History and cycling's mode share in Amsterdam and London

This presentation by René Meijer of the City of Amsterdam has a useful chart showing the transport mode split in Amsterdam by the length of the journey.

I've tried to recreate it for London, using data from the London Travel Demand Survey downloaded from TfL's Romulus website*. It's not possible to make an exact comparison, for a few reasons:

• For London I use data on journey 'stages', which includes things like walking to the bus stop. It's not clear whether the Amsterdam data uses data on stages or on the main mode used for a trip from A to B.
• The journey distance categories are slightly different between the two cities.
• I've included motorcycle journeys in the 'car' category in London, but it's not clear whether or how they are counted in Amsterdam.

With those caveats in mind, here's the London chart.

Note, I've tried to use similar colours, so the reason there's lots of red in the Amsterdam chart and hardly any in the London chart is that lots of people cycle in Amsterdam and hardly anyone cycles in London. The difference is huge, and seems to be made up by a mix of more travel by car and by public transport in London for trips of between 0.5 and 10 km.

London's high public transport share for medium-length trips (predominantly bus until for trips of up to 6km and predominantly Underground/DLR thereafter) is striking, and I think a little under-discussed in the debate about growing cycling. We have a pretty wonderful public transport network that competes against cycling both in terms of attracting passengers and, in the case of buses, for road space. Even though there is definitely scope for huge growth in cycling with the right policies in place, this starting point probably means we're unlikely to match Amsterdam's modal share even for trips of the same length. Cities can't choose their history, and that can make a huge difference to their future.

* You'll need to request a password to access the full site.

Some simple economics of cycling

This post is an attempt to apply some techniques of economic analysis to cycling. There is a growing body of research about cycling, but most of it is from the perspectives of sociology (e.g. these journal papers) or engineering. Those approaches are very valuable (maybe more valuable!), but I think economics might have something to contribute too. However, while economics has had a lot to say about transport in general, as far as I can see this literature has tended to exclude cycling.

But I think taking an economic approach can help us think about some key questions regarding cycling, such as:

• Why has cycling increased in some city centres even but declined in the suburbs?
• Why do (some) people cycle in dangerous or unpleasant conditions?
• On the other hand, why do (most) people not cycle despite the convenience and low cost?
• What are the real costs and benefits of infrastructure projects or other policies to promote cycling?

Quite a lot of it will seem like simple common sense, or just bloody obvious. But my aim isn't so much to generate any startling insights as to show that cycling can be analysed using the same economic techniques as other modes of transport, and that future research using these techniques may be able to tell us useful things.

Time and money
At its simplest level (which is very much the level this post is written at, since I'm not really an economist), transport economics treats the choice of transport mode for a given journey as a function of a limited number of variables. Generally, the two most important are considered to be speed and cost. Reasonably enough, people are assumed to want to get to their destination as quickly and cheaply as possible.

The standard approach to comparing the speed of transport modes is to use door-to-door times, taking into account any time spent not going anywhere, such as sitting in traffic or waiting for a bus. In economic terms, people consider both the fixed and variable time costs of travel. The chart below illustrates this approach, showing (on the vertical axis) the estimated time taken to travel various distances (on the horizontal axis) in a typical urban environment [1].

Walking is the slowest mode of transport in terms of average speed (as indicated by the slope of its line), but for extremely short trips it is the fastest because there isn't any 'fixed' amount of time you have to spend getting ready, you just start walking. Cycling is the fastest mode for trips of a few kilometres despite having a lower moving speed than cars, because car users are thought to spend more time walking to and from their vehicles. Note, the average moving speed for cars in this chart is about 16km/h, which might seem very low, but that's because it's the average speed and takes into account congestion encountered in urban areas. Because bikes can filter through traffic jams cycling speed is much less affected by congestion and so is almost as quick as driving in this comparison.

This is a highly stylised comparison and you can obviously quibble with any of it, but for our purposes it makes two key points:
1. When thinking about transport speed people consider the total journey time rather than just the average moving speed.
2. On that basis, cycling may be the quickest mode for most short journeys in urban areas. Of course, if congestion is not an issue than the higher speed of cars will make it more attractive, while in extremely congested places cycling will look even better.

What about the monetary cost of travel? This is a bit more complicated, mainly due to the fixed costs involved in owning a car or bike (much less in the case of a bike). But if you set them aside and focus only only on per-trip costs, the important point for our purposes is that the cost of a bike trip is much lower than for any other mode except walking. Even if you take fixed costs into account it's likely that cycling still works out second cheapest for most people.

So if we stop at this point and consider only time and monetary costs, it looks like cycling should be an extremely attractive option for most short urban journeys. It's always cheap, and in congested traffic it's pretty quick. On this basis, cycling should be very popular in congested cities. So why isn't it?

Comfort and safety
Well, with some people it is. A growing number of Londoners, for example, choose cycling over other modes. But even in London cycling's mode share is low compared to its potential, and across the UK as a whole we still cycle very few short or medium-length trips compared to the Dutch or the Germans.

Transport economists have often acknowledged that people take other factors into account when choosing how to travel, with a key factor being the safety or comfort involved. Why 'comfort'? Because people tend to cite safety fears as a reason for not cycling, despite the 'objective' cycling casualty rates being very low even in London and the net health impact probably still positive. This suggests that it's not just the frequency of collisions that people are concerned about but also the frequency of near misses, and the generalised physical and mental discomfort of mixing with motor traffic.

Recently, some researchers have tried to understand how our own safety depends not just on our own choice of mode but also the choices of those around us. For example, in this article Anderson and Auffhammer provide evidence that people in the US are choosing bigger cars in part to protect themselves from the risks posed by everyone else's cars. The result is an 'arms race' in which the average US car becomes (a) bigger and (b) more dangerous to those around it.

Anderson and Auffhammer don't make this connection, but I think you can extend their model to cover cycling, treating bikes as basically extremely light vehicles that offer little or no protection in a collision.  Viewed that way, the choice to drive rather than cycle is similar to the decision to get a bigger car - for many people it is a deviation from their preferred option, motivated by fears over their own safety, that ultimately makes the roads less safe for everyone else too. Played out over a long enough period, it's easy to see how these individual decisions can cascade into a widespread rejection of cycling due to safety fears, as opposed to just the attractiveness of other modes or changing trip patterns.

However, this dynamic won't be the same everywhere. In places where most trips are short and the alternatives are very costly or slow, cycling may still be popular. That's why cycling rates tend to be highest in the centres of big cities - even though there's a lot of vehicle traffic it tends to be slow moving, which makes it less attractive as an alternative means of transport. The mixing of vehicles and bikes, however, means that cycling casualty rates may also be higher in these areas unless safer facilities for cyclists are specifically provided.

So to summarise, cycling has a cost advantage over other modes of transport (except walking, maybe) which is relatively fixed regardless of the conditions, but its relative benefits in terms of speed and safety/comfort are highly dependent on external conditions. Cycling will be quicker than other forms of road transport if traffic is congested, but unless traffic is so congested as to be stationary then the requirement to mix with heavy traffic will also make cycling less safe.

I think this helps make sense of trends such as the fall in cycling in Outer London and the simultaneous rise in Inner London. In Outer London the increasing availability and use of cars and their relatively high speed has cancelled out much of cycling's cost advantage and decreased its safety, while in Inner London high congestion and the falling number of cars has made cycling more attractive despite a still relatively high casualty rate.

Applications

One of the benefits of setting out the cycling decision in an economic framework is that it can allows us to estimate the effects of time, cost and comfort, and that in turn should enable us to make better decisions around infrastructure. Right now TfL are really struggling to properly analyse the costs and benefits of cycling infrastructure, apparently because their models don't incorporate any estimates of mode-switching in response to infrastructure changes, so cycling schemes look poor value for money in contexts where few people currently cycle, i.e. where such schemes may be most needed.

But if we can estimate a statistical model of mode choice with three factors (time, cost and comfort), then we can get at the question of how much additional cycling we should expect to see if we change the infrastructure on a particular route in a way that changes one or more of these factors simultaneously. Such a model could be expressed as something like the following:

Pij = axij + byij+ czij

where Pij is the probability of person i choosing to cycle trip j, x is time, y is cost, z is the comfort or cycling level of service, and a, b and c are the relevant elasticities of cycling with respect to each factor. This specification accounts for the fact that infrastructure elasticities will vary by person and by trip, although you could start with some averages.

Estimating this equation from observational data would be challenging, because every trip and every person are different and because simply gathering data on how people perceive safety is so difficult. So you would probably have to start by plugging in average parameter values from existing research about attitudes to generic types of infrastructure and road layout, assumptions about average speed, and so on.  

That way, you can start generating reasonable estimates of how many people might use a particular cycling facility in future, rather than just going on how many people currently cycle the same route. And that in turn means that you can get better estimates of the wider impacts of cycling, for example on how transport changes affect the 'market potential' of particular locations. This kind of analysis is already commonplace for vehicle and rail transport but not for cycling, due I think to a combination of the particular theoretical and empirical challenges posed by cycling and plain old cultural prejudice against what is still seen as an 'outsider' activity.

[1] Source: http://www.infrastructure.gov.au/infrastructure/mcu/urbanpolicy/files/ACTIVE_TRAVEL_DISCUSSION.pdf

Cycling fatality rate about five times as high in London as in Berlin

The other day Danny from Cyclists in the City linked to the city of Berlin's new plan for improving cycling conditions. I haven't been to Berlin for years and don't know what it's like to cycle there, but on the face of it the strategy looks pretty good.

What the strategy documents also allow us to do is compare how dangerous cycling in Berlin is to cycling in London, using the fatality rate per kilometre cycled. Berlin's strategy says (in German) there are 1.5 million cycling trips  a day, at an average of 3.7 km a trip, giving a total of just over 2 billion km cycled per year. And in the last three years there were 26 recorded cycling fatalities or 8.7 per year, giving an annual rate of 0.43 fatalities every 100 million km (or to put it the other way, over 230 million km cycled for each one fatality).

We can get comparable figures for London from the annual Travel in London report and road casualties reports. According to the data on trips per day (table 3.5) and average distance per trip (figure 2.5) in the most recent Travel in London report, there were 0.5 million cycling trips a day in London over the last three years, at an average distance of 3.1 km per day, giving a total of 553 million km cycled per year in London. There were also 39 cyclist fatalities in this period or an average of 13 per year, giving an annual rate of 2.35 fatalities every 100 million km (or just over 40 million km cycled for each fatality).

So it looks like the fatality rate for cycling in London is about five times as high as in Berlin. Note, these figures shouldn't be subject to the same concerns over casualty recording as the serious injury rates I calculated before, as fatalities are much less subject to under-recording than injuries.

Here's a table with all the numbers:

	London				Berlin
	2009	2010	2011	2009-11	2008-10
Trips per day (millions)	0.47	0.49	0.50	0.49	1.50
Distance per trip (km)	2.99	3.31	3.04	3.11	3.70
Total distance per year (m km)	512	591	555	553	2,026
Fatalities in period	13	10	16	39	26
Fatalities per year	13	10	16	13	8.7
Annual fatalities per 100m km	2.54	1.69	2.88	2.35	0.43
Million km per fatality	39	59	35	43	234

The problem with 'transport poverty'

The RAC Foundation have been getting some good press coverage with their argument that we should be very concerned about the 'transport poverty' experienced by low-income households who own cars. Their preferred solution is a big cut in fuel duty, as merely "tinkering" with the rate would be akin to "rearranging the deck chairs on the Titanic" according to their chair Stephen Glaister.

The RACF's argument is based on these statistics from the ONS, showing that there are around 800,000 households in the UK who are in the poorest 10% of households according to disposable income and who own a car, and that these households spend an average of £45 a week on transport, including an average of £16 a week on motor fuel. As these households all have a weekly income of less than £168 (see the top row of the table) that means that many of them are spending more than a quarter of their income on transport. The RACF calls this 'transport poverty' and thinks the way to deal with it is to make fuel cheaper.

There are several problems with this argument. First, what the RACF don't tell you is that only 31% of households in the poorest tenth of the income distribution actually own a car, compared to 96% in the richest tenth (see p.9 here). So if 'transport poverty' due to fuel costs is a problem, it is a problem only for a minority of the poor.

Second, it is likely that many of those the RACF say are in transport poverty aren't really that poor after all. Households with very low reported incomes are often there because they have suffered a temporary drop in incomes, but they could still be otherwise reasonably well off. As these academics point out,

for some of those at the very bottom of the income distribution, a recorded very low income should not be taken as a sign of more general lack of resources... It might reflect the fact that some individuals experience very low income for a relatively short period of time, but that they maintain their spending at some sort of long-run level: for example, someone between jobs (who could have a 0 or very low income if measured over a sufficiently short period), or someone making a loss in their selfemployment business (which would count as a negative income).

So it is very likely that many of the low-income households who own cars are only temporarily low-income. But even if we accept that these households really are poor in the usual sense and that enough of them own cars for this to be an issue (no matter how contradictory those two statements might seem), there is a third big problem with the RAC Foundation's argument. The ONS figures they cite indicate that car-owning households in the poorest 10% spend around £13m a week on fuel (830,000 households with an average weekly spend of £16), compared to spending on fuel by all households with cars of around £640m a week (19.7 million households with an average weekly spend of £32.50).

That means the households in 'transport poverty' account for just two per cent of total motor fuel spending in the UK. By contrast, car-owning households in the top 10%, who all have disposable household incomes of over £57,000 a year, account for 21% of total motor fuel spending. So any cut in fuel tax aimed at reducing 'transport poverty' would overwhelmingly benefit the better off.

Cutting fuel duty would be an extremely bad way to reduce poverty, especially if the money has to come from elsewhere. If cuts in fuel duty were paid for reducing benefits then you would be directly transferring money from poor to rich. If the RAC Foundation are really interested in reducing 'transport poverty' then they would be better off arguing for reductions in the cost of transport modes used mostly by the poor (i.e. the bus). But the best and most tried-and-tested way to reduce poverty of any kind is to just give poor people more money.

Mapping cyclist casualty concentrations in London

I've produced some maps of London cycling casualties on this blog before but it's hard to come up with an image that manages to capture both the scale and the spatial specificity of the issue. It matters where cycling casualties happen and it also matters how many there are in a particular area, but at the London-wide level there are so many that the most straightforward visualisation techniques just aren't adequate, even before you start trying to understand the underlying patterns of exposure and causation.

Here, for example, is a simple map showing one dot for every recorded cycling casualty in London over the five years to 2011 (the most recent year available). It includes fatal, serious and slight casualties, with the latter category by far the largest.

This map does tell us some useful things: cycling casualties are heavily concentrated in central London, and judging by the linear patterns seem to be common on major roads too. But there are so many casualties in the centre that the image becomes overwhelmed and you lose a sense of either scale or space. Some kind of aggregation would help.

Following the steps outlined on the Mapbox blog here, I generated a 'heatmap' of cycling casualties in Quantum GIS. A heatmap is a technique that uses spatial interpolation to predict the number of events (in this case, cycling casualties) in a small part of an area based on the actual number observed in or close to that area. It smooths out the pattern a bit and highlights with variations of colour the locations with the greatest concentrations.

If anything, this shows even more clearly how many casualties there are on London's major roads (including many of the ancient Roman routes shown recently on the Mapping London blog). I thought it could do with a bit more clarity, so, again following the Mapbox tutorial, I added some definition using contour lines, and a legend.  Note that the heatmap model 'predicts' 140 casualties in the deepest red cells and zero in the deep blue squares that cover most of the suburbs.

Maybe the contours look a bit messy at the London-wide scale but when you zoom into the city centre I think they help, partly by overcoming the blockiness of the heatmap results. In the map below I've labelled the areas in central London with the greatest concentrations of cycling casualties in the last five years. As you can see, the Elephant and Castle area is pretty clearly the worst in terms of the number of casualties, while there is a cluster of areas just north of the river also focused on major junctions.

Finally, here's a version with the street network showing underneath.

Animating London's population change 1801-2011

Over the last couple of hundred years London's population has grown and spread out on a vast scale. This process has involved a remarkable deconcentration of population from the centre to the suburbs: in 1801 the vast majority of its million people were crammed into a few central boroughs, with the square mile of the City of London holding 129,000 people. The population of the city grew by more than 5 million over the next century with the vast majority of that growth in the inner suburbs opened up by public transport. Over the course of the 20th century suburbanisation accelerated with the advent of the car, only for population growth to pick up again in the centre in the last few decades.

Summing up all this change in a single graphic is quite a challenge, so instead I've made this simple animated map, illustrating the changes in population using a dot-density approach where every dot represents 2,000 people. The dots are randomly distributed at borough level at the point of each Census starting in 1801 and ending in 2011, so they are not meant to represent the exact locations of individual settlements.

The maps were made in R using data from the Census (downloaded from the London Datastore and updated with 2011 data) and boundaries from the Ordnance Survey. The animation was done in UnFREEz as I couldn't get the native animation in R to work.

What the congestion charge did

The London congestion charge was ten years old this week, which provoked a bit of discussion about it, most notable for the absence of anyone seriously calling for it to be abolished. As Adam Bienkov points out, its introduction in 2003 was by contrast preceded by an avalanche of criticism and predictions of doom, much of it motivated more by political grievance than by evidence or principle.

I think the best way to get a sense of the C-charge's impact is to look at Transport for London's data (here, under 'Central London Peak Count') showing how people travelled into central London during the weekday morning rush-hour between 1978 and 2011. The chart below shows the trend for people arriving by car or motorcycle only (for some reason TfL don't separate the two out).

The number of people entering central London by car (and motorbike) has clearly been trending downwards since the early 1980s, but just as clearly there was a very big drop in the early 2000s. What's really interesting is that although there was a big dros (of about 20,000) in 2003, the first year of the C-charge, that was preceded by two years of almost equally big drops in 2001 and 2002. I don't know very much about what transport policy was like back then but given that the same TfL data shows a concurrent spike upwards in bus ridership it does look rather like a generalised 'Livingstone effect' rather than something limited to the congestion charge alone, though obviously that was a very important part of it.

More recently the decline in car traffic has slowed a bit and in 2011 there was even a small increase, though hardly a noteworthy one. And if anyone is dissatisfied with current congestion levels in London, as for example the AA seem to be, the obvious answer is to campaign vigorously for an increase in the charge.

A couple of maps showing where in London cycling commuting did and didn't grow between 2001 and 2011

A couple of weeks ago I posted a map showing the mix of commuting modes by ward in London from the 2011 Census, and this week I'd like to focus on the change in cycling levels since 2001. The maps below (click here and here for PDF versions) show the change in cycling, again at ward level, first the change in the number of people cycling to work in each ward and then the change in cycling's share of all commuting. I've included both because the numerical increase is interesting but can be distorted by differences in population growth.

In both maps the yellow areas represent areas of decline, the greens no change or moderate growth, and the blues higher rates of growth.

Growth in cycling commuting is concentrated in inner London and the South West, with the greatest increase in both proportional and absolute terms in Hackney. A couple of other patterns jump out: the rich boroughs of Westminster and Kensington and Chelsea are exceptions to the inner London rule, showing very little growth in cycling over the last decade, probably due to a combination of car-friendly policies and demographic factors. Newham also stands out as showing very little growth, suggesting that the river Lee or rather its crossings may be quite a significant barrier, hopefully something the new superhighway will address. South of the river there look to be pockets of high growth, in numerical terms at least, along the routes of Cycle Superhighways 7 and 8 and in a few other areas.

But while most of inner London saw pretty good growth in cycling over the decade, Hackney is clearly the star of the show. In 2001 4,940 people in Hackney said they cycled to work. By 2011 that had more than trebled to 17,312. Hackney's workforce grew at the same time, but there was a big increase in cycling's commuting mode share too, from 6.8% in 2001 to 15.4% in 2011 (in both cases excluding those working from home). In five wards cycling's mode share grew by more than ten percentage points. As Cyclists in the City pointed out, more people commute by bike in Hackney than by car or van.

There's an interesting debate to be had about whether the Hackney trend is due to demographics (an influx of young carless people), the emergence of a fairly localised pro-cycling sub-culture, or Hackney council's approach to road safety. Danny says here it's about policies and I've no doubt they're an important factor, but in my admittedly partial experience cycling in Hackney is no better than in Islington so I suspect demographics and culture have played a role too. We may know a bit more when more detailed Census data on who is cycling where comes out later in the year.

Dangerous driving and London's draft policing plan

The Mayor's Office for Policing and Crime (MOPAC) are currently consulting on a draft Police and Crime plan for London, to cover the period 2013-17. You can read the draft here and there's a short survey about it here. The London Assembly are carrying out a review into the draft plan too, which you can read about here. The MOPAC consultation runs until 6 March while you have until 15 February if you want to tell the Assembly your views.

The top priority in the draft plan is to:

Hold the Metropolitan Police to account for delivering the Mayor’s goal of driving down the key crimes of burglary, vandalism, theft of, and theft from motor vehicles, violence with injury, robbery and theft from the person by a total of 20%.

That list of key crimes notably excludes any mention of dangerous driving, as does the document as a whole. This is despite concerns over dangerous driving and road safety featuring prominently in previous police consultations:

• The 2010/11 Metropolitan Police Service annual report said that  "Road safety has featured consistently in the top five public priorities for policing, with speeding and dangerous driving major concerns".
• TfL's draft road safety plan said "The Metropolitan Police Authority's 'Have Your Say on Policing in London' consultation, which ran between June and November 2010, show that traffic and road related issues are the top priority for those who took part. Particular concerns identified in the consultation focus on road safety issues."

If anyone thinks London's policing plan should include something about improving road safety and tackling dangerous driving, then please do respond to the consultation (and to the Assembly review, while you're at it), saying so.

The colour of London's commute

Today saw the release of detailed Census data on, among other things, the mode of transport those in work use to get to work. One interesting aspect of this is the rising level of cycling in London, as described here by Cyclists in the City. I'll probably be looking at that later in the week, but first here is a map which attempts to summarise the transport mix across all of London in a single image.

(Click to embiggen, and higher-quality PDF here)

What the map shows is the mix of transport to work of residents living in each part of London*, using ONS data at Middle Super Output Area (MSOA) level. Each MSOA is given an RGB colour determined by the modal share, with red colours representing travel by car, taxi or motorbike, blue travel by public transport and green cycling or walking.

The result is a fairly simple pattern, with motor vehicles predominating on London's fringes, public transport in the inner suburbs and cycling and walking in the very centre. Those tendrils of blue reaching out presumably represent major public transport links.

A few details about the mapping technique for anyone who's interested: I was inspired to use the RGB approach by James Cheshire's map of election results and after some trial and error found a fairly simple way to do it in R which I can provide more details of to anyone who asks. The data and boundaries are both from ONS, the former downloaded from Neighbourhood Statistics. The maps exclude those people of working age who are not in work, who work from home or who use some form of transport so strange that ONS only describe it as 'other'.

* Edited this to make it clear that the map is based on place of residence, following @santacreu34's helpful comment.