From vulnerability to resilience: long-term livelihood recovery in rural China after the 2008 Wenchuan earthquake
University of Delaware
Disaster recovery is considered as the least understood aspect of emergency management by both practitioners and researchers. In the past few decades, the concept of recovery has evolved from a linear process focused on the physical aspects to a multi-dimensional, dynamic social process that reflects the different stages and patterns of human activities. However, the lack of experiences from developing countries rather than the United States alone limits the theorization effort of disaster recovery. This dissertation contributes the knowledge of disaster recovery from the recovery experience in China after the 2008 Wenchuan earthquake, and a sustainable development perspective is integrated in the study. The Sustainable Livelihoods Approach (SLA) which is developed and widely used in the anti-poverty and development areas is adopted as the research design and theoretical guidance framework in the study series. The SLA framework, especially the five livelihoods assets (financial capital, human capital, social capital, natural capital and physical capital) are used as the main measurements of recovery. Meanwhile, the perception of recovery is also included as one dimension of recovery in the dissertation. The data used in this analysis mainly comes from three data collection efforts. The first one is conducted in June, 2008, about one month after the earthquake and disaster response data is collected through semi-structured interviews. A second data collection effort which adopted a mixed method is implemented in January, 2009. Both quantitative household surveys and semi-structured qualitative interviews are used for data collection. A stratified sampling method is used for the questionnaire survey, and semi-structured in-depth interviews are implemented with both community leaders and community members. In total, 515 household questionnaire surveys and 50 in-depth interviews from nine villages within three towns are obtained for analysis. The third data collection effort occurs in the summer of 2012 and follows up on the same households and villages from the 2009 effort with similar data collection patterns. Finally, information from 415 households is followed and another 37 qualitative interviews are obtained in the summer of 2012. The qualitative data is coded and analyzed using QSR Nvivo 10 and the quantitative data is analyzed by Stata 12.1. The qualitative data reveals that there are six stages of post-disaster activities within these families: self-protection, safety information seeking/rescue, family reunion and temporary sheltering with uncertainty, self-rescue and waiting in temporary shelters, housing reconstruction/repair and livelihood recovery. For the most burdensome housing reconstruction, the unified reconstruction with counterpart assistance, the unified reconstruction without outside help, combined-reconstruction, and self-reconstruction are the four patterns of reconstruction. The major determinants of household reconstruction decision making include the available individual/household resources, higher level government support, and community collective actions. In terms of livelihood assets, the comparisons of the results between the situation in the early recovery period (i.e., 2009) and three years later (i.e., 2012) indicate that all the five livelihood assets: financial capital, human capital, physical capital, natural capital and social capital have increased. With the increase of these livelihood assets, the inequalities of the financial capital, physical capital, natural capital and human capital also enlarged. On the contrary, the social capital gap has decreased in 2012 compared with the situation in 2009. The financial capital structure changes also reflect the changes of people's livelihood strategies since the earthquake: people earn more from their salaries than on-farm activities. The physical capital which reflects the housing condition has the biggest improvement in the disaster recovery process. Overall, most of the 2008 earthquake survivors have recovered from the housing recovery aspect, no matter in terms of estimated house value, housing structure or in terms of habitable space. Though the social capital is narrated as useful for housing recovery in the qualitative interviews, the quantitative models doesn't support such hypothesis. On the contrary, financial capital and human capital's effects on housing reconstruction are significant. Meanwhile, the government assistance is found to play an important role of facilitating housing recovery. In terms of recovery perception, about 70% of our survey respondents report that they have recovered from the disaster impact while another 30% say that they have not recovered yet. The financial capital has a consistently significant positive effect on perceived recovery while other four livelihood capitals' effects are not statistically significant. The results also show that the current livelihood assets are stronger predictors of recovery perception than the changes of livelihood assets, which may indicate that psychological-related recovery perception may be more determined by current well-being status rather than absolute changes. The final chapter discusses the theory and practice contribution, as well as some possible further research agenda of disaster recovery.