Improving Chinese Transnational Enterprises’ Environmental Performance: Does Government’s Green Finance Support Matter?

Abstract

With the goal of promoting Chinese enterprises’ green development in overseas markets, China issued the Guidance on Promoting Green Belt and Road in 2017. The Guidance is the world’s first government document specifically designed to provide green finance support for domestic enterprises in overseas markets. Based on data on heavily polluting enterprises listed on the Shanghai and Shenzhen A-share markets from 2008 to 2021, this paper applies a Difference-in-Differences (DID) approach to explore the impact of this document on Chinese transnational enterprises’ environmental performance. Our empirical results reveal that government’s green finance support significantly improves Chinese transnational enterprises’ environmental performance by financing their overseas projects. Such effects are heterogeneous in enterprises with different natures, management levels, and profitability. From mechanism checks, we discovered that the Guidance impacts Chinese transnational enterprises’ green behaviors through both financial and institutional channels. Our study proposes corresponding policy suggestions and offers government policymakers insights into designing effective financial support for its enterprises’ green development in the global market.

1. Introduction

China’s overseas investment has surged since the country’s entry into the World Trade Organization in 2001. The Belt and Road Initiative (BRI), introduced in 2013, further accelerated the international expansion of Chinese enterprises. By 2016, China’s Outward Foreign Direct Investment (OFDI) exceeded its Foreign Direct Investment (FDI) for the first time, making China a net exporter of capital. By the end of 2021, China’s cumulative net stock of OFDI reached 2.78 trillion USD, ranking third globally. Chinese enterprises have evolved their investment methods from direct factory establishments to resource investments, mergers and acquisitions, and strategic alliances. Heavily polluting industries, such as manufacturing, mining, and construction, have become key areas of non-financial outward investment due to domestic overcapacity and slowing growth rates. Numerous Chinese manufacturing enterprises have established their subsidiaries in most countries around the world (Figure 1). The total overseas operating revenue of the top 500 Chinese manufacturing enterprises in 2023 reached 7.2 trillion yuan, an increase of 19.2% over the previous year.

Environmental challenges are significant for Chinese transnational enterprises, particularly in manufacturing industries, when operating overseas. During the 1980s and 1990s, China had lax environmental regulations, prioritizing economic growth and employment over environmental concerns. This led to severe domestic pollution. However, in the 2010s, China began tightening its environmental standards, raising pollution control costs and causing some enterprises to close or transform. Consequently, some polluting enterprises moved their operations abroad. Despite contributing to economic growth in host countries through infrastructure and energy projects, these enterprises often faced complaints and protests due to environmental pollution. The environmental abatement cost during project implementation is high and enterprises choose to ignore the problems caused by pollution. Therefore, Chinese transnational enterprises’ environmental performance does not always receive recognition from local governments, communities, and media.

The negative environmental image of Chinese transnational enterprises has hindered the BRI. Some projects have been postponed or canceled over environmental concerns. To improve corporate environmental performance, enterprises need to increase their environmental expenditure. The rising costs may weaken their competitiveness in overseas markets and reduce their profits. In response, the Chinese government launched the Guidance on Promoting Green Belt and Road (the Guidance) in 2017 (https://www.yidaiyilu.gov.cn/wcm.files/upload/CMSydylgw/201705/201705080205025.pdf, accessed on 19 January 2025), advocating for green and environmentally friendly initiatives. In the Guidance, the Chinese government has issued green finance support specifically for Chinese enterprises’ overseas projects, aiming to help these enterprises improve their environmental performance.

Our study uses the Guidance as a case study to evaluate whether it has enhanced the environmental performance of Chinese transnational enterprises. The paper is structured as follows: Section 2 reviews the literature on green finance and environmental performance. Section 3 outlines the background of the Guidance. Section 4 describes the data, identification strategy, and empirical model. Section 5 discusses the impact of the Guidance on the environmental performance of Chinese transnational enterprises and presents robustness tests, mechanism analysis, and heterogeneity effects. Section 6 concludes the study.

2. Literature Review

Current literature suggests a significant relationship between green finance and environmental improvement. National studies indicate that credit market reforms can significantly reduce air pollution emissions in some countries [1]. Additionally, there is a negative correlation between levels of green finance and carbon dioxide emissions in the United States, Sweden, the United Kingdom, and Switzerland [2]. Using data from 40 European countries, Afzal et al. found that the expansion of green credit effectively reduced carbon dioxide emissions, greenhouse gas emissions, and natural resource consumption [3]. At the bank level, green loans have been shown to improve banks’ reputation and financial performance and reduce issues with non-performing loans [4,5]. The ability of banks to gather information and supervise applicants is crucial for the success of green finance policies [6]. Corporate-level evidence suggests that companies with adequate funds perform better in implementing environmental responsibilities. Amore and Bendensen found that companies receiving external green financing exhibit strong environmental performance [7].

In recent years, there has been a significant increase in the literature on the impact of green finance on the environmental performance of Chinese enterprises. Green credit plays a positive role in guiding enterprises toward green development [8,9,10,11,12,13], promoting sustainable productivity [14,15,16] and green innovation [17]. Some studies suggest that Chinese government policies related to green finance significantly reduce energy intensity [14], improve renewable energy utilization efficiency [18], and promote low-carbon development [8]. Muganyi et al. analyzed data from 290 Chinese cities between 2011 and 2018 and found that improvements in green finance significantly reduced sulfur dioxide emissions, industrial gas and smoke emissions, and total sulfur dioxide production [19]. However, some scholars maintain a cautious perspective. Cai analyzed data from the papermaking, mining, and electricity industries and concluded that green credit policies had a restraining effect on environmental performance [20]. Wei et al. indicated that green credit might not improve the financial performance and operational efficiency of enterprises in the energy conservation and environmental protection industry [21]. Zhang et al. found that the green credit policy did not fully achieve its intended impact on improving corporate environmental performance [22].

The mechanisms through which green finance affects corporate environmental performance have garnered significant scholarly attention. The mediating roles of corporate financing, debt, social responsibility, and green innovation have been explored. Green finance policies redistribute financial resources between traditional and environmentally friendly industries. With green finance policies, commercial banks consider corporate green behavior as a crucial factor when issuing loans, rather than focusing solely on financial solvency [23]. These policies have a signaling effect, influencing investors’ decisions [24,25,26], and can provide extensive financing for enterprises in the environmental or energy-saving industries [27,28]. Financial institutions incentivize enterprises to invest in clean industries and strengthen environmental protection through specific loan terms and repayment methods of green loans [29,30]. In terms of corporate responsibility, green finance requirements force companies to pay attention to environmental issues [28]. Based on data from Chinese listed companies, Si and Cao concluded that green finance enhances corporate environmental and social responsibility [31]. Lu et al. explored how governmental regulation shaped financial institutions’ greenwashing behaviors in green finance [32]. Das et al. discussed the impact of greenwashing on sustainable environmental technologies and green financing for promoting environmental sustainability [33]. Furthermore, green finance can provide low-cost funding, stimulating enterprises to develop and adopt environmentally friendly technologies and enabling long-term research and development. Sun et al. used the PSM-DID model and found that green finance support promotes green innovation in enterprises [34], which has been confirmed by other scholars [35,36]. The interaction between green finance and technological advancements efficiently improved the green transition of Chinese enterprises [37,38].

For heavily polluting enterprises, green finance primarily exerts its impact by affecting financing constraints [39]. Banks consider environmental and sustainability factors when reviewing credit applications. Poor environmental performance can lead to higher financing costs. Consequently, green finance significantly inhibits the investment and financing behavior of heavily polluting enterprises [23,36]. Green loans may encourage enterprises to improve their environmental performance by reducing their debt scale, increasing borrowing costs, and shortening loan terms [12,29].

However, so far, there has been no literature on whether and how a government’s green finance policy can effectively improve its enterprises’ environmental performance in overseas markets. Considering the research gap, this study aims to explore the impacts of the Chinese government’s green finance support on Chinese transnational enterprises’ environmental performance. The contributions of this study are multiple. First, it explores whether a positive correlation between the government’s green finance policy and transnational enterprises’ environmental performance exists. Second, it provides empirical evidence on the mediating effect of financing constraints and debt structure and thereby validates the government’s green finance policy’s influence path on corporate environmental performance. Furthermore, we consider China’s unique management system for state-owned enterprise leaders as an important mechanism through which the Chinese government impacts corporate environment performance. Finally, based on empirical results, our study offers policymakers insights into designing effective green finance policies for improving enterprises’ environmental performance in the global market.

3. Background

In 2013, the Belt and Road Initiative (BRI) proposed by the Chinese government focused on large-scale infrastructure construction projects, which significantly impacted the environments of various countries. While these infrastructure projects undertaken by Chinese enterprises could boost the economic growth of host countries, they might negatively affect the local environment. Due to insufficient environmental awareness and a lack of understanding of environmental protection standards and legislation in host countries, some Chinese enterprises failed to conduct professional environmental risk assessments in the early stages of project investment. This oversight led to local opposition, obstruction, and even prosecution by international environmental organizations, resulting in risks of project shutdown or cancellation (

Table 1. Disputes arising from environmental issues in overseas projects of Chinese enterprises between 2014 and 2017.

Project	Time	Country	Reasons for Disputes
Nicaraguan Grand Canal Project	2014	Nicaragua	Impact on land and water resources
Bambas Copper Mine Project	2015	Peru	Polluted water source
China-Pakistan Economic Corridor	2015	Pakistan	Environmental protection issues and project site selection
Kashmir Hydropower Project	2015	Pakistan	Ecological crisis caused by hydropower station construction
Tar Desert Coal Fired Power Plant Project	2015	Pakistan	Disputes caused by environmental pollution and water resources
Lahore Urban Light Rail Project	2016	Pakistan	Environmental disputes caused by land acquisition and demolition
Colombo Port City Project	2015	Sri Lanka	Impact on the coastal environment
Nagonha mining project	2015	Mozambique	Blocking of the connecting channel between the atoll lake and the ocean
Gandamara Power Plant Project	2016	Bangladesh	Environmental pollution and land acquisition
Suamahamani Gold Mine Project	2016	Madagascar	Pollution of farmland
Nairobi National Park Elevated Railway Project	2016	Kenya	Wildlife habitats disturbed
Tagaung Taung Nickel Resource Project	2016	Myanmar	Solid waste pollution
Letpadaung Copper Mine Project	2017	Myanmar	Polluted water sources

).

In 2011, the Myitsone Hydropower Station project in Myanmar, undertaken by a Chinese company, was suspended by the local government. The suspension was due to the undisclosed negative environmental impacts in the project’s environmental impact assessment report. In 2018, equipment belonging to a Chinese gold smelting enterprise in Kyrgyzstan was destroyed by over a thousand residents, who believed the factory was harming the local environment. In 2019, the Ram Coal Fired Power Plant project in Kenya, invested in by the Industrial and Commercial Bank of China, faced criticism from the local population for insufficient public participation in environmental impact assessments. Consequently, the Kenyan court revoked the environmental impact assessment permit, leading to the project’s shelving. Since 2014, over half of the coal-fired power plant projects that China planned to support with $160 billion under the Belt and Road Initiative have been canceled or put on hold due to environmental protection considerations by host countries.

To achieve sustainable development in the global market, Chinese transnational enterprises are trying to improve their environmental performance in overseas projects by continuously increasing their green behavior. According to a report by the United Nations Development Programme in 2015 (Figure 2), the top three green behaviors favored by Chinese companies in overseas markets are: reducing pollution emissions by purchasing and installing pollution control equipment, investing in environmentally friendly products to reduce environmental impact, and regularly evaluating the environmental impact of production and operation. These green behaviors are costly, leading to enterprises actively seeking green financing support from the government.

Although green finance started relatively late in China, it has developed rapidly. In 2012, the China Banking Regulatory Commission issued the Green Credit Guidelines, which aimed to direct credit resources toward environment-friendly industries and the green economy, curbing the expansion of heavily polluting industries. The implementation of the Green Credit Guidelines has improved the innovation output of heavily polluting enterprises in the domestic market, enabling these enterprises to promote industrial structure upgrading by enhancing their innovation capabilities [40,41]. According to data from the People’s Bank of China, by the end of 2023, the balance of domestic and foreign currency green loans in China had increased to 30.08 trillion yuan.

With the rising pollution issues of Chinese enterprises in overseas markets and the initial success of green finance policies in the domestic market, the Chinese government launched the Guidance on Promoting Green Belt and Road in 2017. This document implemented green finance policies specifically for the overseas projects of Chinese enterprises for the first time. Measures to provide green funding support for these projects include: (1) Encouraging qualified Belt and Road green projects to apply for support from existing funding channels like the Nnational Green Development Fund and Public-Private-Partnership (PPP) financing support fund. (2) Utilizing existing financial institutions like the China Development Bank and the Export-Import Bank of China to form a multi-channel investment system and long-term mechanism for integrating funds from central and local governments and society. (3) Leveraging the unique advantages of policy-based financial institutions to guide and channel funds from various parties to support the green Belt and Road Initiative. (4) Providing positive support for Belt and Road green projects through existing international multilateral and bilateral cooperative institutions and funds, such as the Silk Road Fund, South-South Cooperation Assistance Fund, China-ASEAN Investment Cooperation Fund, China-Central and Eastern Europe Investment Cooperation Fund, China-ASEAN Maritime Cooperation Fund, Special Fund for Asian Regional Cooperation, and Lancang-Mekong Cooperation Special Fund.

Since the implementation of the Guidance, Chinese transnational enterprises, especially heavily polluting enterprises, have received substantial financial support to improve their environmental performance. From 2017 to 2020, the Export-Import Bank of China and the National Development Bank provided over $15 billion in financing support to Belt and Road renewable energy projects. This includes the photovoltaic power plant project in Hohuy Province, Argentina, financed by the Export-Import Bank of China, and the Karot Hydropower Project in Pakistan, supported by the National Development Bank. The Silk Road Fund financed Chinese enterprises’ projects in over 70 countries and regions, with its investment amount exceeding $23.5 billion by the end of 2023.

In the following section, we establish an econometric model to empirically test whether green financial support specified in the Guidance effectively improved Chinese transnational enterprises’ environmental performance. In addition, we conduct robustness tests, heterogeneity analysis, and mechanism checks to strengthen the baseline results and thereby offer a valuable and complete framework.

4. Econometric Specification

4.1. Data and Variables

Our study focused on heavily polluting enterprises listed on the Shanghai and Shenzhen A-share markets from 2008 to 2021. According to the China Securities Regulatory Commission Listed Company Industry Classification Guidelines (2012) (http://www.csrc.gov.cn/csrc/c100103/c1452025/content.shtml, accessed on 19 January 2025), heavily polluting enterprises are those involved in sectors such as coal mining and washing, oil and gas mining, ferrous and non-ferrous metal mining and dressing, textile industry, leather, fur, feathers and products, shoe manufacturing, paper and paper products, petroleum processing, coking and nuclear fuel processing, chemical raw materials and products manufacturing, chemical fiber manufacturing, rubber and plastic products, non-metallic mineral products, ferrous and non-ferrous metal smelting and rolling processing, electricity and heat production and supply, thermal power, steel, cement, electrolytic aluminum, coal, metallurgy, building materials, mining, chemical, petrochemical, pharmaceutical, paper, fermentation, sugar, vegetable oil processing, brewing, and textile and leather industries.

The sample was processed as follows: (1) Exclude ST (Special Treatment) companies during the sample period; (2) Exclude enterprises with missing main variables; (3) Exclude insolvent enterprises. After applying these criteria, the final sample consisted of 132 eligible listed heavily polluting enterprises. Of these, 52 are transnational enterprises that have conducted business in overseas markets, while the remaining 80 enterprises have limited their operations to the domestic market in China.

The financial data for the enterprises in our sample were obtained from the China Stock Market & Accounting Research (CSMAR) database and corporate financial reports. ESG rating data were sourced from the Bloomberg database, which evaluates companies based on the quantity of relevant data disclosed in their public reports. Companies that excel in environmental protection, social responsibility, and governance receive higher ESG scores. The definitions and descriptive statistics of the variables used in this study are presented in

Table 2. Variable Description.

Variable Type	Variable Symbol	Variable Definition	N	Mean	SD
Dependent variables	ENV	Environment Score	1848	9.836	13.651
	SOC	Society Score	1848	12.701	8.032
	GOV	Governance Score	1848	60.645	14.852
	ESG	ESG Score	1848	27.75	10.377
Independent variable	Treated	Overseas presence dummy variable	1848	0.401	0.49
Control variables	SIZE	Enterprise Size	1848	23.262	1.281
	DER	Debt-to-Equity Ratio	1848	0.473	0.189
	PFA	Proportion of Fixed Assets	1848	0.354	0.177
	RGR	Revenue Growth Rate	1848	0.146	0.307
	TQR	Tobin’s Q Ratio	1848	1.89	1.382
	PID	Proportion of Independent Directors	1848	37.171	5.496
	SC	Shareholder Concentration	1848	40.241	16.231
	SRD	Shareholding Ratio Difference	1848	0.295	0.297
	ON	Ownership Nature	1848	0.693	0.461
	TPSR	Two Powers Separation Ratio	1848	7.173	8.766
	OBE	Overseas Background of Executives	1848	0.548	0.498
	ABF	Audit by Big Four	1848	0.137	0.344
	NPMTA	Net Profit Margin on Total Assets	1848	0.059	0.065
	ROE	Return on Equity	1848	0.102	0.118
	NPMS	Net Profit Margin on Sales	1848	0.09	0.128
Mediating variables	KZ	KZ Index	1848	1.273	1.968
	WW	WW Index	1848	−1.069	0.067
	FC	FC Index	1848	0.253	0.223
	CR	Current Ratio	1848	1.684	1.963
	QR	Quick Ratio	1848	1.249	1.682
	CFR	Cash Flow Ratio	1848	0.084	0.068

.

4.2. Model

In this study, the Difference-in-Differences (DID) model was applied to assess how the Guidance influenced corporate environmental performance. The difference-in-differences approach is a statistical technique used in econometrics and quantitative research in the social sciences that attempts to mimic an experimental research design using observational study data, by studying the differential effect of a treatment on a ’treatment group’ versus a ’control group’ in a natural experiment. It calculates the effect of a treatment (i.e., an explanatory variable or an independent variable) on an outcome (i.e., a response variable or dependent variable) by comparing the average change over time in the outcome variable for the treatment group to the average change over time for the control group. Our DID model leverages the exogenous changes introduced by the Guidance and compares the before-and-after effects on the affected enterprises (the treatment group) with those of firms that were not affected by this change (the control group). The baseline model is structured as follows:

$${ENV}_{it}=\alpha +\beta { {Treated}_i\times Policy}_t+{\eta }^{\prime }{X}_{it}+{\mu }_i+{\tau }_t+{\epsilon }_{it}$$

(1)

where $ENV$ represents the environmental performance of an enterprise. $Treated$ is a dummy variable indicating whether an enterprise conducts business in overseas markets (1 if true, 0 otherwise). $Policy$ is a time dummy variable representing the implementation of the Guidance, set to 1 for the years 2017 and after, and 0 otherwise. $X$ denotes a vector of control variables, which includes SIZE (the measure of a company’s scale or total assets), DER (a financial ratio indicating the relative proportion of shareholders’ equity and debt used to finance a company’s assets), PFA (the proportion of fixed assets), RGR (the percentage increase in an enterprise’s total revenue over a specific period), TQR (a ratio that compares an enterprise’s market value to the replacement cost of its assets), PID (the percentage of directors on a company’s board who are not affiliated with the company), SC (the concentration of ownership measured by the number of shares held by the largest shareholder relative to the total shares), and SRD (the ratio of the second-largest shareholder’s ownership to the largest shareholder’s ownership). Subscripts $i$ and $t$ represent the enterprise and year, respectively, with ${\mu }_i$ and ${\tau }_t$ indicating enterprise fixed effects and year fixed effects.

5. Empirical Results and Discussion

5.1. Baseline Results

Table 3. Baseline results.

	(1)	(2)	(3)	(4)	(5)	(6)
	ENV	ENV	ENV	ENV	ENV	ENV
DID	20.32 ***	11.72 ***	13.48 ***	9.05 ***	9.52 ***	9.31 ***
	(0.90)	(1.26)	(0.93)	(1.15)	(0.95)	(0.95)
Controls	No	No	Yes	Yes	No	Yes
Firm Effect	Yes	No	Yes	No	Yes	Yes
Year Effect	No	Yes	No	Yes	Yes	Yes
N	1848	1848	1848	1848	1848	1848
F	510.25	85.94	96.09	41.38	99.87	14.44
r2_a	0.48	0.38	0.59	0.46	0.65	0.66

Standard errors in parentheses: *** p < 0.01.

presents the estimation results of Equation (1). Columns (1) and (2) present the results without controlling any of the control variables. In Columns (3) and (4), we add control variables and control the firm-fixed effect and time-fixed effect, respectively. In Columns (5) and (6), to capture the different effects of the Guidance within enterprises across time, we control for both the individual fixed effect and time fixed effect, and in Column (6) we add control variables. In all the columns in

Table 4. Testing the Common Trend Assumption with Equation (2).

	ENV		ENV
$\left({Treated}_i\times D_t^{-4}\right)$	−$0.91$	$\left({Treated}_i\times D_t^1\right)$	4.56 ***
	(1.15)		(1.96)
$\left({Treated}_i\times D_t^{-3}\right)$	−2.22	$\left({Treated}_i\times D_t^2\right)$	4.10 ***
	(1.71)		(2.02)
$\left({Treated}_i\times D_t^{-2}\right)$	−2.13	$\left({Treated}_i\times D_t^3\right)$	9.01 ***
	(1.74)		(1.64)
$\left({Treated}_i\times D_t^{-1}\right)$	0.17	$\left({Treated}_i\times D_t^4\right)$	19.36 ***
	(1.71)		(2.21)
Controls	Yes		Yes
Firm Effect	Yes		Yes
Year Effect	Yes		Yes
N	528		528

Standard errors in parentheses, *** p < 0.01.

, our estimations of $\beta$ are significantly positive. While both firm fixed effect and time fixed effect are controlled, estimators of the DID coefficient in Columns (5) and (6) are 9.52 and 9.31, respectively, and are significant at the 1% level. Results in Table 3 suggest that the implementation of the Guidance significantly raised the environment scores of Chinese transnational enterprises in the sample. Based on this baseline result, we believe that the Chinese government’s green finance support is efficient in improving Chinese transnational enterprises’ environmental performance.

5.2. Testing Common Trends for DID Specification

The DID analysis conducted above relies on the assumption that Chinese enterprises without overseas presence (control group) serve as appropriate counterfactuals for predicting the environmental performance of Chinese translational enterprises (treatment group) in the absence of the Guidance. To validate this assumption, it is crucial that a common trend exists between the treated and control groups prior to the implementation of the treatment. We rigorously test this common trend assumption using a formal event study method with 4-year leads and 4-year lags [42], as outlined below:

$${ENV}_{it}=\alpha +{\sum }_{j=-4, j \ne 0}^4{\delta }_j\left({Treated}_i\times D_t^j\right)+{\mu }_i+{\tau }_t+{\eta }^{\prime }{X}_{it}+{\epsilon }_{it}$$

(2)

where $D_t^j$ is an indicator variable equal to one for the $jth$ year before (j < 0) or after (j > 0) the Guidance, and otherwise zero. At the endpoints, $D_t^{-4}$equals one for all years that are 4 or more years before the Guidance, while $D_t^4$ equals one for all years that are 4 or more years after the Guidance. For the purpose of visual comparison among estimated coefficients, we normalize to zero. The remaining specification is identical to the DID model in Equation (1). The test results presented in Table 4 generally support the assumption of common trends.

The results of all four estimated coefficients of $\left({Treated}_i\times D_t^j\right)$ (where j < 0) are not statistically different from zero, and all four estimated coefficients of $\left({Treated}_i\times D_t^j\right)$ (where j > 0) are far from the periods when the regulations were announced. Figure 3 plots the estimated coefficients of $\left({Treated}_i\times D_t^j\right)$for Equation (2), along with 95% confidence intervals. The Guidance has a swift impact on Chinese transnational enterprises, indicating that they promptly improve their environmental performance following the implementation of the Guidance.

5.3. Robustness Test

We conduct the robustness tests by replacing the dependent variable and shortening the sample period. First, as the scores of ESG, SOC, and GOV are highly correlated with the scores of ENV, we replace the dependent variable ENV with ESG, SOC, and GOV, respectively. The results are shown in

Table 5. Results for robustness check (replacing the dependent variable).

	(1)	(2)	(3)	(4)	(5)	(6)
	ESG	ESG	SOC	SOC	GOV	GOV
DID	4.14 ***	3.97 ***	3.05 ***	2.92 ***	3.13 ***	3.31 ***
	(0.47)	(0.47)	(0.50)	(0.50)	(0.56)	(0.56)
Controls	No	Yes	No	Yes	No	Yes
Firm Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
N	1848	1848	1848	1848	1848	1848
F	76.66	12.66	37.07	6.23	0.06	1.69
r2_a	0.83	0.84	0.69	0.69	0.82	0.82

Standard errors in parentheses: *** p < 0.01.

, which remain substantially the same as the basic regression results. Second, from 2020 to 2022, governments worldwide implemented a series of policies to cope with the COVID-19 pandemic, which may have a significant impact on corporate production and environmental performance. Thus, considering that the COVID-19 pandemic may affect the stability of sample data, we shorten the sample period to 2008–2019 and exclude the data after 2019. The test results in

Table 6. Results for robustness check (shortening the sample period).

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	ESG	ESG	SOC	SOC	GOV	GOV	ENV	ENV
DID	3.18 ***	3.12 ***	2.10 ***	1.99 ***	2.73 ***	2.69 ***	7.29 ***	7.44 ***
	(0.43)	(0.43)	(0.46)	(0.47)	(0.55)	(0.56)	(0.92)	(0.92)
Controls	No	Yes	No	Yes	No	Yes	No	Yes
Firm Effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	1584	1584	1584	1584	1584	1584	1584	1584
F	54.89	9.41	20.69	3.96	0.29	1.50	62.88	8.61
r2_a	0.82	0.82	0.67	0.67	0.79	0.79	0.60	0.60

Standard errors in parentheses: *** p < 0.01.

show that the Treated × Policy coefficients are still significantly positive in all the columns. Therefore, the empirical results in Table 5 and Table 6 strengthen the robustness of the baseline results.

5.4. Heterogeneity Analysis

The regression results and robustness test of the previous section indicate that the Guidance has a significant and positive effect on improving Chinese transnational enterprises’ environmental performance. To further investigate the heterogeneous effects, our study divides the sample based on the ownership, management, and profitability of enterprises.

The anticipated impact of the Guidance on corporate environmental performance may vary depending on the nature of the enterprise. The nature of an enterprise includes whether it is a state-owned enterprise, whether it is in the manufacturing industry, and whether it is a high-tech enterprise. State-owned enterprises and non-state-owned enterprises differ in organizational structure and sources of funds. Compared to non-state enterprises, state-owned enterprises may receive more attention and support from the government and corresponding policies. Manufacturing enterprises as a whole face more difficulties in improving environmental performance than other enterprises, while high-tech enterprises are generally more sensitive to government policies than traditional enterprises. The sample groups corresponding to Columns (1) to (6) in Panel A of

Table 7. Heterogeneity test (enterprise nature).

Variable	Env
	(1)	(2)	(3)	(4)	(5)	(6)
Panel A: Enterprise nature
DID	17.89 ***	6.49 ***	5.51 ***	10.24 ***	16.21 ***	4.68 ***
	(2.08)	(1.07)	(1.67)	(1.22)	(1.45)	(1.23)
Controls	No	Yes	No	Yes	No	Yes
Firm Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
N	504	1344	770	1078	756	1092
F	11.09	6.87	4.55	10.89	15.49	4.42
r2_a	0.68	0.66	0.67	0.65	0.65	0.67
Panel B: Enterprise management
DID	10.22 ***	8.17 ***	10.24 ***	5.51 ***	16.21 ***	4.68 ***
	(1.49)	(1.45)	(1.67)	(1.22)	(1.45)	(1.23)
Controls	Yes	Yes	Yes	Yes	Yes	Yes
Firm Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
N	854	994	560	1324	756	1092
F	7.64	9.99	4.55	10.89	15.49	4.42
r2_a	0.66	0.70	0.67	0.65	0.65	0.67
Panel C: Enterprise profitability
DID	7.93 ***	10.22 ***	7.35 ***	10.13 ***	9.67 ***	10.15 ***
	(1.20)	(1.58)	(1.36)	(1.32)	(1.23)	(1.51)
Controls	Yes	Yes	Yes	Yes	Yes	Yes
Firm Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
N	1106	742	910	938	1064	784
F	9.66	10.88	9.39	9.30	11.20	6.99
r2_a	0.64	0.70	0.65	0.67	0.66	0.66

Standard errors in parentheses: *** p < 0.01.

are state-owned enterprises, non-state-owned enterprises, manufacturing enterprises, non-manufacturing enterprises, high-tech enterprises, and non-high-tech enterprises. The estimators of the Treated × Policy coefficients in all columns are positive and significant, indicating that the Guidance has a positive effect on the environmental performance of enterprises with different natures.

When studying the impact of the Guidance on corporate environmental performance, the heterogeneity of corporate management is an important consideration. Enterprises’ management levels may impact their awareness of environmental protection, maintenance of corporate reputation, and compliance with laws, and these differences could affect policy effectiveness. Therefore, we focus on the heterogeneity of enterprises’ management, which includes High separation of two rights (whether the separation ratio of the capital ownership and the capital operation rights of an enterprise is higher than the average enterprise in the sample), Overseas background of executives (whether top executives of an enterprise have overseas backgrounds), and Audit by Big Four (whether an enterprise has been audited by one of the big four accounting firms). The sample groups corresponding to Columns (1) to (6) in Panel B of Table 7 are enterprises whose capital ownership and capital operation rights are highly separated, enterprises whose capital ownership and capital operation rights are not highly separated, enterprises whose top executives have overseas backgrounds, enterprises whose top executives have no overseas backgrounds, enterprises that have been audited by one of the big four accounting firms, and enterprises that have not been audited by one of the big four accounting firms, respectively. The results in all columns are consistent with the baseline result. Additionally, the results demonstrate that the Guidance has a higher impact on enterprises with higher management levels to improve their environmental performance.

We further examine the profitability heterogeneity between Chinese transnational enterprises and domestic enterprises. The method of testing is to group and regress the baseline model according to three profitability classifications, which include Net Profit Margin on Total Assets (the ratio of net profit to total assets, indicating profitability), Return on Equity (a measure of profitability that calculates how much profit an enterprise generates with the money shareholders have invested), and Net Profit Margin on Sales (the ratio of net profit to total sales). The sample groups corresponding to Columns (1) to (6) in Panel C of Table 7 are enterprises without high net profit margins on total assets, enterprises with high net profit margins on total assets, enterprises without high return on equity, enterprises with high return on equity, enterprises without net profit margin on sales, and enterprises with net profit margin on sales. The empirical results indicate that the Guidance is more helpful to enterprises with high profitability (above the mean) than other enterprises in ameliorating their environmental performance.

5.5. Placebo Test

To mitigate the possibility of spurious empirical results in this study, we conducted in-space placebo tests. Specifically, the in-space placebo test is performed by randomly assigning firms as transnational enterprises. Since the pseudo-treatment group was randomly generated, the announcement of the Guidance should not have a significant impact on corporate environmental performance. Therefore, the regression coefficient of the DID interaction term Treated × Policy should be near zero; otherwise, it would indicate a bias in the model specification. Accordingly, we repeated the above random process 100 times for model estimation and plotted the kernel density graph of the estimated coefficients of Treated_i × Policy_t (Figure 4). The study found that the mean of the estimated coefficients under the random process was close to zero. Meanwhile, the regression coefficient of the treatment variable (the value is 9.31, shown in Column 6 of Table 3) fell within the range of low probability events in the kernel density graph of the aforementioned placebo test.

Furthermore, we conducted mixed (time and space) placebo tests. By using four pre-time periods and randomly selected firms for interaction, we generated false policy interaction terms. The regression coefficient of the DID interaction term Treated_i × Policy_t^j should also be near zero. We repeated the above random process 100 times for model estimation and obtained the kernel density graph of the estimated coefficients of Treated_i × Policy_t^j (Figure 5). It was found that the mean of the estimated coefficients under the random process was also close to 0. In other words, the impact of the Guidance on enterprises’ environmental performance is not a random event. The research conclusions of this study are reliable and robust.

5.6. Mechanism Check

(1)

Financial channel

The funding demand for improving environmental performance by Chinese transnational enterprises is enormous. Various uncertainties caused by overseas risks put these enterprises under higher financing constraints than their peers. The Guidance has implemented multiple financing channels for these enterprises, therefore it may improve their environmental performance in overseas markets by alleviating their financing constraints.

We introduce several financial constraint indexes as mediating variables, including the KZ Index [43,44], WW Index [45], and FC Index to measure the financial environment and financing costs. The results in all columns in Panel A of

Table 8. Results from mechanism check (financial channel).

	(1)	(2)	(3)	(4)	(5)	(6)
Panel A: Financial constraint
Variables	KZ	KZ	WW	WW	FC	FC
DID	−0.13 ***	−0.02 **	−0.00 *	−0.00 *	0.02 **	0.02 **
	(0.03)	(0.01)	(0.00)	(0.00)	(0.01)	(0.01)
Controls	No	Yes	No	Yes	No	Yes
Firm Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
N	1848	1848	1848	1848	1848	1848
F	1.01	75.24	1.53	70.44	4.15	83.18
r2_a	0.55	0.70	0.82	0.90	0.79	0.87
Panel B: Debt structure
Variables	CR	CR	QR	QR	CFR	CFR
DID	0.06 ***	−0.01 ***	0.09 *	0.02 **	0.01 **	0.01 **
	(0.02)	(0.00)	(0.05)	(0.01)	(0.01)	(0.01)
Controls	No	Yes	No	Yes	No	Yes
Firm Effect	Yes	Yes	Yes	Yes	Yes	Yes
Year Effect	Yes	Yes	Yes	Yes	Yes	Yes
N	1848	1848	1848	1848	1848	1848
F	0.37	44.54	0.98	43.05	5.57	17.09

Standard errors in parentheses: * p < 0.1, ** p < 0.05, *** p < 0.01.

show that the Guidance significantly decreases the financial constraints. This suggests that Chinese transnational enterprises could have more funds to improve their environmental performance in overseas markets.

It is difficult for Chinese transnational enterprises with poor debt structures to have the motivation and ability to improve their environmental performance, especially those with huge short-term debt. The green loans provided in the Guidance are mostly long-term debts of 20–30 years, with low interest rates, which allows the borrowing enterprises to have a relatively light debt burden and strong monetization ability. Thus, the Guidance may improve the environmental performance of enterprises in the treated group by optimizing their debt structure.

We introduce several debt structure indexes as mediating variables, including Current Ratio (a liquidity ratio that measures a company’s ability to cover its short-term obligations with its short-term assets), Quick Ratio (a liquidity ratio that measures a company’s ability to cover its short-term obligations with its most liquid assets), and Cash Flow Ratio (a ratio that measures a company’s ability to cover its short-term obligations with its operating cash flow). The results in Panel B of Table 8 show that the Guidance significantly reduces the enterprises’ debt burden and improves solvency. It implies that Chinese transnational enterprises could improve their environmental performance as their debt structures are optimized.

(2)

Institutional channel

China’s unique management system for state-owned enterprise leaders is an important institutional channel through which government policy impacts corporate actions. Most of the providers (Chinese banks and other financial institutions) and beneficiaries (Chinese transnational enterprises) of the green finance policies issued by the Guidance are solely state-owned enterprises, state-holding enterprises (In China, state-holding enterprises refer to enterprises that have a relatively high proportion of national capital and are controlled by the state), and state-owned joint stock enterprises (In state-owned joint stock enterprises in China, the government is only an ordinary shareholder of the enterprise and is regulated by the Company Law). Taking banks as an example. The banks that provide green financial services stipulated in the Guidance are mainly large state-owned banks, including the Industrial and Commercial Bank of China, Agricultural Bank of China, Bank of China, China Construction Bank, Bank of Communications, as well as China Development Bank, and the China Export Import Bank.

In China, the government agency responsible for managing different types of state-owned enterprises is the State-Owned Assets Supervision and Administration Commission (SASAC). It is directly under the State Council, at the ministerial level, representing the state in fulfilling the responsibilities of investors. The management of state-owned enterprises by the SASAC includes the management and supervision of state-owned assets, as well as the management of the responsible persons of state-owned enterprises. The SASAC determines the appointment, dismissal, and salaries of state-owned enterprise leaders through evaluations of their performance (Figure 6). One of the most important performance evaluation criteria is whether the leaders actively respond to national policies (The State-owned Assets Supervision and Administration Commission issued the Interim Measures for the Performance Assessment of Central Enterprise Leaders in 2010 (Chapter 1, Article 5, https://www.gov.cn/gongbao/content/2010/content_1629140.htm, accessed on 19 January 2025) and the Measures for the Performance Assessment of Central Enterprise Leaders in 2019 (Chapter 1, Article 3, http://www.sasac.gov.cn/n2588035/n22302962/n22302967/c22305409/content.html, accessed on 19 January 2025)). The Guidance is an important policy launched by the State Council. Therefore, the efforts of bank executives to launch corresponding green financial services are helpful for improving their performance, while the responsible persons of polluting enterprises also have the motivation to fully profit from these green financial services. For career advancement, leaders of banks and enterprises are highly motivated to implement the Guidance’s financial policies, improving the environmental performance of Chinese transnational enterprises.

Briefly, the impact mechanism of the Guidance on Chinese transnational enterprises is multi-channel (Figure 7). From the perspective of the financing channel, the Guidance supports enterprises with green loans and funds, reducing financing constraints and optimizing debt structures. This enables increased green investment, enhancing environmental performance. From the perspective of institutional channels, the management system for leaders of Chinese state-owned enterprises plays a crucial role. Considering personal career development, leaders of state-owned financial institutions and enterprises have a strong motivation to implement the policies proposed in the Guidance. As a result, financial institutions would actively provide green financial support for Chinese transnational enterprises’ projects in overseas markets, and the corresponding enterprises would strive to improve their environmental performance through this support. It is worth noting that state-owned enterprises currently account for a high proportion of Chinese enterprises in overseas markets, therefore, the impact of the Chinese government’s green finance policy on corporate environmental performance through institutional channels is evident. With the gradual growth of Chinese private enterprises in overseas markets, the Chinese government would mainly rely on financial channels to influence the environmental performance of these enterprises when implementing green finance policies.

6. Conclusions

The Guidance on Promoting Green Belt and Road issued by the Chinese government in 2017 is the world’s first government document specifically designed to provide green finance support for domestic enterprises in overseas markets. Studying the impact of this document on corporate environmental performance is meaningful for governments, transnational enterprises, and countries where projects supported by green finance are located. This paper employs an experimental method to estimate such an impact using matched data from 2008 to 2021. We apply a DID (Difference-in-Differences) method and address the endogeneity problem to derive a reliable causal effect.

Our findings indicate that the Guidance significantly enhances Chinese transnational enterprises’ environmental performance. Further analysis reveals that the impact of the Guidance is heterogeneous among enterprises. Specifically, state-owned enterprises exhibit better environmental performance post-Guidance compared to their non-state-owned counterparts. This result is confirmed by the relevant mechanism check. Additionally, manufacturing enterprises and high-tech enterprises show greater improvements in their environmental performance compared to non-manufacturing and non-high-tech enterprises, respectively. Among all the Chinese transnational enterprises in our sample, those with modern management structures or high profitability could benefit more from the Guidance in terms of improving corporate environmental performance.

Furthermore, we explore the mechanism through which the Guidance impacts Chinese transnational enterprises’ environmental performance. The results from our quantitative and qualitative analysis indicate that the Guidance promotes corporate environmental responsibility through both financial and institutional channels. Financial channel refers to the green financial support provided by the Guidance, including green loans and green funds. Chinese transnational enterprises could obtain funds for their overseas projects through relevant green finance support. Such support alleviates the financing constraints and optimizes the debt structure of enterprises, thereby enabling enterprises to have more funds to increase green expenses and improve their environmental performance. Institutional channel refers to the incentive effect of the Guidance as a government document on the leaders of state-owned financial institutions and enterprises. Under the unique management system for state-owned enterprise leaders in China, implementing the provisions of the Guidance is helpful for the personal career development of leaders of state-owned banks and enterprises. Therefore, leaders of state-owned banks would proactively develop and launch diverse green finance services for Chinese transnational enterprises, while leaders of state-owned enterprises would actively seek relevant green finance support for their overseas projects, thereby improving their environmental performance. Although the results of the implementation of the Guidance in China are encouraging, there exist some potential challenges in replicating this green policy in non-Chinese settings. Firstly, green finance support for enterprises is based on the government’s fiscal revenue, and not all countries have abundant fiscal revenue and foreign exchange reserves like China. Secondly, the Chinese government has strong control over state-owned enterprises and their leaders, which may not apply to every country. Therefore, the discussion on whether such a policy is applicable to other countries needs to take into account the fiscal revenue and state-owned enterprise management system of the corresponding countries.

The existing literature on the interaction between finance and corporate green behavior primarily focuses on a single national market. This paper provides new insights for a government to improve its transnational enterprises’ environmental performance in the global market through green finance policies. The conclusion of our study offers some policy implications for government policymakers. Firstly, transnational enterprises’ environmental performance is crucial for their sustainable development in overseas markets. A government should recognize the importance of green finance support for its transnational enterprises to improve their environmental performance. Therefore, the government could facilitate its financial institutions to provide green financing to the enterprises’ overseas projects. Secondly, the government can encourage financial institutions to make state-owned enterprises, manufacturing enterprises, and high-tech enterprises the focus of providing green loans and green funds, as the environmental performance improvement of these types of enterprises is more significant due to the intervention of green financing. Similarly, the government may make the corporate management system reform and profitability improvement a prerequisite for providing green financial support to corresponding enterprises. Thirdly, for overseas projects of enterprises with high financing constraints and a high proportion of short-term debt, the government should provide more policy support to guide the financial institutions to provide green financing for these projects. By alleviating financing constraints and optimizing debt structure through green financing, these enterprises would improve their environmental performance to a greater extent.

This study has several limitations. First, given data availability, this paper selects the Environmental Score of ESG as the main indicator to measure corporate environmental performance. This is only a single indicator, and we could consider other indicators including pollutant emissions and the number of green patents of enterprises for future studies. Second, in evaluating the effect of the Guidance on corporate environmental performance, more external variables should be included as control variables in the model to reduce the possibility of endogeneity. Finally, for countries like China, where a large number of financial institutions and transnational enterprises are state-owned, the government could actively utilize its influence on the personal career development of bank and corporate leaders to effectively facilitate the connection between providers and users of green finance services and further enhance the impact of green finance on corporate environmental performance. Therefore, a detailed comparison between China and other emerging countries like India and Vietnam should be considered in future studies to verify whether the results of this paper could be applicable to other countries.