We use propensity score matching methods to quantify the effects of past self-employment experience on subsequent earnings in dependent employment using data on the population of Danish men observed between 1990 and 1996. Our results generally confirm existing studies in that we find that a spell of self-employment is associated with lower hourly wages compared to workers who were consecutively wage-employed. We also show, however, that this effect disappears—and even becomes positive in some settings—for formerly self-employed who find dependent employment in the same sector as their self-employment sector. Hence, the on average negative effect of self-employment is rather caused by sector switching than by the self-employment experience per se. Moreover, formerly self-employed who either enjoyed a high income or hired at least one worker during their self-employment spell receive wages in subsequent dependent employment that are at least as high as for individuals who have been consecutively wage-employed.