Comparing labor markets in the United States and Germany as Europe’s largest economy over the period from 1980−2004 uncovers three stylized differences: (1) Germany’s mean transition rates from unemployment to employment (UE) were lower by a factor of 5 and transition rates from employment to unemployment (EU) were lower by a factor of 4. (2) The volatility of the UE rate was equal in both countries, but the EU rate was 2.3 times more volatile in Germany. (3) In Germany EU flows contributed 60−70% to unemployment volatility, whereas in the U.S. they contributed only 30−40%. Using a search and matching model we show theoretically that the joint analysis of first and second moments offers general identification restrictions on the underlying causes for these differences. We find that a lower efficiency in the matching process can consistently explain the facts while alternative explanations such as employment protection, the benefit system, union power, or rigid earnings can not. We document that a lower matching efficiency due to lower occupational and regional mobility in Germany finds strong support in the data. Finally, we show that the highlighted matching friction leads in the model calibrated to the German economy to a substantial amplification and propagation of shocks.