Scholarlink

<p>This paper investigates the effects of artificial intelligence on the labor markets of US states. The empirical analysis is based on the local projections method, where weekly state-level labor market conditions and Google search trends of "artificial intelligence" as a <i>field of study</i> are used. Robust to the consideration of state fixed effects and time fixed effects, the empirical results based on cumulative impulse responses show that positive shocks to "artificial intelligence" search trends deteriorate state-level labor market conditions for up to one month (consistent with the <i>displacement effect</i>), whereas they start improving state-level labor market conditions after six months (consistent with the <i>productivity effect</i>). Regarding the magnitudes, one standard deviation of a shock to state-level "artificial intelligence" search trends deteriorates (improves) state-level labor market conditions that corresponds to about <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="12197_2025_9732_Article_IEq1.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="43" /> </InlineMediaObject> <EquationSource Format="TEX">\(0.01\%\)</EquationSource> </InlineEquation> of a reduction (<InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="12197_2025_9732_Article_IEq2.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="43" /> </InlineMediaObject> <EquationSource Format="TEX">\(0.36\%\)</EquationSource> </InlineEquation> of an increase) in the state-level annual growth rate of real gross domestic product in cumulative terms after one month (year). Important policy suggestions follow.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial intelligence and labor markets: evidence from google trends

  • 期刊论文

  • 发表日期：

    2025年9月18日

查看全文
  • Hakan Yilmazkuday

摘要

This paper investigates the effects of artificial intelligence on the labor markets of US states. The empirical analysis is based on the local projections method, where weekly state-level labor market conditions and Google search trends of "artificial intelligence" as a field of study are used. Robust to the consideration of state fixed effects and time fixed effects, the empirical results based on cumulative impulse responses show that positive shocks to "artificial intelligence" search trends deteriorate state-level labor market conditions for up to one month (consistent with the displacement effect), whereas they start improving state-level labor market conditions after six months (consistent with the productivity effect). Regarding the magnitudes, one standard deviation of a shock to state-level "artificial intelligence" search trends deteriorates (improves) state-level labor market conditions that corresponds to about \(0.01\%\) of a reduction ( \(0.36\%\) of an increase) in the state-level annual growth rate of real gross domestic product in cumulative terms after one month (year). Important policy suggestions follow.