Economics of CrimeA Standard Application

Gary Becker's famous article on the economics of crime and punishment postulates that potential criminals assess both the probability of being caught and punished and the expected size of the punishment.

Early work on the economics of crime by Becker, Shavell, Polinski, and others wondered whether, if criminals are risk-neutral rational maximizers, it could be more efficient for society to get the same deterrence by reducing the resources used to detect and capture criminals but by increasing the expected size of the punishment. If the probability of capture and conviction decreased, but the punishment increased, then the expected costs for the potential criminal would remain the same, but society might be able to get by with fewer scarce resources devoted to law enforcement. For various reasons and for various crimes, it didn't work [see here for one; and here for a more formal reason].

The upshot has been that more resources are again flowing back into front-line detection, arrest, and conviction to increase the probabilities that people will be caught and punished. In a very simple application of this principle, the City of Ottawa hired more traffic enforcment officers and has begun issuing more traffic violation tickets (primarily for speeding, but also for other traffic violations).

Late in 2003, police received approval to hire 18 new traffic/escort officers, bringing to 30 the number of officers dedicated to traffic enforcement. The program began Jan. 4, 2004.

The 18 new enforcement positions has accounted for one-quarter of all traffic tickets issued, an average of almost 200 per officer. The unit, in total, was responsible for 29 per cent of the 51-per-cent increase in traffic fines, meaning other front-line officers picked up 21 per cent of the total increase.

The effects of the increased resources devoted to increasing the probabilities of detection and conviction have been very high.

Almost 51 per cent more tickets were handed out last year, from 86,917 in 2003 to 130,416 in 2004. The 2004 figure is double the total for 2002.
About 70 per cent of all offences are for speeding.

Furthermore,

Fatal crashes for 2003 to 2004 dropped 18 per cent, from 33 to 27. ... In 2004, there were 18,000 serious collisions, down by 1,500.

The reduction in auto fatalities and serious collisions may be due to many other variables, but it appears to have been substantial.

Keep in mind that the punishment for moving violations involves not just the $100 - $400 fine, but also the increased auto insurance premiums that follow a moving violation conviction. Nevertheless, the size of the punishment has not changed in Ottawa. Only the size of the expected punishment has changed in a probablistic sense. And it appears to have been effective.

Was it cost-justified? That depends on the value you attach to human lives, human time lost due to accidents, etc. My hunch is yes.