Monday 15 October 2012

20 mph zones do not increase the number of cyclists

(Above) Cycling friendly? Streets in 20 mph zones in the London Borough of Waltham Forest and Norwich.

A recurring characteristic of vehicular cycle campaigning is to cherry-pick an aspect of Dutch cycling success and then misapply it to British roads in the form of a syllogism. For example: There are many Dutch cyclists and Dutch cyclists are very safe therefore if we can just get many people cycling in Britain then cycling will be very safe here too. This, of course, is the ‘safety in numbers’ thesis, which remains at the heart of the CTC’s campaigning and which has been repeatedly debunked by As Easy As Riding a Bike, most recently here and here.

But a close second is the thesis that speed reduction is the golden key which will unlock all that suppressed cycling out there. At its most emphatic it is asserted that speed reduction, and that alone, is the ONLY thing that will redress the balance and give people confidence to cycle on the roads. 

The CTC argues that Lower speed limits are also linked with increased cycling and walking. In the Netherlands 30kmh (18.5mph) covers 75% of the residential street network and is deemed a safe speed for cyclists, pedestrians and light vehicles to mix. The objection to that analysis is that in the Netherlands cyclists do not mix with ‘light vehicles’ under the same conditions as British cyclists. Traffic volume is only a fraction of what a cyclist can expect to encounter on a British street and the lay-out of the street is very different, with rat-running prevented and a range of measures giving cyclists clear priority.

The danger of cherry-picking the Netherlands commitment to low speeds in residential areas, noting the high rate of cycling, and then applying it to Britain is shown by this recent example:

‘Cycletopia’ - an imaginary town depicting 15 real life examples of promoting, protecting and inspiring cycling in the UK - has been created by cycling charity, CTC, and one of the key parts of Cycletopia is the 20mph zone, inspired by Lancashire County Council’s 20 mph limits. 

CTC Cycling Development Officer in Lancaster Damian Bonsall said: “Lancashire is powering ahead with introducing a speed limit of 20 mph to residential areas over the whole county. This is a fantastic example of a council spearheading change on our roads and it is a speed at which most people feel safe to mix with motor traffic when cycling. 

In Britain? Oh really? Where is the hard evidence for that? None is cited and as far as I am aware there is none, and the CTC has never supplied any. This involves either the misapplication of the Dutch example to the utterly different conditions to be found on car-sodden British urban streets or it is wishful thinking masquerading as fact - something the UK cycle campaign industry is very good at.

It’s important to make that distinction between the fact that in KSI terms 20 mph zones are objectively safer for cyclists and the reality that KSI statistics [i.e. annual figures for those killed and seriously injured in road crashes] are not in themselves a good index of actual danger. In any case, the crucial issue is not statistics relating to deaths and injuries but whether or not cyclists feel subjectively safe in a 20 mph zone, and whether or not those zones provide cyclists with convenient, direct routes which are not subordinated to the convenience and greater priority of motor vehicles. As a resident of the London Borough of Waltham Forest I have good reason to be sceptical of such claims. I certainly don’t feel safe or relaxed cycling in local 20 mph zones, because I have to share them with psychopathic rat-runners in white vans and minicabs – drivers who cannot bear to be behind a cyclist even for five seconds, and who aggressively seek to overtake in highly unsuitable circumstances. Indeed, most of my worst experiences locally have been on so-called ‘quiet routes’.

Waltham Forest has plenty of 20 mph zones across large networks of residential streets. It also has a cycling modal share of less than one per cent. If it was really true that 20 mph zones unlocked cycling’s potential then one would expect a significantly higher modal share and a clear trend of increased cycling, year after year. No such trend has ever been identified by the council’s cycling counts, which indicate at best stagnation and at worst shrinkage. I have argued all this before, at some length.

Unfortunately the false claims for a connection between 20 mph zones and increased cycling has now become accepted wisdom, so much so that it is even embedded in transport planning. The Northampton Town Transport Strategy states Inside the inner ring road many [cycling] routes cross the town centre, which is a safe cycling environment as it has a 20 mph limit. In one sense this is specious reasoning (a cyclist travelling at 20 mph who is hit by a car coming the other way at 20 mph will be suffering the force of a 40 mph impact, unlike a pedestrian). I once came across the aftermath of a spectacular crash in a local 20 mph zone. In the past year another driver ‘lost control’ at almost the same location, crashing into the house on the corner of Grove Road and Eden Road E17, causing major structural damage. And once you start looking, there’s no shortage of evidence of crashes in 20 mph zones which, if they involve no injury, will never be officially recorded as a “road accident”.

But even if it was true that a 20 mph limit guaranteed a safe cycling environment it doesn’t automatically make cycling attractive. Even though central Northampton offers cyclists a reduced chance of death and serious injury, the town’s low modal share for cycling indicates that this is not enough to produce a surge in cycling. Nor is it likely to when around £150 million is about to be spent making travel by car in and around Northampton more attractive, with increased car parking space in the city centre (see previous post, below).

But there is one city which is currently being trumpeted as the great cycling success story, where 20 mph zones are concerned:

Slowing speed limits from 30 mph to 20 mph contributes to increasing cycling and walking by up to 12%.  

Bristol City Council report that walking and cycling increased by up to 12%

I never trust statistics of this sort, which lump cycling and walking together. Modal share for walking anywhere tends to be quite robust and subject to little change. People will walk because they have to or because it is more attractive to walk than to use another mode, and nothing that local authorities can do will stop them. Thousands – probably tens of thousands - of car owners walk every day to Leytonstone, Leyton, Walthamstow and Blackhorse Road tube stations, because parking is not available there or because it is but would add significantly to the cost of commuting into central London. Cycling is an altogether more fragile mode, probably because it is a far more frightening and dangerous mode, even though I know a certain sort of cycle campaigner is fond of proving the opposite with statistics. The fact is walking never feels as dangerous as cycling. That’s because as a pedestrian you are segregated from motorists. The experience of cycling in Britain is, I would argue, far more stressful, inconvenient and unattractive than other transport modes, which is why of all them it has probably the highest rate of ‘churn’.

The fragility of cycling, even in a country which is the second most successful cycling state in Europe, is underlined by a little-noticed update to one of David Hembrow’s Copenhagen posts, where he notes

A 15% decline over two years in the busiest street in the city

part of a wider, long-term trend in which

cycling dropped from between 18 and 19% to 16% of journeys within Denmark. 

The reason for this decline seems fairly obvious. Danish cycling infrastructure is inferior to Dutch cycling infrastructure because it fails to prioritise cyclists in the same way and because it involves far more subjectively unsafe vehicular cycling.

In the case of Bristol, the Twenty’s Plenty website is plainly misrepresenting the results of the Bristol study. If you bother to look at the fine detail of the Bristol analysis you will find this:

Overall, the number of people that said they never cycle has remained constant at around 66% (inner south area) and 60% (inner east area). [Section 8.13]

So, if the questionnaires are to be believed, being included in a neighbourhood-wide 20 mph zone has not made a blind bit of difference to the residents of these two areas, as far as encouraging them to take up cycling is concerned.

It appears that designating neighbourhoods as 20 mph zones did not reduce local residents’ feelings of threatening behaviour by drivers. Among the “direct communications” [from residents] received during the pilot one of the four most common themes was

Request for more enforcement especially where drivers were acting dangerously. 

The Bristol analysis also includes this cogent and very important qualification:

Casualty and traffic monitoring data is unpredictable over short periods. The numbers are very small and the study period very short, so it is not yet possible to properly assess the impact. 

Got that? In reality the methodological flaws in the Bristol analysis are quite blatant. All we get is percentages, never the numerical data that those percentages refer to. It is also quite clear that the sampling was far too small to carry any weight or significance. The analysis involved cycling and walking counts, but Bristol Council is reluctant to release the figures which might allow anyone to make sense of them. I’m reminded of a London borough which experienced

a 30 per cent increase in cycle trips between 1996 and 2002 and an 83 per cent increase in cycling in the eight year period 1998-2006, showing a success rate even better than Mayor Livingstone’s target of an 80 per cent increase in cycling in the period 2001-2010. 

By 2005 this borough had achieved a modal share of around 2 per cent on most roads and was aiming for 4 per cent modal share for trips within the borough. This borough has won no less than four awards from the London Cycling Campaign, as well as a Bike to School award and a London Transport award. This borough is rightly proud that ‘the number of cyclists is increasing as the new facilities are being introduced’ and is not ashamed to assert the exciting possibility that in this cycling wonderland, cycling might even become as popular as in Holland. Yes, this is irrefutably one of the leading local authorities in London 

Yes, in the London Borough of Waltham Forest numerous 20 mph zones and a fabulous commitment to the London Cycle Network have delivered a modal share of, er, just point eight of one per cent. Cycling has declined or at best stagnated; it has not grown, despite all the promotional froth.

In Bristol, the counts – whatever they are - indicate that

an increase in cycling and walking has taken place however this might indicate that this increase has come from non-residents, or from people who already cycle increasing their number of trips.

Well it might do. After all, Transport for London once quietly conceded that Evidence suggests that the growth in cycle travel between 2001 and 2008 was largely caused by cyclists increasing their cycle trip-making. There is no evidence of a net increase in the number of cyclists overall 

Or, alternatively, it might indicate that the serious methodological flaws of the survey indicate nothing more than statistical noise, from which anyone can extract any of the contradictory results they want to to prove their chosen case. For example, I don’t find this result remotely plausible:

In the inner south walking levels for those walk “most days” actually FELL by 12 % but there was an 8% increase in residents who said they walked “every week” (8.14)

I’m sure these wildly fluctuating percentage figures are worthless and distorted by tiny, unrepresentative sampling, but I can’t prove it because Bristol isn’t releasing the data that would allow an informed and critical analysis.

We do however learn that the sampling was inadequate in terms of its timing, since it relies on nothing more substantial than

manual counts conducted in August 2009 prior to implementation, and in August 2010 for the inner south area (after 2 months of operation) and in August 2011 for the inner east area (after 10 months of operation). Counts were taken on both a weekday and a weekend. 

Even this is lamentably vague in its detail (how many counts? where? for what hours?).

In direct contradiction to everything the Twenty’s Plenty website claims, Bristol concedes that

Early indications are that overall levels of walking and cycling activity across the pilot areas have increased both at weekends and on weekdays. However, it is not possible to confidently state that these changes were due solely to the introduction of the new speed limit. 

None of this should be a surprise.

As Joe Dunckley has observed

Bristol’s authorities can not claim to be administering a cycling city 


the highways department are as obsessed as any other with traffic flow and junction capacity; transport officers are just as obsessed with buses; parking enforcement just as powerless to keep cycle lanes clear; police just as indifferent to dangerous driving. It’s not a cycling city, it’s a city with a cycling department. 

You can read the fine detail of the Bristol report here

20 mph zones involve residential areas, not major though routes. The reason why vehicular cycling campaigners are so desperate to believe in 20 mph zones is because they still cling to the delusion that vehicular cycling can deliver mass cycling. They will happily prioritise anything except segregation. But as As Easy as Riding a Bike pointed out some time ago

This is the kind of thing the Dutch and Danes did at first. They put in the really useful cycle facilities in the places people really needed them on the main roads. They established the primary cycling network. That is the thing that really gets cycling up at first, and establishes in people’s minds the viability of the cycling option, with quality, high-profile provision. 


Routes through ‘back streets’ are rarely faster (and indeed, often just as unpleasant, in that you will often encounter belligerent drivers on streets narrowed by parked cars). and he derides the idea that we should instead go for the ‘low-hanging fruit’ like lower speed limits, stricter enforcement of road traffic laws, and so on – the kind of ‘Dutch’ solutions that Wright wants LCC to campaign on, apparently because we’re just never going to get proper Dutch-style cycle paths on our main roads. 

Exactly. The British cycling campaign establishment still clings to the notion that lots of little disconnected initatives will somehow get enough new cyclists to create a 'tipping point', after which only then can much better infrastructure be asked for, with public opinion strongly behind it. There are all kinds of problems with that scenario and specifically there is no reason whatever to believe that 20 mph zones in isolation will bring about any increase in the numbers of people cycling. For this reason, although Twenty's Plenty is obviously worth supporting as a casualty- and danger-reduction measure, it should not form the foundation of any plan which aims to increase cycling numbers.

The Dutch example indicates that to significantly increase modal share you first of all need to create a bike grid in urban centres, i.e. segregated cycle tracks on primary routes into town and city centres. That should be every cycle campaigner’s number one campaign objective. Everything else is secondary and amounts to nothing more than ‘pepper potting’ – at best localised improvements which fail to deliver meaningful cycling networks. Having established in a very visible way the primacy and attractiveness of cycling – build it and they will come – you then reinforce this cycling success with infrastructure which prevents rat-running through residential areas (which might involve road closures, or complicated and circuitous one-way systems), backed up by speed restrictions, contraflow lanes, traffic calming, and other strategies.

Simply cherry-picking one of these subsidiary strategies and trying to apply it to Britain’s car-sodden streets while hoping for success on the Dutch model is a futile undertaking. In so far as it makes exaggerated claims for encouraging new cyclists, the Twenty’s Plenty campaign is one such exercise in futility, and its popularity plainly rests on a twin appeal. For vehicular cyclists it offers another campaigning method that remains in the ghetto of ‘sharing the road’, while for local authorities it allows them to posture as friends of the cyclist and the pedestrian while doing absolutely nothing to curb the volume, presence or parking of motor vehicles. Because at the end of the day, 20 mph zones on the British model do not deliver safe, convenient, attractive cycling.