Evidence-based policy is the most meaningless phrase in politics today

Ryan Bourne
Follow Ryan
PLENTY of meaningless phrases dominate our public discourse. Think of “the global race” or “a fair share”. The former is a good soundbite, but who exactly are we racing and what does winning entail? History has shown us that prosperity is not a zero-sum game, so the idea our economic progress is someone else’s loss is bizarre.

A fair share is even vaguer. Fairness is inherently subjective, and how much governments should intervene to correct for perceived unfairness is debatable. Yet politicians like Margaret Hodge, with her battles over corporation tax, have helpfully decided that they should define what it means for all of us, using the term incessantly.

But the most infuriating and meaningless phrase of all is the lament that we should have “more evidence-based policy” (usually met with knowing nods of approval). Don’t get me wrong, I like a good piece of evidence as much as the next man or woman. And policy should be rooted in an understanding of the facts. Yet there are three good reasons why I’ve decided this phrase has lost all meaning in policy discussion.

First, evidence-based policy tells you nothing about the assumptions behind what you are trying to achieve. This often means policymakers from left and right speak across each other on the same issue. A recent paper from Karel Mertens of Cornell University, for example, showed that marginal tax rate cuts enhance economic growth but also widen pre-tax income inequality. Now, everyone pays lip service for the need for more to be done to “boost growth”, yet there has been a strange dearth of left-wing economists or commentators holding up the paper as a guide to policy. Why? At best they have a different idea of what is important. If your aim is to minimise inequality, the above research has hugely different policy implications than if you want to maximise growth.

Second, most public policy evidence is not clear-cut because, unlike in science, it is much more difficult to undertake control trials. Take the “austerity” versus “stimulus” debate. Many compare the economic performance of recent years for the US and UK, and conclude that Barack Obama’s fiscal stimulus has been vindicated with higher GDP growth. They talk about it as if it has been a clear-cut experiment. Yet there are hugely different underlying conditions for the two countries (think shale gas, tax, regulation), and the US always seems to recover better than the UK. That’s why we’ll still be debating this 100 years from now.

Finally, some people just say they want evidence-based policy when they clearly don’t. The Independent columnist Owen Jones, for example, tweeted the other day that the CLASS think-tank he was part of intended to bring “evidence-backed policies into debate”. Yet Jones himself continuously agitates for rent controls, which are one of the few policies all trained economists agree is a terrible idea – leading to lower quantity and quality of housing.

For these reasons, I recoil whenever I see or hear a public figure use the term, because it is usually used as a device to shut down debate rather than enhance it. Yes, evidence is important. But so are underlying aims and values. That’s why we need ideas in politics just as much as data.

Ryan Bourne is head of economic research at the Centre for Policy Studies.