I worked at an actuarial firm for a few years, and passed the exams required to be an associate actuary in the US. However, anyone can claim anything on the internet, and I shouldn't and can't claim to be all knowing about actuarial accounting, but I do think I can claim the following:
1) In order to account for something, you have to be able to measure it. Promising a population healthcare until they die requires you to know how much healthcare they will need and how much healthcare will cost, and since this is until they die, you need to know these numbers 30 to 70 years into the future.
I don't think anyone is capable of seeing that far into the future with any kind of reasonable certainty, especially not with healthcare. Therefore, it can't even be possible to "properly" account for this kind of promise, and consider the fact that pretty much every other non taxpayer funded organization has slashed and removed post retirement obligations due to skyrocketing healthcare costs and the inability to fund them.
2) According to the text of the legislation
https://www.govtrack.us/congress/bills/109/hr6407/text
Sec. 8909a. Postal Service Retiree Health Benefit Fund
The law requires funding of accrued benefits. I don't think it's unreasonable to require someone to set aside funds for something that has already been promised. In fact, I think the opposite is ridiculous, which is the way most government agencies work, and the USPS has worked with their pay as you go nonsense:
https://about.usps.com/who-we-are/financials/annual-reports/...
Basically, "pay" someone whatever you want now, and then actually pay for it later by raising taxes is what that means to me. I also consider it buying votes from the union members without having to increase current taxes. However, the fact that this type of law only went into effect for USPS does make me think some lobbyist of Fedex/UPS did a fantastic job.
3) The hundreds of billions of dollars of unfunded healthcare and defined benefit pensions from governments are evidence enough that whatever is going on in the benefit accounting world is not right. I know from sitting in on meetings that the people smart enough to calculate the liabilities aren't actually on the board, that's usually reserved for the uneducated policeman or firefighters and city council members who are exchanging union votes for benefits. Then they tell the actuarial what % return on investment to use to make the liabilities look smaller. Then the actuary puts a little notice in their report that says they don't agree with the unreasonable 8% return on investment assumption, but what difference does that make?
Also, in the pension consulting world, having funding for only 80% of the accrued liabilities is considered "fully funded". That is laughable. Look up what your city/county/state's funding ratios are.
In general, I simply point out the fact that no one can predict, with any reasonable accuracy, that far out into the future for something so complicated (lifespans, healthcare, investment returns, etc), so you can't really account for the liabilities in the first place. And from the fact that pretty much every single defined benefit pension that private organizations offered is no longer offered (due to strict funding requirements from ERISA and PPA laws, conveniently exempting taxpayer funded organizations), and the taxpayer funded ones that are very underfunded, you can surmise that the whole thing doesn't work. Too much money at play, too many ephemeral decision makers at the table who can pass the buck and reap the rewards, which all adds up to too much agency risk. Which, funnily enough, is never taken into account with these long term obligations.
How does this effect you in day to day life? Enjoy the deteriorating subway service, school systems, roadways, water quality, air quality, etc. as all that increased tax money goes towards these unfunded benefits for retired people.