Economists are chewing over a puzzle that could have profound implications for the future path of US interest rates: has the Federal Reserve changed its thinking about the relationship between unemployment and inflation?
To be more precise: does the Fed now think that the US economy can operate with unemployment as low as 4.5 per cent in the long term without generating inflation?
Or does it simply think the inflationary effect of unsustainably low unemployment now takes longer to percolate through the economy than it did in the past?
This puzzle has preoccupied economists ever since Ben Bernanke presented the Fed’s monetary policy report to Congress earlier this month.
The report shows Fed policymakers think unemployment will be between 4.5 and 4.75 per cent this year and next, while inflation will decline from between 2 and 2.25 per cent this year to between 1.75 and 2 per cent in 2008.
This is intriguing, because unemployment at 4.5 to 4.75 per cent is lower than most estimates of what economists call the “non-accelerating inflation rate of unemployment” (Nairu) – the rate of unemployment below which inflation picks up.
Mr Bernanke did not explain how the Fed reconciled its unemployment and inflation forecasts.
The enigma only deepened last week, when Fed-watchers scoured the latest set of Fed minutes for clues as to how its reasoning evolved, and found none.
It is of course possible that the Fed no longer thinks the Nairu concept is a useful guide to policy. Certainly, many Fed policymakers talk that way.
Steve Cecchetti, a professor at Brandeis university, says they prefer broader measures of resource utilisation to estimates based on unemployment alone.
“I would think in terms of potential growth rates,” he says. “What matters are output gaps rather than unemployment rates.”
Yet this is at best only a partial answer to the puzzle.
Joblessness is the most important element of any measure of slack in the economy, and the notion of an unsustainably low unemployment rate is embedded in the staff models that guide policymakers’ decisions.
Given this, one possible interpretation of the Fed forecasts is that the Fed still feels the Nairu concept is relevant, but that unemployment could stay as low as 4.5 per cent without generating accelerating inflation.
Larry Meyer, a former Fed governor now chairman of Macroeconomic Advisers, says the Fed staff’s estimate of the sustainable unemployment rate is around 5 per cent, but officials understand that this estimate is “imprecise and time-varying”.
Fed policymakers – who are a bit more optimistic on inflation and unemployment than the staff – think of the sustainable rate as lying somewhere between 4.5 per cent and 5 per cent.
There are, however, other possible explanations. The Fed may still think unemployment at 4.5 to 4.75 per cent is below the sustainable non-inflationary rate, but believe the effect will be masked over the next two years by the unwinding of temporary factors that boosted inflation over the past couple of years: oil, commodity prices and accelerating rents.
Peter Hooper, chief economist at Deutsche Bank Securities, says: “Their models may give them relatively favourable results over the next year or two, but also show that if you extend out further the transitory effects drop out, and you have a risk of inflation beginning to rise”.
What makes it terribly hard to tell whether the sustainable rate of unemployment has fallen is that something else is going on at the same time.
Economists call this the flattening of the Phillips curve – the line that describes the short-term trade-off between unemployment and inflation.
For reasons largely to do with better anchored inflation expectations and the impact of globalisation, inflation no longer picks up as rapidly as it did before when the economy is operating beyond its sustainable capacity.
“It is so slow,” says Mr Meyer. “We are talking tenths of a percentage point a year. With all the random shocks going on, how can you follow that?”
Bruce Kasman, chief economist at JPMorgan, says a flat Phillips curve cuts both ways.
It means inflation is slower to rise when the economy is overheating, but also implies that it takes a more brutal increase in unemployment to bring inflation back down if it does get stuck at a higher-than-acceptable level.
Investors hoping for a definitive answer as to which of these different explanations the Fed believes may be disappointed.
In all probability, Mr Bernanke himself does not know whether the sustainable unemployment rate has fallen, or whether the combination of beneficial short-term factors and a flatter Phillips curve is masking the fact that the underlying economy is overheating.
This is likely to mean two things for interest rate policy.
First, the Fed will be reluctant to raise rates pre-emptively simply because unemployment is low, without corroborating signs of wage-push inflation.
Second, unless unemployment rises, it will continue to take inflation risks seriously long after the monthly inflation data improves.
