs and pro¬ jects, thus became the answer to the growing gulf between the consumption desires and the limited resources of LDCs worldwide. Moreover, in pursuit of political, social, or 4 economic objectives, the donor nations were as willing to provide this financial assistance as the recipients were to receive it. Like the effect of domestic calls for increased in¬ vestment, foreign economic assistance added new momentum to the recurrent cost problem by magnifying the number of projects which could be constructed. At the same time, donors meting out such assistance largely ignored the ac¬ companying need for recurrent budget growth as a means for supporting operation and maintenance costs associated with these projects. The majority of studies to date have, in fact, emphasized the foreign economic assistance process, specifically the relationship between donor and recipient, as one of the primary causes of unmet recurrent costs associated with investment projects in LDCs. Peter. Heller was reportedly the first to link the recurrent cost problem to a failure by both donor nations and recipient country budgetary authorities to appreciate the dynamic fiscal commitments and ramifications engendered by public in¬ vestment decisions within the context of prevailing re¬ source and politically-induced budgetary constraints in the 2 public sector. In Heller's estimation, a propensity by donors to stress aid for development purposes, coupled with a willingness and ability by LDCs to solicit and accept growing amounts of external assistance for development on donor terms, have together contributed to the recurrent cost problem by enabling or inducing developing countries to increase the overall level of their investment without giving adequate consideration to their budgets' ability to sustain the heightened levels of operation and maintenance- 3 related recurrent costs which new investment implies. K.B. Griffen and J. L. Enos in 1970, Judith Tendler in 1975, and Heller, again in 1979, also indicated that for¬ eign economic assistance may play a role in effecting an overall change in the composition of investment programs to 4 the general disadvantage of many LDCs. In this context, the recipient, seeking to maximize the amount of available aid funds, often agrees to conditions attached to these funds by the donor. This process leads to an increased transfer of available aid funds and to a loss of recipient control in project selection, scope, and design. This, in turn, may be manifested in improper project scale and in projects which are overly capital and import-intensive. Such projects may then become dependent for long afterward upon materials and spare parts whose recurrent funding 5 needs may exceed the capabilities of the domestic budget. Conditional aid, or tied aid (aid based solely upon compliance with pre-specified donor end-uses) on the part of donors has been further implicated in the recurrent cost problem by Clive Gray and Andre Martens in their study, and by A. Jennings in his review of the phenomenon. Both works emphasized changing donor preference from investment in 6 physical infrastructure and industry to investment in the so-called 'social' sectors comprising, notably, education, health, urbanization, agricultural production, and assorted vehicles for rural development as a prominent contributing factor to unmet recurrent costs. Specifically, because social sector projects require significantly higher recur¬ rent expenditure per unit of investment than do those in industry, major shifts in favor of social infrastructure will sharply increase the overall recurrent cost commitment 6 of affected nations. While overall levels of donor-induced project-related recurrent costs have increased among LDCs, the ability or willingness of these nations to generate sufficient reve¬ nues to cover them has not. This may be due, in part, to fiscal policies which constrain both potential and existing domestic revenues and favor objects-of-expenditure that collectively preempt revenue otherwise needed to fund project-related recurrent costs. From the revenue aspect, Harold Hinrichs initially indicated limited overall domestic financing capacity as a 7 fundamental factor in the recurrent cost problem in LDCs. Low per capita income, high rates of consumption, low and in some instances, negative growth, and fluctuating or deteriorating external environments have each contributed to low savings ratios and shrinking or static tax bases. 7 Further, efforts at supporting expenditure by private sec¬ tor saving or money creation have proven inadequate or counterproductive. These characteristics are compounded by a general but less obvious tendency by LDC governments to pursue policies, intentionally or unintentionally, which may limit revenue on the one hand or constrain potential revenue sources on the other. With respect to the limita¬ tion of revenue, several studies have described the govern¬ ments in LDCs as unwilling to mount adequate tax programs coincident with their rates of economic growth and infla¬ tion. In this light, over reliance on indirect taxes, which are typically less progressive and more inelastic than direct taxes, and lags in tax collection have also been implicated in the inability of revenues to keep pace 8 with inflation and budgetary needs. As a result, govern¬ ments have been exceedingly hard-pressed in meeting the 9 needs of their recurrent budgets. Finally, a study conducted under the auspices of the Organization for Economic Cooperation and Development (OECD) in Paris has implicated government policies inhibit¬ ing the growth of the tax base as being crucial to the overall fiscal situation of LDCs. In this regard, the preemption of domestic credit and the undervaluation of the cost of capital and foreign exchange were singled out as affecting the potential growth of the tax base in a nega- 10 tive fashion. 8 In view of such poor revenue performance, the tendency to attribute the shortfall in project-related operation and maintenance expenditures to an overall lack of finance is very real indeed. Yet, the problem can be attributed as much to expenditure policies as to insufficient revenue. Specifically, while certain comprehensive ceilings to do¬ mestic revenue for any given year do, in most cases, exist, government revenues are fungible in the sense that gov¬ ernments can allocate their domestic revenue to whatever objects-of-expenditure they choose within the confines of an overall ceiling. While this limit may be too low to enable the government to operate and maintain every unit of the public sector's installed capacity at its designed level of productivity, a single project cannot be said to suffer a priori from insufficient financing of its opera- 11 tion and maintenance. Rather, an apparent lack of such financing must mean that the government, in light of its revenue constraint, has decided to allocate the correspond¬ ing amount of revenue to objects-of-expenditure which it considers as having a higher priority. Gray and Martens, in commenting on a study conducted on recurrent costs in West Africa, listed the civil service payroll, social in¬ surance obligations, and service payments on the public debt as falling under this category of high priority or 'politically imperative' expenditure. Governments, faced with a number of expenditure categories must, for political 9 reasons, provide for these expenditure imperatives before available revenue can be used to support other expendi¬ tures. Within the context of the recurrent budget as a whole, these necessary expenditures are termed 'committed' by Gray and Martens and are considered to be of an order of magnitude larger than all other claims on the public 12 purse. Accordingly, the remaining 'uncommitted' resour¬ ces, which in addition to other expenditures, support project-related operation and maintenance, will be limited by budgetary policies which increment the committed expen¬ diture categories. Further, committed expenditures may receive preference in the budget's response to inflation. Subsequently, the real levels of uncommitted revenue will be reduced and, given other demands on this revenue, few resources will remain for project-related operation and maintenance. With this line of reasoning in mind, underfi¬ nanced project-related recurrent costs must be interpreted as a by-product not only of intentional or unintentional limitations to overall domestic financing capacity but also of the fiscal behavior patterns which these limitations engender. Overall, this type of fiscal behavior, i.e., the wil¬ lingness to accept continuing inflows of aid without regard for the budgetary impact of the recurrent costs associated with the development projects which such aid supports, may suggest that the real objective of aid-assisted development 10 is not necessarily social improvement or growth. Rather, the disbursement of aid by donors and its acceptance by LDC governments may occur primarily because of its potential political value for the donors or, in the case of the recipients, its prestige value or its utility in perpetuat¬ ing parochial interests. The following chapters will consider at length many of the factors presented thus far. Chapter Two will discuss some of the more notable reasons behind public sector investment spending among LDCs, focusing primarily on pop¬ ulation and modernization effects. Chapter Three will discuss the reasons behind the limited capabilities of LDC governments to fully finance their investment needs. In this regard, the case for foreign economic assistance as the primary vehicle for closing the gap between public sector investment demand and available public sector re¬ sources will be explicated. Chapter Four will discuss the negative consequences for the overall recurrent cost burden of increased amounts of foreign economic assistance in the budgets of developing nations with special reference to the interrelationship between the donor and recipient and its effect upon project design, selection, and scope. Chapter Five will discuss those specific government policies which constrain available tax revenue or increments to tax reve¬ nue and those government policies which favor revenues which support committed expenditure over those which (at 11 least potentially) support project-related recurrent costs. Finally, Chapter Six will apply the concepts discussed in these chapters to five nations of the West African Sahel and two non-Sahelian West African nations. Specific defi¬ nitions related to recurrent costs are presented in the Appendix. 12 ENDNOTES 1. A. Jennings, "The Recurrent Cost Problem in the Least Developed Countries," Journal of Development Studies 19 (July 1983), p. 504. 2. Peter Heller, "Public Investment in LDC's With Recurrent Cost Constraint: The Kenyan Case," Quarterly Journal of Economics 65 (May 1974), p. 251. 3. Heller, "The Underfinancing of Recurrent Develop¬ ment Costs," Finance and Development 16 (March 1979), p. 39. 4. K. B. Griffen and J. L. Enos, "Foreign Assistance: Objectives and Consequences," Economic Development and Cultural Change 18 (April 1970) , pp. 313-327; Judith Tendler, Inside Foreign Aid (Baltimore: The Johns Hopkins University Press, 1975) ? Heller, "The Underfinancing of Recurrent Development Costs," Finance and Development 16 (March 1979), pp. 38-41. 5. Griffen and Enos, op. cit., p. 323. 6. Clive Gray and Andre Martens, "The Political Econ¬ omy of the 'Recurrent Cost Problem' in the West African Sahel," World Development 11 (1983), pp. 101-117; A. Jen¬ nings, "The Recurrent Cost Problem in the Least Developed Countries," Journal of Development Studies 19 (July 1983), pp. 504-521. 7. Harold Hinrichs, A General Theory of Tax Structure Change During Economic Development (Boston: Cambridge University Press, 1966) , p. 7. 8. Bijan Aghevli and Mohsin Kahn, "Government Defi¬ cits and the Inflationary Process in Developing Countries," International Monetary Fund Staff Papers 25 (1978), p. 391. 9. Heller, op. cit., p. 39; A. Jennings, op. cit., p. 507, 512. 10. Club du Sahel/CILSS, Recurrent Costs of Develop¬ ment Programs in the Countries of the Sahel: Analysis and Recommendations (Paris: OECD, 1980). 11. Gray and Martens, op. cit., p. 102. 12. Ibid., p. 102. 13 CHAPTER 2 SPENDING PRESSURES IN DEVELOPING COUNTRIES In analyzing spending pressures in LDCs, one could start by looking at the movements of recorded expenditure relative to GNP over time. Here, while it is easy to establish that there have been sharp upward movements in such data for most LDCs since independence, little concrete interpretation as to the reasons behind spending can be gained from this approach. Essentially, all that can be shown is what was actually spent by governments, a mere confirmation that some pressure to spend exists. We must, therefore, out of necessity, confine ourselves to an ap¬ proach which is primarily theoretical and macroeconomic. Population and Modernization Effects There are two major vantage points for considering the motivational factors behind public sector spending: the effect which increasing populations exert on expen¬ diture and the effect of the modernization process itself. 14 Population Effects The importance of population in any analysis of public sector spending pressures among LDCs has long been appreci¬ ated. For many LDCs, apart from overall size, the age structure, the rapidity of increase, and the geographical concentration of population have all been mentioned as possible explanations of the relative growth of the public sector. For example, I.J. Coffman and D. J. Mahar consi¬ dered the age structure of the population to have been an important factor in public expenditure growth in six Carib- 1 bean countries during the postwar period. High birth rates between 45 and 50 per thousand for Africa and some parts of Asia (as compared with 15 and 20 per thousand in the developed market economies) which exceed morbidity rates by as much as two or three times, also present signi¬ ficant problems for LDC budgets. This rapid net rate of population growth has had, and will continue to have, obvious repercussions on the need for roads, hospitals and clinics, sewers, water supply, and other public works and services. Moreover, high growth rates also denote an over¬ all increase in the number of children and a subsequent increase in the need to expand educational and health- 2 related facilities. Finally, rapid increases in popula¬ tion are frequently associated with movements from rural to urban areas. As standards of public services tend to be 15 higher in the latter, this is likely to mean a more 3 than proportionate increase in government outlay. Modernization Effects In conjunction with these population increases, the process of modernization and development as a whole has exerted a primary influence on spending in the public sector. Karl Deutsch termed the expenditure-promoting component within the modernization process as 'social mo¬ bilization, ' or the shift from traditional patterns of life to modern ones, and gave it two distinct parts: breaking old ties, habits, and commitments, and finding and adopting 4 new ones. According to him, this transition, over time, produces a growing gulf between the needs for and;capabili¬ ties of government and its services. This has been made possible, in part, by the emergence of outside influences and the penetration of these influences into the society of the developing country. In more elaborate terms, Ragnar Nurkse called attention to a modernization-related 'inter¬ national demonstration effect' that causes people in poor countries to emulate the consumption standards of rich 5 countries. Specifically, knowledge of or contact with new consumption patterns in the process of development leads to a realization of previously unrecognized possibilities. They widen the horizon of imagination and desire and 16 promote entirely new styles and rates of consumption. As Nurkse states: New products constantly emerge from the course of technical progress, which modify existing ways of life and frequently become necessities... the presence or mere knowledge of new goods and new methods of consumption tends to raise the gener¬ al propensity to consume. New goods, whether home-made or imported, become part of the stan¬ dard of living, becoming indispensable or at least desirable, and are actively desired as the standard of living rises.° The international demonstration effect applies particularly to government expenditures. Political leaders of LDCs if they are to satisfy their sponsors (elites with various special interests) or their electorates, must be cognizant of the growing political pressures which these individuals or groups can impose. Calls for parity with developed market economies will certainly lead to a heightened demand for government services and more kinds of government activities. This fact will necessitate that government officials in LDCs today not have the same attitudes toward spending for education, health, and various economic and social services that existed in the industrial countries when they were at comparable stages of development. On the contrary, their attitudes will lead to heightened levels of expenditure fueled, in part, by an unwillingness amongst the elite and the citizenry to accept the standards that 7 existed in the past. Thus, both the size and scope of government will increase. Alison Martin and W. Arthur 17 Lewis, in an analysis of public sector revenue and expendi¬ ture, asserted that the externally-influenced shift from traditional societies to modernity, encompassing urbaniza¬ tion, industrialization, secularization, democratization, education, and media penetration, has, indeed, led to a different conception of the role of the state by rulers, the elites which support them, and, to some extent, their 8 electorates. As a result, expenditures in areas such as administration, health, education, and public works pro¬ jects have increased as these nations became infected with the 'progressiveness' of their rulers, their sponsors, or their electorates. This changing role concept has been reinforced by what A. R. Prest refers to as the emergence of 'strong local opinions' among the citizenry of LDCs. Again, education, modern communication, international migration, and renewed foreign presence in these countries all contributed to a demonstration effect that has led to pressures on governments to provide the same kinds of basic infrastructure and public services that exist in more wealthy countries regardless of their levels of economic development and ability to respond to the resulting spen- 9 ding pressures. Thus, over time and within the context of modernize tion, developing country rulers, elites, and electorates have come to advocate a maximum as opposed to a minimum level of government as the shortest route to social change. 18 This is particularly true of socialist or centrally-planned economies. Specifically, it has been hypothesized that countries with a relatively greater degree of government participation in their economies are subject to greater 10 spending pressures. However, with the exception of spen¬ ding on education, much debate remains as to whether or not ideology and the character of the political system notice- 11 ably influence government spending. One fact has emerged from all of these considerations. It is that the cumulative effect of all of those influences mentioned thus far has been a public which has come to rely more and more on government for a broad range of goods and services. Under these circumstances, it may be more diffi¬ cult for a government to restrain expenditures and reverse the process as the public learns to consider government 12 assistance its right. Summary As the population of developing countries increased and as the citizenry of these nations became more aware of the differential that existed between their standard of living and that of the developed market economies, signifi¬ cant pressures for modernization, in the form of public sector projects, began to arise. While LDC governments have responded to these pressures for expenditure in in¬ vestment for social development, they have not, for the 19 most part, been able to mobilize sufficient resources to fully cover the costs of such expenditures. The following chapter will discuss some of the problems of LDC develop¬ ment resource mobilization and indicate the part that for¬ eign official development assistance plays in meeting pub¬ lic sector investment demand. 20 ENDNOTES 1. I.J. Coffman and D.J. Maharf "The Growth of Pub¬ lic Expenditures in Selected Developing Nations: Six Carib¬ bean Countries, 1940-65," Public Finance 26 (January 1971), pp. 57-74. 2. A.R. Prest, Public Finance in Underdeveloped Countries (New York: John Wiley and Sons, 1972), p. 14. 3. Prest, op. cit., p. 14? Jack Diamond, "Wagner's 'Law' and the Developing Countries," Developing Economies 15 (March 1977), pp. 41-46. 4. Karl W. Deutsch, "Social Mobilization and Politi¬ cal Development," American Political Science Review 55 (September 1961), pp. 493-514. 5. Ragnar Nurkse, Problems of Capital Formation in Underdeveloped Countries (New York: Oxford University Press, 1953) , pp. 61-65. 6. Ibid., pp. 61-62. 7. Ibid., pp. 58-59. 8. Alison Martin and W. Arthur Lewis, "Patterns of Public Revenue and Expenditure" pp. 203-244, quoted in H. Hinrichs, A General Theory of Tax Structure Change During Economic Development (Boston: Cambridge University Press, 1966) , p. 9. 9. Prest, op. cit., pp. 14-15. 10. Thomas K. Morrison, "Structural Determinants of Government Budget Deficits in Developing Countries," World Development 10 (1982), pp. 469-472. 11. Frederic L. Pryor, Public Expenditures in Commu¬ nist and Capitalist Nations (Irwin: 1966) , pp. 285-292? Alan A. Tait and Peter S. Heller, "Government Employment and Pay: Some International Comparisons," unpublished paper (Washington, D.C.: International Monetary Fund), quoted in Richard Goode, Government Finance in Developing Economies (Washington, D.C.: Brookings Institution, 1984), pp. 52-53. 12. Morrison, op. cit., p. 470. 21 CHAPTER 3 FINANCING INVESTMENT NEEDS IN LDCs Large amounts of capital are needed if a country is to provide or produce the levels of investment-related goods and services required to meet the needs of increasingly vocal and rapidly growing populations. However, for many LDCs, public sector resources to support this growing de¬ mand are not adequate. As a result, governments must incur deficits and seek financing for their investment programs either from private domestic or public and private external sources. Such avenues of finance include borrowing of private savings, money creation, and foreign economic as¬ sistance. This chapter will explore the reasons behind the public sector resource shortage and establish the subse¬ quent need for external aid. The Public Sector Savings-Investment Gap It is instructive to view the role of foreign economic assistance and its relationship to the public sector deficit in terms of the public sector savings-investment gap. 22 A given public sector investment program can be fi¬ nanced through four sources: 1. public sector savings (budgetary savings); 2. private sector savings? 3. money creation; and 4. foreign savings. Assuming, initially, that foreign savings are exogenous, the basic fiscal problem for policy makers is to save adequate public sector resources to finance the investment budget after current budget needs have been met. If these saved resources are not of sufficient magnitude to meet the level of scheduled public investment. The result is the public sector savings-investment gap. This gap must subse¬ quently be rectified either by augmenting public sector savings, or by financing from monetary authorities and/or domestic commercial banks. Financing Investment by Public Sector Saving Public sector savings arise almost wholly from an excess of total disposable tax revenues over public con¬ sumption expenditure (representing a condition of budgetary surplus). However, this outcome, is more the exception than the rule, given the rapid growth of public sector consumption expenditure and the fact that it is not easy to increase tax revenue in LDCs to the degree needed to match spending. Low per capita incomes, which allow a 23 much smaller margin for taxation after subsistence needs are met, a heavy reliance on indirect taxes, which are not progressive and which are less responsive to economic changes and more sensitive to external shocks, and deteri¬ orating or highly volatile external trade environments have, in part, constrained the ability of many LDC govern¬ ments to mobilize tax revenues sufficient to meet the needs of their budgets. This fact is reflected in the compara- 1 tively low tax ratios of most LDCs. Whereas tax ratios in the twenty-one countries of the OECD (Organization for Economic Cooperation and Develop¬ ment) ordinarily averaged about thirty-four percent between 1972 and 1976, they were considerably less among LDCs. In a study conducted in 1979, average tax ratios for forty- seven developing nations from 1972-76 ranged from 8.8 per¬ cent to 15.4 percent for the lowest 15 and middle 32 coun- 2 tries in a 62 nation sample, while another group of devel¬ oping nations worldwide experienced an average growth of their tax ratios of only six percent between 1976 and 3 1982. Consideration must be given as to whether the rel¬ atively low level of taxation in relation to the economies of LDCs is due to their inadequate use of taxable capacity, and hence tax effort, or whether responsibility lies with some other less apparent factor(s). Elucidation of this question can be gained by examining historical data related 24 to tax efforts in selected developing countries. General determinations of tax effort are based upon the fact that different countries have different types of tax base or tax 'handles.' These tax handles depend mainly on the struc¬ ture of the economy. Hence, the amount of tax revenue that can be mobilized depends quite significantly on the char¬ acter of the economy. Once the tax handles are identified, the question as to whether the country is making adequate effort to realize its tax potential may be raised. One methodology developed in the literature is to express the ratio of tax revenue to GDP as a function of certain key factors which affect tax handles. Then, using regression techniques employing cross-country data, predict the poten¬ tial ratio for each country. Finally, an international tax comparison (ITC) index is obtained by dividing the actual ratio by the potential ratio. Countries with an index exceeding 1 are considered to be making satisfacto¬ ry tax effort and their fiscal problems are said to be 4 caused by other factors. The same study which characterized the tax ratios of LDCs found that the ITC indices for forty-four developing nations average 1.00 from 1969-71, 0.98 from 1972-1976, and 0.99 from 1978-80 with slightly over half the nations in the latter time frame experiencing an ITC index greater 5 than 1. These results indicate that the inability of LDC governments to raise adequate revenues must be due, in 25 part, to several factors other than inadequate tax effort. Some of these factors may include economic or government- imposed limitations to tax base growth and fiscal systems that are characterized by lags in tax collection, wide¬ spread tax evasion, or unresponsiveness to economic growth. The combination of low tax ratios and acceptable tax efforts would indicate that LDC governments have neither generally been in the position to significantly increase tax revenue nor to keep this revenue in line with growing levels of public consumption. In this light, while tax ratios in LDCs typically rose only marginally over the 1960s and 6 1970s, public sector consumption expenditures expanded over the same period, from eight percent of GDP in 1960 to 7 eleven percent of GDP by 1980. This overall rise in public consumption coupled with only moderate increases in tax ratios and, hence, tax revenues, has meant that the contribution of government saving to total domestic saving, and therefore to public sector investment, has been margin¬ al at best, a realization which becomes most apparent in 8 the abundance of government budget deficits for many LDCs. Here, the average government deficit as a percent of GDP for some fifty-eight developing countries from 1970 to 1983 9 was 3.4 with no average yearly surpluses for the period. 26 Financing Investment by Private Sector Saving Structural Restrictions. Private sector saving can 10 also be a source of capital formation in LDCs. However, like budgetary saving, opportunities for government borrow¬ ing at home from nonbank lenders are severely limited due to low levels of income and high propensities to consume on the part of the population and to other general limitations or disincentives to saving such as inadequate financial markets or artificially low interest rates. For capital formation to take place, adequate saving in the economy must exist. Resources available in an economy are used for the production of consumption goods as well as capital goods. Aggregate income is the value of both of these categories of goods. Hence, if at any time the supply of capital goods is to be increased, part of the resources used for the production of consumption goods has to be released and directed into the making of capital goods. Since private savings constitute the difference between income and consumption, the extent to which the supply of consumption goods is restricted measures the addition of private saving to the economy. Data complied by the World Bank for thirty-one devel¬ oping nations worldwide indicate that LDCs have, for the most part, restrained growth in private consumption over 27 long periods, experiencing only a 0.2 percent average in- 11 crease from the period 1960-70 to the period 1970-80. However, private domestic consumption did not grow at sig¬ nificantly slower rates than did GDP. In fact, the growth rate advantage of gross domestic product over this variable significantly slowed from 37.5 percent for the period 1960- 12 70 to 2.9 percent for the period 1970-80. Moreover, the share of private consumption in GDP from 1960-80 actually 13 rose, if only slightly, from 83 to 84 percent. These considerations point to the same conclusion: that private savings could not have increased significantly over the period and may even have decreased. This is ultimately reflected in the fact that the share of gross domestic \ saving in GDP from 1960-70 to 1970-80 periods actually 14 dropped from 9 percent to 7 percent. Because they include some of the poorest in the world, the statistical construction for twenty-three sub-Saharan African nations is even less encouraging. While the aver¬ age annual growth rate of private domestic consumption among these countries between the periods 1960-70 and 1970- 82 slowed by 20 percent (from 3.6 to 3.0 percent), the growth rate margin of GDP over private domestic consumption 15 plummeted from +11 percent to -40 percent. Gross domes¬ tic savings' share in GDP also decreased during the same 16 period from 12 percent to 5 percent. 28 Financial Sector Restrictions. As capital formation in an economy is dependent upon adequate savings, so, too, is it dependent upon the agencies through which these resources can be mobilized and channeled into investment. It is through the securing of public savings that investors (in this case the government) get command over real resources which can then be used for the production of capital goods. Savings are transferred to the investors through the selling of securities. In many developing countries, however, this process is not easily accom¬ plished. Few persons are accustomed to buying securities of any kind and savers are often highly skeptical of gov¬ ernment securities. Time is required to establish the institutions and practices that make possible wide volun¬ tary purchases of government securities. For the most part, then, the banking and financial institutions and the markets which could facilitate this process are few in number in LDCs. Also, bonds or securities are not popular as widespread confidence may be lacking in the financial 17 and political stability of the government. To add to these circumstances, the chronic presence of inflation (among LDCs and developed countries alike) has left little incentive for investment in government securities. As Richard Goode states: Recently, the spread of inflation to other coun¬ tries, together with wide fluctuations in nominal 29 interest rates, has damaged the market for gov¬ ernment bonds in the small number of countries in which it previously flourished.^ Few countries made more than marginal adjustment in nominal deposit rates in relation to these changes in inflation. As a result, real interest rates after the early 1970s turned significantly negative, particularly in many Asian 19 and Latin American nations. Negative real rates of in¬ terest were even more common and enduring in some African countries in the period after 1975, with real interest rates for 1976, 1978 and 1980 averaging -10.2, -14.0, and 20 -8.4 percent, respectively. The negative nature of real interest rates will affect the growth of the financial system as a whole and, ultimately, the growth in available public savings for investment purposes. In sum, opportunities for government borrowing at home from domestic non-bank lenders are limited in most LDCs by low saving stemming from rising shares of private con¬ sumption in GDP and, more importantly, marked declines in the growth rate advantages of GDP over such consumption. Also, saving is discouraged by incipient or shrinking fi¬ nancial markets caused by artificially-imposed low interest rates. On this basis and without corrective measures, it is unlikely that private savings can be considered as an adequate source of investment finance in LDCs. 30 Financing Investment by Money Creation Demands for public spending generated by the policies and aspirations of governments, together with the difficul¬ ties of taxation and borrowing at home, have caused many governments to resort to money creation. Financing by money creation appeals to weak governments because, unlike taxation, no administrative machinery is required and no 21 administrative cost is incurred. Money creation and borrowing are distinct in principle but may be confused because the most important kind of money creation takes the legal form of borrowing from the central bank. While there may be some exceptions, government net borrowing from the central bank generally does not displace other lending but results in an addition to currency and deposits. The balance sheet of the central bank initially shows equal increases in assets in the form of claims on government and in liabilities in the form of government deposits and currency. As the government makes payments to contractors and employees, there is an increase in deposits and curren¬ cy in the hands of enterprises and households, in effect, representing a net increase in the money supply. Government borrowing from domestic commercial banks may or may not entail a net addition to the money stock. 22 If the banks have no excess reserves, i.e., they are fully loaned up, no additional money will be created by 31 government borrowing. In this situation, government bor¬ rowing displaces loans made by the banks. If, however, the commercial banks have excess reserves, they can lend to the government without contracting their credit to their regu¬ lar customers and a net addition to the money supply oc¬ curs. This increase in the supply of money results in a comparable increase in aggregate spending as the new incre¬ ment to the stock of money is used to buy additional goods and services. In both cases, borrowing from the central bank (mone¬ tary authorities) and borrowing from domestic commercial banks with excess reserves, inflation may, or may not, be the major result. If the increase in the money supply does not exceed the quantity that the population at large de¬ mands at stable prices, money creation to finance the government deficit will not be inflationary. However, when financing of government expenditures by money creation exceeds the noninflationary limit, total spending in the country becomes greater than production valued at stable prices or, in other words, the rate of expansion in the supply of money in the economy exceeds the rate of expansion of the demand for money. The resulting imbalance will generate significant inflationary pressure in the economy which, in turn, will affect resource allocation and 23 income distribution. Inflation also affects tax revenue and government 32 expenditures. These effects must also be taken into con¬ sideration when assessing the possible net fiscal contribu¬ tion of financing by money creation. With regard to tax revenue, governments have long recognized inflation as an alternative means of securing resources for the state. All that is required is that the stock of money be expanded at a sufficiently rapid rate to result in increases in the general price level and that people be willing to hold some money balances even as the value of these holdings decline. Inflation then acts as a tax on holdings of money. With regard to inflation taxation, it is important to consider the elasticity of tax liabilities (specific tax classifications) and an additional factor, the length of time between tax accrual and tax payment, or the collection 24 lag. Inflation will reduce or augment the real value of liabilities for specific taxes and all other taxes with an elasticity in relation to nominal income below or above 1, respectively. Unfortunately, few LDCs have been able to secure tax systems with overall elasticity equal to or greater than 1 because of a heavy reliance on indirect taxes which are considered to be less elastic than revenue generated from direct taxes on income, profit, goods, and 25 services. If tax liabilities are not indexed, the real value of tax collections will be reduced because of the collection 33 lag. At any given inflation ratef the longer the lag, the greater the reduction in the real value of collection; and with any given lag, the faster the inflation, the greater the reduction. As an example, if prices are rising by one percent a month, a three-month lag in tax collection will rescue the real value of payment by about three percent. In this scenario, the government's loss of real tax revenue because of the collection lag can exceed the real resources that it obtained from central bank financing. Such was the 26 conclusion in a study conducted by Vito Tanzi in 1978. Inflation may also have an enhanced affect on govern¬ ment expenditures and, therefore, a detrimental affect on the size of budget deficits. For expenditures, the automa¬ tic response to inflation is less important than discre¬ tionary adjustments, at least in the short run. According to Richard Goode: Salaries and wages, a major component of expend¬ iture, usually lag prices in the early stages of inflation and later are subject to irregular adjustments, with periods of declining real com¬ pensation followed by large corrective increases. Conversely, governments cannot control the prices they pay for most other goods and services in the same way, but within limits they can alter the real volume of purchases.27 Bijan B. Aghevli and Mohsin S. Kahn have advanced the hypothesis that government expenditures are adjusted more quickly to inflation than are taxes, with the result that the budget deficit is enlarged. The financing of this inflation-induced deficit would, in turn, increase the 34 money supply and generate further inflation. Thus, "the increase in the supply of money would both cause inflation and be the result thereof," the process becoming self- 28 perpetuating. Thus, money creation as a means of financing investment appears for all practical purposes to be inappropriate. Despite this fact, however, LDC governments have resorted to this form of financing far more than borrowing from the non-bank public. Data compiled by the International Mone¬ tary Fund for fifteen developing nations list average fi¬ nancing by money creation as a percentage of total finan¬ cing for 1975 and 1981 as 29.6 percent and 33.8 percent, 29 respectively. Financing from the non-bank public for these same nations constituted only 10.1 percent of 30 total financing for 1975 and 9.6 percent for 1981. Summary Clearly, LDC governments have not been in a position to finance their deficits by relying upon their own domes¬ tic resources. Public sector saving was seen to be nonex¬ istent in most cases, while nonbank and money-related fi¬ nancing of total government yearly deficits were, together, not sufficient to meet the entire budget deficit. More¬ over, monetary financing of deficits was seen to lead to inflation, in most cases, defeating the purposes of this type of financing. Thus, it remains for resources from 35 external sources to bridge the gap between budgetary needs and available domestic resources. This influx of foreign budgetary assistance will have serious ramifications both for recurrent costs and, ultimately, for the ability of developing nations to meet these costs. How this occurs will be the subject of the next chapter. 36 ENDNOTES 1. The level of taxation or tax ratio is customarily measured by the ratio of tax revenue to gross domestic product. The tax ratio depends upon taxable capacity which, in turn, depends upon the ability of people to pay and the ability of governments to collect. Tax effort is the degree to which taxable capacity is used. Richard Goode, Government Finance in Developing Economies (Washing¬ ton, D.C.: The Brookings Institution, 1984), p. 84-85. 2. Alan A. Tait, Wilfred M. Gratz, and Barry J. Eichengreen, "International Comparisons of Taxation for Selected Developing Countries, 1972-76," International Monetary Fund Staff Papers 26 (March 1979) , pp. 123-156. These figures represent unweighted arithmetic means. The tax ratios were computed using GDP. 3. International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985), p. 80. Tax ratio growth was computed using GDP. The average tax ratio for 1976 was 18.26 while the average tax ratio for 1982 was 19.37. 4. Tait, Gratz, and Eichengreen, op. cit., p. 125. 5. Ibid., p. 125. 6. Available statistics list the average tax ratio growth rate for 39 developing nations worldwide from 1973 to 1983 as approximately 0.7 percent. International Mone¬ tary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985), table on General Government Tax Revenue, p. 80. 7. The World Bank, World Development Report, 1982 (Washington,D.C., 1982), Table 5, pp. 118-119. 8. Malcolm Gillis, Dwight Perkins, Michael Roemer, Donald R. Snodgrass, Economics of Development (New York: W.W. Norton and Company, 1983), p. 276. 9. International Monetary Fund, International Finan¬ cial Statistics Yearbook, 1985 (Washington, D.C., 1985), pi 144. The statistics are even less encouraging for 39 sub-Saharan African nations with average deficits of -4.8 percent in 1972 and -5.5 percent in 1981. 37 10. Gillis, et al., op. cit., pp. 268-269, 276. The extent of private sector saving in the financing of domes¬ tic investment is not as easily resolved numerically as is public sector saving. Statistics are either unavailable or highly aggregative. Therefore, any analysis of this aspect of total domestic saving must be conducted using some degree of inference. 11. The World Bank, op. cit.. Table 4, p. 116. 12. Ibid., Table 2, p. 112 and Table 4, p. 116. Growth rate margin of GDP over private domestic consumption com¬ puted using the equation: Y - C = S where Y = gross domes¬ tic product, C = private domestic consumption, and S = private domestic savings. 13. Ibid., Table 5, p. 118. 14. Ibid. 15. The World Bank, Toward Sustained Development in Sub-Saharan Africa: A Joint Program of Action (Washington, D.C., 1984), Table 2, p. 58 and Table 4, p. 60. 16. Ibid. Table 5, p. 61. 17. M. L. Jhingan, The Economics of Development and Planning, 15th ed. (New Delhi: Vikas Publishing House, PVT., LTD., 1982), p. 292. 18. Richard Goode, Government Finance in Developing Economies (Washington, D.C.: The Brookings Institution, 1984) , p. 198. 19. Gillis, et al., op. cit., p. 337, 352. 20. Ibid. This information was adapted from Table 13.5, p. 353. The real rate of interest is equal to the nominal rate of interest minus inflation. 21. J. M. Keynes, The Collective Writings of John Maynard Keynes, Vol 4., A Tract on Monetary Reform (London: Macmillan for the Royal Economic Society, 1971) , quoted in Richard Goode, Government Finance in Developing Economies (Washington, D.C.: The Brookings Institution, 1984), p. 8. 22. Excess reserves = actual reserves - required reserves. 23. The World Bank, "Public Finance in Egypt: Its Structure and Trends," World Bank Staff Working Papers Number 639 (1984), p. 76. 38 24. For an elaborate consideration of the role of the tax lag in inflation taxation see Bijan B. Aghevli and Mohsin S. Khan, "Government Deficits and the Inflationary Process in Developing Countries," International Monetary Fund Staff Papers 25 (September 1978), pp. 383-414. 25. Gillis, et al., op. cit., p. 324. See also Richard Goode, Government Finance in Developing Economies (Washing¬ ton, D.C.: The Brookings Institution, 1984) , p. 9l, Table 4-2. Of fourteen less developed countries, 41.6 percent of their total revenue was accrued from international trade receipts and 17.0 percent from income and profits. This compares with 3.7 percent and 33.3 percent, respectively, for 20 industrial market economies. 26. Vito Tanzi, "Inflation, Lags in Collection, and the Real Value of Tax Revenue," International Monetary Fund Staff Papers 24 (March 1977), pp. 154-167. 27. Richard Goode, op. cit., p. 223. 28. Bijan B. Aghevli and Mohsin S. Khan, "Government Deficits and the Inflationary Process in Developing Coun¬ tries," International Monetary Fund Staff Papers 25 (Sep¬ tember 1978), p. 409. 29. International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985). Financing by money creation was computed by adding lines 17.2 and 17.3 and dividing this total by line 15. 30. Ibid. Financing from the non-bank public was computed by dividing line 17.1 by line 15. 39 CHAPTER 4 OFFICIAL DEVELOPMENT ASSISTANCE AND RECURRENT COSTS Because deficit financing from domestic sources is limited for most LDCs, governments have turned to external sources as a means of meeting total budget needs. In this respect, the primary vehicle for support has been official development assistance (ODA) from abroad. However, while permitting current expenditures and investment to proceed at planned levels, such official assitance presents some very real implications for the overall fiscal situation of the typical LDC in that it affects both the long run, aggregate level of project-related recurrent costs and, hence, the expenditures required to support them, and the potential revenue which could be used to meet such costs. From the expenditure perspective, both the level of ODA and the process which effects its transfer, act to increase project-related recurrent costs. From the revenue perspec¬ tive, ODA can supplant rather than supplement fiscal re¬ sources, discouraging self-reliance and encouraging LDC governments to reduce their levels of taxation and borrow¬ ing. Further, ODA can encourage such governments to remove resources from their current budgets (which supports pro¬ ject-related recurrent costs) to support their investment 40 budgets. Hence, the influx of ODA into a given country both increases long-run, aggregate project-related recur¬ rent costs and decreases the fiscal resources used to meet such costs. This chapter will explore each of the major expendi¬ ture and revenue effects of ODA on the ability of LDCs to meet recurrent costs associated with their investment pro¬ grams over the long run. In the course of this discussion, particular emphasis will be given to the interaction be¬ tween these donor and recipient entities involved in the transfer of aid. The involvement of the U.S. Agency for International Development (AID) in this process will be used to provide descriptive examples of the impact of this interaction. Expenditure Effects As evident from the discussions of the preceding chap¬ ter, the opportunities for LDC governments to generate sufficient revenue to meet their total (current plus capi¬ tal) budget needs are limited. Hence, in order to meet planned development goals, such governments must solicit the help of foreign governments for investment capital to supplement their own development resources. In the pro¬ cess, both the level and the character of investment will be altered. As this occurs, the recurrent cost burden will also change. 41 Increased Investment In view of the limited fiscal capabilities of LDCs and in light of the continuing emphasis on the primacy of investment as an engine for growth or income redistribu¬ tion, ODA has come to comprise more and more of the fiscal makeup of LDCs. Total official aid flows from all sources to LDCs worldwide have, in fact, increased, nominally, 537 percent, from $4.94 billion in 1960 to $31.66 billion in 1 1980. This growing conspicuousness of ODA among LDCs as a group portends an overall increase in the recurrent costs associated with their public sector investments over time. Obviously, once the productivity of any given project is reliant upon future outlays for its operation and mainte¬ nance, any increase in the amount of overall investment will be accompanied by a similar increase in the recurrent 2 costs associated with this investment. This process, by itself, is not harmful. The deleterious effect of ODA on these recurrent costs arises from the ability of such assistance to enhance a given country's investment budget without directly encouraging the growth of recurrent reve- 3 nues to support such costs. Thus, the effect of ODA has been, and continues to be, a sharp expansion in the long run, aggregate recurrent cost commitments of LDCs. In view of the limited resource¬ generating capabilities of such nations, and in the face of 42 other pressures on the recurrent budget (to be described in the following chapter), these commitments will be difficult to meet. Changes in the Content of Investment The role of ODA in altering the content of investment in LDCs to include projects higher in recurrent costs can best be interpreted as originating from the two primary operational elements of the aid transfer process acting separately or in concert. These are: -The donor and its institutional-level transfer of aid; and -The recipient and its response to aid. From the perspective of the donor, it is the very nature of its operational environment and how this environment af¬ fects the manner in which its organizational output gets defined, which may be the most blameworthy where contribu¬ tions to the recurrent cost problem are concerned. In this regard, donors will be forced to adapt to their institu¬ tional environments to ensure organizational vitality, in the process losing substantial control over their own policies. The role of the recipient in exacerbating this problem will be seen as important chiefly in the manner in which it responds to overtures or controls made or exerted by the donor within the context of its institutional envi¬ ronment. Specifically, recipient countries will alter both their development policies and their fiscal practices to 43 meet such externally-influenced donor guidelines, thereby maximizing their eligibility for foreign assistance—prac¬ tices which both increase the long run, aggregate recurrent cost burden and reduce revenues available to meet these costs, respectively. The mechanics of this process are most easily understood by examining the environment and the institutions involved in American aid transfer. Donor Characteristics and External Incursions. The organizational autonomy or political clout of the foreign assistance agency is weakened because of the nature and location of its task environment i.e., where the majority of its programs are instituted and the bulk of its funds spent, lies outside the experience of those agencies which seek to regulate or constrain it. As a result, supervising agencies are not readily capable of evaluating aid-dispen¬ sing organizations on their own terms i.e., they do not employ specialized criteria apart from that used to guage the performance of all other domestic money-spending insti¬ tutions. Rather, the same evaluation norms that are used to assess the performance of domestically-oriented agencies are applied to those responsible for disbursing foreign aid. This practice originates not only in the inability of supervisory agencies to understand the nuances and mechan¬ ics of the aid process but, more importantly, their seeming unwillingness to comprehend and adjust to the specific regulatory needs which the aid process engenders. 44 The plight of the foreign assistance agency is further compounded by virture of the fact that at least some of the agency's workload is contributed by its beneficiary, in this case, the recipient country government. This fact worsens the donor's predicament in relation to its supervi¬ sory agencies since the donor's output can be largely determined by groups of individuals outside its purview while at the same time being subject to the constraints of the supervisory agencies without benefit of special criter¬ ia or domestic political support. Both of these elements, the uniqueness and the exter- nalization of the task environment, thus work to the detri¬ ment of the foreign assistance agency in the United States. The former does so by weakening its primary means for budgetary vitality and procedural autonomy and the latter by presenting a high degree of uncertainty and unfamiliar¬ ity to those agencies charged with its oversight. Thus, oversight or supervisory agencies, in an effort to regain a certain degree of control over the externalized task environment and, hence, over the foreign assistance institution, will intercede in the development assistance agency's everyday operations, imposing standards and con¬ straints similar to those applied to other bureaucracies, but in a more rigorous fashion, in essence redefining the relatively unknown task environment of the donor agency in 45 terms they can understand. This is summed up in Judith Tendler's Inside Foreign Aid: The watchdogs [supervisors] seem to compensate for their lack of familiarity with the world in which foreign aid projects are built by making a stricter and more comprehensive application of routine check-and balance criteria—as if to make up for the loss of one sense faculty by the in¬ creased use of another. This increased rigor¬ ousness, of course, only emphasizes the inappli¬ cability of such criteria to areas outside the domain for which the criteria where devised.^ In sum, both the distant location of the foreign aid bureaucracy's beneficiaries and, more importantly, the uniqueness of its task environment, remove the 'scene of action' from the watchdog entities' world of experience. This leads to increased suspicion of the donor agencies by the supervisory agencies who must, in turn, step up their regulatory efforts in order to bring the aid bureaucracy's task environment into line with their realm of experience— a process accomplished by penetrating the everyday activi¬ ties of the donors and seeking to control most of the phases of the aid-transfer process. Such incursions into the donor agency's affairs can proceed virtually unopposed because the normal means of support to prevent unnecessary and inappropriate controls, i.e., conventional, domestic constituencies, cannot readily be utilized. In light of the complications arising from the pecu¬ liar nature of the task environment of foreign assistance, and in order to ensure organizational vitality, these donor 46 agencies must attempt to align their procedures and pro¬ grams in such a way as to satisfy, to the greatest degree possible, those agencies responsible for their supervision. When donor organizations accede to the demands of their environments, the externally-imposed standards and con¬ straints which characterize these environments are said to be 'institutionalized' into the decision making process of the donor entity. As a result, even such key project select¬ ion variables as economic return and the ability of recip¬ ient country governments to sustain these projects over the long run are frequently overlooked. As this 'institution¬ alization' of external standards and constraints proceeds, agencies lose most of the control over their capacity to provide meaningful assistance to developing nations while, at the same time, recipient countries may lose valuable input into those projects which may be instituted. In the long run, a suboptimization of aid transfer can occur as aid-dispensing agencies engaged in institutionalization become more preoccupied with protecting themselves from incursions by the supervisors than with enunciating and pursuing their original goals. The Impact of External Incursions. The most notable impact of institutional incursions into American aid agen¬ cies is related specificially to legislation designed to protect private or public U.S. interests. However, despite the fact that this legislation poses some restrictions for 47 donors, it is not so much the nature of this legislation but the use of it by supervisory agencies to expand the scope of their influence in and, hence, control over, the everyday operations and policy making processes of the aid agencies. This fact is most critical when considering the potential effect on the content of investment. In this respect, Tendler found that certain government entities: ...specifically charged with policing [the] legi¬ slative constraints frequently ended up having a power over the organization considerably greater than the original [legislative] constraint intended—a power that spread into areas where may have had no authority Thus, while the constraints in themselves posed one problem for donor agencies, even greater impacts arose from the power over these agencies gained by the policing entity "in the form of its daily presence—actual, expected or 6 feared—on the agencies' work scene." While this type of incursion is typical among the Washington bureaucracy, it is especially tough on assistance organizations "since they 7 are peculiarly unequippped for such struggle." Conse¬ quently, agencies must "give in to the desires of the other [controlling] entity[ies] or alter [their] decisions to 8 [or reduce the impact of] expected incursions." In this way, incursions into agencies responsible for the disbursal of American aid (such as the Agency for Interna¬ tional Development) may lead to a significant displacement of their goals. Moreover, donor agencies come to concern 48 themselves not so much with accusations by Congress and other audit organizations that they are subverting their own goals as with the warding off of incursions and criter¬ ion by other controlling agencies. This is accomplished by coming to identify with the very interests of those agen¬ cies the donors are trying to fend off, even when such behavior supports interests or goals counter to those of the aid agency. In the course of such bureaucratic adaptive behavior on the part of aid agencies, such agencies relinquish a great deal of their control as to the nature and scope of development projects they wish to see instituted. Subse¬ quently, if criteria used to select projects are dominated by outside influences, the propensity for such projects to possess intensified recurrent costs is high. Adaptive Responses to Incursions. Two vantage points must be considered when examining how adaptive responses to external incursions on donor agencies affect project selec¬ tion and, hence, project-related recurrent costs. First and foremost is the specific nature of the adaptive respon¬ ses used by donor agencies to minimize the potential nega tive repercussions of noncompliance with the wishes of supervisory agencies. Second is the manner in which recip¬ ient countries respond to these externally-evoked adapta¬ tions. 49 When such adaptive responses, on the part of donors and recipients alike, become incorporated into the aid transfer environment, the supply of fundable projects will effectively be diminished. In view of this fact, both donors and recipients will perceive assistance funds as becoming increasingly abundant. Subsequently, because funds are abundant and fundable projects are not, economic rationing as the primary criterion in project selection, scope, and design will not receive adequate consideration. Rather, donor agencies will incorporate other criteria, such as those which maximize the use of funds or which allow project proposals to be moved into their final planning stages expeditiously, into their aid disbursal procedures. When this evolves, an inefficient or inappro¬ priate allocation of aid resources is said to occur as large, capital and import-intensive projects are incorpor¬ ated into the development plans of LDCs, in turn, raising long run aggregate recurrent costs. Because their own development resources are scarce, recipient country govern¬ ments must, in turn, rearrange their development priorities to maximize their eligibility for official development assistance. As Friedmann et al. observed: "To qualify for the financing, recipients may place emphasis on projects that meet the requirements of the lending agencies but which may not be of high priority in relation to their 9 needs." This form of adaptive behavior also contributes 50 to the inception and incorporation of large, import and capital-intensive projects into the development plans of recipient countries. There are essentially three major adaptive behaviors which development assistance agencies engage to satisfy the demands of their oversight agencies: a pressing need to accomplish spending i.e.f to effect a transfer of aid funds? a penchant for financing only the foreign exchange or import costs of a project? and a redirection of the project generation function toward the donor rather than the recipient. The output of the development assistance organization seems to have been defined in terms of the total amount of resources successfully transferred during any given period, regardless of its ostensible mission to promote develop¬ ment. This redefinition of organizational output origi¬ nates both from within the organization, in the form of management's response to external supervision, and from external sources which have some interest in the disburse¬ ment of development assistance. Such redefinition has caused a 'quantity-at-any cost' approach to the task of transferring development assistance precluding such all- important project selection criteria as scarcity, applica- 10 bility, and productivity. From within the organization itself, the linking of output with the movement of money pervades the standards of procedure, the incentive • • • 51 system, the career motivations, and the work environment of 11 the foreign assistance entity." Specifically, "...new loan officers are considered bright and energetic if they 12 are good at moving money." Further, "...country mission directors know they must move certain amounts of money in order to prove their worth and can distinguish themselves 13 further by moving even more." An emphasis on money- moving does, in fact, satisfy management and employee needs because, if one is financing projects, "...the staff input on any particular project will obviously not increase 14 proportionately with the amount of money to be lent." Hence, economies of scale will come into play when the maximum possible amount of money is moved. The need to move money also stems from influences outside the agency. These influences are associated speci¬ fically with the need to utilize all appropriated resources L before the end of the fiscal year and with constraints imposed by agencies external to the donor organization (such as Congressional reviews and appropriations) which hamper the release of funds among aid disbursing agencies more than other domestic money-spending agencies. This has profound implications for project selection. By way of explanation, since large projects utilize greater portions of the available 'aid pool,' they are more likely to be chosen under such circumstances than smaller, perhaps more 52 socially useful, ones. Further, larger (or more technical¬ ly involved) projects will hold greater appeal in the eyes of those responsible for project generation because a lar¬ ger project is more efficient than smaller ones, requiring 15 less staff time per dollar transferred. Supporting evi¬ dence of donor institutions advocating large project scale as a means of satisfying pressure groups in their own bureaucracies was noted in a study conducted by the Club du 16 Sahel for eight specific nations in Africa. In this way, "a donor organization's sense of mission [comes to] relate not necessarily to economic development but to the commit- 17 ment of resources, the moving of money." Indeed, in view of such internal motivating factors and external pressures, a development assistance agency's annual appropriations may come to be viewed by its management and its employees as an inert mass which must be moved even at the expense of traditional criteria used for meting out funds. This need to accomplish spending instead of rationing subverts the use of scarcity as a prime project selection criterion and, subsequently, "decisions regarding projects cannot help but 18 be influenced by this absence of a sense of scarcity." Another constraint placed upon U.S. development assis¬ tance organizations by external oversight agencies is that requiring such organizations to finance only the foreign exchange or import costs of their development projects. Specifically, the U.S. aid program is required by procedure 53 to limit most of its project financing to import costs and is subject to external controls by the Treasury, State, and Commerce Departments regarding the amount of acceptable local cost financing in proposed AID project loans. In this regard, statistics show that more than eighty percent of total AID funds continue to be spent in the United 19 States. Moreover, this import-cost constraint is not confined solely to U.S. bilateral assistance. The World Bank's Articles of Agreement also limit financing to the foreign exchange cost of projects, accounting for about 20 seventy-five percent of its total disbursable funds. The rationale of supervisory agencies in requiring projects with high import content stems from such diverse ends as the improvement of the balance-of-payments situa¬ tion, in the enhancement of specific export industries or sectors, and the control by donor governments over the fiscal and developmental courses of the LDCs receiving the aid. However, the most important reason for the policy of requiring donor agencies to finance only the import costs of projects, at least along the lines of this discussion, is that it removes a great deal of the uncertainty indigen¬ ous to the aid transfer process by ensuring confidence over 21 the sale of 'one's own product.' This emphasis on famil¬ iar equipment financing means that supervisory agencies can easily understand the nature of the aid being transferred. 54 ensuring quick disbursal of loan finds and limited project 22 supervision problems. If outside controls dictate to the foreign assistance agency that only projects high in import or foreign ex¬ change content will be acceptable and again, if funds are abundant while acceptable projects are not, then these factors, together with a prevailing pressure to move funds, will necessitate an increase in the number of import¬ intensive projects in the inventories of donor agencies and, likewise, in the development programs of LDCs. The emphasis on import-intensiveness, by nature, leads to capi¬ tal-intensiveness as well for the simple reason that impor¬ ted equipment is a product of the relative labor scarcity of the developed world. The implications for long run, aggregate recurrent costs should be quite clear, i.e., import and capital-intensive projects may become dependent for long afterward, on materials and spare parts from 23 capital rich exporting countries. The last major adaptive response which development assistance organizations engage to reduce the uncertain¬ ty of the aid transfer process and satisfy their oversight agencies is the redirection of the major portion of the project generation function from the recipient country or entity to the donor organization. In development assistance, one of the items which contributes most toward uncertainty is the beneficiary's 55 input. This input is often inexact or incomplete as most developing countries do not have an existing infrastructure of sufficient capacity to produce "the kind of bureaucratic output required to [generate and] qualify for [projects] 24 and later monitor such assistance." This inability on the part of the recipient to adequately generate sound projects thus necessitates that the development assistance agency reassume the project generation function to avoid potential inefficiency and loss of control. Such relocation allows the agency to accomplish two goals: the achievement of control over the "administrative and financial conditions under which recip¬ ient countries develop and operate specific sectors of their economy" and "the [enhancement of] the reputation for effectiveness of the financing agencies, thereby facilita- 25 ting future contributions by taxpayers or investors." In this case, the former exists only to serve the latter goal of establishing donor agency efficiency and integrity in the eyes of its supervisory agencies, and the Washington bureaucracy and American constituency at large. In sum, if the task environment of development assis¬ tance is uncertain because it is once removed from the donor, then a reduction of this uncertainty must certainly be achieved by bringing the random occurrences of this environment within the realm of understanding and predicta¬ bility. This can be accomplished by reassuming the major 56 portion of responsibility and control over the inception, character, and scope of development projects. In this way, donors will be free to opt for the larger, more import and, hence, capital-intensive projects which meet their goal of efficiency and which satisfy their supervisors. Like the donor, the recipient also tends to perceive the supply of development assistance as abundant, if not 26 infinite. At least part of the reason for this percep¬ tion of abundance stems from the fact that the borrower relates the supply of foreign assistance to the size of the single demand that he is making and does not consider the demands of the other developing nations. Subsequently, ...the goods (foreign financing) are his for the asking, as long as he can pay the 'price.' In his eyes, the price is the putting together of a project proposal(s) which qualifies [or which qualify] for such financing.2? In this regard, qualified project proposals will, by cir¬ cumstance, come to include large size and import and/or capital-intensiveness. Revenue Effects Taxation and Borrowing, and Leakage from the Recurrent Budget When the perceived abundance of foreign development assistance is contrasted with the borrower's acute percep¬ tion of the finite nature of his own domestic finances, the tendency to economize on the latter in favor of the former becomes pronounced. Like the rearrangement of priorities 57 in development, this practice is also contrary to the goal of economic self reliance as it minimizes the need for fiscal responsibility, one of the primary long-term goals of foreign aid. Specifically, when 'abundant' funding for the import costs of a given project is presented to a capital-scarce country, the rationale is that this country will be "highly motivated to raise the complementary local cost resources for the project on its own—something it might not have done without the incentive of development 28 financing." It has been suggested by a number of studies that this is not the case. While aid-supplied foreign exchange is designed to function as a complement to the domestically (fiscally)-generated resources for develop¬ ment projects, it is seen by many developing countries as an alternative to these revenue supports. Hence, govern¬ ments, finding abundant resources abroad, tend to reduce savings and refrain from raising taxes. Such was the 29 finding of Hollis Chenery and Alan Strout in 1966 and K. B. Griffen and J. L. Enos in another study conducted in 30 that same year. The fiscal impact of aid becomes even more apparent in an econometric analysis conducted by Heller in 1975. In this study Heller showed an inverse relationship between incoming official development assistance and the level of taxation—taxation being reduced by nineteen percent for every dollar of aid received. Moreover, foreign assistance 58 in the form of loans was found to pull nonloan resources from the recurrent budget rather than induce recipients to 31 augment this budget by their own means. Finally, Vel Pillai, in another study conducted in 1982, also demonstrated the negative effect on revenues exerted by foreign aid, concluding that foreign aid was, in fact, substituted for domestic revenues, the remainder supporting 32 investment. Summary Clearly, official development assistance has exerted a pervasive, and in many respects, detrimental effect on the ability of developing countries to meet the recurrent costs associated with their development projects. This effect occurs through the combined efforts of both the donor and the recipient as both attempt to maximize their utility, the donor by granting the greatest amount of aid possible, and the recipient by obtaining the greatest amount of available aid subject to the criteria set forth by the donor. The two main manifestations of ODA, impacts from the expenditure side and impacts from the revenue side of the 'budgetary equation,1 were also shown to be inextric¬ ably linked to the interaction between donor and recipient. From the vantage point of expenditures, ODA increases pro¬ ject-related recurrent costs by increasing the overall 59 amount of new investment being incorporated into the devel¬ opment programs of LDCs and by significantly altering the content of these programs to include a greater number of large, capital and import-intensive projects. When this increase is matched with the demonstrated manner of fiscal response to foreign aid on the part of the recipient, (revenues available to meet recurrent costs being significantly curtailed), the ability of developing nations to successfully cover their project-related recur¬ rent costs diminishes. This reduced capability is further constrained by other non-ODA-related domestic fiscal policies and practices. These policies and practices will be the subject of the next chapter. 60 ENDNOTES 1. Malcolm Gillis, Dwight Perkins, Michael Roemer, Donald R. Snodgrass, Economics of Development (New York: W.W. Norton and Company, 1983). Table 14.1, p. 367. Offi¬ cial aid from all sources is taken to mean Bilateral Devel¬ opment Assistance of the DAC (Development Assistance Com¬ mittee of the OECD) consisting of 17 industrialized aid donors, bilateral OPEC (Organization of Petroleum Exporting Countries) and multilateral agencies. Flows from CMEA (Council of Mutual Economic Assistance, Eastern Europe and the Soviet Union have been deliberately excluded. 2. Peter Heller, "A Model of Public Fiscal Behavior in Developing Countries: Aid, Investment, and Taxation," American Economic Review 65 (June 1975), pp. 435-441, and "The Underfinancing of Recurrent Development Costs," Fi¬ nance and Development 16 (March 1979), p. 39. 3. Heller, "The Underfinancing of Recurrent Develop¬ ment Costs," Finance and Development 16 (March 1979), p. 39. 4. Judith Tendler, Inside Foreign Aid (Baltimore: The Johns Hopkins University Press, 1975) , p. 41. 5. Ibid., p. 42. 6. Ibid., p. 44. 7. Ibid. 8. Ibid. 9. Wolfgang Friedmann, George Kalmanoff, and Robert Meagher, International Financial Aid (New York: Columbia University Press, 1966), p. 406. 10. Raymond Mikesell, The Economics of Foreign Aid (Chicago: Aldine Publishing Company, 1968) , p. 142. 11. Tendler, op. cit., p. 56. • C M i—1 Ibid., p. 57. 13. Ibid. 14. Ibid., p. • 00 00 15. Ibid., p. 95. 61 16. Club du Sahel/CILSS, Recurrent Costs of Develop¬ ment Programs in Countries of the Sahel: Analysis and Recommendations (Paris: OECD, 1980), p. 295. 17. Tendler, op. cit., p. 89. 18. Ibid., p. 87. 19. Ibid. 20. Ibid., p. 73. 21. Ibid., p. 74. 22. Ibid. 23. K. B. Griffin and J. L. Enos, "Foreign Assistance Objectives and Consequences," Economic Development and Cultural Change 18 (April 1970), p. 323. 24. Tendler, op. cit., p. 105. 25. Ibid., p. 103. 26. Ibid., pp. 54-55, 79-80, 102. 27. Club du Sahel/CILSS, op. cit., p. 103. 28. Tendler, op. cit., p. 74. 29. Hollis Chenery and Alan Strout, "Foreign Assis¬ tance and Economic Development," American Economic Review 56 (September 1966), p. 679-733. 30. K.B. Griffin and J.L. Enos, "Foreign Assistance: Objectives and Consequences," Economic Development and Cultural Change 18 (April 1970), p. 321. 31. Heller, "A Model of Public Fiscal Behavior in Developing Countries: Aid, Investment, and Taxation," American Economic Review 65 (June 1975), pp. 429-45. 32. Vel Pillai, "External Economic Dependence and Fiscal Policy," The Journal of Development Studies 19 (October 1982), p. 16. 62 CHAPTER 5 MEETING RECURRENT COSTS: SHRINKING FISCAL RESOURCES In addition to the response of the recipient to for- reign economic assistance, there are certain pronounced governmental policies and budgetary practices, specifically characteristic of many LDCs, which act alone or in concert, to reduce the ability of such nations to adequately meet the recurrent costs associated with their development pro¬ jects. Such policies and practices can best be understood both from the vantage points of their constraining effect on current domestic revenue and, on the expenditure side, from their preemption of available fiscal resources for objects-of-expenditure other than those ultimately contri¬ buting to operation and maintenance funding for development projects. This chapter will explore such revenue and ex¬ penditure effects stemming from government policies and practices as they relate, ultimately, to the ability of LDCs to meet project-related recurrent costs. A brief an¬ alysis of the selective response to inflation by budget authorities in LDCs will also be made. One of the chief constraints on current revenue for LDCs is their inability to effect an expansion in their tax bases, given their comparatively low, and in some cases. While these sluggish 63 1 negative economic growth rates. rates are important in understanding the current revenue picture (in this case tax revenue) of most developing nationsf it is those specific governmental policies which prevent tax base growth, and therefore the augmentation of tax revenue, which must be considered as paramount when analyzing a reduction in capacity to meet project recurrent costs. In this regard, Clive Gray and Andre Martens listed the preemption of domestic credit, the undervaluation of the cost of capital, and the undervaluation of the cost of foreign exchange by the public sector, as the chief govern- 2 ment-induced determinants of economic stagnation. Specif¬ ically, preemption of domestic credit by the government limits private sector investment, which has obvious ramifi¬ cations for tax base growth. The undervaluation of the cost of capital and foreign exchange also impede tax base growth by limiting investment in the economy and by limi¬ ting exports, both of which can be taxed. Further, foreign exchange undervaluation also encourages imports, limiting foreign exchange earnings and leading to overinvestment in some sectors of the economy and underinvestment in others. It should be noted, though, that such constraints to the tax base, and hence to domestic revenue, are not suf¬ ficient in themselves to account for unmet recurrent costs. Government revenues are fungible which means that governments 64 are free to allocate receipts to whatever objects-of-expen- diture they so choose, within an overall ceiling. While this overall ceiling may be too low, as it almost invari¬ ably is, to enable the government to operate and maintain every unit of the public sector's installed capacity, a single project cannot be said to suffer a priori from insufficient financing of its operation and maintenance costs. Rather, an apparent lack of such financing must mean that the government has decided to allocate the cor¬ responding amount of uncommitted receipts to different objects-of-expenditure which it considers as having a higher priority or a greater degree of political utility than project operation and maintenance. Gray and Martens, in commenting on a study conducted under the auspices of the OECD's Club du Sahel, referred to this preemption of potential revenue for project operation and maintenance by such key 'politically-imperative' objects-of-expenditure as the civil service payroll? social insurance obligations including public sector pensions; service payments on the public debt; and, to an extent, government subsidies associated primarily with state-owned 3 enterprises. Thus, besides the constraints on the revenue side of the budget equation imposed by certain predominant govern¬ mental policies, potential revenue is further squeezed by specific budgetary practices which favor the use of revenue 65 for expenditures other than project-related recurrent costs. External to the budget paradigm related above, but nevertheless germane to the shrinkage of resources poten¬ tially used to support project recurrent costs, is the role of inflation. As related in Chapter Three, Aghevli and Khan advanced the argument that government expenditures in 4 LDCs adjust to inflation faster than revenues. Given the pressure to support committed expenditures, this tendency can only intensify the budgetary squeeze on uncommitted revenue. Thus, in inflationary times, the question becomes one of whether or not LDC governments will further squeeze potential uncommitted revenue by utilizing an active (in¬ flation minimizing) fiscal policy for committed expendi¬ tures at the expense of uncommitted expenditures which may not benefit from such a policy. Revenue Constraints Government Policies and the Tax Base Because LDCs have limited capacities to enhance tax revenues by utilizing existing tax structures, the major avenue for tax revenue augmentation is to increase the size of the tax base. However, the practicality of this option is limited as most of these nations suffer from low or negative economic growth rates—rates which, since they reflect economic vitality, directly influence the size of 66 the overall tax base. While there may be a great number of varied causes for this economic sluggishness, governments can be involved in at least three ways. These include: 1. the preemption of domestic credit? 2. the undervaluation of the cost of capital? and 3. the undervaluation of the cost of foreign exchange. The Preemption of Domestic Credit. In many countries, up to eighty percent of total available bank credit gets 5 appropriated by the government. This practice stems from the need for LDC authorities to subsidize enterprises which it owns or controls and which usually incur operating deficits. In fact, public sector enterprises have not traditionally performed well budgetarily in a host of na¬ tions. As Gillis et al., relate: ...large state enterprise sectors persistently ran deficits that had to be financed from gener¬ al governnment revenues. At times even single enterprises have accummulated losses and exter¬ nal debt obligations large enough to cripple development efforts across all fields of govern¬ mental activity.o When available credit is siphoned off to such a pro¬ nounced degree, resources left for private sector invest¬ ment are significantly curtailed. With impaired investment potential, little new domestic capital expansion (which can be taxed) can occur. The Undervaluation of Capital. The second major gov¬ ernment-induced limitation to economic growth and, hence. 67 tax base expansion, is the undervaluation of the cost of capital within the developing country. This occurs as governments impose artificial ceilings on nominal interest rates in the belief that these low rates will foster an enhanced investment climate. In developed and developing countries alike, policy makers have often viewed low nomi¬ nal rates of interest as essential for expansion of invest¬ ment. Indeed, so long as the supply of investible funds is unlimited and inflation remains low, such interest rates will foster all types of investment activities. If, how¬ ever, inflation rises above the nominal interest rate (or, conversely, if the nominal rate lags behind inflation) then the real interest rate becomes negative. Since borrowers as well as depositors respond ultimately to real, not nominal rates, the materialization of negative real inter¬ est rates will defeat any efforts at increasing capital investment and domestic saving, respectively. Enhancement of the investment climate is not the only motivation for governments where interest rate ceilings are concerned. The imposition of interest rate ceilings have also been attributed to a general distrust of the market mechanism by many LDC governments, and a desire to exercise greater control over the entire financial system by limi¬ ting access to political favorites. As interest rate cei¬ lings facilitate these ends, they also contribute to the contraction of markets, posing even greater barriers to 68 investment by low to medium income investors, further con¬ tributing to sluggish economic growth. Aversion toward market control over the price of capital will lead to its nonprice rationing, ultimately channeling it into less productive investment opportuni¬ ties. Subsequently, the investment programs of developing countries become skewed in favor of less productive in- 7 vestments and potential economic growth is curtailed. Thus, inflation-induced or government-imposed finan¬ cial repression through the undervaluation of the cost of capital tends to lower the marginal efficiency of invest¬ ment (skewing the content of development programs) and, ultimately, retards growth in national income and, hence, growth in the tax base as well. The Undervaluation of Foreign Exchange. The third and final government-induced limitation to economic growth and, hence, tax base expansion, is the undervaluation of the cost of foreign exchange in the developing country. This occurs as governments in these nations accede to the de¬ mands of their politically powerful constituents who desire a wide spectrum of imports, readily available and at little 8 cost. To answer this need, governments will purposely overvalue their official exchange rates, in effect overval¬ uing their national currencies in relation to other nations and undervaluing the cost of foreign exchange expressed in 9 their currencies. 69 Evidence of this trend can be obtained by observing 10 cross-country comparisons of the net foreign assets of developing nations worldwide. If a pattern of continuously declining net foreign assets can be identified, then it can be said that a great degree of disequilibrium exists in these countries1 exchange markets and that the cost of 11 foreign exchange is being undervalued. Examination of International Monetary Fund statistics would indicate that the trend of declining net foreign assets is, indeed, widespread. Of the thirty non-Communist nations classified 12 by the World Bank as low-income econmies, fourteen experienced consecutive declining net foreign assets of six years or more since 1974, ten nations with seven or more 13 years. Like the preemption of capital and the undervaluation of its cost, the undervaluation of the cost of foreign exchange also presents several ramifications for the size of the tax base. Specifically, overvaluation of the ex¬ change rate gives little encouragement to exporting firms, increases imports and, in the absence of a government-wide import substitution program, discourages investment acti¬ vity in industries which could produce the imported items 14 domestically. In short, then, with such a large emphasis on imports brought about by the undervaluation of the cost 70 of foreign exchange, there is little incentive for inves¬ tors to expand into trade-related industries. This stric¬ ture to investment, in addition to the fact that less tax revenue can be gleaned from exports, has obvious repercus¬ sions for the tax base. It should be clear from the foregoing presentation, that expansion of the tax bases in developing countries has been, a priori, significantly curtailed by specific government policies which appeal to or satisfy parochial interests. In the absence of changes, the expansion of tax bases in countries subscribing to such policies is far from assured. Expenditure Inducers Government Policies and Committed Expenditures All foreign assistance aside, the constraints to reve¬ nue explained in the previous section, together with a number of other economically-related causes, act in concert to place an overall ceiling on the resources available to any given developing country to meet its public sector expenditures. To assume, however, that unmet project- related recurrent costs may be due to a lack of finance 15 could be premature. Because of the fungibility of public sector resources, governments retain a maximum de¬ gree of choice as to how to allocate these funds to achieve specific political and economic objectives. 71 In light of this overall resource limitation, govern¬ ments will allocate revenues, "...in principle, out of a concern to maximize national social welfare, up to the point where the last [currency unit] is equalized in each 16 expenditure category...." In this regard, "...a govern¬ ment may consider that the resources available have been expended in an optimal manner even if [project] operation and maintenance costs have not received adequate finan- 17 cing." Since any budget process is highly political, such bias against project operation and maintenance is certainly a very real possibility. This is made evident in the Club du Sahel's report: ...politicians, technocrats, and managers of public and private enterprises may have divergent views on the priority to be accorded to operation and maintenance expenditures relative to expendi¬ tures on public investment and general admini¬ stration. .. [and] about the desirability of pre¬ serving public investments whose operation and maintenance costs have become very onerous. This problem of how to allocate the scarce fiscal resources is compounded by a number of public sector expenditures 19 which are not easily reducible. When these expenditures are taken into consideration, uncommitted receipts which could potentially go toward financing project operation and 20 maintenance are essentially lost. These expenditure categories are considered below: Civil Service Wages and Salaries. Over-ambition by developing country governments in expanding their range of 72 commitments has been listed as one of the major causes of 21 unmet project-related recurrent costs. Foremost among these commitments is the burgeoning size of the public service and its subsequent increasing role in the total recurrent budgets of developing nations. This outcome can be attributed, at least in part, to the common practice of guaranteeing employment to virtually all graduates of sec- 22 ondary and higher education. According to the Club du Sahel, this practice became necessary at independence to accommodate the rapid efflux of colonial personnel? how¬ ever, it has continued as constituents have come to insist on a public service which supports "employment in the public service for new generations of educated young peo- 23 pie." In this light, a World Bank study of seven coun¬ tries (including three less developed) found that the pub¬ lic sector employs between forty and seventy-four percent 24 of those recorded in paid employment. The role of public sector employment commitments in total recurrent expendi¬ tures of developing countries is also predominant. Statis¬ tics extracted from the International Monetary Fund's Gov¬ ernment Finance Statistics Yearbook list the ratio of wages and salaries for public sector employees to recurrent expen¬ diture for nineteen developing nations as ranging from a low of 21.5 percent to a high of 57.9 percent with an average level of 40.2 percent (see Table 1). 73 Table 1. Public Sector Wages and Salaries as a Percentage of Total Current Expenditure, Nineteen Nation World Sample, 1975, 1983 Country 1976 1980 Bolivia 58.2 52.9 Botswana 40.9 37.7 Cameroon 48.8 41.2 Dominican Republic 61.8 57.9 Egypt 21.7 23.2 Guatemala 51.2 53.9 Indonesia 27.3 25.6 Kenya 44.4 37.4 Lesotho 37.9 42.6 Liberia 38.5 50.5 Malawi 21.6 27.2 Mauritius 34.6 37.4 Philippines 30.6 33.2 Senegal 48.2 47.5 Sri Lanka 29.6 21.5 Swaziland 53.3 55.3 Morocco 42.1 48.8 Thailand 33.0 38.9 Zaire 45.0 31.0 Average 40.5 40.2 Source: International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985), specific country tables. Social Insurance Obligations. To a much smaller de¬ gree, public sector social insurance obligations which are not easily reducible, also make demands upon recurrent budget resources. This expenditure category includes so¬ cial security and other forms of assistance related to health benefits and retirement both for public sector em¬ ployees and the population at large. Statistics describing the actual extent of such expenditures in the budgets of 74 developing countries are of limited availability. However, data derived from the International Monetary Fund list the average share of social security and assistance in total current expenditure for eleven developing nations as 4.7 percent for 1975, increasing 23.4 percent to 5.8 percent for 1983 (see Table 2). Table 2. Public Sector Social Insurance as a Percentage of Total Current Expenditure, Eleven Nation World Sample, 1975, 1983 Country 1975 1983 Bolivia 2.0 18.2 Dominican Republic 12.5 11.4 Egypt 10.1 14.3 Kenya 0.3 0.2 Lesotho 3.5 1.5 Liberia 0.0 1.4 Philippines 1.8 3.5 Sri Lanka 3.5 1.2 Morocco 3.8 3.6 Thailand 5.0 3.4 Zaire 9.1 4.7 Average 4.7 5.8 Source: International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985), specific country tables. 75 Interest Payments on the Public Debt. Another budge¬ tary area which is rigid on the down side is interest payments on the public debt. This obligation arises large¬ ly from past government budget deficits supported by both internal and external borrowing. In this regard, the aver¬ age annual percentage change in outstanding long-term debt for seventy-six developing countries increased from 18.4 percent for the period 1970-73 to 21.0 percent for the 25 period 1973-80. This same group of nations was confron¬ ted with average interest rates of 2.8 percent for 1970, increasing to 4.9 percent by 1984. Average total interest payments on long-term debt for forty-five nations in the above group was $17.3 million for 1970, increasing to 26 $204.5 million by 1984, a rise of 1,082 percent. The irreversibility of this trend is described by Gillis et al. Thus: Because interest on government debt reflects past decisions, both on deficit finance and external borrowing, this item of expenditure is difficult to reduce.... On the contrary, prospects are that through the 1980's a larger number of LDCs will be more heavily involved in external bor¬ rowing than in the two previous decades. In¬ terest costs as a proportion of government expen¬ ditures are likely to rise in many nations.2? The average proportion of interest payments in current expenditure for the same group of developing nations used in examining wages and salaries and social insurance pay¬ ments has, in fact, risen 120 percent from 5.5 percent 76 of current expenditure in 1975 to 13.3 percent of current expenditure by 1983 (see Table 3). Table 3. Public Sector Debt Service as a Percentage of Total Current Expenditure, Nineteen Nation World Sample 1975, 1983 Country 1975 1983 Bolivia 3.3 6.3 Botswana 6.5 4.1 Cameroon 1.6 4.2 Dominican Republic 2.0 10.7 Egypt 3.4 9.3 Guatemala 7.5 8.8 Indonesia 2.6 14.2 Kenya 7.3 17.2 Lesotho 1.2 13.8 Liberia 5.5 16.4 Malawi 10.1 19.0 Mauritius 5.1 24.0 Philippines 6.5 14.5 Senegal 2.6 8.9 Sri Lanka 14.4 28.0 Swaziland 4.3 6.3 Morocco 4.3 14.7 Thailand 10.6 15.8 Zaire 6.4 15.8 Average 5.5 13.3 Source: International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985), specific country tables. Subsidization of State-Owned Enterprises. Finally, with regard to objects-of-expenditure rigid on the down side, the Club du Sahel lists the subsidization of state- owned enterprises as a significant constraint to revenues otherwise available as potential supports for project re¬ current costs. 77 State-owned enterprises have come to dominate large segments of many LDC economies. The extent of this in¬ volvement is described by Gillis et al. Thus: In countries as diverse as Bangladesh, Bolivia, and Mexico, the share of state-owned enterprises in annual gross investment outside of agricul¬ ture has been upward of 75 percent....Public enterprises have in recent years contributed about 40 percent of GDP in Bolivia, 25 percent in Pakistan and Ghana, 12 percent in India, and 11 percent in Korea.28 In addition to the preemption of up to eighty percent of available bank credit, the efficacy of public enter¬ prises in developing countries is also questionable speci¬ fically because of concerns about inefficient management and inappropriate investments made by controlling offi¬ cials. Further, "substantial evidence exists that public corporations are often not profitable and represent signif¬ icant drains on the government budget in the form of subsi- 29 dies." In this regard, subsidized services such as food, rural electrification, refined petroleum products, transpor¬ tation, health, and education are not easy to reduce as "the public learns to depend on the government for the services that it provides and to consider subsidies their 30 right." Efforts to cut back on expenditures for food subsidies have led, in fact, to rioting in Sri Lanka, Egypt, and Turkey and reduction of subsidies on gasoline have also led to severe social disturbances in Indonesia, 31 Columbia, and Thailand. 78 According to the International Monetary Fund, total involvement of subsidies in current expenditure of our group of nineteen developing nations was 19.7 percent for 32 1975 decreasing slightly to 19.1 percent by 1983 (see Table 4). Table 4. Public Sector Subsidies as a Percentage of Total Current Expenditure, Nineteen Nation World Sample 1975, 1983 Country 1975 1983 Bolivia 17.1 21.9 Botswana 18.0 34.3 Cameroon 11.3 13.5 Dominican Republic 18.0 16.0 Egypt 53.9 41.9 Guatemala 17.9 17.7 Indonesia 13.7 22.8 Kenya 24.4 14.9 Lesotho 15.8 13.0 Liberia 6.3 9.5 Malawi 14.8 8.9 Mauritius 42.8 30.5 Philippines 10.7 13.5 Senegal 16.2 22.4 Sri Lanka 44.3 30.9 Swaziland 1.7 12.2 Morocco 32.1 18.8 Thailand 21.5 7.4 Zaire 12.7 12.9 Average 19.7 19.1 Source: International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985), specific country tables. Revenue Constraints and Expenditure Inducers: The Uncommitted Revenue Squeeze If an overall ceiling to revenue exists for most developing countries, and if commitments for wages and salaries, social insurance, debt service, and subsidies are to be considered as not easily reducible or reallocable, then the amount of total available uncommitted revenue for project-related recurrent costs must, at least, be restric¬ ted. Reduction of expenditure commitments would mean that a given government has opted to hold back on civil service recruitment, allow emoluments and subsidies to lag, and to ask creditors for some debt abatement. If these policies do not present a viable alternative to developing country governments, then resources for closing the gap between project operation and maintenance needs and available reve¬ nues to meet these needs must be generated through in¬ creased taxation, which seems unlikely based upon evidence presented earlier, or else reallocated from other public uses, essentially the purchase of other goods and services that complement the services of government staff and enable them to perform their assigned tasks. While 'other goods and services' includes all expenditures related to the operation and maintenance of development projects, it also encompasses administrative and operating expenses of the government and non-personnel-related expenditures for de¬ fense. With such demands and again in light of constraints 80 to committed expenditures and to increased tax revenue, it seems unlikely that augmentation of 'other goods and ser¬ vices' or reallocation from within this expenditure cate¬ gory present realistic alternatives in ensuring that pro¬ ject-related operation and maintenance costs are adequately met. These facts become apparent when an examination of our group of nineteen developing nations is made. Again, the IMF's Government Finance Statistic Yearbook indicates a 15.3 percent average decrease in the share of 'other goods and services' as a proportion of current expenditure be¬ tween 1975 and 1983 (see Table 5). When a composite statistical comparison is taken with regard to all of the areas of committed expenditure and uncommitted expenditure, the picture becomes clearer. Table 6 lists the specific expenditure categories discussed earlier (with the exception of social insurance) and com¬ pares the changing relationship between committed and un¬ committed expenditures. In this regard, the total share of committed expenditure in current expenditure increased 10.3 percent from 1975 to 1983 while uncommitted expenditures' share decreased 15.3 percent for the same time period. When inflation is considered for the same period, it becomes apparent that the response of fiscal policy to this economic variable differs with respect to committed vs. uncommitted expenditure. Specifically, in real terms 81 Table 5. Other Goods and Services as a Percentage of Total Current Expenditure, Nineteen Nation World Sample 1975, 1983 Country 1975 1983 Bolivia 32.0 18.9 Botswana 31.2 23.9 Cameroon 38.4 40.8 Dominican Republic 19.3 16.0 Egypt 21.0 25.6 Guatemala 22.2 19.5 Indonesia 40.1 19.8 Kenya 23.6 30.1 Lesotho 30.6 30.6 Liberia 38.5 23.6 Malawi 53.4 44.8 Mauritius 17.5 8.1 Philippines 28.8 23.0 Senegal 33.0 21.1 Sri Lanka 11.7 19.6 Swaziland 40.6 26.1 Morocco 21.5 14.7 Thailand 32.6 37.1 Zaire 35.9 40.3 Average 30.1 25.5 Source: International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985, specific country tables. (after inflation), committed expenditure increased one percent between 1975 and 1983 while uncommitted expendi- 33 tures decreased nearly thirty percent. 82 Table 6. Summary of the Shares of Committed and Uncommit¬ ted Expenditure in Total Current Expenditure, Nineteen Nation World Sample, (percent) 1975, 1983 1975 1983 Percent Change Committed Expenditure3, Wages and Salaries 40.6 40.2 - <1.0 Debt Service (Interest Payments) 5.5 13.3 + 141.8 Subsidies 19.7 19.1 - 3.0 Total 65.8 72.6 + 10.3 Uncommitted Expenditure: Other Goods and Services Total 30.1 25.5 - 15.3 Source: International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985), specific country tables. aDoes not include social insurance due to insufficient data. 83 Summary In light of the general overall constraint to public sector revenue among many developing countries worldwide, continued government preemption of domestic credit, manipu¬ lation of the prevailing interest rate, and the overvalua¬ tion of exchange rates only reduce even further the overall amount of resources, in the form of tax revenue, which could potentially fund project-related recurrent costs. When these economic and fiscally-imposed limits to revenue encounter growing political imperatives toward ex¬ penditures which are particularly rigid on the down side, few resources remain for financing project-related recur¬ rent costs. This 'squeeze' of available fiscal resources for project operation and maintenance is refelcted in the growing share of committed expenditures and the diminishing share of uncommitted expenditures in the total current budgets of many LDCs worldwide. Finally, the constraint to uncommitted revenue is further intensified by an apparently selective approach toward inflation budgeting made by LDC governments with respect to committed vs. uncommitted bud¬ get expenditures. These limitations, together with the effects of ODA on project-related recurrent costs among the nations of the West African Sahel, will be the subject of the next chap¬ ter. 84 ENDNOTES 1. World Bank, World Development Report, 1982 (Washington D.C., 1982), Table 1, p. 180 and Table 2, p. 182. 2. Clive Gray and Andre Martens, "The Political Econ¬ omy of the 'Recurrent Cost Problem' in the West African Sahel," World Development 11 (1983), p. 112. 3. Ibid., p. 106, 112. 4. Bijan Aghevli and Mohsin Khan, "Government Defi¬ cits and the Inflationary Process in Developing Countries," International Monetary Fund Staff Papers 25 (September 1978), p. 391. 5. Club du Sahel/CILSS, Recurrent Costs of Develop- ment Programs of the Sahel: Analysis and Recommendations (Paris: OECD, 1980), p. 253. 6. Malcolm Gillis, Dwight Perkins, Michael Roemer, Donald Snodgrass, Economics of Development (New York, W.W. Norton and Company, 1983), p. 298. 7. Ibid., p. 354. 8. Ibid., p. 442. 9. Club du Sahel/CILSS, op. cit.. P- 254. 10. foreign Net foreign assets are defined assets of monetary authorities as and the sum of the the foreign assets of deposits money banks less the sum of monetary authorities' and deposit money banks' foreign liabilities. International Monetary Fund, International Financial Statistics Yearbook, 1985 (Washington, D.C., 1985), country tables, line 31n. 11. Disequilibrium here denotes a drawdown of foreign assets greater than that being replaced. Since imports must be paid for in foreign currency, the disequilibrium reflects greater imports than exports for a given nation. 12. The World Bank, World Development Report, 1986 (Washington, D.C., 1986), Annex. 13. International Monetary Fund, International Finan¬ cial Statistics Yearbooks 1985 (Washington, D.C., 1985). Various country tables. 85 • r H Club du Sahel/CILSS, op. cit., p. 257. 15. Ibid., p. 199. 16. Ibid., p. 223. 17. Ibid. 18. Ibid., pp. 223-224. 19. Aghevli and Khan, op. cit., p. 391 • 20. Peter Heller, "The Underfinancing i of Recurrent Development Costs," Finance and Development p. 39. 16 (1979), 21. A. Jennings, "The Recurrent Cost Problem in the Least Developed Countries", Journal of Development Studies (July 1983) , P- 513. 22. Club du Sahel/CILSS , op. cit., p. 261 • 23. Ibid., P- 261. ro • Jennings, op. cit., p. 513. ' 25. The World Bank, op. cit., Table A. 10, p. 159. 26. Ibid., Table 19, p. 216 and Table 17, p. 212. • r- CM Gillis et al., op. cit., p. 305. 28. Ibid., P- 567. 29. Thomas K. Morrison, "Structural Determinants of Government Budget Deficits in Developing Countries," World Development 10 (1982), p. 470. 30. Ibid. 31. Gillis, et al., op. cit., p. 306. 32. International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985), various country tables. 33. Ibid. Uncommitted expenditure i.e., 'Other Goods and Services' is one of the 'Current Expenditures by Eco¬ nomic Type' listed in Table C of the country tables in the IMF's Government Finance Statistics Yearbook, the others being 'Wages and Salaries,' 'Interest Payments' (Debt Ser¬ vice), and 'Subsidies and Other Current Transfers.' 86 CHAPTER 6 LIMITATIONS TOWARD FINANCING RECURRENT COSTS IN WEST AFRICA AND THE SAHEL The potential causes behind the inability of develop¬ ing countries to adequately meet their project-related recurrent costs can be better understood when the economic, political, and fiscal characteristics or policies discussed in the previous chapters are applied to specific regions or nations. In this regard, while the phenomenon of unmet project-related recurrent costs has manifested itself in much of the developing world, it is particularly character¬ istic of Western Africa and nations of the West African 1 Sahel. This chapter will discuss the recurrent cost problem in five of the eight nations which comprise the Sahel (Burkina Faso, Senegal, The Gambia, Mali, and Niger) and two non-Sahelian West African nations (Ghana and Togo) from the standpoint of the public sector savings-investment gap discussed earlier, which may affect national abilities to meet project-related operation and maintenance expendi¬ tures. Having demonstrated some of the key limitations to revenue and inducers to expenditures which, together, en¬ gender this gap, consideration will be given to the options of financing the gap through private sector saving and 87 money creation. The role of ODA in rectifying the differ¬ ence between public saving and other means of domestic finance on the one hand, and investment needs on the other, will then be considered as will the ramifications of this aid for project-related recurrent costs. Finally, the public sector-inspired squeeze to uncommitted revenues used 2 to support such costs will be considered. Why West Africa and The Sahel? Economically, many of the nations of West Africa and the Sahel are among the poorest in the world. Of the fifteen nations that comprise West Africa, ten fall within the World Bank's definition of low-income countries for 1982 (per capita GNP of $390 or less). Of these fifteen nations, six are Sahelian and four of these also fall 3 within the confines of this definition. Per capita GNP for the region as a whole averaged $413 for 1982, rising slightly to $436 by 1984. When the middle-income oil impor¬ ting and exporting nations (Mauritania, Senegal, The Ivory Coast, Liberia and Nigeria) are not considered, the average 4 drops to $294 for 1982 and $264 for 1984. Per capita GNP for the seven nations of the Sahel for 1982, excluding the middle-income oil importers, Mauritania and Senegal, aver¬ aged even less at $228. While complete data for non¬ middle-income Sahelian nations was not available for 1984, figures for Mali, Burkina Faso, and Niger indicate an 88 5 average per capita GNP of only $163. When all available Sahelian nations in the data set are included (for a total of seven in 1982 and five in 1984), per capita GNP averaged $300 and $264, respectively, well below the World Bank national transition point from low to middle-income sta- 6 tus. Finally, the average 1982 per capita GNPs for the five-nation Sahel and the seven-nation composite samples were $390 and $321, respectively. Again, these averages are much lower when the oil-importing nation of Senegal is removed—GNP dropping to $265 for the Sahel sample and $293 for the select West African sample (seven nations minus Senegal). This series of statistics give some scope to the developmental status of West Africa and the Sahel when it is compared with the average per capita GNP for 1982 for all 39 Sub-Saharan African nations—among the least devel¬ oped in the world—of $491, for all low and lower middle- income developing countries worldwide of $560, and for the industrial market economies, with an average per capita GNP of $11,070 (see Table 7). Further statistical inquiry indicates that the average annual growth rate of per capita GNP for West Africa as a whole for the period 1960-82 was 0.9 percent, shrinking to 0.6 percent when the middle-income oil importing and expor¬ ting nations are excluded. The average per capita 1960-82 GNP growth rate for the five Sahelian and seven composite T a bl e 7. B a s ic E c o n o m ic a n d D e m o g r a p h ic I n di ca to rs fo r t he S e v e n - C o m p o s it e S a m p l e a n d O t h e r S e l e c t E c o n o m ic o r G e o g r a p h ic 89 w a a o u <0 o z U 41 O ^ ktf 'O u « mm 3 • W 3 — U « 4J O 3 0 U “v m a. *o U 4i O V IS «f C U O w 0 X u o « c w y ^ y e 3 « 9% < <• o ^ — m INI — O ♦ ♦ 4- -O >0 n 4J 44 n 4 y 44 •H rH (0 J r- 9 P a 0) 43 4-» f0 O o •H TJ CO G >1 03 43 CO -p •H 04 03 CO p O O o •H a P 03 E u C O M u o •H TJ G 43 O O 04 ■P •H 03 03 44 P i-H 03 CP 0) 2 O 1 03 i G U G 03 O > 03 •H 03 G -P in 03 03 U 03 O G •H T3 EH E W O P c TJ O 0 G M4 o 03 CO w £ 03 44 4-> U O i-H •H 03 03 'V i-H 03 c 03 X l-H W ^ «M 9 •O O O O CO M U bo ►N • « V U U be -• GO Q) rH ^3 CO EH e u M ^4 2 e « • o ^ *J • ^4 -f I- 3 U T Q O ^ < i i Ai 9 « < “O a be • ^4 U O • ^4 »» a. w «<= 5. r> ^ m — vfi oo r^ n r4 — ^4 ^ -4 U 2 2 I m v ^ 3 w u w CO O S |S ^ 2 .3 -OB 3 ' O —I _< • . . »i o «» — a u mt^^seueo O O-fieor-eoao m n s+4 n JQ u o it o — o u Q •> C 22- C w 5C w J: a o 41 U to a. as c as V —• to w 103 the early 1970s, have been subjected to diminishing market shares and declining prices. Specifically, between the periods 1961-63 and 1980-82, commodities comprising the greatest share of the economic bases of these nations (maize, cocoa, groundnuts, groundnut oil, and cotton) ex¬ perienced an average decline in market share of forty-eight percent. Similarly, within roughly two decades (1961-70 and 1970-82), these commodities experienced an average annual decline in the growth rate of their prices of 1,387 percent (see Table 10). This decline in the price of major cash exports is very much reflected in the terms of trade which, between 1979 and 1982, depreciated an average of thirty percent among twelve nations in West Africa and twenty-two percent among the seven nations of the composite sample (see Table 10). The loss of income (as a percent of GDP) between 1971 and 1981 associated with this deteriora- 28 tion averaged 1.1 percent annually for West Africa (ex¬ cluding Nigeria) and 0.3 percent for the seven nations of 29 the composite sample. Subsequently, not only do gov¬ ernments lose potential tax revenue from the declining mar¬ ket shares of their exports (less export-related taxation), they also lose revenue from the loss to GDP which is later reflected in a constricted tax base. Finally, potential export earnings and the subsequent ability to finance many of the imports associated with public sector current T a b l e 1 0. S e l e c t P r i n c i p a l E x p o r t C o m m o d i t i e s o f t h e S e v e n - N a t i o n S a m p l e : C h a n g e in M a r k e t S h a r e a n d P r i c e C h a n g e i n A v e r a g e a n n u a l w o r l d m a r k e t s h a r e g r o w t h r a t e 104 CM CO I <4-1 O 1 •o CM o -C « o o n- 1 1 1 ■H C b b 1 U V 60 U > NO c <0 ON CM CO I * s 9\ OJ w —< o e •H 0) u u a u w 0.0 o I vO ON « PM 0 CO •H I h ^ O »H ^ CO < « ON 6 d a) QUO b Irf *J «a « £ o.m CO w vO to I 1 • J3 9 CO NO ON *8 1 8 *o I I I CM I ON *4- I m l ON f'. I CO I CO I 00 '«• I •N •H •H O b * f-H —4 co n 0) O J3 o .3 d at 60 . 60 X 60 X PM •H •b O rH O b 2: d H •* d H . d u <0 60 o 60 o d d o 60 . « * CO 01 . CO 9 bo U d d d d d d d 44 o o d •b 01 •H PM 4) ■H PM b u H « 43 to 43 to 43 9 a a d a d 03 4) » d d d . d d i-M a . O b o •H b u •H a d ■H o 41 44 41 44 O a 3 « 60 60 4) b 60 41 b 60 <0 to JS o 3 6 £ •H 3S 9 oa •H as « O H O d • • 4) CL CL a fb 4> •H ... O P ^■N -if CM CO -if CM •—4 AJ AJ 41 •• CO 9 9 60 4) ON 4) 4) d d d U f—4 »b d ■d T3 b b 43 43 d o d d « 4) 9 . d d 8 44 AJ 9 o 9 o N ■H $ O to •Jb H ad 43 o O b b • u u O O X a I • O CO 00 NO 105 and capital consumption are lost as well, as the prices of exports relative to imports drop. While the tax systems of the Sahelian sample nations appear to be sound, with the exception of the over reliance on trade-related taxes, there are some problems with the administration of these systems. Substantial tax evasion has been noted and can be attributed to the insufficient number of trained tax collectors. Further, extensive smug¬ gling which is facilitated by the large expanse of some of 30 the countries, has also been documented. The three primary limitations to tax base growth— government-sponsored preemption of credit, the undervalua¬ tion of the cost of capital, and the undervaluation of the cost of foreign exchange that were noted in the previous chapter, are particularly characteristic of many of the nations of the composite sample. Here, government's aver- 31 age net share of total available domestic credit in¬ creased 173 percent in absolute terms from -39 percent to 32 +29 percent between 1972 and 1983. This statistic is most likely much higher when the claims of government- supported official enterprises are included. For example, when public sector enterprises are added to government's share of total available credit in Gambia, the share in¬ creases from twenty-four percent to sixty-three percent for 33 1983. Without this support, many of these enterprises would succumb to huge operating deficits. This point is 106 reinforced when considering the CILSS report on recurrent costs in the Sahel. Citing one unnamed country, the report states that "the responsible ministry considers that, out of twenty-six enterprises which it controls, sixteen would [not have been able] to survive in 1980 without government 34 subsidization." Like the preemption of capital, the governments of the region have also opted to keep the interest rates on time and savings deposits below the rate of inflation, in ef¬ fect, yielding negative real rates of interest which dis¬ courage overall potential investment, thereby preempting tax base growth. In brief, as a result of these negative real rates of interest, the real size of the financial system shrinks leading to a reduction in the real supply of credit which, in turn, constricts investment in productive assets that could conceivably contribute to the tax base. This tendency toward the undervaluation of the cost of capital and the implied shrinkage of the financial system is borne out when the real interest rates of five of the seven composite nations are considered. In the period 1974-79, real interest rates on time or saving deposits for these nations ranged from a low of -10.4 percent for Niger to a high of -5.5 percent for The Gambia, with no single nation presenting a positive result, indicating a sub¬ stantial lag of nominal interest rates behind consumer prices (see Table 11). T a b l e 11 . N o m i n a l a n d R e a l I n t e r e s t R a t e s f o r t h e F i v e - N a t i o n S a h e l i a n S u b s e t o f t h e S e v e n - N a t i o n C o m p o s i t e S a m p l e , 1 9 7 4 - 7 9 107 00 CM to t-i 00 O O '-o O. 4-t — cfl to a oo C6 .a (0 fH 3 3 3 3 •H C C Li e •H a /-v 3 CM 3 Li w 3 bo 3 b0 c 0 3 3 3 M J= CO 3 3 C > O < 3 to > c0 00 >> Li 4-1 a 3 o o • 3 a LM O CO 4J CO o u 4-1 G 3 M Li 3 u 3 05 VTN CO ON o s 3 CO u O 3 CO C ON O - 3 a c O Li §2 C - w V) •H • Ll T3 3 C Pi 3 —'Pu 3 e * o C/0 «H CO AJ J 3 W T3 o c ^ 3 *H B 3 B Z O 3 O C/0 3 Z 3 -a *0 c -Q 3 3 iH a U -H ^ 0 CO 3 ►> • • rH Cfi 03 S5 Li 3 •• O i-H C/0 3 X 3 C/0 VA CO ON o • a c 0 1 •H x: 03 3 Ta bl e 12 , N e t F o r e ig n A s s e t s o f th e Se ve n- Na ti on Co mp os it e Sa mp le , 19 70 -8 3 108 o • U G m W ftu *4 U r* G m Q <9 * — * o 9 *4 U G 9 •*4 U 0u o ^ e ^ 0) bo (5 w d > (0 c 1 M is'* S > A* o w o (H -P -P o c c V o •p a a c 2 -P 0) M (0 O V 0) ^ > 2: gj o C Qi b0*H u O /-N o & OJ o d M W -P G Q) -H O 60 > •H T3 d 3 S'- 0) f—I 0) ^3 o g rt -H fl W rH 0) > 3 0) <0 P< H G G O O ON ON I o 00 rN -G- c^- i o vn £ ON Cv. CO CO • VA NO I CO • VA NO CA ON S VA C^ NO o W 1 C^- -G* • • • VA co NO >A -G* NO ▼H NO CA -G- • • • ^H VA NO NO 1 CO C^ ON « CO 00 ON VA CM CM ON »A CA $ CM CA CM 1 O CA P CM vA O CO C^- P CA P CM CA CM CM o Cs- O CN- CM o co ON CN. CM •G* M «a 8> O 60 c P O JS 25 H CJ CO CO o CM CA O •ri ■P > -H (x« • P C 0 o U 0 P« c p 0 U d a • p w p p G •G 23 0 U • tH Gi O d fA 60 3 P o C p 0 rH B P P P n a 0 > c G P P 1 s to hO d g5 •H > o d p 0} s: X M Oj o “d p c a d 0 C9 O O P P CUP 0 P 0 3 X P« 0 0 (9 JS O P G •s I' rH iH P 0 G O S rH d G •H C P -H S O s a G O P G (9 (9 HH P -•3 8 * 3 H »H G 3 rH P wd *o T a b l e 15 . K e y B u d g e t I n d i c a t o r s a n d t h e P u b l i c S e c t o r S a v i n g s - I n v e s t m e n t G a p f o r t h e S e v e n - N a t i o n C o m p o s i t e S a m p l e , 1 9 8 0 114 -P i c O M w <3)£> •HOMS ,-1 C ft ^3 O -H W rj 3 0) > 0) tJO ft W rj > W C ■P ft -P -P C O C C m C 0 9 g a 8- o & to 0) d bp W +> B 0) ■*-! O bp > 33s q H 0) .0 0 3 rt ft 9 WHO) > 2 p W ft M 3 o u rv rv vrv 00 o • • • • • • vr* vrv o VTv rv 00 O- Ov C^V ▼H C^- I v£> * Ov I o & o- Cv. CM 'A C^- $ £v- O cn ■B o w & 3 Ov Ov Ov 5 ON £ £ o £ M & 3* 00 Ov CM Cv- B- 00 CM Ov vO O vO vO vO vO O b0 o ft CO Ov CM I 00 -B- 00 CM CM Ov C^ vrv Ov CM a a o CO r> i J G ® o > > r* 01 W O Q w CO 0) o *-• u CL O E <0 M-i in o 0) 4J 4J C -H 0) (0 U O U CL O E cu o u 03 c (0 o 03 *H < 03 o z O I c cu > 0) w -p 0) c T3 C 0 03 £ -P CL P z o U *p g-i 4J o c 0) -P E C +j 0) CO U 0) p > CD C CU ft < o o u as e « ■ 4-) oo « ■ 8 SJ- •9r 00 « 0 ^ 4J ^ c ft* O U z O U u u u v 1M 00 9S <0 -Jt t-c QO -4 OH s- I- X CM —• 00 ov 5- O rt in IM vO m — CM _ 00 00 »■> O' •3 *1 e -i o o a « B « *3 VI eg b. • CO -O z a ■ 1 V* e <9 <« *1 V *-* k* o bC > £ JC V ft) ■x «0 o k* bo V C Db V) bo 3 ft) o 00 z H in H c o — v a. a. > H H eg o V U D T. « 0*1 £ u •** (B C k* t/3 •** *J ^ 3 icS *1 ec •o 3 Z1 H e eo o — *J BO u-. x: > ip HJ w I- “O U Oft) o OJ U-. ft) 6C ft) 3 U O a H /3 3 ^-ft)bC«f- > ^ ^ ft) ft) «ru P u- r: 3 440 u *0 ft) 121 fifty-three percent among five of the seven nations for 1982 and financed no less than fifty percent of the deficit in all but one nation, Ghana. These data are summarized in Table 19. Table 19. Total Loans from External Sources as a Percent of Total Deficit Financing for the Seven Nation Composite Sample, 1982 Mali Niger The Gambia Senegal Burkina Faso Ghana Togo Avg. 89.6 85.6 76.4 50.0 62.0 8.0 53.7 52.6 Source: International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985), specific country tables. While some of this aid is geared to support current budget expenditure, most is destined to augment scarce gov¬ ernment resources for investment, thereby making it poss¬ ible for authorities to increase their overall development programs above the levels normally attainable by utilizing scarce domestic fiscal resources alone. While these external aid flows have allowed most of the composite sample countries to realize their planned level of development goals, to an extent, the long run efficacy of their programs and projects pursuant to these goals has been compromised (as have the goals themselves) by declining productivity and deterioration in capital stock resulting from an inability by governments to gener¬ ate the corresponding fiscal resources needed to operate 122 and maintain, at design levels, a growing number of fixed capital assets. In short, external aid flows have exacer¬ bated the operation and maintenance (recurrent cost) prob¬ lem by permitting governments in the sample countries to increase their level of investment without encouraging the concomitant growth of recurrent revenues needed to support 51 the cost of operating and maintaining it. The Fiscally-Induced Uncommitted Revenue Squeeze The revenue squeeze, brought about by the combination of revenue constraints and expenditure imperatives dis¬ cussed earlier, makes the provision of adequate recurrent cost funding for projects difficult at best by reducing the level of uncommitted fiscal resources which, among other things, support such expenditures. This condition is in¬ tensified by the fact that most donors refuse to provide aid for the explicit financing of project-related recurrent costs for fear of open-ended commitments on their part and the accompanying compromise of the goal of recipient coun- 52 try budgetary self-reliance. Despite these restrictions, the fungibility of available fiscal resources still pro¬ vides for the opportunity to meet project-related recurrent costs. Thus, when a project's operating capacity falls below desired design levels or, in the case of a self- sufficient project, when its direct cost recovery falls below target, the government has the option of picking up 123 the shortfall (or the deficit) by releasing uncommitted fiscal resources, thereby maintaining the specified design level of operation and maintenance or, conversely, holding the corresponding amount of revenue for other uses, in ef¬ fect allowing design operation and maintenance levels to lapse. Even in light of fiscal constraints and reluctant donor support, studies have shown that no case exists whereby all of the uncommitted portion of a host govern¬ ment's budget would be exhausted meeting a given project's 53 recurrent deficit. Subsequently, an unwillingness to adequately provide for such a project's recurrent cost funding must mean that the authorities are expressing a degree of deliberate fiscal preference in using their scarce uncommitted revenue to support objects-of-expendi- ture considered by them to have a greater degree of utili¬ ty, fiscal or otherwise, than project-related operation and maintenance. This tendency is best understood by consid¬ ering the concept of opportunity cost. This concept is relative and concerns the amount or degree of one activity, e.g. non-project-related recurrent cost financing, which must be given up in order to engage in another activity (such as project-related recurrent cost financing). If an activity has a high opportunity cost, then more must be given up to engage in it than if it were to have a low opportunity cost. 124 The overall constraints to uncommitted revenue and the definitive demands made upon this revenue by public sector operations other than project operation and maintenance e.g.f general government administration, means that such revenue will possess a very high opportunity cost. Subse¬ quently, budget authorities find it very hard, indeed, to relinquish scarce uncommitted resources to support unmet recurrent costs. As Gray and Martens explain: ...any increment to uncommitted government rev¬ enue can activate idle public sector capacity, generating social benefits far in excess of the incremental expenditure. It is this phenomenon which makes for a high opportunity cost of uncom¬ mitted revenue, [particularly among the nations] in the Sahel; and it thus frequently happens that the only values [of project recurrent expendi¬ tures and uncommitted revenues potentially able to support project recurrent costs that are accep- ^ table to developing country governments] are zero. Accordingly, in the case of many development projects, ...no level of [uncommitted] expenditure is per¬ ceived by the host country budgetary authority to yield benefits commensurate with the opportun¬ ity cost of the resources...that must be allocated [from other goods and services] to cover any [unmet project-related recurrent] costs.^ Thus, the lack of support for project-related opera¬ tion and maintenance represents an overall fiscal response both to the shortage of uncommitted revenue and to the need to use such revenue for other expenditures deemed more im¬ portant by budgetary authorities. 125 Uncommitted Revenues as a Support Mechanism for Recurrent Costs The scope of uncommitted revenue (which translates to expenditures on 'other goods and services' in the Interna¬ tional Monetary Fund's Government Finance Statistics Year¬ book) in the total current budgets of the sample countries between the mid to late 1970s and the early 1980s ranged from a low average of 15.4 percent for Burkina Faso over a seven-year period (1977-83) to a high average of 48.3 per- 56 cent for The Gambia between 1973 and 1978. Average yearly growth rates of the uncommitted portion of the bud- getss also ranged from a low of -12.9 percent for Togo to a high of 7.5 percent for The Gambia, with four of the seven nations (Burkina Faso, Niger, Togo, and Ghana) experiencing negative growth rates. Average involvement of uncommitted expenditures in the current budget for the entire sample was 26.6 percent with an overall average yearly growth rate of -1.0 percent. Specific country breakdowns of the invol¬ vement of uncommitted revenue in the current budget and the growth rate for this involvement are delineated in Table 20. This table presents an obvious indication that, with the exception of The Gambia, the role of uncommitted expen¬ diture in the current budgets of the composite sample coun¬ tries is minor in relation to that of committed expen¬ diture. In more succinct terms, committed expenditures are receiving, on average, nearly three times the share of T a bl e 20 . U n c o m m it te d E x pe nd it ur e a s a P e r c e n t o f Cu rr en t E x pe nd it ur e, A v e r a ge A n n u a l Ch an ge o f th e P r o po rt io n o f U n c o m m it te d R e v e n u e in Cu rr en t E x pe nd it ur e fo r th e Se ve n- Na ti on Co mp os it e Sa mp le , V a r io us Y e a r s 126 0) CO 00 <0 • • M OO M0 < CO 00 • ON O rH CM co M0 CM rH 1 CO • a\ CM CM 00 O' rH m 1 oo • ON •H CM CO o vO rH oo • rH rH CM O' M0 CT\ 'O r^» • ON rH rH m CO 00 CM r>. • O' rH rH M0 O' r^. 1 • 0^ m rH rH oo O' v£) CO r>. • O' rH rH 0) u u a 0 ai 0) U M CO *H M c T3 3 to c u JC 3 s cs ca o) 0) co < CO 00 • O' CO rH rH o O' CM 00 i CO • O' 'O rH rH m sT rH o 1 oo • O' m rH rH r-» o 00 1 OO • O' vO rH rH 00 • O' O' CO r* • O' 40 rH rH CM CO oo 00 • r-' • O' m rH rH O' rH i— rH • • O' vO rH rH 0) LI M C 3 0) LI Li CO •H M e -a 3 a) c u J: (U CL 'Ll O X o rH 0) a o LI 3 (0 •3 C 3 C ca X > c PQ C3 ca OJ < ft 0) co o 00 ca • Ll CM o < CO rH oo • O' rH rH CM 00 00 CM '3- 0O • O' O' rH rH o m CM rH CO 1 oo • O' m rH CM >T in o sr 00 • O' r* rH rH o CM O' n 1 C" • O' •3- *H CM rH 00 rH f'* • O' sT rH CM C oo ca ca 3 < *H T a bl e 20 (c on ti nu ed ). 127 0) to cn m < 00 vO f'. • *H ON 'S’ o • O' m rH LO 2 6 .5 vO ON • o> •>3’ 00 • CN m NO 1 r'. • ON m »H m NO »—♦ *H CS • ON 'd' iH m NO en '3- • ON iH rH sr 0) 4J u a o o V U M 00 •H M c •O 3 CO C O •C a) a •H a) -H v o 3 S o.’O 00 M o I c 03 03 O C3 0) h Q. 0) a a. 03 4= C O K > a H 9 < o 'O’ CO • ON m rH CM rH CO ON CM 1 • ON NO rH CM co ON 00 ON l r» • ON 00 rH CM CM • m rH rH • i ON rH CO •4r rH NO NO rH • ON o rH co 03 4J 34 c 3 03 3 3-1 34 00 *H 34 c •a s 3 G O JS 03 a. M-I u x o rH 03 3 4-1 3 *3 C 03 C 03 03 34 C 44 44 O 3 3 C 4-3 34 44 3 •H 03 T4 3 Cl 0 G. *3 00 34 C 3 3 03 0 3 3 34 a 00 U PL 3 •H G w X > G z z a 3 < TH 3 00 rH ON 3 • 34 CM 3 CM rH > i < CO co 00 • ON m rH rH oo CM CO 1 ao • ON NO rH CM CO • rH rH oo | CO • ON rH rH CO NO NO ON CO rH r- • ON rH CM O 00 m 1 • ON rH CM OH rH r^. ON CM r-~ • 1 ON 'J' rH CO 3 44 34 S 3 3 3 44 34 00 •H 34 c •3 3 3 C Cl JS 3 o a. «44 x o rH 3 3 44 3 •3 C 3 C 3 3 34 G 44 44 cl 3 3 C 44 34 44 3 •H 3 *H 3 U 0 G> *3 00 34 0 C 3 3 o 0 3 3 34 CL 00 O CL 3 o G M X > 6 H Z 3 3 <3 *H T a bl e 20 (c on ti nu ed ). 128 0) 00 iH CM CJ • • M vO O C fH CM oo • ON m •H CM ON vO o 00 00 • ON CO . *H CM CM • fH ON vO • ON fH CM 1 2 . 6 oo C" i r>» • ON •o rH CM 1 6 . 8 r» i • ON ON rH CM m NO rH i • ON fH co CO vO m 00 rH • ON vO rH CM 0) U M a 3 0) 4) O M oo •H M C *a s 4 s u X O 9 4 0) «< •a c 3 • U* ir» >* oo o> f3 «H 4J (U m a o X u a) a u 3 O to 129 total recurrent resources than are uncommitted expendi- 57 tures. Further, in an examination of the expenditure patterns of four of the sample countries over five to seven-year periods, it appears that committed expenditures increased at a faster rate in absolute nominal terms (142 percent on average) than uncommitted expenditures (see 58 Table 21). Specifically, Burkina Faso's absolute average yearly nominal change advantage was 24.8 percent, Senegal's 42.8 percent, and Niger's 54.9 percent. The remaining na¬ tion was Togo which had an absolute average yearly nominal change advantage of 444.0 percent. Among these nations, consideration was also given to the growth rate advantage, in percentage points, of one type of expenditure over the other. This advantage was derived when yearly changes for total committed expenditures were compared with yearly changes in uncommitted expenditure, averages taken, and the differences between these averages examined. Here, Burkina Faso's growth rate advantage was 2.9 percentage points, Senegal's 5.9 percentage points, Niger's 7.9 percentage points, and Togo's, 18.2 percentage points. This compara¬ tively lower level of emphasis placed upon uncommitted expenditure by the sample nations, both in absolute terms and in terms of yearly growth rates, means that it is more scarce in relation to committed expenditure. Hence, the financing of unmet recurrent costs associated with any given project from uncommitted resources poses a higher T a bl e 21 . N o m in al Ch an ge in Co mm it te d E x pe nd it ur e v s . N o m in al Ch an ge in Un co mm it te d E x pe nd it ur e fo r th e Se ve n- Na ti on Co mp os it e Sa mp le (p er ce nt ), V a r io us Y e a r s 130 < Cl 00 O' • • • CM o 00 CM a> 1 < cn 00 *n 00 • • CM m vO OO rH O' CM 00 *» CM • • CO oo OO CM 1 O' rH 00 «n vr • • O cn •o 00 >o O' rH o 00 m • • O' rH m O' rH O' rH 'T • • oo o rH rH O' 41 « -a 00 00 41 C *3 M C Ml M CO 41 eg u z MI e z H e O Ml H o a «H *H i •H a • H o • cO w a *o rt O *3 00 c o c e c c V H O 41 H 3 41 c a a. a a. •V OCX OCX v> Z H 41 Z H 4) 41 00 cn m • CM *o 3 CM •H > < o oo CM cn • • O' CM m CM CH O' rH O' • • oo CM rH r«. -T O' rH oo r** cn rH **^1. • • r>. 00 O' O' rH r>* rH z • z NO rH r>. rH CM O' rH 3 4) *3 00 00 41 C -3 X C Ml *C 3 4) 3 Ml Z Ml C z H e O Ml "H O | H H ^ i * > > < < < CO CO WO oo • • CM NO o 00 1 ON PH CM oo r-' NO • • •H co CM 00 PH 1 ON co rH oo oo lO CO 00 ON o 2 • • • • • • CM ON PH o cn CO 00 CM * 2 o CM 00 1 in rH O' i ON ON PH rH PH PH ON 2 00 ON ro f'* IT* r** 00 • • • • • • o> m 00 ON CO m rH o ON ON PH rH rH oo m ON 00 r-'. UO NO 2 ON • • • • • • 00 ON oo m CM on rH i^» •H PH rH t". CM 2 ON ON ON PH rH rH oo Mt PH o ON PH r>. IT> 2 • • **>N* • • • • sr vO r. r* cn ON CM CM rH CM CM ON 1 ON ON PH rH PH 0) eg oi (3 U C3 0) CO u JO u c J= -H c 2 u C 2 -H c C3 2 u C 2 fl C O *j •H ■H O *j ■H o a •H •H O u U B ^H •H O | ■H g 2 •H g rH Q • fH O • PH Q • r-4 O • a PH Q • <-4 O • 3 < rH o > CM 00 cn rH * • •X n o 00 cn >o o O' •O rH Vi cl 01 *H X CO CM o • • w o O' o o 00 'S- vO •H O' 4J w •H o O 00 vO *3- 4J • • cn O' «cr CM 0) O' o c m c O' •H OO vO (b • • 00 'O fH u r^. >T CM c O' • • o cn o 00 O' O' •» »H T3 C r>. m rH • • X vO n 00 u cn fH c O' 4J *H H 3 c O 04 0) E M O D U +J •H TJ C 4J 0) 4J cn •H E Pi G a rcj -P JC -H U G rH QJ (0 Oi 0) X « w CO Pi (0 a) >4 co 2 O •H Pi < m CO m 00 '"'‘Nl, • • CM ON QO 1 CM ON 1 rH CM CO CO • • rH 00 vO CO iH O' *H »H 00 o rH • • o n o 00 1 ON »H o 00 p^ NO '■■’•Si, • • ON VO o 1 1 ON •H ON •H o • 00 n ON rH ON *H CO o cn • • rH p*. •H rH ON rH •o 0) o •o n 4J (0 0) 01 at 4-i CO 00 U c 00 *H C p4 C 4-1 *H C 0 *H « "H a S cs JC Q . JZ o • C u B -a o a *a «H o c a c rH CJ Oi rH 3 a X 04 C K a es *H « PS -H 01 04 00 CO o •H u • • 01 ON n > < cn 00 CM rH '■***N* • • CM NO NO 00 1 ON rH CNI CO rH m • • rH rH m CO rH CM ON • 1 rH rH oo m « • o CM QO 00 1 CO ON *H o CO >o CM **■*%* • • ON CM p^ 1 ON rH ON P^ m NO • • 00 o P^ ON rH •o 04 •O M *J M 04 0) 04 4J 00 4-1 c 00 < O ao ON o • • ON rH m rH rH ON •H ON -T NO • • 00 m m cn i ON rH QO P^ r^> ON • • r^. rH o i 1 ON »H PM P^ CM • • sO r— rH r- 1 i ON rH •a 04 *0 K *J K 04 04 0) 4-1 00 4J C 00 1-1 c C 4J TH C B TH cd «H a B JZ g . JZ o • U B *3 0 0*0 IH O C c c 04 rH CJ 04 rH 3 01 00 id a. a o. •H 04 c X oi c x z PS -n 01 PS -n 04 T a bl e 22 (c on ti nu ed ). 136 1 > i > < C < CO 00 00 • • CM CM 00 1 O' rH CM 0O o r>. • • rH sO oo rH 1 O' rH cn rH oo QO •H CM 00 'T rH O' • • • • • CM o rH o rH rH O' CO CM m oo • rH i o\ l O' O' »H rH rH CN| o 00 O' O' 00 oo CM CM CM • • • • • • n oo O' m CO cO CM o 00 l CM CM m r^. i o ON 1 O' 1 O' •H rH rH O' vO 00 00 O oo rx CM • • • • -x. • • ON cn O oo 'S- m M3 *H CO r*. CM CM r- rH O' O' O' i rH rH rH O' 00 m CO O O' rH o • • • • • 00 CM rH OO O CM vO r^. rH rH rH cn 1 cn O' O' i 1 O' rH rH rH 00 rx. *o vO . to rH rH . vO 'J- cO r-» CM cn O cn (~« CM rx. 1 r* rH O' 1 O' O' fH rH rH •a TJ •a 01 0) 4) -o w 4-» M •o M LJ M •a *M Li 0) 0) 0) U (U 0) 0) Lt 4) 4) 4) *j oo u a C0-H d to Li B to *H B <4 00 U B to -H c C « «H c g -H B LJ CN 00 O 00 «k • • •H rH o CO rH CM o a\ U CO 0) fH p* 00 cn m • • (0 o NO u 00 NO m •H ON 1 i VJ •H 0) •H VI o CO 00 m 4J • • C/1 ON m m CM '4- 0) ON i u ■H c CO c ON ■H r>. VO 00 tM • • 00 ON CM VI 1 cn c ON 1 4) B c Vi 00 4) m m > • • O fH o o fH CM ON * iH •a a 9 r^- PM ON cn • • >N vO CM 00 U 00 ON cO ON 1 1 VI iH 4) e o NO X r^- o ON • m oo m CO on tH C3 ON i i O •H <0 •a 6 4) Vi T3 W *J M 4) 0> 0) 4) 4J V) 00 Vi C oo-H a a C VJ -H C B *H M tO «H CO g M a • 43 O • o B TJ y o no • • ca o c c c 4) c •—10 4) rM9 4) O « C0 O. CO O. Vi x: 4) C X cu e x 9 o Ptf *H 4) Ptf -H 4) O CO 138 and salaries, pensions, and subsidies and the consumer price index. He writes: [Wages and salaries] and pensions are often re¬ lated to an index and the volume of subsidies to government-owned enterprises and to other enter¬ prises will rise depending on the extent of price restraint sought by government in essential goods and the rate of inflation in other traditionally subsidized sectors.62 Premchand goes on to state that, in deference to these committed expenditure categories, "the major sector of government expenditures that is adversely affected [will 63 be] the general category of other goods and equipment." Further, losses to uncommitted expenditures as a result of inflation will be higher in countries where the public sector's imports are large, as is the case for a number of 64 the composite sample countries. Summary Clearly, among several of the nations of the composite sample, uncommitted budgetary resources, both in absolute and real terms, and in terms of their growth rates, are subject to a number of constraints, particularly in the four-nation subset of the seven-nation composite sample just considered. These constraints are initially engen¬ dered by the overall economic and fiscal situations which confront the nations of the composite sample. More import¬ antly, they are forced by the deliberate fiscal policies and practices which these nations pursue as a consequence 139 of their revenue constraints. Faced with such economic or fiscal liabilities and forced to maintain strong commit¬ ments toward personnel, debt service, subsidies, and gen¬ eral operating expenses, the goverments of the composite sample countries are faced with a situation whereby little revenue can be earmarked for the support of project opera¬ tion and maintenance. In light of these considerations and in view of the economic and fiscal constraints presented earlier, project- related recurrent costs may go unmet as governments react to externally-imposed situations or create policies which, effectively, restrict the choices that can be made in the budget process. 140 ENDNOTES 1. Clive Gray and Andre Martens, "The Political Econo my of the 'Recurrent Cost Problem' in the West African Sahel," World Development 11 (1983), p. 101. 2. It should be noted that no case studies reflecting the potential effect on recurrent costs exerted by the interaction between the donor and the recipient covered in Chapter Four could be obtained for West Africa and the Sahel. However, in light of the mechanics of the foreign aid process described in that chapter, it will be assumed, a priori, that similiar phenomena are occurring among the nations under consideration. 3. The World Bank, World Development Report, 1982 (Washington, D.C., 1982), Table 1, p. 57. 4. Comparisons between 1982 and 1984 do not translate exactly since no data were available for The Gambia and Guinea-Bissau in the 1984 sampling. 5. The World Bank, op. cit.. Table 1, p. 180. See also The World Bank, Toward Sustained Development in Sub- Saharan Africa: A Joint Program of Action (Washington, D.C. , 1984), Table 1, p. 57. 6. Ibid. 7. Noel Lateef, Crisis in the Sahel: A Case Study in Development Cooperation (Boulder, Co.: Westview Press, 1980) , p. 57. 8. This number exceeds 100 due to some countries' use of universal education which allows pupils above and below the official school age. 9. Peter Heller, "The Underfinancing of Recurrent Development Costs," Finance and Development 16 (March 1979), p. 38. 10. rom The World Bank Operations Evaluation Department, quoted in John Wilhelm and Jerry Feinstein, U.S. Foreign Assistance: Investment of Folly? (New York: traeger Scientific, 1984), Table 14.1, p. 200. 11. bid. 141 12. Club du Sahel/CILSS, Symposium on Recurrent Costs in the Sahel (Ouagadougou, Upper Volta: OECD, 1982) , p. 15. The notable exceptions were from a forest plantation in Mali, a fishing boat motorization program in Senegal, and road maintenance in Senegal and Niger. 13. The countries here were not specified. The agri¬ cultural project was a French-supported cotton production scheme in Chad. 14. Gray and Martens, op. cit., p. 102. 15. Club du Sahel/CILSS, op. cit., p. 107. 16. Ibid. 17. International Monetary Fund, Government Finance Statistics Yearbooks Volume IX, 1985 (Washington, D.C., 1985), various country tables. 18. Ibid., various country tables. 19. A. Jennings, "The Recurrent Cost Problem in the Least Developed Countries," Journal of Development Studies 19 (July 1983) , p. 512. 20. The World Bank, Toward Sustained Development in Sub-Saharan Africa: A Joint Program of Action (Washington> D.C., 1984), Table 5, p. 61. 21. Club du Sahel/CILSS, Recurrent Costs of Develop¬ ment Programs in the Countries of the Sahel: Analysis and Recommendations (Paris: OECD, 1980), p. 242. Data were unavailable for Togo and Ghana. 22. Ibid. 23. Club du Sahel/CILSS, Symposium on Recurrent Costs in the Sahel (Ouagadougou, Upper Volta: OECD, 1982) , pp. 417-430. 24. Club du Sahel/CILSS, Recurrent Costs of Develop¬ ment Programs in the Countries of the Sahel: Analysis And Recommendations (Paris: OECD, 1980), p. 246. 142 25. A.A. Tait, W.L.M. Gratz, B.Y. Eichengreen, "Inter¬ national Comparisons of Taxation for Selected Developing Countries, 1972-76," International Monetary Fund Staff Papers 26 (March, 1979), pp. 123-156. All countries were based upon GNP except Niger which is based upon GDP. Ghana's average tax ratio for 1972-76 was 14.2 and Togo's was 12.4 The ITC indeces for Ghana and Togo were 0.976 and 0.703, respectively. 26. Ibid. 27. D. Sykes Wilford and Walton T. Wilford, "On Reve¬ nue Performance and Revenue Income Stability in the Third World," Economic Development and Cultural Change 26 (April 1978), pp. 80-81. 28. Ibid., Table 2.2, p. 23. This figure was based upon twelve nations. Data unavailable for Guinea-Bissau and Guinea. 29. Ibid. The latter figure excludes Togo. 30. Club du Sahel/CILSS, op. cit., p. 248. 31. Government's average net share of total available domestic credit excludes claims by official entities. 32. International Monetary Fund, International Finan¬ cial Statistics Yearbook, 1985 (Washington, D.C., 1985), specific country Monetary Survey tables, line 31n. Data was obtained for all seven nations. 33. Ibid., specific country Monetary Survey table, line 32b. 34. Club du Sahel/CILSS, op. cit., p. 253. 35. Ibid., p. 254. 36. Douglas Rimmer, The Economies of West Africa (New York: St. Martin's Press, 1984), p. 138. 37. International Monetary Fund, op. cit., various country tables. Imports were configured as c.i.f. (cost, insurance, and freight). 38. The World Bank , op. cit.. Table 7, p. 63. 39. Ibid., Table 13, p. 69. 40. Club du Sahel/CILSS, op. cit., p. 261. 143 41. Malcolm Gillis, Dwight Perkins, Michael Roemer, and Donald Snodgrass, Economics of Development (New York: W.W. Norton and Company, Inc., 1983), p. 305. 42. Club du Sahel/CILSS, op. cit., p. 262. 43. Rimmer, op. cit., p. 225. 44. The World Bank, op. cit. This series of figures excludes Burkina Faso. The information on private consump¬ tion was obtained from Table 4. The information on GDP growth rates was derived from Table 2. 45. Ibid., Table 5, p. 61. Data were excluded for Burkina Faso and The Gambia. 46. Club du Sahel/CILSS, op. cit., p. 241. 47. While official development assistance includes loans made to developing countries on 'soft' or conces¬ sional terms plus grants (outright gifts), only loans are considered here for purposes of deficit finance since they alone come within the definition of means used to finance the deficit. 48. The World Bank, op. cit., Table 18, p. 77. All of these figures were rounded. 49. The World Bank, World Development Report, 1986 (Washington, D.C., 1986), Table 21, p. 220. These figures were rounded. 50. Ibid. 51. Heller, op. cit., p. 39. 52. Gray and Martens, op. cit.p. 104? Club du Sahel/ CILSS, op. cit., p. 311. 53• Ibid., p. 102. 54. Gray and Martens, op. cit., p. 107. 55. Ibid. 56. International Monetary Fund, Government Finance Statistics Yearbook, 1985 (Washington, D.C., 1985), various country tables. The years do not correspond in all cases due to the sporadic nature of available data. 144 57. The advantage of committed expenditures over un¬ committed expenditures was derived by noting the averages for uncommitted expenditures for each of the seven nations in Table 20. An average of these averages was then taken and subtracted from 100. The result shows that committed expenditures received an average 73.4 percent of current expenditure while uuncommitted expenditures received the balance (26.6 percent). 58. This figure was obtained by adding the absolute committed expenditure advantages of Burkina Faso, Senegal, Niger, and Togo (listed in Table 21) then averaging them. 59. This figure includes military non-personnel- related expenditures. 60. Defined as the expenditure on 'other goods and services' vs. the expenditure on 'wages and salaries.' International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C., 1985), country table C, line 1.4 divided by line 1.1. 61. International Monetary Fund, Government Finance Statistics Yearbook, Volume IX (Washington, D.C.), various country tables. 62. A. Premchand, Government Budgeting and Expenditure Controls; Theory and Practice (Washington, D.C.: Inter¬ national Monetary Fund, 1983), p. 239. 63. Ibid. 64. Ibid., Table 13, p. 241. 145 CHAPTER 7 CONCLUSION Insofar as the nations of the composite sample can improve their domestic resource mobilization, their pro¬ ject-related recurrent cost problem may be eased. However, given many of the structural weaknesses of these countries, raising revenues through domestic savings and user charges will be limited. Further, as most of them seem implicitly to reject inflationary financing of their deficits, their only recourse is the possibility of effecting an increase in their tax revenues. While some improvement in tax levying and tax collection performance would assist in achieving this end, the bulk of such new revenues must come from a general growth in the tax base in real terms. Such growth, in turn, must await fundamental policy changes on the part of governments? notably, reductions in the tenden¬ cy to appropriate large shares of scarce capital and to undervalue the cost of capital and foreign exchange. It is important to note that, even if tax revenues are augmented, by whatever method, there can be no assurance that law makers or administrators will utilize such increments to rectify the recuurent costs of development projects. How¬ ever, the tendency to squeeze out uncommitted revenues. 146 which support these costs, in favor of those supporting committed expenditures, must certainly be eased as this condition arises mostly from the knowledge of an overall revenue constraint and the need to support committed expen¬ ditures. In this light and pending such expansion in the tax base, the nations of the composite sample would do well to pursue actions designed to reduce committed expenditures. Here, efforts could begin by placing more control of gov¬ ernment operations in the hands of the private sector, thereby reducing huge expenditure commitments, in the form of subsidies, to publicly-owned enterprises. Committed expenditures, in the form of wages, salaries, and pensions, could also be reduced by making current civil service systems more competitive and entry more difficult. Donor- recipient cooperation in the rescheduling of debt would also reduce overall committed expenditures and ease some of the pressure on uncommitted revenue. Fiscal considerations and the composite sample coun¬ tries aside, a general alleviation of the recurrent cost problem among LDCs worldwide must come, first and foremost, from the framework and environments within which develop¬ ment projects are conceived, planned, and executed. Speci¬ fically, both the donor and the recipient authorities must come to terms with the purpose of aid and the desired effects of projects related to development. 147 As was.demonstrated, donor organizations, at least in the United States have, to an extent, been the pawn of other agencies with wide interests and more political clout. Subsequently, the intended objective of development on the part of the aid agencies has been compromised either because other American agencies, charged with the oversight of donors, balk at the unfamiliarity of the aid transfer environment, or because other considerations, more crucial to American interests, come into play. Such considerations do lend themselves to potential solutions. What may be needed is a setting in which the uncertainty of the aid transfer environment does not threaten the organizational health of the donor. If donor organizations could somehow gain independence from the apparent 'overextension of au¬ thority' by agencies charged with their supervision, then the donors and their overseers would not be so compelled to impose order on that world. This end, however, ought not to be construed as an argument for licentiousness. Final control and accountability are essential. What is neces¬ sary is the minimization of control by external organiza¬ tions whose goals are contradictory to those of the aid agency. Admittedly, such changes would require a great deal of rethinking on the part of the Washington bureaucracy as to the true objectives of aid. This may be virtually imposs¬ ible given the wide range of interests involved in the 148 definition and execution of American foreign policy, mili¬ tary or commercial. It may, however, be of some value to those bodies which generate this policy to realize how their interaction with donor agencies may affect project- related recurrent costs, particularly if viable associa¬ tions with the nations receiving development assistance are to be maintained. Congress might be in a position to enact legislation which might address some of the concerns fos¬ tered by the institutional aid environment—legislation designed to curb the the extent of imposition which over¬ sight agencies are currently exerting on donors. If such changes were to be instituted, and assuming that donors are made more aware of the recurrent costs associated with specific projects, then the tendency to transfer large amounts of aid or to impose large, import and/or capital- intensive projects on recipients, regardless of their value toward promoting development and their implications for recurrent costs, must certainly be reduced. Likewise, recipients would be able to retain more sovereignty as to the direction of development by retaining the bulk of the project generation function with the assistance, rather than the dominance, of aid agencies. In this way, too, potential long run recurrent costs may be avoided or re¬ duced. Further, recipients would not be as compelled to 149 accept 'aid for its own sake' but might seek such assis¬ tance on terms more agreeable to their own development priorities. Pending such fundamental reform in the policy making institutions of the donor countries, there are several changes in policy which could be instituted, either alone or in concert with one another, which may alleviate the problem of unmet project-related recurrent costs. At its simplest, the recurrent cost problem of the developing countries could be alleviated by reducing the amount of official development assistance being transferred to them. This would slow the total number of new projects being constructed in both the short and long run. However, such an approach is naive in that the progress of social and economic development would be arrested. The remaining alternative, then, is to determine ways of applying the present flow of resources from donor to recipient more efficaciously. This end will require the mutual participa¬ tion of both parties involved in the aid transfer process. Donors and their supervisory agencies have several options open to them. These include those that relate to the redefinition of aid and those that relate to the de¬ sign, execution, and follow-up phases of projects. With respect to the redefinition of aid, donors might liberalize the current shape that aid takes. They could begin by 150 providing aid to replace domestic resources from the uncom¬ mitted portion of the budget currently used to support general administrative services not associated with eco¬ nomic development, e.g., construction of office buildings and other capital expenses. This would alleviate the 'squeeze' to uncommitted revenue. They could provide di¬ rect aid for the recurrent costs of development projects. However, as was demonstrated, this approach appears unlike¬ ly given current donor attitudes. In view of this reluct¬ ance, donors might link aid to the objective of reducing overall poverty and resource scarcity rather than reducing project-related recurrent costs. Specifically, donors would commit themselves toward reimbursing, over a given period of years, a certain portion of a country's total current budget for a particular sector or set of activities 1 within a sector, such as health or education. If this were to occur, project-related recurrent costs might be reduced as emphasis shifts from donor-supervised projects to whole sectors, potentially eliminating much of the dama¬ ging interaction between donor and recipient described in Chapter Four. With respect to the design, execution, and follow-up phases of projects, donors might seek to reduce the utili¬ zation of inappropriate technologies, such as overly tech¬ nical or capital-intensive components in development pro¬ jects. Further, they might seek to reduce the scale of 151 projects to more manageable dimensions. Donors could also reclassify some recurrent costs associated with projects as part of the project-initiation phase rather than the opera¬ tion phase, thereby reducing the total magnitude of the recurrent cost burden. Further, as the concept of 'in¬ stalled capacity' implies more subjectivity, donors might seek to extend the project 'operating-capacity' horizon to allow more time for projects to become productive. Donors might also give more attention to post-project budget plan¬ ning, i.e., including the recipient's ability to meet fu¬ ture projects' recurrent budget demands in their inception and design phase. Moreover, they might seek to normalize their development strategies over several years to reduce the impact of the changing priorities of different politi¬ cal adminstrations, thereby ensuring continuity. Apart from the redefinition of aid and the project institution and follow-up phases, donors might alleviate the uncommitted revenue squeeze by increasing the oppor¬ tunity cost of aid to that of uncommitted revenue. As was demonstrated, most ODA is tied to specific end-uses. Thus, recipients can either accept the aid for specific projects or see the donors take it elsewhere. The more insistent the donor is with respect to the particular end-uses for which it will provide aid , the lower will be the opportu¬ nity cost attached by recipients to this aid and to meeting recurrent costs associated with aid-sponsored development 152 projects. In short, recipient governments will allow at least part of the capacity installed with the help of tied aid to deteriorate, preferring to use their own scarce, uncommitted revenues (with high opportunity cost) for other expenditure objects to which they give priority. One way to avoid these circumstances would be to reduce the amount of tied aid, in effect, raising the opportunity cost of the remaining aid. Similarly, recipient governments could, in their planning stages, minimize the approval of low prior¬ ity aid-financed projects, i.e., stop accepting aid for its own sake and concentrate on using whatever aid meets their own development criteria. There are several other actions recipients might take to reduce their project-related recurrent costs. Chief among these would be the closer integration of their plan¬ ning and budgeting functions. To explain, those respon¬ sible for budgeting will have different motives than those responsible for planning the course of development. Budge- ters are efficiency oriented while planners are growth oriented. Further, budgeters tend to be unrealistic in calculating total revenue and assume that current revenues are predictable. This is not usually the case given the 2 trade-related nature of the tax systems of most LDCs. Also, planners have not been characteristically aware of the implications that specific projects pose for the cur¬ rent budget. Finally, the planning and budgeting processes 153 are frequently not synchronized with one another, making many investment requests untimely. On this basis, develop¬ ment plans will tend to exceed the capabilities of current budgets to support the long run, aggregate recurrent costs which the projects associated with development plans engen¬ der. Closer integration of development plans with budget cycles and with budgetary capabilities would thus lay the groundwork for the future minimization of project-related recurrent costs by ensuring that budgets will be capable of supporting such costs. Recipients, in cooperation with donors, might also concentrate on a more careful consideration of public ex¬ penditure end-uses, particularly with respect to which ac¬ tivities should be preserved or expanded in the interests of social welfare. Recurrent cost aid could then be mini¬ mized and would be channelled to those activities on the priority list meeting these criteria. Clearly, much of the fundamental framework in which development decisions are made must change. This will require, first and foremost, a willingness on the part of both the donors and the recipients to recognize that the inability or unwillingness to consider project-related recurrent costs has some very real implications for the progress of development among LDCs worldwide. Secondly, both parties must recognize how their interaction negative¬ ly affects these costs. 154 In addition to fundamental changes in the aid concep¬ tualization and transfer process, American donors must take the lead in designing positive means for the long term reduction of project-related recurrent costs via the redefinition of aid vehicles, the reconsideration of the design, execution, and follow-up phases of projects, and the increasing of the opportunity cost of aid. This latter point can come only as donors and Congress, alike, and other bilateral and multilateral aid-disburing agencies worldwide work toward the relaxation of specific end-uses for aid. Again, such changes must hinge upon difficult political considerations—considerations which lie at the very heart of the rationale behind aid transfer. Certainly, the vi¬ tality of future development efforts will be compromised if such actions are not taken and project-related recurrent costs continue to increase without hope of being met. 155 ENDNOTES 1. Club du Sahel/CILSS, Recurrent Costs of Develop¬ ment Programs in the Countries of the Sahel; Analysis and Recommendations (Paris: OECD, 1980), p. 309. 2. Naomi Caiden and Aaron Wildavsky, Planning and Budgeting in Poor Countries (New York: John Wiley and Sons, 1976), p. 242. A 156 REFERENCES CITED Aghevli, Bijan and Kahn, Mohsin. "Government Deficits and the Inflationary Process in Developing Countries." International Monetary Fund Staff Papers 25 (1978). Caiden, Naomi and Wildavsky, Aaron. Planning and Budgeting in Poor Countries. New York: John Wiley and Sons, (1976). Chenery, Hollis and Strout, Alan. "Foreign Assistance and Economic Development." American Economic Review 56 (September 1966) . Club du Sahel/CILSS. Recurrent Costs of Development Programs in the Countries of the Sahel: Analysis and Recommendations. Paris: OECD, 1980. . Symposium on Recurrent Costs in the Sahel. Ouagadougou, Upper Volta: OECD, 1982. Deutsch, Karl. W. "Social Mobilization and Political Development." American Political Science Review 55 (September 1961). Friedmann, Wolfgang; Kalmanoff, George; Meagher, Robert International Financial Aid. New York: Columbia University Press, 1966. Gillis, Malcolm; Perkins, Dwight; Roemer, Michael; and Snodgrass, Donald. Economics of Development. New York: W.W. Norton and Company, 1983. Goffman, I.J., and Mahar, D.J. "The Growth of Public Ex¬ penditures in Selected Developing Nations: Six Carib¬ bean Countries, 1940-65." Public Finance 26 (January 1971). Goode, Richard. Government Finance in Developing Econo¬ mies. Washington, D.C.: The Brookings Institution, 1984. Gray, Clive and Martens, Andre. "The Political Economy of the 'Recurrent Cost Problem' in the West African Sahel." World Development 11 (1983). Griffin, K.B., and Enos, J.L. "Foreign Assistance: Objec¬ tives and Consequences." Economic Development and Cultural Change 18 (April 1970). 157 Heller, Peter. "Public Investment in LDCs with Recurrent Cost Constraint: The Kenyan Case." Quarterly Journal of Economics 65 (May 1974). . "A Model of Public Fiscal Behavior in De¬ veloping Countries: Aid, Investment, and Taxation." American Economic Review 65 (June 1975). • . "The Underfinancing of Recurrent Development Costs." Finance and Development 16 (March 1979). Hinrichs, Harold. A General Theory of Tax Structure Change During Economic Development. Boston: Cambridge Uni¬ versity Press, 1966. International Monetary Fund. Government Finance Statistics Yearbook. Washington, D.C., 1985. _. International Financial Statistics Yearbook. 1985 Washington, D.C., 1985. Jennings, A. "The Recurrent Cost Problem in the Least De¬ veloped Countries." Journal of Development Studies 19 (July 1983) . Jhingan, M.L. The Economics of Development and Planning. 15th ed. New Delhi: Vikas Publishing House, PVT., LTD., 1982. Lateef, Noel. Crisis in the Sahel: A Case Study in Devel¬ opment Cooperation. Boulder, Co.: Westview Press, 1980. Mikesell, Raymond. The Economics of Foreign Aid. Chicago: Aldine Publishing Company, 1968. Morrison, Thomas K. "Structural Determinants of Government Budget Deficits in Developing Countries." World Devel¬ opment 10 (1982). Nurske, Ragnar. Problems of Capital Formation in Underde¬ veloped Countries. New York: Oxford University Press, 1953. . - Organization for Economic Cooperation and Development, Development Assistance Committee. 1979 Review. Paris, 1980. Pillai, Vel. "External Economic Dependence and Fiscal Pol¬ icy." The Journal of Development Studies 19 (October 1982). 158 Premchand, A. Government Budgeting and Expenditure Con¬ trols: Theory and Practice. Washington, D.C.: Inter¬ national Monetary Fund, 1983. Prest, A.R. Public Finance in Underdeveloped Countries. New York: John Wiley and Sons, 1972. Rimmer, Douglas. The Economies of West Africa. New York: St. Martin's Press, 1984. Tait, Alan A.? Gratz, Wilfred? Eichengreen, Barry J. "International Comparisons of Taxation for Selected Developing Countries." International Monetary Fund Staff Papers 26 (March 1979). Tanzi, Vito. "Inflation, Lags in Collection and the Real Value of Tax Revenue." International Monetary Fund Staff Papers 24 (March 1977) . Tendler, Judith. Inside Foreign Aid. Baltimore: The Johns Hopkins University Press, 1975. Wilford, D. Sykes and Wilford, Walton T. "On Revenue Per formance and Revenue Income Stability in the Third World." Economic Development and Cultural Change 26 (April 1978). World Bank. Public Finance in Egypt: Its Structure and Trends. World Bank Staff Working Papers. No. 639 (1984). . Toward Sustained Development in Sub-Saharan Africa: A Joint Program of Action. Washington, D.C., 1984. . World Development Report, 1982. Washington, D.C., 1982. • World Development Report, 1986. Washington, D.C., 1986. 159 APPENDIX DEFINITIONS Observing international development practice concerning official development assistance (ODA), one notes that the concepts of recurrent and non-recurrent costs and expendi¬ tures have different meanings as used by one or another donor agency or aid-receiving country. The purpose of this section is to review definitions of each of these items that have been conceptually acceptable to both interna¬ tional finance agencies and aid recipients. It should be noted here that specific classifications of recurrent and non-recurrent costs and expenditures by type and by eco¬ nomic sector will not be presented and are not necessary to achieve an understanding of the motivating factors behind the recurrent cost problem. Within the context of the guidelines on Local and Recurrent Cost Financing adopted by the Organization for Economic Cooperation and Development's Development Assis¬ tance Committee (DAC) on May 3rd 1979 and hence endorsed by the major developed market economy governments, recur¬ rent cost financing was defined as: ... financing needs from specified development projects/programs for procurement of goods and services (including salaries of local personnel) required for operating and maintaining a given 160 project/program during and after completion of the initial financing...** This formal definition, however, is not functionally correct in its entirety and tends to confuse any assess¬ ment of the overall impact of recurrent costs on developing country budgets. Specifically, because it included the concept of recurrent costs during the initial financing (start-up) phase as well as those costs characteristic of the operating phase, the DAC's interpretation tended toward incorrectly characterizing, and therefore magnifying, the overall level of project-related recurrent costs among developing countries. It was for this reason that The Working Group on Recurrent Costs in the West African Sahel took issue with the DAC's interpretation. Formed under the auspices of the OECD's Club du Sahel secretariat and the secretariat of the eight-member asso¬ ciation of Sahelian governments known as the CILSS or Comite Permanent Inter-Etats de Lutte Centre dans le Sahel (Permanent Committee for Drought Relief in the Sahel) the Working Group convened four times in Ouagadougou, capital of Burkina Faso (formerly Upper Volta) and headquarters of the CILSS, between January 1979 and June 1980. The Group's final report, issued in August 1980, was based on consul¬ tant-prepared case studies of 29 projects distributed among seven sectors and five countries and may perhaps qualify as the first systematic effort to apply the DAC's definition 161 of recurrent costs to a specific region of the developing world. In its deliberations, the Group formulated its own definition of recurrent costs designed to eliminate some of the errors of the DAC's earlier characterization. This definition is outlined below: The set of annual flows of gross expenditure of the government and its agencies, in local curren¬ cy or foreign exchange, undertaken in order to generate socio-economic benefits in connection with the operation and maintenance of a unit of installed capacity, regardless of the source of finance of the expenditure in question, domestic or foreign.^ Conversely, the Working Group defined non-recurrent costs as: All expenditures incurred in connection with establishing this capacity, regardless of the nature of the expenditure (purchase of capital goods or current imports, personnel payments etc.), the type of payment (in foreign exchange or local currency) , and the source of financing.-^ Having distinguished between the two, it is instruc¬ tive to briefly discuss some of the key constituent ele¬ ments of the definition of recurrent costs/expenditures as adopted by the Working Group. The Concept of Gross Cost/Expenditure The term 'gross' indicates that the expenditure is measured before subtracting public sector receipts, in cash or in kind, to which a project may give rise. This 162 approach was used in the belief that recurrent cost anal¬ ysis should focus on the total process of cost generation independent of any mechanism of cost recovery. Looking only at net expenditures would incur the danger of overlooking the nature of the process that generates the expenditure in the first place. Expenditures in Local Currency and Foreign Exchange"" Many individuals have considered that, to be treated as recurrent, an expenditure must take place in local cur¬ rency i.e., be devoted to the purchase of locally produced goods and services. In reality, project operation and main¬ tenance also requires imported goods and services (fuel, spare parts, etc.) which may, as in the case of road main¬ tenance, constitute a major share of recurrent expenditure. Thus, according to the Working Group's definition, expendi¬ tures in both local currency and foreign exchange are treated as recurrent. The Notion of Installed Capacity The reference to 'installed' capacity indicates that we start counting recurrent costs only after a project has been established. In other words, the notion of recurrent costs/expenditure has concrete meaning only after a project 163 has attained its phase of normal operation. All expendi¬ tures undertaken before the commencement of normal opera¬ tion are thus classified in the category of non-recurrent or development expenditures. The Government Accounting Structure and Recurrent Costs The final key element in the"definition of recurrent costs places any given project within the framework of the consolidated yearly government accounts, comprising the accounts of the central government, local authorities and all public sector agencies. It should be noted that, insofar as recurrent costs are financed by foreign aid agencies (an infrequent prac¬ tice) , they are generally not included in the budgets of the government or its repective projects. 164 ENDNOTES 1. Development Assistance Committee, OECD, 1979 Review (Paris: OECD, 1980), p. 177. 2. Club du Sahel/CILSS, Recurrent Costs of Develop¬ ment Programs in the Countries of the Sahel: Analysis and Recommendations (Paris: OECD, 1980), p. 19. 3. bid.