Do tech workers have a reason to love monopolies?

In the United States, federal antitrust law is composed of several complementary statutes, one of the centerpieces of which is Section 2 of the Sherman Act.1 Section 2 prohibits monopolization (and attempted monopolization). Monopolization does not merely mean to posses a monopoly, rather it means (a) “possession of monopoly power in the relevant market”, and (b) “the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident”.2 At base, this prohibits someone in possession of a monopoly from abusing their monopoly position to prevent others from competing with them.

One of the conventional concerns about monopolies is the risk that they are also monopsonies for labor. A monopsony is a company that is the only buyer of a good or service. A monopsony for a particular form of labor means that the company is the only one who employs a particular type of worker (often in a given geographical region). To give a simple example, a hospital which had a monopoly on offering medical care in a given city would also likely have a monopsony for employing healthcare workers. It’s easy to see that a monopsonist employer raises many of the same concerns as a monopolist seller: in the absence of competition, a monopsonist will likely be able to pay lower wages and offer worse terms to prospective employees than they would be able to in a competitive market. As practical examples, these types of concerns appear in the Department of Justice’s successful challenge to the merger of Penguin Random House and Simon & Schuster, focusing on the impact the merger would have on writers of bestsellers.

The other concern one might have about the impact of a monopoly on labor is that one thing monopoly enables is the ability to reduce output, a critique which has been best articulated by Herb Hovenkamp.3 Monopolies can set prices above the competitive equilibrium while maximizing profits, one of the impacts of which will be that less of their product is produced and consumed. A monopoly which produces less goods will need less workers to produce those goods. This can also ripple, recursively, through supply chains: if one company charges more for (and sells fewer) widgets, then a company which purchases those widgets to sell some finished product will need to raise its prices, resulting in fewer sales, and fewer workers required. For this reason, one school of thought holds that monopolies can be bad for workers, even without labor monopsony.

These two concerns represent two theories of why monopolies can be bad for workers: because they create labor monopsonies or because they reduce output and thereby reduce the need for labor. For either of these reasons, the conventional wisdom is that employees do not want their employers to be monopolists.

But what about in tech?

Last year, the Department of Justice obtained a liable verdict against Google, with the court holding that Google violated Section 2 with respect to the markets for general search services and general text advertising. Later this year, the Federal Trade Commission will proceed to trial, alleging that Meta (Facebook) has violated Section 2 with respect to the market for personal social networking services. The Department of Justice and Federal Trade Commission also have antitrust cases at various earlier stages of litigation against Apple, TicketMaster, Visa, RealPage, Google (a second case, focused on advertising markets), and Amazon. While it is clear that there’s no shortage of tech antitrust cases, none involve monopsony for labor claims.4

Why is that? One thing that is evident is that, whatever else is true, in general all the listed companies compete with each other for tech labor. There’s no doubt that Google, for example, is in the same market for tech labor as Amazon and Facebook, even as they’re in many different markets for the products they offer. This is supported by the (unlawful) “no cold call” agreements between several of them: no one thinks Google, Adobe, Intel, and Pixar are in the same product markets, yet they felt it was profitable to collude to reduce labor competition amongst each other. For this reason, it appears that to a first approximation, there are unlikely to be labor monopsony concerns in tech.

What about output reduction? This story is less clear. On the one hand, if one simply asks the question “how many additional employees would Google need to handle double the searches?” the answer is probably not very many: the number of employees needed to handle increasing levels of traffic scales significantly sub-linearly (this is part of what makes digital markets so profitable). On the other hand, part of how you actually get double the number of searches is by having more employees to build new features and increasing the scope of the product. Your employee count still scales sub-linearly in usage, but doubling the features is will get you many more employees than just doubling the amount of traffic. However, in practice a thing we see tech companies with monopoly positions do is enter entirely new markets (e.g., Amazon’s pivot from book store to cloud computing). And, because tech labor is basically fungible, this form of expansion is more job creating for tech labor than expanding the output of the market in which they have dominance. In short, while there may be narrow reductions in the demand for tech labor on specific products due to output reduction from monopolization, viewed more broadly I don’t think tech workers suffer the sort of negative effects of output reduction that Hovenkamp describes.

This provides the possibility that for tech workers, their employers being monopolists may in fact be a best of both worlds situation.5 Tech workers benefit from their employers' monopolies because those monopolies produce excess profits from which tech workers can be paid substantial salaries, while also reducing the necessary quality of their work.6 At the same time, tech workers face effectively no risk of monopsony, and thus benefit from typical competitive dynamics with respect to their employment, while also not being significantly impacted by output reduction. A market with many buyers, where all of them have lots of cash, is a great market in which to be a seller.

Which means we may have an entire class of employees who have a structural incentive to want their employers to be monopolists.

Which brings us to the question of what it all means? As to ongoing antitrust litigation with these companies, it has no bearing – the absence of a labor monopsony proves nothing about the presence of a monopoly. But I believe understanding this dynamic can be very important for those of us who work in these industries, particularly for people who work for alleged monopolists.

If your employer is a monopolist, and that works to your advantage, you may come to the conclusion that any enforcement of antitrust laws is wrong-headed. But absent that incentive, I believe most people recognize the importance of antitrust laws (with plenty of legitimate debate about their structure and scope): competition is the engine of lowering prices and driving innovation (it can also drive non-price improvements, such as better privacy), it’s the mechanism by which surplus accrues to consumers. But if we’re overly attuned to our parochial interests as tech workers, we run the risk of missing this basic truth, after all, as Upton Sinclair observed, “it is difficult to get a man to understand something, when his salary depends upon his not understanding it”.

Therefore, my request is simple: keep the value of competition in mind. Even if at this moment, non-enforcement of laws that protect competition is working to your advantage. While mere possession of a monopoly is not illegal in the United States, it brings with it “the power to control prices or exclude competition”.7

  1. 15 USC § 2 ↩︎

  2. United States v. Grinnell Corp., 384 U.S. 563 (1966) ↩︎

  3. https://lawreview.uchicago.edu/sites/default/files/2023-02/07_SYMP_HOVENKAMP.pdf ↩︎

  4. Attentive readers may recall a 2010 Department of Justice lawsuit against Adobe, Apple, Google, Intel, Intuit, and Pixar regarding allegations that the companies had “no cold call” agreements amongst themselves. While this was an antitrust case, it was brought under Section 1 of the Sherman Act (15 USC § 1), which prohibits conspiracies in restraint of trade, and does not require the violators to be monopolies or to possess any particular market power. ↩︎

  5. Interestingly, Peter Thiel’s book Zero to One makes the argument that founders should aim to create companies that can be monopolies (i.e., that aren’t simply aiming to take a portion of a well established market). While he gestures at the benefits of a monopoly for employees, the thrust of his book is really that monopolies make good businesses. ↩︎

  6. In the absence of competitors, the quality of a product can degrade below what would be possible in a competitive market, because competition is precisely the mechanism that forces improvement. ↩︎

  7. United States v. Du Pont & Co., 351 U.S. 377 (1956). ↩︎