chevron-down Created with Sketch Beta.

    Antitrust Law Journal

    Volume 82, Issue 3

    Recent Trends in Concentration, Competition, and Entry

    Hal R. Varian

    Oct 09, 2019

    49 minutes

    Antitrust Horizontal Agreements Joint Conduct Monopoly Power Pricing Unilateral Conduct Vertical Agreements

    Summary

    • A discussion of typical claims by the press that the United States has a "monopoly problem" including claims such as, increase in industry concentration; decreased competition; an increase in prices, profits, and stock market values; and restriction of potential entrants by the market power of incumbents. 
    • Reviews the Council of Economic Advisers (CES) "Issues Brief" showing a change in market concentration in 12 of 13 sectors. 
    • Provides a critique of the concentration ratio (CR) as a valid way to measure concentration which shows the revenue earned by the 50 largest firm in that industry. 
    • Evidence suggests that the modest increases in concentration over the last 20-30 years appear to be due to increased productivity, with little to no harmful effects on prices and output for customers. 
    • Discussion of several economic theories that could explain "positive feedback" which refers to a force that makes big firms grow and small firms get smaller including; supply-side economies of scale; demand-side economies of scale; and learning by doing.
    Recent Trends in Concentration, Competition, and Entry
    bfk92 via Getty Images

    Jump to:

    In the last few years a number of articles have been published in the popular press that suggest that the United States has a monopoly problem. The details vary, but here are some typical claims.

    1. Concentration: Industry concentration has increased.
    2. Competition: Competition has decreased.
    3. Profits: Prices, profits, and stock market values have increased.
    4. Entry: Incumbents have used their market power to squash potential entrants.
    5. Scale: These forces are particularly prevalent in technology industries because of data-based economies of scale.

    The online information industries have been a particular target for such claims. Much has been made of the fact that a handful of large, profitable U.S. firms occupy major positions in this sector. In this essay I examine a small, but growing economic literature that examines the validity of these claims and finds that they have little empirical support.

    I. Concentration

    In May 2016 the Council of Economic Advisers (CEA) released an Issues Brief that contained a table showing the change in market concentration in 13 broad industry sectors. To measure concentration, the CEA used a metric known as the 50-firm concentration ratio (CR50), which shows the revenue share earned by the 50 largest firms in that industry. As illustrated in Table 1, according to this metric, concentration has increased in 12 of the 13 sectors.

    Carl Shapiro, former U.S. Deputy Assistant Attorney General for Economics and former member of the CEA, has identified several problems with this brief. I single out three issues.

    1. The fifty-firm concentration ratio (CR50) reported in Table 1 is not informative regarding the state of competition. Industrial organization economists generally believe that markets are normally quite competitive with far fewer than fifty firms, so we measure concentration using the Herfindahl Index (HHI) or perhaps the four-firm concentration ratio (CR4)
    2. [T]he two-digit industry groupings in Table 1 are far too broad to assess market power
    3. Economic Census data only report production at domestic establishments. These data do not include imports of manufactured products, which have grown dramatically over the past 20 years.

    Table 1:

    Industry Industry revenue earned by 50 largest firms, 2012 (Billion $) Revenue share earned by 50 largest firms, 2012 Percentage point change in revenue share earned by 50 largest firms, 1997–2012

    Transportation and warehousing

    307.9

    42.1

    11.4

    Retail trade

    1,555.8

    46.9

    11.2

    Finance and insurance

    1,1762.7

    48.5

    9.9

    Wholesale trade

    2,183.1

    27.5

    7.3

    Real estate rental and leasing

    307.9

    42.1

    11.4

    Utilities

    367.7

    69.1

    4.6

    Educational services

    12.1

    22.7

    3.1

    Professional, scientific, and technical services

    278.2

    18.8

    2.6

    Arts, entertainment and recreation

    39.5

    19.6

    -2.2

    Administrative/Support

    159.2

    23.7

    1.6

    Health care and assistance

    350.2

    17.2

    -1.6

    Accommodation and food services

    149.8

    21.2

    0.1

    Other services, non-public

    46.7

    10.9

    -1.9
    Source: COUNCIL OF ECONOMIC ADVISERS, ISSUE BRIEF: BENEFITS OF COMPETITION AND INDICATORS OF MARKET POWER 4 (2016).

     

    To illustrate point (1), consider an industry with 50 firms, each of which has a 2 percent market share. The CR50 would be 100 percent, even though such an industry structure would likely be quite competitive.

    With respect to point (2), consider the "Retail Trade" industry classification, which the CEA identifies as having one of the largest increases in CR50 from 1997 to 2012. That number represents a conglomeration of largely unrelated markets, from car dealers to furniture stores to sporting-goods centers to gas stations. These sub-industries do not compete with one another in any meaningful way.

    As for point (3), during the period 2000-2005, the only U.S. company that assembled smartphones in the United States was Motorola. According to the CEA methodology, the mobile phone assembly industry (in the United States) would have a 100 percent concentration ratio. This is nonsensical, since during this period Motorola had many foreign-based competitors that sold phones in the United States. Consumer electronics, clothing, furniture, automobiles, and many other sectors also face significant foreign competition that is entirely ignored by the CEA methodology.

    In sum, looking at 50-firm concentration ratios at the level of two-digit NAICS codes does not say anything meaningful about increases in concentration.

    In March 2016 The Economist published a chart that depicted changes in the four-firm concentration ratio at the level of four-digit NAICS codes (893 sectors). This data was also from the Economic Census but was far more granular than the 13 sectors used by the CEA. Shapiro examined data underlying this chart and found that there have been some increases in concentration but the increases were quite modest.

    Both the CEA and The Economist examine concentration at the national level, assuming that the relevant geographic market for each industry is the entire United States. However, in many industries, the relevant geographic markets are more localized than the entire United States. Indeed, recent economic research has demonstrated that concentration in local geographic markets has been declining in many industries, even as the nationwide concentration measure has been increasing...

    Continue reading the full article in PDF.

    The author would like to thank Rosie Lipscomb and Alan Freedman for helpful suggestions.