Red Light Cameras, Money, and Safety

No great suprise, but red light cameras seem to cause crashes. T-bones are down and rear-ending is up, but overall, crashes at cameraed intersections in DC rose about 67% faster than at intersections without cameras.

From an economics standpoint, this should make sense, and it seems a reasonable guess that the crash increase comes from the drivers who would otherwise have been the safest drivers.

For a driver who is already willing to risk the costs, burdens, and pains of a collision for some other gain by saving time (costs x risk < anticipated benefit), racing a light makes economic sense, and the driver will do so. There will be a slight decrease in that number when you add the cost of the ticket (costs x risk + $75 < anticipated benefit), but only if the driver knows about the cameras. The presence of tickets in the equation will tend to shift red-light racing to people who are paid large hourly fees (increasing the anticipated benefit) and to people from out of town (decreasing knowledge of risks).

For a normal rational driver, the costs of getting a ticket will be weighed against the other factors such that there is a slight premium on the side of caution, but one which may be outweighed by other factors (e.g., if there is a bus behind him whose brakes have failed, the $75 fine is a small price to pay to avoid being hit). For him, the equation for racing a light becomes something like [Costs (of being hit running the light) x risk (of being hit running)] - [costs (of being hit NOT running the light) x risk (of being hit while stopping)] + $75 < anticipated benefit.

For someone who is normally a low risk-taker in such situations (the "safe" driver or someone whose "dad will kill me if I get a ticket"), someone who has abnormally high costs involved in being rammed in the side (e.g., Pinto drivers ), or someone who does not anticipate the risks of rear-end collisions, the same equation will lead them to decide to stop with greater frequency. However, those with abnormally high costs will not have much added incentive to safety because they already are so strongly inclined to avoid those costs that the $75 ticket will not, by percent, add much to the costs side of the equation and will so not generally lead to many more situations in which they will stop for the light. For a driver with few anticipated benefits or the greatest shame in being ticketed, however, the nearly-certain prospect of the ticket will have the most effect, meaning that the people most consciencious about keeping (and not being caught breaking) traffic laws which are emphasized by enforcement are the ones who suffer the most under red-light cameras because they are now exposed to an inordinate number of collisions, particularly of the rear-ending type.

Of course, if, as the article hints, the District is more interested in revenue from the cameras than in safety, it would make sense that they would not care which drivers are the ones who pay the costs of their enforcement regime.