New Relationships Between Inflation and Unemployment

August 2013

Richard Wagreich – Quantitative Investment Analyst
Alan Levenson – T. Rowe Price Chief Economist

As the Federal Reserve weighs when to begin “tapering” its unprecedented experiment in quantitative easing, it must consider its view of NAIRU—the non-accelerating inflation rate of unemployment. While simple in concept, NAIRU is notoriously difficult to estimate. Structural factors, such as skill mismatches or the availability of jobless benefits, may allow inflation pressures to emerge even at relatively high unemployment levels.

An analysis of the relationship between unemployment and job vacancies (the so-called Beveridge curve) suggests that NAIRU has risen since the 2008–2009 recession, at least in the short term. While the Fed shares this view, it appears to believe the increase is temporary, and its long-term NAIRU estimate remains well below what the data currently suggest. This could add additional uncertainty to Fed policymaking in 2014.

This post reflects five main premises regarding this relationship made by Quantitative Investment Analyst Richard Wagreich and T. Rowe Price’s Chief Economist Alan Levenson, as of August 15, 2013.


The Fed’s Policy Process Assumes a Direct Trade-Off Between Inflation and Unemployment

The Fed’s policy approach assumes that there is a trade-off between inflation and unemployment, such that a reduction in unemployment below the “natural” or full-employment level will cause wage and price inflation to accelerate. This level is known as NAIRU.

While conceptually straightforward, NAIRU is notoriously difficult to estimate. Structural factors—such as a mismatch between available jobs and the skills of the currently unemployed—may allow inflationary pressures to emerge even at relatively high rates of unemployment.

One way to estimate structural unemployment is to compare reported job vacancies with the unemployment rate, a relationship known as the Beveridge curve. Shifts in this curve—i.e., a higher level of unemployment for a given job vacancy rate—may indicate that structural factors have raised NAIRU (Figure 1).



Labor Market Data Point to a Shift in NAIRU Since the End of the “Great Recession”

Labor market data from the end of 2000 through 2012 suggest that the Beveridge curve has shifted in the wake of the deep 2008–2009 recession. While the unemployment rate has declined slowly over the ensuing recovery and expansion, it has remained elevated relative to job vacancies (Figure 2).

Assuming a 5% pre-recession NAIRU (in line with both the consensus among private economists and the Fed’s own model) the shift in the Beveridge curve implies that structural factors may have raised short-term NAIRU by more than a full percentage point. A shift of that magnitude could have significant implications for Fed policy.

The Fed’s current guidance indicates it will maintain near-zero levels of short-term policy rates at least as long as the unemployment rate remains above 6.5%, subject to the Fed’s near-term inflation forecast remaining within ½ percent of its 2% long-run goal. Fed Chairman Ben Bernanke has spoken of 7% as a possible threshold for ending large-scale asset purchases. These targets are designed to be well above NAIRU, reflecting the Fed’s traditional concern about policy lags.

A short-term NAIRU in line with the apparent shift in the Beveridge curve would imply that the Fed may need to reduce monetary accommodation earlier than anticipated in order to maintain an inflation level aligned with its stated policy objective. On the other hand, if the shift is temporary, the Fed could reasonably choose to ignore it.

Extended Unemployment Benefits Are Probably Not the Primary Cause of Higher Structural Unemployment

Several explanations have been offered for the apparent shift in the Beveridge curve. One is the repeated extension of unemployment benefits since the end of the last recession. While federal and state benefits were also extended in past recoveries, current benefit periods are the longest on record. Some analysts have suggested these extensions may have reduced the incentive for the unemployed to pursue less desirable or lower-paying job openings.

Given that roughly half of all unemployed workers (job leavers, new entrants, and those returning to the labor force) are ineligible for jobless benefits, extended benefits are unlikely to be the primary driver of the rise in long-term unemployment. Empirical studies support this conclusion.1

Another explanation for the post-recession rise in structural unemployment could be the severe sectoral labor market adjustments stemming from the financial crisis, the collapse of residential housing construction, and the prolonged retrenchment in government finances.

Sectoral dislocations may create skill mismatches between the pool of unemployed job seekers and available openings. Specialists in residential homebuilding trades, for example, may be less qualified for jobs in health care. This can lead to a temporary rise in structural unemployment.

Sector-Level Data Are Mixed on Whether Structural Unemployment Is Temporary or Permanent

Beveridge curves for selected U.S. economic sectors yield an ambiguous view of whether the rise in structural unemployment is temporary or persistent. T. Rowe Price’s analysis covered the period April 2009 through December 2012 and was based on data available from the Bureau of Labor Statistics (BLS), with seasonal adjustment and smoothing functions applied.

The construction sector experienced some of the largest labor market disruptions during the recession and the ensuing recovery. However, through mid-2012, the Beveridge curve appeared to be reverting toward its pre-recession position—although the vacancy rate has since moved higher (Figure 3).

The manufacturing sector also experienced a deep structural shift during the recession. However, as the job vacancy rate has risen, unemployment has not declined to pre-recession levels, suggesting a persistent increase in structural unemployment in the sector (Figure 4).

While sector-level unemployment and job vacancy data must be interpreted with caution, it appears that structural unemployment has risen in some industries. At least part of this increase persists more than four years after the trough of the last recession.

Most Private Economists, as Well as the Fed, Continue to Place Long-Term NAIRU Significantly Below 6%

The Federal Open Market Committee’s most recent estimate (June 19, 2013) places short-term NAIRU at 5.6%—a level consistent with its stated 6.5% unemployment threshold for considering an initial increase in policy rates. This suggests that Fed officials see any upward shift in structural unemployment as a short-term issue.

Addressing the structural unemployment question in a November 2012 policy speech, Mr. Bernanke remarked that sluggish final demand, not a structural skills mismatch, may have made U.S. firms more selective in their hiring decisions—a factor that presumably will dissipate as the recovery continues.2

Our own view is that NAIRU has risen to 6.0%–6.5%, at least in the short term, as reflected by the shift in the Beveridge curve. This would leave the Fed with significantly less maneuvering room than its current estimates suggest. While wage pressures still appear to be well contained, inflation trends will bear close watching as we move through 2014.



This material is provided for informational purposes only and is not intended to be investment advice or a recommendation to take any particular investment action. The views contained herein are as of August 15, 2013, and may have changed since that time.


1 See, for example, Do Extended Unemployment Benefits Lengthen Unemployment Spells? Evidence from Recent Cycles in the U.S. Labor Market, Federal Reserve Bank of San Francisco, April 2013.

2 Ben Bernanke, The Economic Recovery and Economic Policy, Speech to the Economic Club of New York, November 20, 2012. See in particular footnote 4.