> There's an assumption there that its not possible to design and build things that can be adapative and function across a much longer timespan than a modern big box store's typical lifetime.
But that assumption is, by and large, true. If you look at hundred year old buildings, by modern standards they're extremely flawed. The insulation is typically poor or non-existent. They often had higher ceilings due to the limitations of contemporary lighting and HVAC systems, which are now a liability with respect to heating and cooling.
There is little reason to suspect that progress in building design has been discontinued, so a building from today would be expected to be similarly deficient by the standards of a hundred years from now, which means there is little reason to design it to last that long.
But the biggest flaw is often that the building is the wrong size. Decades ago you needed five stories and now you need ten. Or you needed five stories and now you need one. The obvious solution to that is to move to a bigger or smaller building as the need arises.
> There's also an assumption that the cost of have an imperfectly adapated building is higher than the cost of multiple building efforts over time.
Not at all, because the replacement building doesn't have to be on the same lot. You don't have to knock it down to build a new one consistent with your current needs. You can build a new one and then sell the existing one to someone who may be able to make productive use of it as it is. But then it's not the state reaping the benefits of any unusual longevity.
And that assumption may still be true in many cases regardless. If you have a five story building which over time comes to be surrounded by skyscrapers, it could easily be cost effective to remove it and make the lot available for another skyscraper, or for the new owner to do so and in so doing yield two dividends to the local government -- one from the sale of a lot with a high property value, and then another from the increased property tax revenue from the taller building.
> Claiming that private entities are immune (or at least, less susceptible) to the same problems of temporal discounting that clearly affects both individuals and many north American governments seems without foundation to me.
Temporal discounting isn't a problem, it's what you're supposed to do. An expense today is much more costly than the same dollar value expense in 50 years, because if you don't spend the money today, you get 50 years worth of interest on it (or don't have to pay 50 years in interest on having borrowed it).
> I'm thinking more about what their construction actually represented in terms of a society's willingess to invest (inefficiently!) in the future, and in a more-than-barebones aesthetic.
In which case you're essentially asking for the state to fund architecture as art with tax money. But if that's what the taxpayers want, why don't the buy it directly?
More saliently, spending tens of millions of dollars to re-tile a subway station isn't doing that anyway.