تاثیر بر ترافیک، سهام بازار و تمرکز اتحادیه های خطوط هوایی بر مسیرهای انتخاب شده اروپا- امریکا

This paper examines the impact of airline alliances on traffic of the constituent airlines using an analysis of US Bureau of Transportation Statistics T-100 International Market Data on a monthly basis for five routes to the US from European hubs. The European hubs are Frankfurt and Paris. The period covered is January 1990–December 2003; a sufficiently lengthy period to enable the derivation of good time-series models before the ‘intervention’ of alliance formation and development. The alliances focussed on are Air France and Delta, part of the SkyTeam Alliance and Lufthansa and United Airlines, part of the Star Alliance. It is possible to distinguish code sharing agreements and then the subsequent immunity from US antitrust legislation. It is also possible to suggest some conclusions on the differences in alliance development in the more liberal open skies environments adopted by many European countries with the more traditional, stricter regulated bilaterals that exist in others such as the UK. Competition is examined using the Hirschman-Herfindahl Index so as to throw light on the impact of alliances on market concentration by route.

A number of publications have examined the impact of alliances, in particular, Oum et al. (2000), the Brattle Group (2002) and Morrish and Hamilton (2002). There are expected differences in impact depending on the type of alliance. More recently, Iatrou and Alamdari (2005) have surveyed and reported on the expectations and perceived impacts of alliances. There is an expectation of a positive impact on traffic on a route as well as on the shares of the alliance members and that these impacts will be greater if the participating airlines operate hub and spoke systems based at both the origin and the destination. In addition, these impacts are thought to reach fruition “between 1 and 2 years of the inception of the partnerships” (Iatrou and Alamdari, 2005) and will be greater from the inception of antitrust immunity. Analysis of traffic data from the US Bureau of Transportation Statistics (2005)2 does not yield unambiguous conclusions in accordance with theory or expectations. Data can be analysed year on year or month on month and the picture that emerges is complex, for amongst other things, capacity on the principal routes examined here is changed by both the incumbent airlines and airlines leaving and entering the market and this causes traffic volumes to fluctuate. Nevertheless, both the date of code sharing agreements and of immunity from US antitrust legislation can be firmly stated,3 so it should be possible using Autoregressive Integrated Moving Average Models (ARIMA) with Intervention Analysis to identify both the size and the significance of these influences on traffic by route. These impacts should be able to be identified if they are significant, despite the implausibility of the ceteris paribus assumption; there are many other influences on traffic volumes and market shares by route besides alliance formation and development, for example, the focus of some US carriers on non-hub EU routes with smaller aircraft types. The impact of alliance formation on market concentration and competition can also be identified by the examination of market share data. These impacts are of current concern as the European Commission and the US have reached a provisional agreement on the extension of open skies policy that requires approval by transport ministers in late March 2007 with implementation in October. It is estimated that passenger numbers would be boosted by 26 million as European airlines would be allowed to make transatlantic flights from any EU country, not just their own nation, and more airlines would be able to use London Heathrow (LHR). British Airways (BA) is against the proposals and the UK government might be as changes in ownership rights will not result in, for example, the foundation in the US of Virgin America, contrary to Richard Branson's desires.4 The chief beneficiaries might be those airlines that have a large number of slots at LHR such as Aer Lingus and British Midland International.5

For the parallel alliances, the intervention analysis shows that for CDG–JFK there is no significant effect whereas for FRA–ORD, it appears there was an almost significant positive impact on traffic from code sharing in accordance with the expectations argument, especially as the alliance partners hub at both ends. However, the result does not support the theory of parallel alliances. By contrast, the complementary alliance intervention analyses for CDG/ORY–ORD and FRA–BOS provide weak evidence that traffic and shares fall which counters the predictions of both theory and expectations. The stronger evidence for FRA–JFK is for a significant negative impact on share from code sharing. The FRA– LAX route, by contrast, provides weak evidence of a positive impact on traffic after code sharing and stronger evidence of a gradual positive impact on shares achieved within a year of code sharing. This supports theory and expectations. By contrast, the strongest intervention effect is found for this route where antitrust immunity has a negative impact on alliance market share. This again is contrary to theory and expectations. It seems that the fluctuations in traffic and traffic shares has more to do with the ceteris paribus conditions than with alliance formation and development, despite the expectations of airlines and the hypotheses of theory. The observed traffic is a product of the interaction of demand and supply through changes in frequency and aircraft size. Over a long period of time, the ARIMA models are able to mimic this traffic based implicitly on these underlying variables. As no intervention representing alliance activity is significant, it can be concluded that although supply and pricing may be affected by such activity, the existing ARIMA models are able to deal with the resulting variations in traffic. Whereas it might be expected that alliance activity would lead to changes in traffic that disturbed the status quo, the insignificance of the interventions suggests that this is not so. Alliance activity does not result in changes in traffic beyond what ARIMA models based on past demand and supply can successfully model. The analysis of market concentration is instructive as it shows that competition declined irrespective of alliancetype. The impact of parallel alliances has this expected result. More suggestively, the comparison of the LHR route demonstrates both the lesser market concentration on LHR–JFK, despite the slot constraints at LHR and the debilitating impact that an alliance between BA and AA might have had on competition. It is hard to see how these results can be viewed as compatible with the views of the Brattle Group (2002) or the recent EU view ( BBC News, 2007 ) that the spread of open skies agreements will increase transatlantic traffic. Open skies agreements do not seem to result in either a significant growth in traffic or in increased competition. Indeed, the strength of the alliances could act as a barrier to entry, contrary to the rhetoric that surrounds open skies policies. 15 However, the deliberate focus of this paper on the north Atlantic clearly ignores markets where open skies policies might satisfy their rhetoric when a large volume of latent demand is satisfied as especially restrictive bilaterals are replaced, for example, the possible case of Japan