Don't blame economics for Reinhart and Rogoff's blunder; blame them

How a correlation slowly morphed into a policy prescription.

New Statesman

Samuel Brittain has a piece in today's Financial Times about "the spell of magic numbers" and the latest example of someone falling under its sway, Kenneth Rogoff and Carmen Reinhart:

This episode has lessons well beyond the immediate fiscal debate. Generations of economists worried that their subject was not regarded as a hard science, a syndrome labelled as “physics envy”. Milton Friedman complained that it had not unearthed key numbers or ratios that could be regarded as constants…

The general moral is that economic analysis and policy would benefit from a less credulous acceptance of each purported research finding. It would pay to wait until a number of different studies employing different techniques point in the same direction and have survived professional criticism. In the meanwhile we can go quite a long way with a few well established elementary generalisations and inferences from them.

It's an important piece, because economics – perhaps more than any other social science – has fallen prey to the idea that if we just weigh and measure everything we can, soon we will be able to understand humanity as well as we do nature.

Going even further back into the depths of economics than Brittan does, the very basis of the subject is warped slightly by a need for numbers and mathematics. On the first day of an economics course, you are told that the object of study is utility, preferences and incentives. You draw utility curves, illustrating how someone may prefer two beers and a burger to two burgers and a beer. You discuss the assumptions the subject makes about human behaviour: that we will always prefer things which make us happier, that we are capable of knowing beforehand which is which, that happiness is a one-dimensional variable. And then, you are told that the easiest way to measure preferences is by looking at money.

And so economics shifts from becoming a study of how choices are made to a study of how monetary choices are made. If something affects happiness but is outside the market, it gets pushed out from basic economic study. And so you get the typical left-wing criticisms of economics as a right-wing discipline, which fails to take into account relationships, or society, or love.

But while that's a more fundamental criticism of economics, it's not one which we need to resort to to attack Reinhart and Rogoff. In fact, we don't even need to take Brittan's less dramatic reading of the problem, and accuse Reinhart and Rogoff of chasing after a magic number. Because the thing is, Reinhart and Rogoff's famous paper was actually rather prosaic.

Even before you iron out its flaws, the paper never claims to have found a magic number; nor, in fact, to have discovered any causal link at all, instead merely discussing "association" between high debt-to-GDP ratios and low growth (an association which is pretty close to being a mere mathematical fact – "a/b is likely to be low if b is growing quickly).

The problem came when the economists moved outside the realm of peer-reviewed academic papers and into public policy. That was when a (spurious, we now know) correlation became a magic number.

Take Carmen Reinhart in 2011:

Reinhart echoed Conrad’s point and explained that countries rarely pass the 90 percent debt-to-GDP tipping point precisely because it is dangerous to let that much debt accumulate. She said, “If it is not risky to hit the 90 percent threshold, we would expect a higher incidence.”

Or Kenneth Rogoff in 2012:

In a series of academic papers with Carmen Reinhart – including, most recently, joint work with Vincent Reinhart (“Debt Overhangs: Past and Present”) – we find that very high debt levels of 90% of GDP are a long-term secular drag on economic growth that often lasts for two decades or more. The cumulative costs can be stunning. The average high-debt episodes since 1800 last 23 years and are associated with a growth rate more than one percentage point below the rate typical for periods of lower debt levels. That is, after a quarter-century of high debt, income can be 25% lower than it would have been at normal growth rates.

Economics isn't what failed Reinhart and Rogoff. Being gracious, we could say it's the need for the media and political establishment to have simplified versions of economic arguments to latch onto. Being ungracious, we could say it was the need for Reinhart and Rogoff to boost their profiles by providing simplified versions of economic arguments.