Improving the employment in less-developed areas through low-carbon city pilot policy: A study on old revolutionary areas in China

Abstract

Objectives

Understanding approaches to promoting green development as well as to increasing employment in old revolutionary areas from the perspective of the “dual carbon” goals is important and, hence, needs to be studied.

Methods

Using city and enterprise panel data from 2007 to 2019, this study used a differential model to assess the impact of China's low-carbon city pilot policies on employment in old revolutionary base areas.

Results

The results showed that the low-carbon city pilot policies could significantly reduce the carbon emissions of old revolutionary base areas as well as significantly promote an improvement in employment levels. Specifically, low-carbon policies could promote the green technological innovation of enterprises and optimize the industrial structure through both output effects and factor-substitution effects to create more jobs. In addition, the impact of low-carbon city pilot policies on employment in different industries and factor intensity was heterogeneous, and the employment promotion effect was more significant in low-energy industries and non-labor-intensive enterprises.

Conclusions

The research results can enrich the theoretical basis of regional development planning in China, and provide important references that can be used to both stabilize the employment of residents and promote high-quality development in underdeveloped areas.

Introduction

As the detrimental effects of global climate change on society grow more acute, the transition to a low-carbon development model has become an essential priority for nations across the globe in their pursuit of sustainable development.^1–4 The societal challenges posed by climate change are multifaceted, encompassing severe extreme weather events, the decimation of ecosystems, and the alarming loss of biodiversity.^5–7 In response, the international community has acknowledged the imperative to move away from the conventional model of development characterized by high energy consumption and emissions, and instead, to embark on a trajectory of low-carbon, eco-friendly, and sustainable growth.^8–11 Realizing this low-carbon vision necessitates a paradigm shift in individual mindsets, a reorientation of business practices, a transformation of production methods, technological innovation, and the phasing out of outdated production capacities, among other strategic choices. These changes will inevitably have profound implications for the employment landscape and the roles of associated personnel.^12–15

The adoption of such policies is especially critical for China's historically significant underdeveloped old revolutionary base areas, which played a pivotal role in the revolutionary struggles under the guidance of the Communist Party of China.^16,17 These regions, imbued with profound historical significance and revolutionary spirit, stand as a vital chapter in China's revolutionary narrative, carrying immense political, cultural, and historical weight in the nation's modernization drive.^18,19 Nevertheless, they face a constellation of challenges, including historical and geographical constraints such as economic underdevelopment, inadequate infrastructure, an undiversified industrial base, and significant employment challenges. The acceleration of the green transformation in these areas can mitigate environmental degradation, enhance the quality of life for urban dwellers, catalyze economic vitality, and foster employment and social harmony.^20–22 To advance the green and low-carbon transition, the National Development and Reform Commission issued the Notice on Carrying out the Pilot Work for Low-Carbon Provinces and Regions and for Low-Carbon Cities on July 19, 2010, initiating the first wave of low-carbon city pilots across five provinces and eight municipalities. Subsequent rounds in 2013 and 2017 expanded the initiative, encompassing a total of 71 cities across three provinces and municipalities directly under the Central Government.^23,24 Within this framework, the low-carbon city pilot policy emerges as a strategic linchpin in China's pursuit of environmental stewardship, economic restructuring, and societal well-being. Through initiatives such as the promotion of renewable energy sources, enhancement of energy efficiency, optimization of transportation systems, and the encouragement of green building practices, the policy is designed to curtail urban greenhouse gas emissions while simultaneously driving the optimization and upgrading of the economic structure.^25,26

The deployment of low-carbon city pilot policies in these venerable regions is both a recognition of and a recompense for the historical sacrifices made by the old revolutionary base areas. It represents a pivotal strategy for invigorating their economic and social fabric. The execution of these low-carbon initiatives can catalyze a shift in the industrial structure of these areas, fostering the growth of green industries that align with local resource endowments and environmental attributes, thereby generating employment prospects for the inhabitants.^27,28 Concurrently, the cultivation of a low-carbon economy can enhance the ecological quality of the old revolutionary base areas, safeguard and promote the heritage of red culture, and establish a robust foundation for their sustainable development. Consequently, examining the impact and functioning of low-carbon city pilot policies within these historical regions holds significant theoretical and practical value, advancing the economic and social progress of these storied areas. ²⁹

Drawing on the efficacy of China's low-carbon city pilot initiatives in pertinent sectors, this research utilized panel data from 2007 to 2019, focusing on China's old revolutionary base areas, to explore the impact and underlying mechanisms of these policies on employment within these historic regions. The low-carbon city pilot policies implemented in 2010, 2013, and 2017 were treated as quasi-natural experiments. The study employed a difference-in-differences (DID) model to assess the outcomes of these policies, finding that they significantly curtailed carbon emissions and markedly enhanced employment rates among the workforce. ³⁰ Mechanistically, the implementation of low-carbon city pilot policies can offset the costs incurred by enterprises for emission reductions and can also lengthen and broaden industry chains, thereby boosting labor demand and fostering job creation. ³¹

This study offers several key contributions. Initially, it directs attention to underdeveloped regions characterized by their distinctive historical, cultural, and economic profiles, namely China's old revolutionary areas, to examine the role of low-carbon city pilot policies in enhancing employment opportunities. Theoretically, it broadens the scope of research on low-carbon city pilot policies and complements and refines the application of employment theories in these specialized areas. Furthermore, it investigates the mechanisms by which these policies affect employment in these unique contexts, shedding light on the intricate relationship between low-carbon transitions and job markets, thereby enriching the theoretical corpus on the interplay between low-carbon strategies and employment. Secondly, grounded in theories such as the Porter Hypothesis and factor substitution effects, the empirical research delineates the precise impact of low-carbon city pilot policies on employment enhancement in underdeveloped regions, furnishing robust evidence for local governments to refine their policy frameworks. Thirdly, as exemplars of underdeveloped regions, the successful employment outcomes realized under low-carbon city pilot policies in old revolutionary areas offer valuable insights for other comparable regions aiming to navigate green transformation while securing employment stability. In essence, this study constructs a nuanced analytical framework for understanding the policy-employment nexus in China's old revolutionary areas. Through targeted empirical analysis, it provides scientific validation to comprehend and advance the role of low-carbon city pilot policies in bolstering employment. The findings guide the holistic green transformation of underdeveloped regions and contribute to the betterment of the livelihoods of the inhabitants in old revolutionary areas.

Review of the existing literature and theoretical framework

Literature review

A pilot policy for low-carbon cities represents a strategic approach to environmental governance in China, designed to proactively address climate change at the urban level. Assessing the efficacy of this policy is a central focus of academic inquiry. The implementation of low-carbon city pilot policies has markedly enhanced urban air quality through two primary channels: by curbing enterprise emissions and by reshaping industrial structures, thereby exerting significant pollution mitigation effects. ³² Scholars have delved into the economic implications of these policies, with studies at the enterprise level concentrating on their influence on green innovation, ³³ total factor productivity, and the promotion of high-quality development within enterprises to achieve economic outcomes.^34,35 At the city level, research has primarily centered on evaluating the policy's impact on total factor energy efficiency, ³⁶ the upgrading of industrial structures, and the fostering of green economic growth. ³⁷

The relationship between the transition to a low-carbon economy and employment is multifaceted, intertwining elements of economic development with social and environmental sustainability considerations. ³⁸ As a holistic environmental regulatory policy, China's low-carbon city pilot policy is aimed at achieving a transformation of the economic structure and promoting the harmonious development of citizens’ well-being. However, there is no consensus on the precise impact of low-carbon city pilot policies on employment. One perspective, the employment increase theory, posits that these policies can substantially enhance the labor market by improving the environment, which in turn can boost employment levels, particularly within high-pollution industries. ³⁹ Contrarily, the employment reduction theory suggests that stringent external environmental regulations may lead to increased production costs for enterprises, potentially causing price hikes and reduced market demand. This could result in a downsizing of business operations.^40,41 In response to such adverse scale effects, companies might resort to layoffs to alleviate the financial burden imposed by tougher environmental rules. These regulations can significantly affect industries with high employment absorption rates, potentially disrupting the labor market. ⁴² The stringent constraints imposed by rigorous environmental regulations may also pose challenges for some heavy polluters and small businesses to absorb costs in the short term, potentially leading to forced shutdowns or the relocation of businesses, with consequent effects on local employment. ⁴³ A third viewpoint, the employment uncertainty theory, suggests that the impact of environmental policies on employment is not constant. Incorporating an interactive term analysis method reveals that the effects of environmental policies on employment can vary across different industries and countries. ⁴⁴

In less-developed regions, the employment implications of low-carbon policies are intricate and multifaceted. These areas are characterized by a high demand for economic growth and enhancements in the quality of life for their inhabitants. Typically, they possess a fragile economic foundation and a monolithic industrial structure, leading to a dualistic impact when low-carbon policies are enacted.^45,46 On one hand, such policies generate employment opportunities in underdeveloped regions by fostering the growth of green industries.^47,48 The establishment of wind and solar projects, for instance, not only necessitates the expertise of skilled workers but also offers employment to local residents, thereby stimulating regional economic advancement.^49,50 On the other hand, the pursuit of low-carbon policies may pose employment challenges, especially in regions heavily dependent on high-carbon industries. Concurrently, the adoption of low-carbon policies is contingent upon technology transfer and skills training, both of which are vital for enhancing the competencies of the workforce in less-developed areas. ⁵¹ Through the upskilling of the labor force, low-carbon policies facilitate the transition of workers from traditional, carbon-intensive sectors to more sustainable, low-carbon industries, thereby refining the employment landscape.

In conclusion, the current scholarly discourse predominantly addresses the ways in which low-carbon city pilot policies affect local environmental and economic outcomes. Nevertheless, the exploration of social implications and employment dynamics is notably lacking, and a consensus has yet to be reached. As distinctive underdeveloped regions in China, the old revolutionary base areas are distinguished by their unique historical context and cultural legacy. The majority of research has centered on the economic development lag in these areas, advocating for enhancement through tourism and the promotion of red culture. ⁵² However, the investigation into the efficacy of low-carbon city pilot policies within old revolutionary base areas, particularly their impact on employment and social welfare, is an area ripe for further exploration. This study aimed to address this gap in the literature, focusing on the broader social and employment effects of these policies in these historically significant regions.

Analysis of theoretical mechanisms

Building on the research methodology established by Berman and Bui, ⁵³ who categorized factors of production into variable and quasi-fixed factors, our study examined the mechanism through which a low-carbon city pilot policy influences urban employment in traditional revolutionary areas. Variable factors include labor, raw materials, and production capital, while quasi-fixed factors are inputs constrained by external elements and are not fixed solely for cost minimization. Thus, costs such as equipment purchases and pollution control are classified as quasi-fixed inputs, subject to environmental regulations and policies. In an economy characterized by perfect competition, wherein firms operate using a cost-minimization strategy, their production costs comprise M variable inputs and N quasi-fixed elements. Variable costs can be expressed as follows:

$$\mathrm{PC}=\mathrm{F}(\mathrm{Y},{\mathrm{V}}_1,..,{\mathrm{V}}_{\mathrm{m}},{\mathrm{Q}}_1,\dots ,{\mathrm{Q}}_{\mathrm{n}})$$

(1)

where PC denotes the variable costs, Y denotes the output, represents the input prices of the variable factors, and represents the quasi-fixed-factor inputs. According to the principle of minimizing production costs, firms’ costs are differentiated, and labor demand was approximated using linear equations in this study:

$$\mathrm{L}=\alpha +{\rho }_{\mathrm{Y}}\mathrm{Y}+{\sum }_{\mathrm{m}=1}^{\mathrm{M}}{\gamma }_{\mathrm{m}}{\mathrm{V}}_{\mathrm{m}}+{\sum }_{\mathrm{n}=1}^{\mathrm{N}}{\beta }_{\mathrm{n}}{\mathrm{Q}}_{\mathrm{n}}$$

(2)

where L signifies the labor demand, and the parameter represents the substitution or complementary relationship between environmental management inputs and labor demand as a consequence of the low-carbon city pilot policy. Using R to represent the low-carbon city pilot policy, we captured the policy's incremental effect on employment, while also considering the effects of other variables. Thus, the interaction between the policy and employment was expressed as follows:

$$\mathrm{L}=\mu +\theta \mathrm{R}$$

(3)

By computing the total derivative with respect to R for both sides of equation (3), the impact of the low-carbon city pilot policy, R, on employment can be expressed as follows:

$${\displaystyle \frac{\mathrm{dL}}{\mathrm{dR}}={\rho }_{\mathrm{Y}}{\displaystyle \frac{\mathrm{dY}}{\mathrm{dR}}+{\sum }_{\mathrm{m}=1}^{\mathrm{M}}{\gamma }_{\mathrm{m}}{\displaystyle \frac{{\mathrm{dV}}_{\mathrm{m}}}{\mathrm{dR}}+{\sum }_{\mathrm{n}=1}^{\mathrm{N}}{\beta }_{\mathrm{n}}{\displaystyle \frac{{\mathrm{dQ}}_{\mathrm{n}}}{\mathrm{dR}}}}}}$$

(4)

Owing to strong competition in factor markets, the impact of the low-carbon city pilot policy on variable factor costs was considered insignificant, leading the second term in equation (4) to tend toward 0. This allowed for a more refined simplification of the expression, as follows

$${\displaystyle \frac{\mathrm{dL}}{\mathrm{dR}}={\rho }_{\mathrm{Y}}{\displaystyle \frac{\mathrm{dY}}{\mathrm{dR}}+{\sum }_{\mathrm{n}=1}^{\mathrm{N}}{\beta }_{\mathrm{n}}{\displaystyle \frac{{\mathrm{dQ}}_{\mathrm{n}}}{\mathrm{dR}}}}}$$

(5)

Low-carbon city pilot policies can affect urban employment in old revolutionary areas through output and factor substitution. Equation (5) suggests that a policy's impact on employment in the urban areas of traditional revolutionary regions can be categorized into two primary groups: a direct output effect, whereby the low-carbon city pilot policy directly influences employment levels; and an indirect factor-substitution effect, whereby the policy affects employment by altering the demand for factors of production. The conceptual framework outlined above indicates that a low-carbon city pilot policy affects employment via these separate pathways. Leveraging this analytical foundation, the paper presents Figure 1 to illustrate the policy's employment impact mechanism, before providing a detailed explanation of how these two effects individually affect employment outcomes.

Figure 1.

Theoretical framework of the study.

Mechanism 1

The output effect refers to the manner in which low-carbon city pilot policies foster economic growth and industrial enhancement by shaping enterprise production. This, in turn, boosts the demand for labor and elevates employment levels.

According to the Porter effect, low-carbon city pilot policies affect employment by influencing cost structures and fostering innovation as firms strive to reduce carbon emissions. In terms of the cost hypothesis, a low-carbon pilot initiative prompts firms to implement more eco-friendly and efficient methods, resulting in heightened environmental investment. This may result in higher prices for their products as they strive to maintain profitability, potentially leading to a decline in labor demand. Conversely, the innovation compensation theory suggests that a low-carbon pilot initiative motivates businesses to engage in sustainable technological innovations. Developing new, environmentally friendly products and pursuing new markets could create production demand, thereby enlarging production scales and boosting labor demand, which ultimately contribute to employment expansion.

Mechanism 2

The factor-substitution effect refers to the fact that low-carbon city pilot policies affect employment by influencing the proportion of firms’ factors of production.

The mechanism is primarily governed by two elements: the impact of the low-carbon city pilot policy on the quasi-fixed factors of businesses ( $\mathrm{dQn}/\mathrm{dR}$ ), and the effect of these quasi-fixed factors on employment ( $\beta$ ). Regarding the former, as a low-carbon city pilot program is implemented, firms are driven to increase their spending on pollution control to meet environmental standards. These expenditures are categorized as quasi-fixed factors, leading to a positive effect, denoted as $\mathrm{dQn}/\mathrm{dR}$  > 0. The second aspect represents the impact of quasi-fixed factors on the labor market. A value of 0 implies that these factors do not influence the labor market. However, the impact of quasi-fixed factors on the labor market can vary, being either favorable or adverse, depending on the nature of their relationship with labor—that is, whether it is substitutive ( $\beta$  < 0) or complementary ( $\beta$  > 0). Companies have adopted various countermeasures under strict environmental policy regulations. Some have optimized their production by adopting advanced environmental protection facilities and cleaner production technologies, thereby reducing pollutant emissions and improving productivity. However, this technological advancement could result in job cuts, potentially negatively affecting employment. Conversely, some firms invest in environmental management at the conclusion of the production; this may include adding extra dust removal and filtration systems to mitigate pollution before its release. These measures reduce negative environmental impacts and may create jobs.

Essentially, low-carbon city pilot policies can influence employment through two mechanisms: output effects and factor-substitution effects. The overall effect of low-carbon city pilot policies on the job market is determined by both output and factor-substitution effects. This study employed an empirical research methodology to thoroughly analyze how these two effects jointly shape employment outcomes, aiming to comprehensively evaluate the impact of low-carbon city pilot policies on the employment market. Figure 2 shows the mechanism of the impact of low-carbon city pilot policies on employment.

Figure 2.

The influence mechanism of low-carbon city pilot policy on employment.

Methods

Model setting

The DID econometric method has been widely used to evaluate policy effects in recent years. Policy change is considered a naturally occurring experimental situation. When establishing pilot low-carbon cities in China, the DID method can capture changes at two levels: changes in carbon emissions and employment in old revolutionary base cities before and after implementing a pilot policy, and the difference in the two indicators between pilot and non-pilot old revolutionary base area cities during the same period. After analyzing this double difference, the impact of other policies in the same period and the initial differences between pilot and non-pilot cities can be effectively excluded to accurately measure the net effect of policies on the carbon emissions and employment of cities in old pilot revolutionary base areas. Therefore, China's low-carbon city pilot policies represent an ideal quasi-natural experiment scenario, suitable for using the DID method to evaluate its policy effects.

Given the phased implementation of the low-carbon city pilot policy, this study employed the methodological framework of Wang et al., ⁵⁴ utilizing a multi-period DID method to investigate the possible influences of the policy on employment in historic revolutionary regions. Two observations stood out regarding the implementation of China's low-carbon city pilot policy. First, in pilot cities within traditional revolutionary regions, a noticeable difference in carbon emissions and employment statistics between the pre- and post-policy periods is likely. Second, at all times, discrepancies may exist in the patterns of carbon emissions and employment between pilot and non-pilot cities within old revolutionary areas. By applying regression analysis using a DID model, this study successfully accounted for the impacts of other coexisting policies and the intrinsic differences between pilot and non-pilot cities in old revolutionary base areas before the policy was enacted. This enabled an accurate assessment of the actual effects of the low-carbon city pilot policies on both carbon emissions and the employment market in such regions. Consequently, to ensure the validity of the research, this study first confirmed whether the pilot cities in historic revolutionary areas met the primary goal of carbon emission reduction as outlined by the pilot policy. Subsequently, this study evaluated whether the policy has achieved a notable reduction in carbon emissions. Thereafter, the study determined whether the policy has had a substantial impact on the job market.

1. Model for assessing the emission reduction effect of low-carbon city pilot policies on pilot cities in old revolutionary areas.

Initially, this study evaluated the actual city-level impacts of low-carbon city pilot policies on carbon emission reductions in cities within the old pilot revolutionary areas. For this, the following modeling configurations were employed:

$${\mathrm{CO}}_2\mathrm{ct}={\alpha }_1+{\theta }_1{\mathrm{lccpost}}_{\mathrm{ct}}+{\lambda }_1{\mathrm{Z}}_{\mathrm{ct}}^1+{\eta }_{\mathrm{c}}+{\mu }_{\mathrm{t}}+{\epsilon }_{\mathrm{ct}}^1$$

(6)

In the model, c and t stand for city and year, respectively; CO₂ denotes carbon emissions; lccpost is a binary variable indicating whether a city has adopted a low-carbon city pilot policy; ${\eta }_{\mathrm{c}}$ and ${\mu }_{\mathrm{t}}$ are city fixed effects and year fixed effects, respectively; ${\epsilon }_{\mathrm{ct}}^1$ is a random error term affecting a city's carbon emissions; ${\theta }_1$ is a double-DID estimator, used to assess the effect of low-carbon city pilot policies on carbon emissions in pilot cities within old revolutionary areas; and ${\mathrm{Z}}_{\mathrm{ct}}^1$ represents other control variables.

2. Model for evaluating the impact of low-carbon city pilot policies on city-level employment in old revolutionary areas.

Considering the effectiveness of the low-carbon pilot policy in reducing carbon emissions, we assessed the possible effects of this policy on the labor market from the perspective of a city located in a traditional revolutionary area.

$${\mathrm{citylabor}}_{\mathrm{ct}}={\alpha }_2+{\theta }_2{\mathrm{lccpost}}_{\mathrm{ct}}+{\lambda }_1{\mathrm{Z}}_{\mathrm{ct}}^2+{\eta }_{\mathrm{c}}+{\mu }_{\mathrm{t}}+{\epsilon }_{\mathrm{ct}}^2$$

(7)

The model was formulated as follows: city-level employment (citylabor) was measured by taking the natural logarithm of the average number of employees in a city's labor force; ${\eta }_{\mathrm{c}}$ and ${\mu }_{\mathrm{t}}$ indicate city fixed effects and year fixed effects, respectively; ${\epsilon }_{\mathrm{ct}}^2$ denotes a random error term that affects the level of city employment; ${\theta }_2$ is a double-DID estimator used to gauge the impact of the low-carbon city pilot policy on employment in cities within old pilot revolutionary areas; and ${\mathrm{Z}}_{\mathrm{ct}}^2$ encompasses a collection of control variables.

3. Model for assessing the impact of low-carbon city pilot policies on enterprise-level employment in old revolutionary areas.

$${\mathrm{labor}}_{\mathrm{ct}}={\alpha }_2+{\theta }_3{\mathrm{citylccpost}}_{\mathrm{it}}+{\lambda }_3{\mathrm{X}}_{\mathrm{it}}+{\phi }_{\mathrm{i}}+{\mu }_{\mathrm{t}}+{\epsilon }_{\mathrm{it}}$$

(8)

In the model, i and t indicate the firm and the time period, respectively; labor represents the number of employees within the firm; citylccpost is a binary variable that indicates whether the firm is situated in an old pilot revolutionary area where the low-carbon city pilot policy has been enacted; ${\phi }_{\mathrm{i}}$ and ${\mu }_{\mathrm{t}}$ capture the firm fixed effects and the time fixed effects, respectively; ${\theta }_3$ is a DID method estimator used to assess the effect of the low-carbon city pilot policy on the firm's employment level; and ${\mathrm{X}}_{\mathrm{it}}$ encompasses other control variables.

Variable description

Explanatory variables

Consistent with studies by Khanna et al., ²⁶ Shayegh et al., ⁵⁵ and Wang et al., ⁵⁶ the following variables were selected: carbon emissions ( ${\mathrm{CO}}_2$ ), as the dependent variable used for evaluating the success of emission reduction efforts; citylabor, an independent variable for urban employment, calculated as the natural logarithm of the mean number of urban employees; and labor, the dependent variable for firm-level employment, represented by the natural logarithm of the firm's employee count.

Core explanatory variables

A low-carbon city pilot policy (lccpost), a critical independent variable, was implemented in China by the National Development and Reform Commission in three phases between 2010 and 2017. In subsequent empirical analyses, for cities selected as pilots in the second or third wave and whose provinces were already included in an earlier pilot wave, the city's pilot start year was considered the year of the first batch's initiation. Of note, the second group of pilot cities was announced at the end of 2012, and in 2013, the policy was officially implemented in these cities. This study categorized the implementation of the low-carbon city pilot policy into three key stages: 2010, 2013, and 2017.

Control variables

When the effectiveness of emission reduction was analyzed, the control variables comprised the natural logarithm of regional GDP per capita as an indicator of economic development; the natural logarithm of the total urban population at year end, representing the population scale; the proportion of the value added by secondary industries in regional GDP, used to assess the industrialization level; the ratio of the urban population to the total population, indicating urbanization; and the ratio of coal consumption to total energy consumption, used to gauge the energy structure.

In the context of urban employment, the control variables encompassed the natural logarithm of the city's year-end population, the natural logarithm of the average salary for urban employees, the natural logarithm of the gross regional product per capita, and the natural logarithm of the total retail sales of consumer goods.

For control variables related to firm employment, we used the natural logarithm of the firm's establishment year plus one, as a proxy for firm age; the natural logarithm of the firm's overall assets, indicating the firm size; the average wage of employees, used to measure the wage level within the firm; Tobin's Q, used to assess the firm's long-term growth potential; and the natural logarithm of operating income.

Data sources

To assess how the low-carbon city pilot policies impact city-level carbon emissions and employment, based on equations (6) and (7), this study selected 2007–2019 as the research interval, and selected 164 cities in China's old revolutionary base areas as research samples. Among them, 70 cities were pilot cities and 94 were non-pilot cities. In equation (6), the explained variable is urban carbon emissions in the old revolutionary base area and was represented by the logarithm of urban carbon dioxide emissions, recorded as CO₂. The data were obtained from the China Emission Accounts & Datasets. In equation (7), the explained variable is the urban employment level, expressed as the logarithm of the average number of employees in the city, and was denoted as citylabor. These data and the data of the city-level control variables were derived from the China Urban Statistical Yearbook (2008–2020).

To assess how the pilot policies of low-carbon cities impact enterprise-level employment in old revolutionary base areas, based on equation (8), this study selected A-share listed companies in the Shanghai and Shenzhen stock markets of China from 2007 to 2019 as research samples. To avoid influences from abnormal samples, this study processed the original data using the following four steps: (1) Eliminate the listed enterprises with “ST” and “* ST” in the stock abbreviation; (2) exclude enterprises whose listing status is “ST,” “* ST,” “suspended listing,” or “terminated listing”; (3) eliminate listed companies with substantial missing data; (4) use linear interpolation to complete some of the missing data of individual listed enterprises. In this study, the logarithm of the total number of employees was used to measure enterprise employment, which was recorded as labor. The data for all variables in equation (8) were obtained from the Guotai'an database. China's low-carbon city pilot program was launched in July 2010, November 2012, and January 2017. Considering the implementation time and lag of the pilot policy, and this study's operability, the starting times of the three batches of low-carbon city pilot policies were determined as 2010, 2013, and 2017. The definitions of the variables are shown in Table 1.

Table 1.

Variable definition.

	Variables	Definition
Explained variables	lccpost	The starting time of the three batches of low-carbon city pilot policies
Explaining variables	CO2	Carbon emission
	Citylabor	The logarithm of the average number of employees in a city
	labor	The logarithm of the total number of employees in an enterprise
Control variables	pgdp	The level of economic development is expressed as the logarithm of GDP per capita
	pop	Population size is expressed as the logarithm of the city's total population at the end of the year
	ind	The level of industrialization is measured by the proportion of the added value of the secondary industry to the gross regional product
	urb	The level of urbanization is expressed by the proportion of urban population in the total population
	es	The energy structure is measured by the ratio of coal consumption to total energy consumption
	citywage	It is expressed by the average wage of employees in the city
	rsc	Expressed as the logarithm of total retail sales of consumer goods
	age	The age of the enterprise is expressed by adding 1 to the natural logarithm of the year of establishment of the enterprise
	size	The natural logarithm of a firm's total assets represents its size
	wage	The average salary of employees is used to measure the internal wage level of enterprises
	grow	Tobin's Q is used to measure long-term growth potential
	income	Operating income is expressed as the natural logarithm of operating income

Results

Coupling analysis

This study developed a cohesive framework integrating improvements in livelihoods and environmental protection; this was grounded in the interplay between environmental and employment considerations, alongside the specific geographical context of China's old revolutionary areas. Drawing from data from 2007 to 2019, this study selected 11 secondary indicators in the four dimensions of economic development, social construction, ecological construction, and pollution emissions and adopted the coupling coordination degree model to quantitatively evaluate the coordination level of the two systems of livelihood improvement and environmental protection, which enabled a scientific assessment of their developmental dynamics (Table 2). This method strengthened the theoretical underpinnings of the low-carbon city pilot policy, which aims to drive harmonious progress in energy saving, carbon mitigation, and job creation in traditional revolutionary regions and, at the same time, focuses on improving the living conditions and general well-being of the inhabitants. The pilot low-carbon city initiative provides a conceptual platform for realizing synergetic enhancements in energy efficiency, pollution reduction, and employment opportunities, while ensuring the prosperity and quality of life of residents in these historically important revolutionary areas. To mitigate the impact of disparities in indicator magnitude and units of measurement on the evaluation results, the data underwent standardization. The standardization formulas were as follows: for positive indicators, ${\mathrm{x}}_{\mathrm{i}}={\mathrm{x}}_{\mathrm{i}}-\frac{\min \{{\mathrm{x}}_{\mathrm{i}}\}}{\max \{{\mathrm{x}}_{\mathrm{i}\}}}-\min \{{\mathrm{x}}_{\mathrm{i}}\}$ ; for negative indicators, ${\mathrm{x}}_{\mathrm{i}}=\max \{{\mathrm{x}}_{\mathrm{i}}\}-\frac{{\mathrm{x}}_{\mathrm{i}}}{\max \{{\mathrm{x}}_{\mathrm{i}}\}-\min \{{\mathrm{x}}_{\mathrm{i}}\}}$ . In the formula, $\min \{{\mathrm{x}}_{\mathrm{i}}\}$ and $\max \{{\mathrm{x}}_{\mathrm{i}}\}$ denote the minimum and maximum indicator values, respectively. The indicator's value denotes its level of favorability toward system development; a higher value is more beneficial, while a lower value is more favorable for system development.

Table 2.

Evaluation index system for coupled coordination.

Systems	First indicators	Secondary indicators	Unit	Weights	Characteristic
People's livelihoods	Economic development	GDP per capita	Yuan	0.167775784	+
		Urban residents’ per capita disposable income	Yuan	0.071771902	+
		Consumer goods retail sales	10,000 yuan	0.410231769	+
	Social construction	Urban road space density per capita	Meter	0.067719884	+
		Hospital and health center bed counts	Sheet of paper	0.21251385	+
Environmental protection	Ecological construction	Greening coverage in built-up areas	Percent	0.01273035	+
		Nonhazardous domestic waste disposal rate	Percent	0.025942578	+
	Pollution emission	Industrial sulfur dioxide emissions	Ton	0.004129042	−
		Industrial smoke emissions	Ton	0.000417811	−
		Sewage discharge	Million cubic meters	0.004924477	−
		PM2.5	Mg/m³	0.01273035	−

Note: Weights for indicators are determined using the entropy method. A positive (+) indicates that a higher indicator value is preferable, while a negative (−) signifies that a lower value is more advantageous.

Coupled coordination degree model

To quantify the coupling degree, a linear integrated evaluation model was adopted to determine the integrated evaluation index for each subsystem. The formula for this calculation is as follows:

$$\mathrm{p}(\mathrm{x})={\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{a}}_{\mathrm{i}}{\mathrm{x}}_{\mathrm{i}};\mathrm{e}(\mathrm{y}){=\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{b}}_{\mathrm{i}}{\mathrm{y}}_{\mathrm{i}}$$

(9)

where a_i and b_i represent the calculated weights, and p(x) and e(y) denote the comprehensive development indices for people's livelihoods and the environment, respectively. Higher values of these indices correspond to more advanced system development, whereas lower values indicate less-developed systems. The coupling calculation formula is as follows:

$$\mathrm{C}=2\times [{\displaystyle \frac{\mathrm{p}(\mathrm{x})*\mathrm{e}(\mathrm{y})}{{(\mathrm{p}(\mathrm{x})+\mathrm{e}(\mathrm{y}))}^2}{]}^{\frac{1}{2}}}$$

(10)

$$\mathrm{T}=\alpha \times \mathrm{p}(\mathrm{x})+\beta \times \mathrm{e}(\mathrm{y})$$

(11)

$$\mathrm{D}=\sqrt{\mathrm{C}\times \mathrm{T}}$$

(12)

where C represents the coupling degree; T denotes the comprehensive coordination index; D is the degree of coupled coordination; and p(x) and e(y) represent the livelihood improvement and environmental protection scores, respectively. A larger value indicates a higher level of coordination between the two systems. Parameters α and β represent the uncertain weights that gauge the significance of enhancing people's livelihoods and the effects of environmental conservation, respectively. These weights were integrated with the coupling degree and coordination degree, as specified in Table 3, to establish a hierarchy of the coupling levels.

Coupled coordination degree model

Spatial distribution maps were constructed to visualize the levels of coupling and coordination for 2007 and 2019, to analyze the spatial distribution of the correlation between the enhancement of livelihoods and environmental conservation within China's old revolutionary areas. These maps were developed using the computed data and established grading benchmarks, as shown in Figure 3. The coupling level of the livelihood improvement–environmental protection system in China's old revolutionary areas in 2007 showed some spatial differences (Figure 3). Most regions were in the integration stage, with a high coupling level, indicating that the improvements in livelihoods and environmental protection in these regions were relatively coordinated. However, some regions were in the antagonistic stage, and the coupling level was low, indicating that there were greater conflicts and challenges between livelihood improvement and environmental protection in these regions. Spatially, cities exhibiting strong coupling were concentrated in the eastern and central parts of the country, whereas regions with weaker coupling were mainly found in western China and in some isolated areas. Owing to geographical and historical factors, these regions experienced slower economic growth, and resources for enhancing livelihoods and environmental conservation were scarce. This scarcity caused a strain in the coupling between the two systems. By 2019, the coupling level of the system integrating livelihood improvement and environmental protection in China's old revolutionary areas had advanced significantly overall, with areas of higher coupling concentrations predominantly found in eastern China. The coupling level of the livelihood improvement–environmental protection system in China's old revolutionary regions in 2019 greatly improved, as the whole, and the coupling level was higher. In contrast, the coupling level in the western region remained in the teething stage, while economically developed areas, such as the eastern region, maintained a high coupling level. This might have been due to the weak economic base and limited resources in the western region, as well as insufficient investments in livelihood improvement and environmental protection.

Figure 3.

Trend of spatial evolution of the degree of coupling and the degree of coordination of the system for improving people's livelihoods and environmental protection.

Table 3.

Coupling degree of livelihood improvement and environmental protection and the grading criteria of coupling coordination degree.

Coupling classification		Coupling coordination level classification
Coupling	Coupling level	Degree of coupling coordination	Level of coordination	Coordination phase	Degree of coupling coordination	Level of coordination	Coordination phase
0.0	Uncoupled relationship	(0.0, 0.1)	Extreme disorder	Dysfunctional recession category	(0.6,0.7)	Primary coordination	Coordinated development category
(0.0,0.3)	Low-level coupling	(0.1, 0.2)	Severe disorder		(0.7,0.8)	Intermediate-level coordination
(0.3,0.5)	Antagonistic phase	(0.2, 0.3)	Moderate disorder		(0.8,0.9)	Good coordination
(0.5,0.8)	Break-in period	(0.3, 0.4)	Mild disorder		(0.9,1.0)	Quality coordination
(0.8,1.0)	High-level coupling	(0.4, 0.5)	On the verge of becoming dysfunctional	Transitional development category	–	–	–
1.0	Benign resonance coupling	(0.5, 0.6)	Suing for coordination

As Figure 3 illustrates, in 2007, the overall degree of coordination for the system linking livelihood improvement and environmental protection in China's old revolutionary areas was categorized as transitional development. This included 146 cities on the brink of dysfunction and 70 cities barely achieving coordination. Spatially, the eastern coastal areas exhibited higher levels of coupling coordination, with Shenzhen achieving a level of primary coordination. By contrast, the central, western, and northeastern regions primarily exhibited coupling coordination at a mildly dysfunctional level, with only a few economically advanced cities performing favorably. Most cities in these regions were at the edge of dysfunctional coupling coordination. By 2019, the level of coupling coordination between livelihood enhancement and environmental conservation systems in China's historic revolutionary regions had notably improved, falling within the scope of coordinated development. Specifically, 39 cities reached an intermediate level of coordination, while 136 were at the primary level. Geographically, the eastern region maintained the highest degree of system coupling coordination. Leading in the country in terms of the coupling coordination degree, Shenzhen was classified as being in the high-quality coordination category, followed by Chengdu, Nanjing, and Quanzhou, each falling within the good-coordination category. The coupling coordination degree of systems in most cities within China's central, western, and northeastern regions transitioned from near dysfunction to primary coordination. This progression represented a notable enhancement in the coordination between the two key systems, indicating their transition into a phase of harmonious development. However, only a few cities achieved good and high-quality coordination, suggesting that there is ample scope for enhancing the coupling and coordination between the two systems. Additionally, further improving this coordination is urgently required.

In summary, the degree of synergy of the livelihood improvement–environmental protection system in China's old revolutionary areas was marked by a significant level of coupling but a less pronounced degree of coordination, indicating a potential for comprehensive enhancement in the quality of their development. This suggested that environmental protection significantly enhance residents’ well-being and that the implementation of low-carbon pilot policies can boost employment in old revolutionary areas, thereby promoting the harmonious development of the environment and people's livelihoods.

Analysis of baseline regression results

Test results of the impact of low-carbon city pilot policies on carbon emissions in cities in old revolutionary areas.

Column (1) of Table 4 delineates the effects of the low-carbon city pilot policy on carbon emissions within China's historic revolutionary cities. The coefficient for the dummy variable citylccpost, which indicated the implementation of the policy, was −0.0444 and was statistically significant at the 5% level. This suggested a substantial reduction in carbon emissions, which could be attributed to the policy. The findings suggested that the low-carbon city pilot policy had a substantial effect on reducing carbon emissions in the selected old revolutionary cities. Considering all pertinent factors, the policy was associated with an average reduction of 4.44% in carbon emissions in these pilot cities compared to those of cities not included in the program. Yan et al. ⁵⁷ found that China's low-carbon city pilot program has resulted in a decrease in the pollution levels of particulate matter 2.5 within these cities, thereby contributing to air pollution prevention and control efforts. The outcomes of this study are consistent with those of this article, thus further supporting the notion that such policies are effective in reducing air pollution. This forms the basis for the subsequent section, which explores how low-carbon city pilot policies affect employment rates.

Test results of the impact of low-carbon city pilot policies on urban employment in old revolutionary areas

Table 4.

Impact of pilot low-carbon city policies on carbon emissions and employment.

Variables	CO2	Citylabor
	（1）	（2）
Citylccpost	−0.0444**	0.0602*
	(−2.4945)	−0.0363
Control variable	YES	YES
City, year fixed effects	YES	YES
Observations	3003	3003
R 2	0.9872	0.9709

Notes: Stars (*, **, and ***) are used to indicate statistical significance at the 10%, 5%, and 1% confidence levels, respectively, with the t-statistics listed in parentheses.

Column (2) of Table 4 illustrates the policy's influence on employment in revolutionary regions. The coefficient for citylccpost was 0.0602, demonstrating statistical significance at the 10% confidence interval. This suggested that the low-carbon city pilot policy has had a beneficial impact on employment within the pilot cities of the old revolutionary areas, leading to an estimated employment increase of 6.02% relative to that of non-pilot cities in these areas once other relevant factors are considered. The results showed that the overall employment effect of the low-carbon city pilot policy was positive. Although some jobs were lost, jobs were also created, and the number of jobs created greatly exceeded that of jobs lost.

Test results of the impact of low-carbon city pilot policies on employment in enterprises in old districts

This study builds upon previous studies highlighting the efficacy of China's low-carbon city pilot policy in curbing carbon emissions and fostering employment in its historic revolutionary cities. This study evaluated the policy's influence on corporate employment; the findings are summarized in Table 5. Column (1) shows the estimation results without the control variables, indicating a coefficient of 0.0483 for the lccpost dummy variable, which was statistically significant at the 10% level. Column (2) incorporates the control variables, with the lccpost coefficient declining to 0.0482, which was still significant at the 1% level. The minor decrease in the coefficient when controls were added suggested that these variables affected firm employment. The data in column (2) reveal that employment in firms located within the pilot cities experienced an average increase of 4.82% compared with firms outside these areas. This is because although the pilot policy of low-carbon cities increases the cost of enterprises to a certain extent, enterprises may choose to reduce the scale of production and the number of labor employees, resulting in a decline in employment. However, to achieve sustainable development goals, enterprises must implement green technological innovation. This will increase labor demands and product outputs, and then increase the total employment level of old revolutionary base areas.

Table 5.

Impact of low-carbon city pilot policies on business employment.

Variables	Labor	Labor
	（1）	（2）
Lccpost	0.0483*	0.0482***
	(1.7302)	(2.7412)
Control variable	NO	YES
City, year fixed effects	YES	YES
Observations	13,149	12,894
R2	0.9174	0.9657

Notes: Stars (*, **, and ***) are used to indicate statistical significance at the 10%, 5%, and 1% confidence levels, respectively, with the T-statistics listed in parentheses.

Robustness tests

Parallel-trend test

The parallel-trend hypothesis is important in the DID method—without low-carbon city pilot policies, employment growth trends should be consistent across cities. Accordingly, this study employed the event study methodology, using methods of Jacobson et al. ⁵⁸ and Couch and Placzek, ⁵⁹ to evaluate parallel trends. The model was constructed as follows:

$${\mathrm{Labor}}_{\mathrm{it}}={\alpha }_0+{\alpha }_1{\sum }_{\mathrm{t}=\text{-}3}^4{\rho }_{\mathrm{t}}{\mathrm{D}}_{\mathrm{it}}+{\alpha }_2{\mathrm{X}}_{\mathrm{it}}+{\pi }_{\mathrm{it}}+{\tau }_{\mathrm{t}}+{\epsilon }_{\mathrm{it}}$$

(13)

where D_it is a set of virtual variables, representing the implementation status of the low-carbon city pilot policy in region i in year t. If this policy is implemented, the D_it value is 1; otherwise, it is 0. The settings of the other variables are consistent with equation (8). This paper took t ≤ −4 as the base year, and when t > 0, the coefficient reflects the dynamic effect of the low-carbon city pilot policy on employment in old revolutionary base areas. Figure 4 shows the estimations of the parallel-trend test. Before implementing the low-carbon pilot policy, the estimated coefficients of each period were nonsignificant, indicating no significant difference in employment levels between the treatment and control groups (i.e. the parallel-trend test was passed). After implementing the policy, the low-carbon city pilot policy sustainably and significantly increased employment in old revolutionary base areas and had a dynamic effect.

Figure 4.

Parallel-trends assessment.

Placebo test

To mitigate the impact of omitted variables and investigate whether the baseline regression was affected by unobservable factors, this study adopted the city placebo test outlined by Li et al. ⁶⁰ This involved randomly assigning 123 cities to a treatment group, and the remaining to a control group, using random sampling. The impact on corporate employment was then estimated for these placebo groups. This process was replicated 500 times, and the distribution of the resulting coefficient estimates is shown in Figure 5. The random-matching results showed that the estimated clustering was near 0, and all coefficients had p-values > .1, suggesting that most of the regressions are nonsignificant at the 10% level. This showed that the estimations were not random, and interference from other unobjectionable and random factors was excluded.

Figure 5.

Placebo test.

Exclusion of other policy interferences

To avoid the bias of benchmark regression results caused by other policies affecting enterprises’ employment, this study sorted relevant documents and found that other policies, in the same period, may also affect the employment level of the regional labor; thus, attempts were made to exclude the impact of other relevant policies on employment within a given period.

Demonstration cities for energy conservation and emission reduction. Regarding energy conservation policy, this study included the fiscal policy of energy conservation and emission reduction in the model to exclude its impact on the employment level by combining documents and policy practices. In 2011–2014, China published three lists of comprehensive demonstration cities for energy conservation and emission reduction policies, including 30 pilot cities. To eliminate impacts of energy conservation and emission reduction on employment, dummy variables for energy conservation and emission reduction policies were constructed based on the pilot cities and corresponding policy years, and introduced into the benchmark regression model as control variables. As shown in column (1) of Table 5, the estimated coefficient of the core explanatory variable lccpost was still significantly positive, indicating that the benchmark test results were robust.

National smart city. The development strategy of smart cities is key to effectively integrating urbanization and informatization and promoting sustainable innovation and development. It optimizes the connection between the labor force and employment positions via digital technology and facilitates the cultivation of new urban business forms, thus expanding the employment breadth and depth. In 2012, China launched a smart city pilot program and identified pilot cities in batches over the next three years. To accurately assess the impact of this policy on the results, based on the list of pilot cities and the implementation years, a variable representing the national smart city policy was created and introduced into the baseline regression model for consideration. From column (2) of Table 5, after eliminating interference from smart city pilot policies, the promotion effect of low-carbon city pilot policies on employment remained valid.

Broadband China strategy. Between 2014 and 2016, China implemented the Broadband China Strategic Demonstration City pilot project in three phases. The core of this strategy is strengthening the construction of network infrastructure, which provides superior infrastructure and environmental support for enterprises’ innovation activities. The direct employment impact of the Broadband China strategy is reflected in the development of the regional information technology industry, which creates jobs and increases employment opportunities; indirect effects include the increased attractiveness of a region for science and technology talent through increased innovation activities. To assess the impact of the Broadband China strategy on employment, a dummy variable of the Broadband China urban policy was constructed based on the list of pilot cities and the corresponding implementation years, and added as a control variable to the benchmark model to eliminate influences of the Broadband China strategy on the results. Column (3) of Table 6 shows that the results were unaffected.

Table 6.

Exclusion of other policies.

Variables	Labor	Labor	Labor
	Exclusion of other policies
	(1)	(2)	(3)
Lccpost	0.0482***	0.0482***	0.0482***
	(2.7421)	(2.7444)	(2.7423)
Tobin	0.0086	0.0085	0.0087
	(1.5217)	(1.5125)	(1.5299)
LnSale	0.3124***	0.3124***	0.3124***
	(7.3787)	(7.3771)	(7.3779)
Wage	−0.0548***	−0.0548***	−0.0548***
	(−18.6683)	(−18.6609)	(−18.6666)
Size	0.3854***	0.3855***	0.3855***
	(8.4843)	(8.4821)	(8.4837)
Lnage	0.0797***	0.0797***	0.0797***
	(5.3086)	(5.3112)	(5.3102)
Constant	−3.5245***	−3.5292***	−3.5234***
	(−7.5008)	(−7.5157)	(−7.4994)
Energy-saving and emission reduction demonstration city	Yes	No	No
National smart city	No	Yes	No
Broadband China strategy	No	No	Yes
Firm/year fixed effects	Yes	Yes	Yes
Observations	12,894	12,894	12,894
R 2	0.9657	0.9657	0.9657

Notes: Stars (*, **, and ***) are used to indicate statistical significance at the 10%, 5%, and 1% confidence levels, respectively, with the T-statistics listed in parentheses.

Nonrandom selection of low-carbon city pilots

The sequential DID method technique is valuable for gauging policy change outcomes. However, owing to the inherent attributes of cities, such as geographical location, economic development, and population density, which relate closely to the pilot list of low-carbon cities, differences in these attributes may have different impacts on urban employment over time, leading to certain deviations in the calculation results. This study drew upon the research of Lu et al. ⁶¹ and Edmonds et al., ⁶² integrating the interaction term between city-specific benchmark variables and the linear time trends into our baseline model. This adjustment was made to consider how cities’ intrinsic attributes impact employment over time.

This study employed four variables to represent city characteristics: membership in two control zones, classification as a special economic zone city, status as a provincial capital, and location east of the Hu Huanyong Line. Table 7 presents the estimations. The coefficient for lccpost remained statistically significant, suggesting that the results were consistent and robust even when accounting for potential city-specific differences.

Table 7.

Discussion score regression results for non-randomized selection of pilot cities.

Variables	Labor	Labor	Labor	Labor	Labor
	(3)	(4)	(5)	(6)	(7)
Lccpost	0.0482***	0.0481***	0.0481***	0.0481***	0.0480***
	(2.7417)	(2.7379)	(2.7377)	(2.7389)	(2.7339)
Tobin	0.0086	0.0087	0.0087	0.0087	0.0087
	(1.5254)	(1.5322)	(1.5327)	(1.5289)	(1.5477)
LnSale	0.3124***	0.3125***	0.3123***	0.3124***	0.3124***
	(7.3795)	(7.3812)	(7.3812)	(7.3798)	(7.3885)
Wage	−0.0548***	−0.0548***	−0.0548***	−0.0548***	−0.0548***
	(−18.6652)	(−18.6745)	(−18.6496)	(−18.6656)	(−18.6630)
Size	0.3854***	0.3854***	0.3854***	0.3855***	0.3853***
	(8.4854)	(8.4819)	(8.4916)	(8.4850)	(8.4925)
Lnage	0.0797***	0.0796***	0.0797***	0.0798***	0.0796***
	(5.3115)	(5.3053)	(5.3103)	(5.3147)	(5.3076)
Constant	−3.5243***	−3.5235***	−3.5192***	−3.5126***	−3.5077***
	(−7.5041)	(−7.5044)	(−7.4901)	(−7.4769)	(−7.4743)
Two control areas × time trend	Yes	No	No	No	Yes
Provincial capital × time trend	No	Yes	No	No	Yes
Special economic zone × time trend	No	No	Yes	No	Yes
East of Hu Huanyong line×time trend	No	No	No	Yes	Yes
Firm/year fixed effects	Yes	Yes	Yes	Yes	Yes
Observations	12,894	12,894	12,894	12,894	12,894
R 2	0.9657	0.9657	0.9657	0.9657	0.9657

Notes: Stars (*, **, and ***) are used to indicate statistical significance at the 10%, 5%, and 1% confidence levels, respectively, with the T-statistics listed in parentheses.

PSM-DID and outlier tests

Because the selection of pilot cities in low-carbon cities is determined by the central government and based on the business basis of each region, the pilot representativeness, and the application status, the process is not random; therefore, directly applying the differential method may lead to sample selection bias. To solve this, this study used a DID model combined with propensity score matching to reduce the endogeneity caused by sample selection errors, and conducted a robustness test to verify the results’ reliability. Column (2) of Table 7 presents the outcomes after matching, where the estimated coefficients for the key explanatory variables remain substantially positive and statistically significant at the 1% level. This further supports the positive impact of low-carbon urban policies implemented in these regions on urban employment in old revolutionary base areas.

To avoid the impact of extreme values having a strong influence on the benchmark regression results, this study truncated the research samples of the explained variable labor by 1% and 5% and conducted the regression again. As seen in the regression results in Table 8, the coefficient estimate for lccpost remained statistically significant even after accounting for extreme values. The positive impact of the low-carbon city pilot program on labor force employment levels remained significant, confirming the robustness of the study's main findings.

Table 8.

Regression results of PSM-DID and outlier test.

Variables	(1)	(2)	Outlier test
	Psmdid	Psmdid1	Truncated 1 percent	Truncated 5 percent
	Labor	Labor	(3)	(4)
Lccpost	0.0482***	0.0529***	0.0344**	0.0329*
	(2.7412)	(3.4443)	(2.0146)	(1.9144)
Tobin	0.0086	0.0138**	0.0086	0.0088
	(1.5252)	(2.1181)	(1.4926)	(1.4715)
LnSale	0.3124***	0.3692***	0.3170***	0.3153***
	(7.3791)	(13.5230)	(7.6030)	(7.2274)
Wage	−0.0548***	−0.0560***	−0.0531***	−0.0528***
	(−18.6647)	(−20.3724)	(−18.9877)	(−18.7728)
Size	0.3854***	0.3451***	0.3708***	0.3306***
	(8.4842)	(11.0254)	(8.5390)	(7.6153)
Lnage	0.0797***	0.0547***	0.0857***	0.0826***
	(5.3109)	(3.0317)	(5.8656)	(5.5313)
Constant	−3.5241***	−4.1919***	−3.4537***	−2.8944***
	(−7.5024)	(−8.7249)	(−7.1348)	(−5.6990)
Observations	12,894	6205	12,650	11,598
R2	0.9657	0.9725	0.9637	0.9535

Notes: Stars (*, **, and ***) are used to indicate statistical significance at the 10%, 5%, and 1% confidence levels, respectively, with the T-statistics listed in parentheses.

Analysis of impact mechanisms

Output effect test

This study explored how the low-carbon city pilot policy could influence employment by way of its impact on output, with the city gross regional product serving as an indicator of output variation. The city's gross regional product was used to assess how the low-carbon city pilot policies affected employment, by analyzing shifts in economic output. The results in column (1) of Table 9 demonstrate the impact of the low-carbon city pilot policy on firm production.

Table 9.

Impact of pilot low-carbon city policies on employment through output effects.

Variables	Lngdp	Lngdp	Labor	Labor
	（1）	（2）	（3）	（4）
Lccpost	0.0376***	0.0369***
	(4.4218)	(4.4032)
Tobin	0.0022		0.0083
	(1.2614)		(1.4438)
LnSale	0.0020		0.3116***
	(0.2747)		(7.3784)
Wage	−0.0006		−0.0547***
	(−0.5885)		(−18.7045)
Size	0.0070		0.3840***
	(0.8017)		(8.4917)
Lnage	0.0010		0.0810***
	(0.1379)		(5.4408)
Lngdp			0.1564***	0.2374***
			(3.4484)	(3.0639)
Constant	8.4084***	8.5528***	−4.8073***	5.7266***
	(64.4751)	(2229.7955)	(−8.3980)	(8.6241)
Observations	12,894	13,149	12,894	13,149
R 2	0.9884	0.9885	0.9658	0.9178

Notes: Stars (*, **, and ***) are used to indicate statistical significance at the 10%, 5%, and 1% confidence levels, respectively, with the T-statistics listed in parentheses.

The coefficient of 0.0376 for lccpost indicated a statistically significant effect at the 1% significance level. As depicted in column (3), the coefficient for lngdp (the logarithm of the GDP) was 0.156, demonstrating statistical significance at the 1% level. This indicated that, by implementing the low-carbon city pilot policy, enterprises will adopt green technological innovation to improve product performance and production efficiency for future sustainable development, so that output increases offset the cost of emission reductions, thus increasing the demand for the labor force and promoting jobs.

Factor substitution effects

Factor substitution affects the labor market through the pilot policies of low-carbon cities and, thus, affects employment. Optimizing the industrial structure often means a transfer from traditional and low-value-added industries to high-tech and high-value-added industries, which can create more knowledge- and skill-intensive jobs and improve the employment quality of the labor force. Optimizing the industrial structure increases the proportion of the service industry and plays a crucial role in absorbing labor and employment. To examine this transition, this study used the tertiary sector's share of regional GDP to measure industrial structural optimization. Table 10 presents the assessment results. In Column (1) of Table 9, the estimated coefficient for lccpost is 0.0130, statistically significant at the 1% confidence level. This indicates that the low-carbon city program in the pilot areas promoted a shift toward a service-based industrial structure, increasing a more balanced industrial mix. The third column displays the estimation outcomes regarding the impact of industrial structure optimization on employment. The coefficient for industrial structure optimization is 0.122, statistically significant at the 10% level. This suggests that the low-carbon city pilot policy enhanced employment opportunities by promoting an improved industrial structure. Industrial structure optimization could also extend and expand the industrial chain, and new industrial chain links could create employment opportunities. Therefore, industrial structure optimization positively impacted employment by promoting technological innovation, industrial upgrading, improved labor market adaptability, and industry chain extension.

Table 10.

Empirical results of factor substitution effects.

Variables	IS	IS	Labor	Labor
	(1)	(2)	(3)	(4)
Lccpost	0.0130***	0.0137***
	(2.8973)	(3.0907)
Tobin	0.0028**		0.0083
	(2.0689)		(1.4452)
LnSale	−0.0008		0.3120***
	(−0.1980)		(7.3711)
Wage	−0.0000		−0.0548***
	(−0.0358)		(−18.6038)
Size	−0.0016		0.3852***
	(−0.3304)		(8.4966)
Lnage	0.0094**		0.0801***
	(2.4326)		(5.3491)
IS			0.1220*	0.1251
			(1.8985)	(1.1910)
Constant	0.4929***	0.4782***	−3.5493***	7.7000***
	(7.1405)	(235.3277)	(−7.5347)	(151.3335)
Observations	12,894	13,144	12,894	13,144
R2	0.5540	0.5536	0.9656	0.9174

Notes: Stars (*, **, and ***) are used to indicate statistical significance at the 10%, 5%, and 1% confidence levels, respectively, with the T-statistics listed in parentheses.

Heterogeneity analysis

Pollution intensity of the industry to which the enterprise belongs.

By cross-referencing the pollution-intensive sector classifications from Wu et al. ⁶³ and Zhang et al., ⁶⁴ we identified energy-consuming industries based on the 2012 revised industrial classification framework for listed companies, as per the China Securities Regulatory Commission. The regression analysis results (Table 10) revealed that the low-carbon pilot initiative had a notable positive impact on employment in less energy-intensive sectors. Conversely, the policy's effect on employment in high-energy-consuming industries was statistically insignificant. The empirical analysis confirmed that the low-carbon pilot policy had a threshold effect on firms’ employment incentives, with industries characterized by high energy consumption showing lower efficiency in both energy use and productivity. By contrast, industries with lower energy consumption faced fewer constraints in achieving green and low-carbon transitions. These industries were more adept at exploring diverse low-carbon transformation strategies, yielding better outcomes and gaining a competitive edge in terms of employment enhancements during the transition.

Heterogeneity analysis based on firms’ factor intensity.

Factor intensity reflects the size of a firm's labor demand. To further investigate whether the employment effectiveness of the pilot policies of low-carbon cities differed among enterprises with different factor intensities, following a study by Dai et al. ⁶⁵ regarding the effect of labor-intensive transformation and upgrading on employment, this study categorized labor-intensive industries based on industry codes (columns (3) and (4) of Table 11). The low-carbon city pilot policy had a more pronounced effect on employment in less labor-intensive firms than in more labor-intensive firms, suggesting that expanding the low-carbon sector requires a workforce with high skill and knowledge levels. This demand has prompted a transformation in the labor market, with an increasing need for highly skilled workers. Consequently, employment opportunities in sectors less reliant on labor-intensive activities have expanded. This shift means that economic growth is no longer dependent only on an increase in the number of laborers but also more on technological progress and productivity gains.

Table 11.

Heterogeneity analysis of the impact of pilot low-carbon city policies on employment.

Variables	Labor	Labor	Labor	Labor
	High-energy-consuming industries	Low-energy-consuming industries	Labor-intensive	Non-labor-intensive
Lccpost	0.0029	0.0541***	0.0217	0.0626***
	(0.0803)	(2.8190)	(0.7557)	(2.7503)
Control variables	YES	YES	YES	YES
City fixed effects	YES	YES	YES	YES
Year fixed effects	YES	YES	YES	YES
Constant	−3.4173***	−3.4173***	−4.0597***	−3.1497***
	(−6.7376)	(−6.7376)	(−6.3397)	(−5.0166)
Observations	3089	9781	4154	8592
R 2	0.9735	0.9660	0.9748	0.9669

Notes: Stars (*, **, and ***) are used to indicate statistical significance at the 10%, 5%, and 1% confidence levels, respectively, with the T-statistics listed in parentheses.

Conclusions and relevance

This study employs the low-carbon city pilot policy as a quasi-natural experiment to examine its impact on employment within traditional revolutionary areas. The research yields the following key findings:

The substantial decline in carbon emissions within the old pilot revolutionary base areas has been accompanied by a notable enhancement in employment levels, attributable to the low-carbon city pilot policy. The robustness of these findings was confirmed through rigorous testing, ensuring the reliability of the conclusions.

The low-carbon city pilot policy has catalyzed an increase in labor demand and job creation by incentivizing enterprises to embrace green technological innovations, enhance production efficiency, and expand output. Furthermore, the policy has facilitated the optimization of the industrial structure, particularly the shift toward the service sector, thereby augmenting the need for labor and elevating the quality of employment opportunities.

The heterogeneity analysis revealed that the low-carbon pilot policies exerted a more pronounced effect on employment within low-energy and non-labor-intensive industries, signifying that the policy's influence varies across diverse sectors and enterprise categories. Drawing from these findings, we articulate a set of policy recommendations tailored to address these disparities.

First and foremost, the advancement of low-carbon transitions should be intensified. Traditional revolutionary base areas must persist in their commitment to low-carbon development strategies, bolstering support for green industries, fostering the optimization and elevation of the industrial structure, encouraging green technological innovation, and generating employment opportunities for the local populace. This study's findings indicate that the low-carbon city pilot policy has effectively curtailed urban carbon emissions in these historical regions without adversely affecting employment. On the contrary, it has enhanced the employment landscape, yielding the dual benefits of energy conservation, emission reduction, and job creation. Consequently, the low-carbon city pilot policy has spurred the green transformation of conventional industries, enhanced resource efficiency, and nurtured the growth of sectors such as new energy, energy conservation, environmental protection, and the circular economy, thereby bolstering industrial competitiveness and stabilizing or increasing employment opportunities.

Secondly, it is imperative to refine the employment structure. The low-carbon city pilot policy is particularly targeted at high-energy-consuming industries. Lessons from China's low-carbon city pilots suggest that while the policy fosters employment in low-energy sectors, its impact on high-energy industries is negligible. This discrepancy may stem from the policy's incentive for enterprises to undergo green transformation, a process that necessitates a higher caliber of skilled professionals, thereby facilitating structural changes in the labor market and enhancing employment quality. Pilot cities should tailor low-carbon transformation strategies to the specific needs of various industries, aiming for nuanced management. The cultivation and expansion of green and low-carbon industries should be actively encouraged to generate new jobs. For industries with high-carbon footprints, a strategic approach involving upgrading, transformation, or phased elimination, based on the complexity and pace of change, is essential. Concurrently, for employees in affected industries, comprehensive support in the form of career diversification, relocation, training, and re-employment opportunities must be provided to ensure a seamless transition.

Thirdly, technological innovation has elevated the quality of employment. The pilot initiatives of low-carbon cities in China have demonstrated that the shift toward a green, low-carbon economy can boost employment through both output expansion and factor substitution effects. Innovations in technology and management can further augment enterprise productivity and facilitate factor substitution, thereby amplifying the positive impact of these drivers on employment. Consequently, pilot cities within the traditional revolutionary base areas should proactively establish service platforms for scientific and technological innovation, refine innovation incentive structures, and foster an environment conducive to widespread technological and institutional innovation. By doing so, they will encourage the adoption of innovative practices across society.

This study is not without its limitations. China's traditional revolutionary base areas, often encompassing administrative frontier and provincial remote regions, are characterized by rugged terrain and ecological fragility. The economic underdevelopment in these regions can be attributed to a complex interplay of natural, policy, and market factors. The intricate regional dynamics present challenges in deriving universally applicable conclusions about the enhancement of employment. Furthermore, the present research primarily concentrates on the city-level implications. Future studies should delve into the varying social impacts of low-carbon city pilot policies across different regions and city types, as well as the underlying causes for these discrepancies. Additionally, international comparative research on the outcomes of low-carbon policies in less-developed areas would be beneficial. Such studies could inform and refine China's low-carbon city pilot policies by drawing on global insights and experiences.