This has been the case in several recent reports. It really ought to put to bed the idea that Trump was somehow going to be "good for business" and bring back manufacturing. At best all he's done is create a ton of uncertainty, at worst he's actively harmed massive swaths of American businesses with tariffs, war, and gutting the federal workforce and programs that indirectly supported a ton of businesses.
> In yet another troubling sign for the labor market, the Bureau of Labor Statistics said that it revised down the prior two months by a combined 103,000. May’s jobs total was cut by 66,000 to 129,000 total jobs added, while June’s total was lowered by 37,000 to a total gain of 57,000.
One of the reaction threads I saw elsewhere on social media pointed out the response rates for the surveys that come up with these numbers is plummeting--about a decade ago, we had ~85% response rate, now we're done to ~65% response rate. So the initial quality of these numbers is probably getting a lot worse.
This is how bls data has always worked. And the reason is that they don’t always have the full picture every month and need to account for new data in the stats.
Honest ask because I’ve not generally paid attention…
Is it the case that the numbers always appear more optimistic, and then are revised to be more pessimistic values? Or is it just that the original numbers are off and are revised? Does that make sense?
> They move in both directions regularly, typically in the direction the economic situation is moving
And for good reason! These are surveys. The easiest respondents are those saying “no change.” The latest responses, the ones who had the biggest change. In a booming economy, the biggest changes will be hires. In a troubled one, the biggest changes will be those doing lay-offs.
Add to that: the BLS tries to correct for this effect. Which means at the moment the economy turns, its corrections skew the wrong way. (On the whole, per the above chart, the figures are getting better over time.)
If you can do economics without revision, that is, find the real number of jobs lost or gained or workplace participation in real time, you could make a lot of money. Probably win some prizes, too.
Interesting, this article doesn't seem to mention the layoffs from the tech industry. Which, if you look at the numbers, sort of make up the bulk of it.
Tech sector is not the bulk of it, many sectors doing worse and that is why they got the headlines. "Financial Activities" supersector was terrible this month. Another reason is there is not an umbrella "tech" sector in the NAICS, so it's harder to report on.
I began rapidly recalling the hundreds of sources foreseeing this well in advance, citing it, raising alarm, and being ignored.
I guess we are supposed to take numbers at face value, disregard what generated them and why, accept what they imply as presented, and never question it.
Then we must patiently withhold intelligence, wait for sanctioned admission, then finally consider what we already knew.
Suddenly, Rome collapsed.
Edit: Don't take the flagging to heart. If you dig, you will find a thousand exceptions made where the single-word comment fit the narrative and endured. This comment will be flagged too.
What is also interesting is that the only sector thad added jobs was health care.
This has been the case in several recent reports. It really ought to put to bed the idea that Trump was somehow going to be "good for business" and bring back manufacturing. At best all he's done is create a ton of uncertainty, at worst he's actively harmed massive swaths of American businesses with tariffs, war, and gutting the federal workforce and programs that indirectly supported a ton of businesses.
It was the same in June I think...
Meanwhile Canada added 75,000 new jobs in July. Perhaps Trump's tariff scheme is not working so well after all.
https://www.cbc.ca/lite/story/9.7299225
Note, also:
> In yet another troubling sign for the labor market, the Bureau of Labor Statistics said that it revised down the prior two months by a combined 103,000. May’s jobs total was cut by 66,000 to 129,000 total jobs added, while June’s total was lowered by 37,000 to a total gain of 57,000.
One of the reaction threads I saw elsewhere on social media pointed out the response rates for the surveys that come up with these numbers is plummeting--about a decade ago, we had ~85% response rate, now we're done to ~65% response rate. So the initial quality of these numbers is probably getting a lot worse.
https://xcancel.com/ernietedeschi/status/1951721848393105573 indicates they've been getting better over time.
I'd suspect better computing helps a lot here.
Cut that graph to the last decade, and they don't look like they've been getting better.
The COVID era is likely to be an outlier in many things.
These guys are just nonstop liars. Print the headline, revise when no one is looking.
To be fair, revisions are very normal, because the more reliable data (tax filings, for example) comes in later.
https://xcancel.com/ernietedeschi/status/1951721848393105573
The jobs numbers are notoriously laggy. On the way up, lag means constant upwards revisions. On the way down, lag means constant downwards revisions.
ADP Payrolls are a lower latency indicator but not quite as broad.
This is how bls data has always worked. And the reason is that they don’t always have the full picture every month and need to account for new data in the stats.
Honest ask because I’ve not generally paid attention…
Is it the case that the numbers always appear more optimistic, and then are revised to be more pessimistic values? Or is it just that the original numbers are off and are revised? Does that make sense?
They move in both directions regularly, typically in the direction the economic situation is moving at the time.
Chart: https://xcancel.com/ernietedeschi/status/1951721848393105573
> They move in both directions regularly, typically in the direction the economic situation is moving
And for good reason! These are surveys. The easiest respondents are those saying “no change.” The latest responses, the ones who had the biggest change. In a booming economy, the biggest changes will be hires. In a troubled one, the biggest changes will be those doing lay-offs.
Add to that: the BLS tries to correct for this effect. Which means at the moment the economy turns, its corrections skew the wrong way. (On the whole, per the above chart, the figures are getting better over time.)
If you can do economics without revision, that is, find the real number of jobs lost or gained or workplace participation in real time, you could make a lot of money. Probably win some prizes, too.
Economics is an imperfect science.
Vertical Thinkers: “Who could’ve seen this coming when all the data looks good?!”
Lateral Thinkers: quietly sipping tea, nursing our wounds from layoffs, lamenting another unheard Cassandra moment.
Basically sums up the mood, I find. Don’t worry, we’re always steeping a fresh pot for the new club members when stories like these hit the wire.
Interesting, this article doesn't seem to mention the layoffs from the tech industry. Which, if you look at the numbers, sort of make up the bulk of it.
Tech sector is not the bulk of it, many sectors doing worse and that is why they got the headlines. "Financial Activities" supersector was terrible this month. Another reason is there is not an umbrella "tech" sector in the NAICS, so it's harder to report on.
"sudden"
My thought exactly.
I began rapidly recalling the hundreds of sources foreseeing this well in advance, citing it, raising alarm, and being ignored.
I guess we are supposed to take numbers at face value, disregard what generated them and why, accept what they imply as presented, and never question it.
Then we must patiently withhold intelligence, wait for sanctioned admission, then finally consider what we already knew.
Suddenly, Rome collapsed.
Edit: Don't take the flagging to heart. If you dig, you will find a thousand exceptions made where the single-word comment fit the narrative and endured. This comment will be flagged too.
I predict the numbers will look good after November. Maybe not reality but the numbers.
https://news.ycombinator.com/newsguidelines.html