Transport Economics is a tricky business
Car dependency needs addressing. Sitting is the new scourge etc.
More and more stakeholders are interested in the health impact of policy and strategy decisions across all government depts, and beyond – so called Health in all Policies. This is true in the transport sector.
Folklore says the standard DfT model is predicated on a premise of road capacity is needed to address congestion thus roads always win out. Certainly to an extent. Its this sort of stuff that determines the budget for active travel from which interventions can be funded. The basic issue = the standard DfT model says “build roads to ease congestion, it’s the most efficient thing to do”. Thus we allocate the transport resources to build roads, build roads then we don’t ease congestion and don’t achieve any modal shift.
Going upstream, building a model that leads to different approach
So -going upstream, some research focused effort on the economics and supply issues inherent in this may help. This is compounded by a lot of the stuff I’ve seen of late on cycling coming from DfT is focused on safety and individual level – lets issue green tabbards etc.
This needs to be somewhat more sophisticated than an ROI toolkit that concludes cycling is good for you etc… What’s needed is a lot more nuanced and sophisticated and probably (definitely) needs a HMT perspective.
Resource wise, basically the need is, as I see it, to fundamentally shift govt thinking about allocation of transport budget away from the “build more road to ease congestion model”.
View it as a transport / health ROI toolkit.
This is my effort to describe the sort of stuff that might get attended to in construction of alternative economic modelling scenarios around non car / non road building futures.
It’s not neat nor is it pithy ROI toolkit.
I accept some more thought is needed on how best to play all this stuff into an economic model and an argument about ways to reduce congestion that are not about building more roads. We’ve been hearing the “build more roads” is solution to congestion for about 40 years now. We maybe need a different approach to the economics.
There are five main themes.
1a) The overall approach to ROI needs careful consideration
Obviously we need to get into how we redefine ROI calcs. And how schemes are prioritised on the back of ROI calcs. Obviously current model prioritise devolved transport schemes based on GVA (wages*people) impact alone. Obviously that fails to include their environmental and health and well being impact (obviously interlinked). Maybe a policy and economic effort to monetise health and well being benefit would be worth it. Maybe we should get into industrialising SROI methods (yes I’m dubious of methodology but means / end etc).
The standard economic model used by HMG almost certainly over values shaving short times off car journeys and this is at the expense of under value health benefit of active travel at population scale). My take is the net effect of this is that standard economic model favours road building, which becomes supply led and ends in more cars, thus more disincentive to cycle or walk.
What do we know about the DfT approach to economic modelling
Understanding how he current economic model works is key. My understanding of the key points is thus:
• Transport Appraisal Guidance (TAG) which the DfT use to assess major schemes is very much focused on the Treasury’s Green Book 5 Case Model, placing a significant weight on the economic impacts of investment. In broad terms, this places a strong emphasis on established monetary values for transport user benefits, such as the £’s associated with journey time saving and reliability as well as individual user costs such as reduced fuel costs and personal productive time benefits. It must be noted that there has been a push for wider economic impacts and environmental savings to be included in transport scheme assessment, and although these are considered, the weighting of the Benefit Cost Ratio is established to focus on the transport user benefit. Health impacts are included as part of this assessment but again, have a lower status in the case making given the financial assigned to such benefits. Hence the reason a highway scheme will always demonstrate a high BCR whereas walking and cycling will fall short.
• Local assurance frameworks could alter this, although the treasury to some extent prevent this as they dictate that local assurance needs to relate the TAG (point 1). So the circle continues.
• The prominence of key issues such as the Clean Air Zone which carry large fines. This in turn will shape the BCR process as a do nothing will incur cost. This cost could be classed as a benefit if something was to be done. Basically, if a scheme costs £5 but only delivers £3 of benefit that wouldn’t demonstrate good value for money. However, if by not spending that £5 you incur a cost of £10 (fines/sanctions etc), the benefit should be actually £13. I think it is this element which needs to be picked up in strategic decision making.
• The base case manual (TAG) is very bad at focusing investment in areas with high deprivation giving an uneven spatial dimension to the issue. BCR’s will always be higher in areas of buoyant economic activity therefore investment in areas of high poverty (most likely to have poorer health) places another constraint on selecting which schemes to take forward and potentially isolating those most in need. Bicycles are great social levellers – according to research, mass cycling could increase mobility of the nation’s poorest families by 25%.
1b) Tackling “The economy” question. This will all be “harmful to the economy”.
Nope. But needs debunking.
1. There is a narrative on this, but rarely any data. London has found (Business backs congestion charge) that there is business support for the actions of the London Mayor around the introduction of the congestion charge (arguably the most contentious policy area) from businesses after initial resistance. It is possible that any resistance is related to how the narrative is framed and developed. Half felt it to be working, one in seven to be failing. 71% of the businesses surveyed felt the scheme had no discernible effect on their bottom line. This isn’t an argument for this policy per se, but a reflection on the views of businesses after the implementation of one of the more contentious aspects of air pollution control. There may be issues about beliefs vs evidence, for example with shops and bike lanes. Business always think bike lanes will reduce shoppers as people need to drive to shops. The evidence is that the opposite is true but it doesn’t change peoples minds.
2. The EPA in the USA (Clean Air Act and the Economy. https://www.epa.gov/clean-air-act-overview/clean-air-act-and-economy) have considered the question of the economic benefit of a clean air act at some length – benefit from addressing health related issues, clean air and growing economy are symbiotic, sustainable technology and related industries itself a driver for economic growth. As ever people have completely different word views on this…. And doubtless US Treasury more influenced in a traditional way?
3. This RAND study (AQ related- http://www.rand.org/pubs/research_briefs/RB9952.html ) considered links between actions to improve air quality and economic growth. Pittsburgh – ex steel town. Improving AQ has costs – see as cost vs investment. RAND report noted concern re improving air q having a detrimental effect on economic performance – by being see as “anti business” and “anti car” Finding that Meeting national AQ standards would improve health, less CV events (the principal risk in UK epidemiological context), less repiratory disease – (the principal risk as noted in this RAND report), fewer work lost days (on account of direct work loss of worker being ill, and childcare as result of children being ill). This is quanitifed. Benefit quantified as $128m for reductions in O3 from 2012 level to national standard; or $448m for reducing PM2.5 from 2012 level to national standard. Suggestion that many do vote with feet in terms of location decisions – individuals and businesses. Little hard empirical data. There IS a business impact. If an area is not within the AQ standard – businesses in specific regulated industries might find it more difficulr to locate or grow in that area. Extrapolations from a national study to Pittsburgh – being out of AQ standard led to 8 fewer new business establishments in Pittsburgh within regulated industries. Being out of standard for PM25 and O3 associated with 1,900 and 400 fewer jobs respectively; and $229 / $57m lower output from those industries. Limitations of the model explored. Turning this on its head, if AQ standards ARE met, businesses in regulated industries will find it easier to locate in Pittsburgh – 8 more establishments, 1,900 more jobs / $229m for O3 being met, 400 more jobs / $57m if PM2.5 met.
4.But building bike lanes will kill passing trade for businesses.
Increased levels of cycling can bring benefits for everyone, whether they cycle or not. Bicycles take up far less road space than cars and emit no toxic fumes. They’re good for our high streets: Every Study Ever Conducted on the Impact Converting Street Parking Into Bike Lanes Has on Businesses. This is an annotated, chart-filled review of 12 studies from around the world. For example on New York streets where cycle lanes were introduced average trade rose by a quarter.
5. Making an economic case always trumps the “health” case. Maybe. My stock line here is to go upstream and change our understanding of the economy. It is not “what business does” but the sum total of everything. And we all know the weaknesses of the GVA measure as the indicator of economic buoyancy. It is easy to make a case that the value of health benefits needs to be more highly weighted in these assessments. The costs of active travel is underplayed and should somehow take into account the overall saving to the NHS/service providers (and the economy – see here for notes on the link between healthy population and the economy) if you get someone to walk and extra 10 miles a week etc. These benefits are not included in the modelling.
1c) getting into the issues re cars / road capacity / congestion economics
Im told that the cycling budget for England is about equivalent to the budget for refurbishing 10 miles of the M62……Do we wonder why there’s never any investment in active travel infrastructure…… its cos the standard model comes up with answer of “build roads for cars”
1. This INRIX piece is excellent – Traffic Congestion to Cost the UK Economy More Than £300 Billion Over the Next 16 Years… http://inrix.com/press-releases/traffic-congestion-to-cost-the-uk-economy-more-than-300-billion-over-the-next-16-years/
2. Building more roads doesn’t seem to solve the problem it promised to solve. It’s a bit like building more hospitals? – you can never have too many, but too many give adverse consequence. Effects of new urban motorway infrastructure on road traffic accidents in the local area: a retrospective longitudinal study in Scotland –http://jech.bmj.com/content/70/11/1088.short?rss=1
3.The classic research on whether expanding (trunk) roads caused more driving is from SACTRA, summarised here http://www.bettertransport.org.uk/roads-nowhere/induced-traffic – seems horribly hard to measure conclusively, and contexts vary, but it seems to conclude that broadly more roads = more driving.
4. Economic effects from driving vs cycling studied in Copenhagen – exercise in calculating the externality of car travel. One KM by car costs society about €0.17 society gains €0.16 for each KM cycled. http://www.fastcoexist.com/3046345/how-copenhagen-became-a-cycling-paradise-by-considering-the-full-cost-of-cars and here https://www.sciencedaily.com/releases/2015/05/150512104023.htm – I think they use this to justify significant investment in cycle infrastructure since it offers returns, while encouraging driving has large costs. Copenhagen has shown the most impressive recent increase in cycle rate, even from a high base: up from 36% to 45% in two years (seemingly due to major road disruption from constructing 17 new metro stations at once. A similar lesson to how Groningen started their cycle revolution by disrupting the traffic – something really cheap to invest in first: ‘no entry’ signs!
5.Addressing issues in planning flow simulations. Im told that transport planning model flow simulations simply don’t consider cycling or modal shift to cycles. So the model always assume that if one route is made less convenient to drive, another route will always be used rather than some of the traffic shifting to cycling (or probably walking either) as a mode. Apparently cycles are categorically more difficult to model than cars – not sure why.
6. There’s a mind set issue that building roads always wins transport ROI wise cos it creates most jobs. This is a good thread – https://twitter.com/i/moments/980111715198218245 and this one also a good thread –https://twitter.com/brenttoderian/status/939168127991943168esp see the middle bit – thoughts on job creation through different forms of transport infrastructure. Bikes win.
7. Im genuinely not sure why DfT continue to want to build transport infrastructure designed for energy-intensive, human-powered vehicles that will be gone in a few years (and need to go much sooner if we are to do anything meaningful about carbon emissions), when the demand for it is fizzling out anyway? See this one https://twitter.com/voleospeed/status/992042017344704512?s=21 Largest is the 30% reduction in car miles per head by young people in London over 10 years, but there is a national trend. It looks likely we really have seen ‘peak car’, and the DfT’s forecasts for traffic growth will prove wrong
Getting the model parameters is quite important also, and the underpinning assumptions/ beliefs that drive those parameters
All modelling studies are sensitive to the model parameters. Car speed is about the same as cycle speed.
It’s important to get model parameters right. I’d guess one of the key assumptions is that cars are more efficient as they get you to destinations more quickly, and thus economic productivity is better.
Hold on:- Motorists are only marginally quicker than cyclists, shows DfT modelling (and the cyclists had time added on for parking). And do cyclists really take five minutes to park their bikes? Data is here. On A-roads, motorists average 27mph but only average 11.6mph on urban streets. It’s worth noting that motorways and dual carriageways are 2% total road space, so think through on what form of road is most of that cycling happening. As per below cars are efficient vehicles for travelling between towns (as are trains!) but not so efficient within them.
Play all this into an economic model and an argument about ways to reduce congestion that are not about building more roads. We’ve been hearing the “build more roads” is solution to congestion for about 40 years now. Supply led demand? Certainly to an extent. Its this sort of stuff that determines the budget for active travel from which interventions can be funded….So -going upstream, some research focused effort on the economics and supply issues inherent in this may help.
2) The “health gain” of more folk sweating more in their travel choices
One concern I have in this space is that almost all the “health” models focus on deaths brought forward and lives lost. Understandable… but meaningless in a transport planning context (got to talk about congestion, econ productivity, AQ etc) and a broader local authority perspective (got to talk about avoidable demand for social care…. Why do people use social care – downstream of poorly… how can better transport help etc). Monetising some of the demand for services we commission (which consistently overspend) might change nature. Being explicit re monetised health benefit may revolutionise the type of schemes that are prioritised – health impacts tend to far outweigh all other considerations when measured economically.
Not being advice leads to less folk being healthy – that leads to more demand on the key pressure point for every LA – social care. So in a way its upstream preventive argument for social care demand
1. Probably worth a relook at the Woodcock model. They want to use it commercially so aren’t releasing the secrets any more. Effect of increasing active travel in urban England and Wales on costs to the National Health Service – The Lancet.. Table 5 and fig 1 are the key bits. Obviously numbers dependant on scenarios and assumptions etc, as per the normal, this is worth careful consideration
2. The Woodcock model was taken on, and adapted by TfL – calculating the “health” benefit of transport policy. TfL. TRANSPORT AND HEALTH IN LONDON. The main impacts of London road transport on health. Put a monetary benefit on modal shift. Cant recall off hand whether it was SROI – ie off the back of some measure of healthy life years gained ose something, or cash monetatry benefit. Id guess the former – SROI
TfL report on health impacts of walking and cycling – ¼ of Londoners meet their minimum daily activity targets through active commutes alone (utility walking and cycling). And walking to work can make a substantive contribution to daily physical activity, contributing to reducing preventable deaths, and improving health in the workplace.
I had a helpful chat with Lucy Sanders on it once. Basis of chat was whether it was worth repeating bespoke in a new area or gutting what’s there and making best use. My view is it’s probably not worth the effort of repeating THIS, but it might be in terms of revising the model of ROI to incorporate health return and set the health return in economic terms. This might be worth a piece of commissioned academic work?. The TfL report was in turn a gutted version of something done by the mob in Cambridge – James Woodcock and others.
3.I once blogged on it – Parks & bike lanes vs cath labs & cancer drugs – here re comparing investments in bike lanes vs cancer drugs (hypothetical exercise – may be interesting to see fleshed out with data).
4.Some helpful economic stuff in the NICE guidance on AQ. You need to dig into the appendices to get it
5. Health impact assessment of cycling network expansions in European cities. http://www.ciudadesporlabicicleta.org/2018/01/24/health-impact-assessment-of-cycling-network-expansions-in-european-cities/ & here Health impact assessment of cycling network expansions in European cities. Analysis of data from 167 European cities suggests that the length of cycling infrastructures is associated with a cycling mode share of up to 24.7%. The study also estimates that if all the assessed cities achieved this mode share, over 10,000 premature deaths could be avoided annually.
6. HEAT tool. WHO. Arguably focused on deaths not demand for services.
Framed and couched in transport planner language though, not “health” language (well so Harry Rutter tells me)
In short – Cycling is good for you (Who knew
In more detail –
this paper assesses population impact of achieving Dutch level of cycling leads to
6.5k deaths avoided
0.5y on life expectancy
3% added to GDP
7. Recently published- Ditching the car may reduce your risk of dying from heart disease and stroke by almost a third. It was in BMJ Heart so focused on CVD. Think in similar space re obesity related cancer.
8.see this BMJ study – Association between active commuting and incident cardiovascular disease, cancer, and mortality: prospective cohort study. Again there is a focus on mortality. That’s important, but of more relevance is incident disease – that’s what drives demand for public services. “Commuting by cycling and walking were associated with a lower risk of CVD incidence (cycling 50% less; walking 25% less)”. Taking these numbers, at present only 3% of the UK population commute by bike, while 36% use a car. If we increased cycling in this country to German levels by 2025, we would save £1.8bn in health benefits and £284m thanks to less congestion nationally, so approximately £180 million for Sheffield in health, £2,840,000 in productivity due to less congestion.
More folk moving more in their travel choices is a good thing. No doubts. It needs to be translated in to different languages and narratives. See the caveat at the top of section 2.
3) Various cost of illness studies associated with transport choices and or AQ
See for eg OECD
here from Lancet – Ambient air pollution is responsible for great economic losses. These losses include medical expenditures—an estimated US$21 billion globally in 2015 (ref 4 from the editorial – OECD –see above) lost economic productivity resulting from pollution-related disease and premature death, and the cost of environmental degradation. Im surprised its as low as this
4) The stuff the transport professionals twitch most about – busses / trams – needs to be factored in
Most insecure part of network…. As if it goes under, everyone gets in car and makes it all a lot worse! Etc. So the up and downstream consequences of this
5)Framing things appropriately might make a difference, maybe not to the economic analysis, but to the setting of analytic qs and interpretation of analysis
•Travel and health. 5% of trips are “health related”. Apparently. I disagree. 100% of trips are “health related” – by rote of whether we sweat or not in travel, or whether we contribute to or alleviate air quality in those trips
•Cars are good devices for moving around rural areas. They are poor devices for moving around urban areas
•We see questions of why are parks and bike lanes are an investment for health and well being, but why not roads. Framing the question explicitly in the context of park / obesity link fuels an expectation that park stuff will lead directly to a measurable impact on obesity rates. Obviously it’s not going to happen. We don’t have to justify the existence of a road, but we do have to justify the existence of a bike lane
•Transport for work (seems standard economic model) vs transport. The work aspect is hardwired in the DNA of most documents I see. I get this, it’s understandable. But people travel for all sorts of other reasons, inc many short journeys. That collectively has impact on road capacity and AQ no?
•City centre v city – Seems a reluctance to include much specific stuff in the strategy about making it easier for people to cycle and walk outside the city centre. Cynical view would be that there’s a view people in poor areas don’t cause the congestion issues – they don’t drive, or don’t drive into town for work. We ALL do take lots of short trips, for many other reasons, and thus have cars on roads. Maybe something cultural there about is it a transport to work strategy or a transport strategy.
• I was struck by how much response I had to a tweet after I saw a news item on York city centre / congestion on look North. The item was about an app. The tweet was that York won’t solve its congestion problems by apps, but by more walking + bikes + less cars.
•E bikes – where they fit – Mode of transport vs form of exercise?? All those I know that have bought one now view them with this type of framing (admittedly a small club so far)
•Parochial point to Sheffield -the hilly city myth –Looking (admittedly only eyeballing) on National Data split by town, there is no correlation between topography and %cycling. So many other factors at play. But it’s consistently used in this town as a reason why cycling is low
•But the “we aren’t Copenhagen or Amsterdam” line is also often used. 30 years ago, both were car parks. It is only as it is on account of what they have done. I’m told (never been but a mate of mine spent a year working in Copenhagen) that bike congestion is a serious problem. Nicer problems to have
•But we don’t know what to “do”. We do….. but we ignore it. We wont make cyclists safer by issuing cyclists with bright yellow tabbards. Some infrastructure needed. This review from Scottish Govt struck me as good source of what evidence does tell us on modal share. Similar from Glasgow Uni Centre for Pop Health, especially this synthesis.
We all know theres a need to commission research around economics of this that is NOT focused on individuals as targets (ie do pedometers work, does issueing cyclists with safety bibs make them safer), but focused on a much bigger and more complex environment & totality of road environment – research consider totality not just vulnerable users that focuses the transport discussion far further upstream – research exploring what going upstream means in this context that focuses the discussion on health in all policies. How is it that we use the transport budget to achieve better health
An Economic Case is supported by a Strategic Case so there is greater opportunity to play these issues through the Strategic Case. Money talks and the economic case is often where the BCR makes or breaks the investment decision.
Thus it is important that the economic case is built on sound assumptions.
Addressing all of this requires some change in the status quo, and a willingness to change the status quo.
If we are serious about shifting the emphasis of funds out of building roads, esp given DfT history, I suspect we will need to continually push on that one?
We are not going to change without structural intervention to change the approach to how we think about funding allocation, which is in turn informed by economic models.