By attacking power imbalances, competition policy can steer income to workers and independent merchants who are more inclined to spend than monopoly platforms.
When the economy is contracting, as it is doing sharply now, the tendency of competition authorities, including the trustbusters in FDR’s early days, is to take the foot off the pedal of antitrust enforcement. The purported logic is that antitrust, like any regulation, serves as a constraint on business activity, and thus acts as a drag on economic activity when demand is weak. But what if this logic is backwards?
Consider the following experiment: Under the current antitrust regime, which places coordination rights in the hands of corporations and punishes atomistic suppliers from coordinating in their dealings against dominant platforms, Uber captures about a third of the revenues from each ride in the form of commissions. Suppose a policy intervention—whether a reform to antitrust, labor, or some other law—altered the workplace power imbalance, such that drivers captured an additional 10 percentage points of the fares, reducing Uber’s commissions from 33 to 23 percent. New research on the stimulus checks from the 2020 CARES Act shows lower-income households spent a greater share of the checks than did middle-income households. To the extent that lower-income drivers would spend a greater portion of every incremental dollar than do Uber’s higher-income shareholders, this redistribution of fares towards drivers would mean a greater multiplier effect for the economy every time a passenger takes a ride.
This thought experiment could explain the anemic growth of the US economy, even before the pandemic struck. The labor share of the US economy—the part of the national income allocated to wages—declined by about ten percentage points from the late 1950s (around 66 percent) to the 2010s (around 56 percent), with the biggest drop off occurring in the aughts. Several economists have traced the cause of a declining labor share back to lax antitrust enforcement and the hollowing out of unions; in Europe, those factors have not moved against workers as much and the labor share has held steady. Firms that acquire monopolies in product markets, thanks to lax enforcement of merger law, can wield that same power in their role as buyers of labor. As income shifts away from workers and towards capital, each dollar of revenue generated at a firm is attached to a lower multiplier effect. Hence, slow growth.
Antitrust, at least in its original conception, is not just a tool that disperses economic power from owners to labor, but also from monopoly behemoths to smaller firms. Rather than having one tech giant controlling social media and the associated advertising, imagine we had ten firms instead: Assuming the same revenues, it is not a stretch to conclude industry-wide spending on R&D and labor under the decentralized configuration would be larger.
Indeed, the colossal tech giants have been faulted for having a low propensity to hire workers and spend money on other firms, relative to their valuations. As described in Thomas Philippon’s The Great Reversal, the market value-to-employment ratio of superstar firms began to rise sharply in the 1990s, revealing a declining labor footprint; today’s superstars also contribute less to productivity growth than their counterparts in previous decades. The analog to shifting income from low-income workers to high-income shareholders is that firms enjoying a dominant or monopoly position have no need to invest to stay on the technological cutting edge, as documented by Germán Gutiérrez and Philippon, and instead paying money out to creditors who then sit on it, or to the politicians and think tanks that exist to keep the monopoly intact.
This isn’t to say that antitrust should be the primary tool to stimulate the economy back to health. We have short-run macroeconomic stabilization tools for that, and subsidized loans for ailing businesses alongside generous unemployment benefits for laid-off workers are crucial components of the macroeconomic toolkit. But what antitrust can do is re-engineer the economy by redistributing income downward to smaller, less powerful economic agents who are more inclined to spend, so that we don’t keep having these severe recessions and glacially slow recoveries.
There is a good policy basis to act before an industry tips towards monopoly, as it has in e-commerce, internet search, and social media. We call this “pre-distribution antitrust,” as it suggests a role for competition policy before a market has become monopolized, at which point the winner has greater incentives to win the game against other economic stakeholders. In particular, government agencies, which are major conduits for recovery-related outlays, should look more consciously for opportunities to spend money to promote entry in concentrated sectors of the economy, or in data-driven industries where incumbents enjoy a massive head-start advantage.
For example, Tesla appears to be running away with the electric vehicle segment. In addition to its nationwide network of supercharging stations and its significantly longer battery life compared to entrants, Tesla enjoys an incumbency advantage due to its superior database that relates tweaks to battery design to driving range and overheating. To overcome these entry barriers, Congress should appropriate funding to build a national network of supercharging stations so that Tesla’s nascent rivals can compete more effectively in the short run. Once built by the government, the network could be operated by private entities such as gas stations or convenience stores, so long as they are not affiliated with any electric vehicle maker. Congress should also encourage interoperable charging standards, data sharing, and battery design among electric vehicle entrants, even if that means some coordination in their dealings over these narrow issues. Intervening after a data-driven industry has tipped towards monopoly has proven to be more difficult.
To ensure that society gets the best possible benefits from public recovery expenditures, antitrust agencies should be vigilant in preventing any types of conspiracies among funding recipients against consumers. If we permit recipients of state aid to combine or coordinate and deprive consumers of fruits of competition, then the aid will have been wasted. It was disturbing to witness American Airlines and Jet Blue form a marketing alliance in the wake of the first stimulus. Based on new research by professors José Azar, Martin Schmalz, and Isabel Tecu, concentration in the airline industry is already high, and when you account for common ownership via a modified HHI (MHHI), concentration is off the charts and growing. Agencies should be able to claw back funding as soon as a recipient acts in an anticompetitive way.
Although an across-the-board merger prohibition was considered extreme, tapping the breaks on acquisitions by dominant platforms makes sense; for example, by making any such acquisition presumptively anticompetitive and shifting the burden of proof onto the dominant platform, as proposed by the House Subcommittee on Antitrust last week. Professor Herbert Hovenkamp has a clever idea of only permitting a dominant platform to acquire nonexclusive rights to the acquired technology. The pandemic makes already vulnerable acquisition targets even more vulnerable. When combined with the threat that dominant platforms hang over targets, including appropriation of the target’s business, we could get payments to targets well below fair market value.
Another potential role for antitrust agencies is to serve as advocates for competitive outcomes in front of legislatures and courts. Although workable in theory, a practical concern is that agencies could become captured by entrenched incumbents that oppose any coordination rights for smaller and less powerful actors. We have experienced this firsthand in the Trump Antitrust Division, with Assistant Attorney General Makan Delrahim weighing in via amicus briefs against atomistic suppliers—against drivers in support of Uber and against writers in support of talent agencies—and on behalf of former clients (Qualcomm). One solution could entail winding down the amicus program entirely.
There is a temptation in Washington to call for studies when action is needed. We likely don’t need a study on how the pandemic is affecting the competitive environment in the US—Covid simply accelerated the movement to online in general, and to Amazon in particular. Rather than a study, we need to think about how to keep local retailers alive, and how to prevent Amazon from dominating and controlling them. Empty storefronts and barren malls are the hallmarks of a failed town.
A New Competition Agenda
The next heads of the Federal Trade Commission and of the DOJ Antitrust Division should outline a competition policy agenda to promote a recovery that encourages more competition. There is a tendency to pull back from antitrust enforcement when producers are suffering. But that would wrongheaded. Why should employers, for example, be permitted to engage in a market-allocation scheme for workers simply because the economy is contracting?
To stop these abuses, agency heads should redirect the focus of antitrust enforcement towards worker welfare. Policing non-competes and no-poach agreements and protecting worker mobility should be a priority. Agencies should also focus on adverse worker effects from mergers. DOJ’s open support of the merger between T-Mobile and Sprint, a four-to-three wireless merger that decimated employment in the sector, clearly shows that worker welfare played no role in the analysis.
The Chicago School revolution in antitrust has made worker welfare subservient to consumer welfare, but workers are the ones who spend the money that keeps local economies humming. Thus, worker welfare should be central to any recovery program. And just like low-income households, independent retailers would be more inclined to spend and thereby stimulate the economy than would a monopoly platform. A new competitive landscape, in which workers and smaller merchants achieved more power, would bring about a faster recovery.