Bike-o-nomics: Bike Lanes, Biking and Causality

It is not hard to imagine that building more bike lanes encourages biking.  As I was being driven through Coconut Grove and Coral Gables on Friday I was struck by both the super aggressive drivers and the complete lack of space for bikers.  Needless to say I saw very few, even in and around the campus of the University of Miami.  When I mentioned this to my host, who is an old friend and economist, and speculated that the lack of infrastructure (walking too - no sidewalks anywhere!) probably explained it she said "well, it also gets super hot for a lot of the year and if you try to walk or bike you get immediately drenched in sweat."

Which goes to show that the causality probably goes both ways, no bikes because of no infrastructure and no infrastructure because no bikes.  So it was with some interest that I found out about this paper through Joseph Rose's twitter feed. I had hoped that someone had done something clever with the data to identify causality.  Especially since Joe had tweeted: "Guess What? More Bike Lanes = More Cyclists."

Sadly though, this was not the case. After a long and detailed paper looking at controlled correlations, the authors fess up:

The cross-sectional analysis in our study aims at explaining differences in cycling rates among cities but cannot be used to predict changes over time. Moreover, as in any cross- sectional regression analysis, none of our models can prove causality, although the sig- nificant associations we measured are consistent with the hypothesis that bike paths and lanes encourage more cycling. Our analysis is also limited by its reliance on aggregate, city-level data, which mask variations within cities, among neighborhoods, and individ- uals. The results suggest a statistically significant relationship between bike paths and lanes and cycling at the city level, but results do not permit conclusions about individual travel behavior.

In addition to the inherent limitations of cross-sectional regression analysis and aggregate data, there is a problem of endogeneity among some of the variables in our models. Cycling levels and the extent of the bikeway network almost certainly affect each other, so that causation is probably in both directions. In this paper, we have focused on the role of bike paths and lanes in explaining variation among cities in cycling levels. Conversely, however, high cycling levels might help explain the provision of a large supply of bike paths and lanes. Endogeneity and simultaneous equations bias are potentially serious problems in our regression analysis because the key explanatory variables—bike paths and bike lanes—are also a function of cycling levels, the dependent variable.

I give the authors all the credit for their honesty, and I think there is utility in their paper - they uncover some interesting correlations.  But I fear it will be used by transportation advocates as proof that infrastructure creates more bikers.  I believe this to be true, but this paper not really help you out at all in this argument.  Which is the problem, by the way, with being an economist. This paper would not get published in any decent peer-reviewed economics journal because no one wants to hear that you did not do anything to deal with the causality issue.

Surely there has to be some clever instrument for the provision of bike lanes?  Can anyone think of any? Some kind of natural experiment?  Something? Let's nail this causal link people!

As an aside, the only people I saw biking in South Florida (in delightful weather mind you, high 70s, breezy, low humidity) were folks who seemed to be on bike for lack of other transport alternative not those choosing bikes over other alternatives.  Interesting - wonder if there is a stigma attached to biking in South Florida of if it is just the weather and poor biking infrastructure.