Structural Econometric Modeling Rationales and Examples ...

1 Chapter 64. Structural Econometric Modeling : Rationales AND Examples FROM INDUSTRIAL. ORGANIZATION. PETER C. REISS. Graduate School of Business, Stanford University, Stanford, CA 94305-5015, USA. e-mail: FRANK A. WOLAK. Department of Economics, Stanford University, Stanford, CA 94305-6072, USA. e-mail: Contents Abstract 4280. Keywords 4280. 1. Introduction 4281. 2. Structural models defined 4282. 3. Constructing Structural models 4285. Sources of structure 4285. Why add structure? 4288. Evaluating structure single equation models 4290. Evaluating structure simultaneous equation models 4293. The role of nonexperimental data in Structural Modeling 4301. 4. A framework for Structural Econometric models in IO 4303.

2 The economic model 4304. The stochastic model 4304. Unobserved heterogeneity and agent uncertainty 4305. Optimization errors 4308. Measurement error 4311. Steps to estimation 4312. Structural model epilogue 4314. 5. Demand and cost function estimation under imperfect competition 4315. Using price and quantity data to diagnose collusion 4315. The economic model 4317. Environment and primitives 4317. Handbook of econometrics , Volume 6A. Copyright 2007 Elsevier All rights reserved DOI: (07)06064-3. 4278 Reiss and Wolak Behavior and optimization 4318. The stochastic model 4320. Summary 4324. 6. Market power models more generally 4325. Estimating price cost margins 4326. Identifying and interpreting price cost margins 4329.

3 Summary 4333. 7. Models of differentiated product competition 4334. Neoclassical demand models 4334. Micro-data models 4340. A household-level demand model 4342. Goldberg's economic model 4342. The stochastic model 4344. Results 4347. A product-level demand model 4348. The economic model in BLP 4349. The stochastic model 4350. More on the Econometric assumptions 4353. Functional form assumptions for price 4353. Distribution of consumer heterogeneity 4355. Unobserved product quality 4357. Cost function specifications 4359. Summary 4360. 8. Games with incomplete information: Auctions 4361. Auctions overview 4361. Descriptive models 4363. Structural models 4365. Nonparametric identification and estimation 4368.

4 Further issues 4374. Parametric specifications for auction market equilibria 4375. Why estimate a Structural auction model ? 4379. Extensions of basic auctions models 4381. 9. Games with incomplete information: Principal-agent contracting models 4382. Observables and unobservables 4383. Economic models of regulator utility interactions 4385. Estimating productions functions accounting for private information 4388. Symmetric information model 4391. Asymmetric information model 4391. Econometric model 4393. Estimation results 4397. Further extensions 4398. 10. Market structure and firm turnover 4398. Overview of the issues 4399. Ch. 64: Structural Econometric Modeling 4279. Airline competition and entry 4400.

5 An economic model and data 4402. Modeling profits and competition 4404. The Econometric model 4406. Estimation 4409. Epilogue 4410. 11. Ending remarks 4411. References 4412. 4280 Reiss and Wolak Abstract This chapter explains the logic of Structural Econometric models and compares them to other types of Econometric models. We provide a framework researchers can use to develop and evaluate Structural Econometric models. This framework pays particu- lar attention to describing different sources of unobservables in Structural models. We use our framework to evaluate several literatures in industrial organization economics, including the literatures dealing with market power, product differentiation, auctions, regulation and entry.

6 Keywords Structural Econometric model , market power, auctions, regulation, entry JEL classification: C50, C51, C52, D10, D20, D40. Ch. 64: Structural Econometric Modeling 4281. 1. Introduction The founding members of the Cowles Commission defined econometrics as: a branch of economics in which economic theory and statistical method are fused in the analysis of numerical and institutional data [Hood and Koopmans (1953, p. xv)]. Today econo- mists refer to models that combine explicit economic theories with statistical models as Structural Econometric models. This chapter has three main goals. The first is to explain the logic of Structural econo- metric Modeling . While Structural Econometric models have the logical advantage of detailing the economic and statistical assumptions required to estimate economic quan- tities, the fact that they impose structure does not automatically make them sensible.

7 To be convincing, Structural models minimally must be: (1) flexible statistical descriptions of data; (2) respectful of the economic institutions under consideration; and, (3) sen- sitive to the nonexperimental nature of economic data. When, for example, there is little economic theory on which to build, the empiricist may instead prefer to use non- Structural or descriptive Econometric models. Alternatively, if there is a large body of relevant economic theory, then there may significant benefits to estimating a Structural Econometric model provided the model can satisfy the above demands. A second goal of this chapter is to describe the ingredients of Structural models and how Structural modelers go about evaluating them.

8 Our discussion emphasizes that the process of building a Structural model involves a series of related steps. These steps are by no means formulaic and often involve economic, statistical and practical compro- mises. Understanding when and why Structural modelers must make compromises, and that Structural modelers can disagree on compromises, is important for understanding that Structural Modeling is in part art . For example, Structural modelers often intro- duce conditioning variables that are not explicitly part of the economic theory as a way of controlling for plausible differences across observations. Our third goal is to illustrate how Structural Modeling tradeoffs are made in practice.

9 Specifically, we examine different types of Structural Econometric models developed by industrial organization ( IO ) economists. These models examine such issues as: the extent of market power possessed by firms; the efficiency of alternative market alloca- tion mechanisms ( , different rules for running single and multi-unit auctions); and the empirical implications of information and game-theoretic models. We should em- phasize that this chapter is NOT a comprehensive survey of the IO literature or even a complete discussion of any single topic. Readers interested in a comprehensive re- view of a particular literature should instead begin with the surveys we cite. Our goal is instead to illustrate selectively how IO researchers have used economic and statistical assumptions to identify and estimate economic magnitudes.

10 Our hope is that in doing so, we can provide a better sense of the benefits and limitations of Structural Econometric models. We begin by defining Structural Econometric models and discussing when one would want to use a Structural model . As part of this discussion, we provide a framework 4282 Reiss and Wolak for evaluating the benefits and limitations of Structural models. The remainder of the chapter illustrates some of the practical tradeoffs IO researchers have made. 2. Structural models defined In Structural Econometric models, economic theory is used to develop mathematical statements about how a set of observable endogenous variables, y, are related to an- other set of observable explanatory variables, x.