The Phillips Curve never died – it simply moved, argues Richard Thomas
Economists are unlikely victims of fashion. The closest most get to the catwalk designs is a subtle but daring contrast between the colour of their cords and elbow patches. But intellectually, for all its seriousness, the profession has developed a butterfly tendency to flit between theoretical fads. Fundamental rules and equations are pass.
Economists are being prodded further in this direction by their philosophy colleagues, who espouse post-modernism – a world in which everything is relative, nothing is sacred, and every idea can be dismissed as “socially constructed” (whatever that means).
Not that economists have given up on theories altogether, simply on the last one, whatever it happens to be. The process started in the 1970s when the old models appeared to have broken down and forecasting became a mug’s game. The spectacular economic policy mistakes of that decade, and of the 1980s – most notably in the British boom and bust – finally snuffed out any remaining self- esteem.
This is a great shame. Partly because economics is more fun when it tries to provide answers as well as questions, but mostly because the profession may have given up its traditional tools of diligent analysis, correlation and prediction just when they are coming back into their own.
One of the quintessentially old economist approaches was the Phillips Curve, which shows a stable, inverse relationship between unemployment and wage inflation. It is learned religiously by every economics student and then put safely into the historical, do-not-touch file.
Bill Phillips himself personified the old approach, which assumed that economies could be tracked, graphed and controlled. His curve, unveiled in 1958, took a straightforward approach. Phillips used British data on pay and unemployment between 1861 to 1957, and showed a robust predictable relationship between the two variables. Policymakers had a choice between the two, and could decide where to strike the balance.
Unfortunately, most economies went pear- shaped about a decade after the curve was let loose on the world. Unemployment rose in parallel with inflation, in apparent contradiction of the Phillips hypothesis. A new, ugly word – stagflation – was born.
A new theory, needless to say, popped up just in time. Milton Friedman, a decade after Phillips’s breakthrough, said because workers knew that lower unemployment would cause higher inflation, they would build in the expected inflation to their current pay demands – and inflation would take off.
In the Friedman world, wages and prices could only be contained while unemployment remained above a “natural” rate, later refined as the awkward Non-Accelerating Inflation Rate of Unemployment (Nairu).
The 1970s certainly looked a Nairuish sort of place, so Phillips and his near-century of data, were consigned to an early grave. Now, of course, the Nairu thesis is in trouble, not least in the United States. Unemployment keeps falling: wages have not gone into orbit.
There are still some diehard economists trying, like truffle-hunters, to sniff out the “true” Nairu so that it can be avoided. It isn’t working. Every time they claim to have isolated it – say at a 5% jobless rate – unemployment falls below it and still wages fail to lift off.
Why wage Inflation Will Stay Low, a paper by Ian Shepherdson, chief US economist for HSBC, offers an alternative to sticking blindly to the Nairu or dashing off yet again to another theory. His notion is simple but heretical: exhume the Phillips Curve.
In fact, Shepherdson argues, the Phillips Curve never really died – it simply moved position. At the end of the 1960s and early 1970s, the US Federal Reserve significantly loosened monetary policy, even though inflation was rising. From a peak of 9,25% in 1969 the Federal funds rate was cut to 3,75% by the spring of 1971. This took the real cost of borrowing (the Fed funds rate minus inflation) into negative territory for the first time in post-war history.
By cutting rates as prices rose, the Fed signalled to workers and firms that it was prepared to tolerate inflation. Workers therefore asked for more money (as Friedman predicted) while companies figured that they could accommodate the demands by passing the costs on to consumers through higher price tags.
The result was to significantly weaken, and worsen, the unemployment-price trade-off. And so it remained until Paul Volcker came along. The tough new Fed governor performed the opposite trick in 1979/81, aggressively tightening monetary policy even as inflation eased. Workers and bosses quickly got the new message, that inflation would not be tolerated, and curbed pay demands.
Shepherdson argues that Volcker – a “visionary” – managed to re-establish a stable Phillips-style relationship to the economy, albeit at a high price. For the last decade, inflation has been a minor feature in US pay rounds.
So, Shepherdson argues, it was the 1970s that were the aberration: “The basic association – lower unemployment with greater pressure on earnings – did not disappear into a nightmare Nairu world of ever-accelerating inflation. Instead, the curve simply shifted once, upwards and to the right, in the 1970s. It has since moved more or less back to where it was in the 1950s and 1960s.” The policy significance of this is profound. The Nairu school contends that once a certain point is reached inflation will suddenly become unstoppable. One more step and we might all be off the cliff into the inflationary abyss. By contrast, a Phillips world means we are on a clearly-marked slope, and control the descent.
This does not mean that unemployment can simply keep on falling. Indeed the curve gets steeper as the dole queue shrinks: HSBC estimates that while a 0,5 percentage point rise in the US unemployment rate would knock 0,5 percentage points off wage growth, a 0,5 percentage point fall in unemployment would add closer to 0,7 percentage points to wage growth.
But it does mean there is no need to panic. The US economy is not about to spiral out of control. It is not clear whether the Phillips-curvy world extends beyond the US borders but, given the convergence of many labour and product markets towards American lines it seems likely that, if it doesn’t yet, it will.
And on top of the greater leverage this implies for policy-makers, it should give economists pause for thought too.
In their rush to be the next to produce a new theory, the boffins shouldn’t forget there are some sharp tools available already, including Bill’s curve. If economics is to thrive, it must tend to its roots.