There is a tendency in today’s world of robust, albeit at times skittish, capital flows to developing countries to say that the developing country debt crisis is over, and that, consequently, the challenge of external debt management is not as urgent as it was just a few years back. There are at least three reasons why this is not quite right. First, debt distress remains a central economic reality for the majority of low-income countries. Second, there are new countries, arising out of the former Soviet Union and Yugoslavia, that are embarking on borrowing programs without the necessary infrastructure or regulatory and legal frameworks that permit adequate management of the borrowing process. And, third, there is recidivism; that is, the possibility either that the lessons learned from past practices were inadequately retained, or that the lessons had changed.

The recent crisis in Mexican economic management has shown that it would be premature to judge that, even for middle-income countries, the debt crisis really is over. The Mexican crisis was a failure in debt management. Some elements of the Mexican situation were very different from a decade ago. On the one hand, the country did not face the extreme terms of trade shocks combined with exceedingly high real interest rates which had triggered its default in the early 1980s. On the other hand, Mexico did choose to finance a large current account deficit with short-term borrowing, and it was deploying these resources to finance a boom in consumption. It was this unsustainable borrowing strategy that altered Mexico’s debt profile so rapidly that the government’s policy responses lagged behind the rapid buildup of debt service obligations.

The Mexican authorities financed their deficit mainly by using short-term, domestic debt instruments. Therefore, the crisis initially had the features of a domestic debt problem. However, since the instruments were dollar-linked and held by foreign as well as domestic nationals, the problem was external debt. The rescue package that the international financial community assembled effectively switched the Mexican portfolio back to medium-term, dollar-denominated external debt obligations. This reverse switch was designed to be quite rapid, putting a burden on the authorities to demonstrate not merely that policy adjustment had been effected, but that their ability to manage all the complexities of the capital flows taking place had been restored. The need for sound management of external debt had once again been thrown into stark relief in Mexico.

Mexico is, of course, part of a broader picture demonstrating how immeasurably more complex debt management is becoming in today’s global financial world. There are many more financing instruments available to economic managers than there were when the debt crisis struck in the early 1980s. With liberalization of financial markets and the changing nature of governmental and private sector relations worldwide, there are also many more borrowers than there were in the 1980s. Today, flows from the private sector to private sector beneficiaries constitute over 70 percent of all long-term flows to developing countries. Thus, as will be outlined below, the very definition of debt is changing. Given these changes, the emerging issue is the following: Are management practices and tools—developed in response to the debt crisis of the 1980s—up to the task of managing capital flows in today’s world?

Lessons of Debt Management⁵

The following five interrelated functions need to be mastered if a country’s debt management is to be put on a sound footing.

The policy function. This function involves the coordination among all agencies responsible for formulating borrowing policies and debt strategies. The government needs to be clear and firm at the outset about the limits to borrowing, both in terms of amounts and responsibility for conducting the borrowing, and its end uses. A consensus is also needed on prospects for the sustainability of the borrowing program over time. Judgments are needed about export prospects over the medium term. Relatedly, the likelihood of new funds, the terms of these flows, and whether they will be used to maintain consumption or to invest in production, especially production for export, need to be assessed and agreed upon. This is not an easy task, since the sustainability of flows depends on many variables, controlled by different interests.

The regulatory function. Borrowers must be required to report borrowing activity—both the contracts themselves and the individual transactions thereunder—covering both loan receipts and repayments. Governments must establish efficient mechanisms for handling these reports and acting on their implications. It is important to avoid the conclusion that the changing role of government in the economy would diminish or eliminate the need for regulations that govern borrowing. Removal of ex ante approvals for borrowing actually accentuates the need for clear mandates and responsibilities in reporting each borrowing and the transactions thereunder. Policies and practices for issuing guarantees should be as coherent as for borrowing, since they are contingent liabilities on the government balance sheet.

The operational function. Governments need to choose explicitly where they wish to position themselves on the passive/active debt management spectrum. Passive debt management entails the registration of all loan activity and drawing their implications into balance of payments calculations. Active debt management requires choices on the extent of market penetration and participation, choice of markets, instruments, currencies, and maturities to manage risk. Trading in currencies or interest rates so as to reprofile a debt portfolio is a vastly different set of activities from debt reporting and monitoring. Specialized training is a vital prerequisite for success in this “high tech” area. The possession by traders of specialized skills makes them difficult to manage. Precise guidelines are needed to circumscribe trading activities within the limits imposed by government policy. In other words, government choices concerning the extent of active debt management are strategic, and must be made prior to trading. Simply put, whether to hedge or take positions in currencies and interest rates are choices that should not be left to traders.

Another dimension of the operational function is that these choices have to be consonant with the management of all capital flows. With shifts in the composition of capital flows toward a greater proportion of non-debt flows, the rhythm and timing of obligations created by these flows needs management. After all, profit remittances compete with debt service. If the country’s investment climate is to be kept untarnished, profit remittances become as critical as inflexible debt on the government balance sheet. At present, Africa accounts for about 3 percent of all foreign direct investment; most of the flows are debt flows. With policy reform that liberalizes productive and financial systems, however, this proportion cannot but grow, and the implications of such a change need to be figured into a cohesive framework for the management of all capital flows.

The accounting function. An appropriate accounting framework is needed for defining debt. When capital accounts were controlled and the number of borrowers limited, the definition of debt was relatively straightforward, and indeed somewhat static. The definition consisted of the government/s external and contingent liabilities; foreign exchange to meet private sector debt; medium- and long-term debt owed by the public sector to nonresidents; short-term public sector debt, direct investment; and all private sector debt (both medium- and short-term)—with the balance among these categories continuously shifting over time as economies opened up. Because of the diversification that flows cause, there is now a need to include domestic debt held by foreigners.

Establishing a sound accounting framework, however, is not an end in itself. The organization of the debt management function must reflect the diversity of instruments, sources of capital, and creditors (or investors). Governmental decisions are needed on an appropriate institutional setup to enable sound management of flows. The decisions should cover both the locus of the accounting function and the management information systems to be used. Where capital flows consisted mainly of debt flows to the public sector, the Ministry of Finance and the Central Bank were perfectly suited to manage external debt, which was, as mentioned, largely government or government-guaranteed debt. With the expansion of borrowing by and lending to the private sector, it has become essential for the financial sector at large to be integrated into the system. To facilitate central bank supervision of the commercial banking sector, uniform mechanisms became necessary with which private sector borrowing activity, conducted through commercial banks, could be reported to the central financial authorities.

The statistical function. Databases created in the management of capital flows are very large. Russia’s debt database captures about 5,000 loans. And, of course, under each loan, the transactions are manifold, making database management an exceptionally challenging task, and the trade-offs between coverage and accuracy are not easy ones to make. Comprehensiveness of coverage is essential, but so is the necessity to produce focused and timely reports from it, designed to maximize contributions to policymaking. At first glance, the statistical function appears to be the most pedestrian of the key functions of debt management. But, it is the pillar of the entire system.

At the height of the debt crisis in the 1980s, many institutions (and, indeed, governments) were involved in the development of computer software packages that would facilitate the database management function. The World Bank attempted to create a system that would be compatible with its own debtor reporting system to enable the electronic transfer of existing data without requiring additional staff in reporting countries. Many of these efforts fell by the wayside as database complexities mounted, as requirements for customization of reports increased, and as the software development and maintenance costs rose. Today, two major database systems are available—one developed by the United Nations Conference on Trade and Development, and the other by the Commonwealth Secretariat. These two systems are similar in purpose, but are dissimilar in database construction and design. Governments must choose one system prior to its installation because compatibility of the debt database is paramount, since it must be shared by all participants in debt management (for example, the Ministry of Finance and the Central Bank should have the same system).

A software package, even if it is comprehensive and user-friendly, is a useful management tool in this complex endeavor. But it is not magic. It is important to allocate sufficient resources (staff and equipment) for the training in debt management techniques. Experience has shown that attrition rates for trained staff are high, as they find attractive markets for their skills elsewhere in the government or in the private sector. Government heads should recognize this reality, accept that such developments are inevitable (and not always detrimental if the skills are deployed elsewhere in the economy), and act accordingly by training sections and units of staff, rather than individuals.

An issue related to software development is whether management tools can keep up with the growing complexity of debt management in rapidly changing financial markets. System upgrading can take two or more years. It will be important for the software providers to strive constantly to avoid anachronism.

In reviewing the multiplicity of functions required for effective debt management, the inevitable conclusion is that mastery of all functions is necessary. Governments can, however, be selective in the sequencing of their tasks. Passive debt management is a useful starting point. Simply put, the registration of all loan obligations in a manageable form is an essential prerequisite to trading. One cannot trade assets or liabilities if the origins and destinations of payments due under them are unknown.

Technical Assistance

Sound debt management requires teamwork if all the critical functions are to be mastered. In debt offices, staff attrition rates are high. Therefore, training programs should be designed to target teams to avoid discontinuities attendant on the departure of individuals. A concrete capacity-building program must be designed to integrate technical assistance with training. The rapid evolution of financial markets and instruments, as well as the emergence of new creditor policies and practices exercised in the Paris Club and other fora, all mean that training becomes a quasi-continuous, rather than a discrete, activity.

In the past, training was conducted from afar. This practice was neither efficient nor effective. Throughout the 1980s, both software providers and the World Bank trained staff from the debt office as needed in the use of software packages for debt management and reporting (the Bank’s debtor reporting system). The Africa region is pioneering a new approach. The creation of the ESAIDARM (Eastern and Southern Africa Initiative in Debt and Reserves Management) in early 1993 created a subregional cooperative entity onto which training and technical assistance can be devolved. It will conduct core training, and will provide technical assistance on demand in the entire spectrum of needed areas. The location of the ESAIDARM in the field will shorten response time in providing technical assistance, will foster the cross-fertilization of local experience in debt management, and will use local expertise to the extent possible.

While the decentralization of technical assistance is considered an important ingredient in ensuring the cost-effective provision of required services, it should not lead to isolation. The ESAIDARM must stay abreast of developments in global finance and developments of tools that will facilitate effective debt management.