One of the consequences of the Great Recession we’ve been in since 2008 may be the long-overdue death of the “root causes” theory of crime and criminality. In A Crime Theory Demolished Heather McDonald points out that crime rates are dropping to 50-year lows even as unemployment hits a 70-year high.
While she is right to point out that this makes a joke of “root causes” theory, she doesn’t really propose an alternative to it, other than by waving a hand in the direction of rising incarceration rates. And I don’t believe that explanation; too many of our prisoners are nonviolent drug offenders.
In fact that category alone seems to account for most of the prison population growth over the last couple of decades, which suggests that increased incarceration is suppressing other sorts of crime very little or not at all. (See the update below.)
It’s also been noted recently that the trends in crime rates make nonsense of the notion that civilian firearms cause crime — even so reliable a bellwether of bland bien-pensant liberalism as the Christian Science Monitor has remarked upon this.
Mind you, the continuing fall in crime also falsifies some of the pet theories of social conservatives. They’re prone to chunter on about “defining deviance down” and the coarsening of popular culture as though sales of Grand Theft Auto actually had something to do with rates of grand theft auto. Given that there has been no sign of pop culture reverting to the 1950s (or whatever other era they imagine to have been ideal), this too seems an unsustainable explanation.
So what is actually going on here?
I think we get a clue from the fact, observable in emergencies such as floods and hurricanes, that most people do not instantly become criminals even when social order collapses around them. It takes continuing survival stress comparable to battlefield conditions to beat sociability and mutual trust out of most human beings.
However, I said “most”. Any criminologist will tell you that criminals as a group are also highly deviant in ways that are not criminal. They have very high rates of accidental injury, alcoholism, nicotine addiction, and involvement in automobile collisions. They have poor impulse control. They have high time preference (that is, they find it difficult to defer gratification or regulate their own behavior in light of distant future consequences). And they’re stupid, well below the whole-population average in IQ or whatever other measure of reasoning capacity you apply. I’m going to revive a term from early criminology and refer to these dysfunctional deviants as “jukes”.
One clue to the long-term fall in crime rates may be that most of the juke traits I’ve just described are heritable. Note that this is not exactly the same thing as genetically transmitted; children may to a significant extent acquire them from their families by imitation and learning.
The long-term fall in crime rates suggest that something may have been disrupting the generational transmission of traits associated with criminal deviance. Are there plausible candidates for that something? Are there selective pressures operating against jukeness that have become more pressing since the 1960s?
I think I can name three: ready availability of intoxicants, contraception, and automobiles.
Once I got this far in my thinking I realized that the authors of Freakonomics got there before me on one of these; they argued for a strong forward influence from availability of abortion to decreased crime rates two decades later. And yes, I know that a couple of conservative economists (Steve Sailer and John Lott) think they’ve found fatal flaws in the Levitt/Dubner argument; I’ve read the debate and I think Levitt/Dubner have done an effective job of defending their insight.
But I’m arguing a more general case that subsumes Levitt/Dubner. That is, that modern life makes juke traits more dangerous to reproductive success than they used to be. Automobiles are a good example. Before they became ubiquitous, most people didn’t own anything that they used every single day and that so often rewarded a moment’s inattention with injury or death.
Ready availability of cheap booze and powerful drugs means people with addictive personalities can kill themselves faster. Easy access to contraception and abortion means impulse fucks are less likely to actually produce offspring. More generally, as people gain more control over their lives and faster ways to screw up, the selective consequences from bad judgment and the selective premium on good judgment both increase.
Now that we’ve got this far, maybe incarcerating a lot of nonviolent druggies has actually been somewhat helpful, to the degree that group oversamples jukes and thus suppresses their reproduction. I’m still opposed to it for lots of pragmatic reasons that I won’t go into because they would distract from my main argument here.
This model inverts the traditional form of moralism according to which wealthy, libertine societies breed levels of vice, sloth, and degeneracy unknown to the struggling but virtuous poor. In fact, it suggests that jukes need the tight social controls of a conservative, tradition-bound society to minimize their disadvantages.
This model is testable. I used to live in West Philadelphia, and left for other reasons right around the onset time of the crack-cocaine epidemic there in the early 1980s. Longitudinal studies of crime, addiction, and accident rates there before and after might show that the juke population had seriously thinned itself out by 1990. That would be a worthy investigation for some young social scientist.
UPDATE: One of my regulars informs me that my source on incarceration figures was playing games with the statistics and shows evidence that most of the increase in incarcerations is not in drug offenses. This means that a study to test the adverse selection theory would need to be designed carefully to disentangle that effect from increased incarceration. I don’t know offhand, how to do that.