Ravallion () argues that the Human Development Index (HDI) embeds questionable tradeoffs between the dimensions used to compute the index. To alleviate these problems he proposes the adoption of one of the indices developed by Chakravarty (). In this paper I identify the following paradox: while the Chakravarty indices clearly exhibit more sensible tradeoffs than the HDI, the HDI produces more sensible rankings than the Chakravarty indices. To solve the paradox I identify the axioms behind each methodology responsible for the unintuitive tradeoffs and rankings and illustrate how to develop an index with these questionable axioms removed. This approach can result in methodologies that exhibit more intuitive tradeoffs by design, as it seeks inputs from the public as to what those tradeoffs ought to be, and produces rankings that are more in line with what the HDI wishes to measure: human development and capabilities, as conceptualized by Sen ().