Technology companies based in Seattle or Silicon Valley now account for five out of the five most valuable companies in America, leading to a spate of commentary last year from lawyers like Columbia’s Tim Wu to economists like Harvard’s Kenneth Rogoff arguing that Big Tech has, in some sense, gotten “too big.” And in 2019, politicians are starting to listen.
One of these five, Microsoft, was the subject of considerable antitrust scrutiny in the late 1990s. Another, Google, has been hit by European antitrust regulators for major fines. But in recent years, US regulators have taken an optimistic view of Big Tech’s relentless growth — seeing a world of improving smartphone quality, free online services, and cheap e-commerce as evidence that competitive markets are working as intended.
At the same time, big technology companies have been brutalizing suppliers and competitors (and, not irrelevantly, publishers of journalism) with a range of questionable tactics — including Amazon’s price wars against competitors, Apple’s high-handed management of its own App Store, and Facebook’s long parade of privacy scandals. As the rich get richer, the criticism has become more intense.
Congressional Democrats have formally committed themselves to pushing for more stringent antitrust enforcement, and presidential candidate Sen. Elizabeth Warren has even called to dismantle several of these companies. Few politicians go quite as far as Warren, but most of the Democrats running for president in 2020 have called for more stringent regulation of technology companies and several Republicans including Sens. Ted Cruz (R-TX) and Josh Hawley (R-MO) are even making noises about it.
The issue is complicated by the fact that even though it’s convenient to shorthand Microsoft, Google, Amazon, Apple, and Facebook as “Big Tech,” these five companies are actually structured in very different ways.
Facebook is a pretty narrowly focused company, running three big online services (Instagram, WhatsApp, and Facebook itself) that might superficially look like competitors. Apple is a behemoth of vertical integration, producing hardware, software, and online services that mostly work as an integrated whole. But Google, Amazon, and Microsoft have all become pretty classic conglomerates, with a sprawling array of loosely related businesses under one roof.
Threading the needle is the trick. Antitrust enforcement in the technology sector has been so lax in recent years — while Warren’s proposal is so drastic — that there should be options to tighten the screws that would fall short of exploding multiple major corporations. Meanwhile, many ideas that have garnered at least some political support (reviving old laws aimed at curbing “predatory pricing,” for example) would impact a much larger range of companies than the big five.
Beyond the economic and legal details, there’s also a larger struggle over the cultural meaning of Big Tech. After a period in which technology entrepreneurs in particular were often celebrated as “good guys” of the business world — especially in contrast with the bankers of Wall Street — critics are aiming less at a specific legal point and more at a general sense that the richest companies in the world (and the billionaires who own them) are part of the problem. At the same time, antitrust law is a blunt-force weapon, having been developed over the decades to deal with a set of concerns that only partially overlap with the nexus of issues people have about modern technology conglomerates.
Contemporary antitrust law mostly cares about high prices
The main laws governing antitrust policy in the United States are both very old and very broadly worded.
The Clayton Antitrust Act, for example, is more than 100 years old and its predecessor the Sherman Act is even older. Both arose in an era when increasingly financial sophistication was allowing the creation of large industrial organizations — often with classic Gilded Age generic names like US Steel, Standard Oil, and the American Sugar Refining Company — that dominated their respective industries. Rather than specifying any particular analysis to address worries about monopolization, the Clayton Act simply bars one company from acquiring another when “the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly” and does not offer much in the way of further definition or elaboration of what that means.
In some sense, for example, back when Facebook bought Instagram, that was clearly an acquisition that lessened competition in the sense that both companies made smartphone apps that competed for a supply of time and attention and ad dollars. But Instagram at the time had only 30 million users (and barely a dozen employees). So did it substantially lessen competition?
Both regulators and courts have taken different views as to what that means over the years, but since the 1980s — operating under the influence of the “law and economics” movement, which sought to recast the legal concept of competition in terms of economic efficiency — the judicial system has defined these kinds of antitrust questions in terms of their impact on the welfare of consumers. And while there is more to consumer welfare than price alone, in a practical sense, looking at prices (which are obviously important and also lend themselves to being measured in an objective way) has been the dominant strand of consumer welfare analysis.
This sometimes leads to “anti-monopoly” policies that can sound a little perverse. A few years back, for example, Amazon essentially monopolized the market for e-books. Major book publishers fought back by teaming up to take on the bigger company and the Justice Department filed an antitrust suit against them. Why? Well, Amazon was using its power in the marketplace to keep e-book prices low. The publishers, the government argued, were trying to form a cartel to force Amazon to raise prices. And, indeed, even though the publishers ended up settling with the government, the introduction of more competition into the e-book marketplace (primarily from Apple) has had the impact of making e-books more expensive than they were when Amazon ruled the roost. The standard, in other words, isn’t that one company dominating a market is bad. It’s that it’s bad if a company’s market domination leads to bad outcomes for consumers.
Back to Facebook and Instagram. At the time, few observers saw how significant this deal was. But technology industry analyst Ben Thompson told the Code Conference audience last year that allowing this acquisition was “the greatest regulatory failure of the last 10 years” by allowing Facebook to entrench its dominance of social media. Yet under the contemporary antitrust framework, one might argue there’s no harm to consumers here — Facebook and Instagram are both free, so there’s after all no increase in prices. Yes, the fact that the combined entity is such an advertising juggernaut, pulling in $17 billion last quarter, is a big problem for other companies trying to sell ads (such as publishing companies that use ad revenue to fund actual journalism, for example) — but that’s not necessarily a problem for consumers.
Big tech companies generate a lot of non-price complaints
A now-infamous story Brad Stone recounts in his book about Amazon, The Everything Store, tells of Jeff Bezos’s thuggish behavior toward a startup called Quidsi that was trying to break into the e-commerce business with brands like Diapers.com and Soap.com. Bezos took notice of the company and dispatched an Amazon senior vice president to have lunch with Quidsi’s founders. The SVP’s message was simple: Amazon was thinking of getting into the diaper business and Quidsi should consider selling to Amazon as a way to make it happen.
Quidsi didn’t want to sell. So Bezos made clear it was an “offer you can’t refuse”:
Soon after, Quidsi noticed Amazon dropping prices up to 30 percent on diapers and other baby products. As an experiment, Quidsi executives manipulated their prices and then watched as Amazon’s website changed its prices accordingly. Amazon’s pricing bots — software that carefully monitors other companies’ prices and adjusts Amazon’s to match — were tracking Diapers.com.
This began to worry investors, who became hesitant to put more money into Quidsi. And then Amazon ratcheted up the stakes by rolling out a new service called Amazon Mom that offered huge discounts and free shipping on diapers and other baby supplies. Quidsi executives calculated, according to Stone, that “Amazon was on track to lose $100 million over three months in the diaper category alone.” By November 2010, Quidsi’s board agreed to sell to Amazon. A few years later, the Amazon Mom program ended; a couple of years after that, the Diapers.com brand went away entirely.
Behavior of this sort violates a lot of people’s sense of fair play. But it doesn’t raise prices. Indeed, it arguably lowers them — at least in the short term.
And it’s not the only way Amazon plays hardball. Amazon acts as a marketplace for many third-party vendors, but it also sells Amazon-branded goods under labels like Amazon Basics and Amazon Essentials but also less obvious names including Lark & Ro, North Eleven, and Society New York. European competition authorities suspect that Amazon uses data it gathers in its capacity as a marketplace to develop copycat first-party products, a practice that certainly feels kind of sleazy — though, of course, not entirely different from the house brands you’ll find at any supermarket or department store. But, again, it is not harmful to consumers in any particularly obvious way. Apple, somewhat similarly, generates a steady stream of complaints from iOS developers about its App Store policies.
But in both cases, vendors often feel the large platforms are so dominant that they have no choice but to participate on them.
Facebook, meanwhile, has been besieged by media complaints about privacy and fake news. But not only do approximately a couple billion users derive daily entertainment from the free service, but the service’s sheer ubiquity makes it hard to leave. Google, like Facebook, also uses black box algorithms to surface content to users that can dramatically sway the fortunes of individual publishers’ fates while also raising questions about editorial skew.
These are all concerns that at least could fall under the scope of American antitrust law, but mostly haven’t.
Democrats want to do something on antitrust reform
It was the Reagan administration that entrenched the modern era of price-centric antitrust enforcement as part of a generally business-friendly regulatory stance. Reagan’s broad approach to antitrust policy remained in force through Bill Clinton’s term in the White House, though the Democratic Party’s fundamentally more regulation-friendly philosophy did manifest itself in a significant antitrust case against Microsoft, which ended with a commitment to disentangle the Internet Explorer web browser from the Windows operating system.
Under George W. Bush, enforcement got laxer: Mergers like Sirius and XM in satellite radio and Maytag and Whirlpool in appliances were waived through. Under Obama, regulatory activism came back in style to an extent (proposed takeovers of T-Mobile by AT&T and Sprint were blocked, for example), but the Reagan-era conceptual framework stayed in place.
By Obama’s final year in office, however, his Council of Economic Advisers (CEA) issued a report sounding the alarm about an increase in economic concentration across almost all economic sectors whose deleterious impacts extended beyond higher prices to potentially playing a role in declining investment and stifling wage growth. The CEA report further called for “examination” of “market structure changes throughout the supply chain” — and then in 2017, congressional Democrats formally committed themselves to a new antitrust agenda.
These proposals, turned into legislation by presidential candidate Sen. Amy Klobuchar, do not directly target the technology industry.
But they do embed two big changes that are directly relevant to Big Tech.
- One is that Democrats are calling for new standards that “will prevent not only mergers that unfairly increase prices but also those that unfairly reduce competition — they will ensure that regulators carefully scrutinize whether mergers reduce wages, cut jobs, lower product quality, limit access to services, stifle innovation, or hinder the ability of small businesses and entrepreneurs to compete.”
- The other is that they want to shift the rules so that “the largest mergers would be presumed to be anticompetitive and would be blocked unless the merging firms could establish the benefits of the deal.”
These moves would both dramatically expand the scope of harms that antitrust regulators consider and make life more difficult for merging firms by forcing them to prove themselves innocent rather than leaving the burden on the government. The implications of that kind of thinking for the five big technology giants is obvious. But what the legislation actually does is change merger approval rules. And while today’s big tech companies certainly do their share of acquisitions, nothing in this suite of proposals would necessarily impact the behavior of existing tech giants.
That’s where Elizabeth Warren comes in.
Elizabeth Warren wants to break up giant tech conglomerates
In early March, Warren laid out a proposal that would lead to the breakup of Google, Amazon, and Facebook while imposing some significant restrictions on how Apple and Microsoft do business.
As a sports fan, she explains this in terms of a baseball analogy.
“You can be the umpire in a baseball game or you can have a team in the game,” Warren says, “but you don’t get to be the umpire and have a team in the game.”
Giant tech companies have too much power. My plan to #BreakUpBigTech prevents corporations like Amazon from knocking out the rest of the competition. You can be an umpire, or you can be a player—but you can’t be both. #WarrenTownHall pic.twitter.com/73y1002QVv— Elizabeth Warren (@ewarren) April 23, 2019
In other words, you can’t both be the owner of a technology platform where people find things and make the things that people find on the platforms. One could, however, easily find flaw with this analogy. It’s true that Major League Baseball relies on neutral umpires to guarantee fair play, but the team owners themselves write the rules of the game and collectively hire and fire the umpires who enforce them. Warren-style platform neutrality might be closer to requiring that Major League Baseball be an entirely distinct business from its 30 constituent franchises rather than being controlled by them. (The fact that courts have ruled MLB is exempt from antitrust law makes this a particularly curious example.)
Specifically, under her plan, a company with annual global revenue over $25 billion that “offers to the public an online marketplace, an exchange, or a platform for connecting third parties” would be designated as a “platform utility.” A platform utility would be barred from owning any of the participants on the platform. So no Amazon Basics batteries in the online store, only batteries made by third parties like Duracell. No Google Reviews of local restaurants on the search page, only search results for reviews by Yelp and other third parties.
Smaller companies (those with annual global revenue between $90 million and $25 million) would face a lower regulatory barrier — required to meet a standard of “fair, reasonable, and nondiscriminatory dealing with users” but not required to structurally separate in the same way as the major platforms. European antitrust law already applies something like this standard and has hit Google with fines for favoring Google content in Google searches.
This would not really impact the core business models of Microsoft and Apple in the same way as it would Google, Facebook, and Amazon, but as Warren herself later acknowledged, this would render illegal the current setup of the iOS App Store and Microsoft’s Xbox Store. The fact that the impact on the two most valuable technology companies in the world was not something Warren explicitly noted in her rollout of the policy is perhaps a sign that her proposal is, on some level, more symbolic than practical. Nevertheless, it’s an important marker in what is a major ongoing intra-Democratic clash — with some very important implications.
Democrats’ big argument: Is Big Tech good?
The odds that the next president is going to push a major legislative reform of the antitrust process through Congress are fairly minimal.
But the next president will certainly appoint a director of the Justice Department’s antitrust division. And he or she will also appoint a head of the Federal Trade Commission. These are the key regulatory agencies that decide when the federal government will or won’t take action in antitrust cases. The next president will also appoint federal judges whose rulings make up the substance of antitrust law; a Federal Communications Commission chair whose work, while not technically antitrust per se, is relevant to the interests of the technology industry; and diplomats and trade policy officials whose work (or lack thereof) on behalf of US-based technology giants will make a big difference to their fates.
In that light, a relevant aspect of the 2020 campaign is the fact that while most businesses lean Republican, the behemoths of Silicon Valley have traditionally had fairly close ties to the Democratic Party.
Al Gore is on the board of Apple, and Barack Obama’s EPA director, Lisa Jackson, runs their environmental initiatives. Facebook chief operating officer Sheryl Sandberg worked in Bill Clinton’s administration and is a bit of a protégé of Larry Summers, who was a senior economic policymaker under both Clinton and Obama. Mark Penn, a former top aide to both Bill and Hillary Clinton, was for years Microsoft’s executive vice president of strategy. Obama’s first press secretary, Jay Carney, is now a top Amazon executive. And the ties between the Obama White House and Google are almost too extensive to summarize. But as the Intercept’s David Dayen documented in April 2016, 55 Google employees moved into federal appointments under Obama, while 197 Obama administration appointees landed at Google after leaving the government.
In short, the revolving door to Silicon Valley has largely displaced the one to Wall Street as the cash-in of choice for top Democrats. And as far as major industries go, tech is in some ways a natural partner for Democrats.
The industry is based primarily in blue states, so if it wants to have clout in Washington it’s going need allies on the Democratic side of the aisle. And though all businesses love Republican tax policy, the highly globalized and immigrant-heavy high-tech industry has serious practical problems with Trump’s brand of economic nationalism and major conceptual problems with his nostalgia-inflected view of prosperity.
But conversely, if Democrats really want to take on the concentrated economic and political power of the billionaire class — as many of them say they do — they have to take on the industry that’s made so many of today’s largest fortunes. That could specifically mean antitrust policy, but it could also mean a dozen other things, from higher taxes to more union organizing to rethinking the priorities in American trade policy. Whatever it means, it requires establishing an arm’s-length, or even antagonistic, relationship with technology companies and their top executives rather than the cozy, Obama-era ones.
And a clear antitrust commitment to taking on Big Tech serves, in part, as a strong signal of a turning of the page. Conversely, it’s not a coincidence that Eric Schmidt, Google’s former CEO, threw a massive fundraiser for Joe Biden — the candidate of explicit continuity with the Obama years.
The risk in all this, however, is that big technology companies are not necessarily the best choice of villain for a new era of regulatory activism.
Big Tech is still mostly popular
The volume of backlash against big technology companies from political and media figures can serve to obscure the extent to which most of these companies remain very popular.
Local New York elected officials, including Rep. Alexandria Ocasio-Cortez, scuttled a deal to bring a large Amazon office to Queens in exchange for tax subsidies — but New Yorkers consistently tell pollsters they wish she hadn’t. Indeed, Amazon in particular is wildly popular. When Georgetown University’s Baker Center polled Americans to ask about their confidence in various institutions, Amazon was No. 1 among Democrats and No. 3 among Republicans. Gallup’s most recent poll showed both the “computer industry” and the “internet industry” had net positive favorable ratings. And while recent privacy scandals seem to have hurt Facebook, an Axios/Harris poll showed the other four of the big five to all be in the top half of the corporate popularity rankings.
If you want to make the case for getting tough on big business, old reliables like banking, telecoms, and the pharmaceutical industry seem like softer targets.
And the underlying premise of the new antitrust push helps underscore one possible reason big tech remains popular: These companies are mostly in the business of offering good deals to consumers.
If Google were using its dominance of the search market to gouge people on prices, the company would be wildly unpopular. And it would either lose market share (thus debunking the basic antitrust concern) or be vulnerable to traditional antitrust complaints. But Google isn’t doing that. Search is free. So is Gmail. So are apps like Google Docs and Google Sheets. So is Google’s somewhat miraculous Google Translate software. So is Google Maps.
Google offers a vast bounty of free services in order to maximize its data collection and optimize its advertising capabilities. That ends up annoying a huge range of stakeholders — including those who used to depend on getting a slice of the advertising pie — but the basic value proposition to consumers is really, really good; hence the interest in revising antitrust doctrine.
Similarly, Amazon is credibly accused of hurting suppliers, hurting competitors, and even hurting its own employees — but nobody can deny that it’s a cheap and convenient way to shop for a staggering array of things.
Of course, the proponents of a big shift in antitrust policy would deny the existence of a clear trade-off here. A less concentrated technology marketplace, they hope, would ultimately lead to more innovation and better outcomes for everyone.
And perhaps it would.
But part of the reason the consumer welfare standard has proven so appealing to lawyers, judges, and economists is that it turned antitrust decisions into a set of mathematical calculations about prices and quantities. A merger that seems likely to increase profits by raising prices is bad, but one that seems likely to increase profits by making it easier to lay off redundant workers or squeeze suppliers harder is good. This kind of simple calculus doesn’t capture everyone a reasonable person might care about, but it does set out a fairly clear set of rules for everyone to try to follow.
And consumers both like and enjoy a good deal when they see one. A political promise to stand up to nefarious special interests sounds good, but a promise to take on companies that are offering great bargains is a much dicier proposition.
Recode and Vox have joined forces to uncover and explain how our digital world is changing — and changing us. Subscribe to Recode podcasts to hear Kara Swisher and Peter Kafka lead the tough conversations the technology industry needs today.