Federal Reserve Bank of San Francisco

FRBSF Economic Letter

1996-32 | November 1, 1996

More Economic Letters

Why Is the Philippines Repurchasing Its Brady Bonds?

Mark M. Spiegel

In September, the Philippines announced it would issue $1.9 billion in Eurobonds to finance a repurchase of outstanding Philippine “Brady bonds,” the securities acquired by banks in the Philippines debt restructuring under the Brady Plan in 1992. The Brady Plan was a program of debt reduction partially financed by official institutions to allow highly indebted countries to repurchase debt at a discount. Debt reductions have taken place under the Brady Plan for almost all of the highly indebted nations.

The Philippines transaction involves swapping collateralized debt–the Brady bonds, whose principal is backed by U.S. Treasuries–for non-collateralized bonds–the Eurobonds. Since the principal component of the Brady bonds was collateralized by U.S. Treasury bonds, they were less exposed to the sovereign risk generally associated with loans to the Phillippines. It follows that the Eurobonds the Philippines would be issuing in exchange for the old Brady bonds would bear greater sovereign risk exposure.

The motivation for this exchange is not immediately apparent. The market for Brady bonds is considered a well-functioning financial market. Therefore, under normal circumstances the market should correctly discount the Philippine Brady bonds according to the probability of default on Philippine debt. The transaction would then have no real impact on the Philippines or its creditors. In financial theory, this notion that in frictionless financial markets the mode of finance has no real impact is associated with the well-known Modigliani-Miller theorem (1958).

However, the comments in the popular press indicate that the transaction is expected to have real effects. For example, the Philippine Secretary of State for Finance, Mr. Roberto de Ocampo, has said that the transaction would help the Philippines “shed its image as a rescheduling country.” The Financial Times reported that the transaction was expected to reduce the Philippines’ cost of servicing its $40 billion external debt, without speculating on why this might be the case. Moreover, secondary market prices of Philippine debt obligations were up slightly on the news of the anticipated repurchase. All of these indications that the swap will have real effects suggest that the Modigliani-Miller theorem fails to hold in the Philippine case.

In this Economic Letter, we examine why such a repurchase of its Brady bonds may be in the Philippines’ interest. We first motivate the conditions under which the issue of collateralized debt such as the Brady bonds might be warranted. We argue that those conditions are not consistent with the Modigliani-Miller framework. In particular, we suggest that when a sovereign debtor, such as the Philippines government, has interests that conflict with those of its creditors, it may choose to guarantee a portion of its debt, despite the fact that the issue of collateral is costly. We then discuss how it may be rational to retire this secured debt, as the Philippines is currently doing, when conditions improve.

Sovereign risk and agency problems

One argument for the collateralization of loans is that the existence of collateral can influence the behavior of debtors. In particular, the literature has stressed the ability of collateral to address “agency problems” that are likely to arise in lending situations (Stulz and Johnson 1985). Agency problems arise when the goals of creditors (the principals) differ from the incentives faced by the debtors (the agents). Under such circumstances, the agent may not be able to commit credibly to certain actions that are in the interest of the principal, even though making such a commitment might be in the agent’s interest.

Agency problems are likely to arise in sovereign lending situations. When foreign creditors have a claim on the proceeds of any investment that takes place in a nation, that nation has an incentive to invest less than it would without this debt obligation. The debt obligation itself can therefore distort the investment decisions of the debtor.

Indeed, it was the desire to mitigate these investment distortions that originally motivated the Brady debt reduction. Secured debt does not face such an incentive problem, because creditors will earn the secured portion of debt regardless of the behavior of the debtor. Consequently, collateralization of the loan can mitigate the agency problems in international lending. By mitigating these agency problems, collateral should improve the creditworthiness of debtors and therefore their lending terms. If this change in the terms faced by borrowers is sufficient, borrowers may find it in their interest to issue collateral even though doing so entails real resource costs.

For a developing nation like the Philippines, it is apparent that providing collateral is costly. As a rapidly developing nation, it is likely that the internal rate of return on capital within the Philippines exceeds the world rate of interest. Holding collateral offshore reduces the earnings on these assets relative to what they could earn inside the country. Since collateral consumes real resources, then, it follows that real gains, such as the mitigation of agency problems, are required to motivate their issue.

The following example illustrates how real effects of the swap might arise. Consider the case of two types of borrowers, low-risk and high-risk. They differ in that the high-risk borrower’s government considers default on its sovereign risk obligations less costly than a low-risk borrower’s government. Therefore, the high-risk borrower faces more severe agency problems than the low-risk borrower does. If the agency problems faced by the high-risk borrowers are sufficiently severe, they will be able to improve their terms of debt service by issuing collateralized debt. This implies that the benefits of collateralization from addressing the agency problem between the high-risk borrower and its creditors outweigh the costs.

On the other hand, if the agency problem faced by the low-risk borrower is sufficiently small, the cost of collateralized debt will outweigh its advantages. Since the low-risk borrower considers the costs of default to be higher, he has less of an incentive to underinvest, which increases his probability of default. This mitigates the agency problem associated with sovereign debt and therefore limits the potential gains from collateralization. Therefore, with sufficiently high perceived costs of default, the low-risk borrowers do better by issuing non-collateralized debt.

The outcome where one type of agent behaves differently from another, as is the case here with the low-risk borrower preferring non-collateralized debt and a high-risk borrower preferring collateralized debt, is termed a “separating equilibrium,” because in equilibrium we can distinguish between borrower types by observing their actions. Under these circumstances, a borrower whose type changed from high-risk to low-risk would have a real incentive to convert his outstanding debt claims from secured to non-secured debt. This is precisely what is occurring in the Philippine debt repurchase.

Good news from the Philippines

In light of this discussion, consider the events immediately preceding the decision by the Philippines government to repurchase its Brady debt. Just a few days before the Brady repurchase announcement, Moody’s raised the Philippines’ credit rating to Ba2, stating that the change reflected the country’s improving credit fundamentals, including broad market-oriented reforms that have placed the nation on a higher growth trajectory. These changes include a liberalization of the exchange rate regime, the elimination of non-tariff barriers, the removal of various subsidies and price controls, and the deregulation of investment. Two days later, the Philippines government reached an agreement with the Moro National Liberation Front to end the 25-year religious conflict.

The close timing of the good news about its investment prospects and its decision to repurchase its Brady debt is unlikely to be coincidental. A bank spokesman was quoted as saying that the repurchases reflected the country’s improved economic fundamentals and that credit enhancements such as collateral, were now less important in the eyes of investors (Luce, et al. 1996).

The Philippines rejoins the “low-risk” borrower pool

The recent good news concerning the Philippines appears to have improved the perceptions of international investors concerning its creditworthiness. The Philippines may once more be a nation that can issue non-collateralized debt in international capital markets at favorable terms. As in the illustration of low-risk and high-risk borrowers, a borrower reclassified by the market as low-risk would rationally convert its collateralized debt to non-collateralized debt, as the Philippines is doing through its repurchase of its collateralized Brady debt. It follows that there will be real effects of switching to direct issues of non-collateralized debt.

Note that this does not mean that the Brady bonds were undesirable instruments at the time that they were used in the debt restructurings. At that time, the Philippines was classified by the market as a high-risk country for foreign investment. The collateralized instruments were then the superior ones to issue, in the sense that the terms available to the Philippines at the time on non-collateralized debt may have been undesirable.

Twilight of the Brady market?

The Philippines is the second nation to undertake such a repurchase. Mexico took similar action earlier this year. In addition, Brazil, Argentina, and Venezuela are planning their own repurchases. Mexico is planning an additional repurchase of Brady bonds shortly. The popular press has suggested that the end of the era where the Brady bond market is the most liquid and widely traded emerging-nation debt market may be at hand: “Emerging markets want to get rid of the stigma of being linked to U.S. Treasuries. There’s every chance that the Brady market will cease in a few years from now to be the primary vehicle for emerging market debt” (Luce and Middelmann 1996).

However, our analysis suggests that the gains from retiring securitized debt depend on the change in the perception of the international financial markets concerning a borrower’s creditworthiness. The Philippines’ move towards retiring their securitized debt may be defensible in light of the positive developments in their economy since the time at which the securitized debt was issued. However, the Philippines’ original issue of collateralized debt may have been the best strategy if the Philippines was at that time a sufficiently high-risk borrower. Similarly, it is not apparent that all Brady nations will gain through such a repurchase program. The gains will depend on how much each country’s creditworthiness has improved in the eyes of the international market. The decision to conduct such a program should then be made on a case by case basis. Consequently, it is not clear that a rush to convert Brady bonds is imminent, or that the Brady market will soon shrink in size dramatically.

Mark M. Spiegel
Senior Economist


Luce, Edward, and Conner Middelmann. 1996. “Philippines Plan Marks the Beginning of End for Bradys.” Financial Times (September 3).

___________, ________, and Richard Lapper. 1996. “Philippines Plans $2 bn Bond Issue to Reduce Brady Debt.” Financial Times (September 3)

Modigliani, Franco, and Merton H. Miller. 1958. “The Cost of Capital, Corporation Finance, and the Theory of Investment.” American Economic Review pp. 261-297.

Stulz, Rene M., and Herb Johnson. 1985. “An Analysis of Secured Debt.” Journal of Financial Economics 14, pp. 501-521.

Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.

Please send editorial comments and requests for reprint permission to
Research Library
Attn: Research publications, MS 1140
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120