The burden of glory: Competing for nonmonetary incentives in rank‐order tournaments

• Author: Raja Kali,David Pastoriza,Jean‐François Plante
• Published date: 01 March 2018
• DOI: http://doi.org/10.1111/jems.12233
• Date: 01 March 2018

Received: 10 November2015 Revised: 22 August 2017 Accepted: 31 August 2017

DOI: 10.1111/jems.12233

ORIGINAL ARTICLE

The burden of glory: Competing for nonmonetary incentives

in rank-order tournaments

Raja Kali1David Pastoriza2Jean-François Plante3

1Department of Economics, Sam M.

WaltonCollege of Business, Univer-

sity of Arkansas, Fayetteville,AR, USA

(Email: RKali@walton.uark.edu)

2Department of International Business,

HEC Montréal, Chemin de la Cote-

Sainte-Catherine, Montréal, Canada

(Email: david.pastoriza@hec.ca)

3Department of Decision Sciences, HEC Mon-

tréal, Chemin de la Cote-Sainte-Catherine,

Montréal, Canada (Email: jfplante@hec.ca)

Weare very grateful to the PGA Tour for pro-

vidingus access to ShotLink® data. We thank

the Official WorldGolf Ranking for their data

set,and FRQNT and SSHRC for funding. Our

specialt hanks to AlvaroBeamonte for his

adviceand suppor t.

Abstract

In an environment in which elite, highly paid professionals compete for nonmone-

tary rewards, we find evidence of underperformance. Our analysis suggests that chok-

ing under pressure from high-stakes nonmonetary rewards is behind the underper-

formance. This implies that high stakes nonmonetary rewards can create meaningful

pressure on individuals and lead to worse performance, a distinct issue that has yet

to be adequately examined. These findings come from an examination of the behav-

ior of top U.S. golfers competing to earn a place on the U.S. Ryder Cup team via

their performance in PGA Tour tournaments with differing allocations of Ryder Cup

qualifying points.

1INTRODUCTION

What is the impact of high-stakes nonmonetary incentives on job performance? This is an important question given how widely

used nonmonetary incentives are in business1. A growing body of academic literature has begun to consider this question2.

A primary focus of this literature has been on the role of nonmonetary incentives in principal–agent problems. At the risk of

generalizing, one could say that the conclusion of the literature thus far, both theoretical and empirical, is that nonmonetary

rewards can provide effective incentives, and in some cases may be better than financial incentives at aligning objectives of

agents and principals.

In this paper, we examine an environment in which individuals compete for a high-powered nonmonetary reward associated

with status and social esteem, and find evidence that runs contrary to existing empirical literature on nonmonetary incentives,

albeit in a setting that is entrepreneurial rather than agency. We find that elite, highly paid professionals who compete for

glory underperform, and that underperformance intensifies as pressure increases. Although pr ior research (Ariely, Gneezy,

Loewenstein, & Mazar, 2009) has found that high-powered monetary incentives can result in a decrease in performance, our

finding that such “choking” under pressure can happen with nonmonetary incentives is novel, to the best of our knowledge.

Empirical work on the question of nonmonetary incentives has been primarily experimental, in large part because of the diffi-

culty in finding a real-world setting with a convincing separation of monetary and nonmonetary incentives. What distinguishes

our study is an attempt to examine the importance of nonmonetary incentives in a competitive environment with high stakes,

high-ability, and wealthy agents: the PGA Tour. Because we cannot distinguish between the effects of status, social esteem and

respect (and indeed multiple nonmonetary mechanisms could be at work), we therefore bundle them together under the term

“glory.” Specifically,we use data from the PGA Tour to examine the performance of U.S. players between 1996 and 2012 when

competing to qualify for the Ryder Cup.

The Ryder Cup is the oldest competition of nations in professional golf, involving the two golf superpowers: the United States

and Europe. Players are not paid to participate in the Ryder Cup and there is no monetary award for winning the Ryder Cup.

Moreover, players who qualify for the U.S. Ryder Cup team must make a donation to a charitable foundation. This biennial

102 © 2017 Wiley Periodicals, Inc. J Econ Manage Strat. 2018;27:102–118.wileyonlinelibrary.com/journal/jems

KALI ET AL.103

competition is widely acknowledged to be the most prestigious competition of nations in golf. Playing for one's country is a

great honor for which even very wealthy professional golfers vie (Palmer & Dodson, 1999; Rosaforte, 1996). Being part of an

elite group of golfers who have the privilege of representing their countries, and not a direct monetary gain, is considered the

Ryder Cup's own reward. We use the contest to qualify for the Ryder Cup to providea separation of monetar y and nonmonetary

incentives.

The Ryder Cup qualifying point system allocates a number of points to each PGA Tour tournament. It is designed in such

a way that throughout the qualifying Ryder Cup period (typically two years), two editions of the same PGA Tour tournament

have a different endowment of Ryder Cup points. One year (the year previous to the Ryder Cup) in which there are few Ryder

points and another year (the year of the RyderCup) in which there are many Ryder points while all other aspects of the tournament

are similar. Our identification strategy is based on comparing the performance of players across the same tournament in two

subsequent years. By focusing on blocks of the PGA Tour tournament with similar economic incentives (i.e., prize money

awarded), but different glory incentives (i.e., available Ryder Cup points), we can measure the effect of the latter.

The main results of our paper are as follows. First, the higher the number of Ryder Cup points at stake is, the worse players

play. In other words, we find significant evidence that the desire to attain glory is a burden for player performance. Second, we

find evidence that as the time to qualify runs out and competitive crowding intensifies, players feel more pressure and tend to

underperform. Third, we find that Ryder Cup points have negative effects on both players with and without previous Ryder Cup

participation. Thus, adverse effects are present not only among players who try to attain glory for the first time, but also among

those who try to retain it. Fourth, we find no evidence that players who are not competing to qualify for the U.S. Ryder Cup

team (e.g., low-ability3players) are negatively affected in PGA Tour tournaments with a higher allocation Ryder Cup points.

Our results are thus robust to an out-of-sample test. Fifth, we eliminate the possibility that player underperformance is due to the

adoption of risky strategies. We find that when playersfind themselves in tournaments with a high allocation of Ryder Cup points,

they become risk averse and adopt safe strategies. Sixth, we show that off-the-course income (e.g., corporate endorsements),

is not sensitive to qualifying for the Ryder Cup. Thus, player underperformance is unlikely to be driven by pressure from the

expectation of future monetary rewards, as in Ariely et al. (2009). Finally, we find that even the wealthiest players in the PGA

Tour (i.e., those who earn very high off-the-course incomes) underperform when trying to qualify for the Ryder Cup.

The remainder of the paper is structured as follows. Section 2 motivates our study in the context of the literature on nonmon-

etary incentives and choking under pressure. Section 3 describes the Ryder Cup and the way in which American playersqualify

to represent the United States, which sets the stage for our empirical strategy described in Section 4. Section 5 contains our

primary results. In Section 6, we conduct a series of robustness checks. In Section 7, we consider alternative explanations for

the behavior of players to qualify for the U.S. Ryder Cup Team. Section 8 discusses our results and concludes.

2NONMONETARY INCENTIVES AND PERFORMANCE

Most empirical research on nonmonetary incentives has taken the form of laboratory experiments4. In pioneering work, Ball and

Eckel (1996, 1998) and Ball, Eckel,Grossman, and Zame (2001) directly manipulate status in the lab by artificially awarding high

or low status to subjects. They find that their manipulation affects behavior and those individuals awarded high status in the lab

enjoy favorableeconomic outcomes. Ashraf, Bandiera, and Jack (2014a) conduct a field experiment to examine the effectiveness

of financial and nonfinancial rewards for prosocial tasks. Theyfind t hat agentswho are awarded nonfinancial rewards exert more

effort than those offered financial rewards. Ashraf, Bandiera, and Lee (2014b) and Bradler, Dur, Neckermann, and Non (2016),

Kosfeld and Neckermann (2011) study the impact of status and social recognition on worker performance in field experiments

and find that purely symbolic awards significantly improve performance5.

The theoretical literature on nonmonetary incentives is similarly aligned. Ellingsen and Johanesson (2008) expand the theoret-

ical framework initiated by Benabou and Tirole (2003, 2006) to explainwhy a desire for social esteem can lead to greater effort,

especially when an actor deems the esteem accorded by an audience's approval to be worthwhile. The notion of social esteem

or respect in these studies reflects the idea that individuals care about what others think and is based on self-regarding motives

such as pride and shame. Besley and Ghatak (2008) develop a model showing how the explicit creation of status incentives by

firms can increase effort by creating what they call “motivated” agents. According to these authors, status is associated with the

awarding of a positional good such as a nonmonetary honor (a prize or a title) that only individuals who have produced high

output receive. Status incentives work by creating social divisions. Our notion of glory is closest in spirit to the Besley–Ghatak

notion of a nonmonetary positional good such as an honor.

Although previous results in the literature on nonmonetary incentives indicate by and large a positive effect on performance,

recent developments suggestt hat nonmonetary rewardscan have hidden costs not currently recognized. Specifically, Gubler et al.