Global Gender Gap Index: A stochastic dominance approach

Consistent tests of stochastic dominance efficiency are used to assess the Global Gender Gap Index which is a composite gender gap measure comprising 4 equally weighted dimensions - economic participation & opportunity, educational attainment, health & survival, and political empowerment. Country rankings and index scores vary considerably depending on the choice of dimensional weights. The aim of this article is to use stochastic dominance efficiency in order to examine two extreme scenarios - a best-case scenario where more countries achieve better measured outcomes and a worst-case scenario where more countries achieve poorer measured outcomes. This presents an opportunity to analyze the sensitivity of the index and it allows us to uncover which dimensions have contributed most to closing the gender gap and which ones are still in need of improvement. Shifting more weight towards educational attainment, and health & survival induces the best-case scenario. On the other hand, weighting political empowerment, and economic participation & opportunity more heavily induces the worst-case scenario. In 2018, 68% of the official (equally-weighted) measured gender gap was closed. Under the most optimistic weighting scheme, 90% of the gap closes. And under the most pessimistic weighted scheme, 34% of the gap closes.