The Gains and Losses from Industrial Concentration

Abstract

No field in the industrial organization literature has been as well plowed as the relationship between concentration and profitability.Weiss (19711), in his latest review of this literature, d.iscusses over 4O such studies since 1951, and this is not a complete census.The reader has by this late date earned the right to demand strong justification for a new entrant.Mine is simply that,despite its bulk, the literature fails to inform us how to interpret its main findings.Those findings are well known: with few exceptions, market concentration and industry profitability are positively correlated.since the correlation is usually weak, the literature has tended to become a search for more complex and/or accurately specified relationships.I eschew that approach to focus on a more basic question: if concentration and profitability are indeed related, what market process produces the relationship?The traditional answer has been that high concentration facilitates collusion and hence super marginal-cost pricing, for which some profitability measure is a proxy.Unfortunately, this answer does not logically follow from the usual evidence, so its acceptance by economists and practitioners of antitrust policy is little more than an act of faith.Any profitability measure implies a corresponding difference between prlee and average cost.As a matter of simple arithmetic a causal relationship running from concentration to profitability can operate either through an effect on price (the usual interpretation) or on average cost, or, of course, p. I. Market Structure, Costs and PricesThe possibility that nrket structure and costs are related has of course long been recognized.An unconcentrated industry in which a technological advance produces "natural monopoly" or oligopoly cost conditions will become both more concentrated and more efficient over time.The process by which the old technolo is rendered economically obsolete will also entail a fall (or at least no increase) in price.The price decline need not be great enough to eliminate producer rents, either because the associated increase in concentration permits collusion or because the new technology diffuses slowly enough to leave room in the market for a fringe of' old- technology firms.Whichever force operates, there is a clear dilemma for antitrust policy: attempts to thwart increased concentration will merely waste resources without benefiting consumers.1The concentration-profitability literature has so far given little weight to "natural oligopoly" interpretations of the data.Two reasons for this neglect are important enough to affect the structure of this study: 1.A "natural oligopoly" interpretation is asymmetric.Some technological progress can be scale-reducing.If this sort of' change diffuses slowly enough, or cannot be implemented as efficiently by all firms, the large firms will be the marginal firms and smafl firms will earn rents.Thus we should observe unusually low (or declining) as well as high concentration associated with unusually high rates of return.By and large, such a U-shaped concentration-profits relationship has not been found.One inference could bc that rents to size-specific technical change are unimportant empirically.However, it would be just as easy to conclude that large size-related economies are simply more prevalent.Or, to the extent t.hut such economies are specific to a few oranizaUons, the largc'-size-relat.eIet'onomie; wM I

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 420 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.