In Australia it is estimated that mental illness accounts for around 13 per cent of the total years of healthy life lost to disease yet only 5 per cent of the annual health budget is allocated to services for the mentally ill (Butterworth 2003; Andrews, Hall Teeson and Henderson 1999). There is growing concern regarding the effect of employment, including stress, low-quality work and difficulties in balancing work and family life, and of the effect of unemployment on mental health. This paper uses a summary measure of individuals’ mental health to examine the links between mental health and labour market experiences. The measure is based on the summary score typically derived from the SF-36 instrument included in the first 4 waves of the Household Income and Labour Dynamics Australia survey (HILDA), but is adjusted to correct for seemingly undesirable properties of that measure. Simple analysis of labour force transitions over the first four waves of HILDA suggests that entering or remaining in unemployment does not have the deleterious impact upon mental health that might have been expected. Greater changes in mental health are associated with movements into and out of the labour force altogether. The estimation of multivariate panel models does find evidence of unemployment contributing to lower mental health for those persons who are participating in the labour force. However, larger differences in mental health are identified among those in employment conditional upon characteristics of their work or attitudes towards their jobs, than are found between those in and out of work.