> A few critics bizarrely claimed that one recent paper had refuted my entire enterprise. For them your whole enterprise is also "one recent paper". And your

> A few critics bizarrely claimed that one recent paper had refuted my entire enterprise. For them your whole enterprise is also "one recent paper". And yours is the one with extraordinary claims. I only skimmed the spreadsheets and my criticism isn't quantitative so feel free to ignore it. For one thing signaling provides value, so just treating it as useless isn't fair. You can argue it's not worth it at the current prices, but at some price over 0 it would be worth it. For another thing there's a lot of nonlinear special cases that you throw out of window by looking at aggregates. Scientists and engineers don't grow on trees. For example imagine we save 30 years of education spending but we don't have quantum mechanics and relativity. Are we better off? Unicorns in education have crazy high potential value and can be used to justify a lot of wasted spending.