Productivity Analysis of German Electricity Distribution Utilities

This paper provides a productivity analysis of German electricity distribution companies. It addresses both traditional issues in electricity sector benchmarking, such as the role of scale effects and optimal utility size, as well as new evidence specific to the situation in Germany. Regarding the latter, we consider the potential effects of the three structural variables defined in the association agreements ("Verbändevereinbarung Strom VV II+"): consumer density, grid composition (cable versus aerial lines),and differences between East and West German distribution companies. We use labour, capital, and peak load capacity as inputs, and units sold and the number of customers as output. The data covers 380 (out of 553) German electricity distribution utilities. We apply a non-parametric data envelopment analysis (DEA) with constant returns to scale (CRS) as the main productivity analysis technique, whereas stochastic frontier analysis (SFA) is our verification method. The results suggest that returns to scale play a minor role: only very small utilities have a significant cost advantage. Low customer density is found to affect the efficiency score significantly in the lower third of all observations. The grid composition does not produce systematic effects. Surprisingly, East German utilities feature a higher average efficiency than their West German counterparts. The correlation tests imply a high coherence of the results.