After discussing these two non-debt-weighted approaches, below we focus on the risk exposures of corporate bond indices. Our focus will mainly be on the standard debt-weighted indices when discussing risks, as these indices have been studied more widely than their equal-weighted and fundamental-weighted counterparts. In particular, we will discuss the reliability of credit risk exposure, the stability of duration over time and liquidity risks.
MANAGING CREDIT RISK
The issuer’s willingness and ability to pay the debt and stick to the terms of the obligation is a key factor for fixed income investors. As a consequence, default risk is one of the first considerations in corporate bond indices and should be assessed and reflected in a reliable way. The typical credit risk classification used by bond index providers is straightforward. Index firms separate the bond universe into two parts: investment grade and high-yield corporate bond indices and then use credit rating agencies’ bond ratings to specify credit risk.
Using the ratings scales of the main three agencies: (Moody’s, S&P and Fitch7), the threshold between investment grade and high yield is set at Baa3/BBB-/BBB-. Index providers differ, however, on whether they include or exclude a bond when at the threshold rating, or how they handle a bond with a “split” rating from different agencies.8 Beyond the simple distinction between high yield and investment grade, it is rare to find corporate bond indices that focus on specific ratings categories. Thus investors will not easily find passive investment products which allow a specific sub-segment of the credit market to be captured.
The agency credit ratings underlying all these indices are based on accounting and fundamental data. Ratings are often narrow in focus (an accurate assessment of companies’ pension liabilities, for instance, is often difficult to achieve). Moody’s, S&P and Fitch describe their ratings as a long-term assessment of credit risk through the economic cycle (Altman and Rijken 2004), resulting in an approach that tends to be both stable and backward-looking.
Given this backward-looking approach and the bias towards ratings stability, it is perhaps not surprising that there is substantial empirical evidence that ratings lag the credit spreads observed in the market for corporate bonds. It is not surprising that market-based measures lead ratings changes. Both bond credit spreads and credit default swap (CDS) spreads for corporate issuers are market-based measures for assessing credit risk. There is a debate in academic literature over the information content of each of these measures. It is clear that both measures can be impacted by other effects not related to the pure default risk of the issuer, such as liquidity, supply and demand in the market of the respective instrument and other risk factors (Huang and Huang (2003), Colin-Dufresne, Goldstein, and Martin (2001), Longstaff, Mithal, and Neis (2005), Bao and Pan (2008) (2011), Aunon-Nerin et al. (2002) and Tang and Yan (2007)). However, such market-based measures have been found to be clearly more informative than ratings (Hull and White (2004), Blanco et al. (2005)). Given that investors ought to be concerned more about the current credit risk, rather than the past credit risk of constituents, it’s surprising they continue to rely on indices that make heavy use of agency credit ratings.