In this paper we attempt to identify behavioral differences between public and private not-for-profit hospitals, by exploiting the introduction of the DRG-based payment system in the Italian NHS during the second half of the Nineties. We estimate the technical efficiency of a sample of hospitals for the period 1995-2000 considering an output distance function, and adopting both parametric (COLS and SF) and non-parametric (DEA) approaches. Our results show a convergence of mean efficiency scores between not-for-profit and public hospitals, and seem to suggest that differences in economic performances between competing ownership forms are more the result of the institutional settings in which they operate than the effect of the incentive structures embedded in the different proprietary forms. We also observe a decline in technical efficiency, probably due to policies aimed at reducing hospitalization rates.

No abstract

In this paper we deal with the role of proprietary structure in explaining efficiency within the Italian school industry. We analyze a sample of 497 schools located in Piemonte, a region in the North-Western part of the country, distinguishing between public, private for-profit and private nonprofit schools. In stage one of the analysis, we provide robust estimates of efficiency scores, using the two most widely known techniques in applied works, namely Data Envelopment Analysis (DEA) and Stochastic Frontiers (SF). In stage two, we suggests that proprietary structure matters in explaining efficiency. Nonprofit schools are more efficient than public ones, whereas for-profit counterparts are outperformed by public producers. Moreover, we find that foreign and disabled students affects negatively efficiency, raising concerns for

We exploit Italian law DPR 81/2009, which determines class composition, as an instrument to identify the causal effect of grouping students of different grades into a single class (multigrading) on children cognitive achievement. This article focuses on 7-yearold students-those at the beginning of their formal education. Results suggest that attendance in multigrade classes versus single-grade classes increases students' performance on standardized tests by 15-20 percent of a standard deviation. The positive impact of multigrading only appears for children sharing their class with peers from higher grades and is relatively stronger for students from disadvantaged backgrounds.

No abstract