Credit Default Swaps and Managers’ Voluntary Disclosure

• DOI: http://doi.org/10.1111/1475-679X.12194
• Date: 01 June 2018
• Author: PERVIN SHROFF,DUSHYANTKUMAR VYAS,REGINA WITTENBERG‐MOERMAN,JAE B. KIM
• Published date: 01 June 2018

DOI: 10.1111/1475-679X.12194

Journal of Accounting Research

Vol. 56 No. 3 June 2018

Printed in U.S.A.

Credit Default Swaps and Managers’

Voluntary Disclosure

JAE B. KIM,

∗PERVIN SHROFF,

†DUSHYANTKUMAR VYAS,

‡

AND REGINA WITTENBERG-MOERMAN

§

Received 17 March 2015; accepted 19 October 2017

ABSTRACT

We investigate how the availability of traded credit default swaps (CDSs)

affects the referenced firms’ voluntary disclosure choices. CDSs enable

∗College of Business and Economics, Lehigh University; †Carlson School of Management,

University of Minnesota; ‡Department of Management – UTM & Rotman School of Manage-

ment, University of Toronto; §Marshall School of Business, University of Southern California.

Accepted by Stephen Ryan. We appreciate the helpful comments of two anonymous re-

viewers, Phil Berger, Neil Bhattacharya, Jeffrey Callen, Xia Chen, Maria Correia, Gus De

Franco, Emmanuel De George, Yiwei Dou, Richard Frankel, Frank Gigler,Di Guo (discussant),

JungKoo Kang, Stephen Karolyi (discussant), Hanna Lee (discussant), Christian Leuz, Paul

Ma, Xiumin Martin (discussant), Miguel Minutti-Meza, Jeffrey Ng, Shailendra Pandit (discus-

sant), Joshua Ronen, Sugata Roychowdhury, Doug Skinner, Ewa Sletten, Irem Tuna, Andrew

Winton, Joanna Wu, Biqin Xie, Gaoqing Zhang, Ivy Zhang, Yoonseok Zang, and workshop

participants at Boston College, the Columbia University Burton Workshop, Lehigh University,

London Business School, New York University,the Nor wegian School of Economics, Pennsyl-

vania State University, Rice University, Singapore Management University, Washington Uni-

versity in St. Louis, the University of Calgary, the University of Miami, the University of New

Castle, the University of Rochester, University of Texas at Arlington, the 2014 Financial Eco-

nomics and Accounting Conference, the 2015 European Accounting Association annual and

Financial Accounting and Reporting Section midyear meetings, the 2015 Singapore Manage-

ment University Accounting Symposium, the 2015 EuroFIT conference on syndicated loans at

London Business School, and the 2015 University of Illinois at Chicago accounting research

conference. We are very grateful to Vincent Pham for excellent research assistance. We grate-

fully acknowledge the financial support of Lehigh University, Singapore Management Uni-

versity, the University of Minnesota, the University of Toronto, the University of Chicago, the

University of Southern California, and the Social Sciences and Humanities Research Coun-

cil of Canada. This paper was previously circulated under the titles “Active CDS Trading and

Managers’ Voluntary Disclosure” and “CDS Tradingand Managers’ Voluntary Disclosure.”

953

Copyright C, University of Chicago on behalf of the Accounting Research Center,2017

954 J.B.KIM,P.SHROFF,D.VYAS,AND R.WITTENBERG-MOERMAN

lenders to hedge their credit risk exposure, weakening their incentives to

monitor borrowers. We predict that reduced lender monitoring in turn

leads shareholders to intensify their monitoring and demand increased

voluntary disclosure from managers. Consistent with this expectation, we

find that managers are more likely to issue earnings forecasts and fore-

cast more frequently when traded CDSs reference their firms. We fur-

ther find a stronger impact of CDS availability on firm disclosure when

(1) lenders have higher ability and propensity to hedge credit risk using

CDSs, and (2) lender monitoring incentives and monitoring strength are

weaker. Consistent with an increase in shareholder demand for public in-

formation disclosure induced by a reduction in lender monitoring, we find

a stronger effect of CDSs on voluntary disclosure for firms with higher in-

stitutional ownership and stronger corporate governance. Overall, our find-

ings suggest that firms with traded CDS contracts enhance their volun-

tary disclosure to offset the effect of reduced monitoring by CDS-protected

lenders.

JEL codes: G14; G20; G21; G23; G32; M40; M41

Keywords: CDS market; credit default swaps; CDS trading initiation; bank

monitoring; private lender monitoring; voluntary disclosures; earnings

forecasts; management forecasts

1. Introduction

This study investigates the effect of the availability of traded credit default

swaps (CDSs) on the referenced firms’ voluntary disclosure choices.1The

availability of traded CDSs (CDS availability, hereafter) enables lenders to

distribute credit risk to parties more willing and able to bear it, thereby

enhancing the liquidity and flexibility of individual lenders and the finan-

cial system (Greenspan [2004]). However, by unbundling lenders’ cash

flow and control rights, CDS availability decreases lenders’ incentives to

monitor borrower performance (e.g., Gorton and Pennacchi [1995], Stulz

[2010], Parlour and Winton [2013]). We predict and find that CDS refer-

enced firms enhance their voluntary disclosure to offset this negative effect

of traded CDSs.

Private debt contracts rely on extensive information collection and bor-

rower monitoring by lenders (e.g., Diamond [1984], Fama [1985], Sufi

[2007]). Prior research documents a dilution of lenders’ monitoring in-

centives resulting from credit risk transfer mechanisms (e.g., Pennacchi

[1988], Gorton and Pennacchi [1995]). In contrast to other mechanisms

such as loan sales, CDS contracts do not transfer control or monitor-

ing rights to counterparties (Marsh [2009], Stulz [2010], Parlour and

1A CDS protects the buyer of the contract against default by the reference entity in re-

turn for a periodic payment (the CDS spread) over the term of the contract or up to default

whichever comes first. The buyer is compensated if the reference entity experiences a “credit

event” specified in the contract, such as default, certain types of restructuring, and bankruptcy.

CDS AND MANAGERS’VOLUNTARY DISCLOSURE 955

Winton [2013]). Hence, the use of CDSs to hedge credit risk likely dilutes

the combined monitoring incentives of the original lender and the CDS

writer.

Recent empirical studies find evidence of a significant decline in the

rigor and efficiency of monitoring by CDS-protected lenders. Ashcraft and

Santos [2009] find an increase in the cost of debt financing for risky and

informationally opaque firms after the onset of CDS trading. Amiram et al.

[2017] show that loan syndicate participants demand higher loan share re-

tention by the lead arrangers and higher loan spreads to compensate for

the reduction in lead arranger monitoring incentives after the initiation

of CDS trading. Similarly, Subrahmanyam, Tang, and Wang [2014] find

that borrowers’ credit risk increases after the inception of CDS trading.

Martin and Roychowdhury [2015] show that CDS-referenced firms (CDS

firms, hereafter) report less conservatively. Prior studies also show that

CDS availability undermines the effectiveness of financial covenants—a key

monitoring mechanism in private lending. Shan, Tang, and Winton [2015]

find that loans to CDS firms have looser covenants that are less effective as

“tripwires” for loan contract renegotiations. Chakraborty, Chava, and Gan-

duri [2015] show that CDS-protected lenders are less likely to expend effort

in renegotiating loan contracts and impose weaker investment restrictions

when covenants are violated.

We expect shareholders to increase their information acquisition to mon-

itor managers and thereby compensate for their inability to effectively del-

egate monitoring to CDS-protected lenders. Shareholders, however, lack

access to the private information available to lenders and therefore pri-

marily rely on public disclosures for monitoring. Based on prior findings

that voluntary disclosure facilitates shareholder monitoring of managers

(Bushman and Smith [2001, 2003], Armstrong, Guay, and Weber [2010]),

we hypothesize that managers respond to shareholders’ information de-

mand by increasing voluntary disclosure. This hypothesis is further moti-

vated by evidence in Vashishtha [2014] that equity holders reduce their

demand for voluntary disclosure when lenders increase their monitoring

intensity following covenant violations (i.e., a tradeoff between monitoring

by lenders and shareholders in the opposite direction).

We focus on management earnings forecasts, an important voluntary dis-

closure choice (Beyer et al. [2010]). We test whether CDS availability is

significantly associated with the likelihood and frequency of management

earnings forecasts. Our benchmark specification includes year fixed effects

to control for time variation in the frequencies of CDS trading initiation

and voluntary disclosure. In addition, we include an indicator variable for

whether a firm has CDS availability at any point during our sample period to

control for time-invariant unobservable differences between CDS and non-

CDS firms (e.g., Ashcraft and Santos [2009], Saretto and Tookes [2013],

Subrahmanyam, Tang, and Wang [2014]). The inclusion of this variable al-

lows us to identify the incremental impact of CDS availability on voluntary

disclosure in the post-CDS initiation period.