Why has research productivity in Japan stalled over more than 20 years, especially for regional universities? To clarify the cause of this stagnation, this article analyzes how different forms of resource allocation in national universities affect publication productivity by using the time-series analysis of the top 10% of the highest quality papers (1991-2013) as well as cross-sectional analysis of the National Survey of the Academic Profession conducted in 1992 and 2016. Both national surveys asked professors for a self-reported estimate of publications, the amount of weekly teaching load, fixed budgets, and scientific research funds evaluated on the basis of project proposals.
The main findings of the analyses are: (a) Results using time-series regression analysis show that both budgets for science and technology and the subsidies for operating expenses increased the number of the top 10% count papers significantly, but the weekly hours of teaching load as well as the dummy variable of incorporation of national universities decreased research productivity. (b) The total research funds in 2016 shows even larger and statistically significant effect on the number of articles than that of 1992. This result supports the hypothesis that performance-based competitive funds bring positive change in the motivation and interest of faculty, from the fact that national universities received the majority of their funding from performance -based research funding after 21st century.
(c) Nevertheless, when we differentiate the total budgets of 2016 into the fixed and the performance based research fund, the reduced fixed budgets still increase the number of articles significantly, other variables being constant. This result suggests that the relationship between the fixed budgets and performance based research funds is not a substitute but are complementary. This explains why research productivity in the regional universities is lower than that of the research universities, from the fact that the amount of reduction of fixed funds, which allows free expenditure, is sharper than that of research universities. These findings suggest that Science and Technology policy should be reconsidered to turn around the negative spiral we are in now.