Development Acupuncture: The Network Structure of Multidimensional Poverty and Its Implications

While development literature has come a long way in conceptualizing and measuring poverty multidimensionally, policy interventions to address it remain trapped in fragmented, sector-specific approaches. One of the main challenges in implementing integrated policy responses to multidimensional poverty reduction is understanding how the different dimensions are interlinked and how they jointly evolve over time. For example, this could require disentangling how a person’s health, education, and standards of living all interact in a dynamic sense. Motivated by economic complexity methods and applications, this paper uses network science to propose two new measures to understand the interconnected structure of multidimensional poverty: the Poverty Space (a network that visualizes the interactions among different indicators of poverty) and Poverty Centrality (a measure of the relative importance of each indicator within this network). Applying these measures to 67 developing countries using data from the Oxford Poverty and Human Development Initiative–United Nations Development Programme Global Multidimensional Poverty Index, the paper finds that the structure of multidimensional poverty networks is similar across countries and stable over time. The findings also show that indicators that are more central in the Poverty Space witness a more significant reduction in the censored headcount ratio over time, compared to peripheral indicators. These results are used to demonstrate how the Poverty Space can be applied in policy: using the forward-looking Policy Priority Inference framework to help guide policy choices. Overall, the paper points to the relevance of using network science methods to help to identify key “nodes” in the structure of multidimensional poverty where applied pressure (targeted interventions) could lead to a greater effect on the system as a whole.