Crunching the Crime Numbers: Too Many Cops or Not Enough?

A new study, "The Effect of Police on Crime: New Evidence from U.S. Cities, 1960-2008,"recently applied a sophisticated statistical analysis to two sets of data that are assumed to be closely linked: the number of police officers per capita in a community and the amount of crime that community will suffer.

Aaron Chalfin and Justin McCrary, both from the University of California at Berkeley, ran the federal, state, and municipal numbers for 50 years of crime and policing and concluded:

• If cities want less violent crime, city councils should hire more police. Although seeming common sense, other studies have been skeptical that violent crime rates respond to the number of police officers on the municipal force.
• However, correlations between the size of the police force and the number of property crimes -- burglaries, auto thefts, larcenies, etc. -- aren't as clear. Property crimes in a community aren't markedly reduced by adding more police to the force.
• Overall, a bigger police force is a significant return on investment. Each $1 spent to expand the police force results in about $1.60 in reduced victimization costs. (To determine those costs, the researchers sought to calculate the value of all of the kinds of losses suffered by crime victims.)
• That favorable ratio, however, is skewed by local conditions. In an affluent, low-crime Sunnyvale, every new dollar spent on increasing the number of police reduces per capita victimization costs by only 20 cents. In distressed, high-crime Gary, Indiana, every new dollar would cut victimization costs by $14. The difference is due in large part because police services in Gary are less than $100,000 per officer and Sunnyvale's are almost $300,000.

That finding led the study's authors -- and the news media -- to conclude that some cities might be "over policed," because they pay so much more per capita for law enforcement and get only marginally lower victimization costs.

Based on crime rates, income levels, victimization costs, and policing levels, 12 of the 30 most "over-policed" medium-to-large cities in the nation are in California. Torrance ("over policed" rank 3), Santa Monica (7), Burbank (10), and Glendale (28) are in Los Angeles County. San Bernardino is among the 30 cities -- like Gary, Indiana -- that the study concluded are woefully "under policed."

A closely reasoned cost-benefit analysis from UC researchers isn't likely to persuade city councils to add or subtract police officers from their law enforcement ranks. The study points out that the decision to expand or shrink the number of police officers over the past 50 years has not been closely linked to "demographic factors, the local economy, city budgets, measures of social disorganization, and recent changes in crime rates."

Which suggests that the choice to hire more cops or lay some of them off is driven by factors that aren't reasonable: federal and state funding policies, the power of police unions, the politics of city council members, and the expectations of homeowners.

But then, who could have expected a rational analysis of our fear of crime?

D. J. Waldie, author and historian, writes about Los Angeles twice each week at KCET's SoCal Focus blog.