Balkinization  

Monday, January 10, 2005

stop thief!

Ian Ayres

Years ago, Steve Levitt and I did an empirical study about the impact of Lojack on auto theft. The big surprise was that Lojack had a big deterence effect. For every dollar you spent on Lojack you reduced your neighbors' expected auto-theft losses by about $10.

People spend too much money on "the club" and traditional car alarms that mostly just shift crime they don't reduce it, while Lojack reduces crime because it gives the police a chance to catch the crooks -- the police clearance rate on cars stollen with Lojack is about twice as high as on cars stollen without Lojack.

Levitt and I made a big point out of saying that Lojack doesn't stop your car from being stollen because the thief doesn't know that you have Lojack (it's a hidden transmitter), But in coming back to this issue with Barry Nalebuff, I now see that Lojack does stop your car from being stollen -- because it stops you from stealing it your self. Lojack is an anti-fraud device. Turns out that maybe 20% of all theft is fraud (people steeling their own cars). But the fraud rate in Lojack cars is probably much much smaller. You'd have to be crazy to steal your car with Lojack in it, because you'd know that the police would find you. The commitment not to commit fraud should dreduce the insurance premia that Lojack owners pay on their cars (but doesn't sufficiently).

Another 20% of autotheft may occur because people leave their keys in the car. Leaving your keys in the car isn't just stupid, it's inconsiderate. 20% of police time on autotheft is spent on these crimes, 20% of your insurance premia is because other people leave their keys in the car.

So stepping back, we can see that a lot of auto theft is caused because some people don't take enough precaution (they don't take their keys) and a lot is caused because some people take too much precaution (they use the crime-shifting club). Both of these behaviors are socially inconsiderate and there are simple things that we could do to induce better precaution taking. Barry Nalebuff and I write about several of them in the latest issue of Forbes:

Removing the key . . . impose[s] small inconveniences on the
owner but yield large benefits to society.We need to change social customs.
Folks who make life easier for thieves are imposing costs on the rest of us. The
damage is just like that from secondhand smoke.Modesto, Calif.,which holds the
dubious distinction of having the nation’s highest per capita auto theft rate,
is considering ticketing people who leave their keys in the car.


Comments:

I loved the suggestion of ticketing people who leave keys in their cars, especially if it is true that the net costs of neglected keys (including those externalized onto others) make it worthwhile. I also agree that states should mandate insurance discounts for Lojak. But I had some questions about this. My thinking is probably a mile off, so feel free to shoot holes in it. There must be some positive externality to this; a thief is much more likely to take the car with keys in the ignition, than mine without. The 20% extra I pay for my insurance, then, is also partly a premium I pay for the fact that thieves steal those other cars instead of mine. I see why the Club is inconsiderate of others. But leaving my keys, apart from raising the crime rate, also volunteers the car for theft; which is very considerate to my neighbors, if theirs is LESS likely to be stolen as a result. It's like the opposite of the Club in that respect.

The question for me is whether the 20% would give up car stealing completely if no keys were left in ignitions. It seems possible that thieves prioritize cars with keys there, but in the absence of them, resort to hotwiring (or mugging me for my keys outside my car). The model seems to assume that those 20% would not have been stolen without volunteer keys. I understand that there must be SOME increased crime as a result of having easier targets around, but the 20% is not necessarily that number, is it? And it would seem that this cost needs to be offset by the bizarre but valuable service that the volunteer provides to the rest of us - a decreased risk of our own cars being stolen.

This leads me to a related question: wouldn't the ideal theft reduction strategy be for the police to leave cars sitting around with keys AND Lojak? This would have a triple beneficial effect: first, the police could easily catch these thieves, getting them off the streets and into jail. Second, the thieves are (at least for awhile) more likely to steal these decoy cars than my keyless car, so the police would have steered the theives away from innocent victims. Third, decoys create a "lemons" market for stolen cars over time - just like the deterrent effect from Lojak making every car a "risk," using decoys adds to this. Thieves would eventually become aware of the risk that some cars are police decoys, which lowers the value of all the cars they could steal - stealing becomes less worthwhile. Perhaps they police should start with a few Escalades as decoys.
 

I disagree with characterizing using a "club" as a classic "negative externality" or with the statement that it's "inconsiderate of others." (I should say here that we use a club for our car). At that rate, anything shy of leaving the doors unlocked and a mint on the driver's seat imposes inconsiderate negative externalities on society.

I understand that you mean clubs to some extent simply re-route thieves to pick easier things to steal, in this case cars. But so do a lot of other passive measures commonly used to deter crime: bars on windows, locks on doors. Indeed, if/when effective detection systems/countermeasures for lojacks are devised, they'll just be another "club".

But even in the meantime, I wish everyone in my neighborhood used clubs: then our neighborhood would be viewed as a tougher nut to crack than other, [REVISED: less security-minded :)] neighborhoods. Instead, car thefts are rampant, and the stolen cars are generally un-clubbed. So my neighborhood's failure to use clubs imposes a negative externality on me -- higher risk. I mitigate that with a club -- lower risk. I've generously decided not to charge my community.

Given that clubs are relatively cheap, there is little preventing anyone else using a club as well. This may eventually slightly reduce the utility of mine to me, as the crime-shifting effect goes away, but at the end of the day it's simply another physical obstacle to crime.
 

Two small points:

(1) You write, "a lot of auto theft is caused . . . because some people . . . use the crime-shifting club." This is, of course, internally contradictory (if the club merely shifts theft, it can't also cause it).

(2) "premia": unless you plan to write your entire post in Latin, you may wish to consider using the English word "premiums."

Keep up the good work!
 

Nice blog! I have a engine free generating generator lead search system traffic training site I thought you and your visitors might like.

Click on engine free generating generator lead search system traffic training to check it out. engine free generating generator lead search system traffic training
 

This web site is really a walk-through for all of the info you wanted about this and didn’t know who to ask. Glimpse here, and you’ll definitely discover it. Agen Bola Online
 

Post a Comment

Older Posts
Newer Posts
Home