My answer is along the same lines as Mr. Larivière's.
We often look at the model used by the Natural Sciences and Engineering Research Council, or NSERC. There is also the fact that it helps a pretty good chunk of the research community achieve a certain success rate, which encourages innovation.
Rather than looking at disincentives or incentives in terms of concentrating research, we must also consider how research communities are supported throughout their lifetimes. New researchers as well as mid-career and late-career researchers must be taken into account.
In addition to looking at the amounts provided by research grants, we will also have to look at what type of support is provided to institutions. As stated previously, researchers or research teams can do the work, and they do it even better if they are supported by teams. Then there are the laboratory technicians and professionals, who often have precarious working conditions, but who are nevertheless essential to carrying out research and maintaining infrastructure, be it in the sciences or in the humanities and social sciences. As Mr. Larivière pointed out, it is wrong to say that the humanities and social sciences have no infrastructure. Researchers in the social sciences and humanities count on an increasing number of databases and artificial intelligence.
A researcher must be supported by an ecosystem, by a team made up of research professionals and lab technicians, but also people working on the administrative side. These people support researchers by managing budgets and putting together grant applications, which allows researchers to focus on their main activity.
I would like to give an example of an important issue related to research, i.e., national security. The federal government has provided funding to support institutions in their efforts to ensure the security of their research. However, the way the money was distributed still favours concentration, because it was based on the total funding handed out by the grants councils. So the University of Toronto has received huge amounts of money, but smaller institutions have received $2,000 in funding and sometimes nothing at all to support the expertise required to ensure the security of the research being done.
This shows that resources are distributed inequitably, and this places a disproportionate burden on researchers at small institutions compared to those at bigger institutions, which are supported by a slew of professionals and experts.