In a July 24, 2017 New York Times op-ed, Senate Minority Leader Chuck Schumer (D–NY) promised aggressive antitrust activism as part of his party’s “Better Deal for American Workers.” “We are going to fight to allow regulators to break up big companies if they’re hurting consumers,” Schumer promised. Antitrust laws, he argued, are “padding the pockets of investors but sending costs skyrocketing for everything from cable bills and airline tickets to food and health care.” As Jeff Stein at Vox explained, this Better Deal intends to create “a new federal ‘Trust Buster’ agency … similar in scope to the Consumer Financial Protection Bureau.”

In a 4,000-word column titled "Is Amazon Getting Too Big?" in the Washington Post a few days later, business writer Steven Pearlstein went Schumer one better, arguing that a new "antitrust czar" should not focus narrowly on consumer harm, but should combat "bigness" in general. Pearlstein lauded a Yale Law Journal article by Lina Khan, now a fellow with the Open Markets Institute, that makes the same argument.

Pearlstein was followed by numerous columns and articles containing similar points with increasing frequency and intensity. In a January 16, 2018 Wall Street Journal piece titled "The Antitrust Case against Facebook, Google and Amazon," economics commentator Greg Ip claimed, "A growing number of (nameless) critics think these tech giants need to be broken up or regulated as Standard Oil and AT&T once were." That was followed by The Economist's January 20 cover story, "The New Titans: And How to Tame Them," with Facebook, Google, and Amazon depicted as gigantic scary robots.

Just a few days earlier, Brookings Institution political scientist William Galston and assistant Clara Hendrickson released "A Policy at Peace with Itself: Antitrust Remedies for Our Concentrated, Uncompetitive Economy." The report emphasized that "antitrust is not merely an object of scholarly concern; it has also become an important political talking point." The two authors soon added a follow-up in the Harvard Business Review, "What the Future of U.S. Antitrust Should Look Like."

This article is a critical review of the evidence cited in these calls for a "new antitrust." Specifically, I examine claims regarding the effects of past mergers on prices in several industries, assertions that corporate profitability is evidence of increased market concentration, and estimates of concentration of large firms in broad sectors such as retailing and service, interpreted as diminished competition. I'll end this article with some cautionary lessons from the antitrust suits against IBM (1970s–1980s) and Microsoft (1990s–2000s) about the dangers of confusing imaginative prosecutors with technological forecasters, or of assuming that tech firms with an early lead on some innovation have an invincible advantage over new rivals.

In the renowned 2004 study "Does Antitrust Policy Improve Consumer Welfare? Assessing the Evidence," Brookings Institution scholars Robert Crandall and Clifford Winston found "no evidence that antitrust policy in the areas of monopolization, collusion, and mergers has provided much benefit to consumers and, in some instances, we find evidence that it may have lowered consumer welfare." But consumer welfare is not what drives populist/progressive Better Deal enthusiasts. Since the Chicago School shifted the emphasis of antitrust to consumer welfare, complains Pearlstein, "courts and regulators narrowed their analysis to ask whether it would hurt consumers by raising prices." Pearlstein would like courts and regulators to pay more attention to "leveling the playing field." Kahn likewise argues that "undue focus on consumer welfare is misguided. It betrays legislative history, which reveals that Congress passed antitrust laws to promote a host of political economic ends."

The trouble with grounding policy on legal precedent and political ends, however, is that Congress has passed many laws to promote the special interests of producers at the expense of consumers. Some examples include the creation of the Interstate Commerce Commission (1887), the National Economic Recovery Act (1933), the Robinson–Patman Act (1936), the creation of the Civil Aeronautics Board (1938), price supports for farm and dairy products, and numerous tariffs and regulations designed to benefit influential interest groups and the politicians who represent them. The harm that these initiatives did to consumers is now well understood, though not by all politicians and journalists. That raises the question of why we should unleash another round of consumer harm on the public.

Old Airline Mergers Don't Justify Breaking up Tech Firms

University of Pisa economist Nicola Giocoli examines the economic analysis of antitrust over the period 1939–1974, when it was dominated by the structure–conduct paradigm of Harvard's Edward Mason and his student Joe Bain. By the 1950s and 1960s, that "Harvard School" approach evolved into a more rigid "structuralist" view that market concentration ("oligopoly") could be assumed to facilitate collusion and therefore high prices and profit. Harold Demsetz later showed the lack of evidence for that hypothesis, which helped to advance a consumer-focused price theory approach, dubbed "The Chicago School of Antitrust Analysis" by Richard Posner.

Kahn claimed the Chicago School's "consumer welfare frame has led to higher prices and few efficiencies," citing a collection of studies John Kwoka discusses in his 2014 book Mergers, Merger Control, and Remedies. Galston and Hendrickson praise the book as "a comprehensive study of recent mergers." In reality, 10 of the book's 42 "recent" mergers happened between 1976 and 1987, and 21 others happened in the 1990s. Those old studies were mainly focused on very few industries, including airline and railroad mergers enforced by the Department of Transportation and the Surface Transportation Board, rather than the Justice Department or Federal Trade Commission. Senator Schumer as well as Galston and Hendrikson allude to "recent" airline fares as a reason for tougher antitrust, even though five of the seven airline mergers in Kwoka's book occurred in 1986–1987, and the other two in 1994.

Pearlstein notes that Kwoka's list of higher prices blamed on mergers includes "hotels, car rentals, cable television, and eyeglasses." The goods on that list look as old-fashioned as Kwoka's definition of "professional journal publishing" as involving print only, ignoring electronic publications. Hotels now face stiff competition from Airbnb; rentals cars from Uber; cable companies face "cord-cutting" alternatives such as broadcast HDTV, satellite providers DirecTV and Dish, and internet providers such as Roku, Netflix, Amazon, Hulu, and more. The claim that eyeglass maker Luxottica controls 80% of U.S. optician chain sales ignores the sales made by thousands of independent optometry practices, huge retailers Walmart and Costco, and online retailers Zenni Optical and Warby Parker. It is difficult to imagine how Pearlstein or Kwoka could seriously suggest consumers face monopoly pricing from such industry leaders as Southwest Airlines, Marriott hotels, Enterprise Rent-A-Car, or Costco Optical.

The most recent merger in Kwoka's compilation of supposedly cartelizing mergers was in 2006 when Whirlpool outbid Haier to acquired Maytag. Any suggestion that Whirlpool gained monopoly power from that merger, however, was dealt a decisive blow on January 22, 2018 when the U.S. government imposed 20–50% tariffs on washers and dryers from LG and Samsung. (See "Putting 97 Million Households through the Wringer," p. 8.)

Unless we know what prices would have been if some now-criticized merger had not occurred, we cannot attribute subsequent price changes to a merger rather than rising input costs, quality improvement, or inflation. To get around that, the study of the clothes dryer market used by Kwoka relies on a difference-in-differences simulation that claims the merger raised the price of new-model Whirlpool dryers (but not Maytags) by 17%. That is because prices of the newest (but unimproved?) Whirlpool dryers rose more than the prices of stovetops, chosen as the control. Prices of clothes washers were unaffected and those of older-model appliances fell. Even accepting all that, how could anyone possibly calculate an estimated "average" price increase for the merger as a whole?

When Kwoka goes on to heroically meld 27 such complex price simulations into a single "average," the figure has no clear meaning without weighting, and without standard deviation data to show the estimates are significant. That is one reason FTC economists Michael Vita and David Osinski argue that "Kwoka has drawn inferences and reached conclusions … that are unjustified by his data and his methods." A subsequent exchange between Kwoka and Vita failed to resolve grave doubts about Kwoka's widely publicized conclusions that mergers have been shown to frequently result in higher prices and lower quality, or that the FTC and DOJ have become laxer about merger enforcement in recent years.

In any case, Kwoka's opinion that merger guidelines should be tightened has nothing to do with Ip's notion that "Facebook, Google, and Amazon need to be broken up or regulated as Standard Oil and AT&T once were." Standard Oil's dominance was not ended by breaking up the trust into Rockefeller-owned regional parts (three of which became Exxon-Mobil). It was ended because electricity displaced kerosene for home lighting, and most gasoline for cars soon came from newcomers in Texas and California.

The suggestion that federal regulation was ever an alternative to competition is also misinformed. Regulating industries "like utilities" meant banning competition in airlines in the decades before the industry's 1978 deregulation, in telephones from 1913 to AT&T's breakup in 1983, and in the Postal Service's First-Class mail even today. Federal regulation meant new entrants were prohibited and cutting prices was a federal offense.

If It's Profitable, It Must Be a Monopoly

Demsetz' 1973 survey of the evidence found the Harvard School's "market concentration doctrine" could not demonstrate a causal connection between concentration ratios (such as top-four firms' share of a market) and profitability; Michael Salinger obtained similar results when he revisited the issue in 1990. Continued failures to find a reliable connection between higher profits and concentration led Kwoka and others to search instead for a connection between higher prices and market concentration. Craig Newmark's survey of these newer price-concentration studies found "the depth of their problems seems to be neither widely nor fully appreciated." He concluded that "those problems are serious enough that the price-concentration studies probably should be discarded just like the profit-concentration studies have been."

The latest reincarnation of the market concentration doctrine evades the repeated failures of profit-concentration studies by simply turning the issue upside down—to assert that high profits prove market concentration. In other words, high profits are assumed to be evidence of too much concentration, and concentration is in turn assumed to be evidence of a lack of competition, as in the 1950s and 1960s.

The Economist cover story of March 16, 2016, titled "Too Much of a Good Thing," offered the subtitle: "Profits are too high. American needs a giant dose of competition." To make that point, the opening paragraph noted, "Last year America's airlines made $24 billion." That was bad timing because airline profits fell to $13.5 billion in 2016. But it was also a bad example because the most consistently profitable airlines are upstarts that began as small regional firms: Southwest, Jet Blue, and Alaska/Virgin. In the Investopedia article "Why Airlines Aren't Profitable," Greg McFarlane notes that "from 2002 to 2011, the three largest surviving legacy airlines—American, United, and Delta—each filed for bankruptcy."

A glaring irony of using profitability as proof of monopoly is that the firm most often mentioned as a target of the populist antitrust campaign is Amazon, which is barely profitable. And Apple, the most profitable U.S. company, is not among The Economist's trio of Titans to be tamed. But inconsistency is no problem for antitrust populists, who seem equally comfortable arguing that low profits are also proof of monopoly. Kahn claims that low operating profits prove Amazon is "choosing to price below cost." What low profits actually show is that Amazon has been plowing back its cash flow into capital expenditures, such as cloud computing, a movie studio, a grocery chain, and innovative consumer electronics such as Kindle and Echo.

A widely quoted 2016 Issue Brief from President Obama's Council of Economic Advisers (CEA) includes a graph from then-chairman Jason Furman showing large recent gains in "returns on invested capital" among public nonfinancial firms as calculated by the consulting firm McKinsey & Co. The Furman CEA claimed that this demonstrates a surge in "rents," which was wrongly defined as returns "in excess of historical standards." At McKinsey, however, Mikel Dodd and Werner Rehm explained that returns appear to be growing larger by their measure because invested capital as traditionally measured (plant and equipment) became smaller as the economy shifted from capital-intensive manufacturing to services and software.

Traditionally defined physical capital also looks smaller because traditional accounting practices exclude or undervalue intangible capital such as research and development, design, training, and market branding through advertising. Accountants commonly count investments in intangibles as operating expenses, which makes invested capital look low and return on capital high.

Traditionally defined physical capital also looks smaller because traditional accounting practices exclude or undervalue intangible capital such as R&D, design, training, and market branding through advertising.

Moreover, official government estimates do not confirm Furman's alleged surge in returns on invested capital. A December 2017 Bureau of Economic Analysis report notes that "the profitability of domestic nonfinancial corporations declined for a second year in 2016, but remains above the low point in 2009." The after-tax rate of return on capital for 14 major nonfinancial industries ranged from 6.3% in 2002 to 8.3% in 2012, falling to 7.6% in 2016.

The Economist, Galston and Hendrickson, and others point to a graph of pretax corporate profits rising to 11–12% of gross domestic product since 2010 as evidence of rapidly declining competition. Yet the most obvious of many problems with comparing such profits to GDP is that profits are increasingly global while GDP is domestic. Nearly 70% of Apple's recent profits were foreign, up from 40% a decade ago.

Urooj Kahn, Suresh Nallareddy, and Ethan Rouen find the reason combined foreign and domestic profits have grown faster than GDP is the widening gap between U.S. and corporate tax rates before 2018, "incentivizing firms to invest these profits abroad or hold them as cash." Foreign profits tell us nothing about domestic competition. Neither Apple's high profits nor Amazon's low profits are evidence of monopoly.

Measuring Concentration Requires Some Concentration

Pearlstein wrote, "There is little debate that this cramped [Chicago School] view of antitrust law has resulted in an economy where two-thirds of all industries are more concentrated than they were 20 years ago, according to a study by President Barack Obama's Council of Economic Advisers, and many are dominated by three or four firms."

That is not what the 2016 CEA Brief said. What it said was the largest 50 firms (not "three or four") in 10 out of 13 "industries" (really sectors) had a larger share of sales in 2012 than in 1997. Pearlstein's "two-thirds of all" means 10 out of 13, but the United States has more than 13 industries. In pointlessly broad sectors such as retailing, real estate, and finance, the top 50 firms had a slightly larger share of sales in 2012 than in 1997. The 50 largest in "retailing" accounted for 36.9% of sales in 2012, said the CEA, but that combines McDonald's, Kroeger, Home Depot, and AT&T Wireless as if they were colluding competitors.

Should we fear monopolistic price gouging simply because 50 firms account for a larger share of the nation's enormous number of retail stores, real estate brokers, or finance companies? Of course not.

In making these claims, the CEA and The Economist compared the quinquennial Census of Business data for 1997 and 2012. Galston and Hendrickson cite those two data points as evidence that "concentration is rising throughout the economy" and "competition across the economy is in decline." Note how they mingle the word "sectors" with the word "industries" in the following passage:

An Economist analysis of 893 industries identifies the market share held by the four largest firms within each. It finds that between 1997 and 2012, two-thirds of industries became more concentrated. During this period, the weighted average share of the top four firms in each sector rose from 26 percent to 32 percent….

Research led by MIT economist David Autor confirms this trend. Between 1982 and 2012, the top four firms in the six major sectors of the U.S. economy became steadily and significantly more concentrated. In the manufacturing sector, the sales concentration ratio of the top four firms increased from 38 percent to 43 percent. Retail trade saw its sales concentration double to 30 percent from 15 percent, wholesale trade a change from 22 percent to 28 percent, services a rise from 11 percent to 15 percent, finance a substantial increase from 24 percent to 35 percent, and utilities and transportation a bump from 29 percent to 37 percent.

The North American Industry Classification System (NAICS) uses two- or three-digit codes to designate broad categories of loosely related industries as "sectors" (retailing) or "subsectors" (restaurants). Four-digit codes constitute broad groups of similar industries, and five- and six-digit codes are industries. The cited paper by David Autor et al., like the CEA Brief, is based on useless two-digit sectors.

Galston and Hendrickson propose a "structural presumption" against "excessive sectoral concentration." Yet economists have no theory or evidence to suggest that four-firm concentration ratios (or Herfindahl–Hirschman indexes) within entire sectors could possibly explain competition, pricing, profits, or anything else within relevant markets—which are usually local and include imports. Sectoral concentration ratios are ably challenged in Carl Shapiro's forthcoming paper "Antitrust in a Time of Populism," which Galston and Hendrickson mentioned but apparently didn't appreciate.

The Economist's 893 "industries" are four-digit industry groups, which are more precise than Autor's "six major sectors" or the CEA's 13. Consider that the two-digit code "52" designates the category of "Finance and Insurance" firms, whereas the four-digit subcategory "5211" designates the Federal Reserve and "5232" designates the New York Stock Exchange—both of which are improbable targets for antitrust. The four-digit NAICS groups are still problematic; consider that drug stores, grocery stores, and wholesale clubs have separate four-digit codes, but people buy drugs and groceries at all three. Wired, wireless, and satellite telecommunications are also artificially separated by four-digit NAICS codes.

The Economist's concentration ratio of the top four firms in "full-service restaurants" rose from 8% in 1997 to 9% in 2012, notes Shapiro. Yet nobody would suggest the number of local restaurant choices is shrinking or that new ones aren't popping up constantly. The Economist's national four-firm concentration ratios may suggest the biggest restaurant, supermarket, or hotel chains grew bigger, but that tells us nothing about competition in these inherently local markets.

Galston and Hendrickson write: "In the past two decades, the pace of concentration has been accelerating. The Fortune 500's revenue as a share of GDP has increased from 58 percent to 73 percent." Those figures are from a FiveThirtyEight post by Andrew Flowers, and "past two decades" means the period 1994–2013. Although Fortune also compares the Fortune 500 revenues to GDP (saying it is "two thirds," not 73%), that is a senseless ratio. More and more of the 500's revenues have been foreign sales, not domestic. Gross revenues also involve double-counting because the 500 firms buy from each other. And, as Flowers noted, "the ratio of Fortune 500 profits to all corporate profits shows … no real upward trend."

Galston and Hendrickson present a graph from Census's Business Dynamics Statistics showing enterprise entry and exit rates. This is said to demonstrate "a breakdown in the competitive process [because] … U.S. startup formation rates have fallen dramatically over the last thirty years and … more are failing."

Kahn's wordiest arguments for antitrust suits against large tech companies are not about facts but about theories and predictions. She makes a plea for predictive punishment based on omniscient futurism.

The Census Bureau estimates that there was a total of 6,786,097 business establishments in the United States in 2015, so yearly variations in such a huge number can't be defined as changing concentration. Besides, the entry rate averaged 12.2% from 2002 to 2007, unchanged from the 12.3% average from 1978 to 2015. The graph just shows the Great Recession shrunk the new business startup rate to 9.3% in 2010. The number of establishments in finance, insurance, and real estate fell from 487,868 in 2009 to 430,364 in 2011 because of the mortgage/housing crisis, not a breakdown in the competitive process.

Using Antitrust to Predict and Manage Technology

The old market concentration doctrine is now being used to divert antitrust zeal from Schumer's mundane original targets—such as "cable bills and airline tickets"—to a select trio of high-tech companies (Ip's Amazon, Facebook, and Google, or Lina Khan's Amazon, Apple, and Google). But Facebook isn't profitable because it has a large "share" of some finite market for social media apps. The Economist's February 17, 2016 feature on "The Rise of Superstars," depicted Snapchat as "a fast-disappearing" app slaughtered by Facebook. Yet Snap's user base doubled from 46 million to 94 million between the fourth quarters of 2014 and 2015 according to statistics portal Statistica, and doubled again to 187 million by late 2017. There is no market for social media. Facebook users are free to be users of Twitter, Snapchat, and as many other such sites as they like.

Kahn's wordiest arguments for antitrust suits against large tech companies are not about facts but about theories and predictions. She makes a plea for preemptive punishment based on omniscient futurism. "The current market is not always a good indication of competitive harm," she writes. Antitrust enforcers "have to ask what the future market will look like." Pearlstein likewise thinks that antitrust authorities should have authority to "block Amazon" from competing too effectively with UPS, Oracle, or Comcast in the future.

How could antitrust enforcers' predictions about what might or might not happen in the future be deemed a crime or cause for civil damages? If the law allowed courts to levy huge fines or break up companies on the basis of prosecutors' predictions about the future, the potential for whimsical damages and political corruption would be almost limitless.

We have already experienced extremely costly federal (and European) antitrust cases based largely on incredible predictions about "what the future market will look like"—most obviously in the cases against IBM and Microsoft. IBM was the subject of 13 years of antitrust "investigation" (harassment) before the suit was finally dismissed "without merit" in 1982. Pearlstein imagines "it was the government's aborted [botched?] prosecution of IBM … that made Microsoft possible."

But IBM's decisions to offer three operating systems for the PC and allow Microsoft to sell MS-DOS to Compaq had nothing to do with the government's antitrust crusade against the firm. (The PC wasn't even available until August 1981.) Instead, the case was about IBM's dominance in data processing. That crusade was a well-funded project of Control Data, Honeywell, NCR, and Sperry Rand, competitors of IBM's that hoped to do better in court than they had with customers. An article on the IBM case that I wrote for Reason in 1974 concluded:

There is an irreconcilable conflict between helping competition and helping competitors. Many firms are quietly making a lot of money in data processing by providing better products at lower prices. Others find it easier and more lucrative to sue—proving once again that antitrust laws seek victims without crimes.

The Microsoft case was another example of antitrust prosecutors trying to reengineer technology to make life easier for a successful firm's rivals. "In May 1998," notes Pearlstein, "U.S. attorneys general filed an antitrust suit against Microsoft, which lurks in the background of the current debate." Microsoft was accused of extending its legal dominance in PCs to achieve a monopoly on internet browsers and "middleware" (e.g., media players, email clients, and instant messaging) that could supposedly serve as "alternative platforms" to Windows in some incomprehensible fashion. In reality, the internet itself has proven to be the alternative, and it is platform-independent. Online services don't know or care which operating system you use to fill out tax returns or which media player you use to watch movies. And you don't need Android devices to use Google Docs.

The government's technologically illiterate case against Microsoft (heavily promoted by IBM, Intel, SUN, AOL, etc.) became a decade-long, ever-changing battle waged by prosecutors and judges who were unable to even imagine that Apple, Amazon, and Google could be competitive rivals of Microsoft in hardware, software, or services, or that cellphones and tablets could serve as handy computers.

The Microsoft settlement barred Microsoft from discouraging computer manufacturers from pre-installing non-Microsoft software on new computers, including buggy "bloatware." But the browsers, search engines, or media players preloaded on Windows and Apple devices have been a non-issue because broadband made it easy to install any or all of them on PCs, tablets, and phones regardless of the operating system. Open-source VLC soon became a popular media player and open-source Firefox a popular browser. Microsoft shut down its MSN Messenger in 2014 and Windows' share of smartphone operating systems (a technology that the DOJ considered irrelevant at the time of the Microsoft suit) recently fell to nearly zero. Internet tracker Statcounter estimates that Google Chrome has a 48.8% share of U.S. browser use, Apple's Safari a 32% share, and Windows Edge and Internet Explorer combined have a 9.1% share.

Kahn urges that antitrust now shift focus to Microsoft's previously unnoticed rivals. "Google, Apple, and Amazon have created disruptive technologies that changed the world," she writes. "But the opportunity to compete must remain open for new entrants and smaller competitors that want their chance to change the world."

Rather than offering evidence that new entrants are somehow excluded from markets supposedly dominated by Google, Apple, and Amazon, Pearlstein offered this summary of Khan's theoretical anxieties:

Chicago antitrust theory is ill equipped to deal with high-tech industries, which naturally tend toward winner-take-all competition. In these, most of the expenses are in the form of upfront investments, such as software (think Apple and Microsoft), meaning that the cost of serving additional customers is close to zero…. What this "post-Chicago" economics shows is that in such industries, firms that jump into an early lead can gain such an overwhelming advantage that new rivals find it nearly impossible to enter the market. [My emphasis.]

The assertion that early entrants into high-tech can gain "such an overwhelming advantage that new rivals [find] it nearly impossible to enter the market" is often referred to as "network effects." (See "Debunking the 'Network Effects' Bogeyman," Winter 2017–2018.) Bruce Kobyashi and Timothy Muris call it the "possibility theorem." Anything might be possible in theory, but the "post-Chicago" claim that early tech leaders can't be challenged has been repeatedly proven false. Consider:

  • In personal computers, Apple, Commodore, and Sinclair were first, followed by Apollo and the IBM PC in 1981, Osborne and Sun in 1982, Compaq in 1983, and Dell in 1984. Contrary to what trustbusters predicted, IBM gave up on this market in 2005 when it sold the whole ThinkPad business and brand to the Chinese firm Lenovo.
  • Netscape had an overwhelming dominance in internet browsing in 1995, but that did not deter Opera and Internet Explorer from entering the market that year, nor Firefox in 2002, Safari in 2003, or Google Chrome in 2008.
  • AOL was the dominant internet portal in 1993, but was challenged by Netscape in 1994, Yahoo in 1995, and later by Comcast, Google, Facebook, and many more.
  • AltaVista, Lycos, Infoseek, HotBot, Excite, and Yahoo were meta-search engines that "jumped into an early lead," yet they were soon trumped by Google, Ask, and Bing. Search Engine Watch reports that Google had a 72.5% share of global meta-search in mid-2016, followed by 10.4% for Bing and 7.8% for Yahoo. But meta-search is only part of a much larger universe that includes specialized vertical search engines such as OpenTable, TripAdvisor, Match, Yelp, Houzz, and Expedia, and also comparative shopping engines such as Amazon, eBay, PriceGrabber, NexTag, and Shopzilla.
  • Palm, Nokia, and Motorola jumped into an early lead in personal digital assistants (PDAs) and cellphones, yet were shoved aside by Blackberry, which in turn was shoved aside by Samsung and iPhone. Symbian was the most widely used PDA operating system in the world until 2010, when it was overtaken by Android.
  • iTunes jumped into the early lead in online music, supposedly creating nearly "predatory" competition for compact discs. But in 2011, Savage Beast went public as Pandora streaming service and Stockholm's Spotify launched its U.S. service.
  • Friendster, Linked-in, and Myspace jumped into an early lead in social networking in 2002–2003, followed by Google's surely invincible social network Orkut in January 2004. Yet Facebook was not afraid to jump into that "market" in 2004, nor were YouTube and Reddit in 2005, and Twitter in 2006, followed by Google+, Snapchat, Tumblr, Instagram, Pinterest, etc.

Conclusion

Khan would not only have antitrust czars prosecuting cases based on their technological predictions, but would have them "overseeing concentrations of power that risk precluding real competition." This "structuralist" approach removes all annoying requirements for evidence that competition is impeded in any way. All that would be needed is a prosecutor's perception that any concentration of undefinable "power" might someday risk some undefinable vision of "real competition" or otherwise harm some undefinable "public interest."

Kahn's proposed carte blanche antitrust mandate is an invitation to "political and ideological mischief," a former antitrust official told Pearlstein. The potential for abuse is obvious and dangerous, as Steven Salop and Carl Shapiro explain in a 2017 paper. President Trump threatened Jeff Bezos with "a huge antitrust problem" because Amazon owns the Washington Post, complaining that "he's using that as a tool for political power against me."

If the Democrats' hopes for stepped-up discretionary antitrust enforcement follows the advice of Pearlstein and the Open Markets Institute, it would add paralyzing uncertainty to business plans and decisions. This vision of unbridled antitrust activism is a recipe for judicial caprice. It would encourage interest group meddling in business planning and pricing, invite political corruption, and risk replacing the rule of law with the rule of lawyers.

Readings

  • "A Policy at Peace with Itself: Antitrust Remedies for Our Concentrated, Uncompetitive Economy," by William A. Galston and Clara Hendrickson. Brookings Institution Report, January 5, 2018.
  • "Amazon's Antitrust Paradox," by Lina Khan. Yale Law Journal 126: 710–805 (2017).
  • "Antitrust for Fun and Profit," by Alan Reynolds. Reason, April 1974.
  • "Antitrust in a Time of Populism," by Carl Shapiro. International Journal of Industrial Organization, forthcoming.
  • "Antitrust Policy Relies More Heavily on Beliefs Rather than a Strong Consensus about Facts," by Sam Peltzman. ProMarket (blog) April 7, 2017.
  • "Benefits of Competition and Indicators of Market Power." Council of Economic Advisers Issue Brief, April 2016.
  • "Big Business Is Getting Bigger," by Andrew Flowers. FiveThirtyEight, May 18, 2015.
  • "Chicago, Post-Chicago, and Beyond: Time to Let Go of the 20th Century," by Bruce H. Kobayashi and Timothy J. Muris. Antitrust Law Journal 78: 505–526 (2012).
  • "Does Antitrust Policy Improve Consumer Welfare?" by Robert W. Crandall and Clifford Winston. Journal of Economic Perspectives 17(4): 3–26 (2004).
  • "John Kwoka's Mergers, Merger Control, and Remedies: A Critical Review," by Michael Vita and F. David Osinski. Working paper, December 21, 2016.
  • "Kwoka's Mergers, Merger Control, and Remedies: Rejoinder to Kwoka," by Michael Vita. Working paper, January 29, 2018.
  • "Labor's Share of GDP: Wrong Answers to a Wrong Question," by Alan Reynolds. Cato@Liberty (blog), October 26, 2017.
  • "Mergers, Merger Control, and Remedies: A Response to the FTC Critique," John Kwoka. Antitrust Institute, March 2017.
  • Mergers, Merger Control, and Remedies: A Retrospective Analysis of U.S. Policy, by John Kwoka. MIT Press, 2014.
  • "Microsoft's Appealing Case," by Robert A. Levy, and Alan Reynolds. Cato Institute Policy Analysis no. 385, November 9, 2000.
  • Predatory Pricing in Antitrust Law and Economics: A Historical Perspective, by Nicola Giocoli. Routledge, 2014.
  • "Price-Concentration Studies: There You Go Again," by Craig M. Neumark. Presentation to the DOJ/FTC Merger Workshop, Concentration and Market Shares Panel, February 14, 2004.
  • The Causes and Consequences of Antitrust: A Public Choice Perspective, by Fred S. McChesney and William F. Shughart. University of Chicago Press, 1995.
  • "The Chicago School of Antitrust Analysis," by Richard A. Posner. University of Pennsylvania Law Review 127: 925–948 (1978).
  • "The Concentration–Margins Relationship Reconsidered," by Michael Salinger. Brookings Papers on Economic Activity: Microeconomics, 1990.
  • "The Fall of Labor Share and the Rise of Superstar Firms," by David Autor, David Dorn, Lawrence F. Katz, et al. Working paper, May 1, 2017.
  • "The Market Concentration Doctrine," by Harold Demsetz. AEI–Hoover Policy Study no. 7, August 1973.
  • "The Role of Taxes in the Disconnect between Corporate Performance and Economic Growth," by Urooj Kahn, Suresh Nallareddy, and Ethan Rouen. Harvard Business School Working Paper 18-006, 2017.
  • "What the Future of Antitrust Should Look Like," by William A. Galston and Clara Hendrickson. Harvard Business Review, January 9, 2018.
  • "Whither Antitrust Enforcement in the Trump Administration?" by Steven Salop and Carl Shapiro. The Antitrust Source, February 2017.
  • "Why Airlines Aren't Profitable," by Greg McFarlane. Investopedia, March 17, 2014.