Tags: Affine Models, Credit Default Swaps, Macroeconomic Variables, Credit Risk, Liquidity, More from: Biao Guo, Songtao Wang

It is widely known that credit risk alone cannot explain corporate yield spreads. For instance, Collin-Dufresne, Goldstein, and Martin (2001) and Huang and Huang (2012) show that credit risk only accounts for a small fraction of the observed yield spreads on investment-grade bonds and a larger fraction for speculative-grade bonds. A large number of papers have been dedicated to studying this issue and find that the nondefault component of yield spreads is mainly related to the illiquidity of corporate bonds (see, e.g., Longstaff, Mithal, and Neis (2005), Chen, Lesmond, and Wei (2007), Martell (2008)).

While previous studies have examined the dynamics of corporate bond credit spreads across maturities, little is known about corporate bond liquidity spreads. What is the term structure of corporate bond liquidity spreads? What determines variations in this term structure? According to Basel III, corporate bonds are included in the Level 2 assets whose liquidity needs to be monitored quantitatively1. Thus, understanding these issues is of fundamental importance from a risk management perspective. Also important for risk management is to understand the co-movement between credit spreads and liquidity spreads of corporate bonds. Ericsson and Renault (2006) and Buhler and Trapp (2009) find that credit spreads and liquidity spreads are positively correlated, however, they do not provide an explanation for this stylized fact. So in this paper we try to address the following questions: 1) are liquidity spreads significant for corporate bonds? 2) how do macro factors affect corporate bond credit and liquidity spreads quantitatively? and 3) what might determine the joint movement of the term structures of Treasury yields, credit spreads and liquidity spreads?

This paper studies these and other related issues by investigating in a unified framework the impacts of macroeconomic and financial factors on Treasury yields, corporate credit spreads, corporate liquidity spreads, and their term structures. Specifically, building on Ang and Piazzesi (2003), Wu and Zhang (2008), and others, we develop a no-arbitrage term structure model to link four economic factors representing monetary condition, macroeconomic fundamentals, and financial market volatility to the pricing of Treasury bonds, Credit Default Swaps (CDS), and corporate bonds by assuming that the instantaneous Treasury yield and the instantaneous corporate bond credit and liquidity spreads are affine in the four economic factors2. Compared to the regression approach, the no-arbitrage term structure model has some advantages. First, it generates more interpretable and more stable results than those from regressions. Second, it allows us to make an internally consistent analysis of the impacts of macroeconomic and financial factors on Treasury yields, corporate bond credit and liquidity spreads across the whole spectrum of maturities. Moreover, the estimation results from the no-arbitrage term structure model not only show how the factors affect the term structure but also reveal the reasons behind it. These advantages serve in part as our motivation for using the no-arbitrage framework to study the macroeconomic determinants of the credit and liquidity spreads.

The empirical analysis first focuses on the sample period from July 2002 to November 2007. We extract four economic factors of monetary condition, inflation, real output, and financial market volatility from a set of 12 macroeconomic and financial data series using the Principal Component Analysis (PCA) approach. With the extracted economic factors, we estimate the parameters for Treasury bonds, and using these estimated parameters as given, we then estimate the parameters for CDS and corporate bonds for each rating class.

Many of this paper’s findings are new while others reinforce and clarify previous results. This paper is the first empirical work that studies in a no-arbitrage framework the determinants of the term structure of corporate bond liquidity spreads. Goyenko, Subrahmanyam, and Ukhov (2011) study the joint time-series of illiquidity for different maturities and the determinants of on-the-run and off-the-run illiquidity, but their analysis focuses on Treasury bonds and is based on the regression approach. Ericsson and Renault (2006) study the term structure of corporate bond liquidity spreads without investigating the determinants of the term structure. Different from many previous studies such as Longstaff (2004) and Acharya, Amihud, and Bharath (2013), this paper finds ”flight-to-liquidity” also in non-stress times. Furthermore, this paper provides economic explanations for the negative correlation between the risk free interest rates and corporate yield spreads and the positive correlation between the credit and liquidity components of corporate yield spreads.

The complete paper is available at:
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3306027.