Estimating returns to tenure

This paper asks whether existing estimates of how workers’ wages grow with their time at a company (returns to tenure) are biased when companies have significant labor market power over wages (monopsony power).

Accurately measuring wage growth helps us understand job stability, worker productivity, and fairness in wage policies, which are increasingly important in economic policy debates. Previous research suggested that wage growth estimates could underestimate true returns to tenure if wage changes within firms cause shifts in hiring patterns, affecting who remains employed and how wages evolve.

Using comprehensive data covering all employees in Portugal (‘Quadros de Pessoal’ data set), this paper confirms that previous findings of negative bias are robust. It shows that wage growth estimates underestimate the true value workers gain from longer tenure because they fail to properly account for firm-specific wage shocks—such as unexpected wage increases or decreases—that influence both hiring decisions and employee retention.

The study finds that this bias is notably larger in firms with greater monopsony power—those that face less competition for employees—than in more competitive environments. These results highlight the importance of carefully accounting for company-specific shocks in analyses of wage growth.

Click here to go to the paper by Andy Snell, Jonathan P. Thomas, and Pedro Martins