Achieving Long-Term Debt Sustainability in All Heavily

Indebted Poor Countries (HIPCs)

Bernhard G. Gunter

*

Discussion Paper prepared for the

Intergovernmental Group of 24 (G-24)

XVI Technical Group Meeting in Trinidad and Tobago

February 13-14, 2003

Final version, February 26, 2003

Abstract:

This paper builds on the emerging consensus in the development literature that the

enhanced HIPC initiative does not fully remove the debt overhang in many poor and

highly indebted countries. It examines the six most crucial problems of the enhanced

HIPC Initiative and presents specific suggestions on how the framework of the HIPC

Initiative would need to be changed in order to provide a better basis for long-term debt

sustainability. However, even after the adoption of such changes, the long-term debt

sustainability of HIPCs would remain fragile. The paper thus addresses some of the key

issues related to a new aid architecture and the structural transformation HIPCs must

undergo to achieve long-term debt sustainability.

____________________

* The author is an independent consultant, specializing on issues related to development macroeconomics,

poverty, and debt, currently associated with the New Rules for Global Finance Coalition, the International

Labor Office, and the World Bank. He previously was an Economist in the World Bank’s HIPC unit (1998-

2000) and a Consultant for the World Bank Institute (1995-1998). The views expressed here are those of

the author and should not be associated to any institution. Without implicating, the author expresses thanks

for valuable comments received from Ariel Buira, Nisreen Farhan, Barry Herman, Rajapakse Jayatissa,

Patience Kunene, Aziz Ali Mohammed, Damian Ondo Mane, Henry Northover, Cristian Ossa, and Benu

Schneider. For communications, please contact

Bernhardgunter@earthlink.net

.

ii

Executive Summary

This paper provides a summarized review of the Heavily Indebted Poor Country

(HIPC) Initiative. It provides the historical and analytical background for the adoption of

the HIPC Initiative, whereby it illustrates that there are a variety of approaches and

indicators to analyze debt sustainability. In any case, given that HIPCs continued to face

a debt overhang after repeated debt rescheduling during the 1980s and early 1990s, the

HIPC Initiative was adopted in 1996 with the goal to provide a permanent exit from

repeated debt reschedulings. Though the HIPC Initiative had been enhanced in 1999,

evidence is mounting that the enhanced HIPC Initiative does not even provide short-term

debt sustainability for many of the poor and highly indebted countries.

The paper describes the six most crucial shortcomings of the HIPC Initiative,

which are related to (1) using inappropriate eligibility and debt sustainability criteria, (2)

the use of overly optimistic growth assumptions, (3) an insufficient provision of interim

debt relief, (4) the delivery of some HIPC debt relief through debt rescheduling, (5) lacks

in creditor participation and financing problems, and (6) the use of currency specific

short-term discount rates. To address these six shortcomings, the paper suggests (1) to

revise the HIPC eligibility and debt sustainability indicators, (2) to use lower bounds of

realistic growth assumptions, (3) to provide deeper and broader interim debt relief, (4) to

deliver HIPC debt relief only through debt cancellation, (5) to adjust the burden-sharing

concept, and (6) to use one fixed low discount rate for all NPV calculation.

However, given that the long-term debt sustainability of many poor countries

would remain fragile even after the full adoption of such changes, the paper also

addresses some of the key issues related to a new aid architecture. It shows (1) that more

aid coordination is urgently needed for HIPCs that have not yet reached their decision

point, (2) that it makes sense to substitute some loans with grants, (3) that HIPC debt

relief has thus far neither been frontloaded nor additional, and (4) that 100 percent debt

relief would be feasible as well as desirable for the poorest countries (irrespective of what

their debt levels are). The last section before the conclusion reviews the key issues related

to structural transformation HIPCs need to undergo. It reviews and comments on recent

trends in sectoral transformation, domestic savings and investment, as well as foreign

investment and export performance of the group of HIPCs.

iii

Contents

page

I. Introduction 1

II. Historical and Analytical Background 3

1. The history of traditional debt relief

2. Analytical background

3. The emergence of the HIPC Initiative

4. HIPC-III, PRSP, or international debt workout mechanism.

III. Problems of the Enhanced HIPC Initiative (HIPC-II) 6

1. Inappropriate eligibility criteria

2. Overly optimistic growth assumptions

3. Insufficient provision of interim debt relief

4. Delivery of HIPC debt relief through debt rescheduling

5. Lacking creditor participation and financing problems

6. Currency-specific short-term discount rates

IV. Suggested Changes in the Framework of the HIPC Initiative 15

1. Revision of HIPC eligibility and debt sustainability indicators

2. Using lower bounds of realistic growth assumptions

3. Provision of deeper and broader interim debt relief

4. Delivery of HIPC debt relief only through debt cancellation

5. Adjustments in the burden-sharing concept

6. Using one fixed low discount rate for the NPV calculation

V. Issues Related to a New Aid Architecture 20

1. Aid Coordination for “new” HIPCs

2. Grants versus loans

3. The issue of additionality

4. Is 100 percent debt relief feasible.

5. Is 100 percent debt relief desirable.

VI. Necessary Structural Changes 23

1. Sectoral shares of GDP

2. Long-term trends in domestic savings and domestic investment

3. Long-term trends in foreign investment

4. Export concentration, export price volatility, and contingent facility

VII. Conclusion 30

iv

Boxes

Box 1: Practitioners’ suggestions of debt sustainability indicators 2

Box 2: Commodity risk vulnerability of HIPCs 29

Tables

Table 1: Breakdown of potential costs of the enhanced HIPC Initiative

by main creditors (US$ billion, in 2001 NPV terms) 12

Table 2: Realistic GDP growth rates for the HIPCs 17

Figures

Figure 1: Relationship between debt and GNP growth 3

Figure 2: Paid debt service of 22 HIPCs 10

Figure 3: Official net resource flows, 1990-2001 22

Figure 4: Structural transformation of 42 HIPCs, 1990-2001 24

Figure 5: Structural transformation: share of manufacturing, 1990-2001 24

Figure 6: Percentage share of investment and savings in GDP, 1990-2001 25

Figure 7: Trends in Foreign Direct Investment (FDI) of 42 HIPCs, 1990-2001 26

Figure 8: Trends in Portfolio Direct Investment of 42 HIPCs, 1990-2001 26

Figure 9: Terms of trade of 42 HIPCs, 1990-2001 27

Figure 10: Exports of 42 HIPCs, 1990-2001 28

v

List of Acronyms

CIRR commercial interest reference rate

DSA debt sustainability analysis

EURODAD European Network on Debt and Development

FDI foreign direct investment

FTAP Fair and Transparent Arbitration Procedure

HIPC Heavily Indebted Poor Country

HIPC-II enhanced HIPC Initiative

HIPC-III envisioned further enhanced HIPC Initiative

HPI-I Human Poverty Index for developing countries

G-7 Intergovernmental Group of Seven

G-24 Intergovernmental Group of Twenty-four

GAO General Accounting Office (of the United States)

GDP gross domestic product

GNP gross national product

IBRD International Bank for Reconstruction and Development

IDA International Development Association

IMF International Monetary Fund

MDB multilateral development bank

MDG Millennium Development Goal

NPV net present value

ODA official development assistance

OECD Organisation for Economic Co-operation and Development

OED Operations Evaluations Department (of the World Bank)

PC Paris Club

PRSP Poverty Reduction Strategy Paper

SDR Special Drawing Right

SDRM Sovereign Debt Restructuring Mechanism

UK United Kingdom

UNDP United Nations Development Program

US United States of America

The $ sign refers to the US dollar.

One billion equals one thousand millions.

One trillion equals one thousand billions.

1

I. Introduction

Achieving long-term debt sustainability is a complex and challenging task that

requires a combination of appropriate macroeconomic, structural, investment, and debt

management policies. In poor countries, long-term debt sustainability is also heavily

influenced by external factors, like the terms of trade, donor financing, and the provision

of debt relief. However, there are various ways to define debt sustainability. As is

illustrated in detail in EURODAD (2002), Northover (2001), and Sachs et al. (1999), if

debt sustainability is approached from a human and social development perspective,

1

most of the poorest countries have an unsustainable debt irrespective of what their debt

levels are. The rationale of such a definition of debt sustainability is that countries with

large proportions of its population living below the poverty line have a more urgent need

to spend their resources on poverty reduction than on debt service.

For analytical purposes, a more narrow definition of debt sustainability is whether

a country can meet its current and future debt service obligations in full, without recourse

to debt relief, rescheduling, or accumulation of arrears. Most of the theoretical debt

sustainability literature looks at the debt dynamics over an infinite horizon and derives

then some kind of solvency conditions according to which debt sustainability can be

determined. However, given the practical limitations of such infinite horizon solvency

conditions, practitioners have suggested a variety of more specific debt sustainability

indicators (please see Box 1). Recently, Gunter, Lopez, Ramadas and Wodon (2002) have

developed a simulation tool, called SimSIP Debt, which contains two Excel.-based

worksheets that can be used for the practical analysis of debt sustainability issues.

2

Following the terms of references, this paper concentrates on a review of the

Heavily Indebted Poor Country (HIPC) Initiative.

3

After providing some historical and

analytical background, it reviews some of the key problems of the HIPC Initiative and

makes a variety of concrete suggestions on how the framework of the HIPC Initiative will

need to be adjusted to overcome the current implementation problems. However, even

after the adoption of such changes would the long-term debt sustainability of HIPCs

remain fragile. The paper thus addresses some of the key issues related to (a) a new aid

architecture and (b) the structural transformation HIPCs must undergo to achieve long-

term debt sustainability.

1

The human development approach to debt relief has originally been suggested by Northover, Joyner and

Woodward at CAFOD in 1998; see

http://www.cafod .org.uk/policy/acaf1.shtml

for more details.

2

The Debt Projection Module can be used to simulate a country’s debt sustainability based on initial

conditions and projections for government expenditures, government revenues, and other parameters.

Looking at levels and trends of various debt sustainability indicators, it provides policymakers and other

interested parties with indications whether a country’s debt is likely to be sustainable over a 15-year

horizon. The Deficit-Debt Consistency Module presents a variety of matrices to determine the consistency

of overall budget deficits with a country’s desired level of indebtedness for a given growth rate. This is

especially useful for fiscal aspects of public debt sustainability.

3

This review is more compact and more focused than the World Bank’s (2003) Operations Evaluation

Department review of the HIPC Initiative, especially as the OED review does not make as concrete

suggestions on how to change the framework of the HIPC Initiative as they are presented here.

2

Box 1: Practitioners’ suggestions of debt sustainability indicators

The following are some examples of debt sustainability indicators as they are used

in the World Bank’s Global Development Finance (GDF), the Millennium Development

Goals (MDGs), and the European Union’s (EU) Maastricht Treaty. The debt

sustainability indicators of the HIPC Initiative are described and discussed in more details

below.

•

The World Bank’s Global Development Finance (GDF, formerly World Debt

Tables) classifies external indebtedness based on two ratios, the ratio of the net

present value (NPV) of total external debt (calculated based on all future debt service)

to the three-year backward looking average of gross national product (GNP), and the

ratio of the NPV of total external debt (calculated based on all future debt service) to

the three-year backward looking average of exports of goods and services (including

workers’ remittances). If either ratio exceeds a critical value—80 percent for the NPV

debt to GNP ratio and 220 percent for the NPV debt to exports ratio—the country is

classified as severely indebted. If the critical value is not exceeded but either ratio is

three-fifths or more of the critical value (that is, 48 percent for the present value of

debt service to GNP and 132 percent for the present value of debt service to exports),

the country is classified as moderately indebted. If both ratios are less than three-

fifths of the critical value, the country is classified as less indebted.

a/

•

Within the recently enlarged set of Millennium Development Goals (MDGs), target

15 is defined as to deal comprehensively with the debt problems of developing

countries through national and international measures in order to make debt

sustainable in the long-term. The four indicators for this target are (a) the proportion

of official bilateral HIPC debt cancelled, (b) debt service as a percentage of exports of

goods and services, (c) the proportion of ODA provided as debt relief, and (d) the

number of countries reaching HIPC decision and completion points.

b/

•

The European Union’s (EU) Maastricht Treaty (signed in early 1992) limited the

ratio of government debt to GDP to 60 percent, though it was also agreed that higher

ratios are acceptable as long as the debt to GDP ratio is sufficiently falling over time.

Indeed, most countries of the EU had debt to GDP ratios above 60 percent for most of

the times during the 1990s, and at least three countries (Belgium, Greece, and Italy)

had debt to GDP ratios of more than 100 percent. Anyway, it should be stressed that

the Maastricht Treaty’s debt to GDP ratio should not be interpreted as debt

sustainability indicator, but as convergence criteria set by a group of European

countries that intended to adopt a single currency by the end of 2001.

__________________

a/. Please see

http://www.worldbank.org/prospects/gdf2002/

for more detailed information.

b/. Please see

http://www.undp.org/mdg/

for further information.

Source:

Adapted from Gunter, Bernhard G., Humberto Lopez, Krishnan Ramadas and Quentin

Wodon, SimSIP Debt: A Simulation Tool for Analyzing Debt Sustainability, Washington, DC:

World Bank (July 2002); available on the internet:

http://www.worldbank.org/simsip

.

3

II. Historical and Analytical Background

II.1. The History of Traditional Debt Relief

Since the early 1980s, more and more low-income countries became unable to

pay their scheduled debt service and were thus repeatedly allowed to reschedule their

debt at increasingly concessional terms. Following the Group of Seven (G-7) summits in

Toronto (1988), London (1991), and Naples (1994), the agreements reached at these three

summits constitute what is called “traditional” debt relief, providing, respectively, 33

percent (Toronto terms), 50 percent (London terms), and 67 percent (Naples terms) debt

relief in net present value (NPV) terms on eligible debt.

4

Since this 16

th

Technical Group

Meeting is here in Trinidad, it may be noted that the 1994 adopted Naples terms were

originally proposed by John Major (then UK Chancellor of the Exchequer), at the

September 1990 Commonwealth Finance Ministers' Conference here in Trinidad.

II.2. Analytical Background

Before that, two seminal contributions by Krugman (1988) and Sachs (1989)

showed that debt relief is more efficient than the provision of new loans in cases where

debt has accumulated beyond some critical level (point A in Figure 1). Debt accumulated

beyond point A represents a debt overhang, which stifles private investment, growth and

development. Starting in the early 1990s, a large empirical literature has confirmed the

existence of a debt overhang in many low-income countries.

5

Figure 1: Relationship between debt and GNP growth

GNP growth

debt

A

4

As will be explained in more details in section III.6, the net present value (NPV) is the sum of all future

debt service discounted by currency specific discount rates. Eligible debt was defined as pre-cutoff-date,

non-ODA (official development assistance) bilateral debt. The cutoff-date refers to the date a debtor

country approached the Paris Club for the first time for a debt rescheduling.

5

See Gunter (2002) for a list of the early empirical studies. Bourne and Nicholls (1990) provided one of the

earliest econometric studies for Barbados and Trinidad & Tobago. More recent empirical studies have been

presented at the 2001 WIDER Conference on Debt Relief: see Bigsten, Levin, and Persson (2001);

Chowdhury (2001); Dijkstra and Hermes (2001); Gunter (2001); Hansen (2001); Pattillo, Poirson and Ricci

4

II.3. The Emergence of the HIPC Initiative

Based on these developments and the continuous failure of traditional debt relief

to end the repeated process of debt rescheduling, the International Monetary Fund (IMF)

and the World Bank adopted the Highly Indebted Poor Country (HIPC) Initiative in fall

1996 (following the G-7 summit in Lyon).

6

The key goal of the HIPC Initiative has been

to reduce the public and publicly guaranteed external debt to a level that would allow

HIPCs to permanently exit the process of repeated debt rescheduling. Three years after

launching the HIPC Initiative, it was clear that the original HIPC framework was not

sufficient to provide HIPCs with a permanent exit from repeated debt rescheduling, and

due to public pressure (and the agreement reached at the G-7 summit in Cologne),

7

the

IMF and World Bank formally agreed in September 1999, to enhance the HIPC

framework.

The enhancements provide broader, deeper and faster debt relief mainly through

(i) a lowering of the NPV debt ratios considered by some to provide debt sustainability

(together with a lowering of the minimum thresholds to qualify under the fiscal window),

(ii) replacing the principally fixed three-year period between decision and completion

points

8

by the concept of a floating completion point, and (iii) the provision of interim

debt relief from some creditors during the period between decision and completion

points. Another key enhancement was to link HIPC debt relief to the preparation of a

country-owned poverty reduction strategy paper (PRSP).

9

II.4. HIPC-III, PRSP, and/or International Debt Workout Mechanism.

The period since the adoption of the enhanced framework has been characterized

by three developments. First, evidence has been mounting that even the enhanced HIPC

framework is insufficient to provide even short-term debt sustainability for many of the

poorest countries, of which some are considered to be HIPCs

10

and others are excluded

from the official group of HIPCs. While the many serious problems of the enhanced

(2001); and Serieux and Samy (2001). All these more recent empirical studies are available on the WIDER

website, and will also be available in Addison, Hansen, and Tarp (forthcoming).

6

Limited to eligible HIPCs, the so-called Lyon terms provided 80 percent debt relief in NPV terms on

eligible debt.

7

Limited to eligible HIPCs, the so-called Cologne terms provided usually up to 90 percent debt relief in

NPV terms on eligible debt. However, given that some HIPCs had started the rescheduling process in the

early 1980s, even 100 percent debt relief on eligible debt was sometimes not sufficient to reach the

enhanced HIPC debt sustainability criteria.

8

The HIPC Initiative involves two stages. The first stage is a three-year period during which a HIPC works

in coordination with the support of the World Bank and the IMF to establish a record of good economic

policies and sustained poverty reduction. At the end of this three-year period the IDA and the IMF

determine whether a country's debt level is sustainable. For those countries whose debt burden remains

unsustainable after full use of traditional debt relief mechanisms, a package of debt relief is identified. This

is known as the Decision Point. While full HIPC debt relief will be provided at the Completion Point,

some creditors might provide interim debt relief (the period between the Decision Point and the floating

Completion Point). Under the enhanced framework, the completion point is “floating” as it is tied to the

implementation of key structural reforms and poverty reduction policies.

9

There are some doubts if conditioning debt relief to the preparation of a PRSP can be considered an

enhancement, especially in cases where countries already had poverty reduction strategies in place (e.g.,

Uganda), but were then required to qualify subsequently for the enhanced decision and completion points.

10

For example, see GAO (2000), GAO (2002), Martin (2002), and Serieux (2001).

5

framework (HIPC-II) would more than justify further enhancements (i.e., HIPC-III), the

fact that HIPC-II continues to face severe financing problems have prevented the IMF

and World Bank from suggesting any further enhancements in the HIPC framework

[though there were some discussions, see IMF and International Development

Association (IDA) (2002), pp. 34-36]. Related to the emerging consensus that the HIPC

Initiative is unlikely to provide debt sustainability is the emerging view [see Johnson

(2000) and Martin (2002)] that domestic debt needs to be taken into account when

analyzing debt sustainability.

Second, the IMF and World Bank have started to shift the task of achieving debt

sustainability away from the HIPC Initiative towards the PRSP framework, “within

which the authorities should seek to maintain a sustainable debt burden.”

11

The way this

important change in the key objective of the HIPC Initiative has been introduced,

indicates that IMF and World Bank staff already know that many HIPCs will not achieve

debt sustainability. Indeed, in April 2001, the IMF and IDA (2001) issued a paper that

recognized for the first time that the HIPC Initiative might not achieve long-term debt

sustainability. The paper acknowledged explicitly that the NPV debt-to-export ratio is

projected to remain above 150 percent for 10 years or more for at least 3 HIPCs (Bolivia,

Malawi, and Niger). Surprisingly, the paper nevertheless concluded that the HIPC

Initiative provides a good basis for HIPCs to exit from future debt rescheduling. In

September 2002, the IMF and IDA (2002, p. 32) cemented a new language (which they

began to introduce in April 2001), stating that “the HIPC Initiative is designed to deal

with (…) the existing stock of debt (…) at a given point in time” and “debt relief under

the HIPC Initiative provides a basis, but no guarantee, for long-term debt sustainability in

HIPCs.” While it is pretty obvious that the HIPC Initiative cannot guarantee long-term

debt sustainability, the stated goal of the Initiative was “to achieve a sustainable debt

situation”

12

in the eligible countries at least in the short- and medium term; a goal which

the HIPC Initiative has clearly not achieved for at least some of the countries that have

exited the HIPC process, especially Uganda.

Third, related to recent debt problems in middle-income countries (like Argentina

and Brazil), the last two years have seen an increasing interest in the adoption of a much

broader international debt workout mechanism. While the IMF has called for a Sovereign

Debt Restructuring Mechanism (SDRM), most international advocacy groups have called

for a Fair and Transparent Arbitration Procedure (FTAP).

13

It is not clear yet how the

HIPC Initiative will be related to these broader suggestions, though it is likely that HIPCs

will be covered under such debt workout mechanisms if they continue to face debt

service problems after having completed the HIPC process.

11

See IMF and IDA (2002), p. 39.

12

See World Bank (2003), referring to the original HIPC concept paper of August 1996: The HIPC Debt

Initiative—Elaboration of Key Features and Possible Procedural Steps.

13

For example, see Herman (2003); the latest developments related to the SDRM can be obtained from the

IMF website (

http://www.imf.org

), the latest information on the FTAP can be obtained from various NGO

websites.

6

III. Key Problems of the Enhanced HIPC Initiative (HIPC-II)

While the recent debt relief literature addresses many more problems of the

enhanced HIPC Initiative,

14

the following six are likely the most crucial ones:

•

inappropriate eligibility criteria,

•

unrealistic growth assumptions,

•

insufficient provision of interim debt relief,

•

delivering HIPC debt relief through debt rescheduling,

•

lacks in creditor participation and financing problems, and

•

currency-specific short-term discount rates.

While we will explain each of these six problems in more details below, the

important point is that the combination of these six problems makes it unlikely that the

enhanced HIPC Initiative will remove the debt overhang, hence will not even provide

short-term debt sustainability. There might be some advantages from shifting the goal to

achieve long-term debt sustainability from the HIPC Initiative to the PRSP process;

15

however, it needs to be stressed that this may ultimately prove inefficient and ultimately

ineffective in cases where countries exit the HIPC process before they have reached debt

sustainability. Although increased social expenditures may help to reduce poverty,

private investors are unlikely to invest in a country as long as there are doubts that the

country has achieved debt sustainability. Consequently, the country will continue to grow

far below its potential.

Hence, it is crucial that the HIPC Initiative provides sufficient debt relief to

convince private investors that countries exiting the HIPC process will remain debt

sustainable for the foreseeable future. The recent developments in Uganda clearly show

that this is unfortunately not the case.

16

Only if the debt overhang is removed, is it likely

that increased social expenditures will be complemented by increased long-term private

investment, which is needed to undergo a structural transformation. Without long-term

private investment, it is unlikely that an economy will undergo the necessary structural

transformation (including a diversification of its exports), and thus, the vicious circle of

low levels of private investment, low degrees of export diversification, high vulnerability,

low growth, and high debt ratios is likely to prevail.

III.1. Inappropriate Eligibility Criteria

III.1.a. Eligibility criteria of the enhanced HIPC Initiative

The enhanced HIPC framework defines a country as “heavily indebted” if

traditional debt relief mechanisms are unlikely to reduce a country’s external debt to a

sustainable level. The HIPC framework assumes that a country’s external debt is

14

See for example, Birdsall and Williamson (2002), Gunter (2002), Martin (2002), Roodman (2001), and

the most recent Operation Evaluation Department (OED) review: World Bank (2003).

15

Like for example, a closer alignment of debt reduction with poverty reduction, and a much less explicit

determination of costs to achieve debt sustainability.

16

The most recent HIPC DSA of July 2002 has shown that due to the unanticipated decline in world coffee

prices, Uganda’s NPV debt to export ratio will remain considerably over the 150 percent level for at least

the next 10 years, for some years even above 200 percent. It should also be pointed out that the official

HIPC DSA has not even taken into account that 13 of Uganda’s creditors have not yet agreed to provide

HIPC debt relief; see IMF and World Bank (2002), p. 68.

7

sustainable if the NPV debt-to-export ratio is around 150 percent (which is called the

export criterion). In cases where a country has both (a) an export-to-GDP ratio of at least

30 percent and (b) a government revenue-to-GDP ratio of at least 15 percent, the HIPC

framework considers also a fiscal window, whereby it is assumed that a country’s debt is

sustainable if the NPV debt-to-government revenue ratio is around 250 percent. The

criterion for being “poor” is to be an “IDA-only” country, which is defined as a country

that relies on highly concessional financing from the World Bank’s concessional lending-

arm (IDA).

17

Hence, the HIPC initiative was not intended to solve the debt problems of

all countries; instead, it was intended to assist the poorest and most heavily indebted

countries. While questions may be raised about the logic of solving the debt problem of

only the poorest countries, this is not the key criticism related to the HIPC eligibility

criteria. The key criticism related to HIPC eligibility is that the current eligibility criteria

are neither based on a comprehensive measure of poverty nor on a comprehensive

measure of indebtedness.

III.1.b. HIPC eligibility is not based on poverty

The HIPC framework’s poverty criterion (the largely income per capita-

determined “IDA-only” criterion) completely neglects that poverty is a multi-dimensional

phenomena. Even worse, since the “IDA-only” criterion is mainly based on nominal GDP

per capita, the HIPC framework’s poverty criterion does not even allow for differences in

purchasing power. Furthermore, the reclassification of Nigeria, (which is poor and highly

indebted by any standard) from an “IDA-only” country to a so-called blend country has

shown that political and cost factors also played a significant role in the determination of

HIPC eligibility.

III.1.c. HIPC eligibility is not based on indebtedness

There is a widespread critique that the HIPC Initiative uses inappropriate debt

sustainability criteria. For example, Sachs (2000) has expressed the view that the HIPC

sustainability criteria have nothing to do with debt sustainability in any real sense. Others

have stressed that “the ratios of debt and debt service to exports (…) are hard to justify on

theoretical grounds” and that “at the very least, indicators relative to GDP should be

taken as seriously as indicators relative to exports.”

18

While completely ignored in the

HIPC framework, the NPV debt-to-GNP ratio is a good overall indicator of a country’s

indebtedness. It also has the advantages of being (a) less volatile than the NPV debt-to-

exports indicator and (b) more easily available than the NPV debt-to-government revenue

indicator.

In any case, the required thresholds for the fiscal window (the NPV debt-to-

revenue ratio) are far too restrictive, and the history behind this also called “Cote

17

Originally, Nigeria and Equatorial Guinea were considered to be HIPCs, but have been dropped from the

list of HIPCs as they were later on considered to be no more IDA-only countries. On the other hand,

Malawi, The Comoros, and The Gambia have been added, as it became clear that their debt is higher than

initially estimated. For the current list of HIPCs and a more detailed description of the HIPC initiative, see

the World Bank’s HIPC website:

www.worldbank.org/hipc/

.

18

See IMF, Report of the Group of Independent Persons Appointed to Conduct an Evaluation of Certain

Aspects of the Enhanced Structural Adjustment Facility, June 1998, pp. 39-40 (available on IMF website).

8

d’Ivoire” criterion,

19

shows that the HIPC framework’s debt sustainability criteria were

also heavily influenced by political considerations. While more than half of the eligible

HIPCs have a NPV debt-to-revenue ratio of more than 250 percent, most do not qualify

for HIPC debt relief under the fiscal window due to the threshold requirements. While

most of these fiscally unsustainable HIPCs still qualify for HIPC debt relief under the

export criterion, the debt relief provided is usually far below of what would be needed to

obtain fiscal debt sustainability. As a World Bank (2001) report shows,

at least four

countries (Guinea, Mauritania, Niger, and Zambia) will continue to pay more than 20

percent of government revenues as external debt service even after enhanced HIPC debt

relief.

The following remarks show how inappropriate the NPV debt-to-export ratio is to

determine the debt sustainability of low-income countries, and especially of import- and

aid-dependent economies like HIPCs. As is well known, the debt-to-export ratio has been

used for mostly middle-income Latin American countries in the aftermath of the 1982

debt crisis.

20

However, the situation there was different to that of the HIPCs. First, a

substantial part of Latin American debt was private debt, and second, exchange rate

devaluations following the outbreak of the crisis lead to substantial trade surpluses with

which the subsequent negative net resource flow was financed. This obviously cannot be

the solution for HIPCs, and even if HIPCs would be forced to cut their imports and to

increase their exports, such a strategy would not work for HIPCs as HIPC governments

usually get only a small proportion of the export revenues.

21

For example, multinational

enterprises control close to 90 percent of Guinea’s exports and use most of the foreign

exchange earnings for imports of equipment, salaries of expatriate workers, and transfers

of profits. In same cases, exports of HIPCs reflect a large degree of re-exports (the

exports simply pass through the country and no foreign exchange is earned by

anybody).

22

Finally, the way the NPV debt-to-export criterion is currently used in the

HIPC framework discourages an export-led growth strategy, especially in HIPCs where

decision and/or completion points are some time in the future.

23

In conclusion, while the

debt-to-export ratio may have some justification for the determination of an upper limit

of a country’s debt sustainability, it says very little about a government’s ability to repay

its external (as well as domestic) public debt.

III.1.d. Examples of better HIPC eligibility criteria

Using more comprehensive measures of poverty and indebtedness, would give us

a considerably different group of HIPCs. For example, using (a) the UNDP’s human

19

See Martin (2002), p. 3, reporting that the NPV debt-to-revenue ratio “was set at a level just low enough

to include Cote d’Ivoire in HIPC (under pressure from France which was threatening to vote against the

eligibility of other HIPCs), but was accompanied by empirically unjustified sub-criteria which exclude

many other HIPCs.”

20

See Hjertholm (2001) for the analytical history of HIPC debt sustainability targets.

21

These budgetary aspects of the transfer problem have been analyzed by Dani Rodrik and especially by

Helmut Reisen, see Hjertholm (2001) for further references.

22

Thus far, the HIPC framework has not been consistent in either including or excluding re-exports in the

calculation of the debt-to-export criteria.

23

Given that debt relief is in most cases determined on a debt-to-export ratio, the higher a HIPC’s exports,

the lower a HIPC’s debt relief, hence a HIPC has incentives to keep exports low.

9

poverty index for developing countries (HPI-1)

24

and the NPV external debt-to-GNP

ratio

25

as reference criteria for poverty and overall external indebtedness, the following

countries are both poorer and more indebted than the two highest ranking eligible

HIPCs:

26

Algeria, Angola, Cambodia, Cape Verde, Djibouti, El Salvador, Gabon,

Indonesia, Kenya, Lesotho, Mongolia, Morocco, Nepal, Nigeria, Pakistan, Papua New

Guinea, Sri Lanka, Syrian Arab Republic, Vietnam, Yemen, and Zimbabwe.

27

Note that

this list includes (a) all four HIPCs (Angola, Kenya, Viet Nam, and Yemen) that (based

on inappropriate debt sustainability criteria) are expected to have a sustainable debt, and

(b) six members of the (G-24).

28

There are obviously many other criteria that could be

considered and would give us different results in terms of country-eligibility. It is not

argued that all the countries listed above should be added to the current list of eligible

HIPCs. The point however is that the current HIPC eligibility criteria neither cover the

poorest nor the highest indebted countries.

III.2. Overly Optimistic Growth Assumptions

Growth assumptions are important for any long-term debt sustainability analysis.

Overly optimistic growth rates affect the result of the debt sustainability analysis in two

ways: first, they inflate the denominator of any relative debt sustainability indicator, and

second, they underestimate a country’s future financing needs/new debt, which is in the

numerator of any relative debt sustainability indicator. Hence, even slightly over-

optimistic growth rates can result in highly unrealistic projections of low debt ratios. For

example, the GAO (2000) report (p. 15) points out that if Tanzania’s exports grow at an

annual 6.5 percent (instead of the 9 percent projected by the IMF and World Bank),

Tanzania’s debt-to-export ratio could be more than twice of what the joint IMF/World

Bank’s forecast shows for the projection period. Furthermore, the overly optimistic

growth assumptions have led to underestimations of the costs of the enhanced HIPC

Initiative, which will need further corrections in the years to come. Hence, the costs of

the enhanced HIPC Initiative are likely to increase further.

III.3. Insufficient Provision of Interim Debt Relief

The enhancements of the HIPC Initiative of fall 1999 recommended the provision

of interim debt relief (HIPC debt relief provided during the interim period, i.e., the period

between reaching decision and completion points). While it was anticipated that interim

debt relief will provide substantial reduction in actual debt service payments to support

24

The HPI-1 has been designed to address the now-well accepted notion that poverty is multi-dimensional,

and certainly goes far beyond income-poverty; the data has been taken for two consecutive years of the

Human Development Report, 2001 and 2002.

25

The data for the NPV debt to GNP ratios have been taken for two consecutive years of the World Bank’s

Global Development Finance, 2001 and 2002.

26

In terms of poverty, the two highest-ranking (least-poor) eligible HIPCs are Bolivia and Guyana; in terms

of indebtedness, the two highest-ranking (least-indebted in terms of NPV debt-to-GNP) eligible HIPCs are

Burkina Faso and Mozambique.

27

Given that the Human Poverty Index is not available for Angola and Gabon, the Human Development

Index has been used as a substitute.

28

Within the G-24 members, there are currently two countries that have reached their enhanced HIPC

decision point (Ethiopia and Ghana) and two countries that are still expected to be eligible for HIPC debt

relief (Democratic Republic of Congo and Cote d’Ivoire).

10

the ambitious poverty reduction strategies necessary to reach the completion point, the

experience so far has not confirmed this anticipation.

Figure 2 shows that actual 2001 debt service payments of the 22 HIPCs that

reached their Enhanced Decision Point by end-2000 is less than 10 percent lower than the

1998-2000 average. When comparing these actual payments, we should keep in mind that

most HIPCs had increased their actual debt service payments since the adoption of the

HIPC Initiative, as prospective HIPCs were not allowed to accrue arrears. Hence, the

actual 2001 debt service payments for these 22 HIPCs (US$3.4 billion) have been higher

than the average actual debt service payments of these 22 HIPCs just before the adoption

of the HIPC Initiative, 1993-95 (US$3.3 billion).

These comparisons of actual debt service payments show clearly how marginal

interim debt relief has been thus far, especially if considering that by 2001, some HIPCs

had already reached their Completion Points. Indeed, thus far, substantial interim debt

relief has only been provided by the IMF, the World Bank,

29

and the Paris Club.

30

Source: World Bank, Global Development Finance 2003.

Finally, while the amount of interim debt relief was already too little before the

recent slow down of the world economy, it is clearly inadequate in the current economic

situation, and thus, many HIPCs have considerable difficulties to reach their completion

points. As of end-February 2003, that is, more than three years after the adoption of the

enhanced framework (which was supposed to shorten the time to reach the completion

point), only six HIPCs have reached their enhanced completion points.

III.4. Delivery of HIPC Debt Relief Through Debt Rescheduling

Given that the key goal of the HIPC initiative intended to provide a lasting

solution to the problem of repeated debt rescheduling, the clear preference of the HIPC

29

The World Bank’s assistance is heavily back-loaded in nominal terms, increasing from US$318 million

in 2001 to 544 million in 2008; see Appendix Tables 7 of the September 2002 Status Report, IMF and IDA

(2002).

30

The Paris Club has provided some flow rescheduling on eligible bilateral debt.

Figure 2: Paid de bt se rv ice of 22 HIPCs

(that reached enhanced HIPC Decision Point by end-2000

0

1

2

3

4

5

1990 1992 1994 1996 1998 2000

Paid debt service

11

Initiative should have been to either cancel some part of the debt stock or to cancel at

least a part of future debt service payments. However, due to legal and political

constraints of some creditors to cancel debt stock or debt service, it was agreed that

creditors could provide their NPV debt relief through a rescheduling of debt service.

Following previous Paris Club arrangements, this rescheduling applied also to

repayments of official development assistance (ODA), which was to be rescheduled over

40 year (including 16 year grace) at original interest rates.

The main problem with such a rescheduling is that it actually increases the total

debt service a country has to pay in the long-term. Hence, though HIPC debt relief

provided through rescheduling reduces debt service payments in the short-term, it pushes

most of the debt service payments to future, and will thus undermine long-term debt

sustainability (especially if combined with overly optimistic growth rates).

31

It is

unfortunate that this rescheduling problem is largely ignored or misinterpreted in the debt

relief literature. A debt rescheduling may be appropriate for borrowers with temporary

payment problems, but is no solution for HIPCs. To the contrary, a debt rescheduling

makes the situation worse as it increases the cumulative amount of debt service

payments.

32

III.5. Lacking Creditor Participation and Financing Problems

The status of creditor participation and some of the financing problems have been

addressed in some details in the latest HIPC Status Report (see IMF and IDA (2002), pp.

12-21). The degree of creditor litigations against HIPCs is further illustrated in Annex III

of the latest HIPC Status Report. Given that there is broad agreement on the need for full

participation by all creditors to ensure debt sustainability within the HIPC framework,

our presentation here concentrates largely on explanations for the lacks in creditor

participation and related financing problems. Before addressing more specific

participation and financing problems in the following two sub-sections, it is useful to

look at the overall breakdown of the potential costs of the HIPC Initiative (see Table 1).

First, as Table 1 shows, the costs for Paris Club members are estimated at

US$14.3 billion. The costs for other official bilateral creditors are estimated at 3.2 billion.

However, to understand this breakdown, it is important to know that the Paris Club is not

a fixed group of countries, but depends on the debtor country’s main creditors. Hence, in

same cases, the Paris Club includes some developing country creditors (like Brazil,

Israel, South Africa, and Trinidad and Tobago). Indeed, based on the latest available

estimates, the top 12 bilateral creditors include three developing countries (Costa Rica,

Brazil and Guatemala, see Table 1). Second, the claims of most bilateral and commercial

31

A more technical problem is that the rescheduling of ODA debt repayments does usually not provide

much NPV debt reduction. In cases where the original interest rate on ODA is higher than the discount rate,

the rescheduling of repayments of ODA debt actually increases the NPV. The burden to achieve the overall

NPV debt reduction falls thus on non-ODA debt. In addition to the problem of a large portion of non-

eligible debt (see footnotes 4 and 7), this has in some cases contributed to the result that a complete

cancellation of all eligible debt was not sufficient to reach the required NPV debt reduction.

32

Note that Easterly’s (1999) characterization that substantial debt relief has been provided to HIPCs

before the HIPC Initiative is misleading as (a) a large part of traditional debt relief has been rescheduled

(hence, provided a NPV debt reduction but no debt relief on the total debt service) and (b) traditional debt

relief was only provided on eligible debt (pre-cut-off date and non-ODA debt).

12

creditors are actually much higher than the numbers provided in Table 1, since the HIPC

costs are calculated on the assumption that all bilateral and commercial creditors have

already provided a 67 percent debt reduction in NPV terms (which is obviously not the

case).

_______________________________________________________________________

Table 1. Breakdown of Potential Costs of the Enhanced HIPC Initiative

by Main Creditors (US$ billion, in 2001 NPV terms)

Bilateral creditors 17.6 Multilateral creditors 17.9

of which:

of which:

Japan 2.8 World Bank 8.2

France 2.3 African Development Bank/Fund 3.2

Germany 1.4 IMF 2.7

Russia 1.1 Inter-American Development Bank 1.2

Italy 1.0 Other multilateral creditors 2.6

Spain 0.7

United Kingdom 0.6 Commercial creditors 1.6

Costa Rica 0.4

Portugal 0.4

United States 0.3

Brazil 0.3

Guatemala 0.3

Other bilateral creditors 6.0

Source: IMF and IDA (2000) and (2002). Data for individual bilateral creditors are

estimated based on combining data provided in Table 3 of IMF and IDA (2002) with

Tables 6, 10, and 11 from IMF and IDA (2000).

_______________________________________________________________________

III.5.a. Non-participation of some bilateral and commercial creditors

The HIPC Initiative does not constitute international law. HIPCs are therefore

obliged to pay their debt service in full until individual agreements have been signed with

each multilateral, bilateral, and commercial creditor. Given that many creditors are

formally or de facto

33

excluded from the decision making process in the Bretton Woods

institutions (especially commercial creditors), the non-participation of such creditors is

not necessarily a surprise.

Furthermore, no provision has been made to exempt creditors that are themselves

HIPCs from the burden sharing of the HIPC Initiative. Hence, from the total of 24

bilateral creditors that have not yet agreed to deliver HIPC relief, nine are HIPCs, and at

33

While the membership of the IMF and IDA include some 180 countries, most of the developing

countries’ voting power is so marginal that they are “de facto” excluded from the decision making process.

For recent reform proposals of the IMF voting power, see Buira (2002).

13

least another 3 would be HIPCs if using more appropriate eligibility criteria. Due to lacks

in consulting these bilateral and commercial creditors before the adoption of the HIPC

Initiative, a variety of these creditors have either sold their claims on the secondary

market (from which litigations are underway or expected) or have themselves initiated

litigations against HIPCs. Given that most of these creditors have not provided any debt

reduction so far, the claims are usually a multiple of the official HIPC costs. As is the

case with most litigations, looking at the amount of resources creditors and debtors need

to allocate for these litigations, they constitute highly unproductive rent seeking.

III.5.b. Financing short-falls for some parts of multilateral debt relief

Including the potential debt relief costs for Liberia, Somalia, and Sudan, all

multilateral creditors face some financing problems, including the IMF and the World

Bank. The financing constraints are especially severe for the small regional multilateral

development banks (MDBs), of which some did not even have enough resources to

provide their share of original HIPC assistance.

34

Anyway, we can use the World Bank’s

financing plan to illustrate the two key problems related to financing constraints of

MDBs.

First, based on a legal arrangement within the World Bank, it has been determined

that the IDA will cover all of the World Bank’s HIPC assistance, comprising debt relief

on IDA debt as well as debt relief of debt to the World Bank’s non-concessional

financing arm, the International Bank for Reconstruction and Development (IBRD). The

main reason behind this legal provision has been to protect the IBRD’s creditor rating.

Obviously, most other MDBs do not have the ability for similar arrangements, and have

thus experienced deteriorations in their creditor ratings.

Second, given that IDA does currently not have all the money to finance the

World Bank’s share of enhanced HIPC assistance, it was decided to use a pay-as-you-go

approach. Initially, the World Bank’s debt relief will be covered by the HIPC Trust Fund

(to which parts of the IBRD net income had been allocated in the past). However, once

these resources are exhausted, the World Bank’s debt relief will come out of the then

current IDA budget. The World Bank estimates that from 2005 onwards, an annual

US$500 million will need to be taken out of then current IDA budgets. Though World

Bank management stressed that resources taken out of IDA budgets are supposed to be

returned to IDA in subsequent IDA replenishments, there are some realistic fears that

future donor contributions to IDA may not fully compensate the debt relief IDA is

providing on behalf of the World Bank. Hence, it is likely that traditional IDA

development projects will need to be reduced.

34

For example, though Uganda had reached its completion point under the original framework in April

1998, it is still waiting to receive original HIPC assistance from the East African Development Bank and

the PTA Bank.

14

III.6. Currency-Specific Short-Term Discount Rates

The discount rates used to calculate the net present value (NPV)

35

are the

currency-specific commercial interest reference rates (CIRRs), provided by the

Organization for Economic Co-operation (OECD), for its member countries’ currencies,

based on commercial lending rates.

36

We call them “short-term” CIRRs, as they averaged

over the last six-month period before the reference date of the DSA. Before illustrating

how currency-specific short-term CIRRs give rise to arbitrary results with regard to

assistance levels and costs of the HIPC Initiative, we first illustrate how sensitive the

NPV calculations are to differences in discount rates. For example, consider a US$10

million loan repayable at an interest rate of 4 percent over 40 years (including 10 years

grace). Using a discount rate of 6 percent results in an NPV of US$7.5 million. Using a

discount rate of 2 percent results in an NPV of US$13.9 million (nearly twice as much as

with the 6 percent discount rate). In other words, every percentage point difference in the

discount rate implies a change in the NPV of about 16 percent.

The way the HIPC framework decided to use these CIRRs entails four problems:

•

currency-specific short-term CIRRs imply an unfair burden sharing,

•

the averaging of CIRRs over 6 months give rise to high volatility in estimations of

assistance and costs,

•

the averaging of CIRRs over 6 months implies unfair assistance levels, and

•

the lack of clear rules as to what CIRR to use for non-OECD currencies has led to

arbitrary uses of discount rates.

III.6.a. Currency-specific short-term CIRRs imply an unfair burden sharing

Currency-specific discount rates imply a highly unfair burden sharing whereby

booming creditor countries get rewarded and creditor countries facing or recovering from

recession get punished. This is because booming economies have generally higher CIRRs

(like the United States before the recession, where the CIRR for the US dollar averaged

around 6 percent) than countries in or recovering from a major recession (like Japan,

where the CIRR for the Yen averaged at slightly above 2 percent).

III.6.b. Short-term CIRRs cause a high volatility in estimations of assistance and costs

As interest rates are changing over time, any change in the DSA reference date

implies changes in the discount rates, and thus represents changes in NPV of HIPC debt

and HIPC assistance. As Uganda’s reassessment under the enhanced HIPC Initiative

showed, high discount rates and too optimistic export projections can make a country’s

debt look sustainable, even though it is not. (Based on Uganda’s original completion

point DSA, Uganda’s NPV debt-to-export ratio at end-June 1999 was supposed to be 207

35

In order to understand the significance of this concept, consider the example of two country cases.

Country A has a debt outstanding of nominal US$100 million, which is all due in the next year. Country B

has a debt outstanding of nominal US$120 million, which is interest free and due in 10 years. Which

country would you prefer to be. While country A's nominal debt (100 million) is lower than country B's

nominal debt (120 million), the fact that country B has no principal and no interest due for the next 10 years

make country B's debt far more attractive. Indeed, the NPV of country A’s debt would be US$100 million,

while the NPV of country B’s debt would be less than US$70 million (using a discount rate of 6 percent).

36

The CIRR of the Special Drawing Right (SDR) is calculated based on its composite currencies’ CIRRs.

15

percent; however, as Uganda’s enhanced decision point DSA showed, the actual NPV

debt-to-export ratio was 240 percent.) Together with volatile exports (and/or volatile

government revenues) this methodology results in highly unpredictable data on the costs

and assistance levels of the HIPC Initiative.

III.6.c. Averaging of CIRRs over 6 months implies unfair assistance levels

The averaging of CIRRs over 6 months implies unfair assistance levels as HIPCs

assessed during periods of relatively high world interest rates will ceteris paribus receive

less HIPC assistance than countries assessed during periods of relatively low world

interest rates. Using our numerical example above, every percentage point increase in

world interest rates will reduce HIPC debt relief by about 16 percent.

III.6.d. Arbitrary use of discount rates

The lack of clear rules as to what CIRR to use for non-OECD currencies has led

to arbitrary uses of discount rates in a number of cases. For example, the Preliminary

HIPC document of Tanzania (prepared in October 1999) has used the SDR’s

37

CIRR for

all currencies which do not have an established CIRR; however, the Preliminary HIPC

document of Guinea (prepared in December 1999) has used the US dollar’s CIRR (6.23

percent) for the Chinese Yuan, the French Franc’s CIRR (5.35 percent) for currencies

which are pegged to the French Franc, and the SDR’s CIRR (5.25 percent) for the other

currencies which do not have an established CIRR. In any case, the use of OECD

discount rates for non-OECD currencies is even inconsistent with the attempt to use

interest rate differentials to determine the long-term value of currencies.

IV. Suggested Changes in the Framework of the HIPC Initiative

The goal of this section is to make some specific suggestions on changes in the

framework of the HIPC Initiative. Following the six key problems described in the last

section, we suggest specific changes in six areas:

(1)

revisions of HIPC eligibility and debt sustainability indicators,

(2)

the appropriate use of growth projections,

(3)

the provision of interim debt relief,

(4)

the delivery of HIPC debt relief,

(5)

adjustments in the burden-sharing concept, and

(6)

the appropriate use of discount rates for the NPV calculation.

IV.1. Revisions of HIPC Eligibility and Debt Sustainability Indicators

The replacement of the current eligibility criteria with more poverty-focused and

more fiscal sustainability-based eligibility criteria would imply an improvement,

particularly if HIPC debt relief is linked more closely to the achievement of the

Millennium Development Goals (MDGs). This could be achieved through the following

six changes:

37

Special Drawing Right (SDR) is the IMF's standard unit of account, introduced in 1969. IMF member

countries may use SDRs to settle international trade balances and debts if the member country meets a

variety of conditions. The SDR’s value is based on a basket of the US Dollar, the Japanese Yen, the British

pound, and the Euro.

16

•

First, replace the theoretically inappropriate and politically charged “IDA-only”

criterion with a more poverty-focused poverty criterion, like for example, the

UNDP’s Human Poverty Index for developing countries (HPI-1). We do not suggest

to extend HIPC debt relief to all low-income countries as the low-income criteria

(purely based on nominal GDP per capita) does not really target debt relief at poor

countries.

•

Second, eliminate the two threshold ratios for the applicability of the fiscal window

(i.e., the requirements of having an export-to-GDP ratio of at least 30 percent and a

government revenue-to-GDP ratio of at least 15 percent).

38

•

Third, abolish the inappropriate NPV debt-to-export criterion and concentrate instead

on fiscal debt sustainability criteria.

39

The appropriate combination of a NPV debt-to-

GDP indicator

40

and a NPV debt-to-government revenue indicator could–together

with poverty levels and vulnerability factors–be used to assess a HIPC’s long-term

debt sustainability, upon which the cumulative amount of debt relief could be

determined. An appropriately defined debt service-to-government revenue indicator

could–together with criteria for necessary investments of anti-poverty programs–be

used to determine the maximum annual debt service payments each specific HIPC

can bear.

41

•

Fourth, instead of calculating government revenues based on current or a three-year

backward looking average ending with the year previous to the decision/completion

point, it would make more sense to use a much longer backward looking average

ending with the year before the initiative is adopted. Using a backward looking

average that ends with the year previous to the decision/completion point gives

HIPCs incentives to keep revenues low, as higher revenues imply lower HIPC debt

relief. There is no such incentive if government revenues are fixed at a pre-

determined level. Hence, there are easy ways to make use of the fiscal criteria without

giving countries incentives to reduce fiscal revenues.

•

Fifth, it is necessary to deepen debt relief for the poorest countries. Hence, the NPV

debt-to-revenue ratio should be reduced. While some critiques have argued that HIPC

debt relief should not be deepened, arguing that the majority of the world’s poor

people live in countries that are not eligible for HIPC debt relief, the fact that the

majority of the world’s poor people live in countries that are not eligible for HIPC

38

While it has been argued that the thresholds are justified to provide some incentives for countries to

increase their exports-to-GDP and revenue-to-GDP ratios, we need to keep in mind that ratios below the

thresholds usually reflect structural problems, which are unlikely to be overcome in the short-term.

Furthermore, given that countries are required to have undergone at least three years of “successful”

adjustment supported by the Bank and the Fund before reaching the HIPC Decision Point, is difficult to

argue (a) that further incentives are needed to determine HIPC eligibility and/or (b) that without these

thresholds, the HIPC Initiative would reward inefficient countries.

39

The recommendation to focus more on government revenue-related indicators is not new. For example,

nearly 15 years ago, Dittus (1989) had analyzed the budgetary dimension of the debt crisis in low-income

Sub-Saharan Africa and suggested that the debt service-to-revenue ratio be assigned a central role.

40

As Birdsall and Williamson (2002) pointed out, a debt criteria based on GDP would avoid rewarding

countries for having failed to collect taxes. However, there is no need to make this the only criteria.

41

This builds on a recent suggestion made by EURODAD (2002), calling for a country-by-country analysis

of how much debt each country can afford to carry without preempting resources available for spending on

a basic level of social service delivery.

17

debt relief simply reflects inappropriate eligibility criteria; it does not constitute an

argument against deeper debt relief. The fact that IDA-only countries like

Bangladesh, Cambodia, Haiti, Nepal, and Tajikistan have a GDP per capita of less

than one-dollar-a-day, make it obvious that these countries do not have their own

resources to repay their external debt, not now and not in the foreseeable future. The

only reason these countries can service their external debt at present is that they

currently receive new loans that are more than sufficient to refinance old debt.

However, this continuous refinancing of old debt with new debt cannot be considered

to constitute debt sustainability.

•

Finally, it might also make sense to take other vulnerability factors into account, like

for example, export concentration and export price volatility (especially as these

factors usually have considerable fiscal implications). Vulnerability factors were

actually taken into account in the original framework of the HIPC Initiative, but have

been dropped in the enhanced framework as Bank and Fund staffs usually disagreed

on how much more debt relief should be provided due to these country-specific

vulnerabilities. Obviously, the official explanation for disregarding vulnerability

factors was related to simplifying the HIPC framework.

IV.2. Using Lower Bounds of Realistic Growth Assumptions

While the Bank and the Fund have recognized that they have used overly

optimistic growth rates in the past, a review of the most recent HIPC Status Report (IMF

and IDA (2002), see especially Annex IV, pp. 78-84) indicates continuous problems, as

there remains an excessive use of past trends to extrapolate future growth rates.

Obviously, historical data can provide some indications for the future, however, there are

better ways to project growth rates. For example, when making projections on future

growth rates of exports, we should analyze the impact of export price volatility, the

diversification of exports, and a variety of other structural aspects as they will be

analyzed in more details below.

Hi storical Historical Projected 90 Percent

Historical Historical Projected 90 Percent

Growth Standard Growth Lower Bound

Growth Standard Growth Lower Bound

1991-01 Deviation 2002-11 on Growth

1991-01 Deviation 2002-11 on Growth

Country

of Growth

over 2002-11 Country

of Growth

over 2002-11

Angol a

3.1 8.5 1.3 -1.8 Lao PDR

6.5 3.6 3.8 0.7

Benin

4.9 3.3 4.1 1.0 Madagascar 2.9 3.5 3.7 0.6

Bolivia

3.7 3.8 4.2 1.1 Malawi

3.9 5.5 4.1 1.0

Burkina Faso

4.8 3.3 3.1 0.0 Mali

4.3 4.9 3.6 0.5

Burundi

-2.4 6.4 2.0 -1.1 Mauritania 4.3 6.1 3.1 0.0

Cameroon

2.8 6.4 2.8 -0.3 Mozambique 8.3 8.1 4.7 1.6

Chad

2.5 7.6 2.3 -0.8 Nicaragua 4.1 7.1 3.5 0.5

Congo, Dem. Rep. -4.2 6.3 0.4 -2.7 Niger

2.9 6.3 3.5 0.4

Congo, Rep.

-0.2 6.5 0.4 -2.7 Rwanda

1.9 11.8 2.8 -0.3

Cote d'Ivoi re

3.4 5.8 3.8 0.7 Senegal

4.3 4.4 3.6 0.6

Ethiopia

5.4 7.4 5.4 2.3 Sierra Leone -2.6 6.1 1.9 -1.2

Gambia, The

3.5 3.3 3.7 0.7 Sudan

8.1 6.3 5.4 2.3

Ghana

4.2 4.4 3.1 0.0 Tanzania

3.3 1.8 3.5 0.4

Guinea

4.3 1.5 3.0 -0.1 Togo

2.7 6.4 3.1 0.0

Guinea-Bissau 0.7 8.4 2.7 -0.4 Uganda

6.9 3.5 4.8 1.7

Guyana

4.5 5.6 4.8 1.7 Vietnam

7.6 2.1 4.7 1.6

Honduras

3.0 3.2 3.1 0.0 Yemen, Rep. 5.6 2.9 4.3 1.2

Kenya

2.1 5.3 4.1 1.0 Zambia

1.0 4.9 2.4 -0.7

Source: International Finance Corporation (IFC), IFC-website: http://www.ifc.org/.

Table 2: Realistic GDP Growth Rates for the HIPCs

1

1/ Growth projections are currently not available for The Central African Republic, The Comoros, Liberia, Myanmar, Sao Tome and

Principe, and Somalia.

18

Furthermore, given that the amount of debt relief to be provided will need to be

sufficient to convince also skeptical private investors that a country is likely to remain

debt sustainable for the foreseeable future, the framework should determine to use the 90

percent lower bound of realistic growth projections, as they are provided in Table 2. This

would be better than using point estimates of growth projections and stressing that the

inherent volatility in growth rates provides challenges when making projections of

economic growth.

IV.3. Provision of Deeper and Broader Interim Debt Relief

Given that more than half of the 20 countries in the interim period are expected to

show debt ratios in excess of the HIPC sustainability thresholds at the time of their

completion points,

42

the provision of interim debt relief should be stepped up

considerably. A common misunderstanding is that this would increase the costs of the

HIPC Initiative. This is not the case as the total amount of HIPC debt relief is fixed in net

present value terms. Thus, the point of time at which HIPC debt relief is provided is not

important for the creditor. Furthermore, when determining the amount of interim debt

relief, creditors should also look at the amount of domestic debt servicing, most of which

is short-term debt, and thus implies a high fiscal burden, especially during the interim

period.

IV.4. Delivery of HIPC Debt Relief

Following our discussion above, showing that the delivery of HIPC debt relief

through a rescheduling at concessional interest rate does not contribute to the goal of

achieving long-term debt sustainability, it is obvious that HIPC debt relief should only be

delivered through debt or debt service cancellation. To the degree that this may have

negative short-term effects,

43

additional changes in the front-loading of debt service

reduction and/or improved interim debt relief may be required.

IV.5. Adjustments in the Burden-Sharing Concept

The HIPC Initiative’s burden sharing concept is—compared to earlier debt

initiatives—in principle a major step forward as all creditors are supposed to share the

costs of the initiative based on each creditor’s NPV share in outstanding debt. However,

we have seen that the theoretically optimal burden-sharing concept has run into serious

implementation problems. When looking for a better functioning burden-sharing concept,

we should keep in mind that most of the multilateral debt relief (excluding that of the

IMF) is actually financed by bilateral donors, partly through bilateral contributions to the

HIPC Trust Fund and partly through direct contributions to MDBs in form of

replenishments (like IDA replenishments).

44

Given this background, we make the

following suggestion to make the HIPC burden-sharing practice more efficient.

•

First, given that financing constraints are the most severe for small regional MDBs

from the South, whose debt relief will most likely be financed through the HIPC

42

See IMF and IDA (2002), p. 28: Benin, Chad, Ethiopia, The Gambia, Guinea-Bissau, Guyana, Malawi,

Niger, Rwanda, Senegal, and Zambia.

43

Negative short-term effects are especially possible in cases where reschedulings involve a grace period.

44

Though a couple of years old, the best discussion of these issues has been provided by Cline (1997).

19

Trust Fund, these Trust Fund resources should be released immediately and thus

allow the full participation of the currently non-participating MDBs. According to the

latest HIPC Status Report, the total costs to all non-participating MDBs together

would amount to US$46 million in NPV terms, which is a marginal amount compared

to the US$1.7 billion bilateral donors have thus far paid in.

45

•

Second, there are two options for dealing with HIPC creditors. One option would

have been to exclude HIPC creditors from the provision of HIPC debt relief. Besides

the disadvantage of deviating from the equal burden-sharing concept, the problem

with this option is that it implies a reduction in debt relief to the HIPC debtor. The

second option would have been to allocate the amount of debt relief a HIPC creditor

is supposed to provide into a Trust Fund, financed from the amount of debt relief the

HIPC creditor is supposed to receive from its creditors. Note that the financial impact

of this second option is exactly the same for all creditors and debtors as that of the

current burden-sharing concept, however, it would have avoided the costly

negotiations and litigations of HIPCs against HIPCs.

•

Third, consistent with existing Paris Club regulations, the HIPC Initiative should have

adopted a “de minimus” clause, which would have exempted minor creditors from the

provision of HIPC debt relief.

•

Finally, centralized consultations—possibly through the United Nations—may have

been undertaken (and can still be undertaken) with (a) bilateral creditors that are not

members of the IMF and World Bank (mainly Cuba, North Korea, and Taiwan), (b)

bilateral creditors, which had a history of not participating in traditional debt relief,

and (c) all commercial creditors. While such negotiations might not have led to the

full provision of HIPC debt relief, it is likely that these creditors would have provided

more debt relief than under the current situation, where HIPCs were advised to stop

payments to non-participating creditors and now face costly litigations. It should be

noted that the IMF and World Bank have now pledged to help HIPCs with the

litigations they face, however, not only comes this help a little bit late, it is still

provided on a case by case basis (instead of through coordinated and concerted

action).

In conclusion, given the costs HIPCs face from creditor litigations, these four

changes in the burden-sharing concept would have been more effective than the strict

insistence on an equal burden-sharing for all HIPC creditors.

IV.6. Using One Fixed Low Discount Rate for the NPV Calculation

The use of currency-specific discount rates would make sense if we would know

the future interest rate differentials of currencies for the remaining repayment periods and

if these future interest rate differentials would appropriately reflect differences in future

currency values (as the theory of interest rate parity suggests). However, given that both

assumptions are inaccurate, it does not make sense to use currency-specific interest rates

for the NPV calculation. There is some deep-seated belief that currencies of developing

45

Alternatively, given that some bilateral donors have made higher pledges to the HIPC Trust Fund than

they have paid-in yet, it would be sufficient if one bilateral donor would pay-in US$46 million and earmark

the amount for the immediate release to non-participating MDBs.

20

countries are less stable than OECD currencies, however, there is no objective way to

quantify such differences. Historical trends of devaluations may give some indications,

but just because a currency has been weak in the past, this does not necessarily imply

future devaluations. Given that there is no objective way to predict the long-term values

of currencies, it would make more sense to use one fixed discount rate for the NPV

calculation of all debt, independent on what currency the debt is nominated in. Note that

this would eliminate all the four problems associated with the current methodology of

currency-specific short-term discount rates described above.

There are even some indications that various investors don’t care much about the

NPV value of a country’s debt but are more concerned about the nominal value, which

would suggest dropping the whole NPV calculation. However, given that this would

imply some injustice in the provision of debt relief to countries with sharply diverging

maturity structures as well as some injustice for creditors providing debt of different

levels of concessionality, some reasonable discounting through a relative low level of one

discount rate for all currencies (e.g., 3 percent) seems appropriate.

V. Key Issues Related to a New Aid Architecture

Comparing the size of debt relief with the size of aid, debt relief plays a minor

role. Overall debt relief (including traditional debt relief) to HIPCs amounts currently to

about US$1.5 billion a year; gross disbursements from bilateral and multilateral donors

amount to about US$15 billion a year.

46

Hence, were there not the need to remove the

debt overhang, we could easily ignore the whole issue of debt relief and concentrate

instead on the financing of poverty reduction policies.

As Lancaster (2000, p. 80) pointed out, at the end-1990s, there were more than 40

poorly coordinated aid agencies working in numerous African countries, pursuing their

own priorities, and expecting their own separate administrative and procurement

requirements to be met, which have overwhelmed African officials and disrupted

government budgets. While important steps have been taken to better coordinate aid,

there remains a large problem of coordinating aid with debt relief. It is not new to suggest

the coordination of aid and debt relief. For example, Killick and Stevens (1997, p. 172)

suggested that

“mechanisms should aim to coordinate the activities of financial and aid

agencies (…) by merging debt relief and ODA coordination processes, with both

considered together, to form a comprehensive package of financial assistance

against a demonstration of need.”

In this regard, Birdsall and Williamson (2002) have also made significant

contributions. Linking debt relief to a new aid architecture, they point out that further

debt relief will need to be complemented by a larger reinvention of the international aid

architecture. We limit our arguments here to four related aspects: (a) the aid coordination

for HIPCs that have not yet reached their HIPC decision points, (b) the debate on weather

46

See Birdsall and Williamson (2002), p. 70.

21

loans should be replaced by grants, (c) the issue of additionality, and (d) the feasibility

and desirability of 100 percent debt reduction.

V.1. Aid Coordination for HIPCs that Have Not Yet Reached their Decision Points

Given that the amount of HIPC debt relief to be provided by each creditor is

calculated based on each creditor’s share in the total NPV debt at the year previous to the

HIPC decision point, no creditor has currently any incentive to provide new loans to

HIPCs that have not yet reached their decision point. In order to overcome this large

disincentive, it is either necessary to coordinate all new disbursements (which is very

difficult) or to simply change the reference date for the allocation of costs of HIPC debt

relief.

V.2. Grants versus Loans.

If the objective of the current development agenda is to secure long-term debt

sustainability, there can be no doubt that some loans should be replaced by grants.

However, one of the key questions related to switching towards grants is how these

grants can be financed. The grant financing should not lead to a long-term reduction in

the overall availability of financial assistance to the poorest countries, as it will be the

case if IDA is not properly replenished. Obviously, most of this problem would be solved

if all OECD countries would raise their aid flows to the long recommended level of 0.7

percent of GNP, and/or if the international community were to adopt global taxes to

finance development (like the Tobin tax). In any case, as the GAO (2002) report has

shown, even after the full implementation of the suggested switching from loans to

grants, HIPCs would continue to face unsustainable debt, and thus, more debt relief is

needed.

V.3. The Issue of Additionality

There is broad agreement (including by the Bank and the Fund) that HIPC debt

relief needs to be additional to conventional aid in order to effectively reduce poverty.

We have shown in Figure 2 above, that thus far, the HIPC Initiative did not achieve the

desired reduction in actual debt service payments for the 22 HIPCs that reached their

Enhanced HIPC Decision Points by end-2000. Indeed, Figure 3 shows that official net

resource flows to these 22 HIPCs had been declining for most of the 1990s, amounting

currently to 6.9 billion per year (1999-2001 average). Hence current official net resource

flows to the 22 early HIPCs are 22 percent less than what official net resource flows were

to these 22 countries in the early 1990s (amounting to an average 8.9 billion). The picture

is even worse if looking at Sub-Saharan Africa, where official net resource flows have

declined from an average US$16 billion during the early 1990s to less than US$10 billion

in 2001. These are alerting trends, especially if taking into account the large amount of

resources needed to combat the HIV/AIDS epidemic.

47

47

See Oxfam (2002) for linking debt relief to the HIV/AIDS epidemic. Culpeper (2001) presents a detailed

analysis of long-term financing for the poorest countries.

22

Source: World Bank, Global Development Finance 2003.

V.4. Is 100 Percent Debt Relief Feasible.

Given that debt relief is (a) only a small portion of overall aid and (b) not

additional to conventional aid (at least not until now, and unlikely to ever be in the

future), it is obvious that 100 percent debt relief is feasible by simply switching from the

provision of conventional aid to debt relief. The remaining question is thus if 100 percent

debt relief is desirable.

V.5. Is 100 Percent Debt Relief Desirable.

In answering this question, we will differentiate between multilateral and bilateral

debt relief, arguing that 100 percent debt relief is desirable on bilateral debt though not

necessarily on multilateral debt. In addition to differences in (a) financing constraints and

(b) debt structures (maturity and concessionality) between bilateral and multilateral debt,

we argue that the following points support 100 percent bilateral debt relief for a selected

group of extremely poor countries:

•

First, limiting 100 percent debt relief to bilateral debt minimizes the equity problem

of 100 percent debt relief. A 100 percent debt relief on multilateral debt would have

serious consequences on the availability of funds for less indebted though equally

poor countries (like for example Bangladesh), though even there would be options to

reduce the equity effect by extending multilateral debt relief to all countries with the

same level of poverty.

•

Second, HIPC debt relief is supposed to provide approximately 90 percent debt relief

in NPV terms on eligible bilateral debt. Hence, there remains (a) only about 10

percent of eligible bilateral debt in NPV terms, (b) post-cut-off-date debt, and (c)

Figure 3: Official net resource flows

1990-2001

6

9

12

15

18

1990

1992

1994

1996

1998

2000

Sub-Saharan Africa

22 HIPCs that reached Decision Point by end-2000

23

official development assistance (ODA) debt, which would need to be canceled;

amounting all together to only a small portion of overall bilateral aid.

•

Third, to the degree that debt relief constitutes actual reductions on debt service

payments, debt relief is required to be spent fully on poverty reduction measures.

Looking at the desirability of 100 percent debt relief on bilateral debt under these

circumstances, it is difficult to find compelling arguments against the provision of 100

percent bilateral debt relief to a selected group of extremely poor countries. Indeed, some

bilateral creditors (like the United Kingdom and the United States) have recently

promised to provide 100 percent debt relief to HIPCs as well as to other equally poor

countries. Furthermore, it should be pointed out that some other donor countries have

already provided 100 percent debt relief many years ago (like for example Canada,

Switzerland and some of the Nordic countries) when these countries cancelled their ODA

debts and moved to 100 percent grant financing. The provision of 100 percent bilateral

debt relief could continue to be voluntary for each creditor, though civil society groups

will need to push their governments for the adoption of such a policy.

VI. Necessary Structural Changes

Geda (2002) asked the question if the debt problem would be over for African

HIPCs if the HIPC Initiative would write-off all debt owed by African HIPCs. Geda

answered the question with a clear no, arguing that the pessimist answer is arrived at by

examining the historical origins of the African debt crisis and the structural problems the

continent is confronted with. Similarly, Birdsall and Williamson (2002, p. 5) pointed out

that “debt reduction alone is not enough to get development in the poorest countries back

on the rails.” Indeed, excluding the option of 100 percent debt relief followed by 100

percent grant financing, there is broad consensus that no debt relief package would

provide long-term debt sustainability as long as countries fail to pursue sound economic,

social and structural policies that stimulate economic growth and help attract increased

investment, especially from private sources. Given that the IMF and World Bank

emphasize the need of reforms with respect to governance, institutions, and social

policies, we prefer to concentrate here on the missing structural transformation in HIPCs,

whereby we look specifically at sectoral shares of GDP, long-term trends in savings and

investment, and some of the key issues related to high export concentration and high

export price volatility.

VI.1. Sectoral Shares of GDP

As Figures 4 and 5 show, HIPCs have not experienced any significant structural

transformation of their economies. While the last two years seem to indicate a decreasing

share of agriculture and an increasing share of industry, Figure 2 shows clearly that the

increase in the share of industry is not due to an increase in the share of manufacturing, as

it would have been desirable. Instead the downward trend in manufacturing seems to

have accelerated during the last two years (2000 and 2001). The continuous downward

trend in the share of manufacturing shows that HIPCs are not undergoing the structural

transformation that is needed to achieve long-term growth and debt sustainability.

24

Source: World Bank, World Development Indicators (central database, February 2003).

Source: World Bank, World Development Indicators (central database, February 2003).

Figure 4: Structural Transformation

of 42 HIPCs, 1990-2001

0

10

20

30

40

50

1990

1992

1994

1996

1998

2000

agriculture

industry

service

Figure 5: Structural Transformation

Share of Manufacturing of 42 HIPCs, 1990-2001

5

10

15

1990

1992

1994

1996

1998

2000

Percentage share of manufacturing in GDP

25

V.2. Long-Term Trends in Domestic Savings and Domestic Investment

Figure 6 shows the structural transformation of HIPC economies in terms of

domestic savings and investment from 1990 to 2001. While the dotted line of Figure 6

shows an overall positive trend in domestic savings, the share of domestic savings has

recently dropped from 10.5 percent in 2000 to 9.0 percent in 2001. The shares of total

investment also show a positive long-term trend, and without having experienced a recent

decline. While these long-term trends in domestic savings and total domestic investment

are promising, it is troublesome that the shares of fixed private investment show a

declining trend, with a relative sharp drop of more than 2 percentage points from 1998

(10.4 percent) to 1999 (8.3 percent). This clearly indicates that private domestic investors

have not been convinced about the long-term prospects of HIPCs.

Source: World Bank, World Development Indicators (central database, February 2003).

V.3. Long-Term Trends in Foreign Investment

We now look shortly at recent trends of foreign investment for the 42 HIPCs,

whereby Figure 7 shows the recent trends in foreign direct investment and Figure 8

shows the trends in portfolio direct investment. Each figure shows two curves, the

continuous lines show the nominal amounts of FDI and portfolio direct investment, the

dotted lines show the relative shares of the 42 HIPCs relative to world levels of FDI and

portfolio direct investment.

V.3.a. Long-term trends in flows on foreign direct investment (FDI) to HIPCs

Looking at the nominal amounts of FDI to the 42 HIPCs (the continuous line of

Figure 7), we see that FDI flows to the 42 HIPCs have increased in nominal terms

through most of the period (though they experienced a seemingly temporary shock in

2000). While this is a positive development, the share of the 42 HIPCs in world FDI

flows (the dotted line of Figure 7) shows a continuous decrease since 1995, reaching a

marginal share of less than half a percent in 2001. Hence, similar to the experience of

most other low-income countries, HIPCs are more and more marginalized in an

Figure 6: Percentage Shares of Investment

and Savings in GDP, 1990-2001

0

5

10

15

20

25

1990 1992 1994 1996 1998 2000

Total investment as percentage of GDP

Fixed private investment as percentage of GDP

Savings as percentage of GDP

26

increasingly globalized world. Furthermore, it needs to be pointed out that most of the

FDI flows to HIPCs are related to the extraction of natural resources (which are usually

exclaves to the real economy).

Source: World Bank, Global Development Finance 2003; and IMF, International Financial Statistics 2003.

V.3.a. Long-term trends in flows of portfolio direct investment to HIPCs

Looking at portfolio direct investment flows to the 42 HIPCs (depicted in Figure

8), we notice that both, the nominal amounts as well as the relative shares of portfolio

direct investment to the 42 HIPCs are basically non-existent; nominal amounts averaged

about US$65 million during 1990-2001; representing about 0.002 percent of average

world flows of portfolio direct investment during 1990-2001. The only times the group of

42 HIPCs experienced some significant inflows of portfolio direct investment were in

1994 and in 1997 (whereby it is not even sure if these two exceptional peaks are data

errors). Anyway, it is clear that HIPCs are largely excluded from the world flow of

portfolio direct investment (amounting to about US$5 trillion a year), which is largely

consistent with the experience of other low-income countries.

Source: World Bank, Global Development Finance 2003; and IMF, International Financial Statistics 2003.

Figure 7: Tre nds in Foreign Direct

Inve stment (FDI) of 42 HIPCs, 1990-2001

0

2

4

6

8

10

1990 1992 1994 1996 1998 2000

FDI (in US$ billion)

percentage s hare in world FDI

Figure 8: Tre nds in Portfolio Dire ct

Inv e stment of 42 HIPCs, 1990-2001

0.0

0.2

0.4

0.6

0.8

1.0

1990 1992 1994 1996 1998 2000

Portfolio direct investment (in US$ billion)

percentage share in world portfolio direct investment

27

Considering the HIPCs’ weak regulatory framework to deal effectively with

increasingly volatile international capital flows, it could be considered an advantage that

HIPCs do not receive large amounts of short-term capital. However, there are obviously

some costs to HIPCs from being more and more marginalized in terms of world capital

flows. In any case. as was the case for domestic investors, there is some indication that

foreign investors are not yet convinced that HIPCs are likely to achieve debt

sustainability.

V.4. Export Concentration, Export Price Volatility, and Contingent Facility

It is well known that HIPCs have a high degree of export concentration, which

makes them highly vulnerable to export price volatility (see Box 2). Based on 2001 data

provided in IMF and IDA (2002, p. 121), the commodity export dependency (defined as

the ratio of the three main commodities in total exports) averages to about 60 percent, the

average change in export price changes amounted 6.5 percent, and the average change in

the terms of trade was 7.3 percent.

While some recent contributions have made valuable contributions to the analysis

of developing countries vulnerability based on developing countries’ high export

concentration and high commodity price volatility,

48

it is somehow neglected that most

developing countries, especially HIPCs, continue to experience considerable

deteriorations in their terms of trade. This is graphically illustrated in Figure 9, whereby

the straight line represents the linear trend. Woodward (1996) has suggested that there is

a link between deteriorating terms of trade for low-income countries and their high export

concentration. Building on the recent investment-under-uncertainty literature, Gunter

(1998) has shown that a high export concentration is highly correlated to macroeconomic

uncertainty and low investment. At the same time, it is well known, that developing

countries’ exports in agricultural products face significant barriers of trade, which has led

to the call of a “development round” of trade negotiations.

49

Source: World Bank, World Development Indicators (central database, February 2003).

48

See Berthélemy and Söderling (2001), Collier (2002), and Combes and Guillaumont (2002).

49

See Birdsall and Hamoudi (2002), as well as World Bank (2002).

Figure 9: Terms of Trade, 1990-2001

90

95

100

105

110

1990 1992 1994 1996 1998 2000

A verage terms of trade of 42 HIP Cs

Linear (Average terms of trade of 42 HIPCs )

28

Taking all these negative aspects into account, it is actually surprising that the

HIPCs managed to increase their exports from US$35 billion in 1990 to over US$73

billion in 2001 (see the left scale of Figure 10 on the next page). Furthermore, looking at

the right-hand scale of Figure 10, we can see that the HIPCs’ share in world exports

declined from 0.69 percent in 1990 to 0.55 percent in 1995, but had risen sharply from

1995 to 1997 and again from 1998 to 2001, amounting now (2001) to 0.81 percent. While

the HIPCs’ share in world exports remains marginal, the recent trend is impressive. Yet,

it has been questioned how much this increase in exports has truly benefited HIPCs.

Looking at the missing structural transformation of HIPCs, there is some indication that

the increase in HIPCs’ exports is largely due to the extraction of natural resources, which

is unlikely to provide the urgently needed forward and backward linkages to achieve

long-term growth and sustainable development.

Source: World Bank, Global Development Finance 2003; and IMF, International Financial Statistics 2003.

While more research on structural relationships is needed, it is clear that the

HIPCs’ high indebtedness is at least partly due to HIPCs’ economic structures.

Furthermore, suggestions to stabilize world commodity prices and to create a contingency

fund to compensate HIPCs for events beyond their control are also welcome, especially

in the medium-term, however, they are unlikely to constitute a sustainable long-term

solution.

50

It is therefore important that HIPCs find market niches in the industrial and

50

Please see Birdsall and Williamson (2002) for more detail suggestions on a contingency fund. An even

more comprehensive report has recently been issued by the International Task Force (ITF) on Commodity

Risk Management (2002). For a more critical assessment on the viability of price stabilization schemes, see

Cashin, Liang and McDermott (2000).

Figure 10: Exports of 42 HIPCs, 1990-2001

0

20

40

60

80

1990 1992 1994 1996 1998 2000

0.50

0.60

0.70

0.80

0.90

1.00

Exports in US$ billion

percentage share in world exports

29

service sectors that will make them less dependent on (a) world commodity prices and (b)

the industrialized countries’ demand for agricultural goods.

Box 2: Commodity risk vulnerability of HIPCs

All Heavily Indebted Poor Countries (HIPCs)

a/

depended on primary commodities

for more than half of their merchandise export earnings in 1997. Of these, 15 HIPCs

b/

generated more than 90 percent of their merchandise export revenues from a few

commodities such as cocoa and coffee. On average, the commodity share in the total

merchandise exports stands around 84 percent for HIPCs, compared to 72 percent for

IDA countries and 55 percent for developing countries as a whole. In Uganda, about 5

million smallholders and poor households-a quarter of the population-earn their living

from producing coffee.

Many HIPCs are major exporters of commodities, important for industrial

production and domestic consumption in the developed world. About 60 percent of cocoa

in the world is produced in three HIPC countries: Cameroon, Côte d'Ivoire, and Ghana.

HIPCs produce and export about 20 percent of coffee in the world. Prices of these

commodities are particularly volatile. For instance, between 1983 and 1997, cocoa prices

fluctuated between 60 percent and 170 percent of the average price over this period, and

robusta coffee from 40 percent to 195 percent-as against other commodity price

fluctuations from 50 percent to 150 percent over the same period. In addition, many

HIPCs are net importers of food or fuels. For 17 HIPCs

c/

such imports represent more

than 20 percent of merchandise imports.

Vulnerability to commodity price fluctuations also has significant implications for

HIPCs' debt sustainability. Dependence on commodities coupled with high volatility of

prices results in significant fluctuations in export earnings in HIPCs and therefore in debt

indicators. For instance, Uganda's export revenues vary with coffee prices. Taking the

level of Uganda's export earnings in 1985 as 100, this benchmark fell to 47 in 1993

following a sharp fall in coffee prices, and rose to 170 in 1997 after prices rallied. Such

changes have direct impacts on debt indicators and on countries' ability to adhere to debt

sustainability targets.

___________________

a/. Reliable trade data are not available for three out of 40 HIPCs: Democratic Republic of Congo,

Equatorial Guinea, and Liberia.

b/. They are, in the order of dependence on commodities, Mauritania, Chad, São Tomé and Príncipe,

Yemen, Angola, Rwanda, Niger, Congo Republic, Sudan, Guinea-Bissau, Burundi, Somalia, Benin, Ghana,

and Cameroon.

c/. They are, in order of dependence on food imports: Senegal, Yemen, Mauritania, Mali, Burkina Faso,

Chad, Togo, Guinea-Bissau, Mozambique, Congo Republic, Laos, Sierra Leone; and in order of

dependence on energy: Guinea, Uganda, Cote d'Ivoire, Zambia, and Madagascar.

Source:

Copied from International Task Force (ITF) on Commodity Risk Management, Report

on Commodity Price Volatility and Developing Countries, Washington, DC: ITF (2002);

available on the internet:

http://www.itf-commrisk.org/documents/discussionpaper.pdf

.

30

VII. Conclusion

In contrast to the conclusions drawn in the official HIPC documents of the IMF

and World Bank, we have shown that the key shortcomings in the enhanced HIPC

Initiative are unlikely to provide HIPCs with the urgently needed assurance of long-term

debt sustainability. Though most of these problems have been recognized, there is

validated fear that the international community will not address them appropriately but

muddle through by shifting the debt sustainability agenda to the PRSP process and

possibly waiting for an international debt workout mechanism.

In addition to this ineffective approach, which might ultimately fail, it is then

likely that at some time in the future, the HIPCs will be blamed for not having achieved

debt sustainability. Indeed, such language has already been drafted in the recent HIPC

documents: “To maintain long-term debt sustainability beyond the HIPC Initiative,

HIPCs will have to achieve sustained economic growth and export diversification

through sound economic management, and improved governance and institutions.”

51

Obviously, sustained economic growth and a diversification of exports are crucial, but

sound economic management, and improved governance and institutions will not be

sufficient for this to happen. As Geda (2002) has illustrated, we need to think beyond the

HIPC Initiative to solve the structural problems. However, without convincing debt relief

and pro-active policies to overcome structural problems, HIPCs are unlikely to achieve

long-term debt sustainability. Hence, there exists a vicious circle of insufficient debt

relief and low growth, which the HIPC Initiative was supposed to break through.

The long-term agenda to achieve long-term debt sustainability needs to focus

more on a long-term development strategy that fosters the necessary structural

transformation of HIPCs. It is unlikely that HIPCs will undergo the necessary structural

transformation without targeted policy interventions. More specifically, this calls for a

combination of industrial investments, supporting competition policies, and strategic

trade policies that go beyond the existing (though blocked) comparative advantages in

agriculture. An appropriate mix of such policies could lead to the development of

selected industrial bases and service centers that can—at least in the long-term—compete

with those of industrialized countries.

While much of the above may give rise to pessimism, it should be pointed out that

the current enhanced HIPC Initiative is likely to provide debt sustainability for some

countries, especially if future lending is more concessional than in the past and supported

by increased grant financing. While additionality of debt relief would be desired to

reduce poverty even more effectively, the additionality of debt relief is not a necessary

condition to remove the remaining debt overhang, and thus, the additional costs of a

HIPC-III outlined above could be marginal if compared with overall aid budgets. There is

also much more that can be done in developing countries, especially in shifting the few

resources they have towards a national development strategy and away from highly

unproductive expenditures.

51

IMF and IDA (2002), p. 4.

31

References

Addison, Tony, Henrik Hansen, and Finn Tarp (eds.), Debt Relief, New York: Oxford

University Press for UNU/WIDER (forthcoming).

Berthélemy, Jean-Claude and Ludvig Söderling, The Role of Capital Accumulation,

Adjustment and Structural Change for Economic Take-Off: Empirical Evidence from

African Growth Episodes, World Development, Vol. 29, No. 2 (February 2001), pp. 323-

343.

Bigsten, Arne, Jörgen Levin, and Håkan Persson: Debt Relief and Growth: A study of

Zambia and Tanzania, United Nations University/World Institute for Development

Economics Research (UNU/WIDER) Discussion Paper No. 2001/104 (2001).

*

Birdsall, Nancy and Amar Hamoudi, It’s Not about Trade Policy! Commodity

Dependence, Trade, Growth, and Poverty, Washington, DC: Center for Global

Development, Working Paper No. 3 (2002).

Birdsall, Nancy and John Williamson with Brian Deese, Delivering on Debt Relief: From

IMF Gold to a New Aid Architecture, Washington, DC: Center for Global Development

and Institute for International Economics (2002).

Bourne, Compton and Shelton M.A. Nicholls, The Real Debt Service Capacity of

Commonwealth Caribbean Countries: A Preliminary Econometric Study of Barbados and

Trinidad and Tobago, Social and Economic Studies, Vol. 39, No. 4 (1990), pp. 45-82.

Buira, Ariel, Reforming the Governance of the Bretton Woods Institutions, in: The OPEC

Fund for International Development (OPEC Fund), Financing for Development,

Pamphlet Series 33, Vienna: OPEC Fund (2002), pp. 213-255.

Cashin, Paul, Hong Liang, and C. John McDermott, How Persistence are Shocks to

World Commodity Prices., IMF Staff Papers, Vol. 47, No. 2 (2000).

Chowdhury, Abdur R., External Debt and Growth in Developing Countries: A Sensitivity

and Causal Analysis, United Nations University/World Institute for Development

Economics Research (UNU/WIDER) Discussion Paper No. 2001/95 (2001).

*

Cline, William R., Debt Relief for Heavily Indebted Poor Countries: Lessons from the

Debt Crisis of the 1980s, in Zubair Iqbal and Ravi Kanbur (eds.), External Finance for

Low-Income Countries, Washington, DC: International Monetary Fund (1997), pp. 134-

144.

Collier, Paul, The Macroeconomic Repercussions of Agricultural Shocks and their

Implications for Insurance, United Nations University/World Institute for Development

Economics Research (UNU/WIDER) Discussion Paper No. 2002/46 (2002).

*

Combes, Jean-Luis and Patrick Guillaumont, Commodity Price Volatility, Vulnerability

and Development, Policy Development Review, Vol. 20, No. 1 (March 2002), pp. 25-39.

Culpeper, Roy, Capital Volatility and Long-Term Financing for the Poorest Countries,

London: International Development Committee (IDC) (2001); also available on the

internet:

www.adb2001.org/annualmeeting/Seminars/culpeper_paper.pdf

.

Dijkstra, Geske and Niels Hermes, The Uncertainty of Debt Service Payments and

Economic Growth of HIPCs: Is there a Case for Debt Relief., United Nations

32

University/World Institute for Development Economics Research (UNU/WIDER)

Discussion Paper No. 2001/122 (2001).*

Easterly, William, How Did Highly Indebted Poor Countries Become Highly Indebted”

Reviewing Two Decades of Debt Relief, Washington, DC: World Bank Policy Research

Working Paper 2346 (1999); available on the internet:

http://econ.worldbank.org/docs/952.pdf

.

European Network on Debt and Development (EURODAD), Debt Reduction for Poverty

Eradication in the Least Developed Countries: Analysis and Recommendations on LDC

Debt, United Nations University/World Institute for Development Economics Research

(UNU/WIDER) Discussion Paper No. 2001/110 (2001).

*

European Network on Development and Debt (EURODAD), Putting Sustainable

Development First, Brussels: EURODAD (2002); available on the EURODAD website:

www.eurodad.org/

.

Geda, Alemayehu, Debt Issues in Africa: Thinking beyond the HIPC Initiative to Solving

Structural Problems, United Nations University/World Institute for Development

Economics Research (UNU/WIDER) Discussion Paper No. 2002/35 (March 2002).

*

Gunter, Bernhard G., Economic Structure and Investment Under Uncertainty,

unpublished dissertation, Washington, DC: American University, Department of

Economics, May 1998.

Gunter, Bernhard G., Does the HIPC Initiative Achieve its Goal of Debt Sustainability,

United Nations University/World Institute for Development Economics Research

(UNU/WIDER) Discussion Paper No. 2001/100 (September 2001); forthcoming in

Addison, Hansen, and Tarp (eds.), Debt Relief, Oxford University Press.

Gunter, Bernhard G., What’s Wrong With the HIPC Initiative and What’s Next., Policy

Development Review, Vol. 20, No. 1 (March 2002), pp. 5-24.

Gunter, Bernhard G., Humberto Lopez, Krishnan Ramadas and Quentin Wodon, SimSIP

Debt: A Simulation Tool for Analyzing Debt Sustainability, Washington, DC: World

Bank (July 2002); available on the internet:

http://www.worldbank.org/simsip

.

Hansen, Henrik, The Impact of Aid & External Debt on Growth & Investment: Insights

from Cross-Country Regression Analysis, Paper presented at the United Nations

University (UNU)/World Institute for Development Economics Research (WIDER)

Development Conference on Debt Relief, Helsinki (August 2001).

Herman, Barry, Strengthening Sovereign Lending Through Mechanisms for Dialogue

and Debt-Crisis Workout: Issues and Proposals, Paper presented at the 16

th

Technical

Meeting of the G-24 in Port of Spain, Trinidad & Tobago, February 13-14, 2003;

available on the internet:

http://www.g24.org/

.

Hersel, Philip, The London Agreement of 1953 on German External Debt: Lessons for

the HIPC-Initiative, in EURODAD, Taking Stock of Debt Creditor Policy in the Face of

Debtor Poverty, Brussels: EURODAD (1998).

Hjertholm, Peter, Debt Relief and the Rule of Thumb: Analytical History of HIPC Debt

Sustainability Targets, United Nations University/World Institute for Development

Economics Research (UNU/WIDER) Discussion Paper No. 2001/68 (August 2001);

forthcoming in Addison, Hansen, and Tarp (eds.), Debt Relief, Oxford University Press.

33

International Monetary Fund (IMF), International Financial Statistics, Washington, DC:

IMF (various years).

International Monetary Fund (IMF) and International Development Association (IDA),

The Enhanced Initiative for Heavily Indebted Poor Countries – Review of

Implementation, Washington, DC: International Monetary Fund and World Bank

(September 2000).

**

International Monetary Fund (IMF) and International Development Association (IDA),

The Challenge of Maintaining Long-Term Debt Sustainability, Progress Report presented

to the Development Committee and the International Monetary and Financial Committee

(IMFC), Washington, DC: International Monetary Fund and World Bank (April 2001).

**

International Monetary Fund (IMF) and International Development Association (IDA):

Heavily Indebted Poor Countries (HIPC) Initiative: Status of Implementation, Progress

Report presented to the Development Committee and the International Monetary and

Financial Committee (IMFC), Washington, DC: IMF and World Bank (September 23,

2002).

**

International Task Force (ITF) on Commodity Risk Management, Report on Commodity

Price Volatility and Developing Countries, Washington, DC: ITF (2002); available on the

internet:

http://www.itf-commrisk.org/documents/discussionpaper.pdf

.

Johnson, Alison, The Fiscal Burden of Debt, Debt Relief International paper to the

Commonwealth Secretariat seminar on the HIPC Initiative, London: Debt Relief

International (July 2000).

Killick, Tony and Simon Stevens, Assessing the Efficiency of Mechanisms for Dealing

with the Debt Problems of Low-Income Countries, in Zubair Iqbal and Ravi Kanbur

(eds.), External Finance for Low-Income Countries, Washington, DC: International

Monetary Fund (1997), pp. 145-174.

Krugman, Paul, Financing vs. Forgiving a Debt Overhang, Journal of Development

Economics, Vol. 29, No. 3 (November 1988), pp. 253-68.

Martin, Matthew, Debt Relief and Poverty Reduction: Do We Need a HIPC III., Paper

presented to the North-South Institute, Global Finance Governance Initiative Workshop,

(May 1-2, 2002); available on the internet:

http://www.nsi-ins.ca/download/Martin.pdf

.

Northover, Henry, A Human Development Approach to Debt Relief for the World’s Poor

(Update on the 1998 Working Paper by Northover, Joyner, Woodward), London:

CAFOD (December 2001); available on the internet:

http://www.cafod.org.uk/policy/acaf1.shtml

.

Oxfam International, Debt Relief and the HIV/AIDS Crisis in Africa: Does the Heavily

Indebted Poor Countries Initiative Go Far Enough. Oxfam Briefing Paper, No. 25 (June

2002); available on the internet:

http://www.oxfam.org/eng/pdfs/pp0206_no25_debt_relief_and_the_HIV_crisis.pdf

.

Pattillo, Catherine, Helene Poirson, and Luca Ricci, External Debt and Growth, Paper

presented at the United Nations University (UNU)/World Institute for Development

Economics Research (WIDER) Development Conference on Debt Relief, Helsinki

(August 2001).

*

34

Roodman, David Malin, Still Waiting for the Jubilee: Pragmatic Solutions for the Third

World Debt Crisis, Washington, DC: Worldwatch Institute, Paper No. 155 (2001).

Sachs, Jeffrey, The Debt Overhang of Developing Countries, in Jorge de Macedo and

Ronald Findlay, (eds.), Debt, Growth and Stabilisation: Essays in Memory of Carlos

Dias Alejandro, Oxford: Blackwell (1989), pp. 80-102.

Sachs, Jeffrey, The Charade of Debt Sustainability, Financial Times (September 26,

2000), p. 17.

Sachs, Jeffrey, Kwesi Botchwey, Maciej Cuchra, and Sara Sievers, Implementing Debt

Relief for the HIPCs, Cambridge: Center for International Development, Harvard

University (August 1999); available on the internet:

www.cid.harvard.edu/

.

Serieux, John E., The Enhanced HIPC Initiative and Poor Countries: Prospects for a

Permanent Exit, Canadian Journal of Development Studies, Vol. 22, No. 2 (2001), pp.

527-548.

Serieux, John E. and Yiagadeesen Samy, The Debt Service Burden & Growth: Evidence

from Low-Income Countries, Paper presented at the United Nations University

(UNU)/World Institute for Development Economics Research (WIDER) Development

Conference on Debt Relief, Helsinki (August 2001).

*

United Nations Development Program (UNDP), Human Development Report 2001, and

Human Development Report 2002, New York: Oxford University Press (2001) and

(2002).

United States General Accounting Office (GAO), Developing Countries: Debt Relief

Initiative for Poor Countries Facing Challenges, Washington, DC: United States General

Accounting Office (2000).

***

United States General Accounting Office (GAO), Developing Countries: Switching Some

Multilateral Loans to Grants Lessens Poor Country Debt Burdens, Washington, DC:

United States General Accounting Office (April 2002).

***

Woodward, David, Effects of Globalization and Liberalization on Poverty: Concepts and

Issues, in: UNCTAD, Globalization and Liberalization: Effects of International

Economic Relations on Poverty, New York: UNCTAD (1996), pp. 25-128.

World Bank, Global Development Finance, and World Development Indicators,

Washington, DC: World Bank (various years).

World Bank, Globalization, Growth, and Poverty: Building an Inclusive World Economy,

New York: Oxford University Press, 2002.

World Bank, The Heavily Indebted Poor Countries (HIPC) Debt Initiative: An OED

Review, Washington, DC: World Bank, Operations Evaluation Department (OED),

(February 2003); also available on the OED website:

http://www.worldbank.org/oed/

.

*

Available on the WIDER website:

http://www.wider.unu.edu/

**

Available on the HIPC website:

www.worldbank.org/hipc/

***

Available on the GAO website:

www.gao.gov/