The U.S. economy has been experiencing a partial shutdown as recommended by health authorities in response to the coronavirus pandemic. As a result, the second quarter of 2020 will be an unparalleled quarter for U.S. unemployment.
A St. Louis Fed On the Economy blog post published March 24 used different ways of looking at the number of workers with high-contact jobs to obtain projections for unemployment during this period.See Faria-e-Castro, Miguel. “Back-of-the-Envelope Estimates of Next Quarter’s Unemployment Rate.” St. Louis Fed On the Economy Blog, March 24, 2020. The blog post estimated that about 47 million workers could be laid off because of the need for physical distancing to help stem virus transmission. Based on that estimate and a few other assumptions, the unemployment rate in the second quarter would be a staggering 32.1%.This calculation was also based on the assumption that the labor force remains at its February level and that individuals who were unemployed in February remain without a job in the second quarter and do not drop out of the labor force. However, the unemployment rate could range between 10.5% and 40.6%, depending on more optimistic and less optimistic estimates for the number of at-risk workers.
The estimates in that blog post are turning out to be a reasonable proxy for what seems to be occurring in the labor market. Since the post was published, the number of initial claims for unemployment insurance through May 23 has totaled about 40.8 million, which is approaching the unprecedented estimate of 47 million at-risk workers.This total includes seasonally adjusted data for the week ended March 21 through the week ended May 23.
So far in this quarter, the Bureau of Labor Statistics (BLS) reported that the unemployment rate in April was 14.7%. This is in line with the range mentioned in the blog post, and the May and June jobs reports are still to come. Notably, the increase in the unemployment rate from 4.4% in March to 14.7% in April (a jump of 10.3 percentage points) is the largest ever recorded.
While the reported April unemployment rate is within the estimated range, the blog post also included some caveats regarding other factors that may affect the actual unemployment rate. The BLS measures the unemployment rate from a survey of households. If people say that they are not working and also say that they have not searched for a job in the last four weeks or are not available to work, they are categorized as out of the labor force rather than unemployed, and therefore not included in the unemployment rate. One caveat to this longstanding approach is that during a pandemic like COVID-19, people who recently lost their jobs due to state and local requirements to shelter in place or enforce physical distancing may not be looking for work in this environment, which would leave them out of the unemployment calculation.
The labor force participation rate declined by 2.5 percentage points from March to April, which represented the largest monthly drop for this series on record. At the same time, the number of people counted as out of the labor force but wanting a job (though not seeking a job or not available to work) rose significantly, from about 5.5 million in March to 9.9 million in April. The pandemic likely played a role in keeping many people from searching for work in the weeks prior to the survey, as people were asked to stay home during that period.For more details, see “Frequently asked questions: The impact of the coronavirus (COVID-19) pandemic on The Employment Situation for April 2020” from the BLS.
The BLS also reported that some people who were incorrectly categorized as employed but absent from work should have instead been categorized as unemployed on temporary layoff. The BLS estimated that including this particular group of workers would have raised the April unemployment rate by 4.8 percentage points, to 19.5% rather than 14.7%.Ibid, p. 11.
Another caveat mentioned in the blog post is that the calculations did not factor in the effects of fiscal measures on the unemployment rate. For example, under the Paycheck Protection Program (PPP), loans to small businesses are forgivable if the firms keep workers on the payroll, even though they are presumably not working and producing any output during this period. From an economist’s point of view, these workers are also unemployed, and they are essentially receiving a form of government insurance, only from the federal level as opposed to their state unemployment office.
Factoring in the additional workers who could be thought of as unemployed in the BLS survey and from the PPP would leave us with a higher unemployment rate. In fact, it may be closer to the middle of the range mentioned in the blog post (between 10.5% and 40.6%) rather than at the low end of the range, which is where the official unemployment rate for April is.
The second quarter is undoubtedly a bad quarter for the labor market, with some of the worst unemployment numbers the U.S. has seen since the Great Depression. Even during the early 1980s, when the U.S. experienced the worst recession in the postwar era in terms of peak unemployment, the unemployment rate reached a high of 10.8%. The reported rate so far in this quarter is already much higher and would be considerably higher if the factors discussed above are included.
It is important to keep in mind that the shock the economy is experiencing is driven by the pandemic, not by some underlying problem in the economy. Health authorities have asked households and businesses not to work in some high-contact jobs that might spread the virus and, therefore, not to produce as much as normal. Because of that, one could view the partial-shutdown policy as an investment in national health by refraining from ordinary economic production. Consequently, this shock, while having devastating impacts on the daily life of many wage earners and businesses across the country, should be interpreted differently from more traditional shocks that have occurred in the U.S. economy since World War II.