Estimating the Cost Efficiency of Public Service Providers in the Presence of Demand Uncertainty

Public service managers generally make input choices in the face of uncertainty about the demand for their services. However, this is generally not taken into account when estimating cost efficiency. The conventional approach to estimating cost efficiency is based on the assumption that managers choose inputs to minimise the cost of producing observed outputs. However, when demand is unknown at the time input decisions are made, many managers will instead choose inputs to minimize the cost of meeting various output targets. This paper explains how data envelopment analysis (DEA) methods can be used to account for demand uncertainty when estimating cost, technical and allocative efficiency. In doing so, it explains how DEA can be used to estimate the effects of demand uncertainty on costs. The methodology is applied to data on hospital and health service providers in the Australian state of Queensland. We obtain estimates of cost, technical and allocative efficiency that are quite different from the estimates obtained using a conventional approach that ignores demand uncertainty. Our empirical results also indicate that demand uncertainty has a significant effect on hospital costs.