A neat idea, but peeling back into the computations, one has to worry. The details of calculations are suspiciously “mathy” - they are embarrassingly simple, bu

A neat idea, but peeling back into the computations, one has to worry. The details of calculations are suspiciously “mathy” - they are embarrassingly simple, but dressed up in fancier language that seems intended to impress and confuse. For example, take the computation for (product) “Diversity”: https://atlas.cid.harvard.edu/glossary https://atlas.cid.harvard.edu/glossary “Imagine a matrix, M_c,p, in which rows represent different countries and columns represent different products. An element of the matrix is equal to 1 if country C produces product P (with RCA greater than 1), and 0 otherwise. We can measure diversity (and ubiquity) simply by summing over the rows (or columns) of that matrix. Formally, Diversity = k_c,0 = ∑_p (M_cp)” This is an abstruse way of just saying “diversity is the number of unique products a country exports”. Which is fine! But why not say that? This is a common thing in economics, and it should raise our suspicion that the systems are far less sophisticated, and just supposed to bamboozle us to see past the underlying poor assumptions.