摘要
【目的】研究地方政府债务对农村公路建设的影响。【方法】基于省级面板数据,构建地方政府债务对农村公路建设影响的基准模型;通过时间-地区双固定效应回归分析,考虑不同地区异质性之间的相关性,基于非线性相关控制因素构建面板门限模型,并进行门限回归分析。【结果】地方政府债务规模与农村公路建设水平之间呈负相关关系,地方政府债务对农村公路建设的影响在东、中及西部地区之间存在差异,西部与中、东部地区相比,负影响更为显著。经济发展水平对农村公路建设具有单一门限效应,可以减缓地方负债对农村公路建设的负影响,经济发展水平较发达的地区对地方负债的承受能力更强。【结论】鉴于目前地方政府的债务压力和农村公路的建设现状,西部地区的地方政府更应合理地控制债务、构建多元化的融资模式。
[Purposes]The paper aims to study the impact of local government debt on rural road construction.[Methods]Based on the provincial panel data,a basic model of the impact of local government debt on rural road construction was constructed.According to the region-time double fixed-effect regression and the correlation analysis among regional heterogeneity,the panel threshold model was constructed based on nonlinear correlated control factors and threshold regression analysis was carried out.[Findings]There is a negative correlation between the scale of local government debt and the level of rural road construction.The impact of local government debt on rural road construction varies among eastern,central and western regions.Compared to the central and eastern regions,the western region has a more significant negative impact.The level of economic development has a single threshold effect on rural road construction,which can mitigate the negative impact of local debt on rural road construction.The regions with more developed economic level have stronger bearing capacity to local debt.[Conclusions]In view of the current debt pressure of local governments and the status of rural road construction,the western local government should reasonably control debts and build a diversified financing mode.
作者
王首绪
黄玥葳
WANG Shouxu;HUANG Yuewei(School of Traffic and Transportation Engineering,Changsha University of Science&Technology,Changsha 410114,China)
出处
《长沙理工大学学报（自然科学版）》
CAS
2022年第3期96-103,共8页
Journal of Changsha University of Science and Technology:Natural Science
基金
湖南省交通厅科技进步与创新项目(201330)。
关键词
农村公路建设
地方政府债务
地区异质性面板
门限模型
rural road construction
local government debt
regional heterogeneity panel
panel threshold model
