Next paper at #BPEA, by Stephanie Aaronson, @marydalyecon, William Wascher and David Wilcox looks at the cyclicality of labor market outcomes across demographic groups. brookings.edu/bpea-articles/…
I was a discussant, so thought I might show a few slides...
The differences in the cyclical sensitivity of unemployment by race are just striking. A deep recession that might push the white unemployment rate up to 10% can cause the black unemployment rate to rise as high as 20%. Business cycles have very different implications by race.
Interestingly enough, there aren’t big differences by gender. It used to be the case that women’s unemployment was more responsive to the business cycle, but in the great recession, it was men’s unemployment that rose then fell most dramatically.
There’s some pretty stark differences by education. The unemployment rate of high school droupouts varies dramatically over the business cycle. For college grads, the unemployment rate never rose above 5%, even in the depths of the great recession.
The lack of cyclical variation in unemployment among college grads is so striking that I broke it apart. Note the scale: There’s not much variation. And for those with professional degrees or PhD’s, it’s rare for unemployment to rise above 2-1/2%, even in the worst of times.
A strong economy helps less educated folks in every country I’ve ever looked at. This plot shows the gap between the unemployment rate of high school dropouts and college grads, versus the state of the business cycle (u-u*). A strong economy narrows these gaps everywhere.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Payrolls grew a relatively uninteresting (and positive!) +177k in April, and unemployment was unchanged at 4.2%.
This economy is still humming along.
NOTE: This is a reading largely from the pre-tariff period. Still very foggy about what lies ahead.
Revisions were somewhat worrying: March was revised down -43k to +185k. Feb down -15k to +102k.
Three month average payrolls growth -- a useful indicator of the underlying pace of job growth -- is a healthy +155k. That's a pretty great place to be at this point in the cycle.
Nominal wage growth was 0.2% this month, and are up 3.8% over the year. That's probably enough to keep inflation above the Fed's target (and that's before factoring in the effect of tariffs).
Ugh. It's happening. The economy shrank in the first quarter, at an annual rate of -0.3%.
The good news: Consumption and investment remained strong. Think of this as a hard-to-interpret report due to -- **all of this**. Remember, this is the average of Q1, and the real concern is about Q2.
Look into the details, and the GDP report really isn't that bad. (We already know from the jobs data that the economy did okay in Q1.)
@jasonfurman suggested focusing on Real final sales to private domestic purchasers (basically C+I, the reliable parts of GDP) which grew +3.0%
The sharp rise in investment appears to be almost all due to pre-tariff front-running. Investment contributed 3.6%-pts to Q1 GDP growth.
Of that, inventory accumulation was 2.2%-pts.
And an additional 1.1% came from equipment investment (which is what the China tariffs hit).
1. Tariffmageddon isn't over: Lotsa tariffs to account for, but the average tariff rate is only down around one quarter.
2. He's not going to get big wins: Tariffs were low before this mess, and if Trump negotiates competently, they'll be low again. Basically no gain.
You've seen this movie before: It was NAFTA which got relabeled by Trump in 2020, but really barely changed.
3. The rationale for this policy keeps changing. Remember when it was all about bringing manufacturing home? (That was yesterday.) Now it's negotiating deals. Those are fundamentally in tension.
(I'm only going to build a factory in the US if tariffs are likely to persist.)
One thing I've learned to do when I have questions about social security number holders who are age 100 or older is to look up the SSA Inspector General audit report, "Numberholders Age 100 or Older Who Did Not Have Death Information on the Numident."
After all: Is there a principled difference between weighting on age (to ensure that your sample includes youngs and olds) and weighting on past vote (to ensure you get folks from across the political spectrum)?
Both age and past vote are:
- Predetermined (before this poll)
- Non-manipulable
- Though self-reported
- And we have good population estimates to weight them to.
What principle would make one of these a legitimate survey design weight and the other "herding"?