If I had my car stolen, the cops had the perp in their sights and they just let him go, I’d be hopping mad, and I’d be right to.

Even if not chasing them produces higher short-term utility, the incentives are just wrong.

Moreover, eliminating the police pursuit indisputably eliminates the danger that the police cruiser will accidentally kill a bystander, as has happened in the past, here in DC.

As for my gunfire analogy, I read you to be essentially saying that it’s a difference of degree. Cops can outshoot most criminals, too; they’ve generally got more and bigger guns available. The issue is the likelihood of collateral harm to innocent people, which is the same issue at stake as to car chases. I do agree, though, that shooting into a crowd certainly carries much worse odds for bystanders, which weakens the analogy.

Your point 3 has validity, but only if the criminal continues to drive recklessly when not being pursued. As I mention above, I think a rational actor generally has incentives not to do so.

I should clarify that I’m only arguing about the public-safety effects; I do not support the award of damages from the city to the bystanders in this sort of case. This seems like a quintessential case for immunity.

One other harm in not chasing is that sometimes that apparent traffic offender or car thief is actually a much more dangerous criminal, who will hurt or kill people until he is stopped. E.g., many steal cars just to drive them around a little, some steal them to sell to a chop shop (and may be carjackers who are indifferent to the dangers to their victims), and some steal them because they need a getaway car that can’t be traced back to them for high-risk armed robberies. It’s unlikely that a cop who lets a car thief get away is actually letting a serial robber-murderer get away with the evidence of his latest crime, but it’s also pretty unlikely that any innocents will be killed if he chases the car…

Problem with that sentiment: you CAN have it both ways! That is, you can both let the offender go AND get dead civilians.

The way the offender gets away is by speeding and reckless driving, etc. The only effect the police have on that is HOW LONG it gos on and if the person is rewarded (they get away) for the behaviour.

You get feewer CHASES, but not fewer incidents of reckless high speed (in fact, you’d almost certainly get more, as any criminal with half a brain and any confidence in his driving ability would do so). So, you effectively get the worst of both worlds: more time spent at reckless high speed AND fewer criminals caught.

Yay! Let’s do THAT! [/sarcasm]

Oh, and Ted’s last post to Tom (about Tom’s analogy not being applicable) is spot on – I was going to post something similar (but worded less succinctly, I’m sure).

1) Police turn on flashers;
2) Criminal runs;
3) Police follow;
4) Criminal decides whether to give himself up or speed up.

I see where you’re going with the slippery slope argument, but there are critical factual distinctions:

1) Cops can outdrive criminals. They’re driving powerful Fords designed to handle precisely these situations, and have the power of numbers. Very very few criminals escape in a car if the police decide to give chase; when escapes do happen, it’s because the criminal abandons the car and goes on foot.
2) Except in the movies, a cop can’t fire into a crowd and hope to stop a criminal: even one-on-one, cops hit maybe 40% of their shots. Thus, changing the policy to allow cops to fire into crowds would not greatly change criminals’ risk calculations in deciding whether to escape into the crowd.
3) The criminals’ decision to speed endangers the public even if the cops give up the chase in response. The same is not true for the criminal escapes into the crowd and police holster their weapons scenario. Thus, the effect of incentivizing the “try to escape” strategy has a different effect on public safety.

Uh, Justinian–it was the CAR THIEVES who swerved into the pedestrians wile running from their criminal act. Why are the car thieves not the ones on the hook for civil damages? [I assume there were charges of vehicular assault and reckless driving, or even attempted murder{?}, here against–at least–the thief driving!] I know, I KNOW . . . @%$%**@$&# deep pocket theory!!!!!

Seems to me this is “which came first, chicken or egg”? Thieves steal SUV, police find thieves in SUV and give chase, and thieves while trying a getaway run into a crowd of pedestrians. How do police end up being found negligent and pay out money?

Let’s see . . . thieves DON’T steal SUV, cops AREN’T looking for them, DON’T give chase, thieves DON’T swerve into crowd, NO ONE sues to get money from anyone. Make sense?

I’m not shouting about a slippering slope, here, just that the idea of disarming the entire population, and restricting sidearms to an obviously idealized (uncorruptible, undiscriminating, universally responsible, and rapidly-responding) police force is a program that, realized, wouldn’t satisfy its proponents because they still misunderstand who (and not what) is responsible for gun deaths, just as these unfortunate pedestrians don’t comprehend who it is that endangered them.

I’m a pedestrian, myself– I’m very sympathetic, because I’ve had some close calls by drivers, and I know I can’t dent a car that’ll break every bone in my body. I also know, though, that a driver’s only control on other vehicles is to limit the speed of the guy behind him. You can’t make the guy in front go faster unless you shove him.