| Abstract: |
In Japan, as in many developed countries, the government's agencies for
science have implemented several reforms to the scientic research system,
which has concentrated research resources in the top research universities.
However, the growth of research papers has stagnated in Japan during the
2000s. To analyze the reason for this, this paper develops a framework that
decomposes the changes in research output. The framework is based on a model
of universities and is an application of growth accounting that is widely used
in economics. Using the framework, we nd that the change in the allocation of
research funds between universities had only a small effect on research output.
The stagnation in research output during the 2000s was mainly accounted for by
the decrease in research time. We also conduct a counterfactual experiment to
examine how the research output would increase if the misallocation of
research resources were completely removed. |