Green Industry and High-Quality Employment Outcomes in 20 Mountainous Counties of Zhejiang (2010–2023)

Abstract

Promoting green industrial development and enhancing high-quality employment are crucial for advancing county-level economic growth and achieving shared prosperity. This study analyzes the spatiotemporal evolution trends of green industrial development and high-quality employment using panel data from 20 mountainous counties (cities and districts) in Zhejiang Province from 2010 to 2023. It employs panel models to investigate the effects and mechanisms through which green industrial development fosters high-quality employment. The results indicate that, during the study period, both green industry development and high-quality employment exhibited uneven progress across the 20 mountainous counties (cities and districts) in Zhejiang. Mechanism analysis revealed that green industrial development significantly promotes high-quality employment through two pathways: industrial structure upgrading and technological progress. The heterogeneity analysis indicates that the impact of green industrial development on high-quality employment varies significantly across different industrial structures, with counties dominated by the tertiary sector showing more substantial promotion effects. The threshold regression analysis reveals a dual-threshold effect of technological progress in promoting high-quality employment through green industrial development, presenting an approximately J-shaped nonlinear relationship. The research findings provide significant support for the sustainable development of the ecological environment and society by addressing current imbalances between ecological preservation and economic or social growth.

1. Introduction

Against the backdrop of escalating global climate change and deteriorating ecological conditions, the development of green industries has emerged as a vital pathway towards achieving the Sustainable Development Goals [1]. China’s official strategic documents have established clear criteria for green industries. As outlined in the Chinese Green Industry Guidance Catalogue (2023 Edition) [2], green industries include economic activities focused on reducing resource consumption, mitigating environmental pollution, and decreasing greenhouse gas emissions. These efforts support products and services that address climate change and promote ecological conservation. Specifically, the scope of green industries comprises six major categories: energy conservation and environmental protection, clean production, clean energy, ecological environment, green infrastructure upgrades, and green services. Concurrently, high-quality employment serves as a fundamental pillar for enhancing incomes and prosperity among urban and rural workers, narrowing the “three major gaps”, and constitutes the most basic aspect of people’s livelihoods [3]. However, while the rapid expansion of green industries has spawned diverse new occupations and skill requirements, it has also precipitated labor surpluses stemming from the contraction of traditional sectors [4,5]. Consequently, properly balancing the development of green industries with achieving high-quality employment has become a critical task for China today.

Current domestic and international research on green industrial development and high-quality employment primarily focuses on conceptual definitions, measurement methodologies, and analyses of impact mechanisms. Scholars generally recognize green industries as models promoting sustainable production and consumption [6]. At the same time, high-quality employment differs from the ideal of full employment. Specifically, it aims to maximize opportunities while recognizing the natural frictional and structural unemployment in an economy. Additionally, it enhances the qualitative aspects of employment [7]. Regarding the measurement of green industrial development levels, existing research predominantly constructs indicator systems that comprehensively assess performance across three dimensions: green production, green consumption, and green environment [8,9]. Conversely, in measuring employment levels, scholars typically employ unemployment rates as a straightforward gauge of employment scale [10] and the proportion of employment in secondary and tertiary industries as a simple indicator of employment structure [11,12]. Concerning the analysis of employment impact mechanisms, some scholars contend that environmental regulations exert a direct employment creation effect [13], potentially increasing jobs by influencing corporate market share [14,15], production costs [16] and industrial agglomeration levels [17]. Furthermore, technological progress can enhance enterprises’ green governance capabilities, thereby driving employment growth [18]. As research deepens, the academic focus has gradually shifted from employment scale to employment quality. One perspective posits that green and low-carbon development achieves high-quality employment through the transmission effects of productivity gains and structural optimizationoptimization [19]. Other studies reveal that environmental regulations can catalyze green technological innovation [20,21], mitigating the tension between environmental protection and employment by creating green jobs [18], thereby fostering higher-quality employment. Additionally, some scholars contend that environmental regulations exert threshold effects on high-quality employment through human capital and labor market segmentation [22]. Research has been conducted at various scales, regions, and sectors, including national level [7,10,19], provincial level [3,23,24], city cluster level [25], city level [20,26,27], central China region [28], and industrial sectors [18].

While existing research provides a solid foundation for constructing this paper’s indicator system and analyzing impact mechanisms, several shortcomings remain: Firstly, most studies focus on environmental regulation from a policy perspective or examine employment impacts through a comprehensive lens of green and low-carbon development, with limited research from the perspective of green industrial development. However, green industrial development inevitably influences industrial structure, which is intrinsically linked to employment patterns. Secondly, research subjects predominantly concentrate on macro and meso levels, lacking micro-level studies using county-level samples. Counties occupy a pivotal position in bridging national and local levels, serving as critical junctures for economic development, livelihood safeguarding, stability maintenance, and promoting long-term national security. Thirdly, the existing literature primarily examines employment levels through employment scale and employment opportunities, focusing on the breadth of these opportunities. As a result, the comprehensive indicator system for high-quality employment remains underdeveloped. Building upon research on employment opportunities and structure, this paper adds four more dimensions: labor remuneration, employment and social security, employability, and employment environment. This allows a comprehensive assessment of high-quality employment at the county level. The total number of employed people across society is used to measure employment opportunities. This method avoids problems from differing statistical approaches, which can make actual unemployment figures hard to quantify.

The contributions of this paper are primarily reflected in the following three aspects. Firstly, adopting a green industrial development perspective, it systematically analyzes the intrinsic mechanisms through which green industries shape the labor market by influencing industrial structure and technological progress. This expands the research dimensions of employment impact mechanisms at the industrial level, providing new empirical evidence for understanding the micro-level transmission pathways of employment under the Porter Hypothesis. Second, it shifts the research focus to the county level—a crucial micro-governance unit—by examining 20 mountainous counties in Zhejiang Province. Its findings not only inform Zhejiang’s efforts to promote coordinated regional development and common prosperity but also offer a micro-level case study for regions globally grappling with similar trade-offs between ecological conservation and economic growth. Thirdly, a multidimensional comprehensive evaluation framework for high-quality employment at the county level has been developed. This approach transcends traditional research limitations that focus solely on employment scale or a single-dimensional structure. For the first time at the county level, it empirically reveals the dual mediating mechanisms of industrial upgrading and technological progress, alongside their dual threshold effects grounded in technological advancement. Building on this, this paper utilizes panel data from 20 mountainous counties in Zhejiang Province, spanning the period from 2010 to 2023. Utilizing bidirectional fixed-effect, mediation effect, and threshold effects model, this study systematically investigates three questions: What direct impact does green industrial development exert on high-quality employment? What mediating roles do industrial structure upgrading and technological progress play in this context? Does this impact exhibit a nonlinear threshold effect depending on the level of technological progress? This research aims to provide theoretical underpinnings for achieving synergistic high-quality development and employment at the county level.

2. Theoretical Mechanism Analysis

2.1. Direct Effects of Green Industry Development on High-Quality Employment

The direct impact of green industrial development on high-quality employment is a tension-laden process of “creative destruction.” This process encompasses both disruptive substitution effects, such as the replacement of traditional industries and jobs, and innovative compensatory effects, including the creation of new types of employment, skills, and industries. The disruptive substitution effect is primarily driven by the costs associated with environmental regulations and the responses of companies. On the one hand, the Porter hypothesis suggests that stringent energy conservation and emission reduction rules in the early stages of the green transition increase compliance costs for traditional heavy industries, such as chemical manufacturing. This reduces their profit margins and can lead to job losses when production is scaled back [29,30]. Enterprises may face additional costs for pollution control. To offset these costs, they may pass part of the expense to workers. This can result in lower wages and reduced benefits for employees [31,32]. The combined effect of these two mechanisms will suppress improvements in overall regional employment quality in the short term and may induce risks of structural unemployment [19]—that is, unemployment arising from mismatches between labor supply and demand in terms of skills, geography, or industry due to changes in the economic structure. From the perspective of creative compensation effects, firstly, green industries, responding to environmental regulations and market demands, will generate a series of high-productivity, low-pollution green jobs [33]. These emerging positions not only directly expand employment scale but also, being typically embedded within knowledge-intensive and technology-intensive sectors, provide workers with higher-quality employment opportunities. Secondly, the development of green industries, by fostering green and low-carbon lifestyles, can effectively stimulate demand for green consumption [34]. According to Keynesian effective demand theory, expanded consumption demand will trigger production expansion, thereby creating more employment opportunities in related service, manufacturing, and R&D sectors, forming a virtuous cycle from demand-driven growth to job creation. Finally, from a human capital theory perspective, numerous positions in traditional industries are characterized by low skills and low benefits. The advancement of green industries drives industrial restructuring towards high-end and intelligent transformation, objectively requiring and incentivizing the workforce to upgrade skills and accumulate human capital. This process helps alleviate structural unemployment by facilitating labor mobility towards better-paid, more secure, and higher-potential positions, thereby achieving a systemic improvement in employment quality.

The net effect of green industry development on high-quality employment hinges on the trade-off between its job creation and displacement effects. Through top-level design, China has deeply integrated its green transition with its employment-first strategy. By implementing effective skills training and supportive industrial policies, it has mitigated the growing pains of industrial greening, with a net positive effect projected. Accordingly, this paper proposes Research Hypothesis 1.

H1. 

Green industrial development contributes to promoting high-quality employment.

2.2. The Mechanism Through Which Green Industry Development Promotes High-Quality Employment

Green industry development not only exerts a direct influence on high-quality employment but also achieves dual benefits of environmental protection and high-quality employment for workers through the transmission mechanisms of industrial structure optimization and technological advancement.

Green industrial development promotes high-quality employment through industrial structure optimization. Existing research indicates that green industrial development primarily achieves industrial structure optimization via two pathways: transforming traditional industries and fostering emerging industries [35]. Optimizing industrial structure enhances the coordination of factor inputs and output allocation within enterprises, thereby improving resource allocation efficiency and productivity [36], This generates sufficient operating profits to drive increases in job numbers, workplace benefits, and worker skills training, ultimately advancing high-quality employment. Furthermore, according to the Kuznets and Pareto-Clark theorems, labor flows from primary and secondary sectors to the tertiary sector at certain stages of economic development [37], The tertiary sector possesses greater labor absorption capacity while offering higher employment quality. Based on this, Hypothesis 2 is proposed.

H2. 

Green industrial development promotes high-quality employment through industrial structure optimization.

Green industrial development can foster high-quality employment through technological advancement. To foster the development of green industries, well-designed environmental regulations can stimulate corporate innovation and drive sustainable growth. These innovations enhance efficiency through technology, which helps offset compliance costs. This results in a win–win for both environmental protection and technological advancement [38]. According to the theory of economies of scale, technological progress enhances production efficiency and lowers unit costs. As a result, enterprises can expand their production scale and directly stimulate labor demand growth. When corporate profits and competitiveness increase, workers’ remuneration and welfare benefits may also rise [39]. The adoption of green technologies helps enterprises build an environmentally conscious image. It also enables them to gain competitive advantages and expand market share by offering more distinctive green products. This creates additional employment opportunities that meet high-quality standards [40]. Accordingly, this paper proposes Research Hypothesis 3.

H3. 

Green industrial development fosters high-quality employment through technological advancements.

2.3. The Nonlinear Impact of Green Industry Development on High-Quality Employment

As analysed above, green industry development promotes high-quality employment through technological progress. However, its impact varies across counties with differing technological levels, potentially indicating the existence of certain thresholds. The evolution of green technology from introduction and assimilation to independent innovation is not uniform, with critical junctures in capability accumulation. Initially, technology undergoes an ‘introduction and trial-and-error phase’ where its transformation of production systems remains unstable. As technical capabilities gradually build up, the internalization of technology triggers systemic innovation, laying the potential foundation for a sudden shift in employment effects. Secondly, during the low-technology phase, green investment primarily manifests as ‘compliance costs,’ which inhibit corporate scale expansion and yield minimal employment effects. Once technological maturity crosses critical thresholds, mechanisms for cost reduction through productivity gains and market expansion are activated, revealing pronounced employment-boosting effects. Finally, the skill-biased technological progress theory explains how employment structures undergo a qualitative transformation accordingly. At low technological levels, the application of green technology primarily replaces routine operational roles in traditional industries, while creating low-skilled supervisory positions. At this stage, technological progress exhibits low-skill or neutral bias, with its promotion of high-quality employment offset by structural displacement. As green technologies advance, progress shifts from bias towards skill complementarity, generating substantial technical worker roles and driving industrial restructuring. Consequently, the effects of employment promotion steadily intensify. Accordingly, this paper proposes Research Hypothesis 4.

H4. 

The impact of green industry development on high-quality employment exhibits a threshold effect linked to technological progress.

In summary, the mechanism through which green industry development influences high-quality employment is illustrated in Figure 1.

3. Research Design

3.1. Variable Selection

3.1.1. Explanatory and Dependent Variables

The explanatory variable is the Green Industry Development level (GID). Green industry encompasses not merely a single production stage, but rather a comprehensive lifecycle system spanning resource input, manufacturing, consumption, and environmental impact. Drawing upon the research of Zhou Ying et al. [8] and Shi Baofeng et al. [9], and prioritizing scientific rigor, comprehensiveness, significance, and data accessibility, this study constructs a county-level evaluation framework for green industrial development comprising 12 secondary indicators across three dimensions: green industry, green consumption, and green environment. Green industry constitutes the supply side of green industrial development, reflecting the sector’s own degree of green transformation; green consumption represents the demand side, serving as the driving force for green industrial development; while green environment signifies the ecological outcome, measuring the actual improvement effects of green industrial development on the natural environment. Building upon this framework, the entropy weighting method was employed to assign weights to each indicator, as detailed in

Table 1. Indicator System for the Level of Green Industry Development in Counties.

Target Layer	Criterion Layer	Indicator Layer	Weights	Unit	Attribute
Green industry development level	Green industry	The proportion of the tertiary sector’s total output value in GDP	0.0244	%	+
		Industrial electricity consumption	0.0569	10,000 kWh	−
		Industrial smoke (and dust) emissions per unit of GDP	0.0151	ton	−
		Industrial sulphur dioxide emissions per unit of GDP	0.0429	ton	−
	Green consumption	The proportion of science and technology expenditure within the general public budget expenditure	0.1860	%	+
		Domestic electricity consumption for urban and rural residents	0.0438	10,000 kWh	−
		Year-end public bus and tram operational vehicles	0.2803	vehicle	+
		Daily per capita domestic water consumption	0.0136	litre	−
		Rate of harmless treatment of domestic waste	0.0057	%	+
	Green environment	Wastewater treatment rate	0.0445	%	+
		Per capita park green space area	0.2388	square metre	+
		Green coverage rate in built-up areas	0.0480	%	+

.

The dependent variable is the High-quality employment level (HQE). High-quality employment is a goal centered on workers’ wellbeing, advocating for equal opportunities and shared prosperity, thereby achieving simultaneous optimization of both employment scale and structure. This paper builds on previous research [10,11] to create a county-level high-quality employment indicator system with 14 secondary indicators across six dimensions: employment opportunities, employment structure, labor remuneration, employment and social security, employability, and employment environment. Employment opportunities are the basis and starting point for quality jobs. They indicate the likelihood of people finding paid work and reflect the market’s openness. The employment structure illustrates how jobs are distributed across various industries, sectors, and roles. Labor pay represents the key value and incentive for quality jobs, directly affecting workers’ living standards and dignity. Employment and social security provide a reliable foundation and stability for quality jobs. They provide workers with support in the event of unemployment, illness, retirement, or other risks. Employment capability forms the core for growth and ongoing progress in quality jobs, shaping both personal career growth and the market’s ability to adjust to change. The employment environment encompasses the external conditions and support that facilitate quality jobs, including public services, infrastructure, rights protection, and favorable working conditions. These shape workers’ experience and future growth. A positive work environment lowers obstacles and boosts both efficiency and job appeal. Building upon this framework, Entropy weighting was applied to assign weights to each indicator, as detailed in

Table 2. Indicator System for the Level of High-quality employment in Counties.

Target Layer	Criterion Layer	Indicator Layer	Weights	Unit	Attribute
High-quality employment level	Job opportunities	Total employment across society	0.0608	ten thousand people	+
	Employment Structure	Share of the population employed in the secondary sector	0.0084	%	−
	Labor remuneration	Income disparity between urban and rural residents	0.0352	yuan	−
		Urban residents’ income	0.0615	yuan	+
		Rural residents’ income	0.0818	yuan	+
		Growth rate of urban residents’ income	0.0165	%	+
		Rural residents’ income growth rate	0.0974	%	+
	Employment and social security	Number of participants in the basic pension insurance scheme	0.0987	ten thousand people	+
		Number of people covered by basic medical insurance	0.1035	ten thousand people	+
		Number of people covered by unemployment insurance	0.1521	ten thousand people	+
	Employment capacity	The proportion of education expenditure in general public budget expenditure	0.0206	%	+
	Employment environment	Number of doctors per 10,000 people	0.0635	per 10,000 people	+
		Total number of specialized vehicles and equipment for urban sanitation	0.1040	platform	+
		Total holdings of books in public libraries	0.0960	ten thousand volumes	+

.

3.1.2. Mediating Variables

This study selects industrial structure (IS) and technological progress (TP) as mediating variables. Drawing upon Yan Wenjuan et al. [41], the ratio of tertiary to secondary industry output is employed to measure industrial structure. Research and development (R&D) serves as the direct driver of innovation, with R&D expenditure reflecting a region’s level of technological advancement.

3.1.3. Control Variables

Drawing upon existing literature [42,43,44], the following control variables are selected: (1) Urbanization Level (Urban). Higher urbanization increases green job creation, thereby promoting high-quality employment, measured by the urbanization rate. (2) Government Intervention (Gov). Governments can influence high-quality employment through employment and industrial policies, measured by the proportion of local fiscal expenditure relative to regional GDP in the current year. (3) Resource Density (Density). Regions with high resource density tend to foster industrial clusters, which in turn influence high-quality employment. This is measured by population density. (4) Economic Development Level (GDPG). Accelerated economic development typically involves the expansion and establishment of high-tech enterprises, which increases demand for high-quality labor and enhances full employment. The regional GDP growth rate measures this.

3.2. Data Processing and Measurement Methods

This paper consolidates indicator data on green industry development and high-quality employment levels for 20 mountainous counties in Zhejiang Province from 2010 to 2023. Dongtou and Shengsi, two island counties, are excluded due to data availability constraints. Data originate from official publications such as the Zhejiang, Wenzhou, Taizhou, Jinhua, Quzhou, and Lishui Statistical Yearbooks. It also uses publicly available data from the National Economic and Social Development Statistical Bulletins of each county. Based on research objectives and data characteristics, three steps were undertaken. First, a panel dataset was constructed for 2010 to 2023. Directly obtainable data were collated, including sewage treatment rate, domestic waste harmless treatment rate, and green coverage rate in built-up areas. Indirect calculations filled gaps for unavailable data, including the proportion of tertiary industry output value in GDP, the share of education expenditure in general public budget expenditure, and the proportion of the population employed in the secondary industry. Second, missing data points for a few years or counties were interpolated linearly, as this method is suitable for panel data with stable temporal trends. Third, indicators with extensive missing values, unsuitable for interpolation, were removed, excluding years with significant gaps, which resulted in unbalanced panel data. Indicator weights were determined using the entropy weighting method [45]. This method assigns weights automatically based on data dispersion, reducing human bias and suiting analyses with multiple indicators and diverse samples. Its results are unaffected by expert bias, unlike subjective weighting methods such as the Analytic Hierarchy Process. Compared to objective methods like Principal Component Analysis, entropy weighting is more intuitive and preserves the independent economic meaning of each indicator. Robustness tests re-measure explained variables with Principal Component Analysis. Core conclusions remain consistent. Descriptive statistics for each variable are presented in

Table 3. Descriptive Statistics.

Variance	Sample Size	Mean	Standard Error	Minimum	Median	Maximum
HQE	269	0.2670	0.0872	0.0934	0.2620	0.5130
GID	269	0.3310	0.0691	0.2160	0.3190	0.6510
Urban	269	0.3630	0.1400	0.0660	0.3620	0.7540
Gov	269	0.3100	0.1480	0.0912	0.2750	1.0430
Density	269	64.3900	37.2200	7.4530	48.8300	176.3000
GDPG	269	7.7490	2.8340	−4.1000	7.8000	14.3000
IS	269	1.2640	0.6040	0.3040	1.1710	4.3160
TP	269	8.5290	0.6930	6.9640	8.5170	10.1500

.

3.3. Model Construction

3.3.1. Benchmark Regression Model

This study employs county-level data to construct a panel dataset, utilizing a two-way fixed-effects model for regression analysis to examine the impact of green industrial development on high-quality employment. The benchmark regression model is constructed as shown in Equation (1).

$${\mathrm{HQE}}_{\mathrm{it}}=\mathsf{\alpha }+{\mathsf{\beta }}_1{\mathrm{GID}}_{\mathrm{it}}+{\mathsf{\beta }}_2{\mathrm{Control}}_{\mathrm{it}}+{\mathsf{\mu }}_{\mathrm{i}}+{\mathsf{\lambda }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{it}}$$

(1)

where ${\mathrm{HQE}}_{\mathrm{it}}$ denotes the level of high-quality employment in region $i$ during period $t$, ${\mathrm{GID}}_{\mathrm{it}}$ represents the level of green industrial development in region $i$ during period $t$, ${\mathrm{Control}}_{\mathrm{it}}$ is the control variable; $\mathsf{\alpha }$ is the constant term, ${\mathsf{\beta }}_1$ and ${\mathsf{\beta }}_2$ are the regression coefficients for each variable; ${\mathsf{\mu }}_{\mathrm{i}}$ is the individual fixed effect, ${\mathsf{\lambda }}_{\mathrm{t}}$ is the time-fixed effect, and ${\mathsf{\epsilon }}_{\mathrm{it}}$ is the random error term.

3.3.2. Mediating Effect Model

To further investigate the mechanism through which green industrial development influences high-quality employment, this study adopts the two-step regression method proposed by Jiang Ting [46] to test for mediating effects, focusing on whether a relationship exists between the explanatory variable and the mediating variable. The constructed models are expressed as Equations (2) and (3).

$${\mathrm{IS}}_{\mathrm{it}}={\mathsf{\gamma }}_0+{\mathsf{\gamma }}_1{\mathrm{GID}}_{\mathrm{it}}+{\mathsf{\gamma }}_2{\mathrm{Control}}_{\mathrm{it}}+{\mathsf{\mu }}_{\mathrm{i}}+{\mathsf{\lambda }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{it}}$$

(2)

$${\mathrm{TP}}_{\mathrm{it}}={\mathsf{\delta }}_0+{\mathsf{\delta }}_1{\mathrm{GID}}_{\mathrm{it}}+{\mathsf{\delta }}_2{\mathrm{Control}}_{\mathrm{it}}+{\mathsf{\mu }}_{\mathrm{i}}+{\mathsf{\lambda }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{it}}$$

(3)

Among these, ${\mathrm{IS}}_{\mathrm{it}}$ represents the industrial structure, and ${\mathrm{TP}}_{\mathrm{it}}$ represents the technological progress.

3.3.3. Panel Threshold Model

As a quantitative model for analyzing whether structural breaks occur in economic parameters, the panel threshold model not only verifies the relationship between green industrial development and high-quality employment but also calculates the threshold value at which structural breaks occur. If only a single threshold exists, the single-threshold model is employed, as in Equation (4).

$${\mathrm{HQE}}_{\mathrm{it}}=\mathsf{\alpha }+{\mathsf{\theta }}_1{\mathrm{GID}}_{\mathrm{it}}\cdot \mathrm{I}({\mathrm{TP}}_{\mathrm{it}}\le \mathrm{k})+{\mathsf{\theta }}_2{\mathrm{GID}}_{\mathrm{it}}\cdot \mathrm{I}({\mathrm{TP}}_{\mathrm{it}}>\mathrm{k})+{\mathsf{\theta }}_3{\mathrm{Control}}_{\mathrm{it}}+{\mathsf{\mu }}_{\mathrm{i}}+{\mathsf{\lambda }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{it}}$$

(4)

Among these, ${\mathrm{GID}}_{\mathrm{it}}$ represents the core explanatory variable, ${\mathrm{TP}}_{\mathrm{it}}$ represents the threshold moderator variable, and $\mathrm{k}$ represents the threshold value to be estimated.

Should two thresholds exist, Equation (4) is extended to Equation (5).

$$\begin{array}{l}{\mathrm{HQE}}_{\mathrm{it}}=\mathsf{\alpha }+{\mathsf{\theta }}_1{\mathrm{GID}}_{\mathrm{it}}\cdot \mathrm{I}({\mathrm{TP}}_{\mathrm{it}}\le {\mathrm{k}}_1)+{\mathsf{\theta }}_2{\mathrm{GID}}_{\mathrm{it}}\cdot \mathrm{I}({\mathrm{k}}_1<{\mathrm{TP}}_{\mathrm{it}}\le {\mathrm{k}}_2)+\\ {\mathsf{\theta }}_3{\mathrm{GID}}_{\mathrm{it}}\cdot \mathrm{I}({\mathrm{TP}}_{\mathrm{it}}>{\mathrm{k}}_2)+{\mathsf{\theta }}_4{\mathrm{Control}}_{\mathrm{it}}+{\mathsf{\mu }}_{\mathrm{i}}+{\mathsf{\lambda }}_{\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{it}}\end{array}$$

(5)

Among these, ${\mathrm{k}}_1$ and ${\mathrm{k}}_2$ represent two threshold values that divide the total sample into three intervals. ${\mathsf{\theta }}_1$, ${\mathsf{\theta }}_2$ and ${\mathsf{\theta }}_3$ denote the respective coefficients measuring the impact of green industry development on high-quality employment across these three intervals.

4. Overview of the Study Area

The 26 mountainous counties of Zhejiang are primarily distributed in the southwestern region, characterized predominantly by mountainous and hilly terrain, presenting a landscape pattern of ‘eight parts mountain, one part water, and one part farmland’. Covers 26 counties under Quzhou and Lishui cities. Includes counties, cities, districts, and areas such as Chun’an and Yongjia. (as shown in Figure 2). These areas account for approximately 45% of Zhejiang’s land area and house about 15% of its permanent population. This region boasts an abundance of ecological resources, including forests and lakes. These resources provide a solid foundation for developing ‘beautiful economy’ initiatives, including ecotourism and ecological agriculture [34]. However, socio-economic development has long lagged behind the provincial average, constrained by the region’s remote location and inadequate transport infrastructure. To promote coordinated regional development, Zhejiang Province has persistently implemented policies such as the Mountain-Sea Collaboration initiative. It has also established a dynamic adjustment mechanism for “mountainous and island counties”. In 2024, Pingyang County, Kecheng District, and Liandu District were removed from the list, with Dongtou County and Shengsi County added. In 2025, Yongjia County, Cangnan County, and Jiangshan City were further removed, ultimately forming a working system comprising 22 mountainous and island counties. The Central Committee attaches great importance to county governance, recognizing counties as fundamental units of national administration and microfoundations of economic and social development. Their role is indispensable in achieving common prosperity. As a pioneer zone for green development in China, Zhejiang’s mountainous counties offer a typical case study for this research. These counties have explored green transformation under ecological constraints [26].

4.1. Analysis of Temporal Trends

As shown in Figure 3, the level of green industry development across 20 mountainous counties in Zhejiang Province has shown an overall upward trend, rising from 0.287 in 2010 to 0.409 in 2023. This progression can be divided into two distinct phases: a fluctuating increase from 2010 to 2016, with a growth rate of 6.12%, followed by a period of steady acceleration from 2016 to 2023, during which growth reached 34.42%. This indicates significant progress in the region’s green industry development during the observation period. Initially, development was relatively slow due to the absence of targeted provincial policies, relying primarily on local initiatives. Following 2016, Zhejiang Province successively introduced specialized policies such as the Zhejiang Green Industry Cultivation Action Plan and Several Measures to Support High-Quality Ecological Industrial Development in 26 Mountainous Counties. These initiatives integrated green industry development with the construction of a demonstration zone for common prosperity. Furthermore, the ‘mountain-sea industrial chain collaboration’ mechanism facilitated pairing between districts and counties, introducing technology, capital, and management expertise. This approach effectively catalyzed the rapid expansion of the green industry. The level of high-quality employment fluctuated upwards between 2010 and 2021, before experiencing a sharp decline from 2021 to 2023. This downturn stems from the greater ecological constraints on employment within Zhejiang’s 20 mountainous counties, coupled with their excessive reliance on policy subsidies, rendering the employment situation vulnerable. The restructuring of the service sector due to the pandemic in 2021, combined with the overly rapid reduction in policy subsidies, contributed to the decline in high-quality employment.

4.2. Analysis of Spatial Evolution Trends

This paper utilizes ArcGIS 10.8 to spatially visualize the levels of green industrial development and high-quality complete employment across 20 mountainous counties in Zhejiang Province (as shown in Figure 4). Overall, both green industry development and high-quality employment exhibit regional differentiation and hierarchical distribution across counties. In terms of green industrial development, Chun’an County and Kaihua County have long maintained a leading position, while Qingdao County and Jinyun County lag relatively behind. Wuyi County and Yunhe County, however, demonstrated the most significant growth during the study period. The reasons for this are as follows: Chun’an and Kaihua, as national-level key ecological functional zones such as the Qiandao Lake water source area and the Qianjiang Source Protection Area, benefit from stringent ecological compensation and industrial access policies, providing a solid foundation for green development. Qingdao and Jinyun, constrained by short industrial chains and an outdated industrial structure, have seen slow progress in their green transformation. Wuyi County, however, has implemented the Provincial Pilot Implementation Plan for Ecological Industrial Development, focusing on ecological industry and circular economy to drive targeted advancement. Yunhe County has deepened reforms to realise the value of ecological products, leveraging the ‘Two Mountains Cooperative’ and municipal ecological product trading platforms to effectively foster rural green industries. Regarding high-quality employment levels, Chun’an, Qingdao, and Jinyun counties have maintained consistent leadership, while Qingyuan and Jingning counties have long lagged behind. Xianju and Tiantai counties demonstrated the most rapid development during the study period. This may be attributable to the following factors: Chun’an, Qingdao and Jinyun, being adjacent to the core metropolitan area, possess relatively well-developed infrastructure, making them more receptive to industrial spillover effects; Qingyuan and Jingning, constrained by relatively inaccessible transport links, face high logistics costs and weak industrial support systems, severely limiting industrial expansion and employment growth; Xianju and Tiantai, through continuously broadening employment channels and strengthening employment support measures, have effectively driven the growth of high-quality employment locally.

5. Impact of Green Industry Development on High-Quality Employment

5.1. Benchmark Regression Results

The benchmark regression results are presented in

Table 4. Benchmark Regression Results on the Impact of Green Industry Development on High-quality employment in 20 Mountainous Counties of Zhejiang Province.

	(1)	(2)	(3)	(4)	(5)
GID	0.1360 ***	0.1260 ***	0.1270 ***	0.1270 ***	0.1700 ***
	(2.7040)	(2.6740)	(2.7280)	(2.7550)	(3.6830)
Urban		0.1240 ***	0.1180 ***	0.1200 ***	0.1280 ***
		(5.7640)	(5.5620)	(5.6590)	(6.1920)
Gov			−0.0934 ***	−0.0945 ***	−0.1070 ***
			(−2.6040)	(−2.6470)	(−3.0860)
GDPG				0.0020 *	0.0019 *
				(1.8430)	(1.7940)
Density					−0.0006 ***
					(−3.8680)
Constant	0.0943 ***	0.0586 ***	0.0801 ***	0.0534 **	0.0800 ***
	(5.7760)	(3.5480)	(4.3810)	(2.2940)	(3.3870)
Individual fixed effects	YES	YES	YES	YES	YES
Time-fixed effect	YES	YES	YES	YES	YES
Obs	269	269	269	269	269
R2	0.8250	0.8460	0.8510	0.8530	0.8620

Note: Values in parentheses denote t-statistics. *** p < 0.01, ** p < 0.05, * p < 0.1.

. Models (1)–(5) sequentially incorporate control variables. The findings reveal that green industry development consistently exerts a significant positive influence on high-quality employment. Specifically, the coefficient for the level of green industry development ranged between 0.1260 and 0.1700 across different models. This indicates that for each unit increase in the level of green industry development, the level of high-quality full employment significantly improved by approximately 0.1260–0.1700 units. This indicates that green industry development directly promotes high-quality employment, providing preliminary support for H1. This conclusion aligns with Maczulskij (2024) finding that green employment possesses a generative mechanism [19]. The process may begin with the abundant ecological resources in Zhejiang’s 20 mountainous counties, which provide inherent advantages for developing green industries such as ecotourism and eco-agriculture. The development of these industries leads to the creation of numerous green jobs, which in turn drive green technological innovation and promote the optimization and upgrading of industrial structures. These changes collectively enhance high-quality employment. Moreover, the regression coefficients after incorporating all control variables and fixed effects exceed those of the linear relationship between green industrial development and high-quality employment. This indicates that the influence of these control variables obscures the relationship between the two. Consequently, Column (5) more effectively eliminates interference from relevant factors, rendering the relationship between green industrial development and high-quality employment more stable and pronounced.

After controlling for other variables, urbanization levels also exert a significant positive influence on high-quality employment, the coefficient lies between 0.1180 and 0.1280, indicating that newly created green jobs during urbanization processes promote high-quality employment. Government intervention, however, exerts a significant negative impact on high-quality employment, the coefficient lies between −0.0934 and −0.1070. This may stem from governments prioritizing ‘scale leaps’ in green industries while neglecting employment inclusivity, thereby triggering the paradox of ‘industrial upgrading leading to employment downgrading’. The level of economic development exerts a significantly positive influence, indicating that economic advancement stimulates the expansion and establishment of high-tech enterprises, thereby increasing demand for high-quality labor and elevating levels of high-quality employment. Resource concentration, conversely, exhibits a significant negative impact. This may stem from administratively driven excessive resource concentration, triggering dual pressures of the ‘siphon effect’ and ‘capital deepening,’ which undermine the employment inclusiveness of green industries.

5.2. Robustness Tests

To validate the robustness of the benchmark regression results, two distinct methods were employed, with findings presented in

Table 5. Robustness Test Results.

	Replacing Variable Assignment Methods	Tail-End
	(1)	(2)
GID	1.1130 **	0.1820 ***
	(1.9940)	(3.7940)
Urban	1.3530 ***	0.1280 ***
	(5.4210)	(6.1740)
Gov	1.1220 ***	−0.1130 ***
	(−2.6740)	(−3.0290)
GDPG	0.0062	0.0022 *
	(0.4800)	(1.9340)
Density	0.0065 ***	−0.0006 ***
	(−3.3640)	(−3.8350)
Constant	1.9000 ***	0.0763 ***
	(−6.6710)	(3.0660)
Individual fixed effects	YES	YES
Time-fixed effect	YES	YES
Obs	269	269
R2	0.9100	0.8640

Note: Values in parentheses denote t-statistics. *** p < 0.01, ** p < 0.05, * p < 0.1.

.

Firstly, the calculation method for the high-quality employment indicator—the dependent variable—was adjusted. Principal component analysis was used to calculate the composite score, as shown in Column (1). The key finding is that green industry development increases high-quality full employment by a coefficient of 1.1130, which is significant at the 5% level. The direction and significance of this effect are consistent with benchmark results, confirming that the conclusions are robust despite changes in measurement methodology. Secondly, the sample data underwent tail trimming to mitigate the effects of extreme values. Column (2) shows that green industry development positively and significantly affects high-quality employment, with a coefficient of 0.8120 at the 1% level. The positive direction and significance remain, confirming robust results. In summary, the results of both robustness tests support the conclusion that green industrial development can significantly promote high-quality employment. The findings are robust and reliable, and H1 remains valid.

5.3. Endogeneity Testing

To further examine the dynamic relationship between green industrial development and high-quality employment, as well as potential reverse causality and omitted variable bias, this study employs an instrumental variables approach for regression analysis. The results are presented in

Table 6. Results of Endogeneity Tests.

	Explanatory Variable Lagged by One Period (1)	Instrumental Variables Estimation (2)
L.GID	0.869 ***
	(20.75)
GID		0.163 ***
		(2.839)
Urban	−0.0122	0.124 ***
	(−0.712)	(6.082)
Gov	0.00967	−0.0953 ***
	(0.334)	(−2.771)
GDPG	−0.00122	0.00164
	(−1.351)	(1.532)
Density	0.000140	−0.000617 ***
	(1.052)	(−3.849)
Constant	0.0325 *	0.0800 ***
	(1.704)	(3.387)
Individual fixed effects	YES	YES
Time-fixed effect	YES	YES
Obs	249	249
R2	0.853	0.828

Note: Values in parentheses denote t-statistics. *** p < 0.01, ** p < 0.05, * p < 0.1.

. Drawing upon the research approach of Wei et al. [47], the lagged level of green industry development (L.GID) is introduced as an explanatory variable in the regression. Column (1) indicates that the lagged level of green industry development exerts a significant positive influence on the current level of green industry development, the coefficient is 0.869, suggesting strong inertia and path dependence in green industry development. Column (2) employs green industrial development as the explanatory variable and high-quality employment as the dependent variable for instrumental variable estimation. The results demonstrate a significant positive impact of green industrial development on high-quality employment; the coefficient is 0.163. This indicates that, after controlling for endogeneity, the promotional effect of green industrial development on high-quality employment remains robust and reliable. Furthermore, results for other control variables are broadly consistent with the preceding analysis. Test results reveal an unidentifiability test statistic of 153.4390 with a p-value of 0.0000, significantly rejecting the null hypothesis of unidentifiability. The weak instrumental variable test statistic of 430.5040 substantially exceeds the critical value, confirming the strong relevance of the instrument and ruling out weak instrumentation issues.

In summary, the robust instrumental variable regression results further substantiate H1.

5.4. Heterogeneity Analysis

Given the significant variations in industrial structure across counties, this study conducts a heterogeneity analysis centered on industrial composition.

Table 7. Results of Heterogeneity Tests.

	Dominated by the Secondary Sector	Dominated by the Tertiary Sector
	(1)	(2)
GID	0.0800	0.4630 ***
	(1.5980)	(3.6030)
Urban	0.1070 ***	0.1270 **
	(4.7980)	(2.4130)
Gov	−0.0559	−0.3520 ***
	(−1.5830)	(−2.8660)
GDPG	0.0015	0.0013
	(1.0420)	(0.5200)
Density	−0.0005 ***	−0.0006
	(−3.1430)	(−0.3820)
Constant	0.0949 ***	0.0352
	(3.4510)	(0.3510)
Individual fixed effects	YES	YES
Time-fixed effect	YES	YES
Obs	173	96
R2	0.8530	0.8860

Note: Values in parentheses denote t-statistics. *** p < 0.01, ** p < 0.05, * p < 0.1.

examines the heterogeneous impact of green industry development on high-quality employment in counties with differing industrial structures. The 20 mountainous counties in Zhejiang are categorized into those dominated by the tertiary sector and those dominated by the secondary sector (where the secondary sector accounts for over 50% of the industrial structure).

For counties predominantly reliant on secondary industries, the coefficient for the impact of green industry development on high-quality employment stands at 0.0800. Whilst positive, it fails to reach statistical significance, hovering only around the 10% level. Conversely, in counties dominated by the tertiary sector, the impact of green industrial development on high-quality employment significantly intensifies, with the coefficient rising to 0.463 and achieving statistical significance at the 1% level. This indicates that the level of green industrial development exerts differential effects on high-quality employment across different industrial types, with stronger impacts observed in counties possessing more optimized industrial structures. This conclusion aligns with Li and Du (2022)’s finding of structurally heterogeneous effects of environmental regulations on employment [13]. The reason may lie in the fact that industrial-dominated countries often have a concentration of more high-pollution, high-energy-consumption industries. Their industrial green transition requires higher investment in energy conservation, emission reduction, and green technological innovation, thereby constraining the enhancement of high-quality employment to some extent. By contrast, counties dominated by the tertiary sector primarily rely on low-energy-consumption industries, such as services and tourism, which entail lower green transition costs. Concurrently, these counties typically exhibit faster economic development and stronger talent attraction, creating more favorable conditions for advancing green technological innovation and industrial transformation, thereby fostering the growth of high-quality employment.

5.5. Testing the Mechanism of Action

This paper employs the two-step regression method proposed by Jiang Ting [46] to examine the mediating effect, building upon the prior validation of the impact of green industry development on high-quality employment. Existing research generally suggests that industrial structure and technological progress are key factors in achieving high-quality employment [40,41,43], If green industry development facilitates industrial upgrading and technological advancement, this would indirectly validate that industrial structure and technological progress constitute key mechanisms through which green industry development promotes high-quality employment. The regression results are presented in

Table 8. Results of Mediational Effect Tests.

Variable	HQE (1)	IS (2)	TP (3)
GID	0.1700 ***	1.6330 ***	1.6950 ***
	(3.6830)	(2.7000)	(4.2520)
Constant	0.0800 ***	0.1140	7.3440 ***
	(3.3870)	(0.3680)	(39.7400)
Individual fixed effects	YES	YES	YES
Time-fixed effect	YES	YES	YES
Obs	269	269	269
R2	0.8620	0.4340	0.8860

Note: Values in parentheses denote t-statistics. *** p < 0.01, ** p < 0.05, * p < 0.1.

.

Column (2) shows that with industrial structure as the mediator, the coefficient for green industrial development is 1.6330, positive and significant at the 1% level. Green industrial development enhances the share of the tertiary sector (industries that provide services rather than primary or secondary activities), driving the optimization and upgrading of the industrial structure. Column (3) shows that with technological progress as the mediator, the coefficient is 1.695, also positive and significant at the 1% level. Green industrial development strongly promotes technological progress. Together, these results demonstrate that green industrial development can indirectly promote high-quality employment by facilitating the optimization of industrial structure and technological progress, thereby supporting H2 and H3.

5.6. Testing Nonlinear Effects

5.6.1. Threshold Effect Analysis

The preceding theoretical mechanism analysis revealed that the impact of green industries at different technological levels on high-quality employment varies across countries, suggesting the potential existence of thresholds. Therefore, this study employs technological progress as the threshold variable to examine the nonlinear relationship between the two. Prior to conducting threshold regression, a threshold effect analysis was performed. Results presented in

Table 9. Threshold Effect Test Results.

Model	RSS	MSE	Fstat	Prob	Crit10	Crit5	Crit1
Single-threshold	0.4614	0.0023	51.8800	0.0000	12.9980	15.7000	20.1470
Double-threshold	0.3702	0.0019	48.5200	0.0000	12.7730	14.8650	20.4580
Triple-threshold	0.3442	0.0017	14.8900	0.6200	28.8020	31.5960	38.6680

indicate that both single-threshold and double-threshold effects are significant (p = 0.000, F far exceeds the critical value), whereas the triple-threshold effect is insignificant (p = 0.620, F far below the critical value). Consequently, a double-threshold regression model is adopted.

5.6.2. Analysis of Threshold Regression Results

Further exploration using panel threshold regression reveals, as shown in

Table 10. Threshold Regression Estimation Results.

Variable	Estimated Coefficient
TP (TP ≤ 8.0107)	0.0621
	(0.5550)
TP (8.0107 < TP ≤ 8.6739)	0.2980 ***
	(2.9770)
TP (TP > 8.6739)	0.4810 ***
	(4.9510)
Urban	0.1620 ***
	(4.2410)
Gov	0.0427
	(0.8680)
GDPG	−0.0050 ***
	(−3.0720)
Density	−0.000495
	(−1.2820)
Constant	0.1540 ***
	(3.7680)
R2	0.6430

Note: Values in parentheses denote t-statistics. *** p < 0.01, ** p < 0.05, * p < 0.1.

, that the marginal effects of green industries exhibit significant variations across different intervals.

Specifically, when the level of technological progress fails to meet the first threshold (set at 8.0107), the coefficient for green industrial development stands at 0.0621 but is not statistically significant. This suggests that at lower levels of technological advancement, the application of green technologies remains superficial, with limited capacity to reshape the overall employment structure, resulting in insignificant promotional effects. When technological progress levels fall between the first threshold (8.0107) and the second threshold (8.6739), the coefficient for green industry development rises to 0.2980 and becomes statistically significant at the 1% level. This indicates that as technological progress increases, green industry development begins to steadily generate substantial numbers of specialist technical positions, driving optimization of the employment structure. Consequently, the employment promotion effect emerges and steadily intensifies. When technological progress exceeds the second threshold, the coefficient for green industry development further increases to 0.4810, remaining statistically significant at the 1% level. This demonstrates that within higher technological advancement ranges, green industry development not only absorbs large numbers of highly skilled workers but also optimizes regional employment ecosystems through industrial chain spillover effects. Consequently, its role in promoting high-quality employment exhibits an accelerating trajectory. Overall, the impact of green industry development on high-quality employment exhibits a pronounced threshold effect. As technological progress increases, the promotional effect continues to strengthen, exhibiting a distinct J-shaped, nonlinear characteristic.

6. Discussion

This study confirms that green industrial development promotes high-quality employment through two mechanism pathways: industrial structure upgrading and technological progress. This finding resonates with and extends existing theories at the county level in China. Firstly, the mediating role of industrial structure upgrading validates the applicability of the Petty-Clark law in the context of green transition, indicating that labor reallocation occurs alongside the upgrading of industrial structure towards services and greening. The study further reveals that this pathway is particularly pronounced in counties dominated by the tertiary sector. Secondly, the mediating effect of technological progress—particularly its dual-threshold characteristic—provides micro-level evidence from developing economies in support of the classical Porter hypothesis. It further reveals that this innovation-driven employment boost is nonlinear, subject to critical capability accumulation thresholds. This indicates that green technologies must undergo a capability leap from introduction and digestion to independent innovation to fully realize their job creation potential. The heterogeneity analysis indicates that green industrial development exhibits negligible employment promotion effects in secondary-industry-dominated counties. This does not negate the value of green transition but sharply highlights the tangible risk of structural unemployment in heavy industrial regions. This finding aligns with the international discourse on a just transition. It implies that for areas with high reliance on legacy industries, green transition may entail significant job disruption and skills mismatch, with more acute short-term pain. Consequently, mere industrial greening policies, without robust labor market interventions (such as skills retraining and transitional social security), struggle to concurrently enhance employment quality in the short term. This offers a crucial regionally differentiated perspective for understanding and designing just transition policies.

The analysis distils multi-layered policy implications, centered on synergistically advancing green transition alongside employment quality enhancement, while implementing differentiated strategies based on transition stages and county-level foundations. Regarding short-term employment adjustment policies, the primary objective is to mitigate transition shocks and guard against structural unemployment risks. This necessitates a policy focus on the supply side of the labor market, particularly for industrial-dominated counties where green industrial development effects are less pronounced, and regions where technological advancement remains at a low threshold. From the perspective of a long-term green industrial strategy, the core goal is to cultivate endogenous momentum, enabling green industries to become a sustainable source of high-quality employment. It should be noted that this study retains the following limitations, which point to directions for future research:

(1)

The employment-boosting effects of green industrial development revealed herein are significantly attributable to Zhejiang Province’s proactive policy environment. In regions lacking comparable intensive policy support, the employment-generating impact of green industrial development may prove weaker, emerge more slowly, or require overcoming higher initial investment and technological barriers. Future research may focus on two directions: comparative studies and theoretical refinement. Firstly, applying this research framework to regions with varying policy support intensities and developmental stages, using comparative analysis to test and refine the generalizability of this paper’s conclusions. Secondly, incorporating policy support intensity or institutional quality as a moderating variable or precondition into the theoretical model, constructing an integrated analytical framework encompassing ‘policy environment—green industries—high-quality employment’ to provide a more universal explanation for regional differences.

(2)

Due to limited county-level social data, this study used proxy indicators for key constructs. These proxies do not always align well with their theoretical counterparts. Future research should seek more direct measurement tools to develop a more comprehensive and reliable measurement system. The study also uses the lagged level of green industry development as an instrumental variable to address reverse causality. While common in the literature and passing basic tests, this method has limits. Historical variables may be linked to omitted variables, thereby weakening exogeneity. Future research could utilize more exogenous instrumental variables, such as those based on geography or policy shocks, to provide stronger validation.

7. Conclusions

Against the backdrop of an era guided by the concept of sustainable development, aiming for high-quality growth and shared prosperity, this study employs panel data from 20 mountainous counties in Zhejiang Province, spanning the period from 2010 to 2023. Utilizing ArcGIS software, it examines the spatiotemporal evolution of green industrial development and high-quality employment levels. By analyzing both industrial structure upgrading and technological progress, it elucidates the nonlinear impact of green industrial development on high-quality employment and its underlying mechanisms, yielding the following conclusions:

(1)

During the study period, both the green industrial development level and the high-quality employment level in the 20 mountainous counties of Zhejiang exhibited specific imbalances. Regarding green industrial development, the overall trend across the 20 counties shifted from fluctuating growth between 2010 and 2016 to steady advancement thereafter. Chun’an County and Kaihua County consistently outperformed other counties, while Qingdao County and Jinyun County lagged. Wuyi County and Yunhe County demonstrated the most rapid development. Concerning high-quality employment, the overall trend across Zhejiang’s 20 mountainous counties shifted from fluctuating growth between 2010 and 2021 to a sharp decline thereafter. Chun’an, Qingdao, and Jinyun counties consistently outperformed the others, while Qingyuan and Jingning counties lagged. Xianju and Tiantai counties demonstrated the most rapid advancement.

(2)

Impact analysis reveals that green industrial development exerts a significant direct effect in promoting high-quality employment, a conclusion upheld by robustness and endogeneity tests. This suggests that, under the ‘sustainable development’ paradigm, green industrial growth can facilitate the coordinated advancement of both employment scale and quality. Additionally, urbanization levels and economic development levels exert a significant positive influence on high-quality employment, while government intervention and resource concentration exert a significant negative influence.

(3)

Heterogeneity analysis indicates that the impact of green industry development on high-quality employment exhibits significant variation across different industrial structures. It substantially promotes high-quality employment in counties dominated by the tertiary sector yet exerts no substantial effect on those primarily reliant on the secondary sector. This suggests that green industry development exerts a more substantial promotional effect on high-quality employment in regions with more optimized industrial structures.

(4)

Mechanism analysis reveals that green industrial development can stimulate industrial upgrading and technological progress, thereby indirectly empowering high-quality employment. This establishes a causal pathway whereby ‘green industrial development promotes high-quality employment through the dual transmission mechanisms of industrial upgrading and technological progress,’ with both industrial structure and technological advancement exhibiting mediating effects on employment.

(5)

Threshold effect analysis reveals that treating technological progress as a threshold variable indicates a dual-threshold effect in the nonlinear relationship between green industrial development and high-quality employment. When technological progress levels fall below the first threshold (TP ≤ 8.0107), the promotional effect of green industrial development on high-quality employment is insignificant; When technological progress levels fall between the first and second thresholds (8.0107 < TP ≤ 8.673), the promotion effect of green industrial development on high-quality employment begins to become significant; when technological progress levels exceed the second threshold (TP > 8.6739), the promotion effect of green industrial development on high-quality employment significantly intensifies, presenting an overall J-shaped nonlinear relationship.

Based on the above research conclusions, this paper proposes the following recommendations:

First, given the actual characteristics of different counties, a prosperity-oriented approach should be adopted for county-level planning. Develop industries with comparative advantages, pursue full-chain development of rural ‘local specialties’, and cultivate new models integrating primary, secondary, and tertiary industries to address the imbalance and inadequacy in the development of Zhejiang’s 20 mountainous counties. For instance, for southwestern counties, priorities ‘zero-carbon infrastructure gap-filling projects’ to reduce green industry implementation costs, thereby accelerating development in these regions, narrowing disparities, and promoting coordinated growth.

Second, green industrial development boosts high-quality employment. Stronger county-level effects appear in tertiary-dominated economies. Policy must promote both structural optimization and technological leapfrogging. Measures need to match each county’s industrial characteristics. For counties with strong secondary industries, focus on green transformation and upgrading industrial chains. Strictly limit new high-energy-consuming and high-emission projects. Prioritize environmental capacity for green manufacturing, supporting local jobs and skill growth. For counties led by tertiary industries, grow new green business models and improve quality. Upgrade the eco-tourism and wellness sectors. Train skilled staff for rural tourism, nature education, and health management. Third, the dual thresholds for technological progress identified in this study (TP = 8.0107, TP = 8.6739) provide a clear, segmented roadmap for county-level green transition strategies. For counties with technological progress levels below 8.0107, the employment-promoting effects of green industrial development remain insignificant. This suggests that local governments should prioritize overcoming the technological initiation threshold through the introduction of technology, talent training, and foundational R&D investment, rather than hastily pursuing the scale expansion of green industries. For counties in the intermediate stage (8.0107 < TP ≤ 8.6739), green industries already generate significant employment, warranting policies focused on accelerating technological assimilation and industrial chain integration to amplify economies of scale. For technologically advanced countries (TP > 8.6739), green industries have become powerful engines for high-quality employment. The strategic focus should shift toward technology exports and the development of new business models to consolidate competitive advantages and generate regional spillover effects.