Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
A random parameters stochastic frontier model is applied to Italian data in order to evaluate the cost function and efficiency of higher education institutions. The method yields useful information about inter-institutional variation in cost structure and technical efficiency. Returns to scale and scope are evaluated for the typical university, and it is found that these returns are almost ubiquitously decreasing, a finding with clear policy implications.