I introduce a utility-consistent benchmark for international comparisons of real income, the GAIA ('Geary-Allen International Accounts') System. It coincides with the Geary method (which underlies the Penn World Table) when preferences are Leontief and with the EKS method (favoured by OECD) when preferences are homogeneous quadratic. The Geary method seems preferable since it gives a (possibly poor) approximation to a consistent set of international comparisons, whereas the EKS method gives a...