All legislation, regulations and bureaucratic procedures have negative externalities just by virtue of existing. Each little piece makes it just that much harder for someone to find what they're looking for, or makes a process take a little bit more time, or makes it take a little bit more effort to learn the entire thing.
Individually, the marginal cost of each little piece is small, and they're generally introduced for good reasons, but the end result of bloated legal codes, regulatory frameworks and "red tape" is undeniable. For example, "ignorance of the law is no defence", but how many people know the entire legal code? (Answer: nobody)
But what they're doing here is attempting to quantify the externalities, and crucially: taking into account that a tiny cost spread among a large number of people can produce a substantial cost in absolute terms. People are bad at doing this intuitively due to scope insensitivity, and the only real solution is to estimate actual numbers for the number of people affected, the amount of time it costs, and the value of their time, and then multiply them together to come up with a hard number.
And that's exactly what's been done here. You could perhaps quibble about the exact figures, and there are probably other more intangible and difficult to quantify externalities that haven't been taken into consideration, but other than that it's essentially a flawless application of utilitarian ethics.
Frankly, I find that utterly astonishing.
I am guessing that doing this sort of analysis is required by some sort of rules or regulation, and I would love to know when this was introduced, by whom it was introduced, and at what scope this requirement exists. Is this now a requirement for all changes to rules affecting the public by US Federal agencies?
While the fact that the paperwork Reduction Act created this new bureaucratic step is amusing, it served to rectify the earlier situation where a lot of federal paperwork was unreasonably complicated just because it was nobody's job to make it otherwise.