Managerial effect or firm effect: Evidence from the private debt market

• Author: Iftekhar Hasan,Yun Zhu,Bill B. Francis
• Published date: 01 February 2020
• DOI: http://doi.org/10.1111/fire.12196
• Date: 01 February 2020

DOI: 10.1111/fire.12196

ORIGINAL ARTICLE

Managerial effect or firm effect: Evidence from the

privatedebtmarket

Bill B. Francis1Iftekhar Hasan2,3 Yun Zhu 4

1Lally School of Management, Rensselaer

Polytechnic Institute, Troy,New York

2Gabelli School of Business, FordhamUniversity,

New York,New York

3Bank of Finland, Helsinki, Finland

4TobinCollege of Business, St. John's University,

New York,New York

Correspondence

IftekharHasan, Gabelli School of Business, Ford-

hamUniversity, New York,NY 10023.

Bankof Finland, P.O.Box 160, FI 00101,

Helsinki,Finland.

Email:ihasan@fordham.edu

Abstract

This paper provides evidence that the managerial effect is a key

determinant of firms’ cost of capital, in the context of private debt

contracting. Applying the novel empirical method developed by an

earlier study to a large sample that tracks the job movement of top

managers, we find that the managerial effect is a critical and signifi-

cant factor that explains a large part of the variation in loan contract

terms more accurately than firm fixed effects. Additional evidence

shows that banks “follow” managers when theychange jobs and offer

loan contracts with preferential terms to their new firms.

KEYWORDS

firm fixed effect, loan contract,managerial effect

JEL CLASSIFICATIONS

G21, G31, J24

1INTRODUCTION

Managers playa critical role in corporate decision making. The upper-echelon theory suggests that a corporate leader's

managerial characteristics and experience have a profound influence on organizational behavior during the strate-

gic decision-making process and, consequently, affect the growth and performance of the organization (Hambrick &

Mason, 1984). Hambrick (2007) further argues that a leader's cognitive, social, and psychological characteristics can

be strong predictors of strategic behavior.

Influenced by this pioneering theory,a stream of work links the heterogeneity of managerial characteristics, back-

ground, experience, and abilities to corporate behavior,policy, and performance.1Although empirical studies provide

abundant evidence to support the argument that managerial traits are influential in firm performance, the aggregated

1Forexample, MBA degree holders are generally correlated with aggressive strategies, while being born into an early birth cohort is related to more conserva-

tivedecision making (Bertrand & Schoar, 2003). Managers with military or financial backgrounds display conservative or precise characteristics, respectively

(Bamber,Jiang, & Wang, 2010). Nicolosi and Yore (2015) show that a CEO's personal life restructuring, such as marriage and divorce, affects corporate level

deal-making activity and overallfirm riskiness. A number of works further explore the unobserved (at least not explicitly observable to outsiders) attributes

that relate to the managerial effect, including overconfidence (Dezs˝

o & Ross, 2012; Galasso & Simcoe, 2011; Hirshleifer,Low, & Teoh, 2012; Kolasinski & Li,

2013;Malmendier & Tate, 2005; Malmendier,Tate, & Yan,2011), risk aversion (Coles & Li, 2018; Graham, Harvey, & Puri, 2013), media exposure (Malmendier

& Tate,2009), social capital (Javakhadze, Ferris, & French, 2016), and so on. Dezs˝

o and Ross (2012), for example,find that banks grant loans at lower rates to

firms in which managers display optimism by delayingexecution of their vested options. Bunkanwanicha, Fan, and Wiwattanakantang (2013) show that the

marriageof a member of the controlling family adds value to public corporations, and Roussanov and Savor (2014) show that firms with single CEOs experience

higherstock volatility.

Financial Review.2020;55:25–59. wileyonlinelibrary.com/journal/fire c

2019 The Eastern Finance Association 25

26 FRANCIS ET AL.

effect of the managerial impact on the decision-making process has barely been discussed. This aggregated effect,

drawn from managers’ heterogeneity and commonly known as managerial style or the managerial effect, captures the

observable and unobservable time-invariant attributes and nature of each individual manager. It incorporates, but is

not limited to, traits such as culturaland religious background, risk preference, and integrity. A manager with a specific

managerial effect displays consistent characteristicways of decision making, which makes it extraordinarily important

in explaining variations in corporate policies and performance. Forexample, Bertrand and Schoar (2003) find that the

managerial effect is powerful in explaining unknown parts of firm performance and firm decisions and remains per-

sistent for the same manager across firms. The managerial effect associated with better firm performance is usually

present in firms with better governance and is better compensated (Graham, Li, & Qiu, 2012). Dyreng, Hanlon, and

Maydew (2010) and Bamber et al. (2010) relate the managerial effect to firms’ tax avoidance and voluntary financial

disclosure choices and find it to be significant.2

Giventhe importance of the manager's role in corporate decision making, this paper aims to explore the existence of

managerial effect in the context of private loan contracts,to contribute to the literature on if, and how, the managerial

effect is valued by the lenders, and to provide new evidenceon how private debt is priced.

We choose the private loan contract, as it constitutes a perfect setting for investigation of the managerial effect.

First, lenders are information savvy. They will value and incorporate the managerial effect into loan contractingif it

provides additional information. Such information could be drawn from a manager's credit preferences through prior

lending activities. For example, Cronqvist, Makhija, and Yonker (2012) find that a firm's leverage choice is consistent

with the manager's personal choice of debt in her primary home purchase. Second, a top manager's business network

facilitates the efficiency of information flow and mitigates agency conflict (Engelberg, Gao, & Parsons,2013; Hochberg,

Ljungqvist, & Lu, 2007; Kuhnen, 2009). Therefore, when a manager changes her job, her credit preference and prior

bank connections provide additional information for contracts to be extendedto the new firm. Although the incentive

for lenders to incorporateand value the managerial effect and thereby reduce information asymmetry is apparent, this

topic has received scant prior attention in the literature. Thispaper aims to fill the gap.

Empirically,we take a two-step approach. First, we examine whether the managerial effect is a key determinant of a

loan contractand if such an effect is large enough to explain a significant amount of variation in contract terms; second,

we look at whether lenders observe and value the managerial effect and incorporate this information into their debt

contracts.

Using a large data set that tracks the movement of CEOs and CFOs across the Standard & Poor's(S&P) 1500 firms

from 1992 to 2010 and the novel empirical approach developed by Abowd, Kramarz, and Margolis (1999), we find

that the managerial effect is one of the most important factors in explaining a substantial amount of variation in loan

contract terms. For example, the loan-spread-related managerial effect has a standard deviation of 77 basis points.

Regardingits significance, the explained variation from the managerial effect is greater than that from firm fixed effects

and, for some loan terms, is greater than firm controls and year fixed effects combined. Forinstance, the loan-spread-

related managerial effect explains21.7% of the variation in loan spread, which is considerably more than that explained

byfirm fixed effects (6.6%). Likewise, the covenant-related managerial effect explains 29.5% of the variation in number

of covenants, which is more than that explained by firm fixed effects (12.1%) and that explainedby the combination

of firm control and year fixed effects (24.0%). Tostrengthen the findings in a strictly exogenous setting, we follow Fee,

Hadlock, and Pierce (2013) and compare changes in loan terms around CEOs’ exogenousdeaths. Using a difference-in-

differences (DiD) approach, our results are robust and consistent with the main argument.

Tofurther examine whether lenders are aware of the managerial effect when they offer a debt contract to a moving

manager's new firm, we construct a list for each such firm that contains all new lenders coming from the moving man-

ager's prior connections and then compare it to a matching list. The matching list is created by randomly assigning the

moving manager to a firm without a moving manager.We find that 41% of moving managers “bring in” their previously

connected banks to their new firms, compared with only 1.06% with a simple matching approach, and 2.15% with a

2On the other hand, Feeet al. (2013) raise an objection to the existence and significance of the managerial effect. They study exogenous CEO turnovers and

findthat a firm's policy does not display abnormally pronounced variation thereafter.

FRANCIS ET AL.27

propensity-score-matching (PSM) approach. In addition, new managers’ “brought-in” banks offer loan contracts with

more favorable terms, including lower spreads, fewer covenants, and fewer requirements for collateral,among other

factors. This evidence directly supports the argument that managers’ specific information is valued bylenders and that

previous banking relationships reduce information asymmetry.3

Weperform several robustness tests with alternative samples to address sample selection concerns. We find consis-

tent results when the sample is limited to only moving CEOs or is extended to top management teams. Our results are

still robust when we include firms with multiple moving managers in the same yearand isolate the effect of each mover.

We also use two-stage least squares (2SLS) approach to control for the endogeneity of loan price, debt maturity,and

collateral requirement (Berger,Espinosa-Vega, Scott Frame, & Miller,2005, 2011a; Berger, Scott Frame, & Ioannidou,

2011b). The results remain unaffected.

The paper makes several contributions to the literature and provides strong evidencethat the managerial effect

exists in the context of loan contractsand is greatly valued by lenders. To the best of our knowledge, this study is the

first attempt to directly examine the overallimportance of the managerial effect in the context of financial contracts.

Having knowledge of managers’ lending activities in their previous firms is critical for lenders when determining the

terms of a loan contract. Weshow directly that banks value the managerial effect, “follow” managers’ moves, and offer

loan contractsto their new firms with preferential terms, such as lower spreads and fewer covenants, implying a reduc-

tion in information asymmetry.

Second, our work provides new evidence on how private debt is priced. A manager's prior business connections

with banks are valuable to the new firm; not only do prior related banks “follow” the manager's move, but they also

offer contractswith preferential terms that reflect the information content, meaning that a beneficial relationship with

lenders can be created with the hiring of new managers. This evidence significantly enriches the literature on credit

rationing, relationship lending, and lending to opaque firms (Berger, Klapper, & Udell, 2001; Berger & Udell, 1992;

Bharath, Dahiya, Saunders, & Srinivasan, 2011).

Third, this paper adds to the growing literature on the managerial effect. By connecting the latter with the cost of

capital, we identify a mechanism through which the managerial effect directly relates to a firm's decision making and

performance. The paper also fits well into the debate on the significance of the managerial effect and provides strong

evidence that (a) banks “follow” moving managers and (b) CEOs’ exogenous deaths lead to changes in loan contract

terms.

The rest of paper is organized as follows: in Section 2, we describe the method and construct a sample to track the

managerial effect across firms. Section 3 tests our main hypothesis on the existenceand significance of a loan contract-

related managerial effect. In Section 4, we examine whether banks value and incorporate information gained via the

managerial effect into their contracts. Section 5 provides additional tests that link the loan contract-related manage-

rial effect to manager characteristics, personal background, and compensation. Robustness checks are presented in

Section 6, and Section 7 concludes the paper.

2METHOD, SAMPLE, AND MODEL

2.1 Method

To quantify the managerial effect, we face the challenge of separating the managerial fixed effect from firm fixed

effects, which is possible only if the two are not perfectly correlated. The traditional approach4is to use a sample that

tracks only moving managers and the firms that have employedthem. This approach faces two shortcomings: limited

sample size and limited generalizability of the results. Also, it is biased toward movers, who may be of higher quality

3Notethat we use the “brought-in” concept to define the manager-lender connection. Such connections are created with managers’ personal links to lenders

throughprior lending relationships. The commonly discussed relationship lending in the literature refers to the borrower-lender connection at the firm level.

4See,for example, Bertrand and Schoar (2003).