Market conditions, default risk and credit spreads

https://doi.org/10.1016/j.jbankfin.2009.05.018Get rights and content

Abstract

This study empirically examines the impact of the interaction between market and default risk on corporate credit spreads. Using credit default swap (CDS) spreads, we find that average credit spreads decrease in GDP growth rate, but increase in GDP growth volatility and jump risk in the equity market. At the market level, investor sentiment is the most important determinant of credit spreads. At the firm level, credit spreads generally rise with cash flow volatility and beta, with the effect of cash flow beta varying with market conditions. We identify implied volatility as the most significant determinant of default risk among firm-level characteristics. Overall, a major portion of individual credit spreads is accounted for by firm-level determinants of default risk, while macroeconomic variables are directly responsible for a lesser portion.

Introduction

Credit risk and market risk are inherently linked. It has been documented that default probabilities and recovery rates vary through business cycles.1 It is also well known that interest rates and corporate bond yield spreads fluctuate over business cycles, as aggregate and firm-level outputs critically depend on the state of the economy. For instance, Fama and French (1989) find that corporate bond yields rise when economic conditions are weak. However, these empirical findings have not been fully understood in a structural framework. In fact, traditional structural models based on the seminal Merton (1974) model have generally ignored the interaction between market risk and credit risk. Consequently, they have failed to match the levels of observed credits spreads (“the credit spread puzzle”).

This paper examines the intrinsic link between market risk and credit risk inspired by recently developed structural models that directly explore the impact of market risk on credit spreads.2 Specifically, we use individual firms’ credit default swap (CDS) spreads to investigate new empirical implications from these structural models. For instance, in addition to the previously documented negative correlation between GDP growth rate and credit spreads, we show that credit spreads also increase in growth volatility as implied by these models. We further demonstrate that credit spreads decrease with a sentiment measure based on the Conference Board Consumer Confidence Index. Because consumer/investor sentiment is usually negatively correlated with the market-wide risk aversion and uncertainty about future economic growth, this result is consistent with the notion that credit spreads depend on investors’ risk attitude and uncertainty about future economic prospects, as predicted by the models.

A number of existing empirical studies use yield spreads of corporate bond indices or average yield spreads within a particular rating class to characterize the dynamics of credit spreads (see, e.g., Huang and Huang, 2003). This approach may obscure the importance of firm heterogeneity and lead to underestimation of expected losses, as pointed out in Hanson et al. (2008). With aggregate credit spreads, macro variables tend to explain a big portion of their variations over time. However, when we re-examine these relations in a panel regression, we find that firm characteristics traditionally determining default risk account for the bulk of the explanatory power, while a monthly time dummy that captures the time series variation in credit spreads accounts for just about 6% of the overall variation. Much of that explanatory power stems from the macro variables implied by the structural models, such as the sentiment indicator.

Recent models also provide additional cross-sectional predictions. We confirm that, across firms, credit spreads decrease with firm-specific growth rate of cash flows and increase with cash flow volatility, as predicted. More interestingly, we detect an important and time-varying role of cash flow beta, which measures the covariation of the firm-level cash flow with the aggregate output. In particular, the evidence suggests that during economic expansions, a high cash flow beta helps reduce credit spreads, while during economic recessions, a high cash flow beta may increase credit spreads. This pattern highlights the effect of the interaction between market risk and credit risk on the dynamics of credit spreads.

Jarrow and Turnbull (2000) suggest that incorporating macroeconomic variables may improve a reduced-form model of credit spreads. Duffie et al. (2007) use macroeconomic variables, such as industrial production growth, to help better predict corporate defaults. Our study represents an effort in a systematic investigation of the impact of market conditions on firm-level credit spreads in the structural framework provided by the recent theoretical models mentioned above. It also bridges the two strands of literature on credit risk that tend to focus separately on the macro and micro determinants and hence allows us to assess the relative explanatory power of macro and micro variables for firm-level credit spreads and examine the interaction between market conditions and firm characteristics.

The rest of this paper is organized as follows: Section 2 discusses the empirical implications of the recently developed models that incorporate market conditions into defaultable bond pricing. Section 3 introduces the CDS data used for the empirical analysis. Sections 4 Macroeconomic conditions and credit spreads, 5 Firm characteristics and credit spreads present results of the time-series and cross-sectional patterns of credit spreads based on the model implications, respectively. Section 6 concludes.

Section snippets

Empirical implications of recent theories

The recent literature has seen a number of theoretical papers attempting to understand the link between credit spreads and macroeconomic risk. For example, Tang and Yan (2006) investigate the dynamics of firm-level credit spreads by highlighting the role of a firm’s cash flow beta that measures its exposure to macroeconomic risk. They show that incorporating a macroeconomic influence on a firm’s cash flow process helps improve significantly the fit of default probabilities and credit spreads.

Data and sample description

Several data issues make empirical analysis of credit risk difficult. Corporate bond yields are known to contain substantial liquidity and tax premia due to illiquidity of the corporate bond market and different tax treatments between corporate bonds and Treasury bonds. Many corporate bonds also have embedded options, further complicating the measurement of credit spreads based on corporate bond yields. To make the matter worse, there is a debate about an adequate reference for the risk-free

Macroeconomic conditions and credit spreads

In this section, we empirically test several predictions from the structural credit risk models conditioned on macroeconomic variables. Our analysis employs the CDS data for credit spreads and Moody’s KMV EDF data for the default probability measure. Hence, compared to existing empirical studies, our examination has two distinct advantages: high quality firm-level data of credit spreads and default risk, and a theory-motivated hypothesis.

Firm characteristics and credit spreads

Our analysis thus far suggests the importance of firm heterogeneity. There have been many studies that document the role of firm-level characteristics, such as leverage ratio, profitability, and stock volatility, in determining firm-level default probabilities and credit spreads. Theoretically, Tang and Yan (2006) make additional predictions on the effect of firm’s cash flow characteristics on credit spreads and on the interaction between market conditions and credit risk. In this section, we

Conclusion

We empirically examine the effect of market conditions on credit spreads, motivated by the recent structural models that explicitly consider the joint effect of market risk and credit risk. Our study uses large scale firm-level CDS data that allow us to investigate the relative explanatory power of macroeconomic conditions and firm characteristics and the effect of their interactions. We show that the macroeconomic condition accounts for about 6% of the overall variation of credit spreads (11%

Acknowledgements

We are grateful to Til Schuermann (discussant and guest editor) and other participants of the Joint Conference of the Deutsche Bundesbank, the Basel Committee on Banking Supervision and the Journal of Banking and Finance on the Interaction of Market and Credit Risk for useful insights. Two anonymous referees provided detailed suggestions that have helped improve the paper. We thank Lorenzo Garlappi, Shisheng Qu, Sheridan Titman, Efstathios Tompaidis, Sergey Tsyplakov and participants of the

References (44)

  • R.A. Jarrow et al.

    The intersection of market and credit risk

    Journal of Banking and Finance

    (2000)
  • R.A. Korajczyk et al.

    Capital structure choice: Macroeconomic conditions and financial constraints

    Journal of Financial Economics

    (2003)
  • B.A. Minton et al.

    The impact of cash flow volatility on discretionary investment and the costs of debt and equity financing

    Journal of Financial Economics

    (1999)
  • M. Baker et al.

    Investor sentiment and the cross section of stock returns

    Journal of Finance

    (2006)
  • Bhamra, H.S., Kuehn, L., Strebulaev, I.A., 2007. The levered equity risk premium and credit spreads: A unified...
  • R. Blanco et al.

    An empirical analysis of the dynamic relationship between investment grade bonds and credit default swaps

    Journal of Finance

    (2005)
  • R.R. Bliss et al.

    Option-implied risk aversion estimates

    Journal of Finance

    (2004)
  • M. Bruche et al.

    Recovery rates, default probabilities, and the credit cycle

    Journal of Banking and Finance

    (2010)
  • J.Y. Campbell et al.

    Equity volatility and corporate bond yields

    Journal of Finance

    (2003)
  • J.Y. Campbell et al.

    Bad beta, good beta

    American Economic Review

    (2004)
  • Chen, H., 2007. Macroeconomic conditions and the puzzles of credit spreads and capital structure. Working paper,...
  • Chen, L., Collin-Dufresne, P., Goldstein, R.S., forthcoming. On the relation between the credit spread puzzle and the...
  • Cited by (220)

    • CEO overconfidence and the choice of debt issuance

      2024, Journal of Banking and Finance
    • Temporal networks and financial contagion

      2024, Journal of Financial Stability
    View all citing articles on Scopus
    View full text