Efficiency analysis of penitentiary centers in Spain using radial and non-radial DEA and its determinants factors 2015-2020

This study has a twofold objective. On the one hand, the comparative analysis of the efficiency of Spanish prisons structured based on three groups of conflict and size, by non-parametric radial and non-radial DEA methods, for 2015-2020. On the other hand, to determine to what extent exogenous factors such as Covid-19, the change of government or economic growth, have influenced the levels of efficiency, controlling for the prison size within each group. The comparative analyses carried out by the Radial and non-radial DEA estimations show important differences in the results. With the non-radial method, there is a clearly decreasing trend in efficiency over time and a greater dispersion of values in the inter-annual distribution. Therefore, the efficiency values of the Penitentiary Centers according to the group to which they belong in relation to their respective efficient frontiers show greater distances or possibilities for improvement in the case of this method compared to the Radial DEA. As for the results of the second stage, the influence of Covid on the management of the centers has had a negative impact on all groups. The implications derived from mandatory confinement have been reflected in greater organizational complexities, regardless of the type of center, in terms of size and criminal complexity. In this sense, there is a positive relationship between efficiency and the size of the centers. In the case of groups 1 (larger size and conflictive) and 2 (both average size and conflictive), a curvilinear shape (U-shape) is captured where a representative part of centers is in the increasing zone of the trend. In the case of group #3, the behavior is linear. Likewise, the influence of the political party in government has been captured. In the first case, differences in efficiency are found, depending on the legislative period, although the effects of the pandemic and its temporal coincidence, and the delay between political decisions and their subsequent implementation and return of results could have had an influence. Finally, the relationship between efficiency and economic growth is positive.