Navigating uncertainty: How global strategies can drive stakeholders’ investment in green innovation?

The lack of confidence generated by uncertainty is increasingly leading global firms to abandon clean technology projects, threatening both sustainability goals and long-term competitiveness. Recent policy initiatives illustrate the tangible impact of institutional frameworks on renewable energy investments. For example, the Inflation Reduction Act in the United States significantly influenced renewable energy investment flows in global firms, demonstrating how regulatory environments can either boost or constrain green initiatives (White House, 2022). Similarly, regulatory uncertainty in Germany delayed Tesla’s Gigafactory expansion in 2023, generating substantial stakeholder hesitancy (Hancock, 2023). Industry experts have repeatedly emphasized the need for predictability in green investments, noting that uncertainty remains a key obstacle to attracting and sustaining long-term commitments (Hancock, 2023).

Failure to address rising uncertainty in green innovation can have direct consequences for firms, including the abandonment of critical sustainability projects. As regulatory pressures and stakeholders’ demands for transparency intensify, firms that lack effective strategies to manage uncertainty risk losing both competitive advantage and stakeholder trust, thereby undermining their long-term financial and environmental performance.

Green innovation is particularly exposed to uncertainty relative to other types of innovation and investment (e.g. Wang and Bansal, 2012). This vulnerability stems from the paradoxical tension between the long-term benefits of superior green strategies and the need for substantial upfront investments in risky technologies (Berrone et al., 2013; Leyva-de la Hiz et al., 2019b; Pan et al., 2021), which heightens stakeholders’ concerns. Although prior research recognizes the importance of reducing uncertainty to foster green investments (Cheng et al., 2025; Leyva-de la Hiz et al., 2019a; Rubio-Andrés et al., 2023), the risk and uncertainty dimensions of international business models remain insufficiently addressed (Guaita Martínez et al., 2024).

Existing environmental management studies have largely proposed isolated stakeholder-specific mechanisms to mitigate uncertainty, such as employee training, knowledge development and participation in governance mechanisms (Awwad et al., 2025; Khan et al., 2025). However, these approaches offer only partial solutions for encouraging firm-specific investments (FSIs) because they (1) address individual stakeholder groups in isolation, (2) overlook the firm’s international position and (3) fail to distinguish between different types of uncertainty (Hoskisson et al., 2018; Poppo et al., 2016). More fundamentally, treating uncertainty as a single construct conflates distinct governance problems and prevents theorizing about the mechanisms through which governance can credibly support stakeholder commitments; a more fine-grained conceptualization is therefore needed to disentangle the specific sources of uncertainty that shape stakeholders’ investment decisions. In this sense, behavioral uncertainty reflects stakeholders’ concerns about firms’ actions after investments are made, whereas environmental uncertainty relates to doubts about the value of the investments themselves (Hashimoto, 1981; Wang and Barney, 2006). This raises a critical question: which governance devices are best suited to address each type of uncertainty and thereby stimulate green investments in global firms?

Drawing on paradox theory, this study integrates the natural resource-based view (NRBV) and stakeholder theory to propose two governance devices – external monitoring and takeover protection – that directly address uncertainty and enhance green innovation. External monitoring reduces behavioral uncertainty by providing third-party oversight that strengthens stakeholder trust, while takeover protection mitigates environmental uncertainty by shielding firms from takeover threats and enabling long-term green strategies (Amore and Bennedsen, 2016; Rubio-Andrés et al., 2023; Wang et al., 2016; Zhang et al., 2022).

Finally, adopting a co-evolutionary perspective on internationalization, we examine how firms’ international expansion shapes the effectiveness of these devices (Gupta et al., 2012). Internationalization intensifies paradoxical tensions by simultaneously increasing managerial complexity and access to diverse resources (Ahmadova et al., 2023; Bueno-García and Ahmadova, 2024; Zhang et al., 2024), thereby affecting stakeholders’ perceptions of success and failure (Barber et al., 2019; Birkinshaw et al., 2016) and their willingness to engage in firm-specific investments (Tenorio et al., 2021). While greater internationalization can amplify behavioral uncertainty due to increased monitoring challenges, it may also reduce environmental uncertainty by raising the cost of takeover actions (Fang et al., 2019; Fernández-Olmos et al., 2016; Goyer and Del Real, 2014; Lee et al., 2014). Consequently, the effectiveness of external monitoring and takeover protection depends critically on firms’ internationalization.

Overall, this study offers a global strategy perspective on how firms can align governance devices with distinct sources of uncertainty. For managers, the findings provide actionable guidance on designing governance mechanisms that sustain stakeholder confidence and support green innovation in complex international contexts.

Green innovation plays a central role in firms’ environmental performance and long-term competitiveness, as it requires substantial investments in advanced and often risky technologies (e.g. Berrone et al., 2013; Pan et al., 2021).

According to the NRBV, competitive environmental strategies stem from unique internal resources that are difficult to imitate. In this context, firms’ ability to develop green innovation depends critically on stakeholders’ willingness to commit firm-specific investments (FSIs), which constitute a key source of competitive advantage (Barney, 1991; Wang and Barney, 2006). Firms therefore differ in their success in green innovation to the extent that they can engage stakeholders in long-term firm-specific green investments (e.g. Cheng et al., 2024; Li et al., 2024; Takalo and Tooranloo, 2021).

However, stakeholder FSIs in green innovation are particularly exposed to persistent uncertainty. Green innovation generates a fundamental paradox: while long-term environmental strategies promise substantial future benefits, they require immediate and risky investments in specialized resources, which heighten stakeholders’ concerns (Cheng et al., 2025; Rubio-Andrés et al., 2023). As a result, uncertainty becomes a central barrier preventing stakeholders from committing to green FSIs.

This uncertainty manifests in two interrelated forms. Behavioral uncertainty refers to stakeholders’ concerns about firms’ actions after investments are made, reflecting tensions between trust and control (Poppo et al., 2016; Schepker et al., 2014; Wang and Barney, 2006). Environmental uncertainty relates to doubts about the future value of green investments, reflecting tensions between long-term strategic commitment and volatile external conditions (Hashimoto, 1981; Wang and Barney, 2006). Rather than operating independently, these uncertainties tend to coexist and reinforce each other over time, jointly reducing stakeholders’ willingness to engage in FSIs.

Despite broad agreement on the importance of reducing uncertainty to foster green innovation, existing research has primarily focused on stakeholder-specific mechanisms, such as transparency-enhancing actions, employee-focused practices or governance-related interventions (e.g. Awwad et al., 2025; Brush et al., 2012; Jabbour et al., 2010). These approaches offer only partial solutions, as they address uncertainty in isolation, overlook firms’ international context and fail to distinguish between different types of stakeholder uncertainty (Hoskisson et al., 2018; Poppo et al., 2016).

From a global strategy perspective, internationalization further complicates how uncertainty is experienced and managed. Operating across multiple countries increases complexity and monitoring challenges, potentially amplifying behavioral uncertainty (Fang et al., 2019; Lee et al., 2014). At the same time, internationalization can mitigate environmental uncertainty by diversifying risk and raising the cost of takeover threats (Fernández-Olmos et al., 2016; Goyer and Del Real, 2014). Consequently, the effectiveness of governance devices aimed at reducing uncertainty is likely to depend on firms’ degree of internationalization.

For these reasons, integrating the NRBV and stakeholder perspectives within a paradox theory framework provides a more comprehensive explanation of why uncertainty remains a central barrier to stakeholders’ FSIs in green innovation. In the next section, we develop this theoretical lens to examine organizational devices and global strategies that enable firms to manage – rather than eliminate – uncertainty over time and across international contexts.

While the NRBV and stakeholder theory provide a solid foundation to explain why green innovation relies on stakeholders’ FSIs, they do not explicitly theorize the nature of uncertainty that discourages such investments. In this respect, paradox theory offers a complementary and necessary lens to conceptualize uncertainty not as a temporary condition, but as a persistent organizational tension inherent to green innovation, particularly in global firms.

Paradox theory defines uncertainty as the coexistence of competing yet interrelated demands that cannot be resolved through simple trade-offs (Hahn and Knight, 2021; Lewis and Smith, 2022). Green innovation epitomizes such paradoxical conditions: it requires long-term, irreversible investments while operating in volatile regulatory, technological and market environments; it promises collective environmental benefits while exposing individual stakeholders to firm-specific risks. As a result, uncertainty does not disappear once an investment decision is made but rather unfolds dynamically over time.

Hahn and Knight (2021) conceptualize paradoxes as evolving through three interrelated stages: latency, saliency and persistency. Applied to green innovation, uncertainty is initially latent when stakeholders consider whether to engage in firm-specific investments in environmentally oriented projects. Once stakeholders commit resources, uncertainty becomes salient, as concerns regarding managerial opportunism (behavioral uncertainty) and the potential devaluation of investments (environmental uncertainty) materialize. When green investments are deployed across international markets, these tensions become persistent, repeatedly resurfacing as firms adapt to heterogeneous institutional, regulatory and competitive environments.

From this perspective, the behavioral and environmental uncertainties identified in prior research are not isolated risks, but interrelated manifestations of a broader paradox of green investment. Behavioral uncertainty reflects tensions between trust and control in stakeholder–firm relationships, while environmental uncertainty reflects tensions between long-term strategic commitment and short-term environmental volatility. Importantly, these tensions are amplified rather than resolved in global contexts, where geographic dispersion, institutional diversity and cross-border coordination increase the complexity of managing stakeholder expectations.

Lewis and Smith (2022) further argue that managerial responses to paradoxical tensions often fall into “either/or” thinking, leading to defensive reactions and short-term compromises that ultimately intensify uncertainty. In the context of green innovation, firms that prioritize either flexibility or commitment, either global efficiency or local adaptation, may inadvertently reinforce stakeholders’ reluctance to engage in FSIs. In contrast, a “both/and” approach enables firms to simultaneously pursue control and trust, stability and change, and global integration and local responsiveness, thereby reducing the persistence of uncertainty over time.

Consequently, we identify two key devices that directly address these paradoxical tensions by targeting two distinct kinds of uncertainty – behavioral and environmental – and thereby jointly stimulate stakeholders’ firm-specific investments in green innovation: external monitoring and takeover protection. We further examine how a firm’s level of internationalization differentially shapes these uncertainties and moderates these relationships, as greater international scope intensifies paradoxical tensions and requires a global “both/and” logic consistent with a co-evolutionary strategy.

External monitoring involves third-party oversight of corporate activities, ensuring that managerial actions align with a firm’s strategic goals and fostering stakeholder trust (e.g. Hoskisson et al., 2018). In this sense, external monitoring is commonly materialized through the actions of outside auditors who directly scrutinize a specific kind of firm activity (e.g. Gul and Ng, 2018). Thus, external monitoring ensures that stakeholder investments are properly allocated, preventing mismanagement and building trust in the firm’s green initiatives.

External monitoring compels firms to strengthen their environmental strategies as stakeholders increasingly demand transparency and accountability. The implementation of external monitoring will have additional positive effects in reducing behavioral uncertainty as, for example, it will increase firms’ transparency in providing information about their environmental activities (e.g. He et al., 2023; Keresztúri et al., 2025), information that gives stakeholders greater confidence in performing FSIs for green improvements. External monitoring protects stakeholders’ investments against opportunistic behaviors from managers and thus provides confidence to stakeholders to invest in long-term green innovation projects, which can enhance firm’s global competitiveness.

In sum, the more firms implement external monitoring, the more they can attract stakeholder investments in green innovation by reducing behavioral uncertainty. We thus hypothesize:

Meanwhile, takeover protection refers to the defensive measures implemented by companies to detect hostile takeovers or unsolicited acquisition attempts (e.g. Hoskisson et al., 2018). These measures, such as poison pills, golden parachutes, staggered boards and other tactics, are employed by a company’s management or board of directors to prevent potential acquirers from gaining control of the company without their approval (e.g. Karpoff and Wittry, 2018). Takeover protection ensures long-term corporate stability, safeguarding against hostile takeovers and preserving the firm’s focus on sustainable innovation (Karpoff and Wittry, 2018; Wang et al., 2016).

Thus, implementing takeover protection devices will be the best option to reduce the so-called environmental uncertainty by preventing sudden and hostile acquisitions that may devaluate FSIs. Environmental uncertainty appears before FSIs are made and concerns the value of the investments themselves (Hashimoto, 1981; Wang and Barney, 2006) since unexpected changes or unpredictable events in the environment may damage or reduce the value of stakeholders’ investments. Therefore, firms have to implement devices to protect stakeholder FSIs against such events to avoid this uncertainty. In particular, takeover acquisitions constitute a crucial source of environmental uncertainty because they usually imply aggressive cost reductions by incoming managers (Wang et al., 2016).

Once the company has acted against hostile takeovers and is no longer under threat, stakeholders will feel more confident about making long-term investments in green innovation to enhance its sustainability. The literature supports the idea that stakeholders’ investments and influence are relevant to ensuring that firms develop green innovations (Rubio-Andrés et al., 2023). In contrast, a lack of takeover protection can leave a company vulnerable to acquisition by investors who may not prioritize the company’s environmental goals, potentially leading to a reduction in green innovation efforts (Amore and Bennedsen, 2016). This is because green innovation requires long-term strategies, for which corporate governance establishes the necessary guidelines (Hristov and Searcy, 2025). In those companies that are able to establish mechanisms against takeovers, the executives can focus on more long-term strategies without having to worry about defending against external threats, can maximize shareholder value in the short term and can immediately address the needs of the stakeholders (Amore and Bennedsen, 2016; Rubio-Andrés et al., 2023; Zhang et al., 2022). Thus, those companies that put in place measures against hostile takeovers will encourage stakeholders to make FSIs in green innovation due to trust in broader corporate insights regarding sustainability that maximize their long-term value.

Takeover protection strategies not only safeguard firms against external threats but also enable them to focus on long-term green innovation projects, which can enhance their global competitiveness. In sum, when firms implement more takeover protection devices, they will incentivize more stakeholder FSIs to improve green innovation as takeover protection will reduce environmental uncertainty. We thus hypothesize:

Paradoxical tensions in green innovation increase in the extent to which firms internationalize, repeatedly resurfacing as firms adapt to heterogeneous institutional, regulatory and competitive environments (e.g. Ahmadova et al., 2023; Bueno-García and Ahmadova, 2024; Zhang et al., 2024).

Indeed, internationalization should not be understood merely as geographic expansion, but as a dynamic process through which firms continuously realign global integration and local adaptation across heterogeneous markets and stakeholder environments. This concept is co-evolution, which refers to the ability of multinational firms to adapt their strategies, organizational structures, and stakeholder relationships in response to ongoing changes in markets, regulations and societal expectations, while simultaneously shaping those environments through strategic action (Gupta et al., 2012). In the context of green innovation, co-evolution becomes particularly critical because environmental technologies face increased persistent uncertainty arising from regulatory volatility, technological disruption and heterogeneous stakeholder demands across countries.

However, specific co-evolution adaptations can succeed or fail in global expansion, making that internationalization may have different effects on the types of uncertainty depending on how international co-evolution works. When firms go global, salient phase uncertainty – both behavioral and environmental – becomes persistent due to increased paradoxical tensions, tensions that repeatedly resurface as firms try to adapt to multiple international scenarios and that must be continuously managed by global firms (Gupta et al., 2012). Consequently, depending on how firms succeed or not in adapting such activities and strategies to manage different tensions and complexities from operating at more different contexts (Barber et al., 2019; Birkinshaw et al., 2016), such continuous global–local realignment will increase or mitigate different uncertainties for stakeholders. Specifically, we extend the past management literature that highlighted that global expansion can either strengthen or complicate a firm’s environmental strategies (e.g. Ahmadova et al., 2023; Bueno-García and Ahmadova, 2024; Zhang et al., 2024), depending on how it influences stakeholders’ perceptions, and develop the proposition that firm internationalization, when enacted through a co-evolutionary global strategy, will have different impacts on the two types of uncertainty – increasing or reducing uncertainty or doubts among stakeholders about participating in or trusting the company’s environmental commitments – and thus boosting or mitigating the action of the firm’s devices for green innovation.

In general, as firms internationalize, behavioral uncertainty rises, driven by the involvement of diverse managers from multiple countries, increasing the risk of opportunistic behavior. Companies that are internationalized in more countries are more able to rely on more different managers, with more different levels of corruption and political instability and more diffuse regulatory frameworks (Cuervo-Cazurra et al., 2018). When internationalization is approached as a dynamic global adaptation process, firms must continuously realign internal governance mechanisms with increasingly diverse managerial, regulatory and cultural contexts. Failure to co-evolve such mechanisms with global expansion amplifies behavioral uncertainty, as monitoring becomes more complex and opportunistic behavior more difficult to detect (Gupta et al., 2012). Hence, behavioral uncertainty will be higher in firms operating in more global markets, which may lead to fewer FSIs as stakeholders will be less confident about potential opportunistic behavior when there are more managers.

Consequently, companies that are highly internationalized and are operating in many countries can weaken the positive effect of external monitoring on green innovation due to this effect of a firm’s internationalization increasing behavioral uncertainty. Indeed, global operations add complexity and costs to monitoring, making it harder to ensure that environmental strategies are applied consistently across diverse regions (e.g. Lee et al., 2014). As these more complex activities of more global firms will depend on more managers abroad (Cuervo-Cazurra et al., 2018; Tenorio et al., 2021), internationalization may create inconsistencies in green strategies (e.g. Ahmadova et al., 2023; Bueno-García and Ahmadova, 2024; Zhang et al., 2024) due to the greater difficulty in monitoring the diverse managers to ensure that they act in line with the green strategy.

In sum, firms with a higher level of internationalization will experience more behavioral uncertainty because they operate with more managers, increasing the possibility of opportunistic behavior. Accordingly, when firms expand internationally, there is a higher difficulty of a parallel co-evolution of their monitoring and coordination capabilities, and thus the positive role of external monitoring in fostering green innovation is expected to diminish as international scope increases. Thus, we hypothesize:

A higher level of a firm’s internationalization will reduce environmental uncertainty. When internationalization unfolds as a dynamic global adaptation process, firms continuously recalibrate their strategic, organizational and governance arrangements in response to evolving global and local conditions. Such global–local–global adaptation stabilizes long-term investment conditions and reduces exposure to disruptive environmental shocks, including hostile takeovers (Gupta et al., 2012). In general, by operating in multiple countries, firms can diversify their investments, spreading risk and reducing vulnerability to environmental changes (e.g. Ding et al., 2025; Hu et al., 2025). Consequently, since a more diversified firm’s investment portfolio reduces the risk of investment depreciation from environmental changes (Del Bosco and Misani, 2016; Goyer and Del Real, 2014) and decreases stakeholders’ exposure to environmental uncertainty, thereby incentivizing FSIs (Wang and Barney, 2006), just as internationalization may lead to a more diversified firm investment portfolio, it will contribute substantially to reducing environmental uncertainty.

When internationalization unfolds as a co-evolutionary process, firms do not simply diversify their operations across countries but continuously recalibrate their strategic, organizational and governance arrangements in response to evolving global and local conditions. Greater global integration enables deeper localization, which in turn identifies unique local market needs that invite more extensive cooperative relationships with the home country, provide access to technological enablers and lead to yet deeper localization (Wakayama et al., 2012). This global–local–global adaptation stabilizes long-term expectations regarding firm continuity, strategic control and investment horizons, thereby reducing the persistency of environmental uncertainty faced by stakeholders (Gupta et al., 2012). In this context, co-evolution counteracts stakeholders’ concerns that long-term green investments may be devalued by abrupt environmental shocks, such as hostile takeovers or sudden strategic reorientations, which are particularly damaging in green technologies characterized by high sunk costs and long development cycles.

To summarize, firms with a higher level of internationalization enacted through a co-evolutionary global strategy experience lower environmental uncertainty, as the continuous alignment between global integration and local adaptation stabilizes long-term investment conditions. This stability enhances the effectiveness of takeover protection by safeguarding long-term global adaptation commitments against sudden ownership changes, thereby reinforcing stakeholders’ confidence in long-term green projects. Consequently, the positive effect of takeover protection on green innovation is expected to be stronger in firms whose internationalization follows a sustained global–local alignment logic. Therefore, we hypothesize:

To test our research hypotheses, we chose the 500 firms in the S&P500 Index as a representative sample of US multinational firms, given their significant global presence and contribution of over 80% of US equity market capitalization. In particular, these firms operate in 11 different economic sectors, and we analyzed data from 2008 to 2019, a period that captures the key trends in corporate sustainability and green innovation strategies during significant global economic shifts. We collected all the information from the Thompson Reuters Eikon database, delving into the environmental, social and governance (ESG) section and the international sales reports for each firm.

Using these data, we built an unbalanced panel containing 2,181 observations from the 243 different S&P500 firms in our original sample. To ensure the robustness of our results, we conducted a T-test to confirm that there was no significant bias in the final sample (Rogelberg and Stanton, 2007). The T-test revealed no significant difference between the two means (53.497 and 54.506 of green innovation with a p-value of 0.159), providing evidence that the missing data did not lead to a bias in our analysis regarding firms’ level of green innovation.

Measuring our dependent variable required us to capture firms’ investments in developing new products and processes (e.g. Berrone et al., 2013; Pan et al., 2021). We used the Environmental Innovation Score from Thomson Reuters Eikon, a widely recognized index that reflects a firm’s ability to minimize its environmental impact and develop eco-friendly technologies (e.g. Ahmadova et al., 2023) because it indicates “a company’s capacity to reduce environmental costs and burdens for its customers, and thereby creating new market opportunities through new environmental technologies and processes or eco-designed products” (Refinitiv, 2020, p. 22). This index is comprised of 23 subdimensions related to a firm’s actions and investments destined to improve or develop new products and processes to reduce its environmental impact. The values of our dependent variable ranged between 0 and 100, with higher values implying that the company is better at green innovation.

The literature has emphasized the adoption of external rather than internal monitoring to reduce stakeholder uncertainty (e.g. Gul and Ng, 2018). We measured external monitoring using the CSR Reporting External Audit Score from Thomson Reuters Eikon, which assesses the robustness of third-party audits in ensuring the transparency and accuracy of sustainability reporting based on the question “Does the company have an external auditor of its CSR/HandS/sustainability report?” with the following items: (1) “In scope, are the data on external audits of the company’s CSR data or extra financial report considered?”; (2) “Do the data consider an audit in the form of a review conducted by a university, academic, expert, external panel, or a research center?”; (3) “Does the company have web-based CSR reports that are externally audited?”; and (4) “Does the company have an integrated annual report with external audit statements for its environmental and social data?” The values of this independent variable ranged between 0 and 100, with higher values implying that the company is taking more of these actions to implement more rigorous external monitoring.

Takeover protection was measured using the total number of anti-takeover devices implemented by each firm, as reported in the Eikon database, with values ranging from 0 to 12. Some examples of anti-takeover devices implemented by firms are “limitation of director liability,” “limited shareholder rights to call meetings,” “supermajority vote requirement” or “staggered board structure.” The firms in our sampled data reported a minimum of one anti-takeover device and a maximum of 11 anti-takeover devices, so this variable is a continuous measure for which higher values imply that the firm is implementing more devices to ensure higher takeover protection.

We measured internationalization as the percentage of revenue generated from foreign sales, a widely used proxy in management research to capture a firm’s global footprint. Past management studies have agreed that a good proxy to capture this phenomenon is the ratio of the revenue from foreign sales divided by the firm’s total revenue (e.g. Bueno-García and Ahmadova, 2024). Therefore, we downloaded each firm’s portfolio of revenue from sales abroad to aggregate all foreign revenue and then divided it by the revenue from the firm’s total sales to obtain a continuous variable consisting of the percentage of a firm’s revenue from foreign sales. As such, we followed the past literature to capture the level of a firm’s internationalization with the percentage of foreign revenue for each firm per year.

We included control variables such as firm size, age and CO₂ emissions to isolate the effect of external monitoring, takeover protection and internationalization on green innovation. We included a firm’s return on equity (ROE) to control for its financial situation in a similar way to other management studies (e.g. Cheng et al., 2024) since firms with better financial results are more able to invest in green strategies. We used the natural logarithm of the firm’s total assets to capture firm size (e.g. Bueno-García and Ahmadova, 2024) because larger firms have more resources available to improve their green innovation. We introduced firm age as the natural logarithm of the total years since a firm’s foundation (e.g. Leyva-de la Hiz et al., 2019b) as older firms are more prone to care about social requirements. Finally, we controlled for firm emissions measured using the natural logarithm of a firm’s total CO₂ emissions in tons because some firms with better environmental performance do not have an extra incentive to invest in green innovation to improve environmental behavior that is already high.

In addition, we included categorical variables to control for two different effects. First, we followed environmental studies (e.g. Ahmadova et al., 2023; Bueno-García and Ahmadova, 2024) to include economic sector dummies using GICS sectors (i.e. industrials, communication services, consumer discretionary, consumer staples, financials, energy, healthcare, information technology, materials, real estate and utilities) with a variable labeled firm industry, thus taking into account firms’ different economic sectors and the industry’s potential effect on green innovation levels. Second, we controlled for the number of different economic areas abroad where the firm was operating by including regional dummies. This variable ranged between 1 and 4 if the firm was simultaneously operating in one or more areas of the four global markets considered in the international business literature (e.g. Bueno-García and Ahmadova, 2024): the Americas, Europe, Asia and the Pacific and Africa. With this set of control variables, we considered different firm characteristics to analyze accurately the potential effect of our independent variables on firms’ green innovation.

We employed panel regression models using STATA 16 to test our hypotheses, ensuring robust control for time- and firm-specific effects. First, we tested whether the use of panel regressions is appropriate for our data by running the Breusch–Pagan Lagrange multiplier test (Breusch and Pagan, 1980), which revealed that our data indeed behave as a panel distribution (Breusch–Pagan Lagrange multiplier test: prob. > chibar2 = 0.000). Second, we decided to implement a random-effect (RE) model instead of a fixed-effect (FE) model due to the presence of industry dummies in all models, which is allowed in the RE but not in the FE model. The RE model also allows better control of variance components in errors and times, assuming the same slopes and intercepts. However, similar results were obtained with both the RE and the FE model; these can be obtained from the authors upon request. Third, we introduced robust standard errors to avoid heteroscedasticity and serial correlation biases in errors, and we included year dummies to control for temporal effects. Finally, we lagged all our predictor variables since they do not exert an appreciable effect on firms’ environmental strategy until at least a year has passed, so we sampled our dependent variable green innovation for the period 2008 to 2019 and our predictor variables between 2007 and 2018. We did not record data for the years after 2019 because the COVID-19 situation may disturb the robustness of this information.

Table 1 presents the descriptive statistics and Pearson correlations for all the variables in the model, offering an overview of their relationships. In addition, we checked that the variance inflation factors (VIFs) are within acceptable limits, obtaining a set of values ranged between 1.00 and 1.46 with a mean of 1.15, confirming that multicollinearity is not a concern in our analysis.

Table 2 presents our random-effect models used to test the research hypotheses. Model 1 highlights the significant influence of several control variables on green innovation, confirming their role as strong predictors. Model 2 was used to test Hypothesis 1, regarding the positive effect of external monitoring on a firm’s green innovation, and Hypothesis 2, which predicted a positive effect of takeover protection devices on a firm’s green innovation. Although external monitoring showed a positive coefficient, the effect on green innovation was not statistically significant (p-value = 0.306), so we did not find support for Hypothesis 1, which predicted that firms will improve their green innovation levels when they increase the implementation of external monitoring. In Model 2, takeover protection had a positive and significant impact on green innovation (p-value = 0.038), supporting Hypothesis 2.

Model 3 was used to test Hypothesis 3 regarding the moderating effect of a firm’s internationalization on the relationship between external monitoring and green innovation. We found significant support for Hypothesis 3 as the interaction between external monitoring and internationalization negatively affected green innovation (β = −0.004, p-value = 0.035), thus providing statistical support in the first step. To check whether the moderating effect was as predicted, we depict this relationship graphically in Figure 1, which indicates that high levels of firm internationalization weaken the relationship between external monitoring and green innovation. In fact, Figure 1 shows that, as firms become more internationalized, the relationship between external monitoring and green innovation turns negative, indicating lower innovation levels. In this sense, we also see a positive relationship between external monitoring and green innovation if the firms’ internationalization level is low; thus, external monitoring will strengthen green innovation levels if the firm operates with a more national scope, providing support for Hypothesis 3.

Finally, we used Model 4 to test Hypothesis 4 regarding the moderating effect of a firm’s internationalization on the relationship between takeover protection and green innovation levels. We found statistical support for this hypothesis with this model as the interaction coefficient is significant (β = 0.040, p-value = 0.002). Figure 2 shows that the predicted effect in Hypothesis 4 is supported, indicating that higher levels of internationalization enhance the positive impact of takeover protection on green innovation. Hence, when firms implement more takeover protection devices, they will present even higher levels of green innovation if their internationalization level is high, as predicted in Hypothesis 4; moreover, this effect will be mitigated if the firms’ internationalization level is low and thus more takeover protection devices will not have a real impact on green innovation if the firm operates with a more national scope.

In addition, we performed an F-test to check that the explanatory power of the variables introduced in each model had a significant increase and thus that adding our independent and moderating variables elevates the explanatory power of our models. The Wald chi-square test showed significant improvements in explanatory power between models, confirming the robustness of our results: 10.68 (p-value < 0.05) between model 1 and model 2, 4.43 (p-value < 0.05) between model 2 and model 3, and 9.87 (p-value < 0.01) between model 2 and model 4. Together, these results suggest that the inclusion of internationalization as a moderating variable significantly increases the explanatory power of our models, as evidenced by the F-test results.

Our results extend the current knowledge in several key aspects. Recent works explore different factors with the potential to reduce uncertainty for a group of stakeholders, such as implementing digitalization tools (e.g. Chen et al., 2024), innovation platforms’ relational governance (e.g. Sun et al., 2025), employee training (e.g. Awwad et al., 2025) or different governance mechanisms (e.g. Khan et al., 2025) among others, due to indeed uncertainty reduction stimulates stakeholders’ engagement in green innovation (e.g. Cheng et al., 2025; Rubio-Andrés et al., 2023) through stakeholders’ FSIs. However, these mechanisms act as partial solutions for a group of stakeholders, neglecting their effectiveness for different types of uncertainty as well as the relevance of the firm’s international position. Hence, a lack of consensus still remains regarding specific strategies for successfully managing uncertainty in international business models (e.g. Guaita Martínez et al., 2024) considering solutions for all stakeholders, but also considering that different types of uncertainty do exist with different impact on firms’ green innovation (e.g. Khan et al., 2025). We extend the current research by showing how specific devices are effective to boost green innovation since they handle the two key kinds of uncertainty to generate confidence for all stakeholders in global firms: external monitoring, as the best device to reduce behavioral uncertainty and takeover protection to reduce environmental uncertainty, and thus they have a positive impact on firm’s green innovation.

Specifically, our findings show that when firms implement more takeover protection devices, stakeholders will be less worried about the specific environmental uncertainty and more confident about investing in green innovative processes within the firm. This result is in line with recent literature which points out that the protection of firm’s value – such as takeover protection – is essential to promote stakeholders’ confidence and their engagement in firms’ key processes (e.g. Bao et al., 2025; Park et al., 2025) and thus enable firms to develop strategic outcomes to overcome competitors. While external monitoring did not show a significant overall effect on green innovation, it proved effective in firms with lower levels of internationalization. We are in line with previous research that suggested that different ways of monitoring may generate confidence among all stakeholders (e.g. Bello-Pintado et al., 2023; He et al., 2023), for example mechanisms to control green washing (Keresztúri et al., 2025), showing a positive effect but only within less internationalized firms, and thus this device will be more effective in boosting green innovation in national firms.

In this sense, we again point out the relevance of the international position of the firm to extend past literature and understand the effectiveness of the proposed devices to handle different types of uncertainty. The international business literature agrees on the notion that a firm’s internationalization is a key process with a strong link with uncertainty (e.g. Chetty et al., 2024; Ding et al., 2025; Hu et al., 2025), a fact which specially impacts on the firm’s environmental management (e.g. Ahmadova et al., 2023; Li et al., 2024; Ma and Sambharya, 2024) and thus influences stakeholders’ decisions to engage in FSIs to boost green strategies. However, these works do not consider the impact of the firm’s international position on different types of uncertainty and thus how firm internationalization has different implications for the effectiveness of specific devices to handle stakeholders’ uncertainty. Specifically, our results reveal that internationalization weakens the effectiveness of external monitoring in promoting green innovation due to increased managerial complexity and potential opportunistic behavior owing to increased behavioral uncertainty. In contrast, the effectiveness of takeover protection to boost FSIs for green innovation will be higher in more global firms as it can mitigate the risk of sudden takeover actions since more international firms require higher investments from potential acquirers to purchase them.

Hence, these findings provide deeper understanding of existing international business research about why firm’s internationalization may entail mixed effects on firm’s green strategy (e.g. Ahmadova et al., 2023; Bueno-García and Ahmadova, 2024; Zhang et al., 2024) due to increased complexity in handling activities abroad but, in parallel, internationalizing firm’s activities may entail an increase of the firm value and keep it from market changes. In this sense, our findings show how the firm’s internationalization differently impacts on behavioral and environmental uncertainty for all stakeholders and consequently on the effectiveness of specific devices implemented to boost green innovation.

This work provides several theoretical contributions. First, it expands the NRBV (Hart, 1995; Russo and Fouts, 1997) and stakeholder literature (Barney, 1991; Wang and Barney, 2006). Uncertainty clearly shows the coexistence of competing yet interrelated demands (Hahn and Knight, 2021; Lewis and Smith, 2022) even more in green strategy (e.g. Cheng et al., 2025; Rubio-Andrés et al., 2023) which need to be resolved through devices which lead to the development of strategic resources – NRBV through indeed collective commitment and participation from stakeholders – stakeholders FSI. In particular, we showed that uncertainty must be differentiated into behavioral uncertainty (Poppo et al., 2016; Schepker et al., 2014; Wang and Barney, 2006) and environmental uncertainty (Hashimoto, 1981; Wang and Barney, 2006) to be dealt with successfully as a whole construct and hence to design specific devices, mechanisms, strategies and/or firm policies appropriately. In this sense, this work points out the relevance of dealing both uncertainties as interrelated manifestations of a broader paradox, especially for green investment: Building on our theoretical framework, we differentiate uncertainty into behavioral and environmental dimensions, previously developed in Sections 2 and 3, and show that distinct governance devices address each dimension differently. Rather than viewing uncertainty as a temporary or isolated constraint, the paradox lens allows us to conceptualize uncertainty in green innovation as a persistent organizational tension arising from the coexistence of long-term sustainability commitments and stakeholders’ concerns about risk, control and value appropriation. Our results support this view by showing that governance devices do not eliminate uncertainty but instead address different dimensions of it: external monitoring primarily mitigates behavioral uncertainty, whereas takeover protection alleviates environmental uncertainty.

Second, we enrich past environmental management literature which argued that green innovation is particularly affected by uncertainty due to requiring long-term investments in riskier technologies (e.g. Cheng et al., 2025; Rubio-Andrés et al., 2023) that, however, may entail higher long-term returns by positioning the firm as a leader in sustainability (e.g. Berrone et al., 2013; Leyva-de la Hiz et al., 2019b; Pan et al., 2021), extending the conceptualization of green innovation by highlighting this paradoxical nature. Thus, it requires the provision of specific devices to deal with uncertainty across its whole process to successfully manage its paradoxical nature. Indeed, we argue that green innovation follows the three stages of uncertainty –latent, salient and persistent (Hahn and Knight, 2021) – being initially latent when stakeholders consider whether to invest in green projects; salient once investment is performed based on the managerial opportunism (behavioral uncertainty) and the potential devaluation of investments (environmental uncertainty); and persistent when green investments are deployed across international markets. Our work highlights the relevance of a “both/and” approach against “either/or” thinking from managers to face these uncertainties and succeed in boosting green innovation. In sum, this work contributes by shedding light regarding this process to enable firms the development of differential resources for a successfully environmental management, proposing specific devices to deal with uncertainty and thus foster stakeholders FSI.

Moreover, our work provides deeper understanding of existing international business research which showed mixed effects of internationalization on firms’ green strategies (e.g. Ahmadova et al., 2023; Bueno-García and Ahmadova, 2024; Zhang et al., 2024) arguing that firms’ internationalization increases paradoxical tensions due to the increase the complexity of handling activities abroad but, in parallel, internationalizing firm’s activities may entail an increase of the firm value and keep it from market changes. To fix this, we develop a co-evolutionary logic to understand this paradoxical nature of green innovation in international firms with a global strategy lens, due to this concept jointly integrates that firms try to co-evolve by adapting their strategies to multiple contexts but, in parallel, the complexity of succeeding in this process will be higher. As this fact means that co-evolution can succeed or fail in global expansion (Barber et al., 2019; Birkinshaw et al., 2016), we argue that internationalization may have different effects on the types of uncertainty since the possibility of success or failure will differently modify stakeholders’ perceptions to engage in green projects. In particular, this work suggests that firms’ internationalization increases behavioral uncertainty due to the increase the possibility of more opportunistic behavior from managers abroad based on a higher difficulty of a parallel co-evolution of their monitoring capabilities, and so the possibility of co-evolution’s failure increases, leading that the positive effect of external monitoring on green innovation will be mitigated at more international firms. In contrast, firms’ internationalization reduces environmental uncertainty because the firm’s value is better protected against market changes based on the increase of safeguarding co-evolutionary commitments against sudden ownership changes, and so co-evolution is more prone to succeed, leading that the positive effect of takeover protection on green innovation will be higher at more international firms. Together, this work provides key theoretical tools to understand the complex environmental management in global firms through strategic devices.

Our research provides key practical implications. We show that global firms need to adopt specific strategies, such as external monitoring and takeover protection, to address both behavioral and environmental uncertainties effectively and drive green innovation, but also to mitigate the doubts of all kinds of stakeholders to foster their participation in key firm strategies. External monitoring and takeover protection may arise as great devices to deal respectively with behavioral and environmental uncertainty and thus boost stakeholder investments in green processes to outperform competitors, so these devices may be considered by managers as interesting options to foster stakeholder investments, which are essential to help firms successfully develop competitive strategies for green innovation.

Looking forward, our results suggest that managers should move beyond static governance solutions and adopt a co-evolutionary approach to green innovation. Rather than treating external monitoring or takeover protection as fixed devices, executives should continuously recalibrate these mechanisms as firms expand internationally and face evolving regulatory, technological and stakeholder pressures. In this process, the effectiveness of each governance device depends critically on the firm’s level of internationalization, as global expansion simultaneously amplifies behavioral uncertainty and mitigates environmental uncertainty. By embracing a both-and logic – simultaneously ensuring control and flexibility, global consistency and local adaptation – firms can reduce the persistence of uncertainty and sustain stakeholder engagement in long-term green investments. In this sense, successful green innovation depends not only on the selection of appropriate governance mechanisms but also on managers’ ability to orchestrate their evolution over time in line with the firm’s global strategy.

From a global strategy perspective, managing uncertainty in green innovation requires a continuous process of adaptive global alignment between the firm, its stakeholders and the international environments in which it operates. The importance of co-evolution is further underscored by evidence on strategic failure in global firms. Research shows that firms that fail to co-evolve with changing institutional, technological and consumer environments often experience strategic decline or market exit, even when they possess substantial resources and market power (Barber et al., 2019). These failures frequently stem from rigid governance structures, overreliance on past success formulas and an inability to reconcile global integration with local responsiveness. For green technologies, such rigidity can amplify stakeholder uncertainty, as firms appear unable to credibly commit to long-term environmental strategies in the face of changing global conditions.

In contrast, successful multinational firms demonstrate that co-evolutionary strategies can effectively manage persistent uncertainty. Evidence from firms such as Nestlé, GSK and BMW shows that organizations capable of combining global integration with continuous strategic adaptation are better positioned to navigate discontinuous change (Birkinshaw et al., 2016). These firms adopt different modes of adaptation – structural separation, behavioral integration or sequential alternation – while maintaining long-term strategic coherence. Applied to green innovation, this implies that firms must develop dynamic governance arrangements that allow them to adjust monitoring mechanisms, investment protection and stakeholder engagement practices as environmental and institutional conditions evolve across countries. For green technologies, which face rapid regulatory shifts and heterogeneous stakeholder expectations across countries, such co-evolutionary governance becomes a critical capability rather than an optional strategic choice.

From a managerial perspective, our findings suggest that green innovation in international firms requires moving beyond static governance solutions toward a co-evolutionary global strategy that continuously aligns global integration and local adaptation. Green technologies are characterized by high sunk costs, long development cycles and persistent uncertainty, which makes stakeholders particularly sensitive to abrupt environmental shocks, such as hostile takeovers or sudden strategic reversals. Managers should therefore view internationalization not merely as geographic diversification but also as an ongoing co-evolutionary process in which governance structures, monitoring mechanisms and stakeholder relationships are regularly recalibrated across global and local contexts. In this setting, takeover protection plays a critical strategic role by stabilizing long-term commitments and safeguarding co-evolutionary investments against ownership disruptions, thereby reinforcing stakeholders’ confidence to engage in firm-specific investments for green innovation. Firms that fail to co-evolve their international strategies with changing regulatory, market and stakeholder demands risk undermining long-term green investments and, ultimately, exiting markets where green technologies require sustained commitment. Consequently, executives should proactively design international governance architectures that combine adaptive local responsiveness with globally integrated control mechanisms, enabling green innovation strategies to remain credible, resilient and attractive to stakeholders over time.

This work is not free from limitations. Although we propose that specific devices reduce uncertainty and enhance green innovation, we did not measure uncertainty or stakeholder investments (FSIs) directly, which represents a limitation in our study. That is, we did not measure the level of uncertainty or the level of FSI investment as we did not have access to such data.

Additionally, this study focuses on two specific governance devices – external monitoring and takeover protection – and on firm internationalization as the primary contextual factor shaping uncertainty. Other managerial mechanisms and organizational or environmental factors that may further reduce uncertainty and influence green innovation are beyond the scope of our analysis, and therefore we do not explicitly examine additional contextual moderators beyond firm internationalization. While external monitoring is generally effective in reducing uncertainty, its success is highly contingent on the firm’s level of internationalization. In highly global firms, the complexity of cross-border operations can reduce monitoring efficiency. This complexity is reflected in Meta’s ongoing challenges with regulating misinformation across diverse regulatory landscapes, highlighting the difficulties in applying uniform strategies to international contexts.

Building on the limitations of this study, future research could extend its scope by examining a broader set of managerial mechanisms and contextual factors that shape uncertainty management and green innovation. While this research focuses on external monitoring and takeover protection as key governance devices, other managerial practices – such as digital transparency tools, incentive alignment mechanisms, relational governance arrangements or sustainability-oriented leadership practices – may also play a relevant role in reducing uncertainty and fostering stakeholders’ engagement in green innovation.

In addition, future studies could explore contextual dimensions beyond firm internationalization that may condition the effectiveness of uncertainty-reducing devices. Factors such as institutional complexity, regulatory fragmentation, organizational structure or the degree of operational dispersion across countries may differentially shape how governance mechanisms operate and interact with stakeholders’ investment decisions.

Finally, further research could investigate how increasing global complexity affects the efficiency of monitoring mechanisms and how managers adapt, combine or recalibrate governance devices to maintain their effectiveness across heterogeneous international environments. Such work would contribute to a more nuanced understanding of uncertainty management as a contingent and evolving managerial decision process, rather than as a static governance choice.

We would like to sincerely thank each of our two reviewers and our editor who dedicated his or her time and expertise to reviewing our manuscript. They have helped to improve substantially our work.