8.2 Tournaments and promotions

Since the seminal work of Lazear and Rosen (1981) and Lazear (1995), economists have devoted considerable effort to the study of internal labour markets and the promotion “tournaments” that operate within them. The theory of tournaments in internal labour markets provides an integrated theory of compensation at different levels of the management hierarchy within firms, and in particular explains:

• the setting of remuneration primarily on the basis of roles rather than the human capital of the individual (as is demonstrated by the fact that remuneration is assigned to a role before an individual is assigned to the role), and
• the discrete jumps in remuneration that occur when individuals are promoted to higher levels.

Promotion is based more on relative rather than absolute performance, which means that promotion is a tournament in the sense that what matters to workers is whether they outperform their colleagues in the same firm. Further in this promotion tournament, the winner takes all of the prize, which illustrates the point that the compensation at each level is not designed to motivate the individuals at that level, but rather to motivate the individuals below that level to seek promotion.

Dispersion and compression of remuneration levels

The larger the pay spread between different levels of the hierarchy, the stronger the incentives to achieve promotion to the next level in the hierarchy. Large pay spreads may induce high levels of effort. However, large pay spreads may also induce work environments that are unattractive, or that inhibit the participation of certain types of staff or staff of certain ages. The 80-90 hour work week characteristic of financial services and law firms with very big prizes for those achieving the top jobs or becoming partners are an example. In addition, large pay spreads provide incentives for staff to collude and/or sabotage the promotion prospects of others.

Where collusion and sabotage are feasible and/or likely to be highly destructive of long-term productivity in the firm, pay compression (defined as remuneration levels that are less variable than individual performance) may be a superior alternative to large differentials in remuneration. Pay compression also serves to reduce the incentives to invest in lobbying for promotions rather than investing that effort in increasing the output of the firm.

Finally, the dispersion or compression of remuneration will provide staff with more or less insurance against low-productivity outcomes (some part of which may be exogenous to their effort). For example, larger pay spreads mean that staff know that low-productivity outcomes will be reflected very directly in their remuneration, whereas compressed remuneration levels provide insurance against low-productivity outcomes which will also reduce incentives for effort.

Internal labour markets and external recruitment

This literature emphasises the importance of internal labour markets in shielding workers from fluctuations in external labour markets by having a specific (junior) point of entry with internal promotion to more senior positions. Internal labour markets also promote higher levels of investment in firm and industry-specific capital than are optimal, if workers expect to be required to have regular resort to external labour markets which may allocate them to different firms and industries. However, internationally and (we suspect) in New Zealand, external recruitment has become more rather than less common in many sectors of the economy. The trade-offs associated with external as opposed to internal labour market recruitment are also of increasing interest within the New Zealand public sector (as recent publicity about decisions to advertise publicly for senior and ambassadorial appointments at the Ministry of Foreign Affairs makes clear).

Traditional labour-market models have focused on the requirement for specific skills and characteristics as the rationale for recruiting externally, but suggested that this involves a trade-off in respect of the intensity of the incentives provided by internal promotion tournaments. From this perspective, external recruitment increases the moral hazard and monitoring problems of the firm by reducing the intensity of incentives for workers to strive for promotion. In other words, the intensity of competition for promotion among cohorts of existing workers is reduced, if those workers think that the promotion may go to an external recruit rather than being allocated to the best candidate from the internal cohort.

An alternative perspective which has considerable merit is provided by Chen (2005). Chen notes that workers engaged in a rank-order tournament will have incentives to engage both in productive activities (that enhance the value of the firm) and unproductive activities such as “sabotage” of their opponents' performance (see Chen 2003 and Lazear 1989) or collusive agreements to shirk. This has two implications:

• When the potential for sabotage is introduced into the internal labour market it means that rank-order tournaments may induce workers to waste resources on unproductive sabotage, but also may mean that those who have the greatest chance of being promoted may not be those workers who add the most value to the firm.
• Collusion to reduce effort, if it is successful, may substantially reduce the payoff to the firm, but result in each individual having no less probability of winning the internal promotion tournament. When work is carried on in teams, the monitoring conditions required for successful collusion may be met, and collusion may be sustainable.

Outside competition for positions addresses both of these problems.

First, external recruitment reduces the payoff to both productive activities and sabotage, but this in itself does not establish net benefits for external recruitment. Second, sabotage has no value in competing against external candidates, but productive effort remains a useful instrument in this tournament. Thus, while external recruitment does reduce the total effort of existing workers, it results in a net increase in productive work and a net decrease in investments in sabotage activities. Thus, “although external recruitment hurts the ‘morale' of insiders and reduces their total effort, the output of the workers will actually increase” (Chen, 2005:261).

Second, competition from external applicants ensures that workers' probabilities of promotion are reduced by collusion on low effort. When workers collude to shirk, they increase the probability of an external appointment. If the differences in the probability of promotion with and without shirking are sufficiently large, external recruitment may actually assist the monitoring efforts of firms and reduce moral hazard by ensuring that there is a unique equilibrium where all team members choose the high-effort strategy.

Recent work has also assisted in clarifying the precise cause of the higher levels of effort resulting from participation in tournaments. Tournaments may induce higher levels of effort because of:

• Selection effects: higher pay attracts higher quality competitors, so the bigger the prize, the more each competitor will expect that higher performance is needed to win the tournament.
• Competition effects: competition (the “thrill of victory”) alone stimulates higher performance, even in the absence of monetary rewards for the winner.
• Pecuniary incentive effects: each participant in the tournament calculates the marginal cost-benefit of winning, so the higher the prize, the greater the effect on performance.

Coffey and Maloney (2010) show that the tournament model (the pecuniary incentive effect) has predictive power, even when it is possible to control for selection and competition effects. However, they find that all three effects have an impact on performance.