Concentrating research funding a very bad idea

Analysis

By Dr Vincent Larivière

In recent years, we have witnessed a tendency for granting agencies, particularly at the federal level, to concentrate research funding in ‘elite' researchers and students who receive an increasingly high proportion of available funds.

At the Social Science and Humanities Research Council of Canada (SSHRC), for example, the 2011 reform of funding programs increased the total amount that a researcher can request from $250,000 over three years (with a maximum of $100,000 per year), to $500,000 over five years (and a maximum of $200,000 per year). Thus, researchers can obtain higher amounts of funding, even though the total budget of SSHRC has declined in constant dollars.

Consequently, for the last year of the Standard Research Grants (2011), the 135 most funded researchers obtained 10% of the total budget of the program. The same percentage was obtained in the first year of the Insight Program (2012) by only 30 researchers. The total number of funded projects also decreased from 1,366 to 872—even when we combine Insight Development Grants with Insight Grants.

But what about the relationship between funding received by researchers and their output? Are those who receive the largest share of funding actually more productive and have higher scientific impact than those with less funding? And if so, how big is the difference? To answer these questions, we collected data on all research grants received by all Quebec university professors, their scholarly publications and citations received by their articles for 2000-10 period.

The combined analysis of research funding and publications shows that, in each of the three major disciplinary families (natural sciences, medical sciences, and social sciences and humanities) the increase in the number of publications does not follow the increase in funding. In other words, the more funding a researcher receives, the higher the cost-per-paper becomes.

In medical science, for example, the most funded researchers, receiving about $1.4 million annually, contribute to a median of 6.6 articles per year, while researchers receiving $550 000 annually (almost three times less) still contribute 4.1 articles, which is only 38% less. The trends are similar in the two other major disciplinary families.

But what about the impact of scientific research? One could argue that, even if the concentration of research funding generates diminishing returns in terms of research output, maybe it stimulates discoveries that have such high scientific impact that it justifies this concentration.

That is not the case: while the scientific impact of research increases along with funding for researchers who receive less than $100,000 annually, it increases much more slowly when the annual funding received climbs to more than $100,000. Only in medical science do citations increase linearly with research funding, suggesting that in this discipline, there are no decreasing returns to scale associated with the concentration of resources.

In summary, both in terms of the quantity of papers produced and of their scientific impact, the concentration of research funding in the hands of a so-called ‘elite' of researchers generally produces diminishing returns. In a context where financial resources for research are declining in constant dollars, it is important to ask whether the way funding is allocated is optimal.

Our numbers show that it is not the case: a more egalitarian distribution of funds would yield greater collective gains. It should be understood that the main determinant of scientific production is not so much the money invested but, rather the number of researchers at work and, by funding a greater number of researchers, we increase overall research productivity.

Research policies that concentrate financial resources also seem to ignore that there is a certain degree of serendipity associated with scientific discoveries, and by funding the work of as many researchers as possible, we increase the likelihood that some of them make major discoveries.

Dr Vincent Larivière is the Canada Research Chair on the transformation of scholarly communication, assistant professor, École de bibliothéconomie et des sciences de l'information de l'Université de Montréal and associate researcher, Observatoire des sciences et des technologies of UQAM.