Our study examines farmers' crop choice to investigate the motivation behind the recent coffee production decrease in Chumphon, a southern province of Thailand. Through a choice experiment based on a random parameter logit model, which takes into account farmers' heterogeneous preferences for commodities, we find that the constraint on hiring labor induces farmers to refrain from replanting new coffee trees. In addition, the coffee tree aging problem induces the removal of old coffee trees, but does not affect significantly the replanting of new coffee trees when the hiring labor constraint is not imposed in the experiment. This seems to suggest that, since coffee has only one harvesting opportunity in a year, the constraint on hiring harvesting labor amplifies the income uncertainty of coffee management.

The pmpose of this study is to contribute to the understanding of changing farm sizes under cultural fann1and inheritance and transactions, in the context of succession decision. Results show that fannland inheritance strategies, involving the sharing offannland amongst children. will not change :from their parents and this will decrease farm sizes for the following farming households. In the past, farmers could enlarge farm sizes by purchasing farmland, but that shall become more difficult for future generations because of changing economic conditions. Farm training and providing suitable technology for small scale farms of the future is therefore required

This study captures the structural transformation in crop composition in Chumphon Province in southern Thailand at the macro (province) and micro (farm household / plot) levels. Based on the decomposition of crop production value growth into the effects of price, productivity, area reallocation between crops, and enlargement of the total area, production at the province level has been found to be driven by price and productivity effects for the past two decades. This raises the question of how farmers obtain crop price or technical information for crop choice decisions. In our original survey, farm or plot-level crop diversification was studied in villages where coffee production was historically active. We found that the capacity for diversification differs according to farm scale, and information use regarding crop price or farming practices is highly dependent on informal social networks. Furthermore, the role of social relationships in information sharing on price and farming practices of a specific crop is important even for farmers who do not produce that crop, suggesting that social relationships are the effective channel for the promotion of crop diversification and sustainable farming practices.