The most recent Bureau of Labor Statistics Employment Situation Summary was as controversial as any I have seen. It showed a solid improvement, including 2.5 million more jobs and an unemployment rate cut to 13.3%, although analysts had forecasted 19.8% and a 7 million loss instead. I didn’t believe it, calling these outcomes “severely in error,” citing the report as saying “BLS and the Census Bureau are investigating why this misclassification error continues to occur and are taking additional steps to address this issue,” and mentioning the possibility of deliberate distortion. On its release date and soon afterwards, many commentators, from economist Paul Krugman to former Vermont governor Howard Dean, voiced similar concerns.
Since then, though, a flurry of articles, all in the New York Times, have defended these results. Ben Casselman, in his June 8th “No, the Jobs Report Wasn’t Rigged. Here’s What Happened,” called it “all but impossible to manipulate the jobs numbers undetected,” and said that, classification problems notwithstanding, joblessness still dropped in May. The muddle, per Casselman, came from people describing themselves as ““employed but absent from work” – a category meant to reflect vacation, family leave or other temporary absences.” Clearly, many people invoked this label expecting to automatically return to their labor once the pandemic abated or ended; while most will be, all the talk about the coronavirus causing permanent workplace changes suggests a large number of others will not return. And the question remains: Just how bad was the American jobs situation in mid-May?
Two others were released not only in the same place but on the same date. “About Those Jobs Numbers…” from the Dealbook Newsletter, said “the job growth reflected mostly laid-off or furloughed workers returning to their jobs,” and that, per Neil Irwin of the Times, we are seeing an “epic collapse in demand.” Krugman asked “Will the Jobs Report Destroy Jobs?” by impeding further “emergency aid” in part responsible for recent economic improvements, and projected that “there will soon be many more layoffs unless aid comes soon.” “Don’t Cheer Too Soon. Keep an Eye on the Core Jobless Rate,” on June 15th by Jed Kolko, just added to the murkiness, but, echoing Krugman, said that May’s output “could give false hope, and lead to complacency about the labor market from public officials.”
Overall, what can we keep and reject from these updates? Per Casselman, a previous BLS chief said that group is “extremely reluctant to make any changes, in part because doing so would invite charges that the agency was massaging the numbers for political or other reasons.” Indeed we can’t have that, even if the data we did get was poor. Expect no corrections from me or anyone else. But we don’t, either, need to jump to the conclusion that the economic side of the crisis will soon be over. June’s data, which is now being collected, will tell us more, and we can go from there, as well as we can.