Fomby, Paula
Working paper no. 1998-31
Abstract
This paper is intended to enhance the literature on population and the environment by developing a model that takes into account age and gender composition of rural households in order to explain variation in the timing of migration and decisions about land use. In differently structured households, migration patterns may reduce or delay environmental degradation by limiting dependence on the productivity of the land, or these patterns may increase degradation by contributing to extensification with slash and burn methods or intensification with inappropriate inputs.
I show that the arguments that characterize the current population and environment debate capture the importance of migration on the agricultural frontier only marginally. Where migration is considered, researchers often ignore the changing capacity of a household’s members to migrate, either permanently or temporarily, over the course of a household’s life cycle and development. This framework is flawed because it assumes an idealized context with an indefinitely productive frontier or ignores the possibility that local conditions may pressure households into using temporary migration as a survival strategy, or as a means to obtain more advanced inputs. By recognizing the potential for migration patterns to continue after initial resettlement on the frontier, we should be able to better identify the emergence and persistence of land use patterns.
I formulate a model that considers how environmental change drives rural households to modify their income strategies, and the ways in which these strategies are conditioned by the household life cycle. Responses to change then influence future activities which perpetuate environmental change, although at different rates, depending on what resources are available to the household in a given life cycle stage. The model emphasizes the producer/consumer ratio and human capital in a household, the household’s social capital, the condition of the landholding, and the desirability of increasing land use relative to other strategies at a given time as factors that contribute to both the probability of migration and decisions about land use. In addition, changes in household activities that result from migration patterns at different life cycle stages are considered. These include changes in land management, use of inputs, extensification, or land acquisition.
To operationalize this model, I propose collecting a data set that includes event history calendars providing information on the life history of the household, its migration history, and its land use history. I then suggest using event history analysis to determine the hazard rate of the initiation of migration and changes in land use and labor supply. Together, these models can provide information about the way in which farmers are changing their income-generating strategies and the ways in which they are using land. By including the household life cycle, these models can also explain part of the variation in rates of deforestation due to small farmers and predict future rates of deforestation based on the types of households and migration patterns prevailing in a region.