Maybe debt doesn’t matter after all

On Saturday I wrote about the paper by Carmen Reinhart and Ken Rogoff about the chilling effect high levels of debt seem to have on economic growth. Now I’m not so sure. David Hendry, the Oxford econometrician who, for his sins, taught me econometrics (actually, he mostly taught me to spot bad econometrics), writes to point out that:

The UK became a dominant world power with [debt/GNP] ratios between 1 and 2; and the UK grew at its fastest when its debt/GNP ratio was highest, not that any causality can be ascribed to that. But essentially there is almost no relationship I can find, having tried over many years, between debt/GNP (or changes etc.) and growth, unemployment, or inflation over 1860-2000. (see Hendry, DF (2009) `Modelling UK Inflation, 1875-1991′, Journal of Applied Econometrics, 16, 255-275; and Castle, JL and Hendry, DF (2009)`The Long-Run Determinants of UK Wages, 1860–2004′, Journal of Macroeconomics, 31, 5-28 ).

Well, this is beyond my pay grade. I’d back Ken Rogoff in a chess match against Hendry any day, but not so sure about the econometrics. One possible objection is that the definition of “high debt” used by Reinhart and Rogoff (90 per cent of GDP) looks a bit arbitrary. Hendry has numerous more technical concerns.

Here are the graphs Hendry sent me for UK debt ratios and economic performance:

Here are Reinhart and Rogoff in their own words. Here is my previous column about David Hendry’s research. Reinhart and Rogoff will hopefully respond; I’ll get to meet Carmen Reinhart at the Royal Economic Society annual meeting soon, and quite possibly David Hendry will be there too.

Tim Harford’s blog

This blog is no longer updated but it remains open as an archive.

Tim, also known as the Undercover Economist, writes about the economics of everyday life.