On some problems of using the Human Development Index in economic history

This paper provides a theoretical framework that shows that the Human Development Index (HDI) is equivalent to a paternalistic social welfare function: this implies that all alternative HDI formulas used by economic historians merely represent their ethical systems. The problem is neither the choice of the dimensions included in the HDI nor the weighting scheme but the lack of consistency with standard economic theory. A key consequence is that with HDI, "anything goes": using Italy 1861-2016 as a case study, we show how, given the same dataset, the interpretation of Italy's performance is entirely driven by the analyst's preferences.