A blemish or two from the household survey, but all in, the US job market is shrugging off a lot of bad policy. Also, breakevens and a PSA re a new Substack from Justin W.
Trump fired the head of labor statistics because he didn't like the numbers she provided. People who read your posts see anomalies in the correlations between various statistics. Despite this situation, you continue to write as though these were numbers you could trust. Please explain?
Just expressing my wish that you do a post focused on AI and its effect on the employment outlook very specifically for propagandists.
I know this is difficult, because the various propagandist jobs, though they are numerous, must be hidden in about 15 different job titles in the official statistics.
In terms of AI as it exists so far, in general, I have observed it has such a high likelihood of being incorrect, that it is virtually useless to me.
However, for propaganda, "being correct" is known not to be an issue.
Still, exactly how to be incorrect and misleading could be potentially be something only a human can do, so it's not clear to me.
Hoping eventually you and/or econ colleagues at the various institutions can produce something to enlighten those readers interested in this little niche area of economic inquiry.
Since January 2025, the establishment survey (CES) has generally shown moderate but positive payroll growth, while the household survey (CPS) has shown flat-to-declining employment. That’s unusual in duration. Short-term divergences are normal — the surveys measure different things — but a 15-month consistent gap raises structural questions.
Doesn't how "good" a jobs report appears depend not just on absolute numbers of growth, but where jobs are growing v where they are disappearing? I.e. growth in min wage warehouse jobs doesn't really offset high-wage tech jobs although the jobs numbers might look great.
Thanks Jared, just subscribed to Justin’s Platypus!
Didn’t an entire airline just go under?
One possibile explanation is that there are thousands of people employed every day to fix the mistakes made by Artificial intelligence
Trump fired the head of labor statistics because he didn't like the numbers she provided. People who read your posts see anomalies in the correlations between various statistics. Despite this situation, you continue to write as though these were numbers you could trust. Please explain?
Just expressing my wish that you do a post focused on AI and its effect on the employment outlook very specifically for propagandists.
I know this is difficult, because the various propagandist jobs, though they are numerous, must be hidden in about 15 different job titles in the official statistics.
In terms of AI as it exists so far, in general, I have observed it has such a high likelihood of being incorrect, that it is virtually useless to me.
However, for propaganda, "being correct" is known not to be an issue.
Still, exactly how to be incorrect and misleading could be potentially be something only a human can do, so it's not clear to me.
Hoping eventually you and/or econ colleagues at the various institutions can produce something to enlighten those readers interested in this little niche area of economic inquiry.
Thanks.
(😊)
Since January 2025, the establishment survey (CES) has generally shown moderate but positive payroll growth, while the household survey (CPS) has shown flat-to-declining employment. That’s unusual in duration. Short-term divergences are normal — the surveys measure different things — but a 15-month consistent gap raises structural questions.
Doesn't how "good" a jobs report appears depend not just on absolute numbers of growth, but where jobs are growing v where they are disappearing? I.e. growth in min wage warehouse jobs doesn't really offset high-wage tech jobs although the jobs numbers might look great.