Pastoralist strategies for survival in unpredictable environments: A model of herd composition that maximises household viability

Abstract

We present a general model of a camel and small stock, nomadic pastoral system. The model addresses the question of how household wealth should be divided between camels and small stock in order to maximise the long term viability of the household in an environment where rainfall is unpredictable. The modelling technique used is stochastic dynamic programming. The model shows that the optimal ratio of camels to small stock depends on total household wealth. In general, poor households should only invest in small stock. Once herd size is above a critical limit, the household should then invest most of its wealth in camels. This critical herd size depends, mainly, on the subsistence requirement of the household. Once investing in camels, the number of small stock kept depends mainly on the variability of small stock herd growth rate and on their price relative to camels. The importance of changing strategy according to wealth is illustrated by comparing the long term survival chances of households following the optimal policy with households following two other simple strategies which do not depend on wealth. The strategy with the highest mean productivity of the three in the short term is the least successful in the long term. This shows the vital importance of incorporating risk of drought when modelling these systems. This is an advance on previous models of pastoral systems as it enables relevant stochastic processes, and the herders' responses, to be taken into account. This has practical application as the expected long term viability of households, and the possible effects on that viability of particular events, such as a donation of stock or food aid, can be estimated.