A close look at second stage data envelopment analysis using compound error models and the Tobit model

In 2014 the Brazilian Electricity Regulator (ANEEL) evaluated the efficiency of power distribution utilities using Data Envelopment Analysis (DEA). Estimated efficiencies range from 22.46% to 100%. Although environmental information is available in the data set, corrected efficiencies were not investigated. Different second stage models can be applied to adjust for environmental heterogeneity. Although statistical correlation among efficiencies and environmental variables can be easily estimated, corrected efficiencies are subject to the underlying structure of the second stage model. Therefore, different second stage models may achieve different corrected efficiencies. We provide a detailed statistical analysis of the Tobit model and compound error models for second stage analysis. Limitations are described and the corrected efficiencies using these models are evaluated. Potentially, Brazilian power distribution utilities may achieve substantial changes in estimated efficiencies if second stage analysis is used.     

A new look at the two-mode problem

This paper considers the second-best policy problem that arises when auto travel is priced below its marginal cost and there is a substitute mass transit mode. We analyze the problem by combining a model of a rail line based on Kraus and Yoshida (J. Urban Econ. 51 (2002) 170) with the highway bottleneck model. The model involves a transit authority which optimizes, in addition to the fare, two dimensions of transit capacity. These are (1) the number of train units serving the route and (2) the capacity of an individual train unit. Under a very weak condition, second-best optimality involves expanding both dimensions of transit capacity. The larger percentage effect is on train capacity.     

Industrial energy efficiency: A look at the data

It is extremely difficult to make a precise, quantitative assessment of the impact of the myriad of factors affecting the improvement in industrial energy efficiency. It is certainly not correct to conclude that housekeeping measures alone have led to the observed improvement. Changing product mix among four digit SIC industries within the same two digit classification, variations in capacity utilization (returns to scale) and energy price increases as well as technological innovations have all contributed to part of the realized reduction in energy use per dollar value added over the period of investigation. Unfortunately, data limitations as well as modeling weaknesses prohibit an exact delineation of the impact of each of the factors on the increase in energy efficiency. The best that can be done—and quite convincingly so—is to qualitatively show that unequivocally these factors had an impact on the efficiency with which energy was used in the manufacturing process for the ten most energy intensive industries in the period 1971–1976.