This article aims at presenting the recent experience of Italy's first research assessment exercise (VTR, valutazione triennale della ricerca) as an internationally relevant example in highlighting a neglected aspect of the evaluation of economic research, that is, the impact of research evaluation on research practice itself.
It will be shown that particularly (but not exclusively) when financial resources are linked to the outcome of the evaluation, procedures and criteria of assessment may create strong incentives for researchers and research institutions to modify their original aims and strategies. Thus, it is crucial to set clear principles and objectives for economic research and to conduct any research assessment on the basis of these objectives. As the case of Italy shows, when pluralism is not explicitly among these goals, the assessment exercise may result in a marginalization of minority approaches, which instead, ,nay be deemed worthy of survival and cultivation, both by policymakers and the scientific community.
The case of Italy is especially suited for our aims for two reasons: on one hand, pluralism of methods and topics within economics is traditionally well established in the Italian academia (if not in absolute terms, in an international comparison). Therefore, Italy's case is convenient for exemplification but is also relevant per se, at least from the perspective of certain economic approaches. On the other hand, the recent research assessment exercise in Italy (VTR) exhibits certain characteristics that clearly highlight the risks as well as the potentiality of research evaluation, with the aim of preserving and developing heterodox economic approaches, along with providing the stimulus for a lively and healthy debate within the mainstream.
Our findings support the view that if research institutions are encouraged to engage only in the lines of research that are likely to receive the highest rating according to the evaluation criteria adopted within the VTR, a convergence process is to be expected within economics, resulting in a potential disregard of heterodox schools and historical methods, and in favor of mainstream "Anglo-Saxon" approaches and quantitative methods. Ultimately, research pluralism may be harmed. These objections have been highlighted by Lee and Harley (1998), Lee (2007), and Lee and Eisner (2008) with reference to the U.K. Research Assessment Exercise. These works show that evaluations based on the criteria of closeness to mainstream economics, by means of the subsequent allocation of funds, may shape economic research in the middle-to-long run toward the disappearance of non-mainstream research fields. Thus, a critical reflection about the rating and ranking criteria adopted in the evaluation exercise is necessary.
Specifically, we conduct a statistical analysis of the publications evaluated within the VTR, contrasting them to a comparable subsample of the EconLit dataset. Our aim is to highlight systematic patterns in the selection of the publications submitted for evaluation. The underlying hypothesis, attaching relevance to this analysis, is that research institutions in the future will discourage the development of research topics (or approaches) that they deem unsuitable for evaluation because they are less likely to be positively ranked and thus, given the link between evaluation and funding, to contribute to the institutions' budgets.
Our main point is that, if evaluation is implicitly based on the criterion of proximity to the mainstream, as it was done in Italy's case, such behavior on the side of institutions may negatively affect the financing of research projects by nonmainstream economists as well as their hiring and career prospects. On the contrary, we claim that it is advisable and indeed possible to conduct research assessments that prove rigorous in assessing quality and at the same time are respectful of pluralism. …