نوسانات غیر متعارف در برابر نقدینگی؟ شواهدی از بازار اوراق قرضه شرکتی ایالات متحده

Our objective in this paper is to determine empirically the extent to which fixed-income investors are concerned about the relative effects of equity volatility and bond liquidity in the cross-section of corporate bond spreads. Our tests reveal that while both volatility and liquidity effects are significant, volatility, representing ex-ante credit shock, has the first-order impact, and liquidity represented by bond characteristics and price impact measure has the secondary impact on bond spreads. Conditional analysis further reveals that distressed bonds and distress regimes are both associated with significantly higher impact of volatility and liquidity shocks. However, the relative impact of these effects varies conditional on the underlying bond attributes and overall market conditions.

Our objective in this paper is to study the relative impact of idiosyncratic volatility and liquidity on corporate yield spreads (i.e., excess of bond yields over equal maturity benchmark yields) cross-sectionally, and empirically disentangle both the effects. Overall we find that both volatility and liquidity matter separately for the cross-section of bond yields, with the relative importance of volatility and liquidity changing with firm-specific and economic conditions. Idiosyncratic equity volatility refers to the firm-specific risk after controlling for systematic market risk factors, and reflects the residual stock volatility of a firm. In their influential paper, Campbell and Taksler (2003) show that idiosyncratic equity volatility has a significant role in explaining corporate bond spreads even after conditioning for ratings, bond and firm-specific characteristics and macro-economic variables. 2, 3 While idiosyncratic equity volatility captures ex-ante default risk, other studies reveal that credit risk determinants alone cannot adequately explain the levels or changes in the corporate bond spreads (e.g., Collin-Dufresne et al., 2001 and Huang and Huang, 2003). If non-default sources of risk such as illiquidity matter in bond spreads, then by ignoring liquidity, structural models can overprice bonds, resulting in the so-called “credit puzzle” ( Covitz and Downing, 2007 and Driessen, 2005). 4 While corporate debt constitutes a significant proportion of capital structure of firms, the underlying bond market remains highly illiquid.5 Extant work documents the significant effect of illiquidity on bond prices (e.g., Longstaff et al., 2005 and Driessen, 2005; and Chen et al., 2007).6 If individual bond liquidity matters, and is not controlled for, then volatility effects on bond spreads as documented in Campbell and Taksler may be difficult to interpret.7 While higher idiosyncratic equity volatility can imply higher ex-ante credit risk and bond spreads, it is not obvious from the Campbell and Taksler results whether higher spreads are attributable to higher equity volatility, lower bond liquidity, or both. In this paper, we extend Campbell and Taksler (2003) by conditioning for underlying bond liquidity, and exploring the relative contribution of equity volatility and bond liquidity in the cross-sectional pricing of corporate bond spreads. We tease out the volatility impact from the liquidity effects, and examine whether idiosyncratic risk subsumes the information in liquidity in explaining corporate bond prices. We explore the roles of volatility and liquidity on cross-sectional bond prices unconditionally, as well as conditional on several underlying distress features. High-distress issues are defined as bonds with low ratings, low liquidity, or high underlying equity volatility. High-distress periods refer to low-growth or recessionary periods, and periods of high aggregate equity market volatility or low aggregate bond market liquidity in the economy. We employ over 195,000 secondary trades of option-free corporate bonds issued by 818 firms over an 11-year period, 1994–2004. We measure idiosyncratic equity volatility as the variance of multifactor risk-adjusted residual returns, and quantify bond liquidity in terms of price impact of bond trades as well as several underlying bond characteristics.8 Our work is unique in that it focuses on a large sample of corporate bonds over an extended period, uses an exhaustive list of volatility and liquidity variables, and provides a comprehensive study of the volatility and liquidity effects in cross-sectional bond pricing.9 We document two principal findings that remain unaltered under a battery of robustness tests. First, both equity volatility and bond liquidity effects are jointly important, but in terms of relative magnitude, equity volatility has the first-order impact, and bond liquidity has the secondary impact on cross-sectional pricing of bond spreads. Second, conditional cross-sectional tests reveal that, on an absolute basis, distressed bonds and distress regimes experience higher impacts of shocks to volatility as well as liquidity. Further, on a relative basis, volatility effects are dominant for distressed bonds and during high-distress regimes, while liquidity matters more for low-distress bonds and during low-distress regimes. Our study extends the Campbell and Taksler study in many ways. Ignoring firm-specific bond liquidity can lead to model misspecification and, therefore, the Campbell and Taksler framework can imply biases in the ex-ante spreads required by bond investors. Moreover, debt issuers need to evaluate volatility as well as liquidity effects while pricing and timing their bond issues. Not including individual liquidity effects, as in the Campbell and Taksler framework, can bias the pricing and timing decisions of bond issuers. The Campbell and Taksler study is limited to the 1995–1999 period, when the market experienced very high growth induced by the high-tech bubble. By extending the sample to 2004 and including the very distinct post-bubble slowdown period, we hope to uncover the determinants of corporate bond spreads, and the differential impact of volatility and liquidity under different distress regimes. Additionally, our paper extends the extant literature that explores the role of either equity volatility or bond liquidity on corporate bond spreads; while the earlier papers exclusively focus on either of these variables, we emphasize joint focus on both variables in bond pricing and evaluate their relative significance. 10 Taken together, our results imply that the idiosyncratic volatility effect does not subsume the liquidity effect in explaining bond prices, unlike the findings in equity markets ( Spiegel and Wang, 2005). While volatility has a significant effect, liquidity still matters for the cross-sectional bond yields, and the strength of effects change conditional on the underlying portfolio, and overall market conditions. Our findings imply that the Campbell and Taksler framework, and more generally Merton (1974), can be augmented to include unconditional and conditional effects of both volatility and liquidity. Moreover, we emphasize that, since we do not explicitly measure exogenous shocks to either volatility or liquidity, our results are best interpreted as stylized facts, rather than any causal evidence. Our work is further related to several recent papers that focus on disentangling credit and liquidity risks from yield spreads (e.g., Longstaff et al., 2005, Driessen, 2005, Covitz and Downing, 2007 and Beber et al., 2009; and Schwartz, 2010).11 Our work also represents an extension to bond markets of current work that examines the overlap in information content of volatility and liquidity risks in equity returns (e.g., Spiegel and Wang, 2005 and Bali et al., 2005; and Boehme et al., 2006).12 The rest of the paper is structured as follows. Section 2 describes the data and variables. Section 3 presents univariate and bivariate portfolio results. Sections 4 and 5 report the results from unconditional and conditional Fama–MacBeth regressions for bond spreads. Section 6 concludes.

The primary objective of this paper is to explore the relative importance of equity volatility and bond liquidity in the Campbell and Taksler (2003) framework. We extend Campbell and Taksler by (a) conditioning for underlying bond liquidity, (b) evaluating the interaction between idiosyncratic equity volatility and bond liquidity in determining corporate bond spreads, and (c) exploring how such interactions are altered when underlying bond and firm characteristics vary, or when market conditions change. This paper contributes to the literature by providing a better understanding of the relative importance of equity volatility and bond liquidity in the cross-sectional pricing of corporate debt. Our analysis shows that the cross-section of corporate bond spreads depends on both volatility and liquidity, with their relative importance depending on underlying firm and economic conditions. Unconditional Fama–MacBeth regressions reveal that while both volatility and liquidity effects are important in evaluating bond spreads, equity volatility has the first-order impact, and bond liquidity, represented by bond characteristics and bond price impact measures, has the second-order effect on bond spreads. Conditional cross-sectional regressions implemented on sub-samples based on relative distress features, however, reveal very distinct absolute and relative significance of the two effects. Our findings imply that bond pricing models can deliver improved pricing and hedging performance by incorporating the relative significance of equity volatility and bond liquidity effects for different distress portfolios and regimes. Our results also indicate a need for regime-switching models for bond spreads that can better incorporate the differential effects of volatility and liquidity across distress regimes (e.g., Watanabe and Watanabe, 2008; and Acharya et al., 2010). Further, our findings reveal that, unlike the results for equity markets documented by Spiegel and Wang (2005), idiosyncratic risk does not subsume bond liquidity in explaining bond spreads. This implies that corporate bond markets are inherently more illiquid compared to equity markets, thereby rendering it much harder to diversify illiquidity. Finally, our findings on the relative strengths of volatility and liquidity effects can have several practical implications. They can, for example, guide (a) fixed-income traders to better formulate their investment and hedging strategies, (b) debt issuers to better time their debt issuance in order to minimize the cost of borrowing, and (c) policy makers to better address the impact of volatility and liquidity shocks on credit markets.