Lower partial moments as a measure of vulnerability to poverty in Cameroon

In this paper the class of Lower Partial Moments (LPMs) is used for measuring vulnerability as downside risk of household income in rural Cameroon. This class of established and coherent risk measures has been shown to meet a number of desirable properties. Among others, the LPMs fulfill the focus axiom, and for order greater than zero they are in harmony with expected utility theory under the weak assumption of risk aversion. Through combining the vulnerability measure with a portfolio approach it is possible to distinguish different livelihood systems for which the poverty and vulnerability measures are the explicit result of stochastic distributions of single activities in the households' portfolio and their covariance structure. In particular we consider the four major income generating activities in the study area: Sorghum, millet and rice production, and fishing. The results suggest that in the study area fishermen are less affected by adverse effects on income than other livelihood systems, while rice growers are the poorest and most vulnerable. It is also shown that rice and millet growers are suffering from chronic poverty, while transient poverty is more prevalent among the group of sorghum growers and fishermen. This implication is further confirmed by assuming a moving target equal to the mean portfolio income for the calculation of LPMs. The results of the scenario analysis suggest that policy interventions aiming at a reduction of the covariation structure between income flows from different activities are quite promising. -- Vulnerability as expected poverty ; Lower Partial Moments ; portfolio theory ; diversification ; Sub Saharan Africa