Report under section 125(4) of The Fair Trading Act 1973 of The Director General's advice, dated 16 August 2001, to the Secretary of State for Trade and Industry under section 76 of the Act
The Go-Ahead Group PLC (Go-Ahead) is a transport group based almost entirely in the UK, providing bus, coach, taxi, vehicle rental and rail services. Go-Ahead also provides ground handling services at many UK airports. Go-Ahead operates the Thames Trains franchise and owns 65% of the share capital of Govia Limited. In 1999 Go-Ahead had a turnover of 869m euros, approximately [see note 1] £540m.
Keolis (formerly known as Generale de Transport et d'Industrie SA (Via)) provides transport services including bus, coach, rail and light railway services in a number of European countries. Keolis's UK operations consist of a 35% stake in Govia Limited. Keolis's ultimate parent company is SNCF. In 1999 Keolis had a turnover of 1.1bn euros, approximately £684m.
Govia Limited (Govia) is a shell joint venture company formed to acquire and operate train operating companies. It presently operates the Thameslink rail franchise and has been named as the SRA's preferred bidder for the new South Central rail franchise.
Connex South Central Limited (Connex) is a train operating company providing passenger rail services in the South Central franchise area. The present franchise is due to expire in May 2003. In 1999 Connex had a turnover of 473m euros, approximately £294m.
Govia notified its proposed acquisition of the entire share capital of Connex South Central under the ECMR on 15 June 2001. The consideration is expected to be between £30m and £42.5m, depending upon the final price adjustment.
This case was referred back to the UK by the European Commission under Article 9(2)(b) of the ECMR on 20 July 2001, following notification to the European Commission on 15 June 2001. Article 9(2)(b) allows the European Commission to refer to member states cases that affect competition on distinct markets which do not constitute a substantial part of the common market. In such cases it is not necessary to demonstrate a threat to create or strengthen a dominant position as a result of which effective competition will be significantly impeded, which is a requirement of Article 9(2)(a).
In previous cases [see note 2], the European Commission has defined the relevant service market as the supply of public passenger transport services by railway. The MMC found in past reports that a degree of competition may exist from coach services on some routes [see note 3], particularly in the leisure market, although rail would seem to provide stronger competition to coach than coach does to rail.
The MMC also concluded [see note 4] that car travel does not provide a sufficiently strong competitive restraint on rail fares partly because other factors seem to determine rail travel (e.g. quicker journeys, less stressful journeys, lack of access to a car, etc). For airport services, rail may face considerably greater competition from private cars, hire cars and taxis than is generally the case for a number of reasons. First, there is higher car ownership amongst air passengers. Second, private transport offers the convenience of door-to-door travel when carrying luggage. Third, a high proportion of passengers using airports travel in groups of two or more, reducing the cost of using cars/taxis.
Peak-time commuters are likely to be less price-sensitive and more time-sensitive than off-peak users. The SRA regulates the price of most commuter fares, such that the principal issue to consider in this case is the potential effect on off-peak fares.
The relevant geographic market is the specific origin-destination points within the franchise area. Under this possibly narrow definition, five local markets have been identified where the parties overlap. These are:
The parties operate out of different London stations, Connex from primarily Victoria but also Charing Cross/London Bridge, and Thameslink from Blackfriars/London Bridge. While it could be argued that these stations serve different markets, it is thought that there will be a significant percentage of passengers for whom one station is an effective substitute for another. London is therefore considered, for the purposes of analysis, to be the relevant origin/destination point.
Market shares for the five overlapping routes are set out below in Table 1. The particular characteristics of each of these routes will be considered before an assessment is made of the likely effects arising, which will be made in the wider context of competitive constraints on the franchises concerned.
Table One: Market Shares for Origin – Destination Pairs Identified [see note 5]
N/A – Not applicable
Source: The parties (based on Capri/ATOC data)
The Gatwick-Redhill route is very short, around 6 miles. Over this distance car, taxi and bus travel are likely to provide a significant competitive constraint such that the competition issues are unlikely to be significant.
The Sutton-London route is a main commuter route within Greater London. Sutton is in London Transport Zone 5. A large proportion of passengers hold season tickets, so that regulated fares account for a significant proportion of total revenue on this line (e.g. 78% for Thameslink). Off-peak fares are likely to be capped by the cost of one-day travel-cards, which are available after 9.30 am and provide customers with the flexibility to use trains, buses and the underground during their journeys. The loss of on-rail competition on this route does not therefore cause any concern.
The London-Gatwick route is the largest market in revenue terms. The biggest operator on this route will continue to be Gatwick Express, which has a very high market share. Coach services (owned by the operator of the Gatwick Express) also operate on this route and may provide an alternative for leisure travellers. Other constraints arise from the regulatory regime and the business context, which are discussed later.
On the Brighton-Gatwick route, the parties would have a virtual monopoly in rail services. Coach services operate two per hour on this route compared to five per hour for trains. Journey times and fares are similar, which, given the general preference for rail, may indicate that coach acts only as a marginal constraint on off-peak rail fares. About 60% of revenue is derived from regulated fares, while the desire to raise the number of off-peak passengers may act as a significant constraint on the ability to raise off-peak fares.
Brighton-London is a main commuter route with a large proportion of passengers holding season tickets. Regulated fares account for a significant proportion of total revenue on this line (e.g. 68% for Thameslink). Coach services (not owned by Go-Ahead) operate once per hour on this route compared to seven per hour for trains, but coach fares are cheaper. Coach travel may therefore act as a restraint on off-peak rail fares.
The parties are required by the SRA to maintain their current service levels [see note 6] for both the Thameslink and South Central franchises until Summer 2003. Govia has a commercial incentive to behave well during the negotiation of details of the new South Central franchise and during the process of bidding for the new Thameslink franchise.
There is arguably also a business rationale for not raising off-peak fares in real terms. Govia's fixed costs, which comprise payments to Railtrack and the rolling stock leasing companies, amount to some 80% of its costs. Consequently the marginal cost of running an extra service is low. There is, therefore, a commercial incentive to provide extra services at off peak times and to sell spare seats using promotional fares. Other TOCs are in exactly the same position and they have for some time been trying to find ways to increase passenger numbers, both by encouraging more regular travel and by encouraging people to switch from other modes of travel. All TOCs behave in this way even though most of them do not face competition on their main routes. Such behaviour is expected to be a feature of the industry during the period leading to the award of the new Thameslink franchise.
Where two franchise areas overlap, the incumbent of one franchise may have an advantage when bidding for the other franchise. In future bidding for Thameslink, Govia may have better access to information as the incumbent than independent bidders, but the merger would not appear greatly to enhance this possible asymmetry of information, and the SRA is well placed to address this during the franchising process. Govia may also have a small commercial advantage, as it will be able to make certain economies of scale and scope not available to other bidders. However, the SRA does not automatically award franchises to incumbents (as the South Central case illustrates) and is able to take account of these advantages in its assessment of franchise bids. And as a general point, efficiency gains which through competition benefit consumers are desirable.
I have also considered the effect of the transaction on the parties' share of the national passenger franchise market and the effect on the voting structure of the Association of Train Operating Companies (ATOC). No significant concerns arise in either of these areas.
No significant vertical issues arise in this case.
Most third parties, including competitors and representatives of passengers, were relaxed or even positive about this transaction. Only one third party had any concerns. These have been considered above.
This transaction results in common ownership of certain overlapping rail routes in local markets for a period which will probably run until the award of the new Thameslink franchise. The degree of overlap is small relative to the context of the South Central franchise (but not in relation to the Thameslink franchise), and the time period is relatively short in the context of proposals for twenty-year franchises. As the preferred bidder following competition for the new South Central franchise Govia has offered significant benefits for passengers.
As discussed above, price controls, alternative modes of transport and a competing operator will all play some part in constraining behaviour on these routes. With regard to the Brighton-to-Gatwick and Brighton-to-London routes, service frequencies and most commuter fares [see note 7] are already set for both the South Central and Thameslink franchises for the next two years and will remain regulated throughout the period under consideration.
These service commitments and other commercial realities may be expected to discipline Govia's behaviour to a considerable degree. It is nevertheless the case that there will be some short term loss of on-rail competition on the local routes identified above. In the context of this case, however, on balance I do not believe that this justifies referral of the merger. First, Govia is the preferred bidder for the new South Central franchise as a result of competition for that franchise. Second, the possible loss in competition is on routes that are small in relation to the South Central franchise (though not in relation to the Thameslink franchise). This raises issues of proportionality. Third, and significantly, the on-rail competition issues can be considered afresh when the new Thameslink franchise is to be awarded. These considerations in combination lead me to conclude that referral would not be appropriate in this case.
I therefore recommend that you do not refer this transaction to the Competition Commission.
1. The conversion of euros to pounds is based on the official rate of 0.6217 at 31 December 1999.
2. M748 CGEA/Network South Central and M901 GoAhead/GoVia/Thameslink
3. See MMC report into National Express/Midland Mainline report (1996).