Cookie Consent

This site uses cookies for functionality. To see our cookie policy click here.

If you continue to use this site we will assume that you are happy with this.

Alliances: What the Competition Authorities are really looking at Jan/Feb 2019 Download PDF

Cloud Image

For some time now the approval of airline joint ventures by competition authorities throughout the world has become almost routine. Such joint ventures allow airlines to co-ordinate (some may prefer  ‘collude in’) their operations on specified routes, including pricing, scheduling and capacity. In most jurisdictions collusion between companies is prima facie illegal, so any airline joint venture is almost invariably closely examined by the appropriate competition authorities.

They will consider whether any restriction on competition is against the overall interests of consumers, and if it is, they will seek ways to minimise such problems. This usually involves ensuring free market access for competitors, for example via an open skies bilateral agreement. If market access is still limited, perhaps because of slot shortages at capacity constrained airports, the authorities will often insist on the applicant carriers making slots available to potential competitors. There may also be other requirements before approval is granted, such as access to frequent flyer programmes or special prorate arrangements, designed to enable new entrant airlines to compete effectively against dominant players in the market.

This approach has not been without its challenges, and some would argue that an alternative policy is needed. In particular, the dominance of the joint venture partners on the routes identified as presenting competition problems has often meant that new entrant carriers have struggled to survive, or have even been put off from entering the market in the first place. In at least one case the European Commission insisted that the applicant airlines actually found a competitor on a key route before approval could be given to their joint venture. Clearly this is far from a perfect solution and there are signs that the Commission is considering an alternative approach, although it is too early to judge whether this will be possible legally or any more successful.

Nevertheless, the process outlined above remains the way in which airline joint ventures obtain approval for what otherwise would be illegal activity. It is a well-trodden path with which airlines and their advisers are familiar. But it was far from always so. In particular, the original application by British Airways and American Airlines for anti-trust immunity for their trans-Atlantic alliance ran into serious problems, despite initial support from the UK and US Governments. Fierce opposition from Virgin Atlantic and others (remember ‘No Way BA/AA!’) and a number of miscalculations by the applicants resulted in demands for slot concessions at Heathrow which BA/AA could not accept.

A later application again ran into problems and again ended with the competition authorities demanding that a large number of Heathrow slots should be surrendered to competitors. It was only in 2009, some 13 years after the original application, and after the signing of the EU/US trans-Atlantic open skies agreement, that approval was finally given at a price acceptable to the applicants. The European Commission found that six routes were of concern from a competition perspective: London to Dallas/Fort Worth, Boston, Miami, Chicago and New York and Madrid to Miami. The US authorities went along with this conclusion.

It is worthwhile listing the slot concessions made by BA and American, as recent press reports have not been accurate. Initially Delta received two pairs of Heathrow daily slots from BA and one pair from AA, which it used for twice daily Boston and daily Miami services from summer 2011. Neither route proved to be as successful as Delta had hoped and the following year it handed back the Miami and one of the Boston pairs. For summer 2013, it took another pair of slots from BA to be used for a service to Atlanta and for summer 2015, it got a pair from AA (under a separate approval for the merger between American and US Air) for a route between Heathrow and Philadelphia.

Interestingly, after three years the Philadelphia slots revert fully to Delta, available to be used on any route, whereas the slots received under the BA/AA approval are time-limited and tied to individual routes. Despite its fierce fight against the BA/AA alliance, Virgin Atlantic initially didn’t apply for any slots. However, it did receive one daily pair from AA for use on the Heathrow-Miami route from summer 2015. This was obtained when BA and AA sought to extend their joint venture to include Iberia and Finnair, with the alliance now called the Atlantic Joint Business Agreement (AJBA).

Norwegian obtained slots for one day per week for use on a Gatwick-Boston service in summer 2016. (Under the remedy settlements, airlines were required to do their best to get slots from the pool, so it is probably a reasonable assumption that Norwegian were able to obtain suitable slots for all or most of the other days in order to mount a competitive service. Note also that these slots were for Gatwick, while the competition problems identified related to Heathrow routes. This appears to be the first time that a city-pair rather than an airport-pair approach was adopted, which could be significant for future alliance anti-trust applications.) For summer 2018, Norwegian separately obtained slots for two services per week between Gatwick and Boston. It seems that the ten-year approval period will run out in 2020, despite more recent amendments to the AJBA joint venture. The free slots will then have to be handed back to BA and AA, unless further remedies are applied by the competition authorities.

The announcement in October 2018 that the UK Competition and Markets Authority (CMA) planned to carry out an investigation into the AJBA generated extensive coverage in the press. There were specific reasons for this, including the mooted take-over of Norwegian by IAG (IAG has now abandoned this project and is disposing of its small shareholding in Norwegian) and the implications of Brexit (more on this below). However, it is equally the case that the CMA review could be regarded as routine, with the previous ten-year approval period coming to an end, and in the absence of significant market or other changes the joint venture could reasonably be expected to be approved, probably with a continuation of the current slot concessions.

Anything significant changed?

The key question, therefore, is whether anything significant has changed since the 2009 decision which might result in a different conclusion by the competition authorities, whether in London, Brussels or Washington. The answer is that there may indeed be cause for concern on the part of the applicants. The chances of a simple rubber stamp certainly seem unlikely.

Let’s first of all consider the three issues which attracted most attention  at the time of the CMA announcement. Firstly, there is no reason to assume that the authorities will be any less concerned about competition on North Atlantic routes. The market is still dominated by the three major airline alliances and airport congestion is no less of a problem. In particular, despite government commitment to a third runway, additional capacity at Heathrow is several years away and far from certain to be built. At the same time, Gatwick (and even Stansted at peak periods) is filling up fast. Admittedly Virgin Atlantic is now closely tied to Delta and other Skyteam members, with its own application for anti-trust immunity recently approved by Brussels, and is unlikely therefore to object to an AJBA application anything like as rigorously as before. But ironically, that could be a double-edged sword.

Secondly, the suggestion that by intervening so early in the process the CMA was sending some form of signal to IAG about a possible take-over of Norwegian seems unlikely. (“The competition regulator appears to have warned the owner of British Airways against attempting a takeover of Norwegian by launching an investigation into the conglomerate’s power in the transatlantic market,” as The Times put it.) This is not the way competition authorities work.

The fact is that IAG did not control Norwegian and its then 4.6% shareholding was very unlikely to have raised any competition concerns. However, this does not mean that were IAG to have made a formal bid, there would not have been a referral to the CMA. Such a combination would clearly have significant implications for competition across the Atlantic and possibly elsewhere as well, with Norwegian already, for example, carrying more passengers between Europe and New York than BA does.

Thirdly, as ever Brexit looms in the background and almost certainly here lies the explanation for the timing of the CMA’s intervention. As the Authority said itself: ”To prepare for the time when the European Commission may no longer have responsibility for competition in the UK, the CMA has decided to review afresh the competitive impact of the agreement in anticipation of the expiry of the [slot divestiture] commitments.” In the absence of a Brexit withdrawal agreement, EU competition law will no longer apply in the UK from as early as the end of March. If there is a deal, the UK will continue to be subject to EU law until the end of 2020, so even then any implementation of new commitments by the applicant carriers would require UK involvement.

The European Commission may decide to carry out its own investigation in addition to the CMA’s given the broader European coverage of the alliance application. Indeed, it would be surprising if it did not do so, although there is likely to be a degree of co-operation between Brussels and London, even in a post-Brexit world. The US authorities are similarly likely to intervene. (Adding to the uncertainty, of course, is the lack of clarity about the Trump Administration’s attitude to competition policy generally, which not surprisingly seems to be following a more ad hoc approach rather than the established policies of its predecessors.) The important point from the perspective of the outcome of any CMA review, however, is that Brexit will not make any real difference. The UK has committed to continue to pursue competition policies effectively identical to the EU’s, at least in the immediate future.

Competition authorities, whether in the UK, EU or US, will inevitably apply the appropriate law. But the officials involved are only human and it should not come as a surprise that they are likely to investigate an application with more rigour if they are under external pressure to do so. (An element of competition between the various bodies does no harm either.) This was illustrated by Sir Richard Branson’s fight over many years to prevent approval of the BA/AA alliance, a success which surprised most observers (including, it has to be said, many in Virgin). With the apparent support of the UK Government and the desire of the US to replace the hated Bermuda 2 bilateral air services agreement with an open skies deal (a requirement for US airline anti-trust immunity), few expected any serious obstacles in the way of approval for the joint venture.

The high-profile campaign launched by Branson and backed by significant resources ensured that approval was delayed for some 13 years, and when it did come it still had a relatively high price (albeit lower than the previous two tentative approvals) in the form of slot divestitures. Despite complaints from consumer groups, there has not been another campaign on a similar scale against an airline joint venture since Virgin’s, at least until now. If such a campaign were to emerge, it could have serious consequences for all trans-Atlantic alliances. It appears that this is now a distinct possibility. The new player threatening to disturb the status quo is US (relatively) low cost carrier Jetblue, which like Virgin Atlantic in earlier years, has resisted any temptation to join one on the three main alliances, preferring to co-operate via code shares with a large number of foreign airlines serving the US.

The JetBlue case

Two factors in particular seem to be influencing Jetblue’s new involvement in competition cases. First, it has announced an interest in operating trans-Atlantic services in its own right, probably primarily from Boston, with new narrow-bodied Airbus 321LR aircraft when they become available. However, it appears to have concluded that in order to do so profitably, access to the principal European gateways is required, which means in particular getting hold of slots at Paris Charles de Gaulle and especially Heathrow. These are expensive, if they can be obtained at all, unless the competition authorities can be persuaded to force the dominant carriers at these airports to hand some over to new entrants for free.

At the same time, JetBlue’s position in the Boston market has come under increased competitive pressure from Delta. JetBlue accounts for 31% of jet aircraft departures from Boston Logan Airport, far more than any other carrier. American has about 18% and Delta just under 20% of flights, but critically Delta’s presence at the airport is growing rapidly. Boston is one of two markets identified by Delta outside of its core hubs for significant additional investment. The city has been described as JetBlue’s star performer, with the highest margin among the airline’s main focus cities.

The impact of Delta’s expansion at Boston is already being felt and the situation is likely to get worse during 2019, with the legacy carrier launching a number of new domestic routes, mostly to cities already served by JetBlue. In addition, it is expanding its international services from the city, both in its own right and in co-operation with joint venture partners. According to CAPA, as JetBlue works towards its immediate goal of 200 daily Boston departures, increasing the number of gates controlled from 24 to 30, Delta and its partners will operate 152 departures in 2019.

JetBlue is certainly well established in the Boston market, having served the airport now for some 15 years, and is well regarded, especially among business travellers with its Mint premium product. On the Atlantic, assuming it maintains its current aircraft configuration, it will offer significantly more leg room in Economy than the legacy carriers and a business product superior to Norwegian’s. But Delta has the advantage of its sheer overall size and international experience and network, important factors with corporate customers in particular.

Boston is not the only city where Delta has proved to be a problem for JetBlue. The low-cost carrier has struggled to establish itself at Atlanta, where Delta is the dominant airline by a wide margin. According to JetBlue’s Chief Executive Robin Hayes, JetBlue has been forced to operate its ten daily flights “over gates spread over two different concourses…. One airline has control or rights to 147 of the airport’s gates. That’s more than 75% — while JetBlue is not able to lease a single gate. Literally not one.”

This is the background to JetBlue’s objection in November to the application to the US Department of Transportation by Delta, Virgin Atlantic and Air France for anti-trust immunity for their trans-Atlantic joint venture. The objection follows a successful intervention in a similar application by Delta and Aeroméxico in 2016. Hayes commented that JetBlue was “delighted” when approval was limited to just five years, with slot divestitures required at Mexico City Airport. JetBlue was a beneficiary of these slots. Speaking at the Wings Club lunch in New York, Hayes remarked that “the Mexico City decision is a great template to follow when airlines seek to link up and it’s something we urge governments around the globe to consider more aggressively.”

JetBlue has argued that the Delta/Virgin/Air France joint venture will “restrict competition on both sides of the Atlantic” and has encouraged the DOT to carry out a comprehensive analysis of the slot allocation implications. It says that combining Delta, Virgin Atlantic and Air France into a “massive single entity, with all of their slots collectively pooled” would “further restrict JetBlue’s ability to meaningfully serve the United Kingdom and European Union markets.” There is a certain irony in these arguments being employed against Delta and especially Virgin Atlantic, both of whom made a very similar case against the BA/AA alliance.

So far JetBlue’s attention has been focused primarily on the activities of Delta and its partners, but it is unlikely that the opportunities created by other anti-trust immunity applications, such as that by the Atlantic Joint Business Agreement, will have escaped its notice. Applications involving a UK airline in particular will raise the prospect of free slots at Heathrow, and perhaps gain other marketing benefits as well. As Hayes has commented: “We have to offer competitive schedules at airports like Heathrow when people will want to fly. We continue to work on that.”

It seems likely, therefore, that airline joint venture applications can expect to be challenged more than they may have been in the recent past. No doubt the proponents and objectors will argue their different cases about the consumer benefits of these alliances, but one conclusion is clear: the lawyers will be busy.

gnuplot Produced by GNUPLOT 5.3 patchlevel 0 0 20 40 60 80 100 120 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Seats (m) Other skyteam star oneworld LHLC Super-connector Other Other SkyTeam SkyTeam Star Alliance Star Alliance oneworld oneworld LHLC Super-connector ATI Joint Ventures 67%

Note: Super-connectors include Emirates, Qatar, Etihad and THY.

gnuplot Produced by GNUPLOT 5.3 patchlevel 0 0 1 2 3 4 5 6 London-New York New York-Paris Los Angeles-London London-Toronto Chicago-London Frankfurt-New York New York-Tel Aviv Paris-Montreal Boston-London London-San Francisco London-Miami Milan-New York Madrid-New York Amsterdam-New York Dublin-New York London-Orlando New York-Rome London-Washington Los Angeles-Paris Dubai-New York Seats (m) oneworld SkyTeam Star Alliance Other gnuplot_plot_5 54% 13% 64% 26% 71% 4% 68% 39% 79% 31% 64% 64% 39% 24% 38% gnuplot_plot_6 31% 77% 25% 13% 32% 58% 32% 28% 21% 38% 16% 85% 16% 61% 42% 17% 100% gnuplot_plot_7 15% 10% 10% 74% 29% 87% 68% 38% 33% 31% 20% 15% 20% 34% 45% oneworld SkyTeam Star Alliance Other
6 5 4 3 2 1 0 Dubai-New York Los Angeles-Paris London-Washington New York-Rome London-Orlando Dublin-New York Amsterdam-New York Madrid-New York Milan-New York London-Miami London-San Francisco Boston-London Paris-Montreal New York-Tel Aviv Frankfurt-New York Chicago-London London-Toronto London-Los Angeles New York-Paris London-New York Seats (m) oneworld SkyTeam Star Alliance Others gnuplot_plot_5 62% 57% 8% 71% 23% 24% 39% 79% 38% 68% 31% 17% 63% 100% gnuplot_plot_6 38% 22% 70% 14% 28% 44% 21% 17% 32% 49% 65% 21% 16% 16% 93% 100% 100% gnuplot_plot_7 9% 29% 58% 76% 33% 29% 45% 55% oneworld SkyTeam Star Alliance Others
gnuplot Produced by GNUPLOT 5.3 patchlevel 0 gnuplot_plot_1 gnuplot_plot_2 JetBlue Delta American United Southwest Others gnuplot_plot_3 31% 20% 18% 9% 7% 15%

Note: Jet aircraft only.


This is premium content, only available to subscribers.
To access Login or contact