耕地集约利用对粮食产量变化影响的定量分析——以江苏省为例

The impact of intensive use of agricultural land on grain yields:A case study of Jiangsu Province

针对粮食生产力影响因素分层结构的特点,构建粮食单产影响因素的多层线性模型,以江苏省为实证案例,定量研究耕地利用集约度及宏观政策因素对粮食单产变化影响及其内在作用机制。研究结果表明:(1)粮食单产的影响因素是多层次的,耕地利用集约度解释了平均粮食单产差异的57.04%,农业政策因素解释了平均粮食单产差异的42.96%;(2)2001～2008年江苏省的粮食单位面积产出总体呈逐年增加的趋势,其中,劳动集约度和资本集约度分别解释了县级平均粮食单产差异的19.50%和5.68%。(3)所选政策变量中,支农支出、科技支出和农业贷款三个变量对市级平均粮食单产具有显著作用。(4)所选政策变量解释了市级粮食单产均值差异的63.21%,市级层次还有36.79%的"背景效应"未得到解释。

Hierarchical linear models can separate variables of different spatial scales and different management levels, and they explain interactions between explanatory variables. This paper, taking Jiangsu province as a research area, uses hierarchical linear models to analyze the influence and mechanism of agricultural land use intensity and policy factors on grain yields. The results are shown as follows. （1） There are multi-level factors affecting grain productivity per unit area, agricultural land use intensity and policy factors separate- ly explains 57.04% and 42.96% of differences in grain yields. （2） Grain productivity per unit area showed a rising trend from 2001 to 2008 in Jiangsu, of which labor intensity and capital intensity separately explained 19.50% and 5.68% of differences in average grain yield at county level. Land use intensity explained 25.17% of differences in average grain yield at county level, and intensive use of agricultural land is not the most important influ- encing factor that affects grain productivity, indicating that natural quality of land and land quantity has an increasingly important impact on grain production capacity. （3） Three pol- icy variables-agriculture investment, technology investment and agriculture loans-have sig- nificant influence on grain yields. Agriculture investment only explained 11.90% of differ- ences in average grain yields, and agriculture loans and technology investment separately explained 18.25% and 33.06% of differences in average grain yields. （4） All the selected policy variables can only explained 63.21% of differences in average grain yield at munici- pal level.