[pic] | COMMISSION OF THE EUROPEAN COMMUNITIES |

Brussels, 13.12.2006

SEC(2006) 1641

COMMISSION STAFF WORKING DOCUMENT Accompanying the

COMMUNICATION FROM THE COMMISSION TO THE COUNCIL AND THE EUROPEAN PARLIAMENT "Facilitate interoperability of locomotives across the EU” PROPOSAL FOR A DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL amending Directive 2004/49/EC on safety of the Community's railways PROPOSAL FOR A REGULATION OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL amending Regulation (EC) No 881/2004 establishing a European Railway Agency PROPOSAL FOR A DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL on the interoperability of the Community rail system Full impact assessment {COM(2006) 782 final}{COM(2006) 784 final}{COM(2006) 785 final}{COM(2006) 783 final}{SEC(2006) 1640}{SEC(2006) 1641}

COMMISSION STAFF WORKING DOCUMENT

Communication of the Commission to the Council and the European Parliament, "Facilitate the interoperability of locomotives across the EU"

Full impact assessment

Executive summary

The acceptance of railway rolling stock across Member States is a growing concern in the railway sector. While the Interoperability Directives (96/48/EC[1], 2001/16/EC[2], both revised by 2004/50/EC[3]) have as their goal the full compatibility of the European railway network (infrastructure, power supply, rolling stock etc.), convergence towards this target is inevitably gradual, being dictated by renewal investment.

The Interoperability Directives prescribe the way rolling stock is to be accepted on existing infrastructure, while the Railway Safety Directive (2004/49/EC)[4] deals in addition with rolling stock already in use, when it crosses national borders. Whereas current legislation defines the responsibilities and procedures to be followed for various actors, stakeholders in the railway sector are asking for more action and short-term efficiency gains, to curb delays and costs associated with the acceptance of rolling stock across Member States.

The Commission has explored possible solutions to this problem, including voluntary and mandatory options. The feasibility of these options is described and evaluated in this report, which also provides a qualitative evaluation of their impact.

The assessment concludes that voluntary options, favoured by most stakeholders, might be implemented with some involvement from the European Railway Agency (ERA).

With regard to the regulatory options, the analysis shows that some of them are necessary to clarify the role of National Safety Authorities (NSAs) and to strengthen the application of the mutual recognition principle.

On balance, it appears that improved cross-acceptance would benefit both the contracting entities and the NSAs. Although few quantitative estimates could be provided (data are not easy to obtain and assess in this complex technical, commercial and political field), it is clear , in terms of the order of magnitude involved, that the costs of not having cross-acceptance as compared to its benefits fully justify the Commission initiative.

Table of contents

1. Procedural issues and consultation of interested parties 6

1.1. Organisation and timing 6

1.2. Consultation and expertise 6

2. Problem definition 7

2.1. Issue 7

2.2. Analysis of shortcomings 8

2.3. Scope of the problem 12

2.4. Reference scenario (or “Option A”) 12

2.5. Legal basis for EU intervention 13

3. Objectives 14

4. Policy options 15

4.1. Voluntary vs Regulatory Options 15

4.2. Cost drivers affected by the options 15

4.3. Cost estimates 16

4.4. Reference scenario (Option A) 17

5. Analysis of impacts 18

5.1. Social and environmental impact 18

5.2. Scope of impact analysis 18

5.3. Option B 1: publish guidelines for cross-acceptance of existing rolling stock and ask Member States to apply them 18

5.4. Option B2: to assign the Agency the task of classifying national rules and identifying those that can be cross-accepted 21

5.5. Option B3: assign the Agency the role of coordinating parallel acceptance procedures 24

5.6. Option B4: to assign the Agency the role of consultative / appeal body 25

5.7. Option B5: to accelerate the development / revision of the RST TSIs and extend their scope 25

5.8. Option B6: close monitoring of the use of the mutual recognition principle and, where appropriate, the launch of infringement procedures 28

5.9. Option C1: modification of the legislation in order to remove the obligation to authorise rolling stock in each Member State 28

5.10. Option C2: clarify the procedure to be applied for existing rolling stock 29

5.11. Option C3: limit and/or clarify the role of the Member State when authorising rolling stock on its territory 30

5.12. Option C 4: extend the competences of the Agency in order to allow it to deliver authorisations for placing into service which are valid in several Member States 31

5.13. Option C4A: One-stop shop 33

5.14. Option C5: Extend the competences of the Agency to allow it to play a coordination and/or appeal role 34

5.15. Option C6: extend the competences of the Notified Bodies in order to allow them to deliver certificates equivalent to national authorisations 35

5.16. Option C7: extend the competences of the Infrastructure Managers to allow them to deliver certificates equivalent to national authorisations 36

6. Comparing the options 37

7. Monitoring and evaluation 38

7.1. Monitoring prerequisites 38

7.2. Monitoring efforts 38

7.3. Monitoring results 38

Annexes

A Status of transpositions and notifications

B Language survey

C Assessment of options, summary table

D Assessment of options, stakeholders’ views

E Assessment of options by stakeholders, analysis

F One-stop-shop consultation

G Rolling stock fleet with cross-border capabilities

1. PROCEDURAL ISSUES AND CONSULTATION OF INTERESTED PARTIES

1.1. Organisation and timing

End of 2005 Presentation of the results of the Task Force to the Commission

March 2006 Presentation of the results of the Task Force to the Committee and endorsement of these results, including the guidelines produced by the Task Force

April 2006 First meeting of the Impact Assessment Commission Inter-Service Group

27/4/2006 Presentation of a consultation document to stakeholders

01/06/2006 Deadline for submission of position papers by stakeholders, extended to 31/06/06

From August to October 2006:

The Commission carries out the impact assessment exercise with the support of the European Railway Agency (ERA).

1.2. Consultation and expertise

1.2.1. External expertise

External expertise was provided mainly by ERA. Contributions were sought from all major stakeholders, mainly through meetings or by consultation in writing. Main discussions partners were: rolling stock suppliers (Alstom, Bombardier), rolling stock leasers (Mitsui Rail Capital), National Safety Authorities (FR, DE, UK, CZ), sector organisations (UNIFE, CER), the UIC and the Network of Safety Authorities managed by ERA.

1.2.2. Consultation of stakeholders

The first consultation of stakeholders was launched on 27/4/2006.

On 19/10/2006, 14 stakeholders had provided an official response. The responses comprised the following:

Member States | European representative organisations |

Sweden | Notified Bodies (NB Rail) |

Norway | European Rail Freight Association (ERFA) |

Luxembourg | International Union of Private Wagon Owners (UIP) |

UK | Joint Programming Committee (JPC) Rail of the European Committee for Standardisation (CEN) |

Poland | Community of European Railways (CER) |

Netherlands | European Rail Supply Industry (UNIFE) |

European Rail Infrastructure Managers (EIM) |

Deutsche Bahn (DB AG) + VDB |

A workshop was held on 27 April 2006. Almost all organisations and Member States represented at the workshop responded to the consultation. Belgium, the Czech Republic, Germany, France and Slovakia have not yet submitted responses. ERFA submitted its official position on 23 October 2006.

A summary of the responses can be found in Annex E .

An additional consultation was launched by ERA on 15 September 2006 on a further option (“one-stop shop”) previously raised by stakeholders. On 19 October 2006, 11 stakeholders had provided a formal response.

A summary of the responses to this specific consultation can be found in Annex G .

The results of an inquiry concerning the present choice of languages and the volume of documentation exchanged in relation with rolling stock authorisation in the different Member States can be found in Annex B .

1.2.3. Main results

The results of the consultation process confirm the need for action at Community level.

Stakeholder statements reveal the difficulties and uncertainties perceived by the sector in the field of rolling stock authorisation. This would justify more “pedagogic” action to be taken in addition to legal action. The network of national safety authorities set up by ERA would be the ideal platform for such pedagogic action.

2. PROBLEM DEFINITION

2.1. Issue

2.1.1. Definition of cross-acceptance

Cross-acceptance of rolling stock is the process by which a railway vehicle that has received an authorisation to be placed in service in one Member State, following checks against TSIs and/or National Technical Rules (in accordance with Directive 2001/16/EC, Article 16), receives further authorisations in other Member States, taking into due consideration all the checks carried out in the first Member State.

2.1.2. Present situation

At present, cross-acceptance is organised as described below. It should be noted that the Railway Safety Directive explicitly addresses (in its Article 14) the case of in-use rolling stock bearing no EC certificate whatsoever, thus filling the gap in the Interoperability Directives, which only address rolling stock placed in service after the entry into force of these Directives (i.e. new rolling stock).

Firstly, cross-acceptance can be achieved through either harmonisation or the application of the mutual recognition principle.

Under the mutual recognition principle, Member States of destination cannot forbid the sale on their territories of products that are lawfully manufactured or marketed in another Member State and which are not subject to Community harmonisation, even if the products in question were manufactured according to different technical and quality rules than those that must be met for their own products. The only exception to this principle are restrictions imposed by the Member State of destination on the grounds described in Article 30 of the EC Treaty or on the basis of overriding requirements of general public importance as recognised by the Court of Justice’s case law, provided that they are proportionate.

With regard to the application of this principle to rolling stock, a distinction should be made between placing on the market and placing in service. In the latter case, there is a need to ensure that the rolling stock is compatible with the national railway infrastructure.

The mutual recognition principle could be applied to existing rolling stock (not yet covered by the Interoperability Directives), at least for those characteristics not directly linked to specific infrastructures. This is what has been proposed by the Task Force mentioned above and in the Commission proposal for modifying the Railway Safety Directive.

There is therefore a need to ensure legal certainty in the application of this principle to rolling stock and, at the same time, to clarify the process whereby only additional checks linked to the specificity of national infrastructure should be carried out.

Experience demonstrates that the lack of adequate and transparent procedures for cross-acceptance of rolling stock results in delays and costs to railway undertakings and manufacturers when placing rolling stock into service.

Recent bilateral initiatives (e.g. between France and Germany, concerning high-speed rolling stock and then locomotives) have shown that improvements are technically possible, achievable in terms of organisation, and affordable in terms of finance and (above all) expert resources.

2.2. Analysis of shortcomings

The obstacles to rolling stock cross-acceptance are many, and are of a technical, procedural and political nature.

2.2.1. Late or missing notification of national technical rules

The lists of National Technical Rules are still being notified (the deadline was 30/4/2005) under Article 16(3) of the Interoperability Directives. The status of notifications can be found in Annex A . The reasons for late or no notification of national technical rules are being investigated. Preliminary results indicate that:

- Generally, the difficulty in converting expertise-based rolling stock acceptance (as practiced by many traditional operators) into a set of fully documented “national rules”, against which conformity can be assessed by a third party (a Notified Body or other designated body), seems to be an obstacle to the prompt notification of national rules .

- In at least one case, the list has been drawn up by the NSA, but has not been notified by the Ministry for unknown reasons.

2.2.2. Unstable sets of national technical rules

The set of national technical rules (notified or not, published or not) against which rolling stock is checked before being put into service is naturally evolving, and is not “frozen” when the acceptance procedure starts. Railway undertakings and manufacturers are confronted with additional, often completely new requirements in the course of rolling stock development and testing.

2.2.3. Acceptance levels

There are about four levels of rolling stock acceptance in the case of a locomotive:

- Authorisation by the National Safety Authorities (NSA) to place a locomotive into service on a network (Directive 2001/16/EC), with possibly some restrictions;

- Authorisation by the NSA to use the locomotive on specific parts of the network (Directive 2004/49/EC), granted in conjunction with safety certificates;

- Authorisation by the infrastructure manager to run trains (hauled by the locomotive) on given lines, with limits placed on the length, mass, and speed of the train, which may depend on the locomotive series; this is granted at the path contracting stage (as for Directive 2001/14/EC);

- Authorisation at the operational stage, where temporary speed or mass restrictions may still be imposed by IM depending on the actual condition of the infrastructure.

The present report deals only with the first two levels.

2.2.4. Scope of national rules

The Interoperability Directives require the Member States to notify their lists of national technical rules. It does not require the notification of the content of any rule. Some cases of rules that include technical specifications covered by intellectual property rights have been reported.

A detailed study of notified rules would be necessary in order to evaluate the differences in the level of detail (functional requirement vs technical specification) and whether or not they address “essential requirements”.

2.2.5. National safety rules vs national technical rules

It has become apparent that National Safety Authorities (NSA) do not make an adequate distinction between what are national technical rules to be notified under the Interoperability Directives and national safety rules notified under Article 8 of the Railway Safety Directive. There is obviously a clear link between the cross-acceptance requirements in Article 14 of the Railway Safety Directive and the safety certification or authorisation requirements of the Directive, especially in relation to Part B safety certificates.

2.2.6. Rules overlap

National technical rules are not always updated or abolished promptly when a TSI comes into force, since both are supposed to ensure the fulfilment of essential requirements and therefore may result in overlap. This may not be the case.

2.2.7. National rules: conformity assessment

TSI requirements (“basic parameters”) include conformity assessment procedures, while it remains to be checked whether national rules are published together with their respective conformity assessment methods.

Since each Member State designates the bodies in charge of assessing conformity with national rules, no competition comparable to that between Notified Bodies can arise. This situation will be perpetuated if assessment methods are unclear or not agreed.

2.2.8. Maintenance of sets of national rules

Sets of national rules need to be maintained (in the wide sense): technical updates, publication, possibly translation (although not mandated by present EU laws), and abolition when a TSI parameter supersedes the rule. These costs are borne by the publisher, and possibly passed onto users (manufacturers, designated verification bodies). When rules enforce national or international standards, the corresponding costs are borne by industry at large. These are the “fixed costs” related to cross-acceptance.

The availability of expertise may be a problem: with the increased division of responsibilities between Infrastructure Managers (IM), Railway Undertakings (RU), Member States, National Safety Authorities (NSA), Notified Bodies, etc., the availability of system-wide experts may be reduced. This could hamper the revision or elimination of national rules, and more generally slow down the evolution of the railway system (technical aspects as well as rulebooks). Such limitations have been evoked in the past (e.g. Control-Command & Signalling migration). Monitoring of the current notification of national rules might provide some hints concerning the scope of the problem.

2.2.9. Migration period

Since TSIs apply only to new or upgraded subsystems, there will still be non-TSI compliant infrastructure for some significant time. Therefore, national technical rules will remain in use for checking the compatibility of “foreign” rolling stock long after all TSIs have been published.

2.2.10. Insufficient feedback

The monitoring of the correct application of the Interoperability Directives and the TSIs is difficult since the actors (manufacturers, leasers, railway undertakings, keepers) rarely or never resort to arbitration: their main aim is to close the authorisation file as quickly as possible in order to support their business, so tend to avoid litigation. From this we may infer that the perceived cost of missed business opportunities (due to being late in putting rolling stock into service) is generally higher than the experienced cost of “unjustified” additional tests.

2.2.11. Beyond the Interoperability and Safety Directives

There are national regulations that may affect cross-acceptance, for instance labour laws (working conditions), environmental protection laws, etc. Individual cases have been reported, but it is not possible to verify how significantly such additional obstacles impact on interoperability.

2.2.12. The question of language

The nature of the national rules, their format and their language have an important impact on cross-acceptance. See the language survey in Annex B for more detail on the present situation.

Annex B shows that many NSAs require their national language for at least the key documents submitted to them. One NSA ( [5] ) mentioned that dealing with safety cases in the mother tongue of the experts involved was a matter of good practice (the risk induced by the translation of input documents seemed minor in comparison).

An analysis of the Directives as transposed in the different Member States could provide more information on the language issue.

As a result of transposition, the costs and delays for translation might have been shifted from the NSA to the suppliers or the railway undertakings. However, it is likely that NSAs are more flexible in reality than they would admit in the survey.

The discussions about language cannot be separated from those concerning deadlines. For example, the deadline for the NSA to give its opinion on rolling stock acceptance (from the submission of the complete technical file, under Directive 2004/49/EC, Article 14(2)(f)) might be particularly tight, depending on language and “depth” of the checks performed.

2.2.13. Files to be presented to NSA for rolling stock authorisation

The volumes of files exchanged with the different NSAs for the same purpose differ by orders of magnitude : from about 100 to 20 000 pages (with an undetermined amount of drawings or even superfluous information, as one NSA stated).

An examination of the transpositions of the Directives into national laws could help in understanding such differences. This suggests quite different understandings of how tasks and responsibilities are shared between NSAs, Notified Bodies and other bodies.

2.3. Scope of the problem

2.3.1. Powered rolling stock is affected, but not coaches and wagons

In-use wagons and coaches, if able to cross borders, are covered by the RIV and RIC agreements. RIV is no longer maintained by the UIC, but the authorisation for RIV wagons to cross borders has not been withdrawn.

For new or upgraded wagons, the Wagon TSI applies. Its Article 7.6 enforces the principle of mutual recognition:

Once safety certification or authorisation of placing in service is granted for grouped wagons in one Member State, this shall be mutually recognised by all Member States in order to avoid duplication of safety/interoperability checks by Safety Authorities.

For new coaches, RIC registering remains possible (there are no plans to abandon RIC for the time being). For passenger carriages, a similar approach is feasible (once the related TSI is available). Compliance with RIC technical rules might still be recognised as sufficient evidence of conformity with “essential requirements”.

In conclusion, the lack of cross-acceptance concerns powered rolling stock.

2.4. Reference scenario (or “Option A”)

The reference scenario is defined by the application of the Interoperability Directives (96/48/EC, 2001/16/EC, both revised by 2004/50/EC), the Railway Safety Directive (2004/49/EC) and the Agency Regulation (EC/881/2004[6]). It is further defined by bilateral agreements relating to the cross-acceptance of rolling stock, as currently signed by, or negotiated between, National Safety Authorities.

However, this reference scenario is still in evolution, since

1. for the second railway package, national implementations differ or are late — see Annex A describing the current status of notifications and transpositions;

2. most respondents indicated that the reference scenario would not be appropriate in order to improve cross-acceptance of rolling stock.

The current European legislation, in particular the Interoperability Directives (96/48/EC, 2001/16/EC, both revised by 2004/50/EC) and the Railway Safety Directive (2004/49/EC), favours mutual recognition and, consequently, cross-acceptance, in particular through 3 fundamental steps:

- The Interoperability Directives specify that TSIs or, in the absence of TSIs or where infrastructure does not comply with TSI requirements, national technical rules should cover all, and only, the essential requirements relating to the rolling stock subsystem and its interfaces with other subsystems;

- The Interoperability Directives require lists of national technical rules to be notified to the Commission, hence ensuring the necessary transparency of these rules ;

- The Railway Safety Directive requires (among other things) checks to be carried out of safety and technical compatibility (i.e. two essential requirements of the six) when rolling stock is to be used on other networks. It seems that this clause is interpreted broadly by some NSAs, leading to additional checks of debatable justification.

2.5. Legal basis for EU intervention

Where the lack of rolling stock cross-acceptance hinders the interoperability of the European railway system, thus going against not only the Safety and Interoperability Directives but also Article 28 of the Treaty on mutual recognition, EU intervention is justified.

3. OBJECTIVES

Strategic Objectives

The strategic Community objectives in improving the cross-acceptance of railway rolling stock are:

- To complete the internal market, and in particular to ensure the free movement of railway rolling stock within the EU in order to safeguard the commercial interests of manufacturers, railway undertakings and logistics providers, which in turn will revitalise the rail mode of transport;

- To simplify EU and national legislation in line with the “better regulation” objective;

- To lower the costs of doing business and remove unnecessary red tape, both of which are particularly burdensome for SMEs;

- To reduce public expenses by lowering the costs of competent authorities.

All these objectives are a part of the “Lisbon Strategy” to strengthen employment, economic reform and social cohesion as part of a knowledge-based economy.

The strategic objectives can be broken down into specific objectives that will enable the strategic objectives to be met. There are two different aspects of cross-acceptance: one concerns new/upgraded/renewed rolling stock and the other relates to existing rolling stock, both with their own specific objectives:

- For new/upgraded/renewed rolling stock: the Community legislation aims to bring benefits through the development of TSIs and their application. Here, the specific objectives are to reduce the number of national rules imposed by Member States on top of the TSIs and to reduce the time for re-authorisation in each Member State;

- For existing rolling stock: the aim is to set up an EU procedure for the coordinated acceptance of rolling stock, to improve the visibility of national rules, to increase the use of the mutual recognition principle and clarify the issue of existing rolling stock in EU legislation.

4. POLICY OPTIONS

4.1. Voluntary vs Regulatory Options

During the consultation process, various options were presented and discussed:

- 6 voluntary options, numbered B1 to B6, and

- 8 regulatory options, numbered C1 to C7.

It is worth noticing that the voluntary and regulatory options have, to a large extent, overlapping targets, thus providing a genuine alternative.

The options and their possible effects are described in Chapter 5 of the present document.

4.2. Cost drivers affected by the options

Three main cost drivers have been identified in the cross-acceptance process: testing, documentation and technical modifications to the rolling stock. This finding applies primarily to new rolling stock but can be extended to in-use rolling stock, with different proportions for the three cost factors.

4.2.1. Process of integrating national rules into the design of rolling stock

[pic]

The above table shows an example of the the administrative process for a customer order of new rolling stock. The integration of national rules starts long before tendering procedures. Redundant or even contradictory national rules are likely to increase costs significantly.

4.2.2. Testing

Field tests are especially expensive. Worse, additional field tests (mostly on first production units) are likely to add delays to the manufacturing process and to deliveries.

National rules sometimes impose testing on national infrastructure (e.g. track circuits), when experimental conditions are deemed difficult to reproduce.

4.2.3. Documentation

The rolling stock authorisation process generates multiple files: as many as the NSAs, Notified Bodies, and verification bodies involved.

The size of a file is counted, generally, in shelves or cardboard boxes, and may even take up a cabinet.

4.2.4. Technical (design) costs, modifications

Such costs may appear at various stages:

- Before contract execution: difficulty creating “generic” designs suitable for several countries, even where customers are willing to buy “off the shelf”;

- During contract execution: added costs, risk of delays in rolling stock delivery

4.3. Cost estimates

4.3.1. Costs per project and per country

Quantitative information is scarce and sometimes questionable. The key reason is the difficulty in disentangling the costs related to the lack of cross-acceptance from other costs.

UNIFE has suggested acceptance costs of about €8m per country per project (series). This seems to be a “ceiling cost”, as it may include tests related to performance or customer requirements, tests needed for solving “teething” troubles, etc. In fact, since tests (including field tests) always have several purposes, the test cost breakdowns do not allow costs to be allocated to a specific category (performance, acceptance, etc.).

4.3.2. Costs borne by contracting entity

Delays resulting from re-testing result in capital costs. A rule of thumb is: 1 month delay = 1 monthly lease rate for equivalent rolling stock = 1% of purchase cost (this is a very rough estimate to indicate just the order of magnitude).

It is clear, however, that the worst consequence is the loss of business opportunities due to unforeseen delays. Such losses may be greater than the above costs by at least one order of magnitude.

The case of rolling stock leasing companies deserves a mention: since they may not (under current legislation) apply for rolling stock authorisation ( [7] ) , the authorisation question is usually examined after signature of the lease contract, but the railway undertaking is unwilling to risk delayed acceptance and would claim compensation in the event of any delay. Enabling the leasing company to apply for an authorisation would help reduce this risk.

4.3.3. Multiplying factors

The total amount of powered rolling stock allowed to operate in more than one country is shown in Annex H . It accounts for 12%-15% of all rolling stock. This amount is certainly rising, and the renewal of this fleet will place a considerable acceptance burden on NSAs, which they may not be able to cope with satisfactorily.

4.4. Reference scenario (Option A)

4.4.1. Status quo

To continue current policy, which basically consists of the application of Article 14 of the Interoperability Directives for new/upgraded/renewed rolling stock and the application of Article 14 of the Safety Directive for existing rolling stock.

National rules constitute the principal source of problems, without which there would probably be no technical barriers to the cross-acceptance of rolling stock. Member States should eliminate technical barriers that are not justified in accordance with Article 28 EC Treaty.

The Commission should identify (possible) national technical barriers and ensure that unjustified barriers are prevented or eliminated. The competent Member State can either justify the (possible) technical barrier on the basis of Article 30 EC Treaty and demonstrate its proportionality, or eliminate the (possible) technical barrier.

The Commission will monitor the evolution of this problem by examining problems notified by the industry and assessing the national rules notified by Member States under Article 16(3) of the Interoperability Directives, under the Commission Decisions adopting individual TSIs and under Article 8 of the Railway Safety Directive, with the help of technical opinions obtained from the Agency[8].

5. ANALYSIS OF IMPACTS

5.1. Social and environmental impact

The analysis mainly considers the economic impact (on railways, their suppliers and their customers), as improving the cross-acceptance of rolling stock has no direct relevance to social or environmental issues. The different options examined will impact rolling stock authorisation costs and time-to-market in different ways; social or environmental effects will reflect the respective efficiency of each option and amplify their economic consequences, but will most certainly not change their acceptability or ranking.

5.2. Scope of impact analysis

Each option has been examined according to the following criteria (not systematically set out in the present report):

3. Current situation

4. Projected situation, as described in the consultation process

5. Task description and possible task allocation (including administrative aspects, responsibilities, etc.)

6. Legal aspects

7. Transition / timeline / link with other options (including option A)

8. Direct impact: effect on acceptance procedures (costs, delays, barriers)

9. Indirect impact: effect on rolling stock design, development

10. Institutional impact: effect on national rule frameworks

11. Potential impact on European Railway Agency

12. Opportunities and risks

5.3. Option B 1: publish guidelines for cross-acceptance of existing rolling stock and ask Member States to apply them

5.3.1. Current situation

The Task Force set up in 2005 by the Commission has developed guidelines for the cross-acceptance of rolling stock[9]. These guidelines were presented to the Committee set up in accordance with Article 27 of Directive 2004/49/EC in March 2006. They are based on concrete, recent cross-acceptance projects.

5.3.2. Projected situation

5.3.2.1. Guidelines

The Task Force mentioned above recommends the use of a checklist for a common acceptance process. The use of the checklist in the acceptance process should be agreed at the start of the project and managed proactively, which will accelerate the cross-acceptance process and minimise its costs.

There is already experience in using this approach. Accordingly, safety authorities should:

- cooperate in defining the combination of rules to be applied

- optimise the practical tests needed for the approval process (preparing the prototype just once, with all tests to be carried out directly one after the other)

- share the work involved in the process

The main steps in common acceptance are described below.

5.3.2.2. Starting with a joint procedure

For the development and approval of vehicles, manufacturers, railway undertakings and contracting entities should approach the approval bodies from all countries for which they intend to ask approval with a request for a joint procedure, involving them from the start and thereby allowing them to decide on the most efficient path for the approval process.

5.3.2.3. Setting up the common checklist

The Task Force strongly recommends the use of the common checklist by all Member States for cross-acceptance projects. The structure of the common checklist should not be amended. The common checklist should contain all relevant rules of each Member State.

The approval requirements according to the common checklist are divided into the following groups:

Group A : contains internationally accepted standards that, once checked by any member state, do not need further checking for cross-acceptance.

Group B : contains requirements that are currently used in specific countries and that

- might be fit for cross acceptance

- might need further detailed discussion before being moved to A or C, now or later, in general or for a specific country.

- are not undisputable but are undeniably essential and necessary requirements linked to the technical characteristics of the infrastructure for safe and interoperable operation in the country in question.

A vehicle could still run if it does not fulfil the B requirements, just as it is already running elsewhere, if the applicant can prove that a standard equivalent to the relevant national requirements is ensured.

Group C : contains undisputable and undeniably essential and necessary requirements linked to the technical characteristics of the infrastructure of a specific country or network, which always need checking, e.g. loading gauge.

5.3.2.4. Dealing with the results of the checks against the common checklist

The authority that handles "first/type approval" supplies, together with this approval, a “result list” of the Group-A reference documents that a vehicle (type) has been checked against. Where appropriate, it mentions the level of the check or the resulting measurement. It also mentions derogations from this document (e.g. within gauge G2, but cabin steps exceed gauge by 3 cm) and if thereby Group A requirements are met for a specific parameter.

The authorities format the “result list” according to the checklist accompanying this procedure and as an official document (appropriately drafted and signed etc.). It may be a declaration of an EU or national notified body, countersigned by the authority. Authorities of other countries where approval of the vehicle is sought accept the "result list" as full proof of compliance with the Group-A documents referred to.

Similarly, the "result list" mentions — where practicable — whether and how Group B and C items were checked, and which documents these checks refer to (sometimes results can be useful for other authorities; sometimes there might be conflict between requirements or required outcomes.)

For the Group C items, a check according to national practice is necessary, as specified by the national authority. Where possible, use is made of earlier tests and checks.

For the Group B items, a check according to national practice and against national requirements may be necessary. The applicant may deliver arguments that he can provide an alternative solution. The national authority has to provide adequate arguments as to why it cannot accept a positive verification against a different, foreign standard.

The authorities should work together bilaterally to update the A/B/C grouping and to reduce the number of Group B items.

5.3.3. Task description and possible task allocation

The guidelines of the Task Force have already been published by the Commission. They can thus be applied immediately by Member States.

5.3.4. Legal aspects

Under this option, no EU legislation change is envisaged to enforce the application of the guidelines. However, it is possible that Member States may not be allowed to apply these guidelines due to their national rules. This is why the process recommended by the guidelines should ideally be incorporated in EU legislation.

5.3.5. Direct impact: effect on acceptance procedures (costs, delays, barriers)

It has been shown that using the guidelines could reduce the total time needed for the acceptance procedure to about one quarter. The savings in time and money for the applicant are significant.

5.3.6. Potential impact on the European Railway Agency

For reasons of transparency, DB AG and VDB have suggested, in their common position paper, that the result list should be published centrally at European level, e.g. by ERA. This is covered by Option B2.

5.3.7. Opportunities and risks

The benefits of the cooperative relationships envisaged are limited to the Member States involved and do not provide an opportunity to share the best-practice lessons learnt. In addition, there is a risk that the proliferation of such "variable geometry" agreements will not open the market, but contribute to further fragmentation and complication. This partly justifies having this option complemented with the following option B2.

5.4. Option B2: to assign the Agency the task of classifying national rules and identifying those that can be cross-accepted

5.4.1. Current situation

For national safety rules, the legal basis for notification to the Commission and evaluation by the Agency is already there.

For national technical rules in the sense of the Interoperability Directives, there is currently no corresponding legal base.

5.4.2. Projected situation

The Agency should be requested to develop, on the basis of the checklist mentioned under option B1, a common EU table of national rules notified by Member States under Article 16(3) of the Interoperability Directives, under the Commission Decisions adopting individual TSIs and under Article 8 of the Railway Safety Directive.

In addition, the Agency should assess these rules and identify those that can be cross- accepted, and should work towards reducing Group B of the table so that all acceptance requirements are either Group A (mutually recognised by all parties) or Group C (which relates to national acceptance against specific national infrastructure requirements)

The purpose is similar to B1, as far as the classification of national rules is concerned, except that the sorting of rules is conducted centrally (Agency mandate) rather than decentrally. NSAs must of course remain involved, if only because they remain responsible for establishing the reference sets of national rules.

5.4.3. Task description and possible task allocation

This option contains 2 sub-tasks: classification of rules / comparative assessment of rules.

The first task consists, for each Member State, in classifying the national rules according to scope (e.g. by subsystem or interfaces between subsystems) and parameter (e.g. by vehicle clearance gauge, electromagnetic compatibility, etc. — see the guidelines checklist), to clarify what rule should be applied in which case.

It is up to NSAs and Member States to define the range of application (rules should generally define their own scope), but the Agency could be given the job of referencing them within an EU-wide reference table.

The second task consists in carrying out a comparative assessment of the rules, for each technical parameter in the checklist.

One main difficulty is that equivalence of subsets of national rules may not be “transitive” in the mathematical sense (mutual recognition between states X and Y on the one hand, and Y and Z on the other hand, does not automatically result in mutual recognition between X and Z without any further checks). However, the experience with the extension of the FR/DE checklist is encouraging, because these extra checks remain limited once the first mutual recognition agreement has been reached.

For carrying out this task, the translation of some national rules will be necessary

The comparative assessment may also lead to the closure of current open points in existing TSIs. The revision of such TSIs should therefore be mandated to the Agency (no legislative change required).

5.4.4. Legal aspects

The process of notifying national technical rules (in accordance with 2001/16/EC) needs to be clarified.

The identification of cross-acceptable sets of rules by the Agency will necessitate further legislative action at EU level to ensure legal certainty as far as the reference table is concerned.

5.4.5. Transition / timeline / link with other options (including option A)

Checks against Group A requirements (mutually recognised technical requirements) must have official status to provide any practical value to this option.

For conventional rolling stock, it should be possible to address the case of locomotives as a first step. The checklist recommended by the Task Force can be used immediately.

5.4.6. Direct impact: effect on acceptance procedures (costs, delays, barriers)

A somewhat extreme example of benefits has recently been seen by the French NSA in the acceptance of class 66 diesel locomotives for EWS in France: some of the checks made for the acceptance of a similar locomotive (JT42) run by other operators in Germany could be re-used. In such cases, only a “formal” check is performed (existence of verification certificates, etc.) by the French NSA.

5.4.7. Indirect impact: effect on rolling stock design, development

The availability of correspondence and equivalence tables between national rules should greatly facilitate the work of design offices. It may be noted that rolling stock design offices currently employ permanent staff for dealing with national rules (along with standards and TSIs): no details have been provided about manning, but the diversity of languages used for national rules suggests that not a few people are employed, at least part-time, in this area. These staff needs might then be reduced when some of the national rules are either incorporated in TSIs, in harmonised form, or eliminated altogether.

5.4.8. Institutional impact: effect on national rule frameworks

This option may provide an opportunity to remove those technical rules not relating to essential requirements. It may also promote “competition” between equivalent sets of national rules, provoking the “obsolescence” of some of these and thereby bringing about simplification in railway regulations.

5.4.9. Potential impact on the European Railway Agency

The Agency cannot perform this task with its current staff. A project team has to be established, consisting of at least:

- A group of technicians (1 mechanical engineer, 1 electrical engineer, 1 safety analyst), and

- 1 specialist in certification processes, with an operations-oriented background.

Synergy needs to be established with the staff already involved in the classification and evaluation of national safety rules notified under the Railway Safety Directive.

The cost of setting up a central repository for national technical rules (which are far from being static data) can be estimated in the range of €0.2–0.8m. The initial input of the most important national technical rules (in electronic form) has been estimated at €100 000 per Member State, including translation but not verification (which should be performed by the NSAs of the corresponding countries).

5.4.10. Opportunities and risks

Synergy is possible between the development of the TSI on locomotives (already mandated) and the evaluation and classification of national rules (notified under both the Interoperability and Safety Directives). For example, UNIFE has proposed the use of the high-speed rolling stock TSI requirements as a “benchmark” against which current national rules could be examined, until the corresponding TSIs for conventional rail become available ( [10] ) . This is partly justified by the non-segregation of high-speed vs conventional rail systems in Europe, leading to many common requirements. This approach is seen by UNIFE as the most effective. It is also consistent with the way the conventional TSIs are derived.

5.5. Option B3: assign the Agency the role of coordinating parallel acceptance procedures

5.5.1. Current situation

Current legislation does not explicitly mention such a role for the Agency.

5.5.2. Projected situation

Within the Network of Safety Authorities, the Agency can act as a coordinator between parallel acceptance projects. It can also deliver Technical Opinions to improve subsequent acceptance procedures.

In this option, the Agency is asked only to monitor parallel acceptance procedures, since no active coordination role can be envisaged with the current legal framework. This role should be passive but not mute, in the sense that the Agency is expected to provide guidance, if needed.

5.5.3. Legal aspects

There is currently no clear legal base for the Agency to intervene in cross-acceptance projects.

5.5.4. Transition / timeline / link with other options (including option A)

This option could be considered as an adjunct to option B2.

5.5.5. Direct impact: effect on acceptance procedures (costs, delays, barriers)

“Monitoring” implies no direct effect. However, this option could provide the necessary “visibility” to identify, at least, gross infringements of current legislation on rolling stock authorisation.

5.5.6. Potential impact on the European Railway Agency

Any monitoring presupposes some amount of data transmission and handling, plus analysis and reporting. Languages may become an issue in terms of cost and delay. This option does not seem appealing if taken in isolation. However, it could very well have significant synergy effects in conjunction with option B2, if the latter option is selected.

5.6. Option B4: to assign the Agency the role of consultative / appeal body

5.6.1. Current situation

Current regulation already provides for the Agency to play an advisory role at the request of the Commission

5.6.2. Projected situation

Option B4 implies a consultative and appeal role in the case of concrete cross-acceptance projects. This should be taken into account, including the budget implications: it is likely that financial contributions would be required to cover costs, at least if deadlines were associated with the demands of this role (as in option C5).

5.6.3. Task description and possible task allocation

This task could be entrusted to the team of the Agency in charge of option B2, if the latter is selected.

5.6.4. Legal aspects

Appeal: currently, the Agency is not an appeal body, so this is definitely a new task that would require modification of the Agency Regulation. This would fit with option C5.

5.6.5. Transition / timeline / link with other options (including option A)

This option is ancillary to B2.

5.6.6. Direct impact: effect on acceptance procedures (costs, delays, barriers)

The key benefit would be “centralised” expertise; furthermore, case studies would improve understanding and boost the implementation of option B2. However, the impact on acceptance procedures themselves is expected to be moderate, as only an advisory role is envisaged.

5.7. Option B5: to accelerate the development / revision of the RST TSIs and extend their scope

5.7.1. Current situation

All “missing” TSIs concerning the various structural subsystems of railways have now been mandated. Recent TSIs have been published with open points.

5.7.2. Projected situation

The projected situation comprises at least three elements: acceleration of the development (drafting) of TSIs, acceleration of the revision of TSIs, and scope extension.

Acceleration of TSI development:

The experience with developing the first generation of High-Speed TSIs and the first group of Conventional Rail TSIs (control-command and signalling, noise, wagons, operation, telematics applications for freight) shows that the typical time frame for the development of a TSI ready for adoption is three years. There seems to be no easy way to significantly accelerate their development. Moreover, there is a trade-off between “fast development” and “scope extension”.

More importantly, accelerating the development of TSIs might not be very helpful in the short term, since the experience gained with high-speed subsystems shows that the verification of TSI conformity represents typically 1/3 of the verification workload, while conformity to national rules may represent 2/3. This suggests that the publication (option A), classification (option B1) and analysis of national rules (option B2) are the most urgent tasks.

Acceleration of TSI revision:

This seems a natural complement to option B2, and should be bundled with it. The key issue is the existence of open points ( [11] ) . Another criterion for sorting priorities is the feedback from Notified Bodies and NSA experience.

A short survey of the remaining “B” or “C” requirements in the FR/DE cross-acceptance table for locomotives suggests that 3 fields should have priority:

- Operations TSI (many “small” topics concern operations);

- Catenary / pantograph interface and crosswinds;

- Electromagnetic compatibility.

Scope extension of TSIs( [12] ):

The scope of the TSIs could be extended to include all of the functional areas identified by the Task Force in their common checklist such as, for example: Joining techniques, Control systems (software), Pressure-discharge (freight wagons).

This extension seems, at first glance, in contradiction with the “New Approach”. The technical aspects mentioned are covered by European or national standards. Under the current circumstances, referring to a standard in a TSI leads to the mandatory application of this standard, whereas standards are usually written to allow presumption of conformity to essential requirements, among other things. This has generally led to the restriction of the scope of TSIs to those features absolutely needed for technical compatibility of TSI-compliant rolling stock with TSI-compliant infrastructure. However, if a NSA imposes a national rule on a given technical aspect, a decision must be taken on whether to include this aspect in the TSI or to ask the Member State to withdraw the relevant rule (through option B1 or an infringment procedure).

In order to avoid such over-regulation practices, appropriate informative documents, such as, for example, the Guide published by the Commission in 2003 for the application of the High-Speed TSIs under Council Directive 96/48, should be published for each TSI. In this guide, a list of harmonised standards allowing presumption of conformity with the TSI requirements was given. .

To summarise the above:

- Acceleration of development seems no real option;

- Acceleration of revision is part of option A and is recommended.

- Scope extension is, arguably, part of option A. It will nevertheless be considered as a B option, since there are differences in opinion that need to be debated.

5.7.3. Task description and possible task allocation

Operations experts are being recruited by the Agency. Catenary / pantograph interaction should be dealt with by the Agency’s current Working Parties, and will probably require further involvement of standardisation bodies. The same applies to “electromagnetic compatibility”.

In addition, the development and publication of application documents for TSIs should be prepared in parallel, probably by the Agency.

5.7.4. Transition / timeline / link with other options (including option A)

Task linked to B2, for the reasons explained above.

5.7.5. Direct impact: effect on acceptance procedures (costs, delays, barriers)

Experience has shown that where TSIs have been developed, competent authorities can assess more quickly if a rule can be mutually recognised. Accelerating TSI development would also reduce the period of cross-acceptance difficulties.

With some national rules migrating into the TSI “core text”, positive effects would be simplified acceptance of international rolling stock and some reduction in the costs of verification (if national rules are withdrawn in a timely manner).

5.7.6. Indirect impact: effect on rolling stock design, development

Possible further harmonisation.

5.7.7. Institutional impact: effect on national rule frameworks / EU

(Accelerated) removal of existing rules possible.

5.7.8. Potential impact on the European Railway Agency (derived from above items)

Probably increased workload on TSI working parties + network of National Safety Authorities, hence on ERA (management, coordination, etc.).

The re-opening of the OPE TSI is part of the reference scenario (mandate in preparation). The recruitment of suitable experts is ongoing. There is no differential cost impact.

Concerning the catenary / pantograph interface, the current mandates and budget (TSIs for traction units and energy subsystems, for conventional rail) should suffice.

The remaining issue is electromagnetic compatibility, for which recruitment of a high-level expert seems necessary.

5.7.9. Opportunities and risks

The main risk is excessive weight of the technical specifications that become legally binding if included in a TSI, with a detrimental effect on innovation. To be handled by ERA under its normal working procedures.

5.8. Option B6: close monitoring of the use of the mutual recognition principle and, where appropriate, the launch of infringement procedures

5.8.1. Current situation

Abiding by Articles 28 and 30 of the Treaty is not an option but an obligation. The legal framework for launching and handling infringement procedures already exists.

However, as previously explained, the mutual recognition principle has never really been applied in the railway field.

5.8.2. Projected situation

Member States and NSA representatives should be properly trained in the use of the mutual recognition principle. The Commission and the Agency should closely follow up the development of the common EU cross-acceptance table and the reduction of its Group B part.

This option can be seen as a complement to option B2.

5.9. Option C1: modification of the legislation in order to remove the obligation to authorise rolling stock in each Member State

Article 14 of the Interoperability Directives establishes the principle of national authorisation for placing new rolling stock in service. With the entry into force of the Railway Safety Directive, however, the necessity of such an obligation may be questioned because any RU wanting to operate in a country has to obtain a safety certificate, in which the acceptance of rolling stock is verified. It would therefore be possible to remove the national authorisation obligation by:

- keeping in the Interoperability Directives the obligation to be authorised in the first Member State only;

- modifying Article 14 of the Safety Directive in order to limit the scope for verifications carried out by other Member States, for example by adding an exhaustive list of the verifications that are allowed (specific cases identified in a TSI, etc.)

Such an option would help clarify the procedure to be followed in the case of new or renewed/upgraded rolling stock, which has to comply (only) with the Interoperability Directives.

For C1, CER proposes a “type approval” for rolling stock. Provisions for “type approval” have already been included in the wagon TSI, Chapter 6. It is expected that the other rolling stock TSIs will contain similar provisions. Existing passenger coaches with RIC agreement should benefit from similar clauses in the future. The remaining problem would be existing tractive rolling stock (or any that is in the planning phase before publication of the TSI).

This option is not separable from C2, since the rationale for delivering safety certificates (part B) is independent of the rolling stock being new or already in use.

5.10. Option C2: clarify the procedure to be applied for existing rolling stock

5.10.1. Current situation

Article 14 of the Safety Directive is applicable. However, the Commission Task Force has highlighted some difficulties in interpretation.

5.10.2. Projected situation

The Interoperability Directives apply only to new/upgraded/renewed rolling stock. The scope of these Directives should be extended to existing rolling stock, and the procedure to be applied for existing rolling stock should be clarified.

To clarify the procedure, two options are possible:

- use of the Task Force recommendations outlined above (option B1);

- use of a more radical approach by immediately applying the mutual recognition principle: the authorisation granted by one Member State for placing into service is accepted as a basis for certification in the second Member State. Only a very small list of aspects strictly limited to the infrastructure of the second Member State can be checked by the latter’s NSA before granting the new authorisation. The rules applicable to this list of aspects should be clearly identified in the registers of infrastructure kept by Member States or by the ERA.

5.10.3. Legal aspects

The status of in-use rolling stock should be clarified. In-use rolling stock introduced before the Interoperability Directives benefits from de facto authorisation that should be formally recognised as such in order to facilitate its re-use on other networks (the “second hand” market may be of importance to the new Member States).

5.10.4. Direct impact: effect on acceptance procedures (costs, delays, barriers)

Key points seem to be:

- authorisation procedures are open to more applicants, and

- the two parts of the authorisation would get legal status (the network-independent part and the network-dependent part). This clarification will potentially be very effective in facilitating re-usability of the network-independent part, which is now clearly identified.

5.10.5. Indirect impact: effect on rolling stock design, development

Not relevant, since C2 essentially addresses in-use rolling stock.

5.10.6. Institutional impact: effect on national rule frameworks

Not relevant.

5.10.7. Potential impact on the European Railway Agency (derived from previous items)

The Agency would have to provide technical opinions in the event of disputes, at the request of the applicant. It is proposed to keep the proposed ERA cross-acceptance team (see B2) to provide such advice, which will benefit from the experience of initially setting up cross-acceptance tables.

5.10.8. Opportunities and risks

The main risk in rewriting Article 14 of both Directives is the possible impact on the implementation of the current version of the Directives, for which the transposition date was 1 May 2006. As long as the negotiations on the text of the changes are going on, Member States might be tempted to apply their current rules and hold back the changes in legislation necessitated by the Directives.

5.11. Option C3: limit and/or clarify the role of the Member State when authorising rolling stock on its territory

5.11.1. Current situation

As for previous option C1.

5.11.2. Projected situation

Once the technical rules are harmonised, thus when all TSIs are developed, there should be very little scope for imposing additional verifications. In this option, the authorisations remain national, but Article 14 of both the Railway Safety and the Interoperability Directives should be amended to limit and/or clarify the role of the Member State when authorising rolling stock on its territory.

In fact, this option is complementary to option C2.

5.11.3. Opportunities and risks

This option raises the question whether existing agreements such as RIC (for passenger coaches) will be maintained as far as the cross-acceptance rules they contain are concerned ( [13] ) . This is certainly the key economic problem. The current assumption is yes.

A separate problem is the definition of a similar agreement for locomotives or other tractive units, based on the common reference cross-acceptance table resulting from options B1 and B2.

Of course, when no authorisation is required, all the further options below lose relevance where existing rolling stock is concerned.

5.12. Option C 4: extend the competences of the Agency in order to allow it to deliver authorisations for placing into service which are valid in several Member States

5.12.1. Current situation

The NSAs were created following the adoption of the second railway package for the purpose of delivering authorisations in their own Member States.

5.12.2. Projected situation

Within the aviation sector, the European Aviation Safety Agency (EASA) is responsible for establishing European-wide certification standards and for carrying out the certification of aeronautical products, parts and appliances that are manufactured, maintained or used by organisations within the EU. The certificates granted by EASA are valid throughout the EU. In carrying out its role, the EASA is supported by national authorities until it is fully capable of performing its tasks with its own resources. Further, the EASA regulations also establish a Board of Appeal to deal with appeals against decisions by the EASA. The Board is an independent body and the EASA is bound to comply with its decisions.

The ERA could carry out cross-acceptance of rolling stock, with the support of the national safety authorities. In this case, the ERA (or the Community) should also establish an independent board of appeal to deal with appeals against decisions by the ERA.

Advantages: national rules and infrastructure compatibility requirements will be visible at European level, facilitating the process of developing the complete specifications for the TSI. The process of conformity assessment may also be optimised by removing the need for duplicate and onerous documentation, tests and trials, hence making it more cost-effective. The experience and knowledge gained of the process and rules remain “in-house” and retained at European level.

Disadvantages: the involvement of the ERA may be considered intrusive by the Member States. To have a real impact, the ERA must be able to act as a technically competent body and hence its resources and knowledge in the area of rolling stock and compatibility must be strengthened and increased. The increased resources will only be required during the period when the TSI specifications are being completed. Also, national rules are still emerging and hence the time and cost of cross-acceptance may not be reduced.

However, using the EASA as a model is questionable since air transport infrastructure is less country-specific than railway infrastructure.

An alternative would be to keep the existing bodies (NSAs), and add a “one-stop shop” working scheme: see option C 4A.

5.12.3. Task description and possible task allocation

The diversity of national rules and national infrastructures suggests that, at least in a transitional phase, some expertise may remain in the NSAs while responsibility is transferred to the Agency.

5.12.4. Legal aspects

To implement this option, the Railway Safety Directive will require modification. In particular:

- Article 14: modified to assign responsibility for the authorisation of in-use rolling stock to the ERA, supported by the national safety authority

- Article 16 : modified to amend the tasks of the national safety authority

- New article: setting up an independent appeals board

Legal power to coordinate national rules should be given to the Agency, beyond the current mandates (especially national technical rules).

The designation of Notified Bodies and other verification bodies could be affected.

The problem of liabilities should also be discussed in the event of improper “certification” leading to an accident.

5.12.5. Transition / timeline / link with other options (including option A)

The coordination of national (technical) rules is a “predecessor task”, described under option B 2.

Coordination of national rules may still be the key “investment”.

5.12.6. Potential impact on the European Railway Agency

If this option is adopted, it would have major resource implications for ERA, opening up the prospect of requiring a special central unit within ERA and the creation of a network of satellite offices in every Member State with a railway. Opportunities and risks

This option may have a significant impact on the NSAs recently established following the adoption of the second railway package; it may even jeopardise transposition in those Member States that are late and facing an infringement procedure.

It would also have an impact on ERA’s relationship with NSAs, as NSAs could view ERA’s activities in this area with suspicion and this could adversely affect productive working relationships in other activities of ERA.

5.13. Option C4A: One-stop shop

5.13.1. Projected situation

The creation of a one-stop shop for delivering authorisations was mooted in the course of the seminar on 27 April 2006. A one-stop shop does not mean a unique shop, however. Following the model of the TAF[14] TSI concerning the relationships between loaders and railway undertakings, one NSA (e.g. the NSA of the country hosting the awarding entity) could act as the “lead NSA”, and take care of the relationships with others. The lead NSA would then be in an optimum position to coordinate all demands from all NSAs, waiving duplicate checks, etc., something the awarding entity can never do ( [15] ) . Consistency between the various NSA approaches would be ensured; ERA could still play an important auxiliary role (through the network of NSAs it moderates).

5.13.2. Opinions in the sector

Opinions have been formally expressed by: CER, UNIFE, UPI, UIRR, four NSAs (Belgium, Czech Republic, France, Norway) an d one Notified Body (NedTrain).

This option has prompted contrasting reactions, which are summarised in Annex G.

5.13.3. Conclusion

It appears that this option should be rejected for the following main reasons:

- While a “one-stop shop” is a reassuring analogy with other well-founded or successful initiatives, it is not likely to be effective in the short or medium term due to the many prerequisites for making it work ;

- The one-stop shop scheme does not attempt to cure the main problems (heterogeneous national rules and acceptance processes), but only some of its symptoms; attention should not be distracted from the main problems;

- The responsibility of railway undertakings for operational safety makes direct contacts with all the NSAs concerned a necessity; the one-stop shop scheme does not fit here.

5.14. Option C5: Extend the competences of the Agency to allow it to play a coordination and/or appeal role

5.14.1. Current situation

This extension of competences currently has no legal base.

5.14.2. Projected situation

The cross-acceptance of rolling stock is carried out by national safety authorities, as described in Article 14 of the Safety Directive. The ERA may act as an arbitrator and appeals body in decisions and delays during the acceptance process.

Advantage : the existence of a European appeals body will discourage national delays in cross-acceptance. During the investigation of appeals, the ERA also gains insight into cross-acceptance problems at national level, facilitating the TSI development process and without having to extensively increase its technical competency and resources.

Disadvantages : the resolution of an appeal potentially increases delays and the cost of cross-acceptance especially if an appeal is raised at a late stage during the cross-acceptance process. The Railway Safety Directive will also have to be modified.

Another possibility is to use the Agency as a coordinator or one-stop-shop for the authorisation of existing rolling stock, while leaving to the NSAs the concrete verifications, but filtering excessive demands.

5.14.3. Task description and possible task allocation; legal aspects

The nature of appeal s should be clarified.

The first instance could be understood as the NSA in charge of delivering the authorisation: would this mean that neither the Member States nor the Commission would intervene?

5.14.4. Transition / timeline / link with other options (including option A)

Assuming that the Agency will implement the B options, the present option could be considered as a follow-up to B2, since a thorough knowledge and continued updating of the databases containing the national technical rules (and safety rules) are a prerequisite for taking sensible decisions.

5.14.5. Direct impact: effect on acceptance procedures (costs, delays, barriers)

Low (the litigation stage is rarely reached, and then only when the damage is already done).

5.14.6. Potential impact on the European Railway Agency (derived from previous items)

The aim could be to keep the team in charge of national rules, but not to expand it.

5.14.7. Opportunities and risks

The main risk is the impossibility for ERA to manage the corresponding workload, especially if response times are short (as they should be, as far as reasonably practicable).

5.15. Option C6: extend the competences of the Notified Bodies in order to allow them to deliver certificates equivalent to national authorisations

5.15.1. Current situation

Competences are currently clearly separated (see Annex VII of the Interoperability Directives).

5.15.2. Projected situation

The role of Notified Bodies is presently limited to “EC” verification procedures on the basis of TSIs. This role could be easily extended to the verification of compliance with national rules, provided that the criteria of competence and independence are met. In this option, Notified Bodies would be allowed to deliver “complementary national certificates” (CNCs), and the role of the NSA would then be limited to checking the presence of these CNCs.

5.15.3. Task description and possible task allocation

The Notified Bodies will receive a new task, though after notification of the new competences by Member States as provided for in Article 20 of the Interoperability Directives.

5.15.4. Transition / timeline / link with other options (including option A)

As usual, the full publication of national rules remains a precondition.

5.15.5. Legal aspects

Option C 6 requires changes to some provisions of Directives 2004/49/EC and 2001/16/EC.

5.15.6. Direct impact: effect on acceptance procedures (costs, delays, barriers)

Possible reduction in the number of Notified Bodies for a given project.

5.15.7. Opportunities and risks

The rather negative reaction of some stakeholders seems to indicate that they have not fully understood the option and thought that the Notified Body would replace the NSA in delivering authorisations.

5.16. Option C7: extend the competences of the Infrastructure Managers to allow them to deliver certificates equivalent to national authorisations

5.16.1. Current situation

This might already be current practice where there is no NSA yet or where the NSA relies on checks done by the Infrastructure Manager.

5.16.2. Projected situation

This option is similar to the previous one, but with Infrastructure Managers able to deliver Complementary National Certificates if the criteria of competence and independence are met.

5.16.3. Legal aspects

This option seems rather unproblematic when the infrastructure manager is truly independent from any railway undertaking. The problem is that railway undertakings and infrastructure managers always have at least a commercial relationship, and in some cases have an institutional relationship as well (belonging to the same holding company, or through delegations of competences).

However, choosing this option would have an impact on the arrangements for dealing with shared risk as set out in the Railway Safety Directive as well as on the safety certification and safety authorisation requirements. It would move the current emphasis in the Safety Directive on the need for all players to cooperate to deal with shared risk to one where the opinion of the infrastructure manager would be dominant. It also raises the question of how to assess the competence and the independence of the infrastructure manager to act in this way.

5.16.4. Transition / timeline / link with other options (incl. A-option)

Option C2 is a prerequisite (separation of authorisations into a network-independent part and a network-dependent part).

5.16.5. Direct impact: effect on acceptance procedures (costs, delays, barriers)

When RUs are confronted with complex legislation and unclear requirements, how are such shifts in responsibilities supposed to bring any improvement?

Entrusting infrastructure managers with only the infrastructure-relevant part of authorisations may also increase the number of interfaces that applicants have to deal with, at least when rolling stock is first placed in service. If option C2 is implemented for applications in further networks as well, the applicant would have to interface first with the IM then with the NSA, so there is no change in the number of files to be submitted. On the other hand, the NSA would get rid of a difficult task. See opportunities and risks below.

5.16.6. Opportunities and risks

Potential conflict of interest. The general tendency in Europe is to rely on third-party assessments whenever quality or safety is at stake.

6. COMPARING THE OPTIONS

A summary impact assessment table is provided under Annex D .

Based on the essentially qualitative impact assessment set out above and on the reactions of stakeholders, it is proposed to profit from the most promising aspects of several options in terms of best cost/benefit ratio, in particular options B1, B2, C2 and C3.

What is thus proposed is the immediate application of the Task Force recommendations (use of the cross-acceptance guidelines), together with modification of the legislation in force to:

- Apply the principle of mutual recognition to authorisations already delivered in one Member State.

- Limit the scope for possible additional requests by other Member States.

- Ask the Agency to develop a cross-acceptance reference table, compare national rules for each basic parameter and identify rules that can be cross-accepted (with, as a prerequisite, full notification of rules by Member States).

- Give power to the Commission, through the comitology procedure, to adopt the cross-acceptance table as a reference as well as decisions on the equivalence of individual rules.

- Allow the Agency to deliver Technical Opinions when difficulties arise during concrete acceptance processes.

- Simplify the procedure for wagons and passenger cars.

- Add the possibility to group authorisations together (“type” authorisations).

With regard to the voluntary options B1 and B2, this proposal is broadly in line with the majority of opinions expressed by the stakeholders that responded to the Commission’s consultation. The reason for bundling the options is that the resources used for their implementation could also provide some of the benefits of other B options (B3 to B6) without additional resources.

Among the C options (regulatory options), only C2 and C3 are likely to bring significant benefits. Expected savings[16], concerning only certification costs, have been estimated between 200 and 280 M€ for railway undertakings and manufacturers in the next 15 years in the EU. This is based on the assumptions that 35 different types of locomotives are likely to be certified in at least 10 different Member States. Other positive effects are expected such as reduction of time to market for new locomotive types and new railway undertakings, as well as benefits due to reduced pollution and increased road safety as a result of the possible shift from road to rail.

Option C5 (Agency delivering technical opinions in cases of dispute) is not overwhelmingly effective, but might be considered as a follow-up to B1+B2, and should be bundled with maintenance of the database of national technical rules, which will continue evolving.

The notification of national technical rules is a prerequisite for all subsequent work. The ongoing publication of TSIs anyway requires the continuous revision of notified national technical rules.

7. MONITORING AND EVALUATION

Monitoring should check the prerequisites, efforts and results.

7.1. Monitoring prerequisites

S tatus of transpositions and notifications.

7.2. Monitoring efforts

Status of bi- or multilateral agreements, by countries and type of rolling stock (effects of option B2), distinguishing between drafting (Agency responsibility) and signature (Member State responsibility).

7.3. Monitoring results

D irect measurement of the improvement in cross-acceptance is difficult, but two indicators could be envisaged:

- Satisfaction index: a qualitative, yearly survey encompassing all stakeholders (railway undertakings, manufacturers, ROSCOs[17], NSAs, Notified Bodies, etc.).

- Technical indicator: ongoing monitoring of information sent to the Commission under Article 19(2) of the Interoperability Directives concerning “any additional checks” required. Of course, it would be necessary to verify that this clause is properly enforced. A similar indicator could be developed for in-use rolling stock.

ANNEX A

The following table provides an overview of the transposition of Directives 2004/49/EC (safety) and 2004/50/EC (interoperability) as well as information on the notification of national safety rules (Article 8(2) of the Railway Safety Directive) and national technical rules (under Article 16(3) of Directive 2001/16/EC as amended by Article 2 of Directive 2004/50/EC). This overview was last updated on 25/08/2006.

2004/49 | 2004/50 | Art. 8 | Art. 16 (3) |

BE | N | N | Y | N |

CZ | Y | Y | Y | Y |

DK | Y | Y | Y | N |

DE | N | N | Y | N |

EE | N | Y | Y | N |

EL | N | N | Y | N |

ES | N | Y | Y | N |

FR | Y | Y | Y | N |

IE | Y | Y | Y | N |

IT | N | N | Y | N |

CY | - | - | - | - |

LV | Y | Y | Y | N |

LT | Y | Y | Y | Y |

LU | N | N | Y | Y |

HU | Y | Y | Y | N |

MT | - | - | - | - |

NL | N | N | Y | Y |

AT | N | N | Y | N |

PL | Y | Y | Y | Y |

PT | N | N | Y | N |

SI | N | N | Y | N |

SK | Y | Y | Y | N |

FI | Y | Y | Y | Y |

SE | N | N | Y | Y |

UK | Y | Y | Y | Y |

NO | N | N | Y | N |

ANNEX B Questionnaire sent in 9/2006 to all NSAs (EU25) |

1 – Legislation |

1a - In which language(s) are the national technical rules or national safety rules available? |

1b – do these national rules refer to other national documents (e.g. national standards)? |

1c – if yes, in which languages are these documents available? |

2 – Documents provided by the National Safety Authority to the Railway Undertaking |

In which language(s) do you send documents (other than the above) to the railway undertaking? |

3 – Documents provided by the Railway Undertaking to the National Safety Authority |

Under Directive 2004/49/EC, Article 12(3), “All applications for safety certificates shall be submitted in the language required by the safety authority”. Which language(s) do you require or accept? |

4 – Approximate volume of documents |

Responses as of 20/10/2006, summarised: see next page.

Country | 1 – Legislation | 2 – Documents provided by the National Safety Authority to the Railway Undertaking | 3 – Documents provided by the Railway Undertaking to the National Safety Authority | 4 – Approximate volume of documents |

LV | LV | yes | LV | LV | LV | (not mentioned) | In-use RST: 10-15 pages ; new RST : >100 pages |

NO | Norwegian | yes | For RST, only international standards | Norwegian | Norwegian; Swedish, Danish and English accepted | (not mentioned) | Usually 100 to 1000 pages |

SK | Slovak translation required for some documents, e.g. inspection records, authorisations, licenses, decisions, certificates… |

ANNEX C

OPTIONS | BENEFITS | COSTS |

Option code | Description | Effect on cost driver: | Comments | Estimated impact on Agency |

Testing / Documentation / Technical changes of rolling stock | Staff | IT systems | External services | Others |

CER | UNIFE | UIP | UIRR | EIM |

Q1, agree in principle | No. | Looks attractive, but does not change fundamentals | Yes, but option B5 should be applied | Yes, but proper implementation of Directives is a condition | Answers received late, apologies for not inserting them |

Q2, main difficulties | No authority = more bureaucracy | Additional burden on NSAs; work on national technical rules will take time | MS not willing to register wagons on basis of TSIs only |

Q3, other necessary measures | Binding task division, frozen checklist | Make national rules available; reduce language diversity | Existing RIV wagons should keep their approval; rules in 3 languages (FR, EN, DE) |

Q4 quick implementation | Not likely to work | No; several prerequisites |

Q5 benefits | None | No short-term benefits | Yes, to wagon keepers, NSAs and rail freight industry | Help mutual understanding of NSAs; wagon owners no longer left alone |

Q6 costs | Incurred by RU | Will divert useful resources in the short term |

Q7 effect on national rules | Will survive longer | No |

Q8 comments | A step backwards | Prefer short-term, voluntary options | Provide same rights for wagon keepers as for RUs under interop. directive | Adaptation of former regime to new, more neutral environment |

(continued)

Not addressed |

ANNEX G

[pic]

[pic][pic][pic][pic]

[1] Council Directive 96/48/EC of 23 July 1996 on the interoperability of the trans-European high-speed rail system.

[2] Directive 2001/16/EC of the European Parliament and of the Council of 19 March 2001 on the interoperability of the trans-European conventional rail system.

[3] Directive 2004/50/EC of the European Parliament and of the Council of 29 April 2004 amending Council Directive 96/48/EC on the interoperability of the trans-European high-speed rail system and Directive 2001/16/EC of the European Parliament and of the Council on the interoperability of the trans-European conventional rail system.

[4] Directive 2004/49/EC of the European Parliament and of the Council of 29 April 2004 on safety on the Community's railways and amending Council Directive 95/18/EC on the licensing of railway undertakings and Directive 2001/14/EC on the allocation of railway infrastructure capacity and the levying of charges for the use of railway infrastructure and safety certification (Railway Safety Directive).

[5] EPSF (France).

[6] Regulation (EC) No 881/2004 of the European Parliament and of the Council of 29 April 2004 establishing a European Railway Agency (Agency Regulation).

[7] In some countries, however (e.g. France), the national transposition of the Directive extends the notion of contracting entity to leasers, keepers, etc.

[8] ERA launche8QRSqrstu–˜¤©ª÷1 Ôîghi‹˜×Øáå

-

?

C

N

‘

“

”

²

ùõêãõãêãêãêõØÐĸ¯õd a call for tender for a research study on 23 September 2006 to examine in detail, with all the key players involved (suppliers, railway undertakings, infrastructure managers, NSAs), the procedures and criteria for the acceptance of Class 66 freight locomotives in the Member States where an application for authorisation to place into service has been made.ERA believes that this study should provide timely and useful information on the type of decision-making principles currently applied and soon to be applied by national safety authorities under Article 17 of the Railway Safety Directive in the field of rolling stock acceptance. The findings of the study should allow safety authorities to engage in an active exchange of views on the issue with a view to developing harmonised decision-making criteria in this area in the short to medium term. The study is also relevant to ERA’s work under Article 12 of Directive 2004/49/EC looking at the application requirements for safety certification and authorisation as well as its work on the examination of national safety rules and the development of Common Safety Methods (CSMs).

[9] The guidelines are available at the following Internet address http://ec.europa.eu/transport/rail/index_en.html

[10] This also seems to be the current approach of the Netherlands, among others.

[11] For instance, most benefits of the current traffic & operation management TSI depend on the successful closure of the listed open points.

[12] Note: extension of the geographic scope of TSIs is another, unrelated subject.

[13] The wagon TSI already mandates (in Article 7.6) cross-acceptance for new wagons operated in the EU.

[14] Telematics Applications for Freight.

[15] This could increase the pressure towards a limited number of languages for setting out the requirements and producing the files for authorisation. Railway undertakings, especially the small ones, will never be in a position to negotiate a “privilege” here. On the contrary, an NSA acting as a “lead NSA” will be on equal footing with other NSAs, since the lead roles may change. This also applies to other less important issues (formats, etc.).

[16] Source: DB industries

[17] Rolling Stock Companies.

EU25+NO, Year 2006

Total locomotives: 29 222

Crossborder locomotives: 4 344

Relative share: 15%