Printer Friendly

Sunk Cost and Market Structure.

Most industrial organization economists, be they tradition structure-conduct-performance empiricists or game theorists, would likely approve of the central goal of Professor Sutton's book: a reconciliation of the diverse worlds of pure theoretical results based on the tools of modern game theory and the body of observed statistical regularities generated by over thirty years of cross sectional empirical studies. The task is approached with a combination of pure theory, empirical test and detailed cross-country industry studies.

Sutton properly notes that principal shortcomings of the game theoretic approach to understanding markets are the sensitivity of the results of game theoretic models to the details of the model's specification and the possibility of multiple equilibria with unexplained (or unexplainable) factors determining which equilibrium will be observed. Sutton develops a general set of game theoretic models and attempts to identify those results that are robust to changes in the details of the models and lend themselves to empirical test. These "robust" results are then used as the basis for standard cross sectional empirical analysis. The class of models employed could be described as two stage games in which firms first decide whether or not to enter a market (and incur a given exogenous level of sunk cost) and then, in the second stage, the firms interact to determine market price. Equilibrium in the market is defined by a zero profit condition.(1) The generality of the model is provided by an assumed exogenous factor, "the toughness of price competition," which defines the assumptions under which the post entry price (quantity) determination game is played.(2) The model is also generalized from cases of homogeneous products to differentiated products where advertising decisions become a part of the model and the "effectiveness of advertising" becomes a second exogenous factor.

The robust results from this class of models concern the relationship between market size and the equilibrium market structure as measured by concentration. Specifically the results are obtained for the lower bound of equilibrium market concentration as a function of the exogenous sunk cost to market size ratio. The actual level of equilibrium market concentration can exceed the lower bound depending on the exogenous "toughness" of price competition. In the homogenous good case, the model predicts that the lower bound on concentration will decrease monotonically as the required sunk cost becomes smaller relative to the size of the market. In the differentiated product case the same procedures yield a different result. The lower bound on equilibrium concentration is not monotonic. For given exogenous setup cost and toughness of price competition, equilibrium concentration will fall as market size increases but then increase as escalating advertising expenditure becomes the optimal strategy. Sutton then tests his general conclusion using a cross country sample of food manufacturing industries finding the expected relationship between the market size to setup cost ratios and concentration in homogenous product industries (e,g., sugar, salt, flour) and finding the relationship much weaker in differentiated product industries (e.g., soft drinks, beer, frozen foods).

The empirical work presented by Sutton raises a number of questions. Some of these are technical in nature but others are substantive. The first, and the one that is most likely to strike those readers with a more traditional Industrial Organization bent, is the choice of the question itself. In his search for theoretical results that are robust across a wide variety of game theoretic specifications, Sutton finds only the relationship between market size and market structure. Though Industrial Organization economists have studied a variety of relationships along the traditional structure-conduct-performance path, the ultimate interest has rightly been on performance. Though Sutton intentionally stays away from normative conclusion about market performance, the framework that is used makes industry structure totally endogenous. Given the exogenous factors assumed by the model, market structure and thus performance are the logical result. Discussions of how well particular markets work are, in a real sense, irrelevant; you get what you get out of the process and that is in essence the end of the story. Though some may applaud this conclusion, others are likely to be less pleased with the normative and policy implication of this result. The key to this conclusion is Sutton's assumption that the "toughness of price competition" is exogenous. It is through this assumption that Sutton is able to examine a whole class of Nash equilibria that are suggestive of different equilibria market levels of concentration for given setup cost to market size ratios. To accept the model, one is required to believe that concentration plays no role in how firms' pricing strategies and views or rivals' reactions are formulated. That is, for a given market something has determined that all firms in this activity are Cournot players and they will remain Cournot players if there are fifty firms with equal market share or if there are two firms in pure duopoly. The more standard view is that the degree of recognized interdependence of firms and thus their chosen competitive strategies is likely to depend on concentration, not be independent of it.(3)

On the technical side, some may well question the market definition chosen by Sutton for the empirical work. While the product definitions are standard, the geographic definitions are not. The geographic extent of Sutton's markets for the purpose of measuring market concentration are national boundaries. Though Sutton has been careful to exclude markets in which there is substantial cross country trade, the national level of markets is troublesome in that a number of the products (e.g., bread, soft drinks and a few others) are regional or local markets in the U.S. Sutton's concern is with industry concentration; the degree to which one sees a fragmented or consolidated structure across the dimensions he describes as horizontal and vertical product differentiation. The pricing outcomes however must be determined within a market context not an industry context. Finally Sutton faces the standard empiricist's curse: his model dictates that "sunk" cost be used in the empirical work but finds he is limited to data that can only proxy for this concept.

The inclusion of the Alfred Marshall quote on the value of inductive reasoning in economics aside, it appears that the theory often dictates the interpretation of the facts rather than the observations dictating the formulation of the theory. The reader will find in the book, however, a wealth of provocative analysis and a fascinating collection of cross country industry studies.

1. The market equilibrium occurs when the post entry profits of potential entrants into the market would be non-positive given Nash equilibrium pricing (or output) behavior.

2. The extremes of this may be thought of in term of a pure Bertand price setting game in which competitive levels of price will be reached and a Chamberlin type cooperative game in which joint profit maximizing level would be achieved.

3. This view is expressed by as diverse and venerable scholars as George Stigler ("A Theory of Oligopoly," Journal of Political Economy, (February 1964) and F. M. Scherer, (F. M. Scherer and D. Ross, Industrial Market Structure and Economic Performance, Boston: Houghton Mifflin, 1990).
COPYRIGHT 1992 Southern Economic Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Stewart, John F.
Publication:Southern Economic Journal
Article Type:Book Review
Date:Oct 1, 1992
Previous Article:Inequality: The Political Economy of Income Distribution.
Next Article:Methodological Foundations of Macroeconomics: Keynes and Lucas.

Related Articles
Innovation: the attacker's advantage.
Hitting a barrier deep within the planet.
Entry, collusion, and capacity constraints.
How plants allocate photosynthetic products.
Conductive silicone. (Materials).
How to Get Out of Debt.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters