COVID-19 first caused chaos in our labor markets with the lockdowns of 2020, which sent unemployment rates soaring to all-time highs. It has continued to disrupt labor markets into 2022 as worries about health risks have kept workers at home, exasperating labor shortages. Looking forward, as we learn to live with COVID, we will also have to adapt to the effects of long COVID, when symptoms such as fatigue, difficulty breathing and “brain fog” appear after COVID. In this commentary, I attempt to assess the risk to our labor markets from long COVID.
Understanding the impact of long COVID on the labor markets is complicated by the wide range of symptoms that can be present as well as the unpredictable duration of the symptoms. Critically, while long COVID can be moderate and resolve within a few months, symptoms can also be debilitating and linger for years. And long COVID is common. The Household Pulse Survey, run by the U.S. Census Bureau, estimates that, as of Aug. 8, 36 million U.S. adults have had or now have long COVID symptoms.1 This includes 32 million working age adults between the ages of 18 and 64.
Long COVID is defined as symptoms that appear after a COVID infection and last for three months or longer. As such, assumptions are needed to translate statistics counting all Americans who have had long COVID into statistics counting Americans with long COVID now. A report by the Federal Reserve Bank of Minneapolis, using data from the University of Southern California Understanding America Study, reports that 48% of Americans who reported having long COVID describe themselves as fully recovered. If this recovery rate, estimated in May to June 2021, continues to hold, this suggests that 15 million working-age adults now have long COVID.
Well … nothing good for sure. At a minimum, long COVID is adding a lot of uncertainty to an already very uncertain economic picture. Will long COVID cases in the U.S. continue to tick up with each subsequent wave of COVID? Will long COVID symptoms continue to affect workers similar to rates in the past? Will new COVID variants bring even more debilitating long COVID or will treatments continue to improve and help to stanch the impact of long COVID?
Absent a degree in medicine and, more importantly, a crystal ball, I have to make some assumptions about the future evolution of the disease. Without any evidence to the contrary, I assume everything more or less stays the same. In this “Groundhog Day” scenario, I assume that the rate at which new Americans join the unfortunate ranks of those with long-haul COVID is the same as the rate by which Americans recover and drop from those ranks.
But how many of these workers will have symptoms so severe that it will affect their ability to work? The University of Southern California Understanding America Study estimates that 30% of working age adults with long COVID experience symptoms severe enough to impinge on their ability to work. Such estimates are consistent with other studies. For example, a recent study in the Lancet, which drew participants from COVID support groups among other sources, found that 45% of long COVID sufferers had to work reduced hours and that 22.3% were not able to work at all because of continued COVID symptoms. Using these estimates, and accounting for potential upward bias in the Lancet study, I will assume that 15% of long COVID sufferers will be unable to work and another 15% will see their work hours reduced.
As detailed above, my best estimate is that 15 million working age Americans will experience long COVID at any given point in time. About 80% of working age Americans participate in the labor force, bringing us to 12 million potential employees with long COVID. Of these potential employees, I will assume 15% of them are unable to work. In which case 1.8 million Americans will be out of the workforce due to long COVID. This is 1% of the current U.S. labor force. Moreover, another 1.8 million Americans will be working reduced hours.
Now 1% may not sound concerning, but this will have large economic implications. Labor force participation rates declined following COVID and have not yet fully recovered. (I have written more about this previously). Among working age adults, the rate of labor force participation is 0.7 percentage points lower than its pre-pandemic peak of 83.1%.2 A 1 percentage point change in these rates is very significant. Another way to think about this is in comparison to unfilled jobs. The latest estimate from the Bureau of Labor Statistics is that there are now 10.7 million unfilled jobs.3 This is a dramatic departure from historical averages, which have traditionally been below 5 million job openings, and 1.8 million workers represents a shocking 28% of the difference between current job openings and historical averages. Or, if you want a dollar estimate, using average annual U.S. wages, this translates into nearly $90 billion in lost earnings per annum. And, most importantly, this represents 1.8 million lives that are severely upended by long COVID.
The statistics bureau has also been asking Americans whether they did not look for work specifically because of COVID in separate survey called the Current Population Survey. This question gets at the overall impact of COVID on labor force participation, summing up the effects of acute COVID, worries about COVID and long COVID on whether survey respondents can look for work. Interestingly, the bureau estimates that 523,000 Americans who want to work are not able because of COVID, less than a third of my estimate using the pulse household survey.
At a minimum, there is much uncertainty as to the impact of long COVID on labor markets. My estimate of 1.8 million Americans being unable to work with long COVID rests on the assumption that 50% of people who have ever had long COVID have recovered. Perhaps recovery rates are higher? Alternatively, the lower estimate of half a million Americans depends on a survey question that asks specifically about whether COVID affected one’s ability to look for work. It could be that long COVID may impact your ability to find and keep a job but have less of an effect on the job search process.
I don’t know, but it’s worth further investigation. Long COVID is here and is disrupting our labor markets. By reasonable estimates, it is already a factor in the labor market disruptions we face. Moreover, in the absence of a major medical advance, long COVID is going to continue to impact our labor force for the foreseeable future. Given the large economic costs of long COVID, advancing medical treatments should be a national priority. Simultaneously, we need to ensure that our social safety nets will protect these Americans who are unable to work because of long COVID.