The economics of railway safety

This paper reviews the statistics and economics of railway safety in Great Britain, the European Union and the United States, together with some results for Finland and Japan. In these countries railway safety has improved over recent decades. That finding applies both to train accidents and to personal accidents such as persons struck by trains. Fatal train collisions and derailments command most attention even though they are infrequent and account for only a small minority of railway fatalities. Great Britain, the EU and the USA formally espouse conventional cost benefit analysis for the appraisal of railway safety measures, using the same valuations for the prevention of casualties as are used in road safety appraisal. However there are often strong institutional, legal and political pressures towards adopting railway safety measures with safety benefit: cost ratios well below 1. The best-documented examples of this are automatic train protection systems, which are discussed in the paper. Apart from trespassers, the largest group of railway fatalities occur at level crossings, which the paper also discusses. Level crossing safety measures would seem to be an appropriate subject for cost benefit analysis, but there are few case-studies in the literature. Over the last few decades, the railways in many countries have been privatised or deregulated with the aim of improving their economic performance. Such changes have the potential to affect safety. The paper reviews evidence of the effects on safety of railway restructuring in Great Britain, Japan and the United State, and finds no evidence that safety deteriorated.