Table 1: GDP growth rate of developing countries according to the exporter type.

It is not difficult to put our A and B Countries into the above classification. If we switch to the usual geographical grouping we see that the share of export of Africa in world trade has substantially decreased while that of Asia has increased(see UNCTAD Trade and Development Report, 1999). Table 2 presents the GDP growth rates by geographic region for a long period.

Table 2: Real GDP growth rate(1966-2007)

Source: World Bank 1999, Global Economic Prospects for Developing Countries.

In Africa population growths at more than 2.5 per cent, sometimes 3 percent, thus GDP per capita is either stagnant or even decreasing in many countries of the region. The same is true for several countries in all other developing regions outside Asia.

Asia shows the ‘sweet’ face of trade at least with respect to the figures regarding aggregate economic growth. What about the rest of developing world? Is there always a positive relationship between free trade and economic growth?

Was the opinion of Adam Smith, the founding father of economic science and of economic liberalism? Largely so, but with several qualifications which unfortunately are often ignored in international and development economics. Volumes have been written on Smith’s liberalism, but he is no naive supporter of free trade. The benefits of free foreign trade are fundamentally of a dynamic nature and concern the possibility of achieving increasing returns to scale. The static advantages and the efficient allocation of resources according to the natural endowments are only a very partial aspect of the overall process. Moreover when he deals with the relationships between rich and poor nations Smith says that there is no automatic mechanism which guarantees the catching up, or convergence, of the poorer countries towards the level of income of the rich ones.

Wealthy nations have an interest in trading among themselves, because of their rich markets rather than with poor countries(see Smith 1763?, p. 578). Therefore poor countries experience great difficulties in the international market; the first step is the most difficult one, there is a sort of threshold in the process of development. Why this difficulty? The reason lies in the very essence of the process of economic growth. In fact quite often a poor country does not have the resources to adopt the same techniques of production of the richer ones. Smith lists an impressive collection of impediments facing the poor countries in their first steps of a development process. Let us quote and number his own remarks:

Thus development is a process, whose triggering is the most difficult part of the story and of all these impediments the lack of capital is the really binding one. It is clear that for Smith the fundamental issue is not the lack of knowledge, but above all to the lack of the necessary amount of capital, because new technologies often require more capital. Productivity increases and technological progress depend on the accumulation of capital, a process which is difficult to trigger in the countries which are late-comers in international market.

As a matter of fact Tables 1 and 2 lead to rather sad conclusion for Africa A-Countries. In 1990 the World Bank published a very interesting work: Sub-Saharan Africa-From Crisis to Sustainable Development- A Long-Term Perspective Study; the optimism of the title has not been confirmed by subsequent facts, not even in the medium term. As a matter of fact the study itself showed the extremely high dependence of African Countries on the price oscillations of their exports, which are made up by very few commodities. These price changes could generate variations of 5-6 percentage points in income (see World Bank 1989, p. 24). Most SSA countries are ‘trade fragile’, depending on few commodities for they exports and it is difficult to see how the situation may improve rapidly, even in presence of extensive foreign trade liberalisation in the North.

Africa shows the ‘sour’ face of trade and we all hope that the next 15-20 years might be ‘sweeter’ or at least less sour for Africa. Trade liberalisation in the ‘North’ is one of the conditions necessary to favour growth and development in the ‘South’, it is one of the ingredients of an enabling external environment, but other measures are equally important and urgent: foremost among them substantial debt cancellations.

The report by UNCTAD on the 49 least developed countries shows that in the nineties their GDP per head grew on average only 0.4 percent. These countries have also carried on a remarkable liberalisation of their economies, sometime much more significant than that of OECD countries(see UNCTAD 2001).

Time, regionalism and multilateralism

The southern countries should not liberalise their economies with the same timing as those in the North. It is important that they might have access to northern markets, which are large and important ones, but without a temporary protection of their fragile manufacturing sectors they will incur major trade deficits. We must accept the fact that in A-countries manufacturing and high value added service industries need time to build up. Sequencing of liberalisation was a major issue in the late eighties with respect to countries of the former Soviet Union. Some lessons should have been learned from that experience; A-countries are in even weaker conditions than most of the former SU countries were in 1988.

We must also remember that the taxation of external trade is often the major source of revenue for these countries, fiscal reforms are needed in order to shift the collection of revenue from foreign trade to income taxes. But again tax reform cannot take place by a fiat, it is a long run process, sometime a very painful one, as the history of many European countries shows.

Processes of regional economic integration must not be seen as opposed to multilateral free trade agreements, they are part of an overall strategy to give A-countries the necessary time to undertake a process of structural change and to help them to stand on their own feet. Indeed in some cases regional integration could help to reduce the time required for these countries to be able to enter international trade without the need to protect their manufacturing sector.

In particular A-counties should be careful not to liberalise their capital account to early, on this point there is by now an almost unanimous consensus among scholars, even though this does not always translate into satisfactory policy suggestions for these countries. The problem of sequencing is also extremely important from the point of view of building a sustainable finance for development.

3. Unsustainable Debts

Even with trade liberalisation in the ‘North’ many LDcountries, particular in SSA, will continue to undergo both trade and current account deficits; and debt service may provide a major additional impediment to those envisaged by Smith..

Let us examine the external accounts of some SILICs, most of which are also HIPC countries, all of them in SSA, for the eighties and nineties see Table 3; the current account is before official transfers.

We see that:

3. the trade deficit is usually much smaller than the current account deficit.

Point 3 shows that it is not just a trade problem; even with a trade account on balance or in the positive, this group of countries would have had negative current account, mainly because of the external debt service.

The trade account and in particular the current account and their deficits give an indication of the size of the external finance needed by these countries. From the point of view of the ‘foreign exchange constraint’ the current account matters and it is important to notice that the trade deficit is smaller than the current one, thus the trade balance needs less foreign exchange to match its deficit. This is a clear indication of additional fragility by these countries with respect to trade relations only.

Table 4 provides some further interesting indications. The actual debt service has always been a relevant share of exports, in the range of 20%. These countries have a negative, large and increasing net factor income. ODA and aid have substantially increased from the eighties to the nineties; ODA is smaller than Debt Service, DS, until 1985, but then larger from 1986 to 1998. The total debt stock is not large if compared to the overall debt of developing countries, which is in the range of 2500 billion USD.

Table 4: Debt and Aid in 28 countries in SSA 1980-1998(current USD)

Table 4 shows that for our 28 SSA countries there is a remarkable and striking similarity between the figures of DS and ODA.

The figures seem to indicate that aid has been used to service the debt, at lest partially; that there relevant financial net outflows, much larger than the trade account and in the nineties these flows also larger than the DS; that interests on foreign debt are certainly part of this negative net transfer before official aid. As a matter of fact the two columns of ‘ODA plus aid’ and ‘negative net income from abroad’ have strikingly similar magnitudes; a further signal that during those nineteen years aid has been largely employed to cover the foreign exchange gap of these countries.

Although the overall net transfer has remained positive it is clear that aid has been largely used to pay interests. In order for these countries to be able to take advantage from trade liberalisation in the North, ODA has to be fully employed to undertake a process of social and economic transformation instead of being used for repaying past debt. This is a necessary condition for an enabling external environment.

During the eighties and nineties these countries have serviced only part of their scheduled obligations on foreign debt. If the servicing of debt should continue, even without fully meeting the scheduled DS, these SSA countries would have to generate a large positive trade balance in order to overcome the negative current account due to debt service. More important

If the overall current account will continue to be negative then a foreign debt stock will inevitably pile up. This is not completely true: these countries could go on leaving on aid, and could continue to play the role of the poor cousin, not to say that of world beggars.

Apart from aid, no forms of external finance have substantially contributed to the needs of external finance of these countries, neither commercial loans , nor FDI, nor Portfolio Investments as can be seen from the composition of the flows to SSA countries in the nineties (see GDF 2000).

The markets are well aware of the distinction between the two groups of countries, A and B; most FDI go to very few countries, all of them in group B. In 1997, only 2% of FDI went to Sub-Saharan Africa (World Bank, WDR 1999-2000, p. 72). Table 5 shows that in our 28 SSA countries FDI account for less than 1% of GDP. In these countries FDI are less than 6% per cent of Gross Domestic Investments. In order to attract FDI there must be expectations of sustainable economic growth, but given their initial economic conditions, the problem of A-countries is precisely that of getting on a development path.

Table 5 also shows that these countries have very low savings ratios, thus according to the ‘two gaps’ approach they may experience a ‘saving constraint’. Then how can they realistically finance a development process. In a sense and in particular because of the large debt stock these countries are taken in a trap in which both constraints are binding.

We can try to roughly evaluate the investment ratio, I/Y, needed to sustain an increase of 3% in the GDP per capita. If population grows at 3% per year we must have a target growth rate of the economy of 6%. In order to raise the GDP these countries must accumulate capital, K, and hence invest, I. We take an ICOR (Incremental Capital Output Ratio = dk/dy), of around 4 which is consistent with the estimates of the World Bank for SSA. Then these countries should invest every year 24% of their GDP, which is substantially higher than the 16% which is the average I/Y for these countries in 1998. Given the low level of domestic savings these countries will have to resort to foreign savings to finance their growth process.

Let us go back to the notion of enabling external environment: we have two clear policy indications.

First, trade liberalisation in the ‘North’ might help to improve the trade account deficit of these SSA countries, but will not eliminate it in the short to medium-run.

Second, debt cancellation would relieve the negative current account and thus reduce the ‘foreign exchange constraint’ on the accumulation process; ODA might be used to support new investments both in physical capital and in human development.

Debt cancellation is an absolutely necessary measure in order establish minimal conditions for these countries to enter a development process; to make Adam Smith’s first step’, the most difficult one.

However, the two above policies will not bring immediate relieve to the trade and current accounts;

-output and exports diversification need time,

-in order to attract private funds credibility has to be gained, which usually requires both high growth records and political stability,

-increasing the saving ratio through tax reform and a more accountable credit and financial system also requires time.

Moreover after two decades of economic and social crises these countries must achieve high rates of both capital accumulation and human development at the same time, and this may imply an overall ICOR higher than 4, or if we wish further financial means to recover from the economic stagnation of the last twenty years.

Given the above considerations finance for development will remain an absolute necessity, at least for A-countries, even if all the foreign debt should be cancelled, which is not the case yet.

Part II Sustainable Development Finance

A-Countries need foreign financial help. Financial flows may accrue in many different ways, on a range which varies from pure grants to portfolio investments, however A-Countries have much more limited options and almost no direct access to international financial markets.

In a sense A-countries are lucky that portfolio investments and short-term flows do not reach them. Surely these countries do not need volatile funds, at the same time they cannot continue to rely mainly on ODA. Even the Executive Directors of IMF and World Bank now say that A-type of countries should not borrow on non-concessional terms(see IMF-World Bank 2001).

A world of contracts

Which are the arguments advanced by those who do not want to cancel this debt? We can simplify the story by indicating three major reasons:

Of course one could counteract by maintaining that the debt

However debt cancellation is not a pure act benevolence, it should be part of a workable development strategy for the twenty first century. But there is the principle of contracts. Development is about change; from countryside to cities, from agriculture to industry, from the values and norms of a peasant society to those of a society largely based on trade, interests and contracts.

Money borrowing and financial techniques are among the most complicated forms of contracts. The debt story is about the training of people according to the rules of a ‘market and contract society’. Some of this training may be appropriate - good governance, financial solidity, respect of contracts- but financial markets do not account for late comers, they do not give them time to practice and learn. The ‘time dimension’ implicit in most financial transactions is extremely short and far away from the day to day experience of the people in the developing world, sometime even the ruling elites are not equipped for the game. They will learn but with enormous costs.

The game is then unfair and inefficient, but the rules of the game have become ‘principles’, they are like natural laws. If you enter the game then you must not default, and ‘moral hazard’ naturally follows.

This is a major lesson from the debt story with all its reschedulings, Toronto, London, Naples Terms etc.. The various ‘plans’ and ‘terms’ have gone on for almost twenty years without solving the problem. It is time to find out new contracts, which could be transparent, straight and workable, from the very beginning. New principles which may help the debtors but which may also reduce the problems of long negotiations, of free riding and of moral hazard.

4. On Sustainability

The IMF and the World Bank define the external debt sustainability of a country as its ability to "meet the current and future external debt service obligations in full, without recourse to debt rescheduling or the accumulation of arrears and without compromising growth"(italics added). This is a useful point of view and must be shared, but one could add some considerations. The debt story has shown that finance for development is not a short-run issue; debt sustainability is a long-run problem. Not a country has ever achieved economic development in less than a generation. Development lending must be on a long-run time horizon, 25-30 years, with considerable ‘grace periods’.

Let us include the three following criteria in the notion of sustainability; a debt:

Points 2 and 3 are necessary in order to make the whole repayment story, point 1, credible. An increase in social exclusion due to repeated economic and social crises in the short-run may take the country out of the appropriate development path/process and would inevitably jeopardise the scheduled debt service and the overall repayment.

There are social safety nets, which in principles should cope with the social and human bitterness of economic and financial crises, but they do not seem to have worked well in the experience of the Transition Economies.

Moreover there is the problem of new and acceptable conditionalities. A safety net is a buffer mechanism for bad periods, but is not part of the lending/borrowing contract. If stringent social and human development conditions must be attached to the lending contract then the terms of the contract itself must be equitable and realistic from the very beginning. Safety nets should come into the picture only as an extreme measure.

Some traditional analysis of debt sustainability

a. Decreasing trends of debt ratios.

The literature on the sustainability of foreign debt closely derives from the debates on public debt sustainability of the eighties(see for instance Spaventa 1987). The discussion is conducted in strict economic and financial terms, hence is limited to point 1. above.

Let us recall the arithmetic of debt sustainability.

D=overall foreign debt.

X=GDP=Gross Domestic Product.

g_n=(dX/dt)/X is the nominal growth rate.

The change of D/X over time is given by the following expression:

[d(D/X)/dt]=(dD/dt)/X - g_nD/X (1)

Take the following definitions:

E (M) = exports (imports) of goods and non factor services,

NFI = Net Factor Incomes,

NCF = Net Capital Flows (net of changes in reserves),

i_n = nominal interest rate on foreign borrowing.

No such items as foreign direct investments, fully concessional loans, capital flights etc. exist, so that capital flows consist only of new loans and the repayment of previous ones. Net Factor Incomes do not include items other than interest payments abroad - workers' remittances and the repatriation of profits on direct investments are ignored. In the balance of payment the current account balance is equal to capital movements, net of changes in reserves (see UNCTAD 1990, p.37):

E-M+NFI = - NCF (2)

To be precise E-M is the balance of the "non-interest current account", which however largely overlaps with the trade account.

dD/dt is the net change in the debt stock over time: the new loans obtained minus the repayment of previous loans in a given year; hence dD/dt = NCF and NFI = -i_nD, that is interest payments on existing debt. Equation (2) becomes:

E - M - i_nD = - dD/dt

and substituting this expression in (1):

d(D/X)/dt = i_nD/X - gD/X - (E - M)/X.

By re-arrenging the terms and deflating both sides of the equation we obtain:

d(D/X)/dt=(i - g)D/X - (E - M)/X (3)

where i and g are the real interest rate and real growth rate of GDP respectively. For debt to be sustainable D/X must stabilise, that is to say must not increase through time and possibly must decrease. Hence an ever growing D/X is clearly unsustainable, while a moderately decreasing ratio is regarded as sustainable.

The 'debt to GDP' ratio is either constant or declines if d(D/X)/dt is zero or negative. If we take a zero non-interest current account E = M, this brings to the fore the relationship between the interest rate and the growth rate.

Therefore g>i is regarded as the main condition securing the sustainability of a debt.

b. Solvency.

When the public debt is mainly owned by residents it is a matter of distributive agreements between generations. In the case of foreign debt there is a contract between external actors, no children and parents, and actual payments are required every single years, therefore solvency becomes a more immediate concern, particularly when countries experience very low growth rates which may be lower that the real rate of interest. But even this situation does not necessarily lead to default on foreign debt. In a very interesting model Cohen argues that in order to maintain the solvency it is sufficient that the present discounted value of all future payments should be equal to the initial face value of the debt (see Cohen 1985, pp. 142, 162). As Cohen shows this happens if the debt grows at a rate g_d which is strictly less than the real rate of interest:

g_d =(D_t – D _t-1)/D _t-1 = iD<i.

In this case the discounted value of the debt at a certain time t, possibly quite far in the future, will be 0 because:

lim D_t/(1+i)^t = 0

t® ¥

The future payments, whose present value should match the face value of the debt, derive from the trade surplus and are a proportion of the export earnings of the debtor country, hence one can calculate the percentage of future export earnings necessary to guarantee the solvency of a country(see ibid. p. 146-9). In order to have g_d <i it is enough to repay part of the interest payments due each year, rolling over the principals and remaining interest payments. By extending the period of future repayments to infinity we can always identify a share of export earnings small enough to justify the solvency of the debtor country.

This methodology clarifies the theoretical background which characterised the rescheduling approach to the debt crisis adopted by Bretton Wood institutions and by the Paris Club of creditors from the outbreak of the crisis in 1982. The main feature of all reschedulings has been an increase in the maturity. Moreover it is clear that in the case of foreign debt what matters is the ability of the economy to earn foreign exchange and this is the reason why exports come to the fore. Therefore the typical debt ratios are in relation to export rather than to GDP, as for the domestic debt. However, the crucial hypothesis which allows the mechanism to work concerns the time horizon, which becomes a decisive element for the repayment of the entire debt.

In the end sustainability can only be achieved by a decrease of D/X over time, i.e. there must be a horizon of credible partial repayments and export base diversification is a crucial step in the process of economic emancipation.

But the problem is precisely that of providing the external conditions, including the financial ones, which are part of an enabling external environment which may favour the economic transformations necessary to put these countries on a sustainable long-run economic growth path.

c. Thresholds

The HIPC initiative defines sustainability through some debt indicators. In the original framework a debt was regarded as being sustainable if the ratios of Net Present Value, NPV, of debt to exports was in the range of 200-250 percent and the debt service ratio in the range of 20-25 percent of exports. For particularly open economies with a large export base there was also a fiscal indicator of NPV to government revenues of 280 percent. Below these thresholds foreign debt was regarded to be sustainable.

After the July 1999 G8 meeting in Cologne the HIPC initiative was ‘enhanced’ and the indicators were lowered: the NPV of debt to exports ratio is now 150 percent of export and debt service to exports ratio must be in the range of 15-20 percent, while the fiscal ratio is down to a NPV of 250 percent of government revenues. The purpose of HIPC is to bring the actual ratios within those limits after the application of all the debt relief measures included in HIPC itself and in the other debt relief plans already available, such as the Naples Terms etc.

These values are admittedly ‘rules of thumb’, dictated by empirical analysis about countries which have managed to avoid rescheduling.

Notice that in the public debt literature there are thresholds, which have become quite famous thanks to the process of convergence of some European countries towards the European Monetary Union. They are the ‘Maastricht parameters’ and are indicators of a process of convergence: an overall fiscal deficit of 3 percent of GDP and a debt stock to GDP ratio of 60 percent.

All these debt indicators are related, therefore there should be coherent values for the assigned targets.

It is important to notice that from the original HIPC till present day debates the fiscal criterion which relates sustainability to government revenues has gained more and more relevance with respect to the ratios related to exports . The ability to obtain foreign currency through export earnings is no longer the only criterion of sustainability. Usually a debtor country uses public resources to service her foreign debt and this could undermine other public expenditures, sometimes extremely important ones such as those in education and health, therefore sustainability must also be evaluated in relation to the fiscal revenue of a country.

However the projections about future exports are still the crucial element in the Debt Sustainability Analysis, DSA, carried on by the World Bank (see below).

The debt itself creates negative conditions for the growth process.

First, there is the well known problem of the 'debt overhang'. This indicates the general climate of uncertainty which surrounds a heavily indebted economy and discourages investments, moreover businessmen may not be ready to start new enterprises, when they know that the benefits will be reaped by foreign creditors (see Krugman 1988, pp. 254-ff.; Sachs 1989).

Second, the debt service itself is detrimental to growth because it affects national savings and investments in negative way(see Bacha 1990).

The debt service is a foreign claim on domestic savings, and if we assume that it has seniority with respect to investments, then investments must be lower than domestic savings, in other words the debt service crowds out domestic investments (see Cohen 1989, p. 8). In all likelihood a positive debt service reduces both domestic investments and the growth rate below the level they could reach according to the value of domestic savings.

Unfortunately we have already seen a clear example of the extremely negative impact of debt service on economic growth: the so called lost decade of the eighties, when growth rates were negative for many developing countries, particularly in Latin America and Africa. In the years following the outbreak of the debt crisis in August 1982 and for most of the eighties we had negative financial transfer:

"developing countries have been exporting more goods and non-factor services to the developed countries than they receive-a reversal of the pattern before the 1980s"(World Bank 1990-91, p. 9).

The adverse effect of debt and debt service on growth was already clearly recognised in the 1989 World Development Report. There the World Bank analyses the impact of a 20% debt reduction on investment and growth. If

"the reduction in net resource transfer in the form of interest payments associated with the reduction in debt stocks.....is used to import needed investment goods, investment rates would raise several percentage points"(World Bank 1989, p. 21).

This should lead to an additional 1% increase in GDP after three years on top of the spontaneous growth rate and the improvement could even be as large as 2% for Argentina, Brazil, Mexico and Nigeria(ibid.)

So much for the eighties; the following decade did not see such negative net transfers, mainly thanks to grants, at least for Africa(see IMF-World Bank 2001, p. 21), but in terms of growth rates the score was not much better. This is even more worrying considering the fact that GDP per capita had decreased in many developing countries between 1982 and 1990, therefore in purely arithmetical terms a recovery of the growth rate should have been easier.

5. The ‘human factor’ approach to sustainability

A-countries are quite likely to need some form of development finance in the coming years. Is it possible to avoid debt crisis in the future? Is there a debt which can be fully serviced without undermining economic growth and human development? Which are the appropriate indicators for the long-run sustainability of an external debt?

Long before Shakespeare's The Merchant of Venice debt repayment does not seem to provoke emotions particularly favourable to creditors; all the more so when - contrary to Shylock and the venetian merchant Antonio - the borrowers are so much poorer than the lenders. But how to take into account of human development considerations? In the following there is a concise summary of the ‘human factor’ approach to debt sustainability(for an extensive analysis see Vaggi 1993 chapter 7).

Take/ the following definitions:

Y_k = GNP per capita.

POP = population.

k = rate of growth of Y_k,

p = rate of growth of population.

We have the following identities:

Y = Y_kPOP,

y = k + p.

We can introduce population growth and GNP per capita increases, as a proxy of the trend in people's standard living, into a model of external debt sustainability, which could be described as a "human face" approach. k and p are assumed to be constant for the entire maturity of the debt; p is a positive datum, k may be negative, but we can regard a positive k as a policy target. Therefore we define

h = k + p

as the growth rate of Y which satisfies the condition of leading to annual increases of GNP per capita equal to k, we regard h as the 'human factor'. Suppose that at time 0 before the external borrowing originates GNP and GDP are equal magnitudes, X₀=Y₀ and there is no debt service, DS₀=0.

Y_hj = Y₀(1+h)j

is the value of GNP at time j which is necessary to fulfil the target of securing an average growth rate of k in GNP per capita for each year during the period 0-j. We can regard Y_hj as that part of GNP which must be reserved for domestic absorption in order to guarantee a decent increase in the standard of living for the people.

The debt stock

Assume that the 'autonomous' rate of growth of GNP, y, is equal to that of GDP, g; we can calculate which part of the GNP of year j, Yj, can be used to service the debt, without worsening the living conditions of the people inside the debtor country; this is Y_fj=Y_j-Y_hj.

Y_fj= Y₀[(1+g)^j - (1+h)^j] (4)

is the part of GNP which can be "freely" transferred abroad to service the debt, compatible with 'human face' conditions, that is an average increase of k in GNP per capita during the first j years. The sum of all Y_fj for the entire maturity gives the value of the debt stock at time 0 which has a repayment schedule satisfying the 'human factor' h. From (4) we have:

D_f0 = S _j Y_fj/(1+i)_j, with j=1,2,.....,n.

Notice that the sequence on the right-hand side of the above equation implies that servicing the debt allows an increase in GNP per capita in each year during the entire maturity. D_f0 is not the actual value of the debt at time 0 but the present value of the stream of payments which may be sustained according to the 'human factor'. With appropriate manipulations we can calculate the debt/GNP ratio which allows the payment of the debt service, without violating human development conditions:

D_f0/Y₀ = {[1-(1+g)ⁿ/(1+i)ⁿ](1+g)/(i-g)}-{[1-(1+h)ⁿ/(1+i)ⁿ](1+h)/(i-h)} (5)

Equation (5) brings to the fore the relationship between the ‘spontaneous’ rate of growth g and the "human factor" h, which includes the target rate of increase in GNP per capita k and the rate of population increase p. The interest and growth rates, i and g, appear in (5) together with the human factor h. But the crucial relationship is now that between the latter and the growth rate, and it must be g> h. The condition that the real growth rate must be higher than the real rate of interest, g> i, guarantees that the debt to GDP ratio decreases over time, but it is not sufficient to secure sustainability according to the ‘human factor’.

Notice that h is itself a threshold, but a qualitative one, because the human factor approach requires a comparison of h with the expected growth rate g and not with some given ratios which are the same for all countries. h can be different for different countries, depending on population growth rates p, and also on the GNP pr capita increase which is regarded as necessary in order to match the human development targets. In principle a poorer country with a lower Human Development Index may have a larger k than a country which has already achieved an acceptable degree of human development.

For each country one can calculate which proportion of GNP (government revenue, exports) has to be spent on health and education if certain targets of human development have to be achieved in a number of years. This becomes part of the ‘human factor’ and provides an idea of the size of the resources which have to be put aside every year in order to try to achieve some target level in human conditions. The remaining part of the GNP (government revenue, exports) is a residual which can be used to pay interests and repay debts, this part gives an indication of the debt servicing capacity of a country according to the ‘human factor’.

To give an example, a 3% average annual increase in GNP per capita is often regarded as a necessary, not sufficient, condition to reduce poverty(see World Bank 1999-2000, p. 26; UNDP 1997, pp. 73, 109-110). The United Nation 20-20 initiative which sets the cost of poverty reduction at 1% of developing countries income(see UNDP 1997, pp. 112-4) provides a similar target increase in GDP per capita, because in low income countries population growth rates are around 2%.

Then the human factor could be h=p+k=2%+3%=5%.

A debt can be sustained in the long run only if g> h; when this condition is not satisfied the country will not have enough income to achieve human development and at the same time to repay its foreign debt, however small. This country will have to resort to grants only.

Equation (5) satisfies both criteria 2 and 3 of sustainability(see above), it also satisfies criterion 1, the actual ability to repay, provided that the country borrows from abroad a proportion of its GNP not higher than D_f0/Y₀. It is easy to see that with medium and long term maturities, the condition g> h allows a country to borrow very large amount of money. But equation (5) requires that GNP per capita increase at rate k, say 3%, in each year during the entire maturity, and thus rules out economic crisis even in the short run, even in a single year.

It is possible to derive a less stringent human factor formula which only requires that the growth rate g satisfies the human factor conditions set by h on average over the entire repayment period. This condition is satisfied by equation (6), where n is the maturity(see Vaggi 1993, pp.136-37):

D_f’0 = [ (1 + g)ⁿ - (1 + h)^n] n/(1 + i) (6)

Notice that when g > h the value D_f’0 systematically higher than D_f0 from equation (5). In this case there could be short-term crises thus violating point 3. above of the three criteria of sustainability.

The debt service

If g> h and with some technical manipulations it is possible to calculate the debt service DS which can be paid each year during the entire maturity of the debt without violating the ‘human factor’. We derive a formula for the less stringent human factor condition, hence we find the average value of the debt service which is compatible with an average GNP per capita growth of k, say 3%, over a number of years q £ n:

å _j DS_fj /q = { (1+g) + … + (1+g)^q - [ (1+h) +… +(1+h) ^{q ] }} / q with j = 1, ….q

In the second term there are two geometric sequences whose ratios are (1+g) and (1+h) therefore:

å _j DS_fj /q = -{ [ 1-(1+g) ^{q ]} (1+g)/g - [ 1-(1+h) ^q] (1+h)/h} /q (7)

The sustainable debt according to the human factor derives from a discounting procedure like the Net Present Value. Thus the human factor approach too requires a comparison of long run growth rates with the ‘human factor’ h, given a target increase k in GNP per capita. The long run should be the average maturity of the external debt. In order to achieve the necessary structural change in the composition of both output and exports this long run may be in the order of 20-30 years and that should be the length of the repayment period.

As for the NPV also in the human factor approach the evaluation about debt sustainability is based on projections of growth rates, and one must decide which types of growth rates have to be compared with the human factor h.

We see that for 23 of the 35 countries for which projections are available the average expected growth rate is higher than any rate which has been experienced in any past period since 1960. For 10 countries the growth rates used in the projections are 6 per cent or more, and for six countries we have expected rates higher than 7 per cent and among these countries we have Angola, Ethiopia, Mozambique, Congo.

The last but one column of Table 6 shows the average growth rates during the period 1987-97; the actual growth performance in a very close past period. Only in two cases, Uganda and Vietnam, these rates exceed those used in the DSA projections; all other HIPCountries had much less exciting economic performances. We hope that many other countries could repeat the good growth evidence that Vietnam and Uganda had in the eighties, but recent and less recent historical experience suggests some caution. Moreover Table 6 indicates that only for 8 countries out of 41 the economy has grown at its fastest pace in the close past, that is t say in 1987-97. In we consider the decades we see that the sixties and seventies saw much economic performances than the eighties and nineties. During the last decade seven countries had negative growth rates, and many other had very poor economic performance. This is exactly the kind of ‘poverty trap’ from which these countries should emerge and development finance should be geared to help this process.

In development economics it is necessary to be optimistic, but realism is also useful in order to avoid disappointments in the future, but also in order to draft the debt contracts according to really plausible assumptions. This is not just an obvious methodological principle, but an essential requirement for the credibility of the contracts themselves and also of the attached conditionalities.

The ‘human factor’: an example

We use the ‘human factor’ approach to evaluate the log run external debt sustainability of the 41 HIPCountries. Let us take as given the projections of the World Bank about future economic growth which appear in the third column of Table 6, thus we subscribe to the optimistic scenario.

We want to see if in this rosy case the external debt is sustainable according to the ‘human factor’. The fourth column of Table 6 gives the actual population growth rate p during the period 1987-97; this value can be added to a ‘target’ increase in GNP per capita of 1 or 3 percent, k=1 and k=3 respectively. Hence we obtain the corresponding human factor h=p+k for each of the two scenarios. We do not present these values, which however can be easily derived from p. For each country we compare the value of h with the growth rate g taken from the World Bank projections of Table 6. When h> g the debt is clearly unsustainable, this is the qualitative threshold characteristic of the ‘human factor’.

For the 35 countries for which there are projections by the World Bank we se that with a target GNP per capita increase of 3% per year only 14 countries fulfil the condition h£ g, for all the other 21 countries the debt is unsustainable. With k=1% the debt becomes sustainable for all countries but Congo.

If we compare the human factor h with the actual growth rates of the period 1987-97 we see that only 5 countries satisfy the ‘human factor’ criteria with k=3%, while the figure increases to 12 countries if we accept the more modest GNP per head increase of 1%. However from 1987 to 1997 all the HIPCountries had to service the external debt and sometimes they were asked to dedicate a relevant percentage of their GDP to this purpose, as the last column of Table 6 shows.

If we consider an average annual population growth of 2-3% in the years to come we can say that most of these 41 countries do not have any hope to be able to service any future debt according to the ‘human factor’. Vietnam, Uganda, Equatorial Guinea and Lao PDR may be good exceptions if they manage to maintain their excellent growth rates.

7. Conclusions

Scenarios of future growth rates should be closely related to past experience; past growth rates over decades seem to provide a good indication of the growth capacity of a country. This data and the growth experience achieved in a more recent period, like the decade 1987-97, provide some evidence of the initial conditions, from which each country has to commence its development process and any growth scenario should take them into account.

Perhaps economics and history may accelerate, and optimism must be part of any development analysis, but the ‘jump’ must bear some proportion with the past experience and the present conditions.

There are three major types of risk involved in an overoptimistic Debt Sustainability Analysis:

After twenty years of different plans and many false promises debt cancellation appears as a necessity for most low income economies; it must be part of that external enabling environment which should help them to step up into a sustainable development process. But the debt story tells us a lot in terms of the criteria and requirements of development finance. Structural change and export diversification take time and developing countries, particularly those that we have classified as A countries, will need to borrow from abroad. Good domestic governance and accountability are essential aspect of a sustainable external debt, but we also need the appropriate financial instruments. There must be realistic terms in the lending contracts and in the attached conditions, with sensible forecasts about the ability to repay by the borrowing countries. Twenty years have already been wasted from the point of view of economic and human development, let us hope that we have learned some lessons, at last.

Footnotes