The Network Effects of Biking
Urbanists and transit planners know that the more people who bike in a city, the safer its roads (not just for cyclists, but for all road users). Since perceived danger, or lack of safety, is the number one reason cited in most surveys for why people don’t commute by bicycle, increasing safety for cyclists drives more people to cycle, which further increases safety, and so on. As each person joins the ranks of the cyclists in a city, cycling there gets safer, and thereby more valuable, for all of them.
There are a number of reasons why this might be the case, but the basic idea is that when drivers expect to see a cyclist they slow down and drive more carefully across the board. This is why, contrary to most people’s intuition, I find riding a bike on a suburban road to be infinitely more stressful than on the gridlocked streets of New York City.
In addition to cycling becoming safer, the more people bike, the more culturally accepted cycling becomes. The idea of commuting by bike become a lot more attractive when it is associated with people that look like you and not, say, Steve Carell’s character from the 40-Year-Old Virgin.
Both of the above are important factors for bike-sharing companies to consider in their city selection and rollout strategies, as well as for cities that are considering whether to allow bikesharing, both docked and undocked.
Bike sharing contributes another dimension to cycling’s network effect: bike-sharing operators collect a plethora of data on how people use the system to move around the city — origin-destination pairs, popular routes, times of day, etc. — and unlike the TNC space, the culture of the bike share industry is to share this data with cities to enable them to improve their cycling infrastructure e.g. protected bike lanes, traffic calming devices, bike racks, etc. Better cycling infrastructure, precisely targeted where the need is greatest, leads to even more people biking on both individually-owned as well as shared bikes, which leads to bike share companies collecting and sharing more data with the city, which leads to better infrastructure, etc.
Unlike many network effects, however, this appears to be one that accrues to the industry and to the city as a whole. It remains an open question whether any individual company can gain a network effect, one that would appear to be answered in the negative. Could this be an argument for awarding an exclusive franchise? This may be what happened initially with docked bikesharing, but with dockless it seems like it is simply too difficult, whether politically or operationally, for municipalities to keep providers out, without even getting into the question of the quality of service that this sort of arrangement tends to result in (see under: cable companies).
One implication is that if American cities want to encourage cycling as a replacement for single-occupancy vehicles in overly congested urban centers, and are willing to commit infrastructure dollars to it, the virtuous cycle — no pun intended — implies that they have the levers to be able to significantly grow the share of cycling as a transportation mode.
This is a bit unrelated, but I was just listening to a podcast in which one of the guests suggested that AVs were going to have their largest impact on walking and cycling, because of how much safer they would make streets. It’s an interesting thought experiment: when all urban vehicles are automated, will it give every city the equivalent of the cycling infrastructure of a Copenhagen or Amsterdam, and if so, does it have the same impact on mode choice as it has there?
