Overall, the Netherlands and Denmark have 1/4 the rate of bike accidents and fatalities than North America has.

I think separated lanes have a lot to do with cycling popularity. Nobody feels comfortable with big behemoths right behind you.

Convenience and cost are big factors in encouraging cycling, especially among the young. It is hard to combat the comfort advantages of cars, especially in foul weather.

tOM

I think that part of this “safety in numbers” idea is the implementation of traffic calming measures, which I would consider improvements in infrastructure. For safe and efficient routes in a city, we need roads (not necessarily cycleways) with slower traffic, room for cyclists and drivers that are aware that they are sharing the road. It becomes a bit of a chicken and egg game- we need more cyclists on the roads to raise awareness and help slow traffic but we need safe conditions to encourage more cyclists.

It certainly isn’t the infrastructure. When studies have been done of the before and after effects of infrastructrues in cities that already have very high bicycling rates, such as Copenhagen, it is found that all of the facilities increase the probability of accidents for bicycles, even when normalized by changes in the number of bicycles and automobiles due to the infrastructure.

I believe, instead, that this “safety in numbers” effect is almost totally dependent on automobile traffic speeds – I see this correlation in most European cities. So my theory is that as more bicycles are used on the roads, they act as “speed bumps”, slowing down overall traffic.

The events in London match my theory very well.

In London, the congestion zone reduced the number of automobiles, allowing traffic in the zone to go faster. The increase in bicycling didn’t add enough speed bumps to counter the lower density of automobiles, so the serious bicycle/automobile accident rate went up.

1) Safe cycling infrastructure;
2) Efficient cycling infrastructure (that takes you where you need to go); and
3) High cost of automobile use, whether by taxes, parking fees, congestion charges, or both.

London and Stockholm have (3), but they’re still working on (1) and (2), and without them, the demand for cycling will probably be limited. Most American cities have none of these. For that matter, it’s really only a relatively few cities in the world that have all three, and even then it’s rather ironic that the best these cities can do is roughly a forty-to-fifty percent mode share. People love their cars!