Given its belated reaction to the inflation shock, the Federal Reserve has been engaged in a rapid catch-up exercise that has seen it raise interest rates by 375 basis points (3.75 percentage points) so far this year. Potential consequences of the rapid pace and magnitude of the policy tightening are likely to include further turmoil in financial and property markets and a recession in 2023.
A failure to understand the fundamentally unique nature of the labor market recovery following the pandemic recession may have contributed to the Fed’s decision to maintain its ultra-accommodative policy stance for far too long. In hindsight, the inability to discern shifts in the labor market dynamics and in the economy’s underlying productive capacity in a timely manner proved to be costly errors.
Fed policymakers were cognizant that the three recessions that preceded the 2020 pandemic shock had all been characterized by “jobless recoveries” in which marginalized groups had faced considerable hurdles in their attempts to re-enter the workforce. Furthermore, the flattening of the Phillips Curve, especially during the 2010-2019 period, had convinced central bank officials that a tight labor market may not automatically imply higher inflation.
Consequently, many Fed officials had a strongly dovish tilt early on. In a March 2021 speech, the San Francisco Fed president, Mary Daly, observed: “We will need to continually reassess what the labor market is capable of and avoid preemptively tightening monetary policy before millions of Americans have an opportunity to benefit. These efforts are critical to support the broad economy and aid the inclusion of historically less advantaged groups, including people of color, those lacking college degrees, and others who face systemic barriers to equitable employment and wages.”
In retrospect, fear of long-term labor market scarring and concern about a jobless recovery following the pandemic shock proved to be unfounded. Instead, a rapid job market recovery and a persistent shortage of workers stoked wage pressures and contributed to the surge in inflation during the 2021-2022 period.
To understand why the labor market outcome following the 2020 pandemic recession was so different from the prior three recessions, it is necessary to grasp the unique features of the current economic recovery.
David Autor and Daron Acemoglu have noted that many jobs can be classified along two dimensions — routine versus non-routine and cognitive versus manual. Nir Jaimovich notes: “The distinction between cognitive and manual jobs refers to the extent of mental versus physical activity required in a given occupation. The distinction between routine and non-routine jobs refers to whether the job-specific tasks can be performed by following well-defined instructions and procedures.”
Following the 1990-91, 2001 and 2007-09 recessions, aggregate employment recovered at a much slower pace than did aggregate output. Jaimovich and Henry Siu found that, in the past, routine manual jobs were lost at a far greater pace as companies resorted to more aggressive automation and offshoring of such tasks during and after economic slowdowns and this resulted in so-called “jobless recoveries.”
Recent research by Philadelphia Fed economist Shigeru Fujita suggests that the employment recovery from the pandemic recession was radically different from prior recessions. Fujita notes: “The COVID-19 recession is different in that nonroutine manual service jobs, which have become more prevalent over the past few decades, were the ones most severely affected, while the manufacturing sector, which employs a large number of routine manual jobs, performed relatively well.”
Two sectors (manufacturing and construction), characterized by the presence of routine manual jobs, were particular beneficiaries of the post-pandemic behavioral and policy shifts. Increased spending on durable goods during the initial recovery phase and the ongoing efforts to reduce excessive reliance on China-centered global supply chains have both boosted demand for manufacturing-sector workers. Also, the boom in the housing sector observed between summer 2020 and summer 2022 led to a surge in demand for construction workers.
More recently, the shift in spending towards services (leisure, travel and hospitality) has led to a sharp increase in demand for non-routine manual workers. Such jobs are not easily automated. Those who bore the brunt of the initial job losses during the pandemic recession were thus able to reengage with the labor market quickly, often at much higher wages, once the recovery took hold and spending on services surged.
Meanwhile, there is growing evidence that the U.S. natural rate of unemployment and potential output were affected by the pandemic shock. Large-scale resignations and sectoral and geographical reallocations likely generated an increase in frictional and structural unemployment. This, along with negative supply-side shocks (sharp fall in legal immigration, decline in domestic labor supply, supply-chain challenges and energy shocks) likely lowered the potential GDP growth rate in the U.S.
Consequently, actual GDP (boosted by a spike in stimulus-juiced aggregate demand) likely exceeded its potential level by a greater extent (implying significant economic overheating) than the Fed had initially estimated. Furthermore, the slope of the Phillips Curve may have steepened as structural forces (globalization, demographics, labor’s bargaining power, etc.) that had previously contributed to disinflationary trends were now enabling an inflationary surge.
By failing to realize the unusually rapid nature of the labor market recovery and discern the early disappearance of economic slack, and by ignoring the possibility of a re-steepening of the slope of the Phillips Curve, the Fed made a series of costly errors. Belatedly trying to overcome recent mistakes will not only create economic pain but potentially generate new policy errors.
Vivekanand Jayakumar is an associate professor of economics at the University of Tampa.