Achieving environmental sustainability requires an integrated understanding of how economic and political dynamics interact with governance structures. The influence of economic policy uncertainty (EPU), political risk (PRI), and governance (GOV) on sustainability (measured by the load capacity factor, LC) can be explained through several theoretical lenses.

Real Options Theory suggests that uncertainty increases the value of delaying investments in long-term projects (Chakraborty et al., 2025; Dixit and Pindyck, 1994). In contexts of high EPU, firms are likely to postpone or reduce investment in clean energy infrastructure, as unpredictable policies increase financial risk (Bloom, 2009; Wang et al., 2025). This dynamic undermines decarbonization pathways by creating hesitation in adopting green technologies.

Porter's Hypothesis and innovation-driven growth theory argue the opposite: under certain regulatory environments, policy uncertainty can stimulate proactive investment in eco-friendly technologies as firms hedge against future regulatory shifts (Yang et al., 2023). Hence, well-designed policies, even amid uncertainty, can catalyze green innovation and competitive advantage (Pata et al., 2023b; Porter and Van Der Linde, 1995; Yasin et al., 2024).

Institutional Theory emphasizes the importance of strong political institutions for environmental governance (North, 1990). Political instability weakens policy enforcement, enabling regulatory gaps, corruption, and unsustainable practices (Garfinkel and Skaperdas, 2000; Liu et al., 2023). However, under the conflict-resolution and political legitimacy hypotheses, unstable governments may adopt stringent environmental laws to build legitimacy or attract foreign investment, paradoxically enhancing sustainability under certain conditions.

Finally, the environmental governance framework underscores that robust governance reduces corruption, improves regulatory enforcement, and facilitates public–private partnerships (Ostrom, 1990; Ulussever et al., 2024; Wine, 2019). Conversely, weak governance leads to fragmented policies, deterring green investment and exacerbating ecological degradation (Barbier, 2010; Zhang and Wen, 2023).

This study integrates these perspectives into a unifying framework. We posit that EPU and PRI create risks that can either delay or, under certain conditions, accelerate green transitions, while governance provides the institutional backbone that determines the effectiveness of sustainability policies. Moreover, environmental innovation is expected to moderate these effects, reducing the adverse impacts of uncertainty and political instability. Figure 1 illustrates the conceptual framework of this study.

The relationship between uncertainty, governance, and environmental sustainability has been widely debated, though findings remain mixed.

Table 1 presents a range of studies examining the influence of EPU, PRI, or GOV on environmental outcomes. Building on these gaps, this study makes a unique conceptual contribution by integrating economic policy uncertainty (EPU), political risk (PRI), and governance (GOV) into a unified framework for understanding environmental sustainability, measured through the load capacity factor (LC). While prior research has examined these drivers individually, their combined effects on ecological capacity remain largely unexplored. Furthermore, by introducing environmental innovation (EI) as a moderating factor, we extend existing theories—such as real options theory, institutional theory, and the environmental governance framework—by demonstrating how innovation can mitigate the adverse effects of uncertainty and amplify the positive role of governance. This integrated perspective advances theoretical discourse by shifting from single-dimension analyses to a systemic model that captures the interplay of uncertainty, governance, and innovation. To the best of our knowledge, this represents the first attempt to test such a framework empirically in the Canadian context, thereby contributing both to theory and practice in environmental sustainability research.

The present study incorporates annual data from 1990 to 2022 to investigate the impacts of economic policy uncertainty (EPU), political risk index (PRI), and governance (GOV) on environmental sustainability in Canada. The study used the load capacity factor as a proxy for environmental sustainability. The variable of load capacity factor is obtained through biocapacity/ecological footprints. The statistics for all these parameters is downloaded from the Global Footprint Network (GFN, 2023). The selection of explanatory variables is based on their critical role in shaping environmental policies and sustainability outcomes. Economic policy uncertainty (EPU) is a key determinant of investment and policy decisions that impact the adoption of clean technologies and sustainable practices. High levels of uncertainty can discourage long-term investments in environmental initiatives, making it an important factor to examine. The data for EPU is extracted from Economic Policy Uncertainty Index (EPU, 2023). Political risk (PRI) influences environmental sustainability by affecting regulatory stability, enforcement mechanisms, and investment confidence. Higher political risk can either delay or incentivize sustainability efforts, depending on governance effectiveness. The PRI data is sourced from Political Risk Services (PRS, 2023). Governance (GOV) plays a pivotal role in implementing and enforcing environmental regulations. Strong governance structures ensure effective policy execution, while weak governance can lead to regulatory inefficiencies and environmental degradation. To measure governance, this study constructs a composite governance index utilizing principal component analysis (PCA). The index integrates six dimensions of governance, providing a comprehensive measure of institutional effectiveness in driving sustainability. Details of the governance index construction are presented in Tables A1–A3 and Figure A1. The single components of the index are given in Table 2. Further, Table 3 describes the specific information on the variables used in the research.

Before diving into the analysis, we ensure the time plots in Figure 2 have the right visual qualities for the raw data. The purpose is to look for signs of structural breaks, seasonality patterns, and drifts. The figure shows a general pattern of fluctuating values of LC, with a rise in the early 1990s, a decline in the mid-1990s, and a fluctuation in the early 2000s. Around 2000, there seems to be a slight rise in the LC related to technological breakthroughs or infrastructure improvements. Moreover, the EPU pattern demonstrates periods of stability and considerable fluctuation, with major rises in 2020 and falls in 2021 and 2022. Understanding these swings is critical for decision-making and successfully navigating the economic environment. Besides, the PRI graph shows a mostly stable trend with minor oscillations. According to the figure, the GOV experienced relatively stable negative values between 1990 and 1996, suggesting a period of consistent governance deficiencies. However, starting from 1996, there is a noticeable positive shift in the GOV, indicating an improvement in governance quality. It shows a significant upward trend in the GOV from the late 1990s to the early 2000s, reflecting positive governance reforms or initiatives undertaken during this period.

Furthermore, the data shows a generally increasing trend in governance scores from 2002 to 2017, indicating sustained efforts to enhance governance practices and institutions in Canada. Finally, from 1990 to the early 2000s, the EI displays some variability, indicating a mix of positive and negative trends. However, starting from the mid-2000s, there appears to be a general upward trend in the EI, pointing toward increased focus and advancements in environmental innovation in Canada.

Environmental innovation refers to creating and using novel technologies, methods, and remedies to reduce environmental harm, preserve resources, and advance sustainability. It involves developing and implementing inventive methods, goods, and services that help safeguard the environment, mitigate climate change, and conserve natural ecosystems (Skordoulis et al., 2020). In this study, environmental innovation is a moderating variable affecting the link between the independent variables (EPU, PRI, GOV) and the dependent variable (LC). It essentially helps to explain how the relationship between these variables is affected by the level of environmental innovation in Canada. The moderation analysis integrating environmental innovation will assist in creating a more thorough knowledge of how environmental issues interact with economic and governance-related variables to shape Canada's environmental sustainability. Numerous empirical investigations have used patents on environmentally related technologies to represent environmental innovation. Therefore, the current research utilized the log of patents related to environmental technologies as the dependent variable (Abbas et al., 2024). The data was obtained from OECD (2023).

This study investigates the linkages between environmental sustainability (LC), economic policy uncertainty (EPU), political risk (PRI), and governance (GOV) in Canada by applying advanced quantile-based and time–frequency methods. Traditional econometric models such as ARDL, CS-ARDL, DOLS, and FMOLS are widely used in the environmental economics literature; however, they focus primarily on mean relationships and assume linearity and symmetry. These assumptions are restrictive in contexts where relationships may vary across different levels of environmental stress, governance quality, or uncertainty. For instance, EPU may exert a weak influence on sustainability under stable ecological conditions but a much stronger effect under high ecological pressure. Similarly, governance quality may matter more during periods of political instability. Such distributional heterogeneity cannot be adequately captured by mean-based estimators.

To address these limitations, this study employs a three-step strategy: (i) Xiao's (2009) quantile-based cointegration test to assess long-run relationships across quantiles; (ii) the Quantile-on-Quantile Regression (QQR) model of Sim and Zhou (2015) to capture cross-quantile linkages between explanatory and dependent variables; and (iii) the Wavelet Coherence Technique (WTC) to explore dynamic co-movements and lead–lag structures across multiple time horizons. Similar approaches have been used in recent environmental and energy economics research. For example, Ali et al. (2025) used QQR to analyze technology–environment linkages, Musibau et al. (2025) applied it to renewable energy and generation, while Yu et al. (2024) employed it to study CO2 emissions. Likewise, WTC has been used by Zhang W. et al. (2024) and Sun et al. (2023b) to assess energy–economic interactions. These studies highlight that quantile-based and wavelet methods are well-suited for uncovering nonlinear, asymmetric, and time-varying dynamics that are central to sustainability outcomes.

Prior to doing the QQR examination, our aim is to uncover the quantitative characteristics of the series of logarithms. To begin, we review the summary statistics shown in Table 4. The average values suggest that all series maintain a positive mean over the observation period. The standard deviation results show that LC has less volatile behavior than EPU, which is the most volatile. Additionally, none of the variables follow a standard distribution based on the Jarque-Bera values. Linear econometric methods fail to account for non-parametric data when the indicators do not conform to a normal distribution. The non-normal distribution of the variables suggests that quantile approaches may provide excellent results. The study used quantile-based nonparametric techniques because, by taking into consideration nonlinearity and asymmetric interactions, quantile approaches may provide strong evidence of the interactions between non-normally distributed series.

In addition, we have examined the quantile plots, as shown in Figure 4, to investigate the normal distribution features of selected variables thoroughly. The red lines in these Q-Q plots show a standard distribution, whereas the highlighted data points illustrate the real distribution of variables. The disparity between the red lines and the colored data points provides information about the departure from normality. The magnitude of this disparity indicates the level of asymmetry, as elucidated by Ozkan et al. (2024).

Additionally, Box plots are depicted in Figure 5, providing a succinct representation of the summary information. Moreover, Figure 6 illustrates the correlation between the factors.

In order to avoid biased results and make sure the empirical findings are accurate, the reliability of the model is checked by applying the Q stationarity test before the QQR method is used. Nineteen quantiles with values ranging from 0.05 to 0.95 are employed for this purpose. Finding quantile unit roots in data is done by comparing the t-statistic to critical values. The t-statistic and critical value resilience across various quantile ranges is shown in Table 5. Assuming the alternative hypothesis is accepted, the null hypothesis is maintained if the anticipated t-statistic value does not exceed the critical value. At the 5% significance level, this happens for every quantile, where α(τ) = 1. According to the quantile unit root test, unit roots exist at the quantile level for all variables.

The research then moved on to examine the variables' long-term interdependence through a unique QC test (Xiao, 2009). The coefficients represent the uniform norm and the critical values β and γ, while the significance levels of 1%, 5%, and 10% are denoted by CV1, CV2, and CV10, respectively. The test was reliably performed for over 19 quantiles, from 0.95 to 0.05. Table 6 indicates that, at a significance level of 1%, both the β and γ coefficients surpass all critical levels. As a result, the results show cointegration between a subset of the variables, exhibiting a steady, asymmetric, prolonged connection between these variables.

As the subsequent measure, Table 7 displays the results of the nonlinearity test. The results demonstrate that every variable has a multidimensional non-linear structure. Reflecting on these aspects, using a nonlinear approach for further empirical analysis may be the most suitable choice. Hence, the research utilizes the innovative QQR technique.

The core aim of this inquiry is to examine the interrelationships among EPU, PRI, GOV, and LC. Following the confirmation of cointegration among the variables, the study explores how these factors influence environmental sustainability in Canada using QQR, which allows for a detailed understanding of the relationships across the entire distribution of LC. Figures 7–9 present the estimated regression coefficient β1(θ, τ), representing the impact of the τ-th quantile of EPU, PRI, and GOV on the θ-th quantile of LC.

Figure 7 depicts the influence of EPU on LC in Canada, revealing notable heterogeneity across quantiles. At lower LC quantiles, where environmental sustainability is relatively weak, rising EPU negatively affects LC, reflecting increased investment uncertainty that discourages long-term commitments to renewable energy projects and green infrastructure. In such conditions, firms and policymakers may prioritize short-term stability, relying on conventional energy sources and thereby hindering sustainability progress. Conversely, at medium and higher quantiles, the effect of EPU becomes positive and increases in magnitude, suggesting that in more sustainable scenarios, uncertainty can stimulate adaptive policy responses and strategic investment in cleaner energy alternatives. Businesses anticipating long-term uncertainty hedge risks by diversifying into sustainable energy portfolios, while Canada's commitment to global environmental agreements and market-based green policies further supports sustainability efforts. These results are consistent with previous studies that have documented the nuanced positive effects of policy uncertainty under certain economic and institutional conditions (Aslan et al., 2024; Pata et al., 2023b; Villanthenkodath and Pal, 2024).

Figure 8 illustrates the relationship between PRI and LC. At lower quantiles, higher political risk negatively affects sustainability due to heightened uncertainty, inconsistent policies, and reduced investor confidence, which can delay or reduce investment in green energy projects. Interestingly, at medium and higher LC quantiles, PRI exhibits a positive effect. While this may appear counterintuitive, it can be explained by Canada's strong institutional framework and adaptive mechanisms: moderate political risk can trigger proactive government and business responses, leading to resilient strategies that promote sustainability. The presence of environmental innovation further strengthens this effect, mitigating the potential negative impacts of political instability and allowing firms to implement adaptive technologies and practices. This observation aligns with evidence suggesting that robust institutions and innovation capacity can transform political risks into opportunities for sustainable investment (Kartal et al., 2024, 2022; Simionescu et al., 2023).

The effect of GOV on LC is shown in Figure 9. At lower quantiles, weak governance can impede environmental policies, resulting in inefficient resource allocation and regulatory shortcomings that negatively affect LC. As quantiles increase, the impact of governance becomes positive and more pronounced, indicating that economies with stronger institutional frameworks and governance mechanisms are better positioned to implement effective sustainability policies. The increasing magnitude of the coefficient at higher quantiles suggests that well-functioning governance enhances regulatory enforcement, encourages green investment, and supports sustainable infrastructure, ultimately fostering long-term improvements in environmental sustainability (Yadav et al., 2024; Yi et al., 2023; Zhang S. et al., 2024).

By examining the moderating role of environmental innovation, we further elucidated how varying levels of innovation intensity influence the relationships between EPU, PRI, GOV, and LC across the sustainability distribution. Figure 10 shows that the interaction term EPU × EI negatively affects LC at lower quantiles, suggesting that the benefits of innovation may be constrained under conditions of economic uncertainty, reduced investor confidence, and delayed policy implementation. However, at higher quantiles, this interaction becomes positive, highlighting that when innovation reaches a critical threshold or when economic conditions are stable, it can mitigate uncertainty and drive sustainable outcomes. Similarly, Figure 11 illustrates that PRI × EI positively influences LC across most quantiles, with the effect strengthening at higher levels. This indicates that strong environmental innovation can offset the adverse effects of political risk, enabling adaptive strategies that enhance resilience and sustainability. Figure 12 demonstrates that GOV × EI negatively affects LC at lower and medium quantiles but becomes strongly positive at higher quantiles, reflecting that the benefits of innovation are fully realized only when governance structures are robust and effectively implemented. These findings emphasize that environmental innovation amplifies the positive impacts of governance and policy while mitigating risks associated with economic and political uncertainty, thereby supporting more resilient and sustainable environmental outcomes.

Finally, the Wavelet Coherence (WTC) analysis (Figures 13A–C) confirms the dynamic linkages between LC and EPU, PRI, and GOV over time. Warmer colors indicate stronger interdependence, whereas cooler colors denote weaker association. The results show a strong positive correlation between EPU and LC over multiple periods (Figure 13A), a negative correlation between PRI and LC (Figure 13B), and a positive correlation between GOV and LC at medium and high frequencies (Figure 13C). These findings corroborate the QQR results and highlight the critical role of governance, institutional quality, and environmental innovation in shaping sustainable outcomes in Canada under varying conditions of policy and political uncertainty. More over, Figure 14 represents the graphical summary of the key results.

This study employs the innovative Quantile-on-Quantile Regression (QQR) approach to investigate the nonlinear effects of Economic Policy Uncertainty (EPU), Political Risk Index (PRI), and Governance (GOV) on environmental sustainability, proxied by the Load Capacity Factor (LC), in Canada from 1990 to 2022. The findings reveal that the relationships among these institutional and policy variables and sustainability are highly heterogeneous across the distribution of LC, with environmental innovation (EI) playing a critical moderating role. Specifically, while EPU can have both positive and negative effects depending on the sustainability context, PRI generally constrains sustainability at lower quantiles but may foster adaptive responses at higher quantiles when supported by strong EI. Effective governance consistently enhances sustainability, particularly when coupled with innovation. These outcomes are corroborated by Wavelet Coherence (WTC) analysis, which confirms the temporal and frequency-dependent linkages among EPU, PRI, GOV, and LC. Collectively, these findings provide a nuanced understanding of how policy, institutional, and innovation mechanisms interact to shape environmental outcomes in Canada, offering both theoretical and practical contributions to sustainability research.

The empirical evidence underscores the importance of tailored and actionable policy interventions. First, to mitigate the destabilizing effects of political risk and maximize the adaptive potential of policy uncertainty, Canadian policymakers should implement transparent decision-making procedures, strengthen risk management frameworks, and promote policy predictability, particularly in sectors critical to clean energy investment. Second, fostering a robust ecosystem for environmental innovation is essential. This includes targeted financial incentives, tax breaks, and research grants for green technology development, as well as establishing knowledge-sharing platforms for best practices in sustainable business operations. Third, the synergistic relationship between governance and EI highlights the need for institutional reforms that enhance regulatory efficiency, ensure accountability, and integrate technological solutions into environmental monitoring and enforcement. By adopting these measures, Canada can leverage innovation and governance to transform potential policy or political risks into opportunities for long-term environmental sustainability.

While this study provides novel insights, certain limitations should be acknowledged. First, the analysis focuses exclusively on Canada, limiting the generalizability of the findings to other national contexts with different institutional or policy frameworks. Second, the study employs LC as a proxy for environmental sustainability, which, while comprehensive, may not capture all dimensions of environmental performance, such as biodiversity or ecosystem services. Future research could extend this work by incorporating multi-dimensional sustainability indicators, analyzing cross-country comparisons, and exploring the dynamic effects of other institutional or market variables, including climate finance and international policy commitments. Additionally, investigating the causal mechanisms through which environmental innovation interacts with governance and policy uncertainty could provide deeper theoretical understanding and inform more precise policy interventions.