[jey]

"For the "Ramen" type, many of them may not afford America's often high living costs."

I think the algorithm PG proposes is something like this:

1. Produce enough that you would be able to convince someone to fund you. Maybe this takes a prototype, a convincing sales pitch, ramen profitability, etc.

2. Convince someone to [seed] fund you.

3. Move to a place with an environment favorable to startups.

4. Continue working on your product, now with access to investors, other founders, good employees, great weather, ...

The idea is to get to step 4 as fast as possible. In most parts of the world you'd never be able to find good investors or a startup-supporting culture, and that counts for lot. As PG has pointed out, there's a huge variance in startup-friendliness even amongst the major cities in the US.