Policy-based evidence making?

For someone with my politics, reading research from the Institute for Economic Affairs is never less than bracing, not least when you find yourself in sneaking agreement with it.  The recent IEA report - Quack Policy – Abusing Science in the Cause of Paternalism by Jamie Whyte - takes a robustly sceptical cleaver to a herd of sacred cows: minimum alcohol pricing and lower speed limits, for example, fail to trade-off their supposed benefits with the pleasures of drinking and the convenience of driving fast.  In many cases, Whyte argues, policy-makers start with their paternalistic opinions and prejudices, and then find the evidence to support them.  Policy-based evidence making.

At first sight, basing policy on evidence rather than prejudice, blind faith or ideology seems uncontentious, a 'no brainer' even.  But I don't think that scepticism about evidence-based policy should be the reserve of the type of people who will be harrumphing that there is no evidence for man-made climate change until the flood waters start lapping round their ankles.

Evidence-based policy has its roots in the concept of evidence-based medicine, which responded to the tendency of medics (alarmingly commonplace until the 1990s) to base interventions on custom and practice rather than any clinical data about what works. 

The elision from choosing cancer treatments based on their demonstrable impact on specific physiological circumstances, to choosing policies based on predictions of human behaviour is not smooth, however.  To start with, policy interventions are rarely based on controlled, randomised scientific trials that can isolate cause and effect from other factors.  Even where a good result seems to have followed a specific policy - the reduction in heart attack rates following the ban on smoking in pubs, for example - the causal links are not simple.  People and societies are more cussed, diverse and chaotic than cancer cells or bunions.

But there is a more fundamental sense in which evidence-based policy worries me.  It takes the politics out of policy, and creates a technocratic world where efficiency and value-for-money are all; where white-coated analysts can dispassionately assess solutions, tinkering with the apparatus of incentives, nudges and penalties to perfect citizens and society.  Tony Blair’s 1997 mantra – “what matters is what works” – was not just a financier-friendly disavowal of socialist dogma, but also a retreat from conviction politics (until their re-appearance after 9/11).

Evidence-based policy may have progressive aims (safer roads, better health, lower re-offending, fitter, happier, more productive people), but this managerialist approach excludes discussions of principle, of morality, of big ideas.  It cloaks opinions behind assertions of scientific fact. This focus is also inherently conservative; it is about tweaking the current system to optimise the way it moulds individuals’ actions, rather than considering whether it is the system itself that is rotten.