Accounting for caseloads in the efficiency analysis of courts of justice

Many reported applications of data envelopment analysis to courts of justice raise the issue of treatment of caseloads, consisting of the backlog of yet unresolved cases transferred from the previous year and new incoming cases. A common approach, which has some known drawbacks, is to incorporate caseloads as inputs in the model. In this paper, we propose a different methodology addressing this issue. We first assess the efficiency of courts disregarding caseloads and then decompose it into the product of two measures. One is the resolution rate that shows the proportion of caseload (capped by the boundaries of the technology) resolved by the court. The second is the caseload sufficiency factor that reflects the potentially limiting effect of the caseload on the ability of the court to achieve its efficient target. This approach allows straightforward extensions to arbitrary production technologies and different efficiency measures. We illustrate the proposed methodology by an application to a sample of Greek county courts. We show that the proposed methodology has significantly higher discriminating power on court efficiency than the model with caseloads as inputs. It also provides more detailed information about the sources of inefficiency of courts than the latter common approach.