Affordable Access

TRAINING, TENURE, AND PRODUCTIVITY

Authors
Publisher
ScholarlyCommons
Publication Date
Keywords
  • Economics
  • General
Disciplines
  • Economics

Abstract

A good deal is known about the relationship between job tenure and employee earnings, primarily from estimates of wage equations which include a tenure variable. In contrast to the abundance of empirical evidence on the tenure-earnings relationship, very little is known about the relationship between the marginal productivity (MP) of employees and length of service (LOS) to the firm. Moreover, a controversy exists concerning the relative rates of growth of earnings and of MP over the typical employee's career within an industrial enterprise. The theory of firm-specific on-the-job training (OJT) suggests that the tenure-MP profile is steeper than the tenure-earnings profile, whereas certain models in the literature on implicit employment contracts imply that earnings grow more than productivity as a worker accumulates years of service.^ This dissertation explores the relationship between job tenure and employee MP among all workers employed in U.S. manufacturing during the period extending from the early 1960s to the mid-1970s. The method of analysis is the estimation of econometric models of producer technology in which employees are classified by LOS to the current establishment. Empirical implementation of these models yields estimates of the MP of employees classified by LOS, i.e., it provides estimates of parameters identifying the tenure-MP profile. The estimates also enable us to evaluate the macroeconomic impact of shifts in the tenure distribution of employees over the sample period on the average quality of employed labor and on the productivity performance of the U.S. manufacturing sector. The analysis reveals that a substantial reduction in the labor quality and in the growth of productivity was associated with the decline in job tenure, and that this decline is understated by conventional wage-weighted indices of the age distribution of the workforce.^ In order to exploit all of the available information on this subject, and also to provide alternative, comparable estimates of the parameters of an index of labor input in which employees are disaggregated by LOS, two different models of producer technology are formulated and estimated. The labor index parameters are estimated within the framework of a translog factor demand (labor's share) equation, on quarterly time-series data for durables and for nondurables manufacturing. . . . (Author's abstract exceeds stipulated maximum length. Discontinued here with permission of school.) UMI ^

There are no comments yet on this publication. Be the first to share your thoughts.