我国林业全要素生产率地区的收敛性分析

【目的】对我国林业全要素生产率的收敛性进行实证分析，为把握省域林业发展差距和寻求平衡发展提供借鉴。【方法】基于2006—2018年我国省域林业数据，选取林业投入产出指标，采用DEA-Malmquist指数对省域林业全要素生产率进行测算，构建普通面板模型、空间误差模型（SEM）和空间自回归模型（SAR），对林业全要素生产率进行绝对β收敛和条件β收敛检验。【结果】2006—2018年我国林业全要素生产率总体平稳，但2018、2010和2015年林业全要素生产率呈下降趋势，主要原因是受自然灾害和金融危机的影响，生产率低于1.000。同时我国林业全要素生产率存在绝对β收敛和条件β收敛，说明省域林业全要素生产率间的差距不断缩小。在从收敛速度来看，绝对β收敛检验结果表明普通面板模型和半生命周期在加入空间因素后，收敛速度由0.0457变为0.0692（SEM模型）和0.0576（SAR模型）；半生命周期由19.363变为13.541（SEM模型）和15.650（SAR模型）年，收敛速度明显加快，追赶的期限明显缩短，说明空间效应对我国林业全要素生产率地区收敛起到促进作用。条件β收敛检验结果表明，在加入4个变量系数后，林业全要素生产率仍然存在收敛性，且收敛速度较之前均加快，半生命周期较之前也明显缩短，其中造林能力和林业保护能力系数较大，人力资本和外商投资系数较小，发挥作用不明显。可见，各地区林业发展基础水平及人力资本水平、外商投资、造林护林投入等因素均对区域林业全要素生产率收敛都具有促进作用。【建议】各地区应充分发挥自身地理优势和区域特点，因地制宜，促进林业经济快速发展；加大人力资本和外商投资，增强人才队伍建设的专业化和系统化，对外商投资的鼓励和监管并行；同时要打破区域限制，促使要素自由流动，进而实现林业经济与生态价值双平衡。

Convergence of forestry total factor productivity of China

【Objective】This paper conducted empirical research on the convergence of forestry total factor productivity of China, which was of great significance for grasping the development gap of provincial forestry and seeking balanced development.【Method】Based on China's provincial forestry data from 2006 to 2018, this paper first selected the forestry input-output indexes, and used the DEA-Malmquist index method to measure the total factor productivity of provincial forestry. Based on the calculation results, the common panel, spatial error model(SEM) and spatial autoregressive model(SAR) were constructed. The model performed absolute β convergence and conditional β convergence test on forestry total factor productivity.【Result】China's forestry total factor productivity was generally stable from 2006 to 2018, but forestry total factor productivity declined in 2018, 2010, and 2015, mainly due to the impact of natural disasters and financial crises, and the productivity was lower than 1.000. At the same time, China's forestry total factor productivity had absolute β convergence and conditional β convergence, It showed that the gap of provincial forestry total factor productivity has been narrowing. From the point of view of convergence speed, the absolute β convergence test results showed thatafter adding spatial factors into ordinary panel model and half-life cycle, thehalf-life cycle changed from 0.0457 to 0.0692(SEM) and 0.0576(SAR);the catch-up period changed from 19.363 to 13.541(SEM) and 15.650(SAR), the speed of convergence has been greatly accelerated, and the time limit for catching up has been greatly shortened, indicating that the spatial effect has played a role in promoting the regional convergence of China's forestry total factor productivity. The result of the conditional β convergence test showed that after adding the four variable coefficients, the forestry total factor productivity still had convergence, and the convergence speed was faster than before, and the half life cycle was also greatly shorter than before. However, the coefficients of afforestation capacity and forestry protection capacity were large.The coefficients of human capital and foreign investment were small, and the role was not obvious. Therefore, the basic level of forestry development in each region, the level of human capital, foreign investment, afforestation and forest protection, and other factors all played a role in promoting the convergence of regional forestry total factor productivity.【Suggestion】Each region should give full play to its own geographical advantages and regional characteristics, adjust measures to local conditions, promote the rapid development of forestry economy. Increase human capital and foreign investment, pay attention to the professionalization and systematization of talent team building, encourage and supervise foreign investment in parallel, break regional restrictions and promote the free flow of factors, and achieve a balanced forestry economic development and ecological value.