Association for Public Policy Analysis & Management

Most research funding in the United States is awarded on a project-by-project basis based on the assessments of the intellectual merit and potential impacts of the proposed research. The result, is a highly unequal distribution of funding across individuals and institutions. The uneven distribution of funding can plausibly be explained in terms of differences in the creativity and ability of the individual scientists who competing for funding. But the unequal distribution of funds may also affect the productivity of the scientific enterprise. A priori, the production of scientific knowledge could be subject to diseconomies of scale if the transactions costs associated with managing larger laboratories and increased delegation of responsibility increase the costs of research. Alternatively, the concentration of resources may encourage the spreading of fixed cost and an increased division of labor contributing to economies of scale that enable those with more funding to produce disproportionately greater contributions to knowledge.

Determining which factors dominate is an empirical question that has only begun to be examined. Lorsch (2015 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4436771/ ) for example compared the productivity of investigators funded by the National Institute of General Medical Science and concluded that publications per dollar of funding, began to decline for investigators supported by the National Institute of General Medical Sciences with more than about $600,000 in direct costs of funded research per year, a result that led him to call for caps on aggregate funding for individual investigators. We are aware of few other studies that investigate this question.

In the proposed paper we will exploit data on academic chemists at 147 U.S. universities over the period 1990-2009 to examine the relationships between funding levels and knowledge production at the level of academic institutions. Using panel data estimation and controlling for a variety of factors that might affect the production of scientific knowledge we will explore the question of whether there are systematic relationships between the scale of funding and the ability of scientists to convert this funding into knowledge.