Information about Asia and the Pacific Asia y el Pacífico
Journal Issue
Share
Article

The Role of the Public Sector in the Market Economies of Developing Asia

Author(s):
International Monetary Fund
Published Date:
January 1988
Share
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

I. Introduction

Papers on the role of the public sector tend to focus on what the public sector ought to do on the basis of theoretical assumptions about the existence of public goods and externalities, the need to mobilize resources and to create infrastructures, and the need to redistribute income and to stabilize the economy. These discussions are inevitably influenced by ideological considerations and by an implicit concept of the public sector as an all-knowing entity that makes few if any mistakes and that always aims at maximizing the “general” welfare. These normative discussions, that fill the pages of public finance textbooks, are useful for providing a framework for reference. 1/ However, more attention must be directed toward the positive role of the public sector recognizing that governments are run by real, and not fictional, individuals who have biases just like everyone else and who make their share of mistakes, and recognizing that public sectors are not monolitic institutions but are generally made up of hundreds of separate and, to some extent, independent entities. 2/ Thus, it may be more useful to reverse the normal question by focusing on what governments should not have done rather than what they should have done.

To limit the scope of the paper I shall deal only with the following nine countries: Bangladesh, India, Indonesia, Korea, Malaysia, Pakistan, Philippines, Sri Lanka, and Thailand. These countries contain about a fourth of the world’s population.

II. A Sketch of a Positive Theory of Public Sector Intervention

The market economies of Asia have assembled in recent decades an economic record that, in many ways, is the envy of policymakers in other regions of the so-called developing world. Whether the performance of these economies is assessed by the rate of growth, by the rate of inflation, by the growth of exports, or by the fact that by and large most of these economies have managed to avoid the economic crises that have afflicted other regions, the basic conclusion must be that the performance of these countries, taken as a group, has been quite good. 3/ Given this performance, and the fact that the public sector has played a major role in these economies, there must be a presumption that that role must have contributed importantly to these results; or, at a minimum, that it must not have been an obstacle to economic development.

In theory at least there are two polar roles that could be assigned to the public sector: an all embracing one, derived broadly from the work of Karl Marx, and a far more limited one, derived from the work of Adam Smith and his followers. Since we are dealing with market economies, there is a presumption that the optimal role for the public sector would be closer to the one envisaged by Smith than by Marx. However, that role must depend on the weights that the policymakers assign to objectives such as income distribution and stabilization, objectives that were not taken into account by market-oriented classical economists. These other objectives may take several forms. For example, income distribution may refer to distribution by size of income, or between urban and rural dwellers, or between wage earners and those who receive other forms of income, or between the workers and the pensioners, or between the very poor (say, the bottom 10 percent of the income distribution) and the rest. Stabilization may refer to stabilization of output, prices, balance of payments, employment, and so forth.

The greater is the concern of policymakers for these other objectives, the greater the role of the government would be expected to be. 4/ Furthermore, even within the purely allocative, or efficiency-enhancing role emphasized by laissez-faire economists, one could argue that, given the nature of the developing economies and the need to create infrastructures (both social and tangible), given the widespread existence of externalities and distortions, given the much lower quantity and quality of information available to private economic agents than in advanced countries, the optimal role of the public sector must be greater in developing countries than in more advanced economies. Here, however, an important distinction must be made between an optimal “normative” role and an optimal “positive” role.

If governments were made up of policymakers who (i) did not have any objectives other than those consistent with “the public interest”, (ii) had all the information needed to conduct good economic policy, (iii) did not make mistakes, and (iv) had enough control over the public sector’s bureaucracy (tax administrators, employees of public enterprises, etc.) to ensure that no gaps developed between the laws and the effective implementation of those laws, then the relevant role for the public sector would be the “normative” one. In other words the government should intervene in all those cases where there would be a theoretical presumption, on the basis of available knowledge and of rational forecasts, that this intervention would bring results that maximize the public welfare given the resource constraints. 5/

The governments of the Asian countries, as with all governments, are not made up of all-knowing saints but of real-life individuals who, to varying extents, have personal objectives that may at times take precedence over national objectives; and, of course, the national objectives are not always definable or defined. These personal objectives may reflect pressures created by political affiliations, racial, demographic or regional characteristics, class affiliations, friendships, family connections and personal ambitions or greed. When these special interests play a significant role, the results of economic policy and/or the costs of carrying out those policies are likely to differ from those that would have prevailed in the absence of these special interests. More specifically, desirable changes in the national objectives would come at higher, and sometimes at much higher, costs than in the ideal situation. In such cases the optimal “positive” role of the public sector is likely to be somewhat smaller than, and different from, the optimal “normative” role. In other words it would be desirable to reduce the role and, thus, the discretion that policymakers have. Especially in these circumstances, policies ought to be judged by results rather than by declarations of intent.

Governments can pursue their economic objectives through the budget or through regulations. For example, assistance to a producer can be given through a direct subsidy, which requires higher public expenditure, and, thus, higher levels of taxation or borrowing, or, indirectly, by limiting or forbidding the importation of the product produced by this producer. Other types of subsidies can be given through the credit mechanism, through overvalued exchange rates, through so-called tax expenditures, through special permissions for particular activities (permissions denied to other activities). 6/ Therefore, the role of the public sector in the economy cannot be assessed only by the ratios of taxes or public spending to gross domestic product (GDP) as is often done, although these ratios are important. That role should be assessed also by the many regulatory policies that exist in all countries. These regulatory policies may not be correlated (or may even be negatively correlated) with traditional measures of the public sector (share of total taxes or total public expenditure in national income). Unfortunately these regulatory policies are not quantifiable in the same way as tax revenue or public spending so that countries cannot be objectively ranked on this basis.

In most developing countries, the government attempts to determine among other things: (i) who can import, what can be imported, and how much; (ii) the domestic prices at which goods can be imported or exported; (iii) what can be produced domestically and where; (iv) the cost of domestic credit; (v) who gets bank credit, how much, and for what purpose; (vi) who can borrow abroad; and (vii) the prices at which many goods and services can be sold. When all these controls are taken into account, it can be safely stated that the impact of public sector intervention on the economy is generally far greater in developing countries than in developed countries, in spite of their lower tax and expenditure ratios.

There is a fundamental difference between policies that control the economy through the tax-expenditure process and those associated with the regulatory route: there is generally a limit (political or administrative) to taxation and public spending, but there is none to regulations since these can be introduced with very little direct cost. 7/ These can be multiplied ad infinitum and can be extended to all areas of economic activities. Therefore, the potential for economic inefficiency in the use of regulations can, in principle, be far greater than in the use of taxes and public spending. 8/ The efficiency costs to an economy from import quotas, high and differentiated import duties, unrealistic exchange rates, negative interest rates, uneconomic pricing policies, and overregulated economic activities could, under particular circumstances, exceed those associated with inefficient and burdensome tax systems and expenditure programs even though the latter can be very high.

The figure provides a stylized version of the point that I wish to make. The horizontal axis measures the net economic benefit (i.e., net of economic costs) that the country derives from public sector intervention. Here intervention refers to both regulations as well as the tax and spending process. These benefits are positive for some level of intervention and become negative when intervention is carried too far, when it serves special rather than general interests, or when it is based on wrong information, wrong expectations, or wrong policies. The vertical axis measures the degree of public sector intervention, assuming that somehow we have an index that properly measures it. 9/ The curves OA, OB, and OC show how countries can benefit from various degrees of intervention. At points S, T, and V the degree of intervention is optimal given the quality of that intervention. 10/ In other words, at those points the countries derive the maximum net benefit from governmental intervention.

Figure 1

Each curve represents a given country at a given moment in time—i.e., with a given need for public sector intervention and a given set of policymakers. 11/ In the country represented by curve OC the beneficial effect of each level of public sector intervention is much greater than in the country represented by curve OA. This may be due to differences in the efficiency of the private sector in allocating resources and in distributing income; or, it may be due to the fact that the policymakers in country OC would predominantly pursue the public interest while those in country OA would promote policies that ultimately benefit them and their clienteles more than the country at large. 12/ Of course, both of these groups of policymakers will declare that the policies that they are promoting are for the benefit of the country. When intervention is carried beyond the optimal levels represented by S, T, and V, the welfare of the country begins to fall. 13/ Eventually public sector policies may do more economic damage than good (when the curves cross the vertical axis).

Although it is difficult to specify the determinants of the optimal role of the public sector in the real world, they are likely to be related to how well the government is pursuing the traditional economic objectives of allocation of resources, redistribution of income and stabilization of the economy. 14/ Basically the optimal role will depend not only on the results of economic policy, as measured by the objectives pursued, but also on the costs of achieving these objectives.

My basic, and admittedly impressionistic, conclusions about the market economies of developing Asia are the following. I would conclude that by and large economic policy has been better in these countries, taken as a group, than in other regions. However, the relatively good results were achieved at higher costs than was necessary. Putting it differently, intervention by the public sector was carried to a degree somewhat beyond the optimum. The paper emphasizes the allocative role of the public sector; however, some brief reference is also made to redistribution of income and stabilization of the economy.

III. The Role of the Public Sector in Resource Allocation

1. Public spending: level and composition

Table 1 provides information, for the 1981-85 period, for public expenditure in the nine countries considered. Some of the data in this table and in all the others may not be strictly comparable so that caution is warranted. The main objective here is to provide broad impressions rather than precise statistical information. There is a wide range between the almost 45 percent of GDP reached by total public expenditure in Malaysia in 1982 and the level (about 13 percent of GDP) shown by the Philippines in 1984-85. It does not take special expertise to conclude that, from an allocative point of view but not necessarily from a stabilization point of view, public expenditure was probably excessive in the former and too low in the latter. 15/ The levels shown by India and Sri Lanka appear to be relatively high by the criterion of international comparisons. Possibly the same could be said for Indonesia, Pakistan, and Thailand. The validity of these conclusions, however, depends on (i) whether the countries could easily and cheaply finance that level of expenditure, and (ii) whether it was efficiently spent. For example, there is nothing wrong with a high level of public expenditure if it is financed by cheap resources (i.e., efficient taxes and low-cost loans) and if it is spent in a highly efficient way. When these considerations are taken into account, the above conclusions may have to be revised. We shall pay some attention to these issues; however, it is important to emphasize that precise answers are just not possible.

Table 1.Public Expenditure(In percent of GDP)
19811982198319841985
Bangladesh17.419.917.316.116.4
Current6.76.66.67.4
Developmental13.210.79.59.0
India29.732.231.133.633.3
Current10.112.110.911.412.0
Developmental19.620.120.222.221.3
Indonesia23.723.923.520.023.0
Current12.911.811.511.011.7
Developmental10.812.112.09.011.1
Korea22.520.920.620.6
Current16.015.715.616.0
Developmental6.55.25.04.6
Malaysia43.544.638.734.435.2
Current24.426.524.523.426.0
Developmental19.918.614.310.89.3
Pakistan22.522.024.123.924.4
Current14.314.416.517.817.8
Developmental8.27.67.66.16.6
Philippines19.619.717.812.913.2
Current9.710.210.08.49.7
Developmental9.99.57.84.53.5
Sri Lanka33.033.832.530.832.9
Current17.218.317.815.717.1
Developmental15.915.414.815.115.8
Thailand22.323.723.123.424.8
Current14.615.515.816.316.9
Developmental7.98.17.47.07.8
Source: Various government documents and official IMF and IBRD statistics.
Source: Various government documents and official IMF and IBRD statistics.

Table 1 provides also a breakdown of total public expenditure between current and developmental. From a stabilization point of view this distinction is generally not useful since all expenditures contribute to demand pressure and affect the balance of payments. 16/ That distinction is considered useful by many economists and policymakers who believe that a large share of developmental expenditure in total public expenditure is a sign of an economic policy that contributes to growth. However, there is no standardized way of classifying expenditure as current or capital so that what is classified as current in one country may be classified as developmental in another. Furthermore, since it is easier to obtain foreign grants and concessional credit for developmental expenditure than for current expenditure, there is an incentive for countries to make this category look larger than it might be in reality by reclassifying some current expenditure as developmental.

Some aspects of Table 1 merit comment. First, a very large proportion of total public spending is classified as developmental. In India and Bangladesh developmental expenditure far exceeded that for current expenditure; in India it was twice as large. Second, a substantial change over the years characterizes this category compared with current expenditure. Current expenditure is more rigid because it is associated with wages and entitlements that are difficult to change in the short run. Therefore, the impact of financing difficulties often falls on the developmental part of the total. For example, in Malaysia developmental public expenditure dropped from almost 20 percent of GDP in 1981 to about 9 percent in 1985. In Bangladesh it dropped from over 13.2 percent in 1982 to 9 percent in 1985. In Indonesia it dropped from 12.1 percent in 1982 to 9 percent in 1984 and rose again to over 11 percent in 1985. In the Philippines it dropped from 9.9 percent in 1981 to 3.5 percent in 1985. Financing difficulties were often behind these reductions. Third, there was a very low level of developmental expenditure in Korea, which was—ironically—the country with the highest rate of growth.

Although developmental expenditure is not the same as public investment, it is closely related to it. 17/Table 2 shows total gross investment (gross capital formation) as a percent of GDP. It also shows a breakdown of gross investment into public and private for the nine countries. The large role played by the public sector in the capital accumulation of many of these countries is highlighted by this table. In Bangladesh, India, Malaysia, Pakistan, and Sri Lanka the public sector accounted for about half or more than half of total gross investment. In the Philippines and in Thailand the proportion was somewhat lower. In Korea public investment was only about 20 percent of total investment. In many of the countries considered the ratio of gross investment to GDP was high, partly explaining the good growth performance of these countries. 18/

Table 2.Gross Investment(In percent of GDP)
Country19811982198319841985
Bangladesh15.113.612.313.213.0
Public6.26.25.66.15.9
Private8.97.46.77.17.1
India23.924.423.624.425.4
Public11.812.211.212.111.9
Private12.112.212.312.313.5
Indonesia27.125.824.820.319.1
Public5.89.19.56.36.9
Private 1/21.316.715.314.012.2
Korea30.328.629.931.931.2
Public4.34.64.74.84.7
Private26.024.025.227.126.4
Malaysia36.539.238.735.832.6
Public16.719.019.116.216.2
Private20.519.018.318.016.5
Pakistan15.215.315.715.215.0
Public9.29.79.79.08.8
Private6.05.66.06.16.2
Philippines26.125.625.220.116.4
Public9.99.57.44.53.5
Private16.216.117.815.613.0
Sri Lanka27.430.529.025.724.8
Public13.315.513.012.111.8
Private14.115.016.013.613.0
Thailand24.721.023.022.121.2
Public8.77.97.97.98.6
Private16.013.215.114.212.7
Source: See Table 1.

Includes the investment of public enterprises.

Source: See Table 1.

Includes the investment of public enterprises.

The magnitude of the public sector’s gross investment as a share of GDP raises several questions. First, was the investment of the public sector in areas of genuine government concerns such as transportation, irrigation, education, and health; or was it in areas (e.g., manufacturing, agriculture) that, according to the normative theory of public sector intervention, should have been left to the private sector? 19/ Second, was the allocation of public sector investment made on the basis of efficiency criteria or on the basis of other criteria reflecting social, political, or private objectives of some policymakers or public sector employees? Third, how much impact did the cost of raising capital, often from foreign sources, have on the determination of the size of the investment budget and on the choice of specific projects? 20/ Fourth, how much impact did the borrowing to finance this expenditure have on the government’s future bill for interest payments?

2. The efficiency of public enterprises

In several of these countries, historical factors or social consideration contributed or were mainly responsible for the fact that public enterprises were often engaged in activities that are normally left to the private sector in market economies. 21/ For example, in Bangladesh public sector involvement was necessitated by mass exodus of owners and managers at the time of the break-up of Pakistan into two countries in 1971. However, in some of them the government involvement in these activities reflected policy decisions and was perhaps encouraged by the availability of cheap credit to the government or by windfalls from high commodity prices in some periods. More recently there has been a change of attitude that has resulted in some movement towards the privatization of some of these enterprises. In Bangladesh, for example, the share of public investment in manufacturing declined from about 66 percent of total investment in 1980 to less than 40 percent in 1984/85. The trend toward privatization is also apparent in several of other countries. However, in most of these countries government involvement in areas in which it does not have an obvious role to play remains large.

On the basis of available information, it is realistic to conclude that often public sector investment has not followed efficiency criteria. There are too many examples of “white elephants” that have come into existence for reasons that had little to do with efficiency. 22/ In many cases these projects have contributed significantly to the foreign indebtedness of these countries and are likely to reduce their future standards of living. The information available on rates of return to public sector’s investments though limited and not strictly comparable, indicates that, although in these countries the public enterprises (taken as a group) were often not the major drains on the budget as in countries in other regions, the financial rates of return were often very low, especially when the enterprises engaged in oil production are excluded. 23/

According to available information, the recent financial performance of public sector industrial enterprises in Bangladesh shows an overall rate of return of 0.4 percent on the book value of assets in 1985/86. 24/ For the nonfinancial public enterprises taken as a whole, the rate of return on assets was higher at 3.7 percent in 1984/85. However, much of the profits came from enterprises engaged in the extraction of gas.

For India, net profit as percent of capital employed in the public manufacturing and service enterprises of the Central Government was 2.3 percent in 1982/83 and 2.8 percent in 1985/86. However, all the profits came from the 12 enterprises engaged in oil production. As a group, non-oil enterprises have been making losses. State government enterprises have also been running losses. According to a recent World Bank study “In India … the financial performance of public enterprises has been significantly worse than that of private enterprises in the same industry…” 25/

In Malaysia the number of public enterprises increased rapidly in the period between 1980 and 1984. Most of these relied on the government for capital or for guaranteeing external loans. These public enterprises have been running large losses in recent years, in part as a consequence of the fall in commodity prices. Recently steps have been taken to privatize some of them.

In the Philippines the rate of return on equity investment in 15 major public enterprises was estimated at 2.8 percent for the years 1980-84.

In Indonesia the after-tax rate of return on total assets for a sample of 135 nonfinancial public enterprises (excluding Pertamina, the national oil company) was about 1 percent in 1982-85. For the enterprises in the industrial sector the rate of return was close to zero. For all public enterprises combined the share of profits before tax to total assets was 3.0 percent in 1985. However, much of the profits came from financial enterprises and from Pertamina. The after-tax rate of return to Pertamina was 3.5 percent for the 1982-85 period. Losses were particularly significant in enterprises engaged in industry and communications.

In Sri Lanka the returns on many of the massive budgetary capital projects carried out since 1978 have been disappointing. In Pakistan “the returns on capital averaged about 2 percent during the 1970s and slightly over 4 percent during 1980-83, compared to an average rate of inflation over the whole period of about 12 percent.” 26/

When considering these financial rates of return of public enterprises, some factors ought to be kept in mind. First, these are nominal rates of return. Since all of these countries had positive rates of inflation, the real rates of return were often negative. Second, these rates are often inflated by factors such as (i) subsidized credit, (ii) cash grants from the government, (iii) special tax treatments especially in connection with imports, (iv) government implicit or explicit guarantees for foreign borrowing, (v) the provision of monopoly status to the public enterprises, and (vi) the fact that in many cases no deduction from profits are made for depreciation.

Several factors have contributed to the low productivity of public investment. These include (i) the choice of unprofitable investment projects for political or other reasons; (ii) the lower emphasis given to efficiency than to objectives such as employment, redistribution of income, regional development, and so forth; (iii) low capacity utilization; (iv) poor pricing policies; (v) managerial staffing at times based more on political considerations than on managerial ability; (vi) the absence of incentives imposed by competition and by the profit motive; and (vii) for some of them in recent years, the fall in commodity prices.

Whatever the reasons, the net result has been that while the rates of growth of many of these countries have been relatively high, they have been lower than they should have been considering the high level of total investment. In other workds, the incremental capital output ratios (ICOR) in several of these countries have been very high. In India, for example, the ICOR has ranged between 3 and 7 in the first half of the 1980s and was about 5 more recently; in Malaysia it has ranged from a low of about 5 in 1982 to a high of above 12 in 1986; in the Philippines it was 9.8 in 1980-83; and in Thailand it was 4.6 in 1980-83. These figures refer to the economy as a whole and thus apply to total investment. However, since public investment has been a relatively large share of total investment in many of these countries, and since private sector investment has probably been more productive than public sector investment, these figures provide a broad indication of the contribution of public investment to growth. 27/

3. Public spending and the growth of foreign debt

In many of these countries, a good part of investment was financed by foreign borrowing. When, as in Korea, the borrowed capital was used to finance investment projects with high rates of return and high potential for earning foreign exchange, it could be argued that the borrowing was fully justified on allocative grounds. In other words, when the marginal productivity of capital exceeds the marginal cost of borrowing, the country is justified in contracting foreign debt. 28/ However, in several of these countries investments that would generate low or even negative rates of returns were often financed by foreign loans obtained at financing costs higher than the projects’ expected rates of return. 29/ This is an important aspect that deserves further attention.

For the nine countries combined the foreign debt rose from about US$44 billion in 1975 to US$109 billion in 1980 and to US$203 billion in 1986 (see Appendix Table A-1). Interest payments on this debt rose from US$1.4 billion in 1975 to US$13 billion in 1986 (see Appendix Table A-2). Dividing the interest payments by the stock of foreign debt provides a measure of the average cost of foreign borrowing for the region as the whole. This cost was 3.3 percent in 1975, rose to 7.3 percent in 1981, and remained relatively high in the 1981-86 period. The yearly average costs were as follows:

19753.3%19795.3%19836.4%
19763.7%19806.4%19847.1%
19774.0%19817.3%19856.3%
19784.5%19827.0%19866.4%

These are average costs for the whole group. They are, therefore, biased downward by concessional loans which are very important for some of these countries such as Bangladesh, India, and some others (see Appendix Table A-3). The terms on commercially obtained debt were much higher, in some cases as high as 15 percent or even more.

A comparison of these costs with the rates of returns that the public enterprises were obtaining on their investment suggests that the allocation of these foreign resources has been far from optimal. 30/ In order for countries to avoid long-term difficulties on their external borrowing, the uses to which foreign loans are put must generate financial rates of return which, in real terms, are at least equal to the real cost of borrowing. This has not been the case for investments by public enterprises and, as we shall argue below, it is unlikely to have been the case for investments in infrastructures or for those private investments that benefited from generous tax incentives.

The consequence has been a considerable increase in the ratios of external debt (public plus private) to GDP for the majority of these countries (see Table 3). In other words, the stock of external debt has grown much faster than the economy. In some of these countries the external debt to GDP ratio has become disturbingly high. It is not surprising that the ratio of debt service to the earnings of these countries from their export of goods and services has been growing in all of these countries except Korea (see Table 4). 31/ Since a large proportion of the foreign debt is held or guaranteed by the public sector and since much of this debt was contracted to finance investment projects, it is obvious that the role of the public sector in allocating investment has left much to be desired. Foreign borrowing may have created jobs and incomes in the short run but at the cost of reductions in the future disposable incomes of these countries. The servicing of the foreign debts must be seen as a tax that will reduce for years to come the net income available for domestic uses in these countries. 32/

Table 3.Ratio of External Debt to GDP or GNP(In percent)
198119821983198419851986
Bangladesh35.342.239.837.845.0
India12.314.315.517.218.719.5
Indonesia21.123.031.031.135.345.8
Korea49.053.553.152.355.846.9
Malaysia34.047.359.456.569.685.3
Pakistan38.237.243.841.040.744.1
Philippines54.461.672.780.782.195.0
Sri Lanka52.356.258.652.055.760.7
Thailand30.633.334.737.445.343.2
Source: Various IMF, IBRD, and national sources.
Source: Various IMF, IBRD, and national sources.
Table 4.Debt-Service Ratio(In percent of exports of goods and services)
198119821983198419851986
Bangladesh24.528.725.734.043.5
India9.912.114.016.118.019.6
Indonesia11.815.015.518.923.437.3
Korea21.222.620.922.923.924.7
Malaysia8.010.211.113.317.418.8
Pakistan23.525.425.231.736.643.7
Philippines*24.936.632.735.033.033.0
Sri Lanka19.322.324.417.522.327.6
Thailand19.020.923.224.027.427.0
Source: Various IMF, IBRD, and national sources.

The figures for the Philippines for the more recent years are biased downward by the rescheduling of the debt. Without rescheduling the debt-service ratio of that country would have increased more than shown by the table.

Source: Various IMF, IBRD, and national sources.

The figures for the Philippines for the more recent years are biased downward by the rescheduling of the debt. Without rescheduling the debt-service ratio of that country would have increased more than shown by the table.

4. Public expenditure for operation and maintenance

The current expenditure of the public sector has not been as high as the developmental expenditure in the majority of these countries. It was, for example, relatively low in Bangladesh, India, Indonesia, and the Philippines. However, it was high in Malaysia and it was not particularly low in Korea (see Table 1). These levels of current expenditure were achieved in spite of: (i) high expenditure for defense in several of these countries (India, Korea, Pakistan, and Thailand); (ii) fast growing expenditure for interest payment on both domestic and foreign debt; 33/ and (iii) in spite of relatively high levels of spending for education in several of these countries.

The market economies of developing Asia have preferred indirect measures for redistributing income; as a consequence, the public sector’s expenditure for social security, direct welfare payments and subsidies has been low by international standards. For example, this expenditure has been less than 10 percent of total expenditure compared with about 25 percent in Latin America. The level of expenditure on operations and maintenance has also been low in most of these countries.

From much of the evidence available, it can be concluded that these countries have been more successful at building infrastructure, than at its adequate maintainance or use. 34/ Pressures of various kinds have induced the policymakers to give high priority to new investment projects rather than to ensuring that the existing infrastructure is kept in good working conditions and is fully utilized. These pressures have included: (i) the more benevolent international attitude toward “capital” spending compared with “current” spending; 35/ (ii) the greater availability of loans for investment projects than for maintenance expenditure; (iii) the political benefits associated with the opening of new infrastructures or new enterprises; 36/ (iv) the greater facility of planning for investment than for recurrent costs; (v) the traditional attitude of many economists and policymakers that investment is the basic ingredient of growth (see Harrod and Domar models); and, finally, (vi) the possibility that some of those who arrange contracts may at times receive personal pecuniary benefits. 37/

Given the very large sums involved, and the potential benefits to the firms who receive the contracts, there is likely to be greater attempts at bribery associated with the capital expenditure part of public spending than with the current expenditure part. When these attempts succeed, as they occasionally do, part of the capital expenditure is de facto transformed into what could be called a transfer payment. This diversion of investment expenditure benefits those who receive these payments but, by inflating the costs of the investment projects, reduces the rate of return to investment. Often these payments are not referred to as bribes but have acquired more colorful and more neutral names such as “oiling the mechanism,” “tangents,” “under the table payments,” and so on.

Inadequate resources for the operation and maintenance of infrastructures and for the operation of public services have resulted in both deterioration and underutilization of existing capital assets. 38/ A majority of these countries has suffered to varying degrees from this shortcoming. Financial difficulties in some of them have contributed to this problem but cannot be completely blamed for it. At least part of the explanation must be found in the differential benefits that those who make these decisions receive from capital spending as compared with spending on operation and maintenance. One important effect has been a deterioration in the existing public sector capital stock. 39/ Another has been the less-than-full capacity use of existing capital stock.

In the Philippines, for example, in 1985 operation and maintenance expenditures in real terms were 60 percent of their 1977 levels. In Indonesia much of the infrastructure built over the past two decades is reported to be deteriorating and in need of costly rehabilitation. For example, only 40 percent of national and provincial roads are considered in a stable condition. The situation for irrigation canals is similar. Equivalent results have been reported for some of the other countries. A recent study has included Sri Lanka, Philippines, Pakistan, Thailand, India, and Indonesia among the group of countries perceived as having significant recurrent cost problems. 40/

This erosion and/or underutilization of the capital stock of these countries must again be seen in the context of the rapid rise in the stock of foreign debt. It must be recalled that the foreign debt was acquired largely in the process of building (or adding to) that capital stock. Thus, while the foreign debt must be serviced in the future, regardless of what happens to the country’s capital stock, the ability of that capital to generate future income will depend on how well it is maintained and how fully it is used. 41/ In the absence of adequate planning and budgeting for operation and maintenance, borrowing for new investment is not likely to be cost-effective and the increase in the countries’ earning power to repay the accumulated debt will be less rapid than it could have been.

5. The tax systems, the foreign debt, and the rate of return to private investment

One of the major instruments available to policymakers for pursuing their social and economic objectives is the tax system. Tables 5 and 6 provide information on the structure of the tax systems of the countries under considerations. Unfortunately the statistics are not very up to date. However, these statistics do not change much over the short run so that the tables are likely to be fairly representative of the current situation.

Table 5.Selected Asian Countries: Tax Revenue by Type of Tax, Three-Year Averages(In percent of GDP)
Domestic Taxes on Goods and Services
CountriesYearsTotal TaxesIncome TaxesTotalGeneral Sales Turnover, VATExcisesOtherForeign TradeSocial SecurityWealth and PropertyOther
TotalIndividualCorporateOtherTotalImport DutiesExport DutiesOther
Bangladesh1981-837.091.231.232.951.221.700.033.342.900.120.310.19
India 1/1981-8316.262.400.941.320.1410.140.135.114.903.062.930.040.090.180.14
Indonesia 1/1982-8418.5414.950.5013.750.702.271.071.000.200.840.720.120.340.70
Korea1983-8516.794.482.392.098.454.112.531.812.862.860.250.14
Malaysia1982-8421.8110.032.417.614.791.591.801.396.273.582.690.140.14
Pakistan1982-8412.922.355.421.024.405.104.840.110.150.05
Philippines 2/1983-8510.292.490.901.370.224.261.052.101.102.962.630.180.120.360.22
Sri Lanka1982-8417.713.060.892.177.814.792.880.146.524.092.430.06
Thailand 2/1983-8515.023.131.701.447.982.873.771.353.423.060.340.030.37
Sources: IMF, Government Finance Statistics Yearbook, Vol. X, 1986.

State tax revenues are included.

Local tax revenues are included.

Sources: IMF, Government Finance Statistics Yearbook, Vol. X, 1986.

State tax revenues are included.

Local tax revenues are included.

Table 6.Selected Asian Countries: Tax Revenue by Type of Tax, Three-Year Averages(In percent of total tax revenue)
Domestic Taxes on Goods and Services
CountriesYearsTotal TaxesIncome TaxesTotalGeneral sales turnover, VATExcisesOtherForeign TradeSocial SecurityWealth and PropertyOther
TotalIndividualCorporateOtherTotalImport dutiesExport dutiesOther
Bangladesh1981-83100.0015.6415.6437.3715.4721.540.3642.3536.821.504.032.40
India 1/1981-83100.0014.755.808.100.8562.370.8331.4230.1218.8318.000.260.571.130.85
Indonesia 1/1982-84100.0080.612.6974.183.7412.245.765.381.094.543.870.671.823.74
Korea1983-85100.0026.7214.2512.4750.3324.4815.0510.7917.0217.021.470.86
Malaysia1982-84100.0045.9811.0634.9121.927.268.276.3928.7816.4412.350.650.62
Pakistan1982-84100.0018.2241.927.9034.0239.4937.470.881.140.35
Philippines 2/1983-85100.0024.248.7613.362.1141.3010.1120.5010.6928.8525.581.841.443.462.11
Sri Lanka1982-84100.0017.344.9912.3543.9627.0816.060.8136.8822.9113.970.37
Thailand 2/1983-85100.0020.8911.329.5753.1319.1125.078.9622.7920.372.240.172.45
Sources: IMF, Government Finance Statistics Yearbook, Vol. X, 1986.

State tax revenues are included.

Local tax revenues are included.

Sources: IMF, Government Finance Statistics Yearbook, Vol. X, 1986.

State tax revenues are included.

Local tax revenues are included.

A few features of the statistics merit attention. First, the level of corporate income tax revenue in Indonesia and Malaysia is high. These are essentially taxes on mineral exports and especially on oil. They are an unstable source of revenue as both of these countries have recently witnessed. Second, the revenue from income taxes on individuals is relatively low. Third, the contribution of excise taxes to total revenue is higher than that of general sales taxes. This partly reflects historical factors and partly the desire of policymakers to influence the pattern of consumption of the citizens. Fourth, foreign trade taxes still play an important role which may reflect to some extent the desire of policymakers to influence the pattern of imports. Finally, the low level of social security contributions is very low.

Table 7 provides more recent estimates of the ratio of tax revenue to GDP. 42/ No particular trend is noticeable in the ratios shown in the table. These countries cannot be considered overtaxed especially when it is realized that the high tax ratios for Indonesia and Malaysia are largely the result of taxes on the “rents” that these countries receive from their mineral deposits. The comparison of the data in Table 7 with those in Table 1 shows that in several of these countries tax revenue covers only a relatively small fraction of total” public expenditures. The difference is, of course, covered by (a) borrowing, (b) grants, and (c) nontax revenue. For many of the countries considered tax revenue covers broadly the level of current public expenditure.

Table 7.Tax Revenue(In percent of GDP)
19811982198319841985
Bangladesh7.36.86.97.1
India16.416.716.116.516.7
Indonesia21.119.020.820.719.8
Korea18.118.819.418.718.6
Malaysia23.422.024.323.224.4
Pakistan13.713.313.513.912.8
Philippines10.310.110.59.310.1
Sri Lanka16.715.516.219.418.9
Thailand12.812.413.313.713.6
Source: IMF, IBRD, and national sources.
Source: IMF, IBRD, and national sources.

There are two aspects that merit consideration as they are linked with the earlier discussion in this paper. These are the elasticity of these tax systems and their role in encouraging capital accumulation. Table 7 implies that the buoyancy of these tax systems was around 1.0 since the tax ratios do not show any particular trend. 43/ However, Table 7 does not show that these relatively constant ratios were often achieved at great political costs, due to the many discretionary changes that some of these countries needed to make over the years to prevent the level of taxation from falling as a share of GDP, since the elasticity of these tax systems was in many cases much lower than 1.0. 44/ For example, in unpublished studies covering the period under consideration the elasticity has been estimated at 0.7 for Bangladesh, 0.5 for the Philippines, 0.73 for Sri Lanka, and 0.92 for Thailand.

As discussed earlier these countries have borrowed on a large scale to finance their capital expenditure. In the process they have accumulated sizable foreign debts. For some of these countries the external debt is largely public debt. At some point, when the stock of foreign debt reaches a given share of GDP, the net flow of resources between the countries and the rest of the world is likely to change direction. 45/ The higher is the rate of interest, and the more reluctant are lenders to keep lending, the more quickly will this point be reached. While in the earlier period the net flow of resources had been from the foreign lenders to the countries, in the later period it must be from the governments to the foreign creditors. This occurs when the interest payments exceed the difference between new borrowing and amortization for existing debts. When this happens, in the absence of a contraction in noninterest government spending, the ratio of tax revenue to GDP will need to rise to generate the fiscal resources needed to make these payments. 46/ This is the fiscal counterpart of the external debt problem. To make the foreign payment the country must also run a surplus in its trade account.

Countries with low elasticity of their tax systems will face greater difficulties to generate this fiscal surplus through automatic tax increases. They will, thus, have a greater chance of running into financial difficulties, especially if political constraints limit their action on the expenditure side or if these constraints prevent the introduction of discretionary measures of sufficient magnitude to raise taxes to the needed level. As a consequence, it would be desirable for these countries to reform their tax systems to make them acquire a higher built-in elasticity. The desirable size of the elasticity would itself depend to some extent on the rate at which the public debt is to be serviced.

The low elasticity of the tax systems of these countries is not an accident but is, at least in part, the result of explicit policy actions. More specifically, in several of the countries considered, it is the consequence of the erosion of tax bases due to the proliferation of tax incentives granted to many activities. In some of these countries the tax bases have been reduced over the years to small fractions of their potential values. Many of these incentives were granted to private investment and resulted in large subsidies to capital accumulation. A study of these tax incentives concluded that:

“…giving large subsidies to capital to the extent of an overall net subsidy might be questioned on efficiency grounds. The net subsidy drives the required rate of return on capital below the investors’ subjective rate of time preference, thereby creating incentives to overinvestment…” 47/

Another study has calculated the effects of these tax incentives on the cost of capital faced by enterprises in seven of these countries. The results are reported in Table 8. The table shows that in Indonesia and Malaysia firms might have found an investment profitable even when its before-tax rate of return was negative. It is worth noticing that the country that has been most successful in the use of its capital—Korea—is also that with the lowest subsidies. The cost of capital for private investors in this country was much higher than in the other countries; this factor is likely to have brought about a better allocation of private investment and a better use of borrowed funds.

Table 8.Cost of Capital at Zero or Actual Rates of Inflation(In percent)
CountryAt Zero Rate of InflationAt Actual Rate of Inflation
Ongoing firmsPioneer firms
India3.593.342.53
Indonesia2.40-1.802.60
Korea2.917.826.60
Malaysia1.50-0.300.85
Pakistan1.924.534.53
Philippines1.900.880.98
Thailand2.001.601.11
Source: Liam Ebrill, “Taxes and the Cost of Capital: Some Estimates for Developing Countries” (mimeo: March 5, 1984), Table 18.
Source: Liam Ebrill, “Taxes and the Cost of Capital: Some Estimates for Developing Countries” (mimeo: March 5, 1984), Table 18.

Once again we find that the role of the public sector has been an ambiguous one. On one hand it has contributed to a high level of capital accumulation and thus to a high rate of growth; on the other it has led to inefficient use of costly resources. The long-run implications of these policies for the ability of these countries to service their foreign debt without encountering difficulties are, thus, less good than they would have been if the capital had been productively invested and used at full capacity.

Before closing this section it should be mentioned that several of these countries (India, Indonesia, Philippines, Thailand) have recently carried out major tax reforms or are in the process of doing so. These reforms should increase the elasticity of the tax system and should reduce the role of these incentives in misallocating investment and in inducing firms to overinvest.

IV. The Role of the Public Sector in Redistributing Income and in Stabilizing the Economies

I cannot deal in any systematic fashion with the role of the public sector in redistributing income and in stabilizing the economies. I shall thus limit myself to a few general observations.

1. Redistribution

Table 9 provides available statistics on income distribution for six of these countries. Notice that these statistics refer to different years. Korea has the most even distribution although it should be noticed that the data for this country refer only to the urban population so that they are not strictly comparable with the others. In Korea the highest 10 percent had average incomes only 2.5 times the country’s average, while the bottom 20 percent had incomes about 40 percent of the country’s average. In the Philippines, on the other hand, the average income of the highest 10 percent was more than four times the country’s average, while the average income of the bottom 20 percent was only about one fifth of the country’s average.

Table 9.Income Distribution Statistics for Selected Countries
YearSourceCoverage of PopulationConceptQuartile DistributionHighest 10%
12345
Bangladesh1976-77Survey100%Household6.210.915.021.046.932.0
India1975-76Survey99%Family5.09.614.120.950.434.9
Philippines1971Survey100%Family3.57.812.419.856.540.9
Korea1981SurveyUrbanHousehold8.013.017.022.439.624.5
Sri Lanka1980-81Survey100%Household5.910.314.119.949.835.2
Thailand1975-76Survey100%Household5.18.211.717.957.142.8
Source: United Nations: National Accounts Statistics: Compendium of Income Distribution Statistics, Statistical Papers, Series M, No. 79 (United Nations: New York, 1985).
Source: United Nations: National Accounts Statistics: Compendium of Income Distribution Statistics, Statistical Papers, Series M, No. 79 (United Nations: New York, 1985).

In most of these countries the redistribution of income has been an objective that has received considerable attention on the part of policymakers. Little attempt, however, has been made to use the tax system to promote this objective. For example, the role of personal income taxes and of taxes on wealth and property has been limited. In general these governments have preferred to pursue the objective of redistribution through the expenditure side of the budget and through the regulatory route. To their credit they have paid much more attention to the agricultural sectors than has been the case in other developing countries.

One problem with the instruments used (rationing of basic products, subsidies to consumers, subsidies to producers, controlled prices of public utilities, public employment, credit subsidies, etc.) is that it is always very difficult to target them efficiently. Therefore, in many cases the main beneficiaries have not been the very poor but individuals from higher deciles of the income distribution. In fact this seems to be the main thread that one finds in discussions of these programs. 48/ The truly poor have benefited much less than they could have with better targeted programs. But, of course, to achieve this better targeting is always very difficult or very expensive administratively.

2. Stabilization

There have been implications for the stabilization of these economies associated with both the large public spending on capital projects and the expenditure and other policies associated with the attempts to redistribute income. One implication already discussed has been the accumulation of public debt, especially the one owed to foreigners. 49/ The other has been the creation of fiscal imbalances that have in some cases grown very large and have forced the countries to rely on foreign borrowing in the absence of sufficient noninflationary domestic sources. In some cases (India) domestic savings have been mobilized to finance, in a noninflationary manner and without excessive recourse to foreign sources, relatively large fiscal deficits. This mobilization of a large share of domestic saving to finance the fiscal deficit raises the inevitable questions of whether private sector’s activities are crowded out. In the absence of the action by the government, would domestic saving have been the same? And would domestic saving have been invested in higher yielding private sector’s investments? The response to the first of these two question depends in part on whether or not private saving rises to accommodate the larger fiscal deficit. That rise could either be a consequence of the so-called Ricardian equivalence, or of the fact that the government may have created a possibility for many individuals to get a good rate of return to their financial saving. In India, for example, the opening of branches of banks in villages may have facilitated financial saving for individuals who might not have saved otherwise. The answer to the second question depends largely on the relative rate of return to private as compared with public investment. Table 10 provides some estimates of the fiscal balances of these countries. 50/ For many countries the recent levels of fiscal deficits may not be sustainable over the longer run. Once again Korea stands apart from the rest for having by far the lowest fiscal deficit and for having succeeded in reducing it over the years.

Table 10.Fiscal Balance(In percent of GDP)
19761977197819791980198119821983198419851986P
Bangladesh-8.7-11.2-9.2-7.5-7.3-8.2
India-4.6-4.2-5.9-6.0-6.5-5.9-7.6-7.1-8.3-10.5-9.3
Indonesian.a.n.a.n.a.1.22.1-2.3-4.9-2.60.9-2.9-5.6
Korea-1.4-1.7-1.3-1.4-3.2-4.7-4.4-1.6-1.4-1.0-1.6
Malaysia-6.8-7.1-6.2-3.2-6.2-14.9-16.9-10.6-6.6-5.9-10.9
Pakistan-9.5-8.6-8.0-8.8-6.2-5.2-5.3-7.1-6.0-7.7-7.8
Philippines-1.7-1.8-1.2-0.1-1.3-4.0-4.3-2.0-1.8-1.9-4.7
Sri Lanka-9.7-9.9-12.1-11.9-18.2-12.4-14.2-10.8-6.7-9.6-10.0
Thailand-3.2-3.1-3.4-3.4-4.7-3.2-5.8-4.1-3.7-5.1-4.6
Source: IMF, IBRD, and national sources. These figures are strictly not comparable across countries as they may refer to more or less comprehensive definitions of government.p = preliminary
Source: IMF, IBRD, and national sources. These figures are strictly not comparable across countries as they may refer to more or less comprehensive definitions of government.p = preliminary

V. Concluding Remarks on the Asian Countries

This paper has dealt with selected aspects of the role of the public sector in nine important market economies of Asia. In recent years, these were, with some exceptions, fairly successful economies when measured against standard criteria such as growth, inflation, and so forth. It may thus seem presumptious especially for an outsider to pass judgment on the actions of their governments. Furthermore, there have been important recent changes in various areas in some of them. For example, several countries have introduced major tax reforms, have begun to liberalize their financial markets, and have begun to privatize their publicly-held industrial enterprises. These changes are particularly desirable in these countries because, in spite of their diversity and levels of per capita income, all of them have dynamic private sectors and significant portions of their work force are highly skilled and market oriented. The idea that in these countries the government can do better than the private sector in providing entrepreneurial talent must be met with skepticism.

In spite of the important changes that have taken place, much remains to be done. At the risk of stating the obvious, I shall list, in general term, actions that, in my judgment, the governments of these countries could take to bring their role closer to an optimum.

1. As in all countries, it is important that governments provide stable laws and institutions. The “rules of the game” should be clear to all citizens. When these rules are clear, there is less scope for corruption and for their arbitrary application to benefit particular groups or individuals.

2. As economies develop there is progressively less need for intervention in, and regulations of, economic activities. Excessive intervention increases the cost of doing business and generates parallel or underground markets for goods, credit, imports, etc. Thus, a progressive and orderly dismantling of the many regulations that apply to investments, imports, credits, and so forth should take place.

The governments of these countries have been particularly active in regulating the allocation of credit. For example, government lending reached almost 5 percent of GDP in Korea in 1981 (before starting to fall), it reached almost 7 percent of GDP in Malaysia in 1982, and was 2.1 percent of GDP in Sri Lanka in 1984. Also, through its credit policies, incentive legislation, and regulation of investments, the government has played too large a role in determining the area where investment should go, and for sure it has played too large a role in determining the composition of imports and even the users of these imports. Quotas and high tariffs should give way to adjustments in the balance of payments that come from changes in the exchange rate and changes in the macroeconomic policies of the countries. Equally important, government controls on domestic prices should be reduced to the minimum.

3. Tax systems should be simplified in all of these countries. They should reduce, if not eliminate, the incentives that have been provided to particular investments. Broader bases with lower rates should be the objective.

4. The various factors that have brought about low returns to investment should be corrected. In particular, a concerted effort should be made to maintain the existing infrastructure in good condition and to utilize fully the existing capital structure. It is not desirable to add to capital while the existing capital is not fully or efficiently utilized. The privatization of public enterprises which are not in areas of natural monopolies should continue. If these changes bring about a lower capital accumulation, but a better use of existing capital, the countries will benefit from the change since the rapid capital accumulation that has taken place has brought about a rapid accumulation of foreign debt and increasing interest payments.

VI. General Lessons for the Current Debt Strategy

1. The current debt strategy

This paper has discussed the role of the public sector in nine market economies of developing Asia. Although some of these countries have run into financial difficulties in the 1980s, as a group they were relatively successful in pursuing the main objectives of economic policy (growth, low inflation, and so forth). However, it was shown that there were considerable costs associated with that success, costs that would become progressively more important with the passing of time in the absence of a change in policies. Using the experience of these countries as a background, I should now like to draw some general lessons for the current debt strategy for developing countries.

That strategy is postulated on the assumption that, in the absence of a political solution, the debt crisis will become more manageable if (a) countries can maintain high investment levels that would promote a high rate of growth; (b) if foreign creditors make this possible by extending additional loans; and (c) if the countries commit themselves to major structural reforms. In this strategy the maintenance of high investment rates seems to acquire a central role. It is recognized that structural reforms take time but, regardless of the speed at which these reforms are (or can be) phased in, it is assumed that a high investment rate must be maintained.

The major gauge of the external debt situation of a country is, perhaps, the ratio of foreign debt to gross domestic product, D/GDP. 51/ Success in dealing with the debt crisis must result, hopefully, in a declining or, at least, in a constant D/GDP ratio. Such a behavior of the debt/GDP ratio would be a necessary but not a sufficient condition since the decline in the ratio should be brought about by an increase in the denominator (GDP) rather than just by a fall in the numerator (D). Furthermore, a strategy that, in attempting to increase the rate of growth of GDP, increased D by an even larger percentage would not be conducive to a solution of the debt crisis. Yet, this is, implicitly, the strategy that is followed when foreign borrowing is maintained while the major economic reforms necessary to increase the efficiency of the economy are postponed or are carried out in a half-hearted manner.

The reforms should aim to achieve at least three interrelated objectives. First, they should make possible (a) a far better screening of investment projects than has been the case in many countries; (b) a far better utilization of the existing capital stock; and (c) a better allocation of the total resources available to the countries, including labor, capital, land, and so forth.

2. Investment and growth

In Part III of this paper it was argued that in many countries there are strong social, political, and other kinds of pressures that often bring about a project selection, and thus an allocation of the investment budget, that is far from optimal. The wrong projects are often chosen and they are built at excessive costs or with the wrong technology. This, obviously, reduces the rate of return on new capital which may have been obtained through expensive foreign borrowing. It is, thus, imperative that a good international debt strategy must be associated with very stringent criteria for project selection and execution. Somehow, and perhaps with the assistance of international institutions, countries should develop mechanisms that would, to the extent possible, insulate investment decisions from the pressures mentioned earlier. These mechanisms should ensure that only projects with expected (and properly measured) rates of return at least equal and hopefully higher than the marginal cost of borrowing are approved.

If a convincing case cannot be made that a project will generate a higher rate of return than the cost of capital, then there would be no reason to carry out the project. The traditional arguments often made to defend unprofitable investments—that they generate unquantifiable social returns in the form of better income distribution, employment, self-sufficiency, and so forth—should not be allowed to carry much weight, especially when the investments are financed with borrowed capital that must be serviced and repaid. When the public investments are financed by domestic taxes, perhaps an argument could be made on political grounds in favor of taking these social objectives into consideration in the choice of projects, especially when more direct and better-targeted redistributive measures are not feasible. In that case some investment would be seen as a substitute for direct redistributive measures.

While the problem of poor project selection is particularly relevant for public investment (including that of public enterprises), it exists also for private investment, but for different reasons. The reasons for poor project selection in the private sector, apart from the normal ones due to lack of information, incompetence, and so forth, are found (a) in tax laws that implicitly subsidize certain kinds of capital accumulation, especially in the presence of inflation; (b) in fragmented capital markets that, for some borrowers, may provide credit at lower than market costs; (c) in overvalued exchange rates that may artificially lower the cost of imported machineries and equipment for certain investments; and, finally, (d) in policies, such as import restrictions, that may give de facto a monopoly status to some activities. Thus, the efficiency of private investment can be increased mainly by structural reforms that bring domestic prices more in line with their true scarcity values. The argument often made—that shifting investment from the public to the private sector will automatically increase the rate of return to investment—while likely to have some validity, in the absence of major changes in the economy, is less strong than it is generally believed.

A rigorous process of investment selection that followed the above rule for public investment, if accompanied by reforms that remove the distortions to private investment, might, possibly, reduce the level of net investment in some countries. This reduction should not be a cause for concern but, on the contrary, should be welcomed since, over the longer run, countries can only gain if they do not have to service a debt that has financed unproductive activities and that has, thus, been associated with a negative net present value.

At least since the time when W.W. Rostow wrote his influential book, there has been a prevalent view that a minimum amount of investment is essential for growth. 52/ A recent expression of his view is the concept of the “investment core” used by the World Bank which purports to indicate the minimum investment level consistent with growth. While few would deny that there is a connection between investment and growth, especially over the longer run, and there is the circumstantial evidence that high-growth countries generally have a high investment rate, in practice the connection between investment and growth is tenuous at best. 53/ As Peter T. Bauer has put it “… it is clear from much and varied evidence that investment spending is not the primary, much less the decisive determinant of economic performance.” 54/ An important reason for this is likely to be the inefficient way in which the investment budget is allocated. In conclusion, a reduction in the expenditure for those investments with low rates of return must be seen as good news.

3. The utilization of the existing capital stock

Assume that (a) the capital stock of a country has not been properly maintained and is not fully used, and (b) that the allocation of the investment budget is unduly influenced by political and other considerations. Then the country’s total output could increase considerably if its capital stock could be used more fully and more productively, even when the investment rate falls. One of the conclusions of this paper with respect to the nine countries analyzed has been that the preoccupation with new investment has, in a way, distracted the country from maintaining in good working condition the infrastructure already in place. Often funds are diverted to the building of new roads while the existing road network was allowed to deteriorate. New vehicles were acquired while many of those already on hand were unutilized because of needed repairs or lack of spare parts. The same is true for hospitals, power plants, schools, and other infrastructures. In some countries new power plants were built when the existing ones were working at only a small fraction of their full capacity.

This problem seems to be a universal one as various examples mentioned in the World Bank’s World Development Report of 1983 indicate. For example, discussing irrigation the WDR states:

“Public and private investments in irrigation in developing countries have increased dramatically over the past 20 years, reaching about $15 billion in 1980. But the returns are much below their potential: one recent estimate for South and Southeast Asia suggested that an additional 20 million tons of rice, enough to provide the minimum food requirements of 90 million people, could be produced every year with inexpensive improvements in water distributions.” (WDR, 1983, p. 45)

Discussing roads, the WDR writes:

“The worldwide road-building boom of the 1960s and 1970s threatens to become the road-maintenance crisis of the 1980s and 1990s. Over the past ten years, roads in many developing countries have been allowed to deteriorate beyond the point where normal maintenance could be effective… Funds budgeted for highways have been mostly absorbed in expanding rather than maintaining the network.” (Ibid, italics added)

Still citing that report:

“Because of inadequate maintenance budgets, public sector assets are often run down much faster than they would be if routine maintenance were correctly carried out. For example, in Brazil it is estimated that a significant proportion of the federal highway network built in the past ten years already needs major rehabilitation, while in Nigeria most of the roads built in the 1970s had to be rebuilt three to five years later.” (Ibid, pp. 45-46)

As mentioned in Part III of this paper these infrastructures that are allowed to deteriorate at a very rapid pace, and that are often not fully used, 55/ were at times largely financed by foreign borrowing. They represent the assets that the countries bought with the foreign debts. As the rate of return on those assets falls below the interest rates that must be paid on the debts, the countries inevitably run into difficulties. Thus, assuming that the rate of interest on the debt cannot be reduced, or that the debts cannot be cancelled, an improvement in the debt situation can only come if, somehow, a larger rate of return can be squeezed out of the existing capital stock. This will be possible only if the funds are redirected away from new investment and toward a concerted action to maintain and upgrade the existing capital stock. This rather obvious and innocent-sounding conclusion would require profound changes in the way countries operate, in the way political decisions are made, and in the way the international community looks at categories such as current and capital expenditures. Perhaps, for a while the international community should encourage some kind of moratorium (or at least a slowdown) in new projects while reforms are made that would raise the level of efficiency and the use of the stock of capital accumulated with past investments. The moratorium would eventually create bottlenecks which, as Albert Hirschmann argued a long time ago, would send the clearest signals of which new investments would in time be the most productive. 56/

4. Structural reforms

Two general lessons came out of our discussion of the nine Asian countries. First, the need to scrutinize much more rigorously than in the past new investment projects and even to look at them with suspicion. If there is any doubt about the high productivity of a new project, it would be wise not to proceed with it. Second, the need to consciously insure that the capital stock available is properly maintained and efficiently used. In the previous paragraph the need to shift resources from new projects (capital expenditure) to operation and maintenance (a current expenditure) has been emphasized. In the early part of this paper it was argued that operation and maintenance expenses often do not have any strong constituencies in the way that capital expenditure has. Thus the proposed shift in emphasis is not likely to happen automatically. It must be pushed by policymakers who keep the public interest predominantly in mind. In this shift they should be assisted by a shift in attitude on the part of the international community. However, that shift would be only part of the change that is necessary. As Kuznets so eloquently put it two decades ago:

“… the effectiveness of a given stock of resources, embodied in physical capital, in increasing total output is partly a matter of its uses in combination with other resources and partly a matter of availability of such other resources and of the organizational arrangements for bringing them together…. the very choice of particular forms of physical capital, of quality rather than quantity of capital formation, depends upon the existence of institutions that can assure the most effective flow of savings so that they will reach those foci in the productive economy in which additions to capital stock will yield the greatest contribution to long-run growth. Without such an organization some part of saving may be stagnant and lead to no capital formation… or some may be invested in ways that are far from optimal for economic growth.” 57/

This takes us to the issue of structural reforms. Reducing the amount of money wasted on unproductive capital expenditure and maintaining the existing capital stock in good working condition will be important steps in the right direction. However, these steps by themselves are not enough to make an economy achieve its full potential given its total resources (including labor, land, and capital). For this it is important that the factors of production move to the uses where they can produce the highest values, measured at world prices. If government policies have introduced obstacles that have prevented resources from going to the most productive uses (and from producing at their full potential) then those policies need to be changed. There are literally thousands of examples available of such distortions but I will mention only one, and not even an extreme one. In one important country with serious external difficulties, a high tax on the production of the crop in which the country had the highest comparative advantage (cotton) progressively induced a shift in the use of the land toward low-yield but untaxed subsistence crops. It should not have been a surprise that the country’s exports suffered.

Distortions abound in the use of credit, in the use of foreign exchange, in the use of labor, in the use of land, in the use of capital, and so forth. These are introduced by various controls, by taxes, and by other government instruments. Structural reform essentially means reducing, if not eliminating, these distortions. Major dividends are likely to be obtained from these reforms but, as the author has argued elsewhere, each distortion is tied to a vested interest. Thus, structural reforms are not likely to come easily, 58/ and, as the theory of the second-best should have made clear by now, removing piece-meal some of these distortions, while leaving others in place, is no guarantee that much good will come from the change. Thus, the sequencing of structural reforms is an area that requires careful attention.

Because of the theme of this paper, I have emphasized changes that could be made by the developing countries to help solve, or at least alleviate, the debt problem. This emphasis, however, should not be interpreted as implying that only the developing countries should contribute to the solution of the debt problem. There is much that the industrial countries can do to make that problem manageable. As a minimum they must keep their markets open to the exports of the developing countries; they must keep interest rates low and their economies growing through sensible economic policies; they must continue to provide financial assistance while insuring that it is productively used; they must encourage private creditors to show as much flexibility as possible; and they must continue supporting the functions of those international institutions that are at the forefront of the attempt to find a permanent and widely accepted solution to the debt problem.

Table A-1Foreign Debt(In millions of United States dollars)
197519761977197819791980198119821983198419851986
Bangladesh1,6111,9382,2942,7672,9553,7334,4905,0705,0995,8446,4977,538
India12,34513,41214,78915,62015,94817,67019,32820,97023,04924,87528,50731,292
Indonesia10,36312,62614,94015,72916,56818,24120,45225,85430,07031,62137,31642,028
Korea6,0309,54413,37215,44220,27126,63229,27435,90138,92340,61945,74742,905
Malaysia1,9482,7103,3493,7624,7165,7308,25011,89916,27917,75419,15020,565
Pakistan5,9006,8757,4838,4369,0979,94310,92411,33412,48512,73112,83714,758
Philippines3,4114,6185,88810,96913,41917,53321,13024,75124,23724,57425,13723,041
Sri Lanka6027017961,0461,1431,5691,8992,3002,6412,8082,9533,746
Thailand1,6962,4193,3934,8716,6818,19310,35811,54812,93714,81216,36417,114
Total43,90655,84366,30478,64288,798109,244126,105149,627165,720175,638194,508202,987
Source: Various IMF, IBRD, and national sources.
Source: Various IMF, IBRD, and national sources.
Table A-2Interest Payments on Foreign Debt(In millions of United States dollars)
197519761977197819791980198119821983198419851986
Bangladesh35465761697296120119121138138
India299345412461487497450609904909778799
Indonesia4606918147871,1641,3321,6711,9181,9792,4932,5582,904
Korea2253395529311,1542,6093,5363,6323,0673,6883,5463,589
Malaysia811802042383044066418081,1041,2181,3291,465
Pakistan115138197262285329362421584528592661
Philippines1041582574406269751,3741,8501,7892,2381,9821,933
Sri Lanka293235424858121128162149171175
Thailand931061373075516809871,0069521,1081,1521,286
Total1,4412,0352,6653,5294,6886,9589,23810,49210,66012,45212,24612,950
Source: Same as Table A-1.
Source: Same as Table A-1.
Table A-3Effective Interest Rate 1/
197519761977197819791980198119821983198419851986
Bangladesh2.22.42.52.22.31.92.12.42.32.12.11.8
India2.42.62.83.03.12.82.32.93.93.72.72.6
Indonesia4.45.55.45.07.07.38.27.46.67.96.96.9
Korea3.73.64.16.05.79.812.110.17.99.17.88.4
Malaysia4.26.66.16.36.47.17.86.86.86.96.97.1
Pakistan2.12.22.63.13.13.33.33.74.74.14.64.5
Philippines3.03.44.44.04.75.66.57.57.49.17.98.4
Sri Lanka4.84.64.44.04.23.76.45.66.15.35.84.7
Thailand5.54.44.06.38.28.39.58.77.47.57.07.5
Total3.33.74.04.55.36.47.37.06.47.16.36.4
Source: Tables A-1 and A-2.

Effective interest rate defined as interest payment/total debt.

Source: Tables A-1 and A-2.

Effective interest rate defined as interest payment/total debt.

*

An earlier and shorter version of this paper was presented at the Fifth Asian Development Bank’s Development Round Table on “The Role of the Public Sector in Development,” Manila, July 6-10, 1987 and is being published by the Asian Development Review. Part VI dealing with general lessons for the current debt strategy is new. In preparing this paper the author has benefitted from valuable assistance and/or comments from Ke-young Chu, Luc De Wulf, Ved Gandhi, David Goldsbrough, Manuel Guitián, Peter Heller, Richard Hemming, Walter Mahler, Ahsan Mansur, A.F. Mohammed, P.R. Narvekar, Parthasarathi Shome, Susan Schadler, Mohammad Shadman, Nicholas Stern, Douglas Scott, and Wanda Tseng. They are not responsible for the views expressed or for any errors.

1/

For basic references to the normative approach see Richard Musgrave, The Theory of Public Finance (New York: McGraw-Hill, 1959); and A. B. Atkinson and J. E. Stiglitz, Lectures on Public Economics (New York: McGraw-Hill, 1980).

2/

For an earlier attempt by this author to outline a positive theory of public sector behavior, see Vito Tanzi, “Toward a Positive Theory of Public Sector Behavior: An Interpretation of Some Italian Contributions” (FAD Working Paper, 1980).

3/

Of course, given their diversity, one must be aware of the fact that there was considerable diversity of performance among them.

4/

Unless the objectives conflict with each other. Conflicting objectives may imply a lesser role for the government than nonconflicting objectives. I owe this point to Professor Nicholas Stern.

5/

Many decisions that are correct (or “rational”) on the basis of the information available at the time they are made may prove wrong if the environment changes. Our definition of a policy mistake would not extend to these cases.

6/

For example, a zoning regulation that permits a particular use of land for a given activity, but it denies the use of that land to other activities, is a form of hidden subsidy.

7/

For the limits to public spending, see Vito Tanzi, “Is There a Limit to the Size of Fiscal Deficits in Developing Countries?” in Public Finance and Public Debt, edited by Bernard Herber, Proceedings of the 40th Congress of the International Institute of Public Finance, Innsbruck, 1984 (Detroit: Wayne State University Press, 1986), pp. 139-52; for the limits to taxation, see Richard Goode, Government Finance in Developing Countries (Washington, D.C.: The Brookings Institution, 1984), pp. 95-100.

8/

A more political discussion would also emphasize the political costs of regulations.

9/

The World Bank’s World Development Report of 1983 has estimated indices of price distortions for many developing countries, including several from our sample.

10/

It will be noted that, except for the points where intervention is optimal, for each attainable net benefit, say OR, there will be two points on the curves at which that net benefit is attained (D and F). These two points will be consistent with a higher (F) or lower (D) degree of intervention. As the net economic benefits at these two points are the same, one cannot choose between these two levels of intervention on purely economic grounds. The choice between, say, D and F must be made on political grounds. Those who assign a positive weight to economic freedom would prefer option D over option F. But the choice would be a political rather than an economic one.

11/

The need for intervention is likely to depend on the efficiency of private markets in achieving given objectives.

12/

In this theoretical discussion, we assume that there is a social welfare function that indicates how the public interest is affected by public sector intervention.

13/

Of course, when the level of intervention is less than the optimal, the welfare of the country will also be less than possible.

14/

Of course the fundamental political objective of any government is the maintenance of law and order enforced by an honest executive and judiciary.

15/

This statement should not be understood to imply that the Philippines should have increased the level of its public spending regardless of any other consideration. It only means that the economic performance of that country would have improved if it had raised its productive public expenditure by a given amount while financing it by, say, taxes with the lowest distortive effects. The reverse might be true for Malaysia, which could have reduced the lowest priority spending while reducing its use of the most expensive borrowing.

16/

In this connection a more meaningful distinction would be that between real expenditure and transfers.

17/

In some countries developmental expenditure includes some development lending and some current expenditure. Also in some countries, no distinction is made between developmental expenditure and public investment.

18/

See Mario Blejer and Mohsin Khan, “Private Investment in Developing Countries,” Finance and Development, June 1984, for the relationship between the level of investment and growth. See also Simon Kuznets, Toward a Theory of Economic Growth (New York: Norton and Company Inc., 1968). The issue of the relationship between investment and growth is taken up below.

19/

For a review of the normative theory of public sector intervention, see Musgrave, 1959, op. cit., or Atkinson and Stiglitz, 1980, op. cit. Basically, this theory would justify government investment in public goods or in activities with important externalities.

20/

One would expect that the higher the cost of financing, the lower the level of investment, and the more biased the investment budget would be toward projects with a shorter time horizon.

21/

See Mahmood Ali Ayub and Sven Olaf Hegstad, Public Industrial Enterprises, Determinants of Performance (Washington, The World Bank, 1986), Annex I, pp. 56-60.

22/

In some cases these investments were justified on the basis of social cost-benefit analysis that assigned heavy weights to the creation of employment or the redistribution of income.

23/

On the role of public enterprises in developing countries see Robert H. Floyd, Clive S. Gray, and R. P. Short, Public Enterprise in Mixed Economies (Washington, D.C.: IMF, 1984). One could argue that the social, as distinguished from the financial, rates of return may not have been as low since these enterprises were often required to engage in activites aimed at achieving other objectives (employment, income redistribution, etc.).

24/

Please note that the estimates of the rate of return to public enterprises reported in this section are not strictly comparable.

25/

Ayub and Hegstad, op. cit., p. 15.

26/

See Ayub and Hegstad, op. cit., pp. 12-13.

27/

For a discussion of the relative efficiency of public and private enterprises, see Bela Balassa, Public Enterprise in Developing Countries: Issues of Privatization, DRD Discussion Paper (World Bank, May 1987), pp. 9-14; and Ayub and Hegstad, op. cit., pp. 12-17. Both of these studies suggest that public enterprises have been less efficient than private enterprises.

28/

Even in this case prudence is required. Since the capital market is not perfect, to avoid financial difficulties, the country must pay attention to the time profile of the expected returns to the investment and to the time profile of the expected future payments.

29/

At times the borrowing was done by the central government and the resources were transferred to the enterprises as loans or grants. In other cases the government guaranteed the loans.

30/

This comparison minimizes the problem since the rates of return to enterprises were distorted by inflation more than the average costs of foreign borrowing.

31/

Some of this increase was accounted for by the fall in commodity prices and some by changes in the real exchange rates. On the other hand, interest rates generally fell over the period shown by the table.

32/

An argument can be made that the financial returns from infrastructure projects take a long time to materialize. The trouble with this argument is that the time may just be too long and the benefits must be discounted to the present at high rates of discount.

33/

Total public debt—that is, external plus domestic public sector debt—has been rising rapidly in some of these countries (India, Malaysia, Pakistan, Sri Lanka, Thailand) and it has reached very high levels in a few of them.

34/

This is a rather general problem in developing countries. There are too many examples of hospitals or schools that have been built but that have remained unutilized for long periods because of lack of resources for operation and maintenance. And there are examples of new roads being built while old roads are not properly maintained. Furthermore, often additional capacity was being built while existing capacity was not being fully utilized. In many cases such infrastructure was built with borrowed external funds.

35/

Countries have been generally praised when a larger share of their budget is allocated to “capital” rather than to “current” expenditure.

36/

These openings are in all countries often accompanied by well-advertised public ceremonies which bring political benefits to those involved. On the other hand, the routine repairing of roads or the provision of basic materials that make hospitals or schools function adequately rarely attract any attention.

37/

In these cases, the contracting firms (domestic or foreign) are likely to simply inflate the costs of the projects.

38/

Often plants are not fully utilized because no provision has been made for inevitable breakdowns. Vehicles remain idle because of lack of spare parts; hospitals remain closed because money for nurses or medicines is not available, and so on. These are general problems in developing countries but they exist in most countries.

39/

All this means that the investment figures overstate net investment since depreciation is very high.

40/

See Peter S. Heller and Joan E. Aghevli, “The Recurrent Cost Problem: An International Overview” in Recurrent Costs and Agricultural Development, edited by John Howell (London: Overseas Development Institute, 1985) pp. 22-49.

41/

Unlike equity, debt financing comes associated with a future servicing that is not reduced by the poor performance of the investment for which the capital was used.

42/

The sources of Table 7 are different from those for Tables 5 and 6.

43/

The buoyancy is the ratio of the actual percentage change in tax revenue to the percentage change in GDP. This ratio reflects the automatic response of revenue to changes in GDP as well as changes due to discretionary actions by the government.

44/

The elasticity would measure the responsiveness of tax revenue to changes in GDP in the absence of discretionary (policy) changes.

45/

The net flow of resources is the difference between foreign borrowing and interest payment and amortization of foreign debt.

46/

In other words, either government spending or private spending must go down.

47/

See Nils J. Agell, “Subsidy to Capital Through Tax Incentives in the Asian Countries” (mimeo, 1982), p. 32.

48/

See, for example, Eugenio Namor, “issues in the Targeting of Food Subsidies for the Poor: A Survey of the Literature,” IMF Working Paper 87/75 (October 28, 1987).

49/

In some of these countries total debt (domestic plus foreign) has reached very high levels.

50/

Again, these data are not strictly comparable since they may refer to different definitions of the public sector or of the fiscal deficit. They should be used only to observe trends over time.

51/

A close competitor would be the debt-service ratio.

52/

W.W. Rostow, The Stages of Economic Growth (Cambridge: Cambridge University Press, 1960).

53/

For a good discussion of the relationship between growth and investment, see Dennis Anderson, Economic Growth and the Returns to Investment, World Bank Discussion Papers, 12 (June 1987). See also Simon Kuznets, Toward a Theory of Economic Growth (New York: W.W. Norton and Company, Inc., 1965).

54/

Peter T. Bauer, Reality and Rhetoric: Studies in the Economics of Development (Cambridge, Massachusetts: Harvard University Press, 1981).

55/

As the WDR stated “… use of plants and equipment is often extremely low, sometimes only a quarter or a third of the rates achieved by the best maintenance organizations… The lack of spare parts and fuel is often to blame for poor plan utilization” (Ibid, p. 45).

56/

See Albert O. Hirschmann, The Strategy of Economic Development (New Haven: Yale University Press, 1958). See especially Chapters 4 and 5.

57/

Kuznets, op. cit., pp. 37-38.

58/

See Vito Tanzi, “Fiscal Policy, Growth, and Design of Stabilization Programs,” in External Debt, Savings, and Growth in Latin America, ed. by Ana María Mortirena-Mantel (International Monetary Fund and Instituto Torcuato di Tella, 1987).

Other Resources Citing This Publication