CEO Incentives and Corporate Innovation

• DOI: http://doi.org/10.1111/fire.12144
• Published date: 01 May 2018
• Date: 01 May 2018

The Financial Review 53 (2018) 255–300

CEO Incentives and Corporate Innovation

Tu Nguyen

La TrobeUniversity

Abstract

Using scaled wealth-performance sensitivity as my measure of Chief Executive Officer

(CEO) incentives, and utilizing cross-sectional variations in industry innovativeness, product

market competition and firms’ degree of exposure to the market for corporate control for

identification purposes, I find that higher long-term incentives that stem from CEO holdings

of unvested options are associated with greater subsequent corporate innovation in innova-

tive industries, competitive product markets, and firms more exposed to the threat of hostile

takeovers, that is, exactly where incentivizing innovation is a matter of necessity. I address

the endogeneity concerns with systems of simultaneous equations estimated using three-stage

least squares. A possible channel for the observed relation between unvested options-based

incentives and subsequent corporate innovation isthat these incentives encourage managers to

undertake riskier projects to achieve long-term economic benefits.

Keywords: long-term incentives,vesting restrictions, corporate innovation, patents, citations

JEL Classifications: G34, O31

1. Introduction

Holmstrom (1989) refers to innovation projects as risky, unpredictable, long-

term, multistage, labor-intensive and idiosyncratic as they involve the exploration

Correspondingauthor: Department of Economics and Finance, La Trobe University, Bundoora, VIC 3086,

Australia; Phone: +61 3 9479 5608; E-mail: tu.nguyen@latrobe.edu.au.

I am grateful for helpful comments and suggestions from Srinivasan Krishnamurthy (the Editor) and an

anonymous referee. I also thank Jing Zhao, Sandy Suardi, and participants at the National Sun Yat-Sen

University’sSecurities and Financial Markets Conference 2016. All errors remain my own.

C2018 The Eastern Finance Association 255

256 T. Nguyen/The Financial Review 53 (2018) 255–300

of new untested approaches. Therefore, the agency costs associated with innova-

tion is likely to be high and designing incentive schemes to motivate innovation

is particularly demanding. Manso (2011) models the innovation process explicitly

and contrasts incentive schemes that motivate the exploration of new possibilities

and the exploitation of old certainties. Since exploration often involves sacrificing

short-term performance in order to achieve greater long-term success, Manso (2011)

shows that the optimal contract that motivates exploration exhibits tolerance for

failure in the short term and rewards long-term success. This empirical implication

is consistent with Lambert (1986), Holmstrom (1989), and Fudenberg, Holmstrom

and Milgrom (1990), among others. Taken together, these theoretical studies, though

adopting different approaches and focusing on different aspects of innovationactivity,

highlight the importance of properly designed managerial incentives, in particular,

the importance of the long-term focus of these incentives in motivating innovation.

Furthermore, among the sources of equity-based incentives, options havedifferent in-

centive value than stock. Specifically, owing to their convexpayoff function, options

improve incentives to risk-averse agents. Therefore, stock options, as compared with

stock, may encourage managers to accept riskier projects (Lambert, 1986; Jackson

and Lazear, 1991; Feltham and Wu, 2001; Manso, 2011)—a categorythat innovation

projects clearly belong to.

In this study, I empirically examine two research questions that are consistent

with the empirical implications of Manso (2011) in the context of executive compen-

sation and motivating innovation. First, I investigate whether long-term incentives

that stem from the Chief Executive Officer(CEO)’s holdings of unvested options and

restricted stock are higher in firms operating in an environment where opportunities

for innovation abound and/or the needs for innovation are more urgent—innovative

industries and competitive product markets. Second, I examine whether these incen-

tives, in particular,unvested options-based incentives spur corporate innovation. I use

the number of patents granted to a firm and the number of citations received by each

patent as my measures of corporate innovation. For measures of CEO incentives, I

use the scaled wealth-performance sensitivity (the dollar change in the CEO’s wealth

for a percent change in firm value, scaled by annual total pay), advocated by Edmans,

Gabaix and Landier (2009).

With regard to the first question, I document that long-term incentives, in par-

ticular, unvested options-based incentives are significantly higher among firms in

innovative industries as well as among firms facing strong product market competi-

tion. Regarding the second question, the results from the baseline specification show

that a higher level of executive long-term incentives is associated with not only a

higher number of patents but also a higher impact of patents generated in the follow-

ing year. And the most detailed breakdown of incentive measures suggests that this

positive relation is driven by the CEO’s holdings of unvested options. Specifically,

for my sample of 13,279 firm-year observations over the 1992–2006 period, a one-

standard-deviation increase in executive incentives stemming from unvested option

holdings increases the number of year-adjusted patents in the following year by 2.0%

T. Nguyen/The Financial Review 53 (2018) 255–300 257

of its mean and the number of year-adjusted citations per patent in the following year

by 2.9% of its mean. Moreover, utilizing cross-sectional variations in industry inno-

vativeness and product market competition, I document that the relation is observed

only within innovative industries and industries with strong product market compe-

tition, that is, exactly where incentivizing innovation is a matter of necessity. While

the bulk of the CEO’s incentives can be attributed to his/her holdings of unrestricted

stock, short-term (vested) incentives do not appear to be associated with greater inno-

vation output. The evidence is thus consistent with my hypothesis that not all sources

of equity-based incentives are equal when it comes to motivating innovation. The

results also suggest that a traditional pay-for-performance sensitivity measure that

lumps together all types of equity-based compensation would likely fail to capture

any motivating effectof equity-based incentives on innovation. On a slightly different

note, utilizing cross-sectional differences in firms’ degree of exposure to the market

for corporate control, I show evidence consistent with the interpretation that firms

treat the adoption of antitakeover provisions and the design of executive incentive

compensation as substitutes when choosing their corporate governance structure to

motivate exploration and innovation. Overall, I view tests conducted within innova-

tive industries or competitive product markets or among firms more exposed to the

threat of hostile takeovers as more powerful ways to identify the effect of executive

incentives on innovation.

I fully recognize that establishing causality is empirically challenging due to

the potential endogeneity problems. The relation between executive incentives and

corporate innovation is likely determined in a dynamic process. Executive incentives

are structured to encourage value-enhancing efforts which, in turn, get remunerated

through incentive compensation. Alternatively, the relation could be spurious because

some unobserved factors simultaneously affect corporate innovation and executive

incentives. Therefore, I have employed various methods to address the endogeneity

concerns. In particular, I use three-stage least squares (3SLS) to estimate systems

of simultaneous equations that explicitly allow for the joint determination of next

year innovation output and current CEO incentives. I continue to find that unvested

options-based incentives motivate innovation.

Additionally, I show evidence suggesting that a possible channel for the ob-

served relation between unvested options-based incentives and subsequent corporate

innovation is that these incentives provide managers with greater incentives for risk

taking, as reflected in higher future firm risk. The increase in firm risk is achieved, in

part, through such avenues as implementing riskier corporate policies, in particular,

undertaking riskier investment projects, among which are innovation projects that

result in higher innovation output.

This paper is closely related to studies that examine the relation between in-

centives and corporate innovation (see, e.g., Clinch, 1991; Kole, 1997; Yanadori

and Marler, 2006; Lerner and Wulf, 2007; Sheikh, 2012; Ederer and Manso, 2013;

Gopalan, Milbourn, Song and Thakor, 2014). I contributeto this strand of literature by

employing a novel comprehensive incentive measure (Edmans, Gabaix and Landier,