(1)

On 13 December 2019, SJ AB (‘the applicant’) submitted to the Commission a request pursuant to Article 35(1) of Directive 2014/25/EU (‘the request’). The request complies with Article 1(1) of Commission Implementing Decision (EU) 2016/1804 (2).

(2)

The request concerns railway passenger transport in Sweden. In particular, the request covers railway passenger transport that constitutes an activity relating to the operation of networks in accordance with Article 11 of Directive 2014/25/EU, and it is not intended to cover procurement by regional public transport authorities, national public transport authorities or other authorities (3).

(3)

Section 24 of Chapter 3 of the Swedish Act on public procurement in the utilities sectors (4) allows contracting entities to submit requests under Article 34 of Directive 2014/25/EU. The applicant is a contracting entity in accordance with Article 4(2) of Directive 2014/25/EU and it carries out an activity relating to operation of networks providing a service to the public in the field of transport by railway, within the meaning of Article 11 of that Directive.

(4)

The request was accompanied by reasoned and substantiated positions adopted by two independent national authorities that are competent in relation to the activities concerned – the Swedish Transport Agency and the Swedish Competition Authority. Those positions thoroughly analyse the condition for the applicability of Article 34(1) of Directive 2014/25/EU to the activities concerned, in accordance with paragraphs 2 and 3 of that Article. In accordance with point 1(b) of Annex IV to Directive 2014/25/EU, given that free access to the market cannot be presumed on the basis of the first subparagraph of Article 34(3) of that Directive, the Commission is to adopt an Implementing Decision on the request within 130 working days. The initial deadline expires on 3 July 2020 (5).

(5)

The Commission held two conference calls with the representatives of the applicant on 30 March 2020 and 29 May 2020 and the applicant made further submissions on 4 May and on 4 June 2020.

(6)

The applicant is a public undertaking, owned 100 % by the Swedish State which was created in 2001 as one of the six divisions formed by the separation and corporatization of the former historic state owned railways company Affärsverket Statens Järnvägar. The applicant is active in the railway passenger transport sector.

(7)

In 2018, it reported 31,8 million journeys (6) indicating that it operated daily 1200 departures from 284 stations (7). It mainly operates within Sweden, but also runs trains to Oslo, Halden, Narvik and Copenhagen.

(8)

Applicant’s business model has two pillars: providing commercial rail services under its’s own brand and providing rail services procured by regional and national public transport authorities under its own brand or that of the procuring authority (8).

(9)

Rail travel in Sweden is increasing due to population growth and urbanisation, deregulation and internationalisation and increased environmental awareness. Growth in travel compared to the previous year, was 2 percent in 2018 (9) passenger-kilometres.

(10)

Directive 2014/25/EU applies to the award of contracts for the pursuit of activities related to the provision or operation of networks providing a service to the public in the field of transport by railway, automated systems, tramway, trolley bus, bus or cable, unless the activity is exempted pursuant to Article 34 of that Directive.

(11)

Pursuant to Article 34 of Directive 2014/25/EU, contracts intended to enable the performance of an activity to which that Directive applies are not to be subject to that Directive if, in the Member State in which it is carried out, the activity is directly exposed to competition on markets to which access is not restricted. Direct exposure to competition is assessed on the basis of objective criteria, taking account of the specific characteristics of the sector concerned. (10) This assessment is, however, limited by the short deadlines applicable and by the need to rely on the information available to the Commission – either from already available sources or from the information obtained in the context of the application pursuant to Article 35 – which can not be supplemented by more time consuming methods, including, in particular, public inquiries addressed to the economic operators concerned (11).

(12)

Access is deemed to be unrestricted if the Member State has implemented and applied the relevant Union legislation opening a given sector or a part of it. That legislation is listed in Annex III to Directive 2014/25/EU. For the domestic railway services, that Annex does not list any relevant legislation liberalising this sector. Consequently, pursuant to Article 34(3) of Directive 2014/25/EU, free access to the market cannot be presumed, and it must be demonstrated de facto and de jure.

(13)

Direct exposure to competition should be evaluated on the basis of various indicators, none of which is necessarily on its own decisive. In respect of the market concerned by this Decision, market shares constitute one criterion which should be taken into account, along with other criteria, such as entry barriers or intermodal (12) competition.

(14)

This Decision is without prejudice to the application of the rules on competition and to other fields of Union law. In particular, the criteria and the methodology used to assess direct exposure to competition under Article 34 of Directive 2014/25/EU are not necessarily identical to those used to perform an assessment under Article 101 or 102 of the Treaty on the Functioning of the European Union or under Council Regulation (EC) No 139/2004 (13) as confirmed by the General Court (14).

(15)

The aim of this Decision is to establish whether the services concerned by the request are exposed to such a level of competition (in markets to which access is not restricted within the meaning of Article 34 of Directive 2014/25/EU) which will ensure that, also in the absence of the discipline brought about by the detailed procurement rules set out in Directive 2014/25/EU, procurement for the pursuit of the activities concerned will be carried out in a transparent, non-discriminatory manner based on criteria allowing purchasers to identify the solution which overall is the economically most advantageous one.

(16)

Access to a market is deemed to be unrestricted if the Member State concerned has implemented and applied the relevant Union legislation opening a given sector or a part of it. That legislation is listed in Annex III to Directive 2014/25/EU. As regards the rail services, that Annex lists Directive 2012/34/EU of the European Parliament and of the Council (15) rail freight transport and for international rail passenger transport but contains no entry for national passenger transport. Consequently, it has to be demonstrated that access to the market is free de facto and de jure.

(17)

Directive (EU) 2016/2370 of the European Parliament and of the Council (16) extends the liberalisation of the rail sector to domestic rail.

(18)

Sweden has not yet fully transposed (17) Directive (EU) 2016/2370 in its national law – the Railways Act (18).

(19)

In January 2019 the Commission addressed a letter of formal notice to Sweden concerning infringement procedure 2019/0087 for partial transposition of Directive (EU) 2016/2370. On 17 May 2019, Sweden notified its 12th transposition measure for that Directive claiming that it had concluded the transposition. However, on 13 December 2019 (date of receipt of the request) and until June 2020, the case was pending.

(20)

In its position (19) accompanying the request, the Swedish Transport Agency examined whether access to the relevant market is free de facto and de jure, and concluded that this is the case.

(21)

As far as free access de jure is concerned, Sweden has extended its national provisions transposing Directive 2012/34/EU so as to provide for free market access also to domestic rail passenger transport. This kind of transport operations is thus covered by the scope of the Swedish Railways Act [2004:519].

(22)

As specified by the Swedish Transport Agency in this context, the Railways Act requires infrastructure managers to allocation infrastructure capacity to railway undertakings in a competition-neutral and non-discriminatory manner.

(23)

Moreover, it follows from the Railway act (Chapter 6, Section 5) that the details of the available infrastructure and information on the terms for access to and use of the infrastructure and on the procedures and criteria for allocating infrastructure capacity are to be published in the network statement, in accordance with Article 27(2) of Directive 2012/34/EU.

(24)

It can be concluded that the conditions of free market access de jure are met.

(25)

As far as free access de facto is concerned, the Commission notes that competitors started to enter the Swedish railway market progressively already since 1990. Today, along with the applicant, there are other fifteen different companies active in the domestic railway passenger market.

(26)

Based on the legal analysis of the relevant national legislation and the position of the Swedish Transport Agency, the Commission concludes that, while some of the concerns identified in the infringement procedure 2019/0087 still persist, and without prejudice to that infringement procedure, access to the market is be considered de jure and de facto free on the territory of Sweden for the purposes of this Decision.

(27)

In previous merger decisional practice (20), the Commission has found that it may be appropriate to consider the market for railways passenger services under public service obligation (PSO), and the market for commercially operated railways passenger services as distinct markets. This distinction is due to the fact that transport services operated under PSO contracts are characterised by competition for the market, meaning that rail undertakings are competing to be awarded the right to offer transport services on a given route, whereas commercial transport services are characterised by competition in the market, meaning that rail undertakings compete to attract passengers.

(28)

For the purposes of this Decision a route is understood as encompassing all journeys between any of the stations along that route (i.e. all point-to-point combinations along that route). Likewise, a sub-route is understood as a segment of a route, encompassing all journeys between the stations along that sub-route (i.e. all possible point-to-point combinations along that sub-route). End-to-end travel is understood as a (point-to-point) journey between two end stations of a given route.

(29)

The applicant proposes to make this distinction between these two markets as referred to in recital 27.

(30)

The Swedish National Competition Authority (NCA) and the Swedish Transport Agency, in their respective positions on the request, also agree (21) with this distinction.

(31)

PSO contracts are awarded through competitive tender processes, or by means of direct award, through which a contracting entity awards a bidder a contract to operate railway passenger services in a certain area or on one or several specific routes.

(32)

Regional and inter-regional PSO contracts are tendered by regional public transport authorities, while the Swedish Transport Administration manages the tender process for subsidized night trains between Stockholm and northern Sweden.

(33)

The Swedish Transport Agency indicates (22) that the awarding entities provide the trains and are responsible for their maintenance.

(34)

For the purposes of the assessment under this Decision and without prejudice to the application of competition law, the Commission considers that the first relevant market is the market for railway passenger services under a PSO, where competition occurs ‘for the market’.

(35)

The applicant argues that, in Sweden, commercially operated railways passenger services can, in principle, be offered by any operator and cover any routes.

(36)

The applicant claims that it faces competition from PSO operators on the commercially operated routes.

(37)

However, to the understanding of the Commission, commercially operated railway passenger services are generally long-distance transport services. In those circumstances, it is likely that PSO operators providing long-distance railway passenger services compete with commercially operated railway passenger services, to attract passengers. The reason is that, on the demand-side, railway passenger services operated under a PSO and operated commercially may be substitutable. However, it may be less obvious that PSO operators providing regional railway passenger services compete with commercially operated railway passenger services. In its decisional practice, the Commission has found (23) that regional (short distance) and long-distance railway passenger services were distinct markets, notably because the characteristics of the traffic are different, for example, in terms of frequencies and stops.

(38)

[Confidential] (24).

(39)

In this case, however, the Commission considers that market definition can be left open given that the result of the assessment is the same in both scenarios, (i.e. whether commercially operated railway services are considered to be in the same market as long-distance PSO and regional PSO railway services, or whether they are considered to be in the same market only with long-distance PSO railway services).

(40)

In the case of commercial services, it is nevertheless important to clarify if and to what extent intermodal competition is to be taken into account. The applicant claims that competition between railway operators is sufficient to conclude that commercially operated railway passenger services are directly exposed to competition, without having to assess inter-modal competition (25). In this case, however, the exact market definition cannot be left open, as proposed by the applicant, as for some routes the applicant’s market shares indicate a significant market power (if only rail transport is considered), and therefore the ability for the applicant to behave independently of its competitors.

(41)

As regards intermodal competition, the Commission has found (26) in the past that other modes of transport were distinct markets. However, this is ultimately an empirical issue and the Commission has left the definition open in other cases (27).

(42)

Competitive pressure from other modes of transport is further analysed in based on time of travel, prices, comfort and availability. As a general remark, the Commission notes that air travel cannot be presumed to exert competitive pressure on all routes or sub-routes, but potentially only on the end-to-end travel, therefore the economic importance of the end-to-end-travel will be taken into account. Moreover, the competitive assessment shows that other modes of transport are generally not substitutable with railway passenger services operated commercially see section 3.2.3.2).

(43)

For the purposes of the assessment under this Decision, and without prejudice to the application of competition law, the Commission considers that the relevant market is the market for commercially operated railway passenger services where competition occurs ‘in the market’. For the purpose of this decision, this market includes long-distance railway passenger services operated under a PSO. The Commission leaves it open whether regional railway passenger services operated under a PSO are part of the relevant market.

(44)

In past decisions, the Commission has defined (28) this market at the national level.

(45)

The applicant’s proposed market definition is in line with Commission’s practice.

(46)

The NCA also agrees with this approach.

(47)

For the purposes of the assessment under this Decision, and without prejudice to the application of competition law, the Commission considers that the geographic scope of the market for railway passenger services offered through PSO contracts is national.

(48)

In its previous antitrust and merger decisional practice relating to rail passenger transport services, when analysing competition in the market, the Commission has defined the geographic market at the level of an entire national railway network (29), at the level of a route (30), or at the level of a point-to-point travel between an origin and a destination (31), depending on the specificities of each case and on the level at which the conditions of competition are similar or sufficiently homogeneous.

(49)

The applicant submits that the market is delineated on a point of origin to a point of destination basis, as competitive conditions can vary significantly across such point-to-point delineations. The applicant also proposes to leave the market definition open, as the exposure to competition has to be evaluated on a wider basis, since railway operators’ costs are not determined on a point-to point basis but at a wider level. Given the very high number of point-to-point markets, and the impossibility to analyse each of those markets, the applicant proposes to analyse the five most important routes in Sweden (i.e.: Stockholm-Malmö(-Copenhagen), Stockholm-Gothenburg, Gothenburg-Malmö, Stockholm-Sundsvall-Umeå and Stockholm-Karlstad(-Oslo)).

(50)

[Confidential].

(51)

[Confidential].

(52)

[Confidential].

(53)

The Commission notes that there are indications that the market for commercially operated railway passenger services in Sweden can be defined at the national level. First, the regulatory framework (such as the requirements to operate as a rail undertaking) is nationwide. Therefore, conditions of competition are likely to have a certain degree of homogeneity at the national level. Secondly, some providers (such as the applicant or Transdev) compete against each other on different routes in Sweden and they are likely able to deploy their trains on various routes in Sweden. Thirdly, the characteristics inherent to the railway network must be taken into account. In particular, the existence of network effects, that is, the generation of additional traffic by the linkage between various routes, is also an element pointing to the existence of a nation-wide market.

(54)

In view of the foregoing, for the purposes of the assessment under this Decision and without prejudice to application of competition law, the Commission leaves it open whether the geographic scope of the market for commercially operated railway passenger services is national or corresponds to each route since under either of the two possible geographic market definitions the conclusion reached by the Commission is not affected.

(55)

A market definition based on point-to-point travel is excluded, considering that, as explained by the applicant and the NCA (32), it would not be possible to assess the competitive pressure for each individual point-to-point travel that can be potentially defined.

(56)

PSO contracts can be awarded through competitive tenders or through direct award. In general, in Sweden, PSO contracts are awarded following a tender procedure and the direct awards are the exception to that rule.

(57)

The applicant reports (33) on a list of contracts for PSO awarded since 2009. Out of the 20 contracts listed, 3 were directly awarded. Directly awarded contracts represent just [Confidential] of all PSO contracts, in terms of value. The rest of the contracts are competitively tendered. It can be observed that Arriva, Transdev, Vy and MTR are regular bidders along with the applicant. The majority of tender procedures have between 2 and 4 bidders and on the majority of cases, the winner is different from the incumbent.

(58)

Out of 17 PSO competitively tendered contracts, the applicant bid for 11 and won 6 (success rate of 54 %). Other competitors performance shows the following ratios of (win/bids): Arriva 4/7, Transdev 2/9, Keolis 0/4, Vy 3/8 MTR 1/2. The applicant seems to have an average rate of success.

(59)

In terms of revenues, the applicant obtained [Confidential] of the total number of PSO contracts and [Confidential] in terms of value.

(60)

Overall, the applicant appears (34) to hold a moderate share of PSO contracts in Sweden [Confidential] with the second and third largest operators having shares of [Confidential]and [Confidential], respectively. Market shares are however less indicative on the nature of competition in markets subject to tenders or direct awards.

(61)

[Confidential].

(62)

[Confidential]. Moreover, out of the 16 procured PSO contracts 11 changed the operator following tender procedure. In the 15 calls for tenders for which data on all bidders were available, a total of 46 bids were submitted.

(63)

Out of the reported three directly awarded PSO contracts, all three appear (35) to be currently subject to a competitive tender procedure following which new companies will be awarded and start operation.

(64)

[Confidential].

(65)

The NCA concludes (36) that the market for PSO contracts, ‘with the exception of directly awarded contracts to own companies’, is directly exposed to competition in the meaning of Article 34 of Directive 2014/25/EU.

(66)

In order to be successful in a tender for a PSO contract, the railway operators have to present competitive bids. This implies that once a contractor is awarded a PSO contract, its behaviour is constrained by its bid. Consequently, procurement for the pursuit of the activities under the PSO contract, even in the absence of the discipline brought about by the detailed EU procurement rules, will be carried out in a transparent, non-discriminatory manner based on criteria allowing the PSO operator to identify the solution which overall is the economically most advantageous one.

(67)

For the purposes of this Decision and without prejudice to competition law or other fields of Union law, the factors described above, should be taken as an indication of exposure to competition of the activity of provision of railway passenger services offered through a PSO.

(68)

On the overall market, covering commercially operated railway passenger services and railway passenger services operated under PSO contracts (including regional and long-distance railway passenger services operated under a PSO) the applicant had, in terms of revenues a market share of [Confidential]in 2018. In the same year, the next market player was [Confidential] The applicant’s market share constitutes an indication of significant market power.

(69)

On the market for commercially operated railway passengers services (including long-distance, but not regional, railway passenger services operated under a PSO), the applicant holds a significantly larger share than its competitors, which has remained rather stable over the last three years reported (2016 to 2018). In 2018, applicant had [Confidential] in terms of revenues and [Confidential] in terms of passenger-kilometres. Therefore, the applicant’s market shares in the market for commercially operated railway passenger services are also indicative of significant market power.

(70)

The Commission also notes the existence of barriers to entry in the market for commercially operated railways passenger services. These barriers concern notably the necessary high investment costs and associated economic risks for the acquisition of the rolling stock, and technical and regulatory requirements, such as for homologation of rolling stock in a country. [Confidential]. These arguments are valid for the market for commercially operated rail passenger transport services, irrespective of the geographic market definition.

(71)

As regards intermodal competition, in view of the observed differences in terms of price and travel time, it cannot be concluded that travel by air or by bus exert competitive pressure on commercially operated railway passenger services as evidenced below in recitals 93, 102, 120, 126, 149, 168, 174, 189 and 195.

(72)

In view of the factors examined above, the Commission cannot conclude that the activity of provision of commercially operated railway passenger services at national level in Sweden, regardless of how the product market is defined, is directly exposed to competition in the meaning of Article 34(1) of Directive 2014/25/EU.

(73)

In the market analysis at route level (focussing on each of the above-identified five routes), the Commission will first examine the competitive situation when (i) commercially operated railway services are assumed to compete with long distance PSO and regional PSO railway services for passengers, and (ii) commercially operated railway passenger services are assumed to compete with long-distance PSO railway services. Second, it will also be examined if and to what extent competitive pressure from other modes of transport exists in each of the above-mentioned routes.

(74)

[Confidential].

(75)

A. If commercially operated railway passenger services are assumed to compete with long-distance PSO and regional PSO railway services, the situation is as described in the following recitals.

(76)

On this route the applicant has one competitor (operating commercial end-to-end train) Snälltåget (Transdev), who entered the market in 2009. Regional PSO services are operating different sub-routes.

(77)

Transdev has two daily departures compared to the applicant, which has 15 daily departures (plus the night train, which is, however, under a PSO contract). Travel time of Transdev is between 4h 52 minutes and 5h, while the applicant has travel time below 4h 33 minutes. The Commission also notes that the prices of Transdev are much lower than those of the applicant. Moreover, Transdev does not run all the way to Copenhagen and the applicant is currently the only train operator who runs from Stockholm to Copenhagen (37).

(78)

The applicant’s market shares are indicative of significant market power on this route. [Confidential].

(79)

[Confidential].

(80)

[Confidential].

(81)

[Confidential].

(82)

[Confidential].

(83)

[Confidential].

(84)

[Confidential].

(85)

[Confidential].

(86)

[Confidential].

(87)

The applicant and the NCA indicate that in spring 2020 FlixTrain was planning to enter with two daily departures commercially operated trains on the low-cost segment on Stockholm-Malmö route. However, the Commission note that to date FlixTrain does not appear to operate trains on this route. Moreover, given the context of COVID-19 pandemics, it is highly uncertain whether the potential competitive threat by FlixTrain is to materialise in the near future.

(88)

B. If commercially operated railway passenger services are assumed to compete only with long-distance PSO railway passenger services, the Commission notes that the applicant has only one competitor – Transdev, which provides commercial services. There are no long-distance PSO competitors on this route. The applicant enjoys thus an even stronger market position [Confidential].

(89)

The NCA concludes in its position that ‘whilst the competitive pressure on the route in question is such that the applicant is unable to act wholly independently in its pricing practice, it is not sufficiently strong, in the Competition Authority’s position, to conclude that the route is directly exposed to competition’ (38).

(90)

In view of the factors examined above, and if only passenger transport is taken into account, under each possible product market configuration, the Commission cannot conclude that the activity of provision of commercially operated trains on the Stockholm-Malmö (-Copenhagen) route is exposed to competition.

Intermodal competition – Bus travel

(91)

The applicant estimates that, when bus and railway transport services are considered, the intermodal market share for bus services on this route is [Confidential].

(92)

The Commission notes that the travel time by bus is of minimum 9 hours (compared to less than 5 hours by train). FlixBus has two (direct) daily departures to Malmö (and two direct daily departures to Copenhagen) and the prices are between 10 % and 20 % cheaper than the prices of the applicant (second class).

(93)

In view of the differences in travel time, it cannot be concluded that travel by bus exerts competitive pressure on commercially operated railway passenger services on this route.

Intermodal competition – Air travel

(94)

Three airlines companies offer direct flights between Stockholm airports Arlanda (ARL) and Bromma (BRM) and the airports in the Öresund region: Copenhagen (CPH) and Malmö (MMX).

(95)

Arlanda airport is situated 40 km form Stockholm city centre.

(96)

Travel time by air is of 1h 10 minutes (to Copenhagen) and 1h (to Malmö), compared to a travel time by train of 4h 38 minutes (to Copenhagen) and 4h 22 minutes (to Malmö).

(97)

Scandinavian Airlines System (SAS) has eight direct flights to Malmö (ARN-MMX) and 15 daily departures to Copenhagen: (ARN-CPH).

(98)

Norwegian Air Shuttle ASA (Norwegian) has six daily departures to Copenhagen (ARN-CPH)

(99)

BRA Sverige AB (BRA) has 13 daily departures to Malmö (BMA-MMX).

(100)

[Confidential].

(101)

The Commission performed an analysis of actual ticket prices, by train and airplane for this route, and analysed the respective information provided by the Applicant. Prices of the plane tickets are almost double compared with train tickets for STO-MMX, and STO-CPH, for first class and second class, with the exception of Norvegian for STO-CPH only.

(102)

Given the differences observed in terms of travel time, and ticket prices, the Commission cannot conclude that travel by air exert competitive pressure on commercially operated railway passenger services on this route.

Conclusions

(103)

In view of the factors examined above, and even considering inter-modal competition, the Commission cannot conclude that the activity of provision of commercially operated railway services on the Stockholm-Malmö (-Copenhagen) route is directly exposed to competition.

(104)

[Confidential].

(105)

A. If commercially operated railway services are assumed to compete with long-distance PSO and regional PSO railway services for passengers, the situation is as described in the following recitals.

(106)

The commercial operators on this route are the applicant and MTR (end-to-end train). AB Galia (Blå Tåget) discontinued commercial operations on this route in 2019. Regional PSO services are operating different sub-routes.

(107)

MTR has eight (high-speed) daily departures while the applicant has 18 daily departures (high speed) and there are another seven daily departures by PSO regional trains.

(108)

Travel times for the high-speed trains of the applicant and MTR are comparable (3h 05 minutes and 3h 20 minutes respectively), while the applicant’s regional trains have travel times of around 4h 50 minutes. Ticket prices are comparable, with MTR having in general slightly lower prices for both first class and second-class tickets.

(109)

The Applicant’s market shares are indicative of significant market power. [Confidential].

(110)

[Confidential].

(111)

[Confidential].

(112)

[Confidential].

(113)

[Confidential].

(114)

The NCA lists (39) two more factors, which, in its view, led to the conclusion that the activity of the applicant is exposed to competition on this route: the future start of operation of FlixTrain and the pressure from air travel.

(115)

The Commission understands that FlixTrain has been allocated slots for three daily departures from the spring of 2020 on this route. However, the Commission notes that to date FlixTrain does not appear to operate trains on this route. Moreover, given the context of COVID-19 pandemics, it is highly uncertain whether the potential competitive threat by FlixTrain is to materialise in the near future.

(116)

B. If commercially operated railway passenger services are assumed to compete only with long-distance PSO railway passenger services the Commission notes that the situation is very similar with the one described in recital 109, as other regional PSO services accounts for less than [Confidential] in terms of revenues and passenger-kilometres.

(117)

In view of the factors examined above, and if only rail passenger transport is taken into account, under each possible product market configuration, the Commission cannot conclude that the activity of provision of commercially operated railways services on the Stockholm-Gothenburg route is exposed to competition.

Intermodal competition – Bus travel

(118)

The applicant estimates that when bus and railway transport services are considered, the overall intermodal market share for bus services on this route is [Confidential].

(119)

The Commission notes that the travel time by bus is of minimum 6h 30 minutes (as compared to 3h 20 minutes by train). FlixBus has six (direct) daily departures to Gothenburg and the prices are than 30 % cheaper than the prices of second class tickets of the applicant.

(120)

In view of the above, despite the price differences, the differences in travel time, do not allow to conclude that travel by bus exert competitive pressure on commercially operated railway passenger services.

Intermodal competition – Air travel

(121)

Three airline companies offer flights between Stockholm airports Arlanda and Bromma and Gothenburg (GOT) airport.

(122)

SAS has 12 daily departures (ARN-GOT), Norwegian has one or two daily departures (ARN-GOT) and BRA has 12 daily departures (BMA-GOT).

(123)

According to the NCA, transport by air should be considered to exert ‘top-down’ pressure in the premium segment, as the travel time for city centre to city centre is the same which is an important factor for business travellers.

(124)

Direct flight travel time is one hour compared to three hour travel time by train.

(125)

The Commission performed an analysis of actual ticket prices, by train air for this route, and analysed the respective information provided by the applicant. Prices of plane tickets are double the price of train tickets for first class and second-class tickets, with the exception of Norwegian, for second class tickets, where differences in prices are less significant.

(126)

In view of the observed differences in prices, it cannot be concluded that travel by air exert competitive pressure on commercially operated railway passenger services on this route.

Conclusions

(127)

In view of the factors examined above, and even considering intermodal competition, the Commission cannot conclude that the activity of provision of commercially operated railway services the Stockholm-Gothenburg route is directly exposed to competition.

(128)

[Confidential]

(129)

A. If commercially operated railway services are assumed to compete with long-distance PSO and regional PSO railway services for passengers, the situation is described in the following recitals.

(130)

The applicant has currently seven daily departures (commercial end-to-end train) with high-speed trains. Transdev, which operates Öresundståg under a long-distance PSO contract, offers 11 daily departures. Travel time for applicant’s high-speed train is of 2h 30 minutes while Transdev’s train has 3h 10 minutes travel time.

(131)

However, from December 2020 the applicant will take over the long-distance PSO contract Öresundståg and will operate all end-to-end trains on this route following the award of the respective PSO contract for the period 2020 to 2030. Current PSO contract with Transdev expires beginning of December 2020.

(132)

In terms of prices, the Commission notes that, despite having lower travelling times on average, the applicant has lower prices than Transdev (under the PSO contract) both for first and second-class tickets.

(133)

The third competitor on this route is Arriva (Pågatåg), which operates within the southern Skåne region only under a regional PSO contract.

(134)

[Confidential].

(135)

[Confidential].

(136)

[Confidential].

(137)

From December 2020, when the applicant will start operations under the PSO contract on this route, the situation is likely going to change, resulting in the consolidation of its market position. [Confidential].

(138)

The NCA indicates that while it may be debatable that Transdev exerts genuine competitive pressure on the premium segment for end-to-end travel, it noted that SAS offers flights between Gothenburg and Copenhagen which should have impact on applicant’s pricing. The NCA concluded (40) that on this route the activity is directly exposed to competition.

(139)

B. If commercially operated railway passenger services are assumed to compete only with long-distance PSO railway services, the Commission notes that the applicant has only one competitor, Transdev, which provides services under a long-distance PSO contract (Öresundståg) [Confidential].

(140)

In view of the factors examined above, and if only rail passenger transport is taken into account, under each possible product market configuration, the Commission considers that currently the activity of provision of commercially operated railway services on the Gothenburg-Malmö route can be deemed to be directly exposed to competition. However, the situation would need to be re-evaluated once the applicant will start operating the Öresundståg train network.

Intermodal competition – Bus travel

(141)

The Applicant estimates that, when bus and railway transport services are considered, the overall intermodal market share for bus services on this route is [Confidential].

(142)

The Commission notes that the travel time by bus is of 3h 5 minutes and 3h 30 minutes as opposed to 2h 30 minutes an applicant’s train. FlixBus has four (direct) daily departures to Malmö and the prices are at least 25 % cheaper than those of the applicant (second-class tickets).

(143)

In view of the above differences in travel time and taking into account the price difference, it cannot be excluded that travel by bus may exert competitive pressure on commercially operated railway passenger services on this route. However, this question can be left open for the purposes of this Decision, since, regardless of whether travelling by bus competes with train travelling, the conclusion of the Commission on the level of competition in this route does not change.

Intermodal competition – Air travel

(144)

[Confidential]. One possible explanation is that there are no direct flights from Gothenburg to Malmö. However, SAS has five direct flights in winter season and seven direct flights in summer season from Gothenburg to Copenhagen.

(145)

Travel time by air is of 2 h 35 minutes (GOT-MMX), or 45 minutes (GOT-CPH). No direct flight is available on GOT-MMX route, at least one stopover is taken into account for that route.

(146)

According (41) to the NCA, while it is not clear that Öresundståg exerts genuine competitive pressure on the applicant premium segment, air transport by SAS should have an impact on its prices. However, the Commission notes that SAS has no direct flight between Gothenburg and Malmö. All flights between Gothenburg and Malmö involve at least one stopover.

(147)

[Confidential].

(148)

The Commission performed an analysis of actual ticket prices, by train, air for this route, and analysed the respective information provided by the applicant. Travel by air appears to be twice more expensive than railway travel, for both, first and second class.

(149)

In view of the observed differences in prices, the Commission cannot support the conclusion that travel by air exerts competitive pressure on commercially operated railway passenger services on this route.

Conclusion

(150)

In view of the factors examined above, the Commission concludes that the activity of providing commercially operated railways passenger services on the Gothenburg-Malmö route is directly exposed to competition. However, the situation would need to be re-evaluated once the applicant will start operating the Öresundståg train network (currently operated by Transdev under the PSO contract).

(151)

[Confidential].

(152)

A. If commercially operated railway services are assumed to compete with long-distance PSO and regional PSO railway services for passengers, the situation is described in the following recitals.

(153)

This route is currently operated (commercial end-to-end train) by the applicant only. Regional PSO services are operating different sub-routes.

(154)

[Confidential].

(155)

[Confidential].

(156)

[Confidential].

(157)

[Confidential].

(158)

[Confidential].

(159)

[Confidential].

(160)

[Confidential].

(161)

The NCA concludes that the entire route is not exposed to competition, but given the competitive situation on sub-routes and their overall economic importance compared to the entire route, the applicant is unable to act completely independently in its pricing.

(162)

[Confidential].

(163)

B. If commercially operated railway passenger services are assumed to compete only with long-distance PSO railway passenger services, the Commission notes that the applicant is the only operator of commercial railways passenger services on this route and there are no competitors providing long-distance PSO railway services on this route.

(164)

In view of the factors examined above, and if only rail passenger transport is taken into account, under any possible product market configuration, the Commission cannot conclude that the activity of provision of commercially operated railway services on the Stockholm-Sundsvall-Umeå route is exposed to competition.

Intermodal competition – Bus travel

(165)

The Commission notes that (end-to-end) travel time by bus is of more than 10 hours, compared to 6h 20 minutes by high-speed train.

(166)

[Confidential].

(167)

[Confidential].

(168)

In view of the above differences in travel time, it cannot be concluded that travel by bus exerts a competitive pressure on commercially operated railway passenger services on this route.

Intermodal competition – Air travel

(169)

The NCA indicates (42) that this route is too long and travel times too long to allow train travel to compete with air travel.

(170)

SAS provides six daily departures (ARN-UME), Norwegian – five daily departures (ARN-UME) and BRA – five daily departures (BMA-UME).

(171)

The Commission performed an analysis of actual ticket prices, by train, air and bus for this route, and analysed the respective information provided by the applicant.

(172)

Travel time by air is of one hour while travel by (high-speed) train is of minimum 6h 20 minutes.

(173)

Flight ticket prices between Stockholm and Umeå are more expensive than train tickets for the first class tickets. This is also true for second-class tickets, except for Norwegian.

(174)

In view of the above differences in prices and travel time, and given [Confidential], it cannot be concluded that travel by air exert competitive pressure on commercially operated railway passenger services on this route.

Conclusions

(175)

In view of the factors examined above, and even if intermodal competition is considered, the Commission cannot conclude that the activity of providing commercially operated railway passenger services on the Stockholm-Sundsvall-Umeå route is directly exposed to competition.

(176)

[Confidential].

(177)

A. If commercially operated railway services are assumed to compete with long-distance PSO and regional PSO railway services for passengers, the situation is described in the following recitals.

(178)

Travel time from Stockholm to Karlstad is 2h 30 minutes (by high-speed train), and from Stockholm to Oslo is minimum 6 hours (by intercity train). The applicant has eight daily direct trains to Karlstad and two daily direct trains to Oslo.

(179)

This route is currently operated (commercial end-to-end train from Stockholm to Oslo) by the applicant only. Tagab is providing services under a PSO contract between Stockholm and Karlstad.

(180)

The applicant has significant market power according to its market shares. [Confidential]

(181)

[Confidential].

(182)

[Confidential].

(183)

The NCA considers (43) that the route is not directly exposed to competition. However, NCA indicates that both flight and buses exert certain ‘top-down and bottom up price pressure in the same way as for Stockholm Malmö route.’

(184)

B. If commercially operated railway passenger services are assumed to compete only with long-distance PSO railway passenger services, the Commission notes that, the applicant has only one competitor – Tagab, which however does not provide end-to-end services. The applicant enjoys thus an even stronger market position [Confidential].

(185)

In view of the factors examined above, and if only rail passenger transport is taken into account, under any possible product market configuration, the Commission cannot conclude that the activity of provision of commercially operated railway passenger service on the Stockholm-Karlstad (-Oslo) route is exposed to competition.

Intermodal competition – Bus travel

(186)

The Applicant estimates (44) on the overall Stockholm-Karlstad (-Oslo) route, bus services hold [Confidential].

(187)

The Commission notes that Flixbus has three daily departures from Stockholm to Karlstad, where travel time is of 4h 20 minutes (as compared to 3h 50 minutes by train) and the price appears to be half of the price of the second-class train tickets.

(188)

Flixbus also has two direct daily departures from Stockholm to Oslo, where travel time is of 7h 30 minutes (as compared to 6h by train) and the prices are around 30 % cheaper than the second-class train tickets.

(189)

In view of the above overall differences in travel time, and characteristics (i.e. such as comfort) it cannot be concluded that travel by bus exerts a competitive pressure on commercially operated railway passenger services on this route.

Intermodal competition – Air travel

(190)

The applicant argues that the competition from air transport is strong on the end-to-end route.

(191)

SAS has 14 daily departures to Oslo (OSL): ARN-OSL and Norwegian has seven daily departures: ARN-OSL. There are no flights from Stockholm to Karlstad (KAR).

(192)

Travel time by air is of 1 hour 10 minutes while travel by train is of 6 hours.

(193)

The Commission performed an analysis of actual ticket prices, by train, air or this route, and analysed the respective information provided by the applicant.

(194)

Flight tickets between Stockholm and Oslo are more expensive than train tickets for the first class. This is also true for second-class tickets but the variations are smaller for Norwegian for second-class tickets.

(195)

In view of the observed differences in prices and travel time, it cannot be concluded that travel by air exerts competitive pressure on commercially operated railway passenger services on this route.

Conclusions

(196)

In view of the factors examined above, and even if intermodal competition is considered, the Commission cannot conclude that the activity of provision of commercially operated railways services the Stockholm-Karlstad (-Oslo) is directly exposed to competition.

(197)

For the purposes of this Decision and without prejudice to the application of competition law, the findings of the market analysis listed in recitals 56 to 67 should be taken as an indication of exposure to competition within the meaning of Article 34 of the Directive 2014/25/EU of the activity of provision of railway passenger services under a PSO in Sweden. Consequently, since the conditions of unrestricted access to the market set out in Article 34 of Directive 2014/25/EU are met, it should be established that Directive 2014/25/EU does not apply to contracts intended to enable the pursuit of this activity in Sweden.

(198)

In light of the analysis set out in section 3.2.3.2 above, under each possible market configuration, the Commission cannot conclude that the activity of provision of commercially operated railways services in Sweden is directly exposed to competition within the meaning of Article 34 of Directive 2014/25/EU. At national level under each possible market configuration, the applicant’s market shares are indicative of significant market power.

(199)

On a route level, in four out of the five routes assessed the activity of provision of commercially operated railway passenger services is not directly exposed to competition. On those routes the applicant has market shares [Confidential], when commercially operated railway services are assumed to compete with long-distance PSO and regional PSO railway services. The position of applicant is even stronger when commercially operated railway passenger services are assumed to compete with long-distance PSO railway passenger services only.

(200)

On one route only (Gothenburg-Malmö), the activity of provision of commercially operated trains is directly exposed to competition [Confidential].

(201)

[Confidential].

(202)

In the light of the above considerations, for the purposes of this Decision and without prejudice to the application of competition law, the findings of the market analysis listed in recitals 68 to 196 should be taken as an indication of the fact that the activity of provision of commercially operated railway services in Sweden is not directly exposed competition within the meaning of Article 34 of the Directive 2014/25/EU. Consequently, it should be concluded that Directive 2014/25/EU continues to apply to contracts intended to enable the pursuit of that activity in Sweden.

(203)

Following its assessment, the NCA concluded that two routes (i.e. Stockholm-Gothenburg and Gothenburg-Malmö) are directly exposed to competition. [Confidential]. The NCA also indicates that the competitive pressure is expected to increase on the Stockholm-Malmö (-Copenhagen) route if Flixtrain starts operating this route, but that this is not sufficient to conclude that this route is directly exposed to competition. [Confidential]. As regards the remaining two routes (Stockholm-Umeå and Stockholm-Oslo), the NCA concluded that they are not directly exposed to competition but they are, in the NCA’s view, subject to a degree of competitive pressure which prevents the applicant from acting completely independently in its pricing behaviour. [Confidential].

(204)

The NCA concluded that the market for rail passenger transport services in Sweden, including the market for commercially operated railway services (combining all routes), and the provision of railway services under PSO contracts) should be regarded as directly exposed to competition.

(205)

The Commission does not draw this conclusion: as noted above, the Commission concludes that only Gothenbug-Malmö route is directly exposed to competition. [Confidential]. For the other four routes, the Commission could not conclude that are exposed to competition for the reasons set out above. Consequently, even if was possible to draw conclusions on the level of competition at the national level based on the level of competition on the individual routes analysed, the Commission could not draw the same conclusion as the NCA for the reasons set out in recitals 198 to 202.

(206)

On 4 May 2020 the applicant submitted additional elements to explain why, in its opinion, an exemption covering PSO market and three commercially operated routes Stockholm-Gothenburg, Stockholm-Malmö and Malmö-Gothenburg would be justified (45). Additionally it was submitted that an exemption would be justified also for Stockholm-Uppsala route (46).

(207)

[Confidential]

(208)

The applicant stated that since its ‘purchases jointly cover routes where SJ faced significant competition as well as routes where SJ was the sole railway operator, SJ argued that competition would come to bear also on those routes where SJ did not face competition’ (47).

(209)

[Confidential].

(210)

The Commission considers that even if it would agree with the applicants and the NCA’s approach described in recital 209 [Confidential], it could not have reached the conclusion sought by the applicant. In this regard, the Commission has found that the activity of provision of commercially operated trains is exposed to competition on one route only (i.e. Gothenburg-Malmö). Therefore, the Commission considers that an exemption cannot be granted for that route given [Confidential].

(211)

This means that it is not possible to grant an exemption just for that route, because it cannot be excluded that joint purchasers formally made for that route and exempted from the Directive, could be allocated to the other routes where no direct exposure to competition exists. This conclusion is further corroborated by the explanations below.

(212)

In its opinion, NCA explicitly pointed out that ‘many of the inputs and services being procured are joint costs.’ (48)

(213)

[Confidential].

(214)

On 4 June 2020 the applicant made a new submission to explain why, in its opinion, an exemption for a particular point of origin (e.g. Stockholm) to a particular point of destination (e.g. Gothenburg) would be possible under the Directive.

(215)

The applicant argued that the assessment to establish the applicability of Article 34 of Directive 2014/25/EU shall be made with regards to the activity and the geographical reference market, and that in the current case the reference markets are from a point of origin to a point of destination.

(216)

In the same letter the applicant argues that ‘Hence, the methodology used to evaluate whether an individual route is directly exposed to competition is exactly the same as at the national level. In fact, the assessment of competition at the individual route level constitutes a building block to the assessment of competition at national level’ (49).

(217)

[Confidential].

(218)

As noted in recital 216 above, since the Commission found that only one route [Confidential], out of the five most important routes, was directly exposed to competition, the conclusion cannot be that an exemption should be granted at national level.

(219)

The Commission recalls also that the purpose of the competition analysis under Article 34 of Directive 2014/25/EU is to establish the pertinence of application of the public procurement rules, as explained in recital 15. The level of competition on a route such Gothenburg-Malmö which is [Confidential] for the applicant, cannot be presumed to discipline its procurement, in view of the fact that procurement is operated on a joint basis. Moreover, the applicant did not provide any example of procurement specific to this route only.

(220)

Finally, if the exemption were granted for the Gothenburg-Malmö route only, in the circumstances described above, there would be a high risk that a large part of the procurement formally made for that route and exempted from the provisions of the Directive could be effectively used for the other routes where no direct exposure to competition exits.

(221)

The Commission also notes that the NCA did not suggest an exemption on a route by route basis either.

(222)

For the purposes of this decision and without prejudice to the application of competition law, the findings of the market analysis listed in recitals 68 to 221 should be taken as an indication of the fact that the activity of provision of commercially operated railway services is not exposed competition within the meaning of Article 34 of the Directive 2014/25/EU. Consequently, it should be established that Directive 2014/25/EU continues to apply to contracts intended to enable the pursuit of that activity in Sweden.

(223)

This Decision is based on the legal and factual situation as of December 2019 to June 2020 and on the information submitted by the applicant, the Swedish Transport Agency and the NCA and publicly available information.

(224)

Since the services related to the commercially operated railway passenger services should continue to be subject to Directive 2014/25/EU, it is recalled that procurement contracts covering several activities are to be treated in accordance with Article 6 of that Directive. This means that, where a contracting entity is engaged in ‘mixed’ procurement, that is procurement used to support the performance of both activities exempted from the application of Directive 2014/25/EU and activities not exempted from the application of that Directive, it is necessary to take into account the activities for which the contract is principally intended. In the event of such mixed procurement, where the purpose is principally to support activities which are not exempted, the provisions of Directive 2014/25/EU are to be applied. Where it is objectively impossible to determine for which activity the contract is principally intended, the contract is to be awarded in accordance with the rules laid down in Article 6(3) of Directive 2014/25/EU,