REMARKS TO THE ABA SECTION OF ANTITRUST LAW
TRANSPORTATION INDUSTRY COMMITTEE
J. Bruce McDonald
March 31, 2004
Good morning. It is a pleasure to be here with the Transportation Industry Committee. I was especially glad to be invited to speak by Trey Nicoud, who was the client in the first airline antitrust litigation I worked on years ago.
Trey asked me to address the issues that the Antitrust Division sees as important in the transportation industries and to talk about practicing before the Division. Most of you know that our transportation matters are handled by the Transportation, Energy & Agriculture section. I'll start by recollecting the transportation matters that the Division has had over the last year and making some observations on what was important in each. Then I would be happy to take questions.
II. AIRLINE INDUSTRY MATTERS
Most of the large, and interesting, transportation matters are in commercial air travel. But until this past year, we had not seen a successful airline merger in some time. As you will recall, in July 2001 we, along with attorneys general of twelve states, announced our intent to challenge the merger of United Airlines and US Airways, the second- and sixth-largest U.S. carriers.(1) We concluded that United's proposal to divest assets at Reagan National Airport and United's agreement with American Airlines that promised to fly five routes on a nonstop basis were inadequate to replace the competitive pressure that US Airways brought to the marketplace, and would have substituted regulation for competition on key routes. After our announcement, the parties abandoned their merger plans.
In September 2003, Air France and KLM announced that they would merge their ownership and begin coordinating operations, although they would remain nominally separate carriers. This combination of two of Europe's largest four carriers creates the world's largest airline, as measured by revenue.
The standard analysis of an airline merger defines the relevant geographic markets as each of the city pairs that the merging carriers serve. That was our approach in Air France/KLM. You should also know that our analysis did not view the merger as the combination of two carriers that compete on a variety of city pairs, but instead as the combination of two codesharing alliances.
There are four major codesharing alliances that have received limited grants of antitrust immunity from DOT. The immunity generally allows the carriers to coordinate their scheduling and pricing on city pairs that they do not both serve independently, subject to some city-pair carve outs of nonstop overlap markets. The SkyTeam alliance is Air France, Delta, Alitalia, and others. The so-called Wings alliance is KLM and Northwest, including an unimmunized codesharing with Continental. Star is United, Lufthansa, and others, and oneworld is American, British Airways, and others.
Because Air France and KLM are members in competing alliances, the postmerger Air France/KLM would have the incentive and ability to coordinate the activities of the SkyTeam and Wings carriers. Therefore, we analyzed this as a four-to-three worldwide merger of alliances, examining the likely effects of their overlap on hundreds of city pairs. Of course, there are other carriers, not part of any of these four alliances, that also compete on some of these city pairs.
There was only one overlap nonstop city pair touching the U.S., New YorkAmsterdam, so almost all the overlaps were in connecting markets. Air France and KLM compete on many more city pairs that do not affect the United States but do affect Europe. The European Commission and the two carriers reached agreement on a demanding set of commitments by the carriers, covering both intra-Europe and beyond Europe markets. The commitments included a surrender of 47 landing and takeoff slot pairs (forever), a "frequency freeze" (agreement to refrain from increasing frequency on the affected routes to give new entrants a "fair chance" to establish themselves), and assurances by the Dutch and French governments that they would give traffic rights to other carriers wishing to stop over in Amsterdam or Paris.(2) Both the European Commission and the Division began investigating the proposed transaction relatively early, and the parties cooperated by providing data and other information up front. In part because of the EC remedies, the Division was able to close its investigation.
I'll diverge for a moment. Generally speaking, in most large mergers it is a fruitful strategy to begin the work of providing information to the antitrust agency before making the premerger filing that puts the agency on a clock. If you can provide the agency with the information it needs to complete its analysis before making an HSR filing, you may make a Second Request unnecessary.
The proposed arrangement among United Airlines, Mesa Air, and Atlantic Coast Airlines highlighted some of the interesting issues that can arise in acquisitions outside the ordinary course.
Atlantic is a commuter airline. It flies mostly as United Express and has substantial operations at Dulles Airport. Atlantic technically has a codeshare with United, and United controls its scheduling, pricing, and reservations. In July 2003, Atlantic announced that, upon the termination of its contract with the bankrupt United, Atlantic would start a low fare carrier, which Atlantic later announced would be Independence Air. With hub operations at Dulles, Independence would be a direct competitor to United and other carriers in Dulles city pairs.
In October, Mesa Air Group made an unsolicited offer to acquire Atlantic. Mesa announced that, if it acquired Atlantic, it would scrap the Independence Air plan. Mesa and United announced their agreement that Mesa would have the Atlantic planes continue to serve as United Express. From the Division's perspective, this arrangement would have had the same competitive effect as United acquiring its new competitor.
Atlantic did not agree to Mesa's proposed takeover, and Mesa announced that it intended to take steps to replace Atlantic's board of directors with directors favorable to the acquisition. Atlantic sought protection from the federal district court in Washington, D.C. Atlantic claimed that Mesa's takeover attempt violated the securities laws and also alleged that Mesa had conspired with United to prevent Atlantic from launching a low-cost competitor, in violation of Section 1. (Atlantic also alleged that the takeover attempt violated Section 7, but the district court determined Atlantic did not have standing.)
The Division opened an investigation and issued CIDs on December 18, which Atlantic made public within hours. The same day, the district court entered an injunction, forestalling Mesa's consent solicitation and stock exchange offer. In quick succession thereafter, United announced it was terminating its agreement with Mesa, and Mesa ended its effort to take over Atlantic.
I offer two puzzles to show you what keeps the Antitrust Division up at night. First, Mesa's proxy solicitation, if successful, could have allowed Mesa to take over the Atlantic board and scuttle Atlantic's low-cost carrier plan even before Mesa had to make an HSR filing. This suggested that the Division should move quickly, but only if we were reasonably confident that Atlantic actually was starting a low-cost carrier, rather than just using the threat of starting a low-cost carrier to gain leverage in contract negotiations with United. If the latter, then the Division would be putting its thumb on the Atlantic side of the dispute when there could have been no real anticompetitive effect from the United-Mesa arrangement. Obviously, the Division decided that an investigation was appropriate, and today it does appear that Atlantic will continue its plans to launch Independence Air, and United has begun offering low-cost services with Ted.
The second puzzle: suppose that Atlantic had agreed to the Mesa takeover, which would have resulted in reversing Atlantic's planned entry as a new competitor. Might that agreement have violated Section 1? Does that depend on how advanced or likely were Atlantic's plans at the time?
As most of you know, Orbitz is a travel website owned by five major domestic airlines. The Division opened its investigation in March 2000, and the Orbitz service became operational in June 2001. All of the airline owners, and approximately 40 domestic and foreign airlines that are "charter associates," are bound by a "most favored nation" (MFN) clause that was the focus of our investigation. The MFN requires the owners and charter associates to provide Orbitz with any publicly available fares that the carriers list on their own websites or on other online travel sites. The Division analyzed whether the MFN clause was likely to facilitate coordination among the participating airlines or reduce their incentives to discount, resulting in higher fares, or make Orbitz dominant in online air travel distribution.
The Division found that the MFN did not result in higher fares or create such dominance. Since the initiation of the Orbitz service, average airfares have decreased. Additionally, based on information available at the time we closed the investigation, Orbitz had remained the third largest online travel agency for over a year, and, according to a December 2002 Department of Transportation report, had 24 percent of online ticket sales. There are more details regarding our decision to close the investigation which I do not have time to share here. I encourage you to read the statement if you have not done so.(3)
Our concern regarding possible coordination among airlines is not new. In December 1992, the Department sued eight of the largest U.S. airlines and the Airline Tariff Publishing Company (ATP) for price fixing and operating ATP, their jointly-owned fare exchange system, in a way that facilitated collusion. The Division alleged two causes of action. First, the defendant airlines engaged in various combinations and conspiracies with other airline defendants, which included agreements, understandings, and concerted actions to fix prices by increasing fares, eliminating discounted fares, and setting fare restrictions for tickets purchased for domestic air travel. The defendants used ATP's computerized fare exchange system to (1) exchange proposals and negotiate fare changes, (2) trade fare changes in certain markets in exchange for fare changes in other markets, and (3) exchange mutual assurances concerning the level, scope, and timing of fare changes.
Second, the airline defendants and ATP conspired and reached an agreement to create, maintain, operate, and participate in the ATP fare exchange system. The defendants designed and operated the system in a way that unnecessarily facilitated coordinated interaction among them so that they could (1) communicate more effectively with one another about future fare increases, restrictions, and elimination of discounted fares, (2) establish links between proposed fare changes in one or more city-pair markets and proposed changes in other city-pair markets, (3) monitor each other's changes, including changes in fares not available for sale, and (4) reduce uncertainty about each other's pricing intentions.(4)
United Airlines and US Air settled immediately,(5) and the others followed suit in March 1994.(6) The ten year consent decree required the defendants to institute a compliance program and restricted the airlines' ability to communicate proposed changes to fares through ATP. Throughout the period of the decree, the Department has monitored the airlines' pricing conduct for potential decree violations and violations of the Sherman Act.
We were admittedly disappointed with the Tenth Circuit's July 2003 decision in the American Airlines predation case upholding the district court's grant of summary judgment to the defendant, while we appreciate that it did provide some clarity regarding Section 2 standards.(7)
We do not take issue with the court's opinion on several important points of legal policy, however. To the contrary, we agree with the following: predation claims should be approached with caution, but not incredulity; Brooke Group's(8) standards were appropriate for evaluating the predation claim in this case; and, because predation claims could chill aggressive competition, the evidentiary standard in such cases should be high. We part ways with the court, however, on the question of whether the Division's evidence satisfied this difficult evidentiary burden.
We were also pleased that the court accepted our argument that the market-wide average variable cost test insisted upon by the district court can 'obscure a predatory scheme' and so is not always the appropriate measure of cost. The court accepted, at least in principle, our argument that in this case an appropriate test was whether American's cost of adding capacity on a route exceeded the revenues that the additional capacity generated. We were also pleased that the court did not adopt a "meeting competition" defense to Sherman Act claims.
III. COMPETITION ADVOCACY
Let me also discuss a topic that does not get a lot of attention: competition advocacy. On occasion the Division advises its sister agencies on competition questions, both informally and in public comments. Whenever appropriate, the Division will take the opportunity to advocate that sister agencies clear away regulatory and other obstacles to allow competition to operate in the markets that these agencies regulate or oversee. In the last year we have filed comments on transportation issues with the Department of Transportation, the Federal Maritime Commission, and the Surface Transportation Board.
The Department of Transportation regulated the operations of travel agent computer reservations systems (CRSs) for twenty years. When the rules were promulgated, the systems had market power over airlines because they controlled access to an essential distribution outlet, and that power was subject to abuse because the systems were owned by individual airlines. The rules were scheduled for sunset in January 2004.
In early 2003, the Department of Transportation issued a proposed rulemaking, which would have extended and expanded its CRS regulations. In June, the Division submitted extensive comments, generally recommending that most CRS regulation be withdrawn, premised on the observations that many of the rules had proved ineffective, that airline divestiture of CRSs had removed much of the incentive that the rules had needed to counter, and that carriers were increasingly using the internet to bypass CRSs.(9)
In short, DOT accepted the Division's analysis and recommendation. DOT has allowed most of the rules to lapse, but maintained for a transition period two that the Division suggested were appropriate, which were the anti-bias rule and rules prohibiting various forms of most favored nations clauses (such as parity clauses and clauses requiring airlines to make all fares available through the CRS).
It's not all about aviation. In a pending matter, the Federal Maritime Commission (FMC) must decide whether to treat equally cargo shippers that have their own ships (vessel-operating common carriers) and those that do not (non-vessel-owning common carriers). As the name implies, a vessel-owning common carrier owns ships and contracts to ship goods on the ships. A non-vessel-owning carrier UPS is an example generally aggregates smaller cargoes and ships them in space leased, wholesale, from the vessel-owning carrier. Currently, under the Ocean Shipping Reform Act,(10) vessel-owning carriers may carry the cargoes of shippers under contracts that are negotiated separately and the terms of need not be published in a tariff. Last year, non-vessel-owning carriers petitioned for the same right.
In October, the Division filed public comments with the FMC, urging it to grant the petition.(11) We made the points that non-tariffed and "confidential" contracts would give the non-vessel-owning carriers more flexibility to negotiate with individual shippers, would eliminate the disclosure of price information to the non-vessel-owning carriers' competitors the vessel-owning carriers that the tariff system produced, and would save the non-vessel-owning carriers needless costs.
We await the FMC's decision.
Similarly, we sometimes weigh in with comments to the Surface Transportation Board. The Surf Board has jurisdiction, and we don't, to approve railroad acquisitions. The Division recently filed public comments urging the Surf Board to impose conditions on approval of Canadian National's acquisition of railroad assets from Great Lakes Transportation.(12)
This merger would combine the only railroad that transports iron ore and taconite moving from several sites in the Mesabe Range of Minnesota to U.S. steel mills in the Great Lakes region via rail/water and all rail services (Great Lakes' railroad, Duluth, Missabe and Iron Range Railway Company, or DMIR), with the potential rail competitor who can enter these markets with the shortest build-outs to the Mesabe Range sites (Canadian National). Barriers to entry are high, and the Division concluded that the threat of potential entry by a second railroad must be preserved to constrain the merging railroads' market power.
The Division recommended that the STB preserve the current competitive discipline Canadian National provided pre-merger by adopting conditions that put a replacement railroad in Canadian National's pre-merger condition. The Division recommended that the conditions give a replacement railroad certain trackage rights and authorize it to connect its line to the Mesabe Range facilities through build-outs or build-ins. The Division left the precise details to the Board to determine, but emphasized that the trackage rights fees a replacement railroad would pay should be set to permit it to compete with the same marginal costs as the merged railroad.
IV. OTHER TRANSPORTATION MATTERS
Yellow Corporation's $1.1 billion acquisition of Roadway Corporation, seemed, when it was announced, like a big deal. The press gave it unusual attention. One of the TEA staffers observed that this was the "[b]iggest trucking merger in the history of the known universe." Yellow and Roadway were the two largest trucking companies that focus on long-haul less-than-truckload freight. Less-than-truckload carriers consolidate smaller shipments from many customers, often using hub and spoke systems. Long-haul less-than-truckload carriers serve the entire continental United States and generally transport shipments 1,000 or more miles.
After closely examining the industry, we determined that this transaction did not warrant a challenge. We concluded that postmerger there would remain sufficient less-than-truckload alternatives, including truckload consolidation, large multi-regional carriers, regional less-than-truckload partnerships, small package carriers, as well as three other long-haul LTL competitors. These alternatives collectively would prevent a combined Yellow/Roadway from imposing a rate increase or reducing service.
The United States and the European Union have been negotiating a comprehensive agreement on civil air services, which will in part replace the various bilateral agreements between the U.S. and EU member states, including the Open Skies agreements. Delegations from the United States (including representatives from the State, Transportation, and Justice departments), the EC (including representatives from DG-TREN and DG-COMP), the EU member states, and industry (including representatives from airlines and airports) have met in Washington and Brussels four times now.
The agreement would address many important issues, including further liberalization of ownership and control, enhancement of aviation safety and security, and consideration of labor and consumer interests. Of special interest to this group, one of the main goals of the agreement would be to broaden competition and enhance cooperation on competition issues in the trans-Atlantic market. Under the terms of the current U.S. proposal, the U.S. and EU would further liberalize scheduling and pricing practices. The agreement will not, however, permit carriers to operate "cabotage" services, that is, the freedom for foreign carriers to fly city pairs within the other's territory on service with no connection to the home territory. For example, Lufthansa could hold out service to fly from Cairo to Milan to New York, with onward service to Houston, but it would not be permitted to sell and operate solely between New York and Houston. Similarly, while United can fly from New York to Paris to Berlin, it would not be permitted to fly from Berlin to Frankfurt.
This is an extraordinary improvement both for the Europeans (who will shed restrictions that remain under a system that treats each of their carriers as a national carrier rather than a European carrier with broader rights) and for the Americans (who hope to clear away particular obstacles, such as limitations on flights to Heathrow). The U.S. is hopeful to have at least a partial agreement by June 2004.
1. Press Release, Department of Justice, Department of Justice and Several States Will Sue to Stop United Airlines from Acquiring US Airways: Deal Would Result in Higher Air Fares for Businesses and Millions of Consumers (July 27, 2001) (available at http://www.usdoj.gov/atr/public/press_releases/2001/8701.htm).
2. See Commission Decision of February 12, 2004 (Air France/KLM, Case No. Comp/M.3280, December 18, 2003).
3. Press Release, Department of Justice, Statement by Assistant Attorney General R. Hewitt Pate Regarding the Closing of the Orbitz Investigation: Investigation Concludes Orbitz Joint Venture Has Not Harmed Consumers or Reduced Competition (July 31, 2003) (available at //www.usdoj.gov/atr/public/press_releases/2003/201208.htm).
4. United States v. Airline Tariff Publishing Company, Civ. Action No. 92-2854 (SSH) (D.D.C. filed Dec. 12, 1992) (available at http://www.usdoj.gov/atr/cases/f4700/4796.pdf).
5. Final Judgment, United States v. Airline Tariff Publishing Company, Civ. Action No. 92-2854 (SSH) (D.D.C. filed Nov. 1, 1993) (available at http://www.usdoj.gov/atr/cases/f4700/4799.pdf.)
6. Final Judgment, United States v. Airline Tariff Publishing Company, Civ. Action No. 92-2854 (SSH) (D.D.C. filed Aug. 10, 1994) (available at http://www.usdoj.gov/atr/cases/f4800/4801.pdf)
7. U.S. v. AMR Corp., 335 F.3d 1109 (10th Cir. 2003).
8. Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 113 S.Ct. 2578 (1993).
9. Reply Comments of the Department of Justice to Notice of Proposed Rulemaking Computer Reservation System Regulations, Department of Transportation (Docket Nos. OST-97-2881, 0ST-97-3014, OST-98-4775, 0ST-99-58888), June 9, 2003 (available at http://www.usdoj.gov/atr/public/comments/201081.htm).
10. Ocean Shipping Reform Act of 1998, Pub. L. No. 108-198, 112 Stat. 1902 (1998).
11. Comments of the United States Department of Justice on Petition of United Parcel Service for an Exemption Pursuant to Section 16 of the Shipping Act of 1984 to Permit Negotiation, Entry and Performance of Service Contracts, Federal Maritime Commission (Petition No. P3-03), Oct. 10, 2003 (available at http://www.usdoj.gov/atr/public/comments/201372.htm).
12. Reply of the United States Department of Justice to Canadian National Railway Company, and Grand Truck CorporationControlDuluth, Missabe and Iron Range Railway Company, Bessemer and Lake Erie Railroad Company, and The Pittsburgh & Conneaut Dock Company, Surface Transportation Board (Finance Docket No. 34424), Feb. 24, 2004 (available at http://www.usdoj.gov/atr/public/comments/202651.htm).
Transportation Update: Remarks To The ABA Section Of Antitrust Law Transportation Industry Committee
Wednesday, March 31, 2004