The approximate validity of the Ricardian equivalence proposition remains hotly disputed despite a large body of empirical work devoted to this issue. Contradictory conclusions arise at least partially from the fact that a large variety of very different tests has been applied to very different data sets. Moreover, instationarity properties of the data are not properly taken into account in some of the older literature, so that inference may be misguided. This paper collects the most important tests of Ricardian equivalence and applies them to a single data set of quarterly West German macro data. It turns out that there is hardly any evidence against the Ricardian proposition. Paradoxically, the theoretical model that implies Ricardian equivalence, is strongly rejected by the data.