Looky, there's a couple of considerations here.
Jackus - the easy answer is that your in-house standards should be selected from material that meets all the requirements of the relevant pharmacopoeia, and from batches that are as fresh as possible when your company "packs them down" into individual vials for in-house standards. The only thing to beware of is trying to use them beyond the approved shelf life for that manufacturer's raw material, as given in their Drug Master File (DMF).
Bruce - I have a friend who used to be very involved with the production of the BP reference standards, so I have some idea of the work that goes into verifying that they are fit for use when sold.
I don't think you are right to call them a major source of revenue - they recover the costs, mostly. Standards have to be tested about 6-8 times more heavily than usually. It is down to achieving results that can be statistically analysed, you need that much data before you can apply a T test and get significant results.
I do think that the USP people are "control freaks" when it comes to specifications.
There was a joint European-US committee on raw materials which was like watching paint dry for years - they didn't come to any conclusions. Then some bright spark said "let's find the points we can agree on", and suddenly there were a few USP materials which would also meet EP specification without comparison of "the books" - but not the other way round. Why?
The USP people insisted on tests for things not covered in the EP - like checking for solvents not used in manufacture "in case they had been spilled on the material in transit". And they are still insisting on stuff like that.
So you can buy a lot of materials to EP spec, and they are as good as USP materials. But if you want to use them in the US, you have to do additional tests, over the certified values.
There's a logic behind the matter - if you are obsessive. For myself, I do not see the logic of testing a consignment of Magnesium Stearate from Europe for chloroform content, just in case someone spilled chloroform on it in transit. But that's how far the "audit trail" extends these days.
In the same vein, there are substances in the USP standards list that are never available. Just try to buy some Mitomycin C standard ... most requests have been going down a black hole since around 1980. No explanations ...
If you do get some, it won't come in a standard USP vial. They can peel the Japanese label off, but if they open the vial they have to work on it ...
Control Freaks don't have to explain themselves ...
Well, Bruce, I've got away with using non-compendial standards to release products. I could probably do it in this day and age, even with the extra paperwork. But there's not much specifically written on how to do it in pharmaceuticals.
Standards, as you say, have to be appropriately prepared, identified, tested, approved, and stored (LOL I used to work with the man that wrote these paragraphs - he used to be The Inspector, and later was a consultant for the company I worked for).
The pharmaceutical industry was not "first in the field" on this. I learned the "how-to" when preparing secondary standards in an aluminium foundry about 35 years ago. The statistics for verifying the testing are laid down in general standards applicable in all civilised places. They don't say pharmaceuticals in the title ...
Identification - the compendial tests are usually adequate.
Preparation - you do your best to pack your "divisions" under uniform conditions, and always perform final testing on the final packed product,
sampling from beginning, middle and end of the packing run.
Storage - the authorities approved the DMF, they won't argue with the conditions it gives if you exceed them for precision of control (eg. 20 degC incubator is usually OK for a lot of things).
All the big companies have had procedures in place for this sort of work for a long time - that's part of how new products get onto the market. Ten years at least before they come off patent? Think about it...