耕地细碎化对农户化肥使用效率的影响
作者简介:史常亮(1988- ),男,甘肃张掖人,博士,助理研究员,研究方向为农村土地制度。E-mail: scl2313@126.com
收稿日期: 2019-08-23
要求修回日期: 2019-10-22
网络出版日期: 2019-12-28
基金资助
湖南省哲学社会科学基金项目(18YBQ080)
北京市社会科学基金项目(17LJB003)
教育部人文社会科学基金项目(18YJA790122)
国家自然科学基金项目(71973137)
The impact of land fragmentation on farmer's chemical fertilizer use efficiency
Received date: 2019-08-23
Request revised date: 2019-10-22
Online published: 2019-12-28
Copyright
在国家提倡化肥使用零增长、减量化的背景下,如何提高农户化肥使用效率变得至关重要。论文从技术效率角度衡量化肥使用效率,使用回归分析和基于R2的夏普里值分解方法考察了耕地细碎化对农户化肥使用效率的影响方向及程度。结果表明:(1)农户在化肥使用上存在着较高的技术无效率,在保持劳动力、土地等其他要素投入数量不变的情况下,约52.51%的化肥投入可以削减且不影响产出。通过提高农户对化肥的使用和管理水平完全可以做到化肥减量而作物不减产。(2)耕地细碎化对农户化肥使用效率具有显著负影响,是导致农户化肥使用效率低下的一个重要原因,能够解释农户间化肥使用效率差异12.15%,在所有决定因素中位居第四,仅次于村虚拟变量、水田比例和粮食作物种植比例。而且,在细碎化程度越高的西部地区,耕地细碎化对农户化肥使用效率差异的贡献越强。(3)农户在化肥使用效率上的差异主要源于所在村特征和包括经营规模、种植结构、土地质量、是否加入合作社等因素在内的生产经营特征的不同;而农户个体特征受集体决策影响,解释力较弱。
史常亮 , 张益 , 郭焱 , 朱俊峰 . 耕地细碎化对农户化肥使用效率的影响[J]. 自然资源学报, 2019 , 34(12) : 2687 -2700 . DOI: 10.31497/zrzyxb.20191216
In the background that China strongly advocated the chemical fertilizer use reduction and zero growth, how to improve chemical fertilizer use efficiency by farmers has become a crucial issue. This paper measures the chemical fertilizer use efficiency from the perspective of technical efficiency and analyzes the influence direction and degree of land fragmentation on the chemical fertilizer use efficiency through regression analysis and Shapley value framework based on R2. The results show that: (1) The technical inefficiency is high in the use of chemical fertilizers. In the case of other inputs such as unchanged labor and land, about 52.51% of fertilizer input can be reduced with no effect on the yield. Therefore, this could sufficiently achieve the aim of fertilizer reduction with no effect on the yield by improving the management level of farmer's fertilizer use. (2) The land fragmentation has a significant negative impact on chemical fertilizer use efficiency. All other factors being equal, each block of land increase could lead to a fertilizer use efficiency decrease of 0.002. This conclusion is robust to different variable definitions, estimating methods and winsorized sample. In addition, the Shapley value decomposition result shows that land fragmentation is also an important reason for low chemical fertilizer use efficiency of Chinese farmers. On the whole, land fragmentation can conduct 12.15% of farmer's chemical fertilization efficiency variance which is fourth in all the factors only next to village dummy variables, paddy field ratio and food crop planting ratio. Moreover, the contribution of the land fragmentation to chemical fertilizer use efficiency difference is stronger in the western region where the fragmentation is relatively high. (3) The difference in the chemical fertilization efficiency mainly depends on local village characte-ristics and the characteristics of production and management including the management scale, plantation structure, land quality, and whether or not to join the cooperative. However, the explanation of farmer's individual characteristics influenced by collective decision to chemical fertilizer use efficiency is relatively weak.
Table 1 Variable definition and descriptive analysis表1 变量定义及描述统计 |
项目 | 变量名称 | 定义或度量方法 | 均值 | 标准差 |
---|---|---|---|---|
投入产出 | 总产出 | 经营总收入/元 | 13813.84 | 20412.20 |
化肥投入 | 化肥费/元 | 1885.45 | 2660.26 | |
土地投入 | 播种面积/hm2 | 0.92 | 1.48 | |
劳动投入 | 投工量/工日 | 105.81 | 105.64 | |
其他投入 | 除化肥费外的其他费用/元 | 3608.44 | 7577.25 | |
被解释变量 | 化肥使用效率 | 农户化肥使用技术效率,利用SFA方法估计得出 | 0.47 | 0.16 |
核心解释变量 | 耕地细碎化_地块数 | 当年经营耕地的块数/块 | 6.23 | 7.75 |
耕地细碎化_块均面积 | 当年经营耕地面积/地块数/(hm2/块) | 0.17 | 0.46 | |
户主个体特征 | 性别 | 男=1,女=0 | 0.96 | 0.20 |
年龄 | 实际年龄/岁 | 56.24 | 10.25 | |
受教育程度 | 在校年数/年 | 7.12 | 2.51 | |
家庭特征 | 劳动力数量 | 15岁以上劳动年龄人口/人 | 3.35 | 1.23 |
家庭资产 | 年末居住房屋原值/元 | 67674.32 | 85171.53 | |
家庭经营主业 | 农业=1,非农业=0 | 0.90 | 0.30 | |
外出务工劳动力比例 | 家庭劳动力中外出务工劳动力a占比 | 0.17 | 0.22 | |
是否有人受过农业培训 | 是=1,否=0 | 0.15 | 0.36 | |
生产经营特征 | 种植结构 | 粮食作物b播种面积/种植业播种总面积 | 0.20 | 0.24 |
经营规模 | 年末实际经营耕地面积/hm2 | 0.81 | 1.34 | |
土地质量 | 有水田地面积c/经营耕地总面积 | 0.52 | 0.44 | |
是否参加合作社 | 是=1,否=0 | 0.05 | 0.22 |
注:a. 指到本乡镇地域外从业时间超过180日的劳动力。b. 包括小麦、稻谷、玉米、大豆和薯类。c. 为直接垦造的水田和水浇地面积之和。 |
Fig. 1 The situation of land fragmentation图1 耕地细碎化状况 |
Table 2 The parameter estimation results of stochastic frontier production function表2 随机前沿生产函数参数估计结果 |
变量 | OLS | SFA |
---|---|---|
化肥投入(对数) | 0.196***(0.015) | 0.172***(0.014) |
土地投入(对数) | 0.478***(0.019) | 0.567***(0.019) |
劳动投入(对数) | 0.074***(0.012) | 0.049***(0.011) |
其他投入(对数) | 0.239***(0.014) | 0.196***(0.013) |
村虚拟变量 | 已控制 | 已控制 |
截距项 | 4.458***(0.112) | 4.999***(0.105) |
Log likelihood | 33.011 | 126.021 |
R2 | 0.956 | |
单边广义LR检验(H0:σu=0) | 186.019 [0.000] | |
λ=σu/σv | 0.877***(0.010) | |
样本量/个 | 2720 | 2720 |
注:圆括号内为标准误,方括号内为相应检验的概率P值;限于篇幅,略去村虚拟变量的汇报结果;***、**、*分别表示在1%、5%和10%水平上显著,下同。 |
Fig. 2 The kernel density distribution of TE and FE图2 农户生产技术效率与化肥使用效率Kernel密度分布 |
Table 3 The benchmark regression results表3 基准回归结果 |
变量 | (1) 耕地细碎化 | (2) 户主个人特征 | (3) 家庭特征 | (4) 生产经营特征 |
---|---|---|---|---|
耕地细碎化_地块数 | -0.002***(0.001) | -0.002***(0.001) | -0.002***(0.001) | -0.002**(0.001) |
户主性别(男=1) | 0.035(0.022) | 0.033(0.022) | 0.035(0.023) | |
户主年龄(对数) | 0.215(0.446) | 0.306(0.457) | 0.287(0.452) | |
户主年龄平方(对数) | -0.026(0.057) | -0.038(0.059) | -0.036(0.058) | |
户主受教育年限 | -0.001(0.002) | -0.001(0.002) | -0.001(0.002) | |
劳动力数量 | 0.004(0.003) | 0.005(0.004) | ||
家庭资产(对数) | -0.003(0.025) | 0.002(0.026) | ||
家庭经营主业(农业=1) | 0.003(0.129) | 0.030(0.130) | ||
外出务工劳动力比例 | -0.052***(0.020) | -0.055***(0.020) | ||
是否受过农业培训(是=1) | -0.003(0.014) | -0.006(0.014) | ||
粮食作物种植比例 | -0.156**(0.063) | |||
经营耕地面积/100 | -0.213**(0.091) | |||
经营耕地面积平方/10000 | 0.088**(0.037) | |||
水田比例 | 0.063**(0.028) | |||
是否参加合作社(是=1) | 0.029*(0.016) | |||
常数项 | 0.505***(0.001) | 0.037(0.867) | -0.130(0.885) | -0.070(0.871) |
村虚拟变量 | 已控制 | 已控制 | 已控制 | 已控制 |
R2 | 0.085 | 0.087 | 0.090 | 0.108 |
样本量/个 | 2720 | 2720 | 2720 | 2720 |
注:圆括号内为村级聚类稳健标准误;限于篇幅,略去村虚拟变量的汇报结果。 |
Table 4 The robustness test results表4 稳健性检验结果 |
变量 | 更换度量方式 | 更换估计方法 | 极端值缩尾 | ||||
---|---|---|---|---|---|---|---|
(1) OLS | (2) OLS | (3) FLR | (4) Tobit | (5) OLS | |||
耕地细碎化_地块数 | -0.002*** (0.001) | -0.002*** (0.001) | -0.002*** (0.001) | -0.004*** (0.001) | |||
耕地细碎化_块均面积 | 0.003** (0.001) | ||||||
控制变量 | 已控制 | 已控制 | 已控制 | 已控制 | 已控制 | ||
村虚拟变量 | 已控制 | 已控制 | 已控制 | 已控制 | 已控制 | ||
(Pseudo)R2 | 0.107 | 0.108 | 0.008 | 0.134 | 0.110 | ||
样本量 | 2563 | 2720 | 2720 | 2720 | 2720 |
注:1. 圆括号内为村级聚类稳健标准误。2. 第(3)、第(4)列报告的是边际效应系数。3、限于篇幅,略去控制变量、村虚拟变量和常数项的汇报结果。 |
Table 5 The Shapley decomposition results (%)表5 夏普里值分解结果 |
变量 | 总体 | 东部 | 中部 | 西部 |
---|---|---|---|---|
耕地细碎化 | 12.15 | 10.67 | 11.94 | 14.16 |
户主个人特征 | 2.23 | 5.20 | 6.84 | 4.58 |
家庭特征 | 3.60 | 10.61 | 7.93 | 4.16 |
生产经营特征 | 34.43 | 32.54 | 18.76 | 30.20 |
村庄特征 | 47.59 | 40.98 | 54.54 | 46.90 |
注:根据国家统计局对东中西三大地区的划分标准,东部包括辽宁、河北、山东、江苏、福建、广东;中部包括黑龙江、吉林、河南、山西、安徽、江西、湖北、湖南;西部包括内蒙古、四川、重庆、陕西、甘肃、宁夏、青海、新疆、贵州、云南、广西。 |
[1] |
[
|
[2] |
|
[3] |
[
|
[4] |
|
[5] |
|
[6] |
[
|
[7] |
[
|
[8] |
[
|
[9] |
[
|
[10] |
[
|
[11] |
[
|
[12] |
[
|
[13] |
[
|
[14] |
|
[15] |
[
|
[16] |
[
|
[17] |
[
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
[22] |
|
[23] |
[
|
[24] |
|
[25] |
[
|
[26] |
[
|
[27] |
[
|
[28] |
[
|
[29] |
|
/
〈 |
|
〉 |