Chapter 1. Trade Finance and Financial Crises

Jian-Ye Wang, and Márcio Ronci
Published Date:
February 2006
  • ShareShare
Show Summary Details
William R. Cline9

Is There a Problem?

Are declines in externally provided trade credits a serious problem for a country’s crisis resolution efforts? My answer, in short, is yes. Trade credit lines are short term in nature and routinely rolled over under normal circumstances. However, when a country enters into a crisis, foreign lenders tend not to renew such credit lines for fear of being caught up in some form of suspension, or simply because of concern about greater default risk by firms now facing more difficult circumstances. Similarly, official bilateral export credit agencies also tend to curb lending during crises following general prudential guidelines.

It is useful to ask about why supply and demand doesn’t resolve the problem through a modest rise in the risk spread of the credit. Thinking further about this, however, leads toward a powerful conclusion: it is extremely difficult to compensate significant expected capital losses through higher interest on short-term paper. In the Scandinavian banking crisis in the early 1990s, overnight bank rates rising to 500 percent reflected a significant capital loss risk that was being compensated by a meteoric interest rate for a very short period.

Consider, for example, a country that goes from no risk to a 10 percent loss probability. On its 10-year paper, the spread has to rise 150 basis points (Table 1.1). Over the life of the new bond, enough extra interest can be collected (an extra 15 percent cumulative) to compensate the rise in expected loss. But if the country has an expected loss of 10 percent during the next four months, and the interest on a four-month credit has to compensate, then this short-term loan will have to bear a spread of 3,000 basis points, that is, 30 percent. Thirty percent annually on a four-month obligation yields the 10 percent insurance against capital loss. In practice, the lender will simply not issue the loan rather than issue it at an annual interest rate of 30 percent. The basic point is that we can expect short-term credit volumes to be cut back rapidly in a crisis, rather than a smooth adjustment by a minor increase in interest rate.

Table 1.1.Required Spread to Compensate for an Expected Capital Loss of 10 Percent
10-year bond:150 basis points
4-month credit line:3,000 basis points

What is the experience of short-term trade flows during crisis? Figure 1.1 shows the level of total short-term external debt in the year before the crisis and the crisis year for six country cases. The largest drop was in Korea (where short-term debt was even higher the year before what is shown in the figure), but the short-term debt there went far beyond trade credit. The other crisis episodes all show substantial drops in short-term debt, however, and typically a runoff in trade credit was involved. In Argentina, from June 2001 to June 2002 total short-term debt fell by more than 50 percent.

Figure 1.1.Short-Term External Debt During Crisis

(In billions of U.S. dollars)

Source: World Bank.

The above data do not discriminate among types of trade credit. It appears, however, that the most sensitive trade credit is for export prefinance. This is essentially a loan to the export firm not secured against goods already in the process of being shipped. Where there is a letter of credit against actual goods, there is physical collateral so the trade credit holds up better in a crisis.

Do such cutbacks represent a serious problem? Available evidence suggests so. One of the most serious effects of a collapse in trade credits is the adverse impact on exports, which—even after the benefits of the depreciation that accompany most crises—tend to be held back by the absence of sufficient trade credits. The increase in exports is critical not just to provide greater foreign exchange earnings, but also to generate a source of demand that offsets the reduction in domestic demand associated with fiscal tightening and the reaction of households to lower real incomes. On both these counts, therefore, the collapse of trade credits is a major problem and the removal of obstacles that stand in the way of an export recovery should be a top priority.

What Kind of Remedies Might Work?

To the extent that there is widespread agreement that a problem exists, a second key question for consideration relates to the type of remedies that might be put in place. In this regard, remedies typically used in much earlier crises are unlikely to be useful. In particular, the debt crises of the 1980s consisted of obligations owed to commercial banks that were often rescheduled through the London Club framework during times of distress. As there was recognition that countries needed to achieve export growth in order to service the debt, there tended to be side agreements providing that short-term credit lines, especially trade credits, would be maintained at existing levels.

The past tendency to exempt short-term trade credits from reschedulings played a role in encouraging banks to lend in this category more than in longer-term debt. One review of 190 cases of London Club reschedulings from 1979 through 1999 found that in more than two-thirds of the cases, there was no suspension in payments on trade credit. Where there were suspensions, three-fourths of the cases involved trade credit maintenance agreements with no loss of principal or interest. Where there was default with loss—in only 17 of 190 cases—the loss was modest, only 11 percent. This experience has meant that in banks’ internal economic capital allocations there tends to be a lesser risk weight attached to short-term trade credit than to longer term credits.

In today’s capital markets, however, long-term syndicated loans from commercial banks no longer comprise much of the debt. Instead, long-term debt tends to be in the form of bonds, and the opportunities to utilize side agreements protecting trade and short-term credits tend not to arise.

Under these circumstances, special coordinated efforts among international banks to deal with short-term debt can help. Two episodes are prime examples:

  • In Korea in 1998, after much of the short-term debt (which reached some $100 billion and included debt far beyond that related to trade) had already run off, it became necessary to supplement the initial IMF support package with an agreement converting $22 billion in short-term credits of international banks to Korean banks into three-year paper at somewhat higher spreads than before and with the guarantee of the Korean government. This stretching out of maturities offered by this deal proved crucial to halting the downward spiral of confidence and setting the stage for stabilization and adjustment.

  • In Brazil, a less formal arrangement was reached by the international banks in March 1999 to maintain their short-term credit lines (then approximately $25 billion) through August. Some key banks interpreted the agreement to be contingent on the Brazilian government implementing promised policy adjustments. The arrangement was relatively informal in part because the Brazilian authorities were extremely reluctant to engage in anything that appeared to be a forced rescheduling of debt, in order not to jeopardize the credit standing Brazil had rebuilt after the debt crisis of the 1980s. The closest thing to a formal component of the arrangement was a form of joint Brazil-IMF monitoring of credit levels. The arrangement was successful, and the voluntary nature of the approach helped avoid a slide toward default and capital controls—which might have been a greater risk if instead there had been some mandatory form of “stay” imposed on credits.

What Are Appropriate International Policy Responses?

One response by the international community in recent years has been to search for systemic disincentives to short-term capital, with a view to limiting their excessive volatility. Short-term lending developed a negative reputation in the East Asian crisis, and countries such as Chile, which imposed tax disincentives to short-term debt, have enjoyed some insulation from the capital market crises of the late 1990s. In addition, it is notable that the Basel Committee reforms for the capital risk-weightings of banks are considering boosting capital requirements against short-term loans.

The challenge of any such initiatives is to arrive at a balance that provides scope for disincentives to excessive short-term debt while retaining creditor confidence that legitimate trade credit will tend to have less default risk than longer-term credit. This challenge is difficult, as short-term financial flows unrelated to trade can be reported as trade credit for regulatory purposes, as occurred in Brazil in the mid-1990s when more favorable treatment resulted in a ballooning of reported short-term credit beyond traditional levels relative to trade. At the same time, some recent initiatives, such as the first version of the IMF’s proposed Sovereign Debt Restructuring Mechanism, had included elements that could have heightened the perceived risk of legitimate short-term credit, most notably some form of an IMF-sanctioned “standstill” enforced by a “stay of litigation.”

Other responses by the international community can be centered on bilateral credit agencies, which could help by adhering to a code of conduct whereby short-term credit lines—including expiring maturities of originally longer-term credits—would be rolled over or expanded so long as the country is in an IMF-supported adjustment program. Although the U.S. EXIM Bank did have a modest expansion of short-term lines to Korea during that country’s 1998 crisis (amounting to some $900 million), this increase was largely offset by a reduction in long-term credit; and medium- and longer-term credits fell in the crisis cases of Russia and Indonesia. The Japan Bank for International Cooperation played a considerably more active role in providing financing in the principal financial crisis cases. What is broadly needed is a special exception to normal credit-risk practices of bilateral agencies, which understandably curb lending when countries’ credit ratings fall, to provide for emergency support as part of internationally-coordinated adjustment programs.

    Other Resources Citing This Publication