The good news for long-standing practitioners in the field of competition policy—of which I am one—is the increased media attention and policy focus on the extent to which there is the exercise of market power by our domestic “super-star” tech companies, such as Facebook, Amazon, Apple, Netflix and Google (the so-called FAANGs). Truth be told, we antitrust economists have been engaged in carrying out research on this question for several years now. The approach being taken in such research has generally been thoughtful and systematic, which has not always been the case when highly charged economic issues are on the table. In large part this stems from the fact that many of the core questions at hand—such as how can the extent of market power be accurately assessed when a product or service becomes more effective, and thus more sought after, the greater the number of users on a platform--are frankly at the frontier of the economics profession.
That an increasing number of antitrust policymakers—not just in the U.S. but also in other countries—appreciates the research complexities these new market dynamics pose is most welcome. They seem to realize a lot is at stake to accurately understand the benefits and costs of these technologies before policy decisions—one way or another—are taken
The bad news is that in some camps—whether in the economics profession or among policy makers—too little attention is being devoted to assessing the role of the FAANGs as the engines of the country’s innovation in advanced technology and our economy's long-run competitiveness in the international marketplace. It is not enough to focus only on trying to achieve a balance between mitigating market dominance and improving consumers' and workers' welfare--the traditional antitrust or competition policy prism. Rather the more fundamental question to be answered is how to not only enhance competition among U.S. firms, but also at the same time strengthen incentives for national innovation and bolster the U.S’s technological footing globally.
What’s lacking is careful scrutiny of the intersection of policies that both promote market competition and national security. This is particularly the case with respect to conduct by foreign rivals--especially those with state backing—not just in their home markets and third-countries, but as they try to make inroads in the U.S. market. In the worst of cases, U.S. authorities may be making policy decisions related to foreign businesses that contravene both U.S. national security and our firms’ ability to be competitive.
Whether this uneven report card is the result of the antitrust and national security communities’ siloed thinking, reluctance to take a fact-based approach, or an inability to take a proactive stance, is not, as a practical matter, important. What matters is this: consistency in promoting the competitiveness of U.S. firms operating in an increasingly complex set of structures within a globalized marketplace.
Let’s Start with the Good: The antitrust community’s approach to the FAANGS and their far-reaching impact on consumer prices, employment, wages, innovation and our lifestyles.
With few exceptions, there is broad agreement that over the last two decades the omnipresence of the FAANGs means that high-tech industries in the U.S., whether defined in product or geographic terms, have become more concentrated. That is, the number and size distribution of U.S. technology firms selling a particular set of goods and services, or purchasing certain inputs, including the hiring of specific types of workers, has decreased.
Amazon’s e-commerce business--which initially only sold books--is probably the most visible example of increased dominance in the digital retail sales space, an industry that for all intents and purposes didn’t exist prior to 1990. Amazon has genuinely revolutionized shopping in that short timeframe. From a computer anywhere in the world one can buy virtually any retail item without having to go to a store; view instantly data on other customers’ reviews of the item; and after charging the purchase on a credit card, the item will be delivered within days (with nominal or sometimes free shipping); and returning items is just as easy.
The time and cost efficiencies of shopping on Amazon is nothing short of astounding. But many Americans bemoan the rapid growth of Amazon’s business model has meant the disappearance of local, small shops from which to buy consumer goods. Yet it’s also the case that other large players in the retail sales sector—far more established than Amazon, such as Walmart and Target—have also increased their market footprints substantially. But they have done so largely on a brick and mortar basis. In so doing, like Amazon, they have also edged out small retail proprietors.
Moreover, Walmart and Target have themselves begun to adopt an aggressive e-commerce strategy to try to compete head to head with Amazon in various—though hardly all—of its market segments and services. In a word, Walmart and Target quickly learned from, and are now adopting, the ingenuity of Jeff Bezos’ e-commerce business model.
By the same token, as Amazon has expanded and matured, it’s realized that its Seattle-based brick and mortar centralized system of control is no longer proving to be an optimal business paradigm. The result? The company is now in the throes of building a second hub on the East Coast of the U.S. And even more ironic, Amazon brick and mortar bookstores are now appearing in certain cities!
While it would be highly premature to suggest that a fundamental convergence between Amazon, Walmart and Target is underway, it is clear that, at least in this industry, market dynamics are robust, competition is thriving, and innovation is hardly the province of one firm. Still, Amazon’s secret sauce of e-commerce has positioned it to be by far the biggest player in the retail market—by a long shot.
Among the rest of the FAANGs, as well as in other high-tech industries in the U.S. where super-large companies also have become dominant, one sees a similar pattern, although there is considerable variation.
It stands to reason that the market dominance of the FAANGs means they have become a focus of the antitrust authorities.
But from an antitrust perspective, one of the key questions is not simply whether a market is concentrated. Rather, it is whether as result of such dominance, firms operating in such markets have the ability to both charge prices in excess of their costs, earning margins higher than what otherwise would be the case if there were more firms in the market, and erect entry barriers, through a variety of behaviors, that prevent potential rivals from operating successfully in the market and compete down those higher margins.
Needless to say, the critical issue in assessing the competitive impacts of the FAANGs (as is the case for other firms) is what gives rise to concentration that engenders the potential to engage in anti-competitive conduct. There are three broad possibilities.
(i) Is it the result of acquisitions or mergers of otherwise separate firms?
If so, there is little question there is a vital role for antitrust enforcement to prevent such transactions. And that has been the case in practice. For many years, the U.S. has defined certain market share thresholds above which acquisition or mergers will be challenged by the authorities. Most antitrust economists would argue that approach should be vigorously enforced economy-wide, especially in today’s increasingly concentrated markets.
To that end, if one believes Amazon is able to exercise market power today as a result of its past acquisitions of, say, Whole Foods or Zappos, that raises a set of obvious questions: did the antitrust authorities at the time of these acquisitions come to the conclusion that these transactions were benign competitively? Or did they just look the other way? The same argument may apply to Google’s acquisition of YouTube; Facebook’s take-over of WhatsApp; or Microsoft’s purchase of Skype.
(ii) Is it the case there is something truly inherent in a particular industry’s underlying technology that gives rise to extensive “economies of scale” or "economies of scope”— meaning that only an extraordinarily small number of suppliers could ever be commercially viable?
A classic example of such a “natural monopoly” industry is local distribution of natural gas. If multiple companies set out to lay numerous networks of natural gas pipelines in one city to supply neighboring households or firms, in short order they will all find they’re simply unable to recoup their costs of “duplicative facilities”. That is why such industries are regulated: in return for being sanctioned as the sole (or one of a select few) supplier(s) in the market, the prices that can be charged are set by a public authority.
Of course, advances in technology can drastically alter the underpinnings of what was once a natural monopoly into a competitive market. Those old enough to remember when telephony service was once provided only terrestrially through lines strung across telephone poles, will recall the Bell Telephone monopoly regulated by the Federal Communications Commission. The advent of cellular telecommunications turned the monopoly structure of the telephony service market on its head.
In the case of the FAANGs, there are not just the traditional static economies of scale and scope at play; rather, such efficiencies are inherently dynamic. Specifically, the more users that are engaged on their platforms, the greater the gains to each and every user. As a result, the overall value of the network increases exponentially. Perhaps the most obvious examples of this phenomenon are LinkedIn and Facebook. But it also pertains to others, for example, Amazon.
(iii) Is market dominance achieved through organic growth as a result of providing a wholly new product or service?
Organic growth was largely the main story behind the advent of Amazon. Of course, there are now some exceptions to that story, not only as a result of Amazon’s acquisition of Whole Foods and Zappos, but also because the company has begun sales of its own branded goods on its platform—a form of vertical integration. While these acquisitions or the move to integrate vertically may be market reinforcing in and of themselves, the operative question with respect to loss of competition is whether they create incentives for Amazon to use its platform to engage in exclusionary conduct that harms other users of their platforms? That would arise, for example, if Amazon were to exclude any other suppliers of shoes or food to sell their goods on Amazon. Still, there are certainly other on-line marketplaces, such as eBay, and many traditional brick and mortar stores are very active in on-line sales themselves.
Google is a far more nuanced example, since it is hardly a wholly new product or service. But it does illustrate in stark relief a FAANG market that is prone to much churning. Some might argue that Google’s preeminent position in today’s search engine market is the result of, say, Microsoft Edge’s inability to provide the same quality search results through Bing. Others would have it that Google is engaging in predatory practices to prevent Edge from gaining market share. Google’s present dominance certainly looks assured as far as the eye can see. But it’s not as if Mozilla Foxfire and Duck Duck Go are not increasingly playing disruptor roles. It’s also instructive to recall, that prior to Google, Microsoft’s precursor to Edge, Internet Explorer, was the one-time dominant player. In those days, the rival search engine, Netscape, began to lose market share for a variety of reasons. While some of them were the result of poor business strategy decisions by Netscape, it also was tough to compete against Microsoft, which bundled Internet Explorer within its core PC software—a practice that was the subject of significant antitrust suit brought against Microsoft.
In addressing those three main issues some of the most salient questions about the FAANGs are currently the subject of fervent debate:
- What criteria should be used to define the relevant product and geographic market boundaries from which economically meaningful market shares and concentration can be assessed?
- What is the most accurate approach to measure the extent to which prices charged and profits earned by FAANGs are above what otherwise would prevail?
- How much of the slowness of wage growth in the U.S. can be attributed to the market power (“monopsony power”) the FAANGs are exercising in the labor market if there are few other employers present to bid up wages to entice away workers?
- If the FAANGs are creating barriers to entry, how sustainable are they, especially in such sectors where inherently technology is clearly changing rapidly?
- How should the time-saving efficiencies in retail purchasing introduced in the market by firms like Amazon be factored in as a contributor to boosting productivity in the economy?
Suffice it to say, assessing the competitive impacts of the FAANGs is no simple matter. While such firms may well engender economic and social costs, the benefits they provide are unquestionably substantial. Wisely, antitrust officials in the U.S. executive branch are taking a clear-headed, inclusive and interdisciplinary approach in coming to judgments about them. Let us hope it continues. Of course, whether the legislative branch adopts the same strategy may be dubious. (I say this as a former member of a U.S. Senate professional committee staff; so I have seen the “sausage-making” process up close!)
But does the FAANGs’ technological complexity, including new, dynamic forms of economies of scale and scope, mean that our existing antitrust statutes and other forms of enforcement are now wholly obsolete? Most in the antitrust world today would argue that the burden of proof rests with those who would advocate there is the need for new enforcement regimes.
Indeed, in this environment, most antitrust officials, counseled by antitrust economists, are smart to take a cautious, deliberative approach in deciding when (and when not) to apply the statutory authority they possess to the FAANGs. If anything, they seem to have made the judgement—at this juncture at least—that the real concerns expressed by the public today about the FAANGs is their potential to exercise political power rather than erode competition and decrease in consumer welfare. If that is in fact the case, the area where public policy may well need to play a role vis a via the FAANGs is in the realm of campaign finance and corruption statues rather than antitrust law.
Now Let’s Turn to the Bad: The intersection of national security and antitrust.
Regrettably, the thoughtful, proactive approach taken by antitrust policy-makers and the bar in the case of the FAANGs does not get readily carried over in advanced technology sectors where there is a systemic intersection between market competition and national security. There are two dimensions where there are ruptures.
The first concerns the role the antitrust community has played—indeed its perception of how much it should play—in governmental decision-making when foreign entities seek to acquire significant U.S. companies in these sectors.
As is increasingly well known, that process is governed by the Committee on Foreign Investment in the United States (CFIUS), an interagency entity initially launched in 1975 and chaired by the Treasury Department. In the decades since, CFIUS’ authority has been strengthened, especially through the enactment of several statutes, most recently The Foreign Investment Risk Review Modernization Act (FIRRMA), which was signed into law in August 2018 by President Trump.
In part, FIRRMA was enacted because of the extraordinarily rapid increase in the number of proposed acquisitions of domestic firms by foreign parties affiliated in some fashion, including through outright ownership, by states whose objectives, whether implicitly or explicitly, run counter to enhancing the vitality and innovative prowess of U.S. companies that form the core of the international competitiveness of the U.S. economy. In fact, in some instances, the contemplated acquisitions of U.S. firms pose direct threats US national security.
While China is the clearest case in point in this regard, it is by no means the only such state.
But the need for FIRRMA also stemmed from the fact that the criteria underlying CFIUS’ decision-making process were in serious need of being defined with greater specificity and issued far more transparently. Here I speak from experience as a former member of CFIUS during my time in the White House in the 1990s.
It is only fair, of course, that investors in firms in which proposed acquisitions are subject to CFIUS regulation need to be in a position in which to intelligently judge the odds of the extent to which a transaction will be approved by CFIUS outright; approved subject to certain modifications being made (such as divestiture of certain segments) or blocked. Perhaps ironically, these new CFIUS criteria are somewhat akin to the merger guidelines the antitrust authorities have used for years as noted earlier.
The antitrust authorities certainly have a seat at the CFIUS table. After all if a proposed acquisition by a foreign entity pertains to a U.S. supplier of a highly sensitive advanced technology for which there is only one of two other suppliers, or in the extreme, the target is the sole supplier—that is, the domestic market is already highly concentrated—the potential impact of the loss of control and ownership of that firm to a risky foreign entity on the national security of the U.S. is rather self-evident.
Surprisingly, however, the role that the antitrust community has taken in the CFIUS decision-making process to date—whether it’s the antitrust policy officials in CFIUS meetings per se or external antitrust lawyers advising clients on CFIUS transactions—has been limited, and certainly could not be characterized as “proactive”.
In law firms, CFIUS practices and antitrust practices tend to work in their silos. And, within CFIUS meetings, more often than not it is the defense and intelligence agencies, as well as the Treasury and Commerce Departments, that are the main players that opine on the competitive and security risks to the economy that could arise from the loss of domestic supplier if a particular acquisition by a foreign firm, especially one with ties to an adversarial state, were allowed to be consummated.
To say such scenarios, where the antitrust authorities are not front and center is puzzling, would be an understatement. After all, they occupy the prime pole position to render judgments as to what are the potential barriers to entry that could prevent new domestic suppliers to populate the market in question. This is hardly the comparative expertise of Treasury or Commerce, let alone the Pentagon or the CIA.
Yet, competition and national security are inextricably linked. Thus, in their consideration of the economic and social benefits and costs of mergers, acquisitions, market dominance, pricing conduct, etc. that could impair or enhance competition, antitrust officials and lawyers cannot ignore the impacts these actions can have on the country’s national security. This is the logical counterpart to the way CFIUS officials and lawyers should be taking into account changes in market competition when making their national security assessments of inbound foreign investment.
Yet, as a case in point, consider the events of last year when Broadcom, the Singapore-based semiconductor firm with ties to China (at least informally but more likely formally as well) was blocked by CFIUS from acquiring Qualcomm—the dominant and most innovative U.S. firms in that industry, whose mobile chips are widely used in cell phones in the U.S. and who stands at the forefront of developing 5G technology.
While the antitrust aspects underpinning CFIUS’ decision were certainly an element—that is, there isn’t another U.S. firm in the industry of which Qualcomm is a part that comes close to possessing the innovative prowess of Qualcomm and thus no other U.S. firm that could step in if Qualcomm had been acquired by Broadcom—the antitrust authorities and bar themselves were not seen out in front articulating those points as vital dimensions of the rationale for halting Broadcom’s attempt. Without the richness of that explanation, much of the public chalked up CFIUS’ decisions to China bashing.
The second failure to integrate antitrust and security concerns is far more serious. The Federal Trade Commission (FTC) brought a suit against Qualcomm more than two years ago, accusing the company of operating a monopoly in wireless chips, forcing its customers--handset manufacturers like Apple and Samsung--to work with it exclusively and charging excessive licensing fees for its technology.
Even putting aside the merits of the FTC’s suit—which many antitrust economists have had difficulty supporting—the FTC pursued its case as if the CFIUS decision last year prohibiting Broadcom’s takeover of Qualcomm on national security grounds never occurred.
Perhaps even more perverse, one of the star witnesses the FTC enlisted to support its argument was Huawei—the Chinese telecom giant that the U.S. defense and intelligence agencies, as well as the Justice, Treasury and Commerce Departments, among others, uniformly deem to be a prime threat to U.S. national security. It is as if the right arm of the U.S. government is unaware of what its left arm is doing.
Worse still, in early December 2018 Huawei’s CFO, Meng Wanzhou, the daughter of Huawei founder Ren Zhengfei, was arrested in Canada at the behest of the U.S. Department, where she has been charged with fraud and violating U.S. laws of doing business with Iran.
The FTC’s case against Qualcomm went to trial on January 4, 2019 and closing statements were presented on January 29, 2019. The presiding judge, Lucy Koh, rendered her decision on May 21, 2019. She ruled that Qualcomm used its monopoly power to coerce companies into paying “unreasonably high” royalties on Qualcomm’s wireless inventions. In particular, she found that Qualcomm’s practice of requiring handset makers to sign a patent license agreement before the handset maker could purchase their modem chips was deemed anticompetitive.
The decision was a major disappointment for two significant reasons--in both the antitrust and national security camps at the same time!
First, as a number of antitrust economic and legal experts have opined subsequent to Judge Koh's ruling in a number of fora, including in an Amicus Brief (for which I was asked to be a signatory and happily did so) to appeal and seek a reversal of her decision, her arguments were not based on a "rule of reason" approach required under modern antitrust law. Rather, she applied an approach more akin to now-abrogated "per se" rules. And she did so with only generalized evidence and theoretical support.
Second, her ruling failed to close the seemingly ever widening gap between the antitrust and national security establishments in order formulate a logically consistent approach—one devoid of cross purposes—to enhance the long-term competitiveness of the U.S. economy.
This piece draws, in part, from a Keynote Dinner speech given at the Annual Meeting of the New York State Bar Association Antitrust Committee in New York City on January 17, 2019. It was updated subsequent to Judge Koh's May 21, 2019 decision on FTC v Qualcomm.
