A Search for Sanity in Antitrust: Move (Too) Fast, Break (Innovative) Things?

Aurelien Portuese October 28, 2020
October 28, 2020

(Ed. Note: The “Innovation Fact of the Week” appears as a regular feature in each edition of ITIF’s weekly email newsletter. Sign up today.)

A 1978 article titled “A Search for Sanity in Antitrust” published in Fortune argued that antitrust enforcement was fraught with “an oligopoly of opposites” where the regulators welcomed economic efficiency of business but enforced populist decisions that preserved an atomized market structure at the expense of economic efficiency and against scale economies of companies. More than 40 years later, antitrust enforcement is yet again in search of sanity in antitrust where enforcement diverges between, on one hand, an atomized market structure protective of small competitors irrespective of their efficiencies, and on other hand, a tolerance for corporate bigness as long as it can be justified based on superior efficiency. This debate has noisily resuscitated in recent weeks.

Rarely has a week in U.S. antitrust been as historical as the one we have just been through. In choreographed moves, the House Judiciary Committee and the Department of Justice (DOJ) have initiated forceful attacks on big tech companies. The Judiciary Committee published a 450-page report on big tech in which it has recommended breaking up of companies such as Google, Apple, Amazon, and Facebook. A few days later, Attorney General William Barr initiated a lawsuit against Google, the alleged “gatekeeper of the Internet,” in what resembles the historical lawsuit against Microsoft 20 years ago. Inspired by Europe’s first moves against big tech, where reports, legislative proposals, and fines have already been issued, American antitrust appears to be undergoing a massive transformation with momentum for “platform busting” in the digital economy. This platform-busting outbreak has its origins with an influential group of scholars and policy advocates—the so-called “Neo-Brandeisians”—who wish to resurrect the anti-monopoly / anti-bigness sentiments that spurred the populist momentum of the late 19th century when the antitrust laws were adopted. Obviously, big is not always bad (in fact it is usually good), and corporate champions may generate social benefits and collective pride, but the current antitrust moment reneges on decades of improvement in economic knowledge in favor of the originalist antimonopoly, over-interventionist stance, which embodies a substantially different approach to the optimal level of interventionism in the market and innovation processes.

The breakups of big tech companies are explicitly called for in the House Judiciary report and suggested again Google in the DOJ’s lawsuit. Aimed at generating a “forceful antitrust enforcement” (p. 20), the report recommends “structural separations and prohibitions of certain dominant platforms from operating in adjacent lines of business” (p. 20). Structural remedies mean breakups of companies by divestitures of assets with an associated prohibition to subsequently buy out or merge these assets. Structural remedies as regulatory tools impose blatant limitations on fundamental liberties such as property rights, entrepreneurship, and freedom to contract. Thus, antitrust enforcement has wisely been reluctant to hastily resort to such intrusive administration tools, unless an unexceptionally strong case can be made. The first breakup took place in 1911 when the Supreme Court decided, in Standard Oil v United States (221 US 1, 1911), to break up John D. Rockefeller’s Standard Oil Trust into spinoff companies after having being accused of having secretly conspired against consumers and rivals. Such conspiracies are prohibited under Section 1 of the Sherman Act of 1890—the main statute of antitrust laws. The other major breakup took place in 1982 with the Supreme Court’s decision in United States v. AT&T (552 F. Supp. 131, 1982), wherein the telecom company, which enjoyed regulated monopoly status, went through judicially imposed asset divestitures to create seven regional companies. Subsequently, spinoff companies merged back to “baby Bells” as well as “baby Standards,” which emerged after the breakup of Standard Oil. Competition was spurred in long-distance telephony, but only to a limited extent and for a short time, because innovation costs and economic disintegration cannot be underestimated: In particular, the coming of the Internet, cable broadband, and mobile telecommunications as outside innovations created more competition than the breakup could have ever expected and realized.

The breakups of companies, beyond the legal obstacles inherent to the consequences they generate, represent the regulators’ desire to reorganize the market to structure it more optimally, irrespective of the fundamental synergies and network effects that are crucial for these firms’ innovativeness and competitiveness. It may perhaps be the acknowledgment of the detrimental effects of breakups which led the DOJ to settle with Microsoft in 2001 so that no structural remedy was imposed on Microsoft. To ditch the accusations of violation of Section 1 of the Sherman Act against Microsoft, which was alleged to have tied-in its browser with its operating system, the DOJ only imposed a less intrusive behavioral remedy: Microsoft had to remove the default installation in PCs of Microsoft’s browser (Internet Explorer) and Microsoft’s media application (Windows Media Player). These regulatory requirements implicitly acknowledged the unreasonableness of a breakup, and correspondingly, recognized the appropriateness of the behavioral remedies over structural remedies, especially when it comes to the removal of default settings.

Default settings are precisely, again, at the heart of the DOJ’s lawsuit against Google. The accusations indeed echo those formulated 20 years ago against Microsoft, and that formulated a couple of years ago against Google by the European Commission. Be that as it may, default settings are incredibly complicated business common practices when it comes to assessing them from an antitrust perspective. On their anticompetitive effects, one can arguably say that they are established only by dominant players and self-reinforced their very dominance without giving other competitors a chance to compete given the alleged status quo bias of consumers. The DOJ lawsuit takes the controversial view that for Google to require preinstalled default status of its search engine on Android-based mobile devices grants Google “de facto exclusivity,” since consumers do not switch to alternative search engines despite none of the traditional switching costs being present with the search engine.

Of course, consumers’ status quo bias exists as first documented in 1988 by Samuelson and Zeckhauser. Under uncertainty, the rational choice model indeed suggests that risk-averse consumers may reveal some inevitable path-dependency. This rationally minded decision does nevertheless not suggest that it is welfare-increasing or welfare-decreasing if consumers both are satisfied with the current quality of services provided and are reluctant to engage in search-and-find costs for alternatives that may not use marketing strategies to appeal to consumers. Also, contrary to Microsoft’s former practice which created obstacles to exit, the costs and obstacles to leaving Google search engines are almost null since alternative search engines are just “one click away.” Indeed, how could consumers be worse off if they are satisfied with the product, if switching costs for alternative search engines are close to zero, and if rivals do not engage in reaching out to consumers via alternative media? It appears thus rational for consumers to stick to the status quo, especially so when there are minimal learning costs involved. Furthermore, the status quo bias is somehow overcome when one looks at the trends: Only 57 percent of consumers use search engines for online product research. Also, digital apps are increasingly used for accessing the Internet: Google may very well be one of the Internet gateways but not its gatekeeper.

Neither the Standard Oil breakup nor the AT&T breakup can serve as inspirations for a breakup of Google. First, the two historical breakups are precedents Google cannot be compared with. The breakup of Standard Oil Trust took place mainly under Section 1 of the Sherman Act, which is designed to prohibit and prevent price-fixing conspiracies by cartels. Of course, Google is not a cartel and does not conspire with competitors against consumers. If a case can be made against Google, it can only be under Section 2 of the Sherman Act, which prevents monopolization attempts by a single firm—here, Alphabet Inc. the mother company of Google. Thus, one cannot apply the rationale behind Standard Oil to Google as the former organized a “trust” while the latter is a vertically integrated single company. Antitrust law may legitimately bust trusts; it can hardly disintegrate vertically integrated companies. The AT&T breakup equally is an inadequate source of inspiration for the DOJ, since AT&T benefited from a government-enticed monopoly that included regulated prices which had to be abandoned in order to foster competition.

Second, Google has emerged neither out of a trust-organizing business model (à la Standard Oil) nor out of a government enticed monopoly business model (à la AT&T). Google has emerged out of massive investments and innovation to compete with many incumbents (i.e., Yahoo! for search engine, Nokia and Blackberry for mobile OS, etc.). While it may no longer be the “scrappy start-up” as colloquially referred to in the DOJ’s lawsuit, Google has invested massively in its search algorithm to best suit consumer preferences and in its business model so that its success enabled it to enter, not always successfully, adjacent markets to compete with incumbents or to create new markets of its own.

Finally, it appears that breaking up Google would immediately benefit both Apple and Microsoft. Apple would gain because its business model of iPhone’s iOS based on a closed, priceable digital ecosystem will prevail over Google’s ad-funded free-of-charge business model enforced through contractual arrangements. Apple may eventually suffer or benefit as payments from Google will end. Should Apple choose to discontinue its use of Google’s search engine subject to the fact that the financial incentives were truly decisive, then either an inferior quality search engine alternative will be chosen for iPhones, and the price-quality ratio of iPhones will worsen at the expense of Apple’s end-consumers, or Apple could launch its own search engine incurring massive costs of replication of Google at the expense of Apple’s other investments and overall profitability. Should Apple choose to continue to use Google’s search engine despite the lack of financial incentives, then the profitability of each iPhone, iPad, and iWatch would decrease so that either consumers may be charged higher prices or the company’ investments in innovation may decrease, or a mixture of these two detrimental implications will occur. Also, breaking up Google would immediately benefit Microsoft’s Bing search engine. Awkwardly left unaddressed by the House Judiciary report, Microsoft’s strengths and capabilities appear ignored, but the Bing search engine (as well as Microsoft Edge’s browser) and associated ad revenue will correspondingly increase with the regulatory fall of Google should a breakup ever occur. The current competition for Bing to tackle the incumbent Google would be artificially bent so that it favors one competitor over another irrespective of consumer benefits or harms. Moreover, if it is not Bing that will win from Google’s breakup, it may very well be the Chinese Baidu or the other Chinese entrant, Huawei.

The breakup of Google will unleash some innovations as argued by those suggesting it, but it will prevent Google from being the engine of innovation we currently have to “escape” competition by a superior efficiency. Google’s trading partners writ large will be harmed (from Apple to device manufacturers through online advertisers) as well as Google’s consumers. Unleashed innovations might be expected from breakups, but only if the targeted company is neither innovative nor competitive. With billions of dollars spent annually on R&D for innovation, and with hundreds of innovations made reality, the breakup of Google can hardly unleash any sort of innovation from the market, except some rents undeservedly earned by rivals thanks to rent-seeking through strategic lawsuits as we have seen in Europe.

Breakups of companies are provided for by antitrust laws, especially for busting trusts—these inefficient cartels which conspire to harm consumers and stifle innovation. Breakups of companies are extremely unfair for dominant players that earned their success through innovative products, competitive offers, and a long-term vision that discounted short-term benefits in favor of long-term corporate strategy. Breakups may eventually yield much greater costs to consumers and to overall innovation than the expected rivals’ benefits to be generated with short-lived spinoff companies. For sure, breakups of these companies may always generate massive media coverage, popular support—especially when announced two weeks before a historical presidential election. Political tactics force decisions to be rushed out with little gains for consumers or democracy but with great self-inflicted pains imposed on innovation and the quality of regulatory decision-making processes. Searching for sanity in electoral politics may prove as complicated as searching for sanity in modern antitrust enforcement.