A theoretical examination suggests that additional lending should accompany debt reduction to promote a recovery of investment
Since 1982, when the external debt problem surfaced, investment rates have fallen dramatically in most troubled debtor countries, both by historical standards and in relation to other countries. For the group of 15 heavily indebted countries, the average ratio of investment to GDP fell sharply from 24 percent in 1971–81 to 18 percent in 1982–87. This sharp drop in investment in the wake of the debt crisis has been much discussed. The fall in investment has, in turn, been associated with a reduced capacity to service their debts and a drying up of foreign capital flows. An important effect, however, has received much less attention; namely the effect of a heavy debt burden in reducing the incentive to invest. In fact, the disincentive effect may be so strong that it could give rise to a situation in which a reduction in foreign debt would generate such a major improvement in the debtor’s economy that debt repayments would actually increase.
Foreign debt may affect investment through two channels: the “debt overhang” channel and the “credit rationing” channel. Debt overhang occurs where countries are unable to service their debt in full and so actual payments are determined by some negotiation process between the debtor country and its creditors. In these circumstances, the amount of debt payments usually becomes linked to the economic performance of the debtor country rather than by the contractual terms of the debt. In such a case, if the economic performance of the debtor improves, part of the gains will be absorbed by higher debt repayments. That is, at least part of the return to any investment is likely to accrue to creditors (as bigger debt service payments) rather than to the country itself, and thus would not increase the total consumption of the debtor country. The past accumulated debt then acts as a foreign “tax” on current and future production, weakening the incentive to invest and encouraging capital flight even if finance is available. The disincentive effect of debt overhang is also likely to discourage government efforts to undertake adjustment policies and might as well adversely affect private sector incentives to hold domestic assets.
For a detailed theoretical discussion, see “Debt Overhang, Credit Rationing, and Investment,” IMF Working Paper, forthcoming, available from the author.
The credit rationing effect arises in a situation where a nonperforming debtor is unable to obtain any new foreign loans. To achieve an equilibrium of savings and investment, its domestic interest rates may then have to be kept higher than the rates in the international financial markets, thus adversely affecting its ability to invest.
These two effects, although usually associated with each other, may not be present together. Even though a debtor burdened with a debt overhang would normally face credit constraints too, it is conceivable that the debtor may be able to get new financing under some conditions. For example, a country could obtain new loans that are (explicitly or implicitly) senior to the previously outstanding debt, that is, they would have priority in any repayment. It is also possible for a country to be credit constrained but not have a debt overhang. For example, despite a moderate current level of indebtedness, a country could still find lenders unwilling to lend to them. This may occur because of the so-called contagion effects emerging from the difficult situation of other debtor countries, which helps raise the risk of debt repudiation.
The importance of the distinction between these two effects becomes clear when considering the possible reaction of investment and production in the debtor country to alternative means (such as debt reduction or new additional lending) of improving the external debt situation. Although any debt initiative would probably affect both the debt overhang and the credit rationing channel, its effects will be more concentrated on one of the two channels. For example, debt reduction measures (i.e., buybacks of debt by countries at discounted values, debt-equity swaps, or forgiveness of part of interest or principal) will reduce debt overhang but may not improve a debtor country’s access to new loans, and thus may not ease foreign borrowing constraints. New loans, by contrast, while increasing the amount of available foreign financing, may not improve the bargaining position of the debtor country vis-à-vis the “old” creditors, and thus may not help reduce the debt overhang.
Chart 1Benchmark economy Chart 2Simulation of investment paths in a hypothetical debtor economy
The effectiveness of any debt initiative to improve economic conditions in the debtor country will depend on which of the two channels exerts a more direct and powerful effect on investment. If debt overhang has the stronger effect on productive investment, debt reduction measures would generate a larger improvement in investment, but if credit rationing has the stronger effect, new lending would achieve a bigger increase in debtor country investment.
Simulation of the two effects
To isolate and evaluate both the debt overhang and the credit rationing effects on investment, some simulations were conducted based on a simple neoclassical growth model. (While the qualitative results would be true for a wide variety of assumptions, the figures presented below are specific to the model used in the simulations.) A benchmark economy served as a reference point. In this benchmark economy, there is external debt accumulation and timely repayment, without any of the problems associated with debt overhang or restrictions on borrowing. Its dynamic path is plotted in Chart 1. Investment is relatively heavy initially, and then gradually declines, from over 27 percent of GDP to about 22 percent in the steady state. Consumption, which is constant in per capita terms, starts off at 78 percent of GDP and stabilizes at about 71 percent. With no initial foreign debt, and following optimal consumption and investment schedules, debt reaches about 120 percent of GDP. Although this level appears considerable, its accumulation is gradual and runs parallel to the investment process that increases the productive capacity in the debtor economy. Once this stage is reached, the debtor economy uses a non- interest current account surplus of 3.5 percent of its GDP to service its foreign debt.
The effect of debt through the two channels mentioned above is depicted in Chart 2. Investment in the benchmark economy corresponds to the case described above of an economy that is not constrained in any way by its foreign debt level. The other three simulated investment paths correspond to different combinations of debt overhang and credit rationing problems. It is assumed that the debt overhang implies a “tax” on production equivalent to 5 percent of GDP. Also, when the debtor economy is unable to borrow in international markets (when it is credit constrained), the domestic savings-investment equilibrium requires the domestic interest rate to rise by roughly 2 to 2 1/2 percent. The domestic interest rate in fact varies over the period of simulation.
The results for the parameter values used in the simulations indicate that in countries where both debt overhang and credit rationing effects exist, credit rationing appears to exert a stronger adverse effect on productive investment than debt overhang. This arises because of the high domestic interest rate and its negative effect on investment demand. Cutting off foreign borrowing completely has an adverse impact on investment demand similar to that of a debt overhang that imposes a debt service burden of about 20 percent or more of a country’s GDP. This is because investment is determined by the profitability of installing new capital. An increase in the real interest rate of 2 percentage points translates into a 20 percent decrease in the present value of expected profits from an additional unit of capital. Hence, if the debt overhang “tax” is less than 20 percent, the credit rationing effect is dominant.
Chart 3The debt overhang and repayment possibilities
Now consider the case of a country with debt overhang but no credit rationing. Here, a country can freely obtain new international borrowing (which is serviced in full) but would still carry the burden of past debts. This reduces the amount of output available for consumers in the debtor country by a constant fraction. Hence, the relevant interest rate for investment and saving decisions is the international rate. In such a case, the investment level is relatively close to that of the unconstrained benchmark economy.
An interesting result is that old creditors would also benefit from the availability of new loans that are senior to their own claims. The reason is that the new loans permit a reduction in the domestic interest rate, producing an improved economic performance, which generates higher payments on the older debts as well. The degree of the improvement in economic performance depends on the extent to which the debtor economy is being affected by the credit rationing situation. For example, simulations indicate that if the domestic interest rate were initially slightly above 10 percent (compared to a world interest rate of 5 percent), the value of repayments on older claims would increase by about 50 percent when the debtor country is allowed to contract new senior debt.
However, the actual costs imposed by the debt overhang effect may have been underestimated in this model for a number of reasons. First, there might be an important difference between the average and marginal effects of the debt overhang. For example, debt payments may be determined as a percentage of the excess of GDP or exports over a certain level, which may imply that the (marginal) value of the fraction of GDP going to external debt payments may be much higher. Second, if investment projects are irreversible (that is, they cannot be liquidated) the risks associated with debt overhang may generate a postponement of investment decisions. Finally, the debt overhang may impose some indirect costs (not captured by this model) on the debtor economy such as those derived from having to carry on the bargaining process continuously. These bargaining costs arise from both the human resources devoted to the negotiations and the inefficiencies in resource allocation derived from strategic considerations.
The model also investigated the possibility that a reduction in the debt overhang tax increased the total value of debt repayments (see Chart 3). The results indicate that the favorable effects of debt relief on the ability to service debts arise only when the ratio of debt service to GDP in a country is more than 55 percent. On the surface, this result does not appear to support the above thesis as there is hardly any country where the ratio of debt service to GDP is anywhere near this amount. However, one could model debt overhang as a “tax” on production of internationally traded goods only. Then, a debt service burden of 55 percent of the income of the tradable sector, while still very high, may prevail in some countries, and debt relief, in some cases, may improve the ability to service debt and make both debtors and creditors better off.
The simulation results suggest that productive investment in a debtor country would increase as a result of additional lending. It also suggests that measures to overcome debt overhang through debt reduction would contribute to increased investment by reducing the “debt overhang”; however, investment would still remain far from its optimal level as long as the debtor country lacks access to financing from international markets. It is, in principle, better to attack the problem from both sides, combining debt reduction with new lending in order to maximize the effect on productive investment.
Another interesting result is that a different (and senior) type of loan that reduces the credit rationing problem could generate an improvement in the situation of all the parties involved. This poses the question of why creditor banks do not encourage the issuance of senior loans to heavily indebted countries, since that would also enhance their own repayment prospects. Leaving aside legal difficulties and problems of coordination among multiple creditor banks as explanations, it could be argued that it is not possible to implement senior lending from private sources because it is not possible to guarantee that the procedure could not later be repeated by issuing an even newer batch of claims with greater seniority. Thus, some official intervention would be required to lend more credibility to this framework.