Resilience, Shocks and the Dynamics of Well-being Evidence from Malawi∗

We seek to understand vulnerable households’ well-being dynamics and their ability to cope with shocks. We draw on the poverty literature and posit the concept of resilience as a latent variable capturing the stochastic dynamics of households’ well-being, measured using the Coping Strategy Index, a good proxy in a humanitarian emergency. In order to inform our analysis empirically, the ‘Measuring Indicators for Resilience Analysis’ project collected 12 months of data in Malawi. We estimate an auto-regressive model to track households’ experience of subjective shocks over time. We find that households with spatially dispersed fields are less likely to experience the adverse effects of drought, while female headed households are more likely to experience the adverse effects of a family member falling ill. Using a Blundell-Bond estimator, we find that differences in land farmed, gender of household head, and having fields far from home lead to shifts in the distribution of expected well-being outcomes. Finally, we harness machine learning techniques to predict future well-being. We find that crises are concentrated in specific geographic areas, making targeting all the more important. Copyright 2017 by Knippenberg, Jensen & Constas. All rights reserved. Readers may make verbatim copies of this document for non-commercial purposes by any means, provided that this copyright notice appears on all such copies.