Urban myths and the misuse of urban data

Following up my post about the mismeasurement of urbanisation in Egypt, I would highly recommend anyone interested in this area to read David Satterthwaite's article on 'Urban myths and the mis-use of urban data' (pdf).

Satterthwaite's article covers a lot of ground, including alarmism over 'out of control' urbanisation in Africa, the extreme paucity of census data in some areas, and problems in measuring city size and therefore various indicators of city performance. But I'm going to focus on his discussion of how the comparability of the UN's urbanisation statistics is undermined by the use of different definitions of urbanisation at country level. For example:

China’s level of urbanization in 1999 could have been 24 per cent, 31 per cent or 73 per cent, depending on which of three official definitions of urban populations was used. India appears to be a predominantly rural nation, but most of India’s rural population lives in settlements with between 500 and 5,000 inhabitants, which are considered as villages and therefore classified as rural; many more live in settlements with more than 5,000 inhabitants, which are still classified as rural. If these were classified as ‘urban’ (as they would be by some national urban definitions), India would suddenly have a predominantly urban population. 

And then there's Egypt:

in 1996, 18 per cent of Egypt’s population lived in settlements with between 10,000 and 20,000 inhabitants that had many urban characteristics, including significant non-agricultural economies and occupational structures. These  were not classified as urban areas – although they would have been in most other nations. If they were considered urban, this would mean that Egypt was much more urbanized, causing major changes to urban growth rates.

Remember that Egypt's central government had an incentive to systematically under-estimate its urban population, as granting city status to an area meant allocating it more funding and representation in parliament.  

The lesson here is that people should be careful about using data on cross-country comparisons, they should be extra careful when it comes to data on topics like urbanisation where there is no standard definition, and they should probably be extra extra careful about data that is produced by governments like Egypt's. After all, if your argument is that the Egyptian government is dysfunctional, then shouldn't you be at least slightly sceptical about the data that government produces?

Game theory and the risks of cycling vs driving

I've been reading Richard Tay's short article from 2002 on "The Prisoner's Dilemma and Vehicle Safety". Tay uses game theory to examine the implications of different fatality rates for drivers of small cars and four-wheel drives (4WD) in Australia. He finds that even though smaller cars are safer in that a collision between two small cars is less likely to result in fatality than one between two 4WDs, drivers will tend to prefer 4WD because in a collision between a small car and a 4WD the driver of the small car has a much greater risk of death. In the language of game theory it is a classic prisoners' dilemma, set out in the box below (the relative risk rates are taken from Australian data).

I thought I'd try and repeat the exercise for the relative risks of cycling and driving in the UK. I got data on the number of miles travelled and the number deaths or serious injuries (KSI, standing for 'killed or seriously injured') from accidents involving bikes and/or cars from the Department for Transport's 2009 road casualties report (tables 1a and 23c respectively), summarised below.

Using the KSI rates we can produce a new version of Tay's table as follows:

There is a similar outcome to the one Tay found, but the relative risks are much more extreme. In a collision between a bicycle and a car, the bike rider is roughly 100 times more likely to be killed or seriously injured than the car driver (1988 divided by 17). But the risk of a fatality is twice as high for a collision between two cars as for one between two bikes*.

None of this is particularly surprising, I suppose - I think everyone knows that if you're going to be in a road collision, it's better to be in a car than on a bike. But it does show how everyone making that calculation means that as a society we choose a sub-optimal equilibrium - a world in which the vast majority of people drive and the overall risk of death or serious injury is higher than it would be in a hypothetical world in which the vast majority of people rode bikes. That hypothetical world may not be a reasonable prospect, but we could surely do with getting a bit closer to it. Since the relative risks in a bike/car collision are probably not going to change very much, significantly reducing the rate of collisions has to be the aim.

*Incidentally, I used KSIs rather than just deaths as there were zero deaths resulting from bike-to-bike collisions in 2009. 


[Update, 5th May - I've corrected an error in the table, but not one that changes the conclusions]

20mph zones and congestion in London

Cyclists in the City report that the City of London is considering introducing a 20mph zone. This could be transformative if it goes ahead, and not just because of the positive impact on road safety. A 20mph zone would probably also lead to less congestion, because road collisions are responsible for around 28% of congestion in London, and 20mph zones reduce collisions by an average of 37% (p. 50 here). There is little point ever going above 20mph in the City anyway, yet at the moment many drivers do accelerate sharply whenever there is a bit of open road ahead of them. Cutting out that kind of behaviour would be a small but significant step to a saner city.

Over-elastic housing supply

What with all the complaining that people like me do about there not being enough new house building to keep up with demand, it is easy to forget that there can also be such a thing as too much housing construction. If the housing market is working well, there shouldn't be a big problem - new supply feeds into lower prices and builders stop building. But, as the Economist said recently:

when it comes to the biggest house-price bubble in history, theory does not get you very far. In some places the boom was big enough and irrational enough to suppress price signals from lots of new supply. Instead, availability of land simply fed speculative activity, which has made the popping of the bubble much more painful.

Much of the pain is felt in the labour market. The speculative frenzy got lots of people building houses, and then suddenly the last thing anybody wants is a new house built. This is more or less what happened in Nevada, which resulted in this:



It was a similar story in Ireland and Spain - see the Economist story for all the gory details [1]. In all these areas and perhaps particularly in Ireland, there was an additional layer of damage to the government's finances, when the revenues from construction income taxes and new home stamp duty suddenly dried up, contributing to the fiscal black hole Ireland is struggling to fill.

So while it's probably not much consolation to the many, many people who can't afford to buy or rent somewhere suitable, we in the UK may have inadvertently avoided a worse recession through not building too many homes. But I wouldn't go hugging your nearest NIMBY just yet - even though we don't know where the sweet spot of optimal housing supply is, I'm fairly sure we're nowhere near it at the moment.

[1] The OECD have also picked up on the potential problem of over-elastic supply in this working paper. 

Ed Glaeser at the LSE

Ed Glaeser spoke at LSE on Monday night as part of a tour plugging his new book, Triumph of the City. With his slightly ship's-captainy appearance, extravagant gestures and SUDDEN CHANGES IN VOLUME, Glaeser makes for an entertaining speaker and a great salesman for cities, which seems to be the role he has settled on.

His talk was focused on the central message of his book, i.e. the benefits of dense urban areas and the costs of anti-density policies. On the benefits side, he made a convincing argument that better communications technology makes cities more rather than less attractive places to work, because by making routine communications easier they increase the returns to the much richer face-to-face form of communication. This is more true in some industries than in others, and one in particular: Glaeser says "There is no industry in which knowing a little more is worth a lot more than in finance", so for better or worse we shouldn't expect financial services to lose their dominant role in New York or London any time soon.

Another key benefit of density, which after all is just proximity in another guise, is that it allows people to save on transport costs. This, said Glaeser, is the main reason why you tend to see so many poor people in big cities. Their housing costs may be high but their transport costs are low, on top of the greater economic opportunities on offer. So maybe the high measured levels of inequality in cities isn't such a bad thing: in fact, Glaeser declared that there is "more of a problem in the artificial equality of suburban areas than in the inequality of urban areas", which is certainly something to chew over.

The economic benefits of agglomeration have always favoured cities, but high population densities can bring significant costs, most notably in terms of disease, congestion and crime. Life expectancies and general health used to be much worse in cities like New York and London than in the countryside, but thanks in large part to huge public investment in sanitation (and, I would argue, transport improvements that reduced overcrowding in the very centre) the gap has closed and in some cases reversed. Public health is still a huge problem in some of today's megacities, but Glaeser is convinced that similar investments can make all the difference.

Cities in rich countries are today generally much safer places than they used to be, with crime levels having fallen in most countries since the 70s and 80s. Glaeser noted that workers used to effectively get a 'real wage premium' for enduring New York's awful conditions, but now it is the other way around, with people willing to pay huge prices for the privilege of living in formerly bohemian areas like Greenwich Village.

At this point Glaeser had some criticism for Jane Jacobs, who he thought was unduly hostile to densification in her Greenwich Village neighbourhood and places like it. When Jacobs lived there the Village was an attractive, human-scale, organically 'mixed community', and Jacobs wanted it to stay that way. But Glaeser contends that a lack of sufficient new housing supply meant prices in nice, central places like the Village went through the roof, with the result that it is now a profoundly un-mixed community where only hedge fund multi-millionaires can afford to live.

I've got a few comments on this last issue, as I think it's fairly critical for cities like London which are trying for economic growth without excluding the poor and middle classes. The first is that it seems to be more or less impossible to preserve both affordability and the existing neighbourhood built form in the face of strong housing demand: you've got to choose one or the other.

The second point is that the relationship between the two is unstable - after all, if the reaction to a big increase in demand for housing in, say, Islington is to build a huge number of flats, then it isn't the same place it was before and a lot of that demand may disappear, potentially leaving you with a supply overhang and a less attractive environment. That's probably an extreme example as housing supply is never that responsive, but the point is that economists don't seem to me to have a good handle on these interactions as yet.

Lastly, it's fine to argue that places like Greenwich Village shouldn't have made the mistake of restricting new housing supply, but that doesn't answer the question of whether it is possible to undo the 'damage' now that it has become an enclave of the super-rich. Is there really any chance of persuading such a group of local residents to accept major structural change to the area they have paid so much to live in? This matters because if the answer is no then new supply won't be in areas like the Village but in less desirable places, probably more distant from where people want to be. If that's the case, then the very attractiveness of cities may undermine some of their economic and environmental benefits.

Cycling in the City of London isn't getting much safer

Cyclists in the City link to a new report on road safety from the City of London, which for international readers is basically the 'Square Mile' financial district in central London. The report includes this chart showing the trend in the number of cyclists killed or seriously injured ('KSI') on the City's roads.

Looks bad! But wait, the City report says:

However the numbers of cyclists riding into the City has significantly increased, so that the actual casualty rate is falling.

Except they don't try to quantify the 'casualty rate'. So I thought I'd have a go.

Chapter 1 of the City's Draft Local Implementation Plan includes another trend chart, this one showing the number of cyclists counted crossing the City boundary on a single day in a given year [1].

To save you squinting, those numbers are:

1999 7,664

2002 9,565

2005 15,425

2007 16,030

2010 24,888

I combined the two, added a linear interpolation for missing years, and also linear trends for each time series, to get the following.

It's tempting to look at the two parallel trend lines and conclude that the relationship between cycle trips and KSI has not really changed, but that would be wrong, as the cyclist count was starting from a relatively low base so has actually increased by a higher percentage.

Another way to look at it is to isolate the years for when we have both count and KSI data, as in the table below [2]. Unfortunately there are only four of these and the most recent is 2007.

It is hard, and possibly foolhardy, to try and extrapolate a trend from so little data (but the same surely applies to the City's take on the figures). There certainly looks to have been a reduction from 1999 to 2002, but then no sign of any further fall in the five subsequent years.

This tallies somewhat with analysis carried out by Ben Lewis of TfL, who calculated cyclist accident exposure rates in Inner and Central London based on TfL counts at a large number of different sites and found a non-linear relationship between the number of cyclists at a particular location and the accident rate (see chart below).

The dashed vertical lines indicate current levels of cycling in each part of London. What the chart shows is that the accident rate is lower in Central London than in Inner London (perhaps partly due to 'safety in numbers') but is in danger of bottoming out in the absence of substantial additional safety measures.

This trend, or rather lack of trend, should be of great concern to the City. Cyclists accounted for 19 of the 46 KSIs in the City in 2009 (table 3.10 here), and in its draft LIP it has set out one target to grow the number of cyclists entering from 24,888 in 2009 to 36,300 in 2013, and another target to reduce the total number of KSIs to 37 by 2013. But if cycling does grow to 36,300 trips and the accident rate doesn't fall any further, from my table above we would expect to see 27 or so cyclists killed or seriously injured in a given year. It seems unlikely that non-cyclist KSIs (most of them pedestrians) will fall from 27 to only 10 in such a short period, in which case the KSI target won't be met.

In fact, the accident rate for either cyclists or pedestrians is going to have to fall dramatically if the City is going to meet both its target to grow cycling numbers and reduce the numbers killed or seriously injured. There is little sign in the draft LIP that the City is really considering any measures that are going to dramatically improve conditions for cyclists and pedestrians, so right now it seems extremely unlikely that it is going to achieve both of its targets. Something is going to have to give - the question is, will we see the growth in cycling tail off, or continue upwards at the expense of yet more cyclists killed or maimed on the City streets?

[1] There is frustratingly little information given about this count (Is is just cyclists entering the City, or all boundary crossings? Is the count always at the same site? Is it measured just once a year or averaged over several days?), so we just have to assume that the methodology is consistent from year to year.

[2] Note that the size of the figures in the final column isn't that meaningful, since it divides an annual figure (KSI) by a daily figure (cycle count).

The mismeasurement of urban Egypt

The short version of this post is that despite recent reports, urbanisation in Egypt has not gone into reverse in the last 30 years. The Egyptian government systematically under-estimates both the size of and the growth in Egypt's urban population for adminstrative/political reasons. For the full story, see below the fold.

London ethnicity map - Now with bigger blobs

I've been told that the dot-map of London ethnicity is a bit faint, so here's a version with bigger dots! Click for full size.

The red dots represent white people, the blue dots black people, the green represent the (south) Asian categories, and there are a much smaller number of purple ones in there that represent mixed-ethnicity or 'other', such as Chinese.

Big differences in attitudes to speeding by age, gender and income

The Department for Transport have published some fascinating research recently (funded by the previous government, I imagine) which has unfortunately received very little media attention, presumably because they have chosen not to highlight it themselves by spoon-feeding the media with press releases and the like. In December they released the results of a big survey into attitudes towards transport and climate change (which I picked out some cycling-related factoids from) and now we've got the results of a survey into attitudes towards road safety. I'm just going to highlight one particular aspect below, but I would encourage you to read the report and/or the accompanying tables yourself.

According to table A1.12 here, 21% of respondents think the level of police enforcement for speeding is 'too high', 45% 'about right' and 33% 'too low'. Seeing as speeding is also dangerous for drivers and pedestrians and costly for society as a whole, you might wonder why the police don't enforce the law on this more strongly. Then you look at the detailed breakdowns:

• Men are slightly more likely to think there is too much enforcement than too little, while women are three times as likely to think there is too little. 
• Those on low incomes (up to £15k a year) are two to three times as likely to think there is too little enforcement as too much, in contrast to those on higher incomes who are more likely to think there is too much than too little.
• Around half of the elderly (aged 75 or over) think there is too little enforcement of the speed limit, compared to only 12% who think there is too much.
• Drivers are almost evenly split on the issue, while three times as many non-drivers think there is too little enforcement as think there is too much. 

We shouldn't ignore the substantial fraction of people (40% to 50% in most socioeconomic groups) who think current enforcement levels are 'about right', but I think these figures should still be of concern to the police and the government. On one side you've got groups of people who have historically been fairly culturally dominant (i.e who are affluent, male, middle-aged or all three) and who by their own account are also more likely to drive (see table A1.1),  to break speed limits (table A1.36) and to break them by a lot (table A1.63). These people either think the law is either about right or too harsh. On the other side you've got groups who have often been marginalised (women, the elderly, the poor) who are much less likely to drive, who drive more safely when they do drive, and who think that current levels of enforcement are too low. In an era when the state is meant to be actively promoting 'equalities' (as well as, y'know, enforcing the law), I'd suggest that the police and the government should seriously think about whether they are striking the right balance.

Quibbling with Ed Glaeser about housing supply in London

In the course of an interesting interview with The Atlantic on the subject of cities in general and his forthcoming book on cities in particular, Ed Glaeser says:

London has imposed a huge cost on itself by keeping its height so low — far too low relative to what an incredibly exciting, great city it is. That city could be much more affordable than it is now, and much more inclusive. It has rigidly held to its Victorian building style, particularly for residential. They have allowed some commercial high-rises to go up, which is fairly typical — usually there’s more comfort with building up in commercial zones than there is in residential sectors.

That's only partly right. While London has not been building enough homes to make housing much more affordable, a lot is being built, and at high densities. New housing supply has been running at between 25,000 and 30,000 a year since 2003 (see table 15 in this pdf), and building densities have doubled in the last decade, as the chart below shows (source). We have built quite a few high-rise residential towers in the past few years, and there are many more in the pipeline (1).

So I think it's not quite as Prof Glaeser suggests (2), in that we're not still churning out Victorian-style homes (not that Victorian terraces are particularly low-density, mind). But he may still be right that we are not building enough new housing overall, partly due to a concern with not over-densifying existing residential areas - i.e. we're not replacing our existing Victorian terraces with higher-density flats.

If you look at the borough-level supply figures you'll see that the highest supply is in areas with either a lot of ex-industrial land or a relative lack of wealthy home-owners, or both (such as Tower Hamlets). Put another way, it is a lot easier to get homes built where there are either not many neighbours or at least not many of the kind likely to make a lot of noise objecting.

This pattern is worrying for a couple of reasons. Firstly, we will presumably eventually run out of former industrial sites to convert to residential use. Secondly, if incomes keep rising in Inner London and if housing benefit changes help force renters out at the expense of affluent homeowners, we can perhaps expect objections to new housing supply to get more vociferous.

(1) I know, many of the towers at that link are commercial. But many others are at least part-residential.
(2) Note, these figure only go up to 2009. I'm currently working with colleagues in the GLA on the new Annual Monitoring Report, which will have data for 2010.