House of Commons - Transport

Select Committee on Transport Seventh Report

5 SAFETY

183. The enforcement of the safety legislation applying to the railway is the responsibility of Her Majesty's Railway Inspectorate (HMRI) within the Health and Safety Executive (HSE), the executive arm of the Health and Safety Commission (HSC). The HSC regulates safety on the railways using the provisions of the Health and Safety at Work Act 1974 together with a variety of specific rail regulations set out as secondary legislation. The HMRI is independent of the railway industry. Other organisations, such as the British Transport Police, the Crown Prosecution Service, and the new Rail Accident Investigation Branch (RAIB),[315] may also become involved in rail safety when accidents occur. When operational, the RAIB will have responsibility for investigating serious rail accidents, but will have no powers of prosecution.

184. The Rail Safety and Standards Board, a body owned by the railway industry, is responsible for rail safety research, a wide range of industry standards, and leading the industry to improvements in health and safety performance.[316] The Board is not involved directly in the safety regulation of the railways, but as the industry "leader" on safety, works closely with the HSE.

Safety record

185. Rail travel is comparatively safe and is becoming progressively safer,[317] as shown in the following graph:

Chart 2

Passenger and Workforce Fatality Rates since 1946

Source: Network Rail

186. The figures comparing road and rail fatalities are telling. In 2002, 3,431 people were killed on the roads;[318] while according to the HSE "a total of 50 passengers, railway staff and other members of the public" were fatally injured and 256 people who died as a result of trespass and suicide on the railway.[319] The SRA points out that "On average more road users die in accidents each day than rail passengers in a year."[320] There will always be dangers on the railways. The recent tragic death of four railway workers in Cumbria on the 15 February 2004 is evidence of that.[321] It remains the case, however, that for the public the railway is an extremely safe way to travel.

Criticism

187. During this inquiry we received a number of informed criticisms of rail safety matters and the Health and Safety Executive (HSE), the rail safety regulator. Mr Alan Osborne, who recently left the HSE as Director of Rail Safety after less than one year in post, considered that safety standards were being "gold plated" because the criterion for balancing risks and costs ("reasonable practicability") was imprecise and was being applied inconsistently which, he felt, could lead to inappropriately high costs when applying safety measures;[322] that the rail expertise of the HMRI was being seriously diluted;[323] that poor communication by HSE with the rail industry had lead to ignorance about its function and it being held in low esteem by rail "stakeholders"; and there was an inappropriate split between the rail policy and operations branches of the HSE. Mr Osborne considered that the regulation of rail safety should be removed from the HSE; and that an exclusive, legal regulatory framework was required specifically for rail.[324]

188. Network Rail had criticisms of the present safety arrangements, including the inappropriate classification of the industry by the HSE as a "major hazard" industry with the offshore oil, chemical and nuclear industries which, in its view, has contributed to its insurance costs rising fourfold. It considered that the rail safety and standards framework as a whole should be reviewed to ensure consistency and cost effectiveness because of the complexity of the regulations and the number of organisations involved; and that the threat of prosecution in the aftermath of accidents has induced a culture of 'risk aversion' in the industry.

Costs

189. When Dr Kim Howells, the Minister of Transport, gave evidence to us, he pointed out that 100% safety on the railway was not possible.[325] While it was vital that the travelling public should have complete confidence in the railway's ability to deliver them safely to their destination, the costs of achieving almost complete safety were problematic:

326

190. While the HSE,[327] Network Rail,[328]and the RSSB,[329] could readily tell us their administrative budgets for rail safety, no one could provide us with a sum totalling the cost of safety to the industry as a whole. In Network Rail's case this was because "the needs of safety arise in every activity we carry out as a company".[330] We asked the Secretary of State for Transport who said "It is difficult to put a precise figure on it….Frankly, safety should be in with the bricks". [331] In the absence of a credible overall figure for the costs of rail safety - which may be extremely difficult to arrive at - it is impossible to conclude that safety on the railways as a whole may be costing too much, or that insufficient is being spent. This does not mean that judgements on the cost effectiveness of particular projects are not possible. However, we have received evidence that the HSE has caused unnecessary extra cost by "gold-plating".

"GOLD PLATING"

191. While the HSE was adamant that it was not applying an unreasonably high standard of safety,[332] we did receive allegations of rail safety standards being "gold plated" on projects as a result of the unclear criterion for balancing risks and costs, and the fear of prosecution ('risk aversion').[333]

Updating standards

192. We were astonished to hear from Mr Osborne that one cause of inappropriate standards being applied resulted from a decision on the part of the HSE's Director-General to "put on hold" the updating of Railway Safety Principles and Guidance documents with the result that standards applied by the Railway Inspectorate appeared not to have undergone proper risk assessment. The reason for the delay was, apparently, internal HSE budgetary considerations.[334]

193. All safety standards applied by the HSE must be constantly updated and be the result of proper and rigorous risk assessment. Unless this is carried out the HSE's service to its railway customers will be unacceptably diminished. It is also completely unacceptable that the industry should risk incurring huge costs for inappropriate safety measures because of relatively small budgetary considerations in the HSE. Where such costs are incurred there will be an inevitable knock-on to the travelling public who are therefore made to bear at least a part of the costs of the safety regulator's poor practice.

194. Mr Osborne provided several instances of "gold plating" in supplementary evidence, for example, automatic level crossings which are the most expensive in the world; Train Protection Warning System (discussed below); and an inflexible approach to permitting selective door opening on trains which are longer than the platforms serving them.[335] Mr Osborne indicated that he had intervened to change the HSE's 'fairly aggressive' policy of seeking "very expensive" platform alterations in the longer term so that selective door opening was acceptable. He pointed out that the prospect of these expensive alterations had placed new train services at risk because the train operating companies balked at the costs, thereby risking the economic viability of services. Two particular instances of "gold plating" cases give us cause for concern.

"MAJOR HAZARD"

195. When we asked HSE about its classification of the railway as a "major hazard" industry, Dr Sefton, Director of Rail Safety at the HSE, appeared to down-play the classification by stating that "The truth of the matter is that this is simply a shorthand that is used within HSE."[336] In the HSC's Annual Report 2002/3, however, rail features in a section entitled 'Major Hazard Industries' alongside the nuclear, offshore petroleum, gas conveyance, and mining industries.[337] It is clear to us that HSE's classification of the railway as a "major hazard industry" is rather more than simple "shorthand".

196. Network Rail told us that this classification was inappropriate in the light of the safety record of the industry and had an adverse effect upon its insurance premiums which have increased from £20 million to £100 million per annum.[338] Mr Osborne believes that the relevant benchmarks for rail are other transport modes, not separate industries,[339] and instanced the Potters Bar accident where the HSE investigation had been led by a nuclear industry expert whose recommendations appeared to him inappropriate.[340] Mr Len Porter, Chief Executive of the Rail Safety and Standards Board, was not more sympathetic to the present classification "if you want to class the rail industry…as a 'high hazard industry', then you would have to class the highways as 'high hazard' in terms of carrying hazards loads."[341]

197. We probed Dr Sefton about bracketing a rail accident in the UK which, at worst, might involve hundreds of people but was more likely to involve many less, with a major nuclear or chemical accident which would almost certainly involve many thousands, or more. He indicated that the characteristics of major hazard industries were, the rarity of accidents coupled with the scope of the result when such accidents occur, affecting "considerable numbers of people in exactly the same way that a serious train crash would."[342] We asked if the HSE had received representations on the classification point from Network Rail. Dr Sefton was clear that "nobody has come to me and said, 'Will you stop referring to us as a major hazard industry?"'[343] However, we received supplementary evidence from Network Rail that its concern about this had been raised "directly" with representatives of the HSC and the HSE in different forums such as the Rail Industry Advisory Committee and the Safety Advisory Committee.[344] It is extremely disappointing that Network Rail's representations on the inappropriateness of designating the industry as a 'major hazard' appear not to have been brought to the attention of the current HSE Director of Rail Safety.

198. We do not accept that parallels exist between the rail, nuclear and chemical industries. We certainly do not underestimate the extremely serious consequences of any train accident; but the scale of nuclear and chemical accidents is not likely to be similar to those on the railway. In our view, this places in doubt HSE's designation of the railway as a "major hazard" industry which we think could result in a "gold plated" approach being taken to rail safety. The HSE needs urgently to rethink its approach to the classification of rail as a "major hazard" industry and generally to adopt a more nuanced, railway-specific approach to such matters.

TRAIN PROTECTION WARNING SYSTEM

199. Installation of the Train Protection Warning System (TPWS), a safety system designed to stop trains at speeds of up to 75 mph, was completed throughout the UK network at the end of 2003.[345] Mr Armitt, Chief Executive of Network Rail, told us that the cost of TPWS was £575 million.[346] He said that. "the original proposal would have saved approximately 60 lives over 25 years and the consequence of what we have done, the full implementation, is 65 lives over 25 years. The extra five lives have cost the difference between £190 million, which was the original estimate, and £575 million [i.e. £385 million]."[347]

200. The original cost of the project escalated.[348] Dr Sefton denied that this was as a result of any additional HSE requirement over those to which Railtrack had agreed ("though there remained some areas of technical contention where it disagreed with the HSE").[349] However, Mr Osborne, who was directly involved with the TPWS installation during his period as Director of Rail Safety, was quite clear:

350

Mr Osborne influenced the granting by the HSE of exemptions to fit TPWS, but this did not prevent 'significant costs' being incurred.[351]

201. We are not convinced that the HSE's approach to TPWS demonstrated a reasonable balance of costs and safety. Further, we were appalled to see so profound a disagreement between Network Rail and the HSE over the extent of the TPWS system to be fitted on the network.

202. It is vital that the Government, the rail industry and the HSE learn lessons from this episode if the spectacle of such disagreements are not to weaken public confidence in the safety regulation of the industry. There needs to be much closer collaboration between the safety regulator and the industry to ensure consensus on the relative costs and benefits of rail safety systems so that demonstrable and significant safety gains are made for the travelling public and value for money can be achieved .

Safety research

203. The Rail Safety and Standards Board, the industry leader on safety matters, railway standards, and rail safety research, is nevertheless funded by the Strategic Rail Authority via Network Rail until 2006. The Government made £76 million available after the Ladbroke Grove accident to spend on rail safety research over 5 years, and the Board has stewardship of that sum.[352] Thereafter, we were told, funding will be the industry's responsibility.[353]

204. We asked Mr Porter whether the HSE, as safety regulator, might not undertake the work of the Board too. He thought not, pointing out that "We are a member driven body and the HSE is very much a regulator."[354] It was our impression nevertheless that more could be done to co-ordinate the work of the Board and the HSE in the field of rail safety research. The Government funds rail safety research but, in effect, hands the money over to the rail industry to spend. We think that the Government needs to examine the use that is being made of such research funding against the overall objectives for rail safety to ensure that these sums are being put to best use. For example, we wondered why the function of rail safety research could not be taken into the Railway Inspectorate and those sums administered directly in order to promote research which would then inform policy directly for improving rail safety regulation.

205. We were also told of the Board's work to review Railway Group Standards[355] "alongside Network Rail standards and indeed the European technical standards for inter-operability".[356] Dr Sefton said that the standards were "difficult to comply with" and that they proved a "barrier for many people in the industry to getting on with their job."[357]The rationalisation of the numerous railway technical standards is an important initiative. Lord Cullen considered that maintaining distance between the safety regulator and the those setting industry standards was important to maintain the integrity of the regulator's independence and also to prevent standards becoming prescriptive "detailed regulations" rather than "goal setting in nature".[358] We are not in a position to say definitely whether more could be done to further the rationalisation of Railway Standards were the responsibility for these to be taken up by the safety regulator. However, it appears that railway standards may have become over-complex and rigid and that further streamlining is necessary. We would expect the Government's present review of the railway's structure to address this issue.

Conclusion

206. The highly disturbing evidence of our inquiry is that the rail industry's confidence in the HSE is at a low ebb. The industry considers that the HSE does not understand the railway business it is seeking to regulate and may be inflating the costs of safety at a time when the evidence is that rail is a safe mode of travel, and has been getting safer. We were told that the HSE is unimaginative in overhauling its approach to risk in the light of historically low risk levels on the railways, and is institutionally complacent about improving its internal procedures.[359] Mr Alan Osborne expressed fears that unless safety costs take full account of the economic circumstances of the railway, then its continued viability may be in doubt.[360]

207. Lord Cullen's inquiry into the Ladbroke Grove accident called for "a person of outstanding managerial ability, not necessarily with a railway background" to head Her Majesty's Railway Inspectorate.[361] It is a matter of profound concern to us that the person appointed to the key post of Director of Rail Safety, leading the Inspectorate in the HSE, left his post after less than a year with the clear perception that the HSE's regulation of the railway industry was so flawed as to require it to relinquish its responsibility.

208. We heard of fears that the already complex mass of safety regulation, and overlapping interests of various bodies, would be compounded by the imposition of a European safety regime and a European Rail Agency. The evidence of the HSE sought to be reassuring about this,[362] but that message appeared to carry little reassurance to the industry.[363] Astonishingly, we were told the HSE's own research suggested that industry stakeholders are poorly informed about its roles and responsibilities,[364] calling into question its ability to communicate effectively with the industry. This is a disturbing picture of a dysfunctional safety regulator.

209. Safety measures must take proper account of the ability of the industry to pay for them, and be clearly proportionate to the risk. The justification for such measures must be based on cost benefit analysis principles that are agreed across the industry, and such calculations need to be shared between the relevant bodies. Benchmarking to establish best practice needs to be extended to other transport modes, and not only to apparently inappropriate comparators such as the nuclear industry. It is apparent that the HSE has lost the confidence of the industry. It should be a priority for the Government's review of the railways to consider whether Her Majesty's Railway Inspectorate should be removed from the HSE and either made an independent Agency of the Department for Transport, or merged with the new unified rail delivery body we propose in the final Chapter.

315 The establishment of an independent body to investigate railway accidents was a recommendation of Lord Cullen, The Ladbroke Grove Rail Inquiry, Part 2 Report, recommendations 57-73, pp 177-179. The Railways and Transport Safety Act 2003 provided the statutory foundation for the Rail Accident Investigation Branch. Back

316 The establishment of the RSSB is the product of recommendations by Lord Cullen's inquiry, Ladbroke Grove Rail Inquiry, Part 2 Report, recommendations 40-56, pp 175-177. It was established on 1 April 2003, ORR Annual Report 2002-2003, p 19 Back

317 FOR 77 Back

318 National Statistics, Transport Statistics Great Britain 2003, Table 4.15 Back

319 Health and Safety Executive, Railway Safety 2002-03, sections, 'Incidents involving passengers, staff and members of the public' and 'Trespass and vandalism ( or route crime ) ', no pagination Back

320 Everyone's Railway, the wider case for rail, p 7 Back

321 'Runaway wagon kills rail workers', BBC News On Line, 15 February 2004 Back

322 Definitional problems also highlighted by the Chief Executive of the RSSB, Q1799 Back

323 FOR 124. Lord Cullen was concerned that the railway inspectorate should have sufficient numbers with rail expertise, The Ladbroke Grove Rail Inquiry, Part 2 Report, para 9.55. Evidence received from the HSE was that at December 2003, of 123 railway inspectors '55 were Railway Specialist Inspectors, paid a premium for their railway knowledge and experience.' FOR 122B Back

324 FOR 124 Back

325 Q1649 Back

326 Q1650 Back