Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)




This research aims at testing the effects of ownership structure on efficiency of various organizational forms: privately-owned regulated electric utilities (PR), publicly-owned electric utilities (PU), and consumer-owned cooperative electric utilities (CO). Because of the attenuation of their structure of property rights and high enforcement costs, it is expected that all three organizational forms (PR, PU, CO) will show some level of inefficiency. CO is expected to be more efficient than PR and PU since it shows: (a) more homogeneity of interest among principals; (b) more individual ownership claim over the assets; (c) a simpler political market (efficiency as the main issue of concern, and election every year); (d) a stronger monitoring structure when the control activity of the Rural Electrification Administration is considered. PU and PR cannot be discriminated and are expected to be as (in)efficient. The methodology used to test for efficiency differences requires the selection of an appropriate estimator (stochastic frontier model), an objective function (cost minimization), and a functional form to capture the characteristics of technology (Cobb-Douglas production function) from which we can derive the economic models to be estimated (long and short run cost functions and a system of equations). Such a methodology is appropriate since it provides us with an absolute measure of mean technical and allocative efficiencies for each ownership structure considered. Those models are estimated using MLE and Davidon-Fletcher-Powell algorithm using data measured at the plant level from steam generating electric utilities. Results from estimation of those models indicate that: (a) all three types of ownership structures suffer from some degree of inefficiency; (b) estimations of long run models rank CO as the most efficient ownership structure; the difference between CO vs PR and PU is important and statistically significant, moreover, measures of inefficiencies of PR and PU are not significantly different; (c) estimation of short run models produces ambiguous results; the stochastic frontier short run cost function puts PR as the most efficient ownership structure while the system of equations shows identical inefficiency measurement for PR and CO with PU having the highest level of inefficiency; (d) total inefficiency is composed mostly of technical inefficiency for PR, PU, CO; (e) when efficiency measurement is compared for small and large plants, total inefficiency of large plants is nearly twice as much as inefficiency of small plants.