Wolak, Frank Abstract Currently, network industries such as telecommunications, electricity, water delivery and postal delivery are being re-structured and privatized in countries all over the world. Although the desire to improve the efficiency of operation of these industries is the common goal, many very different strategies are pursued in an attempt to achieve this common goal. The lack of consensus about the optimal mechanism for privatization and re-structuring of network industries is in part a result of the absence of a consensus within the economics profession about the structure of production and demand in network industries and the impact of the regulatory process on firm behavior. The purpose of this research is to derive the necessary empirical methods and undertake the data analysis required to move to a consensus on the optimal process for privatizing and re-structuring these industries. This research has to three major goals: (1) to expand the set of empirical methods available to analyze firm and consumer behavior in regulated industries; (2) to develop and implement empirical models of the impact of regulation on firm behavior; (3) to apply these and other methodologies to determine how restructuring should take place to achieve the maximum benefits to consumers. This work will analyze the operation of the telecommunications, electricity, water delivery and postal delivery services industries using these new empirical methodologies. The telecommunications research will focus on household-level demand for telecommunications services rather than aggregate demand; this analysis is important because the debate over optimal re-structuring often focuses on its impact on specific demographic groups such as the elderly and poor. My research will develop and implement applied econometric methodologies which expand the range of applicability of techniques for analyzing consumer demand with zero consumption of some goods. Properly accounting for zero consumption in the demand for telecommunications services is necessary because many households do not consume any of a particular telecommunications service. Examples include local and long-distance telephone service and cable television service. Another common problem which arises in the analysis of household-level demand is the lack of information on the specific prices faced by an individual household. Hence, this work will also develop and implement techniques for incorporating estimated distributions of prices, conditional on household characteristics, into the household-level demand estimation. This allows us to more accurately determine the consumption and consumer welfare implications of proposed telecommunications policy changes. The research on the electricity industry will analyze the actual operation of several restructured electricity markets around the world to determine the impact of the design of a so-called `competitive` electricity market on prices. The England and Wales electricity market and Nord Pool (the Nordic Power Market) are among the markets to be analyzed. This research will formulate and estimate a structural econometric model of bidding behavior by the two large producers in the England and Wales market in order to recover estimates of each firm's underlying cost function. These cost function estimates will then be used to determine the impact of alternative market rules and structures on the prices set by the market. The postal delivery services research is concerned with the changing pattern of household-level demand due to changes in the relative price of this good and the increasing availability of substitute technologies. The proposed work will break out the dimensions of the substitution away from postal delivery services due to the specific competing technologies, such as e-mail and FAX machines. This work will also develop alternative econometric methods for analyzing the demand for infrequently purchased products. The final aspects of this work will study how to recover the structure of production of postal delivery services and the rate of growth of USPS total factor productivity in a manner that accounts for the impact of the postal rate-setting process.
