This paper describes work undertaken as part of a wider project investigating alternative institutional and economic policies to promote pro-poor agricultural growth. It presents a set of models to describe major features of poor rural livelihoods in the communal areas of Zimbabwe (based on 1996 data). It then uses these models to investigate the effects of different shocks and policy changes on the livelihood strategies of, and poverty incidence among, different types of household. Following an introduction, it provides a brief description of smallholder agriculture in Zimbabwe's communal areas and identifies critical issues facing policy with respect to these areas. This sets the agenda for the development and application of farm/household models: it describes their structure, scope, and validation before subjecting them to a variety of different shocks and scenarios. These are chosen to illustrate the capacity of the models to shed light on the effects of alternative policy regimes on heterogeneous communal farm households in Zimbabwe. The final section draws together the main conclusions.