Wissenschaftlicher Vortrag

Cross-Efficiencies: Input vs Output based Effects – Using the DEA Approach for an Evaluation of European Health Systems
Dellnitz, A.
Reucher, E.
Wissenschaftliche Vorträge
Data Envelopment Analysis
DEA2015, 13th International Conference on Data Envelopment Analysis, Braunschweig, 24–27 August.

Classical CCR and BCC models follow the principle of best possible self-assessment of each DMU compared to other economic entities. Another, more objective and peer-oriented approach is based on the determination of so called cross-efficiencies. Here, each DMU is evaluated by a peer’s optimal weights in a more objective way. Whilst under the assumption of constant returns of scale (CRS), calculation of cross-efficiencies is independent of input or output orientation, this is not the case for variable returns of scale (VRS). In this case, some input-oriented cross-efficiencies may even be negative. However, what does this entail in authentic situations? A meaningful application of this, an example taken from the European Health System, convincingly illustrates this phenomenon. Resulting from this, recommendations as to how to overcome negative DMUs' inefficiencies from a given peer's viewpoint will be discussed.