Keeping things simple: why the Human Development Index should not diverge from its equal weights assumption

Using a range of statistical criteria rooted in Information Theory we show that there is little justification for relaxing the equal weights assumption underlying the United Nation's Human Development Index (HDI) even if the true HDI diverges significantly from this assumption. Put differently, the additional model complexity that unequal weights add to the HDI more than counteracts the improvement in goodness-of-fit. This suggests that, in some cases, there may be limited validity in increasing the complexity of a range of other composite sustainability indices.