But how does one determine market concentration to start with? The Herfindahl-Hirschmann index squares the market shares of firms in the industry and adds up the total. But the number of firms in a market does not necessarily impart how low - or high - are barriers to entry. These are determined by the structure of the market, legal and bureaucratic hurdles, the existence, or lack thereof of functioning institutions, and by the possibility to turn an excess profit.
The index suffers from other shortcomings. Often the market is difficult to define. Mergers do not always drive prices higher. University of Chicago economists studying Industrial Organization - the branch of economics that deals with competition - have long advocated a shift of emphasis from market share to - usually temporary - market power. Influential antitrust thinkers, such as Robert Bork, recommended to revise the law to focus solely on consumer welfare.
These - and other insights - were incorporated in a theory of market contestability. Contrary to classical economic thinking, monopolies and oligopolies rarely raise prices for fear of attracting new competitors, went the new school. This is especially true in a "contestable" market - where entry is easy and cheap.
An Oligopolistic firm also fears the price-cutting reaction of its rivals if it reduces prices, goes the Hall, Hitch, and Sweezy theory of the Kinked Demand Curve. If it were to raise prices, its rivals may not follow suit, thus undermining its market share. Stackleberg's amendments to Cournot's Competition model, on the other hand, demonstrate the advantages to a price setter of being a first mover.
In "Economic assessment of oligopolies under the Community Merger Control Regulation, in European Competition law Review (Vol 4, Issue 3), Juan Briones Alonso writes: "At first sight, it seems that ... oligopolists will sooner or later find a way of avoiding competition among themselves, since they are aware that their overall profits are maximized with this strategy. However, the question is much more complex. First of all, collusion without explicit agreements is not easy to achieve. Each supplier might have different views on the level of prices which the demand would sustain, or might have different price preferences according to its cost conditions and market share. A company might think it has certain advantages which its competitors do not have, and would perhaps perceive a conflict between maximising its own profits and maximizing industry profits.
Moreover, if collusive strategies are implemented, and oligopolists manage to raise prices significantly above their competitive level, each oligopolist will be confronted with a conflict between sticking to the tacitly agreed behaviour and increasing its individual profits by 'cheating' on its competitors. Therefore, the question of mutual monitoring and control is a key issue in collusive oligopolies".
Monopolies and oligopolies, went the contestability theory, also refrain from restricting output, lest their market share be snatched by new entrants. In other words, even monopolists behave as though their market was fully competitive, their production and pricing decisions and actions constrained by the "ghosts" of potential and threatening newcomers.
In a CRIEFF Discussion Paper titled "From Walrasian Oligopolies to Natural Monopoly - An Evolutionary Model of Market Structure", the authors argue that: "Under decreasing returns and some fixed cost, the market grows to 'full capacity' at Walrasian equilibrium (oligopolies); on the other hand, if returns are increasing, the unique long run outcome involves a profit-maximising monopolist".
While intellectually tempting, contestability theory has little to do with the rough and tumble world of business.
Contestable markets simply do not exist. Entering a market is never cheap, nor easy. Huge sunk costs are required to counter the network effects of more veteran products as well as the competitors' brand recognition and ability and inclination to collude to set prices.