Taxes on labour income are the most important source of revenue for the governments. There are important frictions in the labour market. These two observations seem quite obvious, and there are probably few people who would intuitively disagree on these points. But what are the consequences for economists that follow out of these two observations? Wouldn't it be interesting and - even more important - relevant for economic policy to check the impact of tax policy on the functioning of the labour market? The answer that this thesis provides to this question is a definite "Yes". This dissertation shows that tax policy might effectively have a non-negligible effect on labour market frictions and thus on economic outcomes. Several topics like human capital, redistribution and the impact of unions are analysed in depth. It is shown that the existence of labour market frictions might in all these cases favour progressive tax schedules. Especially, a comparison to a traditional labour supply model without market imperfections shows that frictions on the labour market might imply a much more redistributive optimal tax schedule.