Economists Gone Wild!

In Slate, libertarian economist Steven E. Landsburg trumpets a popular new study called "Pornography, Rape, and the Internet" by Clemson economist Todd Kendall:

How the Web Prevents Rape
All that Internet porn reduces sex crimes. Really.
By Steven E. Landsburg

Does pornography breed rape? Do violent movies breed violent crime? Quite the opposite, it seems.

First, porn. What happens when more people view more of it? The rise of the Internet offers a gigantic natural experiment. Better yet, because Internet usage caught on at different times in different states, it offers 50 natural experiments.

The bottom line on these experiments is, "More Net access, less rape." A 10 percent increase in Net access yields about a 7.3 percent decrease in reported rapes. States that adopted the Internet quickly saw the biggest declines. And, according to Clemson professor Todd Kendall, the effects remain even after you control for all of the obvious confounding variables, such as alcohol consumption, police presence, poverty and unemployment rates, population density, and so forth.

OK, so we can at least tentatively conclude that Net access reduces rape. But that's a far cry from proving that porn access reduces rape. Maybe rape is down because the rapists are all indoors reading Slate or vandalizing Wikipedia. But professor Kendall points out that there is no similar effect of Internet access on homicide. It's hard to see how Wikipedia can deter rape without deterring other violent crimes at the same time. On the other hand, it's easy to imagine how porn might serve as a substitute for rape.

If not Wikipedia, then what? Maybe rape is down because former rapists have found their true loves on Match.com. But professor Kendall points out that the effects are strongest among 15-year-old to 19-year-old perpetrators—the group least likely to use such dating services.

Moreover, professor Kendall argues that those teenagers are precisely the group that (presumably) relies most heavily on the Internet for access to porn. When you're living with your parents, it's a lot easier to close your browser in a hurry than to hide a stash of magazines. So, the auxiliary evidence is all consistent with the hypothesis that Net access reduces rape because Net access makes it easy to find porn.

I strongly suspect that teenage rapists mostly come from the sizable segment of the population that still didn't have Internet access in 2003.

Yet, there is a much larger issue here.

You'll recognize this theory as Son-of-Abortion-Cuts-Crime: take an old theory that a lot of people, for personal reasons, wish is true, mine through a ton of state-level data, and voila!

What bothers me is that economists don't seem to remember anything when it's time to do simple reality checks on their prize theory.. When Steven D. Levitt and John J. Donohue came up with their famous abortion-crime theory by comparing crime rates in 1985 and 1997, they totally forgot the crack wars that happened in between. When David Card showed that the Mariel Boatlift of Cuban refugees to Miami in 1980 did not cause Miami to have lower wages over the next half decade relative to four other cities, he totally forgot that the Miami economy during his study (1980-1985) was notoriously being chemically stimulated by billions in cocaine smuggling profits.

Worse, all the other economists who read and reviewed these famous studies never noticed what happened during their own lifetimes. I can't find any evidence on Google that anybody mentioned the impact of cocaine in Miami on Card's 1989 study until my VDARE article last summer.

On the other hand, maybe nobody remembers anything and I'm just picking on economists.

Kendall doesn't seem to recall that we ran a massive "natural experiment" in the increased availability of pornography from roughly 1965 to 1975, with the opposite results: the rape rate shot upwards. I was only a kid, but I remember the debate: A study from Sweden or Denmark or somewhere "proved" that legalizing pornography reduced rape. Unfortunately, when this was tried in America (by about 1973, every commercial street corner in suburban Los Angles featured a dozen newsracks for pornographic newspapers), once again Americans didn't behave like Scandinavians.

America's greatest social observer, Tom Wolfe, pointed out in an important mid-1970s article "The Boiler Room and the Computer" (collected in his amazing anthology Mauve Gloves & Madmen) that this more-pornography-equals-less-rape theory is based on a dubious 19th Century Freudian analogy between the libido and that symbol of the age, a steam engine. Without frequent release, pressure builds up and "Look out! She's gonna blow!"

A little introspection among male readers about their own teen years should induce skepticism. Only on the shortest time frame is the boiler room analogy true. On a day-to-day basis, it's more like a feedback loop.

Similarly, why are the sports pages filled with police blotter items about professional athletes, who are presumably not lacking in outlets, being arrested on sexual assault charges?

This doesn't mean that pornography causes rape either. I'm just saying that a better memory would induce more humility and skepticism among freakonomists.

My published articles are archived at iSteve.com -- Steve Sailer