People respond strongly to this in the West also -- "least publishable units", etc.

And yet, at least clinical trials fail here, and we don't have peer-review rings being busted or people throwing bales of money out the window as the police raid them for assisting academic fraud. (To name some recent Chinese examples.)

I would venture to guess that the crappy state of science in e.g. China is just due to the weak institutions/high corruption levels in their society.

Again, what incentives? If science cannot survive some 'weak institutions' abroad, which don't strike me as any worse than, say, the Gilded Age in America (and keep in mind the relative per capita GDPs of China now and, say, the golden age of German science before WWII), how long can one expect it to last?

This is almost mystical wording.

It's gesturing to society-wide factors of morality, values, and personality, yes, since it doesn't seem to be related to more mundane factors like per capita GDP.

As that gets better, science will get better too.

Japan is a case in point here. Almost as bad as China on the trial metric despite over a century of Western-style science and a generally uncorrupt society which went through its growing pains decades ago.