52000DC0125

COMMUNICATION FROM THE COMMISSION TO THE COUNCIL, THE EUROPEAN COMMITTEE OF THE REGIONS - PRIORITIES IN EU ROAD SAFETY PROGRESS REPORT AND RANKING OF ACTIONS

COMMUNICATION FROM THE COMMISSION TO THE COUNCIL, THE EUROPEAN PARLIAMENT, THE ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

PRIORITIES IN EU ROAD SAFETY PROGRESS REPORT AND RANKING OF ACTIONS

Executive Summary

There is no room for complacency in road safety. Although the trend is towards fewer fatalities, the decline is levelling off and over 40 000 citizens of the EU continue to meet a premature death on our roads.

This Communication takes us beyond the Action Programme of 1997-2001, which introduced a socio-economic dimension to road safety with the 'EUR 1 million principle' which for the first time sought to give an economic cost to a fatality and the associated injuries and damage.

It takes up suggestions from the Council and the European Parliament to issue a progress report on the Action Programme. Also, as requested by the Parliament, it gives a ranking to future measures at the Community level, and it makes a Recommendation to decision makers at all levels, to give more weight to the fact that the cost of preventing accidents is generally much less than the economic cost of casualties and damage caused by them.

The Action Programme 1997-2001 can be considered a successful on-going process. Annex 3 contains a detailed and systematic listing of measures passed or under consideration. Much has been achieved in the strategic fields of improving enforcement of traffic rules and regulations, raising public awareness of road safety, and the gathering and dissemination of information on safety issues.

A multi-criteria analysis followed by a cost effective assessment of the actions listed in the Programme 1997-2001, has led to the definition of the following short and medium term priorities in road safety in the EU:

- Continue to work with and develop the European New Car Assessment Programme, (EuroNCAP);

- Campaigns and legislation on seat belts and child restraints;

- Recommendation to the Member States on maximum blood/alcohol levels in traffic;

- Legislation on speed limiters for light commercial vehicles;

- Develop guidelines for 'Black Spot' management (places with a concentration of accidents) and the design of 'forgiving' roadsides (i.e. less likely to cause injury in the event of an accident);

- Legislation on safer car fronts for pedestrians and cyclists.

Beyond these key priorities this Communication lists a further 5 measures for which more research into cost effectiveness is desirable: medical standards for driving licences; standards for driving tests; daytime running lights; effects of medicines on driver behaviour; post accident care.

Three supporting measures also received a high priority: the CARE accident statistics data base; an integrated information system; and, research into vehicle standards and telematics.

Finally, this Communication contains a Recommendation of the Commission encouraging governments, local and regional authorities of the Member States to establishing a practice of calculating costs and effects of road safety measures, to increase investments in these measures and to develop mechanisms that will enable the benefits of road safety measures to be felt more directly by those taking the decisions and bearing the costs of their implementation.

TABLE OF CONTENTS

1. INTRODUCTION

1.1. The EU road safety situation

1.2. The EU road safety programme 1997-2001

1.3. Purpose of this Communication

2. ACHIEVEMENTS OF THE ACTION PROGRAMME SINCE APRIL 1997

2.1. Scope

2.2. Improving enforcement

2.3. Raising public awareness

2.4. Information gathering and dissemination

3. PRIORITIES OF THE EU ROAD SAFETY PROGRAMME

3.1. Methodology

3.2. Multi-criteria analysis

3.3. Cost effectiveness assessment

3.4. Summary of results

4. CONCLUSION

4.1. Short/medium term EU Priorities

4.2. Recommendation of the Commission

Annex 1 : Road Accident Database

Annex 2 : The costs of accidents

Annex 3 : Achievements road safety programme 1997-2001

Annex 4 : Duration of effects of road safety measures

1. Introduction

1.1. The EU road safety situation

In April 1997 the Commission issued a Communication on Road Safety in the EU [1] which included a programme of measures.

[1] COM(97)131 final, 9.4.1997

In this Communication the road safety situation in the EU was addressed on the basis of the expected development of the total number of fatalities, under the scenario of a continuation of past trends.

This would result for the 15 Member States in a reduction of total fatalities from 45,000 in 1995, to 38,000 in 2000, to 32,000 in 2005 and 27,000 in the year 2010.

Annex 1 provides a number of statistics updated until 97/98 that inescapably lead to the following conclusions.

- There have been large reductions in fatalities in the last decade. This is considered by most experts to be mainly due to : - safer cars - increased seat belt use - less drink/driving - traffic calming and speed reduction measures. However, the decreasing trend in fatalities seems to flatten out, so that it is likely that in the year 2000 more than 38,000 fatalities will occur. (Table 1 and 2). Also the other estimations made in 1997 for the later periods were apparently too optimistic.

- Enormous differences still exist between Member States (Table 3) leading once more to the conclusion that there must be a large scope for improvement simply by applying best practice throughout the Union.

- There is no room for complacency, not even in those Member States that have better overall statistics. Tables 4, 5 and 6 present the most vulnerable road users indicating possibilities for improvement by more targeted policies for these categories. Unfortunately, as many national authorities have experienced, measures that would reduce the enormous human misery behind these figures, such as better enforced blood alcohol limits, better speed management and more responsible advertisements still have to be defended against misplaced accusations of "limiting personal freedom".

Improving the trend towards a stronger reduction of fatalities is a responsibility of authorities at all levels, including the EU where the right frameworks should be set.

1.2. The EU road safety programme 1997-2001

As indicated, the second road safety programme entitled "Promoting Road Safety in the EU" set out a programme for the period 1997-2001 and identified more than 60 fields of action classified into 3 broad categories:

1) The gathering and dissemination of information on:

- the development of EU road safety campaigns;

- target fields;

- target groups;

- effective measures (including the potential benefits of measures that could be carried out and an assessment of the effectiveness of measures that had already been taken);

- the implementation and enforcement of legislation;

- the rapid transfer of information and best practice throughout the Community.

2) Initiating and supporting measures to avoid accidents with an emphasis on human factors and its interface with the traffic environment. These measures include legislation, pilot projects and campaigns in order to improve the awareness, skills or physical condition of the road user, and extended control procedures and technical measures aiming at the creation of a more fail-safe environment.

3) Initiating and supporting measures to reduce the consequences of accidents when they occur. In this field, protection of the road user by seatbelts, helmets and vehicles with a higher crashworthiness is envisaged, as well as proposals for the development of EU standards and guidelines for more "forgiving" infrastructure when accidents occur.

In the programme, the introduction of the socio-economic dimension of road safety is seen as a key element of the strategy to accelerate improvements by using increasingly scarce public resources as effectively as possible. This strategy is based on the principle that the high costs of accidents (which were roughly estimated at EUR 1 million per recorded fatality) should be fully taken into account in the safety policies of Member States (Annex 2). This measure reflects only the direct economic costs and the value of lost output of reported road accidents and their consequent casualties. It does not include an estimate for non-reported accidents or for the value of human life, which can be very significant. Indeed, some Member States include a valuation of human life in their national estimates, but others do not on the grounds that valuing life is not possible. Therefore, although the EUR 1 million rule under-estimates the real costs of road accidents it provides an acceptable uniform minimum valuation.

When other Institutions debated the Commission Communication the need for selecting and/or ranking the proposed measures according to different criteria, including their cost effectiveness and/or casualty reduction value, was raised. The Council and Parliament have also invited the Commission to report regularly on the progress of the programme.

1.3. Purpose of this Communication

Firstly, this Communication takes up the suggestions of the Council and of the European Parliament, expressed both during the debate on this matter and on later occasions, to issue a progress report on the action programme presented in 1997.

Secondly, as also requested by the European Parliament in its Resolution, this paper presents a ranking of the proposed measures. This ranking has been carried out on the basis of fatality reduction potential and the cost effectiveness of the actions, but also takes into account a number of other criteria that are more difficult to quantify such as European added value, political feasibility and social acceptance.

Thirdly, as announced in the action programme of 1997, this Communication includes a recommendation for decision making at all levels in the field of road safety, which should give more emphasis to the high economic costs of road accidents in comparison to the costs of the remedial measures. Moreover mechanisms should be developed to bring the benefits of road safety measures closer to the decision-makers.

2. achievements of the action programme since April 1997

2.1. Scope

At the time of the publication of the action programme several actions were already on-going or about to start. It should be pointed out that the programme has worked as a catalyst for specific measures notably by bringing in the economic argument. Also the influence of the programme on other "non Union" actions should not be underestimated.

Annex 3 contains a detailed and systematic listing of measures from the action programme which have been passed or are under consideration as well as some additional measures which although not featuring in the programme, have had an important influence on road safety.

The following paragraphs give a summary description of only the most important activities at EU level since April 1997 in the strategic fields of improving enforcement, raising public awareness and information gathering and dissemination.

2.2. Improving enforcement

Since April 1997 the following actions of the programme in the field of legislation and enforcement have been taken

Legislative actions

Regulation on the second generation of tachographs. Formal adoption by the Council on 24.9.1998 (Reg. 2135/98/EC)

After completion of the technical provisions, this Regulation will introduce, from 2002, digital recording equipment for the registration of driving and rest times. This equipment will facilitate roadside checks and systematic controls at operator's premises.

Proposal for a Directive on the random roadside inspection for commercial vehicles that circulate on the roads of the European Community (Commission proposal COM(98)117 final of 11.3.1998)

This proposal, when adopted, will establish a regime of roadside checks of commercial vehicles, whether or not they are registered in the EU, regarding their safety and environmental performance.

Proposal for a Directive on the harmonisation of examinations for safety advisors for the transport of dangerous goods by road, rail or inland waterways (Commission proposal COM(98)174 final of 19.3.1998)

This proposal will complement the current legislation on road safety advisers (Directive 96/35/EC) by harmonising examination requirements.

Proposal for a Directive relating to the front underrun protection of heavy motor vehicles (COM(99)32 final of 10.2.1999)

This proposal defines technical standards for the type approval of this equipment with the aim of reducing the number of deaths and serious injuries in collisions between lorries and smaller vehicles.

Driving disqualification

On 18 June 1998, the 15 Member States signed a Convention which will give Union-wide effect to driving disqualifications. It covers disqualifications imposed for specific driving offences such as hit-and-run driving, driving under the influence of alcohol, or driving a vehicle faster than the permitted speed. It is based on the principle that the state where the offence was committed notifies the driving disqualification without delay to the central authorities of the state where the driver resides and that the latter acts upon this notification.

Alcohol

Legislation as regards maximum blood alcohol level of drivers

Since April 1997 another 4 Member States (Denmark, Spain, Austria and Germany) have adopted their legislation in order to lower the maximum permitted blood alcohol concentration (BAC) from 0.8 mg/ml to 0.5 mg/ml. This means that, at present, only 4 Member States (Ireland, Italy Luxembourg, and the UK) still maintain a legal limit greater than 0.5 mg/ml. The Commission considers that the differences in limits in the EU undermine the credibility of the lower limits which themselves are based on recent research. Alcohol remains one of the major causes of accidents a fact emphasised once more by the European Parliament during its debate on the road safety programme 1997-2001. The Transport Committee of the new European Parliament has not confirmed the old Commission's proposal (COM (88)707 and 640 final), to introduce a maximum BAC limit of 0.5 mg/ml, and has asked the Commission to make a new proposal which will better reflect developments in this field during the past decade. However, the Commission recognises the principle of subsidiarity in this field and the fact that the majority of the Member States have already moved to 0,5 mg/ml or less. Therefore, instead of a new legislative proposal, the Commission intends to submit a recommendation to Member States emphasising in this context the need for more effective enforcement and international co-operation in prosecuting drink driver offenders as well as the adoption of 0,5 or even lower limits for certain categories.

Drugs and medicines

In view of the increasing concerns regarding accidents involving the use of medicines and illicit drugs, and in order to review the situation on such a complex issue, the High Level Group on Road Safety decided, at the end of 1997, to reactivate the Working Group on Alcohol, Drugs and Medicines and Driving. The Group was asked to identify best practices and formulate recommendations on measures that should be considered at the EU level. The results of this Working Group are expected to be available mid 2000.

A research project (ROSITA) has been launched in the 4th framework programme on the development of a usable roadside test for drugs. The results of a study into the information provided to users of medicines in several Member States highlights the fact that there is still a lack of information concerning their effects on driving.

2.3. Raising Public Awareness

EuroNCAP

In the Communication of 1997 the Commission announced its intention to support the setting up of a New Car Assessment Programme (EuroNCAP) in order to create a market for safety by making available adequate consumer information about the passive (secondary) safety of cars.

The EuroNCAP consortium now involves the European Commission, the UK, German, French, Swedish and Dutch Departments of Transport, the FIA/AIT and their touring clubs, and consumer organisations. Other Member States are preparing to participate soon.

EuroNCAP involves crash testing cars in frontal and side impacts plus an assessment of new car fronts as regards the risk to pedestrians and cyclists. The results of test programmes of seven different classes of car have been published so far.

The Commission's funding of EuroNCAP has enabled a more comprehensive test programme involving a larger selection of cars in each class than would otherwise have been the case. The EuroNCAP programme has had a significant influence on both the design of new cars and the installation of safety equipment in current car models to the benefit of car occupants. Most manufacturers are in direct communication with the programme's co-ordinators in order to time product launches so as to ensure that the latest (and safest) model of car is crash tested.

The programme has raised consumer awareness of safety considerably and this was reflected in the sales of some of the cars that had outstanding test performances.

The Ten Seconds Campaign

The European Commission was a major financial contributor to several publicity campaigns one of which was the "Ten seconds that can save your life" campaign which ran from April to October 1998. This Europe-wide campaign focussed on 4 simple actions - taking less than 10 seconds to perform - which could save someone's life in an accident. The actions were: put on a seat belt; correctly adjust the seat and head restraint; stow loose luggage in the boot; fix children in a safety seat/harness.

Anti Drink Drive videos

The Commission funded the production of a number of videos, based on a Danish concept, which illustrated to young people the negative effects of alcohol on their ability to drive. These videos have been extensively shown in youth clubs etc.

MTV videos

The Commission has worked with the pop/rock music TV station MTV to produce three short videos aimed at young people on the themes of seat belt wearing and driving while under the influence on drink or drugs. The videos were screened on MTV's 4 European networks during the spring and summer of 1999.

2.4. Information gathering and dissemination

CARE database

On the basis of Council Decision 93/704/EC a database has been created of all accident statistics, involving at least personal injury, as recorded by national police forces. This database has proven to be a reliable tool for analysing accidents with fatalities but as regards injuries there is still a problem with underreporting, which is influenced at a national level by different reporting practices, definitions of injury, and different legal requirements on the reporting of accidents involving injury. A progress report was submitted to the Council and the European Parliament in June 1997. The Council in its Resolution of 18 June 1997 included a paragraph inviting the Commission to continue the development of this European accident database.

The first phase of transformation rules to create more common variables from disparate national data has recently been incorporated into the Care system, which will significantly increase the possibilities for producing comparable road accident statistics. Within the Commission existing CARE data has already been used to support policy development. For example:

- The age profile of fatalities and casualties in injury accidents of specific road user groups to support the discussion on maximum BAC limits.

- The distribution of fatalities, in and outside urban areas, by light condition, and calendar month, to support assessment of the effect of changing Summer Time upon road accidents.

- The involvement of heavy goods vehicles and other vehicles in tunnels (for the three Member States which record this information) to support an assessment of the current safety record of tunnels.

- The involvement of heavy goods vehicles and buses in fatal accidents, to support an assessment of the contribution of large vehicles in accidents in relation to maximum BAC limits.

- The incidence of fatal accidents involving fatigue, on the basis of data from 7 Member States who record this information.

- The distribution of heavy goods vehicles in fatal accidents, by day of week, to support the assessment of the possible road safety effects of week-end bans.

- The involvement of 'defective tyres' in fatal and non-fatal accidents to respond to enquiries from the tyre industry about the significance of tyre failure in road accidents.

Day-Time Running Lights (DRL)

There are many different views on this issue, and therefore research was commissioned from the Dutch road safety institute SWOV. Their report found a strong case for DRL on the basis of a positive cost benefit analysis.

Because of the complexity of this issue it was agreed with the Member States to form an expert group to examine the issue of DRL in more detail. This group met within the High Level Group - Road Safety. Although there were some disagreements on details the High Level Group concluded that the time had come to actively promote DRL.

3. Priorities of the eu road safety programme

3.1. Methodology

In its road safety programme 1997-2001 the Commission identified a number of actions to be undertaken to improve road safety. This action programme was discussed by the European Parliament and the Economic and Social Committee and received comments from governmental and non-governmental experts. In this process the identified actions were generally supported although a general desire for more concrete definitions and prioritisation was expressed.

In the process of prioritisation, cost effectiveness analysis is considered by leading experts to be one of the key elements contributing to a "mature" approach to road safety, as a method of rational decision-making based on the best available evidence. A cost effectiveness analysis is an analysis in which the aim is to find the cheapest way of realising a certain policy objective. Cost effectiveness calculations can be placed in a wider social context by comparing the cost of the measure per saved fatality with the benefits represented by one saved fatality (the "EUR 1 million Test"). Such a comparison turns the cost effectiveness analysis into a tool closer to cost benefit analysis which should help policy makers to set priorities in order to maximise efficiency in the use of public finances.

As well as the cost-effectiveness of a measure, an important factor in the selection of actions should be the estimated potential number of casualties that will be saved.

However, in the action programme of 1997 there are several measures that concern gathering and dissemination of information, or research. These are not specific road safety measures but are activities which generate information to both monitor and guide policy. They can only be ranked, rather subjectively, on the basis of their perceived benefit to road safety policymakers but are difficult to assess using cost effectiveness and/or casualty reduction potential as criteria.

Furthermore, there are also a number of qualitative criteria such as social acceptability, the political context, and the effects on other policies that cannot be integrated into an economic analysis, but which nonetheless play an important role in prioritising actions. These criteria can only be applied in a more subjective assessment.

A full cost effectiveness assessment is a complicated process because in many cases sufficient data is not available, requiring further study to be carried out. Therefore, taking into account the above mentioned subjective criteria, it was decided to be more pragmatic by carrying out the first ranking of the priorities in the EU road safety programme in two stages:

- A qualitative prioritisation of the measures. In order to reduce its subjective aspects this strategic approach was carried out by means of a systematic multi-criteria analysis involving the opinions of road safety policy makers in the Member States.

- This was followed by an estimation of the quantitative aspects of the actions that had received a high ranking in the first stage. In a number of specific cases this required a calculation of their casualty reduction potential and their cost effectiveness.

3.2. Multi-criteria analysis

In order to arrive at a strategic prioritisation of the action programme the Representatives of the Member States in the High Level Group on road safety together with a number of leading experts in Europe were asked to give for each proposed action in the programme a score from 1 to 5 for eight different criteria.

The following criteria were selected on the basis of similar exercises in other policy fields (e.g. environment) where a strategic prioritisation was made before an economic analysis could be carried out:

1. European added value: to what extent would implementing the measure at the EU level be consistent with the idea that the EU should establish policies only if this can be done more effectively than at the national or sub national level-

2. Institutional commitment: to what extent does the measure build upon past and present resource commitments by relevant organisations at the European level (e.g. Member State governments, industry associations, pilot programme developers)-

3. Social acceptance: to what extent is the measure viewed as legitimate by the majority of citizens in Member States-

4. Political feasibility: to what extent can the measure easily be decided upon by political decision makers (at the EU, national, regional and local levels), given the possible pressures imposed on their decision making by a variety of organised lobby-groups-

5. Ease of institutional implementation: to what extent can the measure's implementation be effectively performed by the public agencies responsible for implementation given their present strategies and resources (e.g. resources for monitoring and enforcement)-

6. Ease of target group implementation: to what extent can the target groups (e.g. car drivers, car manufacturers, schools, agencies responsible for granting drivers' licences, etc.) effectively implement the measures, given their present strategies and present resource constraints.

7. Lack of negative external effects on third parties (e.g. some types of monitoring could be viewed as having an impact on privacy)

8. Independence of a measure's impact from the implementation of other complementary measures (except in the case whereby implementing this measure is a pre-condition to successfully implement other measures, thus this measure must be implemented prior to other measures, to make the latter effective).

The responses and comments received made clear that this exercise was complicated and it led to a more explicit definition of some of the listed actions in the programme as regards:

- The level at which the action should be initiated (EU, national, local)

- The kind of action that is envisaged (legal, study, exchange of information, campaigns)

The calculation of the ranking of the actions on the basis of the replies received was carried out with different weighing factors for the above criteria in order to test the sensitivity of the model. It appeared that the outcome was not influenced in a major way by these variations.

In order to distinguish priorities for action at the Community level from actions that were deemed to be more useful at national or local level it was decided to make two separate calculations. In a first calculation the criterion "European Added Value" is given a weighting of 50% of the total while all other criteria are equally weighted in the remaining 50%. In a second calculation the European Added Value criterion is not taken into account.

The results of this exercise are presented in clusters of related actions in the following table.

// Top priority at Community level (50% weight for "European added value" in the assessment)

// - Crashworthiness of vehicles [2] (e.g. EuroNCAP; "friendly" design of motor vehicles; safer car fronts for pedestrians and cyclists and front underrun protection)

[2] Those actions having been given the highest priority within the clusters and which have been the subject of a further cost effectiveness analyses are shown in bold.

- Use of seat belts and child restraints; Use of motorcycle/cycle crash helmets

- EU road safety monitoring system; CARE database; CARE plus; Info system on national implementation of measures; Info system on research: Integrated info system

// High priority at Community level (50% weight for "European added value" in the assessment)

// - Speed limiters for heavy vehicles (evaluation), Variable speed messages; Speed limiters for lighter vehicles

- On-trip information RDS-traffic management channel; Emergency call systems; Automatic incident detection and emergency management

- Alcohol; Drugs/medicines

- Medical requirements for driving licences; Testing for driving licences; Disqualification of drivers

- Daytime running lights

// Additional high priorities for action at national/local level (no weight for "European added value" in the assessment)

// - Improved advertising

- Post accident care

- Forgiving roadside design; Black spot management (including safety audits of infrastructure, speed management of infrastructure)

3.3. Cost effectiveness assessment

Scope

For cost effectiveness assessments of actions in the field of road safety, monetary valuations have to be carried out measured in casualties avoided.

Firstly the scope of a potential measure must be defined. Actions on a European level can have a Europe-wide application, such as vehicle type approval measures. However, it is also possible that an action will only apply to selected Member States e.g. targeted campaigns. At the local level the assessment of specific infrastructure improvements can even be calculated per separate measure (e.g. per km of road or per roundabout).

Casualty reduction potential

Once the scope of a measure is defined then the basis for the calculation of its effectiveness can be its casualty reduction potential. The casualty reduction potential of most actions in road safety on a European scale has to be based on statistical analysis, assuming that a number of casualties would not have occurred if a certain contributing factor (e.g. alcohol, lack of seat belt wearing, poor infrastructure) had not been present. The examination of trends in road accident data before and after the introduction of a road safety measure is the usual starting point for analysis.

As agreed under 3.1. the casualty reduction potential of a measure is itself also a criterion for prioritisation and also serves as basis for the calculation of the effectiveness of a measure. However, for reasons of underreporting of injuries in statistics the potential of measures in the following will be expressed in the number of fatalities that can be avoided. For the purpose of this global assessment it is assumed that there is a rather constant relation between fatalities, serious injuries, light injuries and damage.

Effectiveness

The assessment of effectiveness in terms of saved fatalities is in some cases extremely difficult, especially when the actions aim at producing a attitudinal change in road users which may or may not lead to behavioural change.

In order to calculate the effect of a measure, a quantitative and a qualitative correction factor should be applied to its fatality reduction potential. The fatality reduction potential of a measure would be the result of such a measure if there was 100% coverage of the target group (quantitative) and 100% compliance of those that are covered (qualitative). This means that the effectiveness of a measure can be calculated as the product of its fatality reduction potential and the two factors that indicate the degree of coverage and compliance of the measure. For the different kind of measures these two factors should be estimated as precisely as possible, and it is clear that they will differ considerably. For instance, a European wide range of potential infrastructure improvements will, for budgetary reasons, have a low percentage of coverage/implementation but the compliance will be near to 100% as road users will normally not have the option of avoiding using these improvements. The degree of coverage of vehicle improvements depends on the number of vehicles affected by a measure (type approval legislation may make something compulsory only for certain new vehicles) and the compliance will be very high. On the contrary, legislative behavioural measures can have 100% implementation if such a measure affects all road users, but relatively low compliance when this measure is not accepted by the public or not sufficiently enforced. In this context, for the calculation of cost effectiveness of enforcement or campaigns, the quantitative factor coverage should be seen as the proportion of people reached, whilst the compliance stands for the proportion of these people that actually changed their behaviour as a result of such an action.

Costs

The calculation of costs is in many cases complex, because some costs are not easily estimated, especially when a particular action generates or reduces external environmental or congestion costs, and also because there could be influences on other policies that generate costs or benefits. An additional complication in a cost effectiveness analysis when applied on an EU scale is the variation in costs of complementary measures from one country to another as a result of varying purchasing power parity. Only global estimations on the basis of averages and applying conversion factors are possible. Finally, the costs attributed to actions should include all costs made by public or private bodies as the benefits are also calculated on the basis of the full social effects of avoided casualties and material damage.

Time reference

In order to make the cost-effectiveness ratios of different safety measures comparable, it is necessary to relate both the number of prevented accidents, damage and casualties, and the costs of implementing the measure to a certain time frame. This need arises because the relationship between costs and the duration of effects varies considerably between safety measures. The duration of the effects of various measures as estimated in the PROMISING project is reflected in Annex 4.

In order to compare costs of safety measures, the easiest method is to convert them to annual costs as the benefits (e.g. lives saved) are normally also calculated on a yearly basis. Simply dividing the investment costs by the number of years that a safety measure has effect is the easiest method. However, in order to take into account normal interest rates on investments it is, especially for measures that have effects over many years, more correct to convert investment costs to constant annual costs, which, if paid throughout the period it applies to, have the original investment cost as its present value.

As an example, if a measure has its effect over a period of 15 years and the interest rate is 5% then the annual cost are the investment costs divided by 10.4 (and not by 15).

In Annex 4 a complete table with conversion coefficients is listed.

Presentation of results

The way the cost effectiveness of a road safety measure is presented can vary depending on the stakeholder to which this assessment is addressed.

A consumer that buys a safety device or spends extra money for a safer car probably is most interested in the percentage risk reduction he or she pays for. In contrast, local authorities that invest in road improvement schemes or in a targeted campaign wants to see how many fatalities and associated injuries and damage are avoided as a result of this investment.

For the purpose of policy decisions at the national or EU-level such as prioritisation, the cost effectiveness of a measure can be expressed as the cost per saved fatality as an "inverted" cost effectiveness ratio which makes comparison with the EUR 1 million threshold easier.

This leads to the following basic formula:

// Costs of the measure

Cost effectiveness = -------------------------------------------------------------------------- Fatality reduction potential x Coverage x Compliance

Using this methodology the cost effectiveness assessment of specific actions that were given a high priority in the multi-criteria analysis of paragraph 3.2 was carried out. Although this communication principally deals with prioritisation of actions at EU level, for the measures that received a high ranking to be carried out on national or local level a global cost effectiveness estimation was also made. These measures, when cost effective, should be the subject of Community activity in the field of exchange of best practice, and development of guidelines.

3.

>TABLE POSITION>

4. Summary of results

>TABLE POSITION>

>TABLE POSITION>

>TABLE POSITION>

>TABLE POSITION>

>TABLE POSITION>

>TABLE POSITION>

>TABLE POSITION>

Certain measures such as research programmes or supporting actions such as CARE were not the subject of Cost Effectiveness assessments because of their indirect benefits on road safety whilst others (e.g. driver training programmes, post accident care) could not be assessed at this stage because of a lack of comparable and reliable data.

4. Conclusion

4.1. Short/medium term EU Priorities

The summary of results presented in 3.4 above sets out for each of the eight priority road safety measures, identified by road safety experts in the multi-criteria analysis, an assessment of the main components of their cost effectiveness. It was decided to narrow this list down to six short to medium term EU priorities taking into account the following criteria

- the scores of experts in the multi-criteria analysis,

- the estimated fatality reduction effect of each measure,

- and the estimated cost effectiveness of each measure.

The following table provides a relative ranking for each of these criteria for each of the selected priority road safety measures. In terms of the multi-criteria analysis the six measures have been ranked according to whether they were considered to be top priority (1) or medium priority (2). But in terms of fatality reduction and cost effectiveness each measure has been ranked in order from (1) to (6).

It is clear on this basis that the EuroNCAP programme which has done so much to reduce the severity of injury in accidents in recent years is still regarded as the most significant road safety action to improve road safety in the future, but all other priority measures provide the impetus required to significantly improve road safety in the EU in the short to medium term.

>TABLE POSITION>

* MC : Multi Criteria analysis * FRP : Fatality Reduction potential * CE : Cost effectiveness

The next 5 measures are also priorities but more research is needed to assess their cost-effectiveness (the prioritisation is based only on a multi-criteria analysis):

// 1) Medical standards for driving licences

2) Testing for driving licences

3) Daytime running lights

4) Effect of medicines on driver behaviour

5) Post accident care

The following 3 priorities are supporting measures that have received a high priority ranking in the multi-criteria analysis but for which a fatality reduction potential is difficult to define:

// 1) CARE database

2) Integrated information system

3) Research into vehicle standards and telematics

4.2. Recommendation of the Commission

The use of cost effectiveness, or in a later stage cost-benefit, calculations to support road safety should be encouraged at all levels. The tentative calculations made for the purpose of this report demonstrate in many cases that investment in road safety is a first class investment in welfare and well being. The paradox is that a calculation on a higher level (EU, Member State) is on the one hand easier because costs and benefits can be allocated to the same stakeholder (the EU/society) but on the other hand is highly complicated because of the global estimations of the effects. Research is needed in order to facilitate application of cost-effectiveness on all levels.

Finally, it should be emphasised that the distance between the decision-maker that has to pay the costs and the one that profits from the increased safety is probably one of the main reason for the slow progress in road safety.

In cases where the consumer himself benefits (e.g. airbags), there is less need to legislate but only to inform him of the safety benefits, whilst for more altruistic measures (no bull bars), legislation seems to be inevitable.

It seems appropriate in these circumstances to think about the development of alternative incentives such as tax discount, lower insurance schemes, that could accelerate the investments in safety at all levels.

Given the above, the Commission issues the following Recommendation :

// Recommendation of the Commission

The Commission,

Having regard to the number of casualties, injuries and damage caused by road accidents,

Having regard the persistent underestimation of the economic consequences of these accidents when general political priorities are set,

Encourages Member States, regional and local authorities to establish a practice of calculating the costs and effects of road safety measures and where appropriate comparing these with the costs of avoided accidents,

Invites Member States, regional and local authorities

- to increase investment in road safety projects, as these investments are economically highly justified,

- to develop mechanisms that will enable the benefits of road safety measures to be felt more directly by those taking the decisions and bearing the costs of their implementation.

ANNEX 1

European Commission

Transport Directorate General

Unit for safety, technology and environment

CARE

>REFERENCE TO A GRAPHIC>

Road Accident Statistics

Table 1 : Number of persons killed*

>TABLE POSITION>

Sources : CARE - National sources

* 1. Persons killed are all persons killed within 30 days from the day of the accident. For Member States not using this definition, corrective factors were applied - (Gr: 1,18 up to 1995 - F: 1,09 up to 1993 and 1,057 after 1994 - I: 1,078 - P: 1,3)

2. Figures in Italic are based on estimated trends

Road Accident Statistics

Table 2 : Trends in the number of persons killed *

>TABLE POSITION>

* 1. Persons killed are all persons killed within 30 days from the day of the accident. For Member States not using this definition, corrective factors were applied - (Gr: 1,18 up to 1995 - F: 1,09 up to 1993 and 1,057 after 1994 - I: 1,078 - P: 1,3)

2. Figures in Italic are based on estimated trends

Diagram 1

>REFERENCE TO A GRAPHIC>

Road Accident Statistics

Table 3 : Number of persons killed* per million inhabitants

>TABLE POSITION>

Sources : CARE - National sources

* 1. Persons killed are all persons killed within 30 days from the day of the accident. For Member States not using this definition, corrective factors were applied - (Gr: 1,18 up to 1995 - F: 1,09 up to 1993 and 1,057 after 1994 - I: 1,078 - P: 1,3)

2. Figures in Italic are based on estimated trends

>REFERENCE TO A GRAPHIC>

Road Accident Statistics

Diagram 2 : Number of persons killed* per million inhabitants

Sources : CARE - National sources

* Person killed are all persons killed within 30 days from the day of the accident. For Member States not using this definition, corrective factors were applied - (Gr: 1,18 up to 1995 - F: 1,09 up to 1993 and 1,057 after 1994 - I: 1,078 - P: 1,3)

Road Accident Statistics

Table 4 : Children (age less than 10) fatalities*

>TABLE POSITION>

Table 5 : Pedestrians fatalities

>TABLE POSITION>

Table 6 : Cyclists fatalities

>TABLE POSITION>

Sources : CARE - National sources

* Persons killed are all persons killed within 30 days from the day of the accident. For Member States not using this definition corrective factors were applied - (Gr: 1,18 up to 1995 - F: 1,09 up to 1993 and 1,057 after 1994 - I: 1,078 - P: 1,3)

2. GR: Data available for 1995

ANNEX 2

The costs of accidents

The introduction by the Commission of the "1 million ECU test" in the road safety programme 1997-2001 has generated a lot of discussion and therefore it should be emphasised once more that it is calculated as the reported economic costs of all types of road accidents (fatalities, injuries, damage only) divided by the total number of fatalities. At the time of the publication of the report in 1997, this total of hard costs was estimated to be roughly 45 billion ECU whilst the number of fatalities in the EU in 1995 was approximately 45,000.

The latest, more precise, estimations of socio-economic costs of road accidents in the EU, including value of human life and unreported accidents on the basis of the statistics of 1995, were made by the ETSC and are reflected in table 1. Applying the same calculation as before would lead to dividing the total reported "economic" costs of 52 billion EUR (the grey area) by 45,000 resulting in an adjusted value of EUR 1,15 million in 1999.

Table 1: Socio-economic costs of road accidents in the EU (billion EUR)

>TABLE POSITION>

However, this figure is also a clear underestimation of the real costs because of underreporting of non-fatal accidents.

EUR 1 million should only be seen as an indicative value that cannot be contested as too high and therefore the "EUR 1 million Test" is maintained for the purpose of promoting EU road safety.

ANNEX 3

Achievements road safety programme 1997-2001

>TABLE POSITION>

Field I: Information gathering and dissemination

>TABLE POSITION>

Field II: Accident avoidance

>TABLE POSITION>

>TABLE POSITION>

Field III: Reduction of consequences of accidents

>TABLE POSITION>

ANNEX 4

Duration of effects of road safety measures.

Category of safety measure // Typical duration of safety effects (service life)

Land use planning; new residential areas

Road pricing; fuel or vehicle taxation systems

Changing the modal split of travel

Major road investment projects (new roads)

Minor road investment projects

Traffic control by means of highway signs

Traffic control by means of road markings

Upgrading road maintenance

Vehicle safety regulations (for new vehicles)

Driver education and training (new drivers)

Training of children

Public information campaigns

Conventional police enforcement

Automated police enforcement // 25-40 years

1-3 years for use of existing motor vehicles, 10-15 years through vehicle purchasing decisions

Contemporaneous effect only

25-40 years

15-25 years

10-15 years

1-10 years

1 year

10-15 years

1-3 years

1-3 years

Effect only during campaign, or a short time after

Effect only when operated, or a short time after

Effect only when operated, or a short time after

Conversion from investment costs to annual costs (coefficients by which the investment costs should be divided)

>TABLE POSITION>