Memorandum from Mr David Woodward, Freelance Development
Consultant
1. Historical Background
From the early 1980s until the middle of the 1990s, international
efforts to deal with the debt crisis were seriously inadequate.
Initially, this reflected an unrealistic insistence on the part
of the major creditor governments (and the IMF) on dealing with
the crisis as essentially a liquidity problem, even as it became
increasingly apparent that the problem was one of solvency. While
solvency-based elements (i.e., debt and debt-service reduction)
were introduced in debt restructuring packages from the late 1980s,
these proved to be seriously inadequate for low-income countries,
for three main reasons.
— An arbitrary limit was imposed on the extent
of debt reduction by government creditors, based on their own
willingness to reduce debts rather than the extent of reduction
needed to achieve debt sustainability and renewed growth.
— "Opt-out" clauses allowed some creditors
(most notably the US and Japan) to avoid debt reduction altogether,
by providing extended reschudulings at market interest rates.
— No mechanisms were available for the reduction
(or even rescheduling) of debts owed to multilateral agencies,
which represented an ever-increasing proportion of the debts of
many low-income countries, as other financing sources dried up.
In short, the debt strategy up to the mid-1980s was always
too little too late; and the economic damage done to the debtor
economies by their continued over-indebtedness (and to the international
prices of their exports by their export-promotion efforts) weakened
their ability to service their remaining debts, increasing the
debt reduction needed still further. Had the Naples Terms been
offered on bilateral debts (and the IDA debt-reduction facility
been available for commercial debts) in the early 1980s, the debt
crisis in the low-income countries could almost certainly have
been resolved. For many countries, the enhanced (or even the original)
Toronto Terms might well have sufficed.
When it was first announced, the HIPC Initiative appeared
to be a dramatic step forward. As well as offering the possibility
of a reduction of multilateral debts for the first time, it moved
away from the principle of reducing (bilateral) debts by an amount
acceptable to creditors to the principle of reducing total debt
to a level regarded as sustainable.
2. The HIPC Initiative: a Promise Unfulfilled
Sadly, it is becoming increasingly clear that the initial
promise of the HIPC Initiative will not be fulfilled, as a number
of serious shortcomings have become apparent as the Initiative
has unfolded. These are discussed in Attachments B, C and E.[16]
1. The main indicator of debt sustainability used is the
ratio of the net present value (NPV) of public and publicly guaranteed
debt to exports of goods and services. This is a reasonable measure
of the long-term balance of payments effect of the debt, but it
has serious limitations as a measure of sustainability.
— It does not take account of other dimensions
of debt sustainability (i.e., debt relative to fiscal revenue,
production and savings/investment).
— It excludes a number of external liabilities
which, for a given level of exports, have a similar effect on
the balance of payments (notably private non-guaranteed debt and
the stock of foreign direct investment and equity investment).
— In terms of the debt overhang effect on investment
and growth, it appears to be the nominal value of debt rather
than its NPV which has the greater effect.
— The NPV of debt is heavily dependent on the discount
rates used, which fluctuate markedly over time. As a result, the
extent of debt reduction (and thus of the "sustainable"
level of debt) for each country may depend critically on the timing
of the completion point.
— It has also been argued that the discount rates
currently used are too high, although this has yet to be investigated.
— The three-year period over which export figures
are averaged is too short to even out price fluctuations. Again,
this makes the extent of debt reduction unduly dependent on the
timing of the completion point, particularly for countries whose
average export prices fluctuate markedly over time.
2. The level of the NPV/export sustainability threshold is
also open to question.
— The range currently proposed (200-250 per cent)
is essentially arbitrary, and no analytical basis has been provided.
The World Bank stated in 1994, on the basis of historical data,
that NPV/export ratios in excess of 200 per cent had "generally
proven unsustainable over the medium term".[17]
This suggests that a figure substantially below 200 per
cent would be appropriate in the context of the HIPC Initiative,
as private non-guaranteed debt was implicitly included, and the
sustainable level of debt is likely to have been reduced further
since the pre-1994 period by the adverse effects of the debt overhang,
and by increased stocks of foreign direct (and in one or two cases
portfolio) investment.
— Within the 200-250 per cent range, the level
set for each country is based on the application of an essentially
arbitrary collection of "vulnerability indicators",
apparently in a largely subjective way. Again, no analytical justification
has been provided, either for the indicators used, or for the
manner of their application. These indicators do not include,
for example, direct or portfolio investment, nominal debt indicators,
ratios of debt to savings or investment or human development indicators.
— The concept of sustainability now used is a very
limited one: that a country should not need further reschedulings
in the future, although this may be achieved only by continued
access to large-scale balance of payments support. This makes
no reference either to the effects of the post-HIPC debt on the
debtor economy, or to the IMF's traditional concept of balance
of payments viability (which also precludes balance of payments
support).
3. While a fiscal criterion (setting a sustainability threshold
for NPV/government current revenue of 280 per cent) has, quite
appropriately, been introduced, it is subject to very restrictive—and
largely arbitrary—conditions.
— Only countries with exports of 40 per cent or
more of GNP are eligible. Since the ratio of exports to GNP has
no effect on the fiscal sustainability of a given level of debt,
this rule has no obvious justification. It seriously limits the
application of the fiscal criterion, as well as discriminating
strongly against the poorest countries: only eight of the 41 HIPCs
currently qualify; and only one of these is among the poorest
two-thirds of HIPCs.
— The revenues used in this calculation are subject
to a floor of 20 per cent of GDP. This is arbitrary, and in many
cases unrealistic: in Uganda, proclaimed by the World Bank as
the model of adjustment in Sub-Saharan Africa over the last decade,
revenues remained below half this level in 1995; in Rwanda, local
IMF staff saw anything significantly above 10 per cent as unattainable
for the foreseeable future in 1996. In all, around three-fifths
of HIPCs had revenues below this level in 1995; and once again
the poorest are disproportionately affected. Three-quarters of
HIPCs with revenues below the floor are among the poorer half
of this group. (Neither is this relationship limited to HIPCs:
among developing countries as a whole, only four out of 26 countries
with GNP per capita at purchasing power parity below $1,200 had
revenues in excess of 20 per cent of GDP in 1995.)
— The 280 per cent threshold itself is entirely
arbitrary, having been set at such a level as to provide significant
debt reduction for Côte d'Ivoire (for political reasons),
while the restrictions noted above were designed to minimise the
eligibility of other countries (for financial reasons). Once again,
no analytical justification has been provided for the 280 per
cent threshold; and there is no obvious reason to use a different
figure from the NPV/export threshold.
— As with the export criterion, important liabilities
are excluded from the fiscal criterion—in this case, domestic
debts of the government. While their direct inclusion would be
problematic, some allowance should be made for these liabilities
in setting the overall threshold, and possibly in varying it between
countries.
4. Further problems are raised by the issues of timing and
conditionality.
— No significant debt reduction beyond that available
under the Naples Terms Initiative is to be provided until a six-year
track record of structural adjustment has been achieved. This
represents a deliberate delay in the potential benefits of the
HIPC Initiative, as a means of reinforcing the IMF and World Bank's
policy conditionality, at the expense of imposing further unnecessary
costs on debtor countries through the debt overhang effect.
— While some past adjustment is to be counted as
part of this track record, this is far less than the periods for
which the countries concerned have actually been undertaking IMF
and World Bank adjustment programmes. As a result, while many
HIPCs had already undergone more than six years of adjustment
before the Initiative was even announced, only two (Uganda and
Bolivia) and expected to benefit before the year 2000.[18]
— The conditionality of debt reduction under the
Initiative on structural adjustment raises a serious risk of further
delays. Moreover, the appropriateness and effectiveness of structural
adjustment as a development strategy for low-income countries
remains unproven: at best, it leaves room for improvement.
— In both Bolivia and Burkina Faso, the decision
point was delayed pending elections, until the new governments
had been in office long enough to demonstrate that they were satisfactory
to the IMF and World Bank. Apart from raising questions of legitimacy
and sovereignty (since both were democratically elected) there
is a risk that this will delay the completion point for these
countries, and thus the reduction of debt for Bolivia. (Burkina
Faso is in any case unlikely to receive debt reducing under the
HIPC Initiative.)
— These delays leave the debt overhang in place,
and will thus further reduce the sustainable level of debt (as
well as being an obstacle to effective economic reform); but there
is no allowance for any reduction in the sustainability thresholds
over time. Significant delays will also mean that the NPV ratios
are calculated on the basis of data for later years. If exports
(or, for countries benefiting from the fiscal criterion, government
revenues) are increasing in nominal terms, this will increase
the level of debt deemed sustainable, and thus reduce the debt
reduction received by the country concerned.
5. Apart from questions about the level of the sustainability
thresholds, the allocation of the designated debt reduction between
different creditor groups is also potentially problematic.
— The rules envisaged for burden-sharing between
bilateral and multilateral creditors, combined with the current
Paris Club restrictions on debt reduction raise serious doubts
about whether the debt reduction envisaged by the sustainability
rules can be achieved in some cases (e.g., Mozambique, Nicaragua
and Guinea-Bissau). There is a risk that this will leave debt
at a level which is clearly unsustainable even by the HIPC Initiative's
standards. At the very least, in the absence of a definitive solution
to this problem, it is likely to cause delays for some countries,
while the creditors wrangle over their respective contributions.
— There is also a serious question about how debt
reduction by non-Paris Club bilateral creditors and the smaller
multilateral creditors can be enforced. At present, it is simply
assumed that the former will match Paris Club terms, and that
the latter will match the terms offered by the IMF and World Bank.
However, there is no mechanism available for ensuring their compliance;
and there would seem to be a real risk of serious delays (again
extending the debt overhang problem), and potentially of a disruption
of the whole process.
6. Experience to date suggests that there has been a substantial
amount of political manipulation of the HIPC process.
— As noted above, the fiscal criterion was introduced
essentially to provide debt reduction to Côte d'Ivoire,
for which it would not otherwise have qualified, under strong
political pressure from the French government. The rules were
designed, and the threshold set, explicitly to provide a politically
acceptable level of relief to Côte d'Ivoire, while minimising
the additional relief provided to other countries, and thus the
additional cost.
— The lack of transparency in the application of
the vulnerability criteria was exploited to set the NPV/export
thresholds for Uganda and Bolivia at exactly the levels below
which burden-sharing problems would arise, apparently to avoid
a dispute between the multilateral agencies and the Paris Club.
— The World Bank allegedly attempted to suppress
local opposition to its Country Assistance Strategy for Mozambique
by threatening to delay the HIPC process.
3. The Role of the IMF
The IMF appears to have been motivated primarily by a desire
to limit the cost of the HIPC Initiative to itself, and to delay
the point at which it incurs such costs. This is largely because
of the continuing problems in securing agreement to sales of its
gold reserves, which would be necessary to finance a substantial
contribution, due to strong German opposition. The delay in gold
sales has already reduced the maximum amount which could be raised
in this way (net of the compulsory contribution to the Fund's
financial reserves) from $33 billion to $24 billion since they
were first proposed in mid-1994, as a result of the dramatic fall
in the world price. (The possibility of IMF gold sales to finance
debt reduction is discussed in Attachment A.)[19]
This has also led the Fund to "talk down" the potential
cost of the Initiative, and to exaggerate the post-HIPC economic
prospects of potential beneficiaries, by relaxing its traditional
concept of financial viability, and by producing seriously over-optimistic
projections for individual countries. (See Attachment C.)[20]
These have, in particular:
— assumed overall export growth rates, sustained
over the long-term, which are far greater than have historically
been achieved by most low-income countries;
— assumed growth rates of commodity exports (notably
Ugandan coffee) which, if replicated in other exporting countries,
would appear seriously incompatible with the price assumptions
for the same commodities; and
— assumed very large inflows of foreign direct
investment while making little or no allowance for profit remittances
in subsequent years.
There is also no sign of any allowance having been made for
the debt overhang effect on productive investment prior to the
completion point.[21]
4. Overall Assessment
In short, the HIPC Initiative, as originally conceived (or
at least presented) had the potential to provide real benefits
to the highly-indebted poor countries, by finally allowing the
debt reduction available to catch up with what was needed to provide
a basis for sustainable economic growth and human development.
However, this potential has not been realised, primarily due to
the efforts of the creditors to limit their respective shares
of the costs, and to delay debt reduction as a means of enforcing
conditionality and (in the case of the IMF) to delay the point
at which costs are incurred.
As a result, the debt reduction provided will be inadequate;
the economic problems of the poorest countries will be extended
unnecessarily into the future, at considerable economic and human
costs; and it seems almost inevitable that there will ultimately
need to be a new debt initiative to resolve the problem. However,
the inadequacies of the HIPC Initiative as it is currently designed
are concealed by the slow pace of its implementation, coupled
with the IMF's over-optimistic projections.
5. What needs to be done?
Some major changes in the HIPC Initiative are needed, and
needed urgently, if the enormous economic and social costs of
the continuing debt crisis of the poorest developing countries
are finally to be brought to an end. Possible changes are discussed
in Attachment F. 1
As an immediate priority,
— the HIPC process needs to be accelerated considerably
(e.g., by increasing the allowance made for past adjustment in
assessing track records);
— the restrictions on the fiscal criterion (the
exports/GNP rule and the revenue floor) need to be removed;[22]
and
— the burden-sharing rules and restrictions on
Paris Club debt reduction need to be revised, to allow debts to
be reduced to the levels indicated by the sustainability thresholds.
A strong case can also be made for
— lowering the NPV/exports and NPV/revenue thresholds;
— incorporating an automatic lowering of the thresholds
over time, to take account of the effects of the debt overhang;
and
— taking account of currently excluded liabilities
(stocks of direct and portfolio investment and private non-guaranteed
debt in the case of the NPV/export criterion; and domestic government
debt in the case of the fiscal criterion).
Serious consideration should also be given to
— reviewing the interest rates used as discounts
rates in NPV calculations;
— extending the period over which exports are averaged;
— making the application of the vulnerability indicators
in the estimation of the NPV/export criterion more systematic
and transparent, and including human development indicators and
nominal debt indicators;
— introducing a similar system for varying the
NPV/revenue threshold; and
— introducing additional criteria relating to the
ratio of the NPV of debt to production, investment and savings.
If changes are made in the HIPC Initiative to provide additional
debt reduction, then resources will need to be found to pay for
this, from an accounting, but not necessarily from an economic,
point of view.[23] Sales
of IMF gold could make a substantial contribution to meeting the
economic costs, and could usefully be pursued in parallel with
the above changes. Even after the recent fall in gold prices,
selling half of the IMF's gold reserves could double the debt
reduction available under the HIPC Initiative.
It will be easier politically to achieve changes in the HIPC
Initiative sooner rather than later. It is currently expected
that the first country (Uganda) will complete the HIPC process
in April 1998, the second (Bolivia) following later in the year.
Once these agreements have been completed, major changes in the
criteria may require them to be reopened; and there is likely
to be some resistance to this.
This is not an absolute block on further progress: neither
of the 1998 deals would need to be revised unless either the NPV/export
threshold were lowered (or the period over which exports are averaged
were extended); or the export rule and the revenue floor were
removed and the fiscal threshold lowered. Nonetheless, the existence
of completed agreements will be an increasing complication, both
practically and politically, as time progresses; and the process
of securing further changes in the HIPC framework will therefore
become progressively slower and more difficult.
At the same time, in the absence of substantial improvements,
the inadequacy of the HIPC Initiative will increase, as the sustainable
level of debt declines further due to the debt overhang effect
and increases in non-debt liabilities. This suggests, firstly,
that the extent of the changes needed will increase; and secondly,
that growing awareness of the Initiative's inadequacies may make
more radical reforms less unthinkable, even as modifications within
the existing framework become more problematic. At some point,
it may therefore become preferable to shift from seeking to improve
the HIPC Initiative to seeking to replace it or to make more fundamental
changes in its basic principles. One possible alternative approach
to the fiscal criterion is outlined in Attachment G.[24]
6. A Footnote: the HIPC Initiative and Debt Conversion
On a more minor point, the Committee may wish to be aware
that the HIPC Initiative represents a serious complication to
the conversion of bilateral debts for development, environmental
and humanitarian projects. This is partly because of the principle
of reducing debt to a specified level (which implies that converted
debts would otherwise be cancelled, except in particular circumstances);
and partly because of uncertainty about the extent of debt reduction
until the completion point is reached. This is likely to cause
a hiatus in debt conversions for HIPCs. (See Attachment D.) 1
Attachments[25]
A. Woodward, D (1994) "IMF Gold Sales for Debt Relief".
Paper for the Debt Crisis Network (UK), September.
B. Woodward, D (1996) "Debt Sustainability and the Debt
Overhang in Heavily-Indebted Poor Countries: some Comments on
the IMF's View". In EURODAD, World Credit Tables, 1996: Creditors'
Claims on Debtors Exposed. Brussels: European Network on Debt
and Development.
C. Woodward, D (1997) "The HIPC Initiative: Latest Developments".
Note for the Jubilee 2000 Coalition, April.
D. Woodward, D (1997) "Mutually-Beneficial Debt Conversion
Using Bilateral Debt: What, Where and When?". Paper for the
Royal Society for the Protection of Birds, October.
E. Woodward, D (1997) "The HIPC Initiative: Presentation
to the EURODAD Annual Conference, November 1997".
F. Woodward, D (1997) "The HIPC Initiative: What Needs
to be Done?". Note for the Jubilee 2000 Coalition, December.
G. Woodward, D (1997) "Fiscal Criteria for Debt Reduction:
Human-Development-Based Approaches". Note for the Catholic
Fund for Overseas Development, December.
H. Woodward, D (1997) "The HIPC Initiative: Effects
of Changes to the Rules Governing the Fiscal Criterion".
Note for the Jubilee 2000 Coalition, December.
David Woodward
Freelance Development Consultant
13 January 1997
16 Not herewith printed. A copy has been deposited
in the Library (DR 2). Back
17 World
Bank (1994) World Debt Tables, 1994-95: External Finance for
Developing Countries. Washington D.C.: World Bank. Volume
I, p40. Back
18 Burkina
Faso is also expected to be eligible, but is unlikely to qualify
for any debt reduction. Back
19 Not
herewith printed. A copy has been deposited in the Library. Back
20 Not
herewith printed. A copy has been deposited in the Library. Back
21 Until
recently, the IMF questioned the existence of the debt overhang
effect in low-income countries (IMF (1995) Official Financing
for Developing Countries: Washington D.C.: IMF). However,
the analytical basis for this view was at best questionable (Attachment
B [Not herewith printed. A copy has been deposited in the Library.]);
and the Fund has reportedly backed off from this position. A paper
in a recent IMF publication has provided strong evidence in support
of the debt overhang hypothesis in low-income countries (Elbadawi,
I, Ndulu, B and Ndung'u, N (1997) "Debt Overhang and Economic
Growth in Sub-Saharan Africa", in Iqbal, Z and Kanbur, R
(eds.) External Finance for Low-Income Countries. Washington
D.C.: IMF). Back
22 The
potential effects of changes in the rules governing the fiscal
criterion are discussed in Attachment H. (Not herewith printed.
A copy has been deposited in the Library.) Back
23 To
the extent that the debts to be cancelled would not otherwise
have been paid (or would have been paid only by the provision
of grants, or of new loans which would not ultimately have been
paid), there is no economic cost, as the money originally lent
has already, in effect, been lost. If servicing the debts would
otherwise have required highly concessional new loans, the economic
cost of cancelling them would be limited to the NPV of these loans. Back
24 Not
herewith printed. A copy has been deposited in the Library. Back
25 Attachments
not herewith printed. A copy of each has been placed in the Library. Back
| 
