Analysis of the Impact of Outward Foreign Direct Investment of Corporations on Green Innovation in the Context of the Belt and Road Initiative

Abstract

Amid the global trend of green transformation, existing studies have explored the impact of outward foreign direct investment (OFDI) on economic performance and technological innovation. However, micro-level empirical analyses on how OFDI facilitates green technological innovation through capital and knowledge channels remain insufficient. Drawing on data from China’s A-share listed companies during 2007–2022, this study systematically investigates, for the first time, the pathways through which OFDI influences green innovation, and identifies the mediating mechanisms of financing constraints and R&D investment. Employing fixed effects and mediation effect models, the empirical results reveal that OFDI significantly promotes firms’ green technological innovation, with stronger effects observed among state-owned enterprises, among non-polluting firms, and in the context of invention patent applications. This study enriches the theoretical framework of green innovation and provides empirical evidence and actionable insights for corporate “going global” strategies and green transition policy making.

1. Introduction

As the global response to climate change and the pursuit of sustainable development goals continue to advance, green innovation has increasingly become a critical pathway for both nations and enterprises to achieve high-quality development. Outward foreign direct investment (OFDI) serves not only as a strategic means for resource allocation and global expansion, but also as a vital mechanism for the diffusion of green technologies and the facilitation of green transformation.

While existing research has explored OFDI primarily from macro-level perspectives—such as its impacts on home-country economic growth and industrial upgrading—there remains a paucity of studies at the micro level, particularly regarding how OFDI affects firms’ green technological innovation through capital and knowledge spillover channels. This gap becomes especially salient in the context of the Belt and Road Initiative (BRI), which has seen an increasing frequency of OFDI activities by Chinese enterprises and significant changes in the patterns of green technology inflow and outflow, offering a novel research opportunity.

This study aims to address this research gap by conducting a firm-level analysis of the mechanisms through which OFDI influences green technological innovation. It introduces financing constraints and R&D investment as mediating variables to explain the internal pathways. In addition, the paper examines the moderating role of firm heterogeneity in shaping the green effects of OFDI, and explicitly proposes three research questions:

(1)

Does OFDI significantly enhance firms’ green innovation performance?

(2)

Are its effects transmitted through the alleviation of financing constraints and increased R&D investment?

(3)

Do these mechanisms differ significantly across firms with varying characteristics (e.g., state owned vs. privately owned, polluting vs. non-polluting enterprises)?

To explore these questions, this study constructs an analytical framework grounded in the resource-based view (RBV) and institutional theory, and it conducts an empirical analysis using panel data from China’s A-share listed firms from 2007 to 2022. By applying fixed effects and mediation models, this research seeks to provide both theoretical insights and empirical evidence to inform the optimization of green BRI policy frameworks and to support enterprises in their global green strategy deployment.

2. Analysis of the Current Status

2.1. Analysis of the Current Status of OFDI Under the Belt and Road Initiative

2.1.1. Scale of OFDI by Enterprises in Belt and Road Co-Building Countries

Over the past decade since the Belt and Road Initiative (BRI) was proposed, Chinese enterprises’ outward foreign direct investment (OFDI) in countries along the Belt and Road has shown continuous growth. According to relevant data, especially in the last five years, OFDI by Chinese enterprises has demonstrated a stable upward trend, with an investment growth rate of 22.6% in 2023, reaching a historic high. However, in 2019, China’s OFDI saw a 3.8% decline, which was closely related to the general decrease in global OFDI flows from 2017 to 2019. Despite this short-term fluctuation, it does not indicate a decline in China’s OFDI but rather reflects the gradual development of a healthier, more stable, and standardized investment model by Chinese enterprises(See Figure 1).

In the past, Chinese enterprises often encountered issues such as blind decision making and poor management in their outward foreign investments due to insufficient experience and lack of market knowledge. These problems led to investments that failed to effectively meet local market demands or comply with local regulations, which, in turn, affected the enterprises’ subsequent operations and even harmed the reputation of Chinese companies in the international market (Wang S et al., 2023) [1].

To address these challenges, starting in late 2016, the national government strengthened the approval and compliance review process for outward foreign investments. This included rigorous assessments of the feasibility and legality of investment projects. This policy adjustment has encouraged Chinese enterprises to place greater emphasis on compliance, sustainability, and risk management in their outward foreign direct investment activities, which has led to the optimization of investment structures and an improvement in the quality of investments.

2.1.2. Analysis of OFDI by Enterprises of Different Ownership Types

With the advancement of global integration, the amount of outward foreign direct investment (OFDI) by Chinese enterprises has continued to grow. According to the “2022 Annual Statistical Bulletin on China’s Outward Foreign Direct Investment”, by the end of 2022, the number of enterprises involved in OFDI had exceeded 29,000. From the perspective of enterprise types, private enterprises account for the largest share at 33.6%, followed by limited liability companies at 28.7%, joint-stock companies at 13.5%, and foreign-invested enterprises at 5.7%, with state-owned enterprises (SOEs) having the lowest share of only 5.6%(See Figure 2 and Figure 3).

As the main force of China’s economy, SOEs primarily bear the responsibility for national economic construction and public welfare functions, and their outward investment is relatively limited. In contrast, non-state-owned enterprises, with profit making as their main goal, have shown significant enthusiasm for outward investment, especially since the Belt and Road Initiative was proposed by President Xi Jinping in 2013. Since then, the outward investment amount of non-state-owned enterprises has increased substantially, clearly surpassing that of SOEs, and the investment gap has continued to expand.

2.1.3. Analysis of OFDI by Enterprises of Different Pollution Types

In contrast to the still-immature pollution control policies domestically, developed countries and regions impose more stringent environmental regulations on polluting enterprises (Wang, Y., 2020) [2]. This implies that polluting enterprises face higher environmental protection costs when expanding into international markets, which, in turn, leads to a reduction in their profitability. In contrast, non-polluting enterprises, due to their better adherence to environmental regulations, incur lower costs and achieve higher profitability. As a result, under the pressure of environmental compliance, the market share and profits of polluting enterprises gradually decline.

Consequently, the expansion of polluting enterprises into overseas markets is constrained, making it difficult for them to engage in large-scale mergers, acquisitions, or outward foreign direct investments.

According to the data presented in Figure 4, the volume and value of outward foreign direct investment by non-polluting enterprises are notably higher than those of polluting enterprises, with this gap widening annually. This suggests that, in the context of increasingly stringent global environmental regulations, the traditional development model of polluting enterprises requires urgent transformation. In the future, green and environmentally sustainable enterprises are expected to become the dominant force in economic development. Polluting enterprises must, therefore, relocate their production processes to regions with less stringent environmental requirements (Xie, F. and Zhang, B., 2021) [3] and accelerate their transition toward green technologies. Only through this shift can they maintain a competitive advantage in the increasingly intense global market; otherwise, they will face heightened survival challenges.

2.2. Analysis of the Current Status of Green Technological Innovation

2.2.1. Analysis of the Overall Scale of Corporate Green Innovation

As microeconomics entities, enterprises should focus not only on economic benefits but also on environmental benefits, thereby promoting green development (Li S et al., 2022) [4]. Enhancing the green technological innovation capabilities of enterprises is key to addressing the lack of sustainable development momentum. It not only enhances the competitiveness of enterprises but also promotes the coordinated development of the economy and the environment.

From 2007 to 2022, the number of green patent applications by Chinese enterprises involved in outward foreign direct investment (OFDI) significantly increased, especially after the proposal of the Belt and Road Initiative, when the growth became even more pronounced. This indicates that when national policies support overseas investment, enterprises’ OFDI activities have contributed to an increase in green patents. As shown in Figure 5, the total number of green patents independently applied for by Chinese enterprises in 2022 was 41.93 times higher than in 2007. This growth reflects the government’s emphasis on the green economy and its active encouragement of green production, which has driven the rapid growth of corporate green patent applications.

Despite the continuous advancement of green innovation technologies, their implementation still faces numerous challenges that require joint efforts from both the government and enterprises to address (Li J. et al., 2021) [5]. Although the Chinese government has introduced multiple policies to encourage green technological innovation, investment in green technology by enterprises remains insufficient, particularly in overcoming technological bottlenecks and developing new productive capacities, where funding constraints remain a critical challenge (Borojo and DG, 2024) [6].

Currently, small and medium-sized enterprises (SMEs) in China face several obstacles in green technology R&D, including limited access to financing channels, high financing costs, and insufficient R&D funds (Tao, Y et al., 2024) [7]. Most enterprises rely primarily on internal cash flows to fund patent research and development; however, green technology R&D involves long development cycles, substantial capital requirements, and uncertain returns, which dampen firms’ enthusiasm for green innovation (Zhou, K et al., 2023) [8]. Simultaneously, China’s green technology market remains underdeveloped, with limited access to intellectual property information, making it difficult for investors to obtain accurate information and further exacerbating the challenges posed by underdeveloped financing channels.

Moreover, many enterprises have not established sound green technology R&D and talent incentive mechanisms, leading to insufficient R&D motivation and a shortage of talent in green technological innovation. Green technology R&D requires not only strong capabilities but also the ability to keep up with trends and understand policy guidance. Lastly, due to unclear definitions of green innovation technologies, a lack of unified technical evaluation systems and certification standards has resulted in low efficiency in green technological innovation. Although China has introduced legal frameworks for financial products such as green credit and green bonds, the relevant technical standards are still unclear, and third-party platforms cannot carry out effective certification. This has allowed some enterprises to obtain ineffective financing under the guise of green technology innovation, leading to market chaos in financing.

2.2.2. Analysis of Green Innovation Technology Types

From the classification of green patent types in Figure 6, it is clear that the number of green invention patents granted is significantly higher than the number of green utility model patents. This is because enterprises with more invention-type green patents typically demonstrate stronger core competitiveness and sustainable development advantages. Such enterprises will be able to secure a foothold in future intense market competition by utilizing their strong innovation capabilities and R&D advantages to create greater value. Therefore, more and more enterprises are placing greater emphasis on applying for green invention patents, which is why their number is relatively higher.

3. Research Background and Significance

3.1. Research Background

Xi Jinping’s ecological civilization thought provides fundamental guidance and action plans for the high-quality co-construction of the green “Belt and Road”. At the opening ceremony of the Third Belt and Road Forum for International Cooperation, President Xi Jinping particularly emphasized the importance of “promoting green development” and laid out a new strategic blueprint for achieving the green development goals of the “Belt and Road” and promoting harmony between humanity and nature. In recent years, the Chinese government has actively promoted the co-construction of a “Green Silk Road” with countries along the Belt and Road, guiding relevant enterprises to expand overseas markets and encouraging more foreign direct investment to better support the construction of the “Green Silk Road”.

With the end of the COVID-19 pandemic, the pace of Chinese enterprises “going global” has become steadier. In 2022, China’s foreign direct investment (FDI) reached 163.12 billion USD, maintaining the country’s position among the world’s leaders. By the end of 2022, China’s cumulative outward foreign direct investment (OFDI) stock reached USD 2.75 trillion USD, with the country remaining among the top three globally for six consecutive years. Chinese enterprises have invested in more than 80% of countries and regions around the world, covering various industries and achieving remarkable success, effectively promoting global economic recovery.

China adheres to green innovation as a means to promote high-quality development, encouraging enterprises to strengthen technological research and development (R&D) and international cooperation to achieve green technological innovation. Some enterprises have utilized OFDI to obtain capital, alleviate financing pressures, and focus on green technology innovation. This has driven the rise of environmentally friendly enterprises and gradually led to the elimination of high-pollution businesses, with green technology innovation-driven enterprises becoming the mainstream of future development.

Significant differences exist among BRICS countries in terms of institutional development, environmental policy intensity, and platforms for technological cooperation. For example, Brazil and South Africa have recently implemented green tariffs and renewable energy subsidy policies, which have increased both the pressure and incentives for Chinese enterprises to engage in local green transformation. In contrast, Russia demonstrates weaker environmental legislation and policy enforcement, leading some Chinese firms to outsource high-pollution activities to the region. Therefore, the varying intensities of green policies across BRICS countries have a substantial impact on the “green motivation” behind OFDI. Comparative empirical evidence shows that Chinese firms are more likely to engage in high-intensity green technology investments in BRICS countries with clearer and more market-oriented green policies (such as India and Brazil), whereas in countries with more relaxed institutional environments, they exhibit a “pollution haven” investment tendency. These institutional discrepancies exemplify the relevance of institutional theory in the study of green innovation within the context of international OFDI.

While previous research has explored the relationship between the Belt and Road Initiative (BRI) and green development from a macro perspective or examined the impact of OFDI on green total factor productivity at the industry level, there remains a lack of micro-level analysis on how firms realize green technological innovation through OFDI. In particular, there is still a shortage of systematic empirical studies on how financing constraints and R&D investment function as mediating mechanisms in the OFDI–green innovation pathway. Moreover, existing research has seldom incorporated firm-level heterogeneity—such as ownership type and pollution intensity—thus overlooking differences in green innovation policy responsiveness across different types of enterprises.

To address these gaps, this study aims to answer the following research questions:

(1)

Does OFDI significantly promote firms’ green technological innovation?

(2)

Does this effect operate through the alleviation of financing constraints and the increase in R&D investment?

(3)

Do these mechanisms vary by firm ownership, pollution type, or green patent type?

Theoretically, this study integrates the resource-based view with institutional theory to advance research on the drivers of green innovation. Practically, it provides differentiated insights for the development of a green Belt and Road and for firms’ internationalization strategies.

3.2. Research Significance

This study, situated within the context of the Belt and Road Initiative (BRI), adopts a firm-level perspective to systematically examine the mechanisms through which outward foreign direct investment (OFDI) influences green technological innovation. It aims to fill the following research gaps:

First, the existing literature has predominantly focused on the macro level—such as the impact of OFDI on national or industrial green development—while lacking micro-level analyses of firm-specific mechanisms. By using A-share listed firms in China as the research sample, this study explores how OFDI affects corporate green innovation behavior through dual pathways of capital and knowledge.

Second, there is a lack of systematic identification of mediating mechanisms. Although prior studies have mentioned the potential for OFDI to generate technology spillovers, few have provided empirical evidence to clarify how such effects are transmitted through the alleviation of financing constraints and increased R&D investment.

Third, the heterogeneity of firms has been largely overlooked. This study further investigates how the green effects of OFDI vary across different types of firms, including state-owned enterprises, non-polluting firms, and those engaging in different categories of patent innovation. In doing so, it extends the applicability of green innovation impact mechanisms across diverse enterprise contexts.

In summary, this paper not only introduces OFDI as a driving variable into the literature on green innovation, but also integrates mainstream economic theories—namely, the resource-based view and institutional theory—into its analytical framework, thereby offering both theoretical and empirical contributions to the study of the green BRI from the perspective of emerging markets.

(1) Theoretical Significance

Currently, research on how outward foreign direct investment at the micro level impacts green technological innovation is relatively scarce. The existing literature predominantly focuses on the effects of foreign investment on the production technology and economic development of the home country, with limited attention paid to green technological innovation. Moreover, while academic studies have explored corporate green technological innovation from perspectives such as policies, regulations, and environmental constraints, few have directly examined the impact of economic behaviors (such as OFDI) on green technological innovation. Therefore, this paper expands the factors influencing green technological innovation, with particular focus on enterprise cooperation with foreign investments. It considers the role of R&D investment and financing constraints in the mechanism, enriching the theoretical research in this field.

(2) Practical Significance

OFDI is an important tool for China to enhance international cooperation and promote economic modernization, and it is also one of the key pathways for building the “Green Silk Road.” Investigating the green effects of China’s OFDI and its role in promoting green technological innovation holds significant practical importance. This research not only contributes to the high-quality construction of the “Belt and Road” but also provides theoretical foundations and practical experience for further implementing the “Go Global” strategy. Furthermore, by analyzing the relationship and impact pathways between OFDI and green technological innovation through fixed-effects and mediation models, as well as conducting heterogeneity analysis, this paper offers policy recommendations for promoting green innovation and achieving sustainable development in China.

4. Research Methods, Literature Review, and Mechanism Analysis

4.1. Research Methods

This paper primarily employs three methods to investigate the impact of outward foreign direct investment (OFDI) on green technological innovation in enterprises. These methods are as follows:

4.1.1. Literature Analysis

Based on the research framework of this paper, a large volume of literature on OFDI and green technological innovation was collected and analyzed at multiple levels. This analysis helped establish the research framework for the study and provided sufficient theoretical and empirical support for the research.

4.1.2. Model Construction

Based on the research hypotheses, a quantitative model was constructed to examine the relationship between OFDI and green innovation. The model reflects both the direct and deeper influences between the two variables at multiple levels, in order to better understand the inherent relationship, the mediation mechanisms, and the heterogeneity effects.

4.1.3. Empirical Analysis

This paper utilizes STATA17.0 software to conduct baseline regression analysis using the fixed-effects model on panel data from A-share listed companies between 2007 and 2022. In addition, mediation effect tests, robustness checks, and heterogeneity analysis are conducted to further validate the results.

4.2. Literature Review

4.2.1. The Relationship Between Outward Foreign Direct Investment and Green Innovation

Scholars have explored the relationship between outward foreign direct investment (OFDI) and green innovation from various perspectives, including the technology spillover effect, heterogeneity tests, and mechanism tests.

Among them, Wang S. et al. (2023) [1] argue that OFDI by Chinese enterprises significantly enhances the level of green technological innovation in countries along the Belt and Road. Through a mechanism analysis, they found that OFDI promotes green technological innovation in these countries through two main channels: technology spillover effects and growth effects. Additionally, the level of green innovation is closely related to the intensity of cooperation between China and Belt and Road Initiative (BRI) countries, and OFDI has been shown to significantly promote green technological innovation, particularly in high-income BRI countries. Zheng M.G. et al. (2022) [9] found that OFDI has a significant impact on corporate green innovation, with a more pronounced effect on enterprises investing in developed countries and regions, as well as on substantive green innovation. Xing G. and Dong H. (2023) [10] constructed a moderated mediation effect model and discovered that OFDI has a reverse spillover effect on green technological innovation. From the perspective of moderated mediation effects, OFDI has a positive impact on the greening of the production environment of Chinese enterprises. Among these effects, increased domestic R&D investment helps narrow the green technology gap between China and developed countries, while stricter domestic environmental regulations encourage multinational corporations to strengthen cooperation with foreign firms in the field of green technology. Dai L. et al. (2021) [11] used the Super-SBM model to measure green innovation levels and applied the GMM model to test the impact of OFDI on green innovation. The results indicate that, overall, OFDI exerts a significant positive effect on green innovation. However, heterogeneity tests reveal that the impact of OFDI on green innovation varies significantly across different regions. Similarly, Yang, Z.J. and Wang D.Y. (2019) [12] employed the PVAR model and found that OFDI and green total factor productivity exhibit a mutually reinforcing relationship.

In addition to outward foreign direct investment (OFDI), firms’ green innovation is also driven by a range of other factors, including domestic investment (such as government subsidies and green credit), internal technological accumulation, market competition intensity, and policy orientation (Xiao, Z. et al., 2022) [13]. Unlike government subsidies, OFDI not only brings direct capital expansion, but also facilitates technological collaboration, transnational resource integration, and institutional adaptation. It is, thus, characterized by more dynamic capabilities of “knowledge absorption” and “capability building” (Luo, Y. & Tung, R. 2007) [14].

Domestic investment is typically centered on capital injection, often yielding short-term effects. However, its ability to systematically shape firms’ green innovation capacity tends to be limited (Porter M.E. et al., 1995) [15]. In contrast, OFDI contributes to the evolution of green innovation over the medium and long term through mechanisms such as reverse technology spillovers and cross-border network learning (Cohen W.M & Levinthal, D.A. 1990) [16]. This highlights the fact that, compared with other drivers, OFDI operates through more complex yet higher-potential pathways, making it particularly suitable for fostering substantive, invention-oriented green innovation.

However, some scholars argue that OFDI does not significantly contribute to the enhancement of green innovation levels and may even have negative effects. Luo, L.W. and Liang, S.R. (2017) [17] used the DEA method to measure the efficiency of green technological innovation and constructed a spatial econometric model based on the nonlinear CH model. Their findings indicate that OFDI’s R&D capital has a crowding-out effect on domestic investment, hindering improvements in green technological innovation efficiency. Zheng, Q. and Ran G.H. (2018) [18] first calculated green total factor productivity and then tested their results using a dynamic panel model, revealing that OFDI serves as a “stumbling block” to green innovation in China. Xia J. et al. (2016) [19] argue that the level of marketization influences the technological spillover effects of OFDI, and that the positive technology spillover effect of OFDI is only significant when the marketization level is sufficiently high. Alvarado R. et al. (2017) [20] found that only OFDI from high-income countries can effectively promote technological innovation, whereas OFDI from middle- and low-income countries tends to hinder economic development and technological innovation. Chen Y. et al. (2019) [21] discovered that the Belt and Road Initiative (BRI) does not directly stimulate R&D investment in OFDI firms but instead exerts a short-term inhibitory effect, with a more pronounced negative impact on state-owned enterprises (SOEs).

4.2.2. Review of the Research Status

The above literature has discussed the impact of OFDI on green innovation from multiple perspectives. However, due to differences in research perspectives and methodologies, the conclusions drawn have not been consistent. Furthermore, many studies have not fully considered the nonlinear relationships between variables, which may result in biased estimation outcomes. In addition, the existing literature rarely explores in depth the mechanisms through which OFDI affects green innovation in enterprises, or the main channels through which enterprises gain green innovation through OFDI. These are essential for accurately analyzing the green innovation effects generated by OFDI. Therefore, this paper will further investigate the shortcomings of the existing literature.

Despite the valuable insights offered by the existing literature, several limitations remain:

(1)

Most studies on the relationship between OFDI and green innovation focus on the macro or industry level, with insufficient attention to firm-level micro mechanisms;

(2)

The majority of research fails to incorporate financing constraints or R&D investment into the explanatory framework, thereby overlooking the “resource reallocation” pathway of OFDI;

(3)

Few studies systematically interpret corporate behavior by integrating mainstream economic theories such as the resource-based view and institutional theory;

(4)

Many analyses lack heterogeneity tests, ignoring structural differences among firms in their policy responsiveness and technological absorptive capacity.

To address these gaps, this paper introduces firm-level microdata and integrates the transmission mechanisms linking OFDI, financing constraints, and green innovation. By constructing a dual theoretical–empirical framework, it aims to fill the existing research void and contribute to a more comprehensive understanding of how OFDI drives green innovation at the firm level.

4.3. Theoretical Analysis of the Impact of OFDI on Green Innovation in Enterprises

4.3.1. Theoretical Mechanisms: The Impact Pathways of OFDI on Green Innovation

From the perspective of the resource-based view (RBV), outward foreign direct investment (OFDI) represents a strategic behavior through which firms acquire heterogeneous resources from host countries, thereby building the unique resource endowments required for green innovation. Through OFDI, firms can access advanced green technologies, high-quality human capital, and low-carbon management expertise from abroad, and these are then transformed into core competencies that enhance the firms’ green technological innovation capacity (Ma, X. et al., 2024) [22]. This aligns with RBV’s central tenet that sustainable competitive advantage stems from valuable and rare resources (Wernerfelt, B.A 1984) [23]. In addition, the springboard internationalization theory suggests that emerging market firms often use OFDI as a means to overcome domestic resource disadvantages. By engaging in overseas mergers, acquisitions, and collaborations, these firms are able to acquire green patents and specialized talent, thereby achieving a leap in innovation capability. This process, however, requires a strong absorptive capacity to ensure that the externally sourced knowledge can be effectively internalized and transformed into domestic innovation outputs. Therefore, grounded in RBV and supported by organizational learning theory, it is reasonable to expect that the rare knowledge and technical capabilities obtained through OFDI will significantly enhance firms’ green technological innovation (Shao, Y. et al., 2024) [24].

At the same time, institutional theory emphasizes the profound influence of the external institutional environment on corporate behavior. Within the strategic context of OFDI—particularly under initiatives such as the Belt and Road Initiative (BRI)—firms are often subject to strict environmental regulations and green development policies in host countries. These create institutional pressures that compel firms to enhance their environmental awareness and increase investments in green innovation in order to gain legitimacy and recognition from host governments and societies. This dynamic is especially pronounced in the case of multinational operations, where firms must simultaneously comply with institutional norms from both the home and host countries. Under the scrutiny of a broader set of stakeholders, they are expected to fulfill more comprehensive green responsibilities (Yang, Z. et al., 2020) [25]. Drawing on the institutional pressure perspective, Yang, Z. et al. (2020) [25] found that when firms expand into international markets, the pursuit of legitimacy becomes a major driving force that incentivizes stronger engagement in green technological innovation to meet the environmental compliance requirements of cross-border operations. In other words, OFDI embeds firms within pluralistic institutional environments, where they are simultaneously influenced by the stringent environmental standards of host countries and the green development imperatives of the home country. As a result, firms are more likely to adopt green production processes and develop clean technologies to align with these overlapping institutional expectations (Wang, M. et al., 2021) [26]. This also explains why Chinese firms often achieve higher levels of green innovation when investing in countries with stricter environmental regulations.

Beyond resource acquisition and institutional adaptation, the impact of OFDI on green innovation also manifests through several economic mechanisms, including technology spillovers and competitive pressure (Sun, Y. et al., 2023) [27]. On one hand, firms that “go global” gain direct access to emerging green technologies and international knowledge networks, enabling reverse technology spillovers that channel foreign knowledge back to the home country and enhance the firm’s innovation capacity. Empirical research has shown that reverse knowledge spillovers generated by Chinese OFDI significantly contribute to domestic progress in green technologies and improve total factor productivity (Lin X. et al., 2025) [28]. On the other hand, OFDI exposes firms to the rigors of international market competition, where higher standards and broader competitive dynamics compel them to upgrade their R&D capabilities and improve environmental performance (Kshetri, N. 2008) [29]. Some scholars have argued that OFDI activates three core micro-level mechanisms—the scale effect, competitive effect, and human capital effect—which together stimulate parent firms’ green innovation potential (Wang, J. et al., 2024) [30]. The scale effect refers to firms’ ability to expand production and sales through entry into larger markets, thus securing more resources to invest in green R&D. The competitive effect suggests that exposure to international rivals forces firms to accelerate their green technology upgrading to maintain competitiveness (Yang, H. et al., 2011) [31]. The human capital effect implies that multinational operations help firms attract and develop talent with expertise in environmental technologies, strengthening the talent foundation for green innovation.

In summary, multiple theoretical perspectives collectively suggest that OFDI exerts a positive influence on corporate green innovation. Accordingly, this study proposes the following hypothesis:

H1. 

Outward foreign direct investment significantly enhances firms’ green technological innovation. That is, holding other conditions constant, firms engaging in OFDI are expected to generate more green innovation outputs than those that do not.

4.3.2. Mediating Role of Financing Constraints

Insufficient capital investment is one of the key barriers to technological innovation, particularly in the context of green innovation. When firms face severe financing constraints, they often struggle to raise adequate funds for high-risk, long-term green R&D projects. Theoretically, OFDI may help alleviate financial pressure, thereby indirectly promoting investment in green innovation. It is important to recognize that OFDI itself is characterized by long cycles and high risks, which may initially exacerbate financial strain and raise firms’ cost of capital (Zhu, X. et al., 2023) [32]. Some studies have argued that firms with high financing constraints are less likely to undertake OFDI due to the substantial capital requirements associated with overseas projects. However, in the long run, successful OFDI activities can improve firms’ revenue structures and asset positions, thereby broadening their access to external financing. For Chinese firms, diversified overseas investments allow them to acquire scarce resources and explore new markets, which helps reduce operational risks and enhance overall profitability. Once overseas projects begin generating stable cash flows, firms will have more internally available capital for R&D activities, thereby alleviating the funding bottlenecks that often restrict green innovation. Empirical evidence from Shao, Y. et al. (2024) [24] confirms that OFDI can increase firms’ internal capital supply by enhancing overseas operating income, thus providing financial support for green innovation. This finding validates the so-called “surplus enhancement” effect, whereby profits repatriated from international investments and economies of scale are partially redirected toward the development of green technologies in the home country.

Secondly, OFDI may also serve as a positive signal to capital markets, thereby helping firms ease external financing constraints. According to information asymmetry theory, banks and investors often lack complete knowledge about a firm’s future prospects, leading to higher borrowing costs or limited access to credit. If a firm’s “going global” strategy achieves phased success, it can demonstrate international competitiveness and growth potential, thereby strengthening investor confidence. Empirical studies have shown that firms engaged in OFDI are more likely to obtain bank lending and equity financing (Liao, Z. et al., 2024) [33], as international operations are generally viewed as indicators of firm strength and credibility. This signaling effect is particularly pronounced for private firms, which typically face higher financing barriers. For such firms, OFDI can enhance reputation and reduce perceived risks, helping to lower interest rates and improve credit ratings, thereby alleviating financial pressures and enabling greater investment in technological innovation (Borensztein, E. et al., 1998) [34]. In addition, governments have introduced a range of financial support policies—such as concessional loans and credit guarantees—for green OFDI projects under the Belt and Road Initiative (BRI). These policies further reduce financing constraints and encourage firms to increase their investment in green R&D (Teece D.J., 2007) [35].

In sum, OFDI can mitigate financing constraints through two complementary channels: improving internal cash flow and enhancing external financing conditions. Based on this mechanism, we propose the following hypothesis:

H2. 

Financing constraints mediate the relationship between OFDI and green technological innovation. That is, OFDI indirectly promotes green innovation by alleviating firms’ financing constraints.

4.3.3. Mediating Role of R&D Investment

R&D investment serves as the direct driving force behind green technological innovation. Only through sustained R&D activities can firms develop new products and processes that are energy-efficient and environmentally friendly, thereby achieving cumulative green innovation outcomes. Existing studies have confirmed that increasing R&D intensity significantly improves green patent output and green total factor productivity (Dong, F. et al., 2025) [36]. However, numerous factors influence a firm’s level of R&D investment, and OFDI provides an important opportunity to enhance R&D input. On one hand, OFDI enables firms to leverage advanced R&D resources and innovation networks in host countries, thereby improving their R&D efficiency through international collaboration (Kim, S. et al., 2017) [37]. When firms establish overseas R&D centers or engage in joint green technology development with local research institutions, they gain access to the technological expertise, infrastructure, and innovation ecosystems of their partners. This allows them to share the cost and risk of innovation projects. Kim, S. et al. (2017) [37] found that cross-border R&D collaboration generates synergistic effects, which reduce the innovation burden on individual firms and allow them to allocate more financial and human resources to green technological breakthroughs. This suggests that OFDI can effectively expand the boundaries of corporate R&D, enabling firms to harness global innovation resources and improve the efficiency and output of their R&D investments.

On the other hand, OFDI may also stimulate firms’ internal motivation to increase R&D investment. In the context of international market competition, firms must continuously innovate to maintain or strengthen their positions in foreign markets, particularly by adapting to diverse environmental regulations and consumer preferences. Such external pressure pushes firms to allocate more R&D resources to the development of environmentally friendly products that comply with host country laws or align with local market demand (Calel, R. and Dechezleprêtre, A., 2016) [38]. At the same time, the new knowledge and novel ideas acquired through OFDI generate spillover effects within the firm, broadening the perspectives of R&D personnel and fostering fresh thinking in green innovation. This process enhances the effectiveness and relevance of R&D investments (Rodrik, D., 2004) [39]. It is important to note that green innovation typically involves high levels of path dependency and technological uncertainty, meaning that conventional R&D investments alone may not be sufficient to overcome core technical bottlenecks. OFDI provides experimental grounds in foreign markets and valuable opportunities to learn from advanced international experiences, thereby increasing firms’ confidence and willingness to invest more aggressively in R&D for critical environmental technologies. Empirical evidence also shows a positive correlation between OFDI and R&D intensity—firms engaged in OFDI tend to demonstrate significantly higher R&D input levels than purely domestic firms. Micro-level data from Shao, Y. et al. (2024) [24] further support this view: OFDI enhances green technological innovation by increasing firms’ R&D investments.

Based on these insights, we propose the following hypothesis:

H3. 

R&D investment mediates the relationship between OFDI and green technological innovation. That is, OFDI indirectly promotes green innovation by enhancing firms’ level of R&D investment.

The following is a path analysis diagram of the mechanism of action (See Figure 7).

4.4. Data Sources

This study selects panel data from A-share listed companies involved in outward foreign direct investment (OFDI) from 2007 to 2022. To ensure the accuracy and reliability of the empirical results, the following screening principles were applied to the selected sample before conducting the empirical analysis: (1) ST, *ST, and PT companies were excluded; (2) companies registered in tax havens were excluded; (3) companies with zero total assets were excluded; and (4) to mitigate the impact of outliers, the data were truncated at the 1% level. A final sample of 8,790 companies was obtained.

The data used in the empirical analysis primarily come from the following databases: Green patent data are sourced from the CNRDS Listed Company Green Patent Database; data on China’s outward foreign direct investment come from the Ministry of Commerce’s published annual statistical reports; and other company-level data are sourced from the Guotai An Database.

4.5. Variable Definitions and Model Design

4.5.1. Variable Definitions

(1) Dependent Variable: Green Innovation Ability (PAT)

Green patent applications are specifically categorized into green invention patents (INVPAT) and green utility model patents (UTYPAT). This paper uses the number of green patents applied for by enterprises in a given year to measure their green innovation ability, following the green patent indicator construction method of scholars Xu Jia and Cui Jingbo (2020) and Wang Xin and Wang Ying (2021) [40,41]. The specific measurement method is to sum the number of green invention patents and green utility model patents to obtain the total number of green patent applications for the year, and to then apply the logarithm transformation by adding 1 to the total green patent applications.

(2) Independent Variable: Outward Foreign Direct Investment Amount (OFDI)

In previous studies, many papers have used the amount of outward foreign direct investment as an indicator to measure the scale of OFDI. Therefore, in this study, the amount of OFDI is used to measure its impact on green innovation. The measurement method applied is the logarithm of the amount of OFDI. Additionally, in the robustness check, this paper introduces the number of OFDI occurrences (TOFDI) as an alternative to the core independent variable OFDI, which refers to the number of times a company engages in OFDI in a given year, to ensure the robustness of the baseline analysis.

(3) Mediating Variables

Financing Constraints (SA): In this study, the severity of financing constraints is measured using the SA index. The greater the absolute value of SA, the more severe the financing constraints. Since the SA value is negative, it implies that a smaller SA value indicates more severe constraints. This study analyzes how OFDI alleviates financing constraints, further exploring the impact of financing constraints on green innovation and ultimately deriving the impact of OFDI on green innovation.

R&D Investment (RD): The R&D investment level is measured by the logarithm of the R&D investment amount. Existing research shows that increasing R&D investment can promote green technological innovation. This study investigates whether OFDI increases R&D investment and uses R&D investment as a mediating variable to study the effect of OFDI on promoting green innovation.

(4) Control Variables

The control variables in this study include variables reflecting a company’s debt-paying ability and profitability, such as the debt-to-asset ratio and return on equity (ROE); variables reflecting corporate governance ability, such as the number of board members and the proportion of independent directors; and variables reflecting the operational status of the company, such as the book-to-market ratio and the proportion of fixed assets. As shown in

Table 1. Variable definition table.

Variable Type	Variable Name	Symbol	Measurement Method
Dependent Variable	Green Innovation	PAT	Logarithm of the number of green patent applications +1
Independent Variable	Outward Foreign Direct Investment	OFDI	Logarithm of the amount of outward foreign direct investment
Mediating Variable	Financing Constraints	SA	The larger the absolute value, the greater the financing constraint
	R&D Investment	RD	Logarithm of R&D investment
Control Variables	Debt-to-Asset Ratio	LEV	Total liabilities at year-end/Total assets at year-end
	Return on Equity	ROE	Net profit/Average owners’ equity balance
	Book-to-Market Ratio	BM	Book value/Total market value
	Number of Directors	BOARD	Natural logarithm of the number of board members
	Proportion of Independent Directors	INDEP	Number of independent directors/Total number of directors × 100%
	Proportion of Fixed Assets	FIXED	Net fixed assets/Total assets

(variable definition table), the measurement methods for these control variables are detailed.

4.5.2. Model Design

This study employs a fixed effects model, as panel data exhibit clear individual heterogeneity. The fixed effects approach controls for unobservable firm-level characteristics, making it more consistent with the realities of enterprise behavior. In contrast, the random effects model assumes that individual effects are uncorrelated with the explanatory variables, which does not hold in the context of this research.

Additionally, a mediation effect model is used to identify causal pathways among variables, enabling the analysis to explain how OFDI influences green innovation through the mechanisms of financing constraints and R&D investment.

This paper primarily studies the relationship between outward foreign direct investment (OFDI) by Chinese enterprises and green technological innovation in the context of the Belt and Road Initiative. To examine the impact of OFDI on green technological innovation, this paper constructs the following fixed-effects model:

$${\mathrm{P}\mathrm{A}\mathrm{T}}_{\mathrm{i},\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{O}\mathrm{F}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\alpha }}_2{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{t}}+\mathrm{I}\mathrm{n}\mathrm{d}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{y}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(1)

In Equation (1), ${\mathsf{\alpha }}_0$ is the constant term, ${\mathsf{\alpha }}_1$ and ${\mathsf{\alpha }}_2$ are the estimated coefficients of the variables, i represents the outward foreign direct investment enterprises, t represents the year of outward foreign direct investment, PAT represents the green patent applications, which indicates the enterprise’s green innovation capability, OFDI is the core independent variable, which varies across individuals and time, Control represents the set of control variables, and ${\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$ represents the random disturbance term. Additionally, Industry and Year represent the fixed effects for controlling industry and year, respectively.

5. Empirical Analysis

5.1. Descriptive Statistics

Using STATA software, descriptive statistics were performed on the mean, standard deviation, minimum, and maximum values for the selected sample. The specific results are shown in

Table 2. Descriptive statistics.

Variable	Observations	Mean	Standard Deviation	Minimum	Maximum
PAT	8790	0.420	0.836	0	3.871
OFDI	8790	16.220	3.361	5.090	20.970
SA	8790	−3.755	0.273	−4.378	−2.865
RD	8260	17.195	4.509	0	24.630
LEV	8790	0.436	0.196	0.056	0.850
ROE	8790	0.076	0.109	−0.427	0.363
BM	8790	0.637	0.242	0.143	1.173
BOARD	8790	2.133	0.205	1.609	2.708
INDEP	8790	37.756	5.468	33.330	57.140
FIXED	8790	0.207	0.145	0.003	0.645

.

From the table, we can observe that (1) after trimming the data at the 1% level, the maximum value of PAT (the indicator for green innovation ability) is 3.871, the minimum value is 0, and the mean is 0.420. This indicates that some enterprises exhibit strong green innovation capabilities in certain years, while others do not achieve green technological innovation in those years, reflecting significant differences in the green innovation performance of different enterprises. (2) In some years, the maximum value of OFDI (outward foreign direct investment) is 20.97, while the minimum value in other years is 5.09, with a mean of 16.22. This shows a large individual variation in the outward investment amounts between enterprises. (3) The maximum value of R&D investment is 24.63, the minimum is 0, and the mean is 17.195. This suggests that most enterprises allocate some amount of funds to R&D for technological innovation each year, but there are significant differences in R&D investment across enterprises. (4) Additionally, the dispersion of the other variables is relatively good, indicating that the selected sample has strong representativeness.

5.2. Correlation Test

Before conducting the baseline regression analysis, this paper performed a correlation test for the main variables, and the results are shown in

Table 3. Correlation test of main variables.

Variable	PAT	OFDI	SA	RD	LEV	ROE	BM	BOARD	INDEP	FIXED
PAT	1
OFDI	0.102 ***	1
SA	0.116 ***	0.036 ***	1
RD	0.233 ***	0.055 ***	0.102 ***	1
LEV	0.126 ***	0.148 ***	−0.018 *	−0.052 ***	1
ROE	0.039 ***	0.042 ***	0.092 ***	0.046 ***	−0.122 ***	1
BM	0.084 ***	0.180 ***	0.051 ***	−0.043 ***	0.388 ***	−0.151 ***	1
BOARD	0.104 ***	0.081 ***	0.108 ***	−0.015	0.171 ***	0.037 ***	0.166 ***	1
INDEP	0.024 **	0.008	0.060 ***	0.026 **	−0.007	−0.037 ***	0.019 *	−0.496 ***	1
FIXED	0.007	0.082 ***	0.068 ***	0.022 **	0.102 ***	−0.089 ***	0.126 ***	0.157 ***	−0.024 **	1

Note: This table shows the correlation coefficients between the variables in this study. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

. The symbols and numbers in the table represent the direction and magnitude of the correlations between the variables. By observing Table 3, we can see that the correlation coefficients between all variables are less than 0.5, indicating that there is no obvious multicollinearity problem between the main variables under study.

5.3. Baseline Regression Analysis

To examine the impact of outward foreign direct investment (OFDI) on the green innovation capability of enterprises, this paper first conducts a baseline regression of the constructed main model. The regression results are shown in

Table 4. Baseline regression results.

Variable	(1)	(2)	(3)	(4)
OFDI	0.0254 ***	0.0183 ***	0.0246 ***	0.0152 ***
	(0.0026)	(0.0027)	(0.0055)	(0.0047)
LEV		0.4025 ***		0.5287 ***
		(0.0491)		(0.0969)
ROE		0.3577 ***		0.6069 ***
		(0.0818)		(0.1134)
BM		0.0763 *		0.2353 ***
		(0.0403)		(0.0865)
BOARD		0.5023 ***		0.6388 ***
		(0.0512)		(0.1378)
INDEP		0.0132 ***		0.0121 ***
		(0.0019)		(0.0035)
FIXED		−0.1415 **		−0.2528
		(0.0617)		(0.1643)
Constant	0.0079	−1.6664 ***	0.0207	−2.0190 ***
	(0.0437)	(0.1555)	(0.0812)	(0.4146)
Industry	NO	NO	YES	YES
Year	NO	NO	YES	YES
N	8790	8790	8790	8790
adj. R2	0.010	0.037	0.150	0.191

Note: This table shows the baseline regression results for the variables in this study. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively, with clustered robust standard errors in parentheses.

.

From columns (1) and (3) in the table, we can see that, regardless of whether industry and year fixed effects are controlled for, and without adding other control variables, the regression coefficient of OFDI is significantly positive at the 1% level. This indicates that OFDI significantly promotes green technological innovation.

From columns (2) and (4), when control variables such as debt-to-asset ratio, return on equity, book-to-market ratio, number of board members, proportion of independent directors, and proportion of fixed assets are included, the regression coefficient of OFDI remains significantly positive at the 1% level, regardless of whether industry and year fixed effects are controlled for. This further confirms that OFDI has a significant positive impact on green technological innovation, supporting Hypothesis 1.

5.4. Mediation Effect Test

After conducting the baseline regression analysis on the fixed-effects model established earlier, this paper aims to further understand the impact mechanism between outward foreign direct investment (OFDI) and green technological innovation, and whether the effects are positive or negative. To achieve this, a mediation effect model is constructed to further explore the path through which OFDI influences green innovation in enterprises.

As discussed in the theoretical analysis and baseline regression analysis, OFDI has a clear promoting effect on green technological innovation. Through OFDI, enterprises can achieve the effects of alleviating financing constraints and promoting R&D investment, and these two effects can enhance the green innovation capability of enterprises. To verify this mechanism, this paper establishes the following mediation effect model to test whether these two effects exist.

$${\mathrm{P}\mathrm{A}\mathrm{T}}_{\mathrm{i},\mathrm{t}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{O}\mathrm{F}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\alpha }}_2{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{t}}+\mathrm{I}\mathrm{n}\mathrm{d}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{y}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(2)

$${\mathrm{S}\mathrm{A}}_{\mathrm{i},\mathrm{t}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{O}\mathrm{F}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\beta }}_2{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{t}}+\mathrm{I}\mathrm{n}\mathrm{d}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{y}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(3)

$${\mathrm{R}\mathrm{D}}_{\mathrm{i},\mathrm{t}}={\mathsf{\gamma }}_0+{\mathsf{\gamma }}_1{\mathrm{O}\mathrm{F}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\gamma }}_2{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{t}}+\mathrm{I}\mathrm{n}\mathrm{d}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{y}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(4)

$${\mathrm{P}\mathrm{A}\mathrm{T}}_{\mathrm{i},\mathrm{t}}={\mathsf{\theta }}_0+{\mathsf{\theta }}_1{\mathrm{O}\mathrm{F}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{t}}+{{\mathsf{\theta }}_2\mathrm{S}\mathrm{A}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\theta }}_3{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{t}}+\mathrm{I}\mathrm{n}\mathrm{d}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{y}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(5)

$${\mathrm{P}\mathrm{A}\mathrm{T}}_{\mathrm{i},\mathrm{t}}={\mathsf{\mu }}_0+{\mathsf{\mu }}_1{\mathrm{O}\mathrm{F}\mathrm{D}\mathrm{I}}_{\mathrm{i},\mathrm{t}}+{{\mathsf{\mu }}_2\mathrm{R}\mathrm{D}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\mu }}_3{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}_{\mathrm{i},\mathrm{t}}+\mathrm{I}\mathrm{n}\mathrm{d}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{y}+\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(6)

In the model, ${\mathsf{\alpha }}_0$, ${\mathsf{\beta }}_0$, ${\mathsf{\gamma }}_0$, ${\mathsf{\theta }}_0$, and ${\mathsf{\mu }}_0$ are constant terms, and ${\mathsf{\alpha }}_1$, ${\mathsf{\alpha }}_2$, ${\mathsf{\beta }}_1$, ${\mathsf{\beta }}_2$, ${\mathsf{\gamma }}_1$, ${\mathsf{\gamma }}_2$, ${\mathsf{\theta }}_1$, ${\mathsf{\theta }}_2$, ${\mathsf{\mu }}_1$, and ${\mathsf{\mu }}_2$ represent the estimated coefficients for each variable. i represents the outward foreign direct investment enterprises, and t represents the year of outward foreign direct investment. PAT denotes the number of green patent applications, which indicates the enterprise’s green innovation capability. OFDI is the core independent variable, which varies across individuals and time. SA represents financing constraints, RD represents R&D investment, and Control represents the set of control variables. ${\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$ represents the random disturbance term. Additionally, Industry and Year represent the fixed effects for controlling industry and year, respectively.

The mediation effect model constructed in this paper follows the transmission path:

①

As established in the baseline regression analysis using the fixed-effects Equation (2) in the previous section, we have already determined that outward foreign direct investment (OFDI) significantly promotes green technological innovation in enterprises.

②

According to Equation (3), we test whether the regression coefficient ${\mathsf{\beta }}_1$ of the mediating variable SA (financing constraints) with respect to the core independent variable OFDI is significant, which reflects the relationship between outward foreign direct investment (OFDI) and financing constraints. Then, according to Equation (5), we test whether the regression coefficient ${\mathsf{\theta }}_2$ of SA with respect to the dependent variable PAT (green patent applications) is significant, which reflects the relationship between green innovation capability and financing constraints.

③

According to Equation (4), we test whether the regression coefficient ${\mathsf{\gamma }}_1$ of the mediating variable RD (R&D investment) with respect to the core independent variable OFDI is significant, which reflects the relationship between outward foreign direct investment (OFDI) and R&D investment. Then, according to Equation (6), we test whether the regression coefficient ${\mathsf{\mu }}_2$ of RD with respect to the dependent variable PAT (green patent applications) is significant, which reflects the relationship between green innovation capability and R&D investment.

④

If the regression coefficient ${\mathsf{\beta }}_1$ of the mediating variable SA with respect to the core independent variable OFDI is significant, and the regression coefficient ${\mathsf{\theta }}_2$ of SA with respect to the dependent variable PAT is also significant, it can be concluded that outward foreign direct investment (OFDI) can alleviate financing constraints, and the alleviation of these constraints can promote green technological innovation. Based on the conclusions of the previous study, it can be inferred that OFDI promotes green technological innovation. Therefore, the mediation effect model with SA as the mediating variable is valid.

⑤

When the regression coefficient ${\mathsf{\gamma }}_1$ of the mediating variable RD with respect to the core independent variable OFDI is significant, and the regression coefficient ${\mathsf{\mu }}_2$ of RD with respect to the dependent variable PAT is also significant, it can be concluded that outward foreign direct investment (OFDI) promotes an increase in R&D investment, and the increase in R&D investment further promotes green technological innovation. Similarly, based on the previous research conclusions, it can be concluded that OFDI promotes green technological innovation. Therefore, the mediation model with RD as the mediating variable is valid.

Interpretation of the

Table 5. Mechanism test results.

	(1)	(2)	(3)	(4)
Variable	SA	PAT	RD	PAT
OFDI	0.0058 ***	0.0133 ***	0.0814 ***	0.0118 ***
	(0.0017)	(0.0045)	(0.0185)	(0.0046)
LEV	−0.1450 ***	0.5771 ***	0.9362 *	0.5787 ***
	(0.0309)	(0.0984)	(0.4845)	(0.0987)
ROE	0.0872 **	0.5778 ***	2.5007 ***	0.5367 ***
	(0.0394)	(0.1114)	(0.6437)	(0.1100)
BM	0.0247	0.2270 ***	1.4974 ***	0.1959 **
	(0.0289)	(0.0846)	(0.4123)	(0.0854)
BOARD	0.0723 **	0.6147 ***	1.5513 ***	0.6175 ***
	(0.0360)	(0.1334)	(0.4526)	(0.1382)
INDEP	0.0048 ***	0.0105 ***	0.0170	0.0108 ***
	(0.0014)	(0.0034)	(0.0155)	(0.0036)
FIXED	-0.0679	−0.2302	−0.6819	−0.2026
	(0.0561)	(0.1599)	(0.7491)	(0.1663)
SA		0.3333 ***		0.2944 ***
		(0.0979)		(0.0982)
RD				0.0222 ***
				(0.0032)
Constant	−4.1274 ***	−0.6432	10.5399 ***	−1.1257 **
	(0.1172)	(0.4860)	(1.3073)	(0.5057)
Industry	YES	YES	YES	YES
Year	YES	YES	YES	YES
N	8790	8790	8259	8259
adj. R2	0.313	0.199	0.418	0.204

Note: This table shows the results of the mechanism test for the variables in this study. ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively, with clustered robust standard errors in parentheses.

: From the results of the regression in columns (1) and (2), we can observe that when financing constraints are included in the model, the regression coefficient of OFDI is significantly positive at the 1% level in column (1), and the regression coefficient of SA (financing constraints) is also significantly positive at the 1% level in column (2). This indicates that the model with financing constraints as the mediating variable is valid, and the mediation effect is significant. OFDI can alleviate financing constraints, which in turn promotes green technological innovation. This supports Hypothesis 2.

In columns (3) and (4), when R&D investment is added as another variable, the regression coefficient of OFDI remains significantly positive at the 1% level in column (3), and the regression coefficient of RD (R&D investment) is also significantly positive at the 1% level in column (4). This indicates that the model with R&D investment as the mediating variable is valid, and the mediation effect is significant. OFDI promotes an increase in R&D investment, which further fosters green technological innovation. This supports Hypothesis 3.

5.5. Robustness Check

(1) Substituting the Independent Variable

This paper uses the number of green patent applications by enterprises as an indicator to measure green innovation capability. To ensure the robustness of the empirical results, the paper employs a substitution approach for the independent variable. Specifically, the variable for outward foreign direct investment (OFDI) is replaced with the number of times a company makes outward foreign direct investments (denoted as “TOFDI” in this paper). This new variable is then incorporated into the previously constructed fixed-effects Equation (2) for baseline regression, controlling for industry and year fixed effects. The regression results are presented in

Table 6. Robustness check results.

Variable	(1)	(2)	(3)	(4)
TOFDI	0.0269 **
	(0.0132)
L.PAT				0.4441 ***
				(0.0365)
L2.PAT				0.1615 ***
				(0.0350)
OFDI		0.0147 ***		0.0068 **
		(0.0046)		(0.0030)
L.OFDI			0.0151 ***
			(0.0055)
LEV	0.5397 ***	0.5512 ***	0.5686 ***	0.1821 ***
	(0.0993)	(0.0979)	(0.1127)	(0.0637)
ROE	0.6155 ***	0.6994 ***	0.6677 ***	0.2016 ***
	(0.1140)	(0.1186)	(0.1324)	(0.0744)
BM	0.2562 ***	0.2234 ***	0.2648 ***	0.0124
	(0.0876)	(0.0859)	(0.1000)	(0.0485)
BOARD	0.6382 ***	0.6343 ***	0.7530 ***	0.3340 ***
	(0.1395)	(0.1374)	(0.1605)	(0.0883)
INDEP	0.0119 ***	0.0121 ***	0.0129 ***	0.0071 ***
	(0.0035)	(0.0035)	(0.0041)	(0.0027)
FIXED	−0.2475	-0.2667	−0.2766	−0.0078
	(0.1656)	(0.1651)	(0.1915)	(0.0976)
GROWTH		−0.1051 ***
		(0.0263)
Constant	−1.8365 ***	−1.9898 ***	−2.3137 ***	−1.0112 ***
	(0.3993)	(0.4130)	(0.4791)	(0.2778)
Industry	YES	YES	YES
Year	YES	YES	YES
N	8790	8787	6418	4758
adj. R2	0.189	0.192	0.213
AR (1)				0.000
AR (2)				0.482
AR (3)				0.856
Hansen				0.522

Note: This table presents the robustness check results for the variables in this study. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively, with clustered robust standard errors in parentheses.

.

From column (1) of the table, we can see that the regression coefficient for the new independent variable TOFDI is significantly positive at the 5% level. Although the coefficient differs slightly from the original core independent variable OFDI, the regression result remains significantly positive. This suggests that the conclusion from the previous baseline regression, that OFDI significantly promotes green technological innovation, still holds true, and further confirms the robustness and reliability of the baseline regression results.

(2) Adding Control Variables

In previous studies, many scholars have focused on the enterprises themselves, noting that the promotion of green technological innovation through OFDI is not only related to the OFDI behavior but also to the operational conditions of the listed companies. Therefore, this paper adds a new control variable—the enterprise’s main business income growth rate (denoted as “GROWTH”). This variable is incorporated into the previously constructed fixed-effects Equation (1) for baseline regression, controlling for industry and year fixed effects. The regression results are shown in column (2) of Table 6.

From column (2) of the table, we can observe that the new control variable GROWTH has a significantly positive regression coefficient at the 1% level, and the regression coefficient for OFDI remains significantly positive at the 1% level. This result suggests that an increase in the business income growth rate significantly promotes green technological innovation, i.e., a company’s good operational condition benefits its green technological innovation. Therefore, as long as the main business income growth rate remains positive, OFDI continues to significantly promote green technological innovation, further validating the robustness and reliability of the baseline regression results.

(3) Lagging the Independent Variable

Based on the previous empirical research, it is known that OFDI significantly promotes green technological innovation. However, it is also possible that green technological innovation may promote OFDI, suggesting that there could be potential endogeneity issues, i.e., a bidirectional causal relationship between OFDI and green innovation. To address this, the paper lags the independent variable by one period. The core independent variable OFDI is treated as a lagged variable (denoted as “L.OFDI” in this paper), and industry and year fixed effects are still controlled. The baseline regression results are shown in column (3) of Table 6.

From column (3) of the table, we see that L.OFDI is significantly positive at the 1% level. This indicates that the endogeneity issue between OFDI and green technological innovation is not prominent, and OFDI still significantly promotes green technological innovation. This suggests that after addressing the potential endogeneity issue, the research conclusions remain valid, further confirming the robustness of the baseline regression results.

(4) GMM Dynamic Panel Analysis

Since green technological innovation in enterprises tends to exhibit persistence over time, i.e., a serial correlation issue, this paper uses the system GMM estimation to address this issue. From column (4) of Table 6, we can see that the regression coefficient for OFDI is still significantly positive at the 5% level. This shows that even after considering the serial correlation in green technological innovation, OFDI continues to significantly promote green technological innovation. This further validates the stability of the baseline regression results.

5.6. Heterogeneity Analysis

(1) Green Patent Heterogeneity Analysis

This study categorizes corporate green technological innovation into substantive green innovation and superficial green innovation. Substantive green innovation refers to the number of green invention patents (denoted as “INVPAT”) applied for by enterprises in a given year. This type of innovation fundamentally alters the nature of a product and represents a higher level of technological advancement. In contrast, superficial green innovation refers to the number of green utility model patents (denoted as “UTYPAT”) applied for by enterprises in a given year. This form of innovation involves modifications to a product’s external characteristics without altering its core functionality (Wang, X. et al., 2022) [42].

The regression results from the subsample discussion are shown in

Table 7. Green patent heterogeneity test results.

	(1)	(2)
Variable	INVPAT	UTYPAT
OFDI	0.0136 ***	0.0095 ***
	(0.0042)	(0.0032)
LEV	0.4363 ***	0.2978 ***
	(0.0878)	(0.0655)
ROE	0.4963 ***	0.3574 ***
	(0.0937)	(0.0868)
BM	0.2348 ***	0.1213 **
	(0.0778)	(0.0560)
BOARD	0.5977 ***	0.4069 ***
	(0.1372)	(0.1019)
INDEP	0.0115 ***	0.0081 ***
	(0.0031)	(0.0024)
FIXED	−0.3101 **	−0.1252
	(0.1557)	(0.1061)
Constant	−1.9430 ***	−1.2916 ***
	(0.4134)	(0.2924)
Industry	YES	YES
Year	YES	YES
N	8790	8790
adj. R2	0.191	0.175

Note: This table presents the green patent heterogeneity test results for the variables in this study. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively, with clustered robust standard errors in parentheses.

. From columns (1) and (2), we observe that the regression coefficients for OFDI are significantly positive at the 1% level, indicating that outward foreign direct investment (OFDI) promotes both substantive and surface-level green innovation. However, the regression coefficient for substantive green innovation is slightly larger than that for surface-level green innovation, suggesting that OFDI has a more significant effect on substantive green innovation. This may be because enterprises engaging in outward foreign direct investment are more focused on substantive green innovations that enhance their competitive advantage and demonstrate their technological strength, which supports long-term and stable OFDI activities. Although surface-level green innovation increases the number of green patent applications, it does not genuinely enhance the enterprise’s green innovation capabilities. Therefore, enterprises tend to prioritize substantive green innovations and focus more on applying for green invention patents.

(2) Enterprise Ownership Heterogeneity Analysis

The innovation efficiency of enterprises is related to their ownership structure. According to economic classifications, China’s enterprises can be divided into state-owned enterprises (SOEs), private enterprises, joint-stock companies, and other types. This classification is based on the nature of the asset owner. In this paper, the sample is divided into SOEs and non-SOEs to explore the differences in the impact of OFDI on green technological innovation based on ownership structure. The regression results are shown in

Table 8. Enterprise ownership heterogeneity test results.

	(1)	(2)
Variable	PAT	PAT
OFDI	0.0258 ***	0.0088
	(0.0051)	(0.0056)
LEV	0.1123	0.4997 ***
	(0.1029)	(0.1177)
ROE	0.5993 ***	0.5215 ***
	(0.1481)	(0.1080)
BM	0.4582 ***	−0.0446
	(0.0828)	(0.0848)
BOARD	0.5906 ***	0.3559 **
	(0.0880)	(0.1700)
INDEP	0.0132 ***	0.0059
	(0.0030)	(0.0042)
FIXED	−0.1201	−0.3260 *
	(0.1410)	(0.1800)
Constant	−2.1025 ***	−0.9044 *
	(0.2648)	(0.5347)
Industry	YES	YES
Year	YES	YES
N	2880	5696
adj. R2	0.322	0.136

Note: This table presents the enterprise ownership heterogeneity test results for the variables in this study. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively, with clustered robust standard errors in parentheses.

, where column (1) represents the regression results for SOEs and column (2) represents the results for non-SOEs.

From column (1), it can be seen that for SOEs, the regression coefficient of OFDI is significantly positive at the 1% level, indicating that OFDI significantly promotes green technological innovation in SOEs. From column (2), it can be seen that for non-SOEs, the regression coefficient of OFDI is not significant, suggesting that OFDI does not promote green technological innovation in non-SOEs.

This is because state-owned enterprises (SOEs) hold a unique economic position in China, serving as a barometer of government policies (Zhao, J. and Lee, J., 2020) [43]. Consequently, benefiting from ownership policy advantages and substantial government financial support, these enterprises exhibit a higher propensity for outward foreign direct investment (OFDI). Moreover, SOEs possess stronger research teams and a greater number of high-quality research projects, enabling them to rapidly absorb and adopt advanced green technologies from foreign markets (Li, Y. et al., 2024) [44]. Additionally, in terms of financing, non-SOEs face higher borrowing costs from banks and other financial institutions compared to SOEs. This financial disadvantage often leads to funding shortages, which in turn constrains R&D investment in non-SOEs.

(3) Enterprise Pollution Degree Heterogeneity Analysis

To explore whether OFDI has different effects on green technological innovation for enterprises with varying pollution levels, this paper challenges the assumption of homogeneity regarding the pollution levels of enterprises. It classifies listed companies into polluting and non-polluting enterprises, based on the “Environmental Information Disclosure Guidelines” issued by the Ministry of Ecology and Environment of the People’s Republic of China. The regression results are shown in

Table 9. Enterprise pollution degree heterogeneity test results.

	(1)	(2)
Variable	PAT	PAT
OFDI	0.0163 *	0.0154 ***
	(0.0083)	(0.0055)
LEV	0.0514	0.6296 ***
	(0.2199)	(0.1077)
ROE	0.4977 **	0.6204 ***
	(0.2503)	(0.1241)
BM	0.3283 **	0.1877 *
	(0.1624)	(0.1015)
BOARD	0.8339 ***	0.5918 ***
	(0.2597)	(0.1599)
INDEP	0.0152 **	0.0114 ***
	(0.0064)	(0.0040)
FIXED	0.2301	−0.4034 **
	(0.3297)	(0.1913)
Constant	−2.5919 ***	−1.8781 ***
	(0.6996)	(0.4899)
Industry	YES	YES
Year	YES	YES
N	1873	6917
adj. R2	0.240	0.180

Note: This table presents the enterprise pollution degree heterogeneity test results for the variables in this study. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively, with clustered robust standard errors in parentheses.

, with column (1) representing the results for polluting enterprises and column (2) representing the results for non-polluting enterprises.

From column (1), it can be seen that for polluting enterprises, the regression coefficient of OFDI is only significant at the 10% level, indicating that the effect of OFDI on green technological innovation is not very pronounced for polluting enterprises. In contrast, from column (2), it can be seen that for non-polluting enterprises, the regression coefficient of OFDI is significantly positive at the 1% level, indicating that OFDI significantly promotes green technological innovation in non-polluting enterprises. Compared to non-polluting enterprises, the impact of OFDI on promoting green innovation is weaker for polluting enterprises.

This is because operating in international markets is inherently a long-term process, requiring enterprises to continuously expand and gradually adapt to foreign markets. As a result, the costs incurred in overseas expansion are difficult to translate into immediate financial returns in the short run (Xu, L.E. et al., 2021) [45]. Compared to non-polluting enterprises, polluting enterprises engage in outward foreign direct investment (OFDI) primarily to relocate their production processes with high pollution emissions—those that are more likely to face penalties, stricter regulations, or even shutdowns from domestic environmental authorities—to overseas regions. These enterprises tend to shift such activities to less developed countries and regions that have not yet prioritized environmental protection policies (Tolliver, C. et al., 2020; Bai, Y. et al., 2020) [46,47]. As a result, polluting enterprises are more inclined to sustain their regular overseas operations rather than allocate financial resources to green innovation initiatives. Furthermore, investing in green innovation projects entails significantly higher costs and greater risks, making it a less attractive option for polluting enterprises.

6. Discussion

The empirical findings of this study confirm that outward foreign direct investment (OFDI) can effectively promote firms’ green technological innovation. This result aligns with the theoretical framework of the reverse technology spillover effect: The advanced green knowledge and technologies acquired through overseas investments are transmitted back to the home country, thereby enhancing the firm’s domestic innovation capacity. This conclusion is also consistent with prior empirical research. For example, Bai, Y. et al. (2020) [47] found that OFDI by Chinese manufacturing firms in developed countries significantly improved their environmental innovation performance. Hence, this study further supports the notion that China’s “going global” strategy serves as an important pathway for advancing green innovation. Notably, the results do not reveal any negative effects of OFDI—such as resource diversion or innovation hollowing-out—on domestic green innovation. On the contrary, international expansion appears to stimulate innovation momentum and offer new opportunities for sustainable development.

The study also uncovers the internal mechanisms through which OFDI promotes green innovation. The mediation analysis shows that alleviating financing constraints and increasing R&D investment are the two main transmission pathways. This finding is theoretically expected: Engaging in OFDI can expand firms’ access to financial resources (e.g., by tapping into overseas markets or improving creditworthiness), thus easing financial pressures and enabling greater investment in green projects (Shi, X. et al., 2023) [48]. In addition, competition in international markets compels firms to intensify R&D efforts in order to assimilate advanced foreign technologies and maintain competitive advantages. These mechanisms are consistent with both the resource-based view (RBV) and the knowledge-based view (KBV), suggesting that OFDI strengthens firms’ green innovation capacities via dual channels—capital supply and knowledge acquisition. Previous studies (e.g., Wang Xin and Wang Ying, 2021) [41] have also emphasized the importance of reducing financing constraints and boosting R&D for green innovation; this study extends that perspective into the context of OFDI, enriching the theoretical explanation framework.

Furthermore, the heterogeneity analysis provides more nuanced insights into the OFDI–green innovation relationship. This study finds that the effect of OFDI is significantly stronger for substantive green innovation (i.e., invention patents) than for incremental green innovation (i.e., utility model patents). In other words, firms’ overseas expansion primarily contributes to breakthrough innovations with high technological content, while having a limited effect on more incremental improvements. This may be because international technology exchange and competitive pressure are more conducive to radical innovation outcomes rather than merely increasing the number of patents (Shi, J. et al., 2024) [49]. This result echoes macro-level findings by Xu, L.E. et al. (2021) [45], who demonstrated that different types of green innovation have varying impacts on emission reduction, with high-quality innovations more effectively improving environmental performance. The present study confirms, at the firm level, the importance of focusing on high-level green technological innovation, offering strategic guidance for corporate innovation planning.

The findings also highlight the moderating roles of ownership and industry attributes in shaping the green effects of OFDI. Specifically, OFDI has a significant positive effect on the green innovation of state-owned enterprises (SOEs), but not on non-state-owned enterprises. At first glance, this may appear counterintuitive, since private firms are often considered more efficient innovators in competitive markets. However, this result is logically supported by the unique advantages enjoyed by Chinese SOEs. On the one hand, SOEs often operate under policy mandates and benefit from government-backed funding and preferential policies in the context of the “going global” strategy, enabling them to undertake larger-scale and more sustained overseas investments (Zhao, J and Lee, J 2020) [43]. As a result, SOEs are better positioned to leverage OFDI for resource acquisition and green technology R&D (Li, Y. et al., 2024) [44]. In contrast, non-SOEs face higher financing costs and tighter funding constraints. Even when they acquire advanced technologies abroad, they may lack the resources—financial or human—to translate those gains into substantive innovation outcomes. This finding provides a novel perspective for the literature, suggesting that ownership heterogeneity affects the innovation outcomes of OFDI, and underscores the importance of institutional environment and resource endowment.

Importantly, this study also finds industry-level differences in the green innovation benefits derived from OFDI. Compared with high-pollution industry firms, non-polluting firms demonstrate significantly greater improvements in green technologies through OFDI. This may stem from two key reasons. First, some high-pollution firms may engage in OFDI with the primary intention of relocating pollution-intensive production to jurisdictions with lax environmental regulations, thereby reducing compliance costs rather than upgrading technologies (Xie & Zhang, 2021) [3]. In such cases, overseas investments may not translate into improved green innovation capacity but rather serve as a strategy to evade domestic environmental constraints. Second, non-polluting firms typically possess pre-existing green technological advantages and stronger R&D capabilities. As they expand globally, these firms are more willing and able to leverage international resources and market demand to develop new technologies, resulting in more pronounced green innovation outcomes. These differences highlight the critical role of industry characteristics in shaping the OFDI–green innovation relationship and underscore the need to tailor policy design to the specific needs and behavioral patterns of firms across sectors.

In conclusion, the empirical evidence and mechanism analysis presented in this study illustrate the complex and multifaceted relationship between firms’ internationalization strategies and green innovation performance. On the one hand, OFDI generally creates opportunities for green technological advancement; on the other hand, its effects vary depending on firms’ internal resource conditions and external contexts. These findings not only deepen our understanding of the economic consequences of China’s “going global” strategy but also offer new evidence and perspectives for the existing literature. In the following sections, we elaborate on this study’s theoretical contributions and policy implications as well as its limitations and future research directions.

7. Conclusions, Implications, Policy Recommendations, and Limitations

7.1. Research Conclusions

Based on an empirical analysis of panel data from China’s A-share listed firms during 2007–2022, this study yields the following main conclusions:

First, outward foreign direct investment (OFDI) significantly enhances firms’ green technological innovation performance. Against the backdrop of global economic slowdown and intensifying environmental pressures, green innovation has emerged as a key pathway for reconciling economic growth with environmental sustainability. The findings of this study demonstrate that China’s “going global” strategy can effectively enhance firms’ green innovation capabilities through mechanisms such as knowledge spillovers. When engaging in OFDI, firms acquire advanced green technologies and managerial expertise from abroad, while also experiencing regulatory pressure from stricter environmental standards in host countries. These factors jointly stimulate domestic firms to accelerate their green innovation efforts in order to improve competitiveness. This conclusion is consistent with the theory of reverse technology spillovers and contributes empirical evidence to the literature on the economic effects of OFDI.

Second, the alleviation of financing constraints and the increase in R&D investment constitute the primary pathways through which OFDI promotes green innovation. The empirical results reveal that, following OFDI engagement, firms experience reduced financial constraints and expanded financing channels, enabling them to allocate more resources toward green technological R&D. Simultaneously, competitive pressures in overseas markets incentivize firms to increase R&D intensity and invest more capital and effort in the development of new technologies. These findings suggest that financial support and R&D commitment are critical links in realizing the green innovation benefits of OFDI. Lowering firms’ financing costs and improving the efficiency of their R&D input–output processes can help amplify the innovation returns from outward investment.

Third, heterogeneity analysis reveals that the impact of OFDI on green innovation varies significantly across types of innovation and types of firms. On one hand, OFDI has a stronger effect on substantive green innovation—as represented by invention patents—than on incremental or superficial innovation, suggesting that overseas investments are more likely to drive high-quality technological breakthroughs. On the other hand, the green innovation-enhancing effects of OFDI are more pronounced for state-owned enterprises and non-polluting firms, whereas the effects are relatively weaker for non-state-owned and polluting firms. These patterns imply that firm ownership structures and industry characteristics influence the degree to which firms benefit from OFDI. State-owned enterprises, which enjoy strong resource endowments and policy support, and firms in cleaner industries with solid technological foundations, are better positioned to leverage overseas investment for technological upgrading. In contrast, private firms and those in high-pollution sectors may be constrained by limited resources or divergent strategic motivations, thus failing to fully convert OFDI opportunities into innovation gains. These findings offer important insights for understanding the boundary conditions and differential effects in the OFDI–green innovation nexus.

In sum, this study confirms that corporate internationalization through OFDI contributes positively to the development of green technological innovation. It also explicates the underlying mechanisms and conditional factors that shape this relationship. These conclusions not only advance theoretical research at the intersection of international business and green innovation but also provide empirically grounded guidance for both policymakers and enterprise strategists—demonstrating both theoretical significance and practical value.

7.2. Theoretical Implications

This study makes several theoretical contributions to the growing body of literature at the intersection of green technological innovation and firm internationalization:

First, it enriches the literature on OFDI and firm-level green innovation performance. Previous research has primarily examined the relationship between outward foreign direct investment (OFDI) and green development at the macro level or focused on the environmental impacts of OFDI on host countries. However, few studies have systematically explored, from a micro-level perspective, the mechanisms through which OFDI affects green technological innovation in home-country firms. Drawing on a large-scale firm-level dataset, this study provides the first systematic empirical evidence that OFDI by Chinese firms contributes to domestic green innovation performance, thus filling a notable gap in the literature. Our findings suggest that, under the dual backdrop of the Belt and Road Initiative (BRI) and global green transformation, corporate internationalization behavior can serve as a critical driver of green technological advancement. This expands the scope of discussion within the field of international business, extending the effects of OFDI into the realm of environmental technological innovation, and builds a conceptual bridge between international investment activities and green innovation outcomes.

Second, the study identifies the key mechanisms linking OFDI and green innovation through a multidisciplinary theoretical lens. This study integrates perspectives from corporate finance and innovation management into international business research by introducing financing constraints and R&D investment as mediating variables to clarify the internal process through which OFDI affects green innovation. The results demonstrate that OFDI alleviates firms’ financing constraints and enhances their R&D efforts, which, in turn, stimulates green technological innovation. This mechanism-based finding contributes to the theoretical understanding of what drives innovation in firms. It supports the resource-based wiew (RBV) in emphasizing the role of capital access and complements the knowledge-based view (KBV) by underscoring the importance of R&D investment in innovation generation. By identifying both capital supply and technological learning as distinct but complementary channels, this study offers a more detailed theoretical account of how OFDI translates into innovation outputs, addressing a key gap in prior literature concerning the underlying mechanisms.

Third, the study expands the boundary conditions of OFDI–innovation effects by considering firm heterogeneity and innovation type. Existing theories often treat the effects of OFDI on innovation as broadly generalizable. In contrast, the heterogeneity analysis in this study shows that these effects vary significantly depending on firm characteristics and innovation types. Specifically, we find that OFDI primarily promotes high-level green innovation (invention patents) while having a limited impact on lower-level incremental innovation. Additionally, OFDI is more effective for state-owned enterprises (SOEs) and non-polluting firms, but it shows weaker effects on private firms and high-pollution enterprises. These findings contribute to boundary condition theories in the internationalization–innovation literature by showing that the presence of sufficient resources and capabilities (e.g., SOEs’ access to funding and talent) is a prerequisite for realizing the innovation benefits of OFDI. Moreover, the quality dimension of innovation (substantive vs. superficial) should not be overlooked when evaluating OFDI’s innovation outcomes. By incorporating moderating factors such as ownership structure and industrial environment, this study provides a nuanced discussion of the contexts under which OFDI is more likely to foster innovation, offering valuable directions for future research.

Finally, the study integrates the sustainability agenda into international business research, with strong interdisciplinary implications. By focusing on green technological innovation—a key issue in environmental sustainability—this study analyzes OFDI within the broader framework of sustainable development, which remains underexplored in the international business literature. The findings respond to the urgent global call for green growth and low-carbon transition, demonstrating that corporate internationalization and environmental technology advancement can interact in a mutually reinforcing way. This injects a sustainability-oriented perspective into traditional OFDI theory, broadening its scope of application. At the same time, it introduces internationalization as a novel dimension to the fields of environmental economics and innovation management, thereby contributing to cross-disciplinary integration. Against the backdrop of growing global attention to climate change and green development, the theoretical insights offered by this study are both forward-looking and highly relevant to real-world policy and strategy.

7.3. Policy Recommendations

Based on the empirical findings, this study proposes the following targeted policy recommendations regarding the role of OFDI in promoting green innovation and the heterogeneity of firms.

7.3.1. At the Enterprise Level

For enterprises, the first priority is to fully leverage OFDI opportunities to enhance their green technological innovation capabilities. In the face of increasingly stringent international environmental standards and “green trade barriers”, firms should increase investment in R&D for green products and processes. They are encouraged to actively participate in initiatives such as the China Green Standards Program and green technology cooperation projects along the Belt and Road Initiative (BRI), in order to acquire and apply advanced environmental technologies and managerial expertise. In practice, only by continuously accessing frontier knowledge and cultivating talent in green innovation can firms gain a competitive edge in the global marketplace.

Second, firms should pay close attention to strategic financing planning to ensure the sustained advancement of green innovation. Given the long cycles and high risks associated with green technology R&D, stable and sufficient financial support is especially crucial. Enterprises should make full use of national and local green finance incentives—such as issuing green bonds and applying for green credit—to ease financial constraints. A well-designed financial strategy can not only reduce the cost burden of innovation but also enhance firms’ ability to manage the risks of overseas investments.

Finally, firms with different characteristics should adopt differentiated green development strategies. Firms in high-pollution industries should seize the transformative opportunities offered by the green BRI, using OFDI to introduce cleaner production technologies and achieve process improvements and emissions reductions. In contrast, non-polluting firms should continue to capitalize on their green technological strengths, expand into international markets through the BRI, and enhance both their innovation capacity and global brand image. Overall, firms should embed green innovation into their globalization strategies, building “green competitiveness” alongside international expansion.

7.3.2. At the Government Level

Government agencies should provide comprehensive support for enterprises engaging in green innovation-oriented OFDI through policy design, financial incentives, and international cooperation mechanisms.

First, the government should adopt targeted and differentiated policy measures based on firm types and industry characteristics. For high-pollution enterprises, special subsidies, environmental risk guarantees, and other forms of support can incentivize green transformation through OFDI, assisting in technological upgrading and emission reduction. For technologically capable firms—especially SOEs and large private enterprises—the government should encourage participation in international green cooperation projects and enable them to play a demonstrative and catalytic role.

Moreover, the government should foster collaboration between SOEs and private firms by creating cooperation platforms that promote technology sharing and leverage complementary advantages to enhance overall innovation capacity.

Second, fiscal, tax, and science and technology authorities should work together to improve the policy ecosystem for incentivizing green innovation. On one hand, enterprises that actively engage in green R&D and OFDI projects could receive tax reductions, R&D subsidies, and other financial incentives to reduce innovation costs. On the other hand, expert teams and technical officials should be dispatched to provide hands-on guidance to help firms align with international environmental standards and improve production processes, thereby increasing the effectiveness and success rate of green initiatives.

Finally, the government should continue to actively advance the implementation of the green Belt and Road Initiative. This involves integrating green development themes into foreign diplomacy and economic cooperation, guiding financial institutions and private capital toward green OFDI projects. At the same time, the government should strengthen environmental risk assessment and regulatory mechanisms for outbound investment, ensuring that firms achieve both economic and environmental outcomes in their international activities and contribute meaningfully to global green development.

7.3.3. At the Financial Institution Level

Financial institutions should play a guiding role in financing to support the integration of green innovation and OFDI.

Banks and other financial institutions can innovate green financial products and expand credit offerings to support both green projects and overseas investments. For instance, they may introduce dedicated green credit programs, offer preferential interest rate loans, or establish special green development funds to reduce the financing costs associated with green technology R&D and overseas mergers and acquisitions.

In capital markets, the development of instruments such as green bonds and green asset-backed securities should be accelerated to provide enterprises with diversified financing platforms, thereby attracting more long-term capital to the green innovation domain.

Meanwhile, financial institutions can incorporate corporate environmental performance into their credit evaluation and risk management processes. Firms that demonstrate strong performance in green innovation and sustainability may receive preferential access to credit, which not only eases their financing constraints but also incentivizes others to pursue eco-innovation through market mechanisms.

Finally, regulatory bodies should guide the financial system to expand investment and financing support for green projects along the BRI, while improving disclosure requirements and risk controls to ensure that financial resources are effectively allocated to projects with demonstrable environmental benefits. In summary, the active participation of financial institutions can provide robust financial backing for green innovation, accelerate the transition toward a high-end green economy, and help financial institutions achieve strategic transformation and value creation under China’s dual carbon goals.

7.4. Limitations and Future Research

This study also has certain limitations. First, due to data availability constraints, it does not examine the moderating effect of host countries’ green institutional strength on investment direction. Second, green patent applications are used as a proxy for innovation, which may underestimate non-patented green innovation outcomes.

Future research could incorporate multi-source data and introduce variables such as host country institutional indicators and ESG (environmental, social, and governance) ratings to further explore institutional moderation mechanisms and multi-actor pathways for collaborative green innovation.