Transport Safety Performance in the EU – A Statistical Overview

  • July 1, 2003

Executive Summary

Transport crashes in the EU killed about 39,200 EU citizens in 2001, cause over 3.3 million casualties and presently cost over 180 billion Euros, around twice the total EU budget for all activity.

Nowadays road crashes in the EU lead to 97% of all transport deaths and to more than 93% of all transport crash costs and are the leading cause of death and hospital admission for citizens under 50 years. Road crashes cost more than congestion and pollution, or cancer and heart disease.

Comparisons across the modes

Risk comparisons for the EU show that the fatality risk for motorised two-wheelers is the highest of all modes, being on average 20 times higher than for car occupants. Also cycling and walking have on average a 7 to 9 times higher fatality risk per distance travelled than car travel. Road traffic collectively has the highest fatality risk per passenger kilometre of all passenger transport modes.

Rail and air travel are the safest modes per distance travelled, followed by bus. The passengers of trains, bus/coach and planes within the EU have the lowest fatality risk per passenger kilometre. For the average passenger trip in the EU, bus travel has a 10 times lower fatality risk than car travel and air travel within the EU has for the average flight distance about the same fatality risk per passenger kilometre as train travel and both are half as risky as travel by coach. The risks associated with ferry travel fluctuate, but the expected fatality risk is 4 to 8 times that of train travel.


Fatality risk in road transport: the current EU countries

France, Germany, Italy, and Spain account for 68 per cent of the total of 38,935 road fatalities in the EU in 2001. France and the UK have about the same number of inhabitants, but France has about twice the number of fatalities as the UK as its risk is twice as high.

The southern EU countries, France, and Belgium have fatality risks above the average for the EU, and risks in the other EU countries are below-average. A five times higher death rate per motor vehicle kilometre is nowadays present in the worst than the best performing Member States (in 1997 this ratio was 7).

The fatality risk order of EU countries in 2000/2001 changed with respect to 1997, due to noticeable differences between the rates at which national risks have fallen. The six countries with lowest fatality risks have had numerical targets for casualty or fatality reduction in the past decade or decades.

The fatality rate in the EU has decreased almost constantly in an exponential way, by an average annual reduction of 5.3%. Combined with extrapolated traffic growth it leads to a predicted 27,000 fatalities in 2010 within the present countries of the EU. The EU target of reducing fatalities by 50% between 2000 and 2010 (i.e. to 20,000 fatalities in 2010) will only be achieved if the EU takes additional actions that reduce the risk more rapidly than in the past.

Fatality risk in road transport: the ten accession countries

Motor vehicle kilometre data are not available for the accession countries. In order to compare their fatality risks with those of the present EU countries, it is assumed that on average each vehicle drives 10 thousand kilometres per year, which seems approximately correct overall. With this assumption, the average fatality risk of the 10 accession countries is about 3 times higher than the EU average, slightly higher than Greece (worst performing EU country) and 5 times higher than the UK (best performing EU country).

The multiplication of extrapolated trends for fatality rate and vehicle growth predicts a total of 8,625 fatalities in 2010 in the 10 accession countries, where in 2001 about 11,000 road traffic deaths (within 30 days after the crash) are reported.

The safety of motorised two-wheelers in the EU

Users of motorised two-wheelers have the highest fatality risk of all road user groups. Moped and mofa fatalities are predominantly in the southern EU countries. This is also true for motorcycle and scooter fatalities, but also Germany and the UK have relatively many motorcycle and scooter fatalities.

Railway passenger safety in the EU

The fatality risk per billion train passenger kilometre fell from 1970 to 2000 by an average of 5.5% per year. The mean annual number of train passenger fatalities in the EU decreased from about 400 in the early 1970s to around 100 in 1999 and 2000. The rail passenger kilometre in the 15 EU-nations increased from about 200 billion in 1970 to about 300 billion in 2000. The expected fatality risk per billion train passenger kilometre is about 0.35 for 2001/2002 within the EU.

Passenger safety in civil aviation within the EU

Air transport fatality statistics refer mainly to scheduled flights, because air travel fatalities on unscheduled flights are only partially reported by international air transport organisations. A similar practice exists for private plane fatalities – they are neither registered worldwide nor for the EU. Due to large annual variations of air passenger deaths in scheduled flights and the absence of reliable data on unscheduled flights within the EU, the most probable passenger risks in civil aviation within the EU is obtained from the average of (1) the world-wide risk statistics with corrections for the share, duration or distance and occupancy of flights within the EU and (2) the actual observed risks for scheduled flights within the EU, both adjusted for the estimated additional share of unscheduled flights with estimated risks from their partially available data. The passenger fatality risk in civil aviation within the EU is estimated to be about 0.035 per 100 million passenger kilometre, but 16 per 100 million passenger travel hours. Air and train travel within the EU have now on average the same risk per distance. However, planes have much higher risks during the take-off, climb, descent, approach and landing manoeuvres than when cruising. Therefore, air travel trips have the higher risks per distance the shorter the air travel trips are, whereby trips below 600 kilometres ground distance are safer by (high speed) train than by plane. Due to regional differences in the safety procedures and ground communications around airports, the risks of air travel differ markedly according to region of departure and of destination, and less by the home country of the flight operator. The main airport regions within the EU belong to the safest in the world.

Passenger risk by sea transport in EU waters

Casualty risks per ship type and size are given in the ETSC Report of 1999 on “Exposure data for travel risk assessment”. This Report shows that tankers and ships of more than 6 million kg cargo capacity have a casualty risk per million nautical miles that is more than twice as high as for all vessels (about 2.8 compared with about 1.35), while casualty risk for ferries and roll-on/roll-off container ships is almost half of the risk for all vessels.

There were no ferry passenger fatalities in EU waters after the ESTONIA disaster in 1994. The risk levels identified in the ETSC Report of 1999 can now be updated for the longer period 1984-2001 and are thus less dominated by the exceptional number of fatalities in 1994. The updated fatality risks of ferry passengers that are comparable with other transport modes risks become:

fatalities/100 million passenger hours 8.0

fatalities/100 million passenger kilometres 0.25


  1. Road safety needs more priority in the transport policies of EU Member States and the EU, because 97% of all transport fatalities in the EU are caused by road transport. Road transport accounts for 88% of all passenger transport in the EU, but accounts for over 100-times more deaths than all other modes together.
  2. It is recommended that national and EU health policies recognise the relatively high mortality and injury incidence rates for road traffic.
  3. The EU target of 50% road traffic fatality reduction between 2000 and 2010 to about 20,000 fatalities in 2010 will not be achieved unless the EU takes additional actions that reduce the fatality risk more rapidly than in the past. Therefore, it is recommended that further actions within the competence of the EU itself (mainly vehicle safety regulations) are taken and that a EU road safety subsidy fund is created for financial incentives that support and trigger national road safety actions and measures with a proven effectiveness.
  4. Priority setting for transport safety must recognise the very high fatality risk of motorised two-wheelers (15 times the average road risk per kilometre travelled).
  5. It is recommended that the safety of pedestrians and cyclists be improved, because their fatality risks per kilometre are 7 to 9 times higher than for car travel.
  6. Passenger transport policies of the EU and its countries should promote the use of (high speed) trains in long distance trips, because the fatality risk of air travel for ground distances of less than 600 kilometres is higher than for trains.
  7. Intermodal passenger transport policies have to recognise that large differences exist between the risks of travel modes. The safety of walking and cycling needs also to be improved in order to optimise the safety of public transport, due to the high risk of the necessary walking and/or cycling in the ‘before and after’ phases of these trips.
  8. Initiatives to improve the recording of road travel volumes and fatalities are being undertaken by the EU, but progress is lacking for rail travel in the EU, travel on EU waters and inland waterways of the EU, and air travel by unscheduled flights and private planes within the EU. Moreover, serious injury risks for different travel modes are hard to assess because the necessary incident and exposure data are defined differently and not consistently gathered. It is recommended that research and development as well as reporting harmonisation on these matters be initiated by the EU.

Similar recommendations (except the third) were formulated in 1999 in the ETSC Report on “Exposure data for travel risk assessment”, but it is disappointing that they have had so little effect. In view of the EU road traffic fatality reduction target that has been set and the urgency of taking action to achieve the target, the recommendations need to be implemented soon.