In this paper, the small, but persistent interest rate differentials via-à-vis Germany which have existed in Austria, the Netherlands, and Belgium are analysed. These interest differentials may be thought of to consist of three parts: expected exchange rate movements within the band, expected changes of the central rates and a risk premium. Following a similar test as proposed by Svensson, we examine the credibility of the exchange rate policy in these countries. According to this test the Belgian exchange rate policy clearly lacks credibility for most of the period under consideration. There are, however, serious problems applying this test to Belgium due to the dual exchange rate system. For Austria and the Netherlands we calculate interest differentials which are adjusted for expected exchange rate movements within the band. It appears that the differences between the adjusted and the unadjusted interest differentials can be substantial. The Granger-causal relationship between fundamentals like the rate of inflation, the government budget deficit and the current account is also quite different for the adjusted and the unadjusted interest differential, where the fundamentals have the highest explanatory power for the latter measure.