Trump fires labor statistic chief...

[Valcazar]

It is even worse, and it is already causing a stir at work. I know our finance people (and almost any investor/company, btw) rely upon these US numbers amongst other data to help the ceo/cfo etc do long term planning. Thus if those numbers are no longer trustworthy ...God help us.
BTW I double checked Trump's claim that she "inflated the job numbers to help Biden". She actually adjusted him downward by 800k in August 2024, right before the election. Likely not a good time to do it if she was biased for Biden, as she could have waited til November. So Trump's claim is bunk.

I would have to check but I would assume the "never so many corrections in a row in one direction and so large" is a bullshit claim.
But there is another thing to consider. DOGE tore a swath through government before it petered out. Did the BLS get fucked over during that time, making it more prone to error?

 

[Valcazar]

Shaq don't try to quibble.

FACT: We have had major revisions in the jobs data the past few years. In some instances, the revisions came several months later.

I didn't give you any reasons because it is irrelevant to the fact above.

But is that any different than what came before or is it just more widely reported because of how politicized things have become?

 

[Valcazar]

The BLS has been producing the same fucking number as with the past potus. This ptus just did not like the numbers, and he fired the chief to have numbers he likes in the future. End of the fucking story.

Not entirely.
The initial numbers have been getting less accurate as fewer and fewer businesses participate in the survey.
I wouldn't be surprised if there have also been reductions in ability due to DOGE fucking with shit, but would need to dig further.

Trump blaming this on the messenger and insisting that the numbers need to be good for him is bullshit (and, to be fair, something most people expected to happen over time. Trump has always said the numbers were being fucked with because he is someone who believes that the numbers should be fucked with) but that there seems to have been a loss in accuracy on the early estimates is true enough.

 

[WyattEarp]

But is that any different than what came before or is it just more widely reported because of how politicized things have become?

I don't think so, but perhaps you're right.

I caught a story that said the BLS has had a very hard time with job stats since COVID. I think people here know my perspective. If it ain't working, break it. This has nothing to do with Trump. This is just how I think about things.

Most of the anti-Trump crowd is laser focused on any Trump action that can be criticized. For me, the Fed has been flying blind for a few months at an inflection point. While I don't expect an emergency rate cut (it spooks markets), the Fed will move on a big rate cut in September.

I personally could give two shits about what Trump thinks rates should be. However, we can't move on from Jerome Powell soon enough.

 

[WyattEarp]

I wouldn't be surprised if there have also been reductions in ability due to DOGE fucking with shit, but would need to dig further.

I don't think this is a fair statement. BLS had a big miss with the data in pre-DOGE 2024. Regardless of how you view it politically, an election year is not a good time to be popping surprises on jobs data.

In my opinion, this could be a good time for a big change at the BLS. Hopefully, they will be prodded to look at more advanced methods at collecting data. Perhaps the person leading the BLS was doing a good job, but maybe they need new leadership.

 

[Valcazar]

I don't think so, but perhaps you're right.

I caught a story that said the BLS has had a very hard time with job stats since COVID. I think people here know my perspective. If it ain't working, break it. This has nothing to do with Trump. This is just how I think about things.

The thing to break would be excessive reporting and politicization of the initial reports.
If those numbers aren't coming in with useful accuracy then they should be reported as having wild margins for error.

Or, maybe even better, they should be reported as the total jobs and not "jobs added/jobs lost" since those numbers are much more tentative given they are such a tiny percentage of the overall number.
(as mentioned above)

Most of the anti-Trump crowd is laser focused on any Trump action that can be criticized. For me, the Fed has been flying blind for a few months at an inflection point. While I don't expect an emergency rate cut (it spooks markets), the Fed will move on a big rate cut in September.

I personally could give two shits about what Trump thinks rates should be. However, we can't move on from Jerome Powell soon enough.

Trump firing the head of the Fed seems inevitable.
He's only been worried about backlash from rich people he respects and once the economy is bad enough, he won't care about that and just indulge his authoritarian tendencies.

I don't think this is a fair statement. BLS had a big miss with the data in pre-DOGE 2024. Regardless of how you view it politically, an election year is not a good time to be popping surprises on jobs data.

I'm arguing if DOGE fucked with BLS it would make things worse, not that they were the only problem. (The non-response rate seems to be the major problem.)

In my opinion, this could be a good time for a big change at the BLS. Hopefully, they will be prodded to look at more advanced methods at collecting data. Perhaps the person leading the BLS was doing a good job, but maybe they need new leadership.

The question is can they get a better pre-estimate than the current survey? What other signals can they use?
I would have no objection to using a better proxy measure, but any proxy measure will have its weaknesses.

 

[WyattEarp]

Trump firing the head of the Fed seems inevitable.

I knew the moment the numbers were revised the BLS head was gone.

He's only been worried about backlash from rich people he respects and once the economy is bad enough, he won't care about that and just indulge his authoritarian tendencies.

Yes, the government will only answer to Herr Trump.

These takes are odd. I can't think of anything more important than timely and accurate jobs data from the BLS. The statist crowd seems to be arguing why mess with a broken system

 

[WyattEarp]

Trump blaming this on the messenger and insisting that the numbers need to be good for him is bullshit (and, to be fair, something most people expected to happen over time. Trump has always said the numbers were being fucked with because he is someone who believes that the numbers should be fucked with) but that there seems to have been a loss in accuracy on the early estimates is true enough.

I don't think anything nefarious was happening. However, I do think there is a lot of human judgement involved with jobs data. Let's call it "massaging" the numbers. Anytime, you have human judgement you are setting up the potential for bias one way or another. It doesn't even need to be political bias. It can be merely we always adjust the numbers this way thinking.

Now in regards to Fed Chairman Powell, I think he hasn't had a good tenure. High U.S. inflation and indirectly high global inflation was unleashed under his watch. He and other Fed Governors in response now have been leaning on high rates for too long. Erroneous jobs data was giving them cover for holding on to high interest rates.

As I said if the jobs data were accurate, rates would have been cut sooner. The Bank of Canada would have cut rates along with the Fed.
(Higher U.S. rates tend to draw foreign capital and affect exchange rates that's why other central banks watch the Fed's actions.)

 

[jalimon]

But is that any different than what came before or is it just more widely reported because of how politicized things have become?

That's where even guys like you Val have still not yet realized the situation we are in.

Things are not politicized at all.

The Trump regime does not care about politics. They care about controlling everything their way.

The Democrats are being crushed right now because they are irrelevant.

 

[KDK13]

I appreciate everyone's POV. I think the point here is that people who actually know the numbers know the limitations as well. We know at work it's based estimates, and all the usual methodological issues and crticisms found when using surveys etc apply so grains of salt are taken always when we get briefed by the finance guys. However, they also know it's a best estimate, but they also know it employs a consistent methodology from month to month - thus it is useful - and certainly much better than nothing - at identifying trends from month to month or quarter to quarter or year to year. And, there was acknowledgement that it was based upon best faith efforts by a board to get those numbers as correct as possible. (Not the work/opinion of one lady).

But I think the discussion about exactly how accurate are the numbers has shifted it away from the real problem - that is the president of the USA pulls a third world dictator move and then justifies it with a blatant lie or lies.

Now, no one will trust the USA numbers.

Before long, it will be the "record harvest" every year - like the communists used to report - as their people starved to death.

 

[Valcazar]

I knew the moment the numbers were revised the BLS head was gone.

At least you aren't denying that he's going to keep doing this thing where he fires people who aren't "loyal" and just goes with political operatives everywhere, regardless of the legality or sagacity of the action.

Yes, the government will only answer to Herr Trump.

That's what he would like, clearly.
He doesn't pretend otherwise.

These takes are odd. I can't think of anything more important than timely and accurate jobs data from the BLS. The statist crowd seems to be arguing why mess with a broken system

Firing her has nothing to do with fixing the system, though.
The problem with the surveys is a decade or more old.

He is firing her because he claims she is "rigging the numbers" against him.
That is not "I want to fix the collection system" that is "I want the system to report what I want it to report and I will fire people until I get what I want".

Everyone has been worried that virtually all data collecting in the US government was going to get worse and worse under Trump - either through neglect or through outright politicization.
This appears to be specifically the latter, which is always worse because trust can be regained by fixing neglect far more easily than establishing that the numbers are going to be just lies.

 

[jalimon]

The weatherman said last night today would be all sunny and 30 degrees.

Turns it was only 25 and we even had a bit of rain.

i just filed a complaint to have the weatherman fired.

 

[Valcazar]

I don't think anything nefarious was happening.

There really isn't any reason to believe there is.
Which is why Trump claiming that's the reason is so worrying.

However, I do think there is a lot of human judgement involved with jobs data. Let's call it "massaging" the numbers. Anytime, you have human judgement you are setting up the potential for bias one way or another. It doesn't even need to be political bias. It can be merely we always adjust the numbers this way thinking.

Wyatt, are you claiming statistical methods are now "just judgment and massaging"?

You can look at the historical revision pattern in various ways.
https://www.reddit.com/r/dataisbeautiful/comments/1mf08pa https://www.reddit.com/r/Economics/comments/1ew7gzb
You can look at their methodology.

Overview

www.bls.gov

You can look at the response rate.

Household and establishment survey response rates

Household and establishment survey response rates

www.bls.gov

You can look at confidence intervals

Employment by industry, monthly changes, with confidence intervals

www.bls.gov

You can read their own explanation of the revisions.

Why are there revisions to the jobs numbers?

At the beginning of each month, the Bureau of Labor Statistics (BLS) reports the change in payroll employment for the previous month. This estimate of jobs gained or lost over the month is closely watched by policymakers and those who work in financial markets and the media. When the estimate...

www.bls.gov

Can you argue that there is reason for a methodological overhaul?
Very likely you can. In the modern era, there might be better ways to get early samples and their may even be faster ways to get the more solid later results.
All worth looking at.

The issue here, of course, is that all of that has nothing to do with what Trump fired her for.
He fired her because he thinks she's conspiring against him and he wants a political loyalist who will massage the numbers the way he wants them massaged.

Now in regards to Fed Chairman Powell, I think he hasn't had a good tenure. High U.S. inflation and indirectly high global inflation was unleashed under his watch. He and other Fed Governors in response now have been leaning on high rates for too long. Erroneous jobs data was giving them cover for holding on to high interest rates.

Tough.
I mean that.
The choice was made some time ago that the four-year term should be independent.
"I disagree with him" doesn't make the cut.

As I said if the jobs data were accurate, rates would have been cut sooner. The Bank of Canada would have cut rates along with the Fed.
(Higher U.S. rates tend to draw foreign capital and affect exchange rates that's why other central banks watch the Fed's actions.)

The trade off you are asking for is "the President can make Federal policy entirely dependent on his political whims".
Not a position I want people to be in, thanks.

 

[Valcazar]

That's where even guys like you Val have still not yet realized the situation we are in.

Things are not politicized at all.

The Trump regime does not care about politics. They care about controlling everything their way.

You're going to have to elaborate on your statement here.

One - yes, things are politicized.
I'm talking far before Trump alone here.

Controlling everything their own way is caring about politics. (I think you are using politics to talk only about partisanship or possibly electoral politics, which isn't the same as politics as a whole in my view, but I'd prefer you elaborate since I could obviously be wrong.)

The Democrats are being crushed right now because they are irrelevant.

You will need to elaborate on that as well.
What doe you mean by "being crushed" in this context?
I

 

[Valcazar]

But I think the discussion about exactly how accurate are the numbers has shifted it away from the real problem - that is the president of the USA pulls a third world dictator move and then justifies it with a blatant lie or lies.

Now, no one will trust the USA numbers.

Before long, it will be the "record harvest" every year - like the communists used to report - as their people starved to death.

This is exactly the problem, and it isn't even something Trump is bothering to hide.

 

[WyattEarp]

At least you aren't denying that he's going to keep doing this thing where he fires people who aren't "loyal" and just goes with political operatives everywhere, regardless of the legality or sagacity of the action.

That's not at all what I'm saying. Let's not go down that path where you put words into my mouth. Leave that to the less capable minds here.

I was clear in saying that if things aren't going well there will be consequences. This wasn't caused by DOGE. It's been a few years in the making.

Firing her has nothing to do with fixing the system, though.
The problem with the surveys is a decade or more old.

He is firing her because he claims she is "rigging the numbers" against him.
That is not "I want to fix the collection system" that is "I want the system to report what I want it to report and I will fire people until I get what I want".

Everyone has been worried that virtually all data collecting in the US government was going to get worse and worse under Trump - either through neglect or through outright politicization.
This appears to be specifically the latter, which is always worse because trust can be regained by fixing neglect far more easily than establishing that the numbers are going to be just lies.

It would seem to me the jobs numbers are only directionally correct most of the time. They also appear to miss inflection points.

So I don't think they hold as much value as the markets have historically placed on them. And not of much use to the Federal Reserve in calibrating interest rates in a timely basis.

 

[WyattEarp]

Wyatt, are you claiming statistical methods are now "just judgment and massaging"?

The employment statistics go through human processes that require some judgement. It's much like a common debate on TERB and MERB a few years back where people would argue that something was driven by an algorithm and therefore humans had no hand in the result.

If you read the articles you posted, you will see words like estimate (many times over) and assume for the BLS. This is not saying that there is politicization within BLS. The numbers have just become more fluid and the BLS' methodologies tired..

Tough.
I mean that.
The choice was made some time ago that the four-year term should be independent.
"I disagree with him" doesn't make the cut.

"I disagree with him (Powell)" is my personal but widely shared opinion. As I stated with clarity, IMHO Powell's tenure as Fed Chair will not be lauded in posterity. I don't even have to make my case that Powell's description of inflation as "transitory" in 2021 was WRONG. Where I come from, results matter. Powell's results as Fed Chair do not "make the cut".

 

[Valcazar]

That's not at all what I'm saying. Let's not go down that path where you put words into my mouth. Leave that to the less capable minds here.

I was clear in saying that if things aren't going well there will be consequences. This wasn't caused by DOGE. It's been a few years in the making.

"There would be consequences" is a very different statement from the one you actually made.

You said that the moment he got bad numbers, you knew she would be fired.
You also know that firing her has nothing to do with the methods and Trump himself provided the bullshit excuse that she was rigging the numbers against him.

So either you "knew" she would be fired because you are wildly disconnected from reality and made a lucky guess or you "knew" because you know he makes political decisions to fire non-loyalists.

You're not an idiot, so it was the latter.
I don't know why you are pretending to be stupider than you are here.

It would seem to me the jobs numbers are only directionally correct most of the time. They also appear to miss inflection points.

So I don't think they hold as much value as the markets have historically placed on them. And not of much use to the Federal Reserve in calibrating interest rates in a timely basis.

I have no idea what you mean by "directionally correct" here.
I'm glad you agree that firing the BLS head is just wildly political and has nothing to do with the system itself.
The issues with the estimation method are well known and people have said for years that reporting the initial estimates without caveats is bad.
At the same time, people demand fast numbers and so BLS has done the bast they can to provide responsible ones.

We could be having an interesting discussion about revising the system, but instead we are in a world where Trump demands political subservience.

 

[Valcazar]

The employment statistics go through human processes that require some judgement. It's much like a common debate on TERB and MERB a few years back where people would argue that something was driven by an algorithm and therefore humans had no hand in the result.

If you read the articles you posted, you will see words like estimate (many times over) and assume for the BLS. This is not saying that there is politicization within BLS. The numbers have just become more fluid and the BLS' methodologies tired..

Good.
So you admit there is no political shenanigans here.
Just the normal process of statistical work.
So if you admit this, and don't think shenanigans happened, why keep hinting darkly about the possibility of people "massaging" the numbers?
There's no reason to do that except to provide Trump with political cover for his declaration, which is one you have already said you don't agree with.

"I disagree with him (Powell)" is my personal but widely shared opinion. As I stated with clarity, IMHO Powell's tenure as Fed Chair will not be lauded in posterity. I don't even have to make my case that Powell's description of inflation as "transitory" in 2021 was WRONG. Where I come from, results matter. Powell's results as Fed Chair do not "make the cut".

And you can have the opinion you want.
The issue is Trump's threats to fire him for political reasons and replace him with someone more tractable to Trump's position.

Nothing in this thread is about "do we think these systems are performing at their best".
Trump isn't talking about that or justifying his decisions that way.

 

[xmontrealer]

Janet Yellen, the former Treasury secretary and chair of the Federal Reserve, said the firing is not what is expected from the most advanced economy in the world.

"This is the kind of thing you would only expect to see in a banana republic," she said...