Measuring the sustainability of Latin American external debt

Transcription

1 Applied Economics Letters, 2003, 10, Measuring the sustainability of Latin American external debt MARYANN O. KEATING and BARRY P. KEATINGy* Associate Faculty, School of Business and Economics, Indiana University South Bend and ymendoza College of Business, University of Notre Dame, Notre Dame, IN This article, based on a theoretical model of external debt sustainability, compares Latin American GDP growth with the debt service to total debt ratio. A country-bycountry forecast of external debt composition is presented and discussed. External Debt Forecasting is accomplished using yearly data from Global Development Finance Data for the years 1970 to 1999 are used for a Latin American composite and cross-country comparisons between Argentina, Brazil, Chile, Colombia, El Salvador, Guatemala, Mexico, Panama, Peru and Venezuela. I. INTRODUCTION This study forecasts private and foreign direct investment components of external debt in Latin America and applies a theoretical model to determine external debt sustainability. Debt is only sustainable if it can be realistically repaid according to the terms of the contract. For debt repayment to take place, a real transfer must be made. The macroeconomic model, in which the sum of savings, taxes and imports (S þ T þ IM) equals investment, government expenditures, and exports (I þ G þ EX), outlines the cost. Debt repayment must include one or more, of the following: increased exports, reduced government spending, reduced domestic investment, reduced imports, increased taxes and increased saving (decreased consumption). Repayment causes distress unless it is financed out of increased growth in output (GDP). Given default, foreign private investors or foreign taxpayers, guaranteeing the loans, lose. Forecasting here is done using yearly data from Global Development Finance Data for the years 1970 to 1999 provides a Latin American composite of ten countries. Cross-country comparisons are made between Argentina, Brazil, Chile, Colombia, El Salvador, Guatemala, Mexico, Panama, Peru and Venezuela. II. A THEORY OF FOREIGN DEBT IN EMERGING COUNTRIES The Balance of Payments is an accounting statement of financial transactions between one country and the rest of the world. By definition, this statement must balance and consists of a current account and a capital account. In analysing emerging country debt, a current account in deficit indicates imports exceeding exports and/or repayment of interest on external debt. In such a case, net capital flows to the country are positive and total external debt is increasing. The theoretical and mathematical limits to sustainable debt become quantifiable based on the following model: (i) The Nominal Growth Rate is defined: g n ¼ðdGDP=dtÞ=GDP where GDP ¼ Gross Domestic Product. (ii) The change in the ratio of External Total Debt (EDT) to GDP over time is: dðedt=gdpþ=dt ¼½ðdEDT=dtÞ=GDPŠ ½g n ðedt=gdpþš *Corresponding author. barry.p.keating.1@nd.edu Applied Economics Letters ISSN print/issn online ß 2003 Taylor & Francis Ltd DOI: /

2 360 M. O. Keating and B. P. Keating (iii) The Balance of Payments identity is: E M þ NFI ¼ NCF where E ¼ exports of goods and non-factor services; M ¼ imports of goods and non-factor services. NFI (Net Factor Income) is assumed to consist only of interest payments. NCF (Net Capital Flows) represents new loans and repayment of previous ones. (iv) The nominal interest rate on foreign borrowing (i n ) times the External Debt Total can be substituted for outflows of Net Factor Income, ignoring all other factor income. The change in External Debt Total with respect to time (dedt/dt) ¼ NCF, where capital flows consist only of new loans and the repayment of previous ones. Therefore, the Balance of Payments identity can be rewritten as: ðe MÞþð i n EDTÞ ¼ ðdedt=dtþ (v) This new Balance of Payments identity can be substituted into the equation for the change in the ratio of External Total Debt to GDP (ii). dðedt=gdpþ=dt ¼½i n EDT=GDPŠ ½ðg n EDTÞ=GDPŠ ½ðE MÞ=GDPŠ ¼½ði gþðedt=gdpþš ½ðE MÞ=GDPŠ III. LATIN AMERICAN EXTERNAL DEBT The model presented concludes that the growth rate of GDP must exceed the interest rate on external debt for sustainability. Therefore, one must first check average nominal growth rates in Latin America with debt service ratios. Then compare the composition of debt on a country-by-country basis. Hayek in explaining why markets are efficient yet remains vulnerable to monetary disturbances caused by unpredictable foreign capital flows points out that markets operate with an economy of information and are characterized by decentralization and dispersion of knowledge. Standard business forecasting methods are employed, to forecast industry or company sales, to economize on information and permit cross-country comparisons. In 8 out of 11 cases, it was found that forecasts for 1997 were closer to the actual 1997 debt service to total external debt ratios than an average of three previous years. But, as Hayek predicted, no stable ratios of debt composition were found over the course of a few years (1979). This section substitutes dynamic forecasting for traditional static ratio analysis of private to total external debt. Unless it is government guaranteed, private external debt normally represents a lower burden for people in emerging countries and for taxpayers in countries issuing loans. Sustainable debt must be capable of being repaid according to the terms of the contract. Therefore, the ratio of total external debt to GDP must stabilize. Evergrowing EDT/GDP is unsustainable; debt payments cannot be allowed to approach GDP. All other things equal, an increase in exports lowers external debt growth, but foreign exchange earnings from exports are not stable (Stanley and Bunnag, 2001). If the trade balance is assumed to be zero, the debt is sustainable only if the growth rate of GDP is greater than the interest rate. Sustainability is not sufficient to provide for increasing or even constant per capita incomes; it merely ensures the feasibility of repayment. An alternative, more risky, approach is to concentrate on solvency in which the present discounted value of all future payments is equal to the initial face value of the debt. Then the only condition is that the growth of the debt (not GDP) be less than the interest rate. Kiguel argues that the term structure of external debt, not debt service ratios, is the appropriate significant indicator (1999). The term structure is the amount of debt due at a particular time over several years. In this model, extending the period of future repayments and identifying a small share of export earnings assure solvency of the debtor country. However, there is no guarantee that the structural changes needed for export repayments incorporate per capita income improvements or maintenance (Vaggi, 2001). External debt sustainability Columns 1 and 2 in Table 1 list average GDP growth rates and the growth rates of external debt stocks for Column 3 is a proxy for interest payments on external debt. It is the forecasted ratio of debt service to total debt; debt service here includes repayment. If repayment is significant, this differs from the model above, where repayment is treated in the capital account. Note, however, that in all cases, with the exception of El Salvador and Peru, the average growth rate of external debt exceeds GDP growth; therefore, the ratio of external debt to GDP continues to rise. In terms of sustainability, the average growth rate for each country and the area as a whole is less over the period tested than the debt service to debt ratio, an interest rate proxy. The external debt is growing faster than the ability of the countries to service debt; it is, therefore, unsustainable. It may be reassuring to those concerned about the burden of debt to note that poorer countries, such as El Salvador, Guatemala, and Peru pay less debt service to total debt (Column 3, Table 1) than more affluent Latin American countries. However, either in spite of their poverty or due to concessional rates, they accumulate external debt at approximately the same rate as more developed countries.

3 Latin American external debt 361 Table 1. External debt sustainability Nominal GDP Avg. Growth Rate Nominal Average Growth Rate of External Total Debt Ratio Debt Service to External Total Debt: 1999 and 2001 Forecasts Ratio For Direct Investment to Transfers: 1999 and 2001 Forecasts Countries (1) (2) (3) (4) (5) Ratio Private Debt to Debt 1999 and 2001 Forecasts Argentina & & & 0.22 MAPE Brazil & & & 0.52 MAPE * Chile & & & 0.68 MAPE * * 16.21* Colombia & & & 0.48 MAPE * 95.74* El Salvador & & & 0.00 MAPE * Guatemala & & & 0.07 MAPE * * Mexico & & & 0.18 MAPE * 86.19* Panama & & & 0.00 MAPE Peru & & & 0.06 MAPE * Venezuela & & & 0.28 MAPE * Ten Country Composite & & & 0.30 MAPE * Note: A Decomposition Model was used to forecast in those cases where decomposition had a lower MAPE than ARIMA (Wilson and Keating, 2002). Composition of external debt Analysts laud the increase in foreign direct investment (FDI) into Latin America in the 1990s because FDI tends to be less speculative and hence less volatile than portfolio investment. With FDI, stockholders, not the general public, accept loss. Direct Investment may be preferable to other types of capital inflows, but FDI growth in the nineties was accompanied by rapid growth of short-term highly volatile capital flows in and out of Latin America. Table 1 indicates that the ratio of net inflows of foreign direct investment to total transfers cannot be accurately forecasted for a given year (Column 4, Table 1), due in part to the instability of the denominator (net inflows). The use of dynamic forecasting rather than using static ratios at a given point in time emphasizes the vulnerability of countries to capital flows and provides a measurement of risk. MAPE, the mean average percentage error of the forecast, approaches or exceeds 100% in most cases. Column 5, Table 1 presents the ratio of private nonguaranteed loans to total transfers. Lower forecasting errors suggest that this ratio is more stable than DFI and could provide information of the investment climate in a particular country. However, a cross-country comparison of these ratios does not exhibit high negative correlation coefficients, expected if lenders were differentiating between Latin American countries. Forecasting helps in analysing the macroeconomic sustainability of debt (Columns 1, 2, and 3, Table 1). The composition of debt (Columns 4 and 5, Table 1) highlights the role of official agencies in external debt. And the 2001 forecasts correctly anticipate the current business climates in Argentina and Colombia (Column 4, Table 1). Commercial bank lending to Latin America was reduced to slightly more than 10% of total private flows in the early 1990s during which foreign direct and equity investment rose to nearly 40%. Bonds increased to over 20 percent of total transfers (Rojas-Suarez and Weisbrod, 1996). Private to total external debt has increased from the 1980s for Latin America as a whole to 30%, still lower than the 1970 level of 50%. Latin America has not recovered from the closed nationalistic policies of the 1970s and 1980s. During that period, FDI was sought to the extent that it conformed with import substitution rather than to stabilize capital flows and finance research and development. There are benefits to being able to attract private foreign capital of all types. In general, smaller countries are hampered in their ability to attract private capital but

4 362 M. O. Keating and B. P. Keating concessional interest rates, which make external debt more sustainable, foster rather than discourage debt growth. Third parties, multinational organizations as well as host and donor governments, by guaranteeing loans have become major agents of the external debt story. This contributes greatly to the moral hazard of lenders who free ride on monitoring loans. Borrowers, as well, have a decreased incentive to analyse the real cost of servicing debt. IV. CONCLUSION This study ignores the human distress in servicing large external debt. Growth rates are compared with forecasts of debt service to external debt ratios, used here as a proxy for interest rates. The Latin American countries in general do not meet the theoretical test of external debt sustainability using a simple balance of payments identity model. To some extent, the dominance of commercial bank lending to Latin America is being surpassed with foreign direct investment and bonds. Latin America has not yet attained the level of non-guaranteed private to total borrowing that existed in Non-guaranteed private debt, in theory, could be restructured minimizing costs to taxpayers in creditor and debtor countries. However, private creditors have become increasingly difficult to coordinate and can rely on aggressive legal strategy to disrupt a restructuring process (Kruger, 2001). History and theory confirm that the international debt problem is mainly one of low growth and lack of access to international trade. Some lenders and borrowers benefit from debt crises. The time has come to find new contracts that are transparent and workable, ending the free riding of and moral hazard for borrowers and lenders alike. The data and tools available to do this are generally available. Doing so may actually re-ignite growth and international trade, the antidote for excessive external debt in Latin America. REFERENCES Gavin, M., Hausmann, R. and Leiderman, L. (1996) The macroeconomics of capital flows to Latin America, in Volatile Capital Flows: Taming Their Impact on Latin America, (Eds) Hausmann and Rojas-Suarez, Inter- American Development Bank, Washington, DC. Hayek, F. A. (1979) Unemployment and Monetary Policy: Government as Generator of the Business Cycle, Cato Institute, San Francisco. International Monetary Fund (1991) The International Capital Markets: Developments and Prospects, World Economic and Financial Surveys, Washington, DC, 5. Kiguel, M. A. (1999) A note on Argentina s debt management strategy, J. Applied Economics, 11, Kruger, A. (2001) International Financial Architecture for 2002: A New Approach to Sovereign Debt Restructuring, International Monetary Fund [Online] 26 November. Available: [2002, 28 February]. Rojas-Suarez, L. and Weisbrod, S. R. (1996) Achieving Stability in Latin American Financial Markets in the Presence of Volatile Capital Flows. In Volatile Capital Flows: Taming Their Impact on Latin America, (Eds) R. Hausmann and L. Rojas-Suarez, Inter-American Development Bank, Washington, DC. Stanley, D. and Bunnag, S. (2001) A new look at the benefits of diversification; lessons from Central America, J. Applied Economics, 33, Vaggi, G. (2001) Trade and sustainable finance for development, Jubilee þ [Online] (May). Available: [2002, 28 February]. Wilson, J. H. and Keating, B. John Galt Solutions, Inc. (2002) Business Forecasting. McGraw Hill, Boston. World Bank (1999) Global Development Finance 1999, World Bank, Washington DC.

5