In my column for the FT this week I write about a subject which is freshly topical in the US but which takes me back 25 years to the debate in Britain in the 1980s about the right operational target for macroeconomic policy.

To make the case for new stimulus, the Fed needs better arguments. The past few weeks have settled, to my satisfaction at least, a long-running debate on this very topic. Rather than targeting inflation, central banks should keep nominal incomes growing on a pre-announced path: say 5 per cent a year. Nominal gross domestic product is the sum of inflation and growth in real output - and is the variable that monetary stimulus directly drives.

Samuel Brittan made the case for this approach decades ago on this page. The crucial point - how an increase in nominal GDP breaks down between output and inflation - is not something the Fed can determine, or should have to explain. There are pros and cons to this approach, but that is the decisive political virtue of casting the target this way.

When nominal GDP falls below track, monetary stimulus pushes it back. If inflation rises temporarily during catch-up, that is tolerated. In current conditions, this makes all the difference. The new GDP figures showed demand has fallen much further below trend than had been appreciated. With a nominal GDP target, that announcement would have led investors to expect new monetary stimulus. With the implicit inflation target that the Fed is assumed to use, it did not.