I don't think every study is completely corrupt. I wouldn't call it an assumption as much as an assessment that a lot of studies seem to agree with reality pretty well.

Pragmatically I suspect this might be because most of the market don't actually read these things, and are just as happy with made up bs.