Doctoral Dissertation Research in Economics: Development of the Milk and Beef Processing Sectors: Implications for Land Use by Small Farmers in the Amazon

政府的政策促进了亚马孙地区融入国内和国际市场，鼓励建立牛奶和牛肉加工厂。因此，养牛已成为许多小农的首选土地用途。但是，关于牧场生产的可持续性和环境成本的争论仍然激烈。虽然区域加工厂数量的增加增加了农民作为牛奶和牛肉供应商的市场准入，但加工厂对农民土地使用决定的影响不那么明显。在市场条件下，农民采用更密集的畜牧业系统或转换更多的森林，以广泛的牧场的经验证据是显着lacking.This博士论文改进将允许研究者阐明（1）牛奶和牛肉加工厂的位置决策的决定因素和（2）和这些位置的决定对小农户的土地利用选择的影响。具体而言，要求提供资金，以收集详细信息的牛奶和牛肉市场在巴西亚马逊西部的案例研究区域，将结合现有的空间参考三个时期的面板数据集殖民地农民。假设是：（1）加工厂根据供应牛奶或牛肉的农民数量的预期以及牧场生产力可持续性的计算来确定区域;（2）加工厂密度增大带来的市场机会鼓励小农户投资于牧场集约化。本论文将建立在现有的国家科学基金会资助的研究基础上（SES，0452852）题为“亚马逊边境家庭土地使用和经济福利的动态”。通过与农民的访谈和遥感，收集了三个时期的牛奶和牛肉家庭生产、牧场专用土地数量和对牛生产力的投资的小组数据。拟于2006年7月至9月进行的实地调查将通过以下方式补充现有数据：（1）调查研究区域内牛奶和牛肉加工厂的人员，了解选择目前地点的原因、生产结构、农民的供应网络和每个加工厂的市场棚;（2）就加工厂的合同协议和采购方法与农民及其协会中的关键知情人进行面谈;（3）从以往调查中访问的农户中抽取次级样本，收集有关养牛业演变的详细资料（SES，0452852）来自IDARON（Agrosilvopastoril），自1999年以来在研究地点进行了广泛的牛普查的朗多尼亚州政府机构;（4）收集土壤肥力和地形的二级空间信息，这一分析将有助于不断增加的关于将遥感和GPS处理的空间数据与社会学相结合的跨学科文献。来自家庭调查的经济数据和来自对农产品加工厂访谈的市场结构信息。 随机效用框架中的离散选择模型将被用来解释牛奶和牛肉加工厂在边境地区的位置选择。动态面板方法将用于估计牛市场和农民土地使用决策的联合演变，利用数据的空间内容来校正空间依赖性和空间异质性。这项研究将研究牛市场的形成，包括牛奶和牛肉市场的位置和准入，分析农民对采用集约/粗放畜牧系统的反应，并向决策者通报土地利用集约化途径与毁林之间的联系。 从利润最大化的加工厂收集的信息将提供有关牧场可持续性和牛市场未来可能轨迹的见解。拟议的实地考察将丰富现有的面板数据集农民土地利用与牛市场的演变信息，并允许研究人员研究更广泛的研究问题。收集的数据将成为博士论文的核心，并用于期刊出版物和会议演示。

Government policy has facilitated integration of the Amazon into national and international markets, encouraging the establishment of milk and beef processing plants. As a result, cattle ranching have become the preferred land use for many small farmers. But there continues to be a raging debate over the sustainability and environmental cost of pasture creation. While an increase in the number of regional processing plants increases market access of farmers as suppliers of milk and beef, the impact of processing plants on land use decisions of farmers is less obvious. Empirical evidence on the market conditions under which farmers adopt more intensive pastoral systems or convert more forest to extensive pastures is notably lacking.This Doctoral Dissertation Improvement will permit the investigator to elucidate (1) the determinants of location decisions of milk and beef processing plants and (2) and the impact of these location decisions on land use choices of small farmers. Specifically, funding is requested to gather detailed information on the milk and beef markets in a case study region in the western Brazilian Amazon, to be combined with an existing spatially referenced three period panel data set on colonist farmers. The hypotheses are that - (1) processing plants locate in regions based on expectations of the number of farmers supplying milk or beef and calculations of the sustainability of pasture productivity; (2) market opportunities arising from greater density of processing plants encourages small farmers to invest in intensification of pastures.The dissertation will build on existing National Science Foundation funded research (SES, 0452852) entitled 'Dynamics of Household Land Use and Economic Welfare on the Amazon Frontier'. Panel data on household production of milk and beef, amount of land devoted to pastures and investments made in cattle productivity have been collected for three periods through interviews with farmer and remote sensing. The proposed fieldwork to be undertaken in July to September 2006 will supplement existing data by (1) surveying personnel of the milk and beef processing plants serving the study region about the reasons for choosing present location, production structure, supply network of farmers and market shed of each processing plant; (2) interviewing key informants among farmers and their associations about contractual agreements and procurement methods of processing plants; (3) gathering detailed information on evolution of cattle husbandry for a sub-sample of farmers interviewed in previous surveys (SES, 0452852) from IDARON (Agencia de Defesa Sanitaria Agrosilvopastoril), the government agency in the state of Rondonia that has conducted extensive cattle census in the study site since 1999; (4) collecting secondary spatial information on soil fertility and topography.The analysis will contribute to the growing inter-disciplinary literature on combining spatial data processed from remote sensing and GPS with socio-economic data from household surveys and information on market structure from interviews with agro-processing plants. Discrete-choice models in the random utility framework will be used to explain the location choice of milk and beef processing plants in a frontier region. Dynamic panel methods will be used to estimate the joint evolution of cattle markets and farmer' land use decisions, utilizing the spatial content of the data to correct for spatial dependence and spatial heterogeneity.This research will examine the formation of cattle market, including location and access to milk vs. beef markets, analyze farmer response to adopt intensive/extensive pastoral systems and inform policy-makers on the links between avenues of landuse intensification and deforestation. Information collected from profit-maximizing processing plants will provide insights on pasture sustainability and possible future trajectories of the cattle market. Proposed fieldwork will enrich an existing panel dataset on farmer landuse with information on evolution of cattle markets and allow researchers to examine a broader array of research questions. The data collected will be the nucleus of a doctoral dissertation and used in journal publications and conference presentations.