The Local Government Technical Advisers Group (LGTAG) thank the Select Committee for the opportunity to give written evidence and are more than willing to discuss our submission further if required.
1. About us, Introduction and Summary
1.1 The Local Government Technical Advisers Group (LGTAG) is an Association of Technical Officers providing public services in and for Local Authorities. Our officers deal with public services like waste, recycling, street cleansing, parks, recreation, Planning, Council property (including housing) design, construction and maintenance, roads and transport, flooding and coastal protection etc. Our raison d’etre is to share best practice and advise central and local government bodies and representatives on problems and solutions relevant to these subject areas. Our membership is mainly senior staff in London and Metropolitan Boroughs, Counties and Unitary and District Councils as well as Consultants and Service providers where such services are outsourced to the private sector.
1.2 Local Government have statutory responsibilities for Road Safety and for managing nearly 98% of the road network but unfortunately have very little say on the remaining just over 2% of the road network and how Highway’s England’s road network affects the planning, transport, environmental or economic aspects of our towns and villages.
1.3 Of particular relevance to this Select Committee inquiry - we recently submitted evidence to the Inquiry on Major transport infrastructure projects: appraisal and delivery and have in the past submitted evidence to this committee on Road Safety Issues, The Strategic Road programme and on the effects and consequences of induced traffic from capacity increases on Strategic roads over several decades.
1.4 In this evidence we give a response to all five issues raised by the committee. The main themes of our evidence relate to the fact that the Smart Motorway programme is unlikely to provide anything but a short term solution to the congested part of the motorway network and will exacerbate congestion problems on the rest of the road network (98% of the total). Consequently, there will be adverse environmental and health consequences. Furthermore, we are concerned with the high speeds on the motorway network, safety both on and off the strategic network is/could be worsened. We believe there needs to be a much better joined up policy and delivery mechanisms for roads and overall transport policy involving all transport authorities.
2. Responses on issues identified by the Committee
On the detail issues included in your invitation to submit evidence we have the following comments:
2.1 The benefits of smart motorways, for instance to reduce congestion on busy sections
of motorway, and how necessary they are:
2.1.1 We note that in the introductory comments of the call for evidence on the stated rationale for Smart Motorways and All Lane Running of the objective (of the DfT/HE) to reduce congestion. We have consistently disputed that such Smart motorways would normally reduce congestion overall, certainly anywhere near large cities or near ‘busy sections of motorway’. We have provided hard evidence to demonstrate our position to the DfT and indeed this Committee on many occasions - at least dating back to 1988. We recognise that this fact is counter-intuitive especially from the position of behind a steering wheel in congested conditions.
2.1.2 If the committee needs further evidence of this situation, we can provide it as well as links on the LGTAG website to many related matters and submissions. In the meantime, we attach as an Appendix a peer reviewed technical paper which describes how the Smart Motorway programme is very unlikely to reduce congestion overall after a few years of extra induced traffic resulting from the increased capacity. (We would normally have provided a link to this and the more extensive work on the subject, but the World Transport Policy and Practice (WTPP) technical journal is not at present live on the web).
2.1.3 We are particularly concerned about the consequences of the hard shoulder and all lane running on the Strategic Road network on the local authority road network. It is not realistically possible to redesign and develop the rest of the road network (nearly 98% of the total) to cope with the extra generated or induced traffic from the expanded capacity and traffic volumes using the Strategic roads to reach the actual trip destinations nor is it easy to develop necessary policies and strategies to abate/reduce such other traffic as it comes off the motorways.
2.1.4 Furthermore, the extra capacity being added to already high capacity routes makes the whole network rather more sensitive in reliability or resilience terms. Anything going wrong on the motorways or immediate approaches means even more traffic has to use alternative routes which will have an even smaller fraction of the capacity to cope. It is also harder for the emergency services and motorway operators to reach the problem area and sort it out without a continuous hard shoulder. Reliability of the road network, or indeed any transport system, is far more important to industry and individuals than speeding up traffic.
2.1.5 Despite our very strong reservations about the general use of Smart motorways as an attempt to reduce congestion we do recognise that in certain limited locations it may be desirable to add to the capacity of some links. The practicalities of traffic flow for instance with an over or under bridge it may be cost effective to remove a hard shoulder for a short distance.
2.2 The safety of smart motorways, the adequacy of safety measures in place and how
safety could be improved:
2.2.1 We are concerned about the extra traffic on the motorways as described above and the potential to increase accidents resulting from the extra traffic. Stopped vehicles in running lanes are extremely hazardous and we are sure other organisations will be discussing this aspect in their evidence. Similarly, for the automatic detection and warning systems, we are also nervous about the reliability of such warning systems and the notice drivers may take of such systems. Any regular motorway user is aware that some of the other warning systems in use at present get ignored by many drivers and often the incidents needing such warnings are often historic before the warnings are turned off.
2.2.2 We understand that at least some Smart Motorways were introduced with variable signing for use of the hard shoulder only during congested periods and that they were normally signed with 40mph or 50 mph speed limits. We believe the DfT/HE has generally revised the speed limit to 60 mph during all lane running and variable hard shoulder use. A 60 mph limit is potentially 44% more damaging on impact than at 50 mph (from the kinetic energy) and 125% more damaging than at 40 mph (at 70 mph the kinetic energy is over three times that at 40mph). Furthermore, if it is desired to increase capacity, the maximum stable and reliable capacity of a dual carriageway road is likely to be at about 40mph rather than any faster. We also believe part of the rationale for the decision to increase speed limits on such roads is the ‘economic appraisal’ from predicted and calculated travel time savings based on a number of assumptions. Our evidence to you on the other live inquiry - ‘Major transport infrastructure projects: appraisal and delivery’ casts grave doubts about the methodology and meaningfulness of such economic assessments.
2.2.3 Finally, while on safety issues, while local government is not really or directly responsible for improving safety on the Strategic Road system, a common safety problem we experience is that drivers coming off high speed roads find it hard to adapt to the rest of the road network quickly. Hence many authorities experience excess accidents on their networks in the vicinity of the end of ‘motorways’. We are therefore even more concerned about further increase in traffic volumes or speeds using the Strategic road network.
2.2.4 Whether All Lane Running is the most suitable type of smart motorway to roll out or if there are better alternatives:
2.2.5 As mentioned above and in other submissions to this committee, the DfT and other organisations on numerous occasions, LGTAG is not generally in favour of increasing road capacity on the strategic road network. This is especially if such increases in capacity allow increased car commuting, extra CO2, pollution or other health problems (e.g., obesity) arising from excessive car use. We need other strategies and close working across all highway and transport authorities and providers to provide the best system for the public and industry. Any such strategies must include managing our towns, where most trips end, in the most effective manner.
2.2.6 We recognise that it would be difficult to close down quickly and almost completely the present HE program of design and construction of schemes but we would like to see the minimum further damage to our Transport system done by the present programme. In the first instance there are many locations in the country that are not well served by high quality routes and many small towns and villages with major roads flowing through them. Transferring road building resources to such problems would be a significant improvement, produce a finer network which would also improve reliability and resilience.
2.2.7 If we must increase the capacity of existing motorways the hard shoulder running in peak times only with maximum speed limits of 40mph would be less inappropriate and safer.
2.3 Public confidence in using smart motorways and how this could be improved:
2.3.1 One of the scariest driving experiences any member of the public can experience is a breakdown or being stopped in a running lane of a motorway. Punctures and breakdowns do happen and getting to a safe place freewheeling or managed driving quickly is very important.
2.3.2 As implied, we are strongly in support of lower speed limits especially at congested times; vehicle detection and variable camera enforced speed limits would be well worthwhile but preferably without abandoning hard shoulders.
2.4 The impact of smart motorways on the usage and safety of other roads in the
strategic road network:
2.4.1 Other roads in the strategic road network include:
2.4.2 In the first case the situation will be similar to that on many local authority roads where the extra traffic from smart motorways causes extra congestion and also the safety issue described above in para 2.2.3, as drivers take time to adapt to the changed driving conditions. In the second case it is unlikely that Smart motorways will have much impact. In the third case we would draw attention to the impact of induced traffic (see appendix and references) – we would not expect that the Smart motorways would reduce traffic on such alternative routes on the basis of past experience.
2.5 The effectiveness of Highways England’s delivery of the smart motorways
programme, the impact of construction works, and the costs of implementation.
2.6.1 Save for disruption and extra congestion while work is being carried out and diversion of resources from more worthwhile transport interventions, we would not wish to comment further on HE’s delivery programme.
3. Closing Comments
3.1.1 As representatives of the technical side of delivering and managing nearly 98% of the road networks for the country we hope our experiences will be of assistance to the committee in their deliberations. We are available anytime to answer questions and would be pleased to give oral evidence to the committee.
3.1.2 To conclude our main concerns are:
As published in peer reviewed technical journal World Transport Policy and Practice (WTPP). (Journal not presently available on web)
Will the government’s spending on expanding the national road network deliver anything useful? Have they properly taken into account induced traffic and extra congestion likely to be caused elsewhere?
In this paper John Elliott reminds us of earlier studies which have shown the consequences of past reviews of induced traffic and traffic reductions from expanding or contracting road space. The government seemed keen on evidence based strategies and policies and it now seems opportune to reiterate some of the hard and not quite so hard evidence available. It should be added that the core of this evidence has been given to government from 1985 onwards by the author and others; the first evidence of the extent of generated or induced traffic from roadbuilding dates from the 1920s but has appeared from various studies at regular intervals since.
Many of the road schemes being proposed by the UK government in 2015 are in the vicinity of cities and conurbations in places where it is recognised that there is serious congestion. However, all the evidence suggests that that additional road construction especially anywhere near such major urban areas is unlikely to help traffic conditions or the economy.
Pollution and environmental aspects of the government’s road policies are not discussed but are possibly an equally strong reason to challenge the present policies.
Will the government’s spending on expanding the national road network deliver anything useful? Have they properly taken into account induced traffic and extra congestion likely to be caused elsewhere?
There is very strong evidence (but not as widely known as it should be) that road building can increase traffic levels enormously within a few years of opening and is likely to cause more congestion in the area rather than reducing it. Mechanisms which can account for a substantial proportion of the additional traffic are mode and destination change. These are often modelled for larger schemes but rarely do the results of the modelling reproduce what actually happened after opening. Occasionally land use effects are modelled though usually are not; completely new trips and peak narrowing are sometimes mentioned qualitatively. Psychological and social mechanisms are not usually in the competence of traffic models.
Transport planners who want to make estimates of induced traffic can now make an attempt, however this does not appear to correlate well with observations and past experience. Any comparison can be disguised by including an element of ‘natural traffic growth’ which is usually assumed to be the same with or without a scheme.
Extra traffic induced by a scheme will inevitably cause extra congestion at the ends of the new roads (often in urban areas where congestion is at its worst in any case) from the extra traffic. Although the extra traffic and congestion is very visible and usually actually appears 1-5 years after opening, it is seldom quantified or assessed properly.
The December 1994 SACTRA report ‘Trunk Roads and the Generation of Traffic’ (SACTRA, 1994) is probably the most robust evidence and analysis of all this. The SACTRA work itself pulled on a number of different evidence sources - not only the before-and-after traffic counts but also inferences from large numbers of studies of values of time, demand elasticities, differential growth rates, and surveys. It was because all these told a broadly consistent story that SACTRA felt able to be more decisive in its conclusions than the Department of Transport was expecting at the time.
Since then, there has been one completely different type of evidence base as well: this is the complementary literature on traffic reduction following road space reductions (pedestrianisation, bus lanes, disasters, maintenance closures, etc). This is important in its own right, but also significant in giving confidence that the induced traffic effect is not an artefact due to economic growth or similar. The largest empirical compilation is a book Traffic Impact of Highway Capacity Reductions (Cairns et al, 1998) with some updating papers subsequently of which one was awarded a gold medal by the Institution of Civil Engineers.
SACTRA does show that the net effect of road enlargements could actually create more congestion. Mogridge and Thompson really propounded this in their various books and technical papers (Mogridge, 1997; Thompson, 1969; Thompson, 1977) and it certainly can apply most in big cities on radial routes. Explained simply there is a vicious circle --- more road space – more car use - less public transport use – fares go up, frequency goes down – more people transfer to cars etc and the new equilibrium point is a lower level of service in both cars and Public Transport.
In addition to SACTRA (1994) there are a number of other sources on this subject:
It is interesting to note that SACTRA was appointed and given terms of reference by a Conservative Government, which was also in place when its 1994 report was published. The author recalls the then Transport Minister Stephen Norris saying in a short speech at the time that the government would not have built so many roads if they had known about it. (Stephen Norris is aware that his statement has been repeated by the author of this paper)
While this gives robust evidence of roads generating traffic there is also substantial evidence that extra road space and higher speeds do not necessarily provide economic benefits. Reference can also be made to the 1999 SACTRA (SACTRA, 1999) report on roads and the economy which included doubts that the economic benefits could be as high as the calculated time savings.
We also know that the ‘economic benefits’ calculated largely on basis of predicted time savings are highly artificial (Elliott, 2015). The author outlined his basic criticisms of the ‘economic assessments’ in the following terms:
“--- However linking models (which have been highly calibrated or ‘adjusted’ to try and match a base year situation) to predict behaviour change with an economic assessment introduces potential for a highly distorted view of the potential benefits to real travellers or the economy.
Within the ‘economic assessment process there are the following major issues:
-The evaluation is largely based on the difference between two enormous sums of time spent on the network with and without a scheme; each of these sums is based on a large number of assumptions, the process is therefore mathematically very unsound.
-For major road schemes most of the ‘benefits’ appear for the peak traffic times (i.e., largely for car commuting – a mode and time that most highway and planning authorities do not want to encourage – it is noted that the very recent increase in the value of time for commuting is likely to further exacerbate this problem) and for the period 30-60 years in the future (where the assumptions taken are even less real).
The impact of ‘generated’ traffic compared with the so called ‘natural traffic growth’ and its impact outside the proposed scheme is never adequately considered (seldom do the models predict the level of generated traffic caused by the new road fully, also the extra generated traffic causes minor extra widespread congestion and delays outside the immediate study area, this is usually modelled without the junctions and without other traffic that passes through that area but not through the study area – delays at junctions in urban areas increase very rapidly with small increases in traffic.”
It appears that the statements of certain business people and the roads lobby, is being used to make the case for Strategic Road improvements rather than any evidence. It is an easy message to say that widening a road would appear to be a solution to congestion especially from behind the steering wheel of a car in a traffic queue! The real message on transport policy is much more complicated.
Rather than using my own words to describe the history, analysis and effects of generated or induced traffic it is perhaps pertinent to look at Goodwin’s (2006) where he reviewed the literature on this subject, gave practical information but also more than justifying his title: “Induced traffic again. And again. And again”.
Goodwin (2006) and Purnell et al (1999) noted the long history of reports of induced traffic. As far back as 1925 it was identified that the opening of a new section of the Great West Road caused a massive increase in traffic. In Bressey’s report of 1937 it stated “the remarkable manner in which new roads create new traffic”. According to the study, immediately after the Great West Road was opened in 1925 it “carried 4 1 / 2 times more vehicles than the old route was carrying; no diminution, however, occurred in the flow of traffic along the old route,”.
Goodwin (2006) refers to a number of times when newly generated traffic had been reported and its extent confirmed, even appearing in Ministry advice, but every time the issue has been ‘forgotten’ for a number of reasons. He states:
“So 1925, 1937, 1958, 1963, 1968, 1985, 1987, 1988, 1994, 1996, now 2006: for eighty years, every eight years on average there has been the same experience, the same conclusions - even, for goodness sake, more or less the same figures. The evidence has been consistent, recurrent, unchallenged by serious countervailing evidence, but repeatedly forgotten.”
This recurrence about every eight years explains why LTT’s editor suggested Phil Goodwin’s article should be republished in 2014.
In 1985 the GLC team showed every major road constructed in London between 1966 and 1985 generated phenomenal amounts of extra traffic; for example Westway corridor was carrying nearly double the all-day traffic within five years of opening and the peak flows on Blackwall Tunnel doubled in less than a year. The full details are available in Purnell et al (1999); however a later version of the 1985/6 GLC report (no longer available) did investigate modal split impacts on the M11 corridor indicating significant modal shifts from rail to car and rail heading on the tube network. It is notable also in this GLC publication that ‘natural traffic growth’ is minimal in areas without road enlargements.
SACTRA (1994) published its best known report, on what it renamed 'induced' traffic. The average traffic flow on 151 improved roads was 10.4% higher than forecasts which omitted induced traffic, and 16.4% higher than forecast on 85 alternative routes that improvements had been intended to relieve. In a dozen more detailed case studies the measured increase in traffic ranged from 9% to 44% in the short run, and 20% to 178% in the longer run. This fitted in with other evidence on elasticities, and aggregate data. The conclusion was
"an average road improvement, for which traffic growth due to all other factors is forecast correctly, will see an additional [i.e. induced] 10% of base traffic in the short term and 20% in the long term"
The CPRE report (CPRE, 2006) looked in detail at three big schemes on the A27, A34 and M65, and a further ten schemes on the A5, A6, A41, A43, A46, A66, A500 and A1033. These were schemes undertaken 8 years after SACTRA's 1994 report had been finished and accepted. The report included some telling statements:
"Careful scrutiny of the traffic flow data suggests that traffic growth after the scheme opened has been significantly higher than growth on other nearby road corridors or national traffic growth"
"…in all three case studies the current traffic flows are near or already in excess of what was predicted for 2010. In towns with bypasses, such as Newbury and Polegate, the new roads did significantly reduce the town centre traffic levels. However, these reductions are not as great as originally forecast and there has subsequently been regrowth in traffic levels on the bypassed roads. The net effect in combination with the new road is generally a considerable overall increase in traffic".
After noting the Highways Agency's own explanations for the extra traffic growth, (which were intriguingly similar to those rejected by SACTRA 12 years earlier - including wide area reassignment), the report states:
"Nevertheless, in view of the fact that many of the schemes reviewed have demonstrated significant increases in traffic volumes (in the range of 10-35%, within a period of one to two years after opening), there would seem a strong case to consider the issue of induced traffic in more detail in future evaluations".
This last published study did not include schemes near London or major conurbations. Since the GLC 1985 study there have been a number of schemes on the North Circular Road, the completion of the M25 and major enlargements of radial routes – the M40/A40, A13, M1, M2/A2, A41. As a user of most of these routes over the period since 1985 it is notable how volumes changed after the enlargements were carried out. It is also interesting to note Phil Goodwin’s comment on the M25:
“in 1988 after the M25 exceeded its long term forecast traffic growth within months of opening.”
In 2010 Zeibots and Elliott using Australian and UK evidence concluded:
“When taken as a whole, there appears to be a cycle at play where road expansion is advocated to overcome congestion; people in affected neighbourhoods object, saying they want public transport to be improved instead; governments react to public complaint; road expansion policies are put on hold and new policy directions are investigated; congestion continues to be a problem, and eventually road expansion policies creep back into government transport plans so that the cycle begins again. In light of this history, we ask why government administrations in the UK and Australia, and other parts of the world, have continued to increase road capacity as a solution to congestion when all the evidence indicates it generates additional traffic that perpetuates congested conditions?”
Many of the road schemes presently being proposed by the government are in the vicinity of cities and conurbations, in places where it is recognised that there is serious congestion. However, all the evidence suggests that enlarging roads in such places will increase traffic and is likely to cause more congestion in the area, rather than reducing it, within a very few years. Even the arguments on calculated or predicted economic benefits seem very spurious indeed.
While many Transport Planners, especially those who might read WTPP journals, are well aware of the level of induced traffic and its consequences, government should also be well aware and be adjusting policies accordingly. It should be noted that the Local Government Technical Advisers Group has outlined these issues to government on a regular basis - the last occasion was in December 2014 to the House of Commons Scrutiny Committee on the Infrastructure Bill.
Cairns S, Hass-Klau C and Goodwin P (1998) Traffic Impact of Highway Capacity Reductions: Assessment of the Evidence, Landor Publishing, London, ISBN 1 899650 10 5
Elliott, J E (and others) 2015, Modelling; misuses, mistakes and misconceptions - Local Transport Today 686 27th November – 10th December 2015.
Goodwin, P (2006) Induced traffic again. And again. And again published in Local Transport Today August 2006. (Full wording can be found on http://www.lgtag.com/index.php/news/566-tag-evidence-to-hoc-scrutiny-committee-on-infrastructure-bill see Appendix 3.
Local Government Technical Advisers Group December 2014 to the House of Commons Scrutiny Committee on the Infrastructure Bill. Please see: http://www.lgtag.com/index.php/news/566-tag-evidence-to-hoc-scrutiny-committee-on-infrastructure-bill and particularly Appendix 3.
SACTRA (1994) The Standing Advisory Committee on Trunk Road Assessment Trunk Roads and the Generation of Traffic. Chairman Mr DA Wood QC December 1994 HMSO ISBN 0 11 551613 1
Bressey, C.H. (1937) Highway Development Survey HMSO, London.
SACTRA (1999) Standing Advisory Committee on Trunk Road Assessment Transport and the economy (http://webarchive.nationalarchives.gov.uk/20050301192906/http:/dft.gov.uk/stellent/groups/dft_econappr/documents/pdf/dft_econappr_pdf_022512.pdf)
Zeibots, M and John R. Elliott (2010) Urban road building and traffic congestion: What went wrong, World Transport Policy and Practice, 17, 3, 6-26
The author’s thanks to Phil Goodwin for his advice on the understanding of the workings of the 1994 SACTRA review and the use of text from his 2006 article in Local Transport Today.
John Elliott BSc, CEng, MICE, FCIHT, MCMI. Vice Chair National Transport Committee and Communications Officer for the Local Government Technical Advisers Group and Independent Consultant.