CEO Risk‐taking Incentives and Bank Loan Syndicate Structure

• Date: 01 November 2014
• DOI: http://doi.org/10.1111/jbfa.12087
• Published date: 01 November 2014
• Author: Liqiang Chen

Journal of Business Finance & Accounting

Journal of Business Finance & Accounting, 41(9) & (10), 1269–1308, November/December 2014, 0306-686X

doi: 10.1111/jbfa.12087

CEO Risk-taking Incentives and Bank

Loan Syndicate Structure

LIQIANG CHEN*

Abstract: This paper investigates the effects of a borrowing firm’s CEO risk-taking incentives

on the structure of the firm’s syndicated loans. When CEO risk-taking incentives are high,

syndicates are structured to facilitate better due diligence and monitoring efforts. These

syndicates have a smaller number of total lenders and are more concentrated, and lead

arrangers will retain a greater portion of the loan. Moreover, CEO risk-taking incentives have a

lesser effect on the syndicate structure when lead arrangers have a good reputation and a prior

lending relationship with a borrowing firm, while they have a greater effect on the syndicate

structure when borrowing firms have low information transparency, are financially distressed or

have low growth prospects.

Keywords: risk-taking incentives, syndicate loans

1. INTRODUCTION

Over the past 20 years, syndicated loans have become a dominant form of bank lending

in the global corporate financing market, with originations in 2013 surpassing US$ 4.2

trillion (Loan Pricing Corporation). Syndicated loans are made by multiple lenders,

with one or more of the lenders (lead arrangers) playing the role of arranging, pricing

and monitoring such loans. Lead arrangers analyze credit quality, negotiate key terms

with borrowers before inviting a group of banks to participate and are responsible

for allocating loan shares among participating banks (Lin et al., 2012). Although the

lead arrangers perform the traditional role of due diligence as informed lenders,

the loan amount itself is shared with one or more syndicate participant banks (Esty,

2001). Given that lead arrangers in a syndicate hold less than 100% of the debt, other

participant lenders can become concerned about the level of monitoring effort that

is exerted by the lead arrangers because lead arrangers do have an incentive to shirk

their monitoring responsibilities when undertaking most of the monitoring costs and

owning only part of the loan (Holmstrom and Tirole, 1997; Sufi, 2007). In addition,

∗The author is PhD, Assistant Professor from the Department of Finance, IS & MS, Sobey School of

Business, Saint Mary’s University, Halifax, Nova Scotia, Canada. The author is grateful for comments from

an anonymous referee, as well as Jiaping Qiu, Luke Chan, Peter Miu, Mohammad Rahaman and Andrew

W. Stark (the editor). All errors are those of the author. (Paper received September 2014, revised version

accepted September 2014)

Address for correspondence: Liqiang Chen, PhD, Assistant Professor, Department of Finance, IS & MS,

Sobey School of Business, Saint Mary’s University,Halifax, Nova Scotia B3H 3C3, Canada.

e-mail: liqiang.chen@smu.ca

C

2014 John Wiley & Sons Ltd 1269

1270 CHEN

adverse selection problems may arise because loans with higher ex-ante credit risks

are more likely to be syndicated by lead arrangers when lead arrangers have better

information than participant banks do.

Therefore, the syndication process generates conflicts of interest between the

lead arrangers and other participant syndicate members, in addition to the typical

moral hazard problems arising between the borrowing firms and syndicate lenders.

Participant banks’ concerns regarding the lead arrangers’ potential to shirk their

monitoring and due diligence duties are especially relevant in situations in which

borrowing firms require better screening and more intensive monitoring, and as the

literature suggests, those concerns are reflected during the syndication process. For

example, Sufi (2007) that and Lee and Mullineaux (2004) demonstrate that borrowing

firms with higher levels of information asymmetry require a more concentrated

syndicate structure. Ball et al. (2008) reveal that lead arrangers hold a smaller portion

of syndicated loans if borrowing firms’ accounting information can capture credit

quality in a timely fashion. While prior studies have demonstrated that the structure of

syndicated loans will be affected by borrower moral hazard problems, little is known

about the potential sources of the borrower’s moral hazards that might influence the

structure of syndicated loans.

Equity compensation – such as options – can provide risk-taking incentives and

encourage chief executive officers’ (CEOs) risk-seeking behaviors such that risk-

averse CEOs will take on risky but positive net present value (NPV) projects (Coles

et al., 2006). However, creditors might have negative perceptions of such risk-seeking

behaviors because of potential asset substitution effects (Jensen and Meckling, 1976).

Thus, CEO risk-taking incentives can create conflicts of interest between shareholders

and creditors and are an important source of borrower moral hazard problems. As

a result, syndicate lenders should consider CEO risk-taking incentives during the

syndication process, and the ownership of syndicated loans should be structured to

address syndicate lenders’ concerns regarding CEO risk-taking incentives. However,

the literature remains silent with respect to how CEO risk-taking incentives affect the

syndication structure and, if so, to what end.

The purpose of this paper is twofold. First, we focus on borrowing firms’ executive

compensation incentives to investigate how CEOs’ risk-taking incentives influence the

structure of syndicated loans. Following the extant literature (e.g., Guay, 1999; Coles

et al., 2006; Core and Guay, 2002), CEO risk-taking incentives are measured based

on the sensitivity of the value of a CEO’s portfolio to stock return volatility (vega)

and stock price (delta). Similar to recent empirical studies (e.g., Lee and Mullineaux,

2004; Sufi, 2007; Lin et al., 2012), we employ four measures to estimate the structure

of syndicated loans: the total number of lenders in a syndicate, the amount of the

loan held by lead arrangers, the percentage of the loan held by lead arrangers and

a Herfindahl index of lenders’ shares. With a combined sample of syndicate loan

structure information, financial information from borrowing firms and CEO risk-

taking incentives from 1992 to 2010, we find that vega has a significant influence on the

structure of a syndicated loan, whereas delta has no significant effect on the syndication

structure. Specifically, a syndicated loan will have a smaller number of total lenders,

lead arrangers will hold a greater amount and percentage of the syndicate loan,

and the syndicate ownership will be more concentrated as vega increases. The results

suggest that syndicate lenders indeed consider CEO risk-taking incentives when they

C

2014 John Wiley & Sons Ltd

CEO INCENTIVES AND SYNDICATE STRUCTURE 1271

form a syndicate that is structured to ensure more efficient monitoring of borrower

moral hazard.

The second purpose of this paper is to investigate the possible channels through

which CEO risk-taking incentives affect the structure of syndicated bank loans. We

find that vega’s effects are moderated by lead arrangers’ reputation, borrowing firms’

lending relationship with lead arrangers, and the borrowing firms’ information

transparency, financial distress level and growth opportunities. In particular, vega’s

effects on the syndication structure are mitigated if lead arrangers have a good

reputation or a lending relationship with borrowing firms. Conversely, vega’seffects on

the syndication structure are strengthened if borrowing firms have a low information

transparency level, are financially distressed or have low growth opportunities.

One potential concern for our empirical analysis is the issue of endogeneity.

Although the syndicate structure of a loan is unlikely to be a direct determinant

of the CEO risk-taking incentives of the borrowing firm, our empirical results may

be driven by the possibility that some unaccounted-for firm characteristics could

jointly determine CEO risk-taking incentives and syndicate structure. We address this

endogenous issue in the following ways. First, for all of the variables relating to firm

and CEO characteristics, we use lagged values rather than contemporaneous values

in all empirical specifications. The use of lagged values can mitigate endogenous

concerns from the missing variable bias and reverse causality. Second, we use an

instrumental variable approach to address the possible endogeneity of CEO risk-

taking incentives. Following extant empirical studies (e.g., Liu and Mauer, 2011), we

instrument CEO risk-taking incentives with CEO and firm characteristics, such as firm

age, CEO age and CEO tenure. Our empirical results continue to hold when CEO

risk-taking incentives are instrumented with instrumental variables. Third, as argued

by Lin et al. (2012), interaction tests of the moderators that affect the relationship

between CEO risk-taking incentives and the loan syndication structure also help to

alleviate omitted variable bias because an omitted variable is less likely to be correlated

with interaction terms than with linear terms (Raddatz, 2006).

Moreover, we performed several additional tests to check the robustness of our

empirical results. First, we estimate Poisson and Tobit regressions to assess the

robustness of the ordinary least square (OLS) regression results, as some of the

dependent variables are either truncated or in integer form. Both the Poisson and

Tobit regressions generate similar results to those of the OLS regression. Second, we

consider the effects of executive compensation characteristics such as CEO’s inside

debts (i.e., pension and deferred compensation) and the cross-sectional difference of

vega (high vega vs. low vega) on the relationship between CEO risk-taking incentives

and the syndication structure of bank loans. Third, we investigate the effects of lead

arrangers’ dual ownership of borrowing firms’ equity and debt on our main results.

Fourth, we examine corporate governance quality to determine the effects of corpo-

rate governance on the relationship between CEO risk-taking incentives and the syn-

dication structure. Finally, we examine the impact of macro-economic conditions such

as recessions on the main results, as recessions exacerbate lenders’ distress levels and

increase borrowing firms’ financial risks. Following these additional robustness tests,

our main results still hold: CEO risk-taking incentives have a significant impact on

the syndication structure of bank loans. While these additional tests cannot entirely

exclude alternative explanations, they certainly increase the validity of our empirical

results.

C

2014 John Wiley & Sons Ltd