2.1.4. Predicting the post-merger structure

After having delineated the relevant market, the expected post-merger structure needs to be predicted and evaluated. Besides barriers to entry, the most important factor playing a role is the concentration of the market. As already mentioned above, barriers to entry will be dealt with in the next section of this paper. We therefore constrain ourselves to shortly deal with the concentration of the market here. It is determined by (1) the number of competitors and (2) their market shares. The general assumption with regard to concentration ratios is that the lower the number of competitors and the higher their market shares, the higher their oligopolistic interdependence. With it, the capacity to constrain competition either by explicit agreements or by tacit collusion might increase. The most important concentration measure used in antitrust policy is the Hirschman Herfindahl Index, which is defined as the sum of the squared market shares in per cent of the firms in the market. It can thus take on values ranging between 0 and 10,000 (1002).

In principle, drawing on clearly identified values of the HHI might help to improve predictability in merger control. But it should not be overlooked that the simplicity of the indicator also has some setbacks. The degree of concentration in a given market, which the indicator is to express is, of course, not directly linked to a certain behaviour by the competing firms. Depending on the propensity to act competitively – which has also been called the “spirit of competition” – a high degree of concentration can lead to highly competitive behaviour as well as to attempts to circumvent competition. The strategies chosen by competing firms are not exclusively determined by concentration ratios. If “conduct” is not exclusively determined by the concentration of a given market, then the “performance” to be expected in that market cannot be reliably predicted drawing on concentration ratios. It thus remains unclear whether higher concentration ratios will lead to worse results – or to better ones. The following example might illustrate this problem.

Compare a symmetric Cournot duopoly with an asymmetric Stackelberg duopoly13 (Neumann 2000, 148): In the symmetrical Cournot game, both firms have a market share of 50% but the aggregate quantity supplied by them is only 2/3 of the quantity that they would supply under perfect competition. In the Stackelberg game, the leader has a market share of 2/3 and the follower correspondingly of 1/3. The aggregate quantity supplied by the two firms would be ¾ of the quantity that they would supply under perfect competition – and prices would be lower than in the Cournot game. From a welfare economic point of view, one would thus prefer a Stackelberg outcome to a Cournot outcome. Higher values of the HHI indicate trouble. In the example, the value for the Cournot outcome should thus be higher than that for the Stackelberg outcome. The exact opposite is, however, the case: the HHI value for the Cournot game is 5,000 (502 + 502) and that for the Stackelberg game is 5,555 (66.62 + 33.32). Exclusive reliance on the HHI can thus lead to wrong decisions.