And it's really much worse than that. The aim should never be for maximum production of IP "at all costs" - it should be for maximum benefit at minimum cost.

And it's really much worse than that. The aim should never be for maximum production of IP "at all costs" - it should be for maximum benefit at minimum cost. So if an IP horizon of (say) 30 years started to make a tiny but measurable impact on the kind of IP produced - then that's a sign that 30 years is likely waaaay too long! because the benefits are virtually nil, yet the costs are huge - not just culturally, but even economically, due to lack of competition (being an exclusionary right and all). Not to mention there's this straw-man that the alternative to copyright is assumed to implicitly be a complete free-for all. I mean you could think up at least a dozen alternatives - like region limited copyrights; mandatory limited-price licensing; copyrights only for human beings, not legal entities; continually rising registration costs; production subsidies; citation and consumption based subsidies; retrospective rewards based on popular vote or a council of experts, etc etc etc - and you could easily use many of those ideas simultaneously - potentially including a very short old-style copyright term. Also, technology has advanced somewhat since copyright came to be. Numerical models that would have been heinously impractical hundreds of years ago may now be trivially practical; i.e. accounting for each and every consumer of a piece of art may now actually be practical. TL;DR: If society wants to support creators even at the cost to others - an entirely reasonable aim - it's a fallacy to compare copyright-as-is-today to anarchy.