Non-farm payrolls fell by 20.5 million last month, and the unemployment rate is now 14.7%.
This is obviously the steepest decline in labor market conditions ever, and I can almost guarantee that it's the worst that you'll ever see in your lifetime.
This is obviously the steepest decline in labor market conditions ever, and I can almost guarantee that it's the worst that you'll ever see in your lifetime.
It's terrible. Devastating. Awful.
It's also no surprise. The virus made it unsafe to leave the house. As a result millions can't work.
The alternative, of heading in to a virus-stricken workplace is even worse.
It's also no surprise. The virus made it unsafe to leave the house. As a result millions can't work.
The alternative, of heading in to a virus-stricken workplace is even worse.
Today's numbers also largely confirm what economists had pieced together from a range of other indicators. As devastating as today's numbers are, we suspected much of this already.
This is an important nugget of good news. If most of these job losses really are temporary layoffs—and if we can maintain and later restore the connective tissue between workers & their jobs—then the economy could bounce back much more strongly than usual
There's a robust—almost surprising—consistency between the payroll survey which saw job losses of 20.5 million, and the household survey which saw employment falling 22.4 million.
Don’t be fooled by average wages rising nearly 5% in a month. That’s a sign that this recession has hollowed out the ranks of low-paid workers, which mechanically leads the average pay among those of those left behind to be higher. Believe me, no-one’s getting a pay rise. 
Betsey makes an important point: The BLS realizes that there's a (largely technical) problem with its numbers due to interviewers mis-coding responses, leading it to understate the unemployment rate by about 5 percentage points.
Follow this logic, and the misclassification-adjusted unemployment rate is really 20%.
[Note: This isn't the usual conspiracy theory wing-nuttery, nor is it about discouraged workers. This is literally a technical mis-classification that is obvious in the data.]
[Note: This isn't the usual conspiracy theory wing-nuttery, nor is it about discouraged workers. This is literally a technical mis-classification that is obvious in the data.]
Here's the problem: Millions of people suddenly report being absent from work for "other reasons." In reality, it's extremely unusual. According to the definition of unemployment, corona-related absences are temporary layoffs that should be counted as (temporary) unemployed.
Kudos to the BLS for treating this well.
It didn't "fix" the problem, because it doesn't want to start changing people's responses, or change the rules for measuring unemployment. (Imagine the shenanigans that could follow.)
So it reports the numbers, along with the problems.
It didn't "fix" the problem, because it doesn't want to start changing people's responses, or change the rules for measuring unemployment. (Imagine the shenanigans that could follow.)
So it reports the numbers, along with the problems.
Who's getting hurt in this downturn?
It's those on low wages. Here's data not from the BLS, but from a recent paper by Erik Hurst & colleagues.
bfi.uchicago.edu/wp-content/upl…
It's those on low wages. Here's data not from the BLS, but from a recent paper by Erik Hurst & colleagues.
bfi.uchicago.edu/wp-content/upl…
This confirms what I conjectured earlier: The rise in average wages reflect the hollowing out of low-wage jobs during the pandemic.
Hurst & coauthors estimate a "selection-adjusted" wage measure (using ADP data) that adjusts for this.
It confirms: No-one is getting a wage rise.
Hurst & coauthors estimate a "selection-adjusted" wage measure (using ADP data) that adjusts for this.
It confirms: No-one is getting a wage rise.
Erik Hurst—who is as sharp of an economist as anyone I know—emails me some more context. He and a big team of data hounds have been crunching the ADP data, and they believe that today's payrolls number will be revised to be even more negative. Much more negative. Like -27 million
An aside: So much is broken right now, that none of our economic statistics are as reliable as they usually are. Surveys aren't being completed, questionnaires don't ask the right questions, and our data aren't designed to measure or describe a pandemic economy.
Here's the BLS's best attempt at explaining some of the myriad problems/difficulties/changes/possible misinterpretations that might arise just from this jobs report: bls.gov/cps/employment…
Nerdy tweet: One very smart adjustment that the BLS made was to switch from multiplicative seasonal adjustment to additive. This is important, and I hope the @USDOL follows suit so that the seasonal adjustments applied to the initial claims numbers start to make more sense.
Bottom line: The economy is in a massive hole, caused by the reality that being close to others could kill.
Thes only pathway to a robust recovery is to fix the public health crisis. We have no economic tools as powerful as our public health tools.
Thes only pathway to a robust recovery is to fix the public health crisis. We have no economic tools as powerful as our public health tools.
Fix public health and you'll fix the economy.
