Do natural disasters change savings and employment choices? Evidence from Pakistan
This research investigates the economic response of rural households in Pakistan to natural disasters. In particular, the work explores the extent to which households adjust their savings and income strategies in response to floods.
The context – vulnerability to floods
Natural disasters such as floods are particularly harmful in developing countries such as Pakistan, and they disproportionately affect the rural poor, who depend especially heavily on agriculture. Non-farm employment and income are equally vulnerable to climatic events as they are usually tied to agricultural production. Pakistan is highly vulnerable to natural disasters, especially flooding, usually caused by excessive monsoon rainfall and glacial melt. In particular, the floodplains of the Indus River experience recurrent flooding events.
Research scope, findings and policy implications
The authors used a detailed panel dataset (the Pakistan Rural Household Panel Survey) which was assembled concurrently with two major flood events, the August 2011 flood in Sindh province and the August 2012 floods, which affected parts of Balochistan, Punjab and, again, Sindh.
The authors find evidence of an economic response, although adjustments appear to be temporary: affected farmers move away from agriculture as an immediate response to floods, but they return within a year. Although flood exposure lowers savings, adjustments in income strategies help farmers to overcome immediate losses and initiate recovery: flood-affected households allocate a significantly higher proportion of their post-flood income than unaffected households to replenishing livestock and seeds.
The findings have policy implications in terms of strategies to develop non-farm employment opportunities and financing economic migration to reduce income vulnerability.
This work was undertaken as part of the Pathways to Resilience in Semi-Arid Economies (PRISE) project.
ISSN 2515-5717 (Online) – Grantham Research Institute Working Paper series
ISSN 2515-5709 (Online) – CCCEP Working Paper series