This article uses Australian panel data for the years 2001–2009 to estimate returns
to general experience, job and occupational tenure.We pay particular attention to
issues of unobserved heterogeneity bias in our estimations. We find that both
general experience and occupational tenure have statistically and numerically
significant effects on wage outcomes, even after controlling for unobserved
heterogeneity. Job tenure on the other hand only seems to matter in OLS regressions
that do not control for heterogeneity biases. Once these biases are controlled
for, only a modest effect from job tenure remains. The inclusion of occupational
tenure in the estimating equation tends to negate even this modest job tenure
effect. The only exception to this is for workers in large organizations. For these
workers a small but statistically significant effect from job tenure remains, even
once we have controlled for heterogeneity and included occupational tenure in the
estimating equation. The results reported in this article have implications for the
various theories of the labour market that predict upward-sloping wage-job tenure
profiles.