Product Market Competition and Returns to Talent

• Author: Jean‐Etienne Bettignies,Jen Baggs,John Ries
• Published date: 01 September 2013
• DOI: http://doi.org/10.1111/jems.12020
• Date: 01 September 2013

Product Market Competition and Returns to Talent

JEN BAGGS

Gustavson School of Business

University of Victoria

PO Box 1700 STN CSC Victoria BC V8W 2Y2, Canada

jenbaggs@uvic.ca.

JEAN-ETIENNE DE BETTIGNIES

Queen’s School of Business

Queen’s University

Goodes Hall, 143 Union Street Kingston ON K7L 3N6

jdebettignies@business.queensu.ca

JOHN RIES

Sauder School of Business

University of British Columbia

2053 Main Mall Vancouver BC V6R 3J1

john.ries@sauder.ubc.ca

This paper investigates how product market competition influences the wages paid to workers

and the distribution of talent across industries. We develop a model where firms facing different

competitive conditions bid for workers. The model predicts that wages are increasing in talent,

decreasing in competition, and the interaction between talent and competition is positive. In

addition, the most talented workers will be concentrated in competitive industries and talent

dispersion rises with competition. We use linked employee–employer data to test these predictions.

1. Introduction

The relationship between product market competition and economic efficiency has been

a topic of considerable interest in economics. Stigler (1958), for instance, defended the

“survivor principle” and argued that competition has a positive impact on efficiency by

weeding out the weaker firms, leaving only the more efficient in the industry. Bettignies

(2006) showed that competition may change the organization of the firm, leading to

leaner, vertically disintegrated, and economically more efficient, structures. And since

Hart’s (1983) pioneering work, a new literature emerged that focuses on the mitigating

or exacerbating effects of competition on agency costs and, in turn, on efficiency. In this

paper, we establish a different channel through which competition affects efficiency—

worker self-selection.

We develop a model where firms facing different levels of product market compe-

tition bid for workers. Competition influences the wages offered to workers and induces

self-selection. We show that the effect of competition on equilibrium wages is positive

for relatively talented workers and negative for relatively untalented workers. In addi-

tion, competitive industries will end up hiring a mix of highly talented and untalented

We thank the co-editor, two referees, participants at the 2009 CEA meetings, and discussant Joel Sandonis for

helpful comments. We aregrateful to the SSHRC for funding and Statistics Canada for access to the data.

C2013 Wiley Periodicals, Inc.

Journal of Economics & Management Strategy, Volume22, Number 3, Fall 2013, 569–593

570 Journal of Economics & Management Strategy

workers resulting in the dispersion of talent to be highest in competitive industries. We

test these and other predictions using the Workplace and Employee Survey (WES), a

large data set providing information at both the firm and employee levels.

Our theoretical model contributes to the literature on competition and incen-

tives which suggests that competition may have a positive (Hart, 1983; Raith, 2003;

Baggs and Bettignies, 2007), negative (Scharfstein, 1988;Martin,1993), or ambiguous

(Hermalin, 1992; Schmidt, 1997;Vives,2008) effect on worker effort and, as a conse-

quence, firm efficiency. But unlike this prior work which focuses on a single worker

varying effort according to competition levels, here we examine how various workers

differing in talent levels choose between industries of varying degrees of competition.

Our main theoretical contribution is to suggest that the relationship between competition

and efficiency may occur through self-selection of heterogeneously talented workers.

Recent empirical work suggests that talent is differentially rewarded across firms

and industries. Mocan and Tekin (2006), for example, find differencesin human capital,

and how it is rewarded, between for profit and not-for-profit businesses. Freedmanet al.

(2009) find that software firms operating in industries with higher variation in returns

are willing to pay more for “star” workers than firms in less volatile markets. They

argue that firms in high-variance payoff markets value star talent the most, since those

are the firms with the highest potential returns to good project selection, and the largest

losses from poor selection. Assuming star employees lower the probability of selecting

bad projects and raise the probability of good projects, these firms are accordingly more

willing to pay more for talent than firms with less variation in returns. Gibbons et al.

(2005) examine how skill determines both the wage and sectoral employment. In their

model, as the firm and employees become better informed about an employee’s skill

level, wages adjust and employees move between sectors. In an empirical test of their

model they find that high-wage sectors offer higher returns to skill. Weextend this work

to consider the role of competition in determining wages and attracting talent.

This paper also contributes to a small literature examining the effectsof competition

on wages. Guadalupe (2007) uses individual-level wage data in the United Kingdom and

finds that competition increases the returns to skill. Her model takes the distribution of

skills in an industry as given and predicts that wage dispersion is higher in competitive

industries. In her empirical implementation, she associates skills with occupations, with

managers deemed high-skilled workers and production and clerical workers being the

least skilled. Her measures of competition are industry concentration indexes, the 1992

Single European Act, and the 1996 appreciation of the British Pound. Our analysis

differs from Guadalupe in a number of dimensions. While as in her work, we find

that competition increases the returns to skills, we consider firms in different industries

competing for workers. This enables us to examine the distribution of talent across

industries resulting from worker self-selection. In addition, we allow for more or less

talented workers within industries by measuring talent as a function of education,

whether workers were recruited, and promotion histories rather than by occupation

type. Finally, our measures of competition are based on survey questions about the

degree of product differentiation perceived by firms, closely matching the theoretical

construct of our model.

The next section of the paper develops the theoretical model and its empirical

implications. Section 3presents the empirical analysis. In this section, we describe the

unique survey data that enables us to investigate the relationship between worker

talent, product market competition, and wages and discuss the empirical findings. We

summarize and discuss implications of our analysis in the conclusion.