By now the nature of the economy’s market power problem is clear: decades of lax antitrust policy have permitted a concentration of economic power in the hands of dominant corporations, their executives, and their boards of directors not seen since the Gilded Age. This has disadvantaged the economy’s many other stakeholders: consumers, workers, entrepreneurs, communities, and everyone who benefits from economic growth as a whole. The question now is what to do about it.
It’s instructive to look at the airline industry as a case study in policy gone awry. It isn’t just your imagination that flying has become increasingly miserable over the past decade: a wave of mergers has left us with four major airlines who are free to offer us take-it-or-leave it terms—extra fees, smaller seats, fewer routes—while reaping consistent record profits. The bloody face of a passenger being dragged off a United flight last year is only the most memorable image of how unpleasant air travel has become.
How did we get here? The federal antitrust statutes, passed more than a century ago, are supposed to let the government keep big businesses from abusing their power. But, since the 1970s, Supreme Court decisions influenced by conservative economic theories have greatly diluted those laws, leaving the government with very few effective enforcement tools. As I explain in a new research paper for the Roosevelt Institute, the one lever left to antitrust enforcers to guard against consolidation that squeezes out competition—merger review—has not been up to the task. Under existing judicial precedents, there’s no way to rescue the airline sector or the economy as a whole from the structural decrepitude into which they have sunk. Instead, it’s time for new comprehensive federal legislation to revive enforcement against dominant firms in airlines and elsewhere, to undo decades of consolidation, integration, predation, and exploitation and reconstruct an economy that serves everyone’s interests.
Antitrust in an era of deregulation
Commercial air travel used to be highly regulated by the federal Civil Aeronautics Board (CAB), which controlled how routes were allocated, what prices could be charged, and which areas had to be serviced. At the time, the sector was dominated by legacy incumbent carriers at the regional and national levels. Airlines were required to provide affordable service on unprofitable routes, in exchange for retaining their scarce, valuable slots on high-traffic routes where they could earn a profit.
That regulatory apparatus was dismantled during the Carter administration through legislation sponsored by Ted Kennedy and strongly influenced by his staffer, then law professor Stephen Breyer. Deregulation meant that instead of applying to the CAB for permission to charge a certain price for a given route, airlines were free to charge what they thought the market would bear. While the scarcity of landing slots at airports meant there would never be entirely free entry in the sector, deregulation allowed airlines to compete on routes that had previously been divvied up among a few chosen incumbents. The premise of deregulation, following a decade of rising fares in response to oil shocks and government-guaranteed profit margins, was that competition and the “free market” were better at allocating scarce resources than the CAB. Inefficiencies would be washed out of the system and consumers would benefit.
Deregulation did at first seem to deliver on some of its promised benefits. Airfares plummeted on the routes that remained in service (though this was in part due to the combination of increased fuel economy and the decline in oil prices following the end of OPEC’s effective global monopoly). Airlines like Southwest, which began life operating solely within Texas, and was hence never subject to the CAB regime, exploded outward to put pressure on comfortable legacy carriers. In many ways, air travel went from being an elite to a mass phenomenon. But this came with downsides that many deregulation advocates have underplayed: reduced status and job security for airline workers, a greater capacity for incumbents to price-discriminate among passengers. In order to economize after their profits ceased to be guaranteed, legacy carriers designated certain airports as hubs (think of Delta in Atlanta or United in Newark), and most trips would involve passengers passing through those hubs. With fewer direct flights, the quality of the product on offer was reduced. The hub system also effectively segmented the national market into geographically-subdivided monopolies or oligopolies—certain airports became dominated by one or two airlines, and non-hub airports would be serviced by at most a few routes linking them to each airline’s nearest hub.
The division of the national market into regional monopolies had an even more dire impact: Airlines simply stopped service on routes too sparsely traveled to remain profitable, which had the effect of cutting off whole swathes of the country from regular commercial air traffic. Nowadays, even cities as large as Saint Louis and Cincinnati suffer from a lack of service. When Amazon released its request for proposals for cities to house its “HQ2,” it specified that the chosen location would have to have direct air links to its existing headquarters in Seattle, San Francisco, New York, and Washington, D.C. Fairly few cities fit the bill.
Another consequence of deregulation was that running an airline became a far riskier business proposition. Until the last decade, the sector operated in a perpetual boom-and-bust cycle. Price wars and recessions made revenue extremely volatile, while the fixed cost of maintaining a basic level of coverage remained high. For example, airlines need to insure a minimum number of flights reach a given hub in order to fill their flights out of that hub. Each of those incoming flights may be individually unprofitable, but the profitability of the whole system depends on servicing them.
These dynamics are inherent in unregulated, infrastructure-dependent networked industries like telecoms and railroads, and so, as in the Gilded Age, the solution to protect incumbents from hard times was to merge. After the last round of airline bankruptcies in the mid-2000s, the antitrust authorities permitted the remaining firms to consolidate to increase profits. Where there were nine major airlines in 2005 (remember Northwest?), today there are four.
In any industry, consolidation gives a small number of firms market power: the ability to charge customers more (or treat them worse), and pay employees and suppliers less, without fear of being undercut by the competition—because there is no competition. Too much market power is bad for the economy for the same reason that it’s attractive to companies: it allows them to make higher profits without raising quality or efficiency.
Antitrust enforcement gets neutralized
In the late 1990s, the government tried to prevent airlines from abusing their market power. The Department of Justice under Bill Clinton filed suit against American Airlines for predatory pricing at the Dallas-Fort Worth airport. Predatory pricing is when a monopolist attempts to exclude competitors by aggressively reducing prices whenever a new entrant tries to get into the market, eating the temporary losses until the competition is crushed. This was the first major predatory pricing case brought by the DOJ Antitrust Division after the Supreme Court decision in Brooke Group v. Brown and Williamson, in 1993. Before Brooke Group, to prove predatory pricing, the government or any plaintiff just needed to show that a company was charging prices below cost. But following the decision, the DOJ would have to not only show that American Airlines priced its tickets below cost—a relatively simple thing to prove— but also prove that it would be able to “recoup” its losses through later profits. In 2001, a federal judge ruled that the government had failed to meet that burden in its suit against American. The result was to effectively take away an important tool the government could have used to prevent airlines from carving up the market.
Now that they don’t have to worry much about customers going with the competition, airlines have exploited their market power to “unbundle” their fares, forcing customers to pay extra for checked bags, boarding priority, space in overhead bins, rebooking, and so on. Unbundling would be fine if the savings were passed on to consumers in the form of lower fares. But in practice, it appears to only slightly reduce baseline fares, while forcing passengers to pay for services they previously got for free.
Frequent flyer programs are another mechanism for price discrimination, and big data will only make this more extreme: the carriers will know when you have a can’t-miss family event versus surfing fares for a one-off vacation opportunity, and price accordingly. They’ll also know whether you personally are likely to get enticing offers from their competitors—which has a lot to do with your family background, race, and income. All of these strategies were once subject to much closer antitrust scrutiny, but today unilateral conduct—meaning behavior by a single dominant firm—is more or less immune from challenge. Even when airlines seem to be colluding, it’s hard for plaintiffs to prevail. In 2017, a private class-action lawsuit alleging that Delta and AirTran had conspired to introduce bag fees on routes from their mutual Atlanta hub was thrown out of court because the plaintiffs couldn’t prove that the carriers had actually acted in concert. It was possible, the court said, that this was mere “conscious parallelism”— doing the same anticompetitive action (introducing bag fees) without having explicitly planned to do so together. The federal agencies haven’t even tried a similar suit. The perversity of this situation is clear: under existing law, as long as all the airlines worsen their service, none of them can be accused of violating the antitrust laws.
“Economics” in merger review
Decades of narrow Supreme Court precedents have made it all but impossible for the government to police conduct like the coordinated introduction of bag fees, or to reduce geographic concentration by challenging the predatory pricing that preserves it. The only policy lever left is “merger review”. And when deciding to bring (or, more often, not to bring) challenges, the decisive consideration tends to be whether the government can show that airline consolidation is likely to increase fares specifically on routes that are served by fewer carriers as a result of the merger.
In each of the three major airline mergers that have closed since the Great Recession—Delta–Northwest in 2008, United–Continental in 2011, and American Airlines–US Airways in 2014—the Justice Department was presumably unable to show that fares on merger-affected routes would increase relative to less affected routes, and so they declined to challenge any of them.
That doesn’t mean antitrust enforcers weren’t worried about anti-competitive consequences of the mergers. In fact, the DOJ did file suit to block the American–US Airways merger, but it ultimately settled the case in exchange for minor concessions. A 2016 article in ProPublica reported that DOJ professional staff were overruled by political appointees in the matter, and since then, the DOJ has opened an investigation into collusion by all the major airlines—an outcome it predicted would take place in its initial complaint seeking to block the 2014 merger.
That complaint listed many harms to competition that would result from allowing the merger to proceed. But what ultimately ends up deciding the matter at trial is the extremely narrow question of whether increases in concentration on individual routes will cause the fares paid by passengers to increase. If the DOJ can’t make that case, then there’s little they can do to block a merger, and since blocking mergers is the only tool they have available to structure markets, there you have the essential inadequacy of our current antitrust policy.
In a new Roosevelt Institute paper, I show where such an approach can go wrong. I conducted retrospective studies of the three mergers and concluded that direct effects on fares are hard to establish. Concentration on individual routes usually changes relatively little in response to a single merger. The subset of routes in which concentration increases most dominate the estimate of the overall effect of a given merger, and for those few routes, merger effects are mixed. Our analysis shows that fares remained neutral following the Delta-Northwest merger, increased following the Continental-United merger, and declined after American-US Airways.
Does that mean airline consolidation hasn’t been bad for competition? In short, no. The problem is that this narrow approach to evaluating mergers—comparison of routes that are heavily affected by a given merger with those that are not—is necessarily going to have a problem detecting the effect of declining competition overall, whether on fares or any other outcome. But overall loss of competition is exactly what has happened in airlines—so-called “coordinated effects,” as opposed to the behavior of only the parties to a given merger.
In order to investigate the overall effect of consolidation in airlines over the past ten years, as opposed to isolating the effect of individual mergers, I looked at two episodes in which oil prices declined precipitously: one in 2008, the other starting in late 2014. Oil is one of the most important inputs to airlines, and probably their most variable cost. The extent of competition in the sector can therefore be estimated by examining how responsive passenger fares are to fluctuations in that cost: the less fares reflect changes in oil prices, the less competitive pressure the airlines face.
My analysis shows that between the two episodes, the “pass-through” of oil price declines to passenger fares diminished most on the routes that had become more concentrated. This doesn’t pin the blame on any one merger, as merger-by-merger review would have to do, but—along with revelations about airline collusion, the proliferation of unbundling and price discrimination, and general customer dissatisfaction—suggests there’s a big problem with competition in airlines, and that merger-by-merger review of the effect of route-level concentration on fares is insufficient to solve it, especially given all the consolidation that has taken place to date. If we want to solve our airlines problem—not to mention the economy’s overall market power problem—it will require a much more systematic approach, and given the state of the jurisprudence, the only available lever is likely to be legislative.
At a minimum, Congress must address the question of monopoly (and monopsony) power that has already been accumulated, through past mergers and over-permissive conduct—the exact area of law where policy has weakened the most since airline deregulation. It should do that by prohibiting firms with market power from abusing that power through anti-competitive, exclusionary, or exploitative business models. For firms that rely on those business models, it would mandate structural remedies to remove the incentives to engage in them. And it would streamline the legal wrangling necessary to establish a defendant’s liability for anti-competitive schemes by eliminating their ability to claim that such conduct improves competition.
The airline industry is just one example of how our everyday lives have been affected by four decades of weakened competition policy. Unless we have the tools to rectify past policy errors, there’s no hope of unraveling the dire situation in which the economy now finds itself.
