, , , , , , , ,

Hi readers, I wrote this piece back in May, but because of the hectic nature of moving from one continent to another, I neglected to post it until now. All of this is still true, but now I’m much closer to publishing the results of my work abroad. Stay tuned!


When I started to make sense of the Vélib’ and bikeshare literature that I had been reading back in February, I had approached the material poised to do a holistic system evaluation of Vélib’. In fact in 2011, that’s what I pitched to Fulbright that I would undertake during my eight months in Paris. Yet, after reading a lot about Vélib’ and reaching out to other researchers, members of the former users committee, and the system itself, it became more and more evident that I lacked the time, resources, and access necessary to perform such a study. Harrumph.

In my defense, evaluations of bikeshare systems — especially those in France operated by private advertising companies– have proven to be quite difficult. In particular, without adequate access to financial information, it’s hard to see if there is return on investment on the behalf of the operator. That said, fiscal health is only just one part of a bikeshare evaluation; whether the service is meeting demand, subscribers are being retained, car trips are being avoided, and streets becoming more bike-friendly and less congested are all equally valuable considerations with their own requisite data sets that are often difficult to obtain, if they exist at all.  And just how to measure them? Well it seems to me that the jury is still out on a lot of those fronts. As far as I can tell the bikeshare scholarship has yet to adopt measurement techniques that can be universally applied to bikeshare systems to calculate system efficiency, return on investment, readiness for expansion, etc. Maybe after Capital Bikeshare releases the results of its internal evaluation, similar studies will become easier to replicate for other systems.

(Perhaps such measures exist but they are proprietary? Or perhaps I’m being dense and I don’t know where to find these best evaluative practices? Dear readers if you have any insight here, please feel free to send your comments my way.)

All this to say that an evaluative study of Vélib’ could be possible, but it would require a better equipped researcher than yours truly. So back in March, I found myself wondering, what should I do with my time, my Vélib’ knowledge, and my tax-payer funded stipend?

Travel! Well, over the course of reading many papers and articles about Vélib’, I found myself thinking often, yes, it’s cool that there are 20,000 Vélib’s rolling about the streets of Paris, and yes, it’s cool that Vélib’ can boast of almost 250,000 subscribers, but just who are these people pedaling the Vélib’s? Did they ride their own bikes before they could rent them from the Mairie de Paris? Did they give up their cars to lead a low-gasoline lifestyle? Do they tend to be students? Thirtysomethings? Do they live in the same neighborhoods? Where do they ride? Are they mostly commuters? Who are Vélib’s users?

As I see it, a transportation system that is built in part to effect a behavior change, should be evaluated on the basis of its ability to change said behavior. Now I’ll be the first to admit that it is overly reductionist to base the totality of a program’s worth on its ability to convert target audiences into users, but it’s a pretty critical component. Consider if all of Vélib’s annual subscribers rode their own bicycles which they promptly abandoned when Vélib’ became available. In such case, the net effect is an increase in public expenses and increase of greenhouse gas emissions (from the trucks that redistribute the bikes) from before. In other words, who the users are, and what their habits were before they became bikeshare users makes a big difference in the efficacy of the system.

Take, for example, the bikeshare users of Lyon. According to the a survey of Vélo’v  users, in the absence of the bikeshare system, 51% of them would have used public transportation for their trips in question, 37% would have walked, 7% would’ve driven, 2% would have used personal bicycles, and 2% would not have taken the trips,” (“Coûts et avantages des VLS: un bilan socio-économique équilibre” PDF). From this, we can wager that in Lyon, bikeshare is primarily replacing trips that mostly would have been previously served by walking or public transport and not cars. That’s not necessarily a bad thing — it’s possible that those users can save more money, get to their destinations faster, reduce transit congestion, and burn more calories by taking Vélo’v as opposed to walking or taking public transportation — but from this survey sample it’s not evident that bikeshare is significantly reducing car trips or carbon dioxide emissions in Lyon.

Are Vélib’ users the same? What’s their demographic profile? These lingering questions have become my new focus, and what I will produce a paper on by the end of this summer. Of course it would be too easy if Vélib’ provided system data for this kind of demographic data. They don’t release it — at least as far as I can tell — but they did conduct a few publicly available user surveys and recently open their trip data. And the Mairie de Paris produces an annual report on the state of transportation in Paris, which shows trends in how residents get around the city. Hopefully this will lead to an interesting profile of just who Vélib’s users are and how they have shifted the transportation trends in Paris over the past five years.