Against all expectations, the U.S. labor market rebounded sharply in May, adding 2.5 million jobs and bringing the headline unemployment rate down to 13.3%, from 14.7% in April. Consensus expectations leading up to last Friday’s Bureau of Labor Statistics (BLS) release called for a loss of an additional 7.5 million jobs, so analysts were off by a staggering 10 million people. That enormous gap between expectations and reality speaks to the unprecedented volatility and unpredictability of the economic cost of the coronavirus pandemic.
This surprisingly good report is also evidence of the amount of noise in the system. To compile the monthly data, BLS analysts usually conduct interviews via telephone and in person. COVID-19 precautions precluded real interviews in April and May, leading to a response rate 15% below pre-crisis levels. Furthermore, the BLS acknowledged that the peculiar nature of furloughs and temporary absences from work has muddled the definition of unemployment. “As was the case in March and April,” the BLS reports, “household survey interviewers were instructed to classify employed persons absent from work due to coronavirus-related business closures as unemployed on temporary layoff. However, it is apparent that not all such workers were so classified.”
The BLS estimates that adjusting for this misclassification, the unemployment rate for May would have been 3 percentage points higher (that is, closer to 16%). The same categorization error also skewed the April report, understating the unemployment rate by close to 5 percentage points. The bad news, therefore, is that the May unemployment rate is likely understated, and labor market conditions aren’t as good as they appear. The good news is that April data was even more understated, and the rebound in May was therefore even more impressive than headlines would indicate. This classification challenge is not a new development: It occurs particularly during periods of government shutdowns, when public employees are furloughed temporarily, and therefore technically unemployed until the government reopens, although they may not think of themselves as such.
Regardless of noise and data collection challenges, the May employment report demonstrated that, as state and regional economies cautiously reopened last month, there were jobs for workers to go back to. The biggest gains came precisely in those sectors that were most impacted by shutdowns in the first place. Leisure and hospitality led the way as restaurants began to open their doors, restoring over 1.2 million jobs during the month. Construction, education and health services, retail trade, manufacturing and professional services also experienced significant gains. The only sector to suffer a further sharp decline in May was government, driven largely by the continued closure of public schools and the furloughing of associated workers.