Published and Forthcoming

• Measuring Long-Run Marketing Effects and Their Implications for Long-Run Marketing Decision, Marketing Letters, 19(3-4), December 2008, with P. Albuquerque, B. Bronnenberg, J. P. Dube, T. Erdem, D. Hanssens, G. Hitsch, H. Hong, C. Mela, and B. Sun.
  
  Paper     Abstract
  
  
  Abstract: This paper discusses the role of agents' beliefs and their implications for the economic modeling of their behavior, in particular, their behavior over time. The paper also discusses the corresponding planning problems facing both firms and consumers in their current decision making. After a general discussion of the consumer and firm problem, we discuss recent examples of some of the emerging empirical literature on dynamic choice behavior in marketing.


• A Dynamic Model of Consumer Replacement Cycles in the PC Processor Industry
  
  Forthcoming at Marketing Science
  Paper     Abstract
  
  
  Abstract: Product replacement plays a critical role in many high-tech durable goods markets. Since these markets frequently undergo both rapid improvements in quality and falling prices, the consumer's replacement decision is most often due to product obsolescence, as opposed to wear and tear. Managers in these industries know that consumers follow replacement cycles, but little is known about how and why these replacement cycles change over time. And recognizing that consumers do follow such cycles, how should managers alter their own strategies to take advantage of this?

  To address these issues, I develop and estimate a dynamic consumer demand model that explicitly accounts for the replacement decision when consumers are uncertain about both future product price and quality. Using a unique data set from the PC processor industry, I show how to combine aggregate data on sales and product ownership to infer replacement behavior. The results reveal substantial variation in replacement behavior over time. I find that a myopic model of replacement underestimates price elasticities by approximately 30 to 40 percent and overestimates the frequency of replacement by 50 percent. I demonstrate that this heterogeneity in consumer replacement behavior provides an opportunity for managers to tailor their product introduction and pricing strategies to target the particular segment of consumers that is most likely to replace in the near future.



• A New Approach to Estimating the Production Function for Housing
  
  Forthcoming at American Economic Review with Dennis Epple and Holger Sieg
  Paper     Abstract     Data & Code
  
  
  Abstract: Dating at least to the classic works of Alonso, Mills, and Muth, the production function for housing has played a central role in urban economics and local public finance. Estimating housing production functions is, however, challenging. The key problem encountered in estimation is that that the quantity of housing services per dwelling is not observed by the econometrician. Instead, we observe the value of a housing unit, which is the product of the price per unit of housing services and the quantity of housing services per dwelling. This paper provides a new flexible approach for estimating the housing production function which treats housing quantities and prices as latent variables. The empirical analysis is based on a comprehensive database of recently built properties in Allegheny County, Pennsylvania. We find that the new method proposed in this paper works well in the application and provides reasonable estimates for the underlying production function.


• Drs. Muth and Mills meet Dr. Tiebout: Integrating Location-Specific Amenities into Multi-Community Equilibrium Models
  
  Forthcoming at Journal of Regional Science with Dennis Epple and Holger Sieg
  Paper     Abstract
  
  
  Abstract: We consider the problem of integrating spatial amenities into locational equilibrium models with multiple jurisdictions. We provide sufficient conditions under which models that assume a single housing price in each community continue to apply in the presence of location-specific amenities that vary both within and across communities. If these conditions are satisfied, the models, estimation methods, and results in Epple and Sieg (1999) are valid in the presence of (potentially unobserved) location-specific amenities. We also show how to construct sufficient statistics that capture location specific spatial heterogeneity. We apply these techniques using data from the Pittsburgh metropolitan area. We find that these amenity measures capture proximity to important local employment centers as well as heterogeneity in school quality within a given school district.

Working Papers

• Competition and Innovation in the Microprocessor Industry: Does AMD spur Intel to innovate more? (Submitted, revised 7.29.2009)
  with Ron Goettler
  Paper     Abstract
  
  
  Abstract: We propose and estimate a model of dynamic oligopoly with durable goods and endogenous innovation. Firms make dynamic pricing and investment decisions while taking into account the dynamic behavior of consumers who anticipate the product improvements and price declines. The distribution of currently owned products is a state variable that affects current demand and evolves endogenously as consumers make replacement purchases. Our work extends the dynamic oligopoly framework of Ericson and Pakes (1995) to incorporate durable goods. We propose an alternative approach to bounding the state space that is less restrictive of frontier firms and yields an endogenous long-run rate of innovation.

  Using a simulated minimum distance estimator, we estimate the model for the PC microprocessor industry and perform counterfactuals to measure the benefits of competition. Consumer surplus is 2.5 percent higher ($5 billion per year) with AMD than if Intel were a monopolist. Innovation, however, would be higher without AMD. Counterfactuals reveal that consumer surplus can actually increase as the market moves towards monopoly, which suggests that policymakers ought to consider the dynamic trade-off of lower current consumer surplus from higher prices for higher future surplus from more innovation. Comparative statics reveal that competition does induce higher innovation if consumers have sufficiently high preferences for quality and low price sensitivity.

  Note: This paper was previously circulated under the title "Durable Goods Oligopoly with Innovation: Theory and Empirics." This paper was discussed in the March 2008 issue of Survey of Current Business, published by the Bureau of Economic Analysis in the U.S. Department of Commerce.



• A Dynamic Structural Model of Addiction, Promotions, and Permanent Price Cuts
  (submitted) with Baohong Sun
  Paper     Abstract
  
  
  Abstract: Addictive goods fundamentally differ from non-addictive goods: consuming more of an addictive good today reinforces the addiction and increases the likelihood of future consumption. Thus, addiction creates an intertemporal link between a consumer?s past and present decisions, altering their incentives to purchase and to hold inventory. Despite the influence of addiction, its impact on consumer purchase strategies and its implications for firms remain unclear.

  We construct a dynamic structural model with rational addiction and endogenous consumption to investigate how consumers respond differently to temporary versus permanent price promotions for addictive and non-addictive goods. We apply our model to unique consumer panel data on purchases of cigarettes, crackers, and butter. We find that addiction accumulated through past consumption affects decisions for cigarettes but not the two non-addictive categories. Ignoring addiction for cigarettes leads to biased estimates of price sensitivity, inventory holding costs, and stock-out costs. For cigarettes, we find an interesting asymmetry: the temporary consumption elasticity is smaller than the permanent consumption elasticity, but the converse is true for the purchase elasticities. No such asymmetry exists for crackers or butter.

• Product Strategy for Commercial Open Source Software
  with Vineet Kumar and Kannan Srinivasan
  Abstract
  
  Abstract: Commercial open source software (COSS) products---privately developed software based on publicly available code---represent a rapidly growing multi-billion dollar market. A unique aspect of competition between COSS firms is that many use software licenses that require their enhancements to be made publicly available, creating an incentive for some firms to free-ride on the contributions made by others. Despite its success, the COSS industry has received scant attention in the marketing literature. There are several puzzling aspects of the COSS industry that we explain with our model. First, why would profit-maximizing COSS firms contribute to software features knowing that their competitors can take advantage of these contributions? Third, in the presence of free-riding behavior, how can COSS firms make positive profits? Fourth, industry observers note that some COSS products are of comparable or better quality than traditional software products, which raises the questions of how a market based on free riding can produce better quality products.

  We develop a model of competition between COSS firms that rationalizes these puzzling observations in the industry. Our model consists of (1) a duopoly of software firms competing in a vertically differentiated market, where product quality is composed of a mix of features and usability, and (2) a developers' market where firms hire developers who seek to signal their unobservable skills by contributing to open source software. In addition to helping understand the puzzling aspects of this industry, our analysis also reveals several counterintuitive findings. First, when developers skills are unobservable to firms, we find that the profits of all firms as well as consumer surplus can be higher than when skills are observable. Second, we demonstrate that COSS products can be of higher quality than traditional closed source products. Third, when the market is large and signaling is costly, all firms can benefit when they are mandated to make their software code public, even when some firms free-ride completely on the contributions made by others.

Works in Progress

• Charter School Entry in Market Equilibrium
  with Dennis Epple and Maria Marta Ferreyra



• The Dynamics of Variety-Seeking Behavior
  with Jonathan Levav



• Policy Implications of State Space Bounds in Dynamic Oligopoly
  with Ron Goettler



• Categorizing Web Viewership Using Statistical Models of Web Navigation and Text Classification
  with Alan Montgomery
  
Presentation Slides

Older

• The Seven Percent Illusion?
  Paper     Abstract
  
  
  Abstract: From 1995 to 1998, more than 90 percent of mid-range IPOs had underwriter spreads of exactly seven percent, even though evidence suggests that economies of scale in the size of the offering should allow for lower fees. Chen and Ritter (2000) offer collusion as a possible explanation based on empirical analysis and observation. Alternatively, I investigate whether the spreads could be the result of non-cooperative behavior on the part of the underwriters. I build a model that captures several important characteristics of competition in the IPO industry, such as differentiation among underwriters in their reputation and ability to screen potential entrepreneurs. In equilibrium, as the underlying distribution of entrepreneurs becomes increasingly skewed towards larger firms, I demonstrate underwriters may charge approximately the same spread to entrepreneurs of different valuations. Thus, the model produces a pricing pattern that is consistent with observation, and does so in a non-cooperative environment.
• Mining the Web for Multimedia Content
  Presentation Slides    8/02