Levity aside, some of those green categories toward the bottom are probably headed toward a major supply side crunch due to the immigration crackdown
Workforce Economist in Residence at Guild; Senior Fellow at the Burning Glass Institute. I tweet a lot about labor markets, macro, and (sorry) music! Tweets represent my own views.
9,562 followers 2,171 following 3,056 posts
view profile on Bluesky
Levity aside, some of those green categories toward the bottom are probably headed toward a major supply side crunch due to the immigration crackdown
In: therapists Out: chiropractors
It's not good news, but I appreciate @eleanormueller.bsky.social at @semafor.com for chatting with me about August's disappointing jobs report. www.semafor.com/article/09/0...
Unfortunately and understandably not performing anymore. But glad he is still alive!
For those who are interested, here's my preview of the preliminary benchmark estimate we get from the BLS on Tuesday morning. substack.com/home/post/p-...
For those who are interested, here's my preview of the preliminary benchmark estimate we get from the BLS on Tuesday morning. substack.com/home/post/p-...
Sonny Rollins turns 95 today! Tough to pick a favorite track but this one always gets a chuckle out of me youtu.be/LZQWQk13lGA?...
This George Washington guy had some good ideas
Lots of interesting discourse about the recent decline in manufacturing employment But that discourse needs to incorporate the concurrent rise in manufacturing production… (Could be a measurement issue)
***Though increases in long-term unemployment are a concurrent indicator. It's the level that's lagging.
Since people are talking about the rise in long-term unemployment, a reminder that it's a lagging***, not leading indicator. In other words, it tells us what has happened, not what will happen. (It's the result of an extended period of weak hiring and is terrible.)
Just learned that we have these in California
I’m on a bunch of block lists, this is the funniest one
So far, you can still trust them. I don’t know how long that will last.
Post jobs day aural cleanse youtu.be/TjEQQbpjAlg?...
My kids left this under where I work and I kept activating it accidentally while analyzing the jobs report (not recommended)
Stable core, fraying fringes. My recap of August's bad jobs report: macromostly.substack.com/p/another-ba...
It's not a college grads unemployment crisis, it's a young people unemployment crisis. 1/ The headlines have been dominated by how hard it is for recent college grads to find a job. But it's even harder for recent high school grads, and getting rapidly worse. Blame weak hiring.
For those who like to watch things, my take on today’s bad jobs data with an appropriately grey sky backdrop
12/ Long-term unemployment continues to rise, even as short-term unemployment remains stable. Consistent with a low firing, low hiring environment. People who are unlucky enough to lose their jobs get stuck.
Actually, one more thing to add - that whatever the circumstances, it’s an honor to be mentioned in the same breath as Jed and Ernie
🤷♂️
11/ We're still a little lower than the Fed anticipated we would be in their most recent projections (from June), but catching up.
10/ Wage growth has been particularly weak for managerial/supervisory workers recently.
8/ The recent rise in unemployment for black women over the age of 20 is particularly stark, though it's going up for black men too. Less movement for Hispanic women (could be immigration crackdown related...) and white women.
7/ The unemployment rate for African Americans continues to rise - 7.5% in August, the highest in almost 4 years.
6/ Unemployment for people in their late teens fell in August (and is a little lower than a year ago - very noisy series), but that for people in their early 20s rose to its highest level since the spring of 2021.
5/ The most "recessiony" component of unemployment, "due to permanent layoff", was nearly flat in August. New entrants unemployment has been running a little higher in the last few months - consistent with the increase in unemployment among teenagers.
4/ Part time for economic reasons rose a little in August, but about a hair below where it was a year ago.
3/ Prime working age labor force participation rate rises to 83.7%, the highest in almost a year.
2/ On the plus side... prime working age employment population ratio, the flagship indicator of this report rises to 80.7%. No sign of deterioration in this metric. Steady since last fall.
BLS charts: 1/ Another bad jobs report. Weak job gains, and most concerningly the unemployment rate inches up further (now just a hair above 4.3%).
Thanks to @brianscheid.bsky.social at S&P Global Market Intelligence for including me in his JOLTS roundup yesterday! www.spglobal.com/market-intel...
I like that various private sector data provides are developing proxies for employment growth, but IMHO proxies for slack and turnover would be even more useful.
Unfortunately this video exceeds Blue Sky’s time limit, but if you want to watch my preview of tomorrow’s jobs report substack.com/@macromostly...
Since Challenger job cut announcements are generating headlines again, a quick reminder that while they are directionally OK, they massively exaggerate the amplitude of layoff fluctuations. Layoffs are growing slowly or not at all.
“Man, it's tough. It's tough to find a job... But that’s sort of been the picture for the past year. It has not fluctuated very much.” Thanks to @samfields.bsky.social at @marketplace.org for letting me give my two cents on yesterday's JOLTS report. www.marketplace.org/story/2025/0...
My preview of the August jobs report. In one sentence: pay less attention to NFP, more attention to the household survey ratios! Link: macromostly.substack.com/p/bls-jobs-r...
I appreciate Work Shift's @elyseashburn.bsky.social and @paulfain.bsky.social for giving us the space to pontificate, and also my amazing comms partner Hope Pitner.
"Regardless of what the cooling agent is, the question becomes: How do we help young people succeed in an environment where hiring is slowing across the board?" Thanks to @workshift.bsky.social for hosting this op-ed by @guildeducation.bsky.social CEO Bijal Shah & me. workshift.org/what-the-job...
We can look at exhaustion rates which have been very slowly creeping up each quarter - but, big picture, other slightly less timely indicators like the monthly jobs report are more useful to answer your question
6/ Next-week revisions to continuing claims running at about -9K
5/ The nowcast for August unemployment due to permanent layoff is flat or very slightly down.
3/ Initial claims just below my benchmark. Looks like our brief period of layoffs below year-ago levels is over.
gah thanks for pointing this out - I was so buried yesterday I totally forgot to update it
2/ They're just rising at a steady slow rate. No acceleration, no deceleration. Nothing to feel good about, nothing to make you worried.
Claims: 1/ Continuing claims a hair above my benchmark, at 1.94M. (Will get revised down to ~1.93M next week.) Due to residual seasonality there will probably be a small dip over the next few weeks, but flattish for the remainder of the year.
3/ Growth in searches for unemployment insurance (traditionally a nowcast of initial claims) continues to be far higher than that for actual initial claims.
2/ The benchmark for initial claims is 239K - we've been running below it since early June.
Claims: 1/ The benchmark for tomorrow's CC release is 1.930M, which if it materializes would be the lowest since late May. Last week's data will probably get revised down to ~1.945M. Trajectory of the published series will be misleadingly flattish ROY.
5/ Job openings (the most timely part of this report, though also IMHO the least valuable otherwise) have been flattish recently.
4/ Layoffs remain very low by historical standards.
3/ The quit rate is at 2.0% - around where it was in late 2015 or early 2016. As if the unemployment rate was slightly over 5%.
2/ The hiring rate, at 3.3%, is around where it was for much of 2011-13 (or if you like an average unemployment rate for that period, just over 8%). It's a really hard time to find a job.
July JOLTS: 1/ Some mild softening in hiring over the past 2 months. Too early to tell whether it is noise or renewed labor market cooling. Quits pretty steady.
fwiw, we see the exact opposite once a sector bottoms out and starts growing again - a surge in the employment of younger workers (though in construction there was a bit of a barbell with the oldest workers)
The "harder on youth" pattern we see in the @bharatchandar.bsky.social / @erikbryn.bsky.social / Chen AI paper is also quite apparent in construction employment after the housing crash. So much of the impact of downturns (economy-wide or narrower) happens via the hiring channel.
3/ Finally, I'm interested in whether the immigration crackdown intensifies talent shortages in the accommodation & food services sector.
2/ Quasi-JOLTS metrics from July (based on CPS data) looked pretty good, but those are more stale - they mostly come (as do the monthly jobs numbers) from the 1st half of the month. JOLTS turnover data cover the entire month, ergo are "fresher". bsky.app/profile/econ...
We get the July JOLTS data tomorrow morning (10 AM ET). 1/ Top of mind - we've had 12 months of stability in hiring, about 9 months of stability in quits. Both are great barometers of the tug of war between weak labor demand growth and weak labor supply growth.
Watching Back to the Future w/my kids. It’s just as great as you remember it
I realize sharing music from “The Boss” is a bit of a faux pas on Labor Day
I’m not a fan of the Backstreet Boys but can’t begrudge Bruce Springsteen paying tribute to another group of grizzled music veterans who inspired him early in his career youtu.be/USQ697oqkaw?...
I don’t think it’s just a per capita thing
I prefer 1966-67. But 1968-69 is my 2nd favorite
“I asked an American why the US produced and hosted so many legendary physicists / mathematicians” is actually something we should be asking more often
Lots of great Kinks songs but I’m pretty sure Ray Davies never topped this one youtu.be/qG3JyuAUE2E?...
The Death and Taxes black lager is delicious - visually and to some degree flavor-wise reminiscent of Guinness except it’s actually good (Yes, I know Guinness is an ale)
It’s an odd mix of views: brag about foreign investment but express dismay about the net international investment position
They do reduce taxes from other revenues but on net tax revenues go up as long as the tariffs aren’t on the other side of the Laffer Curve
To the degree that before today you believed tariffs would increase (decrease) inflation, the court ruling on IEEPA tariffs (if it stands) will decrease (increase) inflation
100% agreed with everything you wrote! But I do think a material minority of post-January tariffs will remain in place even if SCOTUS supports the appeals court ruling.
"you're killing your father, Larry" - Walter Sobchak
Jokes aside (and SCOTUS ruling dependent), this ruling only affects the IEEPA tariffs. Other tariffs remain in place and to some extent the admin will be able to use other tools to replace the IEEPA tariffs
5/ As I said, I think it's *reasonable but speculative* to assume that the May-July 2025 data will be revised down eventually, whether in early 2026 or in early 2027. But it will be out of scope for the preliminary benchmark estimate that I & Waller are eagerly anticipating.
4/ In early 2026 (not a few weeks), we'll get the annual revision, which will incorporate both the benchmark revision and also non-benchmark revisions to the post-benchmark-date data.
3/ Assuming "what we will learn in a couple of weeks" refers to the preliminary benchmark estimate published on September 9th... that estimate will only cover data from March 2024 to March 2025. It will not cover the May, June or July data.
2/ Waller discusses the possibility that the low May-July NFP gains will *eventually* be revised down to a small negative number, which I think is a reasonable (albeit speculative) argument. However, I think he is also making a stronger argument that I think is wrong...
PSA The preliminary benchmark estimate in a few weeks will not be applied to May-July 2025 data! More: 1/ A pedantic point about Governor (and hopefully, eventual Fed Chair) Waller's revision commentary yesterday that I missed until someone pointed out to me...
If the Fed loses its independence, it will be a great opportunity for the resurgence of the Shadow Open Market Committee: shadowfed.org
This excerpt is interesting, and it’s a signal, I’m just not sure how much of one it is without seeing the data. ADP, my DMs are open! bsky.app/profile/econ...
I think they’re valuable. Just not the only signal and *mostly* a placeholder until the CPS data fills the gap Claims are funny, they agree w/Waller that there is an ongoing mild deterioration in the labor market but not that there’s any recent change
If the fine folks at ADP want to share their weekly data with me, I too will write about it! Waller will be much busier after his appointment as chair and I’ll be glad to step in to plug the gap (Screenshot courtesy of @jonathanjlevin.bsky.social )
4/ If payrolls grow faster for the remainder of the year than they did over the last 3 months, but the unemployment rate rises, I'll be worried if payrolls continue to grow at a piddling rate but the UR stays around 4.2%, I won't be worried
3/ The lower break-even rate of NFP is a bit of a theological argument - we know it's true but don't know exactly where it is. And the only hard proof we have of its location is what is happening to those HH & JOLTS survey data I mentioned
2/ The most important "if" is what happens to measures of slack, especially the ratios in the HH survey (and to a lesser degree the key turnover rates in the JOLTS report, though I worry about revisions with these) More important than what has happened / will happen to NFP
1/ As always I enjoyed reading Governor Waller's speech. He'd be a top-notch choice for the next Fed Chair. That said... I think the argument here has a lot of implicit "ifs"
4/ One thing I'd hypothesized 2 weeks ago is that improvement on recent headcount actions was negatively correlated to firm size (which would mean the aggregate employment improvement would be smaller than the diffusion index suggests). Hypothesis/correlation weaker this time.
3/ Same with future plans for headcount. They're more expansionary than a year ago mostly because fewer firms are planning to cut headcount.
2/ The improvement is coming from fewer employers cutting headcount. We're not seeing an increase in the share of employers expanding headcount, it's about the same as it was a year ago.
Business headcount plans: 1/ The labor market news from the Census Bureau's Business Trends & Outlook Survey remains relatively good. Employers' recent headcount actions are slightly more net-expansionary than they were a year ago.
6/ The nowcast for "unemployment due to permanent layoff" is for little change in August from July.
5/ A recap of next-week revisions for continuing claims. In all likelihood this week's 1.954M will change to 1.945M next week.
4/ The period of lower layoffs was brief, but no signs of layoffs increasing either.
3/ Initial claims were marginally below 2023-24 levels, as well as the benchmark. I expect (low confidence) that they'll drift down for the remainder of the year, barring any large natural disasters like the hurricanes we saw last year.
2/ The ongoing picture of continuing claims is very very slow deterioration. Fluctuations since April have been tiny. I don't like it, but I'm also not particularly worried.
