Addressing High Food Prices: A Household Vulnerability Analysis in Rural Burundi

Abstract

This work tackles the food crisis through two distinct standpoints. On the one hand, international price surge is addressed in its most financial view – i.e. the price generation process - and the related information within connected. On the other, the drawbacks such dynamics may have triggered are analysed in terms of rural household consumption and production. That is to say how this crisis has turned for a great part of the world population from being purely a financial issue into a matter affecting food provision, or the capability to get adequate amounts of food or not. The main research question is whether or not high food prices are pro poor and, if not, which are the households most affected in terms of their vulnerability. This work is organized as follows: the first part reviews the possible implications of soaring food prices in terms of speculation and behavioural aspects related to food stock accumulation. The dynamics of commodity futures prices in terms of stationarity and non-stationarity are then described as well as the role of structural breaks in the price generation process. Finally, structural breaks findings are compared against relevant food policy events and financial crisis turmoil. The second part addresses most specifically the question of high price consequences on rural households, reviewing the literature on first and second-round effects, and linking the main findings with household model theory. The investigation then converges into the field of uncertainty at very low levels of consumption, where the rural households are the most affected in terms of possible drawbacks related to food price inflation. The third part conceptualizes these drawbacks in the general field of vulnerability to poverty, assessing and theorizing the received different definitions and measures. Consequently, arising from these contributions, some major findings are reported in order to sketch an operational background for the proceeding empirical part. The last part addresses the main research issue against the case study in rural Burundi. A multilevel model is presented, to take advantage of the sampling procedure. Full description of the empirical setting is provided, as well as the econometric results. Finally, vulnerability sources influence by soaring prices is measured and broken down, with regards to different household settings.