My Thoughts on the firing of the BLS Commissioner
I spent over 6 years working for the Census Bureau, of which three and a half were spent in Statistical Methods Division. So, I have some background in this stuff. On the one hand, clear procedures should be in place that produce these labor statistics (and the survey is probably done by Census and sponsored by the BLS.). On the other hand, with clear procedures in place, these statistics shouldn't need constant major revision each month. So it seems to me that the probability that something is wrong is high.
Now this doesn't mean that the commissioner of BLS is responsible. But it does say to me that there are likely problems that need to be fixed, and this brings up the question as to why that hadn't happened. So while not completely clear to me, Trump's firing of Erika McEntarfer may be correct. By the way, she has her Ph.D. (in economics) from the same school, Virginia Tech, from which my degrees are from. So, she should be competent.
I’m a supply chain economist so work more with FRED than any BLS data sources directly. These revisions are extremely standard. Ultimately these reports are based on data submitted by businesses which are rarely submitted on-time. We are talking about an initial error rate of 00.06% - these revisions only look large if you ignore how large our working population is.
More importantly, the financial community has to trust the numbers the US government puts out regarding the economy. The independence of central banking and the data they use is one of the most fundamental aspects of financial stability in capitalism. Our bond market relies on investors knowing exactly what risk the US economy carries.
If all of our data and economic decisions become predicated on whatever’s good in the short term for whoever’s in office, it severely degrades trust in the US economy. If you thought the US debt was dangerous now, just wait for when the financial community no longer views our long term stability as a sure-thing.
I spent over 6 years working for the Census Bureau, of which three and a half were spent in Statistical Methods Division. So, I have some background in this stuff. On the one hand, clear procedures should be in place that produce these labor statistics (and the survey is probably done by Census and sponsored by the BLS.). On the other hand, with clear procedures in place,
What do you mean by the bolded? What unusual major revisions are occurring?
Now this doesn't mean that the commissioner of BLS is responsible. But it does say to me that there are likely problems that need to be fixed, and this brings up the question as to why that hadn't happened. So while not completely clear to me, Trump's firing of Erika McEntarfer may be correct.
I obviously have no first hand knowledge of whether this woman is competent, but I don't need any such knowledge to infer that this firing had nothing to do with her competence. If the reported job numbers had been strong -- even to the point of facial implausibility -- Trump would not have fired her, and indeed would be vigorously defending her from criticism.
By the way, she has her Ph.D. (in economics) from the same school, Virginia Tech, from which my degrees are from. So, she should be competent.
I'm guessing that this is a joke. I can't imagine anyone actually thinking this way.
I’m a supply chain economist so work more with FRED than any BLS data sources directly. These revisions are extremely standard. Ultimately these reports are based on data submitted by businesses which are rarely submitted on-time. We are talking about an initial error rate of 00.06% - these revisions only look large if you ignore how large our working population is.More importa
The problem is that the data which is released every month has to go through significant revisions. And when I use the word "significant" here I mean that the revisions affect the markets, and sometimes they affect them by a lot. That says to me that there is something inaccurate of the initial released statistics. Perhaps, the fix, as you imply, is to just wait until more data is received. When I worked at Census, we would say that the initial non-response rate was too high for accurate data.
So, the question I have is why is data like this, that's not accurate, being released every month. Also, a second question is that was the data that was released every month inaccurate if you go back a couple of years or more. (I don't know this answer.)
But it does seem clear to me that this should have been addressed. And perhaps BLS is correctly following all the procedures. But if that's the case, this should have been addressed.
It doesn't seem horrible. If you know the school you went to has high standards for its PhD programs then I'm not sure what's so bad about it. Of course Virginia Tech could have gone way downhill since Mason went vs the lady in question but if she's got a PhD from his school he should be allowed to show a little pride in his alma matter I would think.
It doesn't seem horrible. If you know the school you went to has high standards for its PhD programs then I'm not wrong what's so bad about it. Of course Virginia Tech could have gone way downhill since Mason went vs the lady in question but if she's got a PhD from his school he should be allowed to show a little pride in his alma matter I would think.
It isn't about pride in your alma mater. I'm not taking a shot at Virginia Tech or its PhD programs, the latter of which I know nothing about. My point is that I've known a lot of people who went to a lot of school. All of them would report that their schools or programs included people who had no business being there.
The problem is that the data which is released every month has to go through significant revisions. And when I use the word "significant" here I mean that the revisions affect the markets, and sometimes they affect them by a lot. That says to me that there is something inaccurate of the initial released statistics. Perhaps, the fix, as you imply, is to just wait until more data
The question you’re asking now is fundamentally different than justifying firing a source of independent economic data. It’s like saying “I think the way the NBA is ref’ed is dumb so clearly LeBron isn’t a good athlete.” You’re taking about two entirely different things. People operate within the systems they inhabit.
Let’s take a current example. Suppliers are quickly approaching being unable to use pre-tariff inventory to keep competitive pricing. Across the board companies are seeing significant hits to their gross margins. In the next few months, we will see price increases begin to hit consumers.
Would you rather have that data in a month at 99.9% certainty and have the market price in that 0.1%, or have everyone fly blind for a year trying to do their own (significantly more inaccurate) research until the 99.99% certainty report comes out?
As I understand it they put out the surveys and when it's time to release they go with what they've received back and then the updates/revisions are based on the remaining surveys coming in.
Anecdotally I have several friends who have been laid-off or furloughed very recently in fairly quick succession--which is kinda rare in my circles anyway. Maybe they're using a different kinda rocket fuel these days and the economy is really going gangbusters but you can't feel it 😀
What do you mean by the bolded? What unusual major revisions are occurring?
I think I explained my reasoning here in my post after yours.
I obviously have no first hand knowledge of whether this woman is competent, but I don't need any such knowledge to infer that this firing had nothing to do with her competence. If the reported job numbers had been strong -- even to the point of facial implausibility -- Trump would not have fired her, and indeed would be vigorously defending her from criticism.
This is more important than you think. I worked for Census from 1975 through the end of 1981 and I met many people, some of whom were in high management positions, who were, to be polite, statistically incompetent. And a lot of this had to do with the schools from which their degrees came from. and I assume that things would be worse today.
Also, my best friend at the time, who I went to school with, eventually became the director of the Census Bureau.
I'm guessing that this is a joke. I can't imagine anyone actually thinking this way.
Part of the reason for my leaving Census was simply that I didn't want to deal with incompetent people anymore. Now, not everyone was like this, but my years at Northrop were much better in this area.
It doesn't seem horrible. If you know the school you went to has high standards for its PhD programs then I'm not sure what's so bad about it. Of course Virginia Tech could have gone way downhill since Mason went vs the lady in question but if she's got a PhD from his school he should be allowed to show a little pride in his alma matter I would think.
I got my degrees in 1973 and 1975, and at that time the standards were very high. Now it may be true that today many years later, the standards have gone down some, but I doubt it based on my knowledgeog the current status of Va Tech.
For what it's worth, I am the one who gets the US Census business survey mailings around here, and I throw them directly into the garbage until I get something that sounds moderately threatening.
Anything that says, "Requests your cooperation" or something like that goes right in the bin. I do the exact same thing every time the City of Milwaukee "kindly requests" data about our rental properties, which they can only ever use to hang you on something or increase taxes. **** you, government.
Our little businesses can only skew the numbers by a few dozen people, though.
The question you’re asking now is fundamentally different than justifying firing a source of independent economic data. It’s like saying “I think the way the NBA is ref’ed is dumb so clearly LeBron isn’t a good athlete.” You’re taking about two entirely different things. People operate within the systems they inhabit.Let’s take a current example. Suppliers are quickly approachi
You're making assumptions about the released statistics that neither you or I know. The only thing we really know is that the released statistics over the past year or so (and perhaps longer) have needed a lot of revision, and in my opinion, this needs to be fixed. Also, the fix might be easy or it might be hard. That's something we don't know, and we also don't know if any effort was being made to solve these problems.
This is more important than you think. I worked for Census from 1975 through the end of 1981 and I met many people, some of whom were in high management positions, who were, to be polite, statistically incompetent. And a lot of this had to do with the schools from which their degrees came from. and I assume that things would be worse today. Also, my best friend at the time, who
I don't know what "this" refers to. There is little chance that competence is more important than I think it is, so I'm guessing that you are referring to where someone went to school. That matters in the aggregate, of course. Harvard of course is going to attract more smart people into its undergraduate and graduate programs than random university X, and a higher percentage of its graduates will go on to distinguish themselves in a professional capacity.
But I can honestly say that I have known graduates from almost every school under the sun who were not sufficiently competent imo.
In my experience, if you are a reasonably smart, industrious person in a "top tier" academic or professional environment, and you look at the people around, you probably will conclude that <5% meet any sort of definition of genius, 10-20% have no business being there, and the rest are a lot like you.
For what it's worth, I am the one who gets the US Census business survey mailings around here, and I throw them directly into the garbage until I get something that sounds moderately threatening. Anything that says, "Requests your cooperation" or something like that goes right in the bin. I do the exact same thing every time the City of Milwaukee "kindly requests" data about
I understand where you're coming from and today I think the Census Bureau does too many surveys and their questions are too extensive. But when you don't answer, this contributes to the non-response and then assumptions are made to account for data that comes under this heading.
The assumptions are fine as long as the non-response rate is low. But when it creeps up, it can distort the data. This may be what's happening, and if that's the case, the BLS stattistics perhaps should not be released as quickly.
As I understand it they put out the surveys and when it's time to release they go with what they've received back and then the updates/revisions are based on the remaining surveys coming in.
This is the answer. As long as this is the standard*, there will always be revisions.
*Many government statistics are done this way. The need to know the information NOW is great. Revisions happen all the time. The only reason this particular number at this particular time is newsworthy is the idiot in charge threw a fit.
The assumptions are fine as long as the non-response rate is low. But when it creeps up, it can distort the data. This may be what's happening, and if that's the case, the BLS stattistics perhaps should not be released as quickly.
Many of these reporting times are mandated by statute.
The problem is that the data which is released every month has to go through significant revisions. And when I use the word "significant" here I mean that the revisions affect the markets, and sometimes they affect them by a lot. That says to me that there is something inaccurate of the initial released statistics. Perhaps, the fix, as you imply, is to just wait until more data
Actually the markets has long moved on when revisions occurs .
Majority don’t actually cares much about revision .
Usually we all know those numbers are just estimates .
This is the answer. As long as this is the standard*, there will always be revisions.
*Many government statistics are done this way. The need to know the information NOW is great. Revisions happen all the time. The only reason this particular number at this particular time is newsworthy is the idiot in charge threw a fit. Many of these reporting times are mandated by statute.
What if we went with trump's method for valuing his businesses by feel? Just took a fantastic dump? Economy looks amazing! Rigorous Science 😀
SMH. This was nothing more than trumps usual petty temper tantrum when something he does (****s up the economy in this case) and has to blame it on someone else. And yet his drooling sycophants lap it up again and again.
And..... trump somehow denying the economy is bad because the jobs report is fake, and at the same time saying "see jay powell, i told you so" makes no sense unless you're a lying conman.
Mason, I know you think I’m trolling you, and admittedly up to this post that is probably accurate because I have never read a single post of yours and thought it carried anything of value
But knowing I’m not trolling with the following question:
Why do you think based on your body of work, your transparently obvious political leanings and your unwillingness to accept you are flat wrong about anything you have speculated on with trump’s presidency that you have any right to the benefit of the doubt with absurdist ‘but maybe’ platitudes as it relates to this topic?
I thought I would give an example of incompetence that I experienced many years ago.
In early 1979, when I was still at Census headquarters in Suitland MD, most of my time was spent working on the Annual Housing Survey (AHS), (It still exists today, is now called America's Housing Survey, but is not done every year anymore.)
What had happened is that the Department of Housing and Urban Development (HUD), who sponsored the survey, wanted to add a supplement that would gather information on the needs of older people. Without going into details. it was clear to those of us who worked in Statistical Methods Division that the data collected would be worthless.
To address this, a meeting was set up with some important managers at Census where we would explain why the data would be worthless. This was done and the answer we received went something like this:
We understand what you are saying. But the money for this work has already been received from HUD and because of that, it can't be stopped.
A few months later I transferred to California to do statistical quality control work on the 1980 Census. So, I don't know exactly what happened to the AHS supplement which we tried to kill. But I suspect that the worthless data was collected, and then HUD spent millions of dollars based on the data that was collected.
I obviously have no first hand knowledge of whether this woman is competent, but I don't need any such knowledge to infer that this firing had nothing to do with her competence. If the reported job numbers had been strong -- even to the point of facial implausibility -- Trump would not have fired her, and indeed would be vigorously defending her from criticism.
Don't think this has been pointed out enough and Rococo is 100% right. This firing has nothing to do with competence.
