About . . . . . . Classes . . . . . . Books . . . . . . Vita . . . . . . . Links. . . . . . Blog

by Peter Moskos

December 26, 2015

"The enduring commitment of antipolice progressives to the 'root causes' theory of crime"

This op-ed by Heather Mac Donald is the one I wanted to write. But I didn't. And she did.

The point, one could say rather simply, is that police matter as a force for crime prevention. That simple concept is why I decided to study policing and then became a cop.

In the mid 1990s I got into this gig because an entire academic field said that the crime drop couldn't happen. Crime wouldn't go down until we improved "root causes" and fixed a racist society. By the time I entered graduate school in 1995, it was clear that crime was going down. Something was up. And it wasn't employment and equal opportunity.

This link to Mac Donald's op-ed is behind the Wall Street Journal paywall. To read it all, try googling the headline "Trying to Hide the Rise of Violent Crime" and click through. Excerpts:
An 11% one-year increase in any crime category is massive; an equivalent decrease in homicides would be greeted with high-fives by politicians and police chiefs. Yet the media have tried to repackage that 11% homicide increase as trivial.

Several strategies are employed to play down the jump in homicides. The simplest is to hide the actual figure. An Atlantic magazine article in November, “Debunking the Ferguson Effect,” reports: “Based on their data, the Brennan Center projects that homicides will rise slightly overall from 2014 to 2015.”
A second strategy for brushing off the homicide surge is to contextualize it over a long period. Because homicides haven’t returned to their appalling early 1990s or early 2000s levels, the current crime increase is insignificant.
The most desperate tactic for discounting the homicide increase is to disaggregate the average. ... The “numbers make clear that violent crime is up in some major U.S. cities and down in others.”

If there weren’t variation across the members of a set, no average would be needed. [Zing! Nice one. I always appreciate a snarky line about stats.]
To the Brennan Center and its cheerleaders, the nation’s law-enforcement officials are in the grip of a delusion that prevents them from seeing the halcyon crime picture before their eyes.
FBI Director James Comey noted “a chill wind blowing through American law enforcement over the last year,” and called it “deeply disturbing.”
Obama ... accused Mr. Comey of “cherry-picking data” and ignoring “the facts” on crime in pursuit of a “political agenda.”
Critics of the Ferguson-effect analysis ignore or deny the animosity that the police now face in urban areas.
The St. Louis area includes Ferguson.... The Justice Department later determined that the officer’s use of force was justified, but the damage to the social fabric had already been done.... The media and many politicians decry as racist law-enforcement tools like pedestrian stops and broken-windows policing—the proven method of stopping major crimes by going after minor ones.
Consider that background. Here's the point I've been trying to make:
The puzzle is why these progressives are so intent on denying that such depolicing is occurring and that it is affecting public safety.

The answer lies in the enduring commitment of antipolice progressives to the “root causes” theory of crime. The Brennan Center study closes by hypothesizing that lower incomes, higher poverty rates, falling populations and high unemployment are driving the rising murder rates.... But those aspects of urban life haven’t dramatically worsened over the past year and a half.
To acknowledge the Ferguson effect would be tantamount to acknowledging that police matter, especially when the family and other informal social controls break down.
Many of those who are driving the "there is no Ferguson effect" bandwagon still believe that police are largely irrelevant to crime prevention and, rather than having anything to do with crime prevention, serve primarily as agents of racial oppression. That sentiment lies just under the surface of anti-police protests.

It's not just about "Justice For [fill in the blank of latest person shot by cops]." It's about an ideology that still won't accept that aggressive order-maintenance policing did any good. The "root cause" brigade never accepted that crime could decrease independent of structural changes. That's what I mean when I talk about an ideological opposition to Broken Windows.

So the next time you hear somebody say "crime isn't up" or "there is no Ferguson effect" or "Michael Brown had his hands up" consider that they're not just mistaken about one detail, however important. Instead, consider that they have a fundamentally different ideological view of who police are and what they can do.


bacchys said...

I think there's a Ferguson Effect. It's not because people are somehow doing something wrong. It's because corruption in law enforcement institutions has become public knowledge.

Don't talk about the "Ferguson Effect" without noting what the Ferguson PD was like. They earned this shit.


Matt Ashby said...

There is plenty of good-quality evidence that police activity at crime hotspots can work, but that doesn't mean it's always necessarily a good thing. Once we know it works, the next questions become 'in what circumstances does it work, and is it worth the various types of cost?' I don't know the answer to that because it depends on how you measure the cost. As far as I know we don't have a clear understanding of how different dosages of order-maintenance policing influence community perceptions of police, and how that in turn influences crime and willingness to assist the police. Hopefully everything that's happened over the past year will give researchers the opportunity to better understand these issues.

One thing I disagree with in Mac Donald's piece is her dismissal of the importance of variation around the national homicide rate. Yes, variation is inherent but that doesn't mean it isn't important. In a way she's got it backwards: she suggests that you should ignore the variation because it's inevitable, but actually if anything you should consider ignoring an average if it isn't a good summary of the national picture. It makes no sense to be a slave to a summary statistic when you know that there's no guarantee that an average (of any type) is a good summary. This is particularly true when looking at change over time because there can be so many different trajectories.

Dismissing the variation about the average is particularly worrying because there are plenty of reasons to think that the variation is meaningful. We know that both local communities and policing tactics vary in different cities so why ignore all that information by looking only at the average? It is likely to be much more useful to consider the variation and try to explain it. That might help to explain exactly how police activity influences crime, what exactly police should and shouldn't be doing in particular circumstances.

Jason Taverner said...

Remember the huge crime wave that was predicted to occur at the start of The Great Recession? Another example of the failure of the musings of the "root causes" crowd to conform with reality.

That Fuzzy Bastard said...

There may be a Ferguson effect, but not the one police claim there is. As it becomes increasingly obvious that calling the cops is a good way to get minor troublemakers, or even the people making the call, shot and killed, people are more reluctant to call the cops. There is an increase in crime not because of poverty (though it would be interesting to know why the drug trade has become deadlier), but because the people who are supposed to be controlling crime are bad at their jobs.

Bill Harshaw said...

I'm just confused. The NYTimes reports that arrests are way down, which sounds like "depolicing" but crime is still down. The Post reports that DC police are retiring faster than they can be replaced, partly reflecting the surge in hiring back in the early 90's, the days of crack and crime. The radio reports VA is seeing big jumps in deaths from heroin and opiates. Thinking back, that suggests maybe some of the violence is, as Lanier suggests, drug dealers fighting over changing markets for different drugs (and pot is sort of legal in DC). Baltimore and DC homicides are way up.

Bottom line: I can't buy any national narrative from any side.

Peter Moskos said...

Good point, Matt.

Jason, do you have any cite in mind? Did somebody actually write a piece about that? I'm sure people said it (to me).

Bill, I don't think there's any single narrative. I think both "sides" are guilty of trying to say there's one national trend. To paraphrase Tip O'Neill, all policing is local.

Concerned citizen said...

TFB wrote, " the people who are supposed to be controlling crime are bad at their jobs."

Homicide death rates for blacks were 50% higher in 1950 than they are now.

This reduction in homicide since 1950 occurred despite a subsequent massive proliferation of guns, gangs, drug sales, etc.

It seems reasonable to me that at least some measure of credit should go to law enforcement.

Source: Homicide death rates are from CDC Health U.S. 2013, Table 34.

direwolfc said...

Peter - I agree with you about the progressive obsession with 'root causes' of crime. The tendency to blame police for increases in crime but not credit them for decreases in crime is a convenient heads-i-win-tails-you-lose practice in epistemic closure.

That being said, I just can't get myself to read anything written by Mac Donald, given things she's said in the past. Randomly off the top of my head, in an interview with Mike Pesca in Slate, she gave a totally misrepresentative account of an arrest by the Baltimore police of an individual with a handgun during the week of the riots in late April. Her use of statistics is regularly terrible and self-serving. She reminds me of those FOP presidents that say the most wildly inflammatory things in their enthusiastic defense of the police.

Peter Moskos said...

I find Mac Donald frustrating, because because of the way she frames the debate and is predictable in her conclusion. She's smart. I'm sure she does think there have been some cases in US history of policing done wrong. But I don't know if she's ever expressed what cops have to do to cross the line. It is FOP like.

Yet in general, I agree with about 80 percent of what she says. And still I find even that part I agree with hard to take. After reading her, I want to disagree with pro-cop positions I already believe, simply because of how she frames the debate, is ideologically dismissive of opposing views, and can ignore legitimate criticism.

I wrote this a good while back: http://www.copinthehood.com/2010/06/heather-macdonald-on-stop-and-frisks.html?showComment=1277654901842

All that said, this current piece by her articulates some good points I've been trying to get my head around. (I haven't listened to the Slate interview.)

Jay Livingston said...

Can someone direct me to a definitive article (meta-analysis, review article) on the assertion that "pedestrian stops and broken-windows policing [are]the proven method of stopping major crimes." Or that "good policing over the past two decades produced an extraordinary 50% drop in crime." Yes crime dropped by 50%, but us MacDonald correct in implying that all of that caused by changes in policing?

Also I do not understand her claim that "disaggregating the average" is a "desperate tactic." If crime is up in one city and down in another, why is it desperate to look for other differences between those cities that might account for the crime difference? I would think that such research would be just good social science. In fact, MacDonald herself does precisely that when she says that cities "with significant black populations and where antipolice agitation has been most strident" have had the biggest increases in murder. I assume that her data is more systematic than just "homicides" in two cities, "violent crime" in one city, "shooting victims" in another city, and "shooting incidents" in yet another. (Far be it from me to suggest that MacDonald's selection of different outcome variables for each city is cherry picking.)

Peter Moskos said...

Braga, Welsh, & Schnell (2015) "Can Policing Disorder Reduce Crime? A Systematic Review and Meta-analysis." Journal or Research in Crime and Delinquency. This is probably the best there is. But it won't satisfy critics. Nothing will.

Sousa in the City Journal provides a good passionate (but accurate, thinks me) summary of the opposition. http://www.city-journal.org/2015/eon0825ws.html

Best big picture summary of Broken Windows and Order Maintenance comes from Thacher (2014) in the Oxford Handbook of Police and Policing, "Order Maintenance Policing."

Here's the problem. As you know, there will *never* be a definitive summary proving or disproving such a complicated real-life issue. But you were here. I was in New Jersey. We were both watching. Certainly we agree that police played some role, right? Really, what else major was going on? (getting lead out, maybe?)

Also everywhere Bratton went, Boston, NYC, LA, homicides went down. A lot. And yet some will never give him credit because he gave it to policing. But if Bratton doesn't deserve credit, he's the luckiest SOB on the face of this earth!

I also think it's absurd to argue that police had nothing to do with the crime drop in NYC. At some point, I'm willing to say the burden of proof is on opponents. Prove it wrong.

I think it really does come down to to the issue of whether police matter at all. I think yes. Maybe not as much as Mac Donald says. But whether police are 30 or 80 percent of the crime drop doesn't really concern me as much as what they did that brought about that reduction.

Peter Moskos said...

As to "disaggregating the average," far be it for me to put words in her mouth. And her words may exceed her quantitative reasoning, but I think what she's trying to say is crime -- homicide -- *is* up in a lot of places and overall. So why the hell do so many people put their head in the sand and say it doesn't matter as long as crime (homicide) is also down in some places?

As for me and you, we'd love to look for differences in those cities that might account for the crime difference. I think Mac Donald (cherry picking and all) is saying great, let's do that. But first we have to accept that homicides really are up a lot in some places. But if you (or a headline editor) say crime is up, the response is, "crime isn't spiking!" Or, "historically, it's low." So effing what?! The answer from too many is some variation of "don't create a moral panic" or "No it's not! Look at the Brennan Center, "which seems to spend a lot of effort explaining why we shouldn't be concerned about more murder. Why was Comey pillared for trying to address the issue in a thoughtful way? What is the real issue here? Why the semi-organized denial?

How can we get to the why when can't even agree on the what?

Homicides are up 15 percent this year. While I wouldn't say a "surge" or a "wave," 15 percent is a hell of a spike!

(I'm more sympathetic to those, myself included, who say crime isn't going up in NYC because NYC is a special political case and people predicting imminent hell in a handcart.)

Peter Moskos said...

The Brennan Center says "The 2015 murder rate is projected to be 14.6 percent higher than last year" but there is no need to worry.

I almost hate to bring this up, but I just glanced at the homicide rates since 1925. You know the last time there was a 15 percent annual increase in the homicide rate?


1945 saw a 14 percent increase. 1946 another 12 percent. Those are both to be expected.

1966-1968 saw 3 consecutive double digit percent increases.

Frankly, shouldn't we be a little worried?




Jay Livingston said...

Thanks for the references. I’ll try to find them and take a look. This is one of the research areas where it probably helps to disaggregate even further. That is, I would ask “Which police tactics?” and “Which crimes?” For example, I can see why homicides might be affected by stop-and-frisk because maybe it makes people less willing to walk around carrying a gun. But I don’t see how homicide rates will be lowered by cops arresting fare-beats. Or squeegee men. Or guys selling loosies. Even with a single crime like homicide, I’d want to know which kinds of homicides are affected – domestic, gang, acquaintance arguments, drug disputes, etc.

In the same way, if you really wanted to test the Ferguson-effect idea, you’d have to measure the cause variable (reductions in aggressive policing) in a way that allowed you to differentiate among different cities and then see where crime increased, and which crimes.

As for the politics of crime surges, I was around in the 60s when my liberal friends were saying that the reported increase in crime was merely an increase in reporting and recording and that crime stats were so bad you couldn’t believe them at all. They were wrong of course. The Brennans may turn out to be wrong too, but for the moment, we don’t know whether the increase in some crimes isa blip or a trend, so cautioning against panic isn’t such a bad idea, especially considering what the Muslim-terrorist panic has brought us in the way of policy proposals and political candidates.

Also, the liberals’ criticism of the UCR was accurate, at least in part, and it did lead to the NCVS and other improvements in crime statistics. Ferguson has already led to more information about what the police there do – not so much in the one incident (killing Michael Brown) but in the everyday performance of their jobs. Maybe more and better information, maybe even in some systematic national database, will be an outcome of the current disagreement.

Adrian said...

"Maybe more and better information, maybe even in some systematic national database, will be an outcome of the current disagreement."

My biggest frustration with the #BLM movement is that one of their demands is not DATA. Instead, they assume that everything they feel about policing is correct, thus there is no need to gather data. They've achieved some impressive things in terms of organizing a huge group of people and getting political pressure on Democrats. If they made one of their demands better data collection/reporting on crime and policing, there's a good chance that we might actually understand more about what is going on. But #BLM already knows exactly what's going on! So no need for data...

Ryanaldo said...

some have said legalizing abortion reduced crime, because it allowed people who aren't prepared to be parents to not produce more children.

Peter Moskos said...

Though it makes intuitive sense, that's been debunked for a lot of reasons I don't want to get into.