CEO Incentives, Relationship Lending, and the Cost of Corporate Borrowing

• DOI: http://doi.org/10.1111/fima.12156
• Published date: 01 September 2017
• Date: 01 September 2017
• Author: Liqiang Chen,Jiaping Qiu

CEO Incentives, Relationship Lending,

and the Cost of Corporate Borrowing

Liqiang Chen and Jiaping Qiu∗

We investigate how lending relationships attenuate the conflict of interest between creditors and

shareholders that arises from chief executive officer (CEO) compensation contracts. We find

that lending relationships mitigate the influence of CEO risk-taking incentives on loan spreads,

especially forinformationally opaque firms. In addition, lending relationships attenuate the impact

of CEO risk-taking incentives on maturity and collateral requirements.This article highlights the

importance of bank monitoring through lending relationshipsto mitigate managerial risk-shifting

activities that arise from equity incentives.

Managerial risk-taking incentives may cause conflicts of interest between shareholders and

creditors. For example, shareholders may award risk-averse chief executive officers (CEOs)

greater risk-taking incentives to mitigate underinvestment problems, and creditors may have

concerns over asset substitution activities arising from the risk-taking incentives of the CEO

(see DeFusco, Johnson, and Zorn, 1990; Ortiz-Molina, 2006). However, the conflict can be

mitigated if shareholders and managers are under the close scrutiny of creditors (Rajan, 1992).

A creditor can threaten to reduce or deny future financing to a borrowing firm when the creditor

believes her interests are in danger. The threat is likely to be more effective when the creditor

is also a relationship lender because breaking up with a relationship lender sends a negative

signal to external investors (Boot, 2000). Therefore, relationship lending can provide aneffective

mechanism to enhance bank monitoring because the CEO is discouraged from engaging in

risk-shifting investment activities, even though the CEO may be awarded greater risk-taking

incentives.

In this article, we examine whether relationship lending mitigates potential conflicts of interest

between shareholders and creditors arising from managerial risk-taking incentives embedded in

executive compensation contracts. We conduct our analysis on a merged sample of 13,657 loans

to 1,994 US firms using data from ExecuComp, Compustat, Center for Research in Security

Prices (CRSP), and DealScan from 1992 to 2013. Following the extant literature (e.g., Coles,

Daniel, and Naveen, 2006; Core and Guay, 2002), we measure CEO risk-taking incentives based

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

price (Delta). Veg a is expected to be positively related to CEO risk-taking behavior because, as

Veg a increases, the increase in CEO compensation grows in tandem with stock return volatility.

Delta, however, has an ambiguous effect on CEO risk-taking incentives, as suggested by the

We are grateful to Raghavendra Rau (Editor) and an anonymous referee. Wethank Varouj Aivazian, Luke Chan, Trevor

Chamberlain, Sudipto Sarka, Yuhai Xuan, and seminar participants at McMaster University and the Eastern Finance

Association 2013 meetings for comments. Jiaping Qiu is gratefulto the Social Sciences and Humanities Research Council

of Canada for its financial support.

∗Liqiang Chen is from the Sobey School of Business at Saint Mary’s University in Halifax, Nova Scotia, Canada. Jiaping

Qiu is from the DeGroote Schoolof Business at McMaster University in Hamilton, Ontario, Canada.

Financial Management •Fall 2017 •pages 627 – 654

628 Financial Management rFall 2017

extant literature.1Tomeasure a lending relationship, we use the method outlined by Bharath et al.

(2011). For each loan, we search all previous loan transactions for the borrowing firm in the

five-year window preceding the loan activation date to identify prior lending transactions with

the same lead lender.

We find that Veg a (Delta) is positively (negatively) related to the costs of bank loans measured

using the all-in spread in a loan contract. However, the positive effect ofVe g a on the loan spread is

significantly lower if the borrowerhas a prior relationship with the lender. In contrast, relationship

lending does not moderate Delta’s negative effects on loan spreads. The positiveeffect of Ve g a on

loan spreads is consistent with the notion that Ve g a encourages managerial risk-taking behavior.

Relationship lending has the potential to moderate Veg a ’spositive effects on loan spreads because

enhanced bank monitoring through relationship lending discourages managers from engaging in

asset substitution activities even though CEOs are offered greater equity incentives.

The negative effect of Delta on loan spreads implies that a higher Delta leads to managerial

conservatism in selecting investmentprojects, as suggested by several prior studies.2However,the

insignificant moderating effect of relationship lending on Delta is not totally surprising. Because

firms with higher Delta are associated with conservative investment policies and are less likely

to expropriate from creditors, bank monitoring from relationship lending is not as necessary for

Delta as it is for Veg a . Our findings are also of economic signif icance. For loans borrowed from

banks with which a firm does not have a prior relationship, an increase in the natural logarithm

of Veg a from the first quartile to the third quartile increases the loan cost by 16.91 basis points

(bps). In contrast, for loans borrowed from banks with which the firm has a prior relationship, an

increase in the natural logarithm of Veg a from the first quar tile to the third quartile increases the

loan cost by only 7.25 bps.

Next, we explore the channels through which relationship lending affectsa bank’s sensitivity to

CEO risk-taking incentives. First, because relationship lending mitigates information asymmetry

and enhances monitoring intensity through proprietary information production, the effectiveness

of relationship lending on mitigating borrower moral hazard should be lowerfor infor mationally

transparent firms than for informationally opaque f irms. Indeed, we find that the effects of rela-

tionship lending on the loan spread-Veg a relation are stronger for informationally opaque firms.

Second, because relationship lending reduces monitoring costs through repetitive transactions,

we expect that relationship lenders will offer short-term loans for firms with greater managerial

risk-taking incentives to enhance monitoring intensity. Consistent with this conjecture, we find

that firms with greater risk-taking incentives usually have shorter-term loans and this effect is

stronger for relationship loans than for nonrelationship loans. This finding implies that the re-

duced monitoring costs of relationship lending enable a relationship lender to further shorten the

loan maturity for firms with higher CEO risk-taking incentives to increase monitoring frequency.

Third, prior studies suggest that nonpricing terms such as collateral also help mitigate borrow-

ing firms’ risk-shifting activities. We find that greater risk-taking incentives are associated with

more collateral requirements, though relationship lending can moderate this effect. This finding is

1For example,John and John (1993) show that the riskiness of an investment policy implemented bya manager increases

with Delta. Conversely, Smith and Stulz (1985) argue that Delta is expected to be negatively related to CEO risk-taking

incentives. Specifically, as Delta increases, managers with equity compensation have incentives to reduce the riskiness

levelof investments to maximize personal wealth. Lambert, Larcker, and Verrechia(1991) show that a risk-averse manager

with a large proportion of compensation in the form of firm equity may actually avoid risk.

2Carpenter (2000) shows that equity compensation does not strictly lead to greater risk seeking when a manager’s pay-

performance sensitivity (Delta) is high. Similarly, Ross (2004) shows that a manager’s willingness to take risks may

decrease as Delta increases.