Internal promotion competitions in firms.

Using a sample of skilled workers from a cross section of establishments in four metropolitan areas of the United States, l present evidence suggesting that promotions are determined by relative worker performance. I then estimate a structural model of promotion tournaments (treating as endogenous promotions, worker performance, and the wage spread from promotion) that simultaneously accounts for worker and firm behavior and how the interaction of these behaviors gives rise to promotions. The results are consistent with the predictions of tournament theory that employers set wage spreads to induce optimal performance levels, and that workers are motivated by larger spreads.

1. Introduction

* Since the seminal work on tournament theory by Lazear and Rosen (1981), an extensive theoretical literature has emerged on the subject of promotions as incentive mechanisms. (1) In the tournament model, workers of a given rank in an organization compete for promotion to the next level of the job hierarchy, with the promotion (and associated wage increase) awarded to the worker with the highest performance. The prize is the difference in wages between the post-promotion and pre-promotion jobs, and this is chosen by the employer to induce the optimal level of worker effort in the pre-promotion job. Tournament theory has important implications for the compensation structure of the firm and its relation to worker effort and performance. It provides a theory of career advancement and promotions within firms. Despite the theoretical appeal of the tournament model, its practical relevance as an explanation for promotions remains an open empirical question. Empirical tests of tournament theory have been of two main types. The first type focuses only on the behavior of workers (or agents), testing to see whether tournaments have incentive effects in that larger prizes imply higher levels of performance, usually in the context of sporting events rather than promotion decisions in typical firms. The second type focuses only on the behavior of the firm (or principal), testing to see whether compensation spreads appear to be structured to produce incentives in the manner predicted by tournament theory. In this article I diverge from both streams of previous empirical literature by considering the behavior of workers and firms jointly rather than in isolation. I do this in the context of greatest interest, namely promotion decisions in typical firms. Using a cross-sectional employer dataset containing information on promotions, wage spreads from promotion, worker performance, and worker, firm, and job characteristics, I estimate a structural model treating performance, the wage spread, and promotions as endogenous variables. In contrast to the studies in the first branch of empirical literature that regress a measure of the agent's performance on a spread that is assumed to be exogenous, I treat this spread as endogenous in the performance equation, since it is chosen by the firm to induce the optimal worker effort choice. To my knowledge, this is the first study of tournament theory in which the empirical methodology accounts for the optimizing behavior of both workers and firms, and how these behaviors jointly determine promotion outcomes. I describe how the interaction of worker and firm behavior has testable implications that would be missed by considering only worker behavior or firm behavior individually.

Central to tournament theory is the idea that promotions are determined by relative performance. Competitions based on relative performance, with the highest performer of a given rank winning the promotion, arise when internal hiring policies are combined with fixed job hierarchies. Throughout this article I refer to such situations as internal promotion competitions. Tournament theory takes the notion of an internal promotion competition and adds stronger implications arising from the optimizing behavior of workers and firms, in particular that firms optimally set wages to create incentives. Thus, a promotion tournament is a special case of an internal promotion competition, with additional testable implications.

I present two sets of empirical results in this article. In the first, I provide evidence suggesting that promotions are determined by relative performance for workers in a cross section of establishments. In the second, I estimate a three-equation structural model, finding support for the stronger predictions of tournament theory. A distinguishing feature of the data is the presence of employer-reported worker performance ratings. Such information is rare in datasets that span many establishments. The performance data allow a test of the incentive effects of tournaments in the context of greatest interest, namely promotion decisions in conventional firms. Prendergast (1999) criticizes the empirical incentives literature for what he perceives as its excessive focus on the contracts of workers for whom objective measures of output are readily available (e.g., CEOs, golfers, mutual fund managers, tree cutters, windshield installers, etc.). As Prendergast argues, most people do not work in such jobs. Instead, most workers are evaluated on subjective criteria. Since the analysis in this article is based on a broad cross section of workers for whom the relevant output measure is a subjective performance rating, it contributes results to the empirical incentives literature on the types of "typical jobs" that are rarely studied.

A common approach in empirical analyses of promotions is to study comprehensive data on all of the workers in a single firm, the identity of which is often undisclosed. Examples of such studies include Lazear (1992), Baker, Gibbs, and Holmstrom (1994a, 1994b), Audas, Barmby, and Treble (2004), and DeVaro and Waldman (2005). A small number of influential studies in this vein, most notably those of Baker, Gibbs, and Holmstrom, have significantly shaped the development of new theory on careers, as in the recent studies by Gibbons and Waldman (1999, 2006). In addition to large sample sizes and rich sets of consistently measured variables, the great advantage of single-firm datasets is that there is only one set of firm-specific institutions and procedures operating rather than a multitude of different processes, as is the case in a cross section. Such single-firm studies are useful in identifying the empirical regularities that hold simultaneously in one environment. Their advantages notwithstanding, since case studies are based only on single firms, there is no way of knowing how representative the firm's behavior is. In addition, the single firms that are analyzed tend to be selected nonrandomly, making it difficult to draw general inferences about employer behavior even when pooling the results from multiple case studies. An alternative approach, and the one taken in this article, is to use broader cross sections or panels of employers. Examples of this approach include McCue (1996), DeVaro and Brookshire (2007), and Belzil and Boganno (2005). Empirical work based on such broader samples of workers in a range of firms provides important information that complements the more detailed case studies of individual firms. In estimating a structural model using data spanning the full spectrum of firm sizes and types, industries, and distinct geographic labor markets, I aim to shed light on how well tournament theory describes the general tendencies of employer behavior.

2. Background and previous literature

* Lazear and Rosen (1981) showed that compensation schemes based on workers' ranks within an organization are attractive alternatives to output-contingent contracts, particularly when an employer cannot easily measure a worker's output. More precisely, tournaments induce the same efficient allocation of resources as output-contingent contracts such as piece rates and quotas, when the principal and agents are risk-neutral. In the model, two identical, risk-neutral workers in a firm compete in a low-level job for promotion to a high-level job. (2) The promotion (and its associated higher pay) is awarded to the worker who performs the best in the low-level job during some observation period. The prize from winning the promotion, commonly referred to as the "spread" is the wage difference between the two jobs. The firm chooses the spread before observing worker performance, with the knowledge that workers will then choose effort levels accordingly. Worker performance is a function of effort and a stochastic "luck" component that is independent across workers, with mean zero and variance [theta]. Having observed the wage spread, the worker chooses an effort level to maximize expected utility. The resulting optimal labor supply condition states that the worker chooses the effort level that equates the marginal return of effort to its marginal cost. This optimal labor supply condition and the assumed convexity of the effort cost function give rise to two implications about the optimal effort level. First, effort is increasing in the wage spread. (3) Second, effort is decreasing in [theta]. Intuitively, when random factors over which the worker has no control become more important determinants of the promotion probability, the marginal return to effort declines and the worker's incentives to exert effort are depressed.