Measuring Market Power: Shares and Other Techniques

• Pages: 93-133

93

CHAPTER V

MEASURING MARKET POWER: SHARES AND

OTHER TECHNIQUES

An important part of antitrust analysis is to try to identify whether a

firm or group of firms has or could obtain market power once a relevant

market has been defined. Market share and concentration statistics are

frequently a useful starting point in an assessment of whether a firm or

group of firms have sufficient control over the supply in a market to

exercise market power, but too much attention to share and concentration

statistics can obscure the market characteristics that underlie those

statistics and that determine the actual market power of the firm or group

of firms.1 Market share does not equal market power but rather can be an

initial indication as to whether market power might exist. Identification

of low market shares or concentration can be used as a screen to suggest

that the existence of individual or joint market power is unlikely to exist.

High market shares, however, need not imply the existence of market

power. For example, a high market share may not indicate significant

market power if entry barriers are low, if expansion could be readily

accomplished by incumbent firms, if competitors are well positioned to

expand, or if historical sales are not indicative of a firm’s ability to

compete for future sales.2

1. See U.S. DEP’T OF JUSTICE & FED. TRADE COMM’N, HORIZONTAL

MERGER GUIDELINES § 4 (2010) (“The measurement of market shares and

market concentration is not an end in itself, but is useful to the extent it

illuminates the merger’s likely competitive effects.”), available at

http://www.justice.gov/atr/-public/guidelines/hmg-2010.html [hereinafter

MERGER GUIDELINES]; id. § 5 (“The Agencies evaluate market shares and

concentration in conjunction with other reasonably available and reliable

evidence for the ultimate purpose of determining whether a merger may

substantially lessen competition.”).

2. See, e.g., United States v. Gen. Dynamics Corp., 415 U.S. 486, 493-94

(1974) (explaining that General Dynamics’ share at the time of the

inquiry overstated its power in the coal market because its share was

expected to decline due to its limited reserves); see also The Boeing Co.

and McDonnell/Douglas Corp., No. 971-0051 (F.T.C. 1997) (explaining

that although Boeing had nearly 60 percent of the current large

94 Market Power Handbook

This Chapter first explains the traditional methods of calculating

market shares and concentration levels, and identifies numerous issues to

consider in applying these methods. The Chapter then discusses issues

related to the interpretation of market shares and concentration and the

range of characteristics that may make it more or less likely that market

power exists individually or jointly for a given level of concentration or

market share. Because market shares and concentration are generally at

most an initial indicator of the potential for market power to exist, the

final part of this Chapter discusses other empirical methods for

measuring market power directly.

A. Fundamentals of Market Share and Concentration Data

Calculation

The definition of the relevant market will affect substantially the

calculations of market shares and, as a result, is often the biggest source

of dispute in market share calculations.3 As the definition of the relevant

market broadens, the market shares of the firm or firms in question often

decline as a broader market definition may result in the addition of more

firms in the relevant market. For example, a superpremium ice cream

manufacturer with a 50 percent market share of the superpremium

segment may have a much lower share if the relevant market is defined

as all ice cream (including superpremium, premium, and store brand

products), which would include many more manufacturers.4 On the other

hand, a broader market definition may imply a higher rather than lower

market share for the firm or firms in question.5 Thus, market definition is

a necessary and crucial first step in the measurement of market shares.

________________________

commercial aircraft market, McDonnell Douglas no longer constituted a

meaningful competitive force in the commercial aircraft market).

3. See supra ch. IV for discussion of market definition.

4. This example is adapted from Analysis of Proposed Consent Order to Aid

Public Comment, In re Nestle Holdings, Inc., available at

http://www.ftc.gov/-os/2003/06/dryeranalysis.htm.

5. For example, in a debate about whether HMO and PPO insurance plans

are in the same market, whether a firm’s share increases or decreases

under the broader market definition will depend on the firm’s relative

share of HMO and PPO lives.

Measuring Market Power: Shares and Other Techniques 95

1.

Market Share Calculation

Once a relevant market is properly defined, the calculation of market

shares may appear relatively straightforward; however, a number of

factors impact the reliability and usefulness of any statistics, including

the metrics, data sources, and time period used. Calculation of market

shares must also consider any relationships among the competitors.

a. Choice of Metric

The choice of metric used to calculate market share can be

influential.6 Sales are often seen as the de facto measure of a firm’s

market share. However, market share can also be calculated based on

production, capacity, or volume of reserves. The Merger Guidelines

explain:

In most contexts, the Agencies measure each firm’s market share

based on its actual or projected revenues in the relevant market

. . . . In cases where one unit of a low-priced product can

substitute for one unit of a higher-priced product, unit sales may

measure competitive significance better than revenues . . . .

In markets for homogeneous products, a firm’s competitive

significance may derive principally from its ability and incentive

to rapidly expand production in the relevant market in response

to a price increase or output reduction by others in that market . .

. . In such markets, capacities or reserves may better reflect the

future competitive significance of suppliers than revenues, and

the Agencies may calculate market shares using those measures.7

In theory, if products and associated services are truly homogeneous,

it should not matter whether dollars or units are used, since prices will be

the same across firms or brands. More often, however, products are

either physically differentiated or sold with different services (e.g.,

different amounts of product support, including presale and postsale

service and warranties) resulting in disparate pricing.8 When products are

6. See Gregory J. Werden, Assigning Market Shares, 70 ANTITRUST L.J. 67,

73-78 (2002).

7. MERGER GUIDELINES,supra note 1, § 5.2.

8. See Werden, supra note 6, at 74-75 (explaining the difficulty of finding a

“common denominator” across products, using bread as an example of a