Evidence-based policy making
Passing through into another room in which Womans Hour was playing. Two women, engaged in heated debate on prostitution: one was in favour of banning the industry, the other in favour of liberalising and regulating it, both agreeing that the mischief to be addressed is people-trafficking. Each supported their position with anecdotal cases from countries in which one or other course has been followed.
There’s obviously something distasteful about paying for sex, but it’s impossible to eliminate economic factors entirely. Is it prostitution if a woman puts out after being generously wined and dined? What if he buys her an expensive jewel? Were pay-sex to be even more outlawed, it’s much more likely that it would be driven further underground than that johns would start paying with bling.
But which policy on prostitution is most likely to have the greatest impact in reducing people-trafficking? Both the women arguing on Womans Hour today had very strong opinions, and they referred to countries such as Sweden (which has tightened its laws) and New Zealand (which has liberalised them). So it should be possible to design an impartial study looking at the impact of tighter or more liberal laws. Instead of pitting the two ladies into such opposition, why didn’t Jenni Murray (if it was her chairing the debate) suggest that each of them contribute their evident expertise to the study design? You don’t need to answer; firstly, it wouldn’t make such compelling radio, and secondly, of course, each of them would want the study to reinforce their existing prejudices (which, to each of them, is not prejudice, but established fact).
I felt like shouting, “shut up you miserable harridans, neither of you actually knows!”. Because anecdotes are not evidence, and however strong the evidence, neither of them would be likely to change their position. Yet it is a serious question, and both of them seemed to be knowledgeable. We do need reliable answers to this and to lots of similar questions in social policy, and we can find the evidence with the right design of study. Like clinical trials for medicines, studies need to be designed in advance, and in my opinion this should be done openly so that all shades of opinion and prejudice can influence the design. Web-technology like blogs and wikis are ideal for this sort of open collaborative work. State the purpose of the study (to find out which direction in pay-sex laws, towards liberalisation or towards prohibition, has the biggest impact on people-trafficking); anticipate and design for the obvious corrections – how do you correct for different social attitudes towards women in the respective countries being studied, for example, or the impact of poverty.
Open, evidence-based policy-making can be used in many areas of social, environmental and economic policy. Unfortunately, I don’t think either of the three main parties is yet willing to face up to the fact that almost certainly bobbies on the beat do not reduce crime. Of course, we’d need a properly-designed study to find out – in fact I’m sure that a meta-analysis of many already published studies would confirm it – but what would that do the Laura Norder rhetoric? Yet while political debate and policy-making is driven by rhetoric rather than reason, knee-jerk reactions to media hysteria and a desire to know the answer before the question, more and more people will be turned away from political engagement.