Hidden figures

A flattened curve

Why does low unemployment no longer lift inflation?

EVERY NIGHT at about 10pm the lights of the prisoner-of-war camp in Indonesia would mysteriously dim, to the puzzlement of the Japanese guards. They failed to spot the makeshift immersion heaters, used to brew cups of tea for the inmates, that had been cobbled together by a prisoner from New Zealand, William Phillips. These secret contraptions were just one example of his resourcefulness.

After the second world war he built a “hydraulic” model of the circular flow of income in an economy—a labyrinth of water tanks, valves and pipes that helped earn him an appointment at the London School of Economics. But neither of these exploits is the reason why Phillips is known to every economist today. His fame rests instead on his “quick and dirty” study, published in 1958, documenting a striking, decades-long relationship between British wage inflation and unemployment: the one tended to be high when the other was low. A downward-sloping curve, which he drew largely freehand, illustrated the point. The Phillips curve, as it became known, has been described as “probably the single most important macroeconomic relationship”. It has also been called the “least solid piece of work” he ever did.

The Phillips curve’s solidity and shape has been called into question more than once in the past 60 years, including in the period since the global financial crisis of 2007-09.…