Zombie Statistics on Sex Work
As a former escort, I can say with confidence that nearly everything you know, or think you know, about the money involved in sex work is wrong. And as a former statistician, I know the numbers definitely are.
Last month the UK’s Office for National Statistics (ONS) released their plans to add calculations for the contributions of sex work and drug sales to the GDP, as is already done in several other countries, and will soon be required throughout the EU. The total they suggested? An impressive £10 billion per year. This immediately got the attention of the media, with some outlets suggesting that drugs and prostitution are worth more to the UK economy than agriculture is.
Having had a close look at the methods employed to come up with their number, though, I know this total is way too high. In fact it seems possible the total may be as much as an order of magnitude too high. Certainly the model the ONS produced for the sex work part of the calculation is problematic in several ways.
Estimating the number of sex workers in a given country is an art, not a science. It is a Fermi problem with a history of being misused or misunderstood. The term “sex worker” is frequently conflated with street-based prostitution, or workers who access outreach services. As a result, such numbers have been used to justify funding agendas in nongovernmental organizations like charities that directly benefit from provisions to rehabilitate sex workers.
The most frequently cited statistic cited in the UK is an estimate from 1999 for the European Network for HIV/STD Prevention in Prostitution (EUROPAP) project, derived by Hilary Kinnell. This put the total of UK sex workers at around 80,000. However, Kinnell’s methodology relied on a fair amount of guesswork; while prostitution is legal in Britain, it is also highly stigmatized, meaning the only sex workers who come into contact with outreach services and clinics tend to be on the more vulnerable end. Escorts and call girls largely pass unnoticed, and their numbers and working habits, without their direct input into research figures, can only be guessed at.
Kinnell herself has commented on how surprised she was for the number to have such a long afterlife, given her own caveats about the number.
Perhaps government bodies seized on this number because it is higher than most—such as a 2001 estimate that put the total at a more modest 53,000. Of course, whichever is closer to the truth, they are both over a decade out of date. Given the fluctuations in supply that follow periods of sustained recession as we had in the UK, coupled with the introduction of university fees for British students, it is reasonable to suggest whatever the picture was fifteen years ago is very different now.
Reports suggest that more students have been entering sex work in the past decade in order to offset the cost of their studies, and that the money changing hands for services has gone down during that same time period. While opponents of sex work often suggest “ending demand” as an approach to what is frequently considered a social ill, few seem to have considered the impact of increased supply.
But Kinnell’s oft-cited, outdated statistic wasn’t actually what the ONS used in its latest evaluation—it’s even worse than that. It turns out the ONS got their estimate of the current number of sex workers from Eaves for Women, a charity that works only with women, only in London, and that has come under fire for its slipshod statistical methods in the past. And what about men who pay for sex with other men, which accounts for the majority of male sex workers? As is usual when prostitution is discussed, gay sex workers are being erased from the story.
Nor did the ONS contact any number of prominent sex work researchers. I recently contacted a number of them myself—such as Belinda Brooks-Gordon, a frequent commentator on sex work and public policy, and Teela Sanders, whose work on strippers and other workers in sex-related industries is widely covered by UK newspapers—and all confirmed they had not been approached by the ONS.
The ONS is wrong about sex-work wages, too. They seem to have guessed at the average earnings of a sex worker by reading PunterNet, a website for clients of prostitution. It is also a popular site for female escorts to advertise on. While it might give a rough estimate of some people’s working life, it is subject to inflation, and only represents a select group of workers—many of whom are at the higher end of income. Providing a more accurate estimate of income is difficult, but allowing for the many varieties of sex work, it is safe to assume the average for all sex workers is rather less than what indoor, independent workers make.
Then, for their calculation of how many clients sex workers see in a week, the ONS relied on research conducted in the Netherlands. It doesn’t take a genius to spot that a comparison between sex work in a country where it is legal and heavily regulated, and one where it is legal but still routinely criminalized, is simply not going to be equivalent or useful.
The ONS is also wrong on its estimates of how many clients a sex worker in the UK meets in a week. It “calculated” an eye-watering twenty-five clients per sex worker per week, on average. At no point as a sex worker did I ever see twenty-five clients in one week, much less on “average.” With brothels mostly criminalized (the toleration of saunas in Edinburgh and walkups in London’s SoHo have recently come under attack from police), it is unlikely any full-time sex worker in the UK would see such a turnover. And most sex workers here are not full-time.
Finally, the model also does not make allowances for double-counting. Many sex workers are already declaring their income for tax purposes; they’re just filing it under the vague labels like “massage” or (in my own case) “personal services.”
Concerned about these inaccuracies, I contacted the ONS researchers directly to ask about their model. The researchers told me they were under time pressure to come up with a number, with EU directives demanding a result before September, and therefore did not have the time to contact anyone else about the prostitution figures. Given that at least one mainstream popular science book (mine) has been written on the topic, and that many academics have spoken on the topic in the media, how is this excusable?
When the statistics about the sex work industry are wrong, it matters. For the organizations that rely on people believing the sex industry is enormous to secure funding, the inflation of these numbers is a win. But for the sex workers themselves, who may face more intrusive police and tax investigations fueled by unrealistic depictions of the industry, however, the failure of ONS to correct their heavily flawed model is a disaster.
Already the inaccurate figure “£10 billion” has entered public discussion as a fact. Take note, we have just seen a future zombie statistic in the making.