Empirical studies on job satisfaction have relied on two hypotheses: firstly, that wages are exogenous in a job satisfaction regression and secondly, that appropriate measures of relative wage can be inferred. In this paper we test both assumptions using two cohorts of UK university graduates....