Skip to content

How Do You Solve a Problem Like Amazon?

Fulfillment: Winning and Losing in One-Click America
by Alec MacGillis
Farrar, Straus, and Giroux, 2021, 400 pages

Amazon Unbound: Jeff Bezos and the Invention of a Global Empire
by Brad Stone
Simon & Schuster, 2021, 488 pages

Amazon is the sine qua non of Digital Age economic and social infrastructure. Its scope is massive: an original retail business (boasting a market share of 50 percent of all digital retail) and highly popular Prime subscription service; its indispensable Amazon Marketplace platform for third-party vendors; its market-leading Amazon Web Services cloud infrastructure; its Fulfillment-by-Amazon logistics and delivery service (including branded delivery trucks); its numerous other ventures from Kindle e-books to Alexa to Amazon Studios; its hugely popular Whole Foods supermarket empire (acquired in 2017 for a cool $13.7 billion); and its other Amazon Go and Amazon Books retail stores.

It is, in short, the one-stop shop that seems to bind together our fractured body politic. Or that’s what Jeff Bezos wants Americans to believe.

America’s third-largest company by market capitalization has filled the shoes of predecessors like Standard Oil, Sears, and General Electric as an epoch-defining corporation. Indeed, offering enhanced consumer convenience, Amazon now defines an entire way of life. Amazon’s profits soared nearly 200 percent last year as its national fulfillment infrastructure and rapid Prime delivery serviced a lockdown-stricken and home-confined populace. But in spite of its achievement of seamless logistics, Amazon’s dramatic surge has been neither victimless nor costless. Employment at Amazon is infamously tenuous, and the New York Times recently found the company’s black and Hispanic warehouse workers to be “almost 50 percent more likely to be fired” than their white colleagues.

Shining a spotlight on those victims and costs is the purpose of Alec MacGillis’s impressive recent work of investigative journalism, Fulfillment: Winning and Losing in One-Click America. Considering it alongside Brad Stone’s Amazon Unbound: Jeff Bezos and the Invention of a Global Empire, the reader is left with an unmistakable impression of long-term entrepreneurship combined with raw ambition and a Nietzschean will to power. MacGillis and Stone illuminate not only the more abstract losses of social capital and community wrought by Amazon, but also the tangible losses of brick-and-mortar retail stores—and of human life itself (Amazon suffers about 5.9 serious injuries for every 100 warehouse workers, about 80 percent more on average than its competitors).

Haves and Have-Nots

In 2017, long before the Covid-era ramp-up in Amazon’s warehousing and other “fulfillment” infrastructure (such as trucking), Scott Galloway of NYU’s Stern School of Business already estimated that Amazon would destroy seventy-six thousand retail jobs per annum. On net, it is estimated that Amazon has eliminated about twice as many retail jobs as it has created, depriving downtrodden localities of property tax revenue while bilking city councilmen into doling out grotesque tax subsidies. Taking a broader view of Amazon’s tax subsidy machine and its ever-widening regulatory capture, it is easy to see the truth of Franklin Foer’s observation (quoted by MacGillis) that cutting costs and taxes amounts to an “overriding corporate obsession” for Bezos.

MacGillis draws a striking picture of some of the local, regional, and national costs of the Amazonian makeover. Amazon’s success has played “an outsized role,” he says, in the nation’s “zero-sum sorting”—the broader structural transformation for which our present political paradigms have been largely ill-equipped. Amazon has accelerated the reconfiguration of America into a new world of haves and have-nots.

The rise of Seattle—an erstwhile hub of West Coast grunge transmogrified by Microsoft and Amazon into a preeminent playground of the gentry liberal imperium—encapsulates today’s “haves.” Transformed by its new arrivals, the Seattle MacGillis discovers peddles $200 martinis and a “wizard pub” where patrons can purchase custom-made wands. “The city had few children,” MacGillis writes, “but it had many adults with the disposable income needed to reenact childhood.” At one time a racially diverse city that was home to Jimi Hendrix and a strong middle class, Seattle has been transformed through the ingress of countless technologists and software developers into the city with the second-highest per-capita income behind only San Francisco. Sadly, it has also attained Bay Area-esque income inequality.

The great political bellwether of Ohio, which plays a large role in Fulfillment, best displays the divide between “have” and “have-not” cities. Bezos built up a great deal of Amazon’s infrastructure around the state capital of Columbus, which is home to Ohio’s flagship university and is well-positioned as a budding economic force. On the other hand, smaller peripheral cities such as Dayton and Akron—to say nothing of blue-collar exurbs such as Nelsonville—have been left economically barren and socially denuded, laden with mass despondency and opioid-induced tragedy, sapped of communitarian social capital, and struggling mightily to adapt and survive. MacGillis quotes a lachrymose Tim Grumbacher, former head of the now-defunct Bon-Ton, as saying that “the great thing about retail was face-to-face. . . . Not only with customers, but managers, with the associates that worked there. We had people who walked on the floor for fifty years. It was a community.”

Amazon is the poster child for the great maladies of our era: a division of economy and nation-state into sectors and regions that barely know each other, an insidious emptying of the reservoir of civil society, and the divvying up of the zero-sum economy. No city exemplifies the zero-sum economy quite like Washington, D.C., and the metropolitan area that is home to four of the nation’s six highest-income counties indeed features heavily in Amazon’s story. Amazon’s second headquarters—“HQ2”—is conspicuously located in suburban Virginia. The company chose to ground its East Coast operations there following a highly publicized bidding war; Virginia ultimately offered to shower the company with nearly $600 million in tax breaks and $23 million in cash grants.

Amazon now operates sprawling data centers throughout northern Virginia, extending as far out as the outer rings of exurban Prince William County. Amazon Web Services, the company’s industry-dominant cloud infrastructure service, works with 6,500 government agencies, including the Central Intelligence Agency. Yes, the same company that deplatformed Parler following the January 2021 U.S. Capitol riot also controls large swaths of the internet infrastructure for the nation’s spooks. Key intelligence community assets are exposed to the inherent risks of Amazon’s dominant cloud colossus.

Leadership Principles

Brad Stone’s Amazon Unbound explores the world’s wealthiest man and his 1990s-era vision to launch a global retail juggernaut. From a business and financial vantage point, that vision always entailed forsaking the pursuit of short- and medium-term profits in order to build out brute scale and gobble up as much user data as possible. Scale was, and remains, Amazon’s golden ticket to accomplishing everything it sets out to do. It is the means by which the company exerts pressure across distinct but vaguely related lines of business. It is the very “premise” of Amazon’s business model, as Lina Khan—then a Yale Law student, now Federal Trade Commissioner—put it in her highly influential 2017 Yale Law Journal article, “Amazon’s Antitrust Paradox.”

Bezos’s legacy as a visionary is undeniable, and the portrait that emerges from Stone’s work is that of an assertively hands-on leader. We learn how, in January 2011, the idea that eventually became the Alexa virtual assistant was spawned in a hilariously brief email the CEO sent to a trusted inner circle: “We should build a $20 device with its brains in the cloud that’s completely controlled by our voice.” The boss followed up on his terse missive by largely developing the ensuing project himself. In an episode indicative of the ludicrous anti-worker culture Bezos helped foster over the decades, Stone describes how the company hired temporary workers to spend days on end speaking into microphones in an attempt to slowly improve the quality of the nascent artificial intelligence product. In fairness, perhaps those conditions were at least less draconian than the company’s infamously injury-prone warehouse gigs. As MacGillis puts it, “unsettling questions about the asymmetric distribution of money and power” abound in the Digital Age.

Few episodes illustrate Bezos’s character more than the dramatically overhyped and tawdry “HQ2” national sweepstakes affair—a recurring subject for both MacGillis and Stone. Stone emphasizes Bezos’s micromanager tendencies, detailing how New York City only emerged as a possible final contender after he eschewed the company’s own studies on the matter and instead just followed his own instinct. (The company initially announced that it would split HQ2 between Queens, New York, and Arlington, Virginia, before dropping New York after strong local pushback.) MacGillis points out that Bezos’s focus on New York City and Washington, D.C., the two over-credentialed East Coast crown jewels of the zero-sum economy, was emblematic of the company’s overarching corporate ethos. Bezos could have easily chosen to revitalize a downtrodden city trying to make a comeback in the Digital Age. But HQ2, as MacGillis points out, “was not going to be an uplift project or an effort to knit together the American fabric . . . the company and Jeff Bezos had never provided much reason for anyone to believe that it would be.”

“Leaders have conviction and are tenacious,” states one of Amazon’s corporate “Leadership Principles.” “They do not compromise for the sake of social cohesion.” The ferocious battles waged by the company against warehouse unionization in places such as Bessemer, Alabama, illustrate Bezos’s commitment to these mantras. MacGillis quotes Nick Hanauer, an early Amazon investor who has since become a Bezos skeptic, on this point: “Jeff Bezos is a straight-up libertarian. . . . Jeff’s perspective is the canonical neoliberal perspective: that the only purpose of corporations, the only purpose of shareholders, is to enrich themselves to the exclusion of everyone else. . . . Maximize shareholder value and somehow that will create the common good.” Who needs Ayn Rand when you have Jeff Bezos?

Culture is Downstream from Amazon

MacGillis and Stone’s works have appeared during a time of record-breaking Amazon corporate profits, Bezos’s resignation as CEO of the company, and a broader political realignment that has many conservatives rethinking stale neoliberal orthodoxies as they have perhaps never done before. They key question for them is whether Amazon’s comprehensive business model and internal corporate culture—what New York Post op-ed Editor Sohrab Ahmari dubbedhomo amazonicus”—are, in Ahmari’s words, “compatible with the goods proper to family, faith, and community.”

Amazon, that leading digital driver of the one-click economy, has so much power across so many industries that upstream market power bleeds effortlessly into downstream cultural power. Hardly anyone batted an eyelash when, during Black History Month in February, Amazon suddenly decided to stop streaming the excellent Michael Pack–directed documentary, Created Equal: Clarence Thomas In His Own Words. Perhaps Amazon doesn’t believe Justice Thomas is the kind of black man worth celebrating. It’s Jeff Bezos’s world and we’re just living in it.

Thus the story of the rise of Amazon squarely presents questions pertaining to the role of the state in attending to (as Ahmari put it) the “common good of the whole,” not the “economic autonomy of any one enterprise or industry.” No less a free marketeer than Adam Smith himself would have easily intuited this. Smith decried the dangers inherent in tightly concentrated economic power and the conflation of the state and that power. In The Wealth of Nations, he observed the “strange absurdity” by which an ostensibly private entity, in Smith’s case the East India Company, might take on both the “character of the sovereign” and that of a “merchant.” In this passage, Smith was quite prophetic when it comes to the rise of Amazon, Google, and a handful of other Silicon Valley titans. “The Age of Big Tech, like the age of the robber baron, would be the age of monopoly,” wrote Sen. Josh Hawley (R-Mo.) in his recent book, The Tyranny of Big Tech.

Consider Amazon’s unilateral, unprompted February decision to delete all digital traces of Ryan T. Anderson’s 2018 book, When Harry Became Sally: Responding to the Transgender Moment. Amazon has a roughly 83 percent market share in national e-book sales. When its mandarins decide to eliminate a morally traditionalist book from its digital shelves, there are tangible ramifications far beyond the reach of the one book excised. Not only are readers deprived of a valuable addition to the proverbial “marketplace of ideas” liberals used to love, but authors are discouraged from writing them in the first place.

The chilling effects, in short, are real—and likely dramatic. Facebook and Twitter are properly criticized for their abuses of Section 230, the obscure liability shield provision in the 1996 Communications Decency Act. But the long-term effects of Amazon’s abuses are likely even more dire. Publishers become similarly disinclined to sign on those brave few authors willing to take such a risk to bring their works across the finish line. “If a book has even a small chance of crossing the censors at Amazon,” asked conservative Big Tech critic Rachel Bovard, “what financial incentive do publishers have to give it a green light?”

New Questions concerning Antitrust

A survey of Amazon suggests myriad other possible antitrust concerns. Stone in Amazon Unbound and MacGillis in Fulfillment both detail how the company uses its mammoth scale to exercise muscle over third-party Amazon Marketplace merchants and rivals. The company takes a standard 15 percent (technically ranging from 6 percent to 50 percent) cut from all sales that local suppliers make on the Marketplace, belying the notion that Amazon is merely a selfless conduit. Amazon’s 15 percent cut amounts to a de facto private tax on third-party sales, landing Amazon in Adam Smith’s awkward middle between “sovereign” and “merchant.”

Amazon also infamously culls its Marketplace sellers’ data and then uses that data trove to aggressively promote to online consumers its own near-copy “private label” Amazon Basics products—which range from batteries to pet supplies to bedding to diapers—in lieu of third-party competitors’ offerings. Bezos testified before the House antitrust subcommittee in July 2020, following Wall Street Journal reporting that called into question Amazon’s sanctimonious appeals to detached neutrality. Records obtained by the subcommittee at that time divulged how Amazon had referred to third-party Marketplace sellers as “internal competitors”—not the “partners” the company spoke of in public. Back in 2017, Khan had warned about precisely this threat in her Yale Law Journal article. “The fact that Amazon competes with many of the businesses that are coming to depend on it,” wrote Khan, “creates a host of conflicts of interest that the company can exploit to privilege its own products.” It turns out she could not have been more prescient.

Similarly, Stone in Amazon Unbound explains how Amazon once acquired a pivotal robot manufacturer and simply stopped shipping the robots to competitors. There are likely many other such cases. And perhaps most fundamentally, last year a remarkable 53 percent of American online shoppers began their product searches on Amazon. All of these various actions and data points raise serious questions under the Sherman Antitrust Act, the foundational but too-oft-neglected 1890 antitrust statute.

For decades, American conservatives’ approach to antitrust was dictated by the Chicago school of economics and its attendant law and economics movement. Early Chicago school thinkers understood the need for a vigilant and robust enforcement of antitrust laws. No less an economist than George Stigler, a founding member of the Chicago school, observed in 1952 that “the dissolution of big business is . . . a part of the program necessary to increase the support for a private, competitive enterprise economy.” But the goal of seeking the “dissolution of big business,” and the acute awareness of the threats to individual liberty and human flourishing posed by concentrated corporate power more generally, began to fade in importance as leading mid-to-late twentieth-century Chicago school thinkers like Judges Robert Bork and Richard Posner developed the now-ubiquitous “consumer welfare standard” approach to the judicial interpretation of the antitrust laws.

Bork’s seminal antitrust work, 1978’s The Antitrust Paradox, argued that the corpus of antitrust law ought to be understood and enforced through the prism of efficiency. The way to do this, Bork and Posner argued throughout the Reagan years, was to articulate as the preeminent goal of antitrust law the protection and maximization of “consumer welfare.” In fact, even prior to Reagan’s presidency, The Antitrust Paradox was almost immediately influential. A mere one year after publication, in the 1979 case of Reiter v. Sonotone Corp., the U.S. Supreme Court adopted it tout court: “Congress designed the Sherman Act as a ‘consumer welfare prescription,’” the Court stated.

But while Bork lauded the benefits of a legal enterprise channeled toward the end of economic efficiency, Bork himself was personally no utilitarian fundamentalist. Bovard has persuasively cited The Antitrust Paradox to argue that Bork “unequivocally warned against allowing the ‘economic extravaganza’ [of measurements and quantifications] to distract from the actual [judicial] task of assessing anti-competitive behavior.” Somewhere along the line, however, the consumer welfare standard for antitrust law that Bork essentially created was transmuted in an overtly utilitarian direction, away from his criticisms of excessive reliance on “performance tests and efficiency defenses” in antitrust.

The precise provenance of this doctrinal change is debatable. But regardless, as Khan argued in “Amazon’s Antitrust Paradox,” an antitrust legal injury today “requires showing harm to consumer welfare, generally in the form of price increases and output restrictions.” Or, as Posner himself put it in his law review article “The Chicago School of Antitrust Analysis,” written the same year the Court handed down Reiter, “the proper lens for viewing antitrust problems is price theory.”

In an age of “free” Big Tech services, however—such as searching on Google or adding “friends” on Facebook—a consumer welfare standard of antitrust law rooted in neoclassical price theory is worse than anachronistic—it is utterly feckless. A price-theory-centric consumer welfare standard of antitrust enforcement also dissuades judges from considering factors particularly endemic to the structure of the digital economy, such as the ability of a firm’s vertical integration to exert cross-pressure across distinct product lines due to raw bargaining power.

There is nothing in the text of either the Sherman or Clayton Acts that requires warping and hamstringing the judicial enterprise in this manner. Indeed, if we consider what Sir William Blackstone would have referred to as the statutes’ ratio legis, or “reason of the law” (as I have urged), such a conclusion is dramatically buttressed. After all, as Khan wrote, “Congress enacted antitrust laws to rein in the power of industrial trusts”—not necessarily to maximize an au courant notion of price- and output-determined consumer welfare. That notion would have baffled the Sherman and Clayton Act drafters. The upshot is that whatever benefits a price-theory-centric consumer welfare standard approach may have had in a bygone, non-digitized era, a change of course is now badly needed.

The good news is that many conservatives and elected Republicans are indeed leaving “efficiency”-maximizing utilitarianism in the rearview mirror, and are putting forward concrete ways to improve our extant antitrust regime. Bipartisan lawmakers in the House Judiciary Committee’s antitrust subcommittee, led by Congressmen Ken Buck (R-Colo.) and David Cicilline (D-R.I.), recently unveiled a suite of five innovative antitrust reform bills. Collectively billed as “A Stronger Online Economy: Opportunity, Innovation, and Choice,” the bills range from a ban on “self-preferencing and picking winners and losers online” (a core malfeasance of Amazon) to an overt ban on anti-competitive acquisitions from dominant online platforms to (in what seems like a clear targeting of “Fulfillment-by-Amazon”) a ban on leveraging “control across multiple business lines to self-preference and disadvantage competitors.”

On the Senate side, Senators Mike Lee (R-Utah) and Chuck Grassley (R-Iowa) have introduced a bill that would tighten antitrust enforcement in a handful ways and codify the consumer welfare standard but, crucially, would define that standard more broadly as including not merely price but also “output, quality, innovation, and consumer choice.” Senator Hawley—a biographer of Theodore Roosevelt and leading skeptic of fundamentalist neoliberalism—has similarly introduced the Trust-Busting for the Twenty-First Century Act. That act would go even further, with hard bans on large-cap tech mergers, bans on self-dealing and self-preferencing, and direct stipulation that vertical mergers are not per se exempt from antitrust scrutiny. And, in a nod to Bork’s objections decades ago to “economic extravaganza” myopia, Hawley’s bill would reform “the Sherman and Clayton Acts to make clear that direct evidence of anti-competitive conduct is sufficient to support an antitrust claim, which will allow enforcers to effectively pursue the breakup of dominant firms and prevent antitrust cases from devolving into battles between economists.”

Reasonable minds will disagree on specific policy details, including whether an incremental consumer welfare-standard-tweaking antitrust approach (such as Senator Lee’s) or a more forthrightly neo-Brandeisian/anti-“bigness” approach (such as Senator Hawley’s) is preferable. But the ratio legis of the antitrust laws and the ever-evolving Digital Age’s requirement that common-good-oriented statesmen be adaptable, pragmatic, and prudent suggest that major reforms to the antitrust laws along these lines are now sorely needed.

Finding Fulfillment by Controlling Big Tech

Another possible legal avenue for reining in the Big Tech behemoths comes in the form of the common law doctrine of common carriage. Traditionally, common carrier regulation has applied to “private” services—usually in the realms of communication, transportation, or utilities—that are so “clothed in the public interest” that legislatures and regulators provide the carrot of certain forms of legal immunity (such as tort liability) in mandatory exchange for the stick of having to provide their goods (with limited dispensations) in a nondiscriminatory manner. In effect, when it comes to Big Tech, Congress has actually already provided the carrot: Section 230’s (debatably large) liability shield. But, despite the Big Tech actors’ rapidly growing and impossible-to-ignore list of misdeeds, the concomitant stick has never arrived. That stick could be as straightforward as direct FCC regulation under Title II of the Communications Act of 1934. Title II was most recently in national news during the “net neutrality” debate surrounding common-carrier-style regulation for internet service providers. Alternatively, Congress could legislate—through bills such as Senator Bill Hagerty’s (R-Tenn.) recent 21st Century FREE Speech Act proposal.

The argument for applying common carriage to internet platforms got a big boost in conservative circles earlier this year in the form of Supreme Court Justice Clarence Thomas’s concurrence in Biden v. Knight First Amendment Institute. In a perspective generally skeptical of Silicon Valley’s power, Thomas forthrightly opined, “There is a fair argument that some digital platforms are sufficiently akin to common carriers or places of accommodation to be regulated in this manner.” After all, tech platforms are “at bottom communications platforms, and they ‘carry’ information from one user to another.” The possible harms to the common good of failing to regulate such information-carrying platforms are most acute, Justice Thomas wrote, for those firms exercising “dominant market share”—such as Amazon—which “can impose cataclysmic consequences on authors by . . . blocking a listing.” Given that Congress has already provided the Big Tech platforms with the gratuitous immunization carrot of Section 230 “in derogation of common-law rights and duties,” it’s past time to impose the correlative nondiscrimination “stick.”

To a large extent, however, antitrust enforcement and common carrier regulation are mutually exclusive: it makes very little sense to bust up a tech company, Teddy Roosevelt-style, and then respond to the trustbusting by slapping Title II common carrier regulation on it. While Justice Thomas is right to imply, as he did in Knight First Amendment, that a company’s monopolistic power is not a prerequisite for applying the law of common carriage, it will usually be nonsensical to break up a company for undermining consumer welfare or exercising too much structural market power and then deeming that same stripped-down company to be so utterly essential to the functioning of the macroeconomy—so “clothed in the public interest” or “affected with a public interest,” to use Sir Matthew Hale’s phrase—so as to mandate the strict carrots and sticks of common carriage law.

The proper remedy, or suite of remedies, will also necessarily depend on the nature of the company under consideration. For traditional social media platforms such as Facebook and Twitter, whose very product value for consumers is necessarily dependent on the network effects and positive externalities of “following” others and adding “friends,” statutory nondiscrimination mandates or direct common carrier regulation may be most appropriate. Amending the shoddy 1990s-era draftsmanship of Section 230 is also necessary to curtail ideological censorship.

Unlike Facebook, which relies on a heavy network effect, platforms such as Amazon have become goliaths sprawling across multiple, distinct lines of business through vertical and conglomerate mergers that enable the exertion of anti-competitive cross-leverage pressures. For such entities, straightforward antitrust enforcement is likely the best path. The breakup of Amazon into various constituent parts would involve paying careful attention to ensure that the specific business components most directly abetting anti-competitive cross-leveraging—such as Amazon Web Services and the Fulfillment-by-Amazon logistics and delivery service—are divested and kept separate. That our reigning price-theory-centric consumer welfare approach to antitrust does not support such an outcome is a scathing indictment of that approach.

In the final analysis, Jeff Bezos’s Hydra can be tamed, but it will require legislators and regulators to act diligently. In the interim, an atrophying citizenry yearning to reclaim self-governing democracy from the throes of woke technocracy will wait patiently, hopeful that its government will awake from its slumber, rise to the occasion, and break up Amazon before it is too late.

This article is an American Affairs online exclusive, published August 20, 2021.

Sorry, PDF downloads are available
to subscribers only.


Already subscribed?
Sign In With Your AAJ Account | Sign In with Blink