Schmalensee, R. 1984. Do Markets Differ Much?
Summary by:
Fernández Rodríguez, Sonia
Introduction:
The first thing to be said about the topic under consideration is that there are some basic controversis in
industrial economics about how much does markets differ. This essay try to contribute with some ideas.
It is based in a cross-section study, but it differs from the others because it is a descriptive analysis and
most of this type of studies test hypotheses to apply to all markets.
With this paper, the author wants to show that the cross-section studies don’t need such controversial
hypothesis, so this study is fundamentally based in the importance of various effects. In particular, it provides
estimates of the relative importance of firm, market and market share differences.
Literature review:
There are three sources which give us different points of view in this area:
Classical tradition, which supports that differences among firms are transitory or unimportant (the only
differences are based on escale economies). The central hypotheses is that increases in seller concentration tend
to increase the beneficts of the industry. In the author’s opinión, this point of view is wrong because it is known
that there are differences. Scale economies
Seller concentration and collusion
Correlation between concentration and profitability
Revisionist view: it is anticlassical, and says that all markets are competitive, so within some industries
there are some efficiency differences between sellers (the more important are efficiency differences, the less
equal are market shares).
Managerial, which supports that the differences between firms are based in the managment. Some firms
are better managed tan others.
Methodology and results:
In order to solve the problem, the author makes some basic descriptive models, and an analysis of the
population variance of the rate of return of different operations in industry.
The author also makes an ilustrative scheme that shows the result of the analysis excluding one or more
of the three different effects (firm, industry and share).
From this scheme, the author explains the results observed, which are that while the probability level of
reject the null hypotheses of no firm effects is high (0,292), we can see that firms effects simply don’t exist.
However, we can see in the same way that effects of industry and market share do exist (<0,0001 and 0,0446
respectively). Anyway, according to the value R2 = 0,0023, which is very small, market share effects only are not
important.
In addition, from a correlation analysis, the author gets a negative correlation (-0.089) between the
effects of market share and industry.
Conclusions:
•Firms effects do not exist, which is perfectly consistent with substantial intra-industry
profitability differences.
•Industry efects exist and they are important. This topic supports that the classical focus on
industry-level analysis as against the revisionist tendancy to downplay industry differences.
•Market share effects exist, but do not matter much.
•Industry and market share effects are negatively correlated
