We have asked the question of whether it is possible to increase the flexibility of heavily indebted primary producing countries in meeting their concessional debt service obligations. The answer to this question is that this can in principle be done by augmenting existing concessional debt agreements by a set of floating-for-fixed commodity swaps or swaptions. These should take the form of “variable strike” instruments in which the “floating” price is an estimate of the underlying trend price, probably implemented through a moving average of past prices. Modifications of this sort certainly appear feasible.
However, feasibility does not imply effectiveness. Countries may already have methods for coping with commodity price variability, even at the macroeconomic level, and these may be undermined by hedging arrangements. Of even greater importance, different countries are subject to a large number of shocks (for example, demand shocks, exchange rate changes and political disturbances), and the effects of these shocks may dominate those of commodity price shocks.
The issue of whether a particular country would have benefited from a scheme of this sort, for example over the past decade, is therefore primarily empirical. We have investigated this for ten commodity dependent severely or moderately indebted African countries.
The scale of the benefits varies from country to country. Using a fairly conservative money metric, the proposed instrument turns out to be worth the same as a cut in debt service payments of between zero and 5%. The median value across our sample is between ½% and 2¼% depending on scheme design.
Use of a less conservative risk aversion parameter could raise these estimates by a factor of two or, conceivably, three. However, those schemes which are seen as generating the highest (1%-5%) benefits divorce payments to creditors almost entirely from existing contractual debt service commitments except in so far as these serve as an accounting device to ensure eventual repayment. The additional benefits to the HIPCs in these unconstrained schemes turn out to be at the expense of a disproportionately large disruption of debt service payments to the concessionary lenders, and in particular to IDA.
One is therefore confronted by a choice between schemes which impose relatively low costs on the lenders but generate correspondingly low benefits to the borrowers, and alternatives which generate higher, albeit still modest, benefits, but at the expense of very considerable and potentially costly disruption of flows to lenders.
We have looked in some detail at the reasons for the relatively weak performance of these price-based schemes. The main reason appears to be that movements in international prices are only one of several factors that generate movements in countries’ reported commodity earnings. In general, world prices account for less than half of total revenue movements. Quantity variations and variations in local prices relative to those in world markets also play an important role. In particular, quantity variation is typically at least of the same order of importance as price variation in accounting for variations in countries’ commodity export revenues. Schemes, such as those we consider, which are based on world prices, cannot offer protection against these other factors.
The summary answer to the question we have posed is therefore that it is possible to modify the terms of concessional debt service so that it more closely matches countries’ abilities to make these payments, but this will only be of value to a subset of indebted countries. Schemes based on these principles cannot be a universal panacea and, if they are introduced, this should be on a discretionary, country-by-country, basis. But even where such a scheme would be useful, its value would be relatively modest – we estimate as equivalent to a reduction of less than 10% in concessional debt service payments.
This conclusion has implications for other strategies which attempt to improve the lot of commodity-exporting developing countries by operating purely in terms of world prices. At least at the level of the nation state, the extent of quantity variability is simply too large for such policies to substantially reduce or offset the variability of export revenues (or import expenditures).
To be successful, a scheme must operate directly in terms of revenues. Compensatory finance schemes meet this requirement, but they are problematic in terms of timeliness and implementability. Nevertheless, it may be right that this approach should be reevaluated. There is some evidence that large adverse terms of trade shocks have a disproportionate negative impact on growth25, and if this conclusion can be sustained, it could justify a new “shock” window. One possibility might be to open such a window in the pending IDA 14 replenishment.
An alternative possibility, closer to the principles espoused in the approach we have analyzed here, would be to allow HIPCs to redenominate part or all of their concessional debt on an indexed basis in terms of their own local currencies – see Hausmann and Rigobon (2003).
Because any decline in export earnings would tend to lead to real depreciation of the local currencies (see Cashin et al., 2004), this could provide a degree of automatic protection. However, as with compensatory finance schemes, this gives rise to serious potential moral hazard and adverse selection problems.
There is a growing consensus that commodity-dependent developing countries have received relatively small dividends from the growth in the world economy over the past fifteen years, even though it is not clear whether their poor performance resulted from commodity dependence or whether this dependence was the consequence of a failure to grow through diversification.
Be that as it may, this dependence exacerbates the already severe problems they experience in servicing their debt and threatens to undermine the objectives of the HIPC Initiative. This realization has prompted the resurrection of some old proposals and examination of some new possibilities.
By itself, concern is insufficient and may prove a poor guide for policy. It is important that the policy community is clear as to what it is aiming to achieve. If the objective is to ease debt payments, it may be preferable to consider either further debt relief or further moves towards grant-based assistance rather than attempting to make the service of existing debt more palatable.
On the other hand, if the objective is to provide commodity-exporting countries, or their governments, with a measure of insurance against the impact of adverse revenue fluctuations, it is arguable that this may be better achieved through a compensatory finance mechanism or a shock facility, notwithstanding the well known problems (timeliness, moral hazard etc) that arise in connection with these schemes. These approaches are not mutually incompatible but it appears difficult to achieve both within a single scheme.
25 See Collier and Gunning (1996) and Dehn (2004). Dehn et al. (2004) take a more sceptical view.
By Prof. Christopher L. Gilbert, Prof. Alexandra Tabova
Next: Appendix: Social Welfare and Detailed Results
Summary: Index