What does a successful policy on prostitution look like? The first and obvious answer is “nothing like what the UK is currently doing”. Although the sale and purchase of sex are both legal in the UK, related activities (such as soliciting, kerb crawling, or keeping a brothel) are criminalised. The Crown Prosecution Service includes prostitution within its “violence against women framework” – but despite this, since 2013, more women than men have been targeted under prostitution law.
Whichever side of the ideological lines you fall on when it comes to the sex industry, a system that punishes those it identifies as victims more than those it identifies as perpetrators can only be described as a terrible failure. Which is why new guidance to the police stresses that they have a “responsibility to protect” women in prostitution, and should treat these women “not… as offenders per se but people who may become victims of crime.”
We need to look at how the law is enforced, but also at what that law is. In 2014, the Northern Ireland Assembly passed a law introducing what is known as the Nordic Model, based on the legislation in Sweden: under this, the sale of sex is decriminalised, while the purchase is made illegal. Westminster parliamentarians have voiced support for a similar law in the rest of Britain. But in the same year, another regime informed by very different principles was implemented in one UK city, and more could follow.
Leeds City Council instituted a “managed area” in the Holbeck district in 2014. The enforcement of laws against soliciting, kerb crawling and loitering was suspended, effectively decriminalising prostitution in the area. At the same time, a liaison officer was appointed to work closely with women in prostitution. An “unofficial pilot” begun in June 2014 was followed by an official pilot launched in October 2014. In January 2016, it was reported that Leeds City Council considered the managed area a success and had made it permanent. The announcement was controversial – just weeks earlier, 21-year-old Daria Pionko had been murdered in Holbeck, where she had been selling sex – but Birmingham councillors have nevertheless expressed interest in emulating the project.
One of the problems in deciding which approach works best is a fundamental disagreement on objectives. For abolitionist campaigners (who often advocate the Nordic model), the priority is securing the safety of women in prostitution – but this aim is embedded in an understanding of prostitution as intrinsically harmful to women, both a cause and a consequence of sexual inequality. Opposing them are sex workers’ rights advocates, who see prostitution as a consensually undertaken economic activity made perilous by social stigma, and who consequently promote liberal frameworks such as the decriminalisation in Leeds. Others locate success elsewhere entirely: Birmingham councillor Majid Mahmood has described managed areas as “the only way to get prostitutes away from family homes.” The predominant concern there is not to protect women from violence, but to minimise nuisance for other residents.
Researching prostitution is fraught. Shamed and legally vulnerable, participants might well wish to avoid engaging with official agencies at all. The violence attending the trade can make investigation dangerous: Melissa Farley, executive director of the organisation Prostitution Research and Education (mission statement: “to abolish the institution of prostitution”), recalls a pimp putting a gun to her head in a legal brothel in Nevada.
But beyond this, there is one issue that makes it particularly difficult to get good quality evidence on prostitution: it is impossible to run true experiments to see how your chosen intervention performs. That doesn’t rule out the scientific method entirely, but it does demand a careful approach. In his 1969 essay Reforms as Experiments, Donald Campbell noted that “where randomized treatments are not possible, a self-critical use of quasi-experimental designs is advocated. We must do the best we can with what is available to us.”
You cannot assess an intervention in prostitution as you would a pharmaceutical treatment, with double-blind testing and controls for all variables; but you can design trials in such a way as to get the best possible idea of what (if anything) your intervention does.
Unfortunately, as Campbell explained, not all social policy interventions are informed by this ethos: “Ambiguity, lack of truly comparable comparison base, and lack of concrete evidence all work to increase the administrator’s control over what gets said, or at least to reduce the bite of criticism in the case of actual failure [...] If the political and administrative system has committed itself in advance the correctness of efficacy of its reforms, it cannot tolerate learning of failure.” But as we’ve already seen, when it comes to prostitution, there isn’t even agreement about what “failure” consists of.
Is prostitution an ill to be eliminated, or is the expansion of the sex industry acceptable under some conditions? Are men who pay for sex a problem, or unreasonably stigmatised consumers? The disputes run deep.
Since around 2009, when the campaigning journalist Nick Davies called concerns about trafficking a “moral panic” and compared reporting on the subject to “the now notorious torrent [of misinformation] about Saddam Hussein's weapons”, it’s often been assumed that the sceptical, well-evidenced position is one in favour decriminalisation. It is, however, perfectly possible to produce bad studies in support of liberalised prostitution regimes, by making the mistakes Campbell described in his article. By failing to acknowledge confounding variables, using confusing reporting periods, omitting comparison with a control area not affected by the intervention, and withholding research from public scrutiny, the internal assessment on the Leeds managed area does just that.
The first problem is accessing the information at all: although researchers conducted an analysis of the managed area in 2015, this was not released to the public, and is considered a “confidential internal document” according to one of its authors. However, Leeds City Council did send me an executive summary of that analysis on request, and one of the first obvious oversights is a failure to investigate unintended consequences.
It does not count the number of women in Leeds involved in prostitution before or after the managed zone was introduced, nor does it mention the possibility of trafficking – even though these seems like obvious examples of variables that could be influenced by the policy, and should be monitored. A 2013 analysis in the journal World Development found that where prostitution is decriminalised, “the scale effect dominates the substitution effect”. Or in other words, rather than local women moving into prostitution once legal barriers are removed, women from elsewhere must be imported to fulfil the increased demand.
According to a Mail report on the managed area, this is what happened in Holbeck: “About half the prostitutes are English, some locals and others from Manchester, Sheffield or Liverpool. Some have been here for years; others arrived 16 months ago when the ‘legal’ red light zone was introduced. The other half is European, mostly Romanian, Hungarian and Polish.” (Daria Pionko was Polish.) This observation could be inaccurate, or it could reflect national trends rather than an effect of the managed area – but without the data, and with no comparison to a similar district without the intervention, it’s impossible to know the truth.
The issues with the Leeds policy don’t end with what isn’t measured: there are also problems with the metrics it does track. For example, one of the main objectives of the Leeds project was to increase trust between women in prostitution and the police, and the executive summary claims that this has been achieved (Ugly Mugs and Basis Sex Work Project are both services to which women in prostitution can report crimes against them):
"The second main outcome has been significant evidence of increased Ugly Mugs reporting: A year on under the pilot scheme, 2014/2015, there has been a significant increase in reporting. From 1st April 2014 up until March 31st 2015, 73 reports had been taken by Basis Sex Work Project (formerly Genesis) of which only 4 per cent were not shared at all. A further 46 per cent were shared anonymously and an increase to 50 per cent was reported with full details to the police. In first quarter of 2015/2016 100 per cent of Ugly Mugs reports made were formally reported to the police."
The period “1st April 2014 up until March 31st 2015” is 12 months. The “first quarter of 2015/2016” is only three months. To draw its conclusions, the executive summary relies on two incomparable periods with data collected by different agencies. Moreover, the two reporting periods don’t correspond to the institution of the managed area. Only three months of the first period took place before the unofficial pilot began in June. It is, generously, a stretch to call this “significant evidence of increased Ugly Mugs reporting”.
Dr Teela Sanders, who wrote the assessment, tells me in an email that: “[W]e of course have more details now since March 2015 which demonstrate this objective has been very successfully met.” She also cites 2013-14 reporting data not included in the executive summary: “In 2013/2014 there were 52 reports taken, of which 38 per cent were requested not to be shared with the police, 45 per cent shared anonymously, and only 17 per cent shared their full personal details with the police.”
This does indeed show an increase in women’s willingness to share details with police. But there is also a significant increase in overall reports from 52 in 2013-14, to 73 in 2014-15. Does that increase reflect enhanced trust, an enlarged population of women in prostitution, more violence from men, or some combination of the three? Without figures on the number of women selling sex in the area before and after the introduction of the policy, we cannot begin to say.
In effect, the managed area is not one intervention but at least four related ones: suspension of enforcement against women in prostitution, suspension of enforcement against punters, the introduction of the liaison officer, and the relocation of outdoor prostitution in Leeds to one area. Any one of these, or a combination of them, might be responsible for a particular effect – perhaps liaison officers build trust, but decriminalising punters increases demand. Comparison with a carefully selected control area could have been informative here; without it, all we can do is guess, or accept the managed area wholesale.
However, there is one point on which we can make a fairly secure before-and-after comparison. Measuring non-fatal violence against women in prostitution is difficult because women may be reluctant to report, and police may fail to take it seriously. However, fatal violence is easier to tally. According to Counting Dead Women, which records all women killed by men in the UK, Daria Pionko was the first woman in prostitution to be murdered in Leeds for a considerable time. That doesn't mean that the murder was a consequence of the managed area, but when a death takes place 18 months after a reform following many years with no deaths, it's difficult to claim that the reform has actively improved safety.
In fact, because of the paucity of data made available on the managed area, and the limited analysis to which the available data is put, it’s not possible to say very much at all about it – other than that, as an experiment, it is tragically unilluminating. Tragic, because we need to know more about the effect of social policy on prostitution. Tragic, because the reallocation of resources in Leeds amounts to wasted effort if we can’t know what was actually accomplished. And tragic, because in the process of very little being discovered, a woman affected by the managed area policy was killed.
For opponents of the Leeds policy, there can be no pleasure in these methodological failings, only frustration at a missed opportunity. For supporters of managed areas, there are hard questions to be answered about why such inadequate data has been promoted as proof of success. And for all parties, the problem remains: just what would an effective policy around prostitution aim to do, and how would it work?
