Let foreign airlines fly domestic routes

Removing barriers to entry and increasing competition is the best fix

Reminder: if you’re in the DC area today, join us for happy hour at 6pm! Details here.

---

If you want to fly direct from DC to Bangor, Maine these days, your only option is on American Airlines, even at the peak of the summer travel demand season. Relative to the recent past when United also ran a direct flight, that’s a somewhat unsatisfactory situation.

Two factors differentiate the “you can fly any airline you choose as long as it’s United” kind of situation from a true monopoly. One is that you can always take a connecting flight if you want to. But the other, which is probably more important, is that the right of entry exists. If American put the fare up too high, then United could re-enter the market. Or Southwest could launch a flight from BWI. Or Allegiant and Breeze, two small airlines that currently fly routes from Bangor to various Florida airports, could decide to serve DC. The shadow competition provided by the threat of entry is not as good for consumers as an actual second option. But it is still a meaningful form of market discipline.

At the same time, an interesting fact about the aviation industry is that airlines like RyanAir, EasyJet, and Whizz Air that serve lots of European leisure travelers could not enter this market. Only an American airline can fly between American cities.

Singapore Airlines will take you to LA, San Francisco, or Seattle on the west coast and to New York on the east. But if they wanted to enter the market in transcontinental service, that would be illegal.

This is more or less the situation all over the world, as long as you treat “Europe” as a country rather than a couple dozen independent states. Which is to say that if you want to fly from Milan to Palermo, you can go on ITA (an Italian company) or on RyanAir or EasyJet (not Italian companies), but Spirit cannot enter this market.

The principle of airline nationalism is not on my Top Ten list of problems. But it does pop up on my radar every so often, because people sometimes invoke a lack of competition in the American aviation industry as a big problem without mentioning the most straightforward way to solve it.

Here’s an article from 2016, for example, talking about the Roosevelt Institute’s Felicia Wong and other figures with rising clout on the left, in which Wong and Elizabeth Warren say it’s a huge problem that four airlines control 80 percent of the American market:

Wong thinks it’s no longer accurate to even think of these issues in terms of left versus right. Instead, she holds, real political realignment means a long-term cultural change in the perception of government and its relationship to consolidated power. Wong has been resolute in refusing to draw a bright line, as some progressives would, to rule out bankers, in part because banks are only one element in the pattern. If most people have a hard time understanding or worrying about the concept of “financialization,” they have a much easier time recognizing — as Elizabeth Warren put it in a speech at the New America Foundation last month — that four airlines control 80 percent of American airline seats, three chains own 99 percent of drugstores and four companies sell 85 percent of the beef.

As far as I know, neither Warren nor Roosevelt has ever proposed that we allow foreign airlines to compete on domestic routes. But if I were going to make lack of competition in American aviation one of my headline public policy complaints, I would push for repealing the laws that limit competition in American aviation.

A brief history of airline competition

Back in the days before Jimmy Carter, federal policy toward passenger aviation proceeded on a totally different conceptual basis.

Airlines were seen as a utility-type business that needed utility-type regulation. The core aspect of utility regulation is that we do not expect market competition to work for something like the provision of home electricity. We’re not going to have two or three separate overlapping electrical utility grids serving the same community with customers choosing between them. And you’re definitely not going to have the easy threat of entry into new markets. If there are two different utilities serving Chicago and its suburbs and electricity prices are getting high, nobody is going to come in and built a whole third electrical grid to try to compete.

What you need instead is either for the government to own the electrical grid (as is the case in some places) or else a utility commission to regulate the prices. Large swathes of the country have “deregulated” electricity markets, where residents rely on market competition for the provision of electrons to the grid. But nowhere has multiple competing electrical grids. Whatever the problems with government regulation, the amount of waste and subsidy that would be required to maintain competing grids is way too big for anyone to try it.

Aviation used to be like that, with fare-setting and route-selection governed by the Civil Aeronautics Board. At least, the CAB controlled interstate flying. A handful of American states, notably Texas and California, were large enough to sustain competitive unregulated markets in intrastate air travel. Policymakers in the 1970s noted that intrastate flights in those states had lower fares than comparable flights between northeastern cities. They dismantled the CAB, and unleashed an era of vigorous competition between airlines. An idiosyncratic group of people spent the period 10-15 years ago arguing that airline deregulation was a bust. Before Lina Khan ever went to law school, she co-authored a 2012 article with Philip Longman arguing that introducing competition into the airline market was a failure. The Institute for Local Self-Reliance also took this view, and you would sometimes hear it from labor-aligned groups, since a more competitive aviation market was, in fact, straightforwardly bad for airline unions.

The evidence behind those arguments has always struck me as incredibly flimsy — Derek Thompson summed up the conventional pro-competition view in 2013 and I think it’s utterly convincing. Jeremy Horpedahl also did a more recent version of the conventional take that’s a little more technical and that I think is good.

But it is true that airline competition has declined since its peak. Delta and Northwest merged in 2008, United and Continental merged in 2010, and American Airlines and US Air merged in 2013. Within a five-year span, we went from six “mainline” airlines to just three. Antitrust authorities nitpicked these mergers, requiring small concessions to let them go through, but did not fundamentally object to the consolidation trend in the industry. With less competition, airlines became more stable and more profitable. And smaller startup players like Spirit, Frontier, and JetBlue began to expand somewhat, proving that competition is still alive. But then over the past few years, Spirit looked at merging with Frontier only to decide it actually wanted to merge with JetBlue instead. The Department of Transportation stepped in and said no this time — enough is enough, airlines need to compete.

Foreign airlines could help

In all of these mergers, the merging airlines argue that consolidation would actually be good for competition. They’d tell you, for example, that a combined JetBlue/Spirit, precisely because it’s larger, will have the network effects and economies of scale that it needs to take on the Big Four.

The counterargument, of course, is that you don’t get more competition by having fewer competitors.

These are difficult and somewhat technical matters to judge. I think DOT probably did the right thing by blocking the Spirit deal and probably did the wrong thing by allowing the US/American merger. But I’m also not really sure; you could convince me otherwise with fresh info or analysis.

What seems indisputable, though, is that if we want more competition, we should repeal the rule that prevents foreign airlines from competing with domestic ones. Sometimes people don’t want competition because they think some other thing is more important. With the Jones Act, for example, we have decided that propping up America’s pathetic civilian shipbuilding industry is more important than competition on freight shipping. The bar on foreign airlines running domestic routes is less stringent than Jones-style rules. Airlines can (and do) use planes built in Europe or Brazil or Canada, along with planes built in the United States. But still, we choose to protect the domestic airlines rather than foster competition.

Consider that if a Gulf airline like Emirates were to buy Jet Blue, they would obtain many of the putative upsides of an airline merger. They would suddenly have access to a much larger international network, which would make them a more appealing choice for domestic flights for a certain class of customer, and in turn would strengthen the logic of investing in domestic route expansion. The Gulf airlines themselves, meanwhile, have a lot of success and a lot of capital, but limited opportunities for expansion since there are only so many routes to Doha or the UAE that it makes sense to fly.

Similarly, Spirit could tie up with one of the European low-cost carriers — not so much for the flight network, but for the back-end economies of scale. There’d be no loss of competition here. Spirit and Allegiant don’t compete with RyanAir or EasyJet on any routes because it’s illegal. There are also a couple of airlines dabbling in a low-cost transatlantic model, and there’s the interesting case of Iceland, which flies to both Europe and North America from an island in between. Those are again cases where mergers would strengthen the network (the way merger proponents say) without diminishing competition on any routes (the way critics fear).

A competitive market, if you want one

How large of an effect are we talking about? A paper by Xinlong Tan, Clifford Winston, and Jia Yan says that allowing foreign competition would increase US consumer welfare by about $1.6 billion per year. They characterize the impact as “modest,” and I have to say, their findings somewhat diminished my enthusiasm for this as a cause.

But note an important point. The reason they say the benefits are modest is that “airline competition is already intense.”

In other words, if the American aviation market does not have a competition problem, then eliminating anti-competitive rules only has small benefits. Others see it differently. A report by economists with National Economic Research Associates¹ commissioned by IATA found major competition issues in the global aviation industry. The NERA view is that airline employment would meaningfully expand if we allowed cross-border ownership and foreign-flagged domestic flights. In other words, they think the current status quo doesn’t just limit consumer choice somewhat but that airline oligopolies are meaningfully under-producing airline services relative to a market with fewer barriers to entry.

Of course, that report came out years ago, and the situation is now somewhat different.

It’s hard to offer new flights without pilots available to fly, and there aren’t a lot of unemployed people with a license to fly large airplanes. Of course, you can train new ones. In Europe, pilots need 500 hours of flight time to qualify for a license, while in the United States pilots need 1,500 hours, which makes it much more difficult and expensive to become a pilot with no clear safety benefits. There is also a proposal in Congress to raise the mandatory retirement age for pilots from 65 to 67, which could help a bit. In both cases, pilots and their lobbyists have argued against reform precisely because they don’t want more competition — just like the existing Big Four airlines don’t want competition from foreign brands.

Who wants airline competition?

These reforms could work separately or in tandem. Making it easier to get qualified pilots would, on its own, increase the level of competition between America’s existing large and small airlines by making it easier to expand service. Allowing foreign competitors to enter the market would also increase the level of competition by leveling the playing field and expanding the amount of investment capital available. Doing both simultaneously would have a cumulative impact.

Going back to Longman and Khan’s critique of airline deregulation, one of their main points is that certain markets have lost service in the deregulation era.

As an argument against deregulation, I don’t think this really holds up. They are mostly showing that smaller markets in regions of the country suffering from relative population decline have suffered a reduction in service. There are plenty of other communities — Charlotte, Austin, Phoenix, Las Vegas — that have seen dramatic expansions of airline service. I don’t think it’s a huge mystery why Charlotte (a metro area that has grown six-fold since 1970) has seen improved airline service while things have gotten worse in Cleveland (which has fewer residents than it did 50 years ago). I think a virtue of a market-oriented system rather than a utility-style one is precisely that it allows the airline industry’s service patterns to shift with shifting demand patterns, rather than everything becoming a highly politicized struggle in which service to growing cities like Seattle and Austin is strangled because policymakers are focused on boosting Pittsburgh because it’s pivotal to the electoral college.

Khan has obviously moved on to bigger and better things since her article with Longman.

But the basic question remains: If we see a market with barriers to entry and a lack of competition, do we want to change policy to make the industry more competitive, or do we want to point to the lack of competition as a reason that we need utility-style regulation? Sometimes utility-style regulation is, in fact, the answer. But I think there is a tendency in some quarters to reach too quickly for that solution rather than “remove the barriers to entry.”

In the airline case, an important counterweight is labor unions. Even as unionization rates in the United States have plummeted, aviation remains a highly unionized industry. For various structural reasons, it would almost certainly continue to be a highly unionized industry if opened up to competition. Reducing barriers of entry would mean more pilots, more flight attendants, more baggage handlers, and more union members. But from the standpoint of the existing union members, it’s still bad. As a labor union, your dream scenario is to represent workers in an industry with high barriers to entry and limited competition. Your company extracts monopoly rents from consumers, and then workers leverage the power of the union to ensure that the rents are shared with the staff. In a highly competitive industry, margins are small, there’s less for the union to win, and whole companies end up failing and people need to go find new jobs elsewhere.

The view that competition is bad and we should prioritize stability and rent-sharing instead is one that people rarely articulate explicitly. And yet, when you look at questions like tariffs or the Jones Act or foreign airlines, you see clearly that this is motivating policy. I’m on the side of competition and lower barriers to entry. But at a minimum, I think we would have a clearer and more enlightening discourse if people would join this argument more squarely. If the airline industry suffers from a lack of competition, then obviously we should block anticompetitive mergers. But we should also eliminate rules whose whole purpose is to block competition.

1. Full disclosure: My grandfather founded this company, two of my uncles have been affiliated with them at different times, and I worked there one summer. Alfred Kahn the “father of airline deregulation” was a longtime NERA guy. So while I have zero financial stake in any of this, “NERA consultants fix aviation policy problems” was an important part of family lore when I was growing up.