The impact of digital economy on household commercial insurance participation：Evidence from China
Description
This paper constructs provincial-level digital economy indicators, combines them with data from the 2016 and 2018 China Family Panel Studies (CFPS), and uses Probit and Tobit models to explore the impact of the digital economy on household commercial insurance participation. The results show that the digital economy significantly promotes the possibility of household commercial insurance participation and the degree of participation; the digital economy promotes household commercial insurance participation through channels such as increasing the accessibility of residents' internet use, enhancing social trust, and increasing household income; and the marginal effect of the digital economy in promoting commercial insurance participation is stronger for households with lower education level and lower per capita income. The findings suggest that increasing the level of digitization in economically disadvantaged areas is an important way to promote the development of commercial insurance.
Files
Steps to reproduce
This study used data from three sources. First, the data from the 2016 and 2018 CFPS, which contains demographic characteristics of residents as well as economic and structural characteristics of households, which meets the micro-level data requirements. Second, China's Digital Inclusive Finance Development Index, which serves as one of the indicators for constructing the level of development of the digital economy. Third, the provincial-level data on Internet construction, informatization development, digital transactions, and industrial development published by the National Bureau of Statistics and China Statistical Yearbook, which serves as an important indicator for constructing the level of development of the digital economy. The data were processed as follows: First, we identified the respondents in the CFPS database as heads of households, matched the personal and family information of the heads of households, and retained the sample of heads of households over 16 years of age. Second, we utilized the provincial information disclosed in CFPS, and matched the constructed level of digital economic development as well as provincial characteristic data to the CFPS sample using the year and the province as the key variables. Third, we removed missing values for key variables and shrinked the continuous variables such as commercial insurance expenditure, household consumption expenditure, social capital, household size, net household income per capita, and total household property up and down at the 1% level. The final valid sample was 25,420.