Crash cost estimates for European countries

Deliverable 3.2 of the H2020 project SafetyCube


Wijnen, W.; Weijermars, W.; Vanden Berghe, W.; Schoeters, A.; Bauer, R.; Carnis, L.; Elvik, R.; Theofilatos, A.; Filtness, A.; Reed, S.; Perez, C.; Martensen, H.



Safety CaUsation, Benefits and Efficiency (SafetyCube) is a European Commission supported Horizon 2020 project with the objective of developing an innovative road safety Decision Support System (DSS) that will enable policy-makers and stakeholders to select and implement the most appropriate strategies, measures and cost-effective approaches to reduce casualties of all road user types and all severities.

SafetyCube also aims to support stakeholders in doing an Economic Efficiency Assessment (EEA) of road safety measures. To apply an EEA, information on costs of crashes is needed. This Deliverable provides information on costs of crashes in Europe. First of all, it provides an overview of the cost components that should be included in crash cost estimates and how each cost component should be determined according to the international guidelines and best practices. Second, information on costs of crashes is collected by means of a survey among all EU countries. Third, for some countries not all information is available or costs are not calculated according to the international guidelines. In those cases, comparable estimates are provided by means of value transfer. In that way, we also provide an estimate for the total costs of crashes in the EU.

Although they were developed more than 20 years ago, the COST313 guidelines are still the most comprehensive guidelines for estimating the costs of road crashes. Therefore, these guidelines are the main basis for the framework developed for the cost estimates within SafetyCube, although more recent developments have been taken into account as well. Within SafetyCube, costs of crashes are considered from a socio-economic perspective and the following cost components are taken into account:

  • Medical costs (e.g. costs of transportation to the hospital, costs related to hospital treatment)
  • Costs related to production loss
  • Human costs
  • Costs related to property damage (mainly vehicles)
  • Administrative costs (e.g. police, fire department, insurances)
  • Other costs (funeral costs, congestion costs)

According to the international guidelines; medical costs, costs related to property damage, and administrative costs should be calculated by means of the restitution costs method. Costs related to production loss should be calculated by means of the human capital approach, which implies that the loss of productive human capacities is valued. The (individual) willingness to pay (WTP) approach is generally recommended to estimate human costs, although several other approaches have been developed as well. In Germany and Australia for example, human costs are based on financial compensations that are awarded to road casualties or their relatives in courts or by law. Another approach is to deduct human costs from premiums people pay for life insurances or from public expenditures on improving (road) safety. These alternative approaches typically result in much lower values than those from WTP studies. Within SafetyCube we recommend the (individual) WTP approach to estimate human costs, as this is the most theoretically sound method, in particular for use in cost-benefit analysis, and is common practice in many countries.

Information on costs of crashes in European countries was collected by means of an Excel based questionnaire that was developed together with the InDeV project. Information is collected concerning: costs per casualty and crash by severity level, total costs, costs per component, methods and definitions, and number of casualties. Official cost figures used by governmental organizations were requested. Questionnaires were initially prefilled by a responsible SafetyCube or InDeV partner using available information and then sent to experts in each country for a check and final completion. Data from 31 European countries, out of the 32 involved in the study, was obtained and included in the analysis. Within SafetyCube, the questionnaires were integrated into a SQLite database, consistency checks were carried out, and the data was standardized for currency, inflation and relative income differences.

For all EU countries, except Romania, at least some information on costs of crashes was available. Reported costs per fatality vary between €0.7 million per fatality in Slovakia and €3.0 million per fatality in Austria and tend to be higher in North-West Europe than in South and East Europe. Reported costs per serious injury range from €28,000 in Latvia to €959,000 in Estonia, whereas reported costs per slight injury range from €296 in Latvia to €71,742 in Iceland. When we relate the costs per injury to the costs per fatality, it shows that the costs of a serious injury range from 2.5% to 34% of the costs of a fatality, though for about three quarters of the countries this figure is between 10% and 20%. The costs per slight injury are 0.03% to 4.2% of the costs of a fatality.

The total costs of crashes vary between 0.4% and 4.1% of the Gross Domestic Product (GDP). There is no clear geographical pattern. A better road safety performance should in principle result in lower road crash costs, but we found only a weak positive relation between mortality rate and costs as a percentage of GDP. Exclusion of property damage only (PDO) crashes or other severity levels and not-correcting for underreporting can result in an underestimation of the total costs of crashes. Differences between countries are also due to methodological differences, particularly whether the Willingness to Pay (WTP) method is applied for the calculation of human costs. In countries that use the WTP approach, human costs have a major share (34% to 91%) in the total costs of crashes. In countries that apply an alternative method, the share of human costs in the total costs is much smaller (less than 10%). Also, property damage costs and production losses are major cost components in most countries, whereas medical costs and administrative costs are relatively low.

Injuries appear to have a large share in the total costs (on average 2.4 times higher than the share of fatalities in total costs), but this differs substantially between countries. For countries that included all severity levels, fatalities account for 7.4% to 55% of the costs, serious injuries account for 14% to 77%, slight injuries account for 1.9% to 34% and PDO crashes account for 2.0% to 55%. Possible explanations for this variation include differences in definitions of severity levels and in reporting rates.

Not all countries have information for all cost components and/or all severity levels. Also, not all countries calculate cost estimates according to the international guidelines. Within SafetyCube, the value transfer method is applied to estimate standard cost values per casualty/crash type and to estimate total costs of crashes according to international guidelines for each EU country. The value transfer method uses crash cost estimates from countries whose estimates are consistent with international guidelines to estimate costs for countries that do not have cost information according to the international guidelines. The general approach is that the median (adjusted for purchasing power parity) value per casualty (fatality, serious injury, slight injury), and per crash (fatal, serious injury, slight injury and PDO), for a specific cost component, is determined for a group of countries that use the recommended methods and included all relevant cost items. This median value is used for countries that have not used the recommended method or do not have information at all for that cost component. The ‘standard’ costs of a fatality are estimated at €2.3 million. These costs mainly consist of human cost (€1.6 million) and production loss (€0.7 million). Costs per serious and slight injury are estimated at 13% and 1% of the value of a fatality. Also for injuries human costs are by far the largest cost item. Total costs according to the international guidelines in all EU countries individually as well as the EU in total were calculated. For the 28 EU member states costs are estimated at about €270 billion if the results of the value transfer approach are applied. This corresponds to 1.8% of the GDP. This is still an underestimation, because many countries have not corrected the numbers of casualties/crashes for underreporting. If unreported casualties and crashes are taken into account, we expect that total costs are in the order of magnitude of at least 3% of GDP. The European total costs based on the values given in the survey are almost €200 billion, which clearly shows the importance of adding the missing components, and of using a standard methodology, in estimating total costs of crashes.

For future cost studies in individual countries it is recommended to include all relevant cost items and to use the internationally recommended methods, in order to provide a complete picture of the socio-economic costs and to make costs estimates more comparable across Europe. In addition, we recommend monitoring the socio-economic impact of road crashes on a European-scale as well as the methods used to estimate the costs by repeating the survey on a regular basis. Finally, new research into the costs of serious injuries, particularly human costs, is recommended as information on these costs is very limited yet they have a major impact on the total costs.

Print this page


85 + 26

Gepubliceerd door

European Commission, Brussels