What the Hell Happened With Today’s Shocking Jobs Report?

A server wears a mask and gloves while working, as Hoopers Crab House opens for in-person dining, amid the coronavirus pandemic, on May 30, 2020 in Ocean City, Maryland. - On the evening of May 29, 2020 restaurants and bars in the state of Maryland were opened for outdoor and open air dining. (Photo by Alex Edelman / AFP) (Photo by ALEX EDELMAN/AFP via Getty Images)
The new normal. Alex Edelman/Getty Images

Most economists and investors had expected that Friday’s jobs report would be a train wreck. Throughout late April and early May, historic numbers of Americans filed for unemployment benefits each week, despite states’ attempts to reopen their economies. There were predictions that the unemployment rate would hit 20 percent, on par with the deepest depths of the Great Depression.

Instead, the government reported this morning that the U.S. added 2.5 million jobs and the official unemployment rate fell to 13.3 percent last month. Construction payrolls climbed by 464,000. Employment at food services and drinking places (aka restaurants and bars) rose by 1.4 million. The results left many journalists, such as myself, to tear up their prewritten stories and analysts scrambling for explanations. (“Real-time reax: WHAT THE WHAT???” tweeted the University of Michigan’s Justin Wolfers.) Some pointed to possible measurement errors, or murmured darkly about the possibility that books had been cooked. New York Times columnist Paul Krugman chose to do both.

Meanwhile, Donald Trump took the morning to gloat, holding a press conference where he said that “we’re opening with a bang” and that “we’ll go back to having the greatest economy in the world.” The president proceeded to claim vindication for his entire jobs and public health record, noting that only 100,000 or so Americans had died, rather than the millions who would have perished if he’d done nothing. “We made every decision correctly,” he said.

So, what happened? Are these numbers too good to be true? Not necessarily. There may be some flaws in the data, which the Bureau of Labor Statistics has been transparent about. But they’re consistent with real-time survey results from last month that suggested the jobless rate had plateaued. If you look closely enough, they also subtly tell a somewhat similar story to the weekly unemployment insurance numbers that had led many to expect the worst, though you have to look carefully at those numbers to see how.

But the most important thing to keep in mind of all is that, no matter how you interpret Friday’s new numbers, a crushing number of Americans are still unemployed, and in the end we can only guess when they’ll get their jobs back. There are also plenty of reasons to think that without more aid from the government, we’re stuck in a slow slog back to normalcy.

Now about the data. To start, it’s possible that this month’s figures are just less accurate than usual. Thanks to the plague we’ve been dealing with, the response rate to the government’s regular household survey—which it uses to calculate the unemployment rate—was 15 percentage points lower than usual, while the establishment survey, used to track changes in payrolls, also fell slightly. With fewer people and companies responding, there was greater room for error.

Second, the headline unemployment rate that has been widely reported—that 13.3 number—is almost certainly off by a few percentage points. Just like last month, a large number of Americans who were likely furloughed or had their hours cut to zero because of the outbreak were categorized as employed but not at work for “other reasons.” The Bureau of Labor Statistics says that if you include them, the jobless rate would be about 3 percentage points higher, or 16.3 percent, down from 19.5 percent last month.

Third, some analysts did actually predict this. Surveys by Civis Analytics, published by the New York Times Upshot, had found that the share of Americans who had lost their job since March fell below 12 percent by mid-May, pretty much right in line with Friday’s report.

And finally, the unemployment insurance rolls also suggest that a lot of Americans started going back to work by the middle of last month, when the jobs data was gathered. According to the Department of Labor, the total number of individuals claiming jobless benefits fell by about 1 million the week of May 16, to under 30 million. A lot of people were losing their jobs. But on net, all signs suggest that more were being rehired. In real life terms, it seems like a lot of people went back to work in outdoors industries like construction, and after truly gargantuan layoffs in March and April, restaurants and retail either revved back up a bit in states that officially reopened, or at least started doing some takeout.

It’s also crucial to keep all of this in broad perspective. If you take this new report at face value, there were still about 26 million fewer Americans with jobs in May than in February (21.5 million fewer officially employed, another 4.7 million miscategorized as working). The question is how many of them will be able to get their old positions back, and if not, what we need to do in order to support them in their time of need.

On the first question, there’s both good news and bad news.

The good: Most of the job losses we’ve seen during this crisis still aren’t permanent yet. According to the Bureau of Labor Statistics, about 8 out of every 10 workers who say they’ve lost their job are only on a temporary layoff, meaning they could be called back soon, assuming business picks up at their old employer.

Now the bad: As noted earlier, a lot of businesses are still laying people off, which suggests some might be closing down for good or running out of cash, even as states are beginning to reopen. Last week, 1.6 million Americans (not seasonally adjusted) filed new applications for unemployment insurance, and continuing claims (meaning the number of people processed and potentially receiving benefits) actually edged back up, meaning it’s possible progress stalled out, at least temporarily. Meanwhile, the Census Small Business Pulse survey found that more firms were still letting employees go than rehiring at the end of May, and a greater share were reporting revenue declines than increases.

What this tells us is that the economy is still in a delicate, uncertain place, and we are not necessarily on the verge of a v-shaped bounce back. Yet some conservatives are already trumpeting Friday’s jobs numbers as an excuse to declare victory and urge cutbacks on aid to businesses and the unemployed.

This is an absurd position. Tens of millions of Americans are still out of work, and there’s no guarantee when or if they will get their jobs back. Some businesses have rehired workers with money from the Paycheck Protection Program, and it’s not clear what will happen when that money runs out if, say, sales are still slow. This week’s protests against police brutality may have also set the reopening process back a bit, given the number of stores that have been forced to board up and put further pressure on business’s finances. Meanwhile, simply letting aid like enhanced unemployment insurance completely expire would not only be cruel to those who are still jobless but could also slow down the pace of recovery by cratering consumer spending. The economy isn’t as bad as it could be. But it’s far, far too early to start declaring mission accomplished.

For more of Slate’s news coverage, subscribe to What Next on Apple Podcasts or listen below.