McGill’s Bus Services Limited/Arriva Scotland West Limited ...webarchive.nationalarchives.gov.uk/20140402141250/http:/www... · McGill’s Bus Services Limited/Arriva Scotland West

McGill’s Bus Services Limited/Arriva Scotland West Limited merger inquiry

A report on the completed acquisition by McGill’s Bus Services Limited of the business and assets of Arriva Scotland West Limited

21 September 2012

Competition Commission 2012

Website: www.competition-commission.org.uk

http://www.competition-commission.org.uk/

iii

Members of the Competition Commission who conducted this inquiry

Professor Alasdair Smith (Chairman of the Group)

Robin Aaronson

Phil Evans

Roger Finbow

Chief Executive and Secretary of the Competition Commission

David Saunders

The Competition Commission has excluded from this published version of the final report information which the Inquiry Group considers should be excluded having regard to the three considerations set out in section 244 of the Enterprise Act 2002 (specified

information: considerations relevant to disclosure). The omissions are indicated by []. Non-sensitive wording is indicated in square brackets.

1

Contents Page

Summary .............................................................................................................................. 2 Findings ................................................................................................................................ 5 1. The reference .................................................................................................................. 5 2. The bus industry and the companies ............................................................................... 5

The regulation of local bus services in Scotland .............................................................. 5 The relevant public bodies which regulate local bus services in Scotland .................. 5 Statutory quality bus partnerships .............................................................................. 6 BSOG and concessionary fares................................................................................. 7 Disability Discrimination Act ....................................................................................... 8

The companies ................................................................................................................ 8 McGill’s ...................................................................................................................... 8 Arriva Scotland West ................................................................................................. 9

3. The acquisition .............................................................................................................. 10 The acquisition process ................................................................................................. 10 Rationale for the transaction .......................................................................................... 10 Details of the transaction ............................................................................................... 11

4. The relevant merger situation ........................................................................................ 11 5. The relevant market ...................................................................................................... 13 6. The counterfactual ........................................................................................................ 15

The views of parties ...................................................................................................... 15 Our assessment ............................................................................................................ 16 Conclusion on the counterfactual .................................................................................. 18

7. The competitive assessment ......................................................................................... 18 Existing and potential operators in the Renfrewshire area ............................................. 19

McGill’s and ASW .................................................................................................... 20 FirstGroup ............................................................................................................... 22 Riverside Transport ................................................................................................. 24 Stagecoach ............................................................................................................. 24 Others ..................................................................................................................... 26

The overlap flows between McGill’s and ASW............................................................... 26 Levels of competition pre-acquisition and their likely development over time ................ 28

Impact of head-to-head competition between McGill’s and ASW ............................. 31 Conclusions on competition pre-acquisition ............................................................. 31 Sustainability of head-to-head competition over time ............................................... 32

Impact of the acquisition on actual competition ............................................................. 33 Assessment of each flow overlap ............................................................................ 33 Entry or threat of entry via new services .................................................................. 35 Conclusions on the loss of actual competition ......................................................... 44

Impact of the acquisition on potential competition ......................................................... 45 Loss of potential competition ................................................................................... 45 Loss of potential competition within the Renfrewshire area ...................................... 45 Loss of potential competition outside the Renfrewshire area ................................... 46 Conclusions on the loss of potential competition ..................................................... 47

8. Conclusions on the substantial lessening of competition test ........................................ 47 Appendices A Terms of reference and conduct of the inquiry B The regulation of local bus services in Scotland C Financial information on McGill’s and Arriva Scotland West D Jurisdiction E Market definition F Loss of actual competition G Entry and expansion Glossary

2

Summary

1. On 18 April 2012, the Office of Fair Trading referred the completed acquisition by McGill’s Bus Services Limited (McGill’s) of the business and assets of Arriva Scotland West Limited (ASW) to the Competition Commission (CC) for investigation and report. The reference was made under section 22(1) of the Enterprise Act 2002 (the Act). Our provisional findings were published on 13 August 2012. We are required to publish our final report by 2 October 2012.

2. The acquisition was completed on 26 March 2012. McGill’s acquired the business undertaken by ASW and specified assets including vehicles and depots.

3. McGill’s is a family-owned bus operator that provides primarily commercial local bus services in Inverclyde, Renfrewshire and East Renfrewshire, with some services running into Glasgow. In 2011, McGill’s had around 160 buses, located at two depots: Barrhead, East Renfrewshire, and Greenock, Inverclyde.

4. ASW was a subsidiary of Arriva plc (Arriva) whose ultimate holding company is Deutsche Bahn AG. In the UK, Arriva is the third largest regional bus operator. ASW was Arriva’s only operation in Scotland and provided primarily commercial local bus services in Renfrewshire and East Renfrewshire with some services running into Glasgow. At the time of the acquisition, ASW had 153 buses located at two depots: Johnstone and Inchinnan.

5. Following deregulation of the bus industry in 1985, up to 25 small independent oper-ators delivered local bus services in the Renfrewshire area. Many were of a low quality and the high number of bus services also led to severe congestion in some parts of Paisley. In March 2011, the Paisley Statutory Quality Bus Partnership came into force to improve vehicle standards and reduce emissions, leading to a number of small operators reducing their services or, more commonly, divesting their business, often to McGill’s.

6. There have been recent changes to subsidies in Scotland that have had an impact on the profitability of commercial bus services, particularly in urban areas. The Bus Service Operator’s Grant (BSOG) is a subsidy that reduces mileage costs incurred by local bus operators. Since April 2012, payments have been made on the basis of a rate per eligible kilometre with no link to fuel used or fuel duty. There is also a reimbursement scheme for concessionary passengers in Scotland. The rate of reimbursement reduced in April 2010. Total payments under the scheme have also been capped.

7. We concluded that Renfrewshire was a ‘substantial part of the UK’, that the share of supply test was met and that a relevant merger situation had been created.

8. We concluded that the appropriate market definition was at the level of an individual flow (journey from A to B), including flows that show a broadly similar level of frequency and frequency distribution over time; that flexible modes of transport (including the car) were not included in the product market definition; and that there were separate markets for commercial and supported services. We considered whether rail should be included as a constraint for particular overlap flows.

9. We considered the situation that would have prevailed absent the acquisition (the counterfactual). We concluded that, on balance, the most likely counterfactual was one of McGill’s and ASW continuing to operate commercial bus services in the Renfrewshire area, and that it was less likely that Arriva would sell ASW or, despite ASW’s low levels of profitability, exit the business through closure. However, we

3

formed the view that competition between McGill’s and ASW would have lessened substantially over the foreseeable future (see paragraph 13).

10. The bus services of McGill’s and ASW overlapped primarily in Renfrewshire, as well as in parts of East Renfrewshire and Glasgow. We term this ‘the Renfrewshire area’. We identified 15 McGill’s services on which there was at least one significant overlap flow with ASW services. We estimated that more than three-quarters of McGill’s revenues and concession trips were subject to some degree of competition from at least one of ASW’s services.

11. Pre-acquisition, in addition to McGill’s and ASW, a number of services in the Renfrewshire area were run by First Glasgow, Stagecoach and a few other small bus operators: Riverside Transport, Colchri Coaches and Key Coaches.

12. We assessed the level of competition between McGill’s and ASW pre-acquisition. We concluded that McGill’s and ASW were each other’s closest competitors in the supply of bus services in the Renfrewshire area. Although we could not construct the full competitive interaction between the two, we were told of a number of examples of intense competitive interactions between McGill’s and ASW, representing a substan-tial proportion of McGill’s business in the Renfrewshire area and leading to benefits for passengers in terms of lower fares and higher frequencies. ASW’s operations reports also suggested that ASW monitored and often reacted to McGill’s actions (both in terms of frequency and, less actively, fare changes) more than any other bus operator.

13. We also considered whether such competitive interaction was likely to be sustainable over time. We found that there was sufficiently high demand on some flows in the Renfrewshire area to allow more than one operator to be present. We noted, how-ever, that head-to-head competition between McGill’s and ASW only started in 2009 when McGill’s entered the Renfrewshire area and developed over the period leading up to the acquisition as McGill’s acquired a number of smaller operators. During this time, McGill’s was expanding against a competitor, ASW, that it described as declin-ing. We thought it likely that the level of head-to-head competition that we observed at the time of the acquisition represented an upper bound because, in large part, it still represented ASW’s reaction to McGill’s entry and expansion in the Renfrewshire area. Both McGill’s and ASW’s profitability had been declining for several years. ASW was not making the levels of return sought by the larger bus operators across the UK as a whole. We thought it likely that, in the absence of the acquisition, McGill’s and ASW’s competitive offers would have become less attractive than they had been in the period immediately running up to the acquisition. We therefore formed the view that the competitive interaction between the two companies observed at the time of the acquisition was likely to reduce substantially over the foreseeable future.

14. We concluded that the acquisition would give McGill’s a very large market share on a number of overlap flows, covering at least half of McGill’s revenues/ concession trips. We did not think that other providers already present on these overlap flows were likely to exert a sufficient constraint on McGill’s post-acquisition. They had fewer services, lower frequencies or were only present on some of the overlap flows and they were not, in our view, likely to increase frequencies or, modify routes sufficiently to constrain McGill’s across its network, except as part of a wider entry strategy.

15. We therefore looked at whether providers with nearby facilities who were not currently active on the overlap flows would find it profitable to launch new services to compete on all or part of the overlap flows. We found there to be no structural factors which would be a significant barrier to a bus operator’s ability to enter or expand on

4

the overlap flows identified. However, in our view, issues of route profitability and possible incumbents’ reaction to entry would provide disincentives for smaller oper-ators in particular to enter on new flows. We thought it unlikely that small-scale entry would act as a sufficient constraint.

16. Although no operator told us that it had current plans to expand into the Renfrewshire area, we considered whether entry could become more attractive post-acquisition, or whether McGill’s would be likely to perceive this to be the case. We looked in particu-lar at the nature of McGill’s as incumbent and the likelihood of entry by the two large operators with nearby facilities: First Glasgow and Stagecoach.

17. We noted that, relative to either FirstGroup or Stagecoach, McGill’s is a small, family-owned local bus operator. It does not have the management structure or financial reserves of a large company. This is likely to influence its perception of the constraint posed by the larger operators. We found that it would have strong incentives, at least for the next few years, to avoid providing opportunities for entry by reducing fre-quencies or increasing fares.

18. On balance, we reached the conclusion that McGill’s, as a smaller bus operator, perceived First Glasgow to be a credible threat and that this would be likely to con-strain McGill’s from substantially worsening its offer post-acquisition (relative to the counterfactual). Given the low barriers to entry and expansion, its extensive network, its presence already on some flows in the Renfrewshire area and, in particular, its spare capacity nearby, we had no doubt that First Glasgow could enter new flows in a timely way should a profitable opportunity arise. Given its financial resources, we thought that First Glasgow would be willing to take on McGill’s and be prepared to suffer some short-term losses should McGill’s react to its entry. The past examples of First Glasgow’s entry and expansion suggest that a threat of entry is credible. We also thought that the incentive for one of the larger operators to enter against McGill’s was greater than it would have been against ASW. We therefore thought it likely that McGill’s would perceive First Glasgow as being prepared to enter flows on which it was not currently present. The presence of Stagecoach would, in our view, add to the likelihood that McGill’s behaviour would be constrained post-acquisition.

19. We therefore concluded that there was no substantial loss of actual competition as a result of the acquisition.

20. We also concluded that there was no loss of potential competition from the acqui-sition either within or outside the Renfrewshire area.

21. We concluded that there was no substantial lessening of competition as a result of the acquisition of ASW by McGill’s.

5

Findings

1. The reference

1.1 On 18 April 2012, the Office of Fair Trading referred the completed acquisition by McGill’s of the business and assets of ASW to the CC for investigation and report. The reference was made under section 22(1) of the Act.

1.2 Our terms of reference, together with information on the conduct of the inquiry, are set out in Appendix A. We are required to publish our final report by 2 October 2012.

1.3 This document, together with the appendices, constitutes our final report which we are required to publish under section 38(1) of the Act. Further information, including non-commercially sensitive versions of McGill’s and ASW’s submissions, summaries of evidence from third parties, and our provisional findings published on 13 August 2012, can be found on our website.1

2. The bus industry and the companies

2.1 In this section, we outline the regulatory regime which governs the provision of local bus services in Scotland (see paragraphs 2.2 to 2.14). We then provide a brief out-line of the two companies which are the subject of this inquiry: McGill’s (see para-graphs 2.15 to 2.21) and ASW (see paragraphs 2.22 to 2.26).

The regulation of local bus services in Scotland

2.2 In the following paragraphs we outline:

(a) the relevant public bodies which regulate local bus services in Scotland (see paragraphs 2.3 to 2.6);

(b) the key features of the Statutory Quality Bus Partnerships (SQPs) for Paisley and for Glasgow which affect bus services in Renfrewshire and East Renfrewshire (see paragraphs 2.7 to 2.10);

(c) the public funding of local bus services in Scotland through the Bus Service Operator’s Grant (BSOG) and the concessionary fares regime (see paragraphs 2.11 to 2.13); and

(d) the Disability Discrimination Act 1995 (DDA) (see paragraph 2.14).

A more detailed description of each of these is set out in Appendix B.

The relevant public bodies which regulate local bus services in Scotland

2.3 There are several public bodies involved in the regulation of local bus services in Scotland:

(a) The Scottish Government has specific responsibilities for setting and implement-ing transport strategy and policy in Scotland.2,3

1 www.competition-commission.org.uk/our-work/mcgills-bus-services-limited-arriva-scotland-west-limited-merger-inquiry. 2 Certain reserved matters, such as road safety and licensing, remain the responsibility of the UK Parliament.

http://www.competition-commission.org.uk/our-work/mcgills-bus-services-limited-arriva-scotland-west-limited-merger-inquiry

6

(b) Transport Scotland is the national transport agency responsible for helping to develop transport policy, delivering the Scottish Government’s investment pro-gramme, overseeing the safe and efficient running of the transport system, and running the national concessionary schemes (see paragraph 2.13).

(c) The Regional Transport Partnerships (RTPs) are responsible for planning and delivering transport solutions at a regional level. There are seven of these in Scotland. Strathclyde Partnership for Transport (SPT) is the RTP for the West of Scotland which includes Renfrewshire and East Renfrewshire.4 In addition to being an RTP, SPT is also the Local Transport Authority (LTA) for the West of Scotland.5

2.4 The majority of local bus services in Scotland are commercial with local bus oper-ators developing their own routes and networks and setting their fares and fre-quencies. LTAs, however, have a duty to secure the provision of public transport services that they consider appropriate to meet social needs that would not be available on a purely commercial basis.6

2.5 In order to operate a local bus service in Scotland, a bus operator must apply to the Traffic Commissioner for Scotland (TCS) to obtain a public service vehicle (PSV) operator’s licence7,8 and to register the local bus service. To register a local bus service in Scotland, an operator must provide 14 days’ notice to the Public Transport Authority (PTA)9 and a further 56 days’ notice10 to the TCS. If an operator wishes to vary or cancel a service, the operator must likewise give 56 days’ notice to the TCS and 14 to the PTA. However, an operator cannot terminate a service within 90 days of its start.11,12

2.6 The TCS has the power to take action if an operator no longer meets the conditions of its licence or does not operate services in line with the registration that it made. Between 2008 and 2012, the TCS held 34 completed public inquiries leading to 40 distinct actions being taken, including nine licence revocations.

Statutory quality bus partnerships

2.7 LTAs in Scotland can enter into partnership with local bus operators in order to deliver local bus policy. Under such partnerships the LTA, local authority and bus

- - - - - - - - - - 3 The key legislation governing transport in Scotland is: The Transport Act 1985, Transport (Scotland) Act 2001 and the Transport (Scotland) Act 2005. 4 As the RTP for the West of Scotland, SPT is responsible for planning and delivering transport solutions in its area. It is the owner of the statutory regional transport strategy, but relies on a range of public and private bodies to help it deliver the strategy. As part of its role, SPT monitors standards on its network through its bus inspectors and maintains bus shelters and bus stops on behalf of the councils. 5 In the rest of Scotland, local authorities act as LTAs. 6 We refer to such services as ‘supported services’. 7 The three Acts that govern PSV operator licensing are: Public Passenger Vehicles Act 1981, Transport Act 1985, and Deregulation and Contracting Out Act 1994. The 1981 Act is the primary legislation for PSV operator licensing, but it has been amended by several sections of the Transport Act 1985 and by sections 58–69 and Schedule 14 of the Deregulation and Contracting Out Act 1994. 8 To obtain a PSV licence, a bus operator must be of good repute, have appropriate financial standing, and be professionally competent. 9 SPT is the relevant PTA for Renfrewshire and East Renfrewshire. It told us that the 14 days’ notice allowed the PTA to assess practical aspects of the registration such as stance and stop allocation, whether there were any traffic restrictions which would prevent the route being run as applied for and whether the operator had sufficient vehicles to meet the application’s commit-ments. 10 In special circumstances the TCS can use discretion to accept a shorter period of notice. 11A service can have an earlier termination/change date at the discretion of the TCS, eg if the route is included in a batch of services to which the 56 days’ notice period applies. One aim of the 90-day minimum operation period is to ensure that all registration changes reflect sustainable local bus services and avoid the public facing an unexpected withdrawal of a service. 12 The interaction of the 56- and 90-day notice periods means that an operator must run a new service for a minimum of 34 days before it can put in an application to alter the service.

7

operators enter into agreements to improve the quality of local services in the area, or to reduce traffic congestion, noise or air pollution. SQPs13 differ from voluntary agreements in that only participating operators are allowed to use the scheme facili-ties. An SQP must be open to all operators who meet the required quality standards. Compliance is enforced through the TCS.

2.8 SPT told us that the history of Paisley’s bus market was of a large number of small operators, low-quality vehicles and poor standards of customer service. Paisley town centre was highly congested, leading to environmental issues, high numbers of inci-dents of stand blocking and breaches of bus operator regulations. The TCS similarly told us that there had been significant problems with operator behaviour in Paisley. These issues led to Scotland’s first SQP being introduced in Paisley in 2011.

2.9 The Paisley SQP came into force on 7 March 2011 and will run for five years. The scheme sets minimum standards with which each operator must comply for bus journeys within the relevant area. These standards are shown in Appendix B, Table 5. The majority of McGill’s routes are affected by the Paisley SQP.

2.10 The Glasgow SQP came into force on 1 April 2012 and will run for seven years. The scheme covers the city centre and some 13 of the main arterial roads into the centre from the city boundary (‘Streamline’ corridors). The Glasgow SQP standards are set out in Appendix B, Table7. Seven of McGill’s routes are affected by the Glasgow SQP.14

BSOG and concessionary fares

2.11 BSOG is paid to all operators of registered local bus services. It is a subsidy that reduces mileage costs incurred by local bus operators. In its market investigation into local bus services, the CC found that the payment of BSOG might allow greater fre-quencies on some routes and more chance for marginal routes to run on a commer-cial basis. Therefore, a reduction in BSOG was likely to have the effect of making some previously commercial services unviable, particularly those with relatively high fuel consumption.15

2.12 Until 1 April 2010, BSOG in Scotland was linked to fuel duty, offsetting around 80 per cent of the fuel duty paid by operators. From 1 April 2010, payments were calculated on the eligible kilometres run on local bus services (including dead mileage16), the total volume of fuel used and a predetermined payment rate set by Transport Scotland.17 Since April 2012, payments have been made on the basis of a rate per eligible kilometre (excluding dead mileage). There is no link to fuel used or fuel duty.18 The budget for BSOG in Scotland was £66.5 million for 2010/11 and £60.3 million in 2011/12. The budget for 2012/13 is £50 million.

2.13 The National Bus Travel Concession Scheme for Older and Disabled Persons was introduced in Scotland on 1 April 2006.19 From 8 January 2007, Scotland has also provided a national bus and rail travel concession scheme for young people (National

13 SQPs were introduced in Scotland under the Transport (Scotland) Act 2001. 14 McGill’s routes affected by the Glasgow SQP are: 3; 7; X6; X7 and X7A; 23 and 23A; 38 and 38A; and 26. 15 Local bus services market investigation report, paragraphs 12.21 & 12.22. 16 Dead mileage is the distance a bus runs without earning revenue between a depot and a pick-up point for passengers. 17 An additional incentive was introduced for low-carbon vehicles. 18 This rate is changeable by Transport Scotland on three months’ notice. 19 Part 3 of the Transport Scotland Act 2005. The scheme permits free bus travel within Scotland on any registered bus service to anyone who lives in Scotland, and is aged 60 and over, or meets certain disability criteria, and has applied for and received a National Entitlement Card. Local authorities are responsible for issuing the National Entitlement Card.

8

Concessionary Travel Scheme for Young People).20 Prior to April 2010, operators were reimbursed for concessionary passengers at 73.6 per cent. This was then reduced to 67 per cent. The scheme has a capped level of funding (£180 million in 2011/12 and £187 million for 2012/13). In 2011/12 the level of claims exceeded the cap, resulting in operators being reimbursed at less than their claim level.21

Disability Discrimination Act

2.14 DDA legislation affects all operators of local bus services. It requires all buses to have low-floor access and comes into effect between 1 January 2015 and 1 January 2017 (depending on whether the bus is single or double decked).22

The companies

McGill’s

2.15 McGill’s is a family-owned bus operator that provides local bus services in Inverclyde, Renfrewshire and East Renfrewshire (see Figure 1), with some services running into Glasgow. Prior to the acquisition, McGill’s derived virtually all (over 99 per cent) of its revenue from commercial services.

2.16 McGill’s was originally established in 1933. It was acquired by Arriva in 1997. In 2001, Alex Kean bought the McGill’s company from ASW, together with ASW’s oper-ation in Greenock, Inverclyde, as Arriva withdrew from Greenock to concentrate on its network in Renfrewshire and East Renfrewshire. In 2005, McGill’s was acquired by Arranglen.23 Between 2005 and 2008, McGill’s introduced services connecting Inverclyde and Glasgow. In 2009, it started to run services in the Renfrewshire area, taking over four services from an operator whose licence was revoked.24 It continued to expand in Renfrewshire and East Renfrewshire through further acquisitions of services from other operators25 and growing these services by increasing individual bus capacity and frequency.

2.17 In 2011, McGill’s had around 160 buses, of which approximately 80 were based at its Barrhead depot, East Renfrewshire, with the remainder at its depot at Greenock, Inverclyde. McGill’s told us that its fleet had an average age of eight years and fully complied with the requirements of the Paisley and Glasgow SQPs and the require-ments of the DDA.

2.18 In the financial year to 31 December 2010, McGill’s reported revenues of £11.1 million (2009: £10.1 million) and profit before tax of £0.3 million (2009: £0.5 million). Further details of McGill’s financial performance are set out in Appendix C.

20 The Scheme provides all 16- to 18-year-olds and young full-time volunteers up to the age of 25, who are resident in Scotland, a discount of one-third off the adult single bus fare on local registered buses and long-distance scheduled bus journeys throughout Scotland. 21 [] 22 The deadlines for low-floor access compliance are: single decks weighing less than 7.5T, 1 January 2015; single decks weighing more than 7.5T, 1 January 2016; and double decks, 1 January 2017. 23Arranglen is a private company. Mr James Easdale owns 99.98 per cent of the shares in Arranglen. McGill’s is the largest of Arranglen’s subsidiaries by turnover (2010). 24 McGill’s entered on four routes when First Stop Travel Ltd’s licence was revoked. In addition, in 2010 and 2011 McGill’s acquired: Routes 21 and 68 from Gibson Direct; Route 61 from Travel Direct; Route 26 from Fairway Coaches; Route 68 from Phoenix Buses; Route 64 from Fereneze Travel; and Route 38 from Dickson’s of Erskine. 25 McGill’s told us that these operators would have withdrawn their services from the routes because they would not have upgraded their buses to meet the requirements of the Paisley SQP.

9

2.19 McGill’s Managing Director is Mr Ralph Roberts. He joined McGill’s from ASW in 2010 where he had been Managing Director.

2.20 McGill’s told us that its aim was to become one of the major operators in the West of Scotland with a fleet of [] buses. Its strategy has been to grow its business through the acquisition of services from smaller operators and then increase the passenger numbers and revenue on those routes by providing a higher-quality, customer-focused service (including better reliability and punctuality, better customer service, newer and cleaner buses and better-value tickets). McGill’s noted that it might take several years for a service to achieve its full potential as it took time for passenger demand to build up. McGill’s attributed approximately [] per cent of its growth to acquisitions and [] per cent to organic growth achieved on the acquired routes post-acquisition.

2.21 We note that McGill’s does not have a culture of formal, written communication or decision making.

Arriva Scotland West

2.22 At the time of the acquisition, ASW was a subsidiary of Arriva plc whose ultimate holding company is Deutsche Bahn AG (DB). DB has transport operations in 130 countries and 291,000 employees and in 2011 generated revenue of €38 billion. Within DB, Arriva is the business division responsible for all regional passenger transport outside Germany. Arriva generated revenue of €3 billion in 2011. In the UK, it is the third largest regional bus operator. ASW was Arriva’s only operation in Scotland and provided local bus services in Renfrewshire and East Renfrewshire (see Figure 2) with some services running into Glasgow. ASW derived the vast majority (over 95 per cent) of its revenue from commercial services.

2.23 ASW’s operations were originally part of the state-owned Scottish Bus Group. When the Scottish Bus Group was privatized in 1991, the operations in the Clydeside area (covering Inverclyde, the Isle of Bute and the south side of Glasgow) were bought by the management and employees. These were subsequently sold to British Bus Company, which itself was acquired by Arriva (formerly known as the Cowie Group) in 1996. Arriva was acquired by DB in August 2010.

2.24 At the time of the acquisition, ASW had 153 buses located at two depots: Johnstone and Inchinnan. The average age of the fleet was around [] years.26 Arriva told us that not all its buses met the requirements of the Paisley and Glasgow SQPs. However, during 2010 and 2011 it transferred buses to ASW from other Arriva oper-ations in order to meet these requirements and additional buses were planned to be transferred in 2012 had ASW not been sold. In addition, some buses would have been refurbished and upgraded as necessary to meet the requirements of the DDA.

2.25 In the financial year to December 2011, ASW generated revenue of £16.2 million27 (2010: £18.0 million) and a loss before tax of £0.3 million (2010: profit £0.4 million). Arriva told us that the decrease in revenue in 2011 resulted mainly from the loss of the Glasgow Flyer service concession from Glasgow city centre to Glasgow Airport in July 2011. Further details of ASW’s financial performance are set out in Appendix C.

2.26 ASW was the smallest and least profitable business within Arriva’s UK Bus division. Prior to the sale of ASW to McGill’s, Arriva had been seeking, unsuccessfully, to

26 The average age of Arriva’s UK bus fleet was [] years. 27 Based on 2011 draft accounts

10

expand the business by acquisition, and, at the same time, to restructure its oper-ations to improve its profitability.

3. The acquisition

3.1 In this section, we set out the events leading up to the acquisition, each party’s stated rationale for the transaction and details of the acquisition.

The acquisition process

3.2 McGill’s first approached Arriva in 2007 about acquiring ASW. Arriva told McGill’s that it had no plans to sell ASW. McGill’s made further approaches to Arriva in 2009 and 2010. In 2010, Arriva advised McGill’s that, as a result of the anticipated acqui-sition of Arriva by DB, it was not in a position to discuss McGill’s approach for ASW. In February 2011, McGill’s again approached Arriva and made an indicative offer for ASW. Following negotiations, the parties reached an agreement in principle in August 2011. Contracts were exchanged in December 2011 and the acquisition was completed on 26 March 2012 (see paragraphs 3.10 and 3.11).

3.3 Arriva told us that it had discussions with Stagecoach in 2007 and 2011 as to whether it might be interested in acquiring ASW in exchange for another part of its business, but those discussions did not progress beyond a preliminary stage. Arriva told us that it did not solicit any other offers for ASW as it was focused on improving ASW’s profitability; nor did it receive any other offers.

Rationale for the transaction

3.4 McGill’s told us that the acquisition of ASW was an attractive opportunity to increase its operations in Renfrewshire and that it would follow the same strategy as it had with previous smaller acquisitions (see paragraph 2.20) to improve the quality of the acquired services and increase passenger numbers and revenue. As discussed in paragraph 3.2, it had approached Arriva on several occasions about acquiring ASW.

3.5 McGill’s believed that ASW was under significant financial pressure. It said that ASW needed to invest in its fleet to meet local regulation and national legislation changes. It was facing reductions in BSOG and concessionary fares income, and, due to the general economic conditions, reductions in bus patronage, as well as higher fuel costs. McGill’s said that it thought that ASW would exit the Renfrewshire area28 in around 18 months either through a sale or through closure of its operations.

3.6 McGill’s told us that it was concerned that if ASW exited through closure it could lead to a loss of patronage on ASW’s routes as passengers experienced travel disruption. If, on the other hand, ASW withdrew incrementally from its routes, McGill’s would have sufficient notice to be able to increase its bus fleet and recruit drivers to take over these routes, reducing travel disruption and potential patronage loss.

3.7 McGill’s considered that acquiring ASW was the lower-risk option as it would enable it to expand its fleet and operations quickly and retain patronage on ASW’s routes.

3.8 McGill’s based its valuation of ASW on []. McGill’s told us that it expected to improve profitability through growth in passenger numbers. More passengers would

28 The term ‘Renfrewshire area’ covers the areas where the merged parties overlapped pre-acquisition and may include portions of East Renfrewshire, Glasgow City Centre and Inverclyde.

11

be attracted to McGill’s services as a result of a more reliable service, better cus-tomer service, newer and cleaner buses and better-value tickets than ASW. It would also achieve cost savings from the increased scale of the combined business and the rationalization of services following the acquisition.

3.9 Arriva told us that, while it had not been planning to sell ASW, the offer from McGill’s was attractive in comparison with continuing to own ASW given the relatively low level of profitability of the ASW business and the level of capital employed. Arriva told us that there were several factors that led to this conclusion:

(a) ASW was the smallest business within the UK Bus division and produced the lowest returns. In the financial year ended 31 December 2009, Arriva UK Bus generated total revenues of £961.5 million and achieved an operating profit margin (before goodwill, exceptional items and any group central cost allocations) of 9.5 per cent.29 By comparison, in the same period ASW generated revenues of £17.9 million and achieved an operating profit margin (before goodwill amortiz-ation and exceptional items) of 0.9 per cent (group central cost allocations not disclosed).30

(b) Arriva expected the financial pressure on ASW to increase as a result of austerity measures in the UK, the forthcoming reductions in BSOG and concessionary fares reimbursement and the investment required in the fleet to comply with the requirements of the DDA.

(c) The offer from McGill’s included £[] and realized a gain more easily than could have been achieved by continuing operations.

(d) The sale enabled Arriva to focus more attention on its other, larger, bus oper-ations.

Details of the transaction

3.10 The acquisition was completed on 26 March 2012. McGill’s acquired the business undertaken by ASW and specified assets including vehicles and depots (see para-graph 4.4).31

3.11 The purchase price was £[] million. £[] million was paid on completion and the remaining £[] million was deferred and is payable over [].

4. The relevant merger situation

4.1 In this section, we discuss the relevant merger situation. Further details are set out in Appendix D.

4.2 Under the Act, and pursuant to our terms of reference (see Appendix A), we must investigate and report on whether there is a relevant merger situation.32 If so, we must go on to consider whether the creation of that relevant merger situation has resulted or may be expected to result in a substantial lessening of competition within any market or markets in the UK for goods or services.33

29 Bus market investigation report, Appendix 3.1. 30 ASW statutory accounts for year ended 31 December 2009. 31 McGill’s did not acquire the ASW corporate entity. 32 Sections 23 & 35(1)(a) of the Act. 33 Section 35(1)(b) of the Act.

12

4.3 The concept of a relevant merger situation has two principal elements, set out in section 23 of the Act: first, that the transaction structure is one that involves ‘enter-prises ceasing to be distinct’, and second, that either the ‘turnover test’ or the ‘share of supply test’ is satisfied.

4.4 Under clauses 2 and 3 of the Business Sale Agreement dated 13 December 2011, McGill’s has acquired the following from ASW: [].

4.5 We therefore conclude that the sale and purchase arrangements concerned consti-tutes ‘enterprises ceasing to be distinct’, with operations that were previously separ-ate now being under common ownership.

4.6 The turnover test will apply where the value of the turnover in the UK of the ‘enter-prise being taken over’ exceeds £70 million. The turnover of ASW in the year ended 31 December 2011 was £16.2 million (see paragraph 2.25) and so the turnover test in the Act is not met. We therefore considered the share of supply test.

4.7 The share of supply test applies where, as a result of enterprises ceasing to be distinct, either (a) the merged entity will supply or acquire 25 per cent or more of any goods or services in the UK or a substantial part of the UK, or (b) an existing share of supply of 25 per cent or more will be enlarged.34

4.8 In applying the share of supply test, we considered whether it was satisfied on a national basis or in a substantial part of the UK.

4.9 McGill’s and ASW operate in the West of Scotland. Though the parties are both active in Renfrewshire and East Renfrewshire, the majority of their service overlaps are in Renfrewshire, the reference area. We considered whether the share of supply test was met in a ‘substantial part of the UK’.

4.10 We examined whether the reference area of Renfrewshire could be considered a ‘substantial part of the UK’ within the meaning of the Act. In the House of Lords judgment in R v MMC and another ex parte South Yorkshire Ltd35 (ex parte South Yorkshire), it was held that for a given area to be a substantial part of the UK it must be of such ‘size, character and importance as to make it worthy of consideration’ for merger control purposes.

4.11 McGill’s noted that the land area and population of Renfrewshire were much smaller than that under consideration in ex parte South Yorkshire. Consequently, McGill’s said that some further feature was required for Renfrewshire to be deemed a ‘sub-stantial part of the UK’. McGill’s also said that Renfrewshire was part of the wider Glasgow conurbation rather than a distinct or discrete local market.

4.12 Our view is that the reference area satisfies the criteria adopted in ex parte South Yorkshire and that our approach is consistent with previous CC merger decisions. In our view, Renfrewshire is significant both in its own terms and as part of the wider Glasgow conurbation. Paisley, the administrative centre of Renfrewshire, is 6 miles from Glasgow city centre, with both being important bus destinations. Further, although Renfrewshire is smaller in terms of population and land area than the corresponding figures in ex parte South Yorkshire, it is greater than those in other CC merger decisions, with a population of 170,000 and a land area of 195 sq km.

34 Section23 of the Act. 35 [1993] 1 WLR 23.

13

4.13 In addition, Lord Mustill in ex parte South Yorkshire was of the opinion that the local nature of bus services and their importance to the public was highly relevant. We are of the view that the local bus market in Renfrewshire is of sufficient importance to the Renfrewshire area to merit consideration.36

4.14 We conclude that Renfrewshire is a ‘substantial part of the UK’.

4.15 McGill’s has over 69 per cent of the share of supply of local bus services in Renfrewshire as a result of the acquisition. We conclude that the share of supply test is met and that a relevant merger situation has been created.

5. The relevant market

5.1 In this section, we set out our findings on the relevant market, drawing on the evidence and findings from the CC’s local bus services market investigation and recent merger inquiries. Further details on market definition are set out in Appendix E.

5.2 The purpose of market definition in merger analysis is to provide a framework for the analysis of competitive effects.37 Regardless of our chosen market definition, we take into account broader competitive constraints as appropriate when undertaking our competitive assessment.

5.3 There are normally two dimensions to the relevant market: a product dimension and a geographic dimension. The relevant product market is a set of products that customers consider to be close substitutes. In the case of local bus services, for example, this will reflect different aspects of the offer such as fares, frequencies, capacity and quality, as well the extent to which different modes of transport are substitutes for passengers. The relevant geographic market may, in general terms, be local, regional, national or wider.

5.4 In determining the relevant market, we drew on the evidence and findings from the CC’s local bus market services investigation38 which included Scotland, as well as the findings in the CC’s two most recent merger inquires involving local bus oper-ators: Stagecoach Group Plc/Preston Bus Limited (Stagecoach/Preston)39 and Stagecoach Group Plc/Eastbourne Buses Limited (Stagecoach/Eastbourne).40

5.5 In the local bus services market investigation, the CC concluded that in the short run flexible transport modes such as the car do not constrain bus operators. It concluded that the relevant product market, in certain circumstances, should be widened to include fixed modes of travel such as rail if they offered a sufficiently similar journey proposition to present a viable alternative for passengers.

5.6 In determining the relevant geographic market, the local bus services market investi-gation found no evidence to suggest that customers would substitute to another flow41 in response to a small worsening of an operator’s offer on a particular flow. This suggested that geographic markets should be defined at the level of an individ-ual flow. The local bus services market investigation found that operators with exist-

36 See Appendix D, paragraphs 28–32. 37 CC Merger guidelines, CC2, section 5.2. 38 See www.competition-commission.org.uk/our-work/local-bus-services/final-report-and-appendices-glossary. The scope of the market investigation excluded Northern Ireland and London. 39 See www.competition-commission.org.uk/our-work/directory-of-all-inquiries/stagecoach-preston. 40 See www.competition-commission.org.uk/our-work/directory-of-all-inquiries/stagecoach-group-plc-eastbourne-buses-limited. 41 A flow is any journey or connection between two specified points, say A to B. Flows may correspond to journeys on a route (eg between bus stops).

http://www.competition-commission.org.uk/assets/competitioncommission/docs/pdf/non-inquiry/rep_pub/rules_and_guide/pdf/100916_merger_assessment_guidelines.pdf#5.2.1

http://www.competition-commission.org.uk/our-work/local-bus-services/final-report-and-appendices-glossary

http://www.competition-commission.org.uk/our-work/directory-of-all-inquiries/stagecoach-preston

http://www.competition-commission.org.uk/our-work/directory-of-all-inquiries/stagecoach-group-plc-eastbourne-buses-limited

14

ing services and facilities within approximately a 30-minute drive-time from a flow may constrain incumbent operators. We discuss potential competition in the context of this inquiry in our competitive assessment in Section 7.

5.7 In both Stagecoach/Preston42 and Stagecoach/Eastbourne,43 the CC concluded that:

(a) The appropriate market definition was at the flow level including:

(i) flows which had both their origin and destination within 500m of each other;44 and

(ii) services which had a similar level of frequencies.

(b) Flexible modes—ie car, taxi—should not be included.45

(c) Further segmentation according to different types of customers was not neces-sary.46

(d) The markets for commercial and tendered services were distinct.

5.8 McGill’s told us that, in its view, the market should be much wider. On the product side, McGill’s said that competition from cars should not be ignored and that rail was likely to provide a competitive constraint. McGill’s also argued that the geographic market should take into account all potential flows that could take place from depots located within the Glasgow urban area.47

5.9 We explored whether there were particular characteristics of the Renfrewshire area that suggested that we should alter the market definition established in previous local bus inquiries.48

5.10 We first explored whether consumers’ characteristics and behaviours in the Renfrewshire area showed any difference from the rest of Great Britain in terms of:

(a) the distance travelled to the place of work or study;

(b) the relative proportions of different modes of transport used to travel to the place of work or study;

(c) the distance travelled by bus or minibus; and

(d) the degree of car ownership. 42 Stagecoach/Preston, paragraph 7.20. 43 Stagecoach/Eastbourne, paragraphs 5.30–5.33. 44 See, for instance, Stagecoach/Preston, Appendix I, paragraphs 8–11. 45 Although the demand for car and bus travel interact, the CC’s view is that any constraint from the car on bus travel will only be felt over the long term and following substantial changes in consumers’ travel habits or disposable income rather than following a small worsening of the competitive offer. 46 In Stagecoach/Eastbourne, the CC noted that since competitive constraints were similar for all groups, it seemed pragmatic to aggregate across consumer groups. It also noted that doing so was unlikely to change its conclusions because none of those subgroups appeared likely to switch away from bus travel to any other form of transport following a small worsening of the offer. (Appendix I, paragraph 78.) 47 The CC’s local bus services market investigation defined, in broad terms, an urban area as a town or city and contiguous suburbs. As part of its analysis, the CC defined a set of 239 Urban Areas each corresponding to a city or town (or a group of nearby cities or towns) and any nearby suburbs with highly interconnected bus networks. The Glasgow Urban Area is shown in the local bus market services investigation, Appendix 4.2, p9. 48 We are also aware that there are important differences in approach between a market investigation and a merger inquiry. For example, while in a market investigation covering the UK it made sense to use a pragmatic approach and examine competition at a relatively aggregated level—eg assessing the competitive conditions at the level of the Glasgow Urban Area—this is less appropriate for a merger inquiry on a merger between bus operators providing services in a well-defined area. The findings and conclusions of the market investigation therefore need to be interpreted in the context of a merger inquiry.

15

5.11 We found that there was no significant difference with regard to the distances travelled by bus in Renfrewshire (and East Renfrewshire)49 and the rest of Great Britain, although a higher proportion of consumers travelled very short distances (less than 2 km) in Renfrewshire than in England and Wales. We also found that, based on car ownership, fewer consumers have the option to travel by car in Renfrewshire (but not in East Renfrewshire) than in the rest of Great Britain. This was reflected in the higher proportion of consumers who travelled to work by bus in Renfrewshire than in England and Wales (and to a lesser extent Scotland).

5.12 Overall, we formed the view that although there were some differences between the areas affected by this acquisition and the rest of Great Britain, these did not suggest that we should alter the broad conclusions on market definition established in pre-vious local bus inquiries. Our competitive assessment incorporates considerations of any local specific features that are relevant to this acquisition.

5.13 We conclude, therefore, that our market definition is as follows:

(a) Our starting point for market definition is at the flow level.50 The market includes flows which have both their origins and destination within 500m of each other and that show a broadly similar level of frequency and frequency distribution over time.

(b) Flexible modes of transport (including the car) are not included in the product market definition. We consider whether rail should be included as a constraint for particular overlap flows.

(c) There are separate markets for commercial and supported services.51

5.14 We identified a separate geographic market for each flow in which McGill’s and ASW competed pre-acquisition. In our competitive assessment, we considered the extent of head-to-head competition at the flow level, the constraint from the threat of expan-sion by those operators that are already present on the flow and the threat of entry by operators not on the flow with nearby facilities.

6. The counterfactual

6.1 In carrying out our competitive assessment, we compare the prospects for compe-tition with the acquisition against the competitive situation without the acquisition. The latter is called the ‘counterfactual’.52

6.2 In this section, we first set out the views of parties on the counterfactual regarding ASW. We then discuss our assessment of the counterfactual before reaching our conclusion.

The views of parties

6.3 Arriva told us that historically it operated from six depots in Barrhead, Cumbernauld, Greenock, Inchinnan, Johnstone and Paisley. It ceased trading from Barrhead in

49 We included both Renfrewshire and East Renfrewshire in this comparison. However, given that the majority of flow overlaps between ASW and McGill’s are in Renfrewshire, we have assigned greater weight to the relative characteristics of this area. 50 We have some evidence that a proportion of consumers made network journeys (see Appendix G). However, given the scope of the flow overlaps between McGill’s and ASW pre-acquisition, we did not find it necessary to conclude on this point. 51 Given the small proportion of revenues that the parties obtain from supported services, we did not look at supported services in detail in this merger inquiry. 52 CC2, paragraph 4.3.1.


16

2000, and from Greenock in 2002 after its Inverclyde services were sold to McGill’s. From 2005, ASW operated only from its depots in Inchinnan and Johnstone.

6.4 Arriva told us that ASW’s profit margin had declined sharply since 2009 and was significantly lower than for its UK Bus division as a whole. The offer from McGill’s was therefore attractive in comparison with continuing to own ASW (see paragraph 3.9). If ASW had not been sold to McGill’s, Arriva told us that:

(a) It would have expected to continue the ASW operations, seeking to improve profitability through a continued focus on increasing revenue and reducing costs.

(b) If ASW had not succeeded in making an operating profit, it would have thought more actively about a sale (rather than closure). However, prior to the acquisition, it had not taken the decision to exit and had not sought to market ASW. With the exception of Stagecoach, it believed there were no other large bus operators (such as [] or []) or any medium or small operators that would be likely pur-chasers for ASW. It thought that Stagecoach would have been unlikely to make an attractive offer given the level of the indicative offer received in 2011.

(c) Despite the low forecast levels of profitability for ASW, exiting the business through closure was likely to be less financially attractive to it than a sale, owing to the costs, particularly redundancy payments,53 that would have been incurred. There would also have been a reputational risk to closure given that Arriva had never previously closed one of its UK bus operations.

6.5 McGill’s said54 that, although ASW was not an exiting firm as defined in the CC’s Merger Assessment Guidelines,55 ASW was in decline and would have substantially reduced the frequency of its off-peak services and exited the market in the medium term (around 18 months) (see paragraph 3.5). It said that closure was not an insuper-able problem for Arriva; ASW had declined substantially in the previous ten years from having a fleet of 475 buses run out of eight depots to having 153 buses run out of two depots and it had previously sold operations in Inverclyde and Glasgow.

6.6 Several third parties told us that ASW had an ageing fleet. One small operator said that ASW had seemed to stop trying to compete over the last couple of years. SPT told us that it had the impression that Arriva did not intend to be in the area in the long run. The TCS told us that she was not surprised that ASW had exited. She was not sure how sustainable ASW’s operations were because it had constraints which more local, smaller and medium-sized companies did not have. On the other hand, FirstGroup described ASW as a reasonably large professional organization which had not looked as if it was a failing company.

Our assessment

6.7 We considered whether ASW met the conditions of an exiting firm set out in the Merger Assessment Guidelines. We looked at whether ASW would have exited (through failure or otherwise) and, if so, whether there would have been an alterna-tive purchaser to McGill’s. We also considered what might have happened to McGill’s in the absence of the acquisition.

53 Arriva estimated the redundancy costs at £[]. 54 Submission 16 May 2012, paragraph 6.4. 55 CC2, paragraphs 4.3.8–4.3.18.

http://www.competition-commission.org.uk/assets/competitioncommission/docs/2012/mcgills-arriva-west-scotland-inquiry/mcgills_initial_submission_non_confidential.pdf



17

6.8 As noted in paragraphs 2.26 and 3.9, ASW was the smallest business within Arriva’s UK Bus division and produced the lowest returns. Given the low returns, Arriva sought to minimize the capital employed in ASW. There was limited investment in ASW’s bus fleet as the policy with Arriva’s UK Bus division was to introduce new buses in the most profitable areas with older buses being transferred to less profit-able areas. Consequently the average age of ASW’s fleet was higher than for Arriva’s UK Bus division as a whole (see paragraph 2.24).

6.9 We reviewed the financial performance of ASW over the five years to December 2011 as reported in its statutory accounts. The profit and loss account is summarized in Appendix C, Table 5. ASW’s revenues increased from £17.2 million in 2007 to £18.3 million in 2008 but fell to £18.0 million in 2009 and declined further to £16.2 million in 2011. The operating profit margin increased from 7.3 per cent in 2007 to 8.6 per cent in 2008 but fell to only 0.9 per cent in 2009. After a recovery to 2.4 per cent in 2010, operating profit margin declined further to a loss of 1.7 per cent in 2011.

6.10 Arriva said that, between 2009 and 2011, ASW took several steps to reduce costs and improve revenue including, in particular, moving the back-office functions to regional level, optimizing the route network and restructuring drivers’ working patterns. Arriva told us that the network had been rationalized to three route groups running into Glasgow, supported by two depots. It had been planning a review of ticket types and fare structure to simplify the passenger offering and optimize revenue in 2012.

6.11 ASW’s budget for 2012 shows a return to profitability compared with the small loss in 2011, although the projected operating profit margin was forecast to remain low (at 2.3 per cent). We note that a relatively small cost overrun or drop in revenue would have led to ASW making a loss in 2012 for the second year running.

6.12 Arriva told us that it did not envisage making any cuts to the routes or services that ASW operated given its optimization of the network in January and November 2011, which had structured its network around three route ‘groups’. Arriva said that although some routes were not profitable, ASW would continue to operate those routes as they served as feeders to other, profitable, routes. Arriva also told us that there was no scope to reduce the number of route groups from three to two as the revenues and profits from the remaining routes would not have been sufficient for the business to be viable.

6.13 The high-level medium-term plan prepared by Arriva as part of DB’s financial plan-ning process projected that ASW would achieve a modest rise in revenue and oper-ating profits (see Appendix C, paragraph 32), but would not return to the modest level of operating profits achieved in 2010 until 2015. The plan showed that ASW was forecast to continue to make substantially lower returns than the profit margins for Arriva’s UK bus division.56

6.14 We also considered whether Arriva would have been more likely to have sold ASW to another bus operator than retain it and continue it as an ongoing business. We noted Arriva’s view that there were no other likely purchasers of ASW (see paragraph 6.4(b)). We considered whether either Stagecoach or FirstGroup might have been interested in acquiring ASW. Arriva told us that it would not have reopened discus-sions with Stagecoach if the sale to McGill’s had not proceeded given the level of the indicative offer from Stagecoach in 2011. FirstGroup told us that it had not been

56Arriva told us that the medium-term plan was prepared on the same basis as the management accounts and was an extrapo-lation of the 2012 budget. It assumed that ASW would continue to operate at a similar scale as it had been forecast to do in 2012.

18

invited by Arriva to make an offer for ASW and that it would not have been interested in doing so.

6.15 We also considered what might have happened to McGill’s in the absence of the acquisition. We note that McGill’s profitability had been declining for several years; it too was affected by changes in the regulatory environment, including reductions in BSOG and concessionary fares reimbursement, and also by rising costs. We also note that in 2011, []. McGill’s told us that, in the absence of the acquisition, it would have reduced its operation in Renfrewshire by 18 buses and cut frequencies on Sundays, off-peak periods and evenings to cut costs.57 It also told us that, had ASW been sold to another operator which was, in its view, committed to operating in the Renfrewshire area in the longer term, it would have been likely, at a minimum, to have reduced its operations and might have decided to exit the Renfrewshire area altogether.

Conclusion on the counterfactual

6.16 We conclude that, on balance, the most likely counterfactual is one of McGill’s and ASW continuing to operate commercial bus services in the Renfrewshire area. We found that the sale of ASW was less likely than its continued operation given that, other than preliminary discussions with Stagecoach, Arriva did not receive or solicit any other offers for ASW and there did not appear to be other likely purchasers in the future. We also decided, on balance, that, despite ASW’s poor levels of profitability, it was less likely that Arriva would exit the business by closure than continue its oper-ation, given, in particular, the reputational and redundancy costs that would have been incurred.

6.17 We note, however, that McGill’s and ASW’s profitability had been declining for several years. Both McGill’s and ASW were affected by changes in the regulatory environment (particularly through changes to BSOG and the cap in concessionary fares reimbursement) and rising costs. Arriva told us that it could not reduce the size of ASW’s operations any further; it looked unlikely to deliver more than a very small profit in 2012 and it was not forecast to return to the levels of profitability seen across the Arriva UK bus business as a whole. McGill’s told us that, in the absence of the acquisition, it would have been likely to have reduced its services in Renfrewshire. Our view, therefore, is that, whilst the most likely scenario is that both ASW and McGill’s would have continued to operate in the Renfrewshire area, competition between them would have lessened substantially over the foreseeable future. We discuss the sustainability of McGill’s and ASW’s competitive interaction further in our competitive assessment (see paragraphs 7.48 to 7.52).

7. The competitive assessment

7.1 In order to assess the competitive effects of the acquisition, we have looked at the following:

(a) the existing and potential operators in the Renfrewshire area, including an over-view of their networks (alongside those of McGill’s and ASW) (see paragraphs 7.2 to 7.23);

(b) the overlap flows between McGill’s and ASW (see paragraphs 7.24 to 7.29);

57 We note that, as set out in paragraph 6.12, Arriva told us that, if it reduced its network further, its business would not be viable.

19

(c) levels of competition between McGill’s and ASW pre-acquisition and their likely development over time (see paragraphs 7.30 to 7.52);

(d) the impact of the acquisition on actual competition in the Renfrewshire area, including an assessment of entry (see paragraphs 7.53 to 7.115); and

(e) the impact of the acquisition on potential competition (see paragraphs 7.116 to 7.126).

Existing and potential operators in the Renfrewshire area

7.2 The bus services of McGill’s and ASW overlap primarily in Renfrewshire, as well as in parts of East Renfrewshire and Glasgow. We term this ‘the Renfrewshire area’.58

7.3 Renfrewshire itself has relatively low income and low car ownership compared with the rest of Great Britain. The demand for bus services in the Renfrewshire area is high and therefore offers good opportunities for bus operators. Following deregula-tion of the bus industry in 1985, up to 25 small independent operators delivered local bus services in the area.59

7.4 After deregulation, a large number of local bus operators in the Renfrewshire area, and in particular Paisley, were of a low quality with many providers using old, small minibuses. The high number of bus services also led to severe congestion in some parts of Paisley. As a direct consequence, SPT and Renfrewshire Council introduced a Local Traffic Order in Paisley town centre to reduce congestion and, subsequently, the Paisley SQP to improve vehicle standards and reduce emissions.60.Although, we were told, the investment required was not substantial,61 we were also told that the introduction of the Paisley SQP led directly to a number of small operators exiting, reducing their services, focusing on supported services or private hire or, more commonly, divesting their business, often to McGill’s.62

7.5 In paragraphs 2.11 to 2.13, we noted recent changes to how BSOG is calculated and the budgetary cap on concessionary fares reimbursement. We were told that the impact of such changes on the profitability of commercial bus services had been significant, especially on bus providers that operated in urban areas. Past fuel price increases63 also increased the costs of operating bus services. Operators told us that the market conditions had become much tougher compared with four or five years before. We were told by a number of bus operators that they would have to respond by either increasing fares, reducing frequencies and/or cutting some routes or portion of routes.64

7.6 Overall these changes are the most likely explanation for the consolidation in the bus services that has taken place in the Renfrewshire area over the last few years. Pre-acquisition, in addition to McGill’s and ASW, a number of services in the Renfrewshire area were run by First Glasgow, Stagecoach and a few small bus operators such as Riverside Transport, Colchri Coaches, Key Coaches and others. Based on weekly

58 See footnote to paragraph 3.5. 59 Deregulation was introduced as part of The Transport Act 1985. 60 See paragraphs 2.8 & 2.9. 61 We were told by Arriva that the investment required to make its buses compliant was not substantial and amounted to less than £5,000 depending on the type of bus. However, many small bus operators had small buses which we understand could not be made compliant. 62 McGill’s told us that between November 2010 and June 2011 it had acquired Gibson Direct, Travel Direct, Fairway Coaches, Phoenix Buses, Fereneze Travel and Dixons of Erskine in Renfrewshire. 63 We understand that this was a concern in the past, though it is unclear whether this will remain a concern in the near future. 64 [Some operators] told us that they were reacting to changes by attempting to reduce their reliance on routes which had a high proportion of concession passengers.

20

frequencies at 1 October 2011, ASW had the largest share (41 per cent) followed by McGill’s (28 per cent), First Glasgow (18 per cent), Riverside Transport (8 per cent), Colchri Coaches (1.5 per cent) and others (3.5 per cent).65 We describe these oper-ators and their networks as at 1 October 2011 in paragraphs 7.7 to 7.23.

McGill’s and ASW

7.7 Paragraphs 2.15 to 2.26 and Appendix C give an overview of the financial positions of McGill’s and ASW. Figures 1 and 2 show the extent of McGill’s and ASW’s bus networks pre-acquisition, respectively. The shading indicates the service frequency on the portion of the route.

7.8 Prior to 2009, McGill’s network operated from Inverclyde (including services to Greenock, Largs and Port Glasgow). From 2009, McGill’s expanded into Renfrewshire, primarily by taking over small bus operators. It ran an extensive network around Paisley with connections to Glasgow city centre, Barrhead, Renfrew, Erskine and Johnstone (see Figure 1).

7.9 McGill’s told us that, consistent with its culture (see paragraph 2.21), it did not under-take a detailed analysis of route profitability for all routes; it only did so when there was a performance issue on a particular route. McGill’s told us that certain of its smaller routes were not profitable.66

65 We calculated market shares according to weekly frequencies. As discussed in paragraph 7.27, we took McGill’s services as of 1 October 2011 as a base to identify all the significant overlaps with ASW and other operators. ‘Others’ includes Stagecoach, Key Coaches, Slaemuir Coaches, Harte Buses, Scottish Travel, Henderson Travel and Ayrways Coaches. 66 For example, it refers to its services [] as loss making.

21

FIGURE 1

McGill’s bus network and frequencies (as of 1 October 2011)

Source: CC’s elaboration on data from Traveline.

7.10 ASW’s network extended around Paisley with services linking Glasgow city centre, Paisley, Renfrew, Clydebank, Erskine, Barrhead, Largs and Johnstone (see Figure 2). ASW did not provide services in Inverclyde.

7.11 Arriva provided us with route-costing information for part of 2011. ASW’s route-costing analysis showed that most of its services, including those that faced compe-tition from McGill’s, made a positive profit contribution after the allocation of depot overheads.67 Arriva told us that some of its smaller services which did not make a positive profit contribution served as feeders to other, profitable, services.

67 The calculation of profit contribution after the allocation of depot overheads does not include the allocation of other overheads such as financing costs and goodwill amortization.

22

FIGURE 2

ASW’s bus network and frequencies (as of 1 October 2011)

Source: CC’s elaboration on data from Traveline.

FirstGroup

7.12 FirstGroup is a provider of public transport services in the UK, the Republic of Ireland, Continental Europe and North America. Its turnover for the year ended 31 March 2012 was £6.7 billion (2011: £6.4 billion) and operating profit was £429 million (2011: £457 million). In the UK, its bus division operates 8,000 vehicles and in the financial year ended 31 March 2012 generated turnover of £1,157 million (2011: £1,137 million) and an operating profit of £134 million (2011: £149 million), a margin of 11.6 per cent (2011: 13.1 per cent).68

7.13 At its annual results presentation in May 2012,69 FirstGroup described the perform-ance of its UK bus division as steady although passenger revenue growth was lower in Scotland and the North of England than elsewhere in its business. FirstGroup noted that despite facing challenging trading conditions as a result of deteriorating economic conditions, reduced government subsidies and funding to the industry, and higher fuel costs, it was focused on achieving growth. It said that its historic operating model, including the mileage reductions it had implemented since the 2008/09 finan-cial year in response to the economic downturn, had controlled costs but affected passenger growth. The revised strategy, led by the new management appointed during 2011, emphasized revenue growth which it aimed to achieve by focusing on

68 FirstGroup annual accounts, 2012. 69 Preliminary results presentation dated 23 May 2012.

23

network design, fares and ticketing, local market solutions, effective partnerships and service quality and delivery, as well as the disposal of selected operations in order to focus on those areas with the greatest potential for growth.70 In 2011,71 FirstGroup also announced that it would be investing £160 million in 955 buses over the follow-ing two years.

7.14 In the Glasgow and Renfrewshire areas, FirstGroup operates through its subsidiary First Glasgow (No. 1) Limited (First Glasgow). It has an extensive network in Glasgow with some services also in the Renfrewshire area (see Figure 3 and paragraphs 7.90 to 7.94). In the financial year ended 26 March 2011, First Glasgow reported turnover of £92.4 million (2010: £92.4 million) and an operating profit of £11.2 million (2010: £7.8 million), a margin of 12.1 per cent (2010: 8.4 per cent).72

FIGURE 3

First Glasgow’s network (as of 1 October 2011)

Source: CC’s elaboration on Traveline.

7.15 In our competitive assessment (see Appendix F), we identified the following First Glasgow services as overlapping with McGill’s pre-acquisition (as of 1 October 2011) in the Renfrewshire area: 747, 25/25A, 9/N9, 56/56A, 89, 103, 6, X8/X8A. First Glasgow’s service 9/N9 made up half of all its frequencies in the Renfrewshire area.

70 A subsequent press article reported that FirstGroup had earmarked seven provincial English operations for sale (see Passenger Transport Magazine, 28 June 2012). FirstGroup said that this was unsubstantiated. 71 Preliminary results presentation dated 11 May 2011. 72 First Glasgow (No. 1) Limited annual accounts 2011

24

Riverside Transport

7.16 Riverside Transport was established in 1994 and is based in Barrhead. It operates 15 vehicles and has seven routes in the Renfrewshire area (one of which is a sup-ported service). Riverside Transport told us that it relied substantially on income from concession passengers. On 16 July 2012, it started to operate two new routes []. Since the summer of 2011, it also cancelled or reduced frequencies on various services.73 Overall, between 1 October 2011 and July 2012 Riverside Transport reduced its total weekly frequencies by 17 per cent and it is expected to decline further by a similar amount by October 2012 (see Appendix F, Table 3).

7.17 Figure 4 shows Riverside Transport’s network as it stood as at 1 October 2011.

FIGURE 4

Riverside Transport’s network (as of 1 October 2011)


7.18 In our competitive assessment (see paragraphs 7.55 to 7.59), we identified the following Riverside Transport services as overlapping with McGill’s pre-acquisition (as of 1 October 2011) in the Renfrewshire area: 1, 6, 7, 8. 9, 10, 17, 18, 19.

Stagecoach

7.19 Stagecoach is an operator of bus, coach and rail services in the UK and North America. Its turnover for the financial year ended 30 April 2012 was £2.6 billion (2011: £2.4 billion) and operating profit was £237 million (2011: £240 million). In the UK, its bus division operates 7,900 vehicles and in the financial year ended 30 April

73 It cancelled two services (N.6 in January 2012 and 16 in May 2012) and an additional service (N.9) will cease operating in October 2012. It has also reduced frequencies on two other services (N.7 and 17).

25

2012 generated turnover of £1,140 million (2011: £1,027 million) and an operating profit of £176 million (£147 million), a margin of 15.4 per cent (2011: 14.3 per cent).74

7.20 At the annual results presentation in June 2012,75 Stagecoach described the per-formance of its UK bus division as robust. It noted that challenges such as reduced BSOG, higher fuel costs, government spending constraints and the uncertain econ-omy had been known for some time and had been planned for, and that it expected to at least maintain operating profit in the forthcoming financial year.

7.21 Commercial bus services are a core business for Stagecoach. The company attributed the growth of the business particularly to its focus on offering value-for-money fares, investment in greener buses and the roll-out of smartcard technology. Although its focus is on organic growth, it stated that it would evaluate opportunities to acquire smaller bus operators that complemented its existing regional oper-ations.76

7.22 In the Renfrewshire area, Stagecoach operates a small number of mainly inter-urban77 services through its subsidiary Stagecoach Scotland Limited (see Figure 5). In the year ended 30 April 2011, Stagecoach Scotland Limited reported turnover of £135 million (2010: £142 million) and operating profit of £10.2 million (2010: £21.5 million), a margin of 7.6 per cent (2010: 15.1 per cent).78

74 Stagecoach annual accounts, 2012. 75 Preliminary results presentation dated 26 June 2012. 76 Stagecoach annual accounts, 2012. 77 These are services that run between urban centres. 78 Stagecoach Scotland Limited annual accounts, 2011.

26

FIGURE 5

Stagecoach’s network (as of 1 October 2011)


Others

7.23 We also identified a number of small bus operators whose services overlapped (as at 1 October 2011) with those of McGill’s. These included Colchri Coaches, Key Coaches, Henderson Travel, Harte Buses, Slaemuir Coaches, Scottish Travel and Ayrways Coaches. These were generally low-frequency services and often provided services only in the evenings. Some of these operators provide supported services and vehicles for private hire. McGill’s also told us that Gibson Direct had around 100 buses in Paisley, though it did not at present run commercial services.79

The overlap flows between McGill’s and ASW

7.24 The pre-acquisition networks of McGill’s and ASW are set out in paragraphs 7.7 to 7.11 and Figures 1 and 2. The combination of McGill’s and ASW’s services pre-acquisition (as at 1 October 2011) is shown in Figure 6. This shows that the routes on which ASW and McGill’s services overlapped80 (shown in orange) comprised a substantial part of their commercial bus networks. We discuss these in more detail in the following paragraphs and in Appendix F.

79 Gibson Direct historically ran some commercial services in the Renfrewshire area. McGill’s acquired two of these routes (Services 21 and 68) from Gibson Direct in November 2010. 80 These are exact overlaps, ie they are marked as overlaps when the two services share the same roads. Therefore, they underestimate the degree of overlap as they do not consider overlaps portions of routes that are within 500m of each other.

27

FIGURE 6

The extent of route overlaps between McGill’s and ASW’s networks pre-acquisition (as of 1 October 2011)

Source: CC elaboration on data from Traveline.

7.25 In order to identify the overlaps between McGill’s and ASW, we considered that, all other things being equal, services that start and end at exactly the same location, or within 500m of one another, are likely to be perceived as close substitutes by pas-sengers (see paragraph 5.13).

7.26 We note that bus services are differentiated in a number of ways and that not all bus services will be close substitutes for one another:

(a) Since many passengers will wish to travel at a particular time, services by bus operators that operate at significantly different times are not close substitutes.

(b) Frequency is an important service feature for some passengers. A more frequent service reduces waiting time at the bus stops and reduces overcrowding on buses. Frequency of service provided by one particular operator (as opposed to total frequency along a flow) could be less important for other passengers (eg those purchasing single tickets and concession passengers).

(c) The type of ticket available and the level of the fares may influence a passenger’s decision over which bus operator to use. For example, a passenger who is inter-ested in making multiple trips may prefer an operator that offers a ticket (and a network) that allows them to do so. At a network level, ASW and McGill’s offered a largely similar range and type of tickets which were also similarly priced.

28

(d) Most passengers would prefer a service with a shorter travel time that stops less frequently and/or takes a shorter route between the start and the end of their journey. In particular, express services that make infrequent stops provide quicker travel to passengers than urban buses that make more frequent stops.

(e) Some passengers may have a preference for a better-quality service including better reliability, customer service, and newer and cleaner vehicles with additional features such as Wi-Fi. McGill’s business strategy was to provide such a service. For this to succeed in attracting existing passengers from rivals, passengers would have to plan to catch the better-quality service or be prepared to wait for such buses rather than boarding the first bus to arrive. We have not seen evi-dence that passengers differentiated sufficiently between McGill’s and ASW’s services in terms of quality to suggest that ASW and McGill’s vehicles were not close substitutes pre-acquisition.81

7.27 We used a database drawn from traveline.com82 as the main source of information on the bus services supplied by bus operators in the Renfrewshire area as of 1 October 2011.

7.28 We note that some changes have taken place to services since 1 October 2011 which we have taken into account in our flow-by-flow assessment. First, some operators cancelled, launched or changed the frequencies of some of their services. Second, McGill’s made some changes to its services since the acquisition. The flow-by-flow assessment in Appendix F reports the changes in McGill’s frequencies since full integration of the two businesses on 26 March 2012.83

7.29 Using the methodology set out in Appendix F, we identified 15 McGill’s services on which there was at least one significant overlap flow with ASW services. These are assessed in detail in Appendix F and summarized in paragraphs 7.55 and 7.56.84

Levels of competition pre-acquisition and their likely development over time

7.30 As part of our assessment of the overall level of competition between McGill’s and ASW pre-acquisition, we estimated the proportion of McGill’s revenues from the identified 15 overlap flows. We estimated separately the proportion of revenues (excluding concessionary fares reimbursement) and concession trips.85

7.31 We found that a substantial proportion of revenues and concessions on a large number of McGill’s services pre-acquisition faced competition from one or more of ASW’s services pre-acquisition. On seven of McGill’s services this amounted to all, or almost all, the revenue/concessions. Overall, for the 15 overlapping services in the Renfrewshire area, we estimate that [more than three-quarters] ([] and [] per cent) of McGill’s revenues and concession trips, respectively, were subject to some degree of competition from at least one of ASW’s services.

81 It is possible that some of the vehicles of other operators, especially the smaller ones operating minibuses outside Paisley, may not be seen as close substitutes. 82 The CC contracted Basemap Ltd to geo-code the relevant timetable information from the Traveline database for routes oper-ated by bus operators in the Renfrewshire area in October 2011. This information was then used by the CC in its assessment of overlap. 83 This information is difficult to interpret, in part because we do not have the corresponding information on changes in capacity (although in many cases McGill’s told us that a reduction in frequency had been accompanied by an increase in capacity) and in part because McGill’s has an incentive to maintain frequencies of its services close to the pre-acquisition level during the course of our inquiry. 84 We understand that these are all of McGill’s services in Renfrewshire. McGill’s has an additional 22 services, mostly in the Inverclyde area. 85 The information relates to November 2011, which McGill’s told us was a representative month.

29

7.32 We were unable to carry out the same exercise on revenue and concessions for ASW’s services.86 However, we had documentary evidence to show that ASW regu-larly monitored McGill’s behaviour and reacted to it, indicating that ASW’s behaviour was constrained by the presence of McGill’s services (see paragraph 7.40).87

7.33 As well as the degree of overlap between their services, we looked at a number of other factors to assess the degree of competitive interaction between McGill’s and ASW pre-acquisition. Unfortunately, the information available to us was often in-complete (as referred to in paragraph 2.21, McGill’s was not able to provide relevant internal documents and information; Arriva was also unable to provide us with some data and information related to its ASW business; and we were unable fully to re-construct the history of competitive interaction between McGill’s and ASW through fares and frequencies88). Nonetheless, we first look at some specific examples of competitive interaction between McGill’s and ASW. We then consider further indi-cators of the extent to which ASW and McGill’s might have competed pre-acquisition: timetable and frequency changes; how ASW monitored and reacted to McGill’s behaviour; fares; and vehicle type and quality. We assess the impact of this head-to-head competition on services to passengers. We then reach a conclusion on the competitive interaction between McGill’s and ASW pre-acquisition and discuss how this might have developed over time in the absence of the acquisition.

Examples of competitive interaction

7.34 We note a number of examples of intense competitive interaction between ASW and McGill’s pre-acquisition (for more details, see Appendix F):

(a) McGill’s told us that when it took over service 26 in November 2010 and 61 in February 2011 from other operators, ASW halved its return fare to £1.50 for its competing services. McGill’s reacted by lowering its fares to the level of ASW’s on its second day of operation. At the time of the acquisition, more than a year later, McGill’s return fares remained lower than its single fares for service 61. McGill’s also told us that when it took over service 64 in May 2011, ASW again reduced its return fare for its competing service to £1.50.89 Although we do not know how McGill’s reacted, McGill’s return fare on its 64 service at the time of the acquisition, almost a year later, remained below the price of a single ticket.

(b) ASW’s operations reports also provide a detailed picture of the competitive inter-action between its service 23 and McGill’s X23, mainly via frequency changes from September 2009 to at least August 2010.

(c) ASW’s operations reports also mention that in September 2011 McGill’s offered a ten-week student ticket at £81 and a five-week student ticket at £45 with heavy promotion activities at universities. ASW’s operations reports stated that it would react to that.

7.35 We note that, in total, the services mentioned in these examples of competitive inter-action pre-acquisition represented a substantial proportion (at least [] per cent) of McGill’s revenues in the Renfrewshire area, based on overall revenues in November

86 Arriva told us that it had the Electronic Ticket Machine data which would allow us to perform a similar analysis for ASW. However, it no longer had the relevant software (ie Almex) to extract the relevant data. 87 McGill’s told us that it did not produce similar internal documents for its operations (see paragraph 7.41). 88 We did not have ASW’s changes in fares over the last two years. Nor were we able to see all the changes to frequencies, since most of ASW’s and McGill’s services have been registered as ‘frequent services’ and they therefore were not obliged to formally register changes in frequencies with the TCS. 89 We have no information to establish the fare level on service 64 before the change, but because McGill’s told us that it gener-ally set its fares at a network level it is likely that these were at the level of its services 26 and 61 before they were halved.

30

2011. We discuss how this level of competitive activity might have developed in the absence of the acquisition in paragraphs 7.48 to 7.52.

Timetable and frequency changes

7.36 We were told that frequency was an important focus of competition between com-mercial bus providers. In the local bus services market investigation,90 the CC found that bus operators often competed on frequency where they faced direct competition from a rival. Although some passengers were price sensitive, others were more sensitive to differences in overall frequency level.

7.37 In addition to the example set out in paragraph 7.34(b), we looked at changes in ser-vice registrations to see whether ASW and McGill’s competed with each other pre-acquisition by changing the frequency of their services or by launching new routes in the Renfrewshire area. However, the vast majority of McGill’s and ASW’s services in the Renfrewshire area were registered as high-frequency services.91 Provided that McGill’s and ASW maintained their services at a frequency of at least one bus every 10 minutes, we would not be able to identify any substantial changes in frequencies through changes in registration.

7.38 We looked at ASW’s monthly operations reports for the period 2009 to 2011. These showed a few instances (in addition to that mentioned in paragraph 7.34(b)) where operators competed primarily on service frequency and showed that ASW and its competitors competed less frequently on price (see Appendix F).92

Monitoring

7.39 We looked to see if there was evidence of ASW and McGill’s monitoring each other’s service offering and reacting to each other’s actions. Such monitoring (and reaction) would suggest that they considered each other as close competitors, especially if there were fewer, or no, references to actions by other providers.

7.40 ASW’s monthly operations reports show that over the three-year period 2009 to 2011, McGill’s was, by some margin, the operator most often mentioned as taking actions to which ASW either responded or considered responding (see Appendix F). ASW routinely monitored other local bus operators, such as First Glasgow or Stagecoach, on a monthly basis, although actual behaviour was reported much less frequently than McGill’s either in general or as a specific competitive threat to ASW.93 Competitive interaction with smaller bus operators was reported as competitive con-straints or threats early in 2009 but not in the subsequent period to June 2010 when a number had been acquired by McGill’s or ceased trading. No rail services are mentioned in ASW’s operations reports.

7.41 McGill’s told us that it did not produce an equivalent document to ASW’s operations reports and did not formally monitor other bus operators. However, some of the examples mentioned in paragraphs 7.34 and 7.35 suggest that it engaged in informal monitoring.

90 See local bus services market investigation, Section 6. 91 See Appendix F, Annex 16, Table 1. 92 ASW also made reference to marketing campaigns as an important aspect of competition. 93 First Glasgow was mentioned twice, while Stagecoach was never mentioned—see Appendix F.

31

Fares comparison

7.42 We discussed the importance of competition on frequency rather than price in para-graphs 7.36 to 7.38. However, in paragraph 7.36 we argued that a proportion of passengers are likely to be price sensitive. Paragraph 7.34 also mentioned some examples of price interaction. To assess further whether McGill’s and ASW com-peted on price for these customers, we sought to review bus fares94 from McGill’s and ASW along with ASW’s operations reports. Although the data available was insufficient to construct a history of fares interaction, it suggested that the structure of the two operators’ fares was similar, although ASW fare levels were, at the time of the acquisition, slightly higher.

Vehicle type and quality

7.43 McGill’s told us that the type and quality of bus were important for bus users. We noted that McGill’s bus fleet had a lower average age than ASW’s (see paragraphs 2.17 and 2.24) and we were told that McGill’s bus fleet was generally of a higher quality than ASW’s.

7.44 Arriva told us that, given the relatively low profitability of ASW, the average age of its fleet was high and it cascaded older buses down to ASW. Despite McGill’s stated strategy of investing in better-quality buses, it was unclear to us how much passen-gers valued the quality of buses in making choices between competing operators. The evidence from Arriva’s internal survey showed that over the last three years the quality of ASW’s services did not decline in the eyes of consumers and that the quality of its services was on average aligned with those of other Arriva operations in England. This suggests that passengers did not consider that the lower quality of ASW’s buses was an important factor.

Impact of head-to-head competition between McGill’s and ASW

7.45 Paragraphs 7.34 to 7.44 set out, at a high level, the dimensions over which McGill’s and ASW competed pre-acquisition. We sought to understand the impact on passen-gers of this head-to-head competition between the two companies.

7.46 The lack of historic fares and frequency information for ASW makes it difficult systematically to assess the impact of this head-to-head competition. The examples of competitive interaction set out in paragraph 7.34 give some indication of outcomes for passengers. We also looked at the impact of McGill’s entry on ASW’s fares and frequencies as reported in ASW’s operations reports. They report that McGill’s increased its frequencies and improved the quality of the vehicles compared with those of the operators it had taken over. For example, ASW, in referring to increased competition in Erskine in October 2009, also commented in its operations report that, as a result of McGill’s entry on the route, the market was growing. This suggests to us that the number of passengers on the routes involved has increased as a result of competition between ASW and McGill’s, delivering benefits for passengers.

Conclusions on competition pre-acquisition

7.47 Overall we conclude that, pre-acquisition:

94 Our analysis of bus fares was restricted to the period immediately prior to transaction date of 26 March 2012 as this is the only time for which we had both McGill’s and ASW’s data.

32

(a) McGill’s and ASW were each other’s closest competitors in the supply of bus services in the Renfrewshire area. The overlaps covered a significant proportion of McGill’s business in the Renfrewshire area.

(b) We have seen further indications of competitive interaction between ASW and McGill’s. In particular:

(i) We were told of a number of examples of intense competitive interactions between McGill’s and ASW, leading to benefits for passengers in lower fares and higher frequencies. These examples represent a substantial proportion of McGill’s business in the Renfrewshire area.

(ii) ASW’s operations reports suggest that ASW monitored and reacted to McGill’s actions (both in terms of frequency and, less actively, fare changes) more than any other bus operator.

Sustainability of head-to-head competition over time

7.48 We also considered whether the examples of competitive interaction observed pre-acquisition and discussed in paragraphs 7.34 to 7.46 were likely to be sustainable over time. The CC’s local bus services market investigation found that, over the UK as a whole, head-to-head competition was uncommon, although where it existed, it delivered significant benefits to customers. In some cases, entry was immediately followed by a period of intense competition which then settled down into less intense rivalry or co-existence (albeit with continued head-to-head competition) over the longer term.

7.49 The local bus services market investigation identified the following factors as allowing sustainable head-to-head competition:

(a) an element of differentiation between the services provided by the competing operators;

(b) a sufficiently high level of demand along a route for more than one operator to be able to run a frequent service;

(c) an element of timetable coordination; or

(d) accommodation of a non-threatening rival.95

7.50 We found that there was sufficiently high demand on some flows in the Renfrewshire area to allow more than one operator to be present. We saw examples of ASW and McGill’s competing on higher demand routes and in some cases examples of smaller operators being accommodated as they either run fewer frequencies, differentiate their services by time of operation, or compete only on short flows (see Appendix F).

7.51 We note, however, that head-to-head competition between McGill’s and ASW only started in 2009 when McGill’s entered the Renfrewshire area and developed over the period leading up to the acquisition as McGill’s acquired a number of smaller oper-ators. During this time, McGill’s was expanding against a competitor, ASW, that it described as declining. We thought it likely that the level of head-to-head competition that we observed at the time of the acquisition represented an upper bound because, in large part, it still represented ASW’s reaction to McGill’s entry and expansion in the

95 See local bus services market investigation report, paragraphs 8.66–8.81.

33

Renfrewshire area. As set out in Appendix C, paragraphs 7 and 19, both McGill’s and ASW’s profitability had been declining for several years. []

7.52 Given their recent financial performance, we thought it likely that, in the absence of the acquisition, McGill’s and ASW’s competitive offers would have become less attractive (particularly in frequencies and fares) than they had been in the period immediately running up to the acquisition. The competitive interaction that we observed between McGill’s, ASW and First Glasgow on the route from Paisley to Glasgow, where competition intensified and then settled down, is illustrative of this (see paragraph 7.92). We therefore formed the view that the level of competitive interaction between the two companies observed at the time of the acquisition was likely to reduce substantially over the foreseeable future.

Impact of the acquisition on actual competition

7.53 In this section, we examine whether McGill’s acquisition of ASW would lead to a loss of actual competition in the Renfrewshire area compared with the counterfactual situation described in Section 6.

7.54 We first assess each flow overlap and consider for each whether any of the existing providers operating their existing services might constrain McGill’s post-acquisition (see Table 1). We then consider whether providers that are on the overlap flow but would not exert a sufficient constraint on McGill’s post-acquisition with the level of their pre-acquisition services might increase frequencies or modify their routes to improve coverage of the overlap flow in response to a small worsening of McGill’s offer. In the next section, we consider, at an overall level, whether providers with nearby facilities,96 who are not currently active on the overlap flow, would be likely to find it profitable to launch new services to compete on all or part of the overlap flow. Finally, we reach our conclusion on whether there is a loss of actual competition as a result of the acquisition.

Assessment of each flow overlap

7.55 Table 1 lists the 15 overlap flows. It provides summary information on the relevant flows over which McGill’s and ASW overlapped pre-acquisition; the existing con-straints from head-to-head competition from third parties post-acquisition; and whether existing providers on a route would, in our view, be likely to expand or modify those services to constrain any loss of competition that might otherwise arise as a result of the acquisition. Further details of each flow overlap are set out in Appendix F.

96 The local bus services market investigation concluded that a potential entrant with nearby facilities would have lower entry costs than one without them and would therefore exert a stronger constraint (see local bus services market investigation report, paragraphs 8.103–8.165). We only considered operators with nearby facilities as potential entrants in this merger inquiry.

34

TABLE 1 Summary information on the overlap flows identified between McGill’s and ASW

McGill’s service Overlaps with ASW pre-acquisition Constraints post-acquisition Potential concern? No From/to ASW service no Relevant McGill's flows

21 Paisley–Govan 23, 101, 26, 22/22C Almost all flows ([]% of revenue

& []% concessions) First Glasgow (N747) is present, but has fewer

frequencies and does not cover some flows (compete for []% revenue & []% concessions)

Yes. We did not consider that First Glasgow would increase the fre-quencies of its 747 and/ or modify its route to cover more flows.

68 Paisley–Braehead 101, 26, 22/22C 100% As above As above 26 Paisley–Glenburn 26/26E, 22/22C, 101, 66 100% Only on very limited flows from Riverside (N17 & 10)

and Key Coaches (N3). Much fewer frequencies. Yes

51 Paisley–Auchenback 101, 26, 22/22C 100% Only on very limited flows from Riverside (N17). Much fewer frequencies.

Yes

54 Paisley–Auchenback–Neilston

101, 26/26E, 22/22C, 66 Only Paisley to Auchenback ([]% revenue & []% concessions)

As above Yes

38 Glasgow–Millarston (via Paisley)

36, 38, 7/17, 800, 20, 23, 60, 121

100% Strong from First Glasgow (N9) with substantial frequencies and 100% overlap—limited from ScotRail (for []% revenue & []% concessions)

No

52 Barrhead circular 101, 103 100%, but this is a service with a short circular service while ASW's 101 and 103 have much longer routes. Hence, the degree of com-petitive interaction is unclear.

Very limited (Stagecoach N X44B, Henderson N395 and Ayrways N60)

No

55 Neilston–Silverburn 103 Only Neilston–Nitshill–Auchenback ([]% of revenue & []% conces-sions)

Strong on flows between Silverburn and Nitshill from First Glasgow (N103, 56/56A & 6) with high frequencies. Very limited on flows between Silverburn–Nitshill & Barrhead ([]% revenue & []% concessions).

Yes but limited to increases in fares on flows between Silverburn– Nitshill and Barrhead. Decreases in frequencies appear unlikely.

61 Paisley–Foxbar 60, 61, 20 100% Very limited. Riverside (8 & 10) on few flows. Riverside's new service 26 may offer some constraint, but frequencies are limited.

Yes

64 Asda–Gallowhill 64, 61, 20, 7, 60, 800 Almost all ([]% of revenue & []% concessions)

Very limited from Riverside (8 & 9) with frequencies that were substantial but much lower relative to McGill's. Flows cover []% revenue & []% concessions.

Yes

X23 Erskine–Glasgow 23, 22/22C Almost all Very limited from First Glasgow (X8) and Scottish Travel (521). Limited frequencies and very few flows ([]% revenue & []% concessions)

Yes

X5 Bridge of Weir–Glasgow

36, 7/7E Almost none Strong from First Glasgow (9) with substantially higher frequencies

No

X6A Johnston–Glasgow 7/7E, 17A Almost none ([]% revenue & con-cessions)

Limited No

X7/ X7A

Bridge of Weir–Glasgow

7/7E Almost none (only covering flows from Bridge of Weir and Glasgow city centre)

Strong from First Glasgow (9) with substantially higher frequencies on almost the whole route, but the identified flow.

No (although it could lead to increases in fares from Bridge of Weir to Glasgow city centre).

Source: CC’s calculations and assessment using data from Traveline and registrations from the TCS.

35

7.56 Table 1 shows that:

(a) Some services did not raise potential concerns either because the overlaps were modest (X5, X6, X7/X7A and 52) or because First Glasgow remained a strong competitor post-acquisition (38).

(b) Two services raised potential concerns limited to a few confined flows (54 and 55).

(c) Some services raised potential concerns over a large proportion, or all, of the overlap flows (21, 68, 26, 51, 64, 61 and X23).

7.57 For flows under (b) and (c), as set out in Table 1, we reached a view that the existing services of other providers (eg First Glasgow or Riverside Transport) were unlikely to provide a sufficient constraint post-acquisition, since they had fewer services, lower frequencies or were only present on some of the overlap flows. They were not, in our view, likely to increase frequencies or modify routes to improve coverage of the over-lap flow. Potential concerns over McGill’s increased market share on these flows would, however, only result in an actual concern if entry, or the threat of entry, con-sidered in the next section was insufficient to prevent a small worsening of McGill’s offer compared with the situation set out in the counterfactual.

7.58 We estimated that [at least half] of McGill’s revenues and [] concession trips was represented by the services in paragraph 7.56(b) and (c). Assuming that ASW’s revenues on these overlap flows represented a similar proportion of its revenues, about half of all bus services in the Renfrewshire area might cause a concern as a result of the acquisition, absent any countervailing constraint from the threat of entry.

7.59 Although we have not been provided with details of service by service profitability by McGill’s (see paragraph 7.9), ASW provided us with route-costing information for the first ten months of 2011 (see paragraph 7.11 and Appendix C, Table 8). This shows that ASW’s services that competed head-to-head with McGill’s were those that had the highest revenues among ASW’s services and that they all made a profit contribu-tion (with the exception of ASW’s service 7/17).

Entry or threat of entry via new services

7.60 We looked at whether, post-acquisition, McGill’s behaviour would be likely to be con-strained by entry or the threat of entry by operators not present on the identified overlap flows but which have nearby facilities.

7.61 In assessing whether entry and/or the threat of entry would offset any potential con-cerns that might otherwise arise as a result of the acquisition, we consider whether entry is likely, sufficient and timely (or whether it is perceived to be such, in the case of the threat of entry) in response to a small worsening of the incumbent’s offer relative to the pre-acquisition prices and levels of service.97

7.62 In general, the CC considers it to be rare for the threat of entry to act as a sufficient constraint to offset competitive concerns. However, as set out in our guidelines, this is more likely to be the case where entry would be quick and relatively costless. A constraint from potential entry may arise even if the CC does not expect that entry would actually occur. The important factor is whether the incumbent’s behaviour is constrained by its perception of the threat of entry. This assessment inevitably

97 See CC2, paragraphs 5.8.1–5.8.15.



36

requires an element of judgement, particularly in a merger inquiry which cannot rely to a great extent on documentary evidence.

7.63 In the CC’s local bus services market investigation, it found that incumbent bus oper-ators that did not face head-to-head competition could be constrained by entry, or the threat of entry. In general, the CC considered this to be less of a constraint than head-to-head competition. However, it also noted that the strength of the constraint was dependent on the nature of the potential entrant and on specific local circum-stances.

7.64 In particular, the CC noted that:

the strength of the competitive constraint on incumbents is dependent on the nature of the potential entrants that the incumbent faces. Incum-bent operators will be more likely to restrain their prices and where this is the case will restrain their prices to a greater degree, in anticipation of entry, where there is a potential entrant with relatively lower entry costs. The evidence … shows that entry costs are lower where an entrant has nearby services and facilities, where the entrant is a larger operator relative to the size of the incumbent, where the entrant has a network of services that deliver a comparable network advantage in the local area, and where area-specific barriers to entry are absent. The strength of the constraint on incumbent operators is also likely to be influenced by the nature of post-entry competition.98

7.65 In this section, we examine the likely strength of this constraint in this merger inquiry. We first consider the history of entry (including entry by acquisition) in the Renfrewshire area. We then look at whether operators would be able to enter the overlap flows for which we have identified some potential concerns. We consider whether entry or expansion was likely and, if so, whether it would be timely and sufficient to offset any competitive concerns. We look at the nature of McGill’s as the incumbent relative to that of potential entrants, particularly focusing on the likelihood of entry or expansion by First Glasgow and Stagecoach, both of which already operate some services in the Renfrewshire area (see Figures 3 and 5). Finally, we draw some conclusions on entry and expansion.

History of entry

7.66 We looked at the history of entry in the Renfrewshire area over the past few years. We note (see paragraph 7.3) that the Renfrewshire area shows high demand for bus services and offers good opportunities for operators.

7.67 McGill’s has been the most active commercial bus operator in Renfrewshire. It started providing services in the area in 2009 and, by the time of the acquisition, had a network that was broadly similar in coverage to that of the largest operator in the area, ASW (see paragraph 7.6). It told us that it took over smaller existing providers99 and launched some entirely new services.100

7.68 First Glasgow told us that it had launched four new services in the Renfrewshire area since 2008; two of these101 were not successful and were withdrawn rapidly and two

98 See local bus services market investigation report, paragraph 8.162. 99 McGill’s said that it expanded capacity on the services it took over (by replacing existing vehicles with larger capacity vehicles). 100 McGill’s said that its services X23, 51, 52, 54 and 55 were all new services. 101 Service 601 Silverburn–East Kilbride and Service 901 Glasgow–Largs.

http://www.competition-commission.org.uk/assets/competitioncommission/docs/pdf/inquiry/ref2010/localbus/pdf/00_sections_1_15.pdf

37

continued to operate.102 It recently also cancelled one service and combined two services into one (see paragraph 7.94). We discuss this evidence in more detail in paragraph 7.92.

7.69 Riverside Transport, one of the smaller operators in the area, launched two new services in July 2012. We were told that this [] may therefore be an adjustment to regulatory changes (See Appendix F).103 In general, Riverside Transport has been reducing its supply of commercial services in the Renfrewshire area since 1 October 2011 and will be reducing it further in the next few months (see paragraph 7.16).

7.70 We also note that ASW restructured its network in January and November 2011 undertaking some minor changes to its existing services.104

7.71 Overall, we were cautious in drawing inferences from the history of entry in this case. Competitive conditions were different in the past, with a larger number of smaller operators in the Renfrewshire area. Despite this, there is evidence that entry through the launch of new services has occurred, most notably by McGill’s when ASW was the large incumbent. At the same time, the level of government subsidy (through BSOG and concessionary fares reimbursements) was higher than it is today or is likely to be in the future, which may reduce opportunities for expansion.

The ability to enter

7.72 We looked at several factors which might affect the ability of a bus operator to enter the Renfrewshire area. In particular, we considered whether the requirements of an operator’s licence, the need to access bus stations and bus stands, or the availability of bus depots, were likely to affect a potential entrant’s ability to enter. We consider other possible factors such as network effects, route profitability and the possible reaction of the incumbent in discussing the likelihood of entry. These factors are considered further in Appendix G.

7.73 We saw no evidence to suggest that licence requirements or the need to access bus stations and bus stands would be a barrier to entry in the Renfrewshire area. With regard to bus depots, Arriva told us that it would consider that, as a rule of thumb, a route in the Renfrewshire area could be served by any depot or outstation within a 30-minute drive to the route.105 Both First Glasgow and Stagecoach told us that it was easy to alter, add or remove routes by giving the appropriate notice to SPT and the TCS and that they would be able to enter into the areas of overlap between ASW and McGill’s. First Glasgow said that its Larkfield and Scotstoun depots were both within 3 miles of Glasgow city centre and close to most of the competitive overlaps. It had significant spare capacity at these depots (it currently had approximately 450 vehicles with a total capacity of approximately 600). In addition, over the next few years it would need to build a significant depot in the urban Glasgow area appropriate for the Glasgow business. Stagecoach told us that it had limited spare capacity to expand at its existing depot but that finding additional suitable sites would not be difficult and there were no regulatory barriers to setting up depots.

102 Service 747 Glasgow Airport–Buchanan Bus Station and Service 9 Linwood/Paisley–Buchanan Bus Station. 103 [] 104 Arriva also told us that on other occasions in the past ASW had launched some new services, but that these were relatively minor night bus services on Friday and Saturday nights. 105 This is consistent with the findings of the CC’s local bus services market investigation, Appendix 7.2.

38

7.74 We note that both First Glasgow and Stagecoach have bus depots within 10 miles of the identified overlap flows with at least some spare capacity (see Appendix G, para-graphs 11 to 16).106

7.75 We conclude that there are no structural factors impeding entry in the Renfrewshire area.

Likelihood of entry

7.76 We then considered the likelihood of entry. We looked at three factors which we thought might impact on the likelihood of entry: network effects, expectations of route profitability and expectations of incumbent reactions. We also considered the scale of entry that was most likely to be successful.

7.77 Network effects can occur when a single operator has a more frequent and compre-hensive service on a given route (within-route network effects) or on a network (between-route network effects), so that a passenger with a multi-journey ticket bene-fits compared with a single-journey ticket holder (see Appendix G, paragraphs 17 to 24). Where these effects are substantial, an entrant may need to enter on a substan-tial scale to overcome incumbency advantages.

7.78 Prior to the acquisition, both McGill’s and ASW sold a substantial proportion of return or multi-journey tickets107 and therefore may have benefited from some network effects. However, no operators told us that they would be deterred from entry because of network effects. We note that, particularly in the past, there were a number of smaller operators competing on particular routes who were not able to benefit from between-route network effects and that McGill’s was also able to enter the area and build its own network of services despite the presence of ASW that benefited from network effects. We therefore did not consider network effects to make entry substantially less likely.

7.79 The cost of bringing a route to maturity could impact on the likelihood of entry (see Appendix G, paragraphs 30 to 34). In the local bus services market investigation the CC concluded that the uncertainty over the scale of the cost of bringing a new route to maturity raised the financial risks of entry or expansion. The threat of potential competition from smaller operators with limited resources will therefore generally be a lesser constraint than the threat of potential competition from larger and stronger operators. Local bus operators in the Renfrewshire area told us that it was becoming more difficult to operate bus services profitably owing to a combination of factors, including the impact of the economic environment on consumer spending, rising fuel costs, the changes to the BSOG scheme introduced on 1 April 2012 (see paragraphs 2.11 to 2.13) and the budget cap on concessionary fares reimbursement.

7.80 McGill’s stated that the changes in the concessionary travel budget had reduced its revenue [] in 2011 and that there would be a larger shortfall in 2012. McGill’s also told us that its reduction in BSOG amounted to [] per cent of its revenues. McGill’s and third parties also told us that bus operators in urban areas, such as First Glasgow, were the ones that were worst affected by the changes to BSOG. Stagecoach argued that it would not necessarily be discouraged from entering the Renfrewshire area by the changes in BSOG and concessionary fares reimburse-

106 The 10-mile radius used is much smaller than the 20-mile radius that some operators referred to in the local bus services market investigation. 107 McGill’s told us that [] per cent of passengers used return tickets, [] per cent used multi-journey tickets and [] per cent used multi-operator tickets. ASW did not have precise data on ticket types but a customer satisfaction survey in 2011 (see paragraph 7.44) showed that 10 per cent of passengers used return tickets, and 46 per cent used multi-journey tickets.

39

ment. It considered, however, that it was still too early to evaluate the full impact of the BSOG change on its ability to enter the Renfrewshire area.

7.81 In our view, the regulatory changes (eg BSOG) could provide incentives for operators in dense urban areas, such as First Glasgow, to reduce their dependency on these types of routes. The changes in concessionary fares reimbursement also provide incentives for bus operators to adjust their current services to reduce their depen-dency on concession passengers, [].

7.82 We also note that the minimum standards required for buses by the Paisley and Glasgow SQPs may make entry into the Renfrewshire area more costly by increasing the required investment in vehicles (see paragraphs 2.9 and 2.10). Stagecoach said that the Paisley SQP would not be a factor discouraging it from offering services in Renfrewshire (and indeed that it regarded such schemes positively). In our view, the introduction of SQPs would be likely to discourage entry by small operators but was not an impediment to entry for large bus operators.

7.83 In addition, the threat of a competitive reaction from the incumbent operator may reduce the likelihood of entry because it makes it more difficult for the potential entrant to assess whether the new route will be profitable (see Appendix G, para-graphs 25 to 29). We understand that ASW and First Glasgow reacted strongly to McGill’s entry in the past. First Glasgow reacted to McGill’s takeover and expansion of its service 38 (see paragraph 7.92). Equally, as discussed in paragraph 7.34, ASW reacted strongly to McGill’s acquisitions in the Renfrewshire area by lowering its fares. These examples suggest that although the incumbent reacted to entry, the entrant was able to continue to provide services on the flow it had entered. It also suggests that McGill’s, then the entrant, was not discouraged by the incumbent’s reaction from subsequently entering on additional flows in competition with incum-bents.

7.84 Stagecoach told us that if it identified an attractive commercial opportunity to enter a new route, it would not be discouraged by the potential reaction from incumbent operators. Two smaller operators, however, told us that McGill’s potential reaction to entry made the financial outcome of entry uncertain.

7.85 We conclude that the likelihood of entry for smaller operators would be reduced by the costs of bringing a route to maturity and the credible threat of retaliation by an incumbent. This seemed to us less likely to be an issue for larger operators.

7.86 We also considered the scale of entry that was likely to be successful in constraining McGill’s post-acquisition. We thought it unlikely that small-scale entry—ie entry with a low-frequency service relative to that of the incumbent’s service—would act as a sufficient competitive constraint. We also note that, whilst such entry may be attrac-tive to concession and single fare passengers, it is unlikely to attract passengers who use multi-journey or return tickets. We therefore looked specifically at the likelihood of entry by the two large operators with nearby facilities, First Glasgow and Stagecoach, on particular flows in the Renfrewshire area. Before doing so, however, we discuss the nature of McGill’s as incumbent.

McGill’s as incumbent

7.87 The nature of the incumbent is important in assessing the potential constraint from entry or the threat of entry. McGill’s economic adviser argued that the actions, inten-tions and ability of small local operators were likely to be markedly different from those of large operators that operate nationally. He said that the strength of the potential constraint was likely to be dependent, among other things, on the size of the

40

incumbent relative to the potential entrant. As such, he argued that smaller operators, such as McGill’s post-acquisition, were likely to perceive potential competition as a stronger constraint than would a larger operator.

7.88 We note that, relative to either FirstGroup or Stagecoach, McGill’s is a small bus operator. For example, McGill’s, post-acquisition, is only around one-third of the size of First Glasgow, which itself is part of a much larger group. It does not have the management structure or financial reserves of a large company. This is likely to influence its perception of the constraint posed by the larger operators. In particular, given its recent acquisition of ASW and the large financial commitment it has under-taken, we believe that it would have strong incentives, at least for the next [] (while it continued to pay for the acquisition of ASW), to avoid providing opportunities for entry by reducing frequencies or increasing fares. It would therefore be unlikely to worsen its offer post-acquisition and risk entry by First Glasgow in particular. It might feel particularly vulnerable on flows which have generated sufficient demand to sus-tain a period of head-to-head competition between ASW and McGill’s pre-acquisition. Whilst McGill’s has shown itself to be prepared to compete actively with First Glasgow in the past, it would go to some lengths to avoid provoking large-scale entry on a particular route by First Glasgow in the Renfrewshire area.

7.89 We also note that McGill’s is very different in nature from ASW as the incumbent bus operator in the Renfrewshire area. Arriva, FirstGroup or Stagecoach operate across much of the UK; any attempt to enter each other’s area of operation in the Renfrewshire area could be met with a strong competitive reaction elsewhere in the UK. This is much less true of McGill’s, whose only other operation is in Inverclyde, from where it is unlikely to be able to react to entry by First Glasgow in the Renfrewshire area. Thus the incentive for one of the large operators to enter against McGill’s is greater than it would have been against ASW.

First Glasgow

7.90 We describe FirstGroup and its overall strategy in paragraphs 7.12 and 7.13. In paragraph 7.14, we set out First Glasgow’s operations in the Renfrewshire area. First Glasgow operates approximately 70 services in the Glasgow area and (as of 1 October 2011) eight services in the Renfrewshire area which overlap, at least to some extent, with McGill’s services.108

7.91 Given the proximity of its relatively extensive network, the spare capacity in its nearby depots and the ease with which it could find sites for building new depots (see Appendix G, paragraphs 11 to 16), First Glasgow is an obvious candidate for expan-sion in the Renfrewshire area. It would be able to enter new routes in the Renfrewshire area quickly and at low cost, and, in so doing, would be able to benefit further from any network effects. As set out in paragraph 7.62, we consider entry, or the threat of entry, to be more likely to act as a constraint in such circumstances. We considered that small-scale entry in response to a small worsening of McGill’s offer was unlikely (see paragraph 7.86). However, given the size of its operation, First Glasgow would have the scale to enter with a high-frequency service in response to a small worsen-ing of McGill’s offer on the particular overlap flows. Figure 7 shows First Glasgow’s network and the combined network of McGill’s and ASW as of 1 October 2011.

108 In our flow-by-flow assessment, we identified the following First Glasgow services overlapping with McGill’s services: 747, 25/25A, 9/N9, 56/56A, 89, 103, 6, X8/X8A. We have since been told by First Glasgow that service 25/25A was withdrawn in April 2012. Only First Glasgow’s services 9/N9 and 56/56A had weekly frequencies above the average of either McGill’s or ASW’s services (as of 1 October 2011).

41

FIGURE 7

First Glasgow and the combined ASW and McGill’s networks

Source: CC’s elaboration on Traveline data.

7.92 First Glasgow has launched four new services in Renfrewshire since 2008 (see para-graph 7.68). Two of these (service 601 Silverburn–East Kilbride and service 901 Glasgow–Largs) were unsuccessful and were withdrawn but two (service 747 Glasgow Airport–Buchanan Bus Station and service 9 Linwood/Paisley–Buchanan Bus Station) continue to operate. The route of First Glasgow’s service 9 was served by ASW and First Glasgow when McGill’s took over Dickson’s of Erskine, which operated service 38 with a minibus service. First Glasgow told us that it increased it frequencies from 8 to 12 buses an hour in June 2011 and reduced its fares. This was noted in ASW’s operations reports. First Glasgow stated that it reduced the fre-quency of its service 9 to ten buses per hour after April 2012. McGill’s halved its post-acquisition frequencies on the route (including former ASW services 36 and 38) since full integration (26 March 2012).

7.93 As we discussed in paragraph 7.71, we were cautious in drawing inferences from these examples because of the differences in the competitive environment pre- and post-acquisition. Nevertheless, these examples suggest that First Glasgow was able to enter the Renfrewshire area with two services with significant frequencies. This suggests that entry with a high-frequency service is attractive. We do not have more evidence on the two failed entry attempts by First Glasgow—ie services 601 and 901. In our view, though, the scale of First Glasgow’s existing network in Renfrewshire would allow it to launch—and withdraw—new services more easily than might be the case for a smaller operator.

7.94 Although First Glasgow said that it had the ability to expand in the Renfrewshire area (see Appendix G), it told us that it currently had no plans to provide additional ser-

42

vices in the Renfrewshire area. First Glasgow also said that, []. It provided us with data showing that it had reduced its total mileage in Glasgow by [] per cent between 2009/10 and 2010/11 and by a further [] per cent between 2010/11 and 2011/12. A similar percentage reduction occurred in March and April 2012 compared with the same months of 2011.109 Thus First Glasgow had reduced its overall services in the last three years by almost [] per cent. First Glasgow also told us that it would face some criticism externally if it were seen to expand in the Renfrewshire area whilst reducing its services in Glasgow. First Glasgow also provided us with total mileage data for each of the services it operates in the Renfrewshire area that showed a similar percentage decline between October 2011 and June 2012.110

7.95 Several small operators told us that they considered that First Glasgow would be unlikely to enter at the moment. [Two small operators] said that the impact of regulat-ory changes and the current state of the bus industry had forced First Glasgow to reduce its operations across Scotland. [Two small operators] also told us that First Glasgow was particularly affected by cuts in subsidy because of the large vehicles it ran and its type of operation, neither of which was fuel efficient.

7.96 We considered whether First Glasgow would be likely to enter the Renfrewshire area if a profitable opportunity arose as a result of McGill’s worsening its offer post-acquisition. First Glasgow told the Office of Fair Trading that it would not contemplate entry solely as a result of McGill’s increasing prices by 5 or 10 per cent as most areas were already ‘well bussed’ and a reasonable proportion of passengers were concession passengers (for whom price was irrelevant). The principal consideration would be identification of sufficient demand to enable profitable service operation.

7.97 McGill’s economic adviser111 argued that it was hard to believe that First Glasgow would not enter, if presented with profitable opportunities as a result of a change in behaviour by McGill’s resulting from the acquisition. He argued that the presence of First Glasgow was likely to act as a strong competitive constraint on the actions of McGill’s post‐acquisition. First Glasgow’s attempts to enter the Renfrewshire area in the past, and its subsequent exits, suggested to him that barriers to entry (and exit) were low.

7.98 Regardless of First Glasgow’s likelihood of actual entry, we considered whether McGill’s would perceive First Glasgow to be a credible threat which would constrain McGill’s from substantially worsening its offer on the overlaps identified relative to the counterfactual.

7.99 We recognize that the evidence we have on entry or the threat of entry is incomplete for a number of reasons, most notably the lack of internal documents by McGill’s that could allow us to understand better McGill’s perception of First Glasgow as an entry threat. We asked McGill’s for its perceptions of First Glasgow as an entry threat. It told us that it had fought First Glasgow before and, although it was confident that passengers would choose McGill’s in preference to First Glasgow because of McGill’s better quality of service, it had found that First Glasgow had reacted to McGill’s by doubling or trebling frequencies and slashing fares. It said that First Glasgow could do this because the cost did not matter to it, given its financial

109 For the combined months of March and April the decline between 2011 and 2012 continued at [] per cent. 110 The data provided refers to the weeks starting on 2 October 2011 and 10 June 2012. The data shows a combined reduction of total mileage for all combined services of [] per cent. In particular, First Glasgow has cancelled its service 25/25A, combined services 89 and 90 and reduced their overall mileage, reduced the overall mileage of its services 747 and 9, and increased overall mileage of its X8/X8A. 111 Alex Dickson, www.competition-commission.org.uk/assets/competitioncommission/docs/2012/mcgills-arriva-west-scotland-inquiry/mcgills_initial_submission_annex_1.pdf, paragraphs 39–41.

http://www.competition-commission.org.uk/assets/competitioncommission/docs/2012/mcgills-arriva-west-scotland-inquiry/mcgills_initial_submission_annex_1.pdf

http://www.competition-commission.org.uk/assets/competitioncommission/docs/2012/mcgills-arriva-west-scotland-inquiry/mcgills_initial_submission_annex_1.pdf

43

resources. It also commented that, in its view, First Glasgow was much more likely to enter Renfrewshire in competition with a larger McGill’s than it had been when ASW was the main operator in the area. It told us that First Glasgow would have been reluctant to enter against ASW because of its fear of retaliation in another geographic market. This is not the case with McGill’s and therefore McGill’s could not afford to leave any gaps in its services or opportunities for First Glasgow to enter.

7.100 On balance, we reached the conclusion that McGill’s, as a smaller bus operator, per-ceived First Glasgow to be a credible threat and that this would be likely to constrain McGill’s from substantially worsening its offer post-acquisition. Given the low barriers to entry and expansion, its extensive network, its presence on some flows already in the Renfrewshire area and, in particular, its spare capacity nearby, we had no doubt that First Glasgow could enter new flows in a timely way should a profitable oppor-tunity arise.112 Given its financial resources, we thought that First Glasgow would be willing to take on McGill’s and be prepared to suffer some short-term losses should McGill’s react to its entry. The examples of First Glasgow’s past entry and expansion suggest that a threat of entry is credible. For these reasons, we thought it likely that McGill’s would perceive First Glasgow as being prepared to enter flows on which it was not currently present.

Stagecoach

7.101 We describe Stagecoach and its overall strategy in paragraphs 7.19 to 7.21. In para-graph 7.22, we set out Stagecoach’s operations in and around the Renfrewshire area. Stagecoach is also a large bus operator that might discipline the behaviour of McGill’s post-acquisition via the threat of entry.

7.102 Most of the Stagecoach services in and around the Renfrewshire area are inter-urban services from Ayrshire which either travel through the south side of Glasgow en route to Glasgow city centre or enter the city centre from the M8 motorway. In our flow-by-flow assessment, we found only one overlap flow with McGill’s.113 Unlike First Glasgow, Stagecoach does not have a nearby urban network on which to build by adding new services in the Renfrewshire area in response to a small worsening of McGill’s offer. This would make it more difficult for Stagecoach to overcome between-route network effects.

7.103 Although, like First Glasgow, Stagecoach confirmed that it would be able to enter the Renfrewshire area, it told us that it did not have any current plans to introduce further commercial services within the Renfrewshire area which would overlap with services of ASW and McGill’s.

7.104 Stagecoach also said in its response to the Office of Fair Trading that it would [] if McGill’s increased fares by 5 to 10 per cent. Stagecoach stated that it would adopt a broader view [] such that it would create an attractive opportunity for expansion. Stagecoach stated that a change [] would be one factor which it would take into account when assessing any possibility for expansion.

7.105 We asked McGill’s about its perception of Stagecoach as a potential entrant in Renfrewshire. It said that, given its track record, it would be fearful of it, and that it

112 We also note that it is relatively heavily affected by the reduction in public subsidy and increasing fuel prices as it operates a dense urban network and hence may have an incentive to look for opportunities in less dense urban areas such as the Renfrewshire area. 113 Stagecoach service X44B overlapped with McGill’s services 52 and 55, although it had very limited frequencies (see Appendix F for more details).

44

would be likely to give McGill’s a hard time. McGill’s noted that Stagecoach had nearby depots and could enter aggressively.

7.106 Overall, we thought it less likely that, in comparison with First Glasgow, Stagecoach would enter the Renfrewshire area, or that McGill’s behaviour would be constrained by the threat of entry by Stagecoach alone. Nevertheless, in our view, the presence of Stagecoach would add to the likelihood that McGill’s behaviour would be con-strained post-acquisition.

Conclusions on entry

7.107 We find there to be no structural factors which would be a significant barrier to a bus operator’s ability to enter or expand on the overlap flows identified. However, in our view, issues of route profitability and possible incumbents’ reaction to entry would provide disincentives for smaller operators in particular to enter on new flows.

7.108 No operator told us that it had current plans to expand into the Renfrewshire area. Several operators told us that a small worsening of McGill’s offer post-acquisition would be insufficient to trigger entry. []

7.109 Nevertheless, as set out in paragraph 7.100, on balance we reached the conclusion that McGill’s, as a smaller bus operator, perceived First Glasgow to be a credible threat and that this would be likely to constrain McGill’s from substantially worsening its offer post-acquisition. The presence of Stagecoach would, in our view, add to the likelihood that McGill’s behaviour would be constrained post-acquisition.

Conclusions on the loss of actual competition

7.110 We conclude that, pre-acquisition, McGill’s and ASW were close competitors in the supply of bus services in the Renfrewshire area (see paragraph 7.47). The overlaps covered a significant proportion of McGill’s business in the Renfrewshire area and we saw further evidence of competitive interaction between ASW and McGill’s. ASW monitored McGill’s (both on frequency and, less actively, fare changes) more than any other bus operator.

7.111 However, as set out in paragraphs 7.51 and 7.52, we thought it likely that the level of head-to-head competition that we observed at the time of the acquisition represented an upper bound and would reduce over time. Both McGill’s and ASW’s profitability had been declining for several years. [] We thought it likely that, in the absence of the acquisition, McGill’s and ASW’s offers would have been less attractive (particu-larly on frequencies and fares) than they had been in the period immediately running up to the acquisition.

7.112 We conclude that the acquisition would give McGill’s a very large market share on a number of overlap flows, covering at least half of McGill’s revenues/ concession trips. We did not think that other providers already present on these overlap flows were likely to exert a sufficient constraint on McGill’s post-acquisition since they had fewer services, lower frequencies or were only present on some of the overlap flows. They were not, in our view, likely to increase frequencies or modify routes sufficiently to constrain McGill’s across its network, except as part of a wider entry strategy.

7.113 Our conclusions on entry are set out in paragraphs 7.107 to 7.109. In summary, we found there to be no structural factors which would be a significant barrier to a bus operator’s ability to enter or expand on the overlap flows identified. However, in our

45

view, issues of route profitability and possible incumbents’ reaction to entry would provide disincentives for smaller operators in particular to enter on new flows.

7.114 We conclude that McGill’s perception of the credible threat of entry by First Glasgow, in particular, as well as Stagecoach, was, on balance, likely to be sufficient to prevent McGill’s from substantially worsening its offer post-acquisition compared with the counterfactual.

7.115 We therefore conclude that there was no substantial loss of actual competition as a result of the acquisition.

Impact of the acquisition on potential competition

7.116 In this section, we look at the impact of the acquisition on potential competition both within and outside the Renfrewshire area.

Loss of potential competition

7.117 In assessing the impact of this acquisition on potential competition, we consider that any loss of potential competition matters most when, absent the acquisition, entry or the threat of entry is in the form of expansion via organic growth. We put less weight on acquisitions of existing operators and services because an acquisition replaces one operator with another without affecting the structure of the market. However, we also note that the nature of the operator may affect the way it competes.114

Loss of potential competition within the Renfrewshire area

7.118 Pre-acquisition, ASW and McGill’s competed head to head on flows that represented a substantial proportion of their services in the Renfrewshire area (see paragraph 7.31). Nonetheless, McGill’s and ASW might also have constrained each other on flows where they did not compete with each other because each operator’s behav-iour might have been constrained by the threat of entry by the other. The loss of one independent competitor might therefore lead to a reduction in competition even on flows where the two parties do not overlap in the Renfrewshire area. This assess-ment in this merger inquiry is, therefore, complementary to that on the loss of actual competition in the same area.

7.119 We first assessed whether there was a loss of actual potential competition:115

(a) Arriva told us that had the ASW business not been sold, it would have attempted to improve returns both by seeking further opportunities for expansion and by exploring further rationalization. However, we have not seen evidence that ASW significantly altered its services in the last few years. For example, while First Glasgow and McGill’s had launched several substantial new services in the last three years, ASW had generally modified existing services or launched very low-frequency, complementary, night services on Fridays and Saturdays.

(b) McGill’s had recently expanded in the Renfrewshire area. We were told that this occurred both through takeovers of existing operators and services and through the launch of new services.

114 For example, the operator taking over may be more aggressive than, or benefit from economies of scales not available to, the operator that had been taken over. This may lead the new owner, for example, to expand frequencies and/or capacity for the service. 115 CC2 distinguishes between actual and perceived loss of potential competition. See paragraphs 5.4.13–5.4.18.



46

7.120 Overall, we found that, absent the acquisition, neither operator had plans, and/or appeared particularly likely, to expand further through organic growth in the Renfrewshire area in the future. We also note that given the extent of the overlap pre-acquisition and the fact that the remaining (non-overlapping) services by ASW did not seem on the whole to be particularly profitable (see Appendix C), there would be few opportunities for either McGill’s or ASW to expand further in the Renfrewshire area.

7.121 Second, in relation to any loss of perceived potential competition, we concluded in paragraph 7.40 that ASW perceived McGill’s behaviour as a threat pre-acquisition. McGill’s argued that it did not believe that it had been affected by the presence of ASW on routes where it did not compete with ASW.116 Given that there were only a few services over which McGill’s and ASW did not compete head to head pre-acquisition in the Renfrewshire area, and these were not, as far as we were aware, very profitable, we conclude that there will not be any loss of perceived potential competition.

Loss of potential competition outside the Renfrewshire area

7.122 The acquisition may also lead to a change in both actual and perceived potential competition outside the Renfrewshire area, for example in Glasgow, by reducing the number of large potential entrants (relative to other operators in the Renfrewshire area such as Riverside Transport and Colchri Coaches) based in the Renfrewshire area from two to one.117

7.123 We were not aware of any specific pre-acquisition plans by either McGill’s or ASW to expand organically outside the Renfrewshire area in the future. There is some evi-dence that may suggest that ASW was unlikely to have entered new flows in direct competition with First Glasgow.118 Overall, the absence of concrete plans suggested to us that, absent the acquisition, entry by ASW in competition with First Glasgow in the Glasgow area was not likely. Equally we have not been told of definite plans by McGill’s to expand outside the Renfrewshire area.

7.124 First Glasgow told us that it considered McGill’s, ASW (pre-acquisition) and Stagecoach as potential entrants into the Glasgow urban area, given their size and strength, and that it monitored the Vehicle and Operator Services Agency website (where new registrations, changes in registration and cancellations were published) every day. We have no evidence to establish whether First Glasgow’s behaviour was con-strained by the threat of entry by either ASW or McGill’s. It told us that in order to deter entry it avoided leaving gaps (frequency, quality and/or price) in its overall service offering. Yet it provided evidence to us that it had reduced its total services provided in the Glasgow urban area with [] reductions in total mileage supplied (see paragraph 7.94). Given that it is unlikely that ASW represented an actual or perceived threat to First Glasgow (and possibly other operators outside the Renfrewshire area), pre-acquisition, we found there to be no loss of actual or per-ceived potential competition outside Renfrewshire. However, there may also be a benefit from the acquisition if it made it more likely for McGill’s to enter outside the Renfrewshire area compared with McGill’s and ASW pre-acquisition. We considered the following factors:

116 We have not seen documentary evidence that McGill’s actively monitored ASW but this may simply reflect the fact that it did not record this in written form. 117 We focused on the Glasgow area where First Glasgow has a large presence, although the reasoning could equally apply to other areas. Furthermore, there are also some smaller bus operators in the Renfrewshire area. 118 []

47

(a) As part of the acquisition, McGill’s acquired ASW’s two depots. The Inchinnan depot could in principle make it easier for McGill’s to enter the north-west part of the Glasgow urban area where neither McGill’s nor ASW were present pre-acquisition. However, although it is closer to that area than McGill’s Barrhead depot, and is closer to the Erskine Bridge, it is not much closer to the Clyde Tunnel. McGill’s confirmed to us that this would not have a material impact on the ability to expand outside the Renfrewshire area.

(b) The existence of network effects might make it easier for McGill’s to expand outside the Renfrewshire area. Some passengers might be more likely to be attracted to an additional McGill’s service because of McGill’s larger and more extended network post-acquisition. The extent to which this is significant depends on whether the passengers that would use that service particularly valued the possibility of travelling across a wider network. Given the degree of pre-acquisition overlap between ASW and McGill’s, the acquisition would not sub-stantially extend McGill’s network.

(c) McGill’s larger size post-acquisition may make it easier to finance further expan-sion and/or be less concerned about the reaction of incumbents if it entered in head-to-head competition with them. On the former, we noted that a decision to expand would depend on the expected profitability of the business opportunity considered. The larger size of McGill’s post-acquisition would not necessarily lower its cost of capital in the short run given the impact of the scheduled pay-ments for ASW’s operations. On the latter, we noted that McGill’s entered in competition with First Glasgow’s service 9 pre-acquisition with its new service 38 and that after a strong initial reaction by First Glasgow the latter reduced its frequencies (see paragraph 7.92). Therefore it is unclear to us whether, as a result of the acquisition, McGill’s would further increase its ability to withstand the incumbents’ reaction to entry.

(d) First Glasgow also told us that it considered that McGill’s would initially be focused on integration with ASW but later, as it would face no competition in the Renfrewshire area, it could be slightly more focused on outward expansion.

7.125 In any case, even if the likelihood of entry outside Renfrewshire was increased by the acquisition, we considered that this would take place mostly by either picking up some routes abandoned by First Glasgow or taking over smaller operators that could no longer operate profitably. We considered that any impact on passengers of such activity was likely to be small.

Conclusions on the loss of potential competition

7.126 We conclude that there will be no loss of potential competition from the acquisition either within or outside the Renfrewshire area.

8. Conclusions on the substantial lessening of competition test

8.1 We conclude that there is no substantial lessening of competition as a result of the acquisition of ASW by McGill’s.

McGill’s Bus Services Limited/Arriva Scotland West Limited ...webarchive.nationalarchives.gov.uk/20140402141250/http:/www... · McGill’s Bus Services Limited/Arriva Scotland West

Documents