E.g. similar to Census data -- central data collection managed by govt, generalized anonymous data released for interpretation by anyone.
Statistics is often counterintuitive.
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)
Guess biggest worry would be if BLS actually starts keeping secrets and just publishing numbers they want.
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.
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.
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, ...
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 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!).
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...