The Journal of Productivity Analysis, 3, 277-306 (1992) 0 1992 Kluwer Academic Publishers, Boston. Manufactured in the Netherlands. Efficiency Analysis of Norwegian District Courts* SVERRE A.C. KITTELSEN Centre for Research in Economics and Business Administration, Gaustadallien 21, N-0379, Oslo, Norway and SNF-Oslo FINN R. FBRSUND Department of Economics, University of Oslo, Box 1095, N-0397 Oslo, Norway and SNF-Oslo Abstract This paper examines the efficiency of the Norwegian district courts with the aim of suggesting ways of improving this efficiency. Pooling the observations for the period 1983 to 1988 efficiency measures are calculated for each court using the nonparametric data envelopment analysis (DEA) method. The results show estimates of overman- ning due to technical inefficiency. Comparisons are made between the specialized city courts and the generalized rural courts. When using the yearly observations Malmquist indices are calculated to estimate the rate of produc- tivity change. Finally the paper addresses the questions of how the information from DEA can be used by the courts to become more efficient. 1. Introduction The district courts of Norway are organized as 107separate units, each with a geographical basis that includes one or several municipalities. They are the lowest level in a three-tier judicial system that is administered by the Ministry of Justice. Lately there has been some discussion in the Norwegian press as to the causes of the large backlog of cases that are found, particularly in the cities. The judges and staff of the courts blame insufficient alloca- tion of funds, but the political authorities have not been persuaded to grant more, suggest- ing instead that there is room for efficiency improvement in the courts themselves. One main question is therefore to what extent there are inefficiencies in the current behavior of the district courts, and how any inefficiencies can be eliminated. The purpose of this analysis is threefold: 1. To measure the productive efficiency of the courts. 2. To suggest ways of improving this efficiency. 3. To gather experience with methods that are well suited to 1 and 2 above, with an aim to find out to what extent they are applicable to public sector production generally. The courts are in many ways typical of public service production units. The courts are mainly organized in small units with 5 to 15 employees, although some city courts are sub- stantially larger. The products are numerous and heterogeneous, and each unit has been assigned the task of satisfying demand for judicial services from a defined geographical *The editor of this paper was Knox Lovell.