The issue is the demographic 'pyramid'. Japan passed the point of no return (not enough people of child bearing age) in the 80s.
When there's a baby 'drought', that can be good for a country short-term (less need of schooling/training, and as that 'cohort' gets older, more experienced workers); the problems come when they reach retirement age as there aren't enough working-age people to support the retired - and the retired don't consume much - so a nation needs to sell overseas (critical for someplace like Japan where they need to import food and energy)
I'm pretty sure Japan saw this coming. They've managed to co-locate their manufacturing into places that are still consuming (e.g. auto plants in the US).
But what happens to places like China, which is a few decades behind Japan demographics-wise, but also needs to import food and energy? They've run out of just-off-the-farm cheap labor, and in spite of trying to acquire (or create, but a lot of it has been acquisition, much without regard for intellectual property rights) as much IP as they can to make the labor issue less important, move up the manufacturing 'food chain'.
All that said, I'd agree that unrestrained population growth is a bad thing. Back when my wife and I were of childbearing age, we gave it a pass because we thought Malthus was going to catch up with the human race in our potential kids’ lifetimes. Turns out that hasn't happened, but there's no reason it can't, particularly as a lot of that boom came from not just antibiotics and the green revolution but the globalization that came out of the Bretton Woods system. The US has been paying to support that since '44, but I've been watching it grow less and less interested in it since Reagan. (Best evidence: the naval (as in US Navy) switch from numerous cruiser-class ships capable of blue-water trade overwatch to carrier groups that are not.)
You did a great job of explaining why a stable population is good, but the potential downsides of demographic collapse seemed to require a long response (apologies for the length).