Non-competitive research funding

I think that a very relevant recent news passed us by with little attention by the academic sector. It’s definitely worth reminding it to myself and to the few readers of my blog.

The Ghent University last July approved a non-competitive form of research funding.

“The basic idea is simple: if you are a professor at Ghent University and you have a research assignment, you get basic funding,” Rik explains. “You don’t have to go through a heavy procedure for it. And above all, you don’t have to compete with your colleagues. Moreover, you can spend the money freely, provided you use it for scientific research. Whether the funds are used to finance staff, operations or infrastructure: professors should be able to decide on this autonomously.” (Rector Rik Van de Walle, Ghent University)

Now, that’s an expression of trust and esteem on my professionalism I’d really like to hear from the governance of my university!

I think you all have perceived and experienced the extreme emphasis on competition in our field (I think that almost all forms of research funding comes from competitive calls at different levels: regional, national, European - at least for researchers based in the EU). Success rates are extremely low, plus at many levels the emphasis has shifted from research to its practical application, to the point of making it very difficult to fund basic research. Maybe you never even considered that this is a specific choice, because it has been the status quo for a relatively long time now, but it’s worth reflecting on some of the dark sides of this fixation on competition.

But first, you might wonder if this form of basic funding is so rare. Well, most Italian universities used to have some minor yearly research funding basically of non-competitive nature (maybe with small corrections as a bonus correcting a proportional sharing of allocated resources), but the per capita worth of this funding is extremely low (less than 2 thousand euros per person, per year). I’ve found traces of something similar to Ghent’s initiative at Politecnico di Torino, also documented in a research paper: both the initiative and the research are extremely interesting, although the authors are very honest in saying that the evaluation of the effects of this kind of research funding were limited (their results were not conclusive, if that is even possible). I suspect that this is not a kind of research that is popular (at the institutional level the mantra - or dogma? - of meritocracy - that is quite a different thing from recognizing and respecting merit - is basically undisputable) and cost effective (the paper shows that it takes time to get the data, and that the currently gathered data are probably insufficient to properly describe and analyze the phenomenon)… which is sad since it would instead be very important to do. We have a hope in the fact that at Ghent maybe they will gather data about the implications of their initiatives.

However, getting back to the dark side of the fixation on competition, and looking directly at the elephant in the room, do you know what the Matthew effect is? Maybe, Matthew effect is a bit less well-known than the preferential attachment mechanism so well described by Albert-László Barabási in his book Linked. This mechanism generates a self-reinforcing dynamics that Barabási used to define an algorithm for achieving scale-free networks, which he found to be a good model for reproducing properties of a portion of the World Wide Web. Basically, the idea is that whenever a new link is created from a web page it is much more probable that it leads to some popular website (e.g. Wikipedia) rather than this blog. This inevitably leads to have relatively few large hubs, getting a lot of incoming links, and a larger number of other sites with very few (if any) incoming links. Something similar, although probably not as prominent, happens with paper citations (and maybe even reads). The expression that was somehow substituted by “preferential attachment” was “cumulative advantage”, and that probably evokes very well what the Matthew effect is about. Imagine that a very good young researcher participates a call for research projects, gathering the forces of a very good group of organizations and researchers: she is successful, the project is funded, everybody is happy. Let us consider that there is an equally (or similarly) good young researcher that, in parallel, did apply as well to the same call, with an analogously good team, but was not funded, due to the limited budget for that research funding initiative. Well, the first researcher will inevitably have more chances of achieving further research funding in the future, quite simply because having achieved funds is one of the evaluated aspects of the principal investigator in a research proposal. There is an obvious cumulative advantage, and luckily there are some papers that analyzed this kind of phenomenon. An interesting point of the linked paper is that:

“We find no evidence that winners’ improved funding chances in subsequent competitions are due to achievements enabled by the preceding grant, which suggests that early funding itself is an asset for acquiring later funding.” (Thijs Bol, Mathijs de Vaan, and Arnout van de Rijt

Additional works also explore the relationships between the Matthew effect and the halo effect, but things get too complicated for a single post, maybe I’ll try to say something about it in the future.

I would conclude by saying that research funding policies have a very deep influence on the academic system, and on the lives of people working in (or trying to enter) the academia. People taking these decisions should be much more open to re-evaluating the implications of their choices, since intellectual honesty would imply that people taking so deep considerations of the evaluations of researchers would be totally open to being evaluated as well. On the contrary, I think that there is a single ideological position, disguising as a completely rational one, that is currently at the helm and it seems that the academics are not really inclined to start a mutiny.