Written for presentation at the annual meeting of the Eastern Economics Associations, Baltimore, March 8, 1989
Abstract: This article analyzes the role of risk in the less developed country (LDC) debt crisis. LDC debt, mostly in the form of syndicated Eurocurrency credits, exposed borrowers to the risk of fluctuations in foreign exchange, interest rates, and export prices. When all three of these risk factors moved against LDCs in the early 1980s, the crisis hit. Even the apparent over-lending can be partially explained as the result of banks and LDCs trying to stave off the impact of higher real debt payments. This was not an unforeseeable problem. Thus the question must be asked, why were not more appropriate financial instruments not used to meet LDCs' borrowing needs?
* Department of Economics, Columbia University, New York, NY, 10025.
The thesis I want to advance in this study is that international bank loans laid a risky foundation for financing development in less developed countries (LDCs). By shifting nearly all sources of risk onto debtors, International bank loans created the condition for the possibility of the debt crisis. Once the crisis struck, the initial bailouts and then the Baker Plan of exchanging austerity for new lending led to an even further rise in total outstanding debt. Today's indebtedness of hundreds of millions of Third World workers to a few international banks gives new meaning to Churchill's words of thanks to the R.A.F., "Never before in the history of human conflict, have so many, owed so much, to so few." The best way to understand the risky nature of financing development through international bank loans is to compare the risk characteristics of international bank loans to other forms of international capital flows such as foreign direct investment (FDI) and currency-diversified bonds. (I take as a premises here that international banks played the decisive role in determining that International bank loans would be the predominant debt instrument for North-South lending. See my article, Dodd (1989), for a justification of this point.)
During the first two decades following the second World War, most private capital flows to LDCs were in the form of FDI. Then in the late 1960s, many LDCs as well their state-agencies and state owned enterprises and private businesses began to borrow from international banks in the U.S. and the growing Eurodollar markets. At the same time international banks became eager to lend to LDCs, and Kindleberger (1978, p.23-4) stated this very well.
"The enormous external debt of the developing countries, built up not only since the rise of oil prices but importantly -- a widely ignored fact -- in the several years before that time, as multinational banks swollen with dollars tumbled over one another in trying to uncover new foreign borrowers and practically forced money on the LDCs."
Together this markedly changed the composition of capital flows in the 1970s. Financial capital, mostly in the form of international bank loans, flowed in increasing magnitudes from international capital markets to LDCs. As early as 1970, net flows of financial capital matched that of FDI. In the year after the 1973 oil shock, the sum of international bank loans and international bond issues was 250% that of FDI and outstripped official development assistance for the first time. This trend would continue until the debt crisis came to a head in 1982. The new composition of capital flows restructured the international economic relationship between the industrial countries in the "North" and LDCs in the "South".
The substance of this new relationship was the new distribution of risk and sovereignty. By borrowing in order to augment their own capital stocks, LDCs sought to regain sovereignty over their domestic capital stocks and investments. But borrowing in the form of International bank loans allowed international financial capitalists to shift much of the market risk of investing and borrowing onto LDCs. Thus, after the crisis hit and the IMF arrived, LDC governments discovered that in trying to regain sovereignty over their domestic capital stocks they had lost sovereignty over their monetary and fiscal policies.
Different types of risk are associated with FDI and international bank loans. The total risk on FDI, from the perspective of international investors, breaks down into several sources of risk: business failure; natural disasters (acts of God); foreign exchange fluctuations; expropriation or nationalization; and world business cycle. From the perspective of LDCs, the risk on FDI is of little concern. The risk of business failure, which might be called entrepreneurial risk, is associated with any direct investment in a business enterprise. The returns on foreign investments are generally larger than those on domestic investments because foreign investments are more uncertain than similar activities at home due to the lack of information, especially the unfamiliarity with foreign rules-of-the-game, and the relatively less stable economic environment. Whatever the cause, business risk on FDI is borne by foreign investors. FDI also incurs the risks of 'natural' disasters such as floods, droughts, poor harvests and earthquakes. As a result of the distribution of risk associated with FDI, the social losses resulting from a natural disaster would be diversified or spread abroad. Foreign exchange risk is another source of risk international investors face on FDI. An appreciation of the home country's currency, e.g. the U.S. dollar, or a depreciation of the developing country's currency would reduce the dollar value of foreign profits as they were repatriated to the U.S.
The international monetary system of fixed exchange rates founded at Bretton Woods all but eliminated this source of risk, but since the collapse of that system in 1971, foreign exchange risk has reemerged. Nonetheless, foreign exchange risk is borne entirely by international investors. Foreign exchange risks are sometimes mitigated by locating FDI in the export or import competing sectors where, theoretically, the home currency rate of return is maintained despite inflation and depreciation by the purchasing power parity or the Law of One Price. Yet, even if foreign investments earn "hard" currency in the export sector, host governments often require foreign earnings to be converted at a fixed rate to domestic currency at the central bank. As a result, the foreign exchange risk again fell on the international investors.
Expropriations and nationalizations, especially during the late 1960s and 1970s, are another source of risk. The investment experience of FDI for international investors in the 1970s were described by Naim (1987) as a "time of trouble for international business." Hawkins, et al, (1975) provides a litany of the numerous cases of nationalizations and expropriations. They estimate that foreign investors lost over $2 billion during this turbulent period. This might sound like a trifle now, but the real dollar value of the political significance were not. It seems that the political meaning of the capital losses were not lost on the international investors. These initiatives came primarily from Leftist political movements, although they ended up being administered by Rightist governments (Hawkins, 1975; Dare and Weeks, 1976). Even when these political movements were not entirely successful at seizing state power, they often succeeded in crippling economies and the established political order (e.g. Uruguay, Dominican Republic). At the least they succeeded in placing restrictions on foreign investments. As Naim (1987) put it, "By the early 1970s in Latin America, granting explicit incentives and other forms of preferential treatment toward foreign firms became almost unthinkable."
Another source of risk on FDI arises from the powerful influence of world economic business cycles on LDCs. A foreign led recession would adversely affect many foreign owned businesses since, especially those concentrated in the export sectors of LDCs. On the other hand, the risk borne by LDCs from hosting FDI is relatively small. The impact of either business failure, natural disasters, foreign exchange fluctuations, expropriation or nationalization, and world business cycle would be less than if the enterprises were locally owned. In the event of a complete collapse of a foreign business, the abandoned plant and equipment would remain in the LDC. Expropriations or nationalizations would be a wind fall, at least initially, although there would always be the threat of a cessation of future capital inflows and the likelihood of capital outflows. In the event of an exodus of foreign capital, real physical investments, such as plant and equipment, would have to be sold either to domestic investors or to some other, more "friendly," foreign investors. A widespread effort by foreign investors to liquidate their holdings on the host country's domestic capital market would have the effect of greatly depressing the price of such capital. Retreating foreign investors would suffer further capital losses as they sold their capital for the host country's currency and then flooded world foreign exchange markets with it in an attempt to repatriate their capital. Without doubt, the expectation of just such losses would inhibit foreign investors, as a whole, from withdrawing their capital. Selling the capital to other foreigner investors would merely change the nationality of the foreign investments and would thus have no adverse consequences for the host LDC.
Changes in earnings associated with global business cycles would, in this manner, mean that the cost to LDCs of importing foreign capital would be pro-cyclical. Overall, the risk on FDI falls almost exclusively on the international investors. This, it might be argued, is an appropriate distribution of risk since multinational corporations (MNCs), who are the source of most FDI, are very capable of managing such risk. Risk management by MNCs is made easier because many of the above sources of FDI risk are negatively correlated. For instance, an appreciation of the U.S. dollar (or depreciation of the foreign currency) would on one hand increase the U.S. demand for exports and on the other hand reduce the rate of converting foreign currency earnings back into dollars.
Before making the parallel analysis for International bank loans, an apt definition of International bank loans is in order. International bank loans are generally syndicated Eurocurrency (mostly U.S. dollar), variable rate bank loans of intermediate-term maturity. They contain cross-default clauses and high front-end fees. They are issued either directly to states in LDCs or to private businesses with guaranteed by the state. An estimated 94% of international loans to Latin America were U.S. dollar denominated, and most private International bank loans were also variable rate (World Debt Tables). Syndicated Eurocurrency credits are loans issued by a group of banks which are put together by a "lead bank or core group of banks" from financial centers such as New York. Other large and small banks who join in the agreement become minor partners by purchasing a portion of the value of each loan. Syndication often involve consortium banking practices which consist of the major banks pooling their research and analysis efforts for the purpose of issuing loans. Together this led to a more uniform lending practice, and a more united response by creditors once the crisis hit.
Interest costs on variable rate debt are set according to some "spread" above LIBOR, the prime lending rate in the U.S. or other short-term interest rate. When LIBOR rises (falls), the interest rate charged on outstanding principle and hence the interest payment rises (falls). But a one percent rise in LIBOR does not translate into a one percent rise in payments. Instead, a percentage rise in the base rate raises the interest payment by a percentage that is an exponential function of the magnitude of the interest rate and the maturity of the loan. For example, a one percent rise from 9% to 10% on a 20 year loan raises interest payments approximately 10%. LDC borrowers are also charged substantial front-end fees for initiating the loan and commitments to a credit line (Goodman, 1980). While this is not a source of risk, it is a considerable cost that enters into the calculation of whether the loan is profitable. One factor that does significantly increase the riskiness of International bank loans is their effectively short-term maturity. International bank loans are not short-term assets in respect to their maturity, which range upwards to 15 years and average from 7 to 10 years (World Debt Tables). They are effectively short-term assets because their interest rates are adjusted at frequent intervals in accordance with such short-term rates as LIBOR (Goodman, 1980). This difference between actual and effective maturity is akin to the notion of duration which is of great importance in comparing various lending instruments.
Cross-default clauses mean that borrowers can not default against one single or particular group of lenders, and that the default by one borrower in a country will be treated as a general default. In an ironic twist, the international banks have expropriated a slogan from the Industrial Workers of the World, that a (financial) injury to one is an injury to all. These features strongly shape the distribution of risk between borrower and lenders. International bank loans, as opposed to FDI, do not expose international investors (mostly Eurodollar market banks) to foreign exchange risk. The variable interest rates on international bank loans do not expose them to dollar interest rate or inflation risk.
The return to international bank loans, unlike FDI, are isolated from the costs of natural disasters. Holding loans, instead of equity, means that international investors have fewer worries about labor strife, expropriations and nationalizations. And unlike FDI, a worldwide recession would not necessarily reduce the real return on foreign investments. The only risks they are exposed to are illiquidity of loans and sovereign or default risk. The later is reduced by syndication which institutionalizes the relationship among creditor banks and insures that they would act in concert in order to impose sanctions or take other measures necessary to enforce loan payment. Moreover, most International bank loans are guaranteed by LDC states so that even the risk of default is greatly reduced. It would have been a reasonable presumption by international banks that LDC states, as opposed to private firms, would have much greater ability to meet debt payments because of their taxing power and the growth in exports.
International bank loans from the perspective of LDCs were from the beginning a disaster waiting for a time to happen. They bear the risk of U.S. dollar appreciations, the risk of rising U.S. interest rates, the entire risk of natural disasters and business failures, and the risk associated with global business cycles. And as time tore off the false side-whiskers of fate, all these dangers arrived en masse and, in some cases, stayed. As an indication of past mistakes, efforts in recent years to mitigate the debt crisis have sought, although perhaps unwittingly, to change the features of the debt which are criticized above. Rescheduling has lengthened the maturity of the debt as interest payments have been amortized. The portfolio of debt has become more currency diversified, and attempts have been made to swap and renegotiate existing debt into fixed rate debt. And the creation of a secondary market testifies to the banks' own need for a more liquidity lending instrument.
In order to evaluate the appropriateness of any debt instrument, its properties must be analyzed in terms of borrowing needs. Most LDCs have three basic types of borrowing needs, and each is associated with a different term of maturity: (1) short-term trade credits; (2) intermediate-term adjustment borrowing; (3) long-term development project borrowing. Short-term credit is needed in order to encourage exports. These export credits are in the form of privately and publicly issued, short-term instruments as bankers acceptances, discounted bills of exchange, as well as private and public (states as well as multilateral agencies) sponsored trade credits.
A sizable minority of outstanding LDC indebtedness is made up of rolled-over short-term debt; it was ostensibly borrowed for use as trade credits but has functioned in many cases as balance of payments bridge loans. Intermediate-term credit is needed in order to adjust to real changes in world commodity prices. The clearest example of this was the enormous rise in borrowing that followed the 1974 oil shock. As Martin and Selowsky (1981) have argued, the cost of adjustment in an imperfectly flexible world is inversely related to the length of the adjustment period. The borrowing costs could therefore be justified by the benefits of extending the adjustment period. Long-term credit is needed in order to finance (long-term) investment in social overhead capital and fixed plant and equipment. The gestation period on such investments is often 10 to 20 years, and therefore the costs of borrowing should be fixed for the same length of time (Williamson, 1987, makes this same point). This is what is most often thought of when the term "development finance" is used. These problems of the past suggest a solution in the future. Development should be financed with long-term, fixed-fate, currency diversified loans or bonds that include contingencies for disasters and economic downturns.
Part of this was suggested by Pollack back in 1974. He recommended that cash rich nations and institutions should buy World Bank bonds so that the World Bank could expand its development oriented lending. World Bank lending is generally in the form of intermediate to long-term, fixed rate loans denominated in various currencies. It is also important to note that the World Bank was less strict than the IMF about the conditionality of loans. Thus one might infer that Pollack was recommending long-term, fixed rate development finance. Currency diversification is an important aspect of LDCs borrowing needs because the gains to diversification are double. Primarily it would also reduce debtors' foreign exchange exposure. It would also reduce interest rate risk by smoothing out the inevitable changes in interest rates. Mohl and Sobol (1983) have shown that diversification would have reduced the interest rate variance and effective cost of borrowing by over $30 billion for the years 1979-1982 alone. One road not taken, was that of bond lending. Historically, many people and financial institutions took heavy losses following the debt default and repudiation by LDC in the 1930s (Diaz-Alejandro, 1982). In order for LDCs to use bond financing, there have to be bond underwriters, issuers, and investors (purchasers of bonds). The growth of the Eurobond market, which emerged 22 years ago in 1963, exploded in recent years when the quantity of new bond issued reached $80 billion in 1984 and $133 billion in 1985. Hypothetically, the same Eurodollar market banks which issued most of the International bank loans could certainly have underwritten or purchased Eurobonds issued by developing countries, after all, these banks are not subject to any government regulation of their portfolio holdings.
One immediate conclusion gleaned from the above discussion is that the manner in which development is financed is crucial. This would apply to both LDC workers who are the ultimate debtors or to any new socialist state that is integrated into the world capital market. To close with a misquote from Churchill upon the disaster at Dunkirk (another instance of the problem of repatriating foreign resources), the growing interest in and demand for debt forgiveness signals not the end of the beginning but the beginning of the end.
Dare, Elizabeth & John Weeks. 1976. "The Intensification of the Assault Against the Working Class." Latin American Perspective 3.
Diaz-Alejandro, Carlos. 1982. "Latin America in the 1930s." Yale Economic Growth Center Discussion Paper.
Dodd, Randall. 1989. "How and Why International Banks Issued Mostly Syndicated, Variable Rate Loans to LDCs." Paper to be presented to conference "Global Imbalances" at American University, May, 1989.
Goodman, Laurie. 1980. "The Pricing of Syndicated Eurocurrency Credits." FRBNY Quarterly Review, (Summer).
Hawkins, Robert G., Norman Mintz and Michael Provissiero. 1975. Governmental Takeovers of U.S. Foreign Affiliates: A Post-war Profile, Center for Multinational Studies, #7.
Kindleberger, Charles. 1978. Manias, Crises and Panics: A History of Financial Crises. New York, Basic Books.
Martin, Ricardo and Marcelo Selowsky. 1981. "Energy Prices, Substitution, and Optimal Borrowing in the Short-run." World Bank Staff Paper, #466.
Mohl, Andrew & Dorothy Sobol. 1983. "Effects on LDC debt payments of currency diversification on a trade weighted basis." Quarterly Review, Federal Reserve Bank of New York.
Naim, Moises. 1987. Government regulation of foreign investment: emerging lessons from Latin America, in Richard D. Robinson (ed.), Direct Foreign Investment. New York: Praeger.
Pollack, Gerald. 1974. The economic consequences of the energy crisis. Foreign Affairs (Spring).
Williamson, John. 1987. Political Economy and International Money. New York: NYU Press.
World Bank. World Debt Tables, 1970 -1984.
 ) New York Times, April 20, 1986.