Get the Ricochet app
Ricochet will work fine in a browser on your mobile devices, but for the best experience please download our app.
In the Biden years, the reported job numbers each month were positive. Then about 3 weeks later, there would be a significant downward correction. EVERY MONTH. I just assumed that this was partisan hackery--an attempt to feed fake happy numbers to the media followed by quiet corrections back to the real numbers.
But now the recent jobs report and subsequent correction follow exactly the same pattern as under Biden--only with an even bigger correction than usual. You would think that whatever metrics are being used, they would be adjusted or revised so as not to require issuing these embarrassing subsequent admissions of having issued lousy data every time.
I apologize for insinuating political bias when it is apparently just that those guys are not very good at what they do.
The only real difference is that under Biden, the MSM reported the happy initial numbers and ignored the corrections, but under Trump, the reverse is true.
Discussion (34)
To read this bio, please join Ricochet.
Counting the number of employed people seems like a simple data analysis problem, but the BLS insists on using the survey method they developed when disco was new and exciting. The BLS folks could ask the Treasury folks for a count of people paying withholding or quarterly taxes (almost everyone) how many people are working and get an instant and correct answer, but they don't. Surveys are OK to learn about people not employed, but there is still data I would trust better (eg, unemployment checks written).
I suspect the bad methodology is political. It is better for Dems to hire many thousands of friends to produce data that can manipulated rather then settle on a definition of employed as "people paying income taxes" and running a simple Database query. The Dems don't want to equate paying taxes with being employed, because it offend all those aspiring poets in their base.
CHARTER MEMBER
To read this bio, please join Ricochet.
Random errors should result in a roughly equal number of positive and negative revisions over time. If the revisions when more data becomes available later are predominately in a single direction (up or down), that implies a methodology problem in producing the initial estimates.
To read this bio, please join Ricochet.
I heard someone on a business show yesterday say that many are no longer filling out the surveys that BLS needs to compile the data.
To read this bio, please join Ricochet.
To read this bio, please join Ricochet.
Is there evidence that if they did it effectively it would have utility?
You're seeing five of 34 comments. Sign in, or join for just $5 per month, to read them all and add your own.