This study aims at identifying the determinants of demand for micro-insurance in Cameroon using a methodology based on the analytical framework proposed by Heckman (1979). These are counting models with double selection. First, a biprobit selection model is estimated to determine membership of an association on the one hand, and subscription to a micro-insurance on the other. Second, interest models called counting models are estimated to identify and analyze the factors that affect the number of micro-insurance policies. Data for the study are from the Fourth Cameroon Household Survey (ECAM4), a survey with national coverage conducted by the National Institute of Statistics (NIS) in 2014. The results make it possible to identify significant factors that are positively correlated with membership of an association and subscription to a micro-insurance. These are mainly factors such as level of education, age squared and household size. Conversely, the results show that male gender and the age of the household head significantly and negatively influence membership and subscription. Furthermore, male gender, age squared, household size and insurance premium are positively related to the number of micro-insurance policies purchased by household heads. Finally, age and level of education are negatively correlated with the number of micro-insurance policies purchased. Furthermore, the inverse of the Mills ratio indicates that the number of micro-insurance policies is negatively correlated with unobserved characteristics. In a context of poverty, these results call for several actions by public authorities to promote micro-insurance as a way of achieving universal social security protection.
