The externalities of credit default swaps on stock return synchronicity

• Author: Lu Zhu,Ran Zhao
• Date: 01 January 2020
• Published date: 01 January 2020
• DOI: http://doi.org/10.1002/fut.22045

J Futures Markets. 2020;40:92–125.wileyonlinelibrary.com/journal/fut92

|

© 2019 Wiley Periodicals, Inc.

Received: 16 September 2018

|

Accepted: 9 July 2019

DOI: 10.1002/fut.22045

RESEARCH ARTICLE

The externalities of credit default swaps on stock return

synchronicity

Ran Zhao

1

|

Lu Zhu

2

1

Peter F. Drucker and Masatoshi Ito

School of Management, Claremont

Graduate University, Claremont,

California

2

Department of Finance, College of

Business, California State University,

Long Beach, California

Correspondence

Lu Zhu, Department of Finance, College

of Business, California State University,

Long Beach, 1250 Bellflower Boulevard,

Long Beach, CA 90840.

Email: lu.zhu@csulb.edu

Abstract

We examine the externality effect of customer firms’credit default swap (CDS)

trading on the stock price informativeness of supplier firms. Our empirical

results show that firms with a high proportion of sales to CDS referenced

customers tend to have more firm‐specific embedded information in their stock

prices and thus higher stock price informativeness, which is associated with a

lower level of stock return synchronicity. We provide new evidence of CDS

trading externality on equity market information environments along the

supply chain.

KEYWORDS

credit default swaps, customer–supplier relationship, externalities, information spillover, stock

return synchronicity

JEL CLASSIFICATION

G12; G13; G14

1

|

INTRODUCTION

Credit default swaps (CDSs) are one of the most influential financial innovations in recent decades.

1

They are an

effective risk‐sharing and risk‐taking tool for major financial institutions and hedge funds. Recent literature has linked

the initiation of CDS trading to the corporate decisions of the underlying firms, such as changes in borrowing costs

(Ashcraft & Santos, 2009), leverage ratio (Saretto & Tookes, 2013), probability of bankruptcy (Subrahmanyam, Tang, &

Wang, 2014), accounting reporting practices (Martin & Roychowdhury, 2015), cash holdings (Subrahmanyam, Tang, &

Wang, 2017), and managers’voluntary disclosures (Kim, Shroff, Vyas, & Wittenberg‐Moerman, 2018).

Although there is a growing literature on the impact of CDS trading on the underlying reference firms, however,

research into the spillover effects of CDS trading has been sparse. It is worth noting that only a limited number of firms

have CDS contracts traded on their debt, and most of those are large firms. A much larger number of smaller publicly

traded firms are indirectly linked to CDS trading through their economic connections with CDS referenced firms.

Therefore, the ultimate impact of CDS trading could be much more widespread than we expected. Augustin,

Subrahmanyam, Tang, and Wang (2016) have urged researchers to conduct a more “comprehensive analysis of the

welfare implications of CDS contracts.”In this paper, we examine whether the impact of CDS trading goes beyond the

referenced entities and investigate the information‐spillover effect, or the externality effects of CDS trading.

Business linkages along supply chains are among firms’most significant economic ties. Customers’information

provides valuable indication of their suppliers’performance. For example, in a recent report by Bloomberg, Apple Inc.’s

1

CDS is a financial insurance providing the buyer with protection against losses in the default event. In the event of default, the insurance seller pays for the financial loss the insurance buyer suffers.

Each period, the buyer makes insurance payments to the seller until default or maturity date.

supplier Hon Hai Precision Industry experienced its biggest earning declines since the last quarter of 2008 due to tepid

iPhone 8 sales.

2

For the third quarter of 2017, Hon Hai reported a 39% decline in net income. Its stock price dropped to

$106 on November 14, 2017, from $121.5 on August 7. Given the significant interdependence between suppliers and

customers, a natural question is whether the impact of CDS trading can be extended to the suppliers. In other words,

does CDS trading on customers have any spillover effect on their suppliers? In this paper, we investigate whether CDS

trading on customers enriches the information environment for their suppliers and further reduces the suppliers’stock

return synchronicity.

The literature includes examinations of whether the addition of CDS markets to investors’opportunity set can

provide complementary firm‐specific information about the underlying reference entities to investors, in two aspects.

First, due to informed trading in the CDS market, there is an earlier price discovery in the CDS market than in the stock

market. Consistent with this argument, Acharya and Johnson (2007) and Qiu and Yu (2012) find that informed traders

prefer the CDS market, and thus there is significant incremental information flow from the CDS market to the equity

market, especially when the underlying firms have more bank relationships and have recently experienced credit

events. In addition, Lee, Naranjo, and Sirmans (2014) argue that the momentum in CDS returns provides additional

information that is not contemporaneously reflected in stock prices. Second, CDS trading can promote firm‐specific

information production (Kim et al., 2018). To offset reduced lenders’monitoring, shareholders request more voluntary

disclosure from managers after the introduction of CDS. Regardless of whether it happens through price discovery,

information production, or both, we expect that the CDS market conveys additional information of the underlying CDS

firms to investors. Since the firm‐specific information for CDS referenced customers could be used by investors to

foresee the performance of the economically tied suppliers, the CDS market becomes an additional information source.

Therefore, we expect improved informativeness of the suppliers’stock prices due to additional customer information

from the CDS market.

In the financial economic literature, stock return synchronicity has long been used to measure the amount of firm‐

specific information embedded in stock prices, with more firm‐specific information being associated with lower

synchronicity. Roll (1988) shows that less than 40% of stock return variation can be explained by macromovements, and

the rest is explained by firm‐specific factors. Additional firm‐specific information incorporated into stock price lowers

the comovement between stock price and macrofactors, resulting in lower synchronicity. Following this literature, we

measure stock return synchronicity as one minus the R

2

from an asset‐pricing regression of stocks returns on market

factors and use it as our proxy for stock price informativeness (Crawford, Roulstone, & So, 2012; Fernandes & Ferreira,

2008; Kim & Shi, 2012). If the economic linkage along the supply chain is stronger, the customer information could

better reflect the suppliers’prospects. As such, we conjecture that suppliers’stock return synchronicity decreases more

when the economic connections between customer and suppliers are stronger (i.e., when a larger proportion of

suppliers’revenue is from CDS referenced customers).

To test our hypotheses, we construct data from CDS trading merged with supply–chain relationships and conduct

regression analyses. Consistent with our expectations, we find that the stock return synchronicity of suppliers tends to

be lower when a larger proportion of their sales are from CDS referenced customers. The effect is both statistically and

economically significant. For instance, for a firm at the average SYNC1 (the sample mean is –1.990 from Table 2), a

0.331 reduction in SYNC1 represents a 16.63% decrease in the sample mean SYNC1. In terms of R

2

, a reduction of

SYNC1 by 0.331 lowers R

2

by 7.05% (from 15.99% to 8.94%). Our result is also robust to alternative synchronicity

measures, especially from model‐free informativeness measures such as the analyst forecast error. Overall, our

empirical evidence suggests that customer information conveyed from the CDS market enriches the information

environment of the corresponding suppliers. In other words, CDS trading on customers has substantial spillover effects

on the stock price informativeness of the suppliers.

However, the existing research on the impact of CDS trading all faces the potential endogeneity issue

(Ashcraft & Santos, 2009; Kim et al., 2018; Martin & Roychowdhury, 2015; Saretto & Tookes, 2013;

Subrahmanyam et al., 2014, 2017). One of the concerns is that stock return synchronicity and sales to CDS

reference customers are jointly determined. Another one is that there are omitted variables accounting for the

association between suppliers’stock return synchronicity and sales to CDS referenced customers. To address

these concerns, we follow previous studies and employ three different approaches: (a) A difference‐in‐difference

(DID) analysis, by matching treated and control groups of suppliers. In the matched sample, the treated suppliers

2

https://www.bloomberg.com/news/articles/2017‐11‐14/hon‐hai‐posts‐surprise‐profit‐fall‐amid‐delays‐in‐the‐iphone‐x.

ZHAO AND ZHU

|

93

and control suppliers are similar in firm characteristics but have different customer CDS trading status. We find

that suppliers with CDS referenced customers experiencemoredeclinesinstockreturnsynchronicitythanthose

without. (b) A two‐stage instrumental variable approach, using the foreign exchange derivative hedging position

from customers’lenders or bond underwriters as instrumental variables. The foreign exchange hedging position

is directly linked to the likelihood of trading CDS contracts from customers’lenders or bond underwriters

(satisfying the relevance condition), butitisunlikelytobelinkedtosuppliers’stock return synch ronicity

(satisfying the exclusion condition). The impact on suppliers’stock return synchronicity remains both

statistically and economically significant after instrumenting on the foreign exchange derivative hedging

position. (c) A placebo test, by altering the timing of customers’CDS initiation years using the matched data

sample from the DID analysis. We find that the reduction in suppliers’stock return synchronicity is not driven by

an unobserved trend before the customers’CDS initiation.

Next, we explore the differential effect of the customers’CDS initiation on stock return synchronicity from both

information demand and supply channels. Intuitively, suppliers’synchronicity decreases more when there is

higher demand for supplier information or greater information supply from the customer CDS market. From the

information demand channel, given that poor information environment is associated with higher information

demand, we expect such negative relationship to be stronger for suppliers with more opaque information

environment. From the information supply channel, we conjecture that suppliers’synchronicity decreases more if

they have more financially distressed customers. Acharya and Johnson (2007), Qiu and Yu (2012), and Berndt and

Ostrovnaya (2014) show that the CDS market reveals firm‐specific private information ahead of the stock and

option markets, especially for those riskier firms; thus, more firm‐specific information is revealed, especially for

the financially distressed customers. Consistent with our expectations, our results indeed suggest that the negative

relationship between suppliers’synchronicity and sales to CDS referenced customers is more pronounced for

suppliers with distressed customers and poor information environment. In addition, we find complementary

evidence that the CDS externality effects are more pronounced for suppliers with higher proportions of trade

credits granted to customers, more differentiated outputs and services, and longer trading relationships with

certain customers.

Finally, our findings are also robust to different measures (on both dependent and independent variables),

sample selections, and methodologies. We used alternative measures of the economic linkage with CDS referenced

customers: (a) sales to CDS referenced customers as a proportion of sales to all critical customers, and (b) a dummy

for whether a supplier has CDS referenced customers. The empirical conclusion remains quantitatively consistent

between the alternative measures. The additional change analysis suggests that information externality from

customers’CDS trading remains robust. It is worth noting that our baseline sample includes those suppliers with

CDS references themselves, and the empirical results remain statistically and economically significant if we

exclude the CDS referenced suppliers.

Our paper makes the following contribution. First, it advances the mission of a more “comprehensive analysis of the

welfare implications of CDS contracts”urged by Augustin et al. (2016). Specifically, we show that the impact of CDS

trading can be extended to non‐CDS firms along a supply chain network. Except for Li and Tang (2016), few studies

have examined the externality effects of CDS trading. Li and Tang find that CDS trading on customers has spillover

effects on suppliers’leverage ratios. Our paper complements this study by showing that the customer CDS market

enriches suppliers’information environment and therefore decreases suppliers’stock return synchronicity. Our results

suggest that the CDS market serves as an additional information source, not only for the underlying CDS referenced

firms but for their economically linked firms. Second, our findings are relevant to the debate on the role of the CDS

market. Although the CDS market has been blamed for creating “instruments of destruction”during the 2008 financial

crisis, it can also be beneficial for market participants by enriching the information environment for both suppliers and

customers. Third, our paper contributes to the literature on stock return synchronicity. The empirical results provide

new evidence that both supply chain relationship and financial market development can influence stock return

synchronicity. One important practical implication of our findings is that, to make wiser investing decisions, investors

and financial intermediaries can acquire additional firm‐specific information on their economically linked partners

from the CDS market.

The remainder of the paper proceeds as follows. Section 2 reviews the current literature on the CDS market,

information externality along the supply chain, and stock return synchronicity and develops hypotheses. Section 3

describes the data sources, defines the variables, and provides the empirical designs. Section 4 presents and discusses

empirical results. Section 5 concludes the paper.

94

|

ZHAO AND ZHU