'He was the only genius I ever met," says Alan Bollard, the biographer of pioneering New Zealand economist Bill Phillips.
Somewhat ironically, he is most renowned for something neither he nor Bollard would regard as his best work: the famous Phillips curve.
"The Phillips curve is what every economist knows him for but is not what they should know him for," Bollard says.
A Phillips curve is a graph that quantifies an inverse relationship between unemployment and inflation. The original arose from what Phillips described as a wet weekend's work in 1958, crunching the data on British unemployment and wages.
"I don't think Phillips necessarily saw a causal relationship. He saw a correlation," says Bollard, the previous Reserve Bank governor.
But his work was seized on by policy wonks and ultimately politicians, who saw it as offering a choice of trade-offs: low unemployment if you don't mind high inflation, or price stability at the expense of jobs.
"Some were saying we have a choice here. We can run an economy hot and get low unemployment or run it cool and get low inflation," Bollard says.
Phillips did not endorse this.
• Visionary Kiwi economist on radar at last
"He wasn't a policy nerd and didn't have confidence he knew what was required to change the world.
"He was a theorist but also very practical and he liked putting numbers on things."
Phillips was even more uncomfortable as the Phillips curve became a kind of poster child for Keynesianism, when it came under attack from monetarists.
"That was the most bitter debate ever in economics and I don't think Phillips liked that at all," says Bollard. "He resisted attempts to have him come down on one side or another."
The stagflation of the 1970s - when both unemployment and inflation were high - was widely seen as discrediting the idea of a Phillips curve. But it has subsequently been revived, albeit in modified or augmented form.
It is understood that other things matter for inflation as well - supply shocks like the current oil glut and inflation expectations.
But the insight that there is a relationship between how much spare capacity there is in the economy - of which unemployment is a key component - on the one hand and inflation pressure on the other remains fundamental to how central banks see the world and how they approach their task.
"Today most of the world's central banks would have some form of Phillips curves in their forecasting models and a derived working rule in their policy kits," Bollard writes in his biography of Phillips, A Few Hares to Chase.
He quotes US economist Paul Samuelson as saying "My students are constantly killing off the Phillips curve but like the phoenix it won't stay dead."
Most of Bollard's book is devoted to recounting Phillips' remarkable life.
However he devotes some of it to arguing that what Phillips should be remembered for is not so much the eponymous curve, but for pioneering work on how to stabilise economies, dampening cycles of boom and bust.
He brought an engineer's way of thinking to the problem, devising feedback mechanisms incorporated in a new mathematical model of the economy.
It assumed there is some target level of production where, in modern terminology, the "output gap" is zero, that is, full production with no inflationary bottlenecks.
"His stabilisation policies involved detecting any negative or positive output gap then taking corrective action by altering fiscal or monetary stimulus, allowing for the time this would take."
A further complication arises from the fact that the policy-driven adjustment to demand would affect prices as well as output.
"These price responses could help or hinder the stabilisation process and certainly could confuse the appropriate policy response," Bollard writes.
This is arcane and complex stuff, over the heads of us laymen.
But Phillips' conclusion, more than 60 years ago, that it was "likely a monetary policy based on the principles of automatic regulating systems would be adequate to deal with all but the more severe disturbances to the economic system" has proven quite accurate, Bollard says.
Phillips' work on stabilisation provided the forerunners of "policy rules" developed by later economists, like John Taylor, when central banks moved towards inflation targeting two decades later, and which guide central bankers to this day.
Their advantage is that they allow continuous policy adjustments according to technical criteria and independent of political or budgetary constraints of the kind which complicate fiscal policy.
An important finding for Phillips' work was that an economic system can become unstable.
"For the first time an economist had a scientific way to show that appropriate policies could help correct this, but they could quickly become complex too.
"If policymakers got it wrong, their badly formulated policy interventions could leave an economy worse off," Bollard says.
That is seemly modesty from a former high priest of central banking.
One could argue that with the rise of independent central banks, whose mandate is to deliver price stability, it has become too easy for governments to dodge any responsibility for managing or moderating the economic cycle.
"We have delegated all that to the temple priests over the roads," the politicians seem to say.
"We don't understand how they do it, but we have every confidence."
It's a cop-out.
Debate on this article is now closed.