Disentangling managerial incentives from a dynamic perspective: The role of stock grants

• Author: Ngo Van Long,Amal Hili,Didier Laussel
• Published date: 01 December 2017
• DOI: http://doi.org/10.1111/1468-0106.12243
• Date: 01 December 2017

ORIGINAL MANUSCRIPT

Disentangling managerial incentives from a dynamic

perspective: The role of stock grants

Amal Hili

1

| Didier Laussel

1

| Ngo Van Long

2,3

1

Department of Economics, Aix-Marseille

University, Provence, France

2

Department of Economics, McGill University,

Montreal, Canada

3

Tasmanian School of Business and Economics,

University of Tasmania, Hobart, Tasmania

Correspondence

Ngo Van Long, Department of Economics,McGill

University,Montreal, QC H3A2T7, Canada.

Email: ngo.long@mcgill.ca

Abstract

We analyse the optimal contract between a risk-averse

manager and the initial shareholders in a two-period

model where the manager’s investment effort, carried out

in period 1, and his or her current effort, carried out in

period 2, both impact the second-period profit, so that it

may be difficult to disentangle the incentives for these

two types of effort. We show that stock grants play dif-

ferent roles according to whether the signal of investment

effort is less noisy, or noisier, than that of current effort.

We determine simultaneously the optimal stock grants

and the optimal restrictions on sales of shares.

1|INTRODUCTION

Stock grants are pervasive. In several OECD countries and especially in the United States, they con-

stitute a considerable share of the executive compensation package since the 1950s. The empirical

evidence indicates that the use of stock grants has increased substantially since the 1980s

(Morgenson, 1998; Murphy, 1999; Hall & Murphy, 2003). Indeed, according to Frydman and Saks

(2010), stock grants represent a greater part of total executives’pay, reaching more than 35% of total

compensation in 2005. Moreover, Sabadish and Mishel (2013) estimated that in 2012 the stock

grants to US chief executive officers (CEOs) were 26.5 times greater than in 1978. In addition, the

proportion of US executives owning stocks rose from 57% in 1980 to almost 90% in 1994 (Hall &

Liebman, 1998).

We show in this paper that stock grants, which are pervasive, do have a role to play in the

design of managerial optimal incentive schemes when firms’profits in any period are impacted by

the efforts exerted by its manager(s) in earlier periods. The performance of a firm is, indeed, gener-

ally the outcome of both: (i) the efforts its managers undertake in the current period, for instance, to

control its current operating costs and to optimize its pricing scheme; and (ii) the efforts they

(or their predecessors) have exerted in previous periods, for example, to invest in new technologies,

to explore new markets and to hire more efficient employees. The distinction between those two

types of efforts has been raised in the sharecropping literature (Banerjee & Ghatak, 2004). Indeed,

Received: 26 July 2017 Accepted: 18 August 2017

DOI: 10.1111/1468-0106.12243

Pac Econ Rev. 2017;22:743–771. wileyonlinelibrary.com/journal/paer © 2017 John Wiley & Sons Australia, Ltd 743

Banerjee and Ghatak (2004) consider a two-period tenancy model where the tenant farmer chooses

an unobservable action that enhances the probability of high output in the next period. This action is

termed an ‘investment effort’as its effect on profit is realized with a one-period lag (e.g. developing

a new environmentally-friendly technology or acquiring certified green labels). This type of effort is

distinct from a current action that raises the probability of high output in the current period (e.g. the

operational efforts in farming and irrigating).

In our model, the managers’two types of efforts are not perfectly observable and, generally, the

degrees of precision with which they can be inferred differ, so that different incentives should be

ideally given to them. Obviously, a remuneration scheme in which the managers only receive cash

payments contingent on current profits would not do the trick.

The present paper shows that stock grants are part of the optimal solution to this problem. We

distinguish two types of efforts: an investment effort that is carried out in in period 1 and a second-

period current effort, which both impact period 2 profits.

1

We assume that a signal about the effect

of the former is observed at the beginning of period 2 and reflected in the stock price. We then posit

that the degrees of ‘precision’with which efforts can be inferred differ across the two types of effort,

so that different incentives should be ideally given to investment and current efforts, which both

impact the second-period profit. Finally, we endogenize the firm’s choice of restrictions on the man-

ager’s sales of shares; that is, the optimal contract also specifies the proportion of shares granted to

the manager in period 1 which he or she is allowed to sell at the beginning of period 2. We show

that the optimal two-period contract depends on whether investment effort can be inferred ‘more

precisely’or ‘less precisely’than current effort. In the former case, granting unrestricted shares to

the manager allows the firm to provide a greater incentive to invest. In the latter case, where giving

less incentive to investment effort would be ideally required but is not feasible, restricted stock

grants are a commitment device needed for enforcing a contract in which the same average incentive

is optimally given to both types of effort.

Our paper is related to a number of theoretical papers on stock grants and stock options.

Acharya, John, and Sundaram (2000) present a two-period model where the principal (the owner of

the firm) offers the agent (the manager) an initial quantity of αcall options on the firm’s terminal

value, at the strike price of 1. At the end of period 1, a signal is observed which indicates the likeli-

hood of each of four possible terminal values. This signal is either H(high) or L(low). The proba-

bility that the signal is Hdepends on the manager’s period 1 effort level. The authors investigate the

desirability of resetting the stock options upon receiving the signal. For example, if the signal is L,

should the owner revise the strike price downwards? They compare two benchmark strategies: the

strategy of pre-commitment (i.e. the principal is committed to no resetting) and the strategy of rein-

centivization (resetting the strike price downward to encourage the manager to make more effort in

period 2). They find that when the manager’s utility function is linear in consumption and effort, it

is better for the principal to agree from the outset that he or she will reset the strike price in the

event of receiving the signal L. However, with other functional forms, the authors find that either

benchmark strategy may dominate the other, depending on specific parametrization. Our model and

that of Acharya et al. differ in a number of respects. First, Acharya et al. focus on the case of a risk-

neutral manager, while our model assumes that the manager is risk-averse. (As pointed out by Aseff

and Santos (2005), if the manager is risk-neutral, the first-best solution is to sell the firm to the

1

We acknowledge that one could also include another possible interaction between current and investment efforts: current and invest-

ment efforts, when exerted in the same period t, both affect the aggregate manager’s effort cost in period tin a non-additive way. For

simplicity, however, we suppose here that effort costs are additive, thus abstracting from the (cost-side) substitutability or complemen-

tary of efforts.

744 HILI, LAUSSEL, AND LONG