BLS attributed the drop from April to employersgenerating 2.5 million jobs in May as some states gave the greenlight for limited economic activity. (Image:Thinkstock)

There was quite a Friday surprisefrom the Bureau of Labor Statistics, which said the unemployment rate dropped from 14.7% in April to13.3% in May. 

But economists have called thatfigure into question for a variety of reasons — not the least ofwhich was a "misclassification error" by the BLS itself.

"There is good reason to beskeptical of [the Trump] administration's latest report," saidRobert J. Shapiro, chair of the consultancy Sonecon and a fellow atGeorgetown University, in a blog post Friday forWashingtonMonthly.

The BLS said 20.9 millionAmericans were unemployed in May. It attributed the drop from Aprilto employers generating 2.5 million jobs in May as some states gavethe green light for limited economic activity. 

"But it's not that simple,"Shapiro explained.

Behind the numbers

The jobless rate is determined bythe results of a week-long survey of businesses and households; thelatest poll took place during the week that ended May16.

That same week, BLS also reportedthat nearly 30 million Americans collected unemploymentbenefits. 

"That's a gap of 8,980,415people, enough to raise the May jobless rate 5.7 percentage pointsto 19.0% and much closer to economists' expectations," Shapiroexplained.

What happened?

The BLS counted anyone whoseemployers said they were being paid "even if they were not actuallyat their jobs," Shapiro points out.

In addition, the BLS considersfurloughed workers to be "unemployed on temporary layoff," he adds;however, if the BLS expects these individuals to return to theirprior jobs, it does not include them in the tally of unemployedworkers.

A recent University ofChicago study, though,estimates that 42% of workers furloughed due to the pandemic willnot get their jobs back, Shapiro points out. Plus, just 30% ofthose laid off are expected to get new jobs later in2019.

"With 30 million people receivingunemployment benefits, that analysis is more consistent withunemployment at 19.0% today than 13.3%," explained the former UnderSecretary of Commerce.  

Continuing jobless claims

According to JPMorgan ChiefGlobal Strategist David Kelly, "the measured unemployment rate still likelysharply underestimates the true unemployment rate (as it did inApril)."

The underestimation is tied tothe fact that "the rolls of the unemployed exclude millions who areready, willing and able to work but are not currently looking for ajob because of the pandemic — or who are being incorrectly classified as beingemployed but not at work when they are, in reality, unemployed,"Kelly added in a report Monday.

The change in continuedunemployment claims "should have been a better measure of netemployment changes, since they exclude claims filed but denied andnet out those hired or rehired from the rolls of the unemployed,"he said.

The continuing claims forunemployment benefits grew by 2.8 million between the April and Maysurvey weeks, he adds, to 14.2 million people. 

'Misclassification' by BLS

BLS, which is part of the LaborDepartment, acknowledged Friday that counting furloughed workers asunemployed would have increased the official jobless rate to 16.3%,an improvement from 19.7% in April.

Its note explained that if this "misclassificationerror" had not been made, the "overall unemployment rate would havebeen about 3 percentage points higher thanreported." 

There also was "a large number ofworkers who were classified as employed but absent from work,"according to the BLS.

"As was the case in March andApril, household survey interviewers were instructed to classifyemployed persons absent from work due to coronavirus-relatedbusiness closures as unemployed on temporary layoff," itexplained. 

The group added, though, that "itis apparent that not all such workers were so classified. BLS andthe Census Bureau are investigating why this misclassificationerror continues to occur and are taking additional steps to addressthe issue."

Why the discrepancy?

According to Shapiro, the 9million-person "discrepancy" should be seen in the broader contextof how BLS classifies those in the laborforce. 

"If you are out of work butdidn't look for a job over the previous four weeks, or couldn'twork for family reasons such as caring for children who are nolonger in school, BLS says you are out of the labor force," heexplained. "If you're not in the labor force, you're notunemployed."

In May 2019, BLS said 96.2million Americans weren't in the official labor force. As of May2020, that figure stands at 102.1 million, which is a change of 5.9million.

This "could explain nearlytwo-thirds of the 9 million-person gap between those receivingunemployment benefits and those counted as unemployed," theeconomist explained.

Read more: 

• 5 cities with the biggest, smallest increases inpost-pandemic unemployment rates
• $1.3 trillion down: These COVID-19-era unemploymentstats are not pretty
• Unemployment claims fall as states begin toreopen