Counter‐Credit‐Risk Yield Spreads: A Puzzle in China's Corporate Bond Market

• Author: Xiaoxia Ye,Jian Luo,May Hu
• Date: 01 June 2016
• Published date: 01 June 2016
• DOI: http://doi.org/10.1111/irfi.12079

Counter-Credit-Risk Yield Spreads: A

Puzzle in China’s Corporate Bond

Market

JIAN LUO

†

,XIAOXIA YE

‡

AND MAY HU

§

†

Wang Yanan Institute for Studies in Economics, Xiamen University, Xiamen, China,

‡

Stockholm Business School, Stockholm University, Stockholm, Sweden and

§

Deakin Graduate School of Business, Deakin University, Melbourne, Australia

ABSTRACT

In this paper, using China’s risk-free and corporate zero yields together with

aggregate credit risk measures and various control variables from 2006 to

2013, we document a puzzle of counter-credit-risk corporate yield spreads.

We interpret this puzzle as a symptom of the immaturity of China’s credit

bond market, which reveals a distorted pricing mechanism latent in the fun-

damental of this market. We also find interesting results about relationships

between corporate yield spreads and interest rates and risk premia and the

stock index, and these results are somewhat attributed to this puzzle.

JEL Codes: G12; E43

I. INTRODUCTION

In China, banks have long been the primary source of capital. Total outstanding

bank loans are 10 times larger than that of credit bonds (Ma, 2006; Suzuki et al.,

This paper was previously circulated under the title ‘A Puzzle of Counter-Credit-Risk Corporate Yield

Spreads in China’s Corporate Bond Market’by the first two authors. We would like to thank Hong

Yan (the editor), Hui Chen, Jing-Zhi Huang (discussant at the 2014 CICF), Van Vu (discussant at the

5th FMCGC), and Fan Yu for their great, helpful discussions; Reena Aggarwal, Banita Bissoondoyal-

Bheenick, Robert Durand, Robert Faf , Michael Graham, Bj örn Hagströmer, Mia Hinneri ch, Ai Jun

Hou, Jia Hua, Michael Keefe, Steven Li, Hai Lin, IñakiRodriguez Longarela, Lars Nordén, Leigh Roberts,

Goran Zafirovand seminar participantsat the Stockholm BusinessSchool, the 5th FinancialMarkets and

CorporateGovernance Conference, the2014 China International Conference in Finance,and the VUW

School of Economicsand Finance for their helpfulcomments; Tieliang Guoand Martin Lueken for read-

ing earlier draftsof this paper and their insightful comments; Tingguo Zheng for providing us the (up-

dated) GDP Growth data used in Zheng and Wang (2010); and the Risk Management Institute at the

National University of Singapore for making their aggregate credit risk measures freely available online.

All remaining errors are our o wn.

© 2016 International Review of Finance Ltd. 2016

International Review of Finance, 16:2, 2016: pp. 203–241

DOI: 10.1111/irfi.12079

2008).

1

The Chinese government itself has realized that this highly skewed finan-

cial system accumulates systemic risks. Regulatory reform, with officials clearing

away some of the obstacles that have stood in the way of the development of

the bond market (Mu, 2006; Zhou, 2003; Zhou, 2006), has been constantly

advancing the development of China’s credit bond market. After the People’s

Bank of China (PBC) launched commercial papers (CPs) in 2005, China’s credit

bond market entered into its fast growth period. By 2012, the annual issuance

of the whole market is RMB 2 trillion, which is about seven times the amount in

2005. With this speed of growth, the credit bond market is becoming more and

more important in China. Given China’s increasingly important and influential

role in the global economy,this market will soon become one of the indispensable

members in the international investment community. Therefore, it deserves

timely and careful studies that target not only Chinese but broader research audi-

ence. Hale (2007) and Chen et al. (2011) describe the current status and the recent

developments of China’s credit bond market. Huang and Zhu (2009) provide an

overview of the history of China’s credit bond market, although their focus is

on the whole bond market. To the best of our knowledge, however, there has been

no previous literature that provides quantitative empirical studies on this market.

Using a set of comprehensive zero yield curve data of China’s government

bonds and credit bonds, along with China’s aggregate credit risk measures, mac-

roeconomic variables, and an aggregate liquidity measure from 2006 to 2013, we

surprisingly find a puzzle of counter-credit-risk corporate yield spreads (YS). In

other words, we find a significant and negative relationship between the corpo-

rate YS and aggregate credit risk measures. This result is completely counterin-

tuitive given the fact that a corporate bond market is naturally expected to

discover the price of credit risks in a directly proportional way.

Specifically, we first use a three-factor Gaussian term structure model to cap-

ture the dynamic of the term structure of risk-free interest rates. Then, we

develop a five-factor affine defaultable bond model, in which the first three fac-

tors coming from the interest rate model represent the risk factors of the interest

rates, and the other two factors (CIR processes) capture the corporate-specific

component (the component in the corporate yield spread that is uncorrelated

to the risk-free interest rate factors) implicit in the term structure of corporate

YS. This five-factor model is fitted to the corporate zero yield curve data of bonds

with different ratings. Results of the model allow us to study (i) the relationship

between corporate YS and risk-free interest rates; (iii) the dynamics of the

corporate-specific component; and (iii) the dynamics of corporate-specific risk

premia. We find that the corporate YS are negatively related to the level of the

risk-free interest rates but positively related to the slope and curvature of the

interest rates. The positive relationships with the slope and curvature, which

1‘Credit bond’is a term unique to the Chinese market. It is similar to corporate bond in the gen-

eral sense. However, ‘corporate bond’is used in China to represent a sub-market of the whole

‘credit bond’market. The relationship between different terms will become clear as we move

on to the section describing China’s credit bond market.

International Review of Finance

© 2016 International Review of Finance Ltd. 2016204

contradict theoretical predictions (Collin-Dufresne et al., 2001; Longstaff &

Schwartz, 1995), might be attributed to the counter-credit-risk puzzle.

We then regress the model implied corporate-specific components on the

aggregate credit risk measures. The regression results formally confirm the

counter-credit-risk puzzle: the corporate-specific components are significantly

and negatively related to the aggregate credit risk measures. The results are robust

to different aggregate credit risk measures and to the inclusion of various control

variables. The control variables include the non-performing bank loan rate, the

stock index, the GDP growth rate, and an aggregate bond market liquidity mea-

sure. Next, we regress the model implied corporate-specific risk premia on the

aggregate credit risk measures and the control variables. We find that the risk

premia are not consistently and significantly related to the credit risk, and they

are, however, significantly and negatively related to the GDP growth. This is con-

sistent with the evidence in Adrian et al. (2010) about the relationship between

the GDP growth and themacro risk premium. For higher (lower) ratings, they have

a significantly positive (negative) relationship with the stock index (the non-

performing bank loan rate). This might indicate that for the credit bond

investors, the equity and bank loan markets are comparable substitutes in the

sense that when the equity market and/or bank loan market perform better

(worse), the opportunity cost of investing in the credit bond market becomes

higher (lower), therefore the requested risk premium is higher (lower). The results

of the risk premia help us understand the cause and mechanismbehind the puzzle.

We interpret this puzzle as one of the symptoms of the immaturity of China’s

credit bond market. The liquidity in the secondary market is very low, which pre-

vents the bond prices from revealing the true underlying risks. At the same time,

zero default experience in the past two decades plus the explicit or implicit guar-

antees provided by high profile parent companies or local governments attract

credit risk sensitive capital when the overall credit condition deteriorates. This

distorted pricing mechanism might funnel the credit risks to the credit bond

market instead of diversifying the credit risks. The emergence of the symptom

alerts policy makers to focus more on the secondary market development and

correction of the credit risk pricing mechanism.

The remainder of this paper is organized as follows. Section 2 provides a brief

introduction to China’s credit bond market. Section 3 describes the data, and pre-

sents results of some preliminary analyses, which already reveal the counter-

credit-risk puzzle. In Section 4, we derive the term structure models that are used

to identify the corporate-specific risk factorsand risk premia from risk-free and cor-

porate zeroyields. Section 5 presents formalempirical analyses. Section6 concludes

the paper. Appendicescontain technical details andsupplemental empirical results.

II. CHINA’S CREDIT BOND MARKET

China’s creditbond market started in 1983. While before 2005 thedevelopment of

the market was ratherslow, it has beenrapidly developing since 2005 in whichthe

Counter-Credit-Risk Yield Spreads

© 2016 International Review of Finance Ltd. 2016 205