The traditional structure conduct performance (SCP) industrial organisation model placed great emphasis on the role of market structure. Developments in game theory over the past twenty years have shifted the emphasis of competition policy from market structure to firm conduct. Nevertheless market concentration continues to play an important role particularly in merger analysis. For example, the Competition Authority uses market concentration measures as a filter for identifying possible anti-competitive mergers, as do the US competition agencies.
Mergers that result in high levels of market concentration are not presumed to be anti-competitive. Rather high levels of market concentration are merely the starting point for identifying cases that require detailed scrutiny. The most commonly used measures of market concentration are the Herfindahl Hirschman Index (HHI) and the four firm concentration ratio (CR4). These measures are described below. Note that before considering the degree of market concentration in a specific case, it is first of all necessary to define the relevant market. Market definition is the subject of a separate briefing note.
The CR4 is simply the sum of the market shares of the four largest firms in the market in question. The CR4, however, suffers from certain drawbacks.
There is no obvious justification for focusing on the market share of the top four firms rather than the top three or six. Thus the choice of the CR4 as the appropriate measure of concentration is somewhat arbitrary.
More importantly the CR4 provides only limited information about actual market structure. To take an extreme example a market where the top four firms each had a market share of 22.5% would have a CR4 of 90 but so would a market where one firm had a market share of 60% and there were three firms with 10% each. However, company behaviour may differ greatly in these two situations with very different competitive outcomes.
The HHI is generally considered a superior measure of market concentration. The HHI is the sum of the squares of the market shares of all firms in the market. Unlike the CR4, the HHI is influenced both by the number of firms in the market and differences in their relative size. The value of the HHI decreases as the number of firms in a market rises. Similarly the value of the HHI will be greater the larger the degree of inequality in firm size.
Take the two examples referred to previously, one where there are four firms each with 22.5% of the market and the second where one firm has 60% and three others have a 10% market share. Assuming for simplicity that in both cases the remaining 10% is divided up between two firms each with a 5% market share the HHI works out at 2075 in the first case and 3950 in the latter. The inequality in firm size in the latter case results in a much higher HHI.
The maximum value of the HHI is 10,000, i.e. 1002. The number of firms of equivalent size that corresponds to any specific value of the HHI is 10,000 divided by the actual HHI. Taking the two examples referred to above; a HHI of 2075 is equivalent to a market in which there are 4.8 equally sized firms while a HHI of 3950 equates with a market in which there are 2.5 equally sized firms. Comparing the number of equally sized firms corresponding to the value of the HHI with the actual number of firms in the market provides further information about the structure of a particular market.
One of the difficulties sometimes raised about estimating the HHI is that one needs to have information on the market shares of all firms in the market. While this is strictly true, it is the case that firms with small market shares do not greatly affect the result because larger firms have a much greater weighting in the calculation of the HHI. To take the examples referred to earlier, if the final 10% of the market was divided between 10 firms with 1% each rather than two with 5% each as previously assumed, the HHI figures would come out at 2025 and 3910 or 40 points lower than in the previous examples. In other words if one knows the market shares of all of the large firms then it is possible to make a reasonably accurate estimate of the HHI. When looking at a merger between two firms a and b one can estimate the change in the HHI on the basis of the market shares of just the two merging firms. Pre merger their contribution to the HHI is a2 + b2, post merger their contribution is (a + b)2. The difference between these two figures is 2ab, i.e. twice the product of their market shares.
The Competition Authority’s draft Merger Guidelines provide that where post merger market concentration is below certain levels the merger is unlikely to have any adverse effect on competition. The draft contains two alternative thresholds:
Where the HHI is above 1000 and increases by more than 100 points the market is considered to be moderately concentrated. The draft defines this as an amber zone.
Where the HHI is above 1800 and increases by more than 50 points as a result of the merger, the market is considered highly concentrated and the draft defines this as a red zone.
© CompEcon Limited 2002.
Back to top of page.