Natural and Health Risk, Livelihood Resilience, and Relative Poverty Depth

Reducing relative poverty among rural households is important for inclusive and sustainable development. Using balanced panel data from the China Family Panel Studies (CFPS) for 2016–2022, this study draws on the Sustainable Livelihoods Framework and resilience theory to examine the relationships among natural risk, health risk, livelihood resilience, and relative poverty depth. A two-way fixed-effects model is used to estimate the association between livelihood resilience and relative poverty depth and the moderating role of resilience in the risk–poverty relationship. In the baseline specifications, both natural and health risks are positively and significantly associated with greater relative poverty depth, whereas livelihood resilience is associated with lower poverty depth. The interaction estimates indicate weaker positive associations between both risks and relative poverty at higher levels of resilience, with more stable results for health risk across specifications. The findings are broadly robust to alternative measures, samples, and estimation methods, and controls for other risks and policy environments. Subgroup estimates are generally larger among low-income households, and the negative health-risk interaction is marginally significant among households with elderly members. The association of natural risk with relative poverty is more evident in the short term, whereas the association with health risk is more persistent. Analysis of indirect associations supports a liquidity-constraint pathway for the natural-risk interaction. Evidence for the corresponding health-risk indirect association depends on the confidence-interval procedure, while indirect associations through debt pressure are not supported at the 5% level. The health-risk findings are comparatively stable, whereas the statistical significance of the natural-risk estimates, particularly the interaction term, is sensitive to spatial clustering.