Steven Berry

Estimation of a model of entry in the airline industry

This paper considers the effect of an airlines scale of operation at an airport on the profitability of routes flown out of that airport. The empirical methodology uses the entry decisions of airlines as indicators of underlying profitability; the results extend the empirical literature on airport presence by providing a new set of estimates of the determinants of city-pair profitability. These estimates imply that city-pair profits increase in airport presence and decrease rapidly in the number of entering firms. The literature on empirical models of oligopoly entry is also extended via a focus on the role of differences between firms. Copyright 1992 by The Econometric Society.

Differentiated Products Demand Systems from a Combination of Micro and Macro Data: The New Car Market

In this paper, we consider how rich sources of information on consumer choice can help to identify demand parameters in a widely used class of differentiated products demand models. Most important, we show how to use “second‐choice” data on automotive purchases to obtain good estimates of substitution patterns in the automobile industry. We use our estimates to make out‐of‐sample predictions about important recent changes in industry structure.

Do Mergers Increase Product Variety? Evidence from Radio Broadcasting

Mergers can reduce costs and alter incentives about how to position products, so that theory alone cannot predict whether mergers will increase product variety. We document the effect of mergers on variety by exploiting the natural experiment provided by the 1996 Telecommunications Act. We find that consolidation reduced station entry and increased the number of formats available relative to the number of stations. We find some evidence that increased concentration increases variety absolutely. Based on the programming overlap of jointly owned stations, we can infer that the effects operate through product crowding that is consistent with spatial preemption.

Public radio in the United States: does it correct market failure or cannibalize commercial stations?

Radio signals are pure public goods whose total value to society is the sum of their value to advertisers and listeners. Because broadcasters can capture only part of the value of their product as revenue, there is the potential for a classic problem of underprovision. Small markets have much less commercial program variety than larger markets, suggesting a possible underprovision problem. Public funding of radio broadcasting targets programming in three formats - news, classical music, and jazz - with at least some commercial competition. Whether public support corrects a market failure depends on whether the market would have provided similar services in the absence of public broadcasting. To examine this we ask whether public and commercial classical stations compete for listening share and revenue. We then directly examine whether public stations crowd out commercial stations. We find evidence consistent with the view that public broadcasting crowds out commercial programming in large markets, particularly in classical music and to a lesser extent in jazz. Although the majority of government subsidies to radio broadcasting are allocated to stations without commercial competition in their format (thereby possibly correcting inefficient market underprovision), roughly a quarter of subsidies support direct competition with existing commercial stations.

Chapter 63 Econometric Tools for Analyzing Market Outcomes

Abstract This paper outlines recently developed techniques for estimating the primitives needed to empirically analyze equilibrium interactions and their implications in oligopolistic markets. It is divided into an introduction and three sections; a section on estimating demand functions, a section on estimating production functions, and a section on estimating “dynamic” parameters (parameters estimated through their implications on the choice of controls which determine the distribution of future profits). The introduction provides an overview of how these primitives are used in typical I.O. applications, and explains how the individual sections are structured. The topics of the three sections have all been addressed in prior literature. Consequently each section begins with a review of the problems I.O. researchers encountered in using the prior approaches. The sections then continue with a fairly detailed explanation of the recent techniques and their relationship to the problems with the prior approaches. Hopefully the detail is rich enough to enable the reader to actually program up a version of the techniques and use them to analyze data. We conclude each section with a brief discussion of some of the problems with the more recent techniques. Here the emphasis is on when those problems are likely to be particularly important, and on recent research designed to overcome them when they are.

On the Nonparametric Identification of Nonlinear Simultaneous Equations Models: Comment on Brown (1983) and Roehrig (1988)

This note revisits the identification theorems of B. Brown (1983) and Roehrig (1988). We describe an error in the proofs of the main identification theorems in these papers, and provide an important counterexample to the theorems on the identification of the reduced form. Specifically, the reduced form of a nonseparable simultaneous equations model is not identified even under the assumptions of these papers. We provide conditions under which the reduced form is identified and is recoverable using the distribution of the endogenous variables conditional on the exogenous variables. However, these conditions place substantial limitations on the structural model. We conclude the note with a conjecture that it may be possible to use classical exclusion restrictions to recover some of the key implications of the theorems in more general settings.

Optimal Product Variety in Radio Markets

A vast theoretical literature shows that inefficient market structures may arise in free entry equilibria. The inefficiency may manifest itself in the number, variety, or quality of products. Previous empirical work demonstrated that excessive entry may obtain in local radio markets. Our paper extends that literature by relaxing the assumption that stations are symmetric, allowing instead for endogenous station differentiation along both horizontal and vertical dimensions. Importantly, we allow station quality to be an unobserved station characteristic. We compute the optimal market structures in local radio markets and find that, in most broadcasting formats, a social planner who takes into account the welfare of market participants (stations and advertisers) would eliminate 50%-60% of the stations observed in equilibrium. In 80%-95% of markets that have high quality stations in the observed equilibrium, welfare could be unambiguously improved by converting one such station into low quality broadcasting. In contrast, it is never unambiguously welfare-enhancing to convert an observed low quality station into a high quality one. This suggests local over-provision of quality in the observed equilibrium, in addition to the finding of excessive entry.

Fixed Costs and the Product Market Treatment of Preference Minorities

It is well documented that, in the presence of substantial fixed costs, markets offer preference majorities more variety than preference minorities. This fact alone, however, does not demonstrate the market outcome is in any way biased against preference minorities. In this paper, we clarify the sense in which the market outcome may in fact be biased against preference minorities, and we provide some conditions for such bias to occur. We then estimate the degree of bias in a particular industry using an empirical model of entry into radio broadcasting with two types of listeners, a preference majority and a minority, and the two types of stations targeting those respective listeners. Listening model estimates are used to infer fixed costs, which can then be used to find optimal station configurations as well as the welfare weights on different groups that rationalize the current configuration. The ensuing estimates reveal welfare weights that are 2-3 times higher for whites than blacks, and 1.5-2 times higher for non-Hispanic than Hispanic, listeners. The difference between the black and Hispanic results arises from the different patterns of importing and exporting: Hispanics listen to non-Hispanic-targeted stations more than blacks listen to white-targeted stations; and whites listen to black-targeted stations more than non-Hispanics listen to Spanish-language stations. Researchers and policy makers might add product markets to labor markets and other contexts that warrant attention for disparate treatment of minorities.

Optimal product variety in radio markets

A vast theoretical literature explores inefficient market structures in free-entry equilibria, and previous empirical work demonstrated that excessive entry may obtain in local radio markets. We extend that literature by relaxing the assumption that stations are symmetric, allowing for endogenous horizontal and (unobserved) vertical station differentiation. We find that, in most broadcasting formats, a social planner who takes into account the welfare of market participants eliminates 50%–60% of the observed stations. In 80%–94.9% of markets where high-quality stations are observed, welfare could be unambiguously improved by converting one such station into low-quality broadcasting, suggesting local overprovision of quality.

Empirical Modeling for Economics of the Media: Consumer and Advertiser Demand, Firm Supply and Firm Entry Models for Media Markets

Abstract We present empirical techniques that are both familiar to students of industrial organization and useful for modeling of media markets. We first focus on demand estimation with discussion of various discrete choice models. We then turn to estimation of the demand for advertising. We next turn to the supply side of the market. We discuss the estimation of models using continuous variables that firms choose (such as prices or ad level), followed by a discussion of entry models with both homogeneous and differentiated products. We conclude with a brief discussion of future challenges.