Evidence roundup - the importance of infrastructure in enabling cycling

For this week's roundup we've compiled another of our 'one off' specials - we've previously looked at the evidence for shopping on the high street and how designing for walking and cycling actually boosts business.

This week we are looking at the evidence for the importance of cycling infrastructure, particularly when it increasing cycling levels, and the effectiveness of different kinds of interventions.

Barriers to Cycling

Perhaps the most important starting point when we come to consider these issues is to examine some of the fundamental barriers to cycling uptake. While there are undoubtedly many reasons why people feel unable to cycle for short trips in the UK,  the main barrier to cycling in this country is the perception that our roads are too dangerous and unpleasant, largely due to high volumes and high speeds of motor traffic. They feel subjectively unsafe, and indeed are objectively unsafe for anyone travelling by cycle.

Perhaps unsurprisingly, there is now a large literature body supporting this. The extensive 2011 Understanding Walking and Cycling Report found that

it is clear that traffic is a major deterrent for all but the most committed cyclists. Potential cyclists, recreational (off-road) cyclists and occasional cyclists are discouraged from using their bicycles for everyday urban journeys because of their fear of cars and heavy goods vehicles

'Poor safety' was the main reason expressed for not cycling by 80% of the respondents in the report.

There is Transport for London research which matches these findings. Their 'Attitudes to Cycling' reports consistently reveal that safety worries are the main barrier to cycling uptake -

Safety concerns remain the key deterrent to cycling, far more so than concerns about lack of fitness or cycling ability

... with 80% of those surveyed citing safety as their main deterrent. This is matched by TfL's Analysis of Cycling Potential, which found that

For all groups, including frequent cyclists, safety was the most significant barrier to cycling in general and for specific trips. This suggests that, in order to realise the remaining potential from existing frequent cyclists, practical measures to increase safety and improve the provision of facilities will be the most effective.

We also have national research in the form of the British Social Attitudes survey, which consistently finds that (again) the vast majority of people think the roads are too dangerous for them to cycle on, and (significantly) that this is strongly age-dependent, with elderly people much more likely to express this belief. And this is backed by a 2011 Department for Transport study, Climate Change and Transport Choices, which revealed two-thirds of people who are able to cycle agreed that 'it's too dangerous for me to cycle on the roads', with nearly half expressing complete unwillingness to cycle on the roads.

In both these studies, people did express a willingness to cycle - either agreeing they could cycle for short trips, or expressing a belief that they are the kind of person who could cycle.  In other words, a reluctance to cycle on the roads is not simply down to hostility to cycling; rather, demand for cycling is being suppressed by hostile road conditions. Attitude surveys in a 2002 DfT report revealed that

one of the main reasons for people dismissing cycling as a genuine form of transport was fear of actual and perceived road danger.

There is of course a whole host of other local/city surveys, along with studies carried out by organisations like the London Cycling Campaign and Sustrans, which have very similar findings - that the actual and perceived safety of cycling in motor traffic is the most significant barrier to cycling uptake. These can all be found on the Embassy 'Barriers to Cycling' page.

As we shall see in the next section, this barrier can be more significant for different sections of the population, particularly women, children and the elderly.

2) Preference studies

Traffic danger is clearly the most significant impediment to cycling uptake in the U.K. The question then moves to what kind of interventions best address that danger - what kinds of environments might we expect people to start cycling in. These can be described as 'preference studies' - which (to simplify) involve asking members of the public what types of roads and/or cycling facilities are most preferable. Again, there is a broad academic consensus that people prefer to cycle in environments that insulate them from motor traffic, and from motor traffic danger.

2012 research for Transport for London, carried out by Steer Davies Gleave, found strong preferences for cycling in low traffic environments, or away from motor traffic altogether. These preferences were more strongly expressed by women, and by people with less cycling experience. This is matched by a 2012 study in New Zealand which found that both cyclists and non-cyclists alike 'prefer less interactions with traffic and riding in safer conditions', with a strong preference for protected cycleways for entire journeys.

A 2017 systematic literature review has found that, while preferences for greater separation from motor traffic did vary according to gender and age, these differences were 'quantitative rather than qualitative' and that ' no group preferred integration with motor traffic' -

We have found good evidence that women express stronger preferences for greater segregation from motor vehicles than men. This is within a context of similar overall types of preference, that is, typically very similar hierarchies of preference across genders. As stated by Misra, Watkins, and Le Dantec (2015): “Riders across all cyclist types prefer dedicated cycling facilities and are opposed to high speed traffic and high volume traffic, with little variation based on the classification of the cyclist”. In terms of age, again, there is an overall qualitative similarity between groups, but with some evidence suggesting that older people may have stronger preferences for separated infrastructure.

When it comes to cycling with children, research suggests people prefer 'segregation by kerb, segregation by car parking, shared park routes, and filtered streets' - all of which remove, or greatly reduce, interactions with motor traffic.

At the other end of the age scale, the Cycle BOOM project found significant fear of traffic danger amongst its elderly participants, with particular concerns about sharing space with motor traffic, especially with buses and HGVs. This project also found that elderly people are much less able to perform the standard 'coping mechanisms' used to negotiate the road network by able-bodied and more confident cyclists, such as looking over the shoulder before moving into the centre of the road.

We also have surveys carried out by Cycling Works London, showing that Londoners have a very strong preference for fully protected infrastructure, as opposed to mere painted lanes, and from Halfords which revealed 40% of Britons would cycle more if there were dedicated cycle lanes on every road.

Finally, and hot off the press, we have the Department for Transport's Cycling and Walking Investment Strategy research, which finds that

Off-road or traffic free cycle routes have an important role in changing attitudes of non-cyclists... Other strongly preferred routes include those with substantial physical separation along roads (e.g. with hedge or kerb separation), and on very quiet streets with little or no motor traffic. Evidence8 on differences by age and gender suggest under-represented groups have particularly strong preferences for separation from motor traffic.

So it is important to note both that

  • Across all age groups and gender, people express strong preferences for greater separation for motor traffic.
  • The fact these preferences are stronger amongst women, children and the elderly mean that reducing both actual and perceived danger in the form of exposure to motor traffic is critical to increasing cycling inclusivity.

3) Uptake

But what happens when infrastructure is actually built? Does it actually get used, and does it lead to more cycling?

Perhaps the best piece of evidence in the UK is a study of cycling uptake following construction of a cycleway as part of Cambridge's guided busway, which found that 85% of the increase in cycling was due to the new infrastructure, with the largest effect on physical activity being seen in the people who were least active before the busway opened. There's similar evidence from a study that monitored cycling and walking uptake alongside Sustrans routes, showing that 'at 2-year follow-up residents living 1 kilometer from the new infrastructure reported a 45-minute increase in walking and cycling per week relative to those living 4 kilometers away'.

The school that has the highest level of pupils cycling to school anywhere in the UK just happens to have a good quality cycleway connecting to it, allowing its pupils to cycle to ti without having to deal with motor traffic - Cherwell School in Oxford has a near 60% cycle to school to rate.

There's evidence that a new path in Glasgow (again, motor-traffic free) has shifted people away from travelling by car and by bus, to walking and cycling. Seville is perhaps one of the most famous cities in the world for implementing an extensive (albeit of mixed quality) cycle network, with previously very low cycling levels. The evidence suggests that infrastructure has led to a large (and cost-effective) increase in cycling uptake.

The city of Calgary in Canada has also recently implemented a quick, rudimentary network, which saw a 95% increase in cycling levels on the network after just three months, and across the United States the amount of cycle commuting in large cities correlates strongly with the amount of cycle paths and cycle lanes, whereas weather and public transportation availability did not seem to affect cycling rates.

It's very early days for the new, high quality cycling infrastructure in London, but TfL monitoring suggests that even in just the first five months after opening, there has been more than a 50% increase in the number of people cycling on these roads, with both corridors now moving 5% more people at peak times than they were before.

A study of cities - focusing particularly on the Netherlands - found that, unsurprisingly, good quality design for cycling does indeed encourage more cycling. More generally we can point to what successful cycling cities and towns have in common, notably 'dedicated, fit-for-purpose space for cycling' in the form of either protected space on main roads, quiet streets with low speeds limits and low traffic levels, and motor traffic-free greenway paths away from main highways. Finally the paper 'Making Cycling Irresistible' found that

The most important approach to making cycling safe and convenient in Dutch, Danish and German cities is the provision of separate cycling facilities along heavily travelled roads and at intersections, combined with extensive traffic calming of residential neighbourhoods.

It's also worth noting that infrastructure can reduce gender and age inequality in cycling uptake. A (small-scale) study by Rachel Aldred has found that protected infrastructure may 'contribute to improving the gender balance of cycling, where this is poor', while in countries with extensive cycling networks that minimise interactions with motor traffic like the Netherlands more trips overall are made by women than by men, and cycling is broadly distributed across all age groups.

4) Limitations of other interventions

Finally, we also need to place environment-based measures - filtering, building protecting infrastructure, and generally creating stress-free, comfortable cycling environments - in context, alongside other interventions aimed at promoting cycling, in particular, cycle training.

Cycle training should not be seen as a substitute for infrastructure. While Bikeability training can and does increase confidence, a study has found 'no evidence of short-term effects on cycling frequency or independent cycling.' This is backed up by the 2015 Bikeability Hazard Perception Report, which found that Bikeability training does give children more confidence to negotiate roads, but doesn't appear to lead to any increase in cycling frequency among children.

In other words, cycle training is beneficial, but it does not have any substantive effect on overall cycling levels. Some 2 million British children have received Bikeability since 2007, yet there has been no appreciable nation-wide change in cycling to school levels. Put simply, lack of training is not what is stopping people choosing to cycle.

More generally, it is vitally important to make the distinction between 'Nice to Have' measures and 'Must Have' measures. A cycle network that allows anyone to make journeys in safety and comfort, and with minimal inconvenience, is a 'Must Have'. Without this fundamental intervention, 'Nice to Have' measures will be fundamentally ineffective.

This is explained at length in Enabling Cycling Cities: Ingredients for Success, where 'Nice to Have' measures include all those things that are politically easy to implement, and give the appearance of action, even being visible, yet will not make any significant difference to cycling uptake -

Well intentioned decision makers and activists frequently promote actions that make cycling more comfortable to people who are already riding bicycles. These actions are often nice to have and include printing maps, installing parking racks or bicycle lockers, hosting workshops, mounting racks on buses, painting lines on the pavement of roads with traffic speeds over 40 Kph, signage, and more. They are easier than “must haves” as they usually do not create much conflict, are visible and create the sense of doing something. While the impact of these actions can depend on the context, they typically do not result in many new cyclists on the road, as witnessed by experiences of hundreds of cities.


Furthermore, these measures alone can run the risk of setting plans back when skeptical stakeholders claim that they should eliminate new investment of cycling infrastructure as “it’s evident that we do not have a cycling culture as the number of people riding bikes has not substantially increased following the actions taken.” “Nice to have” actions should be part of a comprehensive strategy that first focuses on “must haves” and seeks to increase and broaden the community of cyclists. Combining them with other initiatives can create a multiplier effect.

We can point to the record of Britain as a whole over the last twenty to thirty years, which has engaged in 'nice to have' measures like signage, training, promotion, maps, painting lines on pavements, and so on, without any significant effect on cycling rates at a national level - indeed, with stagnation and decline in most areas, masked by increases in cities which have started to engage in 'must have' measures.


If there is any evidence you think we may have missed, or that is vitally important, please add it below in the comments. We'd like to compile a substantial evidence-base on this topic!


Great stuff. And sorry as I may have overlooked these when skimreading the excellent article. I find these studies more than useful:

Aldred, R. (2015). Adults’ attitudes towards child cycling: a study of the impact of infrastructure. European Journal of Transport and Infrastructure Research

Preference and intercept
Monsere et al. (2014). Protected Lanes - lessons from the green lanes : evaluating protected bike lanes in the US.  Report NITC-RR-583

Street measurements
Furth, P. G., & Mekuria, M. C. (2013). Network Connectivity and Low-Stress Bicycling. Presented at the TRB for the 2013 Annual Meeting, Transportation Research Record.

These quotes are most essential imo

Pucher & Buehler (2012:351)
No city in Europe or North America has achieved high level of cycling without an extensive network of well-integrated bike lanes and paths that provide separation from motor vehicle traffic. […] Separate cycling facilities are a crucial first step towards increasing cycling and making it socially inclusive.
Pucher, J. R., & Buehler, R. (2012). City cycling: MIT Press.

Pooley et al (2013:176)
It is clear from our research that most non-cyclists and recreational cyclists will only consider cycling regularly if they are segregated from [motor vehicle] traffic
Pooley, C. G. et al. (2013). Promoting Walking and Cycling : New Perspectives on Sustainable Travel. Bristol: Policy Press.