Level SettingYou said that the Labor Department should be renamed the propaganda dept. The reasoning for this is the published preliminary revision below.
Source:
https://www.bls.gov/web/empsit/cesprelbmk.htmA downward revision of 818K is bad. But lets make one thing clear: They're estimating total employment, not job growth. Job growth is the first derivative of total employment, so we get that, too. But it's a number that falls out of the true number they're trying to estimate. When we acknowledge that, the revision is 0.5%, not 40%. Still not great, but lets be clear about what we're talking about. You could take a second or third derivative here - say, acceleration in job growth or jerk in job growth, and the numbers could look even more off, but it doesn't really say anything about the whole situation.
It's also worth noting that they published this preliminary revision out of season. Usually, these come out in Jan/February. Why would they do that if they're a propaganda machine? Because they know that the Federal Reserve sets monetary policy, and Congress sets fiscal policy, in part based on what these reports say. Given that we've been managing a pretty tricky economic recovery for 4 years, it's best that policymakers know about this stuff as soon as possible.
This is a good thing.
Now lets get Socratic.
Are these revisions consistently bad?Luckily, the BLS maintains an archive of these annual revisions.
Source:
https://www.bls.gov/web/empsit/cesbmkarch.htmI looked at the last 8 years of reports and pulled out the magnitude of the revision. Remember, this is the percentage difference in the total US employment between the monthly, sample based surveys that inform the regular monthly job reports, and the more comprehensive annual survey.
- 0.2%
+0.4%
+0.3%
- 0.2%
- 0.3%
- 0.0%
+0.1%
- 0.1%
So no.
The revisions are not consistently negative. Not even during the COVID recovery era. And more importantly, they *are* consistently close to 0%.
Another important question to ask,
If an estimate of a 9 figure number is accurate within +/- 0.5% of the actual number, is this estimate useful for policymaking?Yes.
But maybe the revision is inaccurate, too. Maybe the numbers are all garbage and they've been cooking the books at the BLS forever. Ever considered that?Okay, if the employment situation is dramatically worse than the numbers implied by the jobs report, it should show up in other places, right? How about in weekly new unemployment claims:
This chart basically tells the commonly reported story - lots of people went on unemployment during COVID, but the situation has really been pretty good for the past 2 or 3 years, with weekly unemployment claims consistently below 250K.
For reference, the current weekly unemployment claim numbers are basically the same as what they were before the pandemic. In 2008-2010, the last time that pretty much everyone agreed the economy was really bad, the weekly numbers were consistently 2X-3X higher than they are today.
But isn't this just more government data? They're also cooking the books, right?I suppose so. The numbers do come from the Department of
Labor Propaganda, after all. But at the very bottom of their weekly report on unemployment claims, it says this:
This news release presents the weekly unemployment insurance (UI) claims reported by each state's unemployment insurance program offices.
So this means that 50 separate State unemployment offices have to also be in on the conspiracy. Still technically possible, but it's really hard to keep a secret between 50 people, let alone, 50 departments that are made up of dozens of people.
Maybe - But it's still government data. What is the private sector and third parties saying?The Kaiser Family Foundation says that we have a little bit more people covered today under employer sponsored health plans than we did in 2019. This comports with the idea that employment has basically recovered to pre-pandemic levels.
Source:
https://www.kff.org/other/state-indicator/total-population/?dataView=1&activeTab=graph¤tTimeframe=0&startTimeframe=14&selectedDistributions=employer&selectedRows=%7B%22wrapups%22:%7B%22united-states%22:%7B%7D%7D%7D&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D---
The following report from HiringLab (the research arm of Indeed) concludes that 2023 was surprisingly robust.
Source:
https://www.hiringlab.org/2023/11/15/indeeds-2024-us-jobs-hiring-trends-report/Another from September of this year, describing the latest on the employment situation.
Source:
https://www.hiringlab.org/wp-content/uploads/2024/09/sep-2024-us-labor-market-overview.pdfThe latest report is from HiringLab is lukewarm, but the key thing I want to point out here is that many of their slides directly use BLS data. Does that mean the BLS data is rock solid? No, not in and of itself. But it does tell us that the research arm of Indeed, a subsidiary of Recruit Holdings (a publicly traded company with $20b in annual revenue) thinks its pretty good.
They have a lot of reasons to want the best and most accurate data possible.
---
Here is one of my favorite examples from the Door Dash quarterly earnings report:
600 million door dash orders a quarter! Consistent year over year growth of around 20% for five consecutive quarters.
Paying 2-4X to get cold McDonalds delivered to you in 38 minutes doesn't seem like something people would spring for in a lackluster economy.So why do people feel so crappy about the economy?That's an excellent question. I have a couple of guesses.
1.) People accept 100% of the credit for wage growth and 0% of the blame for inflation.
We just came through an inflationary period due to the COVID pandemic. But in the US, inflation adjusted wages are actually up! People are making more money on a real dollar basis than they were before the pandemic.
Source:
https://fred.stlouisfed.org/series/LES1252881600QBut people don't see it that way. They see themselves working really hard, their wages going up X%, but prices going up Y%. It doesn't matter that X>Y. They deserve to earn X, and Y is eating into that.
2.) Just imagine what MMM would say.
It's the fucking phones. It's social media. And it's overconsumption.
Yes it is an empirical fact that fast food is more expensive than it used to be. But also, see above. 600 million quarterly door dash orders. This is one of many gold plated luxuries that people are increasingly opting into. It affects their perception of affordability and is generally just a poor personal finance decision.
3.) Partisan Hackery
Republicans feelings on the economy were extremely low under Obama, then skyrocketed under Trump. They crashed back down under Biden.
I'll just be frank and direct with this one: This is a completely nonsensical and incoherent opinion. If people truly feel the economy was better during the pandemic, with large sectors shut down, and 20% unemployment, than it is right now, there is nothing I can say to them or show to them to convince them otherwise.
