Does the Low-Carbon Transition Affect the Wage Level of Enterprises? Evidence from China’s Low-Carbon City Pilot Policies

Abstract

ESG (Environmental, Social, and Governance) performance is necessary to establish sustainable corporations. As the scale effect of the market application of low-carbon technologies is yet to be realised, China’s enterprises may find it difficult to balance their environmental (low-carbon transition) and social (increase in wages) responsibilities, and are caught in a governance dilemma. Therefore, in order to test the above hypotheses, we utilised the staggered difference-in-differences method to investigate the wage effects of low-carbon transformations. The results are displayed as follows. First, the low-carbon transition constraint initially increases firms’ labour demand and reduces energy consumption, leading to lower productivity and thus negatively affecting firms’ wages, even though the low-carbon transition can promote internal equity. Second, the impact of the policy on wages is heterogeneous. It has a more pronounced negative impact on enterprise wages in state-owned enterprises, old enterprises, primary and secondary industry enterprises, and low-carbon enterprises, whereas its promoting effect on internal fairness within old enterprises, secondary industry enterprises, and low-carbon enterprises is more significant. Finally, the negative impact of low-carbon policies on corporate wages gradually decreases while strengthening the promoting effect on the internal fairness of employee remuneration. The above results confirm that at this stage, when China’s enterprises are pursuing the environmental goal of low-carbon transition, it will affect their efficiency in the short term, which in turn will lead to lower wages. Moreover, this problem is too difficult to be solved by enterprises alone and requires assistance from the government. Under the ESG concepts, we provide insights into how to coordinate policies to improve living standards and promote low-carbon transformations.

1. Introduction

Recently, the ESG movement has increased in corporations all over the world [1]. ESG measures the sustainability of corporate development through three dimensions: Environmental, Social, and Governance. ESG essentially requires companies to gradually shift from profitability to the pursuit of sustainable development, from maximising shareholders’ interests to taking into account the value of shareholders and other stakeholders, and to taking the initiative to assume social responsibility and develop corresponding systems. When ESG gradually becomes an industry consensus and a global trend, ESG can help companies gain a more favourable position in the market competition. Thus, the ESG performance is necessary to establish sustainable corporations.

Environmental challenges, such as the global greenhouse effect, represent significant issues crucial to the survival and advancement of humanity. China, in aligning with its national circumstances, has assumed international responsibilities by actively advocating for the transition to a green economy. To achieve “carbon neutrality” and “carbon peaking”, China has instituted several policies targeting the reduction in carbon emissions. These policies encompassed three waves of provincial and municipal pilot initiatives commencing in 2010, 2012, and 2017, spanning 81 urban pilot projects across diverse geographical areas, as depicted in Figure 1. These initiatives were designed to address spatial configurations, explore trajectories for low-carbon transformations, advocate for low-carbon policies, and promote the evolution of an ecological civilization through initiatives fostering green and low-carbon development.

After the implementation of these pilot projects, China experienced both environmental and economic advancements. The construction of low-carbon cities has improved air quality [2,3], reduced industrial pollution [4], and encouraged technological innovation [5]. However, against the background of internal structural adjustment and external international changes, China’s economy is facing greater downward pressure. Considering that China’s carbon emissions are highly correlated with economic growth at this stage [6], the low-carbon transition will further increase the downward pressure on the economy.

The state of the economy has a direct impact on the operational performance of companies, which in turn affects the wage levels. In order to cope with the risk of uncertainty brought about by the economic downturn, residents are increasingly prioritising stable income sources, notably salaries, which directly influence their livelihoods. According to China’s National Bureau of Statistics, the share of per capita wage income in disposable income for residents in 2023 was 56.2%.

In other words, at this stage, China’s enterprises practicing ESG concepts will be caught in a dilemma: emphasising the environmental effectiveness of low-carbon transformation may affect the promotion of employees’ wages, while emphasising the social responsibility of safeguarding employees’ wages may shelve the enterprise’s low-carbon transformation. Consequently, it is a huge challenge in governance for companies to ensure that they maintain the living standards of their employees while pursuing a low-carbon transition. Therefore, the research purpose of this paper is to study the impact of the low-carbon transition on the wage level of enterprises.

Theoretically, a low-carbon transition may improve wages. First, the shift to a low-carbon economy has driven technological progress [7], enhanced product competitiveness, and boosted corporate profits, increasing wages for employees [8]. Second, technological innovation increases labour productivity and raises business wages. Finally, the technology spillovers resulting from an enterprise’s technological advancement can help decrease wage gaps within industries and promote external fairness.

However, a low-carbon transition can also increase the operating costs of enterprises as quasifixed expenses [9], reducing their profits and, hence, their wage levels. Meanwhile, as enterprises undergo low-carbon transformations, labour demand may change [3], which could impact enterprise wage levels. Moreover, senior management may leverage their positional authority to negotiate higher remuneration, thereby exacerbating internal disparities in employee compensation.

In summary, the impact of the low-carbon transition on wages is theoretically uncertain and remains to be empirically tested further [10]. Nonetheless, scant research exists on the impact of low-carbon transitions on wage levels. Therefore, a thorough examination of the ramifications of low-carbon transformations on enterprise wage dynamics is imperative to address both its theoretical underpinnings and practical implications.

Based on many existing studies, we gathered microdata between 2007 and 2022 on publicly traded A-share companies on the Shanghai Stock Exchange and Shenzhen Stock Exchange and then used the staggered difference-in-differences (DID) method, in which the treatment and control groups were divided based on whether the city where the company was located was a pilot city or not, to analyse the impact of the low-carbon city pilot (LCCP) policy on these enterprises’ wage levels.

2. Literature Review

We focus on studying the enterprise wage levels in low-carbon cities. This part of the paper summarises two branches of the relevant literature. The first branch includes studies on various aspects of LCCP policies. The second branch mainly focuses on the impact of environmental regulations on the wage level and internal wage inequality.

2.1. Evaluations of LCCP Policies

The development of low-carbon cities offers a practical solution for reducing carbon emissions in urban areas and is an important measure for enhancing national competitiveness. Since the implementation of the LCCP policies in China, the number of related studies has increased rapidly. However, most studies assess only the environmental and economic impacts of such policies. To study the environmental effects of the policy, academics focus on policy impacts on emissions, both carbon emissions intensity and efficiency. Several studies have found that building low-carbon cities effectively improves air quality [11]. Yang et al. [4] found that the policy has the potential to significantly reduce corporate carbon emissions through environmental protection technology and the ability to innovate. X. Liu et al. [12] argued that the policy significantly suppressed urban carbon emissions and explored synergies among LCCP policies, pilot policies for new energy vehicles, and pilot policies for the carbon market. Studies have shown that LCCP policies can also reduce the intensity of carbon emissions [13]. Du et al. [14] studied the impact of policies on carbon emissions efficiency and suggested a spatial spillover effect.

Several researchers have studied the economic impact of policies on technological innovations. Zhu and Lee [15] discovered that LCCP policies promote technological innovation in both pilot cities and surrounding areas. Qu et al. [16] examined policies that encourage innovation in energy technology. Policies not only enhance urban green total factor productivity [17] but also improve green efficiency [18]. According to Cheng et al. [19], low-carbon pilot policies can boost economic growth. Wang et al. [20] argued that policies can boost the digital economy, reduce energy consumption and carbon emissions, and drive green development.

Academic research has predominantly focused on the transition to a low-carbon economy through LCCP policies. Studies investigating the environmental and economic impacts of LCCP policies often concentrate on areas such as emissions reductions, overall productivity, green innovation, and sustainable economic growth. However, consideration for the social impacts of these policies is lacking, particularly the potential effects on wages resulting from the shift to a low-carbon economy.

2.2. Effect of Environmental Regulations on Wages

Environmental regulation is not a new topic. Many scholars have explored the socioeconomic effects of environmental regulations through the labour market response to these regulations (such as changes in the employment structure or employment rate). However, research on how environmental regulations affect wages is still in its early stages, and the conclusions reached thus far are inconsistent.

Some researchers believe that environmental regulations might hurt wages. Henderson [21] discovered that environmental regulations can increase company costs, reducing the production scale and wages as a cost-cutting measure. Mishra and Smyth [22] analysed matched employer–employee data in Shanghai to reveal how firms may transfer the burden of environmental regulations to workers by reducing wages. Qin et al. [23] utilised panel data on cities and firms and found that environmental regulations in highly polluting cities result in lower wages, especially for unskilled workers.

Other scholars argued that environmental regulations lead to higher wages. Berman and Bui [9] discovered that environmental regulations can boost the demand for labour, resulting in increased wages. Gray et al. [24] conducted a study on pulp and paper industry plants in the United States using DID models and found that under environmental regulations, wages increase slightly, and businesses do not reduce wages in place of reducing employment. Additionally, Chakraborty et al. [25] examined the prohibition policy of “azo fuel” in Germany and found that the policy led to changes in the composition of the labour force within firms and increased the wages of managers.

Some researchers argued that environmental regulations have a neutral impact on wages. Chao et al. [26] discovered that environmental regulations affect skilled and unskilled workers’ wages in the trade sector differently, which could reduce wage inequality. On the other hand, Yang and Xu [27] found that labour productivity increased significantly after the implementation of the environmental law; however, the change in actual wage levels was not obvious.

Numerous studies have explored the influence of environmental regulations on wage levels; however, due to variations in researchers’ backgrounds and data sources, consensus findings remain elusive. Studies on wage equality often focused on inequality between workers at different skill levels or genders. Researchers have also studied the effects of salaries on executive-level employees versus regular staff members. The impact mechanism is usually studied from a single perspective, and there is no consensus.

2.3. Literature Summary

Summarising and analysing the existing literature, scholars have achieved fruitful research results in the field of the environmental and economic effects of pilot policies for low-carbon cities. Existing studies have not only explored the effects of the policy in depth but also analysed its mechanism of action in detail. However, research attention to the social effects of pilot low-carbon city policies, especially on the labour market, is still insufficient. While there has been an initial exploration of the literature in the area of labour demand, there is a relative dearth of research on how low-carbon policies affect wages, a key element of the labour market. An in-depth analysis of the impact of policies on wages will help to provide a more comprehensive understanding of the actual effects of policies on the labour market and provide a more solid theoretical basis for policy formulation and optimisation.

The number of studies on the impact of environmental regulation on wage levels is also considerable; however, due to the differences in researchers’ backgrounds, datasets adopted, and models constructed, resulting in a lack of consistency between the findings of the studies, even in the same country, there are significant differences in the results of the analysis of data at the city and enterprise levels, which need to be explored in depth.

In addition, the existing literature tends to adopt a single dimension to study the mechanism of action, isolating labour demand and wage factors, which makes it impossible to form a unified theoretical framework in the analysis of the mechanism of action from a doctrinal point of view.

The possible contributions of this paper are as follows. First, given the specific policy context of LCCP policies, no research has explored its impact on wage levels, wage equity, or its mechanism of action, so this study is the first to explore the effects of such policies on wages, providing a new perspective for researchers. Second, we also examined how the policy affects corporate wages and wage gaps with the aim of theoretically unifying multiple influence mechanisms and eliminating multicollinearity between the variables. Additionally, we delved into the heterogeneity of firms and policies and investigated the trends in the policies’ effects. Finally, our findings elucidate how the transition to low-carbon practices can be aligned with social well-being.

3. Theoretical Mechanism

In a review of the recent literature on wage setting, Card [28] noted a growing belief that companies possess certain rights to set wages. The impact of low-carbon transformations on wages has been associated with three main wage determination theories: marginal productivity wage theory, shared wage theory, and supply and demand equilibrium wage theory (Figure 2).

First, Porter argued that moderate environmental regulations not only facilitate emissions reduction among firms but also incentivise investments in scientific and technological innovation, thereby enhancing production technologies and product competitiveness [23]. Clark’s marginal productivity wage theory underscores the critical influence of marginal productivity in determining wage levels. Theoretically, transitioning to low-carbon practices could elevate wages by stimulating technological innovation.

Second, Martin Weizmann’s shared wage theory points out that employees’ wages are no longer fixed wages linked to working hours but are linked to the company’s operating conditions (profit and other indicators). In other words, profit is also a key indicator for companies to determine wage levels. Berman and Bui [9] noted that environmental regulations increase the cost burden on firms in the form of a quasifixed cost, impacting corporate profits and potentially leading to lower wages. However, Chen et al. [7] found that enterprises should be more capable of technological innovation under the moderate pressure of low-carbon transformations. Gray et al. [24] also discovered that technological progress contributes to increased wages. Moreover, technological innovations can increase corporate profits, leading to higher wages. Therefore, the impact of a low-carbon transition on wages depends on whether it increases or decreases corporate profits.

Third, Marshall’s equilibrium wage theory of supply and demand states that the wage level of businesses is determined by the supply and demand situation in the labour market. Simply put, an increase in employment (demand in the labour market) leads to an increase in wages. To avoid the rising costs resulting from emissions reductions, businesses may decrease their investments in labour. Therefore, wages and employment can have a substitutional relationship. Chakraborty et al. [25] discovered that transitioning to low-carbon practices promotes employment, whereas Zheng et al. [3] found that increased environmental regulations reduce enterprise labour demand.

In summary, the policy’s impact on firm wages is influenced by labour productivity, corporate profits, and labour demand, and this impact can be either positive or negative. Although enterprises under the constraint of the low-carbon transition favour increasing green technology innovation, the current emission reduction technology lacks a scale effect and has yet to have a significant market effect. So, the following hypothesis 1 is formed:

H1: 

Low-carbon transformations currently dampen wage increases.

According to optimal contract theory, shareholders need to design necessary compensation contracts for executives to mitigate the conflicts of delegated agencies. However, executives can still influence contracts through their power to demand more compensation for themselves, leading to internal inequality in companies. Low-carbon transformations drive companies to optimise their resource utilisation and organisational structures, converging executive power to an optimal level and promoting internal fairness. So, the following hypothesis 2 is formed:

H2: 

Low-carbon transformations can improve fairness within corporate wages.

In contrast, the LCCP policy encourages the diffusion of technological progress across industries, reducing the “market power” of innovative companies due to technology spillovers, which results in lower wages. However, companies with weak innovation capabilities can benefit from the externalities of technology and increase productivity levels, thus increasing wages. Therefore, the following hypothesis 3 is formed:

H3: 

Low-carbon transformations can improve wage differentials between industries.

4. Econometric Model and Data Description

4.1. Econometric Model

In this paper, we use the DID approach to identify and examine the wage effects and mechanisms of low-carbon pilot policies.

4.1.1. Emissions Reduction Effect of LCCP Policies at the City Level

We used the staggered DID method to explore the emissions reduction effect of LCCP policies. The model is set as follows:

$${CO}_{2ct}={\alpha }_1+{\theta }_1{citytreated}_{ct}+{\lambda }_1{Z}_{1ct}+{\eta }_c+{\mu }_t+{\epsilon }_{1ct},$$

(1)

where ${CO}_{2ct}$ is the carbon dioxide emissions of city $c$ in year $t$, ${\alpha }_1$ is a constant term, and ${\theta }_1$ is the DID estimator, which is the city-level emissions reduction effect estimate of the LCCP policy. The core explanatory variable ${citytreated}_{ct}$ is a dummy variable that measures whether city $c$ implements LCCP policies in year $t$. The dummy variable takes the value 1 if the city implements the policy and 0 if it does not. $Z_{1ct}$ is a set of control variables that affect the carbon emissions of city $c$ in year $t$. The control variables in this model are population size, economic development level, energy structure, urbanisation level, and industrialisation level. ${\mu }_t$ and ${\eta }_c$ are year and city fixed effects, respectively, and ${\epsilon }_{1ct}$ is a random disturbance item.

4.1.2. Impact of LCCP Policies on Enterprise Wage Levels

We also conducted research at the firm level to assess the influence of policies on firm wage levels. The model settings are as follows:

$${wage}_{it}={\alpha }_2+{\theta }_2{treated}_{it}+{\lambda }_2{X}_{it}+{\phi }_i+{\mu }_t+{\epsilon }_{it}$$

(2)

$${Gap\_in}_{it}={\alpha }_3+{\theta }_3{treated}_{it}+{\lambda }_3{X}_{it}+{\phi }_i+{\mu }_t+{\epsilon }_{it}$$

(3)

$${Gap\_out}_{it}={\alpha }_4+{\theta }_4{treated}_{it}+{\lambda }_4{X}_{it}+{\phi }_i+{\mu }_t+{\epsilon }_{it}$$

(4)

where ${wage}_{it}$ represents the salary level of enterprise $i$ in year $t$, ${wage}_{it}=\mathrm{l}\mathrm{n}\left(\right({grosss alary}_{it}-{total executive salary}_{it})/({labor}_{it}-{{executives}^{\prime } number}_{it}))$; ${Gap\_in}_{it}$ represents the inner wage gap of enterprise $i$ between executives and general employees in year $t$, $Ga{p\_in}_{it}=\mathrm{l}\mathrm{n}({(total executive salary}_{it}/{executive{s}^{\prime } number}_{it})-{wage}_{it})$; and ${Gap\_out}_{it}$ represents the external wage gap between the wage level of enterprise $i$ and its industry, ${Gap\_out}_{it}=\mathrm{l}\mathrm{n}\left({wage}_{it}/{industry wage}_{max}\right)$. ${\theta }_2$, ${\theta }_3$, and ${\theta }_4$ are the DID coefficients used to measure the effects of policy on corporate wages and internal and external wage gaps, respectively, and ${treated}_{it}$ is a dummy variable that measures whether the city in which the enterprise $i$ is located implemented its LCCP policy in year $t$. If so, then the value is 1; otherwise, it is 0. $X_{it}$ is a set of control variables that affect the wages of employees of enterprise $i$ in year $t$. The control variable indicators selected in models (2)–(4) include income tax expenses, enterprise scale, sales expense ratio, growth ability, and operating income. ${\phi }_i$ is the fixed effect of the enterprise, and ${\mu }_t$ is the same as in Equation (1). ${\epsilon }_{it}$ is a random effect that affects the firm’s wages.

4.2. Sample Selection and Data Description

In this paper, panel data from 287 cities in China from 2007 to 2022, including 121 pilot cities and 166 nonpilot cities, are used. In model (1), the explained variable ${CO}_{2ct}$ is the logarithm of carbon emissions from the China Emission Accounts & Datasets (CEADs), and the data for the control variables were collected from the China City Statistical Yearbook.

In addition to the city level, we also explore the wage effect of the LCCP policy at the enterprise level (wage level and wage gap). As samples, we selected A-share companies listed on the Shanghai Stock Exchange and Shenzhen Stock Exchange from 2007 to 2022. The treatment and control groups in the DID method were divided based on whether the city where the firm was located was a pilot city or not. To avoid extreme values and bias the results, three types of enterprise samples were removed: (1) enterprises with “ST” and “*ST” in their stock codes; (2) listed companies whose listing status is “ST”, “*ST”, “suspended from listing”, or “terminated from listing”; and (3) listed companies with serious data shortages. Ultimately, unbalanced panel data from 3084 listed companies were obtained. In models (2)–(4), the explained variables and the control variables were obtained from the China Stock Market Accounting Research (CSMAR) database. We also used some data and code published by Wang and Ge [29]. The variables and their descriptions are shown in

Table 1. Variables and their measures.

	Variable	Variable Descriptions
City level	Economic development	Logarithm of GDP per capita
	Population size	Logarithm of the year-end population of the city
	Industrialisation level	Ratio of the added value of the secondary industry to the GDP of the region
	Urbanisation level	Ratio of urban population to total population
	Energy structure	Ratio of coal consumption to total energy consumption
Enterprise level	Enterprise size	Logarithm of the total assets of the enterprise
	Sales expense ratio	Ratio of sales expenses to operating income
	Income tax expense	Logarithm of 1 plus income tax expenses
	Growth ability	Tobin’s Q
	Operating income	Logarithm of operating income

.

5. Empirical Results

5.1. Baseline Regression Results

5.1.1. Regression Results of the Emissions Reduction Effects of LCCP Policies

The empirical evidence first tests whether the low-carbon pilot city policy has a significant abatement effect, which is the factual basis for the subsequent study. In

Table 2. Emissions reduction effects of LCCP policies.

Variable	CO2
City treated	−0.0295 * (0.0153)
Control variable	Yes
City-year fixed effects	Yes
R2	0.9872

Note: * represents the significance levels of 10%, and the values in brackets are cluster robust standard errors.

, the results of the tests that examined the impact of city-level emissions reduction, incorporating control variables and city–year fixed effects, are presented. The findings indicate that the adoption of the LCCP policy has significantly reduced carbon dioxide emissions in pilot towns by approximately 2.95% compared to those in nonpilot cities. These results align with previous studies in the field.

5.1.2. Regression Results of the Influence of LCCP Policies on Enterprise Wages

In

Table 3. Influence of LCCP policies on enterprise wages.

Variable	Wage (1)	Wage (2)	Gap_In (3)	Gap_Out (4)
Treated	−0.0629 * (0.0327)	−0.0549 * (0.0324)	−0.0500 *** (0.0179)	−0.0016 (0.0033)
Control variable	No	Yes	Yes	Yes
Firm–year fixed effects	Yes	Yes	Yes	Yes
R2	0.7463	0.7301	0.7491	0.6621

Note: * and *** represent the significance levels of 10% and 1%, respectively, and the values in brackets are cluster robust standard errors.

, column (1) presents the test results involving enterprise and year fixed effects without the inclusion of control variables. The estimated DID is −0.0629. Column (2) shows the regression results after adding control variables and firm–year fixed effects to the models. At the 10% significance level, the estimated value of the double difference is −0.0549. These results indicate that the LCCP policy has a significantly negative impact on the enterprise wage level. The result is the same as expected from research hypothesis 1. The result suggests that the lack of a scale effect of green innovation in Chinese enterprises at this stage leads to the input cost being greater than the benefit, which restricts the rise of wages in enterprises.

Furthermore, columns (3) and (4) reveal that the LCCP policy negatively affects the internal fairness of employees’ remuneration but does not have a significant impact on the external fairness of employees’ remuneration. The above results affirm research hypothesis 2. After the implementation of the pilot low-carbon city policy, enterprises are required to adopt more environmental protection measures and technological innovations, and these external hard constraints limit the authority of executives within the enterprises. In addition, the low-carbon transition needs to be driven systematically, which prompts companies to optimise their resource use and organisational structure to promote internal equity. According to previous theoretical analyses, this will reduce wage differentials within firms. However, the above results do not give a definite answer to hypothesis 3. That is, technology spillovers are not currently working. This may be consistent with the reasons for hypothesis 1. Spillovers require a certain market size to be guaranteed, but the market performance of green innovations is limited.

5.2. Parallel Trend Test

To use the staggered DID method, it is important to ensure that there are parallel trends before the policy implementation, which means that the wage changes for the control and treatment groups should remain consistent. To test for parallel trends, we utilised the event analysis approach outlined by Jacobson et al. [30], expressed as follows.

$${wage}_{it}={\alpha }_2+{\sum }_{t=-3}^6{\sigma }_{t1}{D}_{it}+{\lambda }_2{X}_{it}+{\phi }_i+{\mu }_t+{\epsilon }_{it},$$

(5)

$${Gap\_in}_{it}={\alpha }_3+{\sum }_{t=-3}^6{\sigma }_{t2}{D}_{it}+{\lambda }_3{X}_{it}+{\phi }_i+{\mu }_t+{\epsilon }_{it},$$

(6)

where $D_{it}$ is a dummy variable indicating whether the city in which enterprise $i$ is located implemented the LCCP policy in year $t$. ${\sigma }_{t1}$ and ${\sigma }_{t2}$ represent the wage and internal unfairness differences between pilot and nonpilot cities, respectively, and are the coefficients from the parallel trend test. The remaining variables are the same as in Equations (2) and (3).

The available data from four years ago and six years after the policy implementation are not abundant. Therefore, we summarised the data four years before the policy implementation in the −4 period and the data six years after the policy implementation in the 6 period. We used the data from the fourth period before the LCCP policy was implemented as the base period. After analysing the data at the enterprise level, the results of the parallel trend test are presented in Figure 3. The results indicate that before the policy was implemented, the significance test did not meet the 10% threshold. There are no significant differences between the pilot and nonpilot cities. In other words, the parallel trend test is successful, and the benchmark regression results are established.

5.3. Placebo Test

Even if the policy was implemented, the difference between the control and experimental groups could be caused by time. Therefore, we combined the results of the parallel trend test to advance the policy by 3, 4, and 5 years, respectively. The spurious time intervals referred to as time 1, time 2, and time 3 exhibit negative DID coefficients in

Table 4. Placebo test based on time.

Variable	Wage (1)	Gap_In (2)
time 1	−0.0538 (0.0491)	−0.0293 (0.0228)
time 2	−0.0373 (0.0526)	−0.0195 (0.0233)
time 3	−0.0033 (0.0607)	−0.0196 (0.0273)
Control variable	Yes	Yes
Firm–year fixed effects	Yes	Yes
R2	0.7300	0.7490

when Equations (2) and (3) are applied but fail to reach statistical significance at the 10% threshold. This outcome indicates the absence of substantial disparities in terms of time trends between pilot and nonpilot cities. Furthermore, the LCCP policy does not have a statistically significant inhibitory impact on enterprise wages due to time or other accidental factors.

To mitigate the influence of unobserved variables in the baseline regression, we employed a random selection method to designate 121 cities from the dataset as a control group for conducting placebo tests. This procedure was iterated 500 times to generate 500 sets of estimated coefficients and corresponding p values. The findings, depicted in Figure 4, reveal normally distributed density curves of estimated coefficients, with the majority of regressions showing nonsignificance. The statistically significant disparity between the benchmark regression coefficient and the pseudocity regression coefficient suggests that the impact of the LCCP policy on wage and intrafirm inequality is not a low-probability event.

5.4. Robustness Tests

5.4.1. Data Filtering

To avoid the influence of outliers on the regression outcomes, we applied data censoring at 1%, 5%, and 10%.

Table 5. Robustness test on the impact of LCCP policies on enterprise wages and internal fairness of employees’ wages—I.

	Censored 1%		Censored 5%		Censored 10%
Variable	Wage (1)	Gap_In (2)	Wage (3)	Gap_In (4)	Wage (5)	Gap_In (6)
Treated	−0.0549 * (0.0324)	−0.0500 *** (0.0179)	−0.0549 * (0.0324)	−0.0500 *** (0.0179)	−0.0483 * (0.0289)	−0.0500 *** (0.0179)
Control variable	Yes	Yes	Yes	Yes	Yes	Yes
Firm–year fixed effects	Yes	Yes	Yes	Yes	Yes	Yes
R2	0.7301	0.7491	0.7301	0.7491	0.7830	0.7491

Note: * and *** represent the significance levels of 10% and 1%, respectively, and the values in brackets are cluster robust standard errors.

shows that LCCP policies exert a suppressive impact on firm wages, all exhibiting statistical significance at the 10% level across the 1%, 5%, and 10% cutoffs. Additionally, the policies have a significantly negative effect on intrafirm inequality, with all the coefficients passing the significance test at the 1% level. This reaffirms the robustness of the baseline regression outcomes.

5.4.2. Adding Benchmark Variables

To ensure that the regression analysis captures the specific impact of the LCCP policy on firm wages and intrafirm inequality rather than potential confounding effects from the political, economic, and geographic characteristics of cities, we introduced a composite variable termed “two control area”. This variable signifies whether a city is a special economic zone, is situated east of the Hu Huanyong Line, and serves as a provincial capital. These four dummy variables interact with the temporal trend variable in the model specification. The Hu Huanyong Line delineates China’s population development levels and socioeconomic patterns, as illustrated in Figure 5.

$${wage}_{it}={\alpha }_2+{\theta }_2{treated}_{it}+{\lambda }_2{X}_{it}+{\varsigma }_2{O}_c\times {timetreated}_t+{\phi }_i+{\mu }_t+{\epsilon }_{it},$$

(7)

$${Gap\_in}_{it}={\alpha }_3+{\theta }_3{treated}_{it}+{\lambda }_3{X}_{it}+{\varsigma }_3{O}_c\times {timetreated}_t+{\phi }_i+{\mu }_t+{\epsilon }_{it},$$

(8)

In Equations (7) and (8), $O_c$ is the introduced dummy variable, $O_c$ is the time trend, and the regression results are as follows.

Table 6. Robustness test of the impact of LCCP policies on enterprise wages—II.

Variable	Wage
	(1)	(2)	(3)	(4)	(5)
Treated	−0.0565 * (0.0324)	−0.0531 * (0.0323)	−0.0551 * (0.0324)	−0.0551 * (0.0324)	−0.0544 * (0.0324)
Control variable	Yes	Yes	Yes	Yes	Yes
Firm–year fixed effects	Yes	Yes	Yes	Yes	Yes
Two control area × 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 side of Hu Huanyong line × time trend	No	No	No	Yes	Yes
R2	0.7301	0.7302	0.7304	0.7301	0.7309

Note: * represents the significance levels of 10%, and the values in brackets are cluster robust standard errors.

indicates that, on inclusion of the transportation variable, the implementation of the LCCP policy has a statistically significant suppressive impact on enterprise wages, achieving significance at the 10% level. Furthermore, the findings presented in

Table 7. Robustness test of the impact of LCCP policies on the internal fairness of employee remuneration—II.

Variable	Gap_In
	(1)	(2)	(3)	(4)	(5)
Treated	−0.0494 *** (0.0178)	−0.0491 *** (0.0178)	−0.0501 *** (0.0179)	−0.0504 *** (0.0179)	−0.0493 *** (0.0178)
Control variable	Yes	Yes	Yes	Yes	Yes
Firm–year fixed effects	Yes	Yes	Yes	Yes	Yes
Two control area × 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 side of Hu Huanyong line × time trend	No	No	No	Yes	Yes
R2	0.7492	0.7492	0.7492	0.7492	0.7493

Note: *** represents the significance levels of 1%, and the values in brackets are cluster robust standard errors.

consistently pass the significance test at the 1% level, confirming the robustness of the baseline regression outcomes.

5.5. Mechanism Analysis

5.5.1. Role of Labour Demand in the Impact of LCCP Policies on Enterprise Wages

According to the balanced wage theory grounded in supply and demand dynamics within the theoretical framework, enterprises’ low-carbon transformations are expected to induce alterations in labour demand, thereby affecting wages. Consequently, we gauged labour demand and conducted a regression analysis. Column (1) of

Table 8. Test of the influence of LCCP policies on enterprise wages.

Variable	Labour	Reward	Power
	(1)	(2)	(3)
Treated	0.0648 *** (0.0193)	−0.0701 *** (0.0270)	−0.0362 ** (0.0301)
Control variable	Yes	Yes	Yes
Firm–year fixed effects	Yes	Yes	Yes
R2	0.9321	0.7662	0.7544

Note: ** and *** represent the significance levels of 5% and 1%, respectively, and the values in brackets are cluster robust standard errors.

shows the regression results, revealing a statistically significant positive association between LCCP policies and employment. Thus, a trade-off between employment and wages is observed.

5.5.2. Role of Productivity Returns in the Impact of LCCP Policies on Firm Wages

According to the marginal productivity wage theory and the shared wage theory within the theoretical framework, it is posited that advancements in science and technology are conducive to enhancing labour productivity, consequently influencing wage levels. Concurrently, fluctuations in labour productivity can impact corporate profitability, highlighting the interdependent relationship between the two variables. In light of this, we utilised productivity rewards as a comprehensive metric to assess the combined effects of labour productivity and profits on wages influenced by the LCCP policy. To address the issue of multicollinearity among intermediary variables, the natural logarithm of the ratio of corporate profits to workforce size was employed. Column (2) of the regression analysis in Table 8 reveals that LCCP policies have a negative short-term impact on productivity rewards, suggesting that room still exists for improvements in translating the technological innovations brought about by the transformation into productivity returns.

5.5.3. Role of Executive Power in the Influence of LCCP Policies on Internal Unfairness

Due to information asymmetry, shareholders have difficulty fully grasping the efforts and management behaviour of executives who have a self-interest tendency and can use their power to obtain more compensation, thus expanding the unfair distribution of internal corporate compensation. We utilised PCA to synthesise four indicators, namely, board size, executive shareholding, equity concentration, and executive concurrent positions, into an executive power variable. Column (3) in Table 8 demonstrates that the LCCP policy promotes internal fairness by reducing executive power.

6. Heterogeneity Analysis

The enterprises were categorised based on four criteria: ownership type, age, industry classification, and whether the enterprise was a high-carbon emissions entity. Subsequently, an investigation into interenterprise heterogeneity was conducted. In addition to evaluating the policy impacts of the three successive batches of LCCP policies, an examination into evolving trends in these policy effects was undertaken.

6.1. Differences in the Effects of Pilot Policies on Wages Based on Ownership Type

Based on enterprise ownership as the classification criterion, enterprises were categorised into three groups: state-owned enterprises, private enterprises, and foreign-funded enterprises. Regression analysis using Equation (2) was conducted separately for each enterprise type, and the findings are presented in

Table 9. Effects of LCCP policies on wages in firms based on ownership type and company age.

Variable	Wage
	State-Owned Enterprises	Private Enterprises	Foreign Enterprises	Old Enterprises	New Enterprises
	(1)	(2)	(3)	(4)	(5)
Treated	−0.0859 * (0.0499)	0.0086 (0.0431)	−0.0962 (0.1491)	−0.0585 * (0.0331)	−0.0962 (0.1491)
Control variable	Yes	Yes	Yes	Yes	Yes
Firm–year fixed effects	Yes	Yes	Yes	Yes	Yes
R2	0.7145	0.7514	0.7541	0.7270	0.7541

Note: * represents the significance levels of 10%, and the values in brackets are cluster robust standard errors.

.

The results indicate that the introduction of pilot policies in low-carbon cities negatively impacted the wages of state-owned enterprises, a statistically significant finding at the 10% level. In contrast, there was no statistically significant impact on the wages of private or foreign-funded enterprises. This divergence could be attributed to the distinct characteristics inherent to each ownership type. State-owned enterprises are heavily influenced by governmental decisions, leading to substantial investments in innovation and emissions reduction, albeit with delayed productivity returns. Consequently, policies adversely affect state-owned enterprises’ wage levels. In contrast, private enterprises exhibit stronger investment preferences and managerial autonomy and typically invest less in technological innovation than state-owned enterprises do [31]. Technological innovation in enterprises generates positive externalities, with private enterprises potentially benefiting from policy-induced innovation investments by state-owned enterprises, thereby positively impacting their wage levels. Foreign-funded enterprises equipped with advanced technologies and superior operational environments experience comparatively fewer impacts from policy interventions.

6.2. Differences in the Effects of Pilot Policies on Wages Based on Company Age

Based on the age of enterprises as the classification criterion, enterprises were categorised into two groups: old (16 years or older) and new (younger than 16 years). Regression analysis using Equation (2) was conducted separately for each category, with the results presented in Table 9, columns (4) and (5) for old and new enterprises, respectively. The findings indicate that the implementation of pilot policies significantly reduced the wage levels of old enterprises, whereas the impact on the wage levels of new enterprises was not statistically significant. This discrepancy may be attributed to the age effect on enterprise innovation and productivity. New enterprises typically operate with advanced technologies, which may already be near their maximum potential for enhancement. Consequently, the wage levels of old enterprises are more adversely affected by LCCP policies than are those of new enterprises.

6.3. Differences in the Effects of Policies on the Wages of Firms in Different Industries

By utilising the industry classification of enterprises as a basis, firms were categorised into three sectors: primary, secondary, and tertiary. Regression analysis based on Equation (2) was conducted for each sector. The results, presented in

Table 10. Impact of policy on the wages of enterprises with different industries and carbon emission intensities.

Variable	Wage
	Primary Industry	Secondary Industry	Tertiary Industry	High Carbon Enterprises	Low Carbon Enterprises
	(1)	(2)	(3)	(4)	(5)
Treated	−0.3937 ** (0.1895)	−0.0491 *** (0.0187)	−0.0370 (0.0705)	0.0374 (0.1007)	−0.0652 * (0.0340)
Control variable	Yes	Yes	Yes	Yes	Yes
Firm–year fixed effects	Yes	Yes	Yes	Yes	Yes
R2	0.8674	0.7343	0.7614	0.7435	0.7324

Note: *, **, and *** represent the significance levels of 10%, 5%, and 1%, respectively, and the values in brackets are cluster robust standard errors.

, demonstrate that the wage level of enterprises in the tertiary sector was not significantly affected by the LCCP policy; however, the policy notably reduced the wages of enterprises in the primary and secondary sectors. This outcome can be attributed to several factors. First, compared with those in tertiary industries, a larger proportion of primary and secondary sector businesses experience increased production costs due to policy implementation. Second, greater potential exists for technological advancement in the primary and secondary sectors. Consequently, the policy exerts the most substantial effect on the wages of enterprises in the primary and secondary sectors relative to those in the tertiary sector.

6.4. Influence of Policies on the Wages of Different Carbon Emissions Enterprises

Based on variations in carbon emissions across enterprises, firms were classified into high-carbon enterprises (those operating in eight industries with high energy consumption) and low-carbon enterprises (those outside these industries). Regression analysis using Equation (2) was separately conducted for both categories, with the results detailed in Table 10, specifically columns (4) and (5) for high-carbon and low-carbon enterprises, respectively. The findings reveal a statistically significant negative effect of LCCP policies on the wages of low-carbon enterprises, whereas the impact on high-carbon enterprises’ wages is not significant. This disparity likely stems from the differing factor dependencies of enterprises. According to Yamazaki [32], many high-carbon enterprises are characterised by high capital intensity, which mitigates the policy’s impact on their wage levels compared to low-carbon enterprises.

6.5. Differences in the Effects of LCCP Policies on the Internal Unfairness of Employee Remuneration

The influence of the LCCP policy on intrafirm wage inequality exhibits statistically significant negative effects solely among old enterprises, secondary sector enterprises, and low-carbon enterprises, as evidenced by the regression outcomes in

Table 11. Effects of pilot policies on internal unfairness of employees’ remuneration based on company age and carbon emissions intensities.

Variable	Gap_In
	Old Enterprises	New Enterprises	Secondary Industry	High Carbon Enterprises	Low Carbon Enterprises
	(1)	(2)	(3)	(4)	(5)
Treated	−0.0482 *** (0.0182)	−0.0961 (0.0.0918)	−0.0491 ** (0.0194)	−0.0192 (0.0523)	−0.0555 *** (0.0192)
Control variable	Yes	Yes	Yes	Yes	Yes
Firm–year fixed effects	Yes	Yes	Yes	Yes	Yes
R2	0.7506	0.7205	0.7549	0.7815	0.7497

Note: ** and *** represent the significance levels of 5% and 1%, respectively, and the values in brackets are cluster robust standard errors.

. Established enterprises typically possess larger scales and more centralised executive authority, thereby enhancing the policy’s efficacy in promoting internal wage equality. Comparatively, secondary sector enterprises experience heightened policy impacts when contrasted with those of the primary and tertiary industries. Furthermore, senior management within high-carbon enterprises may leverage their authority to secure greater compensation, exacerbating internal wage disparities. Consequently, the policy has a more pronounced inhibitory influence on intrafirm wage inequality within these specified enterprise types.

6.6. Evolution of Effects of LCCP Policies

The implementation of policies constitutes a systematic, intricate, and dynamic process. Exploring the trend in policy effects is an important means of gaining policy experience. We investigated the impact of three waves of LCCP policies on enterprise wages and conducted an analysis of their societal implications, as detailed in

Table 12. Evolution of effects of LCCP policies.

Variable	Wage	Labour	Reward	Gap_In	Power
	(1)	(2)	(3)	(4)	(5)
Treated1	0.0275 (0.0722)	0.0282 (0.0414)	−0.0828 (0.0549)	−0.0392 (0.0322)	−0.0008 (0.0336)
Treated2	−0.1175 *** (0.0435)	0.1100 *** (0.0272)	−0.1281 *** (0.0353)	−0.0456 ** (0.0215)	−0.0567 ** (0.0225)
Treated3	−0.0549 * (0.0324)	0.0648 *** (0.0193)	−0.0701 *** (0.0270)	−0.0500 *** (0.0179)	−0.0362 ** (0.0185)
Control variable	Yes	Yes	Yes	Yes	Yes
Firm–year fixed effects	Yes	Yes	Yes	Yes	Yes

Note: *, **, and *** represent the significance levels of 10%, 5%, and 1%, respectively, and the values in brackets are cluster robust standard errors.

. In column (1), we present regression findings on the wage levels of enterprises under the influence of these policies. The detrimental effects of these policies on enterprise wages diminish as policy implementation progresses. Column (2) displays regression outcomes concerning employment, revealing a declining positive effect of policy implementation on job creation. Column (3) illustrates the regression results on productivity returns, indicating a decreasing negative impact of policy implementation on productivity. This trend arises from enterprises benefitting from technological innovations fostered by policy implementation, gradually translating into improved wage levels. As evidenced in column (4), the policies increasingly suppress intrafirm wage inequality, whereas column (5) suggests a waning negative impact on executive power. This signifies that the LCCP policy promotes the convergence of executive authority to a more equitable level, thereby advancing internal wage fairness.

7. Conclusions and Policy Implications

As the importance of ESG concepts for corporate sustainability is gaining popularity, corporations have to balance environmental and social responsibilities in governance. However, in the face of the “dual-carbon” and “common prosperity” goals put forward by the government, Chinese enterprises may face a dilemma at this stage, that is, it is difficult to strike a balance between low-carbon transformation and higher wages. In order to verify this, we comprehensively analyse the impact of the low-carbon city pilot (LCCP) policy on employees’ wages by using the DID method with the data of A-share listed companies in Shanghai and Shenzhen from 2007 to 2022. This paper uses the LCCP policy implemented in China to denote the low-carbon transition. The underlying assumption is that pilot cities are considered to be ahead of non-pilot cities in the low-carbon transition. Our findings are outlined as follows.

(1)

LCCP policies exhibit a notable suppressive effect on employee wages within businesses. The DID estimate value was −0.0549, indicating statistical significance at the 10% level. Moreover, these policies foster internal fairness in employee compensation, albeit without a significant impact on external fairness.

(2)

The influence of LCCP policies on enterprise wage levels operates through two mechanisms: labour demand and productivity returns. These policies mitigate internal wage disparities by curbing executive authority.

(3)

LCCP policies exhibit significant variations in their impact on wage dynamics across different enterprises. They notably depress wage levels in state-owned enterprises, old enterprises, primary-sector firms, secondary-sector firms, and low-carbon industry enterprises. However, the significant inhibition of internal inequality is observed only in old enterprises, secondary-sector firms, and low-carbon enterprises.

(4)

As the implementation of LCCP policies intensifies, adverse effects on enterprise wage levels are expected to diminish gradually, whereas positive impacts on the internal fairness of employee compensation are expected to increase.

The above results confirm that at this stage, when enterprises in China pursue the environmental protection goal of low-carbon transformation, it will affect their corporate efficiency, which in turn leads to lower wages. Moreover, this problem is difficult to solve by the enterprises themselves alone and requires the help of the Government.

Based on these findings, our policy recommendations are formulated as follows.

The implementation experience of LCCP policies warrants consolidation, advocating the nationwide promotion of low-carbon urban development. Through an examination of the societal ramifications of LCCP policies, a comprehensive evaluation of their impacts in China is attained. The results indicate that although LCCP policies stimulate innovation, their efficacy in enhancing enterprise productivity returns remains suboptimal. Consequently, these policies exert adverse effects on enterprise wage levels in the short term and yield negligible improvements in external fairness. To effectively extend China’s low-carbon urban initiatives nationwide, it is imperative to foster innovation through collaborative efforts spanning the industry, education, and research sectors. This approach expedites the realisation of technological innovations in productivity, thereby augmenting external fairness more promptly and effectively.

The harmonisation of emissions reduction efforts with enterprise wage levels warrants prioritisation. Although low-carbon transformations have significant environmental benefits, careful consideration of their short-term impact on employee remuneration is essential. Against the backdrop of the current postpandemic era, a pressing need exists to underscore the synchronised advancement of low-carbon transformations alongside enhancements in wage levels to safeguard societal welfare. The escalation in investments in scientific and technological innovations has intensified pressures on enterprises. Hence, governmental interventions such as emissions reductions subsidies and innovation grants for enterprises are imperative. Concurrently, efforts should be directed toward developing carbon sinks and implementing stringent carbon emissions controls. Enterprises should be afforded a degree of policy flexibility to facilitate a smoother transition toward low-carbon practices.

Precise low-carbon transformation policies are needed to improve policy effectiveness. The influence of LCCP policies on enterprise wages is characterised by enterprise heterogeneity and industry specificity. Therefore, general policies should be avoided. Rather, policy implementations should be refined for each industry, accounting for their unique characteristics. Furthermore, classifying and promoting the upgrading, transformation, and elimination of various industries is important to enable industries to develop appropriately. When promoting low-carbon city pilot policies across the country, the focus should be on stimulating the vitality of enterprises, stabilising employment, ensuring the living standards of residents, and forming good ecological, economic, and social models.

Of course, our study also suffers from some limitations. The sample of A-share listed companies in Shanghai and Shenzhen selected for this paper at the firm level is relatively representative, but this sample may not provide a complete picture of the response of all firms to the policy. In particular, relative to listed firms, unlisted firms may be more sensitive to policy shocks due to their weaker risk resistance. Unfortunately, due to the unavailability of data on unlisted firms, these firms are not included in the sample of this study. Therefore, future research should aim to collect more comprehensive data for a more detailed analysis.