CEO investment of deferred compensation plans and firm performance

• Author: Giuliano Curatola,Giulia Fantini,Stefano Colonnello,Domenico Rocco Cambrea
• Published date: 01 July 2019
• Date: 01 July 2019
• DOI: http://doi.org/10.1111/jbfa.12382

DOI: 10.1111/jbfa.12382

CEO investment of deferred compensation plans

and firm performance

Domenico Rocco Cambrea1Stefano Colonnello2

Giuliano Curatola3,4 Giulia Fantini5

1Bocconi University, Italy

2Otto-von-GuerickeUniversity Magdeburg and

Halle Institute for Economic Research (IWH),

Kleine Märkerstraße8, D-06108 Halle (Saale),

Germany

3University of Siena, Italy

4Research Center SAFE, Germany

5School of Management, Swansea University,

United Kingdom

Correspondence

StefanoColonnello, Otto-von-Guericke

UniversityMagdeburg and Halle Institute

forEconomic Research (IWH), Kleine Märk-

erstraße8, D-06108 Halle (Saale), Germany.

Email:stefano.colonnello@iwh-halle.de

Fundinginformation

DeutscherAkademischer Austauschdienst

(DAAD),Research Grants – Short-TermGrants,

2015(57130097)

Abstract

We study how US chief executive officers (CEOs) invest their

deferred compensation plans depending on the firm’s profitability.

By looking at the correlation between the CEO’s return on these

plans and the firm’s stock return, we show that deferred compen-

sation is to a large extent invested in the company equity in good

times and divested from it in bad times. The divestment from com-

pany equity in bad times arguably reflects CEOs’ incentive to aban-

don the firm and to invest in alternative instruments to preservethe

value of their deferred compensation plans. This result suggests that

the incentive alignment effects of deferred compensation crucially

depend on the firm’s health status.

KEYWORDS

corporate distress, deferred compensation, executivecompensation

JEL CLASSIFICATION

G32, G34

1INTRODUCTION

Top executives of US public firms receive an important proportion of their compensation in the form of retirement

benefits. Such benefits are akin to debt-likeclaims on the firm and are often called inside debt. As pointed out by Jensen

and Meckling (1976) and Edmans and Liu (2011), inside debt can align the incentives of managers and creditors, thus

making the former more conservative.1

Yet, several studies suggest a more nuanced view of inside debt incentives.Inside debt is composed of pensions

(rank-and-file plans and supplemental executiveretirement plans) and deferred compensation. Anantharaman, Fang,

and Gong (2014) show that the incentive alignment effect of inside debt is driven by supplemental executiveretire-

ment plans—i.e., the inside debt component most exposedto default risk. In contrast, rank-and-file plans are protected

in bankruptcy, and deferred compensation plans, though often formally at risk in bankruptcy, allow some flexibility

1Severalstudies provide evidence consistent with this view. See, e.g., Sundaram and Yermack(2007) and Cassel, Huang, Sanchez, and Stuart (2012).

944 c

2019 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/jbfa JBus Fin Acc. 2019;46:944–976.

CAMBREA ET AL.945

in the schedule of withdrawals, which can predate the retirement date (and the maturity date of outstanding debt).2

Interestingly, deferred compensation plans also allow substantial flexibilityin the investment strategy, which has the

potential to affect managerial risk-taking incentives.

This last aspect has received little attention. An exceptionis Jackson and Honigsberg (2014), who show that a sub-

stantial proportion of deferred compensation is invested in the company stock. Hence, deferred compensation may

providemanagers with an equity-like payoff and intensify, ratherthan diminish, managerial risk-taking incentives. Jack-

son and Honigsberg (2014) analyze the average investment strategy of deferred compensation and do not consider

the possibility that this strategy changes over time. In contrast, we focus on the time-varying investment strategy of

deferred compensation. In particular, we analyze theoretically and empirically the manager’s incentivesto reallocate

his/her deferred compensation away from company stock when the firm enters a period of low expectedprofitability.

We then discuss the possible implications of such a reallocation strategy for firm risk.

We start by developing a theoretical framework in which the CEO endogenously decides the allocation of his/her

deferred compensation and the levelof firm risk. When the firm’s expected profitability is low, the CEO reallocates the

deferred compensation away from companystock toward assets with higher expected payoffs.

Wetest this prediction on a large sample of US public firms over the period 2006–2015 by analyzing CEOs’ compen-

sationpackages. Given that current disclosure rules do not require companies to reveal how CEOs’ deferred compensa-

tion plans are invested,we measure the exposure of deferred compensation to company stock by means of the correla-

tion between deferred compensation returns and stock returns. As a preliminary step, we substantiate the suggestive

evidence of Jackson and Honigsberg (2014) by showing that deferred compensation is indeed significantly linked to

company stocks, as witnessed by the positive, large, and statistically significant correlation exhibitedby a substantial

fraction of CEOs in our sample.

We then focus on the time-varying nature of the correlation between CEO deferred compensation returns and

stock returns, which we use as a proxy for the CEO’s investment strategyover time. We show that such a correlation

declines significantly in bad times (i.e., distressed firm-years), in line with CEOs divesting from the firm stock. We con-

firm this finding using the recent financial crisis as a plausibly exogenousshock to the firm’s distress risk. The decline in

the correlation is likely to reflect CEOs’ desireto abandon the firm during bad times.

Moreover,we do not find evidence that CEOs use private information to time the market by means of their deferred

compensation investmentstrategy. However, CEOs appear to be quickerto react to worsened firm conditions through

the asset allocation of deferred compensation plans than through trading on traditional stock incentives, possibly

because of the more intense scrutiny the latter receive from investors.

Using our theoretical setting, we then discuss the possible implications of CEO investment strategy of deferred

compensation for firm risk. By divesting deferred compensation awayfrom firm equity, the CEO can limit losses in case

of default. In bad times, the reallocation strategy of deferred compensation works as an insurance that allows the CEO

to adopt a sort of gambling for resurrection strategy, increasing the value of equity through higher cash flow volatil-

ity without suffering large expected compensation losses. The model thus suggests that deferred compensation may

induce the CEO to takeon more risk precisely when creditors need more prudent behavior, the effect being stronger if

deferred compensation is large.

Early withdrawals are an alternative/complementaryway to limit expected losses on deferred compensation (Jack-

son & Honigsberg, 2014) and can therefore lead to increased managerial risk-taking incentives too. We show empiri-

cally that in bad times CEOs tend to alter the investment strategyof deferred compensation, rather than withdrawing

part of their deferred compensation. This is possibly (i) because of the tax penalty on early withdrawalsand (ii) because

withdrawals are observable and often criticized by the media, especially when the amount of money involvedis sub-

stantial, while the investment strategy of deferred compensation is harder to observe. In other words, changing the

investmentstrategy of deferred compensation may be associated with lower monetary and reputational costs, making

it more desirable for CEOs than early withdrawals.

2Seereferences in Anantharaman et al. (2014).

946 CAMBREA ET AL.

It is worth noting that indirect equity incentives (i.e., those coming from deferred compensation plans) differ from

traditional direct equity incentives (i.e., those coming from the part of CEO’s wealth directly investedin equity). I n the

model, the CEO reduces the fraction of deferred compensation investedin company stock in bad times and, as a result,

takes more risk. In other words, our theoretical setting predicts a negative relationship between firm risk and indirect

equityincentives. Meanwhile, the impact of direct equity holdings depends on the interaction of two opposite channels.

When the number of directly held shares increases, the CEO has an incentive to take more risk to benefit from the

positive effect on the value of equity.At the same time, a surge in firm risk increases the probability of default, which

reduces the value of deferred compensation. Therefore, the relationship between direct equity incentives and firm risk

hinges on the overalleffect of firm risk on the value of total CEO compensation. When deferredcompensation is over-

invested in company equity,the former channel prevails, and direct equity holdings have a positive effect on firm risk.

Otherwise, the latter channel prevails, and direct equity holdings havea negative effect on firm risk. In the same spirit,

Carpenter (2000) and Ross (2004) show that convexcompensation schemes may have ambiguous effects on CEO risk-

taking incentives. We illustrate that this ambiguity can also be generated by flexibility in the investmentstrategy of

deferred compensation.

2LITERATURE REVIEW

Asubstantial body of work provides evidence compatible with the risk-reducing role of inside debt suggested by Jensen

and Meckling (1976) and Edmans and Liu (2011). Sundaram and Yermack(2007) find a negative relationship between

the ratio of inside debt to inside equity and default risk. Wei and Yermack(2011) show that, after firms’ initial disclo-

sure of top executive retirement plans, bond pricesrise while stock prices decrease. Phan (2014) provides consistent

evidence looking at firms’ mergers and acquisitions (M&A) activity, also documenting that acquiring firms adjust the

weight of inside debt relative to equity holdings following M&As to adjust to the new capital structure. Liu, Mauer,

and Zhang (2014) illustrate that inside debt helps protect creditors by favoring cash hoarding behavior.Li, Rhee, and

Shen (2018) illustrate that firms whose CEOs hold large inside debt holdings tend to issue less convertibledebt, which

aligns with a risk-mitigating role of inside debt provided that convertibles are used by firms to curb risk-shifting. Sri-

vastav,Armitage, and Hagendorff (2014) focus on the banking sector and document that inside debt limits managerial

risk-shifting through a reduction of incentives to divert cash to shareholders. Cassel et al. (2012) report evidence of a

negative relationship between executives’inside debt holdings and the volatility of stock returns.

Anantharaman et al. (2014), however,show that inside debt is effective at reducing the cost of private loans only

when it is actually exposedto default risk. Colonnello, Curatola, and Hoang (2017) extend this result to public debt and

illustrate that low-seniority debt can interact with equity incentives to make CEOs less conservative. Such an unin-

tended increase in managerial risk-taking is concentrated in bad times (Inderst & Pfeil, 2013). Goh and Li (2015) doc-

ument that CEO pensions are unlikely to qualify as actual inside debt in the UK context, but ratheras substitutes for

other performance-sensitive compensation items. Jackson and Honigsberg (2014) question the incentive-alignment

role of deferred compensation plans by documenting that they are on averageheavily invested in firm equity and to a

large extent withdrawnby executives after they leave the firm. Cen and Doukas (2017) look at firm risk-taking and the

CEO’s personal returns on deferred compensation plans, but rather than focusing on their correlation with the firm

stock across different states of the world, the authors examine their unconditional volatility to infer the CEO’s risk

preferences. We complement this literature by studying how the CEO’s exposure to firm risk through deferred com-

pensation changes over time depending on his/her personal portfolio choices. We then discuss how this time-varying

exposure of the CEO to the firm canaffect his/her risk-taking incentives.

Our paper also speaks to the literature on the hedging behavior of executivesin relation to their company equity

holdings. Gao (2010) documents that optimal pay-performance sensitivity decreases in the hedging costs faced by

the CEO. Bettis, Bizjak, and Lemmon (2001) find that managers use derivative instruments such as zero-cost collars

and equity swaps to hedge their positions. Anderson and Puleo (2016) show that managers that hedge themselves by