Pinterest is waaay beyond the early stages of startups that Paul Graham focuses on.

It used to be that it did in fact take millions of VC just to get startups off the ground (70s, 80s, 90s). We could go back there. Would it be better for the ecosystem? Would employees have a net gain, or would there be fewer early stage companies & jobs? I don't know, but these are the macroeconomic questions to consider.

The impact would be reasonably small, depending on which sub-slice of the startup ecosystem you are looking at. Paul Graham doesn't advocate jacking up headcount before you have decently functional business fundamentals. The YC model seems to be that (before the product is proven out) development is done by founders, which costs the same no matter what the prevailing wages are. After that, it's either (1) paid by revenues or (2) paid by VC money -- in the latter case the VCs will adjust expectations and/or the amounts they invest, and in the former, they'll just need to get 2x more productive employees, which is well within the band of talent hired by the marketplace today.

I don't think pg says it's founders-only until you prove product-market fit. Usually it takes some hires and seed funding to run some experiments and show proof. Product-market fit is more of a Series A requirement.

If the cash cost of running experiments goes way up, that may very well mean going back to the old way of doing things, where you essentially had to jump straight to VC money just to get off the ground.

Out of curiosity, what are programmer salaries like in Japan, and how is immigration there?

Read and discover:

http://www.kalzumeus.com/2014/11/07/doing-business-in-japan/

(I could pull an answer to your question out of this -- hint, search the page for a dollar sign -- but if you're reading "out of curiosity" you should just get some coffee or something and read the whole glorious thing, because the number is useless without the context, and the context is very different from the USA.)

I'd forgotten about that. On the whole, it looks like, despite lower levels of immigration, Japanese engineers do not earn more than their US counterparts, which makes sense: the economy is complex, with a lot of different things factoring into how much people make.