Declining Crime

It's well known that crime declined significantly in the 1990s and early 2000s.  And thanks to Steve Levitt, we have a pretty good idea of why.  (See his "Understanding Why Crime Fell in the 1990s: Four Factors That Explain the Decline and Six That Do Not," 18 J. Econ. Persp. 163 (2004).)  The Economist magazine of June 9, 2007, has a lead article in its United States section on current crime trends.  That article has some fascinating new information   For example, although many large cities continue to have historically low levels of crime, a few have seen recent spikes.  The three largest cities -- New York, Los Angeles, and Chicago -- have had continuing declines in serious crime.  Los Angeles (which has had the same police chief, William Bratton, since 2002 as New York had in the 1990s) has experienced remarkable recent drops: burglaries down by 20 percent, murders down by 50 percent, and serious assaults down by more than 33 percent.  According to The Economist, LA is now safer than Salt Lake City.  Oakland, California, by contrast, had a 50 percent increase in homicides in 2006 over 2005.  Philadelphia had 406 homicides in 2006, back to the level of the early 1990s.  Homicides increased in U.S. large cities by 7 percent between 2000 and 2006, but not in the three largest cities, all of which had significant declines.  Indeed, if the declines in New York, Chicago, and Los Angeles had not been so large, the increase in homicides nationwide would have been 11 percent.  Why these differences? 

One thing that does not differentiate New York, Los Angeles, and Chicago from other large cities is the number of police.  (A significant increase in the number of police was one of Levitt's four factors that helped to explain the decline in crime in the 1990s.)  New York's police force has 4,000 people fewer today than in 2000.  Chicago and Los Angeles also have smaller police forces. 

What the three cities share, according to The Economist, are two practices and two demographic facts: first, a new method of allocating police called "putting cops on the dots": second, giving local police commanders responsibility for their own territories; and third, a declining percentage of young and African-American populations.  The "dot" idea, pioneered in New York, is simple.  The police have mapped each incidence of crime by putting a dot where each particular kind of crime occurs.  Where the dots are most numerous are clearly areas that need more policing.  So, New York, Los Angeles, and Chicago are deploying their police in greater numbers where the incidence of crime is greatest.  (However,the number of dots may not be large. Los Angeles deployed more officers to only 5 hot spots last year.) 

Now, i confess that I find this to be an astonishing fact -- astonishing in the sense that it defies credulity to think that only in the last six years have police departments learned where crime is most likely and decided to put more police there.  I would have guessed that this practice was one of very long standing.  But apparently not (and for reasons that it would be interesting to know).

The second practice is local responsibility and accountability.  And that, it turns out, is surprisingly rare.  Some police departments, such as Oakland, do not have neighborhood divisions.  They are, rather, organized on a city-wide basis.  As a result, the localized knowledge of hot spots for crime may be more difficult, but not impossible, to come by. 

The third factor that may help to explain why the three largest cities have experienced declining crime is demography. Each of those three cities is losing young people at a rapid rate.  All three, taken together, lost 200,000 15-24 year-olds between 2000 and 2005.  Males in roughly that age range account for between 40 and 50 percent of most crime in most societies.  Additionally, during the same period those three cities lost 10 percent of their African-American populations, while the percentage of black residents remained the same in other large cities.  Because that age group commits a disproportionate share of crime, having fewer of them means, all other things equal, less crime.  And because African-Americans constitute roughly half of all homicide victims and roughly half of all those who commit homicide, the loss of black population could cause, all other things equal, a reduction in crime, too. 

TSU