Bias in returns to tenure when firm wages and employment comove: A quantitative assessment and solution

It is well known that unless worker-firm match quality is controlled for, reduced-form estimates of returns to firm tenure will be biased. In this paper, we show that there is a further pervasive source of bias, namely, the comovement of firm employment and firm wages. We argue that firm-year fixed effects must be used to eliminate this bias. Estimates from two large-panel data sets from Germany and Portugal show that the bias is empirically important. Finally, we show that the results extend to tenure correlates used in macroeconomics, such as the minimum unemployment rate since joining the firm.