Saturday 5 November 2011

Another myth about population density

Some myths just get repeated again and again and again.

I just read a comment under David Arditti's excellent post "Some more thoughts on Dutch cycling" (please read it) that said "This should be considered in the context of just what a congested country the Netherlands is: a population of 18m crammed into an area the size of Wales, if that, so population per square km is much higher than ours, and even in rural areas you can expect to pass a house every couple of hundred metres and a village every km."

This is an emotive way of putting an argument, using Wales as an example because it's generally thought of as a relatively sparsely populated part of the United Kingdom. However, it's also highly inaccurate because he's exaggerated the population of the Netherlands by more than a million people, and is suggesting that Wales is the same size when actually is only half the size of the Netherlands (20779 km2 vs. 41848 km2).

Guessing at things like this is rather pointless when real figures worth comparing are remarkably easy to find in Wikipedia. England makes an interesting comparison with the Netherlands because as it turns out, the population density of England is 395 people per square km while that of the Netherlands is 403 people per square km. i.e. the densities of these two countries are very nearly exactly the same. What has been achieved in the Netherlands could also be achieved in England if British people want it enough.

What's more, the population density even of Wales, at 140 per square km, is not that much lower than the population density of Drenthe, where we live, which has 183 per square km.

Americans use this excuse even more often than the British, but as I wrote about that before, I won't bother to repeat myself. When considering any of this, bear in mind that US cities are some of the most densely populated in the world, and that population density of cities have no correlations at all with their cycling rates.

Note also that Dutch commutes are the longest in Europe.

Another article which I recommend highly this week is Dutch Pick and Mix. More about Britain on Monday...

For the benefit of a blogger with a comprehension problem who misunderstood this post when he linked to it, nowhere above do I imply that Wales is merely a province. Wales is, of course, a country in its own right.


Kevin Love said...

It is not so much the average population density as how that population is distributed. Many people have written about urban sprawl, but even in rural areas of The Netherlands, densely populated villages abruptly give way to fields with cows grazing.

This is very different from rural areas of the USA, where we see a phenomenon that I call "rural sprawl." In other word, residential, commercial and other non-agricultural land uses outside of towns. So, for example, instead of having shopping in the centre of densely populated towns, major stores are far away from where anyone lives. This forces the rural population to drive cars to do their shopping.

Here in Ontario, strict laws have always prevented rural sprawl.

For example, for the last 10 years of her life, my paternal grandmother lived car-free in the town of Clinton, population 3,082. Two banks, the post office, church, grocery store, bakery, library, three clothing stores, two restaurants, three bars and much else was within a 10 minute walk of where she lived in a flat above a barber shop.

The laws preventing sprawl continue to get even more strict with the passage of Ontario's Places to Grow Act, given Royal Assent on June 13, 2005.


...and click on the "backgrounder" for a description of how, in my opinion, land use planning should be done.

Anonymous said...

even in rural areas you can expect to pass a house every couple of hundred mtres

You know what, that's actually true in a lot of places... but it doesn't mean the countryside is heavily populated. Dutch public planning tends to be highly organised, and through a series of land consolidations in the late 19th and early 20th centuries farmers now tend to own rectangular plots, with the short edge (and the farmhouse itself) bordering on a country lane.

So if you're cycling or driving through the Dutch countryside, you will indeed pass a (farm)house every few hundred metres, but there's more than likely up to a kilometre of empty farmland behind it. See for a fairly typical example here.

Tallycyclist said...

I agree with Kevin's point about population distribution in rural areas of the US, and that stores can often be located far away from most people. But this is a problem in and of itself: why are stores not located more centrally? For some, distance can be a valid excuse. But the main problem is that for most others, they just choose to jump on the "too far" bandwagon as an excuse. Population density is one of those statistics that can easily be presented in a misleading way (and works well on the general public) and so the planners use it as an easy cop out. For urban settings and cities, this reason has almost no validity.

I grew up in an American suburb where there were not even crappy shoulders on most roads and where stores were all scattered in several locations (though distances are not more than 3-6 miles to any). This is a very doable distance for most people, even if they're non-athletes like me and don't work out regularly. When I first started university at FSU as a novice cyclist, my 1.6 mile (1-way) ride to school with one little hill was quite a workout. Later I moved and my commute was 5.3 miles one way and after a few months this longer commute to school doesn't even feel as challenging as the shorter one did in the beginning. And having done this for over 2 years now, I can easily out-pace friends with their road bikes who only occasionally ride, especially on hills.

I've also cycled in Copenhagen twice now, staying with friends living in the suburb of Hvidovre, which was a 7.7 km ride to the center of the city, with separated cycle tracks the whole way. I admit Denmark is 2nd best to the Netherlands, but it's still WAY ahead of anywhere in the US and pretty much all other western countries. And I totally feel that infrastructure is the most important factor. There's a very good reason why my parents never allowed me to walk to my elementary school, which was less than 1 mile and located on the same road that my development branched off of: it was a 45 mph road with no sidewalks and very narrow shoulders, which disappear completely half way to the school. Thanks to the crappy planning, I was relegated to taking the yellow school bus everyday, on a less-than 1-minute ride to school.

townmouse said...

I suppose the more interesting question is how far away people live from the places they're likely to visit (shop, school, work...). I know there are calculations about a neighbourhood's 'walk score' - how many amenities you can reasonably get to on foot - I don't know whether anyone does a similar calculation for its 'cycle score'. Some sociological factors (such as the fact that children in the UK are likely to go to quite distant schools, particularly secondary ones) may also come into play. But the much more interesting point is that once you start thinking about cycling as your primary means of transport, your view of where is a good place to live or open your shop changes too - so if there's less urban sprawl in the Netherlands and other parts of Europe that may be as much because people are using bikes to get around as the other way around. Certainly some of those very thinly spread US suburbs must look very much less attractive now, with higher fuel costs...

David Hembrow said...

Actually, this whole idea that there is less urban sprawl in the Netherlands is itself just another excuse.

As you saw when you were Sally, one of the suburbs of Assen is on the other side of all the big-box shops and an industrial estate from the centre of the city. However, that didn't prevent there being good access by bike.

Also, I dispute that many secondary school children in the UK travel further than Dutch children do. Quite a lot of them cycle more than 20 km each day to get to and from school, and it's not that uncommon to ride more than 40 km a day to get to and from school. The difference is more in how they get to school than the distance.

Much is made of the idea that US suburbs are particularly spread out. However, it remains a fact that 40% of US urban journeys are under 2 miles in length.

Andy in Germany said...

Surely the problem is that population density, hills, weather etc are all variables? I'm not sure you can really compare two places on a like for like basis, nor are we going to get the exact same result in two locations.

I really would like to see the same level of infrastructure in our town as I saw in the Netherlands, and in summer I reckon we'd be on the same level as Assen within the town if we did.

Outside of the town we have a variable: a series of deep valleys which are daunting to even the most enthusiastic of cyclists. There are ways around this, of course, and they include cycle infrastructure and stopping cars from having free range, but ultimately our transport mix will look different to yours: probably a mix of public transport and cycling for most people.

r s thompson said...

is there any realtionship to density and traffic accidents?

Paul M said...

Oh Christ on a bike David! What makes you think my personal observation of the Dutch abroad on David Arditti's blog is a "slight"? It may not accord with what you experience at home but it accords with what I experienced in France. I was not intending to sugest that France is some form of cycling Nirvana, any more than the UK is.

Ok, so I managed to overestimate the population of the Netherlands by a massive 7.57% - it was a fair stab even if not quite accurate. You have also been highly selective in your population density comparison. If you read the Wikipedia entry more carefully, you will find that the denisty of metropolitan Netherlands is actually 491/sqkm - you only get to 403 by including three caribbean islands in the tally. So the Netherlands is 25% more densely populated than England is - my comparison was clearly not as accurate as it could have been but it wasn't intended to be a treatise on geography.

You need to lighten up. I consider myself a cycling advocate and campaigner in a modest way, primarily in the City of London where I work and SW Surrey where I live. I am not trying to contradict you or put you down or diminish your arguments, but sometimes I do feel that the "Go Dutch" school of cycle campaigning (which I 90% support) behaves more like the Society of Jesus than an advocacy group - the smallest departure from the doctrine is heresy, heresy is simply not tolerated.

On that basis, there is little hope of us moving anywhere near the Dutch standards in this country - we can so easily be dismissed as a bunch of cranks, not because of what we believe, but because we appear to believe that this belief is the only true faith.

Ben said...

I am intrigued by your pop.density and cycling rate analysis. I will return to give it a good read!

Jim Moore said...

Paul M
You are the type of crank you describe. You agree with 90% of Dutch cyling? Classic "Dutch pick-and-mix". Go read the link.

"Metropolitan Netherlands"? What is that? The Wiki density of 403 does exclude the Caribbean islands but including them only lowers the density to 400. Oh, you forgot to mention Wales where you were only wrong by about 50%.

Starting a comment with "Christ on a bike" and then telling someone to "lighten up" is scary passive-aggressive behaviour. David left England for better cycling conditions but if I was English I would have left to avoid this common English mannerism that borders on being psychopathic.

You work in the City (not so esteemed and glamorous anymore is it?) and live in whitebread conservative Surrey. No wonder you don't want "get" Dutch cycling or want to see it in England, as it's Socialism on wheels.

Kevin Love said...


I don't view urban sprawl as an "excuse," but a serious land use problem that causes issues for active transportation and living a decent life for children and the elderly.

My grandmother was not a cyclist (at least in the last 10 years of her life), but lived car-free just fine in her town of 3,000 people because she could walk everywhere she needed to go. How many US towns are the same?

In most, those who are too old to drive safely continue to do so because they cannot live a decent life without a car. So the USA has so many horrific crashes cased by people who are unfit to drive and yet fight tooth and nail to keep driving.

Here in Toronto, density is strongly correlated to cycling. Take a look at the map at:

The downtown ridings have high public transit, walking and cycling mode shares. For example, the Riding of Toronto Centre has mode shares of:

38% public transit
34% cycling and walking
26% motorists (drivers and passengers)

The transit, walking and cycling mode shares steadily decline as one goes further away from downtown.

Richard Adamfi said...

I've never heard of the term 'Metropolitan Netherlands' but I assume Paul is making an analogy with 'Metropolitan France', which is a well known term meaning the part of France that is in Europe (i.e. mainland France plus Corsica).

I have long come to the conclusion that the UK is incapable of learning from its neighbours. I am a public transport enthusiast and after visiting NL in the mid 90s I decided I was going to move there because of its PT. I learned Dutch from books and tapes but for various reasons I never made the move.

Just like with cycling, British people (including bus/train enthusiasts) keep coming up with reasons why trains can't connect like in NL, why buses can't stop at the rail station like in NL etc.

I have only just discovered this blog in the last few days but I have been unable to stop reading it in the last few days, catching up with all posts from 2008.

I didn't realise, until reading this blog, how big cycling is in NL, making local buses irrelevant. So I'm now going to learn Dutch again!

Dennis Hindman said...

A very important element to the success of most bicycle sharing systems is station density, and that is difficult to accomplish without population density.

New York City is starting off with a bicycle sharing system next year that will have 600 stations and 10,000 bicycles. The entire cost of implemently and operating this will be paid by the user fees, advertising and sponsorship.

Los Angeles is probably the second most likely city in the U.S. to be able and try to implement a non-publicly funded bicycle sharing system. That is mainly due to the population density and the amount of on-street advertising that is allowed. A larger audience provides more money available for the advertisement, which in turn provides more corporate funding for ads attached to the bike sharing stations.

How much of an impact can bicycle sharing have with increasing cycling in a city? Well, Paris doubled their bicycling within a year after Velib was implemented. The amount of bicycling in Paris was only one or two percent before Velib, but Velib also stimulated private bicycle use as much as the public bicycle sharing.

The city of Hangzhou China has a 50,000 bicycle sharing system that is essentially free to the user and the city plans on increasing it to 175,000 bicycles by 2020. Hangzhou's population of eight million is comparable to New York City. If those 175,000 bikes are rented an average of 3 times a day, that's 525,000 bike rides. If that increases ridership with private bikes by an equal amount, like in Paris, then that would be an increase of over a million bike rides a day, or about 13 percent of the population on bikes directly or indirectly due to the bike sharing program.

The Washington D.C. bicycle sharing system is an example of how installing a bicycle sharing system can spark an interest in bicycling for U.S. cities.

Costs of implementing a substantial bicycle infrastructure, similar to the Netherlands, can be a significant problem. The road lanes in Los Angeles have been deteriorating since WWII from lack of adequate funding. The estimated cost of getting L.A. roads up to a B grade level is about a billion dollars and about forty percent of the 10,600 miles of sidewalks are in need of repair. Without enough funding for roadways or sidewalks, I don't see how bicycle infrastructure can go much beyond on-street paint, Bicycle Boulevards and a few bike paths in Los Angeles.

Joseph said...

Dennis, it may be true that repaving the streets (all 10,000 miles!) in Los Angeles will cost $1 billion, and sidewalk repairs might cost anoth $1/2 billion, with crazy USA construction costs. But that's only $250 or $500 per person in the city; less than a cheap hybrid bike. Meanwhile, we are spending $1 billion just to add a few miles of "carpool" lane in one direction on the 405 freeway.

Compared to freeway construction and transit projects (eg. $4 billion for the Westside subway), bike and pedestrian transportation projects are really cheap. 1000 miles of bike lanes can be painted for only $10 million dollars, or about the cost of 10 new natural gas buses. Move the parked cars over, and you've got hundreds of miles cycletrack for only about twice that cost.

I do agree that Bike Share only works with high densities. In Los Angeles, the area around Downtown, and perhaps a couple pockets around Koreatown, Hollywood and Westwood would be the only places that bikeshare would be profitable. But infrastructure is also needed, to make it work.

And the Netherlands does fine with moderate population density and personally-owned bikes, with bike rentals at intercity railway stations (rather than local bike share)

David Hembrow said...

Paul: Please don't take things so personally. You provided a very good example of someone getting the figures very wrong. That's all. BTW, your home county of Surrey has quite a high density: 678 per square km is a higher density than 9 of the 12 provinces of the Netherlands, more than twice as high as 6 of the 12 provinces, and nearly four times so high as Drenthe where we live.

Kevin: I understand your point about your grandmother. There are many people across the world living in much the same way.

However, I think you are making a very incorrect conclusion from your figures about Toronto. Actually, what the data illustrates is that people who live very close to the city centre tend to walk to the city centre.

That's very nice for those who live in that location, but it's not relevant to those who live so far out that walking becomes too time consuming.

Cycling has the potential to achieve much the same thing over a much larger area. However, it remains a small part of the picture in Toronto because it simply isn't inviting enough. If the city had infrastructure which made it possible for a large number of people slightly further out of the centre to make their journeys by bike then this could become significant.

Richard: I think you'd like it here. The PT is good (and cheap) but the cycling is better.

Dennis: You really must stop this hyping of public bikes. Lots of money is being made by the corporations who put these systems in, and much hot air is expended in trying to justify them by telling people what great value for money they are, but there is still no real evidence of an effect on modal share due to them.

10000 bikes in New York ? So what ? That's one bike per 1000 people. Just as with London, the problem in New York was never a shortage of bicycles, but that people are too damned scared to cycle in the city.

It's interesting that you use Hangzhou as one of your examples. China is an example of a country which has gone backwards in recent years. This is not a success story.

You make a lot of attempts to make out that these bikes come with no price tag. However, that's not really so. There are always expenses arising, and in any case the money which has been raised by unconventional means to hand over to profitable bike share companies could instead have been invested in your infrastructure.

The crumbling infrastructure of Los Angeles would have been a much better place to spend that money than on a hyped up bike share scheme.

Current plans for cycling in the city are abysmal.

Paul M said...

I'll try again - the point I was trying to make, clumsily perhaps but I still think I have been wilfully miscontrued, is that I agree with David A's analysis, that it is not anti-car policy, or unusually high levels of traffic congestion (hence the populaton density observation) or abnormally considerate driver behaviour which gets the Dutch onto their bikes, so it has to be something else, and infrastructure seems the most likely explanation.

And Jime Moore, if you can't embrace a 90% on-sider, and you can't tolerate an element of pick'n'mix, then you are doomed to failure. You have to bring over the mainstream politics, and the popular mood, of a country like the UK if you want it to adopt anything remotely resembling a Dutch model. If you aren't willing to embrace me as a fairly early step on the way, you WILL fail, believe me.

John Romeo Alpha said...

With this post, as well as in your comment on my blog post on a density vs commuting in the US, you've convinced me of a "not proven" verdict for those who assert through a simple bivariate analysis that population density and cycling rates are related. It is a myth. Even looking at the cycling rate in one city in China where the density has remained unchanged and comparing 1985 with 2011 would show the vacuity of the simple bivariate view.

Dennis Hindman said...

According to the review: Infrastructure, programs, and policies to increase bicycling, co-written by John Pucher, the impacts of bicycle sharing programs are hard to access due to the bicycle network being often expanded at the same time. The paper does note that the trips by bike increased from 0.75% to 1.76% in Barcelona, from 1.0% to 2.5% in Paris, and it increased 75% in Lyon after implementing a bicycle sharing program. The authors also stated that "the lack of evidence of a positive effect of some specific interventions is not the same as evidence of a lack of positive effect."

The paper also stated that there was an increase in bicycling when people were given bicycles. It also states that it is not clear which measures are the most effective.

The authors only found four studies that examined bicycle boulevards or cycle tracks. They found positive results, but there were no good estimates of the quantity of increase in bicycling after they were put in.

Could it be that parking policies, traffic calming along with auto and gas taxes also play a major role in getting people to bicycle? Why would someone choose to bicycle, when you can drive up to 50 miles an hour and easily find a parking space when you get there, as you can in the San Fernando Valley of Los Angeles. The cars have relatively low taxation and gas is less than four dollars a gallon.

New York City has cleverly added 250 miles of bike infrastructure in the last four years by having 90% of it funded from outside sources, such as the federal government. That funding seems to have decreased significantly, as the bike improvements are slowing considerably this year.

So, to keep the momentum going, NYC is adding bicycle sharing, which again, will not cost NYC a DIME. All installation, maintenance and cleaning will be borne by the company implementing it, per the contract with the city. Half of any profits will be shared with the city and could be used for infrastructure. Now, how could you do this without this contract, have free standing ads? That would probably not go over so well if they are not attached to a bus shelter, bike sharing kiosk, or bikes.

Judging from Paris, Barcelona and London, I would expect that at least 100,000 people will become yearly subscribers to the NYC bike sharing. In addition, thousands of tourists will use it. That would be about an average of at least 30,000 users a day, which is not insignificant.

Joseph E: The $1 billion to add a car-pool lane to the 405 freeway is a specified project in the half cent Measure R sales tax that voters approved by 70%, when President Obama was elected. The $4 billion for the subway extension was also a project listed under Measure R. There is no specific project mentioned for bicycling in Measure R. Funding for bicycling in Measure R is limted by the 15% set aside for each city to do infrastructure improvements. L.A. decided to use 10% of that for bicycle and pedestrian improvements.

Voters turned down Proposition JJ in 1998 to raise $770 million to fix the sidewalks.

In recent years voters in L.A. county have approved added taxes to build K-12 schools, community colleges, and libraries. But, I sincerely doubt that voters would approve more taxes to add bicycle infrastructure for those 1% who are riding bikes now.

David Hembrow said...

Dennis: I'm not against bike-share, but I dislike the huge amount of (often commercially driven) noise that its proponents make about very small numbers.

Pucher is telling you that only part of the small increases claimed in those cities could possibly be due to bike share. Still none of those places have anything approaching mass cycling.

No study is needed to show that bike share can't significantly alter modal share, just calculate it. Even if, each velib bike were used ten times a day (every day of week, every day of the the year, winter and summer), it couldn't possibly achieve 0.8% of total journeys in Paris due to a lack of bikes. In real life, it can only achieve much less than. It is irresponsible to suggest that this scheme has caused more than a small proportion of the rise in cycling in Paris, where the cycling modal share remains very small.

New York: Urban New York has a population of over 18 million living at a density of over 2000 per square km. They hardly cycle at all, with only 0.6% of commutes by bike.

The Netherlands has 16 million at a density of ~= 400 per square km making 27% of all journeys by bike. 45x the NY cycling rate.

NL: A million bike journeys per hour, 14 million per day.

You say a claimed potential figure of 30000 journeys per day by hire bike is "not insignificant" ?

The difference is 500 times, nearly three orders of magnitude. I call that insignificant.

250 miles (400 km) of cycle infrastructure in New York vs. 29000 km of separate cycle-paths + 7000 km of on-road cycle-lanes in NL. 90 times the length for a similar population or two orders of magnitude, not including 40000 km of 30 km/h roads, many prioritized for cycling.

London: London's own figures show that few bike journeys replace car journeys (< 150 per day) and 84% of registered users have their own bike. This is important because it lends support to anecdotal tales about existing cyclists hiring bikes in order to avoid vandalism and theft. Though figures for use are already very low (< 0.1% of total journeys on hire bikes), the number of new bike journeys due to the scheme is almost certainly much lower again.

Driving: It's not at all difficult here either. I can drive efficiently to the centre of this city and use the cheapest car parking in the Netherlands (very often free, actually) when I get there. It's never full so looking for a space won't be a problem. Despite this convenience for driving, over 40% of journeys here are made by bike.

On the other hand, in the UK the cost of running a car is almost the same as here (roughly $9 per gallon) and car parking is often more expensive than here. However, the cycling rate in Britain remains stubbornly under 2% of journeys.

The only significant difference between the two countries is the infrastructure. Britain is as car oriented as is the US, and has achieved much the same low level of cycling as a result.

You won't attract people en-masse out of cars with a combination of not enough infrastructure and an inadequate number of shared bikes.

Dennis Hindman said...

David: You contrast the cycling rate in the Netherlands to other countries or even cities. What you usually fail to indicate is that at it's lowest cycling rate in the mid 70's, the Dutch had a cycling modal rate of about 18%. It has since increased to about 27%. That is an increase of 9% in about 35 years. Which is an average of a quarter of a percentage point increase per year.

So, if a city starts out with a 1% cycle modal rate and increases it by a half percent point per year, or more, you scoff and state how the Netherlands cycling rate is so much higher than that. Well, the rate of increase per year would be at least double of what the Dutch have achieved over the last 35 years. In many ways it can be harder to increase modal share when you start from 1% compared to if you are already at about 18%. The resistance to change or obtaining funding can be fierce at a 1% modal share and as the cycling rate increases, the support for it grows also.

New York City has done a remarkable job of creating 250 miles of bicycle infrastructure in the last four years, with 90% of the funds coming from outside sources. In other words, they did this with very little of their own money.

One estimate states that New York City has about 200,000 daily bicycle riders. The upcoming bicycle share will have 10,000 bicycles and probably at least 30,000 users per day within a year. That would be an increase of 15% more bike riders within a years time, and it will be done with no money coming from city government. That is a remarkable achievement, if it works, and may be the first bicycle sharing system that makes a profit and that in turn will motivate the operators to expand it. This is much less likely to happen if it is entirely run and funded by the New York City government.

If half of those additional 30,000 bike riders are simply replacing their own bikes with the bike share bikes, that would still be a increase of 7%, again with no city funding. But, the more likely scenerio is that some people will try it out of curiosity, perhaps start biking on a regular basis, and then get their own bike to use instead. You have to get people bicycling in order to change their habit and bicycle sharing can create enough enticement for some people to change their lifestyle.

This is unlikely to have taken place ,in this amount of time, without the addition of the bike sharing. The money used towards getting the bicycle sharing operational cannot be used towards bicycle infrastructure as it is not the city's money. Half of the profits, if there are any, will go to the city and those funds would not be available if it wasn't for the bicycle sharing scheme.

David Hembrow said...

Dennis: I'm not scoffing at anyone. You simply continue to make ludicrous claims. Because of statistical errors, figures always dance up and down a bit from year to year. When you're measuring such small figures as you are, then the error margins are huge.

Keep up a slow increase for ten years and you can reasonably claim there is a trend. However, you can't look at one year vs. another and conclude anything from it. That's simply bad statistical technique.

What's more, there is a very big difference between going between 1% and 2% and between 20% and 21%. The first is much easier than the second. It's much more difficult to increase an already high cycling rate than to increase a low cycling rate.

The reasons for this are very easy to understand:

The first 1% who cycle are the easiest people to get on a bike. They'll do it anyway, whatever the conditions. They are the people who have continued to cycle in the UK / US despite the difficulties.

The second 1% are the second easiest group to get onto bikes. They only needed a tiny bit of extra help in order to join in with the first 1%.

It gets progressively more difficult as the percentage of riders increases as the demands that those riders make are more difficult to satisfy. If they are not satisfied then they will not cycle.

This is why doing just a bit more than nothing only achieves just a bit more than nothing.

Please stop the hype for shared bikes. It's ludicrous for you to write of a system that does not exist that is will result in "probably at least 30,000 users per day within a year" as this is a far higher rate of usage than other similar systems have.

And your "likely scenario" is also not likely. There simply aren't enough bikes to make a huge difference. Even if each bike was used 3 times per day as you suggest, that would provide capacity for each New Yorker to use the bikes for a two way journey only once every 4.5 years. This is not enough to change people's habits !

Remember also that London's scheme was supposed to turn a profit. They even made a press-release once upon a time claiming that this had already happened. However, it has not actually done so as yet.

Making similar claims for a not yet existing system in New York simply marks you out as gullible.

The US also claimed to be boosting cycling back in the 1970s. For instance, Los Angeles made rather inadequate plans for more cycle provision in 1977 which they never actually built. Their new target is still inadequate, and is promised over another 35 years. This is not progress, it's just rhetoric.

It's very simple. If I see results in a few years time from New York, Los Angeles, London, Paris, Barcelona or any other place which at the moment claims it's just about to have a revolution in cycling, then I'll believe it. Until the results are in for these cities, all we have is hype and supposition, and there is a very long history of all these cities producing plenty of hype for themselves.

btw. Figures from New York show that there are just 24000 cycle commuters in New York. In that context, your idea that there are 200000 people riding each day seems more than a little hopeful.

Finally, please tell what your commercial interest is in bike sharing.

Paul M said...

"There are three kinds of lies: lies, damned lies, and statistics." (Mark Twain, attributing to Benjamin Disraeli).

Talking of statistical error, this may not be hugely significant but I suggest you get your facts straight: your Dutch population density of circa 400/skm is derived from a population of 16,733,300 in an area of 41,848 km2 (that obviously trustworthy source of information, Wikipedia)

According to the US Central Intelligence Agency ( I'll leave you to decide for yourself whether you think they are more reliable than Wiki) the population at July 2011 is estimated at 16,847,007 in an area of 41,543 km2, of which 7,650km2 is water, ie 33,893km2 is dry land. That makes 497/km2.

CIA doesn't provide comparable figures for England, as they only do the UK as a whole, but there we have 241,930km2 of which 1,680 is water - clearly the English results could not be skewed to the same extent.

Dennis Hindman said...

I have absolutely zero commercial interest in bike sharing. I do see it as a way to get people to start cycling, on a regular basis, that would not do so otherwise. Whether bike sharing is done in a city by a private firm, or by government, should make no difference, although if it can be profitable for a private firm, then that would give them financial resources and incentive to expand it at a faster rate than a government would typically do.

Other examples of private operators making money off of public transit would be Transmilenio in Bogata Columbia and a bullet train route in Japan. These two governments provided the capital to build the systems, but the buses and trains are privately operated.

The .25% average increase for modal share in the Netherlands was over a 31 year period. Portland Oregon, arguably the most successful large U.S. city to get people bicycling, had a 400% increase in cycling from 1990 to 2009, for an average of 21% increase per year. Although this is cherry picking the statistics or an apple and oranges comparison

The idea that it was comparably easier for Portland to do this, starting from a low cycle rate compared to the Netherlands, is not born out by the at least 7 years that it took to get to the 1.76% cycling rate in 2000. The pace of increase quickened after that. You might want to read Mia Birk's book, Joyride, to find out what a struggle it was to get any progress, when she was Portland's bicycle program manager from 1993-1999.

Portland's recently announced plan to increase the rate of cycling, by building more bicycle boulevards, was denounced by you as something that will never work, but outside of bike lanes in downtown, bicycle boulevards have been the main bicycle infrastructure that they have built. The increase could have been more if other types of infrastructure were used, but that was not feasible at the time, and they now believe that increasing the density of bicycle boulevards will have a big payoff in increasing the cycling rate.

The 200,000 bicycling rate, that I used for New York City, comes from a highend estimate that Transportation Alternatives made in 2009 by using counts made by the city at six locations and extrapolated that it would be the same population to bike use ratio citywide. A previous estimate by them is mentioned as 131,000 on page 9 of the New York City report on bicycle sharing. A recommended way to get a truely accurate count of the total amount of daily bicyclists in New York City would be to have a count at a location for every 15 thousand residents, per page 5 of this bike count done in L.A. Which for New York City, with a population of 8.2 million people, would be 546 locations. That's not likely to happen any time soon. New York City has been counting cyclists at only six locations and their 2011 count was almost 19,000 cyclists.

I used that high estimate of 200,000 daily bike riders to try and dilute the affect of the bicycle sharing as much as possible in my numbers.

I used what is considered a low usage of 3%, in the above New York City report, to state the potential use. Incidently, the London cycle hire is averaging 25,000 daily users with 6,000 bikes, or an average of 4 daily trips per bike. A high useage average of 10 rides per bike would be Barcelona or Dublin.

David Hembrow said...

Dennis: You make two fundamental mistakes. One is to report numbers in a confusing way:

Calling an increase between 1% and 4% of journeys by bike a "400% increase" makes for a great newspaper headline, but it's very misleading. This is a much easier thing to achieve than moving between 20% and 30%, which you'd describe as a mere "50% increase".

Secondly, You must stop confusing effort with results:

That people have had to work very hard to achieve political support in Portland doesn't in itself mean that progress has been large.

The effort required in Portland was almost all on the political side. I'm sure that people worked very very hard on this. However, the results they've achieved on the ground remain minor infrastructural changes which have resulted in only the minor change in cycling rate to just 6% of commutes by bike.

If they are to achieve a high cycling rate, perhaps 20% or 30% of all journeys by bike, then the majority of the work is still ahead of them.

A few more routes to the same quality as they have now is almost certainly not adequate to achieve this.

This is not to say that what was done there is bad. Clearly it's a good thing that they have progressed. However, it took too long to achieve too little. At this rate it is difficult to keep up momentum over a longer period.

I'm standing by what I said. What Portland has done was comparatively easy. It took a long time simply because the time was not spent efficiently. Rather, the process was bogged down by politics.

Where do you imagine that I denounced anything ? This is just nonsense.

I've seen this myself between Britain and here. Achieving one dropped kerb in Cambridge took me months of writing to a politician, and months for the work to actually be carried out, and of course they didn't remove the railings nearby or make the cycle-lane a sensible width. The politician involved then printed a photo of himself smiling next to said kerb on his next election material in order to claim credit for his improving of conditions for cycling.

Here, just as I was starting to think that a cycle-path near our home could perhaps be resurfaced, it was done to a fantastic standard amazingly quickly without anyone even asking. No-one is trying to win political capital from this. It's just normal.

The difference in political effort required for these two jobs was vast. However, the difference in effect was also vast, and that which took least political effort is easily the best of the two.

And more figures:

Your New York figures are guesses at the high end. So are your London figures. That they have achieved 25000 hires on a peak day is not unbelievable. However, it's absolutely not the average. It's not even what London claim is the average. Actually, London's figures that they felt confident enough to press-release to the world show 2.5 million journeys in six months which is about 14000 per day.

Why not just accept the numbers for what they are instead of trying to hype everything up ?

Only by making accurate measurements and avoiding hype can we know what we have and know how to make progress.

A "400%" change between 1% and 4% is much easier to achieve than a "400%" change between 4% and 16%. Talking about either of these as "400%" is not the least bit helpful.

Unknown said...

Many people in my city try to use the population density argument, however it doesn't hold much ground.

In my neighbourhood, I'm on average around 3km away from anything I need on a daily basis.
The city is so compact that I'm never anymore then 6km away from anything else I need.
In the "north-end" (where I live) this is where the vast majority of people live and where the majority of stores/businesses are located.

Unfortunately because there isn't any space to build within the city anymore, we are starting to expand further outside. Even so, from where I live it's only 6km to this newer area.
It's not exactly cycle-friendly however the roads are exceptionally wide and bike lanes are slowly being added to this part of the city.

My city is so compact and for the most part flat, that I see little reason why cycling couldn't be similar to what it's like in the Netherlands. (plus 90% of the people here are either from or have European roots -- with around 7% being Dutch!)

Dennis Hindman said...

I admit that the way I presented it was confusing. Let me try this in a different way.

If we could instantly create a bicycle infrastructure in Los Angeles, at the same quality of any city in the Netherlands, it would likely not, in the very near future, achieve anywhere close to the cycling rate that there is in a Dutch city. A reason for that is that without restrictions on car use and speed, it would still be easier and faster to get around by car to most people ,and given transportation choices, people will usually choose the fastest and easiest method to get somewhare. Then there is the problem that the city of Portland has presented, that about a third of their population would never ride a bike. It's also incredibly cheap to own and operate a car in Los Angeles compared to Europe.

Steve Jobs, the co-founder of Apple, was asked in a interview in 1990, on a PBS show called Nova, what would it take to move from a current computer use of about 20 million to 100 million. Steve's answer in essense was that many of the older people would have to die off and kids in school would need to have access to a computer to get a much larger supply of users.

A Dutch rate of cycling in Los Angeles would similarly require that the older, permanent non-users be replaced by children who have had bicycle safety training at school. That would take years to achieve, even with instantly having a Dutch style bike infrastructure.

A couple of reasons that Portland started getting a much larger increase in bicycling after 2000 was that they were beginning to have a connected network of bicycle infrastructure. Another reason was that as people saw a lot more bicycling, they felt comfortable enough about the idea of cycling themselves. Group think as you will.

Last year, Portland announced that they were going to spend an average of $30 million a year, over the next 20 years, to triple the size of their bike network and to achieve a goal of a 25% bicycliing rate They plan to add 314 miles of bike lanes, buffered bike lanes, cycle tracks and 256 miles of bike boulevards.

Portland has no idea how they are going to find the money to accomplish this goal. The political committment is there, but not the money. Would the city council have agreed to an ambitious plan like this with a 1% commuter cycling rate, instead of the approximate 6% rate at the time of the vote? I would say that the chance of that happening diminishes greatly with a much lower rate of cycling.

If Portland had started out building protected bike infrastructure on busy streets, it would have cost a lot more money per mile and almost certainly have taken a lot more time to achieve the same miles of bike infrastructure network. The cycling rate would likely have been less, as the network would have been much smaller, and perhaps disconnected, due to the limited funds. There also would have been much more resistance to putting in cycle tracks compared to bike lanes. There is also the not-so-small problem of protected on-street bicycle lanes or cycle tracks, to this day, are not approved for use in the manual of uniform traffic control devices or MUTCD, which is the bible to most traffic engineers in the U.S. On-street cycle tracks, or protected bike lanes, are still considered experimental in the U.S. and have only been used in a few cities thus far.

Bicycle sharing and Ciclovia type events are other ways to move the bicycling rate up. Both of these tend to be politically popular and can be used like a snow shovel to reduce the resistance to putting in more bicycle infrastructure. Neither of these methods will directly create high Northern European cycling rates. But, without them, the time to achieve higher cycling rates will likely lengthen. They are simply more tools in the tool-box to keep the cycling rate growing.

Trying to quickly achieve Dutch cycling rates in the U.S. is like trying to get across the Grand Canyon using only popcicle sticks. I'm not saying it's impossible, just that it would take some time and be difficult to do.