No Good Recommendation without Mechanistic Explanation? The Case of Higher Education Governance

The paper aims to show how could be promising to explain differences in performance across national Higher Education Systems by differences in the procedural and substantive policy tools deployed, seen by a mechanistic perspective. Mechanisms understand causation as the complex interplay of conditions, all required for some phenomenon to occur. Therefore, at least it specifies the domain of validity of causal associations - so that even classical "if-then" sequences are assumed to hold "provided" some other conditions. Mechanisms are generally valued as the heuristics capable of more realistic causal accounts. Thus, the paper argues that mechanistic explanations can improve if the explanatory model focuses on causes proximate to an unobserved mechanism, and if the operalization and treatment are approached from a Boolean understanding. Therefore, in explaining policy performance, the focus can fruitfully shifts on actual constraints to actors behavior - such as the ones embedded in instruments and related governance devices, so that designs informed by different mechanisms allow different expectations about the typical performance of the system. Such an explanatory focus has many advantages. Policy tools are manipulable, as they depend on political and administrative decisions; also, they are efficient causes, as they trigger mechanisms at the individual level which directly account for people’s and institutions’ behavior, hence performance. Tool-based explanations therefore can allow for policy learning and transfer more easily than «remote» constitutional, historical, or cultural accounts. The article applies the scheme to evaluate gains and limits of recommendations in the field of governance reforms in higher education policies in comparative perspective; refines and details hypotheses about the causal capacity of different governance designs on the effectiveness of education based on a mechanistic understanding; and organizes them into a research design suitable for testing by Qualitative Comparative Analysis.