Research On The Influence Of Artificial Intelligence On Labor Income Share

The sixth plenary session of the 19 th CPC Central Committee clearly proposed to build a new development pattern,promote high-quality development,comprehensively deepen reform and opening up,and promote common prosperity.In order to achieve the development goal of common prosperity,it is necessary to raise the income level of all people and strive to narrow the income gap,which implies the requirement of increasing the share of labor income.The share of labor income refers to the proportion of total output allocated to labors.According to the owner of the initial income distribution,the total output or total income is allocated to the government,enterprises and labor,in which the proportion of government distribution generally remains stable,and the share of labor income and capital income changes reversely.The share of capital income is occupied by capitalists or entrepreneurs,whose number is relatively small.If the share of capital income increases,the income of capitalists or entrepreneurs increases rapidly,while the per capita income of the labor that accounts for the majority of the population declines.This leads to the expansion of income distribution gap or income distribution polarization.Maintaining or moderately increasing labor income share is of great significance to narrow the income gap and achieve common prosperity.In terms of China's economic and social development,the fourth Industrial Revolution was born.As the engine of industrial digitization and digital industrialization,artificial intelligence not only can change the mode of economic operation,but also have an important effect on income distribution.Studying the influence of artificial intelligence on labor income share is of great theoretical and practical significance for our country to narrow the income gap and achieve common prosperity.This paper deeply studies the internal mechanism of the influence of artificial intelligence on labor income share,and reverses the situation that the role of “capital investment” is overemphasized while the role of “artificial intelligence” is neglected in the field of labor income distribution.The influence of artificial intelligence on labor income share depends on the relative size of its substitution effect and compensation effect.The substitution effect of artificial intelligence on labor increases the ratio of capital to labor in production and reduces the income share of industrial labor.It has spawned new jobs in industry and other industries and created new compensatory demand for labor.Artificial intelligence will affect the capital-labor ratio in production,but the substitution elasticity between capital and labor is different in different industries.Artificial intelligence can increase the marginal output of capital,thus increasing the ratio of capital to output,and reduce the share of labor income in this industry.However,in the industries where the substitution elasticity of capital and labor is small,this effect is not significant,and the distribution effect of artificial intelligence in different industries is different.In addition,artificial intelligence also affects the total labor income share by affecting the industrial structure.The paper further studies the heterogeneity of labor,and the gradual nature of technological progress determines the similarity between artificial intelligence and skill-biased technological progress,which can increase the income share of skilled labor and reduce the income share of non-skilled labor.In order to deeply analyze the specific path that artificial intelligence affects labor income share,the paper decomposes the change of labor income share according to four logic:first,employment,average wage and output change as intermediary influence variable;second,average wage and labor productivity change;third,industrial labor income share and industrial structure change;fourth,heterogeneous labor income share as intermediary influence variable.Based on this,the intermediary path of the influence of artificial intelligence on labor income share is analyzed.This paper empirically tests the contribution of artificial intelligence on the labor income share in various manufacturing industries.Based on the panel data of China's manufacturing industry from 2006 to 2019,the fixed effect model of panel data is used to empirically test the influence of artificial intelligence on labor income share in the industry.The results show that artificial intelligence has a strong substitution effect on employment.The impact on average wages is not significant,but it has a significant positive effect on output,and finally reduces the labor income share in the industry.Artificial intelligence also increases the employment and proportion of skilled labor in industry,increases the income share of skilled labor,and reduces the income share of unskilled labor.The heterogeneity results of different industries show that the effect of artificial intelligence on reducing labor income share is most significant in capital-intensive industries,but not significant in technology-intensive industries.This provides a basis for studying the impact of artificial intelligence on all kinds of labor with different levels of human capital.The paper also studies the influence of artificial intelligence on industrial structure and total labor income share by using inter-provincial panel data and panel econometric model,and finds that the total distribution effect of artificial intelligence is counteracted by the change of industrial structure.Finally,the paper examines the impact of artificial intelligence on the income gap between skilled and unskilled labor,as well as the counteract effect of human capital supply.The main conclusions of this paper are as follows:(1)Artificial intelligence reduces the labor income share of Chinese manufacturing industry,but can increase the income share of skilled labor,significantly increase the employment,wage level and average wage level of all labor force,and widen the income gap between skilled and unskilled labor.The development of artificial intelligence is beneficial to the labor with high human capital,but the increase in the average wage caused by artificial intelligence masks the fact that the pattern of income distribution deteriorates in the short term.(2)Artificial intelligence reduces the share of industrial labor income,but promotes the transformation and upgrading of industrial structure,and this inter-industry effect can increase the share of total labor income.When the inter-industry effect is stronger than the industrial effect,artificial intelligence can significantly increase the share of total labor income.(3)Artificial intelligence has different effects on heterogeneous labor.Artificial intelligence can replace unskilled labor with low level of human capital,increase the demand for skilled labor with high level of human capital,widen the income gap between skilled and unskilled wages.And increase the supply of skilled labor can alleviate the widening income gap between groups.The main innovations of this paper are as follows:(1)At the theoretical level,this paper systematically studies the effect and mechanism of artificial intelligence on labor income share.This paper puts forward the hypothesis that artificial intelligence is skill-biased.The effect of artificial intelligence on reducing labor income share mainly comes from the reduction of non-skilled labor income share,which in essence comes from the substitution effect of artificial intelligence on the employment of non-skilled labor force.The influence of artificial intelligence on employment leads to the change of industrial structure.Artificial intelligence affects not only the labor income share in the industry,but also the labor income share of other industries,especially the labor income share of the service industry.The industrial structure adjusts the effect of artificial intelligence on the total labor income share.(2)In terms of research content,a comprehensive empirical study is carried out from the perspective of industry,inter-industry and heterogeneous labor force.Using the panel data model of manufacturing industry,this paper empirically tests the contribution of artificial intelligence on labor income share in various manufacturing industries in China.Using provincial panel data,this paper studies how the effect of artificial intelligence on total labor income share is counteracted by industrial structure.It also empirically tests the effect of artificial intelligence on skilled,non-skilled labor income share and skill premium.(3)In view of research point,artificial intelligence reduces labor income share of manufacturing industries,mainly because artificial intelligence has strong substitution effect on employment,has no significant effect on wage level,and has obvious positive effect on output.Artificial intelligence reduces the share of labor income.When the labor intensity of the service industry is high,the effect of artificial intelligence on the reduction of labor income share is reversed by the industrial structure,and the service-oriented industrial structure increases the total labor income share.The increasing demand of artificial intelligence on high human capital skilled labor leads to an increase in the wage gap between skilled and unskilled labor.But the rich supply of human capital in developed areas can ease the rising trend of skill premium.