Abstract
Poverty is a complex phenomenon and most of the developing countries are struggling to overcome the problem. Many of the countries lack direct poverty estimates at lower administrative levels due to the limited sample size of the national surveys. These become more essential when countries plan to allocate resources to address poverty at lower administrative level. Small area estimation offers help in this regard. To better address poverty the underlying determinants are important to explore. It is often argued that food insecurity is closely linked with poverty and both share some common set of predictors. This thesis tried to address these overarching issues of rural Bangladesh using latest available data, e.g., Census 2011 and Household Income and Expenditure Survey (HIES)-2010 and primary sample. The analyses found that 32% of the households were absolute poor and 19% were extremely poor in rural Bangladesh. Using ELL and M-Quantile methods this research identified Rangpur division as the poorest one, where Kurigram is the poorest district. Finally, considering both upper and lower poverty lines this research identified the poverty estimates at upazila level of Rangpur division. Two-level random intercept binary logistic regression analyses of HIES-2010 data identified age and education of household head, division, household size, household types, number of dependents, per capita income, household own land, access to electricity, amount of cultivable land, engagement in livestock and agro-forestry, household non-agricultural assets, number of male earner and number of female earner in the family as potential determinants of poverty. Significant community-level variations were observed in the analyses that emphasis the need for special attention on the poor performing communities. This study also investigated the linkage between poverty and food security in rural Bangladesh. For this purpose a primary sample was collected from the poorest district, especially from Kurigram of Rangpur division. Considering 2122 kcal, 40% of the surveyed households were found to be insecure. A strong relationship between food security and poverty was identified in the analysis. Furthermore, education, size of own cultivable lands and occupation were found to interact with poverty, resulting varying levels of influence of poverty on food security. Other determinants of food security were household size, income, engagement in livestock rearing and access to improved variety. Furthermore, the analysis of primary data identified some significant determinants of poverty that were occupation, household size, education of household head, per capita income, size of owned cultivable land, and engagement in livestock rearing. The poorest upazilas of Rangpur division identified in the analysis may have underlying causes, which may be somewhat different than that of other upazilas. Determinants of poverty, especially education, access to electricity, engagement in livestock and agro-forestry, identified in the study should be addressed in the policy formulation. As food security has significant association with poverty, strategies to address poverty and food security should have synergies. Also, other factors influencing food security, especially use of improved varieties and skilled based higher education, should be considered in policy formulation.