Imagine month after month praising “your” economy and strong jobs numbers. Then when the numbers are updated, going on a hissy fit and firing the messenger.
The jobs survey procedure DOES need to be updated and these massive revisions ARE concerning. But they happen under both political parties. Firing the BLS head the literal day of the release is just him throwing a temper tantrum because he didn’t like the result. Now he will just install whoever that will be afraid to publish any result that isn’t a decent jobs number. That’s something that Xi and the CCP does.
How about instead of this knee-jerk, rash action, you create a team to work together to improve how this data is gathered? Or better yet, don’t slash the workforce to bare bones of the teams responsible for gathering the data? That would just make too much sense.
Did it occur to you that the firing happened because the numbers first reported were so egregiously wrong in the to begin with? Trump then rightfully assumes they're correct (because he assumes this person can do their job correctly) and acts accordingly, then the later revision was so massive it makes him look extra bad (vs if they had just reported them correctly to begin with) and it proves that individual's inherent incompetence? If someone's job is to give me a report, they give me that report, I make decisions and speak publicly based on that report (effectively tying it to my reputation and ability to manage expectations), and then it comes out the person whose job it was to produce that report fucked up massively and it subsequently reflects on me both the badness of the actual report and the fact it was so wrong in the first place.
This sounds more like a textbook case in any corporate job that would get someone fired for fucking up like that and it reflecting poorly on senior leadership who relied that person to do the work correctly. Certainly reflects poorly on previous admins that would have this happen and did NOTHING to punish the government employees that clearly failed to do their job. If the job's "too hard to do" because of the complexity required then STOP STEALING MONEY from taxpayers under the guise of pursuing an unachievable task and do away with the entire department.
beep bap boop, does not register in libtard programming override, now must protest and bitch
Any time a government employee messes up and is fired, it must be because the party in charge is corrupt and evil, not because for once a public sector worker was incompetent beyond tolerance.
Yeah, except that Trump cut funding to BLS surveying so he clearly isn't interested in any kind of greater accuracy. And he was begging Powell for rate cuts for months, so he obviously wasn't basing his decisions off of the jobs numbers that formerly appeared strong.
It's laughable how many people like you give Trump the benefit of the doubt time after time when ever a cursory glance at his actions show that, no, this is not any kind of calculated play. He really is just retarded.
I'm about to go to a T7 for biglaw so perhaps the soup I swim in is different but this has been transparently obvious in the law world since the indictments and testimony came out after 2020. Wonder when WSO and finance guys will catch up.
Yeah biglaw's definitely not at all an ideologically biased sector. And as we all know, spending more money on a government dept. is the only way to get good results as history has clearly shown time and again /s
The idea rates shouldn't be cut is purely political with unemployment and inflation as low as it is. Just can't bear to give Trump a perceived win lol
I promise you that biglaw, on aggregate, does not care one way or the other whether Trump "gets a win." For every lib partner there is one strict originalist and two more who would sacrifice all of the Civil Rights Acts for a 2% tax cut. The fact that you think this means you've been poisoned by conservative slop media to hate anything resembling academia.
No, big law isn't ideologically biased. It's amazing to me that basically anyone that disagrees with you is "biased" even if that anyone is "tens of thousands of people working a job that is naturally a draw for more conservative or staid people making huge amounts of money." Big law should be extremely conservative politically, not liberal.
Maybe all these people seem liberal to you because your politics are so far to the right (which is the case) that you'd think William Buckley is woke.
This is insanity. Again, if every person is conspiring against you, then maybe the issue is you.
Jerome Powell is a fucking Trump appointment, so if he's some Deep State operative you seem to claim, so is Trump (obviously). The Fed has been exceptionally consistent in their messaging - they want to see inflation at or consistently closer to 2%, and inflation has been consistently higher than that since Trump took office. More to the point, Trump's tariff policies are deliberately inflationary, so it makes perfect sense to see how those play out.
And as far as firing the person who puts out labor statistics... well, optics matter. You can claim that the BLS director is incompetent all you want, but she was fired after releasing a disappointing jobs report, not because of errors in that report. I understand you are too far down the rabbit hole of supporting Mr Trump to ever change that, because it would absolutely shatter your own ego to be so wrong, but most impartial observers can recognize a pattern when they see one, and the pattern for Mr Trump is to shoot the messenger and the lie about the message.
I don't think the finance world is behind the curve on Trump. We know he's a bullshitter. We're just ahead of the curve in understanding how deep and relentless the government grift is, and thus not inclined to react with pearl-clutching "OMG DICTATOR" every time he upsets the orthodoxy and their precious little rules.
See, I would buy this explanation if it wasn’t for the fact that Trump explicitly said, in writing, that the nonfarm payroll data was intentionally and maliciously manipulated to make him look bad. This is objectively a blatant lie.
1) There are many precedent examples of nonfarm payroll revisions of this magnitude.
2) In my past life, I was at a PE firm and one of the portcos I supported was a payroll company. Their client base was such a good representative sample that we could accurately predict the post-revision nonfarm payroll numbers to a high degree of accuracy (I probably could have made enough to retire early if I traded on the data we had - or have gone to jail). I can tell you with certainty that the revised numbers are mostly accurate. I don’t expect you to give me the benefit of the doubt like you consistently seem to with Trump, so I’d implore you to reach out to anyone in your network who may now work in FP&A or something of the sort at a payroll company. They’ll tell you the same exact thing.
Working to adjust an antiquated and sometimes misleading reporting process is one thing. Firing the head of the agency the day a revised report (which again, has plenty of precedent) comes out is a completely separate thing.
I’d also like to reiterate again that Trump EXPLICITLY said the firing was because the numbers were manipulated for political purposes. In that Tweet, he NEVER mentions fixing the current process or anything of the sort. You are putting words in his mouth.
So because he doesn't explicitly say a reason was part of the consideration in a press release, that rules it out entirely from the discussion as a possibility? We're taking the words of a politician for why they did XYZ at face value? Just because it's a precedent that government employees are almost never fired for their shitty work product, that means it can't be part of the decision-making process?
And that doesn't seem at all naive?
Clearly I’ve struck a chord. I’ve never before seen a wave of MS in a 15-minute span well after the original comment was sent. Just so happens to be the amount needed for the “Most Helpful” tag to go to our friend, and I inherit the their former “Most Controversial” tag.
I kindly disagree. The reason for revision is not simply because the first try was inaccurate, but more necessary data were not available at the moment of first disclosure, and also the granularity can change.
In one sentence, the swing of the job numbers is not due to estimate error or incompetence, but more information flow is available.
Did it occur to you that the numbers might be less well-analyzed than normal because Trump is firing and instituting hiring freezes, limiting the ability of the BLS to collect and analyze the same quantity of data that it used to?
https://www.wsj.com/economy/real-strains-inside-the-bls-made-it-vulnerable-to-trumps-accusations-52857f36?gaa_at=eafs&gaa_n=ASWzDAiu3YRgDw7TA_ZHclb_oFONSq8S_TTYHE6aor9SD6eSuLKhZrZo-vUDPXtKn9I%3D&gaa_ts=68930815&gaa_sig=vg3cH6pK7cB_WojsTO-vqrS_mgFdqjCEGtxrqvn1h3LkQ9p5qrjsAMaQH9rvNoPEmK8vnm40j5QKG2AaTMT7Sg%3D%3D
This stuff is well-documented. But of course, I'm sure you were all about it when Elon was brought on to shoot first and ask questions later with DOGE, and now that the predictable consequences have come home to roost you want to pretend that it's the fault of libs, or some silly childish nonsense like this. Read a book.
It's concerning to me that you are so confident in your poorly-founded opinions. Meanwhile the government is being looted by a reality TV bimbo cosplaying as a mob boss / banana republic dictator. Sad that our country and the public weal has degenerated this far due to the proud ignorance of its people.
But BLS numbers are always adjusted, seasonally and otherwise.
There are the early numbers that are used by policymakers to tweak policies and are usually the ones Wall Street, the Fed, the WH and Congress and the newspapers care about.
Then there are the later ones — the revised numbers — that reflect more complete data collection. These are the “official” historical statistics of the US economy. When you’re building a model in a class on US economic history, these are the numbers you’re working with.
Trump shot the messenger because she delivered bad news. The woman did nothing wrong. Her data was as good as it’s always been and BLS is a world-class collector of information. Best statistical shop in the world. Seriously.
This clown PrivateTechQuityGME comments on every post always with the dumbest take. No wonder he’s a GMEbagholder
These numbers get revised all the time and the surveys that the BLS solicits have had lower response rates since COVID. You need to read more about how the BLS does things.
I strongly suggest that anyone who wants to understand these revisions in more context reads this (no paywall):
natesilver (dot) net/p/trumps-jobs-data-denialism-wont-fool
Great read and very helpful to understand the current environment, thanks for sharing!
Honestly gotta wonder what's the point of this data and reports if every single report has to be revised wildly making the initial numbers worthless. I think it's a big assumption that the second set of numbers are even accurate. If I got a report about something one month and a wildly different 'revised' report the next, I wouldn't trust either- makes me skeptical to the actual process of how the information is generated.
This is not a diss or praise of Trump, just a surface level observation. I worry government data is trusted too much as gospel when the underlying data is shaky, unreliable, or flat out wrong.
100% government is fully of shit across parties and can't be trusted. Of course they're going to claim the policies they pass are creating growth. It's the only reason they're tolerated.
yeah but not even them being full of shit in the sense they are lying / deceitful. Maybe it's just too big and complicated a job for anyone to do well / accurately given its all so complex and limited resources.
It's like with voting. I've always wondered how accurate the counts are. Seem to be accurate enough but we know its not 100% since there are re-counts and they'll often get different numbers. So either the original count was wrong, the new count is still wrong, or both are wrong.
Doesn't even get into corrupt officials with their fingers on the scales- even Putin was 'elected' who knows what the actual count is!
Typically, it's been the case that the jobs numbers get revised up and down bit by bit, and that the upward revisions make up for the downwards revisions. It's just that there's been an absurd number of downwards revisions.
Biggest reason behind this is likely to do with 2 things. 1, prevalence of layoffs and "stealth layoffs" through return to office, which causes downwards revisions later on that often aren't counted(so if Amazon lays off 10k staff, that needs to be revised down in that month at a later time if they aren't replacing those workers). With the recent downwards revisions, it seems like education cuts and DOE cuts + school closures have led to downwards revisions in the education sector, particularly k-12. 2, uber and other gig businesses. These are registered as businesses, so when they're started up those people oftentimes register them as such. When the BLS is counting for new business registrations, they look at these figures of new businesses created, and assume that these new businesses must be hiring a lot of people to get the business rolling. This ends up not being the case, so at a later time those figures get revised down.
The BLS is a constantly rolling machine that is trying to adapt to the market environments that change a lot. It's worth keeping in mind, getting a whole view of the economy is really hard. We didn't know we were in a recession in 2008-2009 until more than a year later when the economic and GDP data was finalized. We have better reports on labor and non farm payroll, called the quarterly census of employment and wages, but that one takes a really long time to compile, and by the time we get that data it's almost too late to act on it.
del - embed fail
I also want to point something out - forgetting who the director of BLS is for the moment and the political aspect - there is one thing that we are forgetting here. BLS is an executive branch department under the Department of Labor. As such, Trump is able to hire or fire within that department as he sees fit. All presidents have the opportunity to bring in staff that they want to run the Executive Branch as they see fit. Imagine for a minute, if Trump was required to keep James Clapper or John Brennan on as DNI and DCIA despite the fact that they didn't align on his foreign policy strategy, or Obama being required to keep Michael Hayden on after W. left office instead of replacing him with Leon Panetta. There is latitude to replace appointees under the Executive Branch as needed.
Now, mind you, I agree that BLS needs to revise the entire process. I think that's a given. However, if 2/3 of the last 24 months need to be revised downward, that's just as telling. Interestingly enough, the majority of Biden era revisions were downward. So the question is why, and what did William Beach and Ericka McEntarfer do to explain why? We can look at certain periods of time ('08 to '10, '11 to '18, and '19 to '21) and contextualize the shifts and revisions, but after that, if we listen to the politicians, '21 and beyond become much harder to contextualize in light of the downward revisions. Even things like the massive Tech layoffs going on now can be factored in, but the relative black box we're looking at doesn't help.
Data has been deteriorating because of lack of staff
https://www.apolloacademy.com/the-quality-of-the-cpi-data-continues-to-… The Quality of CPI Data Continues to Deteriorate - Apollo Academy
That's definitely interesting to see and is disconcerting. I'd love to see the underlying/root cause analysis behind the deterioration. I'd also love to see the trend over time with the new BLS head. A six month sample is telling, but it's only six months. If it continues to trend upward or exceed the 25% threshold by December, I would be concerned over lack of staff to actually get the data properly. Without knowing the ins and outs of BLS data, survey methodology, etc, this could just be a blip vs something more ominous. Even something like sample size and distribution can impact the inputs here. Again, I'm not discrediting the blip, I'm just being cautious about what needs to be done to correct the errors here. For all we know it could be political (e.g., McEntarfer purposely screwing it up to make Trump look bad), issues with staffing, poor response rate, survey sampling/bias/etc. that are creating confounding variable.
I believe it's lack of surveyors collecting data because that department had staff laid off. So they have to estimate based on the survey data they do collect. About 80% sure on that
Just as a follow-up and something I saw this morning, here's an interesting take from South Bay Research (courtesy of ZeroHedge since I can't find it on the SB Research website). It highlights at least one problem that I mentioned earlier, poor data bouyed by lower response rates from small businesses. I tend to agree with their take, McEntarfer was in charge and the buck ultimately stopped with her, but this was not really a "political" targeting a la she was a Biden appointee, but a scapegoat for the larger organizational failings that need to be fixed and highlight broader governmental issues with data collection dating back to prior administrations.
It's not because of a lack of staff. We had even worse accuracy in the previous 4 years. It is a highly politicized government agency and should be viewed as nothing more than propaganda. For fucks sake we have companies like ADP who could whip out a jobs report in 20 min that would be 100x more accurate for one millionth the cost of the BLS and it would add a 10 bagger to the market cap of ADP.
The graph literally shows a gigantic jump in what they assume vs collect. It goes back to 2019 and nothing was as poor quality as the data this year. Did you even click the link?
could they just release numbers on a 2 month lag to ensure greater accuracy?
LOL, the reflexive defending of government beuracrats in this sub is the most soy shit I have ever seen. The BLS is one of the worst government agencies in terms of their numbers and reports. No one except smooth brains takes the numbers that they put out and believes they are even remotely real. It is common for 10%+ month over month adjustments to be made to the jobs report. Can you name a single other organization where being 10% wrong is considered top level performance? Hint, you can't.
Consectetur suscipit molestiae facere nulla hic. Distinctio veniam esse autem consequatur veniam consectetur. Deserunt consequatur eum omnis tenetur rerum officiis.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...
Autem odit libero voluptate consequatur nesciunt earum. Tempore et dolorum quibusdam nulla facilis tempore.