Pakistan's debt problem: changing nature and growing gravity.

Hasan, Parvez

INTRODUCTION

It has been evident for some time that Pakistan's debt burden is extremely onerous. The danger of external debt default first emerged in 1996 towards the end of the second Benazir government. Following the nuclear explosions by first India and then Pakistan and the subsequent imposition of economic sanctions by the Western countries in mid-1998, Pakistan froze the foreign currency deposits, a major source of balance of payments financing in recent years, and went into a technical default on external debt. Following a fresh agreement with the IMF in January 1999, Paris and London Clubs provided substantial debt relief in the form of rescheduling of debt payments due in 1998-99, 1999-2000 and the first half of 2000-1. Despite debt relief, the burden of external debt remains extremely heavy and the danger of default has not disappeared. In any case, the access to international financial markets has been greatly curtailed, if not eliminated, especially because The Paris Club has applied the 'comparability of treatment' to claims of private sector investors. On the domestic side, the heavy burden of servicing public debt has made the much needed fiscal adjustment both difficult and disorderly. The rise in interest payments from 2.2 percent of GDP in 1979-80 to 4.9 percent in 1988-89 and to the peak of 7.3 percent in 1998-99 made reductions in fiscal deficit hard to achieve. As interest payments now account for over 45 percent of government revenues, the fiscal deficit reduction has come mainly at the cost of development spending. Clearly the debt overhang is a major factor in the decline in the investment rate to 15 percent of GDP in 1998-99 and 1999-2000, the lowest level in more than two decades. Unless the debt burden can be brought down to more manageable levels, macro-economic management will remain problematical and growth prospects will remain clouded.

Despite the central importance of understanding our debt problem and developing a strategy to cope with it, there has not been much systematic work (1) on debt issues i.e. the nature of our debt problem, the root causes of the debt build up, and the consequences of debt overhang for the economy. Nor is there a debate on the options, admittedly rather limited, to reduce the debt burden in the context of a medium term economic framework which also ensures a recovery in the growth rate. The present government needs both reasonably precise targets for debt reduction and a clear understanding of the policy elements or variables which will assist it in the attainment of these targets. If the previous government had debt reduction goals and a strategy to achieve them, it did not share it with the public. Meanwhile, GDP growth has fallen to 3.7 percent per annum during 1994-99, the lowest rate for any five year period since the 1950s, and the consequences of the economic slowdown are being widely reflected in reduced industrial profitability, increased losses of state enterprises, slow growth in government revenues, increased incidence of poverty and last but not least a deteriorating employment situation.

This paper is an attempt to partially fill this gap in analysis: it focuses on the extremely serious nature of our debt problem, outlines the variables that explain the build up of past debt, explores economic policy implications of the debt overhang and discusses debt reduction goals and scenarios.

An important first step is to define the problem. There is often confusion between the problem of external debt and the problem of public or government debt. The paper argues that it is helpful analytically if one views Pakistan as having not one but in fact two debt problems. High levels of public and foreign debt are two quite distinct, though in Pakistan's case, closely related aspects of the debt issue. The focus must be on both aspects because it is not the debt per se that matters but the ability to service it. The servicing of public debt which is a charge on the budget poses different kind of issues than the servicing of external debt (including public and publicly guaranteed debt) which imposes a burden on the balance of payments because interest payments and repayment obligations become a first charge on future foreign exchange earnings. The expected growth of budgetary revenues is of central importance for handling public debt while export earnings growth is often critical for keeping the external debt burden under control. The interest rate on borrowing is an important variable for all kinds of debt. But there is one critical difference at least between domestic portion of public debt and external debt (both public and private). Countries have relatively little control on the nominal or the real interest rate charged to them on borrowing abroad. But domestic interest rates, both nominal and real, can up to a point be manipulated by monetary authorities. Thus, national debt issues need to be analysed from the differing perspectives of fiscal and monetary policies and balance of payments management.

There are of course strong overlaps between public and external debt in Pakistan. Most of the external debt is either public or publicly guaranteed debt. Pakistan private sector does not have large liabilities, either short term or long term, to foreigners other than those guaranteed by the government. (The important exceptions are the loans from the IFC of the World Bank Group to the private sector which are not guaranteed by the government.) Also there are strong policy inter connections. The level of fiscal deficits, depending on how they are financed, can influence the levels of current account deficit and foreign borrowings. Inflationary financing of budget deficits not only enlarges balance of payments deficits but also exerts a downward pressure on the external value of the currency. The exchange rate changes, in turn, can significantly alter the burden of public debt; devaluation increases the amount of external debt expressed in local currency and increases the domestic costs of servicing foreign debt. While the policy overlaps must be examined especially in relation to options for future debt management, there is merit conceptually in analysing the past growth and present situation of external and public debt separately.

Section I of this paper discusses the various indicators which are normally used to measure the debt burden, both public debt and external debt, analyses the reasons why debt problems arise, and presents the policy guidelines which are normally used to avoid these problems.

Section II analyses Pakistan's public debt problem and debt burden and discusses them both in terms of their historical evolution and comparative international experience, using the key debt indicators and economic variables highlighted in Section I. Section III discusses the level and trends in external debt, Paris and London Club rescheduling, their impact on Pakistan's debt obligations and the constraints the debt situation places on growth and macro economic management in the medium term. Section IV discusses the consequences of debt overhang and analyses the principal issues of public and external debt strategy including the need for better institutional arrangements for debt monitoring and policy guidelines to avoid debt problems in the future.

I. THE ART OF DEBT MANAGEMENT

Borrowing domestically or abroad is a normal, indeed necessary part, of economic activity. Financial intermediation between lenders and borrowers improves the effectiveness of resource allocation and improves growth prospects by giving credit access to (1) entrepreneurs who are willing to take risk in the process of trying out new ideas and generating high economic returns and (2) the government which is often required to undertake necessary capital spending for social and physical infrastructure development. The economic rationale of debt creation is that borrowers can earn a higher economic return than the cost of invested funds and that these economic returns can be translated into financial returns consistent with interest and repayment obligations.

Debt servicing problems in the private sector arise if the assumptions about the economic benefits of investments do not materialise and/or the financial flows do not match the maturity structure of debt. (2) Short term borrowing for longer-term investments can lead to serious cash flow problems even if investments will have high economic returns. The debt servicing obligations arising from public borrowing domestically or abroad cannot be so easily linked to economic and financial returns from projects and/or the impact of individual investments on foreign exchange earnings or savings. A lot of government infrastructure spending on irrigation, roads, schools, hospitals, research, does not have a direct financial return nor does it lead directly to positive balance of payments effects. This does not mean that the economic returns on the investments undertaken by the government with borrowed funds, either at home or abroad, are not relevant. Indeed, countries most often run into difficulties precisely because the borrowed funds are directed to wasteful or low economic return projects. The point simply is that the guidelines for prudent external or domestic borrowing by the government must be expressed in more macro terms.

Sound debt management, like good economic management in general, is more of an art than a science. The first important, and more judgmental step, is to specify the norms for prudent levels of borrowing in terms of one or more indicators of debt burden. The second more technical but a fundamental step is to develop guidelines based on the relationship between key economic variables most notably the rate of interest, the growth rates of GDP, government revenues, foreign exchange earnings, and the initial levels of fiscal and current account balance of payments deficit (excluding interest payments). Once the norms of acceptable debt burden have been specified, the derivation of technical guidelines can be done mathematically. While there can be a lot of debate on what should be the acceptable levels of debt burden and what precise indicators should be used to measure debt burden, the basic rules which determine the relationship between the growth of debt and the key economic variables cannot be ignored.

In the following paragraphs, the range of measures that are used to monitor debt and their relative usefulness are discussed and the technical rules or guidelines which must be followed to keep debt within manageable limits are outlined, it being possible to develop specific guidelines for each debt burden indicator.

Debt Indicators

Debt burden indicators are of two types (a) stock measures which relate the value of a stock of outstanding debt to the annual level of key economic aggregates and (b) the flow measures which relate the value of annual debt service payments to the same economic aggregates.

The stock measurements of the external debt burdens are generally expressed in terms of ratios to GDP and ratio to annual foreign exchange earnings. Since these stock measures do not take into account the interest rate payable on debt and the maturity structure of the debt--very relevant factors influencing debt service payments--the better and more sophisticated measures give the present value of external debt both as a percentage of GDP and as a percentage of foreign exchange earnings. The use of the discount rate reduces the burden of real debt payments in the outer years and also takes account of the concessionality of the various interest rates. Another stock measure is the ratio of short term debt to total external debt. A high ratio of short term debt sends a danger signal because a number of the debt crises in Latin American and more recently in East Asian countries, notably Thailand and Korea, have been triggered by the inability to roll over short term debt (defined as debt with a maturity of one year or less) as it fell due. In the context of short-term debt or other short-term obligations, the level of foreign exchange reserves is an important indicator of the ability to withstand unexpected foreign exchange pressures. Indeed, many international lenders, notably banks, place an excessive faith in the level of reserves as an indicator of a country's ability to avoid debt problems.

The only widely used measure in Pakistan and elsewhere to judge the stock of public debt (including external debt) is its ratio to GDP. (3) Though it is common practice to measure the burden of public debt (as well as that of external debt) as a proportion of GDP, it makes more sense to use the yardstick of government revenues for monitoring the changes in public debt burden. After all the changes in GDP do not automatically translate into revenues particularly in developing countries like Pakistan where the taxation machinery is weak and the taxation systems are inelastic. It is the expected growth in revenues which provides the capacity to service future debt payments.

But even the best stock measures, total public debt as a proportion of revenues (4) and the present value of external debt as a percentage of exports of goods and services must be supplemented by the appropriate flow measures. In the case of external debt, the most frequently used measure is the total annual debt service as a percentage of exports of goods and services. Public debt payments consist of two parts; debt service payments to foreigners for the external debt owed by the government and interest and principal repayments to residents for the domestic debt held by the public sector. There is some merit in considering only interest payments as domestic debt service burden. The rationale of this asymmetry is that since monetary authorities in the country have control over their own currency, the rolling over of domestic debt can be taken for granted, whereas, short of default, the governments have little control on the repayments obligations to foreigners. (5) But irrespective of the definition of debt service used, the measurement of annual revenues pre-empted by debt service whether domestic or foreign is necessary to gauge the burden of public debt.

The central questions in sound debt management relate to the factors, which influence the changes in debt burden, as measured by various debt indicators, over time. These questions can be studied with the help of technical rules which provide the theoretical underpinnings of the art of good debt management, though as mentioned above, an element of judgment about what constitutes a manageable debt burden in given circumstances will always remain.

Some Rules of Debt Management

Public debt accumulates when loans are used to finance (1) an excess of non interest government expenditures over revenues and capital receipts and (2) interest payments on existing debt. The nominal interest rate payments, however, exaggerate the burden of additions to debt. A portion of nominal debt is normally wiped out by inflation which reduces the burden of real debt and, therefore, a part of nominal interest payments in fact represents repayments of principal. In determining the limits on borrowing, therefore, the focus should be on the real interest rate (nominal interest rate minus the rate of inflation). The real rate of growth of debt is, thus, determined by the primary fiscal deficit (deficit before interest payments) as a proportion of debt and the average real interest rate. If the real rate of growth of debt exceeds the real growth rate of GDP, the debt to GDP ratio will begin to rise and if this excess persists for a long time the growth in debt burden can assume explosive proportions. If the primary deficit is zero, it can be mathematically demonstrated that the ratio of public debt to GDP will not rise as long as the average real interest rate on debt does not exceed the real rate of growth of GDP.

Since it is more appropriate to measure the burden of public debt as a percentage of government revenues rather than as a proportion of GDP, the more important rule about limiting public debt growth must be expressed in relation to revenue growth. If the primary deficit is zero, the ratio of public debt to annual revenues or the ratio of interest payments to annual revenues will not grow as long as the rate of interest does not exceed the rate of growth of revenues. If there is a primary deficit, the total growth in debt (interest rate plus primary deficit expressed as a percentage of debt) must be below the growth of revenues if debt to revenues ratio or the ratio of interest payments to revenue is not to increase.

Similarly in case of external debt, if there is no non-interest current account balance of payments deficit the ratio of external debt or annual interest payments to foreign exchange earnings will grow only if the interest rate on external debt is higher than the growth rate of foreign exchange earnings. If there is a non-interest current account balance of payments deficit, the growth of external debt will be the sum of the interest rate and the size of the non-interest current account balance of payments as a proportion of debt. As long as the growth rate of debt does not exceed the growth rate of earnings, the debt burden as a proportion of foreign exchange earnings will not go up. (6)

These simple rules demonstrate the obviously critical role of the cost of borrowing and the level of deficits (both fiscal and balance of payments) before interest payments on the one hand, and the key indicators of debt service capacity i.e. the growth in foreign exchange earnings, in the case of external debt, and growth in government revenues in the case of public debt, on the other hand. In the case of external debt, in addition to the interest rate on borrowing, the debt maturity structure is also important because the repayments of medium and long term loans are a charge on foreign exchange receipts and it should not be assumed that the short term debt can always be rolled over easily. In highlighting the importance of export growth and expansion in government revenues for debt service capacity, these guidelines essentially point to the underlying need to ensure that investments financed with borrowed funds have good economic returns and that these returns can be translated into exports and/or government revenues. They have thus to be very much integrated as a part of the good overall macro economic management. The guidelines themselves cannot suggest whether the country can afford a rise in the debt burden. In developing countries starting with low levels of savings and investment as well as debt, it makes a great deal of sense to borrow abroad provided the resources are well directed. But a perpetually growing debt burden should send up alarm signals.

II. PAKISTAN'S PUBLIC DEBT BURDEN

The changes in public debt aggregates and the key economic variables influencing the changes in Pakistan's public debt burden over the last two decades are summarised in the Table 1 below. Table 2 traces the development of Pakistan's public debt burden by a variety of debt indicators. Table 3 gives the trend in nominal and real interest on domestic debt. Table 4 gives absolute amounts of Pakistan's public debt and highlights the steadily growing share of external debt.

There are several points that emerge from a systematic examination of the evolution of Pakistan's public debt problem in its historical and comparative international context. First, the debt problem has been in making for a long time. Second, by all indicators of debt burden, the debt problem has continued to grow notwithstanding some fiscal adjustment in the last few years. Third, the nature of public debt problem has changed significantly over the last decade, the debt is now driven largely by interest rate costs and the debt indicators are worsening because the key growth rates of GDP, revenues and exports have all declined sharply. Finally, Pakistan has a far more serious debt burden than almost any Asian country including India. I will turn now to the elaboration of these points both in historical and comparative international context.

Historical Context

In Pakistan the alarm signals about the rising burden of public debt should have gone up a long time ago. During the 11 years of Zia rule, 1977-88, the public debt grew six folds reflecting large and growing fiscal deficits. The debt grew by the average annual rate of 17.7 percent in nominal terms and nearly 10 percent in real terms during this period. The rate of growth of real debt was substantially higher than the growth rate of GDP and exceeded growth of government revenues. The main source of growth in real debt was the large primary deficit (see Table 1). But the cost of borrowing, though low on average, was also increasing steadily during 1977-88 as a large portion of domestic debt was raised through very costly borrowing from non-bank sources (notably saving schemes). As Table 2 shows, the ratio of debt to GDP increased from 57.5 percent in 1976-77 to 77.1 percent in 198788. Interest payments on debt in the budget increased from 13.2 percent of the revenues to 28.4 percent over the period.

The debt problem with which the democratic governments struggled, albeit unsuccessfully, during the last decade was to a considerable extent inherited from the Zia period. The debt burden has been made much worse, however, by the inability or unwillingness of elected leaders to reduce the fiscal deficit significantly, a slowing economy, and last but not least a marked falling off in growth in real revenues in the 1990s. It is not surprising that the debt indicators which relate debt or debt service to revenues have shown much greater deterioration in the 1990s than the indicator relating debt to GDP. While the ratio of public debt to GDP increased further from 82.6 percent in 1989-90 to over 100 percent in 1998-99, the ratio of debt to revenues increased from over 400 percent to 600 percent and the proportion of interest payments to revenues rose to well over 40 percent. (Table 2.) This happened despite the fact that the growth in real debt slowed down mainly because of the acceleration of inflation. The impact of inflation was to offset the substantial rise in nominal interest rates and to keep the real interest rates on government down. The average implied real interest rate on debt in the period up to 1996 was only moderately higher than in the late 1980s despite financial liberalisation. But inflation also dampened the growth in real revenues because the elasticity of government revenues to the price level changes is less than one. Still on balance but for higher inflation, which wiped out a large portion of the nominal debt burden, the debt indicators will have deteriorated even more.

Factors Behind Deepening of Debt Problem

What are the factors, which explain the persistence, indeed the deepening, of the debt problem notwithstanding sharp cutbacks in public spending and the relatively high inflation of the 1990s? (It is not widely realised that total public spending (excluding interest) actually declined over 1990-99, falling from 20.4 to 15.0 percent of GDP over the period, the biggest cut being absorbed by development spending which declined from 6.5 to 3.4 percent of GDP.) As mentioned above, the biggest element in the worsening of the key indices of debt burden was the slow down in the growth of revenues. Real revenues i.e. revenues adjusted for inflation which expanded 9 percent annually during 1977-88 increased only 4.5 percent during 1988-96 and showed zero growth during 1996-99. Since revenue growth slowed at a much faster pace than slowdown in the real rate of growth of debt, the debt to revenue indicators worsened.

To some extent, the sharply slower growth in revenues in the 1990s reflects the general slowdown in the economy. The slower GDP growth in turn reflects serious neglect of investment in human and physical capital in the past two decades, declining effectiveness of resource use in the public sector including the largely publicly controlled banking system, and deep structural problems in industry and exports. Specifically, the fact that since the mid-1980s, a part of the public sector borrowing has been for financing revenue deficits i.e. public consumption has contributed directly to the debt burden. Similarly, the spending on low economic priority projects like the motor way, people's works programmes, convention centre, has limited gains to the economy and contributed little to the ability to service debt.

However, as Table 1 shows, the slowdown in revenues has been greater than in the GDP growth: indeed in the period 1996-99 real revenues have stagnated. Three factors explain this. First, the major reductions in income tax, sales tax and customs duty rates in March 1997 amounted to a tax cut of about 1.5 percent of GDP and towered the tax base. (9) Secondly, the persisting governance problem has weakened the compact between the state and the citizens thus leading to a growing resistance to paying taxes. Thirdly, the increase in revenues in the 1980s relying heavily as it did on foreign trade taxes was not sustainable.

Role of Interest Rates

Till the mid-1990s, the average real cost of borrowing, though rising, has not been, on the whole, a major factor in increasing debt burden. The estimates of average real interest rate paid on public debt in Table 1 have been derived as a residual basis and must be taken only as giving broad orders of magnitude. Still, these estimates show that the average real interest rate which was only 2.1 percent during 1977-88 and 3.5 percent during 1988-96 increased sharply to 6.2 percent during 1996-99. These average numbers, however include both domestic debt and external debt and represent the combined effect of four very different variables, the nominal interest rate on domestic debt, the nominal interest rate on external debt, the domestic rate of inflation, and the real devaluation of the exchange rate which increases the stock of real external debt in the same way as domestic inflation reduces the stock of total debt. We, therefore, need to disentangle the influence of these rather disparate elements on the average real interest rate on Pakistan's public debt to understand why the average interest rate remained rather low till relatively recently and why it now seems to have moved to a much higher level.

The average nominal interest rate on domestic debt rose from 5.0 percent in 1979-80 to 9.9 percent in 1987-88 and has risen almost steadily in the 1990s to the peak of 14.5 percent in 1997-98. (Table 3.) Almost all of the increase in the interest rate in the Zia period was due to the heavy reliance on government borrowing from non-bank sources, mainly various saving schemes such as Khas deposits. Nearly 40 percent of the increase in government domestic debt during 1980-88 was financed from this source by offering very high interest rates and tax exemptions. In the mid 1980s, the interest rate paid on Khas deposits was as high as 14 percent per annum while inflation rate was a little over 5 percent, thus giving a safe after tax return of 9 percent annually. The high guaranteed real rate on government debt helped to mobilise the large level of worker remittances that were coming in and kept inflation low by reducing reliance on money creation for financing fiscal deficits but crowded out the private sector investment (10) and added greatly to the real debt burden. That the government started borrowing for current spending, and that public development spending began to decline in relative terms, starting in the mid-1980s (11) compounded the debt problem by adversely affecting the long term ability to service debt.

Till the late 1980s, however, the bulk of government borrowing was at less than market rates as all interest rates were administratively determined and the government debt was sold in the segmented markets. The large reserve requirements for the banks (5 percent cash liquidity requirement and 30 percent liquid asset requirement) forced them to buy low interest treasury paper. In 1989 as a part of the World Bank supported financial sector reform, (12) it was decided to move to full market based auction programme for government borrowing. This move has been criticised as being premature because it unrealistically assumed a quick reduction in the fiscal deficit. (13) Another view is that the move to market based borrowing merely made the real cost of public borrowing explicit and reduced the banking system subsidisation of public debt. (14) The interesting point is that the real cost of government domestic debt remained low or negative till 1996 notwithstanding the moves to more market based interest rate determination because of the acceleration in inflation. The sharp decline in inflation during the last two years has, however, had the impact of pushing real interest rates on domestic debt to unprecedented levels. As we discuss below, if real interest rates remain at this level, the resolution of the debt problem will become very difficult. But first we must examine the rather involved issue of the real interest rate on foreign debt.

The issue of interest payments on foreign debt has not received much attention at least in relation to the discussion of public debt. Nominal interest payments on foreign debt at Rs 29 billion in 1997-98 were only a fraction of the interest payments on domestic debt of Rs 173 billion. As a percentage of GDP, interest payments on foreign debt have rarely exceeded 1.3 percent. The average annual nominal interest rate on external debt till recently has not exceeded 4 percent because, generally speaking, Pakistan has avoided high cost commercial borrowing. But just as in domestic interest rates, the relevant rate is the real rate of interest on external debt after allowing for international inflation. International inflation already low in the 1980s came down to zero during 1995-99 (see Table 7). Thus the real interest rate on Pakistan's foreign debt has risen steadily. Furthermore, if there is real devaluation of the rupee i.e. a greater depreciation than warranted by the relative rates of domestic and international inflation, the burden of external debt increases because in terms of our analytical framework this rise in burden is counted as a rise in the real interest rate. In Pakistan, there was moderate real appreciation of the rupee during 1977-88, little real change during 1988-97 and significant real devaluation during the last two years. The latter has contributed to the sharp increase in overall real interest rate since 1996. Author's estimates indicate that over the long run of 1980-95, the real average interest on foreign debt, including the cost of modest currency depreciation, was over 3 percent per annum. In contrast, the real interest rate on domestic debt during this period was probably only marginally positive at 1-1.5 percent per annum. This tells a very different story from the nominal figures of interest payments. In the most recent period 1996-99 real interest rate on domestic debt reached the high level of 5 percent per annum (see Table 3). Even so the real interest rate on external debt including the impact of substantial real depreciation was higher. Since the foreign debt is now over 50 percent of public debt, the issue of interest rate on foreign debt has become even more important.

International Comparisons

How does Pakistan's public debt burden compare with other major Asian and industrial countries? The available data would suggest that Pakistan has a far more serious problem than even India which is the only large Asian country that has neglected fiscal adjustment and whose government finances are not in good shape. India's government debt to GDP ratio was only 68 percent in 1997-98. More importantly the ratio of Indian public debt to total government revenue was 370 percent. In the US which had a lax fiscal policy till about four or five years ago, the debt to GDP ratio did not go above 50 percent even at its peak and is currently around 40 percent. At the peak of the fiscal deficit in 1992, outstanding Federal debt in the US was 285 percent of the government revenues. This figure now is closer to 200 percent. In Italy, another country which ran very large fiscal deficits till 1996, public debt was only 400 percent of revenues at its peak in 1996.

III. DIMENSIONS OF PAKISTAN'S EXTERNAL DEBT PROBLEM

That Pakistan's foreign debt problem has become even more serious than the domestic debt problem was reflected in the near default and subsequent rescheduling of external debt. Now, after Paris and London Club rescheduling. Due to near stagnation or decline of exports during the past four years, the burden of debt in relation to exports of goods and services including remittances, a major variable determining the ability to service external debt, has grown sharply. In 1992-93, the ratio of outstanding debt to exports at 242 percent was only a little higher than a decade earlier (see Table 5). By 1998-99, this ratio had risen to 351 percent. The ratio of total debt service to exports of goods and services remained at or below 25 percent till the early 1990s. But this ratio had jumped to the high level of 35 percent by 1996-97 and would have been even higher in 1998-99 if a substantial part of the interest and principal payments had not been rescheduled.

It may seem ironic that the latest comparative international external data from the World Bank, summarised in Table 6, does not bring out the full urgency of Pakistan's debt problem. This table gives the most recent figures (relating to year 1997) available in respect of the various external debt indicators for the largest developing country debtors. These debt indicators including total external debt, present value of external debt as a percentage of GDP and exports of goods and services, total annual debt service as a percentage of exports of goods and services. percentage of public and publicly guaranteed debt service as percentage of central government revenue, and the share of short-term debt in total debt.

In the World Bank classification of indebted countries, Pakistan is termed as a moderately indebted (M) country along with countries like India, Indonesia, The Philippines, Malaysia, Mexico, Thailand, and Turkey whereas Argentina and Brazil are among the severely indebted countries (S), and China, Egypt, Korea belong to the group of less indebted countries. The figures in Table 6 do suggest that the present value of Pakistan's external debt. (203 percent of exports in 1997) and the ratio of total debt service (35.2 percent of exports) are substantially higher than many other moderately indebted countries notably India, Indonesia, Malaysia, Thailand and Turkey. Thus, Pakistan appeared a more of a borderline case even before debt rescheduling which may lead to a downgrade.

There are some problems with the data, however. The 1997 World Bank figures apparently do not fully reflect the short-term debt. But in a more basic sense, the debt indicators in Tables 5 and 6 do not capture the real weaknesses of Pakistan's foreign finances in recent years, very large financial obligations arising out of the extensive use of resident and non-resident foreign currency deposits, large recourse to fixed obligation foreign investment in power to finance energy investments and sizable inflows of portfolio foreign investment which can be volatile. Pakistan's debt crisis was essentially triggered by the unsustainability of the level of the current account balance of payments deficits and the pattern of their financing. The normal debt indicators did not signal a problem because debt financing, as defined, was not particularly excessive especially when allowance is made for the fact that the export stagnation witnessed was not anticipated.

During the eight years 1991-98, Pakistan ran current account balance of payments deficits (before accruals of Resident Foreign Currency Deposits, RFCDs) totaling over $28 billion or an average of about 5.5 percent of GDP. This level of deficit is not sustainable for a decade even with a rapid expansion of exports. In Pakistan the growth of exports and remittances had in fact slowed down markedly in the first half of the 1990s and then stagnated. The alarm bells on the external debt did not ring partly because till 1996 only about half of the .balance of payments financing needed took the form of increase in external debt, the rest being financed form of accruals to resident and non-resident foreign currency deposits and direct, portfolio and IPPs related foreign investment. Net foreign investment inflows in the five years 1993-98 alone amounted to $6.5 billion. Foreign currency deposits increased by $10 billion during 1991-98. The consequences of the large exceptional financing notably RFCDs and foreign investments in the energy sector with guaranteed offtake and guaranteed price for electricity sales on the long term balance of payments situation were apparently not carefully considered either by the Pakistan government or the World Bank and IMF. (16) In general, Pakistan needed external adjustment as much as it needed the fiscal adjustment which has been so much the focus of the IMF agreements for more than a decade. But until rather recently, the balance of payments targets suggested by IMF were much less stringent than the fiscal targets. Indeed the 1997 IMF agreement considered a current account balance of payments deficit of roughly 6 percent of GDP (before accrual to RFCD) quite feasible for 199798. The government, the IMF and the World Bank were too sanguine about the prospects of large additions to resident foreign currency deposits though it had been evident for some time that a major factor in their increase was the large implicit subsidy provided by the State Bank of Pakistan through foreign exchange risk cover at a rate much below the expected depreciation of the rupee. The stagnation in export earnings after 1995 should have also raised concerns about the desirability of large further increases in external obligations. Specifically, the large repayment obligations related to IPPs should have been explored in the context of the medium term balance of payments before making irrevocable commitments. That there was a policy failure on many fronts is evident from the fact that not only was the reliance on external flows excessive but also these flows were used to finance consumption rather than investment. In relation to GDP much larger current account balance of payments deficits have been run in the 1990s compared to the 1980s, nonetheless the investment rate has actually tended to decline in the 1990s.

The lessons from Pakistan's experience with external debt are rather the obvious ones: (a) debt problems can arise also if non debt flows are either unsustainable or too costly (b) debt problems cannot be separated from broader issues of economic strategy and management.

Full impact of the payments to IPPs on the balance of payments has yet to be felt because there is dispute about several contracts and because all projects have not been completed. Notwithstanding this, the investment income payments (including interest payments) had risen from $1.33 billion in 1991-92 to $2.57 billion in 1997-98.

Following the methodology outlined in Section I and used to analyse the problem of public debt, we have estimated the "rates of growth of real external debt and the real cost of external borrowing and contrasted them with the key variables influencing debt burden.

These are presented in Table 7 below: Table 7 Growth and Cost of External Debt (Percent Per Annum) Year 1980-90 1990-95 1995-99 Nominal External Debt Growth 7.7 7.8 3.7 International Inflation 1.9 0.9 0.0 Real External Debt Growth 5.7 6.8 3.7 GDP Growth 6.7 5.7 3.3 Real Export Growth 3.7 6.2 -3.0 Average Nominal Interest Rate 3.9 3.7 4.0 Average Real Interest Rate 2.0 2.8 4.0

The above table indicates that over the last two decades the growth in real external debt was on average around 5.5 percent per annum while the real annual cost of debt while steadily increasing was on average less than 3 percent. The borrowing on this scale and at this relatively low cost should not have normally given rise to debt problems. But as mentioned above, in recent years, the debt flows represented less than half of the total foreign exchange liabilities incurred. Moreover, a large part of these liabilities were either short-term and / or high cost. Furthermore, the long term growth rate of foreign exchange earnings including remittances was less than 4 percent--below the rate of growth of external debt alone--and both GDP and export growth trends have worsened in recent years. As the above table brings out, the relative deterioration in the external debt situation during 1996-99 was pronounced. During this period the real interest rate on external debt went up while real growth in foreign exchange earnings turned negative. Behind these economic aggregates, the real problem has been poor economic management, an over reliance on external resources especially during the last decade, a neglect of domestic savings, less than effective use of borrowed resources and, in extreme cases, borrowing for sustaining consumption rather than investment. Given the enormous current account balance of payments deficits and the very fragile pattern of their financing through resident foreign currency deposits, there was certain inevitability about Pakistan's foreign exchange crisis. The imposition of economic sanctions merely hastened it.

Debt Rescheduling and Relief

Prior to debt relief, debt service payments of $8.2 billion (including short-term loans) were due in 1998-99 with another $8.5 billion due in 1999-2000. These repayments included the maturing institutional non-resident foreign currency deposits of $1.3 and $1.4 billion respectively during 1998-99 and 1999-2000. With identified capital flows of only around $4.0 billion, the disappearance of the additions to resident foreign currency deposits which had provided $1.5 billion in 1997-98 alone, and the unwillingness of the non-resident institutional holders of foreign currency deposits, large arrears built up by the end of 1998 indicating a situation of technical default. It is against this background that Pakistan had to approach the Paris and London Clubs for debt relief and debt rescheduling. Under the Paris Club agreement reached in January 1999, principal and interest payments due on public and publicly guaranteed debt up to the end of calendar year 2000 were agreed to be rescheduled with ODA (soft loam) being rescheduled over 20 years with 10 years grace while other loans such as export credits from bilateral donors were rescheduled over 18 years with 3 years grace. Following the Pads Club agreement, the commercial lenders, principally banks, agreed to reschedule medium term loans and roll over a large portion of short-term loans including institutional foreign currency deposits. The total relief likely to be provided by Pads Club members through bilateral agreements is now estimated at over $3.0 billion with an equivalent amount that will be available from London Club over the three years 1998-2001. Since most of the debt from bilateral sources is of concessional nature, the terms of its rescheduling by the Paris Club, 10 year grace and 20 year repayment, amounted to substantial reduction, probably $2 billion, in the real value of the debt.

But even with a substantial debt relief and some virtual debt reduction, Pakistan's balance of payments position remains precarious and the burden of its total foreign exchange obligations extremely high. Even after debt rescheduling, Pakistan is required to make debt service payments, including to the IMF, of over $6 billion (including roll over of short term capital) in each of the two years, 1999-2000 and 2000-1. This contrasts with total foreign exchange earnings of $1112 billion currently. The normal medium and long term flows in recent years have averaged a little over $4 billion annually including the rather buoyant portfolio and other foreign investment (notably in power projects) and are unlikely to exceed this level in the near future. Indeed if no agreement is achieved with the IMF and the World Bank on outstanding issues, even this level of gross flows could be in jeopardy. But even in the best of circumstances some additional exceptional financing will be needed just to meet normal debt service payments. Because there is urgent need to build up foreign exchange reserves (both in absolute terms and as a proportion of short-term liabilities), the non-interest current account balance of payments must generate surpluses in the foreseeable future. This also implies that the contribution of net income and net transfers from abroad will, at best, be only marginally positive. Thus the level of investment will be almost entirely dependent on the level of domestic savings in sharp contrast to the past when net income from abroad was positive, supplemented domestic savings, and in addition there was an external resource deficit (balance of goods and non-factor services also termed as resource gap was negative). Putting another way, domestic savings financed only about 75 percent of investment during 1990-98, the rest was financed by the external resource deficit. In the near future, domestic savings will have to provide most of the financing for investment because the burden of servicing the debt and foreign investments will keep the net income from abroad marginally negative and the financing constraints will not permit Pakistan to run any significant external resource deficits. Self reliance is now being forced on Pakistan by the serious overhang of external debt and other foreign obligations.

IV. CONSEQUENCES OF DEBT OVERHANG AND ISSUES OF FUTURE DEBT STRATEGY

In the near term, say over the next three to four years, Pakistan will have to live with the macro-economic consequences of the heavy debt. It will mean abiding by the painful conditionality of the World Bank and IMF and managing the economy with limited growth in investment and imports. The foreign exchange constraint will make it difficult to stage a quick recovery in economic growth, urgently needed to create jobs and reverse the trend towards increase in the incidence of poverty. Faced with the prospect of painful adjustment, voices are likely to be raised advocating debt default. Because Pakistan is, for the first time, being forced to generate current account balance of payments surpluses before interest payments, the argument will be that a debt default will enable Pakistan to avoid transferring resources abroad. This misleading argument focuses narrowly and superficially on the short term and ignores the long term harm a debt default will do to Pakistan's development prospects in the dynamic context of rapid globalisation of the world economy and unleashing of broad productivity growth possibilities by the revolution in information technology represented by spread of computers, fast expansion of the internet and declining costs of telecommunications.

Default Not a Real Option

Default is not really an option. (17) It will disrupt normal trade transactions forcing cash foreign exchange dealings, further adversely affect the climate for foreign investment, decrease the confidence in the currency and induce capital flight thus further pressuring the exchange rate, and last but not least will isolate Pakistan from the international community at a time when knowledge is becoming a key determinant of growth. On moral grounds also, there is no case for repudiation of Pakistan's external debt. The bulk of Pakistan's borrowing has been on concessional terms. Unlike some other countries Pakistan has, in general, not been a victim of excessive cost borrowing from commercial sources for projects of dubious value. About 40 percent of the public and publicly guaranteed debt is owed to the IBRD, IDA, and ADB, international organisations which have been and can be an important source of technical and economic advice. That the advice from the international organisations like the IMF and the World Bank has not always been sound, that the economic agenda has been driven excessively by the Brettonwoods organisations, and that Pakistan's own capability to sift advice and argue against unreasonable conditionalities has been in serious decline, are valid points but are not tantamount to a case for disruption of the valuable relationship with the international organisations. Pakistan needs to build up its analytical capacities in economic policy areas but it will continue to need financial and technical assistance from international organisations.

Instead of debating the merits and costs of external default, Pakistan needs to focus on avoiding default at all costs in the short run. At the same time, there is need to develop and articulate a medium and long term strategy which will reduce the burden of both external and public debt significantly say by 2008. Defining the debt reduction goals will not be enough. Institutional mechanisms need to be put into place to monitor and control new debt obligations.

Short and Medium Term Management Issues

Instead of turning its back on the international community, Pakistan needs to demonstrate clearly better macro economic management during the next three to four years. This will require not only living with the resource constraints but also expanding the base of government revenues and exports, reducing or eliminating the excess of current government expenditures over revenues, improving the quality of resource use in the public sector. Better governance and good economic management could create conditions for justifying additional debt relief or even debt forgiveness from 2001 onwards if it was evident that Pakistan's growth was suffering not because of economic mismanagement and / or lack of sufficient domestic resource mobilisation but limited net transfers of resources from abroad.

Further Fiscal Adjustment

Continued fiscal adjustment will need to be the central pillar of better macro economic management. Further steady reduction in the fiscal deficit from 6.1 percent of GDP in 1998-99 to 3-3.5 percent over the next three or four years is necessary to slow down the growth of real public debt. But how this fiscal adjustment is achieved will be critically important for long term growth as well as equity in the society. The limited fiscal adjustment made in the 1990s, largely under pressure from the IMF, was achieved almost entirely at the cost of cut in development spending. Even worse, the excess of government current expenditures over revenues actually increased from 0.6 percent of GDP in 1989-90 to the average of 2.7 percent in the three years 1997-99. That the government has been borrowing for current consumption for more than a decade is a root cause of our financial difficulties and a key factor in the low rate of domestic savings (12-13 percent of GDP--half the rate in India). The objective of fiscal policy should be not only to reduce the deficit further but also to transform negative government saving (18) of over 3.0 percent of GDP to a positive figure of GDP over the next three or four years. If this can be achieved, it will make a dramatic contribution to raising the domestic saving rate directly as well as indirectly by increasing confidence in the currency, reducing the need for frequent devaluations and discouraging capital flight.

Critical Importance of Tax Revenues

Whether this can be achieved will depend on the success in mobilising government revenue. As a result of not very well thought out reduction in tax rates in March 1997 and slow growth of the economy, the tax to GDP ratio, already low, fell to 13 percent in 1997-98 notwithstanding the large windfall gain due to lower import prices of oil.

Pakistan needs urgently not only to recover the ground lost in revenues over the last few years but to generate revenue surpluses to fund neglected development spending. This means raising the overall revenue to GDP ratio from 16.3 percent in 1998-99 to 18-19 percent over the next three or four years. But again revenue needs to be raised in a fashion that the long-term goals of an elastic and a fair system of taxation which promotes growth are not sacrificed.

Contrary to popular notions, the scope for orderly net reduction in government expenditure side does not exist. That financial constraint has taken a heavy toll on development spending is well known. That this trend needs to be reversed is generally conceded. But it is not widely recognised that the real current public spending (excluding interest) increased little over 1990-99 falling from 20.4 to 15.1 as a percentage of GDP. Defence spending which has declined from 6.9 percent of GDP in 1989-90 to 4.9 percent of GDP in 1998-99 can and should be reduced gradually by another percentage point of GDP over the next four or five years. But this reduction will be more than offset by more adequate allocations for the social sectors, specially education, and more adequate funding for maintenance of infrastructure. No doubt there is a great deal of waste in government and substantial downsizing is warranted. But on the other hand, public pay has been declining, social programmes are seriously under funded, and maintenance of roads, buildings and irrigation works has been greatly neglected.

Effectiveness of Public Spending

The purposes of increased social and development spending, laying the basis of self-sustaining long term growth and a broad sharing of growth benefits, will be defeated if the effectiveness of resource use in the public sector cannot be improved. In Pakistan economic growth has slowed down during the last decade both because there was insufficient investment in human and physical capital and the quality of public spending deteriorated either because there was increasing waste and corruption or spending was not focused on high economic and social return projects. In the best of circumstances, the investment rate which touched a new low of 15.0 percent of GDP in 1988-99 will recover only slowly. It may not be possible to restore investment to the average level (18.5 percent of GDP) seen during the last decade. This means that, apart from reducing the leakages from the system due to corruption, hard investment choices will have to be made in all sectors. This underscores the importance of re-establishing credible planning and monitoring processes.

Provided the fiscal policy adjustments outlined above can be successfully implemented, growth should recover slowly and the burden of public debt will begin to decline. But, as discussed below, this decline will normally be gradual. In the meanwhile, the interest payments on public debt estimated at around Rs 200 billion or 7.3 percent of GDP in 1998-99, are likely to remain the largest single item in the budget and will continue to constrain economic management and economic growth in Pakistan in the near future. Given this dire predicament, can more aggressive actions be taken to reduce and lower the burden of interest payments? The options are somewhat limited but merit discussion.

Privatisation Proceeds

The most promising avenue of reducing the outstanding debt is the allocation of the large part of the privatisation proceeds for the retirement of debt. In principle the government decision to do this is there. But privatisation process has moved very slowly. It needs to be accelerated for the sake of improving efficiency and transparency in the economy and could possibly result in revenues of Rs 100-200 billion over the next four or five years. Still in relation to the total public debt burden of nearly Rs 3000 billion, the possible contribution of privatisation proceeds to the retirement of debt will be very modest. More importantly, there are substantial accrued and contingent public sector liabilities which are likely to materialise as budgetary obligations and will add to the debt thus offsetting the whole or part of the relief coming from privatisation proceeds. Firstly, there are the unfunded losses of the public corporations especially WAPDA. Second, a large part of the bank loans arrears are unrecoverable and will become a charge on the budget: possibly one quarter of Rs 200 billion in bad loans of the public sector banks not covered by provisioning will have to be provided by the public exchequer to protect the depositors. Thirdly, there are large government guarantees provided in the case of IPPs which are becoming callable because of the weak financial position of WAPDA. The government needs to assess the impact of all these three elements on the evolution of public debt and it needs to make the analysis of contingent liabilities of the public sector an integral part of the debt management and monitoring system.

Interest Costs

If the amount of outstanding debt cannot be reduced quickly, can the interest costs be reduced? As discussed above, the average real interest rate on Pakistan's debt has not been historically high. As Table 1 indicates, the implied real interest rate while rising gradually did not exceed 2.5 percent during the two decades 1977-96 and the high real interest rate was the primary cause of further sharp growth in debt during the period: the primary balance being positive for the first time. The sharp rise in implied real interest rate reflecting the combined effect of four factors (1) slowing domestic inflation (2) zero international inflation (3) persistence of domestic interest rate at very high nominal level notwithstanding a sharp drop in the rate of inflation and (4) real devaluation which has increased the burden of foreign interest payments in rupees.

Pakistan has no control on the level of international inflation and real devaluation was necessary to remedy the structural weakness in the balance of payments. Lower domestic inflation is desirable and has resulted from efforts to reduce the budget deficit and limit monetary expansion. The main question is why has the response of the nominal interest rates to declining inflation been so slow. Two factors may be noted. First, till recently the interest offered by the government on tax exempt saving instruments remained high and this helped to sustain interest rates. Secondly, the international experience suggests that the real interest rates do not fall as long as uncertainty about the future course of inflation exists and the credibility of the government's macro economic policies is not fully established.

The reduction in the return on saving certificates by 2 percent per annum in May 1999 was a right step. The institution in September 1999 of a 10 percent withholding tax, in line with the tax treatment of deposits, was welcome. Unfortunately, it was reversed the government should go further and seriously consider phasing out fixed return saving certificates. It could then rely entirely on market based borrowing. Such a switch may be specially feasible if the aggregate domestic debt held by the public is not growing much--a situation which will be approached if the fiscal deficit is brought down to 3 percent of GDP and is financed mainly through money creation and net foreign inflows.

A significant reduction in the real cost of government borrowing, now very high, will only take place when the public is convinced that the fiscal adjustment will be sustained both through effective reform of the tax system and a judicious use of government revenues and borrowing. Staying the course of fiscal adjustment is likely to have increasing pay off both in terms of lowering of interest rates and greater stability in the value of the currency. But the degrees of freedom are also limited for Pakistan government. If additional revenues cannot be mobilised through the tax system and more effective use of public resources cannot be ensured, on the one hand growth will continue to suffer and on the other hand, inflationary financing of deficits will become unavoidable. But because external debt now accounts for over 50 percent of Pakistan's public debt, inflationary financing will prove counter productive because it will lead to a depreciation of the rupee which in turn will increase the local cost of servicing the external debt. In other words, the possible gains of an increase in the inflationary tax for financing domestic debt service will be more than offset by the increased cost of external debt service. The degrees of freedom are limited indeed.

Issues in External Debt Management

Strengthening the fiscal situation will certainly help the balance of payments but will not be enough to ward off the looming foreign exchange crisis. Here again a multi-pronged approach is necessary. The dispute with the IPPs must be resolved. The relationship with the IMF and the World Bank must be maintained because in the short and medium term they are the only agencies that can provide sizable net resources to Pakistan. On the domestic front, exports need to be revived and capital flight needs to be discouraged. The conditionality of the IMF and the World Bank on power tariffs, sales taxation and IPPs resolution may appear harsh but is necessary for financial viability. However, financial viability is not an end in itself. The purpose of short and medium term economic adjustment is not to seek a low-level equilibrium but rather to create conditions for more rapid growth. Just as additional government revenues are needed both for reducing deficits and increasing development spending, additional exports will relieve the debt burden but also facilitate imports.

Pakistan's exports are suffering because our export structure is weak and still heavily dependent on cotton based products. Extremely liberal incentives for relatively low value exports such as cotton yam were maintained far too long and as a result export diversification to relatively new and dynamic areas such as electronics and software did not take place. Exports are also being hampered by the low level of human development and great neglect of education. The structural problems in exports need to be tackled but in the short run exchange rate adjustments may be the only way to increase the international competitiveness of our products. Some further real devaluation may be necessary though it increases public debt burden. It will also improve the incentives for import substitution especially in wheat, and edible oils where large import deficits have persisted notwithstanding good agricultural potential in these crops.

A realistic and floating exchange rate will also curb the incentives for capital flight. But to some extent Pakistan's income tax laws encourage capital flight because income from sources outside Pakistan is not taxable. Institution of income tax on global incomes could moderate the desire to hold assets abroad. Finally, the merits of de-facto capital account convertibility in Pakistan need to be re-visited. Will it be disastrous if moneychangers and open market in foreign exchange no longer have official sanction? Limiting the free market in foreign exchange, while keeping the exchange rate realistic and access to foreign exchange for current transactions unrestricted, will make capital transfers abroad more expensive and less convenient and if the tax advantage enjoyed by income abroad is removed, the capital flight will be moderated. But in the ultimate analysis, capital flight is influenced more by the confidence in the currency and political and economic stability than return on assets. Purposeful macro economic management and clear political direction could even result in reversal of capital flight that has taken place.

In order to deal effectively with the problem of external debt in the medium term, four other aspects of debt management policy deserve attention. First, the objective should be to strongly discourage additional short-term resident currency deposits and to phase out the existing ones by converting them to longer-term obligations. The latter should be treated as a part of the external debt. Second, in dealing with external debt, other foreign exchange obligations, notably those related to resident foreign currency deposits and bonds issued in lieu of these deposits and obligations relating to IPPs must be explicitly taken into account in the context of a medium term balance of payments projection. Third, the foreign exchange reserves should be increased from the present level of little over one month to at least two months' of foreign exchange payments. The build up of foreign exchange reserves and the phasing out of short term foreign currency deposits will not necessarily have a negative impact on foreign exchange availability because it is likely to increase the flow of worker remittances and improve the prospect of obtaining short term flows from the international banking system. Finally, the foreign exchange revenues that may be available from privatisation i.e. sales of assets to foreigners should be earmarked for reduction rather than for current use--paralleling the policy for retirement of public debt.

An Over View of Adjustment

Table 8 below presents two economic scenarios for the next four years illustrating the constraints which the reduction in the fiscal and current account balance of payments deficits will place on economic management. The higher scenario B assumes a significant recovery in the government revenue to GDP ratio, a substantial shift towards positive government savings, a relatively high marginal saving rate (25 percent) and a large increase in government social and development spending. Even so the GDP growth is likely to average only 5 percent per annum because the overall investment will not recover to 1997-98 level by 2002-03. There will, however, be a reduction in the ratio of both public debt and external debt in relation to GDP. Since in this high scenario both government revenues and exports are expected to grow substantially faster than GDP, the reductions in public debt to revenues and exports to external debt ratios will be larger. In scenario A, growth remains at the average level of the last four years, government revenues, public savings and the national saving rate recover very slowly, public and overall investment rate stagnates. Furthermore there is much less reduction in the burden of public and external debt in relation to revenues and exports respectively. This underscores the basic point that financial discipline alone will not solve Pakistan debt problems, it will have to be a combination of policies stressing both low fiscal and current account deficits and high growth of revenues, exports, public investment and last but not least a better allocation of resources in the public and private sectors.

Longer Term Considerations

The debt burden will not reach the desired low levels by 2003 even in the successful adjustment scenario. Further debt reduction must be planned now and implemented over 2003-2008. What should the goals be? I believe that public policy should strive to lower the public debt to GDP ratio from around 600 percent in 1998-99 to 300 percent by 2008 and reduce the external debt to foreign exchange earnings ratio to less than 200 percent compared to nearly 350 percent now. Furthermore, debt obligations should be defined to include quasi- debt and contingent liabilities. Only limited progress may be possible by 2003 even under' scenario B. The debt goals can and should be debated but must be made public by the government. But unlike the past, debt management should be made an integral part of macro economic management. All borrowing decisions were ad hoc in the past. The institutional capacity to monitor debt levels, analyse debt management issues and give advice on annual borrowing plans is non-existent. The new economic team should rectify this and set up a high-level debt bureau in the Ministry of Finance to undertake the above mentioned functions. Initially it may be useful to set up a task force to gather the necessary data, analyse the debt issues, agree on debt reduction goals, and suggest the proper placement and functions of a debt bureau. Without appropriate institutional mechanisms and safeguards, the debt problem may again be neglected once the immediate urgency is over.

V. CONCLUSION

Pakistan is now a severely indebted country. Its public debt is close to Rs 3000 billion and exceeds 100 percent of GDP and 600 percent of annual government revenue. These public debt figures do not include (1) the unfunded losses of the public corporations especially WAPDA (2) the large part of the bank loans arrears which are unrecoverable (possibly one quarter of Rs 200 billion in bad loans of the public sector banks and will become a charge on the budget in order to protect the depositors) (3) large government guarantees given to IPPs which could become callable because of the weak financial position of WAPDA. External debt now totals over $35 billion (excluding frozen foreign currency deposits) and is nearly 60 percent of GDP and 300 percent of annual foreign exchange earnings. Government debt service absorbs 60 percent of government revenue and external debt payments constitute around 35 percent of earnings. Despite substantial relief and some debt reduction from Paris and London Clubs totaling over $6 billion, the foreign exchange situation remains very difficult and further re-scheduling and exceptional financing from the IMF, the World Bank and other sources will continue to be needed in the next three years.

It is helpful analytically if one views Pakistan as having not one but in fact two debt problems. High levels of public and foreign debt are two quite distinct, though in Pakistan's case closely related, aspects of the debt issue. Focus must be on both aspects because it is not the debt per se that matters but the ability to service it. The expected growth of revenues is of central importance for handling public debt while export earnings growth is often critical for keeping the external debt burden under control.

There are several points that emerge from a systematic examination of the evolution of Pakistan's public debt problem in its historical and comparative international context. First, the debt problem has been in making for a long time. The debt problem with which the democratic governments struggled, albeit unsuccessfully, during the last decade was to a considerable extent inherited from the Zia period. The debt burden was made much worse, however, by the inability or unwillingness of elected leaders to reduce the fiscal deficit significantly till very recently, a not very judicious use of public resources, and consequently a slowing economy, and a marked falling off in growth in real revenues. It is not surprising that the debt indicators which relate debt or debt service to revenues have shown much greater deterioration in the 1990s than in the 1980s. Second, the debt is now driven largely by interest rate costs. Historically, the real interest rate costs of Pakistan's public debt were low, averaging less than 3 percent per annum during the period up to mid 1990s. But the interest costs rose sharply to over 6 percent during 1996-99 reflecting the combined effect of rising nominal interest rates, slow down of both domestic and international inflation, and real depreciation of the exchange rate which increases the cost of servicing of external debt. Thus real debt has continued to grow even though the primary fiscal balance is now in surplus. Finally, Pakistan has a far more serious debt burden than India which is the only other large Asian country that has neglected fiscal adjustment and whose government finances are not in good shape. India's government debt to GDP ratio was only 68 percent in 1997-98 and its ratio to total government revenue was 370 percent.

Pakistan's foreign debt problem has become even more urgent than its domestic debt problem was reflected in the near default and subsequent rescheduling of external debt. Pakistan debt crisis was essentially triggered by the unsustainability of the level of the current account balance of payments deficits and the pattern of their financing. During the eight years 1991-98, Pakistan ran current account balance of payments deficits (before accruals of Resident Foreign Currency Deposits, RFCDs) totaling over $28 billion or an average of about 5.5 percent of GDP. This level of deficit is not sustainable for a decade even with a rapid expansion of exports. In Pakistan the growth of exports and remittances had in fact slowed down markedly in the first half of the 1990s and then stagnated. The alarm bells on the external debt did not ring partly because till 1996 only about half of the balance of payments financing needed took the form of normal debt flows, the rest being financed by short term foreign currency deposits, volatile portfolio investment and fixed cost direct investment in the power sector.

The lessons from Pakistan's experience with external debt are the rather obvious ones (a) debt problems can arise also if non debt flows are either unsustainable or too costly (b) debt problems cannot be separated from broader issues of economic strategy and management notably trend in savings, exports, and government revenue and quality of public resource use.

In the near term, say over the next two to three years, Pakistan will have to live with the macro-economic consequences of the heavy debt. This will mean abiding by the painful conditionality of the World Bank and IMF and managing the economy with limited growth in investment and imports. The foreign exchange constraint will make it difficult to stage a quick recovery in economic growth, urgently needed to create jobs and reverse the trend towards increase in the incidence of poverty. Faced with the prospect of painful adjustment, voices are likely to be raised advocating debt default. Default is not really an option. It will disrupt normal trade transactions forcing cash foreign exchange dealings, further adversely affect the climate for foreign investment, decrease the confidence in the currency and induce capital flight thus further pressuring the exchange rate, and last but not least will isolate Pakistan from the international community at a time when knowledge is becoming a key determinant of growth. Debt default will do great harm to Pakistan's development prospects in the dynamic context of rapid globalisation of the world economy and unleashing of broad productivity growth possibilities by the revolution in information technology represented by spread of computers, fast expansion of the internet and declining costs of telecommunications.

Instead of debating the merits and costs of external default, Pakistan needs to focus on avoiding default at all costs in the short run. At the same time, there is need to develop and articulate a medium and long term strategy which will reduce the burden of both external and public debt significantly and steadily. Defining the debt reduction goals will not be enough. Institutional mechanisms need to be put into place to monitor and control new debt obligations.

Instead of turning its back on the international community, Pakistan needs to demonstrate clearly better macro economic management during the next three to four years. This will require not only living with the resource constraints but also expanding the base of government revenues and exports, reducing or eliminating the excess of current government expenditures over revenues, improving the quality of resource use in the public sector. Better governance and good economic management could create conditions for justifying additional debt relief or even debt forgiveness from 2001 if it was evident that Pakistan's growth was suffering not because of economic mismanagement and / or lack of sufficient domestic resource mobilisation but limited net transfers of resources from abroad.

Comments

1.

Let me say at the very outset that I shall not talk about the process of growth in debt that has tied the country's economic managers in knots, very tight knots. I shall also not talk about the many virtually insurmountable impediments that reckless fiscal indiscipline has created. I shall also not talk about the broader contours of good governance because we never had it except for few brief periods. I shall also not talk about these things because they would need long hours of nerve-racking debate, for which we don't have time in this particular meeting. Let me start with very crucial general comments. In spite of Dr Pervez Hasan's very strong plea that we don't use the debt-GDP ratio, my reference point is the debt-GDP ratio. I shall come to the revenue and export side later on. If I recall correctly, the total debt of the countries that the IMF classified as having debt servicing difficulties in the 1980s was equal to about 50 percent of the GNP in those countries. In the two scenarios that Dr Hasan has presented in Table 8, the public debt as a percent of GDP remains around 82 percent by year 2000-2003, even if the higher growth is stable. We shall not be out of the debt difficulties in the foreseeable future. If so, then how can the society and country survive in these years? My hunch is that the answer would be to further accelerate the intake of revenues and push harder on exports. If that is the answer, then my follow-up question would be: How do you accelerate revenue mobilisation when the economy is in a state of serious recession? And how do you promote exports when the international marketplace is ruthlessly competitive and subject to restrictive rules of WTO? And our exports are dependent on commodities that have very slow growth, if any!

Let me now switch to some other, more specific points. I shall try my best to keep these points very simple and non-technical. I agree that the present value of external debt--both as a percent of the GDP and as a percent of exchange earnings-would be a relatively better measure, but it all depends on what discount rate we select and what our criteria would be, and on whatever basis.

Secondly, it is true that the expected growth in revenue provides the capacity to service debt. But it is also true in a country like Pakistan that, year after year, our economic managers have painted a rather rosy picture about the expected growth in revenues, and on that basis incurred expenditures and got us into greater and greater trouble. How are we going to change their mentality?

Thirdly, I also agree that an element of judgement about what constitutes a manageable debt burden in given circumstances will always remain. The question is whose judgement should count; that of the political masters, the bureaucrats or the financial and economic experts who have unfortunately been sidetracked for a number of years and still are not at the centre of decision-making?

Fourthly, I am also in quarrel with the statement that in developing countries, starting with low levels of savings and investments as well as debt, it makes a good deal of sense to borrow, provided the resources are well-directed. Now, this statement on page 11 [of the conference paper] is fine, but is not the basic proviso substantially missing in our country? The author himself gives several examples of pervasive misdirection of resources in Pakistan, indicated by the spending on low-income projects like the motorways, Peoples' Works Programme, the Convention Centre, and, if I may add, the MNA lodges, etc., that have limited gains for the country and contributed little to the ability to service its debt. We have lived with this massive misdirection of resources from the social point of view for years and from regime-to-regime. What we need is to stop this behaviour of the political heavyweights. It is easier said than done. I fully agree that Pakistan needs external adjustment as much as it needed the fiscal adjustment that has been so much the focus of the IMF agreements. We pay only scant attention to the external adjustment. The problem was not that the IMF and the World Bank were too sanguine about the prospects of additions to the resident foreign currency accounts. In my judgement, the process of using resident foreign currency accounts to finance current consumption and the ever-increasing current account deficit was fundamentally flawed. I fully agree with Dr Hasan and other experts that, we treated resident foreign deposits with the blessings of the IMF and the World Bank as if they were unrequited transfers. They were short-term liabilities in foreign exchange against which we should have kept a certain amount in reserve in foreign exchange.

Fifthly, Dr Hasan has argued strongly that a debt default is not an option. He debunks the notion of a default enabling Pakistan to avoid transferring its resources abroad. He argues that given the binding foreign exchange constraint and limited growth in investment in imports, Pakistan will have to live with the painful conditionalities of the World Bank and the IMF to manage the economy. My question is: What have we got from the conditionality-driven adjustment operations so far, and from the policy framework papers that were given to us and which were signed off without much thought? How good, relevant, and suitable has been the advice from the Brettonwoods institutions? What do we do to develop a domestic capability to critically review the advice given to us as part of the borrowing process? Finally, Mr Chairman, the key issue is whether it is possible for Pakistan to meet its current debt service obligations and yet gradually get out of the debt trap. when I returned to Pakistan four years ago, this was the central development issue that I raised. Some of my economist friends told me that I was an alarmist. I do plead guilty on one count. I was probably ahead of time. But now the time is not on our side. Let me end my comments by fully concurring with Dr Hasan's sensible suggestions which pertain to including contingent liabilities among debt obligations; making debt management an integral part of macroeconomic management; and developing institutional capacity to monitor the debt level; and devising debt management strategies, including annual borrowing plans. This, in turn, would require a task force, as we have mentioned, and setting up a high-level national debt bureau. Dr Pervez Hasan's paper is an excellent starting-point for our future efforts to come to grips with the intractable debt issue.

Fateh M. Chaudhri

Islamabad.

2.

The burden of external and internal debt on Pakistan is currently the most important economic problem in Pakistan and it has many economic and social implications. The paper by Parvez Hasan exposes the current state of debt situation in Pakistan from various angles, highlighting its dimensions, consequences, constraints and solutions. There is no doubt that the problem has reached alarming proportions during the current decade, especially in the past three years. It is also notable that almost all aspects of the poor state of Pakistan's economic, political and social aspects can somehow be linked to the growing burden of external and internal debt. Many of these aspects, such as political instability, corruption, poor governance and impotency of law lie right at the root of the problem.

On the other hand, problems like economic hardship (which encompasses poverty, unemployment and deteriorating income distribution), financial crisis, budget and current account imbalance have been affected seriously either directly or as a result of economic policy prescriptions that have been negotiated with the lending agencies (IMF, World Bank, etc.). As such it is not surprising that any meaningful study on debt, like the one under consideration, ends-up with an unwinding research on almost all aspects of macroeconomics besides political and social aspects. It should also be realised, therefore, that a proper modeling exercise to analyse the issue is difficult and needs a great deal of art to tie the loose ends up.

Although the paper is not based on any mathematical model, it contains a number of novel ideas. As such it provides an excellent guideline for researchers to develop mathematical models for the analysis of any particular aspect of the debt problem in Pakistan. The analysis is thorough in its scope and it rightly goes beyond some of the obvious aspects of the problem. Given the depth of analysis there is hardly any room to criticize the effort and to find conceptual flaws. However I would like to discuss some of the issues that need to be analysed further.

One of the issues that need to be discussed openly is the ethical dimension of debt management policies. This is especially important for policy-makers who represent Pakistan in negotiating the structural adjustment measures (the so-called conditionalities) with the IMF and other lending agencies. These adjustment measures are meant to enable the borrowing country to stand on its own feet and eventually start paying back the debt. At the heart of most of these measures lies the user-pay principle whereby the users of goods and services are supposed to make a just compensation to the provider. Unfortunately this very principle is hardly followed in distributing the burden of debt in the form of policies designed to narrow down the size of budget deficit. While a large portion of aid-inflows has mostly boon used to benefit only certain segments of society, the burden of the adjustment policies invariably has fallen disproportionately on the poor-to-middle income classes.

The author has rightly pointed out that Pakistan's own capability in arguing against unreasonable IMF conditionalities has been a serious problem area. A careful analysis would indicate that various governments in Pakistan have also been dishonest in implementing the agreed agenda. For example governments in the past have shown a great deal of capability while avoiding the implementation of agricultural income tax and protecting defaulters of bank loans, utility bills, etc. But when it comes to increasing utility rates and the real burden of tax on ordinary citizens they have always linked these measures as pre-conditions to soft loans from the IMF. With the exception of a few half-hearted attempts, the taxation system in Pakistan remains unfair and biased in favour of rich landlords and industrialists, while ordinary citizens have been asked to eat grass to save the nation. This is contrary to expectations that ultimately the objective of all economic policies is to raise the welfare levels of the majority.

Another point that needs to be discussed relates to the proposition that the burden of debt can be reduced by massive privatisation. Although there are merits in privatisation of certain economic activities in which government had no justification to enter in the first place, the author has rightly pointed out and it has also been analysed in detail elsewhere, that the solution to Pakistan's debt problem does not really lie in privatisation. The two key parameters relevant for successful solution of the problem are national savings rate and productivity. The author has discussed in detail many issues such as misdirected fiscal policies leading to inefficient resource allocation, governance, transparency, corruption, etc. that have direct bearing on the two parameters and there is not much need for further elaboration.

The current state of affairs is the accumulated outcome of economic mismanagement over the past five decades. There are no quick-fix remedies to the problem and it is naive to search for a solution without addressing the key issues of low saving rate and low productivity. Furthermore, the solution needs to be sustainable. It is in this perspective that privatisation proceeds alone cannot be expected to solve the problem though they may ease the burden temporarily.

Pakistan has wasted many long years in search of miracles. It is time to recognise that long-term planning is inevitable and there has to be a beginning towards that end. If things are allowed to stand as they are today and if Pakistan survives as a nation, fifty years from now the next generation of economists would still be repeating the same stories of failed policies and missed opportunities. It is the duty of economists and experts in related fields to tell the truth.

Eatzaz Ahmad

Quaid-i-Azam University, Islamabad.

(1) Two papers presented at the January 1999 conference of the PSDE deserve mention. Cite papers.

(2) Pakistan has a private debt problem in the sense that a large proportion of bank loans to the private sector are problem loans or non-performing loans. However, this paper deals with the private debt problem only to the extent that it will impinge on the budget (through the uncovered losses of the publicly owned banks) and the balance of payments (through servicing of private sector loans, whether guaranteed or not).

(3) See Annual Reports of the State Bank of Pakistan, 1994-95 to 1997-98.

(4) Ideally, the total value of public debt should also be adjusted for its present value. But because the domestic debt is not normally concessional, the present value calculations make sense only if external debt is a significant proportion of total public debt.

(5) The State Bank of Pakistan's definition of public debt service in fact excludes repayments of principal on domestic debt. See their Annual Report 1997-98, p.100.

(6) For some guidelines for external borrowing and their mathematical derivation, see World Development Report 1985, p. 53.

(7) Repayments only on external debt.

(8) Repayments due before debt rescheduling.

(9) See Parvez Hasan, Pakistan at the threshold of 21st century: How to shape a better economic future? The Pakistan Development Review, Papers and Proceedings, Winter 1998.

(10) See Parvez Hasan, Pakistan's Economy at the Crossroads: Past Policies and Present Imperatives, Oxford University Press, 1998, p. 41.

(11) Ibid, p. 251.

(12) For details of financial sector reform, see Hasan, pp. 292-97.

(13) Ibid, p. 278.

(14) This is the view held by Dr Muhammad Yaqub, former Governor, State Bank of Pakistan (based on author's discussion with Dr Yaqub).

(15) The main source of data is World Bank. The debt figure include not only public and publicly guaranteed debt, but also private non-guaranteed debt, foreign exchange bearer certificates and non resident institutional foreign currency deposits. The figures do not include resident and non resident foreign currency deposits which totaled about $10 billion in mid 1998. State Bank of Pakistan debt figures for total debt are $4-5 Billion lower for each of the recent years presumably because they do not include foreign exchange bearer certificates and non-resident foreign currency deposits. Surprisingly, however the SBP debt service payments are higher by more than $1 billion than the WB figures for years since 1994. A possible explanation could be that SBP includes principal repayments on short-term debt also.

(16) The State Bank of Pakistan began drawing the attention of the government to the dangers inherent in the heavy and subsidised reliance on RFCDs after August 1996 (see SBP annual report 199798, Chapter VII and appendix on Foreign Currency Deposits). By that time the problem had already become very large. Certainly the State Bank of Pakistan could have been far more aggressive in increasing the fee for forward exchange cover because it took its first large loss (Rs 13 billion or nearly 1 percent of GDP in 1993-94). Furthermore, the need to link foreign exchange reserves to the level of foreign currency deposits should have been pursued more forcefully because already in 1993 there had been a run on foreign currency deposits.

(17) I have benefited from an informal note and discussion on the subject with Mr Azizali F. Mohammed, a former Director of the IMF and a distinguished Pakistani economist.

(18) Defined narrowly as excess of government current expenditure over total revenues. If capital receipts, representing mostly depreciation allowances of the public sector and public corporations, are taken into account the government dissavings will appear to be smaller.

(19) The GDP growth rates figures in both scenarios A and B are assumptions relating to the four year period 1998-99 to 2002-3.

(20) Defined as current account balance of payments deficit.

(21) Defined as excess of consolidated public revenues over current expenditures.

Parvez Hasan is a former Director and Chief Economist of the World Bank. Table 1 Key Economic Variables Influencing Public Debt Year 1977-88 1988-96 1996-99 Nominal Debt Growth (1) 17.8 17.2 15.8 Real Debt Growth (1) 9.6 6.1 6.0 Real GDP Growth (1) 6.7 5.7 2.8 Inflation (1) 7.5 10.5 9.1 Real Govt. Revenues Growth (1) 9.0 4.5 0.0 Fiscal Deficit as Percent of GDP 7.2 7.5 6.9 Primary Deficit as Percent of GDP 4.2 2.0 -0.2 Primary Deficit as Percent of Debt 7.3 2.6 -0.2 Implied Real Interest Rate on Debt (1) 2.1 3.5 6.2 (1)Percent per Annum Note: The fiscal deficit figures after FY 1992 have been adjusted upwards by 1 percent per annum to account for borrowing of four major public corporations WAPDA, OGDC, NHA, and PTC which were previously part of the budget. Table 2 Indicators of Public Debt Burden Public Public Debt Interest Debt as % as % of payments as Year of GDP Revenues % of GDP 1976-77 57.5 423 1.9 1979-80 54.5 357 2.2 1984-85 60.2 367 3.6 1987-88 77.1 445 3.8 1989-90 82.6 431 5.4 1992-93 83.4 467 6.2 1995-96 86.8 505 6.2 1996-97 90.2 564 6.6 1997-98 91.2 557 7.3 1998-99 102.0 624 7.3 Interest Total Debt Total Debt payments as Payments Payments as % of as % of % of Year Revenues GDP Revenues (7) 1976-77 13.7 N.A. 1979-80 13.2 4.6 27.8 1984-85 21.3 4.5 32.3 1987-88 28.4 5.8 39.9 1989-90 28.5 7.6 41.6 1992-93 32.9 8.4 45.1 1995-96 36.0 9.0 54.4 1996-97 41.9 10.7 67.3 1997-98 41.8 10.4 61.5 1998-99 42.6 (8) N.A. N.A. Source: State Bank of Pakistan Annual Reports and Economic Surveys. Table 3 Average Interest Rate on Domestic Debt Nominal Average GDP Deflator Year Interest Percent Real Rate Percent Change Interest Rate 1979-80 5.0 11.0 -6.0 1981-82 6.4 9.4 -3.0 1983-84 7.5 9.7 -2.2 1984-85 7.5 4.5 3.0 1985-86 7.1 3.3 3.8 1986-87 7.0 4.5 2.5 1987-88 8.4 9.6 -1.2 1990-91 8.8 13.1 -4.3 1992-93 11.2 8.7 -2.5 1993-94 12.0 12.9 -0.9 1994-95 10.5 14.2 -3.7 1995-96 12.3 8.0 4.3 1996-97 13.0 13.3 -0.3 1997-98 15.1 7.8 7.3 1998-99 13.7 6.0 7.7 Source: Author's estimates based on Economic Surveys. Table 4 Pakistan's Public Debt Rupees in Billions End of Year Domestic Debt External Debt Total Debt 1972-73 17.8 (26.6) 32.3 (48.3) 50.1 (74.9) 1976-77 32.7 (21.8) 53.6 (35.8) 86.3 (57.6) 1979-80 56.8 (24.3) 70.7 (30.2) 137.5 (54.5) 1980-81 60.1 (21.6) 73.9 (26.7) 134.0 (48.3) 1984-85 143.9 (30.5) 140.2 (29.7) 284.1 (60.2) 1985-86 200.1 (39.0) 186.8 (36.3) 387.6 (75.3) 1986-87 247.3 (43.2) 208.6 (36.4) 455.9 (79.1) 1987-88 288.6 (42.7) 232.4 (34.4) 521.0 (77.1) 1988-89 331.1 (43.0) 299.4 (38.9) 630.5 (81.9) 1989-90 378.3 (44.4) 328.9 (38.4) 707.2 (82.6) 1990-91 445.1 (43.6) 436.3 (38.4) 821.1 (80.5) 1991-92 521.8 (43.1) 436.3 (36.0) 958.1 (79.1) 1992-93 602.4 (44.9) 517.2 (38.6) 1119.6 (83.4) 1993-94 702.0 (44.6) 749.4 (47.6) 1451.4 (92.2) 1994-95 798.6 (42.4) 785.1 (41.7) 1731.7 (84.1) 1995-96 908.9 (42.4) 951.0 (44.4) 1859.9 (86.8) 1996-97 1041.9 (43.3) 1127.3 (46.9) 2169.2 (90.2) 1997-98 1151.4 (41.7) 1366.9 (49.5) 2518.3 (91.2) 1998-99 1389.3 (47.7) 1581.9 (54.3) 2971.2 (102.0) Source: State Bank of Pakistan Annual Reports and RAF International Financial Statistics. Note: The figures in parenthesis are ratios to GDP in current market prices. Table 5 External Debt Burden Total Foreign External Exchange Total Debt Interest Year Debt's (15) Earnings Service Payments 1979-80 9.9 (209) 4.8 0.87 (18.3) 0.38 (7.9) 1985-86 14.9 (229) 6.5 1.64 (25.3) 0.58 (8.9) 1987-88 17.0 (231) 7.4 1.83 (25.0) 0.67 (9.2) 1989-90 20.7 (250) 8.3 1.93 (23.3) 0.84 (10.2) 1990-91 23.4 (249) 9.4 1.96 (20.9) 0.87 (9.2) 1991-92 24.9 (246) 9.8 2.34 (23.8) 0.89 (9.0) 1992-93 24.5 (242) 10.0 2.39 (23.9) 0.87 (8.7) 1993-94 27.3 (278) 9.8 3.48 (35.3) 1.00 (10.1) 1994-95 30.1 (258) 11.7 3.20 (27.4) 1.20 (10.2) 1995-96 29.7 (250) 11.9 3.28 (27.5) 1.19 (10.0) 1996-97 30.9 (260) 11.6 4.07 (35.2) 1.23 (10.9) 1997-98(Est.) 31.4 (373) 11.5 N. A. N. A. 1998-99(Est.) 32.3 (288) 11.2 N. A. N. A. Source: World Bank till 1995-96. The rest are author's estimates based on partial data from various sources. World Bank data for 1995-96 apparently short term and private debt. Table 6 External Debt Indicators of Major Developing Countries Total Present External Value of Debt $ Debt (1997) Billion Indebtedness Percent of Country (1997) Classification GNP Algeria 30.9 M 65 Argentina 123.2 S 38 Brazil 71.5 S 23 Chile 31.4 M 41 China 146.7 L 15 Colombia 31.8 M 33 Egypt 29.8 L 28 India 94.4 M 20 Indonesia 136.2 M 62 Korea 143.4 L 33 Malaysia 47.2 M 47 Mexico 149.7 L 37 Pakistan 29.7 M 37 Philippines 45.4 M 45 Russia 125.6 L 27 Thailand 93.4 M 61 Turkey 91.2 M 43 Present Value of Total Debt Debt (1997) Service (1997) Percent of Percent of Exports of Exports of Goods and Goods and Country Services Services Algeria 181 27.2 Argentina 352 58.7 Brazil 277 57.4 Chile 140 20.4 China 63 8.6 Colombia 178 26.6 Egypt 99 9.0 India 138 19.6 Indonesia 195 30.0 Korea 84 8.6 Malaysia 46 7.5 Mexico 110 32.4 Pakistan 203 35.2 Philippines 293 30.9 Russia 114 6.5 Thailand 120 15.4 Turkey 142 18.4 Public and Publicly Guaranteed Debt Service Percent of Short Term Central Debt as Government Percent of Current Total Debt Country Revenues -1997 Algeria NA 0.5 Argentina 34.2 14.6 Brazil NA 18.6 Chile 7.3 31.6 China NA 21.6 Colombia NA 18.1 Egypt NA 10.0 India 17.0 5.3 Indonesia NA 26.4 Korea 5.5 37.5 Malaysia 12.0 31.6 Mexico NA 19.0 Pakistan 31.5 8.4 Philippines 20.3 22.4 Russia NA 4.9 Thailand 7.0 37.3 Turkey NA 24.8 Source: World Bank, World Development Indicators, 1999. Table 8 Two Scenarios of Economic and Financial Adjustment Figures as percent of GDP Years 1989-90 1997-98 1998-99 GDP Growth 4.6 4.3 3.1 Investment 18.1 17.1 15.0 National Savings 13.6 14.6 11.0 Foreign Savings (20) 4.5 2.7 3.8 Public Savings (21) -0.6 -2.7 -3.0 Private Savings 14.2 17.3 14.0 Government Revenues 18.6 15.8 16.3 Tax Revenues 14.0 13.0 12.7 Total Public Expenditure 25.7 21.5 22.4 Defence 6.8 4.6 4.9 Interest 5.4 7.3 7.3 Total Current Expenditure 19.2 19.5 20.0 Development 6.5 3.9 3.0 Fiscal Deficit 6.5 7.7 6.1 Public Debt 83.0 95.0 102.0 External Debt 40.0 51.0 54.0 2002- 2002- Years 03 A 03 B GDP Growth 4.1 (19) 5.0 Investment 15.0 16.0 National Savings 13.0 14.0 Foreign Savings (20) 2.0 2.0 Public Savings (21) 0.5 1.0 Private Savings 12.5 13.0 Government Revenues 17.5 19.0 Tax Revenues 14.5 16.0 Total Public Expenditure 21.5 22.0 Defence 4.5 4.2 Interest 6.5 6.3 Total Current Expenditure 17.0 18.0 Development 3.5 4.0 Fiscal Deficit 4.0 3.0 Public Debt 95.0 92.0 External Debt 56.0 55.0 Sources: Economic Survey 1997-98 and author's estimates and assumptions. The marginal saving rate is assumed at 25 percent in the scenario A and 33 percent in scenario B.