Determinants of Poverty in Cameroon : A Binomial and Polychotomous Logit Analysis

Undertaking a binomial and polychotomous logit regression, we investigate micro-and-meso determinants of probable moderate and persistent poverty status in Cameroon. Using the ECAM II Cameroon Consumption Household Survey, we identify education, age of household head, fraction of active adult household members, access to infrastructure, among others, as outcomes likely to reduce the poverty status of households. Regionally, residing in the Rural Haut Plateau, Savanna and Rural Forest regions all enhance the likelihood of being poor, revealing the worrisome situation of poverty in rural areas. Overall, factors associated with poverty status using the ordered logit model appear more important when targeting moderate rather than extreme poverty, translating government's failure to appropriately deal with the severity of poverty in its strategies and a likely cause for the slow development observed in Cameroon. Policy wise, public intervention that encourage growth in educational levels of household heads (education), employment, rationalization of household sizes and bridging spatially disparities of poverty among regions, should be rendered operational in effectively attempting to resolve issues pertaining to extreme poverty, in view of curbing the rise in clustering between the two extreme quintiles in welfare distribution, which can be a potential source of underdevelopment