Bernard Harcourt is one of those names that catches my attention whenever I see it — if he’s written something, it’s bound to be interesting and enlightening. Mr. Harcourt is Julius Kreeger Professor of Law and Criminology at the University of Chicago Law School — and he is also an inspired researcher and analyst.
This time I ran across him at the Volokh Conspiracy, one of my regular blog reads. Mr. Harcourt made a series of guest posts there, discussing his recently released study: From the Asylum to the Prison: Rethinking the Incarceration Revolution. The graph at right is taken from that study.
The thick black line at the top of the graph is the aggregate incarceration rate — the sum of prison incarceration and mental institution incarceration. Below that are two separate lines for each form of incarceration. Separately in the study (available at the link above) is a graph showing homicide rates: it is a nearly perfect inverse of the aggregate incarceration rate. In other words, the more people who are incarcerated (no matter where they are incarcerated), the lower the homicide rate. This comes as no surprise to me — if you’re willing to believe that usually the criminal justice system does its job correctly, then of course the more bad guys who are locked up, the less bad things they will do. Doh! But some liberals will be quite puzzled — the usual response to this sort of data will be to wonder why so many people are locked up if the crime rate is low. The problem is that these pesky facts don’t jibe very well with the liberal narrative, (sort of like the relationship between gun ownership and crime rates), so they are, naturally, puzzled.
But the graph at right was a real surprise to me. The biggest surprise was just how many people used to be incarcerated for mental illness — I had no idea that the rate was so high! The highest rate of mental institution occurred in 1948, when one of every 159 adults in the U.S. was in a mental institution. Today that rate is about one in 7,000 adults. By contrast, today’s prison incarceration rate is about one adult in 167 — a very high rate, and in fact the highest in the world — but not as high as our one-time mental institution incarceration rate.
If you stare at that graph for a moment (click on it to get a larger version), you’ll be tempted to think that when we dramatically changed the way in which we incarcerated mentally ill people (a reaction to revelations of horrible abuses in the '60s), we dumped tens of thousands of people onto the streets who later became criminals. You’re tempted because you see the precipitous decline in mental institution incarceration in the late 1960s and early 1970s followed a steady rise in the criminal incarceration rate starting in the mid 1970s. But Mr. Harcourt, seeing the same thing, warns us that this simple explanation cannot be right, because the populations are significantly different. If you read the study, you can get all the details, but basically mental institution incarcerees were considerably more likely to be white, female, and older than the prison incarcerees. The same population didn’t just move from mental institutions to prisons. Mr. Harcourt is as puzzled by this as we are.
But the inverse correlation between incarceration rates and homicide rates is clear and strong. These data strongly suggest that mental institutions and prisons, in their respective times, were both places where we safely “stored” people who were likely to commit homicide — and that both were effective protections against homicide for the rest of society. Of course there are more considerations than simply the homicide rate — I certainly would not advocate going back to the way we used to incarcerate people in mental institutions! Nevertheless, the lesson here is instructive: incarceration works. To many, like myself, this seems obvious — but to many liberals, it is an article of their political faith that incarceration (i.e., punishment and accountability) is an ineffective tool of social policy…
