On July 9, 2021, the Biden administration issued an executive order (EO) to promote competition in the American economy. The order claims that competition has weakened across U.S. industries due to business consolidation and government inaction, enabling large companies to leverage monopoly power over workers, small businesses, and consumers. To address this alleged increase in market power, the EO proposes a host of policy changes.1

The administration’s diagnosis is flawed. The EO fails to ask why market power exists, which is critical to understanding whether and how government should act. Further, the EO conflates concentration with market power and takes a static view of competition, leading it to overemphasize any short-term costs of market power relative to long-term benefits from increased innovation incentivized by short-run monopoly profits.2 And, while acknowledging some government-created barriers to entry, the EO glosses over the fact that many others, such as tariffs, zoning laws, and regulations, are crucial impediments to competition in the U.S. economy. Thus, despite a few sensible proposals, the order would reduce economic efficiency and weaken competition in some areas.

In this brief, we first outline key theoretical points about market power and government policy. We then review specific proposals in the Biden EO. Last, we discuss ways the federal government could improve competition and efficiency by scaling back or eliminating existing policies and programs.

Where Does Market Power Come From? A Brief Theoretical Overview

Four different mechanisms can generate product-market concentration and potentially market power. In considering these mechanisms, we distinguish concentration from market power. Concentration means that a substantial fraction of output in a market comes from one or a few firms. Market power means that prices persistently and substantially exceed marginal costs of production. Concentration is presumptively necessary for market power, but the former can occur without the latter.

Defining concentration can be difficult because it is not always clear what the relevant market should be. One problem is to define which products the market encompasses—for example, are electric vehicles a separate market from internal combustion ones? Another problem is geographical range. Retail stores in one state do not compete with stores from another state, so looking at national concentration figures may be misleading.

Concentration occurs in some industries because innovative or productive companies capture a large share of the market. Evidence suggests this partly accounts for many consolidation trends, including historical cases, such as the Great Atlantic & Pacific Tea Company, Sears, and Kodak, and more recent examples, such as Amazon, Google, and Apple.3

Yet the concentration that results from successful innovation is rarely cause for concern. Over time, the profits earned by such firms spur imitation and entry (including from abroad), which reduces concentration and excess profits. In the 1990s, for example, Nintendo accounted for 90 percent of the U.S. video game industry. Vying for a share of the industry, Sega launched its Sonic the Hedgehog games in the wake of Super Mario hits and, at one point, attained a 55 percent share of the industry.4

Many firms that have acquired a substantial share of the market have done so, in part, because their innovative products or production practices allow them to sell at low prices. Relying on economies of scale, better inventory management, and strategic partnerships with suppliers, Walmart has driven prices down in the retail sector since the 1990s.5 Its prices may have been as much as 17–39 percent lower than the competition, triggering mergers between existing chains and price cutbacks among competitors.6 Retail-sector concentration, in this case, was a response to competitive pressure, with Walmart increasing its market share while competitors had to fuse to withstand competition.

A second mechanism that generates market power is production technologies that exhibit declining costs per unit as the scale of production increases (economies of scale). A related possibility is that each consumer prefers using the same product as (most) others, as with PC operating systems or social media platforms. Such network effects can also foster concentration; together, users might prefer a different service, but switching individually makes no sense. In either case, the resulting firm might be what economists call a natural monopoly: this involves technology that makes it natural (cost effective) to have one or a small number of firms.

Thus, economies of scale or network externalities can generate natural monopolies, which could use their market power to set prices above marginal costs. But using government to limit this market power is tricky. Breaking up natural monopolies implies higher costs and lower consumer welfare because the scale economies and network externalities will be smaller. Requiring natural monopolies to charge prices equal to marginal costs makes it impossible for them to recoup their investments.

Instead, allowing natural monopolies to earn monopoly profits can incentivize innovation and competition, which erode the monopoly profits. From 1950 to 1965, computer services were largely provided by room-sized systems owned by large firms and universities.7 This meant the efficient scale of provision for computer services was likely in the hundreds of users. Innovations, such as integrated circuits (microchips) and microprocessors, enabled a transition to smaller devices, eventually resulting in PCs in the 1980s.8 Economies of scale in the mainframe era were replaced through innovation by more-efficient small-scale systems.

The third possible source of market power is anti-competitive practices by private firms. To be sure, private businesses seek to maximize profits, and their actions may have anti-competitive motivations. But the question for competition policy is whether the actions increase economic efficiency overall, taking into account effects on consumers and on the costs of production.

For example, mergers are a useful tool to harness shifting economies of scale or scope. A review of several cases shows that mergers take advantage of a higher scale of operation. Furthermore, mergers and acquisitions can enhance productivity through various methods.9 Even mergers and acquisitions targeted at quelling competition can incentivize firm entry—entrepreneurs may open new firms to be bought by major incumbents or to provide products no longer provided by the acquired firms.10 And for mergers and acquisitions to generate market power, companies must eliminate the threat of competition along with the actual competitors; if a bought-off competitor can simply reenter the market, the market-power benefits from mergers and acquisitions will be short lived.11 Low-cost airlines are a canonical example: if airline profits grow excessively, new firms can lease aircraft and enter the market, bringing incumbent profits down.

In principle, government could screen all merger and acquisition proposals, denying the anti-competitive ones. This is easier said than done. In practice, competitive and anti-competitive mergers are hard to distinguish ex ante, and this may invite unwelcome political meddling. Even in the current regime, research suggests that antitrust authorities are more likely to approve mergers for firms connected to politicians responsible for congressional oversight of regulators.12

An anti-competitive practice that likely reduces consumer welfare is collusion to raise prices and restrict output.13 But maintaining a cartel requires keeping out new entrants, avoiding defections, and agreeing on each participant’s market share. The economics literature recognizes, at best, a handful of successful cartels.14 Even the long-lasting De Beers stranglehold on the global diamond supply has ceased in the 21st century amid challenges from Russian, Canadian, and Australian producers and from synthetic diamonds.15 The company’s diamond market share plunged from more than 80 percent of the global diamond supply in the 1980s to 29 percent in 2019.16

A different example of an anti-competitive practice is predatory pricing, where a dominant firm reduces prices to expel competitors and then raises prices afterward when in a dominant position. Price wars are hard to win, however, and hard to distinguish from healthy competition. Competitors can temporarily exit the market and reenter once the “predator” raises prices. Moreover, the predator firm must increase production to meet expanded demand even as marginal costs increase, adding to losses. Consumers benefit during the periods of low prices, which partially offsets any losses later.

Clear examples of successful predation are rare.17 Even Standard Oil’s predation is not unambiguous.18 Standard Oil’s market share had been declining before the antitrust charges, and prices did not fall as expected after the government broke up the company.19 Perhaps the clearest examples of predation come from government: public schools charge no direct fee and can withstand losses indefinitely, which raises the bar for private school competition. Even so, they fail to achieve full market domination.20

The final mechanism that generates ongoing market power is government-created barriers to entry. Tariffs, quotas, and “Buy American” laws protect domestic industries from foreign competition. Much regulation, regardless of whether it is sensible on broader cost-benefit grounds, handicaps small firms relative to large ones because compliance costs scale less than one for one with size, inhibiting entry and competition. Government sometimes directly creates and operates a monopoly—think of first-class mail by the U.S. Postal Service or state-backed oil company monopolies such as Venezuela’s PDVSA (Petroleum of Venezuela) or Saudi Arabia’s Aramco. Other government policies, such as occupational licensing, land-use restrictions, public schools, regulation that limits entry (taxis, Uber, and harbor pilot appointments), and advertising regulations, directly restrict competition too.21

To be clear, not all policies that limit competition are necessarily undesirable. A major source of monopoly profits is government protection of intellectual property via patents, copyright, and trademarks. These create monopolies for a period so that innovators can reap the benefits of investments in innovative or creative activity. Carefully balancing the beneficial incentives to innovate with the harmful effects of monopoly power is hard, and some research suggests that the current level and structure of government intellectual property enforcement may be excessive.22 Regardless, the intellectual property example shows that market power is not necessarily bad and that a significant fraction results from government.

Market Power in Biden’s Executive Order

Biden’s executive order (EO) repeatedly points to rising product and labor market concentration in different sectors—including transportation, technology, agriculture, and health care—as a symptom of lack of competition and market power. Besides a few cases where government policies are linked to restricted competition, the EO mostly implies that such concentration stems from and is exacerbated by anti-competitive practices by monopolistic firms. To curtail such practices, the EO takes a heavy-handed regulatory approach to either ban or restrict them.

We dispute not only some of the underlying facts in the EO’s diagnosis but also the alleged anti-competitive nature of many practices. As discussed above, concentration may stem from sources other than anti-competitive behavior, and many of the targeted practices can be economically efficient.

Economic efficiency is concerned with maximizing total economic surplus, a measure of welfare that reflects the difference between the willingness to pay and the willingness to accept of the parties to a transaction. This means our analysis focuses not specifically on consumer welfare but on aggregate welfare, whether it comes from consumers, workers, or shareholders. Distributional impacts are a legitimate concern but are best dealt with through distributional policies directly.

Evaluating Policy Diagnosis and Prescriptions

To begin, the EO’s claims that concentration and market power have been rising everywhere are misleading. While market power has increased in national product markets, evidence suggests it has decreased at the local level.23 Within industries with rising national concentration, concentration has been driven by top firms expanding into new local markets.24 Concentration increases are correlated with productivity and output growth but uncorrelated with changes to prices or overall payroll expenditures.25 Moreover, these trends are not exclusive to the United States, suggesting lax U.S. antitrust enforcement is not to blame.26 Overall, rising national concentration seems to result from top firms productively reaching more markets.27 These firms expand output and lower prices at the relevant product market level for customers, increasing consumer welfare.28

The White House document also mentions evidence that markups (the difference between prices and costs) have increased.29 But the same source notes that higher markups are restricted to sectors where fixed costs have increased, even if these do not fully offset the increased markups. If we take these measurements at face value—and these things are hard to measure—they could be pointing to increased economies of scale, where higher fixed costs are recouped through higher markups.30 EO proposals to restrict or review past mergers in the hospital, technology, and finance sectors might interfere with such economies of scale, leading to higher consumer costs. Measures to curb mergers and acquisitions can also limit the benefits of network externalities.

The specific measures proposed in the EO are mostly misguided, which is not surprising given the flawed diagnosis behind them. The EO contains more than 70 proposals, so we illustrate several of them with key examples.

Aiming at tech companies, the EO encourages the Federal Trade Commission (FTC) to restrict the accumulation of personal data. This is ironic, since the FTC and state attorneys general recently sued Google for preventing the accumulation of such personal data (in the form of third-party cookies) in the Chrome web browser.31 In any case, this measure may hinder innovation because many companies rely on customer targeting for advertisers to make profits (and keep their services free). Restricting accumulation of personal data may also benefit bigger providers that can more easily get around restrictions due to a large user base—again, something critics emphasized about Google’s move to block third-party cookies.32

The EO calls for restoring net neutrality, which requires infrastructure providers to price all data packets traveling through internet pathways identically. But different packets come with different needs—streaming video without interruptions is more important than delivering emails without delay. Thus, outlawing differential treatment can generate inefficient queuing.33 U.S. internet service has mostly functioned well without such rules, outperforming neutrality-regulated European Union providers when handling a pandemic-related surge in traffic.34 Moreover, evidence suggests net neutrality has had no impact on mobile app innovation in countries that implemented it, contra the hopes of neutrality proponents.35

Hoping to protect workers, the EO proposes to ban or limit noncompete agreements, which state that employees must not compete with former employers for a period after their employment contract ends. They also prohibit workers from revealing sensitive business information to other parties.

Even if such agreements restrict worker choices, however, both firms and workers might want to enter them voluntarily. A new worker often requires training to acquire productive skills. Workers also have access to company procedures, methods, and information not safeguarded by intellectual property protections. Besides, nondisclosure agreements can be hard to monitor when a worker is hired by a competing firm and so have modest impact in any case.

If firms cannot use noncompetes, they may hesitate to share productivity-enhancing information or provide adequate training to employees who might quit or demand raises afterward. Evidence indeed suggests noncompete agreements foster risky research and development investment.36 And if noncompetes are valuable to firms but costly to employees, workers will demand higher compensation for signing noncompetes.37

A third group of the EO’s proposed measures aims at consumer protection rather than competition per se. These measures include micromanagement of contract clauses, fee refunds, and labeling standards. In the domestic airline market, ancillary fee hikes are taken by the EO as a symptom of feeble competition. But charging additional baggage fees could be an efficient way to price discriminate users that do not require luggage storage space, reducing costs to other passengers.38 And demanding that airlines refund fees on delayed baggage may not be worth the costs to customers (this translates into higher costs for airlines to be passed along to consumers).39

The EO calls for limiting early termination fees in internet service contracts, increasing risk for tenant-specific investments and leading to other fees. A proposal to restrict landlord-provider exclusivity contracts can shorten the duration of contracts, pushing prices up.

Another subset of consumer protection measures encompasses price transparency.40 The EO seems to imply that market power, at least partially, operates on corporate informational leverage that can be mitigated through government intervention. But it is unclear why market power would translate into uncertainty about prices or product quality. If consumers are risk averse, monopolists could increase profits by providing users with insurance—that is, clear information about products or services offered—while charging a higher price for a fixed quality. Moreover, customer informational needs for products are usually heterogeneous and changing, whereas regulation tends to be homogenous and static. And when companies are unwilling or unable to provide valuable information for customers, third parties have the incentive to do so—take for example Consumer Reports or price-comparison websites such as Google Flights and Trivago. Regulation risks hindering third-party innovation to provide customers with bespoke information.41

A More Effective Approach to Curbing Market Power

The EO proposes some good measures to curb government intervention to increase competition, such as restricting government-mandated occupational licensing and allowing for over-the-counter purchase of hearing aids.42 Additionally, it proposes to restrict drug patents to allow for generic and biosimilar drugs, which can benefit consumers, though the proposal’s effects on innovation should be properly assessed.43

But the EO’s approach mostly relies on government solutions to alleged market-power problems. As we have seen, this is misguided; market-power practices are often hard to distinguish from healthy competition, and markets have internal mechanisms to curb market power in the long run (tearing down a monopoly is profitable). The EO’s overreliance on government interventions means it will likely tackle issues that do not reflect market power, while the upside of correctly intervening where true monopolies are at play is smaller than the EO assumes because of market-correcting mechanisms.

A better approach would be for government to remove more legal barriers to entry and thereby bolster competition.

One obvious example is eliminating tariffs and other nontariff barriers to trade, which would enhance competition rapidly.44 Recent examples of U.S. protectionism include tariffs on solar panels, washing machines, steel, and aluminum.45 Other examples include tariffs and nontariff barriers on sugar, oranges, dairy, peanuts, meat, and fish.46

Many other government programs and policies restrict competition: the U.S. Postal Service’s monopoly on first-class mail, the Jones Act’s restrictions on domestic shipping, taxi and Uber quotas that limit market entry, zoning policies barring businesses in certain locations, childcare regulations, car dealership laws restricting direct sales by manufacturers, and restrictions on telemedicine.47

Government subsidies may also be responsible for artificial market concentration. Besides being undesirable for other reasons, the federal government could decrease artificial concentration by stepping aside and abolishing highly concentrated multibillion-dollar agricultural subsidies.48

Conclusion

Biden’s EO relies on a flawed diagnosis of the state of competition in the U.S. economy. Consolidation is not necessarily evidence of market power, and government inaction is not evidence of decreasing competition. While the administration has proposed a few sensible measures that would likely enhance competition and economic efficiency, most others would add to the distortions caused by government intervention. Instead, the best approach to improving economic efficiency is the elimination of existing government interventions that directly or indirectly limit competition without generating a significant offsetting benefit.