人工智能应用对劳动工资的影响及偏向性研究

Research on the Impact and Bias Effect of Artificial Intelligence Applications on Labor Wages

文章通过引入含有时间分配的效用函数拓展任务模型,分析人工智能应用对劳动工资的影响及作用机制,并运用中国家庭追踪调查数据进行实证检验。研究发现:(1)人工智能应用对劳动工资有负向冲击,主要通过缩短劳动工时和劳动岗位更替实现。一方面,智能机器的应用将劳动者从部分繁重的工作任务中解脱出来,减少劳动者的工作时间,从而降低工资;另一方面,机器的应用会直接替代劳动,技能“折旧效应”使其在劳动力市场上难以获得长期稳定的工作,且机器人大规模应用会弱化企业对劳动的依赖,用工关系的短期化也将引发劳动岗位的频繁更替,对劳动者的工资产生负面影响。(2)人工智能应用对不同技能劳动者工资的影响具有偏向性,劳动者技能水平的提升可以缓解人工智能应用对工资的负向冲击;人工智能应用对工资的负向影响会随着劳动者执行任务的复杂度和认知要求的提高而弱化。文章认为,要制定收入分配调节政策以平抑人工智能应用的冲击;提高劳动者的技能水平和职业转换能力,促使劳动力市场上形成“技能—岗位”匹配的良性互动。

Expanding the task model by introducing a utility function with time allocation,this paper theoretically analyzes the impact and mechanism of artificial intelligence applications on labor wages,and uses China Family Panel Survey(CFPS)data to conduct empirical tests.The study provides two results.On the one hand,artificial intelligence applications have a negative impact on labor wages,which is mainly achieved by shortening labor hours and increasing job turnover.The the application of intelligent machines frees workers from some of the heavy work tasks,cutting the working hours of workers,thereby reducing wages.Meanwhile,the application of machines will directly replace labor,the“depreciation effect”of skills makes it difficult for employees to obtain long-term stable jobs in the labor market,and the large-scale application of robots will weaken the dependence of enterprises on labor.The short-term employment relationship will also lead to frequent job turnover,which will have a negative impact on the wages of workers.One the other hand,the impact of artificial intelligence applications on the wages is skill-biased,which means the improvement of workers’skill levels can alleviate the negative impact of artificial intelligence applications on wages.The negative impact of artificial intelligence applications on wages will also be weakened as the complexity and cognitive requirements of tasks performed by workers increase.The paper holds that it is necessary to formulate income distribution adjustment policies to calm the negative impact of artificial intelligence applications,and to improve the skills level and career transformation ability of workers in order to foster a benign interaction of“skill-job”matching in the labor market.