[pic] | COMMISSION OF THE EUROPEAN COMMUNITIES |

Brussels, 20.12.2007

COM(2007) 832 final

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT AND THE COUNCIL

on experience acquired in the application of Regulation (EC) No 91/2003 of the European Parliament and of the Council of 16 December 2002 on rail transport statistics

TABLE OF CONTENTS

EXECUTIVE SUMMARY 3

1 INTRODUCTION 3

1.1 BACKGROUND AND OBJECTIVES 3

1.2 POLICY CONTEXT 4

1.3 IMPLEMENTING RULES 5

1.4 COVERAGE OF MEMBER STATES AND OTHER COUNTRIES 5

2 IMPLEMENTATION OF THE REGULATION 7

2.1 ISSUES ENCOUNTERED 7

2.1.1 Respect of legal obligations 7

2.1.2 Burden of the implementation and regular data provision for reporting countries 9

2.1.3 Data collection, compilation and validation process 9

2.1.4 Methodological issues 10

2.2 RESULTS 10

2.2.1 Data transmission and validation 10

2.2.2 Methodological guidance 11

2.2.3 Data dissemination 11

3 CONCLUSIONS 13

3.1 SHORT TERM DEVELOPMENTS 14

3.2 LONG TERM DEVELOPMENTS 14

EXECUTIVE SUMMARY

Regulation (EC) No 91/2003 of the European Parliament and of the Council of 16 December 2002 on rail transport statistics[1] specifies in its article 9 that, after data have been collected over a period of three years, the Commission shall submit a report to the European Parliament and to the Council on experience acquired in the application of this Regulation. This report is aimed to comply with this requirement once the first 3 years of data collection have been completed (2003-2005).

After a general description of the background, objectives and policy context of the legal act, some issues linked to the implementation are discussed. The report also presents information on the main results achieved. The final section provides conclusions and possible future developments.

The implementation of this Regulation led to the availability of detailed information on European freight and passenger transport by rail, as well as data on rail accidents and marked an improvement in the data quality and timeliness. Member States have made efforts to fulfil the requirements of the Regulation and these efforts need to continue in order to improve the timeliness and completeness of data provision.

1 INTRODUCTION

1.1 BACKGROUND AND OBJECTIVES

The Regulation 91/2003 was designed to provide the Commission, other EU Institutions, national governments and the general public with comparable, reliable, harmonised, regular and comprehensive statistical data on scale and development of the carriage of freight and passengers by rail in the European Union.

This is achieved by expanding certain essential aspects of the data previously collected by the Council Directive on rail goods transport statistics (80/1177/EEC) to include information on passengers transport, traffic flows and accidents on rail network; information on type of dangerous goods; information on type of transport unit and information on national and international goods and passenger transport at regional level (NUTS2).

The main features of Regulation 91/2003 can therefore be summarized as follows:

- Data provision should cover all railways in the Community.

- Data provision is defined for both freight and passenger transport by rail. The data collected include annual and quarterly statistics on goods and passenger transport.

- Origin-destination data provision obligation for both freight and passengers is geographically defined at NUTS2 level.

- The information collected includes statistics on traffic flows on the rail network for freight and passenger trains.

- Data provision covers the statistics on rail accidents.

1.2 POLICY CONTEXT

The development of the Common Transport Policy requires the best possible knowledge of the extent and evolution of passenger and freight transport by rail. The White Paper “European transport policy for 2010: time to decide”[2] defined the objective of the Community Transport Policy as to help provide Europeans with efficient and effective transportation systems. This objective has been maintained in the Mid-term Review of the White Paper[3] although more oriented towards ensuring sustainable mobility in Europe. The renewed EU Sustainable Development Strategy[4] sets sustainable transport as a key objective for EU. Thus, Community rail transport statistics are used for framing, monitoring and evaluation of EU transport policy and other EU policies such as the renewed Sustainable Development Strategy and the functioning of the internal market.

The importance of these data for EU decision-making is increasing, for instance, following the competence attributed to the Commission on behalf of the 2nd and 3rd railway packages of proposals.

The second railway package consists in three directives (safety, interoperability and market opening directives) and one regulation on the European Railway Agency.

The third package of legal measures includes two directives (a directive on a further opening of the market for international passenger transport by rail and a directive for train driver licences) and one regulation (on the rights and obligations of rail passengers).This information is also used for the measurement of passenger and freight flows and the calculation of the modal split of transport in the EU.

Other examples of the use of these data for policy needs are the following:

- The European Railway Agency (ERA) is also interested in exploiting these data for its work, e.g. for the development of common safety indicators.

- There is also a potential to use these data for environmental policy purposes in comparison with other modes of transport. The data need is growing with the political will to reinforce the railway transport for environmental reasons. These statistics are relevant to TERM transport and environment reporting mechanism, which is designed to inform the EU transport policy makers and is developed by the European Environment Agency in cooperation with Member States experts and the European Commission.

1.3 IMPLEMENTING RULES

Regulation 91/2003 confers implementing powers on the Commission, which are to be exercised via the comitology procedure as set out in Articles 10 and 11. Eurostat has used this procedure to adopt a set of rules covering different aspects of the implementation of the Regulation. These rules are adopted in the form of Commission Regulations after opinion of the Statistical Programme Committee.

So far, two implementing Regulations have been adopted:

- Commission Regulation (EC) No 1192/2003[5]of 3 July 2003, which mainly provides harmonised definitions to be used in railway statistics by all countries providing data but also adapts the content of the annex related to statistics on accidents;

- Commission Regulation (EC) No 332/2007[6]of 27 March 2007, which provides the technical format for data transmission (datasets to be provided with the corresponding structure and codification).

One more Commission Regulation is currently being discussed in the relevant committees. It aims at harmonising a number of definitions and amending Annex H of Regulation 91/2003 for it to be consistent with Annex 1 of the Railway Safety Directive 2004/49/EC.

1.4 COVERAGE OF MEMBER STATES AND OTHER COUNTRIES

Regulation 91/2003 applies directly and in its entirety to all Member States. It does not have to be transposed into national legislation. However, the gradual introduction of the Annexes of Regulation (for certain railway companies, datasets and variables) according to certain rules and for a maximum period of time of three years (until 2005 included) has been introduced.

Concerning Romania and Bulgaria, both countries have declared their willingness to fully comply with the EU legislation in this field, starting from 2006 as a reference year for the data provisions. Romania has provided almost all 2005 data according to the provisions of the Regulation.

Concerning the candidate countries, Croatia and Turkey provide some data, while the Former Yugoslav Republic of Macedonia has not yet submitted any information.

In 2003, Regulation 91/2003 was formally incorporated into Annex XXI (Statistics) of the EEA Agreement by a Decision of the EEA Joint Committee[7]. According to this Decision, Norway has to transmit the data required in the Regulation. Regulation foresees provision for simplified reporting and this is applied for Liechtenstein data[8].

Data based on the Regulation principles will be also transmitted by Switzerland in accordance with the Agreement between the European Community and the Swiss Confederation on cooperation in the field of statistics[9]. Switzerland shall collect the data required by this Regulation from, at the latest, the beginning of year 2006[10].

2 IMPLEMENTATION OF THE REGULATION

2.1 ISSUES ENCOUNTERED

2.1.1 Respect of legal obligations

The compliance with the data provision obligations defined in the Regulation has experienced a positive evolution during the period considered. The evolution trend was mostly caused by the fact that the provisions of the Annexes have been introduced gradually as follows:

- 2003 data – Tables A1, A2, A3 and Annex I were required for the first time;

- 2004 data – Tables A4, A5, A6, A7, A8 and A9, Annex B, Annex C, Annex D, Annex E and Annex H were required for the first time;

- 2005 data – Annex F and Annex G were required for the first time.

Table 1 shows the status of the data delivery for the reference year 2005 by dataset and periodicity. According to Regulation 91/2003:

- datasets related to Annex E (Statistics on goods and passenger transport) are quarterly;

- datasets of Annexes A, B, C, D (Statistics on goods and passenger transport) and H (Statistics on accidents) are annual;

- datasets of Annexes F (Regional statistics on goods and passenger transport) and G (Statistics on traffic flows on the rail network) are five-yearly (and are to be provided beginning with the reference year 2005).

Table 1: Compliance with Regulation requirements regarding timeliness and completeness in 2005 (reference year)

Quarterly data delivery | Annual data delivery | 5 years (beginning 2005) |

T | timeliness problems |

M | some tables are missing |

Q | quality problems, incompleteness of data |

– | not concerned |

Concerning the data provision for 2003, some reporting countries had problems with the delivery of certain datasets (mainly regarding statistics on accidents, etc.) or certain elements (train movements, etc.). The data provision in the two subsequent years has improved, although efforts are still needed in order to improve the periodicity and completeness of the datasets.

2.1.2 Burden of the implementation and regular data provision for reporting countries

In the majority of reporting countries the compliance with the Regulation has meant a straightforward exploitation of the existing data and therefore, no specific measures have been taken, as the existing processes were sufficient to satisfy the requirements. However, in some countries, extensions of the data collection and additional compilation procedures were needed. It has to be mentioned that no financial support was foreseen in the Regulation to cover the cost of its implementation.

Concerning the regular data provision, the workload involved is considered as acceptable by a majority of reporting countries. Nevertheless, in some countries this involves a major burden, particularly for the railway companies as original data providers and for the CNA of countries where the EU liberalisation rules are applied and, as consequence, many new railway companies are active.

In terms of costs, a pilot study launched in 2006 on behalf of the Working Group on European Statistical System Programming and Coordination revealed a very heterogeneous situation, with an average annual cost of producing rail transport statistics varying considerably even among Member States with similar size and market structures. With a very few exceptions, the Member States confirmed that the figures presented in the cost-benefit study[11] are indeed reflecting the total annual cost of producing rail transport statistics according to the Regulation 91/2003.

2.1.3 Data collection, compilation and validation process

The data collection and compilation process in the reporting countries follows a traditional bottom-up flow of information. In general, this flow starts at the railway company who transmits the information to the CNA, which makes the compilation, validation and transmission of the national data to Eurostat.

In some cases, CNA use information from other sources, such as the Ministry of Transport, infrastructure managers or National Safety Authorities, particularly for statistics on traffic flows on the rail network and for statistics on accidents.

All CNA apply validation checks before sending the information to Eurostat, using internally developed control procedures and/or using a pre-validation software tool provided by Eurostat. The validation rules cover many aspects of the process including data format, codification, internal consistency of each dataset, consistency between datasets and variables, etc. and when errors are detected, they are manually corrected before transmitting the data to Eurostat.

2.1.4 Methodological issues

The methodological arrangements of implementing Regulation 91/2003 have provided a common framework to ensure a harmonised data collection and the production of comparable results. During the period considered, the following major methodological issues have been raised:

- Question of inclusion/exclusion of passenger transit;

- Methodological problems in establishing the correspondence table of the UIC station codes with the NUTS2 regions and one common conversion tool for NST/R-NST2000 (CPA, CN) for transport statistics purposes;

- Need for harmonisation of the definitions between Rail Safety Directive 49/2004 and the rail Regulations (together with changes in the data collection range) in order to improve quality of rail accident statistics;

- Methodological help on Annex G – Statistics on traffic flows on the rail network;

- Correspondence problems between Annex I and the data deliverables.

2.2 RESULTS

2.2.1 Data transmission and validation

Each Annex of Regulation 91/2003 provides for specific deadlines for transmission of data by the reporting countries. During the period considered, the compliance with these deadlines has improved.

Regarding the technical arrangements for data transmission, a high level of standardisation has been achieved. Data are transmitted electronically to Eurostat respecting the requested structure of the data files and the record format. This allows for a quick integration of the information into the Eurostat production database and an early detection of basic errors and unknown codes.

Once data have been loaded into the Eurostat database, detailed validation procedures are applied to ensure the quality of the transmitted data. The validation includes the internal consistency of the datasets, the consistency over time, the coherence between datasets and a comparison of the results for traffic between two reporting countries.

The data quality can be judged as good in freight transport (which is based on freight papers). The passenger data cannot reach the same quality level, as a result of different methodologies in data collection in each Member State. Concerning the accident statistics the data are sometimes 'polluted' by suicides and attempted suicides. Parts of accident data still suffer from a lack of national adaptation to the common definitions (probably because of the transition period of five years permitted in the Regulation 91/2003); the harmonised definitions mentioned above will improve the data quality. This statement is based on the implementation of a solid common methodological framework, on the efforts in the compilation and validation procedures at national level as well as on the subsequent quality controls performed at Eurostat before data dissemination.

The majority of Member States consider the level of compliance with the Regulation 91/2003 as good. Most difficulties are encountered for the provision of information at regional level due to poor data sources and aggregation tools, limited access to the data and, in some cases, data processing difficulties. However, in many reporting countries there are ongoing efforts to improve the situation, such as regular contacts with data suppliers to allow completing the data transmission, or development of new data processing systems.

2.2.2 Methodological guidance

Eurostat has produced a Reference Manual for the implementation of the Regulation 91/2003 to provide guidance to Member States and other reporting countries on the implementation of the Regulation. The Reference Manual is updated annually to include the most recent information, documentation or guidelines relevant to the collection of these statistics.

Moreover, the correspondence table of the UIC station codes with the NUTS2 regions was established by Eurostat in cooperation with Member States and UIC.

2.2.3 Data dissemination

According to Article 7 of Regulation 91/2003, the Commission can disseminate all data not declared as confidential by the Member States. In practice, very few countries have used this clause and the large majority of the data delivered to Eurostat can be disseminated. The Annex I data (information on railway undertakings) is confidential based on the provisions of Regulation.

Eurostat releases the data collected through its dissemination database freely and is accessible from the Eurostat web site. There are already 19 tables which are filled in with the data coming from the Annexes A–E and H. This year (2007) the number of tables will increase because of dissemination of the data from Annexes F and G.

Eurostat also disseminates information though specific publications such as the series Statistics in Focus or news releases that provide relevance to the data for the media and the general public.

Based on the data which has been delivered within the provisions of the Regulation, it was possible to publish three Statistics in Focus for 2005 data:

- Rail transport accidents in the European Union in 2004-2005;

- Rail freight transport 2005;

- Passenger transport by rail in 2004-2005.

Table 2 contains some key figures on the European rail transport collected on the basis of this Regulation.

Table 2: Total[12] rail transport of passenger and freight and persons killed in rail accidents by country in 2005

Passenger transport (million pkm) | Change comparing to 2004 (%) | Freight transport (million tkm) | Change comparing to 2004 (%) | Number of persons killed in rail accidents[13] |

Belgium | 8305 | -4.3 | :c[14] | 5.7 14 | 24 |

Czech Republic | 6667 | 1.3 | 14866 | -1.5 | 249 |

Denmark | 5957 | 0.7 | 1976 | -14.9 | 23 |

Germany | 78244 | 7.4 | 95421 | 3.8 | 157 |

Estonia | 248 | 28.6 | 10639 | 1.4 | 21 |

Greece | 1854 | 11.1 | 613 | 3.6 | 26 |

Spain | 21153 | 4.1 | 11635 | -2.0 | 65 |

France | 76886 | 3.4 | 40701 | -9.8 | 79 |

Ireland | 1781 | 12.6 | 303 | -24.1 | 0 |

Italy | 50088 | 1.7 | 22761 | 2.6 | 99 |

Cyprus | - | - | - | - | - |

Latvia | 894 | 10.3 | 19779 | 6.2 | 5 |

Lithuania | 280 | -1.0 | 12457 | 7.0 | 33 |

Luxembourg | 267 | 5.8 | 392 | -33.9 | 0 |

Hungary | 9851 | -3.1 | 9090 | 3.9 | 91 |

Malta | - | - | - | - | - |

Netherlands | 14036 | -0.4 | 5025 | -3.8 | 25 |

Austria | 9236 | 5.0 | 18957 | 1.1 | 44 |

Poland | 17882 | -3.0 | 49972 | -4.5 | 291 |

Portugal | 3809 | 1.5 | 2422 | 6.1 | 99 |

Slovenia | 716 | 3.0 | 3245 | 3.0 | 5 |

Slovakia | 2182 | -2.0 | 9463 | -2.5 | 7 |

Finland | 3478 | 3.8 | 9706 | -4.0 | 22 |

Sweden | 8910 | 3.2 | 21675 | 3.9 | 21 |

United Kingdom | 44415 | 2.5 | 22322 | -1.0 | 74 |

European Union 25 | : | : | : | : | 1464 |

Bulgaria | : | : | : | : | : |

Croatia | 1227 | 5.0 | 2835 | 13.7 | 35 |

FYROM | : | : | : | : | : |

Iceland | - | - | - | - | - |

Liechtenstein | 1 | 0.0 | 17 | -17.3 | 0 |

Norway | 2709 | 3.4 | 3149 | 10.7 | 4 |

Romania | 7960 | -7.6 | 16582 | -2.6 | 36 |

Switzerland[15] | : | : | : | : | : |

Turkey | 5036 | -3.8 | 9077 | -2.7 | 154 |

3 CONCLUSIONS

The experience gained and the results obtained on the implementation of Regulation 91/2003 can be considered, in general, as very positive. The adoption of the Regulation, the work and resources put into setting up the legal requirements, both at national and Eurostat level, have allowed a quick production of comparable and reliable results . This is particularly relevant in the case of the ten Member States that joined the EU in 2004.

More problems concerning rail data collection were encountered in the countries where the railway market is developed (there are many railway operators).

The data provisions in the two subsequent years have improved, although there are still remaining problems, particularly as regards the periodicity and completeness of datasets.

Because of the constant learning process (during the three previous years), all the 2006 quarterly data were already available in May 2007 (all Member States respected the deadline for the quarterly data).

A great progress has been achieved by some countries in order to release the confidentiality restrictions of some tables. At the beginning, 7 countries declared some tables as confidential, but with the contacts of NSIs with railway undertakings, the data could become public. Currently only three countries (AT[16], BE[17] and BG[18]) put some restrictions on the data. The procedure for the disclosure of data is still ongoing in these countries. As a result, the EU aggregates (for some variables) are still not available. This seriously hampers the monitoring of the rail market at European level and thus the usefulness of the datasets.

Most of the issues linked to the interpretation of the Regulation, data collection, compilation, transmission and validation have been addressed and properly tackled. The implementation of this Regulation has led to an increase in the availability of information on the sector as well as an improvement of data quality and timeliness. Regarding the use of the data, the results obtained have an immediate application for policy-making, particularly for the development, implementation and monitoring of the rail transport sector policies at national and EU level. Data dissemination through different public means has contributed to the visibility of the whole process and to provide a return to the citizens for the investment of resources made on it.

3.1 SHORT TERM DEVELOPMENTS

In the short term, the efforts should be concentrated on the following issues:

- Finalise the introduction of harmonized definitions for better comparability of data of different countries and different modes of transport.

- Integration and harmonisation of some rail accident data (which are currently collected separately by ERA and Eurostat) in order to avoid duplication of work and publication of different figures in future.

- Further improvement of data quality for certain elements and of the timeliness is necessary. Some reporting countries need to make additional efforts to supply reliable and comparable data and to improve the timeliness of data provision.

- Dissemination of traffic flows on the rail network and regional data;

- Enlargement of the EU to Bulgaria and Romania. Eurostat will pay particular attention to ensure that these two new Member States comply with the EU legislation on rail transport statistics .

- Full lifting of confidentiality restrictions in order to allow for a timely dissemination of EU aggregates.

3.2 LONG TERM DEVELOPMENTS

The collection of new variables may be foreseen and will be discussed, based on the needs announced by the data users (e.g. the train-km performed divided into diesel and electricity could be useful for environment impact calculations). The possibility of data collection on high-speed passenger trains (number of passengers, passenger-km and train-km) can also be discussed.

However, reporting countries have clearly expressed the problems of collecting more information and the need to duly justify any further burden linked to the collection of statistics in general and on rail transport statistics in particular .

At least two future developments not linked to new data requirements could be envisaged:

- A short reduction on the data transmission period to five months after the end of the period of observation for the annual data on rail passengers, in line with the statistical data collection of other modes of transport.

- Extension of the geographical coverage due to either future EU enlargements or the voluntary involvement of third countries through co-operation agreements.

[1] Regulation (EC) No 91/2003 of the European Parliament and of the Council of 16 December 2002 on rail transport statistics, OJ L 14, 21.1.2003, p.1.

[2] COM(2001) 370 of 12 September 2001.

[3] COM(2006) 314 of 22 June 2006: "Keep Europe moving – Sustainable mobility for our continent. Mid-term review of the European Commission's 2001 Transport White Paper".

[4] Adopted by the European Council on 15-16 June 2006.

[5] Commission Regulation (EC) No 1192/2003 of 3 July 2003 amending Regulation (EC) No 91/2003 of the European Parliament and of the Council on rail transport statistics, OJ L 167, 4.7.2003, p.13.

[6] Commission Regulation (EC) No 332/2007 of 27 March 2007 on the technical arrangements for the transmission of railway transport statistics, OJ L 88, 29.3.2007, p.16.

[7] Decision of the EEA Joint Committee No 12/2004 of 6 February 2004 amending Annex XXI (Statistics) to the EEA Agreement, OJ L 116 of 22 April 2004, pp.63-64.

[8] The Liechtenstein data are provided by Austria, as one (Austrian) railway undertaking operates on Liechtenstein territory only.

[9] Signed on 26 October 2004 and approved by Council Decision of 27 February 2006 on the conclusion of the Agreement between the European Community and the Swiss Confederation on cooperation in the field of statistics, OJ L 90 of 28 March 2006, pp.1-21.

[10] Annex A Transport and Tourism Statistics of the above-mentioned Agreement.

[11] The report was presented and discussed with MS in the WG ESS Programming and Coordination on 19 June 2006 (Item 2.1. on the agenda Review of priorities – results from cost-benefit pilot studies ).

[12] Simplified and detailed reporting, all types of transport (international, national, transit for freight transport only).

[13] Some data may contain suicides.

[14] Data confidential because of the confidentiality clause put on the simplified reporting (small undertakings); the data for big undertakings (detailed reporting) was 8130 million tkm in 2005.

[15] Switzerland shall provide the railway data starting from the 2006 as a reference year (according to the Agreement between the European Community and the Swiss Confederation on cooperation in the field of statistics, OJ L 90 of 28 March 2006, pp.1-21).

[16] Tables A3, F1 and F2 are confidential, tables C1, C2, C3 C4, E2, F3, F4 are under disclosure procedure.

[17] B1 for 2004 and 2005 are confidential.

[18] Tables E1 and E2 – ongoing procedure for disclosure.