Urban Myth: Claim that Federal Housing Programs Cause Crime Spikes is a Crock
Does public housing, including Section 8 vouchers that help poor people pay rent, and other "dispersal" programs, cause crime? A reader commenting on an item posted here at the end of last week cited an article published in The Atlantic magazine in 2008, called "American Murder Mystery," saying pretty much that it does.
In that piece writer Hanna Rosin cited what she said were statistical clusters in Memphis in the first decade of this century to argue that crime was spiking wherever recipients of public housing subsidies settled after being pushed out of the old public housing projects. I remember reading it when it was new and finding it pretty disturbing.
But you know 2008 was a long time ago. If you're going to reach that far back for an argument to use today, you need to keep Googling. If an argument is that old, it's important to see what has happened to it since it was born.
In fact, what you find -- easily available all over the net today -- is solid evidence from serious scholars to show that Rosin's article was bullshit magazine journalism.
First she finds a number of programs designed to reduce or eliminate the old post-World War II projects. She wrongly characterizes those programs as liberal attempts to socially engineer. Most of the big tear-downs of the 1970s and '80s -- Pruitt Igoe in St. Louis, West Dallas in Dallas -- were either initiated under Nixon to save money or ordered by the courts to fight deliberate segregation.
Is putting an end to deliberate segregation the same thing as dispersal? I don't know. Let's leave that one to the philosophers of language.
The big point made by scholarly critics of the Rosin piece is that she committed a typical journalistic mistake in her use of statistics by assuming that "correlation is causation."
Yes, the not-white poor generally dispersed themselves out of traditional ghettos in the first decade of this century. Yes, that included recipients of federal housing aid. Yes, crime increased in some of the areas to which they dispersed.
But Rosin failed to churn up any significant evidence that public housing dispersal had a thing to do with the crime spike. Critics of her piece seemed to be able to find persuasive evidence without too much effort that public housing recipients were a relatively minor slice of the much larger outward migration of the poor from the old ghettos.
Beginning in the 1980s, poverty in this country bounced down and then back up following a pronounced cycle of boom and then bust. Poor people moved around a lot.
We need only look to North Dallas and the inner rim of our own suburbs. Areas that were almost entirely white middle-class a decade ago are now a diverse amalgam of immigrant and American-born black and Hispanic populations with whites, presumably quite mixed in terms of class, from chronically unemployed to fully employed working class and middle class.
That kind of demographic shift is going to bring concomitant changes in social patterns, including crime. But before you can single out the smallest portion of the shift, Section 8 voucher recipients, and blame them for the most severe of social problems, crime, you have to do more work to connect the dots.
Critics have pointed out that Rosin failed to dig her way through court records to find proof that crimes committed in areas with spikes in crime rates were being committed by people receiving housing assistance. Given the relatively small proportion of Section 8 recipients in the outward migration, those recipients would have needed to commit a dramatically outsized portion of the crime to be responsible for the overall upticks cited by Rosin.
Would it be hard to find proof or disproof in criminal justice records? Oh, yeah. You bet. It would take a lot of work. But remember two things: You can get paid a few grand to write a piece like that for a national magazine. So you can afford to break a sweat.
And more important: If you're going to argue from the smallest evidence to make your biggest point, you need to connect a solid line of dots first just so you yourself can be sure you know what you're talking about and you're not just dishing up a provocative piece that will get buzz.
Anyway, the larger point here is about public housing and crime. The Atlantic piece published four years ago was no smoking gun. If anything, it was a shotgun wedding of statistics and shibboleths -- a marriage that didn't last.
Her piece was well written, and she apparently said later she was "just trying to start a conversation." I remember saying that once back during my single years. As I recall, I got slapped anyway.
Get the This Week's Top Stories Newsletter
Every week we collect the latest news, music and arts stories — along with film and food reviews and the best things to do this week — so that you’ll never miss Observer's biggest stories.