摘要
改革开放以来，伴随西部大开发，西部陆海新通道的提出，贵州省经济发展
迅速，其经济增速在全国名列前茅。经济快速发展的同时，贵州城乡收入差距的
问题逐步暴露。16年在杭州举办的G20会议发表了关于数字普惠金融发展理念等
文件，旨在希望借助电子化和数字化技术为社会各阶层提供便利有效的服务和全
面实用的金融商品。据贵州省统计局公布的数据显示从2010年开始，贵州省的
县域城乡差距不断扩大，各区域间的收入差距不同，城乡收入差距受到多种因素
的综合作用。在十四五期间，贵州省将深入实施乡村振兴，大数据，大生态的三
大战略背景下。如何缩小贵州省县域城乡收入差距，提升农村居民的幸福获得
感，缓解城乡收入差距的矛盾变得刻不容缓。
本文以贵州省68个县为研究对象，数据直接采用北京大学数字金融研究中
心公布的2014-2019年的县域数字普惠金融数据。实证研究中直接采用对数化处
理的数字普惠金融指数为核心解释变量，泰尔指数表示的城乡收入差距为被解释
变量，选取产业结构，金融化程度，教育水平，财政支出，经济发展水平五个指
标作为模型控制变量，为研究贵州省县域数字普惠金融与城乡收入差距之间的关
系，分别构建面板回归模型和门槛模型。县域层次的城乡收入差距和数字普惠金
融是多方面因素影响的综合结果，基于空间的视角，尝试纳入各县地理位置构建
空间权重，采用地理加权回归模型研究数字普惠金融对城乡收入差距的关系。
具体，文章在梳理国内外学者关于数字普惠金融对城乡收入差距影响的文献
后，理论分析数字普惠金融对城乡收入差距的作用机制，并提出假设进行实证研究。
实证选取了全省68个县域数据构建面板回归模型分析；其次，以数字普惠金融指
数作为门槛变量构建了面板门槛回归模型；最后基于空间视角，将样本县的地理位
置纳入构建空间权重，选定以泰尔指数表示的城乡收入差距为被解释变量，数字普
惠金融指数为核心解释变量，建立地理加权回归模型研究数字普惠金融与城乡收
入差距的关系，并进一步整合地理加权回归模型和面板回归模型结果进行比较分
析。研究结果表明：（1）贵州省县域数字普惠金融能够缩小城乡收入差距。（2）
在贵州省，数字普惠金融的三个维度中，数字普惠普惠金融的使用深度相对于数字
化程度和覆盖广度，使用深度能够更加显著对贵州县域城乡收入差距起到减缓作
用。（3）贵州省县域数字普惠金融对城乡收入差距的影响在着单一门槛值。（4）
基于空间视角，贵州省各县域的数字普惠金融对缩小城乡收入差距影响程度不同，
并且地理加权回归模型比面板回归模型更加准确和有效。
关键词：县域数字普惠金融；城乡居民收入差距；面板门槛模型；地理加权回归
III

Abstract
Since started with the reform and opening up, with the western development and the
proposal of the new western land and sea channel, the economy of Guizhou Province has
developed rapidly, and its economic growth rate is among the best in the country. With
the rapid economic development, the problem of income gap between urban and rural
areas in Guizhou is gradually exposed. The digital inclusive financial development
conference held in Hangzhou in 2016 put forward corresponding development concepts
and issued corresponding documents, hoping to provide convenient and effective services
and comprehensive and practical financial commodities for all sectors of society with the
help of electronic and digital technology According to the data released by the Guizhou
Provincial Bureau of Statistics since 2010, the gap between urban and rural areas in
Guizhou Province has been expanding, and the income gap between regions and city is
different, and the income gap between urban and rural areas has been affected by a
combination of factors. Such as social economic ect .During the 14th Five-Year Plan
period, Guizhou Province will deeply implement the three major strategies of rural
revitalization, big data and big ecology. In order to decarse the income gap between urban
and rural areas in Guizhou Province, enhance the sense of happiness of rural residents,
and alleviated and decrease the contradiction between urban and rural income gaps has
become urgent.aim to improve rural income.
In this paper, 68 counties in Guizhou Province are taken as the research object, and the
data are directly used from the county digital inclusive finance data released by the Digital
Finance Research Center of Peking University .it is publish from 2014 to 2019. In the
empirical research, the logarithmic digital inclusive financial index is directly used as the
core explanatory variable, the urban and rural income gap represented by the Thiel index
is the interpreted variable, and the five indicators of industrial structure, financialization
degree, education level, government public finance expenditure, there are have economic
development level are selected as model control variables, and the panel regression model
and threshold model are constructed respectively to study the relationship between digital
inclusive finance and urban and rural income gap in Guizhou Province. The urban-rural
income gap and digital inclusive finance at the county level are the comprehensive results
of the influence of many factors, based on the spatial perspective, try to incorporate the
geographical location of each county to construct the spatial weight, and use the
geographical weighted regression model to research the relationship.it is include digital
inclusive finance and urban and rural income gap.
Specifically, after combing through the literature published by scholars at home and
abroad on the impact of digital inclusive finance on urban and rural income gap, this paper
analyzes the mechanism of digital inclusive finance on urban and rural income gap
according to theory, and puts forward hypotheses for empirical research. Empirically, 68
counties in the province were selected to construct a panel regression model analysis;
secondly, a normal panel threshold regression model was constructed with the digital
inclusive financial index as the threshold variable; finally, based on the spatial perspective,
the geographical location of the sample counties was included in the construction spatial
weight, and the urban-rural income gap represented by the Thiel index was selected as
IV

the explanatory variable, and the digital inclusive financial index was selected as the
explanatory variable, and a geographically weighted regression model was established to
study the relationship between digital inclusive finance and urban and rural income gap.
The results of the geo-weighted regression model and the panel regression model are
further integrated for comparative analysis. The results show that: (1) Digital inclusive
finance at the county level in Guizhou Province can narrow the income gap between urban
and rural areas. (2) In Guizhou Province, among the three dimensions of digital inclusive
finance, the depth of use of digital inclusive inclusive finance is relative to the degree of
digitalization and the breadth of coverage, and the depth of use can play a more significant
role in alleviating the income gap between urban and rural areas in Guizhou County. (3)
The impact of digital inclusive finance on the income gap between urban and rural areas
in Guizhou Province is at a single threshold. (4) Based on the spatial perspective, digital
inclusive finance in guizhou province has different degrees of impact on narrowing the
urban-rural income gap, and the geographically weighted regression model is more
accurate and effective than the panel regression model.
Key words: Gui zhou county digital inclusive finance; Income gap between gui zhou
urban and rural residents; Panel threshold model; Geographically weighted regression
V
。。。以下略