DIFFERENTIAL RISK‐TAKING IMPLICATIONS OF PERFORMANCE INCENTIVES FROM STOCK AND STOCK OPTION HOLDINGS

• Author: Tanseli Savaşer,Elif Şişli‐Ciamarra
• Date: 01 September 2019
• DOI: http://doi.org/10.1111/jfir.12190
• Published date: 01 September 2019

The Journal of Financial Research Vol. XLII, No. 3 Pages 609–636 Fall 2019

DOI: 10.1111/jfir.12190

DIFFERENTIAL RISK‐TAKING IMPLICATIONS OF PERFORMANCE

INCENTIVES FROM STOCK AND STOCK OPTION HOLDINGS

Tanseli Savaşer

Vassar College

Elif Şişli‐Ciamarra

Stonehill College

Abstract

We study the risk‐taking implications of managerial pay‐for‐performance incentives

(delta) arising from stock and stock options separately in the United States between

1992 and 2017. The current literature assumes that each unit of delta has an equal

incentive effect on firm performance. Instead, we show that the risk‐reducing effect

of performance incentives is more pronounced for executives whose delta comes

mostly from stock holdings relative to option holdings. Accordingly, we propose a

new measure that takes into account the magnitude of delta from option holdings

relative to delta from stock holdings (source ratio). Our results show that risk taking

increases as this ratio increases.

JEL Classification: G32, G34

I. Introduction

Stock and stock option grants form a significant portion of executive pay packages in

the United States, amounting to one‐third of the total compensation of an average chief

executive officer (CEO). The main motivation for stock‐based compensation is to

provide managers with performance incentives. Once equipped with these incentives,

managers are expected to make value‐generating financial decisions. However, since

the onset of the 2007–2008 financial crisis, stock‐based compensation is under scrutiny

because of the popular belief that it leads to excessive risk taking.

The impact of performance‐sensitive pay on managerial risk taking is not

straightforward. On one hand, pay‐for‐performance incentives may induce risk taking

because projects with higher expected returns are also inherently riskier. On the other hand,

performance‐sensitive compensation may dampen risk taking because risk‐averse managers

may forgo risky projects with high expected returns to preserve their firm‐specific wealth

We are indebted for useful comments and suggestions to Tyler Hull, Lubomir Litov, Gunseli Tumer Alkan,

an anonymous referee, as well as the seminar participants at the 2017 Financial Management Association (FMA)

annual meeting in Boston; Financial Deposit Insurance Corporation (FDIC); Stonehill College, Easton;

University of Massachusetts, Boston; Middle Eastern Technical University, Turkey; Humboldt University,

Berlin; Vassar College, New York; ESSEC, Paris; and Vrije University, Amsterdam. We also thank Frances M.

Foote for editorial help.

609

© 2019 The Southern Finance Association and the Southwestern Finance Association

and human capital (Amihud and Lev 1981; Smith and Stulz 1985; Tufano 1996). Hence, for

a sufficiently risk‐averse manager, an increase in pay‐for‐performance sensitivity may

discourage risk taking (Lambert, Larcker, and Verrecchia 1991; Carpenter 2000; Hall and

Murphy 2002; Ross 2004; Lewellen 2006, Flor, Frimor, and Munk 2014, Dittmann, Yu, and

Zhang 2017). Despite the abundance of theoretical models studying the impact of

performance incentives on managerial risk taking, empirical papers examining this relation

are limited and provide mixed results. Furthermore, although the theory distinguishes

between the stock grants and stock option grants (Flor, Frimor, and Munk 2014), empirical

studies do not distinguish among the pay‐for‐performance incentives (delta) from different

pay sources.

To bridge this gap, we argue that it is necessary to account for the composition of

pay‐for‐performance sensitivity in empirical specifications. This is because performance

incentives from option holdings are coupled with risk‐taking incentives due to their convex

payoff structure, whereas performance incentives provided by stocks promise linear

payoffs and are not accompanied by similarly strong risk incentives. We hypothesize that

themarginaleffectofdeltaonfirmriskshouldbemoredepressiveforaCEOwhosedelta

comes mostly from stock holdings relative to option holdings. We propose a new metric to

assess the risk impact of performance incentives, namely, the ratio of delta provided by

accumulated option holdings to delta provided by accumulated stock holdings (source

ratio). Our proposed measure accounts for the relative importance of options in a CEO’s

pay‐for‐performance sensitivity and measures the additional impact of options on

managerial risk taking beyond the convexity effect captured by vega.

To test our hypothesis, we analyze CEO compensation in U.S. public firms between

1992 and 2017. Our sample period is long enough to capture the variability in our proposed

measure of performance incentives, that is, the source ratio (see Figures I and II in Section

IV). We measure managerial pay‐for‐performance incentives with delta, which is the change

in the dollar value of a CEO’s wealth for a 1% change in stock price (Core and Guay 1999).

We further decompose total pay‐for‐performance incentives into two components: (1) delta

generated by accumulated option holdings and (2) delta generated by accumulated stock

holdings, the ratio of which we call the “source ratio.”Because stock‐based pay also affects

the risk appetite of managers by creating pay‐for‐risk sensitivity, we explicitly control for

risk‐taking incentives in all of our analyses. Our measure of risk‐taking incentives is vega,

the change in the dollar value of a CEO’s wealth for a 0.01 change in the annualized

standard deviation of stock returns (Core and Guay 1999). We measure firm risk using

realized stock return volatility, which is one of the standard measures of risk in the

compensation literature (e.g., Guay 1999; Low 2009; Hayes, Lemmon, and Qiu 2012).

We first separate the firms in our sample into quartiles according to the source

ratio. We find that the relation between delta and firm risk is negative for firms that provide

performance incentives mostly by stock grants. As the relative magnitude of delta from

option holdings increases, the risk‐dampening effect of delta disappears and the relation

between firm risk and delta becomes positive. We also directly estimate the relation

between the source ratio and managerial risk taking. The results are consistent with those

from the quartiles approach: an increase in the source ratio is associated with an increase in

firm risk. We confirm the robustness of this result with various estimation methods:

various fixed effects, instrumental variables, and Arellano–Bond (1991) estimations.

610 The Journal of Financial Research