EC, December 11, 2009, No M.5648

Dear Sir/Madam,

Subject: Case No COMP/M.5648 – OTPP/ MACQUARIE/ BRISTOL AIRPORT
Notification of 6/11/2009 pursuant to Article 4 of Council Regulation No 139/20041 

1. On 6 November 2009, the Commission received a notification of a proposed Concentration pursuant to Article 4 of Council Regulation (EC) No 139/2004 ("EC Merger Regulation") by which the undertakings Ontario Teachers' Pension Plan Board ("OTPP", Canada) and Macquarie European Infrastructure Fund L.P. ("MEIF", Guernsey) which is part of the Macquarie group ("Macquarie", Australia) acquire joint control over Bristol International Airport Limited ("Bristol Airport") within the meaning of Article 3(1)(b) of the EC Merger Regulation.

I. THE PARTIES

2. OTPP is a Canadian pension fund which invests in a wide range of different assets including publicly listed companies, private equity investments and long term fixed income investments such as bonds. OTPP already has a minority stake which confers it joint control over Birmingham International Airport ("Birmingham Airport").

3. MEIF is a wholesale investment fund that was established to invest in infrastructure and related businesses. MEIF is managed by Macquarie Capital Funds (Europe) Limited, which is part of the Macquarie Group. The Macquarie Group is a global provider of banking, financial, advisory, investment and funds management services.

II. THE OPERATION AND THE CONCENTRATION

4. Bristol Airport is a regional airport located in the South West of the UK. It is currently owned 50 per cent by MEIF (indirectly via its wholly-owned subsidiary MEIF Luxembourg Holdings S.A.) and 50 per cent by a special purpose vehicle (Bristol Airport Bermuda Limited, "BABL") which is 70.9 per cent owned by MAp Airports ("MAp") and 29.1 per cent owned by OTPP. OTPP does currently not have any veto rights over key strategic decisions involving BABL and Bristol Airport.

5. The proposed transaction consists in the sale by MAp of its stake in BABL to OTPP.2 As a result of the proposed transaction, OTPP will hold 98 per cent of the equity of BABL and hence control 50 per cent of the shares in Bristol Airport while MEIF will retain its 50 per cent interest. Both OTPP and MEIF will have equal rights as regards the strategic and commercial policy of Bristol Airport, including veto rights over key commercial decisions concerning the appointment of senior management and the determination of the annual business plan and budget.

6. Accordingly, post transaction, Bristol Airport will be jointly controlled by  OTPP  and MEIF. OTPP will thus acquire control over Bristol Airport within the meaning of Article 3(1)(b) of the EC Merger Regulation.

III. COMMUNITY DIMENSION

7. The undertakings concerned have a combined aggregate worldwide turnover of more than EUR 5 billion (Macquarie Group: EUR [...] million; OTPP: EUR [...] million) and both have a Community-wide turnover in excess of EUR 250 million (Macquarie Group: EUR [...] million; OTPP: EUR [...] million). Moreover, Macquarie Group does not generate more than two-thirds of its Community-wide turnover in one and the same Member State. The notified operation therefore has a Community dimension pursuant to Article 1(2) of the  EC Merger Regulation.

IV. MARKET DEFINITION

Relevant product market

8. Both OTPP and Macquarie are active in the management and operation of UK airports. The parties submit that the relevant product market is the provision of aeronautical services, wich may be subdivided into the provision of airport infrastructure services to airlines and the provision of ground handling services to airlines.

9. Such a distinction between the provision of airport infrastructure services and ground handling services3 is in line with previous Commission decisions.4 In addition, given the recent developments in the aviation sector, in particular the increase in the number of low cost operations, the Commission has considered previously also whether the provision of airport infrastructure services could be further sub-segmented (i) on the basis of the type of airline customer, i.e. charter operators, scheduled full service carriers and scheduled low cost carriers, or/and (ii) on the basis of the type of flight (i.e. long haul / short haul).5

10. The market investigation has confirmed that a separate market for the provision of airport infrastructure services needs to be distinguished, but it has not revealed a clear result to which extent a further subdivision of that market is necessary in the present case. However, the precise product market definition can be left open as no serious doubts as to the compatibility of the notified transaction with the internal market would arise under any alternative market definition considered.

Relevant geographic market

11. The parties submit that it is not appropriate to adopt a rigid approach to geographic market definition when assessing competition within the UK airport sector as a fixed geographic market boundary does not take account of the importance of the relative proximity of airports in assessing the degree of competitive constraints they impose on each other.6 However, the parties have provided market share data on various different basis.7

12. As far as the provision of airport infrastructure services to airlines is concerned, the Commission has in previous cases generally left open the exact geographic market definition.8 The Commission has tended to focus on the appropriate size of  the  catchment area of airports, taking into account factors such as the size and density of population, level of wealth and type and size of business in the area. Furthermore, the Commission concluded in a previous case that Bristol Airport is not part of the same geographic market as any of the London airports, as the vast majority of airlines would not move operations from London airport to Bristol airport or vice versa since they were considered as not forming part of the same catchment area.9

13. The market investigation in the present case confirmed that an individual assessment of the relevant airports' catchment areas is required in order to determine the geographic dimension of the relevant market. While some respondents consider that Bristol and Birmingham airports both have catchment areas wider than 2 hours (in particular for long-haul flights), several other respondents consider that the catchment areas of both airports are smaller and only overlap to a limited extent. This is further evidenced by CAA surveys, which show that Birmingham Airport accounts for at most 1 per cent of passengers travelling to or from Bristol's core catchment area of Somerset, and that Bristol Airport accounts for under 0.5 per cent of passengers travelling from  Birmingham Airport's core catchment area of West Midlands county.10

14. The limited overlap between the catchment areas of both airports is underlined by catchment area studies transmitted to the Commission by several airlines. These studies, based on a large customer survey for Birmingham and on the CAA periodic survey for Bristol, show that the catchment area overlap for the two airports is mainly limited to north of Bristol and south of Birmingham, i.e. South Gloucestershire. Moreover, according to a survey provided by an airline operating from Birmingham, passengers originating from Bristol constitute a very limited share of total passengers booking flights in the Birmingham region.

15. The very limited substitutability between Bristol and Birmingham is further evidenced by the fact that a majority of airlines would in the case of permanent price increases by 5%-10% for airport infrastructure services at Bristol or Birmingham Airport not move to the respective other airport.

16. In the present case, the exact determination of the geographic market can be left open as no competition concerns arise under any alternative geographic market definition considered.

V. COMPETITIVE ASSESSMENT 

17. The parties' combined market shares on the basis of passenger numbers (July 2008 – June 2009) do not exceed 26% under all product and geographic market definitions discussed above, unless the geographic scope of the relevant market is limited to the  area that can be reached within a 2h drive time from Bristol Airport.11

18. In such a narrow market, the transaction would lead to combined market shares of  84.3% (airport infrastructure services, Birmingham Airport 52.5%, Bristol Airport 31.8%), 84.4% (short-haul sub-segment, Birmingham Airport 50.8%, Bristol Airport 33.6%) and 95.6% (long-haul sub-segment, Birmingham Airport 78.7%, Bristol  Airport 16.9%) respectively. Similar combined market shares would arise if a further distinction was drawn between charter passengers and scheduled passengers.

19. At the outset, it should be noted that these market shares are based on the number of passengers12  who have used the relevant airport (and not on the number of airlines or  the value of infrastructure services provided to them). The market shares in the present case therefore only indirectly reflect the market position of airports vis-à-vis their airline customers.

20. More importantly, market shares based on fixed drive-time catchment areas are  of limited significance when assessing competition between airports in the UK. This is illustrated in the present case by the fact that Birmingham Airport is a 1:59h drive time from Bristol Airport and would thus not fall into the same market as Bristol Airport if the relevant geographic market was defined as a narrower (e.g. 1:45h drive time) catchment area from Bristol Airport. Conversely, if the relevant market was defined as a larger (e.g. 2:15h drive time) catchment area from Bristol Airport, the parties' combined market share would decrease significantly due to the inclusion of Heathrow airport which is a 2:05h drive time from Bristol Airport.13 Accordingly, it was emphasised in  the market investigation that fixed drive-time isochrones cannot determine a single catchment area for any given airport in the UK.

21. The competitive landscape in the UK is further characterised by a wide choice of airports, which according to a responding airline "affects the catchment area for each airport". Similarly, the UK Competition Commission has recently found that a “continuum of substitution possibilities exists” depending on distance and other characteristics of the relevant airports and that “any market definition beyond a single airport is, to an extent, arbitrary and assessment of market shares is unlikely to be a useful tool in itself for measuring market airport power”.14

22. Against this background, the parties argue that the London airports exert a significant competitive constraint upon Bristol Airport and Birmingham Airport. This submission is to a certain extent confirmed by the statements of several respondents to the market investigation who, as explained in more detail below, consider London Heathrow, Gatwick airport and other London airports as alternatives to Bristol and Birmingham Airport, despite current capacity constraints in particular at London Heathrow and Gatwick airport.15 This illustrates that some constraints emanate from the London airports, in particular for long-haul flights, in line with the statement of several respondents that catchment areas are wider as far as long-haul services are concerned.

23. Furthermore, CAA data evidence that more than 30% of passengers travelling to/from the Somerset county (i.e. the county in which Bristol Airport is located) have used London Heathrow or Gatwick airport in the past. Moreover, for each county situated in the catchment area of Bristol airport16, more than 30% of all passengers travelling to/from those counties have also used London Heathrow or Gatwick airport. As a result of  the  inclusion  of  the  London  airports,  the  parties'  market  share  would decrease considerably and would not exceed 26% under any market definition discussed above. It follows from the above that market shares based on a fixed 2:00h drive time catchment area from Bristol Airport are only of limited significance for the competitive assessment in the present case.

24. Conversely, the competitive constraint that the two airports concerned exert upon each other is of particular importance for the assessment of the present case. The parties  argue in this regard that Bristol Airport and Birmingham Airport only impose a very limited competitive constraint upon each other due to the limited overlap of their catchment areas.

25. The limited mutual competitive constraint between the two airports has been confirmed by the market investigation. Although Bristol Airport and Birmingham Airport are generally considered as competing with each other and two airlines indicated that the transaction might negatively influence prices at both airports, many respondents underlined that the catchment areas of both airports overlap only to a limited extent (North of Bristol, South of Birmingham) and that the degree of competition between them is therefore limited.

26. This is also in line with the submission of most airlines responding to the market investigation that they would not move to Bristol or Birmingham in case of price increases at the respective other airport. Some airlines also indicated that they would rather cease a service at Bristol Airport or Birmingham Airport than move it to the respective other airport and another respondent recently terminated operations at Bristol Airport without changing operations at Birmingham Airport. These findings further evidence the limited constraint that both airports exert upon each other.

27. In addition, all airlines responding to the market investigation consider that for both Bristol Airport and Birmingham Airport several close alternatives exist.

28. More in particular for Bristol airport, Cardiff Airport (approximately 1:30h away from Bristol Airport) and London Heathrow have been mentioned by most airlines as one of the close alternatives to Bristol; further such alternatives mentioned were Exeter, Southampton, Plymouth and Gatwick. It is interesting to note that approximately half of all responding airlines did not mention Birmingham Airport among the  close alternatives to Bristol.

29. Also a competitor emphasised that Bristol Airport focussed on low-cost and charter short-haul business and that Cardiff and Exeter Airports would be an alternative in that regard. This submission is in line with data received from Cardiff Airport and Exeter Airport and the views of further competitors responding to the market investigation; for instance, another competitor stated that Cardiff Airport is "within the established Bristol catchment area, with only a marginal differential in terms of average access time"). Furthermore, Exeter Airport provided examples of long-haul services switching between Bristol and Exeter and vice versa, while Cardiff Airport provided examples of long- and short-haul services switching from Cardiff to Bristol Airport. The market investigation further revealed that capacity is available at Cardiff and Exeter airports for airlines wishing to move their operations to these airports, and Cardiff plans to extend its capacity in the near future.

30. For low-cost operations in particular, Gatwick Airport (which is located 2h41 from Bristol Airport) also represents a certain competitive constraint upon Bristol Airport, in line with the statements of some market participants that non-time sensitive passengers accept longer travel times to the airport.

31. Similarly, as regards Birmingham Airport, the market investigation revealed several close alternatives other than Bristol Airport. A majority of responding airlines  mentioned East Midlands Airport (approximately 45 minutes from Birmingham Airport) as an alternative in particular for short-haul routes; one airline has recently moved certain short-haul routes from Birmingham Airport to East Midlands, due to the latter's considerably lower cost base and free capacity. Further alternatives mentioned were Manchester Airport and London Heathrow (both also for long-haul routes) as well as Stansted, Luton and Gatwick.

32. In the light of the foregoing, i.e. the limited degree of competition between the parties, the availability of a sufficient number of alternative airports and the competitive constraint stemming from the London airports, the proposed concentration does not raise serious doubts with regard to the market for the provision of airport infrastructure services.

VI. CONCLUSION

33. For the above reasons, the Commission has decided not to oppose the notified operation and to declare it compatible with the internal market and with the EEA Agreement. This decision is adopted in application of Article 6(1)(b) of the EC Merger Regulation.

1 OJ L 24, 29.1.2004 p. 1.

2 MAp will retain a 2 per cent stake in BABL post transaction. 

3  The transaction does not lead to an overlap in ground handling as neither party is active in the provision 
of ground handling services at Bristol or Birmingham Airport but such services are operated by external 
operators. 

4 See case IV/M.786 – Birmingham International Airport paragraph 14; case M.1035 Hochtief/Aer 
Rianta/Düsseldorf Airport, paragraph 10; case M.2262 – Flughafen Berlin II, paragraph 13; case 
COMP/M.3823 – MAG/Ferrovial Aeropuertos/Exeter Airport, paragraph 15; case COMP/M.4164 –
Ferrovial/Quebec/GIC/BAA, paragraphs 10 to 14. 

5 Case COMP/M.4164 – Ferrovial/Quebec/GIC/BAA, paragraphs 10 to 14; case COMP/M.5652 – 
GIP/Gatwick Airport, paragraphs 9-11. 

6 The parties refer in this regard to the findings of the UK Competition Commission (the "CC") in its in-
depth market investigation into BAA (which is the leading operator or airports in the UK) and UK airport 
markets ("BAA airports market investigation", published on 19 March 2009). 

7  The parties provided data for the following market delineations: Great Britain; England and Wales; the 
area comprising a 2 ½ hour drive time from both airports; a Birmingham Airport catchment area based on 
a 2 ½ hour drive time; a Bristol Airport catchment area based on a 2 ¾ hour drive time; the area 
comprising a 2 hour drive time from both airports; a Birmingham Airport catchment area based on a 2 
hour drive time; and a Bristol Airport catchment area based on a 2 hour drive time. 

8 See case COMP/M.3823 – MAG/Ferrovial Aeropuertos/Exeter Airport, paragraphs 16 to 20; case 
COMP/M.4164 – Ferrovial/Quebec/GIC/BAA, paragraphs  15 to 22; and case COMP/M.5652 –
GIP/Gatwick Airport, paragraph 15.

9 Case No COMP/M.4164 – Ferroial/Quebec/GIC/BAA. 

10  CAA Passenger survey report 2005, 2006,2007 and 2008. 

11  The parties were unable to calculate market data separately for low-cost carriers. However, as explained 
in more detail in the competitive assessment, no serious doubts as to the transaction's compatibility with 
the internal market arise even if the relevant market was confined to airport infrastructure services 
provided to low-cost airlines. 

12  The use of  passenger numbers is in line with the approach used in past Commission decisions.  

13  Drive times are obtained via www.viamichelin.com.

14 UK CC Report "BAA airports market investigation", published on 19 March 2009, paragraph 2.46. 

15  It appears from the market investigation that London Heathrow is generally considered as congested 
whereas Gatwick airport seems to have some limited spare capacity in particular during off peak periods. 

16  The catchment area comprises those counties where more than 20% of passengers travelling to/from used Bristol airport. This approach is in line with the findings of the UK CC.