Research On The Influence Of China's Industrial Intelligence On Labor Income Inequality

In recent years,a new round of technological revolution represented by Artificial Intelligence,Big Data and Industrial Internet is emerging around the world,and the industrial revolution characterized by informatization,digitization and intelligence is deepening.Many countries attach great importance to the research and application of new technologies,and promote the transformation of firms' production and residents' lifestyles to intelligence.However,while industrial intelligence promotes high-quality economic development,it will also have a huge impact on the labor market.Different from the steam,electrical and information technology in the first three industrial revolutions,the intelligent technology expands the intelligent scope of production activities from physical or manual jobs to brain jobs.Brain jobs such as driving,translation and customer service can be performed by intelligent machines independently,so the intelligent technology may have a more profound impact on the labor market.Industrial intelligence may not only change the heterogeneous labor productivity asymmetrically,but also have differential substitution and creation effects on different jobs,thereby affecting the distribution of labor income.Literature focus on the impact of robots or artificial intelligence on the employment and wage,and generally ignore the differential impact of industrial intelligence on the income of different labor groups,which leads the change of labor income inequality.Moreover,the mechanism by which industrial intelligence affects labor income distribution is ambiguous,and empirical evidence from developing countries is insufficient.Focusing on the research theme of the impact of industrial intelligence on the labor income inequality,this paper first systematically sorts out relevant literature views on industrial intelligence and labor income inequality.Second,this paper introduces heterogeneous jobs and labor to expand the task model,and theoretically examines the mechanism by which industrial intelligence affects labor income inequality.Third,this paper measures industrial intelligence and labor income inequality at the city level in China,and explores their evolution,distribution and correlation characteristics.Fourth,this paper empirically studies the impact of industrial intelligence on the labor income inequality,and examines the mechanisms such as productivity improvement,job alternation,and industrial structure transformation.This paper further analyzes the heterogeneity of individuals,jobs,firms and cities in the impact of industrial intelligence.Finally,this paper provides suggestions about exploiting the production advantages of industrial intelligence and promoting the rational distribution of labor income.The specific research contents and conclusions are as follows:First,this paper constructs a multi-sectoral general equilibrium framework by introducing heterogeneous jobs and labor classification into the task model,and endogenizes the research and development process of artificial intelligence technology,and mathematically deduces the mechanism by which industrial intelligence affects labor income inequality.The results show that industrial intelligence has three effects through automation expansion,new job creation and labor productivity improvement,namely productivity improvement,job alternation and industrial structure transformation.These three effects are the mechanisms by which industrial intelligence affects the income inequality between jobs and skills within the industry and the income inequality between industries.The mechanisms about productivity improvement and job alternation of industrial intelligence do not necessarily increase the labor income inequality,and the direction of impact is mainly limited by the R?D efficiency of artificial intelligence technology.The direction of the impact of the industrial structure transformation mechanism on the labor income inequality depends on the relative size of the substitution elasticity among industries and the substitution elasticity between intelligent machines and workers.Second,this paper builds a statistical index system of industrial intelligence from five aspects including network foundation,platform system,integrated application,innovation capability and security assurance,calculates the level of industrial intelligence through a comprehensive evaluation method,and uses Theil index and Gini coefficient to measure labor income inequality.Based on this,this paper analyzes the evolution,distribution and correlation characteristics of industrial intelligence and labor income inequality.The evolutionary characteristics of industrial intelligence show that the level of industrial intelligence in China has grown rapidly from 2003 to 2019,and the network foundation is relatively complete but the innovation capability is relatively weak.The distribution characteristics of industrial intelligence reveal that the level of industrial intelligence and its development speed are significantly higher in coastal areas,and cities with high aging and low labor protection intensity.The evolutionary characteristics of labor income inequality show that the national labor income inequality has a “three-stage” evolutionary characteristics from 2003 to 2019,and the inequality in high-end and production service industries is more prominent.The distribution characteristics of labor income inequality suggest that the male,old,elementary-educated and outdoor workers have higher inequality.The distribution of labor income is more unequal in large,non-state-owned and high operating cost firms and coastal,high aging and low labor protection intensity cities.The correlation characteristics of industrial intelligence and labor income inequality show that in the old and elementary-educated labor groups,large and high operating cost firms,coastal,high aging and low labor protection intensity cities,industrial intelligence and labor income quality presents a more obvious positive correlation.Third,this paper examines the impact of industrial intelligence on the labor income inequality based on the panel data from China.Baseline results indicate that industrial intelligence in China significantly increases the labor income inequality,and the results are robust even with a stricter standard error and controlling for city and year fixed effects.By replacing the measurement methods of core variable indicators,or changing the data sources for the measurement of labor income inequality,industrial intelligence still shows a significant increasing effect on the labor income inequality.By comparing and examining the impact of industrial intelligence on the income of labor groups with different incomes and demographic characteristics,we find that industrial intelligence increases the income gap between labor groups with different incomes,genders,ages and educational backgrounds,thus confirming the baseline conclusion.In order to deal with the endogeneity of the impact of industrial intelligence on the labor income inequality,this paper not only uses the penetration of robots in the United States and the level of information infrastructure in history as the instrumental variable of industrial intelligence,but also constructs instrumental variable based on heteroscedasticity.All instrumental variables pass the underidentification test,weak instruments test and overidentification test,and the two-stage least squares regression result show that the baseline results remain robust.Fourth,this paper empirically examines the mechanism of industrial intelligence affecting labor income inequality,namely productivity improvement,job alternation and industrial structure transformation.The examination of the productivity improvement mechanism shows that industrial intelligence improves the total factor productivity and labor productivity of different groups asymmetrically,which leads to unequal labor income growth among different groups and thus increasing the labor income inequality.The examination of the job alternation mechanism shows that industrial intelligence not only tends to displace the unskilled and routine jobs,but also tends to create skilled and non-routine jobs,which increases the labor demand for skilled versus unskilled jobs and nonroutine versus routine jobs,and thus widening the labor income gap between different jobs and increasing the labor income inequality.The examination of the industrial structure transformation mechanism suggests that,on the one hand,industrial intelligence promotes the rationalization of the industrial structure,weakens the income gap between industries caused by non-productivity factors,and thus alleviates labor income inequality.On the other hand,industrial intelligence promotes the industrial structure towards the high-end service industry and production service industry with higher labor income,thereby widening the income gap between industries and increasing the overall labor income inequality.Fifth,this paper analyzes the heterogeneity of the impact of industrial intelligence on labor income inequality in terms of individuals,jobs,firms,and cities.The analysis of individual heterogeneity shows that for workers of different genders,ages and educational backgrounds,there are differences in innate endowments,acquired abilities and living environments,which leads to differences in the effect of industrial intelligence on them.Industrial intelligence has a more significant effect on the increasing of income inequality among the male,old and elementary-educated labor.The analysis of job heterogeneity reveals that because of the differences in the comparative advantages of intelligent machines and labor in differentiated work scenarios such as different places,environments and safety,industrial intelligence has a different impact on the labor income inequality.Industrial intelligence has a more significantly increasing effect on the income inequality of labor who works outdoor and in the harsh and hazardous environment.The analysis of firm heterogeneity shows that the impact of industrial intelligence on the labor income inequality is constrained by firms' scale,ownership and operating costs.Industrial intelligence has a more significantly positive impact on the labor income inequality in large,non-state-owned and high operating cost firms.The analysis of city heterogeneity reveals that industrial intelligence has a differentiated impact on the labor income distribution in cities with different locations,aging rates and labor protection intensity.Industrial intelligence has enlarged the distributive gap of labor income more strongly in coastal cities and cities with high aging rates and low labor protection intensity.In order to alleviate the labor income inequality caused by industrial intelligence,this paper provides suggestions from four aspects including the worker,university,firm and government based on the research conclusions.Workers should establish the concept of lifelong learning,update their knowledge reserves in a targeted manner,and enhance their comprehensive capabilities to adapt to the intelligent production environment.Universities should optimize the professional curriculum system according to the actual production needs of the society,build a cutting-edge teaching team,and make full use of information technology to improve teaching efficiency.Firms should strengthen the skills training for the staff and actively carry out the R?D and application activities about intelligent in the direction of human-machine collaboration.While promoting the development of intelligence,the government should rationally plan the development process of intelligence in different regions,and provide social assistance to the unemployed caused by industrial intelligence.