Publications

Co-author: Pierre Collin-Dufresne
 The limit-of-arbitrage story for the difference between CDS and cash-bond implied credit spread.

​Co-authors: Turan Bali and Quan Wen
We investigate the cross-sectional determinants of corporate bonds and find that downside risk is the strongest predictor of future bond returns. We also introduce common risk factors based on the prevalent risk characteristics of corporate bonds -- downside risk, credit risk, and liquidity risk -- and find that these novel bond market factors have economically and statistically significant risk premia, which cannot be explained by the long-established stock and bond market factors. We further show that these newly proposed risk factors outperform all other models considered in the literature in explaining the returns of the industry-sorted and size/maturity-sorted portfolios of corporate bonds.

Co-authors: Robert Goldstein and Fan Yang
A model specifies asset dynamics instead of equity dynamics alone can explain the large spread in prices between put options written on individual banks and options written on the bank index during the financial crisis, even without the government bail-out story. It's important to look at equity and debt together!

​Co-authors: Arvind Krishnamurthy and Charles-Henri Weymuller
We propose a novel measure, "Liquidity Mismatch Index (LMI)," to gauge the mismatch between the market liquidity of assets and the funding liquidity of liabilities for the banking sector. LMI can be used for macro-prudential liquidity stress test, and for predicting cross-sectional bank risk. The outperformance of LMI than Basel III's LCR, NSFR results from the time-varying liquidity sensitivity weights which are driven by market prices.
​ ​​Co-authors: Thomas Philippon and Alexi Savov.      Media: -Bloomberg, New York Times 
​The finance industry has grown, financial markets have become more liquid, information technology has been revolutionized. But have financial market prices become more informative? 

Co-author: Liuren Wu
This paper examines the capability of firm fundamentals in explaining the cross-sectional variation of CDS spreads. 

Co-authors: Pierre Collin-Dufresne, Robert Goldstein, and Jean Helwege 
Reduced form models of default that attribute a large fraction of credit spreads as compensation for credit event risk typically preclude the most plausible economic justification for such risk to be priced, namely, a ``contagious" response of the market portfolio during the credit event. 
​ ​Co-author: Shang-Jin Wei
Strong property rights institutions tend to weaken the sovereign transfer risk, whereas contracting institutions (protection of creditor rights or minority shareholder rights) do not matter.

Co-authors: Eric Ghysels and Jonathan Wright
​We examine the relationship between MIDAS and Kalman filter state space models applied to mixed frequency data. 

Working Papers

We revisit Feldhutter and Schaefer (FS, 2018) who report evidence of a ``credit spread puzzle'' for high-yield but not investment-grade bonds. We show their results are reversed when their model is calibrated using market values of debt rather than book values. More importantly, we demonstrate that calibrating structural models of default to credit spreads rather than to historical default rates provides the statistical power necessary to reject the FS assumption that asset value dynamics follows geometric Brownian motion. A large market price of risk on jumps is required to match historical default rates, thus reaffirming a ``credit spread puzzle'' for investment-grade but not high-yield bonds.
​ Chengtou bond is the soli asset with market prices that can capture the funding cost of Chinese local government debt. In contrast to the U.S. municipal bonds, Chengtou bonds are issued by private corporations but implicitly guaranteed by the local hence central governments, which are reflected by novel risk characteristics---real estate GDP and political risk. One standard deviation increase in local real estate GDP (political risk) corresponds to 10 (9) basis points decrease (increase) in bond yields, respectively. However, conditional on political risk, real estate GDP actually increases bond yields, suggesting that only local governments with low political risk can enjoy the low funding costs driven by high real estate growth. ​
​ We identify the unique role of the government bond lending market in accessing safe assets during periods of market stress. Using a novel database, we provide evidence that safe assets in the lending market have higher demand, higher borrowing cost, and higher usage of non-cash collateral relative to non-safe assets during stressed market conditions. Moreover, we find that market participants are able to obtain safe assets using relatively low-quality non-cash collateral, allowing for collateral transformation. These attributes are important since they increase the velocity of safe assets, and hence alleviate the pressure of safe asset shortages. We show that policy interventions by central banks can help reduce safe asset shortages by returning sought-after safe assets to the market.  
​ Despite common wisdom that equities and bonds are segmented, the organization structure of fund families can offset frictions regarding asset market segmentation. We find that equity funds and corporate bond funds linked within a mutual fund family (sister funds) exhibit a significant co-movement in holdings of commonly-held firms' equities and bonds. In contrast, we do not find such a pattern for funds in different families. We show that the holding co-movement is driven by information sharing among sister funds, and such funds make more profit-enhancing investment decisions on common holdings, compared to stand-alone funds. Our findings suggest that collaboration between equity funds and bond funds improves fund performance.

The security lending market allows institutional investors, such as insurance companies, to lend out holding assets in exchange for cash collateral, an important but understudied source of funding for corporate bond lenders. Since these lenders are also primary investors of corporate bonds, we hypothesize that their preference on certain types of bonds originated from the security lending market can influence corporate financing policies, a demand-driven perspective. Using a novel data on corporate bond lending, we show that lenders’ preference on long-term bonds stimulates firms to issue more such bonds and helps boost bond price. Analysis exploiting a quasi-experiment on the regulation of insurance companies in 2010 supports the causal interpretation. Our results shed new insight on corporate financing policies and bond pricing.​
​      (the Arthur Warga Award for the Best Paper in Fixed Income, SFA Cavalcade Asia Pacific)

​We study whether and how the efficiency in the equity market affects the credit market. We exploit a unique experiment (“Regulation SHO”) which, by exogenously reducing transaction costs for short-sellers, improves efficiency in the equity market through increasing the informativeness of the firm. We show that higher efficiency significantly improves the buy-side investor-paid ratings but have little impact on sell-side issuer-paid ratings, because the change of ratings comes not from a lower probability of distress per se, but from a reduction in the uncertainty about the riskiness of the firm and from an enhancement of governance quality. The effect is stronger for firms rated below investment grades, for firms with low analyst coverage, and for firms with low institutional ownership. 

  • What Bond Lending Reveals? The Role of Informed Demand in Predicting Credit Spread Changes, August 2018
Corporate bond lending market is 2~3 times larger than corporate bond trading market, which makes it contain unique information on top of bond trading market. Using the lending information on both prices and quantities, we construct a proxy of informed demand and study its role in predicting the credit spread changes. We find that informed demand exhibits highly significant predictive power on future credit spread changes, even controlling for credit risk, liquidity risk, and other long-established risk factors in Collin-Dufresne et al. (2001) and Zhang, Zhou, and Zhu (2009). We also nd that lending-
based informed demand only have predictive power in corporate bonds, but not in its derivative contract CDS. Therefore, the informed demand help explain the CDS-Bond basis, especially in normal times when limits-to-arbitrage constraints unbind.

We find a significantly positive (negative) relation between volatility (skewness) and the cross-section of corporate bond returns, whereas kurtosis does not make a robust incremental contribution to the predictability of future bond returns. The risk-adjusted return spreads in volatility (skewness) sorted portfolios represent compensation for differences in the level of idiosyncratic volatility (idiosyncratic skewness) and differential exposure to the common idiosyncratic volatility factor in the bond market. We also show that the new risk factors based on the distributional moments of bond returns represent an important source of common return variation missing from the long-established stock and bond market factors.​

Although China now has one of the largest government bond markets in the world, the market has received relatively little attention and analysis. We describe the history and structure of the market and assess its functioning. We find that trading in individual bonds was historically sparse but has increased markedly in recent years. We find also that certain announcements of macroeconomic news, such as China’s producer price index (PPI) and manufacturing purchasing managers’ index (PMI), have significant effects on yields, even when such yields are measured at a daily level. Despite the increased activity in the market, we are able to reject the null hypothesis of market efficiency under two different tests for four of the most actively traded bonds.

          Media: -Asia Week Guide article, The Art Newspaper, 
We analyze two reasons for export prices to be different across markets, namely quality differentiation and variable markups, and attempt to parse their relative importance and some of their underlying drivers. To overcome the substantial measurement issues in this task, we consider a particular industry as a special case: Chinese fine art. The simplicity of the supply-side of art vis-a-vis marginal cost and the wealth of data on its quality characteristics makes it possible to separately identify the markup and quality components of international relative prices for Chinese artworks. Through this lens, we trace the process of internationalization of Chinese art since the year 2000 and uncover a rich set of facts. We find strong support for quality sorting into international markets at both the level of artist and artwork, as well as substantial markup differences across destinations. Using a structural model of endogenous quality choice by Feenstra and Romalis (2012), we argue that much of the international quality premium is driven by specific distribution costs (whether physical or informational) rather than destination-specific preferences for quality.

We revisit Feldhutter and Schaefer (FS, 2018) who report evidence of a ``credit spread puzzle'' for high-yield but not investment-grade bonds. We show their results are reversed when their model is calibrated using market values of debt rather than book values. More importantly, we demonstrate that calibrating structural models of default to credit spreads rather than to historical default rates provides the statistical power necessary to reject the FS assumption that asset value dynamics follows geometric Brownian motion. A large market price of risk on jumps is required to match historical default rates, thus reaffirming a ``credit spread puzzle'' for investment-grade but not high-yield bonds.Security Lending and Corporate Financing  (with Massimo Massa and Hong Zhang)

Policy Report