Friday, April 19, 2013

Evidence-basted policy

A prominent economic study, supposedly showing that growth slows down once national debt gets above a certain level, has turned out to be based on a few simple errors.

Carmen Reinhart and Ken Rogoff’s research was quoted approvingly by supporters of austerity around the world, including George Osborne in a major speech in early 2010. Osborne has consistently argued that the government needs to reduce its borrowing and debt or risk all kinds of disaster, and this research gave him a handy few paragraphs of material.

But he won’t now reconsider.

Politicians like us to think – and they probably like to think themselves – that they go for evidence-based policy. This makes them look like wise, careful, well-informed pragmatists.

But, too often, they don’t go for evidence-based policy. What they go for is evidence-basted policy.

Here’s how to do it:

  1. Choose your policy.
  2. Scour the publications of friendly think-tanks and academics to find evidence that seems to support your policy.
  3. If you also find evidence against your policy, cut this off and throw it away.
  4. Marinate your policy in the evidence that most complements its taste, and cook as needed, sealing in that delicious evidential flavour.
  5. When your policy is ready to serve, it will be all the more appetising.

All the austerity camp have lost this week is one ingredient for their marinade. But this hardly matters to them: that wasn’t the reason they chose this course. They’ll keep serving it, even if it now tastes a little bitterer. A change of evidence doesn’t mean a change of course.

Sunday, April 07, 2013

UK and international welfare spending

Using OECD data from Declan Gaffney (a great blog post – do read it), I’ve made the two charts below. They look at the amount of money different countries’ governments spend on benefits, excluding pensions and benefits in kind.

The first shows that benefit spending in the UK has fallen as a share of GDP, not risen. Of the 18 countries Declan compares, the UK spending fall was the fourth-largest from 1980 to just before the financial crash, and the fifth-largest from 1980 to 2009.


The second chart shows in more detail how spending has gone up and down. To avoid a mass of criss-crossing lines, I’ve included just six countries. This also shows that the UK welfare state – not counting pensions – is not particularly expensive by international standards.