We extend a recently developed methodology for measuring the efficiency of Decision Making Units (DMUs) in the case of multiple inputs and outputs. The methodology accounts for economies of scope through the use of joint inputs, and explicitly includes information about the allocation of inputs to particular outputs. We focus on possible efficiency gains by reallocating inputs across outputs. We introduce a measure of coordination efficiency, which captures these efficiency gains. We demonstrate the practical usefulness of our methodology through an efficiency analysis of education and research conducted at US universities.

We extend a recently developed methodology for measuring the efficiency of Decision Making Units in the case of multiple inputs and outputs. The methodology accounts for economies of scope through the use of joint inputs, and explicitly includes information about the allocation of inputs to particular outputs. We focus on possible efficiency gains by reallocating inputs across outputs. We introduce a measure of coordination efficiency, which captures these efficiency gains. We demonstrate the practical usefulness of our methodology through an efficiency analysis of education and research conducted at US universities.

We advocate a nonparametric multi-output framework to estimate outputspecific economies of scale and we apply this model to male prisons in England and Wales over the sample period 2009-2012. To estimate output-specific returns to scale in prisons, we consider not only the cost-per-place, but also qualitative outputs such as purposeful out-of-cell activity and successful reintegration. Furthermore, we introduce environmental heterogeneity using the characteristics of the prison(ers). England and Wales offers a unique example to study economies of scale in prisons as the UK has started to build new super-size prisons in order to replace the most outdated prisons.

We reconsider the motivation of Data Envelopment Analysis (DEA), the non-parametric technique that is widely employed for analyzing productive efficiency in academia, the private sector and the public sector. We first argue that the conventional engineering motivation of DEA can be problematic since it often builds on unverifiable production axioms. We then provide a dual viewpoint and highlight the 'behavioral' interpretation of DEA models.We start from a specification of the production objectives while imposing minimal structure on the production possibilities, and construct tools to meaningfully quantify deviations of observed producer behavior from optimizing behavior. This brings to light the economic meaning of DEA, provides guidelines for selecting the appropriate model in practical research settings, and prepares the ground for instituting new DEA models. We hope that our insights will contribute to the further dissemination of DEA, and stimulate public sector applications of DEA that build on its behavioral interpretation.JEL Classification: C14, C61, D21, D24.

No abstract