The easiest place to paint a new bike lane is on an existing wide outside lane (WOL) street. These streets are most likely to be the first to get painted, the “low hanging fruit” as it were.
In many metropolitan areas, the roads that have WOLs are also busy, high-speed arterial streets with a lot of commercial truck traffic. These are not attractive streets for non-cyclists. Painting a bike lane on them will not make a newbie any more likely to ride there. (The targeted user class.) Having trucks roar by at their elbow isn't likely to boost their traffic confidence.
Bike lanes do not legitimize a cyclist's right to the road in motorist's eyes. It tells them that bicycles belong on the edge of the road, out of the way. Should a cyclist need to venture away from the curb, he is considered a scofflaw for leaving the area that the authorities have -at great expense- indicated where he should be riding his bicycle. This perception persists even if there is no bike lane on that particular road!
By indicating that the far right position is where cyclists belong also instructs bicyclists to ride in safety compromised positions on roads without Bike lanes, even where riding further away from the curb is warranted. Motorists get the notion that there is always enough room to pass a cyclist when he is at the edge of the road, and bicycle operators often position themselves poorly at intersections and junctions.
Bike lanes are not the clearest, nor are they the most effective means of conveying the idea that bicycles are legitimate road users. The side effects of “bike lanes as education” are too costly. There are better ways to encourage bicycling.
In what other part of the transportation system are traffic control devices used as advertisement?