The starting point of the research work is the dilemma that higher co-payments in health insurance – as in any insurance – may reduce moral hazard and the demand for health care services, but may also lead to undesired distributional effects by imposing an additional financial burden on persons of low income and/or high health care costs. The study tries to find a way out of this dilemma. The basic idea is to differentiate the co-payment rates in such a way that the allocative effects are preserved as far as possible, but that at the same time every insured person is financially better off than before (with lower – or zero – co-payment rates. In a theoretical part it is argued that the co-payment rates of such a system must be differentiated in two respects, namely with respect to income and with respect to health service costs.The higher the income, the higher the co-payment rate. And the higher the health service costs are, the lower is the co-payment rate. It is shown that – under certain assumptions – such a system of co-payment rates, which leaves every insured better off than before, does exist. The empirical part asked what that would mean for the German public health insurance: What could a system of increased double differentiated co-payment rates look like and what would be the effects on the contribution rate (and indirectly on the labour market). The German situation is simplified by considering five income classes and five health care cost classes (i.e., 25 cases). It is taken into consideration that in Germany co-payments already exist (however on a comparatively low level). Thus, the possibility and the effects of increased co-payments are studied. With increased co-payments the contribution rate can be lowered for two reasons. First, due to the co-payments themselves, and second due to the reduction of demand for health cares services brought about by the increased co-payments. High income earners profit more from reduced contribution rates than low income earners so that the former must bear higher co-payment rates than the latter in order to make everybody better off than before (with lower co-payment rates).The results are: – A system of differentiated co-payment rates can be constructed, also for the concrete case of the German public health insurance, which leaves every insured person financially better off than before. – However, the individual financial improvement is unequally distributed. High income earners with low health care costs profit most. – How far the contribution rate can be lowered depends on the highest (socially accepted) co-payment rate. If this were 80 % (only relevant for high income earners with low health care costs), the contribution rate could be lowered by 2.6 percent-age points. If the highest accepted co-payment rate is 40 %, the contribution rate could be lowered by around 2.0 percentage points. – The labour market policy relevance of this result must be seen in the light of what other realistic reform options in the German public health system might bring about in terms of lowering the contribution rate.