Promotions, State Dependence and Intrafirm Job Mobility: Evidence From Personnel Records

The objective of this paper is to add evidence of job mobility within a firm using personnel records from a single large U.S. corporation, focusing on the determinants of the hazard rates of being promoted to a higher hierarchy level. How to successfully control for unobserved heterogeneity is the goal in this type of analyses, and the approach taken here is to look at the history of the worker, by recognizing state dependence on past promotions. The results show that: (a) the peak of the promotion hazard rates is registered at exactly one year at the current position, even after controlling for observables characteristics and state dependence; (b) males, whites, and more educated workers have a higher probability of being promoted; (c) there is no evidence of “fast tracks†: previous promotions do not imply a higher probability of future promotion, even after controlling for tenure at the current job; and (d) recent new hires seems to have an advantage when competing with insiders for a higher position. These last two findings seem to be at odds with the hypotheses of “job-specific†and “firm-specific†human capital, and to indirectly agree with the “Peter Principle†prediction (Lazear, 2004), which suggests that if workers make efforts beyond their capabilities in order to be promoted, after reaching the desired higher position their incentives to be “over-productive†may decline.