The Department of Transportation granted antitrust immunity to Atlantic alliances that reduced competition on the basis of a single paper written by a United Airlines consultant that argued market concentration only has a positive impact on consumers. Part four of four.


In the space of just a few years, the North Atlantic, the world’s biggest aviation market, was converted from robust competition to a permanent oligopoly/cartel of three collusive alliances. By design, the consolidation of the North Atlantic, in turn, forced a wave of mergers that consolidated the domestic US market (the world’s second-largest) and forced most Transpacific and Latin American long-haul airlines to align with one of the three collusive groups.

This radical consolidation undermined the resiliency the industry would have needed to cope with the major downturns it regularly faces and desperately needs as it faces the unprecedented coronavirus crisis.

In order to drive this radical consolidation, the alliances had to reverse 30 years of aviation competition policies, discussed in Part 2 of this series, that saw level-playing field competition as a critical driver of the innovation needed to drive ongoing improvements in industry efficiency and consumer welfare.

They needed to re-establish the regulatory capture that existed prior to the deregulation/liberalization of the 1980s, where the most powerful airlines could negotiate policies about industry structure and competition through private, backroom discussions with government officials without any public discussion or debate or legislative authorization.

Part 3 of this series described the alliance’s orchestrated 2004-08 applications to the US Department of Transportation (DOT) that would expand the share of the market served by the collusive alliances with antitrust immunity (ATI) from just under half of the North Atlantic to virtually 100 percent, and the major PR program to convince politicians and the media that radical consolidation had nothing to do with these government decisions to reduce competition but was the inevitable result of “market forces.”

“In order to drive a radical consolidation, the alliances had to reverse 30 years of aviation competition policies “

Case-Specific Evidence?

Getting DOT officials to privately agree to reverse policies that were popular and had been highly successful was the easy part. US laws established clear requirements before antitrust immunity (which has the same competitive impact as a merger) could be granted.

None of the radical post-2004 industry consolidations could have occurred if the US government obeyed these laws. The consolidation movement’s biggest challenge was developing a basis that the DOT could use to circumvent and nullify these laws.

These laws required a Clayton Act test of whether the reduced competition would increase market power, including evidence demonstrating the absence of risk that it could harm competition by increasing the ability or incentive to raise prices or reduce output in any relevant market and evidence that markets are fully contestable, so that “…entry into the market would be timely, likely, and sufficient either to deter or to counteract a proposed alliance’s potential for harm.”

The law also clearly stated that the apparent absence of serious competitive issues was not sufficient; airline antitrust immunity requests could not be granted unless “required by the public interest” and “necessary to achieve important public benefits.”

The Joint Venture Guidelines and the Horizontal Merger Guidelines defined the evidentiary standards that claims of public benefits must meet: 

“[the applicants] must substantiate efficiency claims so that the Agency can verify by reasonable means the likelihood and magnitude of each asserted efficiency, how and when each would be achieved (and any costs of doing so) how each would enhance the merged firm’s ability and incentive to compete, and why each would be merger-specific. Efficiency claims will not be considered if they are vague or speculative or otherwise cannot be verified by reasonable means.”

The DOT’s dilemma was that it was determined to reduce intercontinental competition to permanent oligopoly/cartel levels, but had absolutely no objective, verifiable evidence that consolidation would not create pricing power or other forms of artificial market power, could not honestly state that intercontinental markets were contestable (there had been no successful new North Atlantic entry since 1983) and had no evidence demonstrating that high levels of concentration would produce things like lower prices and increased service that could legitimately constitute “public benefits.”

Do Consumers Always Win When Industries Consolidate?

DOT’s decisions in all three ATI cases were based entirely on false carrier claims that the reduction of competition would create significant public benefits.  The DOT knew that these claims were vague and speculative, could not be independently verified, and were not backed by case-specific evidence but willfully misrepresented them as objective research that had been independently vetted and was now widely accepted by objective outsiders.

The DOT knew that these claims could not withstand independent scrutiny so it created an ironclad rule establishing that since these claims were immutable truths that would apply to every possible ATI case, these applicants did not have to obey the requirement for case-specific evidence, and DOT could reject all challenges to these claims, and objective evidence others might present contradicting the Alliance carriers’ claims.

The Alliance carriers’ false antitrust claim was that reducing the number of competitors would automatically produce 15-25 percent consumer price reductions regardless of market conditions. Yes, you read that correctly—a claim that consumers always win big when industries consolidate.

“The Alliances claimed that every possible merger in the industry would reduce prices by 15-25 percent, regardless the impact on concentration.”

This isn’t a claim that a specific merger under unique conditions might increase consumer welfare, backed by case-specific evidence that those unique conditions exist. This is a claim that every possible merger in this industry would reduce prices the same 15-25 percent, regardless of the impact on concentration, specific merger synergies, or whether it occurred during boom times or a recession.

This is a claim that is so absolutely certain that every grant of ATI that reduces competition will reduce prices 15-25 percent that the legal requirements for objective, verifiable, case-specific evidence can be ignored.

Taken at face value, the carriers were arguing that even a merger of the three remaining alliances into a monopoly alliance would meet the legal “public benefit” standard, because every possible antitrust immunity grant would reduce prices by the same 15-25 percent.

The American/British Airways ATI application even included a specific estimate of the increased welfare it would create, claiming that prices would immediately fall by $257 per ticket in all connecting markets currently served on an interline basis by the applicants, creating an annual consumer benefit of $92 million.

The entire claim (known as “double marginalization”) is based on a single 1990 journal article by Jan Brueckner, Distinguished Professor of Economics at the University of California, Irvine, who at the time and throughout this period worked as a consultant for United Airlines, although this fact was never disclosed in any of the case submissions or decisions.

Brueckner’s “double marginalization” theory asserts that airlines are physically incapable of setting rational, revenue-maximizing fares on connecting interline itineraries.

If United and Lufthansa had wanted to offer an interline fare between Athens and Seattle (without antitrust immunity), this theory claims that they would completely ignore what other carriers’ fares were in that market, and set a fare that was the sum of independent calculations of Athens-Frankfurt-Chicago and Chicago-Seattle costs, with each carrier separately adding a profit markup on their leg (thus “double marginalization”).

“Double marginalization,” according to this theory, was a “structural negative externality” that forced interline prices 15-25 percent higher than efficient levels universally across the industry. The only ways to eliminate this “structural negative externality” are merger or full immunity to collude on prices. Thus granting ATI automatically and immediately reduces these fares 15-25 percent every time competition is reduced. Not under certain market conditions, but automatically from each and every ATI grant organized along Northwest/KLM lines, just as night automatically follows day.

This “double marginalization” inefficiency was a complete fabrication. Since the first publication of this claim in 2000, no carriers have ever tried to fix the “problem” of interline prices 15-25 percent above efficient levels. Brueckner’s papers on airline pricing present no objective, verifiable evidence that this critical “structural negative externality” ever actually existed.

The airlines applying for antitrust immunity are the world’s experts on how international airline pricing works, but produced no evidence aside from citations of Brueckner’s paper. These applicants knew full and well that no airlines set short-term prices as a function of route costs while disregarding market conditions and competitive prices, and never used “markups” in setting interline prices.

Brueckner’s statistical analysis is also hugely flawed. Brueckner ran regressions of industry pricing data from time periods highly unrepresentative of market conditions when these ATI applications were filed.

As discussed in Part 2 of this series, it is entirely unsurprising that an analysis of the 1990s transatlantic market identified consumer benefits; fares fell 8 percent in the 1990s while capacity grew 54 percent.

None of the variables in Brueckner’s regressions represent his alleged “structural negative externality” and there is nothing in his statistical analysis supporting his assertion that the elimination of this alleged problem was the single, sole cause of these mid-90s pricing gains. 

Many factors contributed, including the initial gains from market liberalization, larger and more efficient hubs, increasingly efficient long-haul aircraft, robust demand growth, much better connection scheduling, and improved supply/demand conditions.

As discussed in Part 2, the actual pricing gains from the original collusive alliances had been exhausted by the end of the decade (when interline travel became nearly extinct) and no one has found evidence of ATI-driven pricing gains after 1999.

Nothing in Brueckner’s analysis supports his critical claim that the impact observed when ATI was first introduced would always occur under different market conditions.

The DOT never demanded these airlines produce case-specific evidence of their alleged “structural negative externality” because they knew that it did not exist. It nonetheless proceeded to convert this academic malfeasance into an ironclad, absolute regulatory rule.

The absolute rule was needed to block anyone from challenging the large numbers that the DOT was accepting as evidence of “public benefits” and to block anyone from using facts or logic to challenge the claim that every ATI application would automatically produce them.

The DOT justified its ironclad rule using the false claim that the ability of immunized alliances to eliminate the “double marginalization” inefficiency has been documented in the “economic literature.” This “literature” is nothing more than follow-up pieces by Brueckner making the exact same points as the original article.

No other published original research has ever documented the existence of “double marginalization.” In the last 20 years, the DOT has never published any research demonstrating whether consumers actually realized any of the benefits promised by these antitrust immunity applicants.

The “Double Marginalization” Claim

As with other industries, governmental oversight of airlines had long depended on having a large set of competitors with diverse, often adversarial interests. As discussed in Part 2, key aspects of Civil Aeronautics Board decision making in the 1960s may have been “captured” by the largest Trunk airlines but every other airline, airport, union, and government entity could file objections within formal rulemaking procedures.

As competition shrank, regulatory capture became simpler and more powerful because the remaining companies could present their shared interests as accepted conventional wisdom. DOT officials responsible for oversight who wanted to implement “pro-business” policies had no incentive to seek out contrary views and stopped exposing its policies to the public scrutiny of rulemaking processes.

For several years, no one raised any questions about the now shared DOT/alliance beliefs about the virtues of radical consolidation, but when challenges emerged in 2009 DOT simply blew them off. In the AA/BA case, DOT explicitly rejected a detailed challenge to “double marginalization,” even though it acknowledged DOJ comments that the link between “double marginalization” benefits and ATI had never been proven, did not dispute any of the observed flaws in the theory and was unwilling to openly defend any of the logic or analysis that the theory is based on, and had no evidence of any actual consumer gains from the previous ATI awards. It nonetheless accepted the AA/BA $92 million annual consumer benefit claim solely on the basis of the Brueckner paper.

At the beginning of the Obama administration, Christine Varney, the new head of the DOJ’s Antitrust Division, filed an attack on the DOT’s preliminary approval of the expanded United/Continental/Lufthansa ATI application, pointing out DOT’s complete failure to meet any of the legal requirements for case evidence.

DOJ comments included: “the Applicants have made no showing that such entry (that could curb any anti-competitive abuse) would be timely, likely, or sufficient”; “The Applicants present no evidence however, that customers will receive quantitatively or qualitatively different service if Continental receives antitrust immunity, compared to what would be provided if Continental merely interacted with the level of cooperation expected of any member of the broader, non-immunized Star Alliance”; “In DOJ’s view, it is not sufficient, however, merely to point towards claimed benefits; rather the Applicants need to demonstrate that immunity is necessary to achieve them. In this regard, the Applicants fall short”; “The Applicants also suggest, without evidentiary support, that consumers benefit from competition between alliances, particularly immunized alliances”; “DOT does not cite the other information it relies upon to analyze the alliance plans, nor does it explain how Continental, or more significantly consumers, would be harmed by the lack of global immunity.”

The DOT’s approval of the United/Continental/Lufthansa application had consisted of nothing more than a verbatim repetition of the applicants’ claims.

But the DOT refused to acknowledge, much less explain, the evidentiary approaches DOJ has criticized, and defended its right to decide cases on the basis of whatever policies it preferred.

The heated inter-agency battle was resolved when Obama’s chief economic advisor, Lawrence H. Summers, came down firmly on the DOT’s side.

Varney resigned from DOJ roughly a year later. If the DOJ’s position had not been beaten back by the DOT and the Obama White House, the expanded Lufthansa group immunity would not have been approved, the even more problematic British Airways-American immunity would not have been approved, the subsequent domestic US consolidation would probably not have happened, and the alliances would not have been able to solidify into a permanent intercontinental oligopoly/cartel.

The DOT simply began asserting that laws now had totally different meanings than they had prior to the turn of the century.

The Airline Deregulation Act’s focus on maximizing long-term consumer welfare was reinterpreted as the basis for DOT policies favoring the largest, most politically influential companies.

“Open Skies” treaties were no longer designed to increase and protect international airline competition, they were now designed to drastically reduce it.

“The Airline Deregulation Act’s focus on maximizing long-term consumer welfare was reinterpreted as the basis for DOT policies favoring the largest, most politically influential companies.”

DOT’s Antitrust Immunity Decisions

Had the DOT conducted the legally required Clayton Act review, they would have found ample evidence that the radical post-2004 consolidation had already increased artificial market power even before the final cases were decided.

From deregulation until 2003, North Atlantic price trends closely tracked domestic price trends. As shown in Exhibit 7, from 2003 onward, a totally new pattern emerged, with North Atlantic fares rising three times faster than domestic fares.

This fundamental shift in pricing behavior exactly tracks the move towards extreme North Atlantic concentration, which started when Air France announced its intention to acquire KLM, previously the largest single driver of price competition in European long-haul network markets.

The near-term pricing power created by consolidation was much worse than this simple Atlantic/Domestic fare comparison suggests. Under normal competitive conditions, airline fares are highly responsive to changes in capacity.

Domestic fares increased 15 percent since 2003 because the industry did not add capacity. When Atlantic capacity spiked in the late 90s, average fares fell, even though this was the peak of the dot-com business boom. But the market power created on the Atlantic in recent years meant normal supply/demand relationships had been subverted.

Atlantic fares increased 46 percent since 2003, even though capacity also increased by 45 percent. If 2008 capacity levels had been operated under pre-2003 competition levels, 2008 Atlantic fares might well have been 30-40 percent lower than observed, suggesting an annual consumer welfare loss due to increased market power of $9-12 billion. Even if one arbitrarily assumes that only half or less of the observed pricing shift is due to market power, consumer welfare losses have been staggeringly larger than the false claim that the AA/BA ATI grant would reduce fares by $92 million a year.

“Atlantic fares increased 46 percent since 2003, even though capacity also increased 45 percent. If 2008 capacity levels had been operated under pre-2003 competition levels, 2008 Atlantic fares might well have been 30-40 percent lower than observed.”

As was always intended (and discussed in Part 2), the North Atlantic ATI decisions immediately drove further rounds of consolidation, including mergers between all of the US airlines who had jointly filed for immunity, the absorption of the few remaining independent European legacy carriers into the three collusive alliances, and a major reduction in competition across the Pacific.

Additional domestic mergers (Southwest-Airtran and Alaska-Virgin) followed once it was clear that Washington had no problem with combinations where the economics would reduce price competition.

Following classic cartel behavior, the three collusive alliances aggressively attacked any carrier unwilling to submit to their dominance. Once North Atlantic consolidation was finalized, the cartel moved aggressively to lock long-haul carriers from Japan, Korea, Australia, and other major Pacific markets into the three collusive alliances.

They mounted a massive political attack on the three independent Middle Eastern hub carriers (Emirates, Etihad, Qatar) even though their network overlap was extremely small, and many other alliance partners were doing the exact same things that constituted unacceptable market behavior when done by independent non-cartel carriers.

Delta invested in developing new international hubs at Seattle and Boston designed to cripple the smaller independent carriers (Alaska, JetBlue) already operating hubs there.

The radical consolidation of the North Atlantic was designed to create a pool of supra-competitive profits that could be used to strengthen market power and distort competition elsewhere. In addition to funding the war on the Middle Eastern carriers and Delta’s attacks on Alaska and JetBlue, it prevented the smaller domestic LCCs such as Spirit and Frontier (and Airtran, before Southwest acquired it) from competing on price in markets where they had much lower costs than the three alliance carriers.

Until ten years ago, the lower-cost non-hub carriers could readily capture large volume point-to-point markets, because the Legacy hub carriers could not sustain the losses. But (prior to coronavirus) those supra-normal international profits funded those domestic losses, preventing resources from shifting from less efficient to more efficient uses within the industry.

The consolidation of the North Atlantic was designed to create a pool of supra-competitive profits that could be used to strengthen market power and distort competition elsewhere

The Counterrevolution

Well after the development of jet technology and modern airline networks, international competition was still controlled by private backroom deals between government officials and the largest incumbent carriers. For roughly thirty years that regulatory regime was superseded by a much more liberal regime that emphasized longer-term improvements in consumer welfare and industry efficiency instead of the short-term financial results of individual airlines.

Laws were established to protect robust competition under level-playing field conditions, so that consumers and investors would determine marketplace winners and losers.

In less than ten years, that more liberal regime has been effectively destroyed, and the ancien régime has been restored. The regulatory capture problem that helped justify deregulation has returned in an even stronger form. Backroom deals between government officials and the largest incumbent carriers have not just distorted market results, but have destroyed the corporate value of smaller competitors.

Government policies are no longer established by legislation based on extensive public debates and detailed analysis of objective evidence about industry economics.

International aviation presents a useful case example of the utter failure of competition policy and antitrust administration. Clearly written laws were ignored by senior officials of both Republican and Democratic administrations who employed falsehoods and serious misrepresentations without consequence.

Procedural structures were reduced to Potemkin Village facades, isolated objections were easily beaten back, and absolutely no one in the legal or economics professions or the business media seemed the least bit troubled.

Decisions were totally devoid of objective data about prices or cost efficiencies or market contestability and were written by people who appeared willfully indifferent to industry economics and competitive dynamics.  

“The international consolidation movement successfully recaptured industry oversight and successfully gutted existing deregulation laws, bankruptcy rules and antitrust requirements designed to protect broader consumer and industry interests .”

The Coronavirus Crisis

The DOT/Alliance claim that consumers and industry efficiency never required more than three competitors in a market was never backed by a shred of evidence. It was always ludicrous to assume that competition limited to United, Delta, and American—three companies that had just emerged from bankruptcy—would remain stable and evenly balanced forever.

Consolidation into a tiny number of Too-Big-To-Fail companies undermined the resiliency the industry needs to cope with the huge coronavirus revenue collapse.

The large number of competitors that were critical to all prior industry restructurings are gone. In the US, $43 billion in desperately needed cash was stripped via extractive stock buybacks and by inflated executive compensation for the managers who believed that that the industry would never face another serious downturn.

“Consolidation into a tiny number of Too-Big-To-Fail companies undermined the resiliency the industry needs to cope with the huge coronavirus revenue collapse.”

Barring the miraculously rapid development of an effective vaccine, no international airline companies are viable going concerns. Bankruptcy-type processes can work when a small percentage of capacity faces liquidity problems but cannot possibly deal with a situation where worldwide demand has totally evaporated.

Airline capacity and employment worldwide will need to shrink far more than anyone had thought possible. This will mean effective nationalization of the industry (including many suppliers), and the establishment of reorganization processes that convince the taxpayers (who will fund it) that the huge costs and sacrifices will be shared equitably.

Those processes will also need to restore the industrial competitiveness of the late 20th century, in order to drive badly needed innovations and efficiencies and to ensure that scarce capital is allocated to where it can produce the most service, employment, and overall economic benefits.

Unfortunately, the international consolidation movement successfully recaptured industry oversight and successfully gutted existing deregulation laws, bankruptcy rules, and antitrust requirements designed to protect broader consumer and industry interests.

The people who now control the industry are the ones who spent the last 15 years undermining competition, maximizing artificial oligopoly market power in order to generate readily extractable cashflow, misallocating capital to less efficient airlines, and ensuring the hegemony of investors pursuing short-term capital appreciation over customers, workers, suppliers, local communities, and every other longer-term stakeholder.

The industry consolidation movement was initiated so that the owners of the largest airlines (in collusion with officials in Washington and Brussels) did not have to admit responsibility for any of the bad decisions that led to the last major (1998-2000) industry crisis.

There is huge danger that the people who now control the industry will use the “no one could have foreseen coronavirus so you can’t blame us for anything” claim to ensure that industry restructuring does not threaten their control, that the industry’s many structural pre-coronavirus problems are ignored, and that the pain and sacrifice of restructuring is disproportionately borne by labor and outsiders.

There is huge danger that the people who now control the industry will double down on reducing competition and maximizing artificial market power, since that has been their single-minded focus for years. But that approach cannot possibly restore a viable, efficient private sector industry.