EUR-Lex Access to European Union law

Back to EUR-Lex homepage

This document is an excerpt from the EUR-Lex website

Document 52014DC0353

REPORT FROM THE COMMISSION TO THE COUNCIL AND THE EUROPEAN PARLIAMENT Fourth report on monitoring development of the rail market

/* COM/2014/0353 final */


REPORT FROM THE COMMISSION TO THE COUNCIL AND THE EUROPEAN PARLIAMENT Fourth report on monitoring development of the rail market /* COM/2014/0353 final */


Fourth report on monitoring development of the rail market


1............ Evolution of internal market in rail services. 4

1.1......... The objectives of the White Paper on Transport (2011) 4

1.2......... The passenger rail market today. 5

1.3......... Evolution of the passenger rail market 8

1.4......... The rail freight market today. 9

1.5......... Evolution of the rail freight market 9

2............ Evolution of the internal market in services to be supplied to railway undertakings. 11

2.1......... Stations. 11

2.1.1...... Stations across the European Union. 11

2.1.2...... Ownership and management 12

2.1.3...... Access to station facilities by railway undertakings. 14

2.1.4...... Quality of services in stations (including accessibility by persons of reduced mobility) 14

2.2......... Freight terminals, marshalling yards and storage facilities. 16

2.3......... Maintenance facilities. 18

2.4......... Other facilities: port access, relief and refuelling facilities. 19

3............ Framework conditions. 20

3.1......... Infrastructure charging. 20

3.1.1...... Infrastructure charges for freight services. 20

3.1.2...... Infrastructure charges for intercity services. 21

3.1.3...... Infrastructure charges for suburban services. 23

3.1.4...... Infrastructure charges – overall rankings. 23

3.2......... Capacity allocation. 24

3.3......... Investments made in infrastructure. 27

3.4......... Developments as regards prices. 28

The annexes of this report are found in SWD(2014) 186


According to Article 15 (4) of Directive 2012/34 of the European Parliament and of the Council of 21 November 2012 establishing a single European railway area (Recast)[1], the Commission has to report every two years to the European Parliament and the Council on:

a) The evolution of internal market in rail services; and

b) services to be supplied to railway undertakings (Annex II to Directive 2012/34/EU)

c) The framework conditions

d) The state of the Union railway network

e) The utilisation of access rights

f) Barriers to more effective rail services

g) Infrastructure limitations

h) The need for legislation

According to Article 15 (3) of Directive 2012/34/EU, the Commission has to monitor the “use of the networks” and the “evolution of framework conditions in the rail sector”, in particular in respect of:

· Infrastructure charging

· Capacity allocation

· Investments made in infrastructure

· Developments as regards prices

· Quality of rail transport services

· Rail transport services covered by public service contracts

· Licensing

· Degree of market opening

· Harmonisation between Member States

· Development of employment and related social conditions

Directive 2012/34/EU has broadened the scope of the Commission's regular reporting as compared to the previous Directive[2] which now includes the evolution of the internal market of service facilities and framework conditions such as investments in infrastructure, price developments, service quality, public service obligations and the development of employment and related social conditions and that are presented for the first time in the Rail Market Monitoring Scheme (RMMS) report. The Commission's earlier reporting and monitoring obligations were covered under Directive 91/440 as amended by Directives 2001/12 and 2004/51 and implemented under Regulation 91/2003 on rail statistics.

This report is the fourth report on the development of the rail market and is accompanied by a staff working document SWD(2014) 186] providing all the annexes containing most of the data supporting the assessments. The previous reports were published in 2007[3], 2009[4] and 2012[5] and were systematically accompanied by staff working documents containing the statistical annexes.

Data for this report has been collected mostly from the Member States through questionnaires ('RMMS questionnaires' sent in 2011 and 2012) and Eurostat, but has been completed by other sources such as a Eurobarometer survey (service quality), publicly available railway fare information, annual financial reports, data for the State aid scoreboard or data provided by specific stakeholders (e.g. UIC[6], UNIFE[7]…). Data collection in some of the recently added areas such as the internal market for rail services is still being discussed in the context of the RMMS implementing act – as a result, reporting in such areas is still fragmentary.

1.           Evolution of internal market in rail services

1.1.        The objectives of the White Paper on Transport (2011)

The White Paper on Transport (2011)[8] recommends that:

· 30% of the increase of  freight transport on land over 300 km should be carried by rail or waterborne transport by 2030;

· 50% of road freight over 300 km should shift to rail or waterborne transport by 2050;

· The majority of medium-distance passenger transport should go by rail by 2050.


Passenger transport is mostly measured in passenger(s) x km which are called passenger-km or p-km. Train-km refers to the distance actually run by the train.

A train from Paris-Brussels transporting 500 passengers over 300 km will generate 150.000 passenger-km and 300 train-km.

Freight transport uses tonnes x km which are called tonnes-km or t-km.

1.2.        The passenger rail market today

Based on a Flash Eurobarometer survey conducted in 2013[9] with some 28.000 respondents aged more than 15 years, only 12% of Europeans are regular users of trains (14% for suburban trains): 6% of Europeans take the train at least once a week and 6% of them take the train “several times per month” but 32% of them never take the train, although 83% of Europeans live within 30 minutes of a train station[10] (cf. graph 14bis).

Graph 1- Frequency of rail use – national, regional and international trains - 2013

Source: Flash Eurobarometer 382a on Europeans' satisfaction with rail services – Annex 2 of Staff Working Document SWD(2014) 186

It is interesting to note that suburban rail use is much more polarised between a group of heavy users (14%) and non-users (53%) than conventional trains. The group of heavy users is mostly composed of youngsters and young commuters[11], whereas the group of non-users is mostly composed of respondents above 55+ (39% have never used a suburban train).

Graph 2- Frequency of rail use – suburban trains - 2013

Source: Flash Eurobarometer 382a on Europeans' satisfaction with rail services – Annex 2 of Staff Working Document SWD(2014) 186

Rail services are to a very large extent domestic services, which represent 94% of all EU passenger-km. International services only represent 6% of all passenger-km, but are important in Luxembourg (30%), Austria (15%), Belgium (13%), France and Latvia (both 11%).

Source: Eurostat

Source: Member States questionnaires, estimations based on annual reports, UIC and Steer Davies Gleave (Study on 4th railway package); data for EU and EEA (including Norway) are identical, – Annex 5a of Staff Working Document SWD(2014) 186

In terms of market segments, half of European railway journeys can be assimilated to regional and suburban services and half relate to long-distance/intercity or high-speed services (27% of all pass-km in 2011[12]). There are important variations across Member States: the UK is, for instance, mostly a commuter market whereas France, thanks to the TGV, is mostly a long-distance market.

Source: RMMS questionnaires, as well as estimations (sources: UIC datasets, Amadeus, annual reports of CP, FS and RENFE, White Paper on Transport assumptions) – no data available for Lithuania, Latvia and Estonia 

High-speed trains almost exclusively dominate the long-distance market in some Member States: in 2011, in France and Spain respectively, 58% and 49% of total passenger-km were travelled in high-speed trains.

Source: EU transport in figures, Statistical pocketbook 2013, quoting UIC, – Annex 5b of Staff Working Document SWD(2014) 186 – In this graph, high-speed rail transport covers all traffic with high-speed rolling stock (incl. tilting trains able to run 200 km/h). This does not necessarily require high-speed infrastructure.


High-speed train services (e.g. TGV, ICE…) and long-distance conventional train services (e.g. Intercity), which often (but not always) require seat reservation, compete mostly against air transport and, to some extent, cars and coaches. High-speed trains operate (almost always) in dedicated infrastructure – since 1990, high-speed tracks kilometres have increased 6-fold (from 1024 km to 6872 km in 2009[13]) – and generally only stop in sizeable urban agglomerations.

Medium-distance/regional train services (e.g. Inter-Regio) and suburban/commuter train services (e.g. RER, S-Bahn, Cercanias…) compete mostly with cars and have free seating. Suburban/commuter train services are often interconnected with metro networks. These services operate almost exclusively with subsidies and public service contracts and call at a high number of stations.  Suburban services require very often intensive-frequencies railway operations (e.g. a train very 5-15 minutes).

1.3.        Evolution of the passenger rail market

As shown in graph 6, the modal share of rail has been increasing since 2003.

However, despite some progress, the share of rail travel journeys in the EU remains modest compared to other transport modes like car and air. The modal share of rail in 2011 has remained stable at 6,2% compared to 2010.

Source: Eurostat – Annex 3 of Staff Working Document SWD(2014) 186 – Data for 2012 is not yet available

Since 1995, rail travel has grown most compared to other modes in the United Kingdom (+70%), Sweden (+42%), France (+37%) and Belgium (+26%). On the other hand, it has decreased by more than 60% in Greece, Poland, Bulgaria, Romania, Estonia and Lithuania (-90%). Overall, the passenger rail modal share has grown by 16% in the EU15 since 1995 (no data is available for EU25) and by 3% in the EU25 since 2000.

Source: Eurostat

As shown in graph 8 below, domestic rail passenger services, which represented 94% of all passenger traffic in 2011, has grown most in Denmark (+15%), Lithuania (+12%), Luxembourg and UK (+9%) since 2010. Member States like Austria, Italy[14] and Czech Republic, where there is now competition on domestic long-distance lines, have also experienced robust growth (6-8% growth). After years of decreasing rail traffic, Lithuania and Italy have been able to revert to growth. The decline of railways in several Member States of South-East Europe resulted from budget restrictions on public service compensations (-38% in Croatia and Greece).

Source: Eurostat, Member States’ RMMS questionnaires

The European domestic rail passenger market is evolving in several Member States. There are now one or more new entrant unaffiliated railway undertakings competing on the long distance Vienna-Salzburg, Naples-Rome-Milan/Venice/Turin and Prague-Ostrava lines. In the Rome-Milan route, the transport share of rail has grown from 36% in 2008 to 66% in 2012. In these lines, incumbent railway undertakings have experienced an increase of traffic (+10% for the Italian incumbent).

International rail services, which represented 6% of rail passenger traffic in 2012[15], have continued their growth (+25% in the period 2004-2011) by growing some 2% in 2011 and some 13% in 2012. Between 2010 and 2012, international rail traffic has grown most in Finland (+42%) thanks notably to the introduction of high-speed services between Helsinki and St-Petersburg. Growth has been impressive in several Central and Eastern European Member States and also in Germany (+23%), France (+20%) and Italy (+13%), where the new entrant Thello has opened night services between Paris and Venice. High-speed services across the Channel have continued to grow (+5% of international rail traffic in the UK).

Source: Eurostat, Member States’ RMMS questionnaires

There are signs however that international passenger traffic is stagnating in important international rail markets like Belgium and the Netherlands, although several companies have started administrative proceedings or expressed their interest to operate rail services on the London/Paris-Brussels-Köln/Amsterdam axes. Finally, rail international services in crisis-hit Member States have taken a toll. TrainOSE, the Greek incumbent, has halted all international services, while cross-border passenger services have nosedived in Ireland (-78%), Croatia (-75%), Romania (-66%), Bulgaria (-50%), Spain (-24%), Portugal (-13%) and Slovenia (-10%). This could be a sign that subsidised public service conventional trains operating over very long-distances are not competitive vis-à-vis other modes, in particular low-cost airlines[16]– the same seems to have taken place in Italy in the early 2000[17].

1.4.        The rail freight market today

Unlike passenger rail, rail freight is far more international: some 47% of all tonne-km in the EU were international in 2011 (out of which 9% of all tonne-km relate to transit). In Belgium, the Netherlands and in the Baltic States, more than 70% of rail freight is international (originating in Germany and Russia), however, in the UK, only 2% of rail freight is international. In Germany, France and Italy, the other 3 major economies of the EU these percentages reach respectively 39%, 19% and 50%. Finally, some 85% of the Danish rail freight traffic is only in transit.

It is also important to underline that Germany and its rail infrastructure play a central role in rail freight, representing alone 27% of all EU tonne-km, leaving far behind even the second major European rail freight market, Poland (12%)[18]. Germany is also at the very heart of the EU rail network: it is by far the most transited Member State (28% of all transiting tonnes-km) together with Austria (13% of transiting tonnes-km).

1.5.        Evolution of the rail freight market

The share of rail freight among all modes has remained stable since 1995 and reached 10,2% in 2010, before rebounding to 11% in 2011 and 10.9% in 2012. Rail freight has lagged behind the overall growth of freight in the EU – rail freight has only grown 5% in tonne-km since 1995 while the overall growth for all modes has been 22%.

Source: Eurostat– Annex 3 of Staff Working Document SWD(2014) 186

Since 1995, the rail modal share has grown most in Northern Europe and has fallen in Southern and Eastern Europe (less so in the Baltic States). Largest growth rates were recorded in the Netherlands (+76%), Denmark (+71%) and the UK (+66%), but also in Germany, the largest rail freight market in Europe. In spite of a slight growth in Italy (+ 4%), rail freight has decreased in France (-5%) and Spain (-54%).

Source: Eurostat

Since 2007, the year when rail freight services have been opened to competition at EU level, traffic has continued to grow strongly in Denmark (+79%), Austria (+15%), UK (+14%), but also in Romania, Ireland, Portugal and Latvia. Although the economic crisis lasted in Romania longer than in Northern Europe, rail freight grew due to high performing non-affiliated freight operators since 2007.

Intermodal rail freight is growing, but single wagonload is decreasing. The share of intermodal rail freight has grown from 15% to 18% between 2007 and 2011 – albeit mostly in Germany, Ireland and Spain. Intermodal rail freight is stagnating in France and Italy and remains small in Poland and the Baltic States (while growing fast[19]). On the other hand, the single wagonload appears to be decreasing everywhere[20] (Eurostat data series are incomplete). In Germany it went down from 39% of all tonne-km in 2004 to only 26% in 2011. In Poland, it reached only 17% of all tonne-km in 2010.

The portfolio of transported goods (cf. graph) has remained stable and remains concentrated in commodities (agriculture, minerals) or products in their first stages of industrial processing (basic metals, chemicals)[21]. Coal, mineral ores, petroleum products and chemicals represented 57% of all tonne-km. Transport of chemical products is the only segment that has grown in absolute and relative terms since 2007 (+7%). Yet, interestingly this growth has been concentrated in Germany, Scandinavia and the Baltic States - as transport of chemicals has declined in Poland and France (- 38%). It is worth underlining the importance of certain transports in some Member States: the rail transport of coal in Poland is more important than the rail freight market of 20 Member States (taken alone).

Source: Eurostat – "Other*" includes miscellaneous articles (cf. footnote 21)

More than half of the decrease in traffic in 2008-2012 can be explained by specific segment evolutions. In Germany, the increase of chemical and transport equipment transport has not compensated important decreases in the rail freight transport of agricultural products, coke, wood and basic metals. In France, most of the decrease has taken place in chemicals (unlike Germany), basic metals (like Germany) and metal ores, but there has been growth in the transport of grouped goods (included in the segment "other"). Finally, more than half of the decrease in Poland, the second largest rail freight market in the EU, derives from a decrease in the transport of coal.

One also ought to ask whether specialisation of rail freight in commodities and basic industrial products does not make its business cycle particularly vulnerable to economic cycles (evolution of commodity prices), energy policies (choices of specific energy sources) and inventory management (commodities cost less to inventory than finished industrial products). Additionally, to be successful, railway freight needs to move into higher-added value "niches" and increase average transport distances.

2.           Evolution of the internal market in services to be supplied to railway undertakings

Information in this field is still fragmentary. This report focuses on the structures of ownership and management and reports on problems encountered in the access to those facilities that have been declared officially.

2.1.        Stations

2.1.1.     Stations across the European Union

There are some 22.000 stations in the EU[22], out of which some 250 are “big stations” that have more than 25.000 travellers/day. The importance of suburban services, which carry much more passengers, explains that some small-sized Member States like the Netherlands have more big stations than Spain or Italy (Luxembourg has for instance one big station). The fact that some important suburban networks are part or not of the railway network covered by the Railway directives could also play a role[23].  Urban decentralisation may also explain why Germany has some 112 big stations, against some 45 and 38 respectively in the UK and France (with Paris-Nord as the busiest station in Europe).

Source: RMMS questionnaires – Annex 6 of Staff Working Document SWD(2014) 186

Density of stations on the network would also vary from one Member State to the other. On average, the distance between two stations on the rail network does not exceed 5 kilometres in Czech Republic, Slovakia, Greece and Austria. In Finland however, this distance goes up to an average of nearly 28 kilometres.

Source: Eurostat, latest data available for length of lines, EIM, CER, infrastructure managers' network statements, number of stations per MS as provided in RMMS questionnaires

This indicator does not imply that trains will necessary stop at all of these stations. Nor does it give the average distance for European citizens to the nearest station. The latest Eurobarometer actually reveals that the Member States where the highest share of population lives less than 10 minutes away from the nearest station are not those with the highest station density on the network. Luxembourg and Denmark are the two Member States where this proportion is the highest while their average distance between stations is above the EU-25 average. On the contrary, in Czech Republic and Slovakia, the share of population living within 10 minutes of the nearest station is around EU-25 average although these two Member States have the lowest distance between stations. This seems to indicate a discrepancy between the spread of stations throughout the territory and the population repartition.

Graph 14bis- Population to the nearest station

Source: Flash Eurobarometer 382a on Europeans' satisfaction with rail services – (telephone interviews of 28.036 EU citizens above 15 years) – Annex 7 of Staff Working Document SWD(2014) 186

2.1.2.     Ownership and management

In spite of various structures of ownership of stations (cf. Table 1), in most cases stations are strongly linked to the incumbent – either through the incumbent holdings (e.g. Ireland and Poland), a subsidiary thereof (e.g. Germany) or the holding’s infrastructure manager (e.g. Austria, Italy). In many cases, there are complex arrangements of ownership of stations – where the infrastructure manager owns platforms but the railway incumbent owns the terminal (e.g. France, Netherlands and Belgium). In other Member States, stations are owned by independent infrastructure managers (e.g. UK, Spain) or by the national government itself (Portugal, Luxembourg, Slovakia and Bulgaria).

Table 1 – Structures of ownership of stations in Europe

Sources: RMMS questionnaires’ contribution from Member States, own research for missing information (grey background)

Management structures of stations are in general identical to those of ownership. However, in those Member States whose national government own stations (Bulgaria, Slovakia, Portugal, Hungary and Luxemburg), management is entrusted to the infrastructure manager[24]. In France, in spite of co-ownership with the French IM RFF, stations are completely managed by the incumbent (SNCF Gares et Connexions). In the UK, Network Rail owns and operates the top 14 largest stations while the rest is owned by Network Rail but operated under lease by the main franchise operator.

In any case, stations are still owned or managed (or at least partly) by incumbent railway operators in all Member States except Bulgaria, Portugal, Slovakia, Spain and the UK.

2.1.3.     Access to station facilities by railway undertakings

Stations may constitute bottlenecks in particular where important stations are concentrated (e.g. Italian 8 mega-stations or Paris). NTV, the Italian new entrant in the Rome-Milan high-speed line has to operate from Rome-Ostiense (instead of Termini) and Milan-Porta Garibaldi (instead of Centrale). On the other hand, using peripheral stations can also be part of a business strategy: Ouigo, the low-cost rail service between Paris-Lyon provides cheaper fares departing from Marne-la-Vallée (Eurodisney), in the outskirts of Paris.

Ownership and management of stations by incumbents creates situations of suspicions of conflict of interest or actual complaints. NTV has filed a complaint to the Italian competition authority accusing Trenitalia of abuse of dominant position, amongst others as regards the management of advertising in Italian stations.

2.1.4.     Quality of services in stations (including accessibility by persons of reduced mobility)

On average, Europeans are fairly satisfied with stations (satisfaction rates have slightly increased since 2011). The 2013 Flash Eurobarometer survey estimates that 51% of Europeans have “high” or “good” satisfaction levels as regards stations, and conversely 49% have “medium” or “low” levels. Highest levels of satisfaction with stations were reached in the UK (73%), Ireland (71%) and Luxembourg (70%). The below-average satisfaction rates are found in Germany (40%), Italy (34%) and Central- and South-Eastern Europe.

Graph 15 – Satisfaction index of railway stations (2013)

Source: Flash Eurobarometer 382a on Europeans' satisfaction with rail services – (telephone interviews of 28.036 EU citizens above 15 years) – Annex 7 of Staff Working Document SWD(2014) 186

According to the same survey, 68% of Europeans are satisfied with the provision of information about train timetables and 67% with the ease to buy tickets. Europeans are less positive as regards cleanliness of stations (57% of satisfaction) and access to complaint-handling mechanisms (37%). Satisfaction with cleanliness of stations is highest in Luxembourg, Austria (80%) and the UK (79%). The below-average satisfaction rates are found in Germany, Italy and Central- and South-Eastern Europe.

Only 37% of Europeans report “high” or “good” satisfaction levels with all the aspects of accessibility of persons with reduced mobility. Satisfaction is highest in the UK (61%), Ireland (56%) and France (52%). The below-average satisfaction rates are found in Germany, Italy and Central- and South-Eastern Europe.

Graph 16 – Satisfaction index of accessibility of stations (2013)

Source: Flash Eurobarometer 382a on Europeans' satisfaction with rail services – (telephone interviews of 28.036 EU citizens above 15 years) – Annex 7 of Staff Working Document SWD(2014) 186

More specifically, the majority of EU citizens are positive for accessibility of ticket offices and vending machines (51%), but less so with accessibility of platforms (46%) and carriages (42%) , and even much less with pre-journey information on accessibility (39%) or assistance for persons with reduced mobility (37%). Dissatisfaction rates are very high when citizens themselves are directly concerned (40% dissatisfaction with accessibility of platforms and 42% as regards accessibility of carriages).

Questions of accessibility are essential to improve the modal share of rail, in particular in the context of the ageing of the European population. 34% of all Europeans never using the train cited at least one accessibility issue as a reason to explain why they do not do so. Rail appears not to be reaching some 19% of the EU population because of accessibility issues.

2.2.        Freight terminals, marshalling yards and storage facilities

Overall, freight terminals, marshalling yards and storage sidings seem to be mostly owned and managed by incumbents’ holdings (in particular in the important freight markets like Germany, Austria, Poland, Lithuania, Latvia), except in the UK and the Netherlands – where the independent infrastructure manager predominantly owns them. In Portugal, Bulgaria, Luxembourg and Slovakia, they are State-owned but managed by the infrastructure manager. Germany appears to account for the majority of reported freight terminals.[25]

Table 2 – Structures of ownership of freight terminals, marshalling yards and storage facilities in Europe

Sources: RMMS questionnaires’ contribution from Member States, own research for missing information (grey background)

2.3.        Maintenance facilities

Ownership of maintenance facilities remains in most Member States under the responsibility of the incumbent railway groups, with the notable exceptions of Romania, the UK and the Netherlands.

Table 3 – Structures of ownership of maintenance facilities in Europe

Sources: RMMS questionnaires’ contribution from Member States, own research for missing information (grey background)

2.4.        Other facilities: port access, relief and refuelling facilities

The remaining facilities appear to be less controlled by incumbent railway undertakings and witness a much greater involvement of private companies. However, in Germany, Ireland, Latvia and Lithuania, these are mostly related to the railway incumbent.

Table 4 – Structures of ownership of port access, relief and refuelling facilities in Europe

Sources: RMMS questionnaires’ contribution from Member States, own research for missing information (grey background)

3.           Framework conditions

3.1.        Infrastructure charging

The “main” infrastructure managers (cf. Part 3.2) have collected from railway undertakings[26] an estimated € 15,7 billion  of infrastructure charges in 2012 (up by 3% compared to 2011) based on financial accounts[27]. Infrastructure charges appear to have represented 41% of all of the “main” infrastructure manager’s revenues (public funds appear to have represented 48% - cf. Part 3.2).


Railway undertakings pay "infrastructure charges" to infrastructure managers for the use of rail infrastructure. Infrastructure charging impacts rail freight, intercity and suburban services differently. Rail freight is reputedly the most sensitive to variations of track access charges. Track access charges also have an impact on the cost structure of intercity services but only marginally on the final fare. Finally, as far as public service obligations are concerned (whose rail fares are generally regulated anyway), the level of track access charges is embedded in the financial architecture of the railway system.

3.1.1.     Infrastructure charges for freight services

As shown in the graph hereunder, the average track access charges in 2014 for a 1000 tonne freight train range between 1.60 €/train-km and 3.40 €/train-km, except in the Baltic States and Ireland. In the Baltic States, infrastructure managers attract significant hinterland traffic from Russia, which operates over very long distances and has a higher average train mass. Networks in Europe's periphery tend to have very low or, in case of Ireland, very high charges. The same holds for the smallest networks, where border crossing problems are prevalent and therefore freight trains cannot pay high charges.

Source: RMMS questionnaires - no data for Norway; *= for France and Italy, data refers to 2013, as these Member States did not provide data for 2014. – Annex 8 of Staff Working Document SWD(2014) 186

Although rail freight reputedly has difficulty to bear “mark-ups” because of its narrow margins, freight trains seem to pay more for access to tracks than passenger trains - the median of all the Member States’ average track access charge for freight is higher than those of intercity and suburban services[28], even when Ireland and the Baltic States are excluded from the calculation[29]. Finally, it is interesting to note that completely separated networks have generally lower infrastructure charges[30].

The reported freight track access charges for 2014 have decreased in Bulgaria (–36%), Czech Republic (-7%) and the Netherlands (-1%) and have remained stable in 8 Member States (hence decreasing de facto[31]). On the other hand, they have substantially increased in Poland (+13%) and Sweden (+6,8%). With the exception of Poland and the Baltic States' networks, the process of a certain levelling out of charges across Member States continues.

In some cases increases have been specific to some segments: in Germany, in the context of increased congestion, track access charges of feeder lines for freight trains with speeds between 50-100 km/h increased 12%, although the average freight track charge seems only to have increased by 2% (hence in line with inflation) in 2014 compared to 2013.

Comparing the evolution of the European[32] average track-access charge for freight suggests a steady decrease over the last years (-28% between 2008 and 2014). More interestingly, the standard deviation – that is the dispersion of the values from the average – has dropped from 2.26 to just 1. In other words, the different national track-access charges seem to be increasingly converging towards the European average, which is helpful in the context of the development of a Single European Railway Area for freight with not only lower but standardized charges too.

Graph 17bis – Convergence of freight track access charges

3.1.2.     Infrastructure charges for intercity services

As shown in the graph hereunder, the average track access charges in 2014 for a 500 tonne intercity train also vary substantially: running an intercity train in Germany or France (and Belgium) costs on average the double of that in Italy or Spain (which also have high-speed networks) and 5 times more than in the UK, Czech Republic or Sweden. Networks with high speed lines (BE, DE, FR, AT, IT, ES) can be found at the "expensive" end of the scale, together with the Baltic States and Ireland.

Source: RMMS questionnaires; *= for France and Italy, data refers to 2013, as these Member States did not provide data for 2014.**= for Spain, average track access charge for high-speed trains running on the network with speed limited at 260km/h – no data for Norway - – Annex 8 of Staff Working Document SWD(2014) 186

The reported intercity track access charges for 2014 have increased in several Member States – in particular in Poland (+43%), Austria (+23%), Spain (+14%) and Sweden (+8,4%, albeit from a very low base). In Austria, the increase of 23% is based on a high-speed surcharge, which has been rejected by the Austrian regulator on 27 September 2013, further to a complaint from the new entrant Westbahn, but is now being appealed by ÖBB Infrastruktur before the Austrian High Administrative Court.

Finally, as far as the evolution of track access charges for intercity trains is concerned, these have remained stable, but their dispersion has increased.

3.1.3.     Infrastructure charges for suburban services

As shown in the graph hereunder, the average track access charges in 2014 for a 140 tonne suburban train are skewed. French suburban track access charges in 2013 are above € 10 train-km, while in 20 Member States they are all below € 2 train-km. German suburban track access charges are also well above those of most Member States (yet in France they are twice as high as German ones). This situation reflects the structure of financing of railways in France (where regions pay track access charges for regional trains under public service obligations (PSOs) directly to the infrastructure manager, which in turn pays the main railway undertaking SNCF to provide maintenance services on the infrastructure) and Germany (where regional authorities provide PSO subsidies that include means to pay track access charges). Similarly, track access charges for suburban rail services are lowest in the UK and Finland.

Source: RMMS questionnaires No data for Norway; *= for France and Italy, data refers to 2013, as these Member States did not provide data for 2014 - – Annex 8 of Staff Working Document SWD(2014) 186

As far as suburban track access charges are concerned, they seem to have remained stable (slight decrease), but their dispersion has increased.

3.1.4.     Infrastructure charges – overall rankings

Overall, track access charges are on the one hand the lowest in Denmark and Sweden and on the other hand the highest in Germany, the most transited network in Europe, and Lithuania, Latvia and Estonia, which have rail networks rather isolated from the core of Europe[33]. To better reflect market structure, it might be useful and necessary in the future to further refine this analysis and distinguish between regional intercity services and high-speed services (instead of a bulk analysis for intercity trains). Last but not least, the fact that some Member States are well ranked in terms of low level of charges has to be analysed with due care – Member States are under the obligation under Directive 2012/34/EU to adequately fund their infrastructure (low track access charges explain why the UK opts for significant investment grants).

Graph 20 – Lowest TAC – rankings for the 3 segments

Source: RMMS questionnaires - No data for Norway; *= for France and Italy, data refers to 2013, as these Member States did not provide data for 2014. **= for Spain, average track access charge for high-speed trains running on the network with speed limited at 260km/h.

Infrastructure charging appears to be suffering from 3 main problems:

- Important transit networks placed in the centre of the European rail system or carrying hinterland traffic levy higher charges than smaller and peripheral networks. This situation prevails for both freight and passenger trains. It hampers the integration of national rail systems, whereby the high costs of interoperability exacerbates the conditions of cross border traffic.  

- Track access charges in the East of the Union remain higher for freight than for (suburban) passenger transport; which from an economic point of view, suggests an insufficient level of compensation for services under public service obligations, whereby infrastructure managers could be recouping the resulting revenue losses through higher charges for freight services. As a result, rail freight traffic could be becoming less competitive than road and the railway undertakings cannot generate the funds they would need for renewing their wagon fleet.

Finally, it has not been possible to usefully take advantage of data requests as regards stations and facilities charges, traction current and diesel, as Member States' data collection has been rather piecemeal. Station and terminal charges would account for a significant part in the total infrastructure charges where trains travel short distances and stop frequently, in some Member States they account for more than half of total infrastructure charges. Market segments that require intensive use of facilities, such as open access regional passenger trains and single wagon load, display the least intensive level of competition.

3.2.        Capacity allocation


Infrastructure managers allocate path access rights to railway undertakings every year based on the requests made by the latter. Infrastructure managers also get ad hoc path requests, in particular from rail freight companies who cannot predict their services one year in advance. Path requests can be rejected if there is congestion.

Productivity of railway track varies between Member States: At one extreme there are the dense networks of the Netherlands and the UK, under strong pressure of commuter services, followed by Germany, Austria and Belgium, and at the other extreme, the relatively under-used rail networks of Baltic States and South-Eastern Europe. There are four times more trains-km per km of track[34] in the Netherlands than in Bulgaria, Romania and Estonia.

Source : Eurostat, UIC, RMMS (Bulgaria) - – Annex 9 of Staff Working Document SWD(2014) 186


Services under PSO generally have pre-reserved paths, as they require very intensive frequencies throughout the day (in particular suburban trains).   High-speed trains operate (almost always) in dedicated infrastructure and generally only stop in sizeable urban agglomerations. Freight trains operate in many instances with paths that are allocated ad hoc.

For each railway line, there is a theoretical maximum level of trains ("maximal capacity"). However, the difference between the real capacity and the theoretical capacity results from a series of trade-offs like maintenance works, stopping patterns, conflicting junction movements (trains at switches), rolling stock mixes and off route constraints. The shape of rail networks also plays an important role: trains are more easily rerouted in mesh networks (e.g. Germany) than in star-shaped networks (e.g. France, Spain)

The European rail networks are predominantly used by passenger trains (78% of all train-km), although there are variations in terms of types of services between Member States. Trains running on the intensely used networks of the UK and the Netherlands, but also in Ireland and Luxemburg, are mostly passenger trains. Trains in the sparsely used networks of the Baltic States will mostly be freight trains.

Source: Eurostat, DB, ISTAT, INSEE – situation in 2011; EU average excludes DK, HU and GR for which no data was found

Only 5 EU Member States (Germany, Denmark, the Netherlands, Romania and the UK) and Norway have declared part of their infrastructure to be “congested” according to their responses to the RMMS questionnaire. In total 1324 km of lines[35]  (0,6% of the whole EU lines) have been declared congested and the greatest bottleneck appears to be located in absolute terms in Germany (399 km), but Denmark and Romania have important bottlenecks.

Most Member States have opted for the prioritisation of PSOs (cf. table 5), services with direct value for society and high frequency services – which often in practice cover mostly commuter services. EU market access law allows for the prioritisation of path allocations in favour of services with value for society, PSOs and international rail freight services. France has not reported any prioritisation of rail services in the context of the RMMS. This prioritisation covers more than 85% of all train-kms in the Netherlands, the UK, Luxembourg and Ireland (cf. graph 23).

Table 5 – Type of priorities in path allocation

Source: RMMS questionnaires

Source: RMMS questionnaires

Finally, a first attempt to take stock of the quantity of rejected paths, suggests that France (4,1 %) and Poland (1,3 %) are experiencing the largest numbers of path rejections. In France, which has not established service priorities, most rejections have actually concerned local and regional services (2,3% of all path requests of local and regional services, which represented 42% of all rejections), but the segment most hit have been domestic and international freight (18% and 13% of path rejected respectively). Path rejections have also been reported in Germany, Netherlands, Norway and Hungary (all less than 0,1%).

3.3.        Investments made in infrastructure

The total amount of reported State grants for rail infrastructure managers varies according to various sources from which it can be estimated (financial accounts of infrastructure managers, State aid scoreboard and RMMS questionnaire on the compensation of multi-annual contracts) and the gaps in data series, but remains overall around € 18-21 billion in 2012.

All but 7 Member States (AT, CZ, EE, FI, GR, LV and PL) have concluded multi-annual contracts with their infrastructure managers. Such contracts cover an equivalent of 73% of the entire EU rail infrastructure and they last on average 5 years (in Spain they last for 2 years, whereas in Luxembourg, the multi-annual contract extends as long as 2024). There’s a wide variety of performance indicators.  It is interesting to note that several Member States in Central and South-Eastern Europe use “train speed” as a performance incentive whereas many congested networks (the Netherlands, Germany, and Belgium) use punctuality or delays as performance indicators.

As illustrated in Annex 10b of the Staff Working Document SWD(2014) 186 accompanying this report, in terms of investments in the network, slightly less than € 29 billion appear to have been invested in the conventional network in 2012 (some 7% more than 2011) and some € 34,5 billion in the whole rail network (including high-speed). As far as the conventional network is concerned, the shares of maintenance (29%), enhancement (36%) and renewal (35%) appear to be roughly equal in 2012 (slightly less so in 2011, where renewals reached 39%).


There are varying definitions of 'maintenance', 'enhancement' and 'renewal'. However, broadly speaking, 'enhancement' covers in general the extension and modernisation of infrastructure through for instance new technologies (e.g. ERTMS, replacing level crossings with underpasses or overpasses),   'renewal' covers the replacement of assets putting the infrastructure back in the condition of when it was new (e.g. replacing sleepers, ballast or rails, renewing a bridge) and 'maintenance' refers to actions that ensure the functioning and extend the life time of existing assets (e.g. grinding, tamping, pruning trees and shrubs at the track side).

Finally, railway-related projects financed by EU funds, either under TEN-T or structural and cohesion funds, have amounted to some 22 billion EUR through the period 2007-2013, hence some 3 billion EUR/year, representing some 2% of the EU annual budget.

The total EU funding for railway infrastructure under the TEN-T 2007-2013 framework programme allocated by the end of 2013 reached some 4.4 billion EUR for rail (including ERTMS), corresponding to 65% of all TEN-T funds allocated by the end of 2013. 8 Member States (IT, FR, DE, AT, ES, SE, BE and DK) attracted 87% of all TEN-T funding for rail during the period 2007-2013, whilst the remaining 19 Member States absorbed 587 million EUR (hence each of them absorbed less than 110 million EUR).

Graph 24a – Allocated TEN-T funding for rail per Member State by the end of 2013

Source: Innovation and Networks Executive Agency (INEA) - *Member States with less than 100 million EUR of TEN-T funding (taken individually – these Member States have totalled 587 million EUR of funds allocated to them).

Most of the funding of rail projects between 2007 and 2012 took place through the structural and cohesion funds (17 billion EUR). The major recipients have been Italy, Poland, Spain, the Czech Republic and Hungary (all in the range of 2 billion EUR or more). 

Source: European Commission, DG REGIO

As a result, for the period 2007-2012, rail projects selected for EU funding were concentrated in Italy, Spain, Poland and the Czech Republic. Member States like Denmark, Sweden and the UK remained largely aside of project funding.

Graph 24c – Earmarked EU funding for selected rail and ERTMS projects, 2007-2012 (million EUR)

Source: Innovation and Networks Executive Agency (INEA), European Commission, DG REGIO

3.4.        Developments as regards prices

Nominal prices for railway services have increased by 4% in 2012 compared to 2011, based on the harmonised consumer price index (HCPI) – which includes urban transport. Major increases took place in Central and South-Eastern Europe (in Slovakia, the increase reached 35%). In Sweden, prices have decreased by 1%.

Source: Eurostat

These aforementioned variations are part of a similar trend. Since 2005, the reference year for the HCPI, rail prices have been increasing by more than 50% in most Southern- and Central-Eastern Europe. In the UK and Italy, they have increased by more than 40% in nominal terms[36]. On the other hand, in Sweden, rail fares have increased only by 3,7% in nominal terms.

Source: Eurostat

However, rail fares since 2005 have increased less than other modes of transport. In fact, in the EU27 rail fares have increased by 0,15 percentage point less than transport prices overall. This is particularly important as regards Sweden and the UK, where transport prices have increased by 17 and 15 percentage points more than rail prices. On the other hand, in Southern- and Central-Eastern Europe, the high rail price increases have overtaken the price increases of other modes. In Germany, rail price increases have been in line with transport prices increases.

Source: Eurostat

Rail fares have grown in line with the prices of the operation of transport equipment, but it is interesting to note that since 2005 the price of fuel has increased by 12 percentage points more than rail fares. However, in Portugal and several Central- and South-Eastern Member States, rail fares have increased more than fuel prices (with extremes of 50 percentage points like in Latvia). In most "old" Member States (and Poland), rail fares increased less than fuel prices. In Belgium and Sweden, fuel has increased by more than 30 percentage points more than rail fares.

Source: Eurostat

Beyond this macro-economic outlook, it is necessary to recall that the variation of fares greatly depends on the structure of financing of the railway market. Public service obligations normally have regulated prices, whereas commercial services have unregulated prices. In some Member States, the public service obligations cover the entire territory (cf. infra). This being said, in the UK, which falls in this category, unregulated prices co-exist with the PSO regulated prices. In the Netherlands, the incumbent NS appears to be free to set fares. Finally, it is important to underline that most long-distance and international services are commercial services (cf. infra).

Rail fares in some commercial lines can vary strongly in relative terms throughout the EU and one should underline that, from a consumer perspective, day returns in some routes remain costly, even if sometimes fidelity rail cards can halve fares[37] (cf. infra). Business class same-day returns in the Paris-London, Madrid-Barcelona, Cologne-Munich have been found to cost around 400 EUR, based on a price survey made by Commission services in February 2013[38]. Similarly, booking a weekend trip 2 weeks in advance between Paris and London can still cost 260 EUR and an immediate departure from Madrid to Barcelona can cost 173 EUR. The fares of railway companies in commercial lines are also influenced by competition of other modes (air and road transport). It also appears that in some lines like London-Paris the vast majority of fares are reserved at lower prices with reservations taking place between 6 weeks and 4 months in advance or more, but at the same time, it is unclear if demand is influencing the booking times or if pricing structures are ultimately influencing demand. Finally, according to German authorities, the overall fares in Germany would fluctuate between 0,18 and 0,66 EUR/km, notably because of the effects of fidelity cards.

The average day return business class fare in February 2013 varied from 0,13 EUR/km on the Prague-Ostrava line (where 3 companies are competing against each other) up to 1,81 EUR/km on the Paris-London line. Similarly, leisure fares[39] evolved in a similar fashion from 0,09 EUR/km on the Prague-Ostrava line till 0,86 EUR/km on the Paris-London line. Interestingly, Ouigo – SNCF low-cost service from Paris’ suburb of Marne-la-Vallée – and the Italian high-speed operators were the cheapest high-speed services at around 0,25 EUR/km,  half the price of the TGV and ICE services in France and Germany (0,40/0,45 EUR/km), consequently well below international services in the PBKA[40] square or France-Germany lines, where no competition has yet materialised (0,60 EUR/km).

Source: Commission’s services price research and own calculations – cf. annexed data, data collection on 19 February, 8 March and 1st April 2013,– Annex 11 of Staff Working Document SWD(2014) 186

For public service obligations, it is useless to compare fares, as these are regulated. It is more useful to look at the financing ratio of passengers versus public transport authorities.

[1]               OJ L 343, 14.12.2012, p. 32.

[2]               Directive 2001/14/EC of the European Parliament and of the Council of 26 February 2001 on the allocation of railway infrastructure capacity and the levying of charges for the use of railway infrastructure, OJ L 75, 15.3.2001, p. 29.

[3]               On 18 October 2007 the European Commission adopted a Communication to the Council and the European Parliament on monitoring development of the rail market COM(2007) 609, accompanied by staff working document SEC(2007)1323;

[4]               On 18 December 2009, the European Commission adopted a Report to the Council and the European Parliament on monitoring development of the rail market COM(2009)676, accompanied by Commission Staff Working Document SEC(2009)1687

[5]               On 21 August 2012, the European Commission adopted the third Report to the Council and the European Parliament on monitoring development of the rail market COM(2012)0459, accompanied by Commission Staff Working Document SWD(2012)246 final/2

[6]               Union Internationale des Chemins de Fer (UIC)

[7]               European association of railway equipment manufacturers

[8]               White Paper – Roadmap to a Single European Transport Area – Towards a competitive and resource efficient transport system COM (2011) 0144 final.

[9]               Flash Eurobarometer 382a on Europeans' satisfaction with rail services, published on 16 December 2012. – 28.036 interviews were conducted over the telephone (some 1000 respondents per Member State) with citizens aged more than 15 years. For more details, refer to Annex 1

[10]             Ibid.

[11]             Students travelling to work, school or university

[12]             EU transport in figures, Statistical pocketbook, 2013, p.52

[13]             EU transport in figures, Statistical pocketbook 2013

[14]             Italian data, as provided in January 2014 by Italian authorities – includes international traffic.

[15]             This estimation  was made by using a mix of Eurostat and RMMS statistics (completing gaps in Eurostat series)

[16]             Air traffic in the Lisbon-Madrid route has continued to (slightly) grow over the period 2009-2011

[17]             International passenger rail traffic has decreased by 50% in Italy since 2004

[18]             France (8%), Italy and UK (5%)

[19]             In Poland, volumes of intermodal have doubled since 2007 (source: Eurostat)

[20]             CER (2013), Rail Freight Status Report 2013 (p.37 fig. 24), reports that the share of single wagon load would have decreased from 41% in 2002 to 31% in 2008, based on a variety of sources (Eurostat, McKinsey, XRail)

[21]             The 33% of tonne-km representing "Other" (cf. graph 12) includes often finished products in intermodal traffic.

[22]             There is no data for Romania, Poland, Finland and Portugal.

[23]             The definition of this indicator could be further refined in the context of the RMMS implementing act

[24]             In Luxembourg, the infrastructure manager is part of the incumbent’s holding

[25] No data for UK, Spain, Finland, Latvia and Belgium.

[26]             In some Member States, it is difficult to distinguish between public funds and infrastructure charges.  In France, regions pay themselves the so-called « redevance d’accès » of the rail services under public service obligations that they purchase from SNCF. To be able to map the financial flows from railway undertakings to infrastructure managers, the payment of the redevance d’accès has been assimilated to a subsidy.

[27]             Financial accounts may differ from regulatory accounts, which fall under the supervision of national regulators

[28]             Median of the average track access charges in the Member States for 1000 tonne freight trains is 2,31 versus 1,81 and 1,30 for respectively intercity and suburban services.

[29]             Withdrawing the average freight track access charges of the 4 more “expensive” networks (Baltic States and Ireland) brings down the median for freight to 2,12 (versus 1,51 and 1,29 for respectively intercity and suburban) and reduce the variance to 0,98 (which is then lower than those of suburban and intercity trains).

[30]             The median freight track access charge of completely separated networks (BG, CZ, DK, EE, ES, FI, GR, NL, RO, SE, SK and UK) was 1,93 EUR/train-km compared to other networks ; similarly the median rank was 10th against 15th for other networks. The main exceptions are LU and SI, which have low freight track access charges.

[31]             Denmark, Spain, Finland, Hungary, Luxembourg, Latvia, Romania and Slovakia – all these Member States have had inflation based on the 12-month November 2013 inflation rate; in Greece, based on that same indicator, because of deflation (-0,7% inflation), stable track access have de facto increased. 

[32]             Ireland and the Baltic States are excluded from this calculation since they are isolated from the rest of the European network. All data come from the RMMS questionnaires

[33]             Denmark ranks on average as the second cheapest rail network (Sweden the fourth cheapest), whereas Latvia overall ranks 24th as the most expensive (and Germany ranks 21st cheapest or 4th most expensive).

[34]             This calculation includes track managed by UIC members, whereas other track, notably so called 'industrial track', is not considered in this calculation.

[35]             Congestion declarations: UK 551 km, Germany 399 km, Romania 170 km, Norway 70 km and the Netherlands 47 km.

[36]             Data from the UK regulator (ORR) shows that rail fares over the same period have evolved in different ways: between 2005 and 2012, ""advance" tickets in the London suburban services have only increased by 14%, while "off peak" and "anytime" tickets have increased 44% and 42% respectively.

[37]             Heute vom Gleis gegenüber, Der Spiegel 14/2003 -  30.03.2013

[38]             The methodology of this fare analysis is provided in the Commission staff working document annexed to this report and is still subject to further refinement with the Member States in the context of the Rail Market Monitoring working group of the Single European Railway Area Committee.

[39]             The average was calculated between the fare of a citytrip reserved 2-weeks in advance and an immediate departure

[40]             PBKA stands for Paris-Brussels-Köln (Cologne)-Amsterdam


Fourth report on monitoring development of the rail market


3.5......... Quality of rail transport services. 44

3.5.1...... Rail services versus other services. 44

3.5.2...... Quality of services. 44

3.5.3...... Frequency. 47

3.5.4...... Punctuality. 47

3.6......... Rail transport services covered by public service obligations (PSOs) 48

3.6.1...... Public service obligations and railway segments. 48

3.6.2...... Financing public service obligations. 49

3.6.3...... Competitive tendering of public service obligations. 51

3.7......... Licensing of railway undertakings. 52

3.8......... Degree of market opening. 53

3.9......... Harmonisation between Member States. 54

3.10....... Development of employment and social conditions. 56

3.10.1.... Employment in rail 56

3.10.2.... Socio-demographic structure of the rail labour market 57

3.10.3.... Training activities. 58

3.10.4.... Other aspects of working conditions. 58

4............ State of the EU network and infrastructure limitations. 59

4.5......... Variation of infrastructure. 59

4.6......... Management of infrastructure. 59

4.7......... Electrification. 59

4.8......... Train movements. 59

5............ Utilisation of access rights. 60

5.5......... Railway freight 60

5.6......... Passenger railways – regional and suburban segments. 61

5.6.1...... Passenger railways – all segments. 61

5.6.2...... Regional and suburban segments. 61

5.6.3...... Rail high-speed and long-distance segments. 61

5.7......... Passenger railways – international services. 62

6............ Barriers to more effective rail services. 62

7............ Conclusions. 62

The annexes of this report are found in SWD(2014) 186

3.5.        Quality of rail transport services

3.5.1.     Rail services versus other services

Rail services continued to be quite badly ranked in the Consumer Scoreboard, which compares several types of services in the internal market. Rail ranked 27th in 2012 compared to other internal market services in terms of the consumer market performance indicator[41] - only real estate, mortgage and investment products fare worse, whereas airlines, postal services and urban transport rank far better than rail.

3.5.2.     Quality of services


Regulation (EC) No 1371/2007 on rail passengers' rights and obligations[42] lays down the following minimum service quality standards in its Annex III:

- Information and tickets

- Punctuality of services, and general principles to cope with disruptions to services

- Cancellation of services

- Cleanliness of rolling stock and station facilities

- Customer satisfaction survey

- Complaint handling, refunds and compensation for non-compliance with service quality standards

- Assistance provided to disabled persons and persons with reduced mobility  Overall satisfaction

According to the Consumer Scoreboard, rail also ranks 27th out of 30 internal market services in terms of overall satisfaction (15% of consumers rate rail services between 0 and 4 in scale of 0 to 10)[43].

Based on the Eurobarometer survey 2013, where 28,036 citizens in the EU were interviewed (some 1,000 interviews per Member State), it appears that only 58% of EU citizens are highly or fairly satisfied with the level with rail services.

Graph 31 – Satisfaction index of railway stations and travels (2013)

Source: Flash Eurobarometer 382a on Europeans' satisfaction with rail services -– (telephone interviews of 28,036 EU citizens above 15 years) – Annex 12 of Staff Working Document SWD(2014) 186

In terms of overall satisfaction, Finland and the UK top the satisfaction index with more than 75% of user satisfaction, whereas Bulgaria and Estonia have less than 30% satisfied users. Italy has also a very low satisfaction rate (36%). More than 50% of respondents are satisfied in Germany and Sweden, albeit below the EU average (55%).

As far as most indicators are concerned, it is essential to underline that in general lowest satisfaction rates are found in Italy and Central-/Southern-Eastern European Member States.  Satisfaction with retail transactions[44]

68 % of Europeans are satisfied with the provision of information about train timetables (16% dissatisfaction rate). As far as conventional trains[45] are concerned, satisfaction has slightly increased since 2011 (3 percentage points), with big jumps in Poland and the Netherlands (satisfaction has increased by 16 and 12 percentage points respectively).

67% of Europeans are satisfied with the ease of buying tickets in stations (17% dissatisfaction rate – although up to 37% in Germany).  The satisfaction rate[46] has remained stable since 2011 (78% satisfaction) with big jumps in Austria and Greece (14 and 10 percentage points increase), but worryingly dissatisfactory in Italy, Denmark and Slovenia (all more than 10 percentage points decrease). Dissatisfaction is slightly higher in rural villages (19% dissatisfaction) than in cities (14% dissatisfaction). Europeans affected by accessibility issues (e.g. persons with reduced mobility) are also slightly more dissatisfied (20% dissatisfaction).

Only 36% of Europeans are satisfied with complaint mechanisms (18% dissatisfaction –up to 31% in Italy). Satisfaction with complaint handling has jumped by 10 percentage points since 2011 – showing the first effects of the implementation of Regulation (EC) No 1371/2007. Satisfaction has jumped by more than 20 percentage points in 4 Member States (France, Latvia, Finland and Spain) and by 10-20 percentage points in 7 others. Satisfaction has only decreased by more than 5 percentage points in the Czech Republic, Italy and Estonia. It is important to underline that 54% of consumers that experienced a problem with trains complained to the railway undertaking or a third party, which is below the 70% average for all services.[47]

Most Europeans are satisfied with the availability of through tickets (58%), with highest levels being reached in France, Belgium and Finland (all above 70%), but also the UK and Germany. As with other satisfaction rates, Italy/Central/South-Eastern European Member States underperforms (but also Sweden and Austria).  Satisfaction with on-board service[48]

Satisfaction with provision of information during train journeys, in particular in case of delays has remained stable but insufficient (less than 50% of satisfaction). The highest rates of satisfaction are found in the UK (70%), Finland and Ireland. Highest rates of dissatisfaction are found in France (47%) and Germany (42%). Since 2011, as far as conventional trains are concerned, great improvements have taken place in Finland, Poland and the Netherlands (all above 9 percentage points increase).

Satisfaction with availability of staff: 58% of Europeans are satisfied with the availability of staff in their Member State. Highest levels of satisfaction are found in Belgium, Finland and Luxembourg (all above 70%), whereas Germany, (32%) Ireland and France have the highest degrees of dissatisfaction (all above 28%).

Satisfaction with cleanliness and maintenance of rolling stock has remained insufficient.  Less than half of Europeans (48%) are satisfied with the cleanliness of railway carriages, including toilets. Finland, Ireland and the UK top satisfaction (above 68%), while Italy, Romania and Bulgaria experience the lowest levels of satisfaction (together with Germany and Central/South-Eastern Europe). Since 2011, on conventional lines, major increases took place in Austria, Poland and the Czech Republic (all have increased by more than 10 percentage points) and major decreases in Italy, Portugal and Latvia (all have decreased by more than 10 percentage points).

Satisfaction with bicycle access to trains is highest in Denmark (47%) and the UK (44%). The highest levels of dissatisfaction are found in Romania (28%), France (24%) and Germany (20%).

3.5.3.     Frequency

Satisfaction with frequency is essential to attract travellers to rail – as time is with price the most critical factor affecting travel consumer decisions[49]. Overall, 59% of Europeans are satisfied with frequency according to the Eurobarometer survey. The UK, Sweden and the Netherlands have the lowest dissatisfaction rates for frequency. Italy, Central-/Southern-Eastern Europe have the highest dissatisfaction rates (as in previous surveys). France and Germany have polarised opinions – satisfaction with frequency is above-average yet so is dissatisfaction.

Source: Flash Eurobarometer 382a on Europeans' satisfaction with rail services - – Annex 12 of Staff Working Document SWD(2014) 186

In terms of frequencies in important high-speed lines, there are now some 4 trains/hour in the Italian high-speed network, which has the advantage of crossing most important cities in a single axis. The Paris-Nantes, Paris-Lyon and Frankfurt-Köln lines have also reached at least 3 trains/hour. On the other hand, otherwise important lines like Paris-London (which links the two major cities of the EU) are still at 1,5 trains/hour.

Overall, in terms of surveyed networks (cf. Analysis on fares), the highest frequencies are reached in the lines with more than 1 operator like the Italian high-speed network, Vienna-Salzburg and Prague-Ostrava with up to 2,5 trains/hour. The frequencies in Paris-Benelux/Germany remain under 1 train/hour (on average).

Source: European Rail Timetable, Summer 2013 edition

3.5.4.     Punctuality

According to the Eurobarometer survey, dissatisfaction[50] with punctuality and reliability is the greatest in France (47%), Germany (42%) and Italy (38%) and the lowest in the UK, Estonia and Lithuania. Conversely, satisfaction with punctuality is the greatest in Ireland, Latvia, Austria and the UK (above 73%).

Source: Flash Eurobarometer 382a on Europeans' satisfaction with rail services - – Annex 12 of Staff Working Document SWD(2014) 186

Local and regional trains have been very punctual in Latvia, Portugal, Austria and Lithuania – while Belgium and Hungary – two important commuter markets - perform badly[51]. Sweden and Italy also fall below 90% of punctuality.

Long-distance trains have been very punctual in Finland, Denmark and Italy while Poland and Lithuania perform badly. The punctuality rates of Germany and Portugal appear as lower because they have been accounted from a 5 minutes threshold, although there is a high degree of dissatisfaction of German travellers.

Sources: RMMS questionnaires and Trafikverket for Sweden (excludes suburban services)– data for Germany and Portugal covers delays above 6 and 3 minutes (respectively – cf. annex 14 of Staff Working Document SWD(2014) 186

Sources: RMMS questionnaires and Trafikverket for Sweden – data for Germany and Portugal covers delays above 6 and 5 minutes (respectively) - annex 14 of Staff Working Document SWD(2014) 186]

There are interesting contrasts in punctuality rates. In Sweden and Italy, long-distance trains have been very punctual in contrast to local trains. In Portugal and Lithuania, it has been exactly the opposite. None of the rates appear to explain the high degree of dissatisfaction with punctuality and reliability in France. Finally, as far as high-speed services are concerned, AVEs in Spain have reached a punctuality rate of 99,2%, whereas in the more congested networks of France, TGVs have reached a 91% rate (and 85% for Thalys services in Belgium). Finally, as far as combined freight is concerned, it appears that punctuality has reached 67% in 2011, with still 19% of trains delayed by more than 3 hours and 4% by more than 24 hours[52].

In terms of reliability, although series of data are incomplete, Hungary performs worst with some 14% of trains cancelled.  Denmark, France, the UK and Norway had all some 2-3% of local trains cancelled. The UK and Norway presented  similar percentages for long-distance services. High-speed services appear to be highly reliable: less than 0,2% of cancellations in France and Italy.

3.6.        Rail transport services covered by public service obligations (PSOs)

3.6.1.     Public service obligations and railway segments

Public service obligations (PSOs) cover some 62-65% of all EU passenger-km and some 74% of train-km in 2011-2012. The number of passenger-kilometres under PSO has changed in France (with the inclusion of the Trains d'équilibre du Territoire, increasing from 31% to 43% of all domestic passenger-kilometres) and Finland (where they increased from 14% in 2010 to 45% in 2012).

Source: RMMS questionnaires - annex 15 of Staff Working Document SWD(2014) 186

The relative share of rail services under PSOs and commercial services varies for the different market segments. In the UK, where most rail services are commuter services (cf. graph 4), PSOs cover 99% of all domestic passenger-kilometres. This situation contrasts with France, which is mostly a high-speed market and where PSOs only cover 38% of all passenger-kilometres. 

All suburban and regional services (corresponding to 49% of all passenger-km) in Europe appear to be covered by PSOs. High-speed services, which represented some 27% of all EU passenger-km, do not seem to be operated under PSOs in any of the Member States, except maybe in some domestic line sections in the Netherlands and Austria (e.g. Salzburg-Innsbruck). As a result, it can be assumed that the two-thirds of the conventional long-distance services are operated as PSOs. Finally, PSOs can cover the entire domestic networks, either with one or several public service contracts, of some small-sized Member States like the Netherlands, Belgium, Hungary, Denmark and Ireland – or bigger Member States as it is the case of the United Kingdom.

Source: RMMS questionnaires, own research, State aid scoreboard; the situation in Poland could be slightly different with some commercial services running on regional lines

3.6.2.     Financing public service obligations

In 2011-2012 some 18 billion EUR[53] of public support have been granted to railway PSOs in the EU, with alone some 6 billion EUR in Germany, 4,5 billion in France and 2 billion in Italy. Direct public support in the UK and the Netherlands is very low as these Member States cross-finance their loss-making services through profitable services falling in the same public service contracts for rail services and revenues from passenger fares (cf. infra). This partly explains why Luxembourg or France provide some 18-17 EUR/train-km[54] of subsidies whereas in the UK and the Netherlands public support is as low as 0,1-0,2 EUR/train-km. As in France, the level of net support also varies in Germany[55]. Finally, there are also some differences in the number of passengers per train in European PSOs: in the Netherlands, France, Belgium, Italy and the UK there are some 120 passengers per train, whereas in the Czech Republic, Lithuania and Luxembourg, there are less than 60 passengers per train.

Source: RMMS questionnaires, own research, State aid scoreboard, annex 15 of Staff Working Document SWD(2014) 186

PSO costs have also been covered by passenger fares, which have amounted to at least 21 billion EUR in the EU. As explained, in the UK or the Netherlands, passenger fares cover 99% of the PSO costs, whereas in Bulgaria, Germany, Italy and the Czech Republic, the taxpayer has supported more than 70% of PSO costs.

Graph 41 – Who is paying public service obligations? (2012)

Source: RMMS questionnaires, own research, annual reports of incumbents


Based on the information provided by Member States in the RMMS and the analysis of financial accounts of railway undertakings (RUs) and infrastructure managers (IMs), it is possible to map the main estimated financial flows (presented in the graph hereunder in € billion  for the year 2012). Passengers have purchased from RUs some € 38 billion of rail passenger services and businesses have purchased some € 11-13 billion of rail freight services. RUs have paid some € 15 billion of track access charges to IMs, which have received some € 18 billion of State grants. The compensation of PSOs has amounted to some € 18-20 billion. Last but not least, IMs receive some € 4 billion of other revenue (electricity, real estate…).

Graph 42 – Financial flows in the rail sector (in billion EUR)

Source: Annual reports, RMMS questionnaires

3.6.3.     Competitive tendering of public service obligations

Several Member States are tendering their public service contracts for rail services. Some are doing so on a large scale (the UK, Germany, Sweden,), whereas some others have only done so for a few contracts (Poland, Italy, Denmark, Portugal, the Netherlands, the Czech Republic and Slovakia)[56]. In 2011, the UK launched the renewal of 5 franchises, while Germany and Slovakia reported the competitive tendering of respectively 18 and 1 public service contract , of a smaller size, though.  

Interestingly, there seems to be an emerging internal market for public service contracts (PSCs). A growing number of these PSCs for rail are published in the OJEU (TED website) as any other public procurement contract. The number of contract notices[57] for PSCs published in 2012 reached 41, hence a doubling since 2006 – overall 205 PSCs have been published since then. The vast majority have been German tenders (some 113), but this might be also due to the size of contracts and there are indications that almost all German PSCs are published in the OJEU[58]. The publication of contract award notices and ex-ante voluntary notices[59] has doubled since 2010, improving therefore transparency.

Table 6 -Public service contracts published in OJEU in 2012 per Member State

|| BG || CZ || DE || DK || IT || NL || PL || SE || UK || Total

Contract notices || 3 || 1 || 113 || 1 || 2 || 3 || 62 || 13 || 7 || 205

Source: OJEU database TED

Source: OJEU database TED

The PSCs published in the OJEU present an interesting sample of PSCs overall in the EU. In this context, it is interesting to note that 76% of published PSCs used “better value for money”[60] as the award criteria and only 23% used price as the only award criteria.

Based on a study of the European social partners in rail, it appears that transfer of personnel is required in Czech Republic, Denmark, Spain, France, Italy, Netherlands, UK and Norway and is optional in Austria, Bulgaria, Germany, Ireland, Poland and Sweden[61]. Additionally, in Austria, Denmark, Germany Netherlands and Sweden, where there are sectoral agreements, it is not strictly necessary to prescribe social criteria in tender documents, as social standards apply to all operators.

3.7.        Licensing of railway undertakings

Germany continues to top by far the number of licences with 417 licences granted, followed by Poland. It is interesting to note that while all German and Dutch licences are reported to the European Railway Agency (ERA), more UK licences are notified to national authorities than to ERA (probably because most railway operations are confined to the UK, given the low degree of cross-border freight traffic and the fact that so far only Eurostar provides cross-border passenger services). Another interesting peculiarity of the UK is the issue of licences for ‘operators of last resort’ – in case of bankruptcy of franchise[62] operators or prolongation of tendering procedures. The number of licences is obviously very low in those Member States where the incumbent still has a monopoly on domestic passenger services.

Sources: European Railway Agency (ERADIS database),          RMMS questionnaires filled by Member States - annex 16 of Staff Working Document SWD(2014) 186

Most licences reported to ERA are for freight (51%) – in particular in Germany and Poland.  According to ERA, some 200 licences have been reported for passenger rail, whereas some 550 concern only freight – and 336 covered both.

Sources: European Railway Agency (ERADIS database)- annex 16 of Staff Working Document SWD(2014) 186

3.8.        Degree of market opening

Under EU law, the market of railway freight services have been open since 2007 and the one of international passenger services since 2010. For railway markets that are not open, it is important to underline that there is currently no obligation at EU level to open passenger domestic markets and that these represent 94% of all passenger-km in the EU.

In the context of the impact assessment of the 4th railway package, Commission services have estimated that in 2010, some 40% of the EU domestic passenger railway market has been so far accessible to new entrants. Only two Member States (Sweden and the UK) have fully opened their commercial services and services under public service obligations, and across the EU 40% of all passenger-kilometres are closed to competition. Repeating this exercise in 2012, based on 2010 assumptions,[63] leads to the same results.

Graph 46 – Rail market structure (2012)

Source: RMMS questionnaires, impact assessment 4th railway package, CER (2010), own estimations

10 Member States (with asterisk in graph), representing 20% of all passenger-kilometres, have opened markets in a way that allows commercial services in open access to co-exist with directly awarded PSCs. Further to the Bundesgerichtshof[64] decision calling for generalised competitive tendering, Germany will not be part of this group anymore. In Estonia, Latvia, Lithuania and Slovakia, full open access co-exists with directly awarded PSCs covering all rail services. PSCs in these Member States should be de jure competitively tendered, however de facto only the incumbent participated. 

This percentage has not changed substantially since no further opening has been reported, except for rail tourist services in Spain. It is important to underline that press reports[65] suggest that the Czech Republic is considering increasing the tendering of public service contracts and Spain ponders to open to competition some of its commercial services. It is also important to underline that in Germany, the number of train-kilometres in PSO that have been competitively tendered has progressed to up to 51%.

In spite of this situation, European railway undertakings have been pursuing their internationalisation. The shares of foreign sales of the main European railway groups have become significant over the years:

· DB appears to have made 42% of its turnover in 2012 outside Germany (as it controls inter alia Arriva and also the main rail freight undertakings in the UK, Denmark and the Netherlands),

· SNCF derives 24% of its turnover from outside France (even though the French market is fully closed to foreign competition), most notably through its subsidiary Keolis (for  suburban trains) and its shares in NTV, Westbahn, Eurostar and Thalys.

· NS, operating in a Dutch passenger market that is almost fully closed to competition, appears to be making 38% of its turnover outside the Netherlands through its subsidiary Abellio.

· Trenitalia is also active outside Italy (Netinera, the German subsidiary of Trenitalia represents some 5% of the latter’s turnover).

3.9.        Harmonisation between Member States

Rail Market Scoreboard

17 rail-related Directives have been adopted since 2000. As of November 2013, the transposition period of 15 of them has expired while it remains open for two (Directives 2012/34 and 2013/9). The transposition deficit indicator – which was developed for the purpose of the Single Market Scoreboard[66] – indicates the percentage of Directives not transposed by a Member State through national measures duly notified to the Commission, in relation to the total number of Directives that should have been transposed by the cut-off date (10th of November or 10th of April, as the Single Market Scoreboard is published twice a year). The transposition deficit of the rail Directives appears significantly higher than the general one for the Internal Market.

Source: European Commission

The rail related transposition deficit was initially of 56%, which reflects the low timely transposition of the first railway package into national laws (due April 2003). All of the 2004's entrants had transposed it when joining which, along with the absence of new Directives to be transposed by November 2004 and 2005, led to a decrease. The two peaks of 2006 and 2010 reveal a low on-time transposition of the second and third railway packages. Although the deficit was practically resorbed in 2013, it may rise again with the expiry of the transposition period for Directive 2013/9 amending Directive 2008/57 (due in January 2014). As of late November 2013, only eight Members States had notified national measures (BG, HR, IE, GR, IT, LV, PL, SE).

Source: European Commission

Overall, the on-time transposition is low. Since 2000, only 32% of rail Directives were transposed on time while 6% required more than two additional years to be transposed. If looking at Member States individually, Romania performs best with 60% of on-time transposition. On the other hand, Austria and Portugal have not transposed one single Directive on time since 2000.

Average delays for overdue Directives show that in four Member States it takes on average 18 "extra" months for rail Directives to be transposed (NL, LU, DE, UK). If considering market Directives only, it goes beyond two years in one Member State (NL).

Source: European Commission; **- as explained in footnote 67, the average overall delays for internal market reflect a general trend

Although subject to some caution, the comparison of this data with the general trend for all of the internal market Directives[67] shows that 21 Member States perform poorer for rail markets Directives than they do in other fields. This figure goes down to 13 Member States if only taking into consideration rail interoperability and safety Directives. The difference between rail and general delays is particularly important for the afore-mentioned four Member States (NL, LU, DE, UK), but also for SE, SI and GR. On the contrary, five Member States perform better on rail Directives than they do for the internal market overall (EE, SK, FI, RO, BG).

Graph 50 – Infringements in rail legislation

Source: European Commission

The Commission has initiated 134 infringement procedures on rail related provisions: 130 on Directives and 4 other on Regulations. By topic, 32% are on rail market provisions and 68% on interoperability and safety. Yet, an analysis by type of infringements shows that market provisions account for a disproportional 56% of all non-conformity and incorrect application infringements. Conversely, 95% of all non-notification infringements deal with interoperability and safety Directives.

For reasons of late or incorrect transposition, Poland, Germany and France are the Member Stares against which the most infringements were initiated. On the other hand Bulgaria and Luxembourg have had the least.

However, France and Italy come first in terms of infringements for non-conformity of the transposing measures or incorrect application of the rail related Directives. Netherlands, Luxembourg and the UK, three of the slowest Member States to transpose rail Directives, have had only one non-conformity/incorrect application infringement initiated against them.

Graph 51 – Infringements per Member State and per topic

Source: European Commission - (*) infringements dismissed by European Court of Justice are not included

Source: European Commission - (*) infringements dismissed by European Court of Justice are not included

3.10.      Development of employment and social conditions

3.10.1.   Employment in rail

Based on data provided by Member States in the RMMS questionnaires, it appears that some 912,000 persons have been employed either in railway undertakings (561,000) or in infrastructure management (351,000)[68]. Employment in these two types of railway entities seems to have diminished by 4% in 2012 compared to 2011. Interestingly, as shown in the graph hereunder, the overall percentage of staff in infrastructure management is generally higher in South and Eastern Europe and lower in Northern Europe.

Source: RMMS questionnaires

**/*for France, the share of jobs as provided by France for the infrastructure management comprises both employees of SNCF Infrastructures and Réseau Ferré de France under (**) and exclusively Réseau Ferré de France (*).

Furthermore, new entrants appear to be employing some 118,000 (21%) of the total. There appear to be also some 118.000 train drivers (21% of all employees) across the EU, incumbents and new entrants combined. 

Finally, 4 of the 5 most staff intensive infrastructure managements (LU, AT, BE, SE, SK) in terms of jobs per kilometres of infrastructure all operate small sized networks, whereby different levels of outsourcing (infrastructure construction and maintenance) and different scopes of activities remain out of consideration.

Source: RMMS questionnaires, annual reports RFIREFER/RFF

3.10.2.   Socio-demographic structure of the rail labour market [69]

In all European Member States, the proportion of men working in the rail sector is superior to the male share of the overall active population. This overrepresentation is among the highest in south European Member States  such as GR, ES and IT but also in LU and BE. In GR and LU, more than 90% of all staff are men. A survey organised by the European social partners in the rail sector[70] among EU railway undertakings estimates the share of women at 19.5%, but indicates that women are greatly underrepresented in some professions like drivers, where only 1.4% of the workforce is female (in managerial positions, 18% of the workforce is female). According to this survey, the situation tends to be better in Eastern Europe (in Latvia, 58% of engineers are female).

The rail sector is further characterized by an acute underrepresentation of young people (less than 30 years old). In all Member States where data is available, the share of staff under the age of 30 (blue bar on the graph) is significantly lower than the proportion of young people in the overall active population (light grey area on the graph). The percentage of people under the age of 30 is the lowest in southern Member States  (GR, ES and PT) but also in PL or CZ. On the other hand, FR, LV and LU have the largest proportion of young rail staff, albeit still lower than the national average. Conversely, the share of rail staff over the age of 50 (pink and red bars on the graph) exceed the according proportion in the overall active population (dark grey area) in all the Member States studied except RO. In GR and FI, more than half of the rail staff is over 50 years old. In another 13 Member States, staff over 50 account for one third or more of the total (AT, BE, BG, CZ, DE, EE,IT, LT, LV, NL, PL, PT, SK). In France, the proportion is only slightly over the national average (27.5% in the rail sector against 26.8% nationally). Although data is missing for countries such as SE and UK where the rail sector appears to be particularly dynamic, this section seems to indicate an overall ageing of the rail labour force that may lead to a staff shortage once the current staff reach retirement age.

Source: UIC (2010)

In most of the studied Member States, the percentage of recently recruited staff is low generally due to downsizing of the labour force in line with other economic sectors. In only 3 Member States is the share of staff with less than 5 years of experience superior to the share of staff with more than 30 years (LT, LV and RO). On the contrary, in 4 Member States, the largest of all groups by seniority is the +30 years one (BE, FI, DE and LU). In GR, PL and ES, more than 80% of the staff has over 20 years of seniority. In another 10 Member States, this proportion is above 50% (SI, BG, PT, AT, DE, CZ, SK, HU, BE, FI). Due to the overall ageing of the labour force, this low replenishment of staff may have adverse effects for the sector, which may imply it has to downsize.

These evolutions plead for a thorough monitoring in terms of statistics, notably to determine which professions are most affected by this greying.

Source: UIC (2010)

3.10.3.   Training activities

BG, DK, FI, GR, HU, LT, NL, PT and UK reported in their RMMS questionnaires their main training activities. In most Member States, training activities have been focused on drivers training and safety. The following schemes deserve attention:

· In DK, railway undertakings and infrastructure managers have developed a common training programme to support competence acceptance between employers.

· In NL, a new train drivers school, fully independent from railway undertakings, has been accredited by the Ministry of Education in 2011. 37 students graduated in spring 2013.

· In UK, the main infrastructure manager, Network Rail, created the cross-rail industry paid graduate programme "Track and Train", where recent graduates are placed 3 times for 6 months in different areas of Network Rail and also in railway undertakings.  

3.10.4.   Other aspects of working conditions

A study of the European Foundation for the Improvement of Living and Working Conditions[71] provides some anecdotal evidence of the evolution of wages, temporary contracts and flexible working. Although there appear to have been increases in wage inequalities in Germany between old and new employees, there were actual wage increases in Belgium, France, Italy, Lithuania and Sweden. The same study indicates that temporary work remains very marginal in the rail sector, except in Slovenia, where most contracts are temporary and are addressed by a collective agreement. Initiatives appear to have taken in Slovakia, Spain and, Czech Republic as regards flexible working time.

4.           State of the EU network and infrastructure limitations

4.5.        Variation of infrastructure

Between 2009-2011, EU railway infrastructure has grown by 882km (+0,4%) reaching 216.297 km. This increase masks variations among Member States: 981km and 602km of railway infrastructure were added in respectively France and Spain, whereas 335km, 203km and 138km were withdrawn in Austria, Bulgaria and Germany. In relative terms, the greatest growth was recorded in the Netherlands (+130km, hence +5%) and the largest decrease in Estonia (-127km, hence – 14%).

Source: EU Transport in figures, 2013, Statistical Pocketbook 2013

Since 1995[72], some 12.958km of rail infrastructure have been abandoned, with the largest decreases taking place either in the Eastern Member States (Poland, Latvia, Estonia) or in Germany (-8412km or a loss of 20%), and the largest increases taking place in Slovenia/Croatia (+18%) and Spain (+1624km or +11%).

Source: EU Transport in figures, 2013, Statistical Pocketbook 2013

4.6.        Management of infrastructure

The structures of management of infrastructure have not evolved during the period 2011-2012, except in Belgium, where the NMBS-SNCB Holding has been dismantled. Infrabel and SNCB-NMBS are now two separate entities that have a common subsidiary in charge of human resources.

4.7.        Electrification

Overall, in terms of electrification, in 2011, only 53,2% of the EU rail infrastructure is electrified. There again, there are important variations between Member States: electrification is low in the UK, the Czech Republic (33% of the network) and extremely low in Greece (17%) and the Baltic Member States (6%).

4.8.        Train movements

The analysis of train movements in TEN segments[73] underlines the expected importance of rail freight traffic around cities, their marshalling yards and ports. In 2010, 44 out of the 56 EU rail segments with freight traffic above 60,000 trains/year[74] (i.e.164 trains/day) were located in Germany and 7 were located in Austria (all in the Vienna-Salzburg axis). In Germany, the most important segments are around the marshalling yards of Maschen (Hamburg), Seelze (Hanover) and Oberhausen (Duisburg). Important movements of freight trains are recorded around cities like Cologne, Warsaw, Krakow and Innsbruck.

The analysis of train movements also suggests that the important freight traffic in Germany is feeding the port of Hamburg, and that Antwerp and Rotterdam are lagging behind in terms of rail freight connections. Rail traffic around the port of Rotterdam only reached 32,600 trains/year (i.e. 89 trains/day)[75]- i.e. Rotterdam rail freight only represented 25% of Hamburg’s rail freight, where some 140.000 trains/year (i.e. 383 trains/day) circulated. Traffic in the rail freight dedicated Betuwe line reaches "only" some 18.000 trains/year (50 trains/day). These conclusions seem to be confirmed by a study of UIC which shows that the rail share of hinterland transport in Hamburg reached 36.8% against 11% and 12% for respectively Rotterdam and Antwerp[76].

Similar trends can be seen for passenger rail: the most important passenger rail movements are found in the vicinity of important central stations with important commuter railway networks, like London and Paris, but also in many German cities (Berlin, Frankfurt, Cologne, Stuttgart) and important European cities (Amsterdam, Barcelona, Helsinki, Stockholm and Vienna), which all have traffic above 200.000 trains/year (some 550 trains/day). Interestingly, segments around Rome and Madrid remain at 140.000 trains/year, while Luxembourg-City manages 75.000 trains/year, i.e. as much as a 2-million inhabitant city like Budapest.

This suggests that cities with important suburban train networks overlapping with the remaining train infrastructure are or could become bottlenecks, either for high-speed services[77] or freight operations[78]. Similarly, in rail freight, data suggests that traffic flows from the Ruhrgebiet to the Netherlands (and probably Belgium) are well below those reaching out Hamburg, who happens to have the largest marshalling yard in Europe. It might well be interesting to analyse the impact of these North-South rail freight traffic flows on the overall congestion of the German network, in particular with the need to complete on the German side interconnections with Belgium and the Netherlands.

5.           Utilisation of access rights

5.5.        Railway freight

The rail freight market was opened to competition in 2007. Since then new entrants have emerged in all but 3 Member States (Finland, Ireland and Lithuania) and the total share of freight new entrants[79] in 2012 is estimated at 28%. DB, the German incumbent, is now the main railway operator in Denmark and the Netherlands. Finally, in the UK and Romania, the principal railway undertaking has less than 50% of the market.

Source: RMMS contributions from Member States; *for  Spain, data is based on the RENFE annual report and the RMMS questionnaire, whereas for Sweden, data is for 2010; ** includes VFLI, a subsidiary of incumbent; *** DB is the main railway freight operator; **** not available - annex 19 of Staff Working Document SWD(2014) 186

5.6.        Passenger railways – regional and suburban segments

5.6.1.     Passenger railways – all segments

Overall, the total market share of all but the principal railway undertaking reaches 23%. However, it is important to distinguish the situation of real new entrants, which have been awarded contracts through competitive tendering and/or through open access rights, from the one of other "non-incumbent" railway undertakings to which exclusive rights for commercial services or public service obligations have been directly awarded. This is important, for instance, to distinguish the situation of rail in Poland, where non-incumbents are regional operators, to which public service contracts have been directly awarded, and the situation in the UK, where franchises have been tendered. In these conditions, the market share of new entrants is estimated at 21% for the whole EU.

Source: RMMS contributions from Member States; RO= all non-incumbents appear to be regional operators; annex 19 of Staff Working Document SWD(2014) 186

5.6.2.     Regional and suburban segments

The share of new entrants in regional and suburban rail is the greatest in those Member States that have opted for competitive tendering for public service contracts completely (the UK, Sweden) or extensively (Germany, Romania).  The same also applies to Italy, Denmark, the Netherlands, the Czech Republic, Austria and Poland, where some public service contracts have been tendered. Hungary is a special case as there are two incumbents. In the UK, the incumbent has been respectively completely dismantled and franchises were awarded through competitive tendering. In Poland, the regional incumbent was dismantled and replaced by in-house operators owned by Polish regions (through direct award). There are no new entrants in the regional services inter alia in France, Spain and Belgium. Finally, in the context of a broadening of PSO tendering in the Czech Republic, Arriva has been doing some pilot tests services in open access in Prague suburban services.

Source: RMMS contributions from Member States, completed with own calculations based on UIC, CER, annual reports, annex 19 of Staff Working Document SWD(2014) 186 *there are 2 incumbents - this is the market of the small one; **no data for Sweden

5.6.3.     Rail high-speed and long-distance segments

The share of new entrants in long-distance services is the greatest in the UK, where the incumbent has been dismantled, and Estonia, where the public service contract for long-distance services has been awarded to a new entrant. There are new entrants in Sweden (Veolia, BluTag), Italy (Italo-NTV in the Italian high-speed network), the Czech Republic (LeoExpress and Regiojet, both in the Prague-Ostrava route) and Austria (Westbahn in the Vienna-Salzburg route).

A number of companies have made known their intentions to open domestic commercial services in several Member States. MTR, part of the broad MTR Group from Hong Kong, has indicated its intention to start new services on the Stockholm-Goteborg line. Leo Express/DLA applied to enter the Warsaw-Krakow/Katowice and Warsaw-Szczecin routes, but the application was refused on administrative grounds[80].

Source: RMMS contributions from Member States, completed with own calculations based on UIC, CER, annual reports, annex 19 of Staff Working Document SWD(2014) 186 *all new entrants appear to be regional new entrants ** no data for Sweden

5.7.        Passenger railways – international services

Thello a joint venture between Veolia and Trenitalia operating night services between Paris and Venice is the only new entrant in international services. DB has indicated that it intends to operate as a competitor to Eurostar on the Brussels-London line (originating in Frankfurt). Thello has also applied for a Belgian licence and the city of The Hague appears to have signed an agreement with DB Arriva to link it with Brussels. Finally, thanks to the opening of the France-Spain high-speed connection, SNCF and RENFE are now offering Paris-Barcelona and Madrid-Marseille services.

6.           Barriers to more effective rail services

The Commission has identified the lack of opening of domestic passenger rail markets (which cover 94% of all passenger-km) and the inadequate portfolio of functions for infrastructure managers as the two main obstacles in the area of market access for a fully functioning Single European Railway Area. On the 30th January 2013, the Commission adopted the 4th railway package proposals to modify Directive 2012/34/EU and Regulation (EC) No 1370/2007 on public passenger transport services by rail and by road[81] to fulfil these objectives. 

7.           Conclusions

The assessment of this report confirms the analysis made by the Commission when submitting the 4th railway package proposals: the level of satisfaction with railway services leaves great room for improvement, many high-speed infrastructures are under-used (and their usage could be increased thanks to open access) and it is necessary to ensure the efficiency of the important amounts of public funds devoted to rail (some € 36 billion for infrastructure grants and public service obligations) to ensure the sector's long-term viability in a context of constrained public budgets. The assessment of this report also confirms the need to proceed with the adoption of several implementing acts under Directive 2012/34/EU (direct costs, economic equilibrium of public service obligations, RMMS statistics, template for licences…).  Furthermore, it also shows that measures to improve accessibility of persons with reduced mobility are likely to have a positive impact for ridership overall. The completion of railway freight corridors can help rail to increase its average transport distances (which make rail more cost-effective and thus more competitive). Finally, this report also supports a research and innovation programme under 'Shift2Rail' devoted to the improvement of the quality of rail services, the reduction in the life-cycle cost of railway transport, and an overall increase in reliability in the different rail market segments, including rail freight (which needs to move more fluidly throughout Europe carrying more high-added value products).

[41]             The Consumer Scoreboard’s Market Performance Indicator is a composite index that takes into account 4 key aspects of consumer experience (easiness to compare services, consumer’s trust that seller complies with consumer protection law, problems experienced and overall consumer satisfaction).

[42]             OJ L 315, 3.12.2007, p. 14.

[43]             Consumer Scoreboard 2012, page 27 – the internal market averages are: 9% of 0-4 rating and 37% of  5-7 rating (for rail, the latter is 40%)

[44]             For more information, see also Annex 12 of Staff Working Document SWD(2014) 186

[45]             The Flash Eurobarometer conducted in 2013 on the quality of rail services follows up a similar exercise performed in 2011 – however, the Flash Eurobarometer of 2011 does not cover suburban services. As a consequence, comparison between the two surveys is only done for national, international and regional services ("conventional").

[46]             Ibid.

[47]             Consumer Scoreboard 2012, page 24

[48]             Cf. Annex 12 of Staff Working Document SWD(2014) 186

[49]             Although the focus is on frequency, other time-related variables play an important role in traveller's decision such as waiting time, distance to station, in-vehicle journey time, transfers, ...

[50]             i.e. respondents that are fairly or very dissatisfied with punctuality and reliability – excludes respondents without opinion

[51]             Ireland, Czech Republic and Estonia did not provide any data.

[52]             UIRR, Road-Rail Transport: new developments and best practices, 55. Session of the UNECE Working Party on Intermodal Transport and  Logistics, 6-7 November 2011, available at – this data seems to be confirmed by CER (2013) Rail Freight Status Report 2013, where less than 70% of freight trains arrived within 15 minutes of their scheduled time during the period 2008-2012.

[53]             17,8 billion in 2011 (no data for Slovenia and Finland) and 18,4 billion in 2012 (no data for The Netherlands)

[54]             The "net public support" to a 140 tonne suburban train in France is estimated to amount only to 7.1 EUR/train-km, as 10.9 EUR/train-km serve to pay track access charges.

[55]             The "net public support" to a 140 tonne suburban train in Germany is estimated to amount only to 4.8 EUR/train-km (deducting track access charges).

[56]             The number of tendered contracts in a Member State also depends on the contract size. There are only 19 franchises in the UK, whereas based on the Mofair report, there are at least 31 railway undertakings operating one or several public service contracts in Germany.

[57]             Contract notices have been interpreted as calls for competition and not notices used only for transparency (for which ex ante voluntary notices or contract award notices are best suited)

[58]             The German RMMS questionnaire reports the tendering of 49 PSCs in 2009-2011, whereas there are 45 PSCs in OJEU/TED during the same period.

[59]             As far as rail is concerned, the number of contract award notices for negotiated and accelerated procedures without prior publication and voluntary ex-ante notices has doubled since 2009: there were on average 12 publications in 2006-2009 and 22 for the period 2010-2012.

[60]             Social criteria, if used, are regrouped statistically under this award criteria

[61]             CER-ETF (2012), Social Aspects and the protection of staff in case of change of railway operator: the current situation, pp.66-67 – in some Member States there are provisions of transfer of staff for urban transport (which explains why some Member States that did not open their PSOs to tendering are listed.

[62]             Franchises are UK public service contracts

[63]             These assumptions have been based mostly on CER (2010) Public service in rail transport in the European Union: an overview

[64]             Bundesgerichtshof Beschluss X ZB4/10 vom 8. Februar 2011 S-Bahn Verkehr Rhein/Ruhr

[65]             Railway Gazette (2013), Czech passenger market opening in sight, 17.10.2013

[66]             Single Market Scoreboard:

[67]             Due to the limited number of rail Directives (15) compared to internal market's (over 1500), the calculation method for average delays of transposition differ. While the internal market scoreboard calculates the average delays at a cut-off date each year, the rail statistics show the average 'net' delays for the 15 rail Directives (i.e. between the end the transposition period and actual notification by a Member State). The graph therefore indicates a general trend rather than a statistical-based accuracy. Sources: internal market scoreboard n°12,14bis, 15,16,20,21,22,23,24. The indicator only catches Directives not transposed on-time (there was a single case of full non-transposition, but where the non-transposed Directive was repealed in the meantime).

[68]             At least some 63,100 additional persons are working with railway service facilities – however, numbers remain insufficiently reliable to provide an EU-wide estimate.

[69]             Unless otherwise mentioned, all data in this section comes from the 2010's International Railway Statistics published by UIC (International Union of Railways). It provides no comprehensive datasets for IE, DK, UK and SE, which could therefore not be included in the age structure and seniority studies. As for the structure per gender, data were provided by Eurostat for SE and UK but not for IE and DK. These two Member States are thus completely excluded from the scope of this section. Moreover UIC datasets for a certain number of Member States deviates significantly from the figures indicated in the RMMS questionnaires completed directly by the Member States and transmitted to the European Commission's services. For these Member States, the indicators developed in this section then ought to be interpreted with caution. These Member States are: DE (inclusion of DB Schenker in the UIC statistics), NL (Infrastructure manager ProRail not included in UIC), AT, BE, PL, RO (fragmental UIC datasets).

[70]             CER (2014), Results of the 2013 questionnaire on the development of women employment in the railway sector in Europe, published on 20.01.2014. This survey was a joint exercise between the European social partners in rail:  employers (railway undertakings represented by CER, the Community of European railways) and workers, (represented by ETF, European Transport Workers Federation)

[71]             Employment and industrial relations in the railway sector, study of the European Foundation for the Improvement of Living and Working Conditions, 2012

[72]             The first TEN decision dates back from 1996.

[73]             Data on traffic in TEN infrastructures and, in most cases in non-TEN, is collected by Eurostat as part of Annex G of Regulation 91/2003 every 5 years (2005 and 2010, which was just recently released). Data for Belgium and Greece is missing for 2010, data for Bulgaria and Romania was not recorded in 2005, as these were not yet EU Member States.

[74]             30,000 in each segment direction – hence 60.000 trains in both directions.

[75]             Data for Antwerp is not available in 2010.

[76]             UIC (2012): 2012 Report on Combined transport in Europe, December 2012, p.80

[77]             As indicated under quality of services, frequencies of high-speed services still have room for growth (except in France).

[78]             Swedish rail freight undertakings have identified Copenhagen, which manages 180.000 trains/year, as a major difficult cross point. In Germany, the highest increases of track access charges took place on regional lines.

[79]             Given that the freight sector does not have any more public service obligations and is legally completely open, "Railway undertakings other than the principal undertaking" give a good estimation of the market penetration of new entrants (since the market leader is almost always the incumbent). In the UK, further to the dismantling of British Rail, it could be argued that all rail freight undertakings are new entrants.

[80]             PKP PLK rejects Leo Express access bid, International Rail Journal, 22.07.2013 – It appears that the application was rejected as the regulator had not  been consulted on the impact of those services on the economic equilibrium of public service obligations.

[81] OJ L 315, 3.12.2007, p. 1.