Chapter

Appendix II

Author(s):
International Monetary Fund
Published Date:
December 1995
Share
  • ShareShare
Show Summary Details

The Fiscal Burden of External Debt

Interest and concern have increased recently about the fiscal burden of external debt, in particular that of the heavily indebted poor countries. Clearly, the burden of servicing external public debt puts demands on budgetary resources and can contribute to a need for fiscal adjustment; this in turn can lower private savings, which could adversely affect growth and thereby a country’s future debt-servicing capacity. This section provides factual background material on the fiscal burden of external debt service for the 41 heavily indebted poor developing countries.102 The main findings are:

  • On the basis of 1994 data, about half of the heavily indebted poor countries face scheduled external debt-service payments exceeding one half of annual government revenue (excluding grants). For 13 of these countries, scheduled debt service exceeded toial annual government revenue.103 However, actual debl service paid on average was only one third of scheduled debt service due to debt relief or the accumulation of arrears. In addition, foreign grants substantially alleviated the debt-service burden in most countries, as these added to government revenue resources equivalent to (or exceeding) the actual debt service most of these countries paid.

  • For most of the heavily indebted poor countries, the assessment of the debt-service burden on the basis of fiscal indicators closely follows that based on external indicators. However, some countries with low external debt-service ratios have high fiscal debt-service burden indicators. In some of these countries, this reflects relatively low revenue-to-GDP ratios.

  • Among the 13 severely indebted countries where scheduled debt service exceeded annual government revenue, a preliminary analysis of fiscal sustainability, based on a purely illustrative and hypothetical framework, indicates that for about half of these countries, the fiscal policy stance observed in 1994 could not avert a further increase in the external debt-to-revenue ratio from already very high levels. Further analysis of the fiscal sustainability of external debt would need to be made on a country-specific basis.

After a brief overview, indicators of the fiscal burden of external debl service in relation to government revenue and expenditure are analyzed below. This debl burden assessment is then compared with an assessment on the basis of the more usual debt service-to-exports ratios. Finally, a stylized framework is applied to a limited set of countries deemed to have a heavy debt-service burden on a fiscal basis as a first look at fiscal sustainability; the framework is described in more detail at the end of this section.

Overview

Total public and publicly guaranteed external debt of the heavily indebted poor countries in 1994 was estimated at $206 billion. Most of the total was medium-and long-term ($189 billion), and three quarters of this was on concessional terms (Table A5). Short-term debt was small at $15 billion, and private sector publicly guaranteed debt, at $2 billion, was equivalent to only about 1 percent of medium- and long-term debt. These countries had arrears of $54 billion, equivalent to over 25 percent of total debt.

The degree of public sector external indebtedness varied widely among countries in 1994. Total externa) debt ranged from about one year’s budget revenue (Myanmar) to well over 20 years’ revenue (Guinea-Bissau, Mozambique, Nicaragua, Sao Tom6 and Principe, Sudan, and Zaire; see Table A6 and Chart A6). After accounting for the devaluation of the CFA franc, public sector external debt on average rose to about 11 years of government revenue by the end of 1994, up from an average of 9 years of government revenue in the 3 preceding years.

External Debt-Service Burden Compared to Government Revenues

For the 41 heavily indebted poor countries examined, scheduled debt service (SDS) on external public debt before debt relief was equivalent, on average, to 90 percent of government revenue (before grants) during 1994, up from 84 percent between 1990 and 1993 (see Table A6, Chart A7, and Box A4).104Table A6 categorizes countries according to the severity of their SDS burden in 1994: 12 countries had SDS obligations of over one half of one year’s government revenue, and a further 13 countries had SDS obligations exceeding annual government revenue. Actual debt service represented, on average, only one third of SDS in 1994, as a result of reschedulings (Mozambique, Sao Tome and Principe, and Zambia) and, in some cases, the continued accumulation of arrears (Equatorial Guinea, Guinea-Bissau, Madagascar, Nicaragua, Nigeria, Sudan, and Zaire).

Table A5.Heavily Indebted Pour Countries: Total External Public and Publicly Guaranteed Debt, End-1994
Consolidated Public Sector Debt 1
Medium andPublicly GuaranteedConcessional
long-termShort-termPrivate DebtArrearsRatio 2
(In millions of U.S. dollars, unless otherwise indicated)
Angola9.2175.1270.32
Benin31,36811--0.96
Bolivia34,17025046-0.59
Burkina Faso1,326--350.88
Burundi1,16911.00
Cameroon7.0281961.0220.42
Central African Republic895-510.91
Chad709--0.83
Congo 43,2180.41
Cote d’Ivoire 513,0721,379 6-4.6360.27
Equatorial Guinea267--540.68
Ethiopia 74,1670.82
Ghana4,720175-0.84
Guinea2,7883021890.84
Guinea-Bissau77043-3030.81
Guyana1,9887591820.72
Honduras 33,7761250.47
Kenya5,506780.60
Lao People’s Democratic Republic 4554---1.00
Liberia0.53
Madagascar3,97662-1,6320.57
Mali2,950--36 60.98
Mauritania32,216108520.82
Mozambique 85,404-120.68
Myanmar5,35912-1.4980.92
Nicaragua11.3034436.0520.39
Niger1,347170.70
Nigeria31,189--9.3040.04
Rwanda866-1611.00
Sao Tome and Principe21447-990.90
Senegal 43,18252-850.71
Sierra Leone782413-220.72
Somalia2,040 90.82
Sudan10,1357,9141,48010,8460.52
Tanzania5,7051102540.73
Togo101,387--2160.75
Uganda3.578782240.80
Vietnam 114,467--1.2530.91
Yemen. Republic of8,4434463,5250.63
Zaire11,7682,1215.8220.37
Zambia6,259652-1,4870.61
Total189.27414,6341,77454,2970.76
Sources: World Debt Tables 1993 for concessional ratios; and IMF staff estimates.

Public sector operations consolidated at the central government level unless otherwise indicated. Includes debt to the IMF. Includes arrears.

Ratio of outstanding concessional debt to total public and publicly guaranteed debt.

Consolidated public sector includes state and local governments, state enterprises, and the financial public sector.

Includes state and local governments.

Includes the amortization, stabilization, and social security funds.

1993

Includes public enterprises.

Includes provincial governments.

1990

Includes the amortization and stabilization funds and some local governments.

Excludes debt to Russia, which is under negotiation.

Sources: World Debt Tables 1993 for concessional ratios; and IMF staff estimates.

Public sector operations consolidated at the central government level unless otherwise indicated. Includes debt to the IMF. Includes arrears.

Ratio of outstanding concessional debt to total public and publicly guaranteed debt.

Consolidated public sector includes state and local governments, state enterprises, and the financial public sector.

Includes state and local governments.

Includes the amortization, stabilization, and social security funds.

1993

Includes public enterprises.

Includes provincial governments.

1990

Includes the amortization and stabilization funds and some local governments.

Excludes debt to Russia, which is under negotiation.

Chart A6.Heavily Indebted Poor Countries: External Debt and Government Revenue, 1994

Source: IMF staff estimates.

Note: Countries ranked according to their scheduled external debt service-to-government revenue ratios.

1 Excludes Zaire.

There are, however, marked variations among countries. Countries such as Cote d’lvoire and Honduras, which are not among the countries with the highest SDS ratios, have high actual debt-service ratios as most of their scheduled debt service in 1994 was actually paid (see Table A6 and Chart A7). For other countries (e.g., Burkina Faso and Sierra Leone), the actual debt service is greater than scheduled debt service, reflecting the payment of arrears. In many countries, high SDS-to-revenue or actual debt-service ratios reflected low revenue-to-GDP ratios (see Table A6 and Chart A8).105

The SDS-to-revenue ratio reflects the degree of concessionality of the debt (see Table A5, last column). Concessional debt as a proportion of total debt is as low as 4 percent for some countries (Nigeria) while others have only concessional debt (Burundi, the Lao People’s Democratic Republic, and Rwanda). Thus, a country such as Chad, for which most of the debt is on highly concessional terms, has a moderate SDS-to-revenue ratio despite a very high debt-to-revenue ratio. In contrast, countries such as Angola and Nigeria have SDS-to-revenue ratios close to unity, despite relatively low debt-to-re venue ratios, partly due to the lower degree of concessionality of their debts.

For most of the heavily indebted poor countries, grant receipts are substantial and an important addition to government revenue, although most grants are project related. In 1994, grant receipts on average were equivalent to 40 percent of government revenue, somewhat higher than the 1990-93 average. At the same time, foreign grants represented, on average, 42 percent of SDS between 1990 and 1994 and were equivalent to actual debt service paid in 1994.106

Chart A7.Heavily Indebted Poor Countries: External Debt Service and Grants in Relation to Government Revenue, 1994

Source: IMF staff estimates.

Note: Countries ranked according to their scheduled external debt service-to-government revenue ratios.

Box A4.Fiscal Indicators of the External Debt-Service Burden

The scheduled external debt service (SDS )-to-revenue ratio shows a government’s capacity to repay as scheduled. Government revenue, as defined in this ratio, does not include foreign grants; however, it does include one-time proceeds, such as those from privatization. Foreign grants are excluded because they are uncertain and not a permanent source of government revenue. Also, foreign grants are often earmarked, for example to finance imports. Table A6, however, does report foreign grants separately in relation to government revenue because the foreign exchange earned through grants has constituted a significant resource for debt service for many countries.

The SDS-to-government revenue ratio has to be as-sessed together with the government revenue-to-GDP ratio in order to capture the government’s ability to ap-propriate real resources from the private economy (Chart A8). A high SDS-to-revenue ratio may simply indicate a government’s difficulties in collecting revenue.

The actual external debt service-to-revenue ratio captures the cash impact of debt servicing. It takes into account debt relief provided by debt-service cancellations and reschedulings as well as the effects of arrears incurred. A shortcoming of this measure is that, given the uncertain nature of debl relief, lower actual debt-service payments may simply reflect an unwillingness to pay.

The SDS-to-expenditure ratio measures the burden of servicing debt vis-a-vis other expenditures. It shows how much the debt burden constrains current and capital expenditure commitments.

Both the SDS-to-revenue and the SDS-to-expenditure ratios are based on the consolidated public sector debt, including short-term debt and debt to the IMF. Private sector publicly guaranteed debts are not included because they are a potential rather than actual fiscal burden. For many countries, though, the fiscal burden of external debt might increase once publicly guaranteed debt is taken into account. These ratios may overstate the fiscal burden for two reasons: (i) they include repayments to the IMF in public sector debt service. Given the monetary character of the IMF, obligations to the IMF tire norm ally obligations of the central bank; and (ii) they compare the external debt-service burden of the consolidated public sector to central government revenues and expenditure. Normally, public enterprises should be able to service their debts from their own revenues, although these debts remain a residual liability of the government.

The main shortcoming of debt-service ratios is the fact that they fail to capture anticipated changes in the debt burden that may result from such factors as expected debt relief or anticipated tax reforms. In addition, SDS ratios do not capture amounts in arrears, which in some cases may be substantial.

Table A6.Heavily Indebted Poor Countries: External Debt, Government Revenue, and Grants, 1990-94 1
ScheduledActualGovernment Revenue
DebtDebt ServiceDebt ServiceForeign Grants(Excluding Grants)
AverageAverageAverageAverageAverage
1990-9319941990-9319941990-9319941990-9319941990-931994
(Ratio to government revenue, excluding grant)(In percent of GDP)
Low (< 0.2) 2
Lao People’s Democratic Republic 33.142.790.100.060.100.060.350.471113
Myanmar1.771.080.110.070.030.040.020.0287
Vietnam42.091.200.410.150.180.070.030.031724
Moderate (0.2-0.5) 2
Benin55.817.250.420.460.140.230.220.161213
Bolivia53.303.160.390.310.210.210.090.112224
Burkina Faso3.136.500.140.280.190.440.380.311211
Burundi5.427.370.250.330.250.330.520.211716
Chad5.1712.630.190.480.070.260.942.5597
Ethiopia64.004.470.390.310.310.321317
Ghana4.523.600.390.290.390.290.240.141324
Kenya2.662.340.340.290.260.290.060.052731
Senegal33.125.960.350.500.220.340.070.281811
Uganda11.89 76.660.740.310.48 70.270.870.45810
Yemen, Republic of3.392.580.330.240.110.050.01-2619
High to severe (> 0.5)2
Angola 83.876.210.800.980.220.17--2736
Central African Republic6.2313.700.340.680.150.370.691.0197
Congo34.867.870.630.390.010.012325
Cote d’Ivoire 96.498.610.820.910.300.590.020.042222
Guinea6.548.790.560.740.380.430.280.341410
Guyana 1014.4710.821.400.920.680.520.050.053634
Honduras54.614.740.540.740.370.430.070.042423
Mali7.8613.000.520.750.180.490.740.771312
Mauritania512.2813.921.120.870.280.310.110.082323
Rwanda3.5219.470.140.800.510.25125
Tanzania9.629.060.750.740.260.330.300.132122
Togo 115.3011.830.520.990.170.180.080.061712
Severe (>1)2
Cameroon3.648.220.471.110.120.400.020.01139
Equatorial Guinea8.5712.280.791.110.180.261.160.712119
Guinea-Bissau23.6027.891.481.440.190.431.281.181412
Madagascar12.8217.741.501.660.370.3698
Mozambique1217.86720.961.681.830.280.360.871.222118
Nicaragua33.9031.203.064.150.440.660.550.241920
Niger6.5814.110.541.100.170.740.580.6896
Nigeria5.476.971.071.050.650.35--1811
Sao Tome and Principe24.1339.751.601.830.330.300.771.561923
Sierra Leone17.4310.402.101.110.381.360.080.301214
Sudan19.9228.151.561.890.050.0589
Zaire20.9379.592.006.080.140.080.150.08610
Zambia10.8210.291.941.861.400.920.430.381319
Simple Average8.9711.17 130.840.89 130.280.360.340.381616
Weighted Average 1410.0810.34131.061.05130.310.310.180.161616
Source: IMF staff estimates.

Based on consolidated public sector debt, including debt to the IMF. Public sector operations consolidated at the central government level unless otherwise indicated. The categorization is based on the scheduled external debt service-to-revenue ratio (1994), Data for Liberia and Somalia are not available.

Refers to the 1994 scheduled external debt service-to-rcventie ratio.

Includes state and local governments.

Excludes debt to Russia, which is under negotiation.

Consolidated public sector includes state and local governments, state enterprises, and the financial public sector.

Includes public enterprises.

Based on the 1992-93 average.

Data for 1990 are not available.

Includes the amortization, stabilization, and social security funds.

Data on debt to private creditors are not available for 1991 through 1993,

Includes the amortization and stabilization funds and some local governments.

Includes provincial governments.

Excludes Zaire.

Weighting based on consolidated public sector indebtedness.

Source: IMF staff estimates.

Based on consolidated public sector debt, including debt to the IMF. Public sector operations consolidated at the central government level unless otherwise indicated. The categorization is based on the scheduled external debt service-to-revenue ratio (1994), Data for Liberia and Somalia are not available.

Refers to the 1994 scheduled external debt service-to-rcventie ratio.

Includes state and local governments.

Excludes debt to Russia, which is under negotiation.

Consolidated public sector includes state and local governments, state enterprises, and the financial public sector.

Includes public enterprises.

Based on the 1992-93 average.

Data for 1990 are not available.

Includes the amortization, stabilization, and social security funds.

Data on debt to private creditors are not available for 1991 through 1993,

Includes the amortization and stabilization funds and some local governments.

Includes provincial governments.

Excludes Zaire.

Weighting based on consolidated public sector indebtedness.

Chart A8.Heavily Indebted Poor Countries: Government Revenue in Relation to GDP, 1994

Source: IMF staff estimates.

Note: Countries ranked according to their scheduled external debt service-to-government revenue ratios.

External Debt-Service Burden Compared to Government Expenditures

A similar picture of the debt-service burden across countries emerges when this burden is measured by the SDS-to-current expenditure ratio. 107Table A7 categorizes countries with low, moderate, high, or severe debt-service burdens on the basis of their SDS-to-current expenditure ratio. On average in 1994, scheduled external debt service was equivalent to about 70 percent of current expenditures for the heavily indebted poor countries, with external interest payments accounting for one third of current expenditures. In the most extreme cases of Nicaragua and Zaire, total scheduled debt service was equivalent to four to six times current expenditures.

Fiscal Compared with External Measures of the External Debt-Service Burden

Generally, there is a high positive correlation (0.82) between SDS-to-revenue and SDS-to-exports of goods and services ratios (Chart A9 and Table A8).108

However, some countries with relatively high external indicators of the debt-service burden have relatively low fiscal indicators, such as Ethiopia and the Republic of Yemen. Equally, there are countries with relatively low export ratios, but that require a relatively high proportion of their fiscal revenues to service their external debt.

For the Central African Republic, Cameroon, Chad, and Niger, the low correlation between export and revenue ratios is clearly due to low revenues (less than 10 percent of GDP).109 In Equatorial Guinea, the Congo, and Tanzania, fiscal revenues are about 20 percent of GDP, but the proportionally stronger balance of payments position reflects relatively more favorable private saving behavior.

Theoretical considerations that may help explain why some of these countries with higher fiscal than external burden indicators may have resorted relatively more to external financing are their comparatively low government revenues—due to inefficient tax systems or weak implementation—and/or relatively high public expenditures. Another possible explanation is limited domestic financing opportunities due to such factors as the small size of the domestic capital market, the high default and political risk perceived by potential bond buyers, or policies that constrain the market determination of interest rates.110

Table A7.Heavily Indebted Poor Countries: External Debt Service and Government Expenditure,1990-94 1
Scheduled Debt Service toInterest Payments on External Debt in Relation to
Current ExpenditureCapital ExpenditureCurrent Expenditure 2Total Interest Payments 3
Average 1990-931994Average 1990-93199419941994
(Ratios)
Low (0.2)4
Lao People’s Democratic Republic 50.100.060.120.060.030.49
Myanmar0.100.060.400.260.020.30
Yemen. Republic of0.250.121.381.150.020.15
Vietnam0.390.171.450.520.080.69
Moderate (0.2-0.5)4
Benin60.340.451.050.910.200.86
Bolivia60.440.350.900.840.140.92
Burkina Faso0.130.240.240.340.090.81
Burundi0.270.310.380.640.080.79
Central African Republic0.220.390.330.480.180.82
Chad0.110.230.190.180.080.85
Congo50.474.970.24
Ethiopia70.280.300.650.450.22
Ghana0.420.370.640.610.090.28
Kenya0.320.361.701.250.110.29
Rwanda0.090.220.191.090.070.34
Senegal50.400.481.671.550.180.82
Uganda0.590.330.600.420.090.82
High to severe (>0.5)4
Angola0.430.634.2812.880.180.97
Cameroon0.380.681.706.110.320.89
Cote d’Ivoire 80.580.795.243.920.300.87
Guinea0.680.820.740.950.180.95
Honduras60.640.651.551.180.280.76
Mali0.460.650.700.730.150.91
Niger0.350.520.841.610.170.95
Nigeria0.870.754.352.480.310.59
Sao Tome and Principe0.930.780.690.580.351.00
Sierra Leone1.380.995.533.340.440.77
Tanzania0.630.623.083.810.200.25
Togo90.380.542.185.340.230.93
Severe (>1)4
Equatorial Guinea0.751.050.531.080.360.93
Guinea-Bissau1.151.230.740.900.381.00
Guyana1.071.025.583.640.500.42
Madagascar1.261.001.711.880.220.95
Mauritania61.231.133.052.080.450.35
Mozambique101.531.411.611.320.430.96
Nicaragua2.354.0919.5210.111.410.93
Sudan1.261.964.868.780.55
Zaire0.526.243.411.990.15
Zambia1.381.525.234.140.61
Simple Average0.650.73112.342.44110.300.69
Weighted Average120.770.92113.843.61110.430.55
Source: IMF staff estimates.

Based on consolidated public sector debt, including debt to the IMF. Public sector operations consolidated at the central government level unless otherwise indicated. The categorization is based on the scheduled external debt service-to-current expenditure ratio. Data for Liberia and Somalia are not available.

Based on medium- and long-term public sector debt.

Based on budgetary data. Includes interest payments on domestic debt.

Refers to the 1994 scheduled debt service-to-current expenditure ratio.

Includes state and local governments.

Consolidated public sector includes state and local governments, state enterprises, and the financial public sector.

Includes public enterprises.

Includes the amortization, stabilization, and social security funds.

Includes the amortization and stabilization funds and some local governments.

Includes provincial governments.

Excludes Zaire.

Weighting based on consolidated public sector indebtedness.

Source: IMF staff estimates.

Based on consolidated public sector debt, including debt to the IMF. Public sector operations consolidated at the central government level unless otherwise indicated. The categorization is based on the scheduled external debt service-to-current expenditure ratio. Data for Liberia and Somalia are not available.

Based on medium- and long-term public sector debt.

Based on budgetary data. Includes interest payments on domestic debt.

Refers to the 1994 scheduled debt service-to-current expenditure ratio.

Includes state and local governments.

Consolidated public sector includes state and local governments, state enterprises, and the financial public sector.

Includes public enterprises.

Includes the amortization, stabilization, and social security funds.

Includes the amortization and stabilization funds and some local governments.

Includes provincial governments.

Excludes Zaire.

Weighting based on consolidated public sector indebtedness.

Chart A9.Heavily Indebted Poor Countries: Government Revenue in Relation to GDP, 1994

Source: IMF staff estimates.

1 Ranking based on scheduled external debt service-to-government revenue ratio.

2 Ranking based on the scheduled external debt service-to-exports of goods and services ratio.

An Illustrative Approach to Fiscal Sustainability

The existence of fiscal deficits does not necessarily imply that the ratio of debt to government revenue will grow over time. This ratio will grow inexorably, however, if a government runs primary deficits and the interest rate exceeds the growth rate of public revenues. How much and how fast this ratio changes, how much debt the private sector and the rest of the world is willing to hold, and the starting level of indebtedness are important determinants of the fiscal sustainability of a country’s external debt.

This subsection provides an illustrative assessment of fiscal sustainability for a group of 13 countries with a severe debt burden on the basis of fiscal ratios.111 The assessment is based on a stylized framework (Box A5). For each country, a base case has been defined with a plausible set of assumptions about key macroeconomic variables (Box A6).112 This rules out the possibility of achieving a lower debt-to-revenue ratio by relying on an unsustainable exchange rate policy to extract resources from the private sector.113 The key base case assumptions include no monetary financing, nonmon-etary domestic financing limited to a rollover of interest due. and a constant external debt-to-revenue ratio. This assessment represents a first illustrative exercise that simply asks whether the 1994 fiscal stance of the countries analyzed would have been sufficiently strong to keep the 1994 external debt-to-revenue ratio unchanged; this does not imply any judgment of the optimality of this ratio. For a more thorough assessment of sustainability, further work would be needed analyzing on a case-by-case basis each country in its specific macroeconomic context and external environment.

Table A8.Heavily Indebted Poor Countries: External and Fiscal Measures of the External Debt-Service Burden, 19941
Scheduled External Debt Service Relative to
Government revenue2Exports of goods and services
(Ratio)(Ranking)(Ratio)(Ranking)
Lao People’s Democratic Republic30.0610.041
Myanmar0.0720.3313
Vietnam0.1530.112
Yemen. Republic of0.2440.3717
Burkina Faso0.2850.184
Ghana0.2960.3111
Kenya0.2970.269
Ethiopia30.3180.4526
Uganda0.3190.2710
Bolivia 40.31100.3716
Burundi0.33110.3615
Benin40.46120.226
Chad0.48130.153
Senegal50.51140.205
Honduras40.54150.3414
Congo50.63160.237
Central African Republic0.68170.238
Tanzania0.74180.3212
Guinea0.74190.3818
Mali0.75200.4022
Rwanda0.8021
Mauritania40.87220.4424
Cote d’lvoire60.91230.4123
Guyana0.92240.4827
Angola0.98250.5529
Togo70.99260.39
Nigeria1.05270.4828
Niger1,10280.3920
Cameroon1.11290.4425
Sierra Leonel.ll300.6631
Equatorial Guinea1.11310.3919
Guinea-Bissau1.44320.9333
Madagascar1.66330.6230
Sao Tome and Principe1.83340.9834
Mozambique81.83351.3836
Zambia1.86361.0635
Sudan1.89371.9937
Nicaragua4.15383.5138
Zaire6.08390.7632
Source: IMF staff estimates.

Data for Liberia and Somalia are not available.

Based on central government obligations, unless otherwise indicated.

Includes public enterprises.

Consolidated public sector includes state and local governments, state enterprises, and the financial public sector.

Includes state and local governments.

Includes the amortization, stabilization, and social security funds.

Includes the amortization and stabilization funds and some local governments.

Includes provincial governments.

Source: IMF staff estimates.

Data for Liberia and Somalia are not available.

Based on central government obligations, unless otherwise indicated.

Includes public enterprises.

Consolidated public sector includes state and local governments, state enterprises, and the financial public sector.

Includes state and local governments.

Includes the amortization, stabilization, and social security funds.

Includes the amortization and stabilization funds and some local governments.

Includes provincial governments.

Box A5.Fiscal Sustainability Framework

The framework compares the current fiscal stance against the stance required to stabilize the external debt-to-revenue ratio at the base year. The indicator of fiscal sustainability is based on the primary gap needed to achieve stability of the external public debt-to-government revenue ratio. The primary gap is defined as the difference between the actual primary balance and the primary balance needed for sustain-ability. A positive gap indicates an inconsistency be-tween the actual fiscal stance and the stability of the debt-to-revenue ratio. The required balance is defined based on

  • a solvency constraint that requires budgeted ex-penditures to equal the sum of domestic revenue (in-flationary and noninflationary) and borrowing (domestic or foreign), and

  • a binding external debt target.

The second constraint is necessary to ensure that the government does not pursue an unsustainably expansionary fiscal policy financed by foreign bor-rowing. The assumption made here is simply dtat the target is to keep the external debt-to-revenue ratio constant, without making a judgment on whether this is a sufficiently ambitious target.

For the most basic illustration of this framework, consider a country with zero revenue growth, a 4 percent average interest rate on external debt, a debt-to-revenue ratio of I, and no domestic financing. Such a country needs to produce primary fiscal surpluses of 4 percent of revenue a year to maintain the external debt-to-revenue ratio constant.

The framework does not imply that the stability of the external debt-to-revenue ratio is necessarily optimal. Reduction of external debt exposure may be required in many cases. In particular, there might be a need to increase public savings by more than the amount suggested by this sustainability criterion alone.

Several factors affect the assessment of fiscal sus-tainability of external debt: anticipated tax reforms, windfalls from export taxes (such as thai recently observed among coffee exporters), the rate of interest on new foreign currency-denominated borrowing, the de-gree of debt concessionality, the evolution of foreign grants, the extent to which external debt is reschedu-lable. and the exchange rate policy. Because changes in the exchange rale may alter substantially the primary balance required to stabilize the debt-to-revenue ratio, Tables A9 and A10 decompose the impact that exchange rate changes have on the primary gap.

Cross-country comparisons such as the one in this appendix may be biased by the different forms in which individual countries receive foreign support, for example, by concessional interest rates, grants, or debt relief measures. An in-depth assessment of the fiscal sustainability of the external debt of a country would need to be based on a comprehensive macro-economic and Fiscal framework applied on a country-by-country basis.

Box A6.Base Case Assumptions for Fiscal Sustainability

The main base case assumptions are

  • nonmonetary domestic financing is limited to a rollover of domestic interest due. The ratio of domestic debt to revenue would vary according to a growth rate given by the real domestic interest rate discounted by the real growth rate of government revenue. There is no monetary financing;

  • no changes in

  • —the ratio of concessional to total external public debt,

  • —the nominal concessional and market interest rates on foreign borrowing by the public sector, and

  • —the inflation rate:

  • no foreign grants; and

  • exchange rate changes set equal to the domestic-foreign inflation differential.

Nine of the countries114 with a severe debt-service burden show a gap in 1994 between their actual fiscal primary balance and the balance that would have been required to keep the external debt-to-revenue ratio differential would remove Equatorial Guinea, Sierra Leone, and Sudan from this group, but add Nigeria. It is striking that under the base case, all countries with current IMF arrangements (Equatorial Guinea, Guinea-Bissau, Nicaragua, Sierra Leone, and Zambia)—except for Mozambique—had negative primary gaps, that is, a fiscal stance that would not contribute to an increase in the external debt-to-re venue ratio. Stable (Table A9).115 Setting the exchange rate variation during the year equal to the domestic-foreign inflation differential would remove Equatorial Guinea, Sierra Leone, and Sudan from this group, but add Nigeria. It is striking that under the base case, all countries with current IMF arrangements (Equatorial Guinea, Guinea-Bissau, Nicaragua, Sierra Leone, and Zambia)—except for Mozambique—had negative primary gaps, that is, a fiscal stance that would not contribute to an increase in the external debt-to-revenue ratio.

These results should be viewed with caution. For many countries, the sustainability assessment for 1994 is expected to differ from their future fiscal outlook to the extent that changes in government revenue or debt relief are anticipated. For example, in the case of Sierra Leone, revenues were depressed in 1994/95 by civil conflict; a recovery is expected in subsequent years together with more debt relief. The results are heavily dependent on the fiscal position in 1994: for some countries, the 1994 results reflected exceptional revenue efforts (such as the recovery of arrears in the case of Zambia). Further analysis based on a case-by-case approach would have to take into account the particular fiscal circumstances of each country.

Stylized Framework

This subsection applies and extends the work of Buiter (1994) on fiscal sustainability of debt.116 In particular, among the extensions provided, the analysis that follows differentiates between stabilizing the external vis-a-vis the total debt ratio, incorporates the effeet of a varying degree of concessionality of foreign borrowing, and uses an alternative income variable defined by total government revenue rather than the usual GDP. In addition, the following framework isolates the effects of foreign grants in financing of the public sector: generalizes the criteria for fiscal sustainability, allowing for the possibility of targeting a particular debt ratio, and accounts for possible changes in the exchange rate that may induce variations in the government’s net worth without changes in the fiscal stance.

Table A9Heavily Indebted Pour Countries: Stylized Fiscal Sustainahility Analysis of External Debt—Summary
Primary BalancePrimary Gap
1990-931994Total 1Base case 2
(Ratios to government revenue)
Cameroon-0.200.062.641.68
Equatorial Guinea-1.27-0.694.87-1.19
Guinea-Bissau0.130.26-3.98-5.99
Madagascar-0.65-0.754.521.61
Mozambique-0.99-1.594.392.46
Nicaragua-0.480.21-0.51-1.96
Niger-0.91-1.386.952.10
Nigeria0.09-0.10-0.570.77
Sao Tome and Principe-2.41-3.6613.100.97
Sierra Leone-0.29-0.221.18-0.24
Sudan-0.93-0.091.28-6.31
Zaire-2.490.0246.377.45
Zambia0.120.35-3.63-3.22
Source: IMF staff estimates.

As defined in the stylized framework. Debt target given by the debt-to-revenue ratio at the beginning of 1994.

Assumes that changes in the exchange rate follow the domestic-foreign inflation differential.

Source: IMF staff estimates.

As defined in the stylized framework. Debt target given by the debt-to-revenue ratio at the beginning of 1994.

Assumes that changes in the exchange rate follow the domestic-foreign inflation differential.

The following definition of the government budget constraint, expressed in domestic currency, is used:

Where the fiscal balance may be financed through foreign debt, domestic debt, or some degree of monetization.

  • St, is the primary balance, defined as public sector revenue (tax and nontax revenue, excluding grants and seigniorage), less expenditures, exclusive of total interest payments

  • It, is the nominal interest rate on government bonds denominated in domestic currency (Bd):

  • I*,t f is the nominal interest rate on foreign currency-denominated borrowing by the public sector (B*); and

  • Bt is the stock of nonmonetary financial debt at period t, excluding official foreign exchange reserves. The stock is composed of domestic (Bd,t) and external (B*,t) liabilities, the latter converted at the average nominal market exchange rate Et

  • Ht is the stock of monetary liabilities, or base money.

The following expression for the required budget surplus based on a target (external or total) debt-to-revenue ratio is derived by denoting by lower cases the above variables expressed in percent of total government revenue, and substituting, as appropriate, for the additional definitions below:

Where

  • i c,tis the nominal concessional interest rate on foreign borrowing by the public sector;

  • α is the ratio of foreign concessional to total external debt;

  • e t is the rate of devaluation of the nominal exchange rate;

  • πt is the growth rate for the deflator of government revenue

  • Gt is the growth rate of real government revenue;

  • σt is seigniorage, defined as the change in Ht in percent of total government revenue;

  • nt is the domestic currency equivalent of foreign grants to the public sector as a proportion of total government revenue

  • t is the targeted degree of reduction in the debt-torevenue ratio;

and

Where rt is the real domestic interest rate.

Table A10.Heavily Indebted Poor Countries: Stylized Fiscal Sustainability Analysis of External Debt
Required Primary BalanceRequired Primary Balance Under Alternative Assumptions
Due toInterestWorldDebt toIncluding
Exchange rateConcessionalRevenueInflationRevenuegrants and
Total1variationBase casefactor3Burden factor4Factor5Factor6TargetIncludingconcessional
(a)=(b)+(c)(b)2(c)=(d)+(e)(d)(e)=[(f)/(g)-1]*(h)(f)(g)(h)grantsterms7
(Government revenue ratio; unless otherwise indicated)
Cameroon2.700.971.73-0.091.821.421.055.102.692.55
Equatorial Guinea4.186.06-1.88-0.08-1.801.101.427.903.473.10
Guinea-Bissau-3.712.01-5.72-0.10-5.630.851.0430.21-4.89-4.92
Madagascar3.772.910.860.181.041.161.0712.203.413,24
Mozambique2.801.930.88-0.301.181.121.0517.031.591.43
Nicaragua-0.301.45-1.75-0.21-1.550.991.0530.33-0.54-0.88
Niger5.574850.72-0.160.881.171.057.754.904.79
Nigeria0.671.340.670.010.681.141.036.170.67-0.86
Sao Tome and Principe9,4412.13-2.69-0.07-2.620.961.0629.477.887.87
Sierra Leone0.961.42-0.46-0.05-0.410.870.983.720.680.65
Sudan1.197.596.40-0.12-6.280.811 0725.86
Zaire46.3938.927.47-0.337.811.311.0329.0846.3145.13
Zambia-3.28-0.41-2.87-0.09-2.770.811.0213.09-3.65-3.71
Source: IMF staff estimates

As defined in the stylized framework. Debt target given by the debt-to-revenue ratio at the beginning of 1994.

Other than the variation assumed under the base case.

Based on actual degree of debt concessionality

Reflects real revenue growth, average nominal foreign interest rate, and inflation in partner countries

Foreign interest and real government revenue growth rate differential.

As measured by the change in export unit value of trading partners.

Assumes all external debt to be on concessional terms.

Source: IMF staff estimates

As defined in the stylized framework. Debt target given by the debt-to-revenue ratio at the beginning of 1994.

Other than the variation assumed under the base case.

Based on actual degree of debt concessionality

Reflects real revenue growth, average nominal foreign interest rate, and inflation in partner countries

Foreign interest and real government revenue growth rate differential.

As measured by the change in export unit value of trading partners.

Assumes all external debt to be on concessional terms.

Note that all the terms in the expression for the required primary surplus (equation (3)) have a straightforward interpretation. The first term on the right hand side determines the sustainable (required) primary surplus based on the previous stock of external debt, the external interest rate, the rate of devaluation, the real growth rate of government revenue, and the inflation rate.117 This term is corrected by the proportion of concessional debt and the difference between concessional interest rates and market rates. The second term, d, (equation (4)), captures the increase in domestic debt ratios above the levels explained by the domestic interest and government revenue growth rates.

The primary surplus required to target the ratio of external debt to government revenue is derived by set-ting d, equal to zero. In this case, nonmonetary domestic financing is limited to the amount of interest payments due. The ratio of domestic debt to revenue wouid vary as long as the growth rate of government revenue differs from the domestic interest rate.118 Targeting of the total debt-to-revenue ratio requires dt to be nonzero, with bdt =bdt-1 yielding the following additional term in the expression for the sustainable primary surplus:

Table A10 derives, from equation (3), the sustainable primary balance for 13 countries with SDS-to-revenue ratios greater than 1 in 1994. The table decomposes the required primary balance in two components: one is due to the difference between the average foreign interest rate and the real government revenue growth rate; the other is due to the impact of exchange rate deviations from the purchasing power parity during the year. In addition, the table shows the required primary balance under two alternative assumptions: including grants as revenue, and replacing all nonconcessional debts with concessional ones.

The results show that in almost every case, because of the sensitivity of the debt-to-revenue ratio to changes in debt valuation, the primary balances required to stabilize it are very large, some being strongly positive, others strongly negative. Therefore, attempting to stabilize the debt-to-revenue ratio in any given year would not be a realistic goal. However, sta-bilizing or reducing the ratio over time would be a reasonable goal. The results also show the difficulty of stabilizing the debt ratio when its initial level is high. In Equatorial Guinea, for instance, to maintain a debt-to-revenue ratio equal to 8 the required primary balance was equivalent to 4 times government revenue in 1994. The exchange rate devaluation that took place during the year required a primary balance equal to 6 times government revenue to maintain a stable debt-to-revenue ratio. Excluding the exchange rate impact, other factors—such as the real growth in government revenue in excess of the average foreign interest rate, and accounting for the existing degree of debt concessionality—would have allowed a reduction in the debt-to-go vein men t revenue ratio equivalent to almost twice government revenue (Table A10).

The group is composed of the 32 countries that are classified by the World Bank as severely indebted low-income countries, an additional 7 rescheduling countries that have received concessional treatment from the Paris Club, and 2 lower middle-income countries that have recently become IDA-only (Angola and the Congo).

Cameroon, Equatorial Guinea, Guinea-Bissau, Madagascar, Mozambique, Nicaragua, Niger, Nigeria, Sao Tome and Principe, Sierra Leone, Sudan, Zaire, and Zambia.

Data for Liberia and Somalia are not available.

For example, Guinea-Bissau, Madagascar, Niger, Nigeria, Sierra Leone, Sudan, and Zaire.

In addition, new disbursements from creditors can alleviate the cash impact of a country’s debt-service obligations by de facto rolling over debt.

The correlation between the SDS-to-current expenditure ratio and the SDS-to-revenue ratio is high (0.94).

As a result of a correlation, as would be expected, between government revenues and exports.

The programs supported by IMF resources in these countries in 1994 all focused on measures to enhance the revenue base and increase revenues.

For a complete discussion, see “Public Debt and Fiscal Policy in Developing Countries,” by Vito Tanzi and Mario I. Blejer, in The Economics of Public Debt, ed. by Kenneth J. Arrow and Michael J. Boskin (New York: St. Martin’s Press, 1988), pp. 230-63.

Cameroon, Equatorial Guinea, Guinea-Bissau, Madagascar, Mozambique, Nicaragua, Niger, Nigeria, Sao Tome and Principe, Sierra Leone, Sudan, Zaire, and Zambia.

See Olivier Blanchard, “Suggestions For a New Set of Fiscal Indicators,” OECD Department of Economics and Statistics Working Paper No. 79 (Paris: Organization for Economic Cooperation and Development, April 1990).

For the application of a related framework to oil-producing countries see Claire Liuksila, Alejandro Garcia, and Sheila Bassett, “Fiscal Policy Sustainability in Oil-Producing Countries,” IMF Working Paper, WP/94/137 (Washington: International Monetary Fund, November 1994); to industrial countries see Chap. IV, World Economic Outlook, World Economic and Financial Surveys (Washington: International Monetary Fund, October 1993); to India see Karen Parker and Steffen Kastner, “A Framework for Assessing Fiscal Sustainability and External Viability, with an Application to India,” IMF Working Paper, WP/93/78 (Washington: International Monetary Fund, October 1993).

Cameroon, Equatorial Guinea, Madagascar, Mozambique, Niger, Sảo Tomé and Príncipe, Sierra Leone, Sudan, and Zaîre.

Allowing for changes in the base case assumptions regarding foreign grants or the degree of concessionality would not change substantially the fiscal sustainability assessment for any of the countries considered (Table A9, and stylized framework below).

Buiter, Willem H., “Indicators of Fiscal Sustainability” (mimeographed, Washington: International Monetary Fund, August 1994).

Assumed to be equal to the growth rate of the deflator for government revenue.

Under the alternative assumption of an absence of nonmonetary domestic financing, the ratio of domestic debt to revenue would vary according to the revenue growth rate, with positive rates implying a declining ratio. In addition to stabilizing the external debt-to-revenue ratio, the required primary surplus would have to provide for the payment of the domestic interest due (dt =—rt&f-1), since no domestic financing would be allowed.

    Other Resources Citing This Publication