Photo via Flickr user Chris Beckett
For months now, civil servants at the UK's Office of National Statistics (ONS) have been wrestling with Britain’s sex workers. Thanks to their efforts, the first "official" figures were released last week showing how much Britons spend on drugs and prostitution. Which turns out to be quite a lot.
The ONS reckon we spend about £11 billion ($17.8 billion) per year on drugs ($10.8 billion) and prostitution ($7 billion). As the Telegraph pointed out, it’s similar to the total amount Britons spent on wine and beer in 2013—or, as an Iranian news source gleefully noted, similar to the size of the UK agriculture industry.
Why are we suddenly measuring this? In a word, the EU. If you’re a European Union member state, one of the things you have to do is report your gross national income—roughly speaking, how much money your economy makes in a year. That figure gets used as part of the calculation that says how much you have to pay into the EU to fund it. The richer a country you are, the more money you pay in.
Measuring the total amount of money flowing through Britain isn’t easy, and over the years the EU has tried to standardize how countries work it out. One of the big problems has been measuring black markets, which isn’t reported to the tax man for reasons that should be pretty bloody obvious. So as of 2014, member states are expected to estimate the size of two key parts of the black market—illegal drugs and what is called "prostitution activity."
In short, the European Union is sticking its nose into Britain’s sex workers, and potentially changing the amount the UK has to pay in EU funding as a result.
But hang on a minute. If the black market isn’t recorded, how on earth do you measure this stuff in the first place? That challenge was handed down to the ONS, which published an initial report on the effort back in May. The short answer: There’s not much data to go on, what there is looks pretty weak, and trying to come up with a good estimate is extremely tricky.
Examining drugs first, you could look at seizures by the police, as some researchers have done. The problem with that, as the ONS point out, is that drug seizures are sporadic and the government explicitly tells people, “The number of drug seizures made and quantity of drugs seized should not be taken as measures of drug prevalence in England and Wales.” A better approach is to take the number of users reported in various studies, but then you’re dealing with self-reported figures, and people aren’t always honest or accurate in surveys.
Of course even if you know how many people take drugs, you still need to figure out how many, how much they’re paying, and what proportion of drugs are manufactured in the UK. To do this, you have to make a whole bunch of assumptions, some of which are pretty dodgy.
For example, the ONS says, “We assume that the purity-adjusted amount of drugs consumed by the average user remains constant over time. This assumption is probably false, but we have no data to challenge it.” On cannabis, “We assume that half of cannabis sold in the UK is imported and half home-grown. This is an arbitrary assumption.”
It’s the same for sex work. The ONS used research by Eaves, a charity that focuses on victims of sex trafficking. Eaves estimates that there were 7,000 off-street sex workers in London in 2004. To get a figure for the whole country they just scaled the figures up, assuming the number of prostitutes per person is the same whether you’re in Camden or Cornwall. Then to get figures for 2014, they assumed “that the number of prostitutes has the same pattern through time as the 16+ male population. This is a weak assumption based on the market for prostitutes' services. It is necessary because we have no time series data for the number of prostitutes."
That still doesn’t tell you how much income they’re generating. To get that, researchers looked at the PunterNet escort directory to figure out an average price—assumed to be the same across the whole country—and then made a whole bunch of assumptions based on studies in the Netherlands about how many clients sex workers see in a typical week, and how many weeks are worked over the course of an average year. That’s assumed to have remained the same since 2004. Under the "justification" column for these assumptions, the ONS put down “no information available.”
It doesn’t exactly inspire confidence, does it? It’s not that the ONS is bad at this—the officials doing the best they can in the circumstances—but the questions are almost impossible to deal with without a big, national, large-scale effort. That raises an important question: If these figures are largely guesswork and bullshit to begin with, how does it help anyone to include them in the EU’s statistics?
Follow Martin Robbins on Twitter.