Balkinization  

Tuesday, August 16, 2011

Bill Bratton in London: Time for a Reality Check

Bernard E. Harcourt

Great news from the LAPD: Violent crime in Los Angeles fell almost 10% for the first six months of 2011, with an 8% drop in homicides. Crimes tied to gangs fell by an astounding 16%. Serious crimes—including robberies, rapes, burglaries, and thefts—were down by about 8%. LAPD’s Bill Bratton must be doing a fabulous job. Oops! Bill Bratton’s not chief anymore! Not since October 2009. Well then, how can we explain those sharp drops in crime? And who is taking credit?

As you've heard, there’s been a lot of talk about Bill Bratton in London these days. Based on several statistical studies, I’ve written somewhat critically in the past about Bratton taking credit for crime drops both in the Los Angeles Times and in the New York Times, and elsewhere. So naturally I’ve been curious about the recent developments abroad. I decided to take a quick look at Bratton’s most recent record to see if there is anything new.

On Bratton’s LAPD experience, there is no serious statistical study regarding his effect on crime—with the exception of Ian Ayres’ thorough study documenting racial profiling by the LAPD under Bratton’s tenure in 2003 and 2004. (Some people think racial profiling should be a crime, in which case the only reliable statistical study suggests the Bratton effect was, sadly, positive).

All that is left in LA, then, are bald assertions based on raw crime rates. The raw crime data from LA are themselves inconclusive—and, as I will get to in a minute, cannot be relied on anyway.

On the raw data. Under Bratton’s tenure, from 2003 to 2009, homicides fell 36.2% from 10.5 to 6.7 per 100,000, and all violent crimes fell 37.6% from 842.7 to 526.0 per 100,000. That’s impressive, but hold on. A decade earlier, when the LAPD was essentially defunct, during the same six-year period, from 1993 to 1999, homicides fell 56.9% from 21.1 to 9.1 per 100,000, and all violent crime fell 47.4% from 1,673.6 to 880.9 per 100,000. That’s a lot higher rate! In other words, when the LAPD was staggering without any leadership and headed toward federal court receivership—navigating between the Rodney King beating in 1991 and the Rampart scandal in the late 1990s—crime was dropping at a faster rate than when Bratton was in charge ten years later!

Not only that, but it was dropping even faster in LA in certain respects than in New York City, where Bratton was in charge (at least from 1993 to 1996). As I showed here, a straight comparison of homicide and robbery rates between 1991 and 1998 reveals that, although New York City was a strong performer, with declines in homicide and robbery rates of 70.6 percent and 60.1 percent respectively, Los Angeles experienced a greater decline in its robbery rate (60.9 percent); in addition, San Diego experienced larger declines in homicide and robbery rates (76.4 percent and 62.6 percent respectively), Boston experienced a comparable decline in its homicide rate (69.3 percent), and San Antonio experienced a comparable decline in its robbery rate (59.1 percent). Other major cities also experienced impressive declines in their homicide and robbery rates, including Houston (61.3 percent and 48.5 percent respectively) and Dallas (52.4 percent and 50.7 percent respectively).

What does all this tell us?

First, there are independent reasons why big cities—our former "homicide capitals"—saw some of the sharper drops in crime. Jens Ludwig and I explain it in our article in 2006. We call it Newton’s Law of Crime: What goes up must come down (and what goes up the most tends to come down the most). Statisticians call this “reversion to the mean”—a slightly more technical term. But essentially, when you take into account the sharp rise in crime in particular neighborhoods (primarily due to the crack trade homicides and violence of the mid-1980s to early 1990s), the correlations between NYPD policing under Bratton (or probably LAPD policing without Bratton) and the sharp crime drops in the 1990s vanish.

In other words, second, it makes no sense to use raw crime data and draw inferences. Comparing raw rates of crime is a shell game.

Third, it’s preposterous for police officials to take credit for these kinds of massive crime drops when there have been larger national forces radically changing the penal landscape of the USA over the past two plus decades. Policing strategies are going to have marginal effects compared to the national level forces such as mass incarceration, drug-use trends, police force numbers, demographics, and societal and ideational changes—many of these are discussed elegantly by my colleague Steve Levit in this paper.

In the end, I’ve never truly understood how we came to believe that the Bratton years were a period of order-maintenance in the Big Apple. To be sure, crime decreased sharply, as it would to the present day two decades after Bratton’s departure. But complaints of police misconduct also rose sharply, as I document here. From 1993 to 1996, under Chief Bratton, allegations of police misconduct rose steadily by 68% and the number of complaints of police misconduct rose by 55%. Court filings against the police in civil rights claims for abusive conduct rose 75% from 1994 to 1998. And those rising allegation and complaint rates stick even when we hold constant a sharply increasing police force and increasing total adult arrest rates. Similarly, racial profiling was going strong under Chief Bratton in LA. So were those periods of order or disorderliness? I suppose the answer depends on which end of the stick you found yourself.

Comments:

Bill Bratton began the myth of Bill Bratton here in the Boston area where he grew up. Egos in the Boston Police Department got in the way and he lost. But the myth persisted and it combined with the myth of Mayor Giuliani in New York City with the Broken Windows Theory; but this clash of myths sent Bratton on his way and the Giuliani myth won out. But the myth of Bratton continued on to LA as described in the post.

I recently saw on TV for the umpteenth time Robert Morse in "How to Succeed in Business Without Really Trying." I thought of Bratton and of so many other myths, including fictional, as I watched Morse perform - some of the music and dancing was entertaining. Bratton's myth was somewhat dormant until revived by the London riots. The myth of Giuliani is also dormant, although he may continue as a fringe GOP candidate to help keep his brand - myth - alive.

As to how myths begin, I come back to Boston and George Frazier's "Harvard Blues" recorded by Count Basie and his orchestra, with vocals by Jimmy Rushing. (A great recording, by the way.) Boston's own George Frazier (those interested should read his bio "Another Man's Poison") wrote the words, about a Harvard student who made a myth of himself, "Rinehart, Rinehart ...." George Avakian's album notes are a "must read."

So now can we expect the myth of Bill Bratton to resolve the long festering underlying issues involved with the London riots? It should be noted that England's (unwritten-in-one-place) constitution does not have a Second Amendment equivalent. For years we have heard that London Bobbies don't cary guns, which may now be a myth. But surely the situation in London may be a bit foggy for the Bratton myth. Perhaps a band of merry men would be preferable, like Monty Python's Flying Circus, that at least knows the territory and the language. But Cameron, with his now many problems politically, will probably pull a Maggie Thatcher by getting tough, as if that will address the underlying problems.
 

Today's NYTimes editorial "Wrong Answers in Britain" is very critical of PM Cameron as he picks on the poor, perhaps (my view) to shift attention from the richer hackers that Cameron hired to serve in his government. Of course, Cameron's austerity program hurts the poor in Britain more than the richer hacker society.

Segue to Paul Krugman's Blog at the NYTimes for how the Bank of England is looking at the austerity program of Cameron.

Supermyth Bratton to the rescue?
 

This post offers one view of the evidence, but there is another.

Regression to the mean is a well documented phenomenon in criminology, but the longer a crime decline continues, the less likely it can be regarded as simply a regression to the mean. Thus, the fact that Los Angeles's crime decline had begun more than five years before Bratton arrived sharply reduces the likelihood that its continuation was no more than regression to the mean.

Consider New York. In 1990, its homicide rate was at the average for large cities. During the following decade, its decline in each category of index crime approximately doubled the rest of the country, and it outperformed every one of the nation's fifteen largest cities. The decline is not explained by incarceration rates, abortion, the economy, or decline in demand for drugs. Declines were especially large for homicide -- where the statistics are widely considered reliable -- and were concentrated in crime in public location, where policing is most likely to be effective. Studies (including Levitt's work) found a relationship between the size of the police force and the crime decline. The decline was correlated as well with changes in New York's policing strategy. Even your co-authbor Jens Ludwig has published work establishing the efficacy of aggressive patrol of the type used in New York in reducing crime. Perhaps our most eminent criminologist, Franklin Zimring, in his book The Great American Crime Decline, offers a far different view of the New York data than is found in this post. As for my own view, it can be found here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1645436

Bratton surely does not deserve all the credit. Regression to the mean was part of the story, and changes in policing strategy began before Bratton's arrival and continued after his departure in both New York and Los Angeles. Still, the magnitude and duration of the crime decline (lasting through a much longer period than was studied in the Harcourt & Ludwig paper) strongly argues against regarding New York as reflecting no more a regression to the mean.

Larry Rosenthal
Chapman University School of Law
 

Post a Comment

Older Posts
Newer Posts
Home