摘要
数字经济带来的注意力配置和由此产生的负外部性并没有引起足够关注,其社会福利效应可能被高估。基于2012-2018年中国219个地级及以上城市数据,利用爬虫技术和文本挖掘法以百度指数和词频统计分别刻画公众和政府对社会福利的注意力配置,实证检验了数字经济的社会福利效应。研究发现,数字经济显著提高了社会福利,且存在类型异质性、非线性递增和空间溢出的特征;数字经济提高了福利注意力配置,但注意力过度配置将引致负外部性,抑制数字经济的社会福利效应;采用双重差分模型作为“赛马”和面板门槛模型作为“反证”再次检验了这一机制:在福利注意力配置的最优区间内,数字经济可以缓解福利注意力配置的负外部性,改善社会福利。由此可见,通过数字经济来改善城市社会福利,应当充分考虑无形资源的最优配置,有效释放数字红利和政策效益。
The attention allocation caused by digital economy and its negative externalities have not attracted enough attention,and its social welfare effect may be overestimated.Based on the data of 219 prefecture-level and above cities in China from 2012 to 2018,this paper uses crawler technology and text mining method to depict the attention allocation of public and government to social welfare by Baidu index and word frequency statistics,and empirically tests the social welfare effect of digital economy.It is found that the digital economy has significantly improved social welfare,and it has the characteristics of type heterogeneity,nonlinear increasing and spatial overflow.Digital economy improves the allocation of welfare attention,but excessive allocation of attention leads to negative externalities and inhibits the social welfare effect of digital economy.The double difference model is used as"horse racing"and the panel threshold model is used as"counter-evidence"to test this mechanism again:in the optimal range of welfare attention allocation,digital economy can alleviate the negative externalities of welfare attention allocation and improve social welfare.Therefore,to improve social welfare through digital economy,we should fully consider the optimal allocation of intangible resources and effectively release digital dividends and policy benefits.
作者
黄险峰
付成林
高鑫
HUANG Xianfeng;FU Chenglin;GAO Xin(School of Economics,Liaoning University,Shenyang 110136,China)
出处
《经济与管理评论》
CSSCI
北大核心
2023年第5期124-137,共14页
Review of Economy and Management
基金
全国教育科学规划课题重点课题“基于经济增长视角的新时代高等教育结构优化研究”(DFA200302)。
关键词
数字经济
注意力配置
社会福利
遮掩效应
资源配置
digital economy
attention allocation
social welfare
masking effect
resource allocation