How industrial robots affect energy intensity: Empirical evidence from Chinese firms
How industrial robots affect energy intensity: Empirical evidence from Chinese firms

Summary/Abstract: Research background: While prior research has examined multiple determinants of energy intensity (EI), firm-level impacts of industrial robots (IR) remain under-explored. As IR technologies witness exponential growth, not only in China but across advanced and emerging economies, their potential to reshape production efficiency and energy consumption patterns at the firm level is increasingly critical. This study addresses a global policy and scholarly imperative: understanding how automation interacts with environmental performance amid accelerating climate goals. Given China's dual commitment to carbon neutrality by 2060 and its status as the world's largest IR market, this context offers unique empirical research--the findings have broad relevance for countries navigating digital transformation and decarbonization simultaneously. Purpose of the article: This study aims to dissect the direct, indirect, and non-linear impacts of IR adoption on firm-level EI using a sample of 1,286 listed manufacturing enterprises in China from 2011-2022. Methods: The empirical strategy integrates fixed effects to control for time-invariant firm heterogeneity and year-specific shocks, and panel quantile regression to capture distributional heterogeneity. Findings & value added: Findings show IR adoption significantly reduces firm-level EI, with IR's squared term exhibiting a threshold effect: EI declines substantially only after IR reaches a developmental threshold. Environmental regulations moderate this relationship, diminishing and even reversing IR's effect to positive. Panel quantile regression results show these effects are significant mainly at medium-high EI levels, absent at low levels. Heterogeneity analysis further indicates that threshold and moderating effects are pronounced in high-energy-consuming and state-owned firms at medium-high EI, but nonexistent in low-energy-consuming and non-state-owned firms. This study advances the theoretical understanding of technology-environment interactions by demonstrating that the energy-saving potential of automation is conditional on scale, institutional context, and baseline energy use--offering a universal framework applicable beyond China to inform sustainable industrial policy and corporate investment strategies worldwide.

• Details
• Contents