In France, coach travel has been slowly opening up to competition since 2011, when reforms were introduced to allow operators to serve inter-regional routes (previously banned) provided they met all of the following conditions.
The destination served was part of an international route and the service did not jeopardise the public service obligation (PSO) contract (existing rail or coach routes).
Domestic coach transport did not account for more than 50% of total annual coach passengers on the route and 50% of the turnover.
Coaches served only one city per administrative region.
In addition, routes have to be approved by the Ministry of Transport following consultation with the relevant regional and district (‘département’) administrations about the economic impact on any existing public service contract.
However, the implementation of this approach has suffered from several shortcomings. First, as the French competition authority recognised in its analysis of the French coach market, decisions to ban a new coach route often lacked economic justification. Typically, the main criterion for making such decisions appears to have been whether existing rail services also served the proposed route. Indeed, in each case there appears to have been little assessment of the potential degree of substitutability between the different modes of transport. For example, some regional administrations gave the existence of high-speed rail links as a reason for rejecting the coach route application, without assessing the extent of competition that would be likely to occur on the route between high-speed rail and coach travel.
Second, the national decision-making process may have been subject to conflicts of interest, with the state being both a decision-maker and a shareholder in the national rail operator, SNCF. In addition, the regional administrations submitting their opinion on the proposed coach services have often themselves been involved in financing the rail infrastructure. Indeed, a number of regions have argued that the route applications should be rejected in order to protect their existing investments in rail infrastructure.
Not surprisingly, the economic analysis assessing the extent to which coach services are a threat to other public transport services has therefore been somewhat limited thus far. This, together with a lack of transparency in the decision-making process, has arguably introduced considerable uncertainty for coach operators—for example, in their ability to predict the likelihood of new routes being approved—and has therefore hindered the attractiveness of the sector to new providers/operators.
The July 2015 reform
The coach market reform passed in July 2015 is likely to address some of these concerns, albeit to different degrees, depending on the length of the coach routes. In particular:
with routes longer than 100km being liberalised, coach operators will now be able to compete freely with rail services for medium- and long-distance routes without needing to go through an approval process;
for routes on which the distance between two stops is less than 100km, it needs to be shown that the entrant will not jeopardise the PSO contract for regional rail services. This will require an economic analysis of the substitutability between coach and rail services.
The economic test that will be applied to routes shorter than 100km will now be carried out by ARAFER rather than a government department. The regulator’s independence, and previous experience with substitutability tests in the context of international rail passenger transport liberalisation, should enable more transparency and alleviate concerns about conflicts of interest. In its assessment, the regulator will also take into account the net economic benefit of allowing a new entrant.
The degree of substitutability between rail and coach services is critical in how the reform might affect the way consumers travel in the future. The experience in Germany’s recently liberalised coach market sheds further light on the extent of diversion from rail to road. The evidence indicates that coach services compete with rail as well as cars/driving: 30–44% of new coach customers acquired since liberalisation in 2013 previously used rail; 30–41% previously used cars (including car pooling); and only 10% were new customers. This is indicative of a significant degree of diversion. Nevertheless, rail remains the first choice for most travellers for long-distance journeys, with 129m rail passengers on these routes compared with 17m–19m coach passengers in 2014.
In France, if consumers view the two modes of transport as similar in terms of their travel time, comfort, reliability and schedule, the fare difference will be the deciding factor when choosing which to use. Where this is the case, coaches are likely to benefit from high passenger demand as they can operate at significantly lower costs than passenger trains, and, hence, charge lower fares. For example, the French competition authority notes that, for routes of less than 200km, the average travel time is similar for coach and rail. This might incentivise passengers to choose the coach as it is cheaper and as fast as rail. To protect SNCF from an unsustainable PSO, an economic test will be conducted for routes of less than 100km.
The economic test
The principle of the test
The aim of the economic test to be undertaken by ARAFER is to assess whether entry by a coach operator would potentially threaten the viability of a public service rail route. Although the precise design of the test is not known at this stage, it is likely to draw on the economic test applied by ARAFER to assess the impact of rail liberalisation on SNCF’s PSO.
In particular, in October 2013, the French rail regulator (then known as ARAF) approved a service proposed by an Italian rail operator, Thello, to transport passengers by rail in France. The economic test aimed to assess whether Thello’s new service would distort the economic balance of the contract signed between SNCF and the regional administration. The test focused mainly on the degree of substitutability between the two services given a range of characteristics, including fares, time schedule, travel time, buying platform (online, at the train station, on the coach, etc.), and the location of coach and train stations.
The new economic test for coach routes (conducted by the same regulator, ARAFER) is likely to be similar, and is therefore likely to take into account any positive effects of allowing new coach routes, for the following stakeholders:
passengers—e.g. faster connections, larger schedule range, lower fares;
the regional administration—e.g. wider economic impact due to better transport connectivity;
SNCF—e.g. through its entry into the coach market, or spillover effects by picking up coach passengers at some stage in their itinerary, benefiting from the increase in passenger traffic.
Similar economic tests have been conducted elsewhere in Europe. In the UK, for example, the Office of Rail and Road (ORR) conducts a ‘not primarily abstractive’ test to determine whether to allow open-access operators to introduce a new route in competition with franchised rail operators. The purpose of the test is to assess the extent to which the open-access operator will generate new revenues (by creating new demand), rather than simply taking revenues away from existing franchised operators. The ORR uses well-defined steps and models in approving new routes.
Debates around the reform and the economic test
As currently drafted in the law, the conduct of the test and its effect are still unclear on a number of points.
The distance threshold
The distance threshold for regulated routes has been contentious because if it were set too high, a significant part of the market would no longer be liberalised; but a low-distance threshold would create distortions for PSO operators, as argued by the French Senate (‘Sénat’) and the French competition authority, which advocated in favour of a 200km threshold. Ultimately, the government limited regulation to 100km.
While this level should not create distortion in many French regions, it might still reduce the attractiveness of entry in some. For instance, along the south-east coast, which is a popular summer destination, some cities are within less than 100km of each other. A coach route serving Saint-Tropez, Nice, Cannes and Monaco would need to be assessed through the test. The test therefore has the potential to generate, albeit limited, ‘threshold effects’—i.e. to reduce incentives to open some routes instead of others or to change the stops on the existing routes.
The test also reduces the flexibility of coach operators in some regions to address demand at peak times or for specific events—for instance, ski resorts during the winter season. The opportunity to increase supply and offer cheaper prices when it is most needed might therefore be limited.
How many entrants are sustainable?
In the Thello Decision, the test seems to have been conducted on a case-by-case basis. However, multiple entrants are unlikely to enter the market and ask for regulatory approval at the same time. If the market can sustain only a limited number of entrants, how should these entrants be selected? It is not yet clear how the regulator will proceed, as it would not necessarily be optimal to select on a first-come, first-served basis. SNCF is also likely to wish to enter the market on some routes with its own subsidiary, affecting the net impact on its own profitability and viability in the test.
In the event of simultaneous entry, ARAFER could draw on the UK experience. When several operators pass the not primarily abstractive test and are bidding to use the same capacity, in principle the ORR uses a cost–benefit analysis to select the operator that would make best use of the available capacity. This involves comparing the net present values of the different applications, and is one option that could be used in the French case for comparing simultaneous applications.
SNCF in between opportunities and threats
The French coach market liberalisation offers potential opportunities for coach operators, but its success will partly depend on how the economic test to approve new routes is implemented. At the same time, the liberalisation could affect SNCF’s ability to meet its PSO, with coach operators potentially cherry-picking the most profitable routes. The route authorisation process will therefore need to adequately balance the benefits of additional (inter-modal) competition with the potential impact on the long-term viability of the PSO.
A task force of rail representatives and experts concluded that, given the public contribution to the operating costs of rail on some routes—estimated at €0.108 per passenger km (approximately €27 for 250km)—coach services should be substituted for rail on the least-used routes. This reform offers SNCF an opportunity to partly reduce its burden of long-distance rail costs (the forecast deficit for which is estimated to be €450m in 2015, €40m of which would be due to coach liberalisation), while still meeting its PSO. Also, SNCF already operates 2,000 regular TER (regional express train) routes by coach (25% of the routes) and already has some experience of coach transport through its subsidiary, iDBus. In addition, SNCF is likely to benefit from the market expansion with an increase in passengers using SNCF transport for connections.
As illustrated by the German example in the box, rail and coach operators can co-exist and SNCF could, as Deutsche Bahn did, find its place in both market segments.
The German experience
The long-distance passenger coach market in Germany was liberalised in 2013. Before this, new coach routes were authorised only where existing rail coverage was insufficient (with the exception of some historical routes to/from Berlin, and international coach services). Consequently, many routes were offered at night (when there is typically no rail alternative) or in order to serve airports. The reform lifted the rail monopoly1 and the number of coach routes grew following liberalisation from 86 as at 31 December 2012 to 285 as at 31 December 2014.
The reform triggered a dynamic process with considerable entry and exit. Competition is strong, and it is likely that only a handful of coach operators will end up serving most of the coach routes. Competitive pressure on rail is equally strong: in 2014, two long-distance rail services were offered by private rail undertakings—Hamburg–Cologne (run by Hamburg-Köln-Express, HKX) and Leipzig–Berlin–Rostock (run by Veolia)—both of which competed directly with the corresponding coach route, with Veolia exiting its route at the end of 2014 as it was no longer profitable.2 In Germany, coach travel is significantly cheaper than train: in more than 94% of cases, coach fares are cheaper than train fares, and fares are on average 139% more expensive for trains than for coaches.3
Finally, liberalisation has also provided an opportunity for rail operating companies to enter the coach market and re-optimise existing rail connections. For example, the German rail incumbent, Deutsche Bahn, is active in the German long-distance coach market through its share in the long-distance coach provider, Berlin Linien Bus.
Note: 1 Rail remains protected on routes below 50km in most cases. See Personenbeförderungsgesetz (PBefG), §42a. 2 Bundesamt für Güterverkehr (2015), ‘Marktbeobachtung Güterverkehr: Marktanalyse des Fernbuslinienverkehrs 2014’, 28 January. Spiegel Online (2014), ‘Leipzig-Berlin-Rostock: Bahn-Konkurrent Veolia stellt Fernzug Interconnex ein’, 15 October. 3 Verkehrsclub Deutschland (2014), ‘VCD Bahntest 2014/2015—Die Bahn im Vergleich mit Fernlinienbus und Auto’, December.
Overall, the success of the reform will partly depend on the regulator’s ability to protect the viability of the PSO, while also offering flexibility to coach operators to address demand where it is most needed.
Contact: Pascale Déchamps
 Autorité de la concurrence (2013), ‘Quel avenir pour le transport longue distance par autocar en France?’, p. 58.
 Autorité de la concurrence (2014), ‘Avis no14-A-05 du 27 février 2014 relatif au fonctionnement concurrentiel du marché du transport interrégional régulier par autocar’, p. 80.
 ARAF (2013), ‘Avis no2013-022 du 8 octobre 2013’.
 Bundesamt für Güterverkehr (2015), ‘Marktbeobachtung Güterverkehr. Marktanalyse des Fernbuslinienverkehrs 2014’, 28 January.
 As opposed to local public transportation.
 Federal Statistical Office Germany (2015), ‘2014: Fahrgastaufkommen in Bussen und Bahnen weiter auf Wachstumskurs’, press release no. 121, 7 April.
 Operating costs of regional express trains are around three times higher than those of coaches. The difference in cost is even larger when infrastructure costs are taken into account. See Certu, Sétra, Cete de Lyon and Cete Nord Picardie (2013), ‘Premiers éléments de réflexion sur la pertinence des modes fer et routes pour les dessertes nationales’.
 Autorité de la concurrence (2014), ‘Avis no14-A-05 du 27 février 2014 relatif au fonctionnement concurrentiel du marché du transport interrégional régulier par autocar’, p. 76.
 For existing routes the French competition authority has compared rail and coach fares and concluded that coach fares are cheaper and more stable, as coach operators seem to be reluctant to use yield management due to the high price sensitivity of their customers. See Autorité de la concurrence (2014), ‘Avis no14-A-05 du 27 février 2014 relatif au fonctionnement concurrentiel du marché du transport interrégional régulier par autocar’, p. 32.
 A study by SNCF’s coach subsidiary of its own customers concluded that 23% of passengers were people who would not otherwise travel by rail.
 ARAF (2013), ‘Avis no2013-022 du 8 octobre 2013’.