Pathways out of poverty through the lens of development resilience: an agent-based simulation

Abstract

Poverty alleviation for smallholders must consider the increasingly varied and intertwined impacts of climate change and globalization. This calls for a resilience perspective that includes eradication of poverty and resilience enhancement under extreme events and shocks. Applying the framework of development resilience, we constructed an agent-based model based on small farming households in the Amazon Delta region in Brazil, and we used it to identify pathways out of poverty and sources for resilience among these households. The model allows us to explore the nonlinearity and heterogeneous nature of the smallholder livelihood systems, including how different household characteristics and livelihood strategies contribute to divergent livelihood outcomes. Using a unique yet simple tracking method, we were able to show the stochastic dynamics of individual household livelihoods in the face of various shocks, and how these households move in and out of different states of poverty over time (i.e., extremely poor, chronic poor, and nonpoor). By comparing traits of households that ended up in different states, we showed the need for targeted interventions for alternative livelihood strategies and key resources improvement. Different from conventional poverty alleviation programs, our findings emphasize empowering smallholders with different livelihood options. This has practical implications in terms of identifying leverage points in smallholder livelihood systems (e.g., livelihood strategy, land resources) that government and other agencies can use to intervene more effectively for households to become prosperous.

Keywords

• agent-based model
• development resilience
• livelihood
• poverty
• poverty trap
• smallholder