I have a couple of comments on Matt's proposed solutions. First of all, cameras. You're right in that I gave it short shrift in writing the essay and undeservedly so. What follows is stuff that I probably would've included if there were a higher word limit. I'm guessing you're referring to the speed cameras that are everywhere in Montgomery County. I concede that they're great speeding deterrents -- for the half mile or so of road that they cover. Everyone I know slows down when they approach a camera zone, sometimes to the point where you see pedestrians whizzing past them, but upon exiting the zone, they put the pedal to the metal and go back to moving at whatever speed they were traveling at before the cameras. Nonetheless, this works in the context of speeding when the cameras are placed in certain areas, e.g. near schools where children cross the street and whatnot, where a half-mile zone of non-speeding is enough to produce a significant improvement in public safety. But getting a camera to read a vehicle's speed is easier than having it detect the number of occupants in a vehicle. Some factors that make detecting the number of occupants in a vehicle difficult (factors that don't complicate reading vehicles' speeds) are the speed of the vehicle, less than optimal lighting conditions, obstacles like headrests and windshields, and the size/position of occupants (e.g. babies and nappers). Infrared technologies are riddled with similar problems. Even if an accurate technology were to develop, it'd probably be prohibitively expensive, especially given that there are thousands of miles of HOV lanes in the US. I won't get into the images of an Orwellian society conjured by a mass surveillance system of cameras that take pictures/video of occupants in vehicles.

I'm not sure how much it costs to install fake cameras along a highway, but I'm guessing that would be pretty expensive too since you'd need enough of them to prevent people from simply ducking in and out of lanes to avoid them. Likewise with empty police cars. Violators can switch out of the HOV lane when they see a cop car and switch back in when it's safely out of sight. Barriers that make it harder to switch in and out of the HOV lane might prevent all this, but they have their own problems. Commuters also usually travel the same way everyday, so they'll probably catch on quickly if all they see are empty cop cars.

I think, in the end, most modes of enforcement can't be justified based on their cost and the relative harmlessness of violating HOV rules. Yes, hefty fines give police an incentive to really crack down on HOV violators, and the income from the fines subsidizes such methods. However, there's only so much a cop can do on a highway of fast-moving vehicles. While a cop chases one violator, others go unnoticed. You'd need a shitload of cops to catch a sizable number of violators. It gets to a point where you have to ask, is it worth it? Social controls are low-cost and low-maintenance (once embedded in people's minds), and have the potential to reach the entire country. -Brandon