Although popular in some circles, efficiency wage models of the labour market have proved surprisingly difficult to test and direct evidence for the central tenets of the theory is rare to non-existent. In this paper we propose a simple test of the Shapiro-Stiglitz shirking model which is based on the following idea. In the traditional search model the distribution of accepted wages should be truncated below by the reservation wage. But, if shirking is important then the employer will never want to employ a worker at their reservation wage and the distribution of accepted wages should be truncated below by the reservation wage plus something. That something is a measure of the importance of shirking. We test this prediction using data from the UK Survey of Incomes In and Out of Work. The results are not particularly supportive of the shirking model.
