Can Benford's Law explain CEO pay?

• Author: Shibashish Mukherjee
• Date: 01 March 2018
• DOI: http://doi.org/10.1111/corg.12195
• Published date: 01 March 2018

ORIGINAL ARTICLE

Can Benford's Law explain CEO pay?

Shibashish Mukherjee

University of Groningen

Correspondence

Mukherjee Shibashish, University of

Groningen, Faculty of Economics and

Business, Department of Accounting, Room

No. 5411‐0817, Nettelbosje 2, 9747AE

Groningen, The Netherlands.

Email: shibashish.mukherjee@rug.nl

JEL Classification: G34; M4; M12; M21

Abstract

Manuscript Type Empirical

Research Question/Issue This study applies the statistical properties of Benford's Law to

CEO pay. Benford's “Law”states that in an unbiased dataset, the first digit values are usually

unequally allocated when considering the logical expectations of equal distribution. In this study

we question whether the striking empirical properties of Benford's Law could be used to analyze

the negotiating power and preferences of CEOs. We argue that performance‐based or market‐

determined compensations should follow Benford's Law, demonstrating no direct negotiation

by the CEOs. Conversely, deviation from Benford's Law could reveal CEO negotiating power or

even preference.

Research Findings/Insights Our analysis shows that market‐determined “Option Fair

Value”(the dollar value of stock options when exercised) conforms closely to Benford's Law, as

opposed to “Salary”, which is fully negotiated. “Bonus”,“Option Award”, and “Total Compensa-

tion”are generally also largely consistent with Benford's Law, but with some exceptions. We

interpret these exceptions as negotiation by the CEOs. Surprisingly, we found that CEOs prefer

to be paid in round figure values, especially “5”. We use Benford's Law to study the negotiating

powers of CEOs vs. that of other executives. Finally, we compare the negotiating tactics of CEOs

before and after SOX and analyze the impact of firm size on their compensation.

Theoretical/Academic Implications This study introduces Benford's Law and its applica-

tions within the corporate governance literature.

Practitioner/Policy Implications This method could be used by academics, industry and

regulators to uncover compensation patterns within large business departments and/or organiza-

tions or even entire industry segments.

KEYWORDS

Corporate Governance, Benford's Law, Executive Compensation, Negotiating Power, Preference

1|INTRODUCTION

CEO pay has emerged as a prolific field of theoretical and empirical

research (Bebchuk, Cremers, & Peyer, 2011; Bebchuk, Fried, & Walker,

2002; Jensen & Meckling, 1976; Jensen & Murphy, 1990b; Murphy,

1999), and because of the diverse research methods offered to study

the many aspects of CEO negotiations, the compensation literature

has benefitedfrom being on the frontiers ofthis domain (Boyd, Haynes,

& Zona, 2011; McNulty, Zattoni, & Douglas, 2013; Shleifer & Vishny,

1997; Vives,2006). Despite the intense interest in the theme, however,

several lacunae still remain in our understanding of CEO pay. Chief

among them are the indicators that could detect the preferences of

CEOs and the intensity with which they negotiate their compensation

with boards.In this paper, we applya novel method that allowsus to dis-

tinguish market‐determined pay from negotiated CEO compensation.

This distinction is important because the more CEOs negotiate their

pay, the more is revealed about the power balance between this group of

professionals and the compensation committees of company boards

(Adams, Almeida,& Ferreira, 2005; Bebchuk et al., 2002; Finkelstein, 1992).

Executive compensation—which in all US‐listed firms is partly per-

formance‐based—depends on predetermined payment arrangements

as well as a number of accounting and market‐based performance indi-

cators. A CEO can negotiate the fixed part of his/her pay, but the real-

ization of performance‐linked compensation is ideally dependent on

Received: 13 March 2015 Revised: 23 March 2016 Accepted: 26 December 2016

DOI: 10.1111/corg.12195

Corp Govern Int Rev. 2018;26:143–156. © 2017 John Wiley & Sons Ltdwileyonlinelibrary.com/journal/corg 143