能源转型对就业结构的影响机制——基于分组分位数回归的异质性分析

The Impact Mechanism of the Energy Transition on Employment Structure—A Heterogeneity Analysis Based on Grouped Quantile Regression

摘要: 在全球气候治理与“双碳”目标推动下，能源转型深刻影响中国就业规模、结构与质量，且对不同地区、产业和技能劳动者的影响存在异质性。本文基于2014~2023年中国30个省份面板数据，利用统计年鉴、百度指数和CEADs碳核算数据库等，从“绿色创造、转型冲击、服务升级、质量升级”四个维度构建就业结构指标体系，并构建能源转型指数PETI，采用面板分位数回归和分组分位数回归，考察能源转型对就业结构的影响及其区域、资源禀赋和政策干预层面的异质性。研究发现：第一，能源转型对就业结构的影响具有显著分布异质性，对低碳就业和生产性服务业的拉动主要集中在高分位省份，对高碳依赖就业的抑制在低中分位省份更明显。第二，资源禀赋异质性突出，非资源型省份呈现“绿色创造 + 传统削减 + 服务升级”的良性调整，高碳资源富集省则出现“绿色受抑、高碳未减”的结构性困境。第三，政策干预能有效缓冲转型冲击，高干预组高碳就业负向冲击更小，对生产性服务业和高薪技术服务业的拉动更强。第四，区域异质性显著，东部转型红利最突出，中部阵痛最剧烈，西部高薪技术岗位增长迅速，东北整体影响最弱。因此，能源转型对就业结构的影响并非“一刀切”，应制定差异化支持政策：对资源型省份建立公平转型机制和再就业培训体系，对中西部加大绿色产业培育，对东部持续引领技术创新，并强化政策干预在缓冲冲击、放大红利方面的作用。

Abstract: Under the global climate governance and the “dual carbon” goals, the energy transition has profoundly affected the scale, structure, and quality of employment in China, with heterogeneous impacts across regions, industries, and skill levels. Based on panel data from 30 Chinese provinces over the period 2014~2023, and drawing on statistical yearbooks, Baidu Index, and the CEADs carbon accounting database, this paper constructs an employment structure indicator system from four dimensions: “green creation, transition shock, service upgrading, and quality upgrading” and develops an energy transition index (PETI). Using panel quantile regression and grouped quantile regression, it examines the impact of the energy transition on employment structure and its heterogeneity across regional, resource-endowment, and policy-intervention dimensions. The findings are as follows. First, the impact of the energy transition on employment structure exhibits significant distributional heterogeneity: its stimulating effect on low-carbon employment and producer services is mainly concentrated in high-quantile provinces, while its suppressive effect on high-carbon-dependent employment is more pronounced in low- and middle-quantile provinces. Second, resource-endowment heterogeneity is prominent: in non-resource-based provinces, the energy transition displays a benign adjustment pattern of “green creation , traditional reduction, service upgrading,” whereas in high-carbon resource-rich provinces, a structural dilemma of “green suppression alongside undiminished high-carbon employment” emerges in the early stage of transition. Third, policy intervention effectively buffers transition shocks: in the high-intervention group, the negative impact on high-carbon employment is smaller, and the stimulating effects on producer services and high-wage technical services are significantly stronger than in the low-intervention group. Fourth, regional heterogeneity is significant: the eastern region enjoys the most prominent transition dividends, the central region experiences the most severe transition pains, the western region shows rapid growth in high-wage technical positions, and the northeastern region exhibits the weakest overall impact. These findings indicate that the impact of the energy transition on employment structure is not “one-size-fits-all” but varies significantly with regional conditions, resource endowments, and policy intervention intensity. Therefore, differentiated transition support policies should be formulated: establish fair transition mechanisms and retraining systems for resource-based provinces, intensify green industry cultivation in the central and western regions, sustain eastern leadership in technological innovation, and strengthen the positive role of policy intervention in buffering shocks and amplifying dividends.