The mystery of the vanishing jobs

If bonds are back today, why wait for tomorrow?

At PGM, our fixed income strategies help investors uncover hidden value and unlock opportunities.

Whether you're looking to enhance your income or diversify your portfolio, our broad range of strategies bring together local expertise and deep credit research to help you achieve your long-term goals.

PGM, our investments shape tomorrow, today.

Pushkin.

The unemployment rate in the United States is 4.3%.

And that is a pretty good number.

It's lower than the historical average, and it's not a million miles.

from what economists call full employment.

Unfortunately, there is another signal coming from the labor market, which we don't like.

The economy is creating fewer jobs every month.

Today on the show, what is the message from the labor market?

And in particular, are we in trouble right now?

This is Unhedged, the markets and finance podcast from the Financial Times and Pushkin.

I am Rob Armstrong, and I'm coming to you from a crisp, glittering, and autumnal New York City, where I am joined by the Unhedged Newsletters expert on the employment market in America, Aiden Reiter.

Aiden, welcome to the show.

Thanks for having me.

So we have sort of two separate issues to talk about.

One is that the monthly job numbers are sluggish.

Kind of last three, four months, we've seen additions of in the low tens of thousands.

And then we got this whopping annual revision to the last year's worth of numbers.

So there's a lot of kind of tricky stuff going on here.

Let's start with the monthly numbers.

What are we learning there?

Yeah, so we've gotten some less than ideal monthly job reports in the last two months.

So in July, things came way under expectations.

The market had a proper freak out.

Yes.

Less than ideal is a powerful phrase, Aiden.

This is like the kind of phrase my doctor uses when he sees my blood work.

It's like your cholesterol is less than ideal.

Well, I think it's intentionally vague here as it is there.

Okay.

Because

maybe, like your doctor, we have no idea what's going on.

Okay, good.

But anyway, the numbers are not what we want to see.

They're not what we want to see.

And especially this past month in August, we got a very low report.

That was 22,000.

And on top of those, every time you get a monthly job report, you also get revisions for the two months prior.

Right.

And those revisions have actually been what really moved the market.

Yes.

So now, with the revisions, so that's the most recent estimates we have, the average has been 29,000 jobs added over the last three months, which is astonishingly low.

A year ago, it was like 100,000 a month or something like that.

150, 160, within the 200s.

It's a big slowdown in jobs added.

It's a big slowdown.

And actually, this revision showed that in the past three months, at one point, we actually had negative job growth.

That means the job market shrink.

Actually, shrank.

Yes.

And that was post-revision, though.

It didn't look that way when it first came out.

When it first came out.

And then tell us about this big annual revision.

What is that?

I find that kind of confusing.

Yeah, so it's important to remember in this broader context, right?

So Trump has attacked the Bureau of Labor Statistics, said these numbers are faked,

these revisions are too big, and sparked a lot of concern.

Yeah.

The reality is it is super duper hard to count how many jobs are added in the U.S.

economy.

It is a big economy.

It's a big country, 300 million people or whatever in our country.

And you're trying to figure out who has jobs and who doesn't.

350.

And actually, we don't even know how many people, really.

That's a whole other conversation we'll get into.

But the point is, it's very hard to do.

The way they do this is with really clever statistics, right?

And clever modeling.

You have two big surveys.

There's a current population survey, which is how you find the unemployment rate.

And that is 60,000 households.

Households you call.

Call people.

You're calling people.

Then there's the current employment survey, which is, you know, you call offices, companies, some companies required to report, et cetera.

What you need to do and able to do that is chase down these numbers, right?

You get people to answer the phone, give you the numbers.

And on top of that, you have to cleverly impute the areas where you're missing people answering the phone.

Yes.

Both those things require a ton of resources.

Right.

Unfortunately, with our era of smartphones and spam calls, et cetera, the response rate has been just rapidly declining over the last five years, especially, and even 20 years before.

It's hard to get people on the phone.

Now, is that true for the household survey as well as the business survey?

Both of them are falling.

Both the response rates are falling.

I think you could get the businesses on the phone at least.

It's hard.

It's hard.

I know everybody has a thousand spam calls a day.

Yes.

You really want them to respond.

Everybody has a thousand emails.

Like, it's hard to get people on the phone.

Is what you're saying here that what we're looking at in this slowdown in job creation,

might we be looking at a data error here?

That's not necessarily what I'm saying.

What I'm saying is there has been this attack on the data bureaus of the United States.

In reality, they actually do face real issues, right?

They actually have seen their budgets decrease over time as this issue has gone up.

But at the same time, the numbers they put out are still incredibly, incredibly reliable.

The difference is, the first time you get that number, it's not that timely.

It's always an estimate.

There's always going to be revisions going back.

And those revisions are reliable.

The reality is, though, as you get worse and worse data up front, few people answering the phone, less resources to track them down, those revisions have just gotten larger over time.

And that's what's been frustrating people.

So the final numbers are still pretty reliable.

Everything's very reliable.

But the revisions are growing.

The revisions

are first cut of the data isn't that good, but we get there in the end.

I guess that's kind of the picture.

I mean, one could argue we're not getting there in the end because this is a very difficult labor market.

But for all intents and purposes, as you look at us compared to our peer economies, our data is still very, very good.

So if that is true, Aiden,

and here I'm looking at a chart going back a couple of years of jobs being added, and the chart is going kind of down and to the right.

Can we now feel pretty confident that the U.S.

jobs market is getting worse?

Yes and no.

So let's start with the revision you asked about before.

The big one.

The big one.

We just got the annual revision.

The annual revision is at the end of every year, you get states' unemployment insurance records, and those are much more reliable because people who are unemployed, they want the insurance, they go,

et cetera.

You may not pick up the phone, but you're damn sure going to try to get your unemployment check.

Absolutely.

So we get them later on than we get the original jobs data.

Every August, there was a revision for the job numbers from the previous March to the March before, right?

So we just got the revisions from March 2024 to March 2025.

Yes.

And it was the biggest preliminary revision of all time.

It was a 911,000 jobs.

Fewer were added than we originally thought.

Interestingly, the market didn't really freak out on this.

It was interesting.

I thought when I first saw that 900,000 plus number, I was like, oh boy.

here we go.

And then nothing kind of happened.

It was strange.

Yeah, and I think there's three reasons for that.

The first is last year, we also had a really big revision.

It was 818,000 jobs.

That's huge.

The reason it happened is this issue with how the modeling is done.

We don't have to get too far into the details.

It seems to be the same issue that happened this year again.

People just knew that was going to be a big issue this year.

There was going to be a big revision.

Got it.

Those numbers have been awry since COVID for a couple of reasons.

The second is the labor data is already looking bad, right?

As we just said, the job market's slowing already.

This doesn't change the picture that much.

The third is every year this is just a preliminary revision.

It tends to be revised up.

It goes down and then it comes back up again.

Exactly.

So people expect it to come up.

And then the final piece is those numbers were March 2024 to March 2025.

That's the Biden administration and then the pre-tariff Trump administration.

So it's just a very different world.

It's very hard to extrapolate between the two periods.

Yes, dinosaurs walked the earth

back then.

So we have a picture now where a weakening job market, meaning less job creation

specifically, is coming into clearer and clearer focus over time.

Absolutely.

Is that evidence that the labor market is faltering, or are there other things we need to keep in mind here?

That's where it gets really tricky.

So on its surface, that definitely shows a labor market faltering, unemployment rate ticking up, job growth ticking down.

And that generally means the economy is poor.

Yes, and that is why people have moved so adamantly for the Fed to cut rates soon.

But there's a few things that are not totally normal about the current moment we're in.

The first we should talk about is the supply of labor.

Workers.

So we've had this huge crackdown in immigration.

It is very hard to tell how many immigrants are in the United States and working in the United States, partially because they don't answer the phone as much, theoretically, right?

They probably answer the phone even less than normal people because they're worried about who they're talking to at the other end of the line in this political environment.

Exactly.

That's conjecture.

We don't really know.

But the point is, like, we...

It stands to reason.

It stands to reason.

So, and, you know, there has been a a very remarkable measured slowdown of illegal crossings at the southern border, and then a lot of anecdotal evidence that some people have left either because they were deported or they just don't want to deal with this political environment.

So we assume that the labor force has shrank a little bit.

That means the break-even number.

So the amount of jobs you need each month to like feed a healthy economy is probably way lower than it used to be.

Yes.

So that makes it hard to make sense of, are these numbers good or are these numbers bad?

So is it too much of a simplification to say it's a smaller country now than it was population-wise, than it was a couple of months ago?

And it stands to reason that a smaller country would create fewer jobs

for any given rate of economic growth.

So the kind of underlying economic growth rate might be the same and the number of jobs created might be less.

Exactly.

Right.

That is totally the right way to think of this.

We just don't know how many new jobs is that new break-even rate.

Yes.

Because it's really hard to tell how many people are in the U.S.

economy and how many of those people are working and looking for jobs and looking for legal or illegal jobs, et cetera, et cetera.

Okay, you know a lot about this.

I'm going to force you to guess.

Give me what you think is the break-even job creation number, monthly job creation number, and your margin for error.

around that number.

I'm putting you on the spot.

No one wants to answer that question, which is why it is fun for me to ask you.

it's it is a really hard question.

I'll start by saying I spoke with somebody who studies this very extensively, Wendy Edelberg at the Brookings Institution.

And she said that the U.S.

market investors might have to get used to a world where zero job growth is a good number.

I'm not saying that yet.

Yes.

I'll say probably somewhere in the like 20 to 50 range,

right?

Somewhere.

I'll say like 35 with a 15,000 give or take.

Yes.

Yeah.

Yeah.

And so we're only a little bit shy of that level.

Yeah.

And I know I said that fully on vibes, but

we do trade in vibes here at Unhedged.

I think that is fine.

Yeah.

So it's really hard to tell how many jobs that you need to add every month because these numbers, not only the BLS numbers we talked about before, but the population immigration numbers are just really hard to understand.

Trevor Burrus: So what's your gut feeling?

I have gut feelings about this too.

Actually, let's call them semi-gut feelings because they are informed by some data and some study of the economy.

But what is your feeling?

Is

the jobs picture telling us that the economy itself is slowing, that activity levels are slowing in the country?

Again, the number of jobs added each month falling definitely suggests that things are worse than they were, right?

That's undisputed.

But I think there's other questions we need to ask, such as how many new jobs are being created that people just haven't filled yet, right?

So is there new labor demand on top of this?

Yes.

And we were getting mixed signals there.

We recently got the Jolts report, which showed for the first time since COVID, there are more people looking for jobs than there are new jobs available.

And for our less nerdy listeners at home, the Jolts survey is the job openings and labor turnover survey.

Yeah, that's a really bad sign.

But if you look at the labor demand, I would say it's really bad.

Like a ratio of about one for that number is historically not that bad, but it's worse than it has been in recent months.

Yes, it's worse than it has been in years.

Trending the wrong direction.

But you know, other indicators of job demand and job growth are like around sideways.

They're not too bad.

So it's really hard to tell what's at fault here.

Yeah, I want to throw out some things that make me feel good about the picture.

Corporate revenues and corporate profits look okay.

They're still growing.

It's hard to screech into a major unemployment event.

a recession or whatever you want to call it when profits and sales in corporate america are pretty good.

Growth of those things are pretty good, and they are pretty good.

Companies are investing.

CapEx is pretty good.

And

the balance sheets of corporations are pretty good.

They're not super over-leveraged.

So they're not, at the first sign of trouble, they won't have to fire people.

Same thing with households.

Yeah, same thing.

And the households are pretty good too.

So that will keep household demand up.

So it just feels like

we don't have the major components of a recession out there,

which would be profits and sales of companies falling year over year.

You'd see CapEx falling through the floor.

It's sort of doing the opposite, actually.

Yeah, definitely spurred up by chips and data centers.

Yeah, data centers, but there's other kinds of CapEx out there.

So it's like

the things you would be looking for in an economy that was really going through the floor, I just don't see them out there.

So that makes me feel like

the jobs numbers maybe are worse than the underlying economy in some way.

As weird a thing is that to say.

Like you might ask, what is the point of an economy if not to employ people?

But like, I'm just not seeing the really ominous, terrible stuff out there.

Yeah, I think that's very fair.

It is

also worth looking at the inflation numbers.

So we just got CPI this morning.

It came in hot, right?

2.9, 3.1 core.

That's pretty hot.

Yes.

If that is not all from tariffs and it doesn't seem like it all seems to be a tariff passsover, that's because people are buying goods

and spending on money.

That suggests a hot economy.

Yeah, not a cold one.

Exactly.

That did come in around expectations, so it's not like we overshot in any meaningful way, but it's just a different situation, right?

We're not necessarily in obviously a slowdown.

You just have this really murky labor market that the Fed and the market has to sort through.

I was playing devil's advocate on one side, on the good side.

Now I'm going to play it on the bad side.

The thing that I worry about with the economy that would kind of stitch some of these divergent points we've made together is that

it's a very uneven economy.

Exactly.

So we know that households...

close to the bottom of the income and wealth scales are performing very differently than households towards the top and that the top is contributing a lot of the demand in this economy.

And I was spooked earlier this week when a subprime auto lender

went bust, this thing called tricolor, tricolor, I don't know how to pronounce it.

Three colors.

Three colors.

And, you know, it loans money to people with iffy credit to buy cars.

And it went bust.

And so I wonder if part of what we are seeing in a jobs market that doesn't look that great, but an overall economy that looks solid solid is a very uneven economy, one that is thriving at the high end and struggling at the low end.

I think you could argue that that's been the case for the past five years.

I mean, we've talked a lot about the K-shaped economy for a long time.

And if that is the case now, so it's more proof that the U.S.

economy can only be resilient because its wealthiest consumers do the lion's share of consumption.

And that's not necessarily a good economy, a healthy economy, because when those people start not spending, and that could happen at any point, then you're really, really in trouble.

Aaron Powell,

in the last few minutes, Aiden, we've sketched a rather complicated picture.

There are serious question marks around the jobs market, but it does appear to be weakening.

At the same time, the overall economy, or the aggregate economy, if you will, looks okay on many of the measures we use.

But we're worried that it might be very uneven, that the distribution might be putting us in a dangerous position.

And finally, inflation is kind of warm.

This demands that we ask, what is the Fed to do here?

If you were the chair of the Fed, Aiden, a job for which I think you would be eminently qualified.

So kind.

What would you do here?

What do they need to do given this complex data?

I mean, it's really, really hard to tell, right?

The Fed is really loath to kick off inflation again after fighting it down for years and still not falling.

And not getting all the way to their 2% target.

We're banging around 3%, 3% plus percent here.

Yeah, since the last Fed statement, which came out at Jackson Hole, it seems like they're more okay with higher inflation going forward.

Yes.

So they might be able to get

a lot of money at the margin, not in the high threes.

So it's really, really hard to say.

But

the picture has not changed, really, since Chair Powell said at Jackson Hole that signs are pointing towards a cut, really.

Yes.

He didn't say it outright, but he heavily suggested.

Right.

So we're going to get a cut.

I think we're all pretty confident we're going to get.

The market is still very confident.

Even after today's hot CPI, it's pricing in

a higher than 25 basis point cut.

So there's some little chance of a 50 in there.

Yeah, there's some people still betting on the 50.

Right.

But I think my guess is we get 25,

but a dovish 25, where they signal that it is not the end of the cutting or anything.

Which is not what the market thinks right now.

The market right now expects there to be cuts of 25 in each of the successive meetings this year.

So October and December as well as September.

Listeners, we'll be right back with long and short.

Speaking of alternatives, PGIM's monthly podcast discussing trends and strategies in alternative investing.

The history and evolution of the Australian superannuation funds.

It's interesting in so many ways.

They actually mandate savings into retirement plans, and they've become some of the biggest global investors in private markets now.

And their members have seen tremendous growth of their portfolios.

Hear the full conversation on PJM's podcast, Speaking of Alternatives.

Listeners, welcome back.

This is Long and Short, the portion of the show where we go long things we like and short things we don't like.

Aiden, are you long or short something?

It's not that I'm long

Larry Ellison being the richest man in the world.

I'm just long Elon Musk no longer being the

So that means you're kind of against greed, but it's like envy means something like, no, I don't, I shouldn't.

I don't have a strong feeling of any way towards Larry Ellison.

I just think it is kind of funny.

Yeah.

Yeah, I don't know how the vibes are over at Musk World Headquarters.

Changing

tack quite seriously, I am long tweed.

Today was the first day I wore a tweed jacket this fall.

Tweed is my favorite material, and it was great to put it on.

Although I'll tell you this, Aiden, I caught a glimpse of myself in a shop window walking to work and I said, who is that old high school English teacher I see there?

But, you know, maybe that's the look I'm going for.

In another life, that really could have happened.

Listeners, we'll be back in your feed next week.

Until then, stay sharp out there.

Unhedged is produced by Trina Menino and edited by Bryant Erstadt.

Our executive producer is Jacob Goldstein.

Topher Forge is the FT's acting co-head of audio.

Special thanks to Laura Clark, Alistair Mackey, Greta Cohn, and Natalie Sadler.

FT Premium subscribers can get the Unhedged newsletter for free.

A 30-day free trial is available to everyone else.

Just go to FT.com/slash unhedged offer.

I'm Rob Armstrong.

Thanks for listening.