Antitrust for Airlines
J. Bruce McDonald
November 3, 2005
It is a pleasure to be here with the leaders of regional aviation in the U.S. The service that your airlines provide is increasingly important, most obviously for the 500 airports served only by a regional airline and the twenty percent of all U.S. passengers whom you carry to and from home each year. Those of you who are successful low cost carriers are doing the traveling public a great service, and you are setting an example of efficient competition that is being followed by carriers all over the U.S. and in places as far away, literally and figuratively, as China, where a fledgling regional airline industry is winning over customers and even government regulators.
As a customer of regional airlines since childhood, I'm a big fan. I grew up near College Station, Texas, home of Easterwood Airport, where the only scheduled air service was provided by Davis Airlines. For about $20 half that for students or military flying standby Guy Davis would take you to Dallas Love Field or Houston Intercontinental. Checking in no less than 20 minutes before flight time, Mrs. Davis took your ticket and, in her beautiful Texas voice, announced the flight. "Davis Airlines announcing Flight 101 to Houston. Gate A. Y'all can get on board now." Passengers took their own bags onto the tarmac, where Mr. Davis loaded them into a little Piper Navajo. "Thank you, Mr. Davis." When all the passengers, and that was never too many, were on board, Mr. Davis climbed in behind you, closed the door, and took the pilot's seat. It was one of the biggest thrills of my teenage life to have arrived early enough to be first in line, and to have Mr. Davis send me up to the last seat on the right, the co-pilot's seat. Those were the days.
That's enough from my childhood, but again I'm glad to be here.
Debbie McElroy asked me to speak about airline merger policy and the effects of airline bankruptcies, high fuel prices, and the other problems that unfortunately have become business-as-usual in your industry.
The Department of Justice Antitrust Division is responsible for enforcing the federal antitrust laws, and we have the authority to review mergers to determine if they may lessen competition. In that context, you will not be surprised that DOJ does not have a merger policy that is specific to the airline business, or any other particular industry for that matter. This is not meant to sound contrarian; you should be pleased that the Justice Department does not have an aviation grand plan. As for most U.S. industries, and for aviation since deregulation, the organizing principle is competition. And therefore the market should identify good merger partners, produce winners and losers, and of course decide what airlines fly where and how much they charge. The government antitrust role is to maintain competitive markets by seeking to block anticompetitive mergers in court, and otherwise to stay out of the way.
DOJ does have a view on how to determine whether a merger may lessen competition. Our approach is outlined in our Horizontal Merger Guidelines, available on the DOJ website. And our lawyers and economists have as much experience analyzing airline markets as anyone in government. I hope my explanation of our legal analysis seems sensible when viewed from your business perspective.
The statute that applies to mergers is Clayton Act § 7, which prohibits mergers and acquisitions that may substantially lessen competition in any U.S. market. In reviewing a planned merger, DOJ works to predict whether the combination will create or enhance market power or facilitate the exercise of market power. Market power is the ability to profitably raise prices over the long term, without losing sales such that the price increases become unprofitable. The goal of antitrust enforcement is to protect consumers by maintaining competitive markets, which produce high quality and low prices.
The so-called "relevant market" in which we evaluate whether a particular merger will lessen competition is not the whole industry. Rather, we have to look at the markets in which passengers buy air travel. These markets are the particular origin and destination city pairs (and occasionally airport pairs) on which passengers fly. It is competition in particular city pair markets that is relevant to competition for passengers. If a passenger wants to fly from Washington to Myrtle Beach, he's pleased that Air Wisconsin is competing to keep service up and prices down. And it doesn't help that passenger that American Eagle flies to Buffalo, much less that Horizon is competing from Santa Barbara to Portland. If all carriers flying Washington to Myrtle Beach merge and raise prices, the passenger cannot turn to American Eagle's Buffalo service or Horizon's Santa Barbara-Portland service for a competitive alternative.
The relevant market can actually be more narrow than a city pair. As airlines know well, not all passengers are the same. Some passengers will always fly nonstop, and be willing to pay more for it, while others will accept the inconvenience of stops to get a lower fare. For some business travelers, the availability of connect service may be irrelevant, because they will never accept the inconvenience. For many leisure travelers, either is a possibility. Airlines take this difference into account in their pricing. Antitrust analysis takes this "business" verus "leisure" distinction into account and considers the effect of mergers in nonstop city pair markets and also the effect in nonstop and connect markets together.
Reviewing any particular merger, we first identify the city pairs in which the merging carriers both provide service. It also may be appropriate to consider markets in which the two both are likely to provide service in the future, such that today they are "potential" competitors. Having identified the overlapping city pairs, it is possible to calculate the market shares of the merging carriers and any competitors. We often calculate shares using data that carriers regularly provide to DOT. Having calculated shares, we then will focus the investigation on the relevant markets in which the merger will significantly increase concentration.
Among hub-and-spoke network carriers, it is most likely that competitive problems will be found in a merger between carriers with hubs at the same airport or at airports in the same metropolitan area. These carriers are likely to serve many of the same city pairs; especially in spoke markets, they may be the only two carriers with few or no others providing service. Similarly, carriers with hubs in nearby cities may be the dominant carriers, usually on a connecting basis, for a significant number of city pairs in their region.
Comparing premerger and postmerger market shares and gauging the change in concentration are a good first step. The next step is to consider possible entry. Assume DOJ has determined that the merger is likely to give the merging carriers the ability to raise prices in a city pair market. We would then ask whether the price increase will attract new competition. If the answer is yes, a new entrant is likely to begin serving the city pair and this will deter a price increase, then that resolves DOJ's concerns about that market.
Whether other carriers will think it profitable to enter to replace the lost competition will depend on the situation in that market. Let me suggest some factors that could affect our view on whether entry is likely.
If a new entrant must challenge a hub carrier at one of its hubs, such that the entrant will itself have to build a hub at that city, it may be unwilling to afford this unless it has a lower cost structure than the merging hub carrier. I'll note that some successful LCCs have developed their own hub and spoke models for entry, including Frontier, America West, AirTran and to some extent Southwest.
The time horizon DOJ considers is two years; if entry may occur within two years, we will consider it in the analysis. Therefore, if entry is, as our guidelines put it, "timely, likely and sufficient" to counteract the competitive problem presented by the merger, then we will not challenge the merger.
In discussions like these, at this point folks are eager to ask, "What would DOJ think about a merger of World Air and Behemoth Air?" or "Behemoth Air and Puddlehop?" The answer is, "I don't know," because we haven't analyzed that merger, and again we do not have an aviation grand plan. But to give more life to these remarks, let me discuss some of the airline combinations we have addressed.
The last major domestic airline merger DOJ investigated was in 2000-2001, United's proposed acquisition of US Airways. United-USAir provides several examples of competitive problems that can arise in airline mergers:
Competitive problems like these convinced us that this merger would substantially lessen competition in numerous markets. DOJ announced it would sue to block this transaction, and United and USAir abandoned the deal.
We have a very different example in this year's merger of America West and USAir, in which DOJ announced, after a short investigation, that it would not take any action. America West-USAir was largely an end-to-end combination. America West operates primarily in the western United States, with hubs in Phoenix and Las Vegas. US Airways operates primarily in the eastern United States, with hubs in Philadelphia, Pittsburgh and Charlotte and substantial presences in Washington and New York City. There's a reason some clever investment banker code-named this deal "Project Barbell." There were few nonstop overlaps, and even there other carriers also provided service. DOJ has found that integration of airlines with complementary networks like these can achieve efficiencies that benefit consumers. We believed that the America West-USAir combination, creating the fifth largest domestic carrier, could enable the merged airline to offer U.S. consumers better service to more destinations throughout the country.
The America West-USAir merger is an example of the kinds of mergers that may easily avoid antitrust problem. This may be a good model for an antitrust-ready merger of regional carriers that seek to combine to expand their footprint and use their low cost structure to compete more vigorously against legacy carriers (although each merger must be analyzed on its own facts, and I am not aware of any that has been proposed).
Can federal antitrust enforcement take into account the unique problems faced by today's airline industry? Some have argued that airline consolidation is urgently needed to restore "rational" pricing and that antitrust gets in the way of that. Similarly, in recent years, some have argued that the antitrust laws, drafted in the days of oil trusts and railroad mergers, cannot be suited for dynamic technology industries. My view is that every industry has its eccentricities and antitrust analysis has to be flexible enough to take them into account. In any event, we do have antitrust laws, so we should ask how antitrust analysis is affected by the prevalence of bankrupt carriers and by skyrocketing costs in the industry.
If bankruptcy is a factor in our merger analysis, obviously it is because one of the merging parties is in or near bankruptcy. This is relevant because, if a firm is failing, then on balance it may be better to allow even an anticompetitive merger than to watch the failing firm's assets leave the market and be unavailable to any competitor. In such a situation, we will allow an otherwise anticompetitive merger to go forward without challenge only if the following conditions are met: the failing firm will be unable to pay its bills in the near future, the failing firm could not successfully reorganize in bankruptcy, the firm has tried to sell itself to someone else, in a combination that will not lessen competition as much, and without the acquisition the firm's assets will exit the market.
In less dire circumstances, the poor condition of a firm that is not to the point of failing may be a sign that the firm is not going to be as much of a competitive factors in the future as in the past, and our mergers analysis will take that into account.
These are strict requirements. Some have suggested that the antitrust enforcement should be less strict in an industry in chronically poor financial health.
Speaking from the consumer perspective, bankruptcy in particular is not so bad. It provides a great opportunity for a firm to shed costs, emerging as a leaner competitor, ready again to compete vigorously with a lower cost structure and perhaps lower prices.
Someone taking the industry perspective might point out that some firms never emerge from bankruptcy. In that case, even an anticompetitive merger may have been better, preserving the firm as an ongoing business, than a liquidation, which may not quickly if at all return the assets to productive use. Our merger analysis tries to take this into account; where the evidence suggests a failing firm will not be able successfully to reorganize, we agree it may be better to be acquired by a competitor than to suffer liquidation. A good example of this is American Airlines' acquisition of its competitor TWA, which was bankrupt, out of money, and without another buyer.
Again from the consumer perspective, the airline industry is providing reliable and safe air travel at low prices, luxurious and safe if you want to pay for it. The industry today is very competitive, even if the carriers themselves and their stakeholders lenders, aircraft lessors, labor are less profitable than at times in the past.
What about skyrocketing fuel prices? There is no general rule that suggests that increasing costs of operation always affects merger analysis in a particular way. However, if a merger would help the two firms lower their combined costs of operation, we take that merger "efficiency" into account. If a merger brings efficiencies that will enhance the combined firms' ability to compete, and the benefits from that enhanced competition will outweigh the merger's likely anticompetitive effect, then on balance the merger is procompetitive. You could describe scenarios in which an airline merger lowered total fuel costs, and that could be an efficiency to take into account in evaluating the overall effect of the merger.
You asked me to address the question, Are skyrocketing fuel costs the result of price fixing in the oil business? DOJ is on the watch for possible collusion, especially as prices have climbed so much following the Gulf Coast hurricanes. Collusion is one possible explanation for increased fuel prices, but the more plausible explanations are worldwide crude prices and refining capacity shortages. Price fixing is one of the categories of antitrust violation that are so plainly anticompetitive that DOJ prosecutes the conduct criminally. Both civil fines and jail time are available to punish price fixing. If DOJ were to develop evidence of price fixing by gasoline or jet fuel suppliers, we would pursue that conduct vigorously to bring criminal charges.
You also asked me to discuss the problem of airport congestion. As you know, a handful of U.S. airports have been so congested that slot controls have been put in place. These rights were originally allocated to incumbent carriers based on existing service, and the airlines present at the creation of the slot system now hold shares of the slots that may be disproportionate to the share of traffic they would have absent slot control.
How should the government address the congestion problem and the best allocation of slots? At the extremes, the suggestions are to build more runways or remove all slot controls and let the airlines work it out with their passengers. Less radically, the DOJ has recommended developing a market-based solution. Try, for example, annually reallocating some fraction of the slots, which would be surrendered by the current slot holders and acquired in an auction to the highest bidders. This would help ensure that slots are put into the hands of the carriers that can put them to the most valuable use. If enough slots are reallocated at once, an auction creates the possibility that a regional carrier or a group of them could acquire enough slots to challenge large carriers on a particular route or in a regional network.
Peak period pricing is another system that uses economic incentives to encourage more efficient use of takeoff and landing rights.
The best solution may differ among airports. The DOJ Antitrust Division plans to continue working on this problem with DOT and FAA, which are open to considering market-based solutions.
You in the airline business live in interesting times, in the language of the Chinese curse. In addition to the issues we have discussed today, you are faced with the challenges of more direct competition against the major carriers, increased security requirements, the difficulties of operating a regional fleet (and regional jets) efficiently as costs increase, proposed government funding and legislative changes (like repeal of the Wright Amendment), and the possibility of more foreign investment in U.S. airlines.
I am confident there will always be role for regional airlines, low cost entrepreneurs without which many of us would be stuck on the ground.
Antitrust For Airlines
Thursday, November 3, 2005
Updated June 25, 2015