05 September 2009

Outcomes a better policy driver than advocacy research

In his discussion about child-rearing techniques Frank Furedi tells us:
The experience of the past tells us that when science is used to provide legitimacy to speculation, there is always a possibility that we will find what we are looking for. Researchers sometimes admit that they are looking for research-based evidence to justify a cause. ... Most of what goes by the name of parenting research is best described as advocacy research. Advocacy research does not set out to discover what's not known it seeks to convince and influence public opinion. [Emphasis in original] Paranoid Parenting: abandon your anxieties and be a good parent (pages 156, 160)
Of course it's not only in parenting that you can generate figures that confirm your prejudices. My own field, economics, is another obvious case in point. It's an important consideration in the world of policymaking. Mr Furedi raises the questions of breastfeeding or smacking children. Similar reservations arise in other fields: crime, for instance, or mental health. What exactly is the role of government when research findings are inconclusive and controversial?

Social Policy Bond could help. They would stipulate a targeted outcome, rather than the supposed means of achieving it. So if society's goal is, for instance, to reduce lung cancer rates, then a bond issue would - probably - lead to the sort of restrictions on smoking that we now see. But if the goal is to reduce crimes committed by young people, where the scientific evidence on raising children is far more equivocal, investors in bonds would have to investigate and explore a much wider range of alternative approaches, the nature of which need not be specified in advance. Advocacy research would be regarded with far more skepticism than currently, because investors in bonds targeting youth crime would be looking for results, rather than confirmation of their prejudices. They would have powerful incentives to find only the most efficient ways of reducing youth crime, and to that extent would be impartial in their observations and conclusions.

Social Policy Bonds' focus on outcomes would have the same impartiality in helping achieve a wide range of social goals where research findings are vague or hotly disputed, and so we are collectively paralysed into inaction. They include larger ones about which I have written many times before, such as climate change or the prevention of natural or man-made catastrophes.


ZenTiger said...

I often see an approach to a problem apparently making things worse. For example, rising pregnancy and abortion rates in the UK by under 16's, in spite of a huge sex education drive over the last several years.

In light of a failure to produce the promised outcome, their solution was to increase funding even higher. More free condoms, younger sex education, more advertising.

I think the entire approach is flawed, and only makes things worse.

How would your policy bond concept facilitate a recognition that the approach is flawed given the outcome hasn't been achieved?

(Sorry, point me to another post if the answer is relevant, rather than retype a response here)

Ronnie Horesh said...

Thanks Zen Tiger for your comment. I agree with you about the existing approach and its flaws. Typically a government agency looks at something like abortion rates or teen pregnancy rates and thinks it knows best how to reduce them. It identifies a cause (in this instance, lack of education and condoms) and channels other people's money into its preferred solution. When that doesn't work, it does more of the same. The Social Policy Bond mechanism is quite different. It would set explicit targets for abortion rates and teen pregnancy. It would reward their achievement however they are achieved. The sum of money it would devote to the purpose would be stipulated in advance and capped. Purchasers of the bonds would have incentives to seek out and implement the most efficient ways of reducing abortion and teen pregnancy rates. These ways could include a mass condom helicoptering and leafletting operation, but it would be up to bondholders to work out which ways are going to achieve the best result for each taxpayer dollar. Crucially, investors in the bonds would have incentives to terminate useless programmes. Efficiency - as well as transparency and a cap on expenditure - are built into the Social Policy Bond mechanism.

ZenTiger said...

Thanks for that explanation, I understand the concept now. (Actually, I just needed my memory jogged as it has been a while since I first read about this [here amongst other places], and your blog had slipped off the reading list when I changed browsers).

Good to see you still blogging.