Corporate credit limits for fixed income portfolios

01 Oct 2014

Fixed income portfolio managers and risk managers constantly grapple with the question of how to size their corporate credit trades. Their task is made more difficult by the fact that corporate credit events are rare, particularly among Investment Grade bonds, and that tail risk is not well captured by most multifactor risk models. In this article, we propose a simple, but effective, method for sizing credit trades based on their spread. In particular, we model the cross-sectional behavior of corporate spreads, estimate the expected shortfall of monthly spread returns, and use our results, along with some observations on the duration of corporate bonds, to derive a simple upper bound on the permissible exposure to any single issuer in a credit portfolio.

The method has been applied successfully to Investment Grade and HighYield fixed income portfolios in both developed and emerging markets, and has proven its worth in daily use by protecting portfolios against disproportionate idiosyncratic losses, while allowing portfolio managers sufficient flexibility to express their investment views with clarity. Its use is not confined to limits on issuers—our method is easily extended to create limits on a portfolio’s exposures to individual industries, sectors, countries, and regions.

Introduction and overview

Fixed income portfolio managers and their risk managers constantly grapple with the problem of setting appropriate issuer limits for corporate credit trades. However, in spite of its fundamental importance in the management of credit portfolios, most approaches to setting issuer limits are surprisingly crude. More often than not, issuer limits are based simply on credit ratings, with a maximum exposure being defined for each rating class, e.g., AAA: 5%, AA: 4%, A: 2%, and BBB: 1%. But credit ratings can be inaccurate, and can also be slow to reflect changes in credit quality, as exemplified by the AAA ratings accorded Enron prior to 2001, WorldCom prior to 2007, and most European sovereigns prior to 2010.

In this article, we describe an alternative approach to setting corporate credit limits that is based on issuer spreads. It is rooted in the widely observed fact that many measures of spread risk, such as spread volatility and the distribution of spread changes, are proportional to spread levels. This observation can be combined with realistic assumptions about the duration of corporate bonds, achievable information ratios, and tolerable levels of loss to derive a useful upper bound on the exposure to a single issuer.

The remainder of this article is organized as follows. We first survey the literature on the empirical behavior of corporate spreads, and present an independent body of evidence that supports the proportional spread change models of Tauren (1999), Ben Dor et al. (2007), and Benzschawel and Lee (2011).We then use these results to guide the development of a position sizing methodology and a simple formula for determining the maximum exposure to a single issuer. Next, armed with some assumptions about the duration of corporate bonds and the information ratios of actively managed portfolios, we select appropriate values for certain parameters in the formula. Finally, we summarize our results and present some numerical examples that illustrate the use of our formula in setting credit limits in both Investment Grade and High Yield portfolios.

Download to read more