American Economic Association

Protecting Competition in the American EconomyAuthor(s): Carl Shapiro

Source: The Journal of Economic Perspectives , Vol. 33, No. 3 (Summer 2019), pp. 69-93
Published by: American Economic Association

Stable URL: https://www.jstor.org/stable/10.2307/26732322

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

American Economic Association is collaborating with JSTOR to digitize, preserve and extend
access to The Journal of Economic Perspectives

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Journal of Economic Perspectives—Volume 33, Number 3—Summer 2019—Pages 69–93

Protecting Competition in the American

Economy: Merger Control, Tech Titans,
Labor Markets
Carl Shapiro

I
n the United States, we have a robust set of antitrust laws and antitrust institu-
tions designed to protect and promote competition. These laws date back to
the passage of the Sherman Act in 1890, supplemented in 1914 by the Clayton
Act and the Federal Trade Commission (FTC) Act. They are enforced by the Anti-
trust Division of the Department of Justice (DOJ) and by the FTC, together with
private antitrust litigation, in which plaintiffs are awarded three times any damages
they have suffered. For more than a century, a rich body of case law interpreting
these statutes has grown up, heavily influenced by economic research and economic
evidence. Indeed, over the twentieth century, the United States led the world in
creating and implementing competition policies to control cartels and mergers and
to rein in monopoly power.
Yet evidence is mounting that the largest US firms account for a growing share
of economic activity, and that profits and price/cost margins at these firms have
grown sharply in recent decades. Meanwhile, the economic might of the largest
tech firms seems to grow without bound. Have our antitrust laws and ­institutions
failed us?

■ Carl Shapiro is a Professor at the Haas School of Business, University of California, Berkeley,
California. He served as the chief economist at the Antitrust Division of the US Department
of Justice during 1995–1996 and 2011–2013 and as a member of the President’s Council of
Economic Advisers during 2011–2012. He has also served as an economic consultant for the
government and for private parties on a variety of antitrust matters over the years. His email
address is [email protected].
†
For supplementary materials such as appendices, datasets, and author disclosure statements, see the
article page at
https://doi.org/10.1257/jep.33.3.69 doi=10.1257/jep.33.3.69

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
70 Journal of Economic Perspectives

This article makes the case that we need to reinvigorate antitrust enforcement in
the United States in three areas. The clearest area where antitrust enforcement has
been overly lax is the treatment of mergers. The accumulated evidence indicates that
competition would be protected and promoted if the Department of Justice and the
Federal Trade Commission were willing and able to block more horizontal mergers.
The second area where antitrust enforcement has become inadequate is the treat-
ment of exclusionary conduct by dominant firms. The fundamental problem in this
area is that the Supreme Court has, over the past 40 years, dramatically narrowed the
reach of the Sherman Act. The third area concerns the market power of employers
as buyers in labor markets. Historically, antitrust enforcement has largely ignored
labor markets. Greater antitrust attention and oversight are warranted, although it
is too soon to know whether more robust antitrust enforcement in US labor markets
would make a significant difference for the wages earned by employees as a group.
Before discussing antitrust policy in these three areas, it is helpful to lay some
groundwork by briefly summarizing some of the evidence that has been accumu-
lating regarding competition and market power in the US economy. Baker (2019)
skillfully reviews this evidence in greater detail. Largely on the basis of that evidence,
he too advocates for stronger antitrust enforcement.
First, there is clear evidence that corporate profits have risen significantly over
the past few decades. The share of US GDP accounted for by corporate profits rose
by about half from 1985 to 2016, from around 7.5 to over 11 percent (Shapiro
2018). Barkai (2017) argues that excess profits, meaning the return to capital above
the level required to attract investors, have risen sharply as the risk-free rate of
return to capital has fallen. Furman and Orszag (2018) report that the premium
on the return to all private capital over safe assets has risen from about 200 basis
points in 1985 to more than 800 basis points in 2015, and that the return to capital
has become much more skewed among US publicly traded nonfinancial firms. The
current market capitalizations of the leading US firms further indicate that investors
expect these high profits to persist. High and persistent economic profits suggest
substantial and durable market power.
Second, there is evidence that price/cost ratios in the United States have risen in
recent decades.1 De Loecker, Eeckhout, and Unger (2018) report a sizable increase
in the weighted average ratio of price to marginal cost for publicly traded firms
in the United States, from 1.21 in 1980 to 1.61 in 2016, with most of that increase
occurring at the top of the distribution. However, measuring price and marginal
cost accurately on the basis of public financial accounting data is extremely diffi-
cult, especially because large, publicly traded firms produce many products and
services.2 Traina (2018) finds that the ratio of price to marginal cost at nonfinancial,
nonutility, publicly traded firms in the United States rose very modestly from 1980

1
Industrial organization economists have long looked at price/cost margins as indicators of market
power. For early reviews and contributions, see Schmalensee (1989), Salinger (1990), and Hall (1988).
2
The issues surrounding the proper measurement of price/cost ratios are addressed in detail in the
articles by Basu; Syverson; and Berry, Gaynor, and Scott Morton in a companion symposium in this issue.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Carl Shapiro 71

to 2016, from 1.1 to 1.15, if one includes marketing and management expenses
when measuring marginal costs. Hall (2018) finds that the weighted average ratio
of price to marginal cost increased from 1.12 in 1988 to 1.39 in 2015. The Interna-
tional Monetary Fund (2019), looking across a number of advanced economies,
finds that the ratio of price to marginal cost rose by about 8 percent between 2000
and 2016, with that increase concentrated among a small fraction of highly produc-
tive and innovative firms. Calligaris, Criscuolo, and Marcolin (2018) obtain similar
findings. The antitrust analysis below accepts as a stylized fact that the ratio of price
to marginal cost has generally risen over the past 20 to 30 years at the largest, most
efficient US firms.
Third, there is convincing evidence that larger, more efficient firms have been
growing at the expense of their smaller, less efficient rivals, causing various measures
of broad industry concentration in the US economy to increase.3 Autor et al.
(2017a, 185) state, “Our hypothesis is that technology or market conditions—or
their ­interaction—have evolved to increasingly concentrate sales among firms with
­superior products or higher productivity, thereby enabling the most successful firms
to control a larger market share.” They call these more efficient firms “­superstar
firms.” Van Reenen (2018) reaches a similar conclusion, based on extensive empir-
ical evidence: “In recent decades the differences between firms in terms of their
relative sales, productivity and wages appear to have increased in the US and many
other industrialized countries.” Bessen (2017) links increases in broad industry
concentration to the adoption of information technology. Looking across broad
industries, Ganapati (2018) finds a positive correlation between increases in the
share of revenue captured by the top four firms and growth in productivity and real
output. Crouzet and Eberly (2019) find that the growing role of intangible capital
is associated both with the rising share of the largest firms and with productivity
gains. None of these conclusions should be surprising, given the extensive literature
showing very large and durable differences in efficiency across firms, even in a given
industry (Syverson 2011; Van Reenen 2018). The growth of superstar firms may also
explain in part the long-term decline in business dynamism (Decker et al. 2016,
2017). The analysis below accepts as a stylized fact that the advantages enjoyed by effi-
cient market leaders over followers and entrants have grown considerably over time.
Fourth, labor’s share of GDP has significantly declined since the 1980s. Autor
et al. (2017a, b) show that this decline is due to the reallocation of activity toward
firms with low and declining labor shares, namely the superstar firms that have

3
This Darwinian mechanism does not appear to be operating as strongly in Europe as in the United
States. Europe has not experienced a similar increase in the share of economic activity accounted for by
the largest firms. Gutiérrez and Philippon (2018) report that from 1995 to 2015, the weighted average
Herfindahl–Hirschman index (HHI) for the United States measured at the broad sector level rose from
about 500 to about 650, while the comparable HHI in each of ten large EU countries fell from about 800
to about 600. (They find lower HHIs in the EU overall than in individual member countries.) Likewise,
Valletti (2018) finds no increase in concentration in EU countries from 2010 to 2015 at the broad sector
level. Note that these measures of concentration are uninformative regarding market power because
these broad sectors bear no relation to relevant product markets.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
72 Journal of Economic Perspectives

gained market share in recent years. The declining labor share also naturally raises
the question of whether employers have growing market power in labor markets,
especially given the declining role of unions, driving down wages and exacerbating
inequality in the distribution of income and wealth.
None of these trends necessarily indicates that US antitrust policy has been
deficient. Indeed, the fact that price/cost margins have risen in many high-income
countries suggests that growing economies of scale and globalization are the cause,
not domestic policy changes. Nonetheless, these trends compel us to take a closer
look at recent antitrust enforcement in the United States, to ask whether stronger
enforcement is now needed, and to see how that can be accomplished.
This article does not address one core aspect of antitrust enforcement: the
prohibition on cartels and price-fixing. The Department of Justice regularly brings
criminal charges against individuals who engage in price-fixing. Companies found
to have participated in cartels are subject to fines assessed by the government plus
liability for treble damages to customers who were overcharged. Over the past ten
years, the DOJ’s Antitrust Division has assessed roughly $10 billion in criminal fines
and penalties (Department of Justice 2019) and has thrown quite a few executives
in jail. This experience shows that antitrust vigilance regarding cartels is vital to
a competitive economy. Cartels continue to form and persist in many industries.
Levenstein and Suslow (2006, 2011) find that the average duration of cartels is
about eight years. Improved detection of active cartels would do much to promote
competition.

Merger Control

Under the Clayton Act, mergers that “may substantially lessen competition”
are illegal in the United States. Merger control works together with the criminal
prohibition on cartels to protect competition. Without merger control, rivals could
achieve collusive outcomes by merging.
In 2017, about 15,000 merger and acquisition deals were announced in the
United States, representing about $2 trillion in total value. Merger control policy
greatly affects the set of deals that are proposed as well as which deals obtain anti-
trust clearance and are consummated.
The Department of Justice and the Federal Trade Commission have the
authority to investigate and challenge proposed mergers before the merging parties
are permitted to consummate their merger. If the DOJ or FTC can convince a federal
judge that a merger is anticompetitive, that merger is blocked. Economic analysis
plays a central role in this process. The DOJ and the FTC publish Horizontal Merger
Guidelines that explain to the business community and the courts how they analyze
horizontal mergers (Department of Justice and Federal Trade Commission 2010).
In 2017, 2,052 proposed transactions were reported to the antitrust authori-
ties, of which 51 received an in-depth investigation and 21 were subject to an
enforcement action (Federal Trade Commission and Department of Justice 2018).

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Protecting Competition in the American Economy 73

Challenged deals often go forward after the merging parties agree to a remedy,
usually an asset divestiture designed to preserve competition. Some deals are aban-
doned in the face of an antitrust challenge, such as the proposed merger between
AT&T and T-Mobile in 2011 and the one involving Halliburton and Baker Hughes
in 2016. Very few deals are litigated in court.
The fundamental challenge for merger control is that it is a predictive exercise:
if one is seeking to identify the subset of proposed mergers that “may substantially
lessen competition,” one must assess the likely competitive effects of a proposed
merger before it is consummated.

The Gradual Weakening of Merger Control

Fifty years ago, predictions of merger effects were based largely on market
shares, and merger control was very strict. In 1963, relying heavily on the economics
literature, the Supreme Court held that any merger producing a firm that controls an
“undue percentage share” of the market and that “results in a significant increase in
the concentration of firms in that market” is “inherently likely to lessen competition
substantially” (United States v. Philadelphia National Bank, 374 US 321 [1963]). Such a
merger would be presumed to be illegal “in the absence of evidence clearly showing
that the merger is not likely to have such anticompetitive effects.” This established a
“structural presumption” against mergers based on market concentration.
As a prominent—some would say infamous—example of the structural
presumption in action, the Supreme Court upheld the Department of Justice chal-
lenge to a merger between two grocery chains with a combined market share of
7.5 percent in the retail grocery market in the Los Angeles area (United States v. Von’s
Grocery, 384 US 270 [1966]). One year later, the Court ruled out efficiencies as a
reason for allowing a merger, stating that “[p]ossible economies cannot be used as a
defense to illegality” (FTC v. Procter & Gamble, 386 US 568 [1967]). See Hovenkamp
and Shapiro (2018) and Werden (2018) for discussions of merger law in the 1960s.
The merger enforcement policies of the Department of Justice during the
1960s and 1970s reflected these Supreme Court rulings. When the DOJ published
its first Merger Guidelines in 1968, they focused very heavily on the market shares
of the merging firms. They stated, for example, that the DOJ “will ordinarily chal-
lenge” a merger between two firms with 5 percent market share each, or between a
firm with a 20 percent market share and a firm with a 2 percent market share.
Since 1968, merger enforcement has evolved significantly along two distinct
dimensions. First, the level of market concentration required to trigger the struc-
tural presumption has risen. Few if any antitrust economists today would favor
applying the low thresholds found in the 1968 Merger Guidelines, given what we
now know about the effects of horizontal mergers. Second, merger analysis now
puts less weight on market shares and more weight on other evidence to predict the
competitive effects of a merger. This shift reflects the accumulation of experience at
the Department of Justice and the Federal Trade Commission along with the recog-
nition that each industry has unique features. In Shapiro (2010), I explain this shift
and how it took place.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
74 Journal of Economic Perspectives

Remarkably, the treatment of proposed mergers under US antitrust law has

become much more lenient without Congress changing the substantive standard
to be used for merger review and without any updated guidance from the Supreme
Court, which has not heard a merger case since 1974. Rather, these changes have
resulted from a dynamic involving the lower courts, the antitrust agencies, antitrust
lawyers, and economists.
That process began in earnest in 1982, when the Department of Justice dramat-
ically revised the Merger Guidelines, giving less weight to market shares and raising
the threshold levels of concentration that would trigger the structural presumption.
Within a decade, the courts followed the DOJ’s lead. In 1990, the most influen-
tial lower court, the DC Circuit, rather brazenly departed from the precedent
established by the Supreme Court in the 1960s, stating that “[e]vidence of market
concentration simply provides a convenient starting point for a broader inquiry
into future competitiveness” (United States v. Baker Hughes, 908 F.2d 981, at 984).
This ruling weakened the structural presumption, making it harder for the DOJ and
Federal Trade Commission to block mergers.
The Department of Justice, joined by the Federal Trade Commission, further
revised its Horizontal Merger Guidelines in 1992, 1997, and 2010.4 With each revi-
sion, less weight was given to market shares and greater weight was attached to more
direct evidence about how competition has taken place in the industry and how
the merger would likely alter that competition. The 1992 guidelines introduced
“unilateral effects” into the analysis, shifting attention to the loss of direct competi-
tion between the merging firms and away from the overall structure of the market.
The 2010 guidelines introduced the concepts of upward pricing pressure, merger
simulation, and bidding and auction models into the analysis of unilateral effects.
They also addressed nonprice dimensions of merger analysis, including product
variety and innovation. In Shapiro (2010), I describe how the 2010 guidelines built
on decades of experience with the economic analysis of the competitive effects of
mergers, and in Shapiro (2012), I emphasize the importance of competition in
spurring innovation.
In principle, the antitrust agencies can more accurately distinguish procompet-
itive horizontal mergers from anticompetitive ones by undertaking detailed analyses
of how a proposed merger is likely to alter competition. However, as the Depart-
ment of Justice and the Federal Trade Commission have put more weight on direct
evidence of competitive effects and less weight on market shares, a gap has opened
up between how they evaluate mergers and how the courts do so. Economists at
the antitrust agencies engage in sophisticated analysis, interacting with economists
hired by the merging parties. But if a DOJ or FTC merger challenge is litigated, a
generalist judge may then be faced with conflicting expert testimony that is difficult
to decipher.

4
I led the team at the Department of Justice that drafted the 2010 Horizontal Merger Guidelines. Joseph
Farrell, also a professor at the University of California, Berkeley, led the team at the Federal Trade
Commission.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Carl Shapiro 75

The result is that the antitrust authorities still rely heavily on market definition,
market shares, and the structural presumption to make their case in court, even
when their enforcement decisions are based on other economic evidence, such as
bidding data, upward pricing pressure, or merger simulations. But the structural
presumption has become weaker as the lower courts place less weight on market
concentration and increasingly look for direct evidence of the likely effects of
proposed mergers (Hovenkamp and Shapiro 2018). The net result is that the anti-
trust authorities have found it harder to prevail in court, causing them to be more
cautious in the mergers they challenge. Merging firms understand this, and the
mix of proposed mergers has adjusted accordingly. This is now highly problematic,
given the mounting evidence cited above about rising profits, widening price/cost
margins, and the rise of superstar firms.

The Need for Stronger Merger Enforcement in an Economy with Superstar Firms
Accumulating evidence in two broad categories points to the need for more
stringent horizontal merger enforcement policy in the United States: (1) evidence
showing that the largest and most successful US firms have increasing market power
and (2) evidence from merger retrospectives.5
The evidence cited above shows that superstar firms are highly profitable
owing to durable competitive advantages they enjoy over their smaller rivals and
over entrants, which cannot easily or quickly replicate their assets and capabili-
ties. These are precisely the conditions under which mergers involving successful
established firms are most likely to lessen competition and harm customers. We
also know that higher price/cost margins cause the unilateral price effects from
horizontal mergers to be more harmful to customers. Likewise, research and devel-
opment competition spurs innovation if future sales are contestable (Shapiro 2012;
Federico, Scott Morton, and Shapiro forthcoming), so a merger between two firms
investing to develop competing products is likely to slow down innovation.
The case for stronger merger enforcement is perfectly consistent with a conclu-
sion that the rise of superstar firms has largely resulted from the normal competitive
process in the presence of growing economies of scale, increased globalization, and
dramatic improvements in information technology, combined with large and persis-
tent differences across firms in their ability to adopt new technologies and adjust to
changing market conditions. After all, we expect a healthy competitive process to
result in the most efficient firms gaining market share through internal growth and
earning above-normal profits. As emphasized by Valletti and Zenger (2019), merger
controls preserve that competition.
Contrary to many popular views, the case for stronger merger enforcement
does not rest on evidence showing that various broad US industries have become
more concentrated over time. Measures of industry concentration based on data

5
Azar, Schmalz, and Tecu (2018) and Schmalz (2018) argue that growing common ownership of rivals by
financial firms has weakened rivalry in many oligopolistic markets. If this claim finds additional support
in future research, it would provide an additional basis for a more stringent merger control policy.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
76 Journal of Economic Perspectives

from the US Economic Census are simply not very informative for merger analysis
because these data are available only at an aggregated level. The modest increases
in concentration observed when using these data confirm that the largest firms
are responsible for a greater portion of economic activity in many industries, but
they tell us very little about concentration in properly defined relevant antitrust
markets (Shapiro 2018).
As one important illustration of this point, a great many real-world markets—
such as hospital services, supermarkets, and automobile dealers—are local. Reported
changes in national concentration can vastly overstate or understate changes in
concentration in local markets. A local merger creating a local monopoly would
not even show up in the national measures. Conversely, if large regional firms are
growing and entering many local markets, their entry would cause concentration
in those local markets to fall. Rossi-Hansberg, Sarte, and Trachter (2018) report
that “the positive trend observed in national product-market concentration between
1990 and 2014 becomes a negative trend when we focus on measures of local
concentration.”
As another illustration, North American Industry Classification System (NAICS)
3254 is “Pharmaceutical and Medicine Manufacturing,” which encompasses a very
large number of drugs that are not substitutes for each other. Reported changes in
concentration in this four-digit industry tell us little or nothing about concentration
in the supply of drugs to treat specific ailments. One simply cannot detect overall
trends in concentration in properly defined relevant markets using data from the
Economic Census.
Furthermore, it is important to remember that an increase in concentration in a
properly defined relevant market does not prove that competition in that market has
declined. Consider an unconcentrated market in which a few of the many suppliers
become more efficient and gain share by offering lower prices and improved prod-
ucts, causing concentration to rise. That increase in concentration clearly goes
hand in hand with customer benefits and reflects the competitive process at work,
not a decline in competition. Industrial organization economists have understood
this fundamental point for at least 50 years, and probably much longer (as one
leading example, see Demsetz 1973). To properly interpret an observed increase
in market concentration, one must understand what caused concentration to rise.
Rising concentration resulting from increased efficiency by one or a few firms
reflects the competitive process at work; rising concentration caused by mergers
may well reflect a decline in competition. Distinguishing one fact pattern from the
other requires looking at properly defined individual markets and examining what
has actually happened over time in those markets.
In any event, regardless of whether one loves or loathes the rise of superstar
firms, the implications for merger analysis are unambiguous: proposed horizontal
mergers involving highly successful firms should be greeted with considerable
skepticism.
The second broad category of evidence supporting more robust horizontal
merger enforcement comes from merger retrospectives: studies of the economic

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Protecting Competition in the American Economy 77

effects of consummated mergers. The most convincing studies identify a control

product or geography and use a difference-in-difference methodology to isolate the
merger’s effect on prices. For example, Ashenfelter and Hosken (2010) look at five
mergers this way, four of which resulted in price increases for consumers. Likewise,
Whirlpool’s acquisition of Maytag led to relatively large price increases for clothes
dryers (Ashenfelter, Hosken, and Weinberg 2013). Blonigen and Pierce (2016)
find that mergers in the manufacturing sector have generally been associated with
increases in markups but not with increases in productivity.
Mergers in the health-care sector have been especially harmful to competition.
Gaynor and Town (2012) report that hospital mergers in concentrated markets
typically lead to price increases of at least 20 percent compared with control hospi-
tals. This literature also shows that hospital competition improves the quality of
care. Gaynor (2018) presents evidence that “consolidation between close competi-
tors leads to substantial price increases for hospitals, insurers, and physicians,
without offsetting gains in improved quality or enhanced efficiency.” The Center
for ­American Progress (Gee and Gurwitz 2018) assembles a range of evidence also
supporting this conclusion.
Kwoka (2014) provides the most comprehensive review of merger retrospec-
tives. Most mergers in these studies are older ones that took place in a few industries
where data were available: banking, hospitals, airlines, petroleum, and journal
publishing. Kwoka reports that “most studied mergers result in competitive harm,
usually in the form of higher prices” (158). Ashenfelter, Hosken, and Weinberg
(2014) review these studies, stating that “[t]he empirical evidence that mergers can
cause economically significant increases in price is overwhelming” (S78).
While the overall body of evidence from merger retrospectives, standing alone,
does not allow us to predict with confidence the effects of any given merger, it does
indicate that merger enforcement has been too lax over the past 25 years.

How to Reinvigorate Merger Enforcement

Merger enforcement can be strengthened in a number of ways, if the Depart-
ment of Justice and the Federal Trade Commission choose to move in that direction
and if the courts cooperate.
First, the structural presumption against mergers that increase concentration
in a properly defined relevant market could be strengthened. For example, after
the government has defined a relevant market based on substantial evidence, the
merging parties could be required to present clear and convincing evidence to
contest that definition of the market. Similarly, once the government has estab-
lished the structural presumption, the merging parties could be required to present
clear and convincing evidence to rebut the presumption. Claims by the merging
parties that growth by small rivals or entry of new firms into the market will quickly
and effectively restore any competition lost due to the merger could be greeted with
greater skepticism.
Second, more weight could be placed on evidence that the merging parties
are significant direct competitors, without the necessity of defining a relevant

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
78 Journal of Economic Perspectives

market and measuring market shares in that market, and without requiring the
government to quantify the harm to customers that the merger will cause. For
example, strong evidence that customers have often obtained lower prices as a
result of direct competition between the two merging firms could, at least in some
cases, be regarded as sufficient for the government to meet its initial burden of
showing that the merger “may substantially lessen competition.”
Third, the agencies and the courts could express greater wariness when a
dominant incumbent firm seeks to acquire a firm operating in an adjacent market,
especially if the target firm is well positioned to challenge the incumbent’s posi-
tion in the foreseeable future. In the language of antitrust law, this would involve
lowering the evidentiary requirements necessary for the government to prevail in
a merger case based on a loss of “potential competition.” For example, the govern-
ment could meet its initial burden by showing that the target firm is reasonably
likely to become a rival to the acquiring firm in the foreseeable future, even if the
target firm has not yet made specific plans to do so. This change would reduce the
ability of powerful firms to acquire potential rivals before they mature into actual
rivals, without stopping them from making acquisitions to improve their offerings
or to challenge other firms with entrenched positions.
This change would be especially consequential as applied to dominant firms
in the tech sector. Under this standard, Facebook’s acquisitions of Instagram
and WhatsApp might well have been blocked, if these firms were seen as well
placed to mature into rivals to Facebook as social media platforms, and Google’s
earlier acquisitions of YouTube and DoubleClick would at least have warranted
greater scrutiny. But it seems unlikely that Amazon’s acquisition of Whole Foods
or Google’s acquisition of Nest would have raised serious issues even under this
stricter standard. More acquisitions by the tech titans involving important inputs
or complements could be challenged, but it is unclear how the courts would
respond to cases involving such vertical or complementary mergers.
Fourth, the courts could insist that any “fixes” to proposed mergers result in a
market structure that preserves competition. This would involve skepticism about
divestitures designed to obtain antitrust approval that do not make good business
sense, and placing little or no weight on behavioral commitments by the merged
entity, such as a commitment not to raise price.
Lastly, the Department of Justice and the Federal Trade Commission could be
given more resources to investigate and challenge mergers. With more resources,
the antitrust agencies could look more closely at more suspect proposed mergers.
They also could investigate more consummated mergers to see whether they have
harmed competition or are likely to do so, including mergers that were below the
size-of-transaction reporting threshold, which is $90 million in 2019. Cunningham,
Ederer, and Ma (2018) find that acquired pharmaceutical projects are less likely
to be developed when they overlap with the product portfolio of the acquiring
firm; these “killer acquisitions” disproportionately occur just below the reporting
threshold. Wollmann (2019) also provides worrisome evidence about mergers
taking place just below the threshold.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Carl Shapiro 79

As a practical matter, the case law relating to mergers evolves very slowly, with
substantial lags following advances in economic learning and then changes in
Department of Justice and Federal Trade Commission merger enforcement poli-
cies. Whether the current judiciary has the appetite to support stronger merger
enforcement remains to be seen. If not, or if that route is too slow, Congress would
need to pass new legislation codifying changes such as the ones suggested above—a
heavy lift to be sure.

Antitrust and the Tech Titans

The most talked-about antitrust question of the day is whether and how antitrust
should act to limit the economic power of the largest tech firms, often identified
as Amazon, Apple, Facebook, and Google. Should they be broken up? Forced to
modify their business practices and pay fines for their past sins? Watched carefully?
Left alone and applauded?
A first step toward answering these questions is to recognize that the goal of
antitrust policy is to protect and promote competition. Antitrust is not designed
or equipped to deal with many of the major social and political problems associ-
ated with the tech titans, including threats to consumer privacy and data security,
or with the spread of hateful speech and fake news. Indeed, it is not clear that
more competition would provide consumers with greater privacy or would better
combat information disorder; unregulated competition might instead trigger a race
to the bottom, and many smaller firms might be harder to regulate than a few large
ones. Addressing these major problems requires sector-specific regulation, which is
beyond the scope of this article.
Three important and insightful reports have recently been released addressing
antitrust in the digital economy: one by the United Kingdom (Furman et al. 2019),
one by the European Commission (Crémer, de Montjoye, and Schweitzer 2019),
and one by the Stigler Center at the University of Chicago (Scott Morton et al.
2019). All three reports conclude that antitrust can and should do more to promote
competition in the digital era, while staying true to its focus on competition issues.
All three reports call for regulation to address other public policy issues relating to
digital platforms.
Within the realm of antitrust, it is important to understand that under the
Sherman Act, it is not illegal for a company to have a monopoly, so long as that
position was achieved by offering customers attractive products and services. There
is a broad consensus behind this approach, because it would be illogical to urge
companies to compete and then tell them they have broken the law when they do
so successfully. Judge Learned Hand famously captured this idea in United States v.
Aluminum Company of America (148 F.2d 416, Second Circuit [1945]):

A single producer may be the survivor out of a group of active competitors,

merely by virtue of his superior skill, foresight and industry. In such cases a

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
80 Journal of Economic Perspectives

strong argument can be made that, although the result may expose the public
to the evils of monopoly, the Act does not mean to condemn the resultant of
those very forces which it is its prime object to foster: finis opus coronat. The
successful competitor, having been urged to compete, must not be turned
upon when he wins.

Following this core principle, the basic antitrust question for each tech titan is
whether that company has engaged in practices that go beyond competition on
the merits and are likely to (1) exclude its rivals and fortify its market position
or (2) extend its power to adjacent markets. If so, a remedy is needed to restore
competition. A behavioral remedy imposes limits and obligations on the company’s
conduct; this was the outcome in the Microsoft case 20 years ago. A structural
remedy breaks up the company; this was the result in the AT&T case in the 1980s.
Talk of breaking up the tech titans without reference to a specific antitrust violation
is putting a very large cart before the horse.6

The Shrinking Scope of the Sherman Act

The portion of the Sherman Act dealing with monopolies is remarkably
broad—and vague. Section 2 states, “Every person who shall monopolize, or attempt
to monopolize . . . any part of the trade or commerce among the several States . . .
shall be deemed guilty of a felony.”
From the outset, it was clear that the courts would play a major role in inter-
preting the broad language of the Sherman Act. As discussed in the companion
piece in this symposium by Naomi Lamoreaux, the role of antitrust in the American
economy has waxed and waned depending on judicial rulings, which have evolved
in response to economic learning and changing market conditions as well as polit-
ical forces and the makeup of the Supreme Court.
For many years, the Supreme Court recognized the expansive nature of the
antitrust statutes. In 1958, the Court stated, “The Sherman Act was designed to be
a comprehensive charter of economic liberty aimed at preserving free and unfet-
tered competition as the rule of trade” (Northern Pacific Railway v. United States,
356 US 1 [1958], at 4). In 1972, the Court stated, “Antitrust laws in general, and
the Sherman Act in particular, are the Magna Carta of free enterprise. They are as
important to the preservation of economic freedom and our free-enterprise system
as the Bill of Rights is to the protection of our fundamental personal freedoms”
(United States v. Topco, 405 US 596 [1972], at 610).
The high-water mark for antitrust in the United States was reached during the
1960s. Since then, the Supreme Court has substantially narrowed the scope of the
antitrust laws. This narrowing took place along multiple dimensions.

6
As discussed above, a breakup to unwind a prior anticompetitive acquisition could well be a suitable
remedy and would not be novel as a legal matter. This section considers exclusionary conduct by domi-
nant firms, not mergers.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Protecting Competition in the American Economy 81

A number of business practices that previously were treated as automatically or

per se anticompetitive are now evaluated on a case-by-case “Rule of Reason” basis.
For example, it used to be per se illegal for a manufacturer to assign territories
to its distributors and to prevent one distributor from selling outside its assigned
territory. But in 1977, the Supreme Court ruled that this practice would in the
future be evaluated using the Rule of Reason (Continental TV v. GTE Sylvania,
433 US 36 [1977], overruling United States v. Arnold Schwinn and Company, 388 US
365 [1967]). Likewise, retail price maintenance—when a manufacturer prohibits a
retailer from selling its products below a specified price—used to be per se illegal.
In 2007, the Court ruled that it would be evaluated using the Rule of Reason (Leegin
Creative Leather Products v. PSKS, 551 US 877 [2007], overruling Dr. Miles Medical
Company v. John D. Park & Sons Company, 220 US 373 [1911]).
The Court also erected obstacles to antitrust plaintiffs in situations when the
Rule of Reason is applied. For example, in 1993 the Court ruled that a plaintiff in a
predatory pricing case had to show that the monopolist was selling below cost and
that the monopolist would be able to recoup the losses resulting from this below-cost
pricing by charging higher prices in the future (Brooke Group v. Brown & Williamson
Tobacco Corporation, 509 US 209 [1993]). Under this standard, a predatory pricing
case by the Department of Justice against American Airlines failed (United States v.
AMR Corporation, 335 F.3d 1109, Tenth Circuit [2003]). Similarly, in 2004, the Court
significantly narrowed the set of circumstances in which an antitrust plaintiff could
win on the basis of a monopolist’s refusal to sell an essential input to a rival (Verizon
Communications v. Law Offices of Curtis V. Trinko, 540 US 398 [2004], narrowing Aspen
Skiing v. Aspen Highlands Skiing, 472 US 585 [1985]).7
Collectively, these and other cases represent a significant backing away from
antitrust by the Supreme Court. Baker (2015) argues forcefully that many of
these judicial decisions were based on erroneous assumptions, including these:
that markets self-correct through entry, that oligopolists compete and cartels are
unstable, and that business practices prevalent in competitive markets cannot harm
competition. This history suggests that it will be challenging for the government
to bring a successful Sherman Act case against the tech titans. But ultimately that
will depend on the specific facts of the case. A strong case can withstand these
headwinds.
The Microsoft case (United States v. Microsoft, 253 F.3d 34, DC Circuit [1998])
provides the best guide to what constitutes monopolization in a high-tech setting.
This case should be encouraging for those in favor of antitrust action against the tech
titans. Microsoft was found to have monopolized the market for ­Intel-compatible
operating systems for personal computers, on the basis of conduct that excluded
the Netscape browser and Java software, which together might have facilitated entry

7
Based on Trinko, the Court subsequently limited the circumstances under which a plaintiff can bring
an antitrust case based on a price squeeze (see Pacific Bell Telephone Company v. LinkLine Communications,
555 US 438 [2009]).

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
82 Journal of Economic Perspectives

and thus eroded the monopoly power of Microsoft Windows. For an overview of the
arguments, see the three-paper symposium in the Spring 2001 issue of this journal.

Applying Antitrust Principles to the Tech Titans

We are now ready to look more closely at Google, Amazon, Facebook, and
Apple. These four “GAFA” firms have received by far the most antitrust attention
of late as they have become economically and socially important. However, they are
not the only firms with powerful positions in the information economy. One could
also look for exclusionary conduct by other large high-tech firms, including Micro-
soft, Oracle, IBM, Salesforce, Adobe, and Uber. Expanding to telecom and media,
the list could include AT&T, Verizon, Comcast, Netflix, and Disney. One also could
conduct a similar exercise in other industries, such as pharmaceuticals, semicon-
ductors, or airlines.
The point of this discussion is not to offer a view on whether any of these four
GAFA firms have violated US antitrust law, but rather to show how antitrust principles
can be applied to these tech titans. The analysis is illustrative and is not intended to
capture all plausible antitrust cases that might be brought against these firms. Each of
these companies is assumed to have sufficient market power to be subject to Section
2 of the Sherman Act. However, is important to remember that these firms’ products
and services, and their business models, are very different from one another, so it
makes no sense to lump them together. Any antitrust analysis must be done company
by company, based on that company’s practices. Here we briefly consider the potential
application of three antitrust doctrines to the tech titans: predatory pricing, exclusion
of nascent threats, and extension of market power into adjacent markets.
A monopolist that engages in predatory pricing—that is, pricing below cost to
drive rivals out of the market so it can then recoup those losses by raising prices to
monopoly levels—can be guilty of monopolization. One can ask whether Amazon’s
(or Uber’s) conduct falls into this category, as some have alleged.
Amazon’s core online retailing business has clearly generated enormous bene-
fits to consumers by offering low prices, a huge variety of products, and reliable
and speedy delivery. Amazon has been a highly successful company, putting great
competitive pressure on other retailers. Under Supreme Court precedent, a preda-
tory pricing case against Amazon would fail unless Amazon were shown to have
priced below cost. Furthermore, showing the prospect of future harm to consumers
would also be a necessary element of any case. Requiring plaintiffs in predatory
pricing cases to show some prospect of harm to customers, not just harm to compet-
itors, is critical to the coherence of antitrust policy, to avoid chilling legitimate price
competition.8 Showing such harm to consumers from Amazon’s aggressive pricing

8
In an earlier era, the grocery chain A&P offered consumers a wide range of products at low prices,
putting pressure on many smaller grocery stores. A&P was successfully prosecuted by the US Department
of Justice, an action now widely seen as misguided (Muris and Neuchterlein 2018). Later, Walmart was
the innovator in retailing, putting great competitive pressure on smaller retailers around the country.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Carl Shapiro 83

and growth strategy would appear to be difficult. Hemphill and Weiser (2018) offer
a road map to bringing and deciding predatory pricing cases.
A tech titan also could be challenged for exclusion of nascent threats, which means
using its dominant position to exclude products or services that it fears may grow
to threaten its core business. This was the basic economic logic of the case against
Microsoft 20 years ago. The Netscape browser and Java were the potential threats to
Microsoft Windows. The Microsoft case established antitrust liability for a dominant
firm that excludes rivals, even if the threats they pose are “nascent.” But the reach
of the Microsoft case is unclear, since the Supreme Court subsequently ruled in the
Trinko case that a dominant firm normally has no duty to deal with its rivals. If the
Supreme Court applies Trinko broadly to the tech titans, then separate regulation
might be needed to impose on the tech titans mandated interconnection or data
sharing with rivals.
One way a dominant firm can exclude rivals is by refusing to sell its product to
customers who also purchase from its rivals. In 1951, the Supreme Court ruled that
such “exclusive dealing” violated Section 2 of the Sherman Act. That case involved a
dominant local newspaper that refused to accept advertisements from those who also
placed advertisements on the local radio station (Lorain Journal v. United States, 342
US 143 [1951]). A tech titan putting up obstacles to customers seeking to also use
rival products could easily face liability under this precedent. As a recent example
of exclusionary conduct, Facebook blocked Vine, a video sharing app launched by
Twitter in January 2013, from accessing Facebook user data (O’Sullivan and Gold
2018). This prevented Facebook users from inviting their Facebook friends to join
Vine. Facebook was applying its policy of restricting access to apps that replicated
Facebook’s core functionality. In response to the Vine episode becoming public,
Facebook stated that it was dropping this policy (Facebook 2018), which appears
difficult to defend. Twitter discontinued the Vine mobile app in October 2016.
A third category of candidate antitrust cases against the tech titans involves
allegations that a firm is abusing its dominant position to expand into adjacent markets.
Several tech titans already face cases that focus on whether they have favored their
own service over a rival service in an adjacent market.
The European Commission is investigating whether Amazon is using data
collected about third-party sellers on its platform to guide its own product offerings
in competition against those third-party sellers. Such conduct could be problematic,
but note that when one firm simply imitates its rivals, that is normally an important
channel for the diffusion of new ideas, so long as the imitation does not involve any
breach of contract or infringement of intellectual property rights. Looking more
broadly, Amazon could be accused of favoring its own products over those of third
parties selling on its platform, by giving its own products preferred placement on
Amazon’s website or by charging third parties excessive rates to be fulfilled or sold
through Amazon. This last type of case would be very difficult in the United States
under Supreme Court precedent.
Apple has been accused of discriminating against rivals who rely on the Apple
platform to reach consumers. In March 2019, the music streaming service Spotify

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
84 Journal of Economic Perspectives

filed an antitrust complaint at the European Commission against Apple (Ek 2019).
Spotify objected to the 30 percent fee that Apple charges on certain purchases
made through Apple’s payment system and claimed that Apple had locked Spotify
out of Apple Watch. Spotify asserted that it should receive the same treatment at
the Apple App Store given to Apple’s competing service, Apple Music. In response,
Apple claimed that it had worked closely with Spotify for years and that Spotify was
not willing to abide by the same rules that apply to all apps on the App Store, which
Apple regards as necessary for the operation and security of the App Store. Apple
further claimed that Apple approved Spotify for the Apple Watch and that Spotify
has been the leading app in the Watch Music category. The European Commission
is opening an investigation in response to Spotify’s complaint.
The Spotify complaint illustrates the tensions that arise when the company
controlling a platform also offers its own services on that platform. Indeed, the
boundary between the “platform” and services running on that platform can be
fuzzy and can change over time. Similar issues will surely arise for other applications
that rely on Apple’s App Store to reach customers. For example, Apple recently
removed several parental control apps from the App Store. These apps provide
alternatives to Apple’s own screen-time control tools. Apple explained that it took
this action to protect users’ privacy and security, but an antitrust complaint here
would not be shocking (Apple 2019).
We know a lot more about what a case of this type might look like against Google,
because the European Commission issued an antitrust decision in June 2017 against
Google involving Google Shopping, including a €2.42 billion ($2.7 billion) fine.9
Google displays advertisements when users enter queries into the Google search
engine that relate to commercial products. For example, Figure 1 shows what one
sees on a desktop computer if one searches for “Nikon Cameras” on Google.
All of the images displayed in Figure 1 are sponsored search results; that is, they
are advertisements paid for by online merchants. Google calls these “Product
Listing Ads.” A user who clicks on one of these ads is directed to the website of the
online merchant sponsoring that ad, and that sponsor pays a fee to Google. The
first link to NikonUSA.com is also an advertisement, in text form. The second link
to NikonUSA.com is a generic search result generated by Google’s algorithm, not an
advertisement. More generic search results, not shown in Figure 1, follow.
Many press reports have left the impression that the European Commission
case was about Google biasing its search algorithm by demoting its rivals, but
that is not correct. The European Commission fact sheet states, “The Commission
Decision does not object to the design of Google’s generic search algorithms or
to demotions as such, nor to the way that Google displays or organizes its search
results pages (e.g., the display of a box with comparison shopping results displayed
prominently in a rich, attractive format)” (European Commission 2017). Instead,
the European Commission “objects to the fact that Google has leveraged its

9
I served as an economic expert for Google in the Google Shopping case. The views expressed here are
my own.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Protecting Competition in the American Economy 85

Figure 1
Result of a Search for “Nikon Cameras” on Google

Source: Unaltered screen capture of a Google search for “Nikon Cameras” performed by the author in
November 2018.
Notes: All of the images in the main column are sponsored search results; that is, they are advertisements
paid for by online merchants. A user who clicks on one of these ads is directed to the website of the online
merchant sponsoring that ad, and that sponsor pays a fee to Google. The first link to NikonUSA.com is
also an advertisement, in text form. The second link to NikonUSA.com is a generic search result generated
by Google’s algorithm, not an advertisement. Additional generic search results, not shown in the figure,
follow.

market dominance in general internet search into a separate market, comparison

shopping. Google abused its market dominance as a search engine to promote
its own comparison shopping service in search results, whilst demoting those
of rivals.”
According to the European Commission, Google did this by displaying
Product Listing Ads, such as those shown in Figure 1. This is a peculiar claim,
because those Product Listing Ads are very much like the text ads that Google
has shown for years, and the European Commission does not object to ads that
use text rather than images. Plus, as the European Commission recognizes, there
is nothing wrong from a competitive perspective when a content provider earns
revenue by selling advertisements. The newspaper and radio industries have done
that for a very long time. Furthermore, it is not apparent how the Product Listing
Ads “promote” Google’s comparison shopping service, since a user who clicks
on one of those ads is directed to the merchant’s website, not to the stand-alone
Google Shopping site.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
86 Journal of Economic Perspectives

Lessons
We can distill several lessons regarding the role and limits of antitrust in control-
ling the tech titans.
First, many of the deepest concerns about the tech titans, including privacy,
data security, and information disorder, do not directly involve competition issues.
Sector-specific regulation is overdue and badly needed to address these problems.
Antitrust cannot solve all manner of economic and social problems and should not
be expected to do so.
Second, those who would like to see the tech titans broken up on antitrust
grounds are likely to be disappointed, since antitrust does not condemn monopoly
as such. Antitrust liability requires that a dominant firm abuse its power in some way,
such as by excluding rivals. When liability is found, a suitable remedy is designed to
restore the competition lost due to the illegal acts.
Third, establishing that a dominant firm has abused its position has become
much harder in the United States over the past 40 years, under a series of decisions
by the Supreme Court. This has left antitrust enforcement in the United States
notably weaker than in the European Union. That is unlikely to change much in the
near future, so look to Brussels for much of the action.
Fourth, many of the antitrust cases against the tech titans in the years ahead will
most likely involve allegations that these firms have used their “platforms” to favor
their own services over rival services. These cases will be complex, and they will be
risky for plaintiffs in the United States given Sherman Act case law and the current
makeup of the Supreme Court.10
Despite these obstacles, ongoing antitrust oversight and vigilant antitrust
enforcement toward the tech titans is critical and can make a real difference, mostly
through deterrence rather than litigation. Antitrust can and should prevent the
tech titans from entrenching their economic power by engaging in exclusionary
conduct that weakens the competitive pressures they face from rivals offering new
and disruptive products and services. At the same time, antitrust should take care not
to discourage the tech titans from competing with each other, as Microsoft has done
with Bing against Google and as Google has done with Android against Apple iOS.

Antitrust in Labor Markets

Antitrust enforcement in the United States has largely focused on the market
power of sellers. However, the Sherman Act applies with equal force to the market

10
Adding to concerns, the Supreme Court recently issued a worrisome and poorly reasoned decision
involving payment systems that could greatly complicate any case brought in a “two-sided market” (Ohio
v. American Express, 138 S. Ct. 2274 [2018]). Justice Breyer’s dissent in this case eviscerates the majority
opinion. While it is not yet clear how broadly the lower courts will read the American Express decision, all
four of the tech titans arguably operate in “two-sided markets.” Apple connects users with application
developers, Facebook and Google connect advertisers with users, and Amazon connects manufacturers
and merchants with consumers.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Carl Shapiro 87

power of buyers. When Weyerhaeuser was sued as a buyer for bidding up the price of
sawlogs in the Pacific Northwest to a level that prevented rival sawmills from being
profitable, the Supreme Court ruled that the legal standard for predatory pricing
also applies to cases of predatory bidding (Weyerhaeuser v. Ross-Simmons Hardwood
Lumber, 549 US 312 [2007]). Last year, the Federal Trade Commission (2018a) held
hearings to explore monopsony and buyer power in the US economy.
Likewise, Section 7 of the Clayton Act bans mergers that may create or
enhance buyer market power. For example, in 2018, the Federal Trade Commis-
sion challenged the acquisition by Grifols of Biotest, a challenge based in part on
the allegation that the acquisition would lessen competition to purchase human
plasma, leading to lower fees for people providing plasma (see the case summary
at Federal Trade Commission 2018b). Hemphill and Rose (2018) discuss ways to
strengthen antitrust enforcement relating to mergers that harm sellers.
The Sherman Act as applied to labor markets prohibits agreements among
employers to refrain from competing to hire workers, while the Clayton Act prohibits
mergers between employers that may substantially lessen competition in the hiring
of workers. Antitrust is generally associated with keeping consumer prices down by
controlling seller market power, but antitrust applies equally to keeping workers’
wages up by controlling employer market power.
It seems clear cut that many labor markets depart rather significantly from the
textbook model of perfect competition, in which employers are wage takers and
face a highly elastic supply of labor. Labor markets are generally defined according
to an occupation and a local geographic area (as emphasized by Moretti 2011).
With costs of job search and costs of geographical mobility, employers will have
some degree of buyer power. Manning (2011) surveys the literature and concludes
that “labor markets are pervasively imperfectly competitive.” Employers commonly
share ­relationship-specific rents with workers, so employees working at more produc-
tive firms earn higher wages (see, for example, Kline et al. 2017; Card et al. 2018).
Some local US labor markets are highly concentrated on the employer side, but
that is not the situation for most workers. Azar, Marinescu, and Steinbaum (2017)
use data from CareerBuilder.com to calculate labor market concentration in some
8,000 selected labor markets in the United States. They define these labor markets
according to occupation and geography, such as “legal secretaries in the Denver
area.” On average, 20 employers post job vacancies on CareerBuilder.com in a given
market in a given quarter. They calculate an employer’s share in the labor market
based on the number of vacancies listed by that employer at CareerBuilder.com
in a given quarter, and they measure market concentration based on the Herfin-
dahl–Hirschman index (HHI) on the employer side of the market. Weighting the
geographic markets by population, the overall mean HHI is 1,691, which antitrust
economists would classify as moderately concentrated. This method is likely to
overestimate labor market concentration, because only about 35 percent of job
openings nationally are listed on CareerBuilder.com.
Antitrust enforcement in labor markets has historically been extremely limited.
As discussed by Naidu, Posner, and Weyl (2018), this most likely reflects the view

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
88 Journal of Economic Perspectives

that most labor markets are reasonably competitive and that most employers face
effective competition to attract and retain workers, combined with a view that some
combination of unions, regulations, and lawsuits will help protect workers. That
overall conclusion is probably true, but antitrust can still play a role in labor markets
in two ways: by considering employer power in labor markets in selected mergers
and by addressing anticompetitive agreements in labor markets.

Merger Policy That Considers Labor Markets

A merger that may substantially lessen competition among employers to hire
workers is illegal under the Clayton Act. Marinescu and Hovenkamp (2018, 1) note
that no merger has ever been blocked on these grounds and infer that “the anti-
trust law against anticompetitive mergers affecting employment markets is certainly
underenforced, very likely by a significant amount.” Prager and Schmitt (2019) find
that hospital mergers resulting in large increases in concentration in markets for
skilled workers, including nurses and pharmacy workers, lead to lower wages.
The Horizontal Merger Guidelines (Department of Justice and Federal Trade
Commission 2010, sec. 12) explain how the government evaluates mergers that may
enhance buyer power. The government could define a relevant labor market and
demonstrate that the merger in question would cause that market to become signifi-
cantly more concentrated. The merging parties might then try to show that the
affected workers have many alternative options for employment. For further details
on this type of analysis, see Marinescu and Hovenkamp (2018) and Naidu, Posner,
and Weyl (2018).
Two thorny issues are likely to arise if the government begins challenging
mergers on the basis of harm to competition in labor markets. First, in cases where
the merging parties assert that the merger will reduce their labor costs, the court
may need to determine whether to credit these reduced costs as an efficiency gain
or instead treat them as the exercise of buyer power in labor markets.11 Second, if a
merger is expected to benefit consumers but harm workers, the court may need to
determine whether and how to balance the interests of these two groups. Marinescu
and Hovenkamp (2018) argue that under current law, a merger that harms workers
by lessening competition in the labor market would not be saved by also offering
benefits to consumers.
If the antitrust authorities seriously want to explore the possibility of chal-
lenging mergers on the basis of harm to competition in labor markets, developing
a quick and efficient means of identifying mergers that involve a significant overlap
in plausible labor markets would be a good first step.

Anticompetitive Labor Market Practices: No-Poach and No-Hire Agreements

Section 1 of the Sherman Act prohibits agreements among employers to
refrain from competing to hire workers, just as it prohibits traditional cartels among

11
Anthem’s claimed purchasing efficiencies were rejected in United States v. Anthem, 855 F.3d 345 (2017).

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Protecting Competition in the American Economy 89

product-market rivals. This raises questions about no-poach and no-hire agreements
that arise in certain labor markets.
In a prominent “no-poach” case, the Department of Justice (2010) sued Adobe,
Apple, Google, Intel, Intuit, and Pixar for entering into agreements not to recruit
certain workers from each other.12 When Apple CEO Steve Jobs learned that Google
was trying to recruit employees from Apple’s Safari team, Jobs threatened Google
co-founder Sergey Brin, stating that “if you hire a single one of these people, that
means war.” In response, Google’s CEO Eric Schmidt stopped all efforts at Google
to recruit anyone from Apple. When this was conveyed to Apple, Apple recipro-
cated (Koh 2014). Later, when a Google recruiter contacted an Apple employee,
Jobs complained to Schmidt, who apologized and made a public example out of
that recruiter, who was terminated within the hour.
The Department of Justice and the Federal Trade Commission later released
Antitrust Guidance for Human Resources Professionals, stating that “[g]oing forward,
the DOJ intends to proceed criminally against naked wage-fixing or no-poaching
agreements. These types of agreements eliminate competition in the same irre-
deemable way as agreements to fix product prices or allocate customers, which
have traditionally been criminally investigated and prosecuted as hardcore cartel
conduct” (Department of Justice and Federal Trade Commission 2016). Notice that
this guidance refers to “naked wage-fixing or no-poaching agreements.” A no-poach
agreement between two or more companies could be justified if those companies
are engaged in legitimate joint activity, such as a joint venture to develop new prod-
ucts, and if the no-poach agreement is confined to employees involved in that joint
activity, especially if the joint activity involves training these employees or providing
them with access to confidential information.
No-hire agreements are common in the franchise sector. Krueger and
Ashenfelter (2018) report that in 58 percent of major franchisors’ contracts with
franchisees, including McDonald’s, Burger King, and Jiffy Lube, one franchisee is
prohibited from hiring workers from another franchisee in the same chain. They
find that no-hire agreements are more common in low-wage, high-turnover indus-
tries and have become more common over the past 20 years.
Some limited no-hire provisions of this type could be justified if they provide
an incentive for franchisees to invest in workers, giving them human capital that is
specific to the franchisor but not to the franchisee. As a result, these agreements
are more difficult to challenge under antitrust than are “naked” no-hire agreements.
Krueger and Posner (2018) describe a court case involving Jack-in-the-Box in which
such a challenge failed. Under the Rule of Reason analysis typically used in antitrust
to analyze agreements of this type, two important considerations will be how signifi-
cantly these agreements restrict the number of employment options available to
workers and whether they have depressed wages. A quick look may be sufficient to

12
I was the chief economist at the Department of Justice when this case was brought.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
90 Journal of Economic Perspectives

determine that a no-hire provision has no real efficiency justification and tends to
suppress wages.

Conclusion

American antitrust laws date from a time when changes in transportation,

communications, and manufacturing technologies generated unprecedented econ-
omies of scale, fueling the rise of industrial behemoths. Today, dramatic advances
in information technology, combined with globalization, are fueling the growth of
large and efficient “superstar firms” that are capturing a growing share of economic
activity. The emergence of the tech titans is especially dramatic.
These economic conditions call for a reinvigoration of antitrust enforcement
in the United States to promote competition, protect consumers and workers, and
spur economic growth. These valuable aims can be achieved by taking a tougher
stance toward mergers involving market leaders and by vigilantly preventing domi-
nant firms from engaging in conduct that excludes their rivals. However, moving
in that direction is a slow process, requiring the antitrust enforcement agencies to
take the lead and convince inertial and possibly skeptical courts to follow. Those
who expect dramatic and rapid changes in antitrust will be disappointed, unless
new legislation is passed. Likewise, those who expect antitrust to solve problems
unrelated to competition will be disappointed. Stronger antitrust enforcement,
while needed, is not a substitute for badly needed regulations directed at reducing
the political influence of corporations, protecting privacy and data security, and
limiting the spread of disinformation.

■ I thank Jonathan Baker, Aaron Edlin, Joseph Farrell, Steven Salop, Fiona Scott Morton,
and Steve Tadelis for their valuable comments on earlier drafts.

References

Apple. 2019. “The Facts about Parental Control Ashenfelter, Orley C., Daniel S. Hosken, and
Apps.” Apple Press Release, April 28, 2019. Michael C. Weinberg. 2013. “The Price Effects of
https://www.apple.com/newsroom/2019/04/ a Large Merger of Manufacturers: A Case Study
the-facts-about-parental-control-apps/. of Maytag-Whirlpool.” American Economic Journal:
Ashenfelter, Orley C., and Daniel S. Hosken. Economic Policy 5(1): 239–61.
2010. “The Effect of Mergers on Consumer Prices: Ashenfelter, Orley C., Daniel S. Hosken, and
Evidence from Five Mergers on the Enforcement Michael C. Weinberg. 2014. “Did Robert Bork
Margin.” Journal of Law and Economics 53(3): Understate the Competitive Impact of Mergers?
417–66. Evidence from Consummated Mergers.” Journal

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Carl Shapiro 91

of Law and Economics 57(S3): S67–100. research.

Autor, David, David Dorn, Lawrence F. Katz, Decker, Ryan A., John Haltiwanger, Ron S.
Christina Patterson, and John Van Reenen. 2017a. Jarmin, and Javier Miranda. 2016. “Declining
“Concentrating on the Fall of the Labor Share.” Business Dynamism: What We Know and the
American Economic Review 107(5): 180–85. Way Forward.” American Economic Review 106(5):
Autor, David, David Dorn, Lawrence F. Katz, 203–7.
Christina Patterson, and John Van Reenen. 2017b. Decker, Ryan A., John Haltiwanger, Ron S.
“The Fall of the Labor Share and the Rise of Jarmin, and Javier Miranda. 2017. “Declining
Superstar Firms.” NBER Working Paper 23396. Dynamism, Allocative Efficiency, and the Produc-
Azar, José, Ioana Marinescu, and Marshall I. tivity Slowdown.” American Economic Review
Steinbaum. 2017. “Labor Market Concentration.” 107(5): 322–26.
NBER Working Paper 24147. Demsetz, Harold. 1973. “Industry Structure,
Azar, José, Martin C. Schmalz, and Isabel Market Rivalry, and Public Policy.” Journal of Law
Tecu. 2018. “Anticompetitive Effects of Common and Economics 16(1): 1–9.
Ownership.” Journal of Finance 73(4): 1513–64. Department of Justice. 2010. “Justice Depart-
Baker, Jonathan B. 2015. “Taking the Error ment Requires Six High Tech Companies to
Out of ‘Error Cost’ Analysis: What’s Wrong with Stop Entering into Anticompetitive Employee
Antitrust’s Right.” Antitrust Law Journal 80(1): Solicitation Agreements.” Department of
1–38. Justice Press Release 10-1076, September
Baker, Jonathan B. 2019. The Antitrust Para- 24, 2010. https://www.justice.gov/opa/pr/
digm: Restoring a Competitive Economy. Cambridge, justice-department-requires-six-high-tech-compa-
MA: Harvard University Press. nies-stop-entering-anticompetitive-employee.
Barkai, Simcha. 2017. “Declining Labor and Department of Justice. 2019. Criminal Enforce-
Capital Shares.” http://facultyresearch.london. ment Trends Charts through Fiscal Year 2018. https://
edu/docs/BarkaiDecliningLaborCapital.pdf. www.justice.gov/atr/criminal-enforcement-fine-
Bessen, James E. 2017. “Information Tech- and-jail-charts.
nology and Industry Concentration.” Boston Department of Justice and Federal Trade
University School of Law, Law and Economics Commission. 2010. Horizontal Merger Guidelines.
Research Paper 17-41. https://ssrn.com/ https://www.justice.gov/atr/file/810276/down-
abstract=3044730. load.
Blonigen, Bruce A., and Justin R. Pierce. 2016. Department of Justice and Federal Trade
“Evidence for the Effects of Mergers on Market Commission. 2016. Antitrust Guidance for Human
Power and Efficiency.” NBER Working Paper Resources Professionals. https://www.justice.gov/
22750. atr/file/903511/download.
Calligaris, Sara, Chiara Criscuolo, and Ek, Daniel. 2019. “Consumers and Innova-
Luca Marcolin. 2018. “Mark-Ups in the tors Win on a Level Playing Field.” Spotify Press
Digital Era.” OECD Science, Technology and Release, March 13, 2019. https://newsroom.
Industry Working Paper 2018/10. https://doi. spotify.com/2019-03-13/consumers-and-innova-
org/10.1787/4efe2d25-en. tors-win-on-a-level-playing-field/.
Card, David, Ana Rute Cardoso, Joerg Heining, European Commission. 2017. “Antitrust:
and Patrick Kline. 2018. “Firms and Labor Market Commission Fines Google €2.42 Billion for
Inequality: Evidence and Some Theory.” Journal of Abusing Dominance as Search Engine by Giving
Labor Economics 36(S1): S13–70. Illegal Advantage to Own Comparison Shopping
Crémer, Jacques, Yves-Alexandre de Montjoye, Service—Factsheet.” European Commission Fact-
and Heike Schweitzer. 2019. Competition Policy for sheet, June 27, 2017. http://europa.eu/rapid/
the Digital Era. Brussels: European Commission. press-release_MEMO-17-1785_en.htm.
Crouzet, Nicolas, and Janice C. Eberly. 2019. Facebook. 2018. “Response to Six4Three Docu-
“Understanding Weak Capital Investment: The ments.” Facebook Press Release, December 5,
Role of Market Concentration and Intangibles.” 2018. https://newsroom.fb.com/news/2018/12/
NBER Working Paper 25869. response-to-six4three-documents/.
Cunningham, Colleen, Florian Ederer, and Federal Trade Commission. 2018a. “FTC
Song Ma. 2018. “Killer Acquisitions.” https:// Hearing #2 on Competition and Consumer
ssrn.com/abstract=3241707. Protection in the 21st Century: Monopsony
De Loecker, Jan, Jan Eeckhout, and Gabriel and the State of U.S. Antitrust Law.” September
Unger. 2018. “The Rise of Market Power and 21, 2018. https://www.ftc.gov/news-events/
the Macroeconomic Implications.” Available at events-calendar/2018/09/ftc-hearing-2-competi-
https://sites.google.com/site/deloeckerjan/ tion-consumer-protection-21st-century.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
92 Journal of Economic Perspectives

Federal Trade Commission. 2018b. “In the Competitive Than US Markets: A Study of Institu-
Matter of Grifols, S.A., a Corporation, and tional Drift.” NBER Working Paper 24700.
Grifols Shared Services North America, Inc., a Hall, Robert E. 1988. “The Relation between
Corporation.” FTC Matter/File No. 181 0081, Price and Marginal Cost in U.S. Industry.” Journal
updated September 18, 2018. https://www.ftc. of Political Economy 96(5): 921–47.
gov/enforcement/cases-proceedings/181-0081/ Hall, Robert E. 2018. “Using Empirical
grifols-sa-grifols-shared-services-north-america- Marginal Cost to Measure Market Power in the
inc-matter. US Economy.” NBER Working Paper 25251.
Federal Trade Commission and Department Hemphill, C. Scott, and Nancy L. Rose. 2018.
of Justice. 2018. “Hart-Scott-Rodino Annual “Mergers That Harm Sellers.” Yale Law Journal
Report: Fiscal Year 2017.” Available at https:// 127(7): 2078–109.
www.ftc.gov/policy/reports/policy-reports/ Hemphill, C. Scott, and Philip J. Weiser.
annual-competition-reports. 2018. “Beyond Brooke Group: Bringing Reality to
Federico, Giulio, Fiona Scott Morton, and Carl the Law of Predatory Pricing.” Yale Law Journal
Shapiro. Forthcoming. “Antitrust and Innovation: 127(7): 2048–77.
Welcoming and Protecting Disruption.” Chap. Hovenkamp, Herbert J., and Carl Shapiro.
4 in Innovation Policy and the Economy, vol. 20, 2018. “Horizontal Mergers, Market Structure,
edited by Josh Lerner and Scott Stern. Chicago: and Burdens of Proof.” Yale Law Journal 127(7):
University of Chicago Press. https://www.nber. 1996–2025.
org/chapters/c14261. International Monetary Fund. 2019. “The
Furman, Jason, Diane Coyle, Amelia Fletcher, Rise of Corporate Market Power and Its Macro-
Philip Marsden, and Derek McAuley. 2019. economic Effects.” Chap. 2 in World Economic
Unlocking Digital Competition: Report of the Digital Outlook: Growth Slowdown, Precarious Recovery.
Competition Expert Panel. London: Digital Compe- April 2019. Washington, DC: International
tition Expert Panel. Monetary Fund.
Furman, Jason, and Peter Orszag. 2018. “A Kline, Patrick, Neviana Petkova, Heidi
Firm-Level Perspective on the Role of Rents in the Williams, and Owen Zidar. 2017. “Who Profits
Rise of Inequality.” In Toward a Just Society: Joseph from Patents? Rent-Sharing at Innovative Firms.”
Stiglitz and Twenty-First Century Economics, edited Institute for Research on Labor and Employment
by Martin Guzman, 19–47. New York: Columbia Working Paper 107-17. http://irle.berkeley.edu/
University Press. files/2017/Who-Profits-from-Patents.pdf.
Ganapati, Sharat. 2018. “Oligopolies, Prices, Koh, Lucy H. 2014. “In Re: High-Tech
Output, and Productivity.” https://ssrn.com/ Employee Antitrust Litigation, Order Denying
abstract=3030966. Plaintiffs’ Motion for Preliminary Approval
Gaynor, Martin. 2018. “Examining the Impact of Settlements with Adobe, Apple, Google,
of Health Care Consolidation.” Statement before and Intel.” US District Court for the Northern
the Committee on Energy and Commerce District of California, San Jose Division, Case
Oversight and Investigations Subcommittee, 11-CV-02509-LHK, Filing 974, August 8, 2014.
US House of Representatives, February 14, Krueger, Alan B., and Orley Ashenfelter. 2018.
2018. https://docs.house.gov/meetings/IF/ “Theory and Evidence on Employer Collusion
IF02/20180214/106855/HHRG-115-IF02-Wstate- in the Franchise Sector.” NBER Working Paper
GaynorM-20180214.pdf. 24831.
Gaynor, Martin, and Robert Town. 2012. “The Krueger, Alan B., and Eric A. Posner.
Impact of Hospital Consolidation: Update.” 2018. “A Proposal for Protecting Low-Income
Robert Wood Johnson Foundation, Synthesis Workers from Monopsony and Collusion.”
Project Policy Brief 9. https://www.rwjf.org/ Hamilton Project Policy Proposal 2018-05.
content/dam/farm/reports/issue_briefs/2012/ http://www.hamiltonproject.org/assets/files/
rwjf73261. protecting_low_income_workers_from_monop-
Gee, Emily, and Ethan Gurwitz. 2018. Provider sony_collusion_krueger_posner_pp.pdf.
Consolidation Drives Up Health Care Costs: Policy Kwoka, John. 2014. Mergers, Merger Control,
Recommendations to Curb Abuses of Market Power and and Remedies: A Retrospective Analysis of U.S. Policy.
Protect Patients. Washington, DC: Center for Amer- Cambridge, MA: MIT Press.
ican Progress. https://cdn.americanprogress. Levenstein, Margaret C., and Valerie Y. Suslow.
org/content/uploads/2018/12/04110830/ 2006. “What Determines Cartel Success?” Journal
Consolidation-HealthCare-Costs.pdf. of Economic Literature 44(1): 43–95.
Gutiérrez, Germán, and Thomas Philippon. Levenstein, Margaret C., and Valerie Y. Suslow.
2018. “How EU Markets Became More 2011. “Breaking Up is Hard to Do: Determinants

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms
 Protecting Competition in the American Economy 93

of Cartel Duration.” Journal of Law and Economics 413–48.

54(2): 455–92. Scott Morton, Fiona, Pascal Bouvier, Ariel
Manning, Alan. 2011. “Imperfect Competition Ezrachi, Bruno Jullien, Roberta Katz, Gene
in the Labor Market.” Chap. 11 in Handbook of Kimmelman, A. Douglas Melamed, and Jamie
Labor Economics, vol. 4b, edited by Orley Ashen- Morgenstern. 2019. Committee for the Study of
felter and David Card, 973–1041. Amsterdam: Digital Platforms: Market Structure and Antitrust
Elsevier. Subcommittee Report. Draft. Chicago: Stigler Center
Marinescu, Ioana Elena, and Herbert Hoven- for the Study of the Economy and the State,
kamp. 2018. “Anticompetitive Mergers in Labor University of Chicago Booth School of Business.
Markets.” University of Pennsylvania Institute https://research.chicagobooth.edu/-/media/
for Law and Economics Research Paper 18-8. research/stigler/pdfs/market-structure---report-
https://ssrn.com/abstract=3124483. as-of-15-may-2019.pdf.
Moretti, Enrico. 2011. “Local Labor Markets.” Shapiro, Carl. 2010. “The 2010 Horizontal
Chap. 14 in Handbook of Labor Economics, vol. Merger Guidelines: From Hedgehog to Fox in
4b, edited by Orley Ashenfelter and David Card, Forty Years.” Antitrust Law Journal 77(1): 701–59.
1237–313. Amsterdam: Elsevier. Shapiro, Carl. 2012. “Competition and Innova-
Muris, Timothy J., and Jonathan E. Neuchter- tion: Did Arrow Hit the Bull’s Eye?” Chap. 7 in
lein. 2018. “Antitrust in the Internet Era: The The Rate and Direction of Inventive Activity Revisited,
Legacy of United States v. A&P.” George Mason edited by Josh Lerner and Scott Stern, 404–10.
Law and Economics Research Paper 18-15. Chicago: University of Chicago Press.
https://ssrn.com/abstract=3186569. Shapiro, Carl. 2018. “Antitrust in a Time of
Naidu, Suresh, Eric A. Posner, and Glen Weyl. Populism.” International Journal of Industrial Orga-
2018. “Antitrust Remedies for Labor Market nization 61(1): 714–48.
Power.” Harvard Law Review 132(2): 536–601. Syverson, Chad. 2011. “What Determines
O’Sullivan, Donie, and Hadas Gold. Productivity?” Journal of Economic Literature 49(2):
2018. “Facebook Internal Emails Show Zuck- 326–65.
erberg Targeting Competitor Vine.” CNN Traina, James. 2018. “Is Aggregate Market
Business, December 5, 2018. https://www.cnn. Power Increasing? Production Trends Using
com/2018/12/05/media/facebook-six4three- Financial Statements.” https://ssrn.com/
internal-documents-emails/index.html. abstract=3120849.
Prager, Elena, and Matthew Schmitt. 2019. Valletti, Tommaso. 2018. “Concentration
“Employer Consolidation and Wages: Evidence Trends.” Brussels: European Commission.
from Hospitals.” Washington Center for Equi- https://www.ecb.europa.eu/pub/conferences/
table Growth Working Paper. https://ssrn.com/ shared/pdf/20180618_ecb_forum_on_central_
abstract=3391889. banking/Valletti_Tommaso_Presentation.pdf.
Rossi-Hansberg, Esteban, Pierre-Daniel Sarte, Valletti, Tommaso M., and Hans Zenger.
and Nicholas Trachter. 2018. “Diverging Trends 2019. “Increasing Market Power and Merger
in National and Local Concentration.” NBER Control.” https://papers.ssrn.com/sol3/papers.
Working Paper 25066. cfm?abstract_id=3387999.
Salinger, Michael. 1990. “The Concentration- Van Reenen, John. 2018. “Increasing Differ-
Margins Relationship Reconsidered.” Brookings ences between Firms: Market Power and the
Papers on Economic Activity 21(Microeconomics): Macro-Economy.” Centre for Economic Perfor-
297–335. mance Discussion Paper 1576. http://cep.lse.
Schmalensee, Richard. 1989. “Inter-Industry ac.uk/pubs/download/dp1576.pdf.
Studies of Structure and Performance.” Chap. Werden, Gregory J. 2018. “The 1968 Merger
16 in Handbook of Industrial Organization, vol. Guidelines: In Praise of Committing to Restraint.”
2, edited by Richard Schmalensee and Robert Review of Industrial Organization 53(3): 445–52.
Willig, 951–1009. Amsterdam: Elsevier. Wollmann, Thomas G. 2019. “Stealth Consoli-
Schmalz, Martin C. 2018. “Common-Owner- dation: Evidence from an Amendment to the
ship Concentration and Corporate Conduct.” Hart-Scott-Rodino Act.” American Economic Review:
Annual Review of Financial Economics 10(1): Insights 1(1): 77–94.

This content downloaded from

14.139.86.166 on Sun, 30 Jul 2023 09:19:55 +00:00
All use subject to https://about.jstor.org/terms