O Comportamento do Preço da Terra Agrícola: um modelo de painel de dados espaciais

This study analyzes the behavior of the agricultural land price in the municipalities of Paraná State over the period 1999-2011. After literature review of the main studies of this field, the determinants of land price and methodology to be used were identified. For this purpose, we estimate the empirical model for price of agricultural land through spatial data panel. The results show that main statistically significant variables to explain the price of agricultural land in Paraná State are the gross value of agricultural production, asset valuation, total financing in agriculture, ratio municipal investment income, percentage of planted soybeans in the municipality and the incorporation of spatial dependence on land prices. These variables can explain around 84% of the prices and indicate that the data presented great intrapanel fit, and the spatial effects would be more intense in the municipality itself and in relation to the whole. The proposed model, specially, advanced in structural terms, by incorporating data from all 399 municipalities in Paraná. Empirically, the study advanced by estimating through panel data and considering the influence of spatial effects on the agricultural land price, highlighting the importance of the location and neighborhood in the price determination.

本研究分析了1999-2011年期间巴拉那州（Paraná State）各城市农业用地价格的变动趋势。通过对该领域主要研究的文献综述，明确了土地价格的决定因素及拟采用的研究方法。为此，我们采用空间面板数据（spatial data panel）方法估计农业用地价格的实证模型。结果表明，解释巴拉那州农业用地价格的主要统计显著变量包括农业生产总值、资产估值、农业总融资额、市政投资收益比、城市大豆种植比例，以及土地价格空间依赖性的纳入。这些变量可解释约84%的价格变动，表明数据具有良好的面板内拟合度，且空间效应在城市自身及整体层面更为显著。所提出的模型尤其在结构层面有所创新，纳入了巴拉那州全部399个城市的数据。实证方面，本研究通过面板数据估计并考虑空间效应对农业用地价格的影响，突出了区位及邻里因素在价格决定中的重要性。