In this paper, a short survey is given on the contents and some problems of the theory of optimal currency areas. In addition, a new criterion for the assessment of the optimality of a currency area is proposed: the ability of member states of a monetary union to adjust their economic system to changing internal and external conditions.This extension of the conventional theory is inspired by a phenomenon that can be observed for example in Germany but also in other European countries: In spite of a principle consensus on the necessity of major reforms on such fields as taxes, social security and labour markets, these reforms are not being implemented. The ability to reform, however, is of particular importance in a monetary union when asymmetric shocks occur. It is plausible to assume that today those determinants of national competitiveness that are under the control of economic policy are of major importance. With this assumption, the ability to adjust these determinants smoothly to a changing economic environment is a possible substitution for the loss of the adjustment instrument of the nominal exchange rate. The conventional theory with the focus on private agents' behaviour - e.g. in the context of labour mobility or wage flexibility - seems to have overlooked the importance of economic policy in the adjustment process.From this extension of the conventional theory there results scepticism in regard to the tendency to a further centralisation of economic policy in Europe, as for example on the field of employment policy. EMU means that there is an increasing demand for instruments to cope with asymmetric shocks. Thus, there tends to be also an increasing need for nationally differentiated approaches of economic policy. Therefore, centralisation - which makes national differentiation more difficult - could prove to be counterproductive.