But what about the evidence? The historical record turns out to be fascinating. The last 20 years of UK statistics provide a decisive refutation of the Summers-Wolf position. They contain two lengthy periods in which fiscal policy was consistently moving in one direction. According to data from the International Monetary Fund, between 1993 and 2000 the UK's cyclically-adjusted budget position was transformed from a deficit of 6.1 per cent of GDP to a surplus of 1.5 per cent. Every year was "deflationary", in Keynesian terms, in this seven-year period. Yet the economy enjoyed strong growth and falling unemployment. 

By contrast, from 2001 to 2009, the cyclically-adjusted budget position slithered from a surplus of 0.6 per cent of GDP to a deficit of 8.5 per cent. With two minor exceptions, every year was "expansionary", according to Keynesian terminology. And what happened to economic activity over these eight years? The answer is that, whereas in 2001 the UK was as near to full employment as it had been for a generation, by 2009 it was suffering from the after-effects of the worst recession since the 1930s. Again, the Keynesian view is contradicted. 

And do the American numbers have a different message? Summers became Treasury Secretary after a prolonged and remarkably successful effort under the two Clinton administrations to bring the US public  finances back to health after the fiscal extravaganzas of the Republican period. Just as in the UK, in virtually every year of the Clinton presidency fiscal policy was restrictive and output grew at an above-trend rate. Summers and Wolf may mock the proposition of "expansionary fiscal contractions". But the data show expansionary fiscal contractions to be far more common in recent Anglo-American experience than the deflationary fiscal contractions which the textbooks lead them and other Keynesians to expect.