Industrial Organization

Industrial Organization

January 27-28, 2017
Matthew Gentzkow of Stanford University and Robin S. Lee of Harvard University, Organizers

Ali Hortaçsu, the University of Chicago and NBER; Fernando Luco and Dongni Zhu, Texas A&M University; and Steven L. Puller, Texas A&M University and NBER

Does Strategic Ability Affect Efficiency? Evidence from Electricity Markets

Standard oligopoly models of short-run price competition predict that large firms can exercise market power and generate inefficiencies. Inefficiency, however, can arise from other sources as well, such as the presence of heterogeneity in strategic sophistication. This paper studies such a setting in the Texas electricity market, in which bidding behavior of some firms persistently and significantly deviates from Nash-equilibrium bidding. Hortaçsu, Luco, Puller, and Zhu leverage a unique dataset that contains information on bids and valuations to estimate the level of strategic sophistication of specific firms in the market. They do this by embedding a Cognitive Hierarchy model into a structural model of bidding into auctions. They show that larger firms have higher levels of strategic sophistication than smaller firms, and there is significant heterogeneity across firms. The researchers then use the estimated distribution of types of strategic sophistication to perform counterfactual calculations about market efficiency under different scenarios that increase strategic sophistication of low-type firms either exogenously or through mergers with more sophisticated firms. They find that exogenously increasing sophistication of small firms increases productive efficiency. Furthermore, mergers that create no cost synergies and increase concentration may nevertheless increase efficiency by improving the strategic sophistication of smaller firms.


Avi Goldfarb, the University of Toronto and NBER, and Mo Xiao, the University of Arizona

Transitory Shocks, Limited Attention, and a Firm's Decision to Exit

This paper investigates bounded rationality in a high-stakes business setting: a restaurant owner's decision to exit. Combining a 20-year quarterly panel on the alcohol revenues from every restaurant in Texas with weather data, Goldfarb and Xiao document that, given the same revenue record, restaurants with experienced owners are more likely to exit from business after unusually good weather (and stay in business after unusually bad weather) than those with inexperienced owners. This descriptive evidence motivates a structural model of belief formation and exit decisions. The researchers' model allows an owner to pay limited attention to transitory shocks, thus misinterpreting revenue signals. The researchers find that the vast majority of restaurant owners pay little attention to transitory shocks. The prevalence of inattention is due to the high cost of casting full attention continuously: for the 738 restaurants that could have made better decisions, the cost of paying full attention would have been about $1,500 per quarter for a median restaurant. Owners' pre-existing experiences in the industry before opening new restaurants reduce these costs substantially. For the 25,575 restaurants in the data, a median restaurant with three years' owner experiences has the cost of paying attention lowered by $1,200 per quarter.


David Atkin, MIT and NBER, and Dave Donaldson, Stanford University and NBER

Who's Getting Globalized? The Size and Implications of Intra-national Trade Costs (NBER Working Paper No. 21439)

How large are the intra-national trade costs that separate consumers in remote locations of developing countries from global markets? What do those barriers imply for the intra-national incidence of the gains from falling international trade barriers? Atkin and Donaldson develop a new methodology for answering these questions and apply it to newly collected CPI micro-data from Ethiopia and Nigeria (as well as to the U.S.). In order to overcome three well-known challenges that arise when using price gaps to estimate trade costs, the researchers: (i) work exclusively with a sample of goods that are identified at the barcode-level (to mitigate bias due to unobserved quality differences over space); (ii) collect novel data on the origin location of each product in their sample (to focus only on the pairs of locations that actually identify trade costs); and (iii) use estimates of cost pass-through to correct for mark-ups that potentially vary over space (to extract trade costs from price variation in an environment with potentially oligopolistic intermediaries). Without these corrections, the researchers find that their estimates of the cost of distance would be biased downwards by a factor of approximately four. Their preferred estimates imply that the effect of log distance on trade costs within Ethiopia or Nigeria is four to five times larger than in the U.S. They also use their pass-through estimates to calculate the incidence of surplus increases due to falling world prices. They find that intermediaries capture the majority of the surplus, and that their share is even higher in distant locations, suggesting that remote consumers see only a small part of the gains from falling international trade barriers.


Bruce Blonigen, the University of Oregon and NBER, and Justin R. Pierce, Federal Reserve Board

Evidence for the Effects of Mergers on Market Power and Efficiency (NBER Working Paper No. 22750)

Study of the impact of mergers and acquisitions (M&As) on productivity and market power has been complicated by the difficulty of separating these two effects. Blonigen and Pierce use newly-developed techniques to separately estimate productivity and markups across a wide range of industries using detailed plant-level data. Employing a difference-in-differences framework, the researchers find that M&As are associated with increases in average markups, but find little evidence for effects on plant-level productivity. They also examine whether M&As increase efficiency through reallocation of production to more efficient plants or through reductions in administrative operations, but again find little evidence for these channels, on average. The results are robust to a range of approaches to address the endogeneity of firms' merger decisions.


Takuo Sugaya, Stanford University, and Alexander Wolitzky, MIT

Maintaining Privacy in Cartels

It is conventional wisdom that transparency in cartels — monitoring of competitors' prices, sales, and profits — facilitates collusion. However, in several recent cases cartels have instead gone out of their way to preserve the privacy of their participants' actions and outcomes. Towards explaining this behavior, Sugaya and Wolitzky show that cartels can sometimes sustain higher profits when actions and outcomes are only observed privately. They provide conditions under which maintaining privacy is optimal for cartels that follow the home-market principle of encouraging firms to act as local monopolies while refraining from competing in each other's markets. On the other hand, they also show that transparency is always optimal when firms' profit functions satisfy a certain concavity condition that holds in linear Bertrand and Cournot competition. The researchers thus give a unified theory of both why transparency usually seems to facilitate collusion and why cartels sometimes benefit from maintaining privacy.


Adam Kapor, Columbia University; Christopher Neilson, Princeton University and NBER; and Seth Zimmerman, the University of Chicago and NBER

Heterogeneous Beliefs and School Choice Mechanisms

This paper studies how welfare and academic outcomes in centralized school choice depend on the assignment mechanism when participants are not fully informed. Kapor, Neilson, and Zimmerman use a survey of school choice participants to estimate a model of school choice that incorporates preference heterogeneity, strategic behavior, and subjective beliefs about admissions chances. The researchers evaluate the equilibrium effects of switching to a strategy-proof deferred acceptance algorithm, and of improving the information available to households. Subjective beliefs differ from rational expectations values and predict choice behavior. Switching to a deferred acceptance algorithm raises welfare, but has limited effects on the distribution of test scores.


Pietro Tebaldi, the University of Chicago

Estimating Equilibrium in Health Insurance Exchanges: Price Competition and Subsidy Design under the ACA

To design premium subsidies in a health insurance market it is necessary to estimate consumer demand, cost, and study how different subsidy schemes affect insurers' incentives. Combining individual-level enrollment data and plan-level claims from the Californian ACA marketplace with a model of insurance demand and insurers' competition, Tebaldi estimates demand and cost primitives, and assesses equilibrium outcomes under alternative subsidy designs. He finds that younger households are significantly more price sensitive and cheaper to cover. Given this, counterfactual simulations show that tailoring subsidies to age leads to equilibria where all buyers are better off and per-person public spending is lower.