I did not question that the varying story of how resource use shifted throughout history. As you correctly pointed out these shifts are extremely relevant for a

I did not question that the varying story of how resource use shifted throughout history. As you correctly pointed out these shifts are extremely relevant for accurate projections though. It however changes nothing of the fact that exponential growth in a finite system will eventually run into saturation. If you want to grow your corps revenue by X% each year you will have to sell more. Selling more means making more, making more means using more resources. Now you can reduce the resource use per part with clever designs, different materials or by seeing planned obsolescence as a feature (which will mean customers have to buy a replacement earlier and thus the actual resource use of the product over its lifecycle gets in fact worse). So you optimize and optimize and then you have reached the moment where there is nothing left to remove, but you still need to grow, so you just keep using more resources. Or if you are lucky eventually a new material or design innovation comes around, that again allows you to improve resource use. But your resource use is not going to be zero and you deplete something else on the finite planet. And you need to grow and use more of it each year. If you're lost on a raft in the ocean and first eat the bread and once the bread is gone switch to cookies that does not mean you won't eventually run out of stuff to eat. The solution is to tap into resources that regrow and not use more than is regrowing, but that doesn't jive well with infinite growth. I am an electical engineer and I know pretty well how exponential resource use works in complex systems. If a part of my system would like to exponentially grow its resource use infinitly I'd either consider it defective or use the fact that it can't grow infinitly by bounds I set on purpose.