This paper reviews the literature on job-related training and the effects of these investments for different groups of individuals. The paper also elaborates on the theories, empirical explanations, and policy implications that can be drawn from these findings. Employer-provided training is by far the most important source of further education and training after an individual enters the labour market. A substantial portion of these human capital investments are financed by firms and it appears that the contribution by individuals are in most circumstances relatively modest. At the same time, substantial gains for individuals participating in training are documented in a large number of studies. The benefits are not only confined to wage returns as research has also shown that training leads to increased internal employability and job-security; and external labour market effects such as higher labour participation rates, lower unemployment, and shorter unemployment periods. Training is not equally distributed among employees. Older, low skilled workers, and to some extent female workers typically receive less training than other groups of employees. However, we do not find any clear-cut evidence that returns to training varies with gender, educational or skills levels, which suggests that inequalities do not arise because of differences in returns to training, but are more a consequence of inequalities of the distribution of training investments. The findings of this review further suggest that the returns to training are higher in the case that it is financed by the employer and that the returns to training are substantially higher for those leaving for a new employer. Employer-financed training appears, however, to lower the probability of an individual leaving for a new job elsewhere. The analysis of the distribution of returns to training reveals that although individuals benefit from these investments, the employer reaps most of the returns to training which suggests that the productivity effects are substantially larger than wage effects.