Is the "natural rate" of unemployment an out-of-date concept?
What evidence do we have for a natural rate of unemployment in an economy? That is one of the key questions facing central bankers as first the Federal Reserve and now the Bank of England have adopted a fixed unemployment target under the so-called Evans rule.
A new paper from Roger Farmer, professor and chair of the economics department at UCLA, seeks to answer that question. In his enquiries over whether the new-Keynesian Natural Rate Hypothesis (NRH) holds Farmer goes back to the data to see whether the empirical evidence supports the theory.
The key starting point for any discussion on the “natural rate” of unemployment is the Phillips curve. Named for the economist A.W.H. Phillips, it illustrates an inverse relationship between the rate of inflation and the unemployment rate – allowing economists to theorise that a “natural” equilibrium could be reached with a stead state of unemployment and gradually rising prices. After the US abandoned the Gold Standard in 1971 the relationship Phillips identified between unemployment and inflation appeared to break down, prompting Edmund Phelps and Milton Friedman to develop the theory.
The contribution of Phelps and Friedman was that although over the long term unemployment will move to the natural rate equilibrium, in the short run incorrect inflation expectations of market participants meant that the unemployment rate could differ from the natural rate. In other words they posited the notion that the Phillips curve was essentially a valid concept, but that expectations matter.Below is a convention chart of an expectations-adjusted Phillips curve, as they suggested:
By working out at what rate of unemployment wages can be expected to start exerting a price push, it was thought monetary policymakers could assess when policy tightening or loosening would be appropriate. The expectations-adjusted model also stressed the importance of central bank inflation targets in order to keep inflation expectations sufficiently well anchored.
As Friedman wrote in 1968:
“A lower level of unemployment [than the natural rate] is an indication that there is an excess demand for labor that will produce upward pressure on real wage rates. A higher level of unemployment is an indication that there is an excess supply of labor that will produce downward pressure on real wage rates.”
This helps to explain why both the Fed and Bank picked unemployment targets under the Forward Guidance framework. It also tells us that the US’s 6.5% and UK’s 7% targets are both some way above what the Federal Open Market Committee and the Monetary Policy Committee believe the “natural rate”, or the “nonaccelerating inflation rate of unemployment” (NAIRU) as it is now known, to be.
Yet is this assumption that there is a natural rate of unemployment at which inflation will start to rise supported by the available data?
Bloomberg’s Evan Soltas provides an interesting example of how focus on the NAIRU can lead to incorrect policy assumptions. Back in 1996 Janet Yellen, now widely touted by the blogosphere as their favoured pick for Fed chair, was president of the Federal Reserve Bank of San Francisco. By December of that year the unemployment rate was 5.4%, well below the assumed level of NAIRU, leading Yellen to argue:
“To my mind, labor markets are undeniably tight. We should be careful not to lull ourselves into a false sense of security about incipient wage pressures."
But, Soltas notes, the much-prophesised wage push never arrives. This despite the unemployment rate continuing to fall to a low of 3.8% in 2000. Yellen concluded that the Reagan labour reforms of the 1970s and impact of globalisation had pushed the natural rate of unemployment below its anticipated level.
So what is the data telling us about the NAIRU?Farmer took a look at the average inflation and average unemployment rate by decade from the 1950s to the 2000s. His findings are below:
As the author notes the shape of the data indicates “no tendency for the points to lie around a vertical line and, if anything, the long run Phillips curve revealed by this chart is upward sloping, and closer to being horizontal than vertical”. Of course, proponents of the natural rate hypothesis could claim that the natural rate changes over time, but since they do not provide any theoretical framework for how this would work the idea has “no predictive content”. Or as Farmer puts it:
“A theory like this, which cannot be falsified by any set of observations, is closer to religion than science.”
Under the new-Keynesian model there can be no involuntary unemployment. As such it cannot explain the secular fall in average weekly hours worked that has persisted since the mid-1960s. Moves in the labour force participation rate also appear not to be strongly correlated with the business cycle – suggesting the current trend is not cyclical in nature.
Instead of looking for an ever-shifting “natural rate” Farmer suggests that high involuntary unemployment can persist as an equilibrium and importantly that “the equilibrium that prevails is selected by the animal spirits of market participants”. It is possible, therefore, that in the aftermath of an economic shock the point at which wages start exerting a cost push can be substantially lower due to the collapse in risk appetite and consequent overwhelming incentive to save.
Though I have not yet tested his suggested belief function model against the UK’s experience since the onset of the Great Recession, I think this paper provides an important warning against relying on past assumptions about the relationship between unemployment and inflation in addressing the current economic malaise. It might also suggest that the central bank unemployment thresholds set under forward guidance may be an unhappy compromise between hawks and doves that offers a poor guide to the amount of slack in the underlying economy.
Further ReadingThe Natural Rate Hypothesis: An Idea Past Its Sell-by Date - Roger E.A. Farmer
The Financialisation of Labour - Frances Coppola
Underemployment as a Challenge to Orthodox Economics - David Spencer