The Village Voice looks at underage prostitution statistics

By Felix Salmon
July 2, 2011

I respect, in theory, the way that Village Voice Media has decided to try and bring some quantifiable common sense to the question of how many underage prostitutes there are in the US. The main thesis of the 4,400-word article is compelling: the universally slung-around statistic that the number is between 100,000 and 300,000 is downright false.

VVM is open about the fact that it has a dog in this fight: a significant portion of its revenues come from adult classified ads, both in print and online. But its detailed reporting shows three different ways how far the crusaders against underage prostitution have veered into scaremongering.

First is that 100,000 to 300,000 figure, repeated ad nauseam in respectable outlets and largely uncontested. The original source of the number turns out to be a non-peer-reviewed paper from 2001 which attempts to add up all the children “‘at risk’ of commercial sexual exploitation” — a number which includes enormous numbers of children who will never become prostitutes at all:

These estimates reported in Exhibit ES.2a reflect what we believe to the number of children in the United States “at risk” of commercial sexual exploitation, i.e., children who because of their unique circum- stances as runaways, thrownaways, victims of physical or sexual abuse, users of psychotropic drugs, members of sexual minority groups, illegally trafficked children, children who cross international borders in search of cheap drugs and sex, and other illicit fare, are at special risk of sexual exploitation. The numbers presented in these exhibits do not, therefore, reflect the actual number of cases of the CSEC in the United States but, rather, what we estimate to be the number of children “at risk” of commercial sexual exploitation.

The choice of scare quotes here is a bit weird, until you read the VVM story, which talks to David Finkelhor, professor of sociology at the University of New Hampshire and director of Crimes Against Children Research Center. Finkelhor read an early draft of the paper and encouraged the authors to add in the “at risk” language: originally, it wasn’t there at all. Finkelhor also says that the paper “has no scientific credibility”.

But what’s undeniable is that the numbers in the paper include every child who ever runs away from home, even for a day or two; it’s silly to use them as a guide to the actual number of child prostitutes.

I do think that the VVM story should have linked to the paper — and, too, should have mentioned that the 100,000 to 300,000 range doesn’t actually appear anywhere in it. They do keep a running tally of the number of children at risk, and then multiply it by 88% to get a “medium scenario”, and by 75% to get a “low scenario”. The “medium scenario” ends up at 286,506 kids, while the “low scenario” is 244,181.

Importantly, the multipliers don’t seek to correct for the possibility that the estimates might be too high; they only seek to correct for double-counting, where the same kid appears in more than one of the 14 different at-risk categories. If you wanted to come up with a lower bound, one easy way to do that would be to simply add up categories 1, 3, and 4: runaway youth from home, runaway youth from juvenile and other institutions, and homeless children not elsewhere counted. Those three categories don’t overlap, and if you add them up you get to 156,676.

If you used this paper to generate a range of numbers, then, the obvious range would be 150,000 to 300,000 — something like that. It’s not at all clear where the 100,000 number comes from, although it might conceivably be something to do with this:

At the time of completing work on this report, a new study of the incidence of runaway and throw- naway children in the United States (NISMART 2) was nearing completion (Hanson, 2000). Inasmuch as 60% of all the children we estimate to be at risk of commercial sexual exploitation fall within the “run- away” and “thrownaway” categories (Rows 1, 2 and 3 of Exhibit ES.2a), the findings from this updated national incidence study of runaway and thrownaway children—but not directly of children involved in commercial sexual exploitation— is expected to have a significant impact on our estimates of the number of children at risk of commercial sexual exploitation. Preliminary discussions with investigators associ- ated with NISMART-2 suggest that the number of runaway and thrownaway children may have declined by as much as 30%-40% between 1988 and 2000.

In any case, the paper nowhere says or implies that it is generating an estimate for the total number of underage prostitutes. That reading got added on later, probably by someone reading a second- or third-hand report of what the paper says.

VVM could have stopped there, but they didn’t. Rather than simply debunk the number they found, they ran their own numbers, by looking at how many children were arrested for prostitution in the nation’s 37 largest cities over the past ten years. (The 2001 paper took a similar geographical approach: it was based on numbers from just 17 US cities.) It turns out that over the course of a decade, there were just 8,263 arrests — 826 per year. No matter how low the proportion of child prostitutes who end up getting arrested any given year, there’s no way that a universe of more than 100,000 prostitutes could generate a mere 826 arrests.

And then, most tellingly, VVM looked at the size of the child-prostitution industry — not the illegal part, but the legal bit.

In 2005 and 2006, the federal government spent $50 million primarily to fund law enforcement task forces involving U.S. Attorneys, local police, FBI and Department of Homeland Security agents, and various nonprofits. The task forces were created to put an end to sex and labor trafficking in America. Today, there are more than 40 such task forces, from Boston to Anchorage, each typically funded with $450,000 for three-year terms…

The Internet Crimes Against Children (ICAC) task forces, also composed of local and federal law enforcement agencies, have investigated child pornography and prostitution cases since 1998. Generally, the units receive tens of millions of dollars annually. As part of the government stimulus package, Uncle Sam handed out $75 million to ICAC groups in 2009.

In the past eight years, Congress has spent $200 million on child pornography in America and another $180 million on all domestic trafficking involving sex or labor.

The wording here (“Congress has spent $200 million on child pornography”) might be a bit unfortunate, but the point is a good one: all of this money has managed to find somewhere between 100 and 250 underage prostitutes per year. And when they’re found, they’re often arrested, rather than being given access to the help they need:

Seattle is one of the few places in the nation with a shelter devoted to underage prostitutes. Despite the obvious need, the city manages the program without federal funding…

Although Congress has spent hundreds of millions in tax-generated money to fight human trafficking, it has yet to spend a penny to shelter and counsel those boys and girls in America who are, in fact, underage prostitutes.

The VVM story, then, finds and demolishes the stated number, gets an empirical basis for the actual number, and makes a powerful point about how current initiatives are a way of spending too much money on exactly the wrong thing. If there were hundreds of thousands of underage prostitutes in the US, then the Congressional appropriations would make sense. But given the actual numbers, that money would be much better spent on shelters.

There are, however, big weaknesses with the piece. For one, it gratuitously attacks Ashton Kutcher, a smart person who’s making the world a better place, in an unpleasantly ad hominem manner. Kutcher is not the problem here. And it needs a lot more serious discussion of VVM’s own ethics with regard to running adult classifieds, including classifieds which turn out to be advertising underage prostitutes. You can argue about the efficacy of Kutcher’s campaign, but he’s not making the problem worse. VVM, meanwhile, is a non-negligible part of the problem, and needs be a lot more honest about its own place in the child-prostitution ecosystem.

The result of all this has been a destructive Twitter war with Kutcher, which has already resulted, among other things, in American Airlines pulling ads from VVM websites. VVM, in other words, could hardly have engineered a higher heat-to-light ratio if they’d tried. All of which makes the article look less a serious investigation, and more a noxious publicity stunt. If VVM is willing to examine its own behavior with regard to child prostitution in detail, then this road might have been well worth traveling. But if they just want to take potshots at Ashton Kutcher, I do wonder whether they will ultimately achieve anything at all, beyond a general notoriety.

14 comments

Comments are closed.