The cyclicality of real wages has important implications for the validity of competing business cycle theories. However, the empirical evidence on the aggregate level is inconclusive. Using a threshold vector autoregressive model for the US and Germany to condition the relationship between real wages and business fluctuations on the phase of the cycle, it is demonstrated that the inconclusive evidence is not only caused by measurement problems, estimation method and composition bias as discussed in the literature. In addition, one should also consider whether the economy is in an upswing or a downswing. In general, the evidence for countercyclical wages is stronger in Germany than for the US, but taken together there is no clear systematic pattern.