Maize diversity, variety attributes, and farmers' choices in southeastern Guanajuato, Mexico

The first purpose of this study is to examine farmers' demand for varieties of maize landraces by applying a choice model in which variety attributes and features of the region of production, in addition to the farm household characteristics that are usually analyzed, determine area shares allocated among varieties. Variety attributes are their performance characteristics as evaluated by farmers who have grown them for many years, and generally they cannot be observed in the shelled grain found in marketplaces. By stratifying the analysis according to the potential productivity and infrastructure development of regions within the study area, we are able to test qualitative hypotheses related to the dynamics of development. The second purpose of the study is to investigate empirically the relationship between farmers' demand for variety diversity and the genetic diversity of maize landraces in the farmers' communities. The theory of impure public goods shows why farmers will generally choose to grow varieties with less crop genetic diversity than is viewed as desirable by society. Even if farmers were aware of changes in the supply of distinct maize materials in the community and considered in their own decisions the decisions made by other farmers in their communities, they could not completely control the level of genetic diversity among the maize populations that they grow. Maize is predominantly a cross-pollinating species, and unless fields are isolated or flowering times are distinct, the pollen of one maize population may reach any nearby population. Growing maize inevitably generates an "externality" in terms of crop genetic diversity. The framework we use also explicity recognizes that the set and structure of genetic resources grown by farmers in any region or at any given point of time reflects a complex process of interaction between biophysical, biological, and socioeconomic factors, each operating at different levels of aggregation or "scales" of analysis. The importance of scale has been well established in ecology and environmental sciences.