Abstract
In this paper we seek a robust methodology for the measurement of the relative public sector efficiency of 19 OECD countries over the period 1980–2000. We estimate relative efficiency scores for five disaggregated accounts of public spending as well as for aggregate public spending. Then, we use a semi-parametric econometric method to isolate the impact of government inefficiency from the inefficiency arising from the socioeconomic environment and luck. To verify the validity of our index, we use it to examine a number of well-established relationships in the public choice literature, which have only been tested using local government data.