Livelihood goals, livelihood strategy selection and household welfare in rural China: A utility maximization perspective : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

Abstract

China has the world’s second largest rural population and dramatic development in the rural economy. Rural households’ livelihoods have changed with economic transformation and face challenges from income equality, relative poverty, climate change and the continuing influence of COVID-19. In this situation, how to realize inclusive, sustainable and resilient livelihoods for rural residents is an urgent problem. Rural households’ livelihood decision largely results from unobserved heterogeneous goals and intentions and directly observable, different rural household characteristics. Previous studies explored the factors influencing rural households’ choice of a livelihood strategy and livelihood outcome but mainly focused on the impacts of external macro factors. However, livelihood goals are emphasized as the main base for rural households to allocate livelihood assets and are the critical orientations of livelihood strategy selection. To date, it is still poorly understood how rural households’ livelihood goals influence their livelihood strategy selection and whether different livelihood goals and their potential interaction effect with livelihood strategy selection influence rural household welfare. Using the panel data from the 2010-2018 China Family Panel Studies, this study explores the relationship between livelihood goals, livelihood strategy selection and rural household welfare. After deleting cases with missing values in the main and control variables, 23,967 rural cases are used for analysis in the study. To explore the linkages between livelihood goals and livelihood strategy selection, this study constructs measurements of three livelihood goals (the survival, security and self-respect goals) and then calculates an index for each goal. Livelihood strategies are divided into three groups (agricultural, non-agricultural, and diversification) based on rural households’ income sources. The multinomial logit model with fixed effects empirically estimates the relationship between livelihood goal indices and livelihood strategy selection. The random effects and the correlated random effects models are used to check the robustness. The two-step instrumental probit regression model is used to address the endogeneity problem. The results show that rural households with a survival goal are more likely to choose the diversification and agricultural strategies but less likely to select the non-agricultural strategy. In contrast, the non-agricultural strategy is the priority when rural households pursue security or self-respect maximization. To address potential endogeneity in the dynamic panel model, the generalized method of moments model is used to investigate the impact of livelihood goals and strategy selection on rural household welfare. The unconditional quantile regression with fixed effects model is employed to analyse the heterogeneous effects on household welfare. The results show that rural households with a higher livelihood goal do not obtain higher welfare. Rural households adopting the non-agricultural and diversification strategies have higher welfare than other strategies. The interaction effects show that rural households can get higher welfare if their livelihood goals and livelihood strategy selection match. The unconditional quantile regression result shows that households’ livelihood goals and livelihood strategy selection affect household welfare across different quantiles.