Issues in the Applied Analysis of Markets

National Bureau Of Economic Research Inc, Cambridge MA

Investigators

Abstract

Abstract: Issues in the Applied Analysis of Markets. This proposal has several parts. The first is to extend prior models of demand systems in characteristic space. These models were primarily developed to provide a tool that enables the user to analyze (own and cross) price and characteristic elasticities of demand. The same models can (and have) been used to evaluate the demand for and the utility derived from new goods, but their implications in those contexts are questionable (these include applica- tions like evaluating the returns to innovations, or constructing price indices). This is because some of the model's properties vis a vis the evaluation of new goods are "counterintuitve". We consider a change in the demand model that gets rid of these coun- terintuitive implications, and then compare alternative ways of evaluating new goods. One of the alternatives is \hedonic" analysis; a technique which is currently used to correct the CPI for new goods bias. In doing this com- parison I will provide a detailed analysis of alternative ways of computing hedonic corrections. Hopefully this will be of use to the statistical agencies. The second section of the proposal seeks to extend methods for computing and analyzing dynamic games to allow for asymmetric information. We focus on models which can detect and analyze collusive outcomes, and show how the learning and artificial intelligence literatures can help provide a simple tool for such an analysis. The third section of the proposal outlines a new and relatively simple procedure for estimation and subsequent empirical analysis of oligopolistic markets when either asymmetric information, or dynamic considerations, are important. I hope to explore the range of application and performance of this technique. The technique was developed while I was trying to sort out research strategies for two new empirical projects, one on the demand for pharmaceuticals and one on deregulated electric utility markets. The two final sections of the proposal outline these projects. Both empirical projects are joint with colleagues who have worked exten- sively on the industries studied. Our goal in the project on the demand for pharmaceuticals is to understand how past experience, marketing, and infor- mation on product performance, interact with other market characteristics to shape demand. We are particularly interested in the impact of advertis- ing on demand and on welfare, and on how recent institutional changes, like the development of the Over the Counter market for some durgs, impact on demand and welfare. The project on deregulated British electric utility markets will initially focus on getting estimates of start up and capital costs. This should enable us to provide measures of profitability that take account of these costs. This, in turn, will enable us to quantify the incentives to invest in more generating capacity that are provided by the institutions which determine the newly derregulated market outcomes. Later we hope to engage in a more in depth analysis of the relationship between different aspects of these institutions and the price and quantity allocations that are likely to be generated by the market.