Modelling embedded risk in peasant agriculture: methodological insights from northern Malawi

Using a linear-programming model of farming systems in northern Malawi, the conditions under which peasant farmhousehold models may need to allow for embedded risk are investigated. Tactical, sequential responses to uncertainty are found to be more important to labour-scarce households with limited access to capital and to credit markets. Compared with semi-sequential programming, discrete stochastic programming (DSP) provided more efficient solutions for problems involving embedded risk. There may be intuitive advantages in presenting results from DSP models in terms of a semisequential strategy. © 1999 Elsevier Science B.V. All rights reserved.