> Instead of firing the commissioner, the president should be giving the agency a raise in the form of a bigger budget. The goal should be to restore public trust in government statistics, not undermine it.
When you believe that government should not be providing many services, or doing most of what it currently does overall, why would you want to bolster trust in government statistics? Those statistics might contradict the administration, which is not a goal of the administration and its backers.
I understand why articles like this are written and why we have these conversations but it all feels like we're sane washing this administration. Any other administration (especially a Democrat) firing the head of BLS due to unsavory numbers would be a scandal on its own. Yet, we're seven months into this one with multiple corruption scandals flooding the zone. It just seems crazy that we're here justifying basic concepts of government/public service and not outright calling this out for what it is.
> but it all feels like we're sane washing this administration
This is not a defense of the admin or their actions, but if there's something we should have learned well over the past 100+ years of history, it's not to assume insanity on the part of people who disagree with you. It feels good to assume that we are so correct that anyone disagreeing with us must be insane, but it's a deeply unproductive (and often counter-productive) way to interact with people.
Personally, I think they think that places like the BLS are stacked with "deep state" people that are trying to sabotage the current administration. I think that's mostly absurd, but they don't, and without evidence either way it's a matter of opinion (I personally lean heavily on things like Hanlon's Razor and trying to gauge "likeliness" rather than assuming the best or worst). If you believe as they do, then cleaning house is not only good but necessary, so the actions aren't insane. If we don't try to (in good faith) understand their beliefs/motivations, and just assume they are just randomly pulling triggers, not only will we only further entrench partisan divides (nothing alienates somebody more than feeling they aren't being properly understood), but we hinder our own ability to predict and prepare for the future.
The prior comment is not using the definition of sane/insane as in: "the leadership of the White House is composed of people that are insane." Rather: "the decisions and policies set forth by the White House are insane." That's an important distinction.
To "sane-wash" is to take an extraordinary break from liberal order/american historical norms and treat such break as normal politics.
I completely agree that there are rational actors at the White House who have an incorrect sense of a Deep State at the BLS, and are taking what they would describe as a rational step and firing it's leader. But simply because they are rational does not mean that their subsequent decisions are rational.
If am a rational person and I have an incorrect belief that my house has cockroaches, then I burn down my house to get of the cockroaches, then the news media reports I "took bold new steps to rid my house of cockroaches," I have been sane washed.
I think you're giving them too much credit. I don't think they care if the numbers are legit. They care about the optics. They're happy to lie about what they believe if it fits the optics they want to project.
For something like this, all the data are there. They show their work. The administration could try to check and see if the numbers are correct, but they don't.
I'm not saying MAGA supporters are insane or that Trump himself is insane. I'm merely pointing out that the response to the extraordinary levels of corruption in this administration is insane along with the actions themselves considering what scandals plagued previous administrations.
You can try to steelman their view all you want, what we're seeing is bold-faced corruption: Trump coin crypto investor dinners, Trump mobile, receiving a $400M jet from Qatar, Tech executives donating to him in various ways to curry favor, and revoking security clearances from law firms who represent things he doesn't like. Just to name a few.
Luckily, we don't need to "try in good faith" to understand their motivations. They published a manifesto about it (Project 2025) and are systematically going down that list of sweeping changes. These changes don't happen in a bubble.
I think they think that government is inherently bad (with the exception of the military and certain law enforcement) and reducing it is automatically good. I think this because I heard Rush Limbaugh saying it for three hours a day every day for years. It’s not hard to figure out their motivations. Just listen to them. Understanding those motivations can be tough, due to a combination of very different values and a reliance on various facts that happen to be untrue, but you can at least figure out the first level pretty easily.
Unfortunately there are lots of people on the sidelines that watch them screaming "censorship!", "persecution!", "injustice!" and fall for it. Any act against them is met with whining and fake outrage, and it works politically.
You can see this in action even in here, a forum where most people are more well-educated than 90% of the global population, they still swallow those cries as if this administration is the victim.
What is even more amusing and weird is how much the same side likes to scream "stop being a victim" while continuously playing into victimising themselves.
The liberal can't argue with inauthentic memes. They must give the benefit of the doubt by their own rules. The conservative sees this as an exploit that they can use to win the game every time. Because they see it as a game, amoral, based on scoring symbolic points for their preexisting tribal identity ("This family has always rooted for Liverpool"). They see it as a Conflict, not even of competing interests, but an inherent conflict of identity. The liberal insists on seeing it as a disagreement between friends, or an opportunity for education. An earnest Mistake. All of their rhetorical tools are based on persuading someone reasoning in good faith.
Exactly, it drives me insane that people are still thinking this administration is in any way acting in good faith. The argument you quote above is assuming an awful lot of good faith when this administration continually lies and violates court orders.
Hell they lied about Trumps weight and height on his physical.
Exactly. Instead of battling with their own internal departments when they show lack-luster numbers in good faith, they can just blame the "agendas" of the non-blessed private sector companies who show the same data.
Even better, they can exult the virtues of select private sector companies who show "good" "approved" data.
What? No, no conflict of interest that members of the President's immediate family happens to hold board seats in those "approved" companies.
Should point out this is not the first time the BLS came under serious fire, politically. When the Republicans regained control of the Senate in 1995, they set up a commission (Boskin commission) that said inflation had been overstated, and henceforth cut Social Security cost of living adjustments.
Cuts will continue until the statistics improve and match the desired narrative! Wish that was a /s but it's just what's happening now and happened then.
(Great post/article!) I was so confused as to why you said counterpoint -- it is deep in the article, if at all.
> If the government’s jobs data is considered biased or unreliable, Wall Street will have other places to look. ADP reports figures on private payrolls, for instance. (And it also tells a bearish story: ADP showed a net loss of jobs in June, for instance.) Meanwhile, Gallup once tracked employment numbers on a weekly basis based on its large-scale surveys and could resume that effort. Or investment banks like Goldman Sachs might conclude they could have a competitive edge by tracking their own economic data.
But in the following paragraph he points out that the numbers aren't likely to be as accurate, comprehensive, or as reassuring as numbers generated by the government. Overall though it is a really good article worth a share of its own.
We have been seeing news reporting, media, and education private organizations knuckling under from pressure by the administration - is there any reason to believe that we would not see the same from data reporting companies
Fair point, but private organizations would at least have a financial incentive to give accurate numbers since that's they're business, if they earn a reputation of unreliability then there'd be no reason to pay them, you could just trust the government numbers for free.
There's significant evidence contradicting this hypothesis. See industry analysts like Gartner, IDC, etc. who all ask tech firms to "pay to play" for better rankings in their reports. As well as the ratings agencies like Moody's, S&P's, and Fitch during the 2008 financial crisis. These ratings agencies were paid by the banks selling CDOs, MBS, and other debt derivatives that were especially tied to the US housing market. The agencies were incentivized to not downgrade those products.
> See industry analysts like Gartner, IDC, etc. who all ask tech firms to "pay to play" for better rankings in their reports.
Do you have any evidence of this happening? I've been involved on the side of the firm in many of these analyst reports, and I've never seen this happen. What /does/ happen is that to be listed at all, e.g. to have the analyst invest time/resources to review you, you have to pay. They don't exhaustively review every company in the segment or every product in the segment, regardless of size and market position. They only review those who pay to be reviewed or are large enough in market position that they are must-reviews (almost all of which also pay). But I've never seen the rankings/outcome dictated by pay at all. Much to my chagrin sometimes, the analyst will lower your ranking due to something I may have felt was a small issue, but they considered a large issue. Analyst Relations is a major investment area for large tech companies, and given my exposure over the last nearly decade, I've never seen any form of quid pro quo / money changing hands.
>> Fair point, but private organizations would at least have a financial incentive to give accurate numbers since that's they're business
Private organizations will quickly start providing numbers that their customers want to see. Remember the risk ratings agencies prior to 2008? Yeah, finance guys want accurate risk assessments internally, but they always want good ratings on the stuff they're selling. Good ratings are a lubricant on transactions and since ratings are paid for by people making transactions there is more pressure in one direction.
You pay for the business that provides the numbers you want. Accuracy is desired only in setting a baseline to be below. Reducing social security is the desired outcome not accuracy.
> financial incentive to give accurate numbers since that's they're business
The Trump admin has already shown a complete willingness to exert financial pressure on companies and organizations to compel compliance and our hypothetical private BLS replacement is not magically insulated from those pressures and threats.
> you could just trust the government numbers for free.
They're clearly in the middle of an attempt to cook the numbers to make the President look good.
If those statistics are even available. Part of the employment numbers is a 60,000 household survey they conduct with the help of the census bureau I doubt there's going to be a consistent survey of that scale done on the private dime. Not to mention all the statistics businesses are required legally to submit to the government that these hypothetical private replacements would never have access to.
Or also, all firms will say what you want to hear, like modern ratings agencies already do. Because otherwise you won't ever shop at the ones who give you bad numbers, and market pressure will turn them all into a ratings shop.
What I really would like to see is a historical trend analysis seeing whether BLS and nongovernment estimates (public or private) start to decouple about now.
If you have enough sources you presumably at least can start to make observations about which sources agree and which disagree.
Actually I presume something like this already exists, but I'm not an economist so I don't know.
With this and SCOTUS neutering Humphrey’s Executor (with Federal Reserve penumbras -- for now) I'm wondering if there will be a push in the future to establish a 4th branch (call it the deep state branch) to protect what should be non-political stuff like the federal reserve and BLS statistics. Like the judiciary, nominated by President, confirmed by the senate, funded by congress, impeachable, limited executive authority, etc.
Interestingly, this is what the EU is. After WW2 it was decided to set things up so that European countries no longer could wage war against each other. Hence starting out with coal and steel (no war without those).
In such a setup, lack of democratic oversight is not a bug but a feature. It prevents democratically elected but possibly not entirely stable governments from doing something stupid. The drawback, of course, is that it also prevents democratically elected and intensely well meaning governments from doing the right thing if the supranational entity does not wish to facilitate that right thing.
This setup works (after a fashion) for a group of small and fairly weak countries, but for a single large and powerful country like the US there simply isn’t a supranational entity powerful enough to reign it in.
If we're launching a fourth branch of government, it should be explicitly aimed at enforcing the Constitution on the government. Having the judiciary try to do that and all the other adjudication is not working.
>I'm wondering if there will be a push in the future to establish a 4th branch (call it the deep state branch) to protect what should be non-political stuff like the federal reserve and BLS statistics.
Now you have four problems, as it were.
If the thesis is that the government can't be trusted to run its own affairs, then more government isn't a solution.
The actual "fourth branch" of government is supposed to be the people, either giving a shit and being engaged in their civic duty and not electing tyrants for entertainment value or to "own the libs," or else doing the other thing that Americans love to talk about doing.
We thought we already had that for a bunch of stuff. Turns out, words on paper don't matter if Congress says nothing, the Executive goes rampant, and the SCOTUS either stays suspiciously quiet or gives an approving nod and wink here and there.
Given the climate created by the TACO-ing over tariffs its not surprising that these numbers are inconsistent.
And Trump keeps rewriting history about the numbers impacting elections. The last number before elections were the Oct numbers released on 1-Nov-24. That was abysmal at 12k jobs. Trump even complained about in his campaign speeches. But never let truth stand in the way of Trump and his supporters.
don't you worry, people in the public sector who want everything to be private will make sure that the BLS becomes shit enough that the private alternative will end up being appealing, we've seen that millions of times before
With the shoot the messenger management style. Will the BLS ever be trustworthy again? What happens to markets when we can't trust any government statistics.
Maybe the best move at this point is to move toward the BLS releasing all the raw data, but do not publish conclusions of their own? Then at least you have a central "unbiased" (if that's ever an actual state) agency responsible for collection, and leave it up to the more polarized outlets/statisticians/... to produce their own interpretations.
E.g. similar to Census data -- central data collection managed by govt, generalized anonymous data released for interpretation by anyone.
The BLS gathers data to support its methodology. If the methodology were biased, having access to the raw data would not solve the problem. You could always claim "BLS is ignoring X".
Sure, but the alternative is implicitly trusting the conclusions and interpretations the govt agency, who are (almost assuredly) humans with biases and not impartial robots.
It's a lot harder to attack general best practice statistic collection than it is to throw shade at an editorial. (not that certain administrations have issues attacking best practice science)
Issue also arise "easily" when the people charged with producing editorials are political appointees who are easy targets for subsequent administrations looking to shift blame. (seems we're at this stage now)
Was on news yesterday, not sure if will happen. But Trump now also does not trust the Census and wants to do another one. So then what happens when we can't even trust the Census? And Totals are just gamed for Congressional seats.
Its not that he doesn't trust the census. It's that they want a census marking US citizens vs. immigrants and illegal immigrants. It doesn't matter what they say so much, more what the end goal is.
Voting districts with higher populations receive a larger number of representatives. Currently we do not differentiate and legally the census defines these numbers which is very hard to change. If a census is run with these identified you have strong legal arguments with the "Current" Supreme Court to remove those immigrants/illegal immigrants from the representative totals.
The side benefit for this admin is many legal immigrants and some residents become more untrustworthy of those style questions and also do not answer again lowering representation in those districts further.
The Census is inaccurate if the goal is legally removing those people from the representative counts.
Well, in this one tiny aspect, Trump may have a point. If we're determining voting districts, and non-citizens can't vote, then using only citizens to determine the size of the voting districts makes some sense.
But trying to do it now makes no rational sense, and neither does the Texas redistricting move. It's all about the 2026 election, and given how he's trying every possible angle, no matter how unprecedented, I think he knows he's in trouble.
Yeah we live in a crazy world. Honestly the fact that they don't trust it probably lends more credence toward their results -- it's extremely unfortunate that many of these agencies/systems/... live under the executive branch's whims and not as part of congressionally-mandated programs where it'd be more difficult to meddle.
The current 2-party system with fixed terms almost lends itself to promoting the political appointee system where entire department head counts turn over after a change in ruling party. If you're a "career" politician it's a no-brainer to fill these positions wholesale with people who are favorable to your world view, or donated to your administration, or spend lots of money at your hotel chain, ...
Sounds like he doesn't want non-citizens to be counted. I'd rather they get counted, but maybe not used for determining seats in congress. It's kind of important to know how many people are present in an area and how that changes over time. Deporting illegals (like it or dislike it) is a separate thing from knowing how many people are in a state. Come to think of it, even if you're in favor of the deportations, you gotta have accurate data to know where to deploy resources. This desire to fake the data is just bizarre.
Yes, if the executive branch is trustworthy. We need to send a clear signal to the Republican Party that this sort of general behavior is not acceptable to Americans.
See, what they really need to do is release the politically-desired results in press releases while sending the real information using steganographic concealment in the page's logo. We must construct a Truman Show around Dear Leader for our own survival. /s
Previous users of this data will seek alternative, less precise signals elsewhere and likely use conservative, pessimistic guesstimates that will tend to be risk-adverse and not encourage expansion or investment.
I mean, they already don't. Investors aren't stupid. Half of financial analysis is converting financial metrics from GAAP to other more meaningful structures that allow for proper forecasting and comparison. They spend billions on alternative data, trying to get an edge. They take nothing at face value.
I think the government stuff is a bit of a farce. They have an incentive like any other institution. For instance, inflation metrics are tied to trillions of dollars in terms of retirement benefits or other programs that are indexed to inflation. They have a gigantic incentive to play with the numbers. And I'm 100% convinced that they do. Does anyone believe food prices went up only 28% since pre-covid? Did something go from costing $4 to $5? No, most things close to doubled. But they're able to play with the numbers changing the basket (beef getting more expensive, we'll just swap that out with Tofu!).
The basket is determined by what people actually buy, rather than what they would like to buy. And they are about two years behind, so the current basket is based on surveys done in 2023.
So based on your experience, your food budget went up about 28% from pre-covid?
They have a lot of discretion, it's not formulaic. If it was, ironically it would be great to privatize! Here are the specs, and provide proof. Easy.
Here is from the world bank talking about tight and loose specifications:
> The CPI Manual refers to tight and loose specifications of product or item varieties. Tight specifications better facilitate the calculation of average prices and ensure comparability of items across countries, which is the aim of ICP price collection. Accordingly, the ICP stipulates strict requirements for the selection of items for pricing. These may include restricting the choice of brand or the size of an item by weight, volume, or quantity for example. For the CPI, tight specifications tend to be useful for electronic and other items with high rates of turnover.
Again, look at the facts:
1. It's complicated and not easily observable
2. There is some discretion
3. Politics and trillions in government spending is on stake
That’s the point. When Trump disagrees with facts the facts must be destroyed. When people are actually trying to solve problems they desire more information, not less.
I think this pattern of behavior needs to be cast as incompetence and cowardice rather than immorality. Avoiding transparency, rigorous analysis, and competitive disagreement and review is a sign that one cannot make a convincing argument, or does not have courage to do so. I don't think they care about the morality of it, or see themselves as morally justified.
Everyone needs trustworthy employment data as this affects every single business in the country. If you don't know what the real employment rate is and you see your business going down you might think the problem is your business and not the whole economy.
You can't run a modern country without collecting real world data about the country's economy.
The data is backward looking, sometimes by as much as 3 months, rarely less than a month old. The only way the "data" knows there is a reccession is if it stared 3/4 of a year ago.
If you are a CEO or running a business and you are blaming your business decisions on the stats from 3 months ago that came out today you should get out of the chair and let someone with a brain sit in it.
> Private companies are in the business of making a profit, and to do that they need to attract customers. Some customers are more profitable than others. As a result, a company’s data will tend to reflect the needs of its clients — it won’t capture the economy as whole.
Honestly, it is a very, very good paragraph. There is a slavish devotion to the idea that markets are efficient and that is the most desirable outcome in all situations. If you are in the business of being a government and running a country for all of your citizens, you need all the data, not just select aspects of it, and not just those that tell a particular story.
There is an infinite amount of data. And most of it is useless. Governments do not need to know how many fry cooks read exactly 3 novels from authors who names end with a letter in the first half of the alphabet in the last year.
So that is why it is important to understand that the government needs data for it's purposes (and not more than that). Restricting data to the narrow purposes of whichever company is willing to pay to produce select economic statistics will leave everyone worse off.
>Restricting data to the narrow purposes of whichever company
This isn't what the article said. It essentially said the data would be based off what people who use the data need. If the government needs some data it would still be able to get it.
Does the Fed use BLS data for that or BEA employment data (BEA does use the BLS data as an input, so they aren't completely independent, but they also make adjustments to make them consistent with other BEA measures and I wonder if changes in BLS methodology or reliability will simply produce greater divergence between BLS and BEA employment information).
I mean, the Fed uses BEA PCE data and not BLS CPI data for inflation, so it would seem consistent for them to use BEA employment data as well.
Of course. if manipulating the source of the popular headline figures doesn't have the desired impact on the Fed, Trump would eventually probably exert the same pressure on BEA.
Karl Marx used the British government Blue Books he found in the Reading Room of the British Museum to write Capital. If the British government wasn't publishing Blue Books in the 19th century, he would have had more difficulty forming his theories.
> Instead of firing the commissioner, the president should be giving the agency a raise in the form of a bigger budget. The goal should be to restore public trust in government statistics, not undermine it.
When you believe that government should not be providing many services, or doing most of what it currently does overall, why would you want to bolster trust in government statistics? Those statistics might contradict the administration, which is not a goal of the administration and its backers.
I understand why articles like this are written and why we have these conversations but it all feels like we're sane washing this administration. Any other administration (especially a Democrat) firing the head of BLS due to unsavory numbers would be a scandal on its own. Yet, we're seven months into this one with multiple corruption scandals flooding the zone. It just seems crazy that we're here justifying basic concepts of government/public service and not outright calling this out for what it is.
> but it all feels like we're sane washing this administration
This is not a defense of the admin or their actions, but if there's something we should have learned well over the past 100+ years of history, it's not to assume insanity on the part of people who disagree with you. It feels good to assume that we are so correct that anyone disagreeing with us must be insane, but it's a deeply unproductive (and often counter-productive) way to interact with people.
Personally, I think they think that places like the BLS are stacked with "deep state" people that are trying to sabotage the current administration. I think that's mostly absurd, but they don't, and without evidence either way it's a matter of opinion (I personally lean heavily on things like Hanlon's Razor and trying to gauge "likeliness" rather than assuming the best or worst). If you believe as they do, then cleaning house is not only good but necessary, so the actions aren't insane. If we don't try to (in good faith) understand their beliefs/motivations, and just assume they are just randomly pulling triggers, not only will we only further entrench partisan divides (nothing alienates somebody more than feeling they aren't being properly understood), but we hinder our own ability to predict and prepare for the future.
The prior comment is not using the definition of sane/insane as in: "the leadership of the White House is composed of people that are insane." Rather: "the decisions and policies set forth by the White House are insane." That's an important distinction.
To "sane-wash" is to take an extraordinary break from liberal order/american historical norms and treat such break as normal politics.
I completely agree that there are rational actors at the White House who have an incorrect sense of a Deep State at the BLS, and are taking what they would describe as a rational step and firing it's leader. But simply because they are rational does not mean that their subsequent decisions are rational.
If am a rational person and I have an incorrect belief that my house has cockroaches, then I burn down my house to get of the cockroaches, then the news media reports I "took bold new steps to rid my house of cockroaches," I have been sane washed.
I think you're giving them too much credit. I don't think they care if the numbers are legit. They care about the optics. They're happy to lie about what they believe if it fits the optics they want to project.
For something like this, all the data are there. They show their work. The administration could try to check and see if the numbers are correct, but they don't.
I'm not saying MAGA supporters are insane or that Trump himself is insane. I'm merely pointing out that the response to the extraordinary levels of corruption in this administration is insane along with the actions themselves considering what scandals plagued previous administrations.
You can try to steelman their view all you want, what we're seeing is bold-faced corruption: Trump coin crypto investor dinners, Trump mobile, receiving a $400M jet from Qatar, Tech executives donating to him in various ways to curry favor, and revoking security clearances from law firms who represent things he doesn't like. Just to name a few.
Thanks, with that clarification, we are actually in perfect agreement
Luckily, we don't need to "try in good faith" to understand their motivations. They published a manifesto about it (Project 2025) and are systematically going down that list of sweeping changes. These changes don't happen in a bubble.
I think they think that government is inherently bad (with the exception of the military and certain law enforcement) and reducing it is automatically good. I think this because I heard Rush Limbaugh saying it for three hours a day every day for years. It’s not hard to figure out their motivations. Just listen to them. Understanding those motivations can be tough, due to a combination of very different values and a reliance on various facts that happen to be untrue, but you can at least figure out the first level pretty easily.
Right. That's a good deal of what gets them away with it. You have to crush this stuff while it's small.
Unfortunately there are lots of people on the sidelines that watch them screaming "censorship!", "persecution!", "injustice!" and fall for it. Any act against them is met with whining and fake outrage, and it works politically.
You can see this in action even in here, a forum where most people are more well-educated than 90% of the global population, they still swallow those cries as if this administration is the victim.
What is even more amusing and weird is how much the same side likes to scream "stop being a victim" while continuously playing into victimising themselves.
"The card says moops"
https://www.youtube.com/watch?v=xMabpBvtXr4
The liberal can't argue with inauthentic memes. They must give the benefit of the doubt by their own rules. The conservative sees this as an exploit that they can use to win the game every time. Because they see it as a game, amoral, based on scoring symbolic points for their preexisting tribal identity ("This family has always rooted for Liverpool"). They see it as a Conflict, not even of competing interests, but an inherent conflict of identity. The liberal insists on seeing it as a disagreement between friends, or an opportunity for education. An earnest Mistake. All of their rhetorical tools are based on persuading someone reasoning in good faith.
"Liberal" means refusing smugly on principle to even try to fight, then crying about how unfair it is to keep catching all these ass kickings.
Nobody likes a loser. So if you want people to like your politics, don't be a loser.
Exactly, it drives me insane that people are still thinking this administration is in any way acting in good faith. The argument you quote above is assuming an awful lot of good faith when this administration continually lies and violates court orders.
Hell they lied about Trumps weight and height on his physical.
Exactly. Instead of battling with their own internal departments when they show lack-luster numbers in good faith, they can just blame the "agendas" of the non-blessed private sector companies who show the same data.
Even better, they can exult the virtues of select private sector companies who show "good" "approved" data.
What? No, no conflict of interest that members of the President's immediate family happens to hold board seats in those "approved" companies.
Should point out this is not the first time the BLS came under serious fire, politically. When the Republicans regained control of the Senate in 1995, they set up a commission (Boskin commission) that said inflation had been overstated, and henceforth cut Social Security cost of living adjustments.
Cuts will continue until the statistics improve and match the desired narrative! Wish that was a /s but it's just what's happening now and happened then.
Counterpoint: https://www.natesilver.net/p/trumps-jobs-data-denialism-wont...
(Great post/article!) I was so confused as to why you said counterpoint -- it is deep in the article, if at all.
> If the government’s jobs data is considered biased or unreliable, Wall Street will have other places to look. ADP reports figures on private payrolls, for instance. (And it also tells a bearish story: ADP showed a net loss of jobs in June, for instance.) Meanwhile, Gallup once tracked employment numbers on a weekly basis based on its large-scale surveys and could resume that effort. Or investment banks like Goldman Sachs might conclude they could have a competitive edge by tracking their own economic data.
But in the following paragraph he points out that the numbers aren't likely to be as accurate, comprehensive, or as reassuring as numbers generated by the government. Overall though it is a really good article worth a share of its own.
Doesn't seem to be a counterpoint?
Presumably private firms will become a necessary component of economic projections since BLS statistics will be less trustworthy going forward.
We have been seeing news reporting, media, and education private organizations knuckling under from pressure by the administration - is there any reason to believe that we would not see the same from data reporting companies
Fair point, but private organizations would at least have a financial incentive to give accurate numbers since that's they're business, if they earn a reputation of unreliability then there'd be no reason to pay them, you could just trust the government numbers for free.
There's significant evidence contradicting this hypothesis. See industry analysts like Gartner, IDC, etc. who all ask tech firms to "pay to play" for better rankings in their reports. As well as the ratings agencies like Moody's, S&P's, and Fitch during the 2008 financial crisis. These ratings agencies were paid by the banks selling CDOs, MBS, and other debt derivatives that were especially tied to the US housing market. The agencies were incentivized to not downgrade those products.
https://en.wikipedia.org/wiki/Credit_rating_agencies_and_the...
> See industry analysts like Gartner, IDC, etc. who all ask tech firms to "pay to play" for better rankings in their reports.
Do you have any evidence of this happening? I've been involved on the side of the firm in many of these analyst reports, and I've never seen this happen. What /does/ happen is that to be listed at all, e.g. to have the analyst invest time/resources to review you, you have to pay. They don't exhaustively review every company in the segment or every product in the segment, regardless of size and market position. They only review those who pay to be reviewed or are large enough in market position that they are must-reviews (almost all of which also pay). But I've never seen the rankings/outcome dictated by pay at all. Much to my chagrin sometimes, the analyst will lower your ranking due to something I may have felt was a small issue, but they considered a large issue. Analyst Relations is a major investment area for large tech companies, and given my exposure over the last nearly decade, I've never seen any form of quid pro quo / money changing hands.
This makes sense. Thanks for the info.
>> Fair point, but private organizations would at least have a financial incentive to give accurate numbers since that's they're business
Private organizations will quickly start providing numbers that their customers want to see. Remember the risk ratings agencies prior to 2008? Yeah, finance guys want accurate risk assessments internally, but they always want good ratings on the stuff they're selling. Good ratings are a lubricant on transactions and since ratings are paid for by people making transactions there is more pressure in one direction.
You pay for the business that provides the numbers you want. Accuracy is desired only in setting a baseline to be below. Reducing social security is the desired outcome not accuracy.
> financial incentive to give accurate numbers since that's they're business
The Trump admin has already shown a complete willingness to exert financial pressure on companies and organizations to compel compliance and our hypothetical private BLS replacement is not magically insulated from those pressures and threats.
> you could just trust the government numbers for free.
They're clearly in the middle of an attempt to cook the numbers to make the President look good.
Love it. You can then just pick the numbers from the firm that is saying what you want to hear.
Yeah, it sucks for the public, but I assume if you're working in FP&A your company will pay for accurate statistics.
If those statistics are even available. Part of the employment numbers is a 60,000 household survey they conduct with the help of the census bureau I doubt there's going to be a consistent survey of that scale done on the private dime. Not to mention all the statistics businesses are required legally to submit to the government that these hypothetical private replacements would never have access to.
Wait, is that what credit rating agencies are? But credit rating agencies also need data from somewhere...
Or also, all firms will say what you want to hear, like modern ratings agencies already do. Because otherwise you won't ever shop at the ones who give you bad numbers, and market pressure will turn them all into a ratings shop.
"Bad" numbers are what companies are paying for.
If I'm a sailor paying for weather data related to my trip, I want to know about the dangerous weather.
What I really would like to see is a historical trend analysis seeing whether BLS and nongovernment estimates (public or private) start to decouple about now.
If you have enough sources you presumably at least can start to make observations about which sources agree and which disagree.
Actually I presume something like this already exists, but I'm not an economist so I don't know.
It will turn out like LIBOR and they will provide numbers that are producing profit for them.
https://archive.ph/el7mM
With this and SCOTUS neutering Humphrey’s Executor (with Federal Reserve penumbras -- for now) I'm wondering if there will be a push in the future to establish a 4th branch (call it the deep state branch) to protect what should be non-political stuff like the federal reserve and BLS statistics. Like the judiciary, nominated by President, confirmed by the senate, funded by congress, impeachable, limited executive authority, etc.
Interestingly, this is what the EU is. After WW2 it was decided to set things up so that European countries no longer could wage war against each other. Hence starting out with coal and steel (no war without those).
In such a setup, lack of democratic oversight is not a bug but a feature. It prevents democratically elected but possibly not entirely stable governments from doing something stupid. The drawback, of course, is that it also prevents democratically elected and intensely well meaning governments from doing the right thing if the supranational entity does not wish to facilitate that right thing.
This setup works (after a fashion) for a group of small and fairly weak countries, but for a single large and powerful country like the US there simply isn’t a supranational entity powerful enough to reign it in.
If we're launching a fourth branch of government, it should be explicitly aimed at enforcing the Constitution on the government. Having the judiciary try to do that and all the other adjudication is not working.
>I'm wondering if there will be a push in the future to establish a 4th branch (call it the deep state branch) to protect what should be non-political stuff like the federal reserve and BLS statistics.
Now you have four problems, as it were.
If the thesis is that the government can't be trusted to run its own affairs, then more government isn't a solution.
The actual "fourth branch" of government is supposed to be the people, either giving a shit and being engaged in their civic duty and not electing tyrants for entertainment value or to "own the libs," or else doing the other thing that Americans love to talk about doing.
We thought we already had that for a bunch of stuff. Turns out, words on paper don't matter if Congress says nothing, the Executive goes rampant, and the SCOTUS either stays suspiciously quiet or gives an approving nod and wink here and there.
Nate Silver has a piece on this: https://www.natesilver.net/p/trumps-jobs-data-denialism-wont...
The point being that revisions can be huge.
Given the climate created by the TACO-ing over tariffs its not surprising that these numbers are inconsistent.
And Trump keeps rewriting history about the numbers impacting elections. The last number before elections were the Oct numbers released on 1-Nov-24. That was abysmal at 12k jobs. Trump even complained about in his campaign speeches. But never let truth stand in the way of Trump and his supporters.
https://www.bls.gov/news.release/archives/empsit_11012024.ht...
don't you worry, people in the public sector who want everything to be private will make sure that the BLS becomes shit enough that the private alternative will end up being appealing, we've seen that millions of times before
With the shoot the messenger management style. Will the BLS ever be trustworthy again? What happens to markets when we can't trust any government statistics.
Maybe the best move at this point is to move toward the BLS releasing all the raw data, but do not publish conclusions of their own? Then at least you have a central "unbiased" (if that's ever an actual state) agency responsible for collection, and leave it up to the more polarized outlets/statisticians/... to produce their own interpretations.
E.g. similar to Census data -- central data collection managed by govt, generalized anonymous data released for interpretation by anyone.
The BLS gathers data to support its methodology. If the methodology were biased, having access to the raw data would not solve the problem. You could always claim "BLS is ignoring X".
BLS already does this. It's not the most friendly interface, but you can directly query the database from its own website here: https://data.bls.gov/dataQuery/search. There's also a public API: https://www.bls.gov/developers/home.htm.
That opens the door for lots of Wrong conclusions either lack of knowledge and maliciousness.
Statistics is often counterintuitive.
See also: The Vaccine Adverse Event Reporting System (VAERS)
Sure, but the alternative is implicitly trusting the conclusions and interpretations the govt agency, who are (almost assuredly) humans with biases and not impartial robots.
It's a lot harder to attack general best practice statistic collection than it is to throw shade at an editorial. (not that certain administrations have issues attacking best practice science)
Issue also arise "easily" when the people charged with producing editorials are political appointees who are easy targets for subsequent administrations looking to shift blame. (seems we're at this stage now)
Except that they already do publish all the data, and their methods.
Was on news yesterday, not sure if will happen. But Trump now also does not trust the Census and wants to do another one. So then what happens when we can't even trust the Census? And Totals are just gamed for Congressional seats.
Its not that he doesn't trust the census. It's that they want a census marking US citizens vs. immigrants and illegal immigrants. It doesn't matter what they say so much, more what the end goal is.
Voting districts with higher populations receive a larger number of representatives. Currently we do not differentiate and legally the census defines these numbers which is very hard to change. If a census is run with these identified you have strong legal arguments with the "Current" Supreme Court to remove those immigrants/illegal immigrants from the representative totals.
The side benefit for this admin is many legal immigrants and some residents become more untrustworthy of those style questions and also do not answer again lowering representation in those districts further.
The Census is inaccurate if the goal is legally removing those people from the representative counts.
Well, in this one tiny aspect, Trump may have a point. If we're determining voting districts, and non-citizens can't vote, then using only citizens to determine the size of the voting districts makes some sense.
But trying to do it now makes no rational sense, and neither does the Texas redistricting move. It's all about the 2026 election, and given how he's trying every possible angle, no matter how unprecedented, I think he knows he's in trouble.
Yeah we live in a crazy world. Honestly the fact that they don't trust it probably lends more credence toward their results -- it's extremely unfortunate that many of these agencies/systems/... live under the executive branch's whims and not as part of congressionally-mandated programs where it'd be more difficult to meddle.
The current 2-party system with fixed terms almost lends itself to promoting the political appointee system where entire department head counts turn over after a change in ruling party. If you're a "career" politician it's a no-brainer to fill these positions wholesale with people who are favorable to your world view, or donated to your administration, or spend lots of money at your hotel chain, ...
Sounds like he doesn't want non-citizens to be counted. I'd rather they get counted, but maybe not used for determining seats in congress. It's kind of important to know how many people are present in an area and how that changes over time. Deporting illegals (like it or dislike it) is a separate thing from knowing how many people are in a state. Come to think of it, even if you're in favor of the deportations, you gotta have accurate data to know where to deploy resources. This desire to fake the data is just bizarre.
Yes, if the executive branch is trustworthy. We need to send a clear signal to the Republican Party that this sort of general behavior is not acceptable to Americans.
They don’t care
Sadly, I think you’re right. And a nation of citizens who don’t care about good government is a nation that is in collapse.
See, what they really need to do is release the politically-desired results in press releases while sending the real information using steganographic concealment in the page's logo. We must construct a Truman Show around Dear Leader for our own survival. /s
Previous users of this data will seek alternative, less precise signals elsewhere and likely use conservative, pessimistic guesstimates that will tend to be risk-adverse and not encourage expansion or investment.
I mean, they already don't. Investors aren't stupid. Half of financial analysis is converting financial metrics from GAAP to other more meaningful structures that allow for proper forecasting and comparison. They spend billions on alternative data, trying to get an edge. They take nothing at face value.
I think the government stuff is a bit of a farce. They have an incentive like any other institution. For instance, inflation metrics are tied to trillions of dollars in terms of retirement benefits or other programs that are indexed to inflation. They have a gigantic incentive to play with the numbers. And I'm 100% convinced that they do. Does anyone believe food prices went up only 28% since pre-covid? Did something go from costing $4 to $5? No, most things close to doubled. But they're able to play with the numbers changing the basket (beef getting more expensive, we'll just swap that out with Tofu!).
The basket is determined by what people actually buy, rather than what they would like to buy. And they are about two years behind, so the current basket is based on surveys done in 2023.
So based on your experience, your food budget went up about 28% from pre-covid?
They have a lot of discretion, it's not formulaic. If it was, ironically it would be great to privatize! Here are the specs, and provide proof. Easy.
Here is from the world bank talking about tight and loose specifications:
> The CPI Manual refers to tight and loose specifications of product or item varieties. Tight specifications better facilitate the calculation of average prices and ensure comparability of items across countries, which is the aim of ICP price collection. Accordingly, the ICP stipulates strict requirements for the selection of items for pricing. These may include restricting the choice of brand or the size of an item by weight, volume, or quantity for example. For the CPI, tight specifications tend to be useful for electronic and other items with high rates of turnover.
Again, look at the facts:
1. It's complicated and not easily observable
2. There is some discretion
3. Politics and trillions in government spending is on stake
4. The numbers don't match personal experience
https://thedocs.worldbank.org/en/doc/2b29c1445d7fa006e5f4ca0...
Everybody's personal experience will be different. I can't speak to yours.
What?
That’s the point. When Trump disagrees with facts the facts must be destroyed. When people are actually trying to solve problems they desire more information, not less.
I think this pattern of behavior needs to be cast as incompetence and cowardice rather than immorality. Avoiding transparency, rigorous analysis, and competitive disagreement and review is a sign that one cannot make a convincing argument, or does not have courage to do so. I don't think they care about the morality of it, or see themselves as morally justified.
> incompetence and cowardice rather than immorality
Why not all three?
Lol. I do think it's all three, I just think the last doesn't really resonate with some people.
The problem is the BLS was putting out wildly incorrect facts
Yes, they had that bias towards reality that authoritarians can’t stand.
They put out estimations and numbers based on reported data.
When the reported data changed they changed their data too.
That‘s not putting out wildly incorrect facts that’s putting out data based on currently known facts
That claim has never been backed up with evidence.
[dead]
[dead]
>As a result, a company’s data will tend to reflect the needs of its clients — it won’t capture the economy as whole.
If no one needs some data doesn't that imply capturing it is a waste of time and energy?
Everyone needs trustworthy employment data as this affects every single business in the country. If you don't know what the real employment rate is and you see your business going down you might think the problem is your business and not the whole economy.
You can't run a modern country without collecting real world data about the country's economy.
> real world data about the country's economy
The data is backward looking, sometimes by as much as 3 months, rarely less than a month old. The only way the "data" knows there is a reccession is if it stared 3/4 of a year ago.
If you are a CEO or running a business and you are blaming your business decisions on the stats from 3 months ago that came out today you should get out of the chair and let someone with a brain sit in it.
how would that change your actions? Would it make you feel better about your business failing?
If employment is strong, you might add marketing, because maybe you're not doing enough to get the money that is clearly being passed around.
If employment is weak, you might cut staff yourself, or maybe cut costs elsewhere.
You might change your product mix to serve a different set of customers. You could change pricing.
No. It is not about feeling better.
The whole paragraph explains it:
> Private companies are in the business of making a profit, and to do that they need to attract customers. Some customers are more profitable than others. As a result, a company’s data will tend to reflect the needs of its clients — it won’t capture the economy as whole.
Honestly, it is a very, very good paragraph. There is a slavish devotion to the idea that markets are efficient and that is the most desirable outcome in all situations. If you are in the business of being a government and running a country for all of your citizens, you need all the data, not just select aspects of it, and not just those that tell a particular story.
>you need all the data
There is an infinite amount of data. And most of it is useless. Governments do not need to know how many fry cooks read exactly 3 novels from authors who names end with a letter in the first half of the alphabet in the last year.
Reductio ad absurdum, sure.
So that is why it is important to understand that the government needs data for it's purposes (and not more than that). Restricting data to the narrow purposes of whichever company is willing to pay to produce select economic statistics will leave everyone worse off.
>Restricting data to the narrow purposes of whichever company
This isn't what the article said. It essentially said the data would be based off what people who use the data need. If the government needs some data it would still be able to get it.
There is so much missing in what you're responding with, and I'm not even sure you're contradicting what's being said.
Lets just start with something simple: Are you assuming that the government only needs what people are willing to pay for?
The federal reserve certainly needs that data when they are trying to decide on interest rate adjustments.
Does the Fed use BLS data for that or BEA employment data (BEA does use the BLS data as an input, so they aren't completely independent, but they also make adjustments to make them consistent with other BEA measures and I wonder if changes in BLS methodology or reliability will simply produce greater divergence between BLS and BEA employment information).
I mean, the Fed uses BEA PCE data and not BLS CPI data for inflation, so it would seem consistent for them to use BEA employment data as well.
Of course. if manipulating the source of the popular headline figures doesn't have the desired impact on the Fed, Trump would eventually probably exert the same pressure on BEA.
Soon they won't need any data, they can just ask Trump what the rate should be.
People who are not the clients could need accurate information about how the total market is doing?
Karl Marx used the British government Blue Books he found in the Reading Room of the British Museum to write Capital. If the British government wasn't publishing Blue Books in the 19th century, he would have had more difficulty forming his theories.