The authors document the research output of 34 central banks from 1990 to 2003, and use proxies of research inputs to measure the research productivity of central banks over this period. Results are obtained with and without controlling for quality and for policy relevance. The authors find that, overall, central banks have been hiring more researchers and publishing more research since 1990, with the United States accounting for more than half of all published central bank research output, although the European Central Bank is rapidly establishing itself as an important research centre. When controlling for research quality and relevance, the authors generally find that there is no clear relationship between the size of an institution and its productivity. They also find preliminary evidence of positive correlations between the policy relevance and the scientific quality of central bank research. There is only very weak evidence of a positive correlation between the quantity of external partnerships and the productivity of researchers in central banks.