2024-12-29-世界银行-拉丁美洲和加勒比地区易受气候风险引发的贫困影响（英）_59页_15mb

报告摘要

Vulnerability to Climate Risk–Induced Poverty in Latin America and the Caribbean

1. Introduction & Methodology

The Sixth Assessment Report (AR6) of the IPCC highlights that climate risk arises from the convergence of climate hazards, vulnerability, and exposure. Following Cardona et al. (2012) and IPCC (2007), hazard refers to the possible future occurrence of physical events with adverse effects; exposure refers to the inventory of elements in an area where hazard events may occur. Vulnerability refers to the susceptibility to, or inability to cope with, adverse effects of climate change, including climate variability and extremes.

This study provides a comprehensive analysis of the vulnerability to poverty in a climate risk framework. Using household-level data from SEDLAC, along with natural disaster and climate data (e.g., EM-DAT,ThinkHazard!, CCKP), the authors model per capita household income and identify poor households within the vulnerability to poverty framework. The framework considers climate disaster risk, idiosyncratic risk, community factors, and specific household characteristics. Both risks are modeled through the shape of the per capita income distribution, estimating the expected variance. The expected mean income is estimated based on household and community characteristics.

The authors explicitly consider climate disaster risk as an aggregate shock by using natural disasters and extreme weather data. This allows classification of households into three categories: poverty-induced (PI) if their expected mean income is below the poverty line; risk-induced (RI) if their expected mean income is above but they are still vulnerable; and climate risk-induced (CRI) specifically for those pushed into poverty by climate-related events.

2. Data

The study uses five years of survey data per country when available (e.g., 2011-2017), harmonized and standardized household survey data from SEDLAC. Natural disaster and extreme weather data are sourced from EM-DAT, ThinkHazard!, and CCKP. Climate change data includes historical trends (e.g., temperature and precipitation) from CCKP, and drought indexes (SPEI) at various administrative levels.

3. Results

3.1. Regional Vulnerability

The incidence of vulnerability to poverty varies widely across countries, ranging from 2% in Chile to 64% in Guatemala (household incidence). Climatically, Guatemala and Honduras have the highest vulnerability, while Uruguay and Chile exhibit the lowest.

Approximately half the region's population is vulnerable to poverty, with significant heterogeneity within countries. El Salvador showed the largest gap between estimated poverty incidence and vulnerability to poverty.

3.2. Poverty & Risk-Induced Vulnerability

Out of the approximately 184 million people vulnerable to poverty in the region, 109.5 million are PI (poverty-induced), and 74.5 million are RI (risk-induced). Almost 14.5 million are CRI (climate risk-induced), with 60 million being IRI (idiosyncratic risk-induced). Idiosyncratic risk is consistently more prominent than climate risk across all countries.

3.3. Spatial Distribution

Bi-maps show vulnerability hotspots, such as high concentrations in Guatemala (753 out of 1,000 people) and Honduras (693 out of 1,000), often linked to climate hazards.

3.4. Social Assistance Coverage

Coverage varies significantly: almost all vulnerabilities in Peru receive some social assistance, while many remain uncovered for CRI households in countries like Colombia and El Salvador.

3.5. Climate Risk Decomposition

The study decomposes risk into sudden-onset (floods, landslides, cyclones/hurricanes, wildfires) and slow-onset (droughts, climate change indicators). While overall vulnerability incidence doesn't differ significantly, the sources of risk differ by country.

4. Conclusions

Approximately 184 million people in the region are vulnerable to poverty, with varying degrees of PI, RI, CRI, and IRI. Climate hazards and disaster risk significantly influence vulnerability, with strong geographic patterns. The study recommends targeted policies based on geographic distribution and integrating climate and hazard data into vulnerability assessments.