Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Marko Bjegovic Profile picture
@MBjegovic
Feb 3, 2023 • 24 tweets • 13 min read • Read on X
Jan NFP came in 200K+ above the highest Wall Street estimate, UR came in lower than expected, while AHE came mostly in line.

How come Wall Street analysts were all so off?

What does that say about the labor mkt?

Let's dig deeper.

A thread.

1/24

#Fed

Before analyzing the final data, let me explain adjustments and revisions BLS does every Jan:
1) revisions to Establishment Survey data and
2) adjustments to population estimates of the Household Survey

#employment

2/24
NFP revisions were done only to reflect revisions of total #employment on Mar 2022.

This means that the Philly #Fed report that found 1.1M overstated jobs in Q2 is not reflected in these revisions.

This won't get reflected until Feb 2024.

3/24
philadelphiafed.org/-/media/frbp/a…
IOW we won't know the real situation with the latest NFPs for yrs, and once we find out, it probably won't matter anymore bc at that point sth else entirely will be at play.

Still annual NFP revision in Jan added 311K jobs to the already published data in 2022.

4/24 Image
Population adjustments are always done as if the whole population changes in just one M (Jan).

No prior data is revised.

For that reason Jan and Dec figures in Household Survey are not easily comparable.

Jan adjustments didn't affect UR but greatly overstated employment.

5/24 Image
On top of the already mentioned annual revisions to Establishment Survey and population adjustments for the Household Survey, BLS is constantly revising its prior data.

This can reflect some meaningful changes in all 3 figures (NFP, UR and AHE) M to M.

6/24
As the published data keeps changing all the time it is tough to do forecasts with any degree of certainty, especially that of NFP.

Therefore it's not unusual to see M where all Wall Street analysts miss the headline number by a wide margin:


7/24
As @Lavorgnanomics said, NFP are nothing but a random number.

I would add that, although many get hyped over it, NFP don't tell us where #inflation is/will be nor it should be used as a determinant for the #Fed's policy in this environment.

8/24
Although these figures will almost certainly get revised lower in the coming M, let's work with what we have for now.

There were strong job gains almost across the board with only mining and finance seeing a slower growth, and IT & utilities seeing (timid) negative growth

9/24 Image
UR ticked down (-0.1 pp) to the cycle low of 3.4%.

Some (including the #Fed) will certainly try to use this as a non-#recession argument but we've had recessions with lower UR.

Also this doesn't mean low UR will cause #inflation. More on this:

10/24
Image
As usual, AHE are the most important detail of the employment report.

Most would interpret the wage growth number as encouraging bc 3.6% annualized is where the #Fed wants it to be (within the level consistent with 2% #inflation).

But...

11/24
...average workweek went up from 34.4 to 34.7.

Adjusted for the longer workweek, AHE are actually up a staggering 14.1% annualized.

Now this number is clearly off and should be regarded as noise bc there is no way wage growth went up by that much in Jan.

12/24
For that reason I suspect in the coming M we will see one of the following:
1) Jan AHE gets revised down compared to Dec
2) Jan workweek gets revised down compared to Dec or
3) both 1) and 2)

13/24
Adding to the confusion are the already mentioned BLS revisions of the prior data.

Surprisingly Jan and Mar of 2021 and Feb of 2022 were revised to reflect 0% MoM which is highly unusual (there should have been at least .sth% change +/-).

14/24 Image
Indeed, if you look at some of my prior charts of the same period, they look somewhat different than the latest one.

BLS is constantly doing revisions of both more recent, as well as some older data painting a (sometimes even substantially) different picture from M to M.

15/24 ImageImage
Given the mentioned problems (revisions and unusually higher workweek showing unrealistic numbers), I decided to leave Jan AHE unadjusted for the longer workweek in my analysis.

Also as mentioned on page 13, this will likely get revised to reflect more realistic numbers.

16/24
Probably the most important take from the Jan employment report:

AHE have been below or within the level consistent with 2% #inflation for the last 3M in a row.

#Fed

17/24 Image
3MMA MoM annualized AHE in:

1) Oct (4.1%) were were just above the higher bound
2) Nov (3.4%) were within
3) Dec (2.6%) were below
4) Jan (2.2%) were well below

the #Fed's LT wage growth target.

For the last 2M wage growth has been lower than the #Fed wants it to be.

18/24 Image
The main #Fed narratives have been disproved by the data:

1) #inflation - "high and sticky" narrative became "services inflation" - but 86%+ and 98% of the CPI disinflated or deflated MoM in Nov and Dec respectively. More details here:

19/24
2) labor market - in his latest comments Powell acknowledged wage growth was not a driver of #inflation

3) no-#recession - such scenario is virtually impossible given what the #Fed has done so far. More in a thread:


20/24
After the #Fed's latest meeting it is clear we are getting near the pause in rate hikes.

Despite what some have been saying, whether the #Fed pauses now or goes 25 in Mar and then pauses or goes 25 in Mar and May and then pauses, makes little difference.

However...

21/24
... there are still some crucial Qs:
1) For how long will the #Fed pause?
2) When will the #Fed cut and by how much?
3) Will #inflation reaccelerate later in the yr like some have said?
4) Will we have a #recession this yr?
5) How will the mkts react to all of that?

22/24
These threads take a lot of time and effort to write.

If you like the content, please love and retweet to help me spread the message.

23/24
I will answer all of these important Qs in more detail at my next

Pick Marko's Brain Workshop Tuesday (Feb 7) at 1 pm ET.

Message me to reserve your spot!

24/24

• • •

Missing some Tweet in this thread? You can try to force a refresh
　

Keep Current with Marko Bjegovic

Marko Bjegovic Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @MBjegovic

Marko Bjegovic Profile picture
Apr 10
At first glance Mar #CPI looks hot.

Some have even said the #Fed should hike rates from here based on the last 3 months of CPI.

Is #inflation reaccelerating?

Let's delve into details.

A CPI thread.

1/9
Headline (+3.477% YoY) and core #CPI (+3.801%) were 8 bps and 7 bps higher than I expected respectively.

My Mar estimate is available here (for free): .

My CPI estimates remain exceptionally accurate with an average error of -2 bps for headline and -4 bps for core.

#inflation

2/9substack.com/home/post/p-14…Image
Image
Headline #CPI was again up +0.6% MoM NSA, double of what we had in Mar 2023, but still lower than both Mar 2022 (+1.3%) and Mar 2021 (+0.7%).

The strength came from:
1) gasoline prices,
2) seasonality and
3) couple of lagged components.

Gasoline prices added +21 bps, shelter added another +21 bps and motor vehicle insurance added another +8 bps to the headline CPI.

So, +50 bps MoM came from just these 3 components where:
gasoline prices are volatile and could be up one month and down another,
shelter is showing persistent numbers that have no foundation in the real world, and
motor vehicle insurance unexpectedly went up by the most since mid-2020 making Mar the 10th largest MoM jump on record (data going back to 1969).

These 3 components make about 40% of the CPI but were responsible for almost 80% of its MoM gain in Mar.

You know where I'm getting at...

The majority of the index (roughly 60%) is only up +0.1 pp NSA which is negative SA (#deflation).

So, not only #inflation is not accelerating but it was down further from Feb.

Headline was only +3 bps above what was expected.

Unexpectedly high increase in motor vehicle insurance alone is responsible for the headline CPI miss.

3/9Image
Read 9 tweets
Marko Bjegovic Profile picture
Jan 11
Many think Dec #CPI disappointed but actually it wasn't that much hotter than expectations like news headlines may suggest.

In the details, CPI remains pretty cool.

Let's delve deeper under the hood.

A CPI thread.

#inflation

1/9
On the surface headline #CPI was 0.1 pp higher than my estimates for MoM and 0.2 pp higher than my estimates for YoY.

Core was 0.1 pp higher than both my MoM and YoY estimates.



#inflation

2/9
News suggest that headline was up +0.3% MoM to +3.4% YoY which is quite misleading.

Headline #CPI (NSA) fell -0.1% MoM which is an outright #deflation for the 3rd month in a row.

How is it possible that headline was down MoM but up +0.3 pp YoY from +3.1%?

Base effects.

Headline wasn't up more than consensus due to MoM increase, but due to lower-than-expected MoM decline. That MoM decline was still significant though.

#inflation

3/9Image
Read 9 tweets
Marko Bjegovic Profile picture
Jan 5
Dec Employment Report looks hot on the surface with
💥NFP above +200K
💥UR not going up from 3.7% and
💥AHE running at +0.4% MoM for the 2nd month in a row.

However, there is much more weakness under the hood.

Let's dive deeper into the numbers.

A thread.

1/8
Dec NFP was +216K but Nov and Oct were revised down by a total of -71K with Oct now matching Jun for the lowest NFP number in 3 yrs of only +105K.

So far, almost all months (save Jul) were revised down in 2023 with total downward revision close to -0.5M.

If Dec is revised down in line with the avg revision in the last 3 months (-48K) NFP could end up in line with +170K consensus.

So the total downward revision in 2023 would equal all job gains in Q4.

Downward revisions of such magnitude and persistence don't happen in either strong nor tight labor mkt like some, including the #Fed, have been describing it.

2/8Image
Private NFP confirm this weakness even more.

Coincidentally, private NFP exactly matched the ADP number this month (+164K) but Nov and Oct were also revised down by a total of -55K.

Oct is now only +44K, the lowest since Dec 2020.

Private NFP were revised down in all months of 2023 by a total of -0.6M.

This is almost 5x the 3MMA for private NFP.

To have 5 months of private NFP taken away by revisions over 11 months is virtually unheard of.

Worth to repeat, this is a sign of weakness, not strength.

3/8Image
Read 8 tweets
Marko Bjegovic Profile picture
Sep 13, 2023
Headline CPI at +3.7% YoY (it was +3.0% 2M ago) which is 0.1 pp higher than expected.

Core CPI at +0.3% MoM, again +0.1 pp higher than expected.

Are there reasons for us and the Fed to worry?

Let's delve deeper.

A CPI thread.

1/9
For accountability purposes, I'll say that headline came +0.1 pp higher on YoY basis and +0.1 pp higher on MoM basis than my estimates, while MoM headline and YoY core came in line with my estimates.



2/9x.com/MBjegovic/stat…
Now every1 seems to talk about gasoline prices lifting headline into sky but in reality headline was +0.4% MoM (NSA), not a lot above +0.3% in Jun.

While SA number shows +0.5 pp difference between Jun and Aug, NSA shows only +0.1 pp difference.

IOW upside in headline has been much more about SA than actual numbers itself.

3/9
 Image
Read 9 tweets
Marko Bjegovic Profile picture
Aug 29, 2023
Lots of chatter around the latest #JOLTS number that went down to 8.827M, the lowest level since Mar 2021 (8.399M).

Important to note a few things:
1) despite the #Fed frequently saying it, this number was never representative of real labor demand bc it has included many fake job openings (IOW it has overstated labor demand by a wide margin which was especially evident after the lockdowns in 2020/2021)
2) right now JOLTS vs unemployed is 1.5 and the Fed wants to see it at 1. However, there is no guarantee this number will stop falling once it equates with unemployed ppl
3) by the time 2) happens this may actually mean that unemployed ppl are already way above JOLTS bc of 1)
4) if 3) is true this also means that #UR will go way above the Fed's desired level
5) Fed's desired UR is recessionary. IOW they have been expecting and hoping for a #recession while trying to convince every1 #softlanding is their base case

A short thread.

1/6
When the #Fed gets misled by inaccurate numbers like #JOLTS or lagging numbers like #CPI/#PCE one thing is certain - they will overdo it, no matter which direction they go.

And that is what they have done.

First they were 1 yr late to start tightening.

If they don't hike from here they will be 1 yr late to stop raising rates.

If they hike one more time this yr like they say they will then they will be even more late.

In this 1 yr so far they hiked by an unnecessary 300 bps.

At one point chickens will come home to roost.

When the lags from all of these (mostly unnecessary) hikes play out, we will get a long (maybe even severe) #recession.

2/6
The #Fed is trying to convince every1 how they got everything under control bc they can always cut rates if things go haywire.

But #recession is likely to be long bc:
1) many at the Fed still has a hiking bias and recession will likely start long before they realize the damage they've created making them miss the time to reverse course even more
2) lags work both ways meaning it will take a long time for these cuts to take effect just like it has taken for the hiking lags to play out

IOW it would be the best thing if recession started yesterday bc the longer we wait, the more damaging it will be.

For any1 thinking whether #softlanding is possible at this time, my answer is hell no – there is no way to escape the economic damage the Fed has created with their too high rates.

In the last 1.5 yrs they hiked by the most since 1980.

Including B/S this is the most severe tightening of the MP in history and there are still ppl thinking there won’t be any recession that follows?

If much less severe MP tightenings couldn’t pull a soft landing, how on Earth could the Fed be able to pull it off this time?

And the best thing is that they talk about it like it’s a done deal despite the fact that they have never done it with such tightening, not even remotely close.

3/6
Read 6 tweets
Marko Bjegovic Profile picture
Aug 9, 2023
Tomorrow we will get Jul #CPI.

Lots of worries around gasoline prices and what will they mean for the #inflation in both July and going forward, as this will be a determinant of the #Fed's policy going forward.

So, where will Jul CPI print at?

A thread.

#inflation

1/7
Before we get to the #CPI estimates, oil prices have (finally) broken to new 2023 highs tdy.

Still, that doesn't necessarily mean that we will see higher #inflation going forward.

More on this in a days old tweet:


2/7
My #CPI estimates are the same for MoM and 0.1 lower for YoY than consensus.

Also they are 0.2 pp lower for both MoM and YoY estimates from the Cleveland #Fed.

#inflation

3/7 Image
Read 7 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

Send Email!

:(