Do wages rise with job seniority? - The Swiss case

Abstract

Whether or not seniority has a substantial effect on wages has been the subject of much controversy in the past decade, mainly in the U.S. Several economists have noted that unobserved heterogeneity across individuals and across job matches may produce inconsistent OLS-estimates of the effect of tenure on wages and turnover. Hence, labor economists have put forward two empirical strategies to deal with unobserved heterogeneity: Altonji and Shakotko (1987) use an instrumental variable for tenure, which is uncorrelated with the individual and job-specific component of the error term, but highly correlated with job tenure, whereas Topel?s (1991) basic idea is that within-job wage growth combines the returns to general and job-specific experience. These two empirical strategies revealed different returns to seniority and experience for the U.S. labor market. Our goal is, on one hand, to use the different methodologies for the Swiss labor market and, on the other hand, to evaluate the sources of these differences. Thus, we replicate these methods with Swiss data (Swiss Labor Force Survey, SLFS). In a first step, we estimate returns to tenure and experience with the standard regression method, OLS. Subsequently, we apply the Topel and the Altonji/Shakotko estimator, and use different specifications for each. We find that (i) Topel?s approach delivers similar returns to tenure to OLS, i.e. about 8% within ten years of job seniority, while the Altonji/Shakotko method delivers substantially lower returns (4%). (ii) Returns to tenure are minor in Switzerland compared to the U.S.