Green finance and tourism-led economic growth

Economic growth (EG) is essential as it drives progress and fosters prosperity (Alexander and Gleeson, 2018; Ali et al., 2021). However, the reliance on carbon-based fuels in the pursuit of EG has adversely affected economies and the environment (Hantoko et al., 2024; Rauf et al., 2023). In many cases, economic development continues to depend heavily on fossil fuels and other environmentally harmful energy sources (Loredana et al., 2024), leading to the degradation of environmental quality. The total heating caused by harmful energy sources has left the environment devastated, depleted resources (Zavala et al., 2023), created land-related problems (Lungarska and Chakir, 2024) and caused overall environmental deterioration, issues that have now become global concerns requiring immediate attention (Ali et al., 2025a, b). The continued reliance on hazardous energy sources has also constrained economic development in leading tourist destinations, causing several of these economies to face persistent environmental pressures, including rising CO₂ emissions from transport and hospitality sectors, increased waste generation linked to high tourist inflows, coastal ecosystem degradation and intensified energy demand during peak tourism seasons. The environmental burden is specifically pronounced in major Mediterranean tourism destinations, where tourism activity has been associated with higher carbon intensity and escalating ecological stress. As a result, sustainable tourism practices and green financing have been proposed as viable solutions to mitigate these environmental pressures.

Sustainable tourism practices have emerged as a prominent and rapidly growing sector worldwide (Annamalah et al., 2023), as tourism is a key contributor to the Sustainable Development Goals (SDGs) agenda, supporting environmental, social and economic sustainability. According to the World Travel and Tourism Council (2023), it has accounted for 9.1% of global GDP, up 23% from 2022 but still 4.1% below 2019. It has also increased the workforce, with 27 million new jobs worldwide, a 9.1% increase from 2022, though still 1.4% below pre-pandemic employment. Furthermore, the literature suggests that tourism positively influences host countries' economic growth by increasing their gross domestic product (Vanhove, 2022), job creation (Purwono et al., 2024) and expanding the diversity of exported goods, among other factors, and, as such, tourism development should be regarded as a vital contributor to economic prosperity. Globally, travel is considered a key driver of EG (Katircioglu et al., 2014), and environmentally responsible tourism is increasingly viewed as an effective approach to mitigating global warming (Shang et al., 2023). Additionally, green financing contributes to reducing CO₂ emissions, improving environmental standards and boosting economic growth by supporting the production of energy efficient goods and services with low CO₂ emissions, thereby addressing climate change (Sharif et al., 2022). Several studies have demonstrated that green funds help mitigate climate change, making them a significant instrument in the global environmental agenda (Ali and Seraj, 2022). Furthermore, developments in clean energy mitigate reliance on energy sources that are responsible for producing excess CO₂, thereby supporting a sustainable economy (Hassan et al., 2022).

Though the literature on tourism-led-growth has expanded significantly, several critical gaps remain unaddressed (Igunnu and Ali, 2025; Zheng et al., 2023). First, from an empirical perspective, prior studies have largely examined TOR, GF, CO₂ and foreign direct investment (FDI) separately, with limited attention given to their integrated effects on EG within leading tourist economies. Existing studies predominantly focus on either the tourism-growth nexus or the environment-growth relationship in isolation, thereby providing fragmented evidence regarding the interaction between tourism expansion, sustainable finance and environmental sustainability. Second, from a theoretical perspective, previous studies have generally relied on single theoretical frameworks, such as the tourism-led growth hypothesis or endogenous growth theory, without integrating environmental sustainability considerations into a unified analytical framework. Consequently, insufficient attention has been devoted to understanding how TOR, GF, CO₂ and FDI jointly shape sustainable EG dynamics in tourism-dependent economies. Third, from a methodological perspective, much of the existing literature has employed conventional mean-based estimators, including OLS, fixed effects and standard panel regression approaches but these methods might fail to capture heterogeneous effects across different levels of EG and may produce biased estimates in the presence of cross-sectional dependence, slope heterogeneity and distributional asymmetry. Therefore, there remains a need for more robust econometric techniques capable of identifying conditional and quantile-specific relationships among the variables. Figure 1 shows the map of the top tourist nations investigated in this study.

Against this background, this study investigates the impacts of TOR, GF, CO₂ and FDI on EG in leading tourist economies, namely Spain, the United States, China, Italy, Turkey, Mexico, Thailand, Germany and the United Kingdom. To achieve this objective, the RALS-EG cointegration approach and the MMQR estimator to capture long-run and heterogeneous relationships across different quantiles of EG have been employed.

This study makes three contributions to existing literature. First, theoretically, it integrates the tourism-led growth hypothesis, green growth theory and endogenous growth into a cohesive framework to clarify the dynamics between TOR, GF, CO₂ and EG. Second, empirically, the study provides novel evidence on the tourism-finance-growth nexus in leading tourist economies using the MMQR approach. Unlike previous studies that focus on the average effects, the findings reveal heterogeneous effects across different levels of economic growth, specifically, the empirical findings expose: 1) the preliminary tests outcomes confirm the evidence of cross-sectional dependence among the variables, while unit root confirms a mixed order of integration; 2) the MMQR findings demonstrate that TOR and CO₂ exert a positive and statistically significant influence of EG across all quantiles. In addition, GF positively promotes economic growth across most of the quantiles, whereas FDI exhibits a stronger growth enhancing effect in higher quantiles, suggesting greater benefits in leading tourist economies; 3) the robustness analysis largely confirms the baseline findings; and 4) causality testing indicates the unidirectional causality running from TOR to EG and from EG to FDI, while directional causal relationships exist between GF and EG, as well as between CO₂ and EG. Third, methodologically, by utilizing the RALS-EG and MMQR approaches, the study accounts for cross-sectional dependence, slope heterogeneity and distributional heterogeneity, thereby providing more robust and reliable estimates compared to conventional means-based estimators (Ali et al., 2022). Figure 2 shows the trends and characteristics of the indicators used in the study.

The paper is structured as follows: Section 2 reviews literature; Section 3 describes the methodology; Section 4 presents findings and discussion; and Section 5 concludes with policy recommendations.

An integrated approach to the theoretical foundation of this study combines the Tourism-Led Growth Hypothesis (TLGH), green growth theory and endogenous growth theory, providing an explanation for the reciprocal relationships among TOR, GF, environmental quality and EG. Synthesizing these perspectives, rather than considering them in isolation, offers a stronger justification for the empirical model and the expected relationship among the variables. The TLGH posits that TOR development is a significant catalyst for EG by generating foreign exchange, creating employment opportunities and promoting infrastructure development that facilitates wider economic progress (Balaguer and Cantavella-Jordá, 2002). In this context, TOR is expected to contribute positively to EG, thus, β₁ > 0 is expected. Additionally, Popp (2012) explained that green growth theory adds more dimensions by arguing that the economy can grow while staying within the ecosystem's boundaries through its resources and eco-positive financial mechanisms. In this regard, GF promotes productive investment that also generates environmental benefits by supporting eco-friendly technologies and the development of sustainable business. Hence, GF is anticipated to yield positive EG, that is, β₂ > 0.

Nonetheless, CO₂ emissions represent a significant constraint on EG. While some strands of the literature suggest emissions may rise during the early stages of development due to industrialization and expanding economic activity, such growth is not sustainable in the long run. This is largely because environmental degradation imposes sustainable economic costs, including adverse health effects, increased regulatory burdens and resource depletion. From the perspective of environment-growth theory, CO₂ emissions are therefore considered economically counterproductive, meaning β₃ < 0. In addition, endogenous growth theory explains the positive impacts of FDI as a channel for technological diffusion, as well as capital and human resources accumulation (Baldwin et al., 2005). FDI contributes to increased productivity and enhances the competitiveness of the host economy, and is therefore expected to have a positive impact on EG; thus, β₄ > 0 is anticipated.

The integration of these theories depicts a coherent framework. Within this context, TOR serves as the primary growth engine, while GF facilitates investment in both the TOR sector and broader economy. Moreover, the CO₂ emissions gauge the environmental-economic and tourism growth trade-off, while at the same time, FDI supports technological advancement, which may either mitigate or exacerbate environmental impacts depending on the nature of the technology adopted. Therefore, the empirical framework conceptualizes EG as a function of TOR, GF, the environmental condition and FDI.

The integration of these theories provides a more comprehensive explanation of the interconnected relationships among TOR, GF, CO₂, FDI and EG. Within this framework, TOR acts as a major driver of economic expansion, while GF supports environmentally sustainable investment and cleaner production activities. At the same time, CO₂ emissions capture the environmental costs associated with economic and tourism expansion, whereas FDI facilitates technological spillovers, capital accumulation and productivity enhancement. Consequently, sustainable economic growth depends on the interaction between tourism development, environmental sustainability policies, financial systems and external investment flows.

It is widely recognized that TOR drives EG in many countries and the relationship between the two is widely studied in contemporary literature, such as Zheng et al. (2023) in BRICS, who found that TOR has a crucial role in promoting EG. Similarly, further studies in New Zealand (Jaforullah, 2015), Morocco (Huseynli, 2022), South Africa (Huseynli, 2022) and Hong Kong (Wang et al., 2022) asserted that TOR had a potentially beneficial impact on the economic development of the countries, although Wang et al. (2022) also highlighted a bidirectional causal link between TOR and EG. On the contrary, Alhowaish (2016) conducted a study in the Gulf Cooperation Council (GCC) nations from 1995 to 2012, the outcome of which differed from prior studies where panel Granger Causality methodology was utilized and determined that there is no causal connection between TOR and economic development in GCC nations. Moreover, research conducted by Nosheen et al. (2021) on Asian economies affirmed that TOR boosts EG at the cost of the environment. Likewise, Katircioglu et al. (2014) revealed a direct connection between the influx of visitors and the substantial increase in CO₂ emissions into the atmosphere and found that TOR contributes to EG but pollutes the environment. Hence, to state that TOR boosts the economy while lowering environmental stressors is inconclusive and there is a need for further investigation into how TOR creates a balance between EG and the environment.

Extensive research has delved into the diverse facets and efficacy of GF, with two notions observed in the literature, the first being that GF boosts economic growth. For example, Falcone (2020) and Zhang et al. (2021) discovered that GF substantially enhances economic development by assisting enterprises causing pollution with their transition into environmentally sustainable practices within the context of current investment and financing. Likewise, in China, Yin and Xu (2022) claimed that GF gadgets have the potential to act as catalysts for EG and development, while Xu et al. (2023) found that green technological innovation and environmental regulation significantly support net-zero emission targets in G20 economies. Similarly, Ali and Sultanuzzaman (2026) showed that green transition policies and governance quality promote renewable energy penetration in the Next Eleven economies, although FDI and industrial growth intensified carbon-intensive production structures. Moreover, Tolliver et al. (2021) and Tan et al. (2023) concluded that GF can further bolster EG and green productivity, while, on the other hand, Sarwar et al. (2021) used the two-step system GMM estimator and found that finance and human capital positively contribute to EG in emerging economies. Tiba and Belaid (2020) revealed the bi-causal connection between CO₂ and economic openness. Likewise, Nguyen Thi Truc and Le Thanh (2025) reported that climate-related financial policies reduce financial risks in economies with strong regulatory systems, although transitional challenges remain evident in developing economies.

Historically, CO₂ and EG have been closely linked (Ali et al., 2024a, b; Igunnu and Ali, 2025). As nations become wealthier, their energy consumption tends to rise, often driven by fossil fuels (Hussain et al., 2021), leading to increased CO₂ emissions. However, there has been a notable shift in recent years, with scholars consistently debating the notion of economic advancement invariably leading to environmental degradation. Recent scientific evidence suggests a significant positive connection between CO₂ and EG (Wang et al., 2022; Abbas et al., 2022; Wen et al., 2022). For instance, in 147 economies (Li et al., 2021), 26 European Union nations (Ren et al., 2021) and in Bangladesh, India, Nepal, Pakistan and Sri Lanka (Khan et al., 2022), scholars have illustrated the positive and substantial link in both factors and activities of an individual economy can have a favorable impact on the amount of CO₂ emissions, particularly in both China and India (Wang and Zhou, 2020) and Bangladesh (Karim et al., 2026). In India, various studies consistently indicated a positive link between CO₂ and EG (Raihan and Tuspekova, 2022) and Peter et al. (2026) observed that incremental rises in EG lead to an increase in environmental degradation in Nigeria, specifying that resources that aided economic growth and development have also led to the deterioration of environmental quality. Using the Driscoll–Kraay estimator for Asian economies, Tran et al. (2026) identified a U-shaped relationship between EG and CO₂ emissions, thereby challenging the Environmental Kuznets Curve hypothesis. Their findings suggest that industrialization and energy-intensive activities contribute to rising emissions at higher levels of economic development, indicating that the growth-environment relationship differs across economies and development stages.

The relationship between FDI and EG has attracted significant interest from researchers worldwide (Vo et al., 2019) and has been thoroughly examined using data from individual countries as well as cross-country samples in the literature, despite unanimous agreement on empirical findings not yet being reached. Iamsiraroj and Ulubaşoğlu (2015) evaluated the connection between FDI and economic development in 140 nations and found that FDI accelerates EG. Concurrently, it was concluded that in Eurozone countries (Pegkas, 2015) as well as 124 other nations worldwide (Iamsiraroj, 2016), FDI contributes to economic development. Using Augmented Mean Group (AMG) and Common Correlated Effects Mean Group (CCEMG) estimators, Ali et al. (2023) found FDI, trade openness, EG and research and development expenditures significantly promoted technological innovation in BRICS economies, and confirmed the existence of long-run causality between FDI and innovation activities, supporting the argument that FDI enhances technological spillovers and productivity growth.

This study investigates the effects of TOR, GT and FDI as environmental buffers and CO₂ emissions as environmental stressors on EG. The fundamental econometric framework is delineated below:

In Equation (1), $EGit$ refers to economic growth (Kyara et al., 2021), $TORit$ denotes tourism (Balsalobre-Lorente et al., 2023), $GFit$ refers to green finance (Ali et al., 2024a, b), $CO2it$ denotes carbon emissions (Ren et al., 2021) and $FDIit$ refers to foreign direct investment (Iamsiraroj, 2016). Moreover, $β0$⁠, $β1,β2,β3,and β4$ represent intercepts and coefficients of TOR, GF, CO₂ and FDI, while the i, t and $μit$ denote the leading tourist economies, study period (2000–2023) and error term, respectively. The summary of variables is presented in Table 1 and Figure 3 presents the graph of kernel density.

First, this study employs the CD test to assess the cross-sectional dependency and to determine which stationary test to utilize. Without conducting the CD test, the study's conclusions may exhibit inconsistency, unreliability and subjectivity, potentially leading to biased, skewed and less robust findings (Ali and Seraj, 2022). The equation form of the CD test is:

Second, based on the evidence from the CD test outcomes, the use of first-generation unit root tests may lead to estimation inefficiency. Therefore, this study employs the Cross-sectionally Augmented Dicky Fuller (CADF) and Cross-sectionally Augmented IPS (CIPS) tests, which are second-generation unit root tests designed to address cross-sectional dependence. The formula employed for the CADF and CIPS are presented in Equations (6) and (8), respectively, as shown below:

In Equation (4), $Δϑit$ represents the intercept, while, $Xit$⁠, denotes TOR, GF, CO₂ and FDI, Δ shows the difference operator, T denotes the time dimension and $∝it$ is the stochastic error term.

Third, the Slope Heterogeneity (SH) analysis is utilized to evaluate the changes in slope of coefficients. Ignoring the SH test could potentially lead to the misinterpretation of regression results, resulting in inaccurate hypothesis testing. Mathematically, SH is denoted as under:

Also, the equation for the adjusted delta in the SH test is as under:

Finally, the RALS-EG econometric technique developed by Lee et al. (2015) is employed to investigate cointegration among research variables, followed by Khalid et al. (2024) and Ali et al. (2024a, b). The RALS-EG offers several advantages, including reducing power loss, producing robust and efficient estimates, and improving least squares estimations under non-normal error distributions.

The Engle-Granger (EG) Test is a two-step cointegration procedure based on a residual-based framework and conventional t-statistics. In the first step, the stationarity properties and variable order of integration are examined using unit root tests to ensure the variables are integrated of the same order, typically I(1).

The ADF test is used to evaluate the level of integration after acquiring the residuals (ût).

When the error distribution of the test equation is non-normal, higher-order residual moments indicate deviations from normality. In such a case, the RALS approach can generate more robust, reliable and efficient estimates. To implement the RALS procedures, Equation (9) is modified as follows:

Whilst (ût) signifies white noise,

Furthermore, the following formula illustrates the RALS proposed by Meng et al. (2017).

in which, m_i = T⁻¹ $∑i=1T$ ̂$utj$⁠. Subsequently, by adding $wˆt$ in Equation (9), RALS is formulated in Equation (11) as detailed below

Furthermore, Equation (12) is employed to ascertain the three exponentially varying ranges of t-statistics

Based on the outcomes of the preliminary tests, this study utilizes the MMQR analyzer proposed by Machado and Silva (2019) to capture the conditional heterogeneity of the explained variable. Unlike standard quantile regression approaches, MMQR addresses unobserved variation and yields more reliable estimates when cross-sectional dependence and distributional asymmetries are present (Khalid et al., 2026). This renders it especially appropriate for panel data analysis, where the influence of explanatory variables is heterogeneous through many aspects of the conditional distribution (Ali et al., 2025a, b).

The model's fixed influence and V_it are represented by $∂i$ and $Ai$⁠, denoting the differentiable transformation of the m vector of Y. The following is the mathematical presentation

Furthermore, in the context of quantiles, estimations are derived utilizing the following procedure

$Yit$ represents the predictor facets vector in Equation (15), encompassing TOR, GF, CO and FDI. The dependent quantile spread of the internal aspect in Equation (15) is situated on the left. The distinguishing feature of the individual effects estimate methodology, in contrast to other LS-fixed effects methodologies, is the absence of an intercept term. The variables are supposed to exhibit time-invariance, indicating that the disparities between the units are presumed to remain constant. The subsequent solution is optimized to derive the quantiles of the collected data.

This study uses the FGLS method to determine the connection among the variables and to assess the robustness of MMQR estimations. The FGLS is considered effective due to its ability to address potential serial correlations, heteroscedasticity and cross-sectional dependence (Bai et al., 2021), and the Dumitrescu-Hurlin (DH) test is used to examine the direction of causality among the study variables with the overall methodological framework presented in Figure 4.

Table 2 presents descriptive statistics of the data. It is observed that all variables exhibit positive mean values, with EG recording the highest mean (28.255). In terms of the range, EG shows the highest values, whereas FDI records the lowest. Furthermore, Table 3 reports the outcomes of the CD, CIPS, CADF and RALS-ADF tests and shows the findings indicate the variables exhibit cross-sectional dependence and mixed order of integration.

Moreover, Table 4 reports the findings of the SH test, where both delta and adjusted delta statistics are at the 1% level which is significant as it provides evidence of slope heterogeneity. Cointegration is then assessed using the RALS-EG estimator, with results presented in Table 5 and the t-statistic (−5.49253) below the critical value (−4.80), confirming the presence of a long-run cointegration relationship among the variables.

The findings of the MMQR estimation are reported in Table 6. The findings indicate the coefficients of TOR across all quantiles (q1 to q9) are positive and statistically significant, suggesting a robust positive relationship between TOR and EG. Moreover, the magnitude of the coefficient exhibits an increasing trend across higher quantiles, implying that economies experiencing higher levels of growth derive greater benefits from TOR over time. These outcomes support the tourism-led-growth hypothesis, affirming that TOR substantially contributes to EG. These results are consistent with Razzaq et al. (2023), who argue that promoting TOR boosts EG in the top ten GDP nations. However, the findings are in contrast with those of Nyasha et al. (2021), who could not find any connection between TOR and EG in low- and middle-income African nations. The stronger positive effects of TOR across higher quantiles may reflect the greater capacity of high-growth tourist economies to efficiently utilize TOR revenues through advanced infrastructure, diversified service industries and stronger institutional support. In leading tourist nations, such as Spain, the United States and the United Kingdom, TOR contributes through direct revenue generation as well as employment creation, transportation development, hospitality expansion and spillover effects across related sectors.

Furthermore, the MMQR findings indicate that GF exerts a positive and statistically substantial effect on EG across the most quantiles (q2 to q9), which suggests that an increase in GF contributes to an increase in the EG in leading tourist nations. Moreover, the magnitude of the coefficients from q2 to q9 increases gradually, implying GF plays a stronger role in promoting EG at higher levels of economic development. The findings are consistent with prior studies by Tan et al. (2023) and Yin and Xu (2022), which highlight that promoting the adoption of environmentally sustainable investment practices in investment policies substantially supports economic development. In addition, the results are consistent with the green growth strategy, which emphasizes the need for governments to harmonize economic expansion with environmental sustainability. This may be accomplished by strengthening the eco-efficiency of economic development and optimizing the synergistic relationship between the environment and the economy. However, insignificant effects observed at the lowest quantile suggest the effectiveness of GF may be constrained by structural and institutional limitations in lower growth economies, where financial systems and environmental investment are developing.

Furthermore, the MMQR findings demonstrate that CO₂ positively influences EG across all quantiles. However, the magnitude of the coefficients declines progressively from q1 to q9, suggesting that the growth-enhancing effect of CO₂ emissions weakens at higher levels of economic development. This may reflect the gradual transition from fossil fuel dependence toward cleaner and renewable energy sources, supported by increasing investment in GF. These findings are consistent with Khan et al. (2022) and Muhammad (2019), who found that CO₂ emissions contribute to EG despite their adverse environmental consequences. Similarly, industrial and tourism-related activities in leading tourist economies have supported economic expansion while simultaneously increasing energy demand and CO₂ (McGee and York, 2018; Zhang et al., 2017). However, this positive relationship reflects a clear growth-environment trade-off, where EG is partly achieved at the expense of environmental sustainability. Therefore, governments should strengthen environmental regulations, promote cleaner technologies and encourage sustainable production systems to reduce long-run environmental risks.

Finally, the MMQR findings confirm FDI favorably influences EG across upper quantiles (q6 to q9) which demonstrates FDI contributes to greater economic performance at higher levels of economic development. Meanwhile, at lower and middle quantiles (q1 to q5), FDI has no meaningful influence on EG. However, these findings confirm the hypothesis that higher FDI would lead to better economic performance, findings in line with those of Hong Hiep et al. (2023) and Tran et al. (2023), who found FDI has a positive effect on provincial EG in Vietnam, which underscores how important FDI is in fostering growth among selected countries. In addition, findings support the endogenous growth theory, which posits that FDI fosters economic development by augmenting both capital accumulation and productivity. Consequently, governments must encourage FDI inflows for technologies, quality management approaches and access to international markets, which ultimately boosts economic development through increasing efficiency and expanding production capacity.

To further strengthen the robustness of the findings, additional analyses were conducted by incorporating global shock variables and heterogeneity-based estimations, as presented in Table 7. Specifically, dummy variables capturing the global financial crisis and the COVID-19 pandemic were introduced into the baseline model to account for potential structural disruptions and global economic shocks affecting tourism-led economies. The findings reveal the COVID-19 shock exerted a positive and statistically significant effect on EG, whereas the global financial crisis variable remained statistically insignificant. Furthermore, the sample was divided into large and small tourist economies based on GDP levels to capture cross-country heterogeneity. The results indicated TOR maintains a positive and significant effect on EG across both groups, although the magnitude was stronger among small economies and that GF positively contributes to growth in large economies but becomes insignificant in smaller economies, suggesting differences in financial market development and absorptive capacity. In addition, CO₂ emissions remain positively associated with EG across all models, reinforcing the existence of a growth-environment trade-off. Finally, the Driscoll–Kraay estimation was employed to further address cross-sectional dependence, heteroskedasticity and autocorrelation, with the outcomes remaining broadly consistent with the baseline estimations, thereby confirming the robustness and reliability of the main findings.

To enhance the robustness of the empirical analysis, this study further employs the Dumitrescu-Hurlin panel causality test to determine the direction of causality among the variables, as reported in Table 8. The findings reveal two unidirectional and two bidirectional causal relationships, with specific evidence of unidirectional causality running from FDI to EG, as well as from TOR to EG. Additionally, bidirectional causality is found between GF and EG and between CO₂ and EG.

The study examines the effects of TOR, GF, CO₂ and FDI on EG in leading tourist economies by employing the MMQR model to capture the heterogeneous effects across different quantiles and the FGLS estimator to ensure the robustness and reliability of the results.

The outcomes suggest that TOR promotes EG, thereby supporting the tourism-led-growth hypothesis. Furthermore, CO₂ emissions exhibit a concerning pattern across all quantiles. Industrial and manufacturing expansions, which has driven EG in leading tourist economies over the recent decades, have also increased demand for energy, thereby contributing to CO₂ emissions. In addition, the results show that GF plays a dual role by reducing CO₂ emissions while simultaneously promoting EG. Thus, GF helps alleviate the adverse environmental impacts associated with CO₂ emissions while enhancing EG. Finally, FDI is observed to have a positive effect on EG, particularly in the upper quantiles, indicating its stronger effect in the long run.

Based on the findings, several practical policy implications are proposed. First, since TOR is found to promote EG, governments should consider TOR as an effective policy instrument for fostering economic expansion. To attract international tourists, it is essential to preserve and sustainably manage indigenous TOR destinations and ensure the safety and security of international visitors, which requires the effective implementation of relevant regulations and policies. Governments may also introduce TOR certification programs, invest in eco-friendly infrastructure and adopt digital TOR promotion strategies to enhance competitiveness and ensure the long-term resilience of the TOR sector.

Second, many challenges affecting the environment caused by the world's leading tourist countries can be tackled by channeling the economy toward sustainability, since GF is effective in fostering sustainable EG. Therefore, governments of the world's leading tourist countries have a responsibility to promote green financing and sustain EG by offering tax incentives and subsidies for investment in renewable energy and sustainable infrastructure. It should also be considered that governments need to support these initiatives by expanding green bond markets, establishing sustainability-linked financing frameworks and enforcing environmental standards within tourism-related investments to ensure alignment with long-term sustainability objectives.

Third, governments should create a framework for attracting FDI by strengthening local infrastructure and supporting political, economic and legal stability, which will create better conditions for local businesses, pushing them to improve a country's economy. Alongside this, targeted policies, such as investment facilitation programs, sector-specific incentives for sustainable industries and streamlined regulatory frameworks, should be implemented to attract high-quality and environmentally responsible FDIs. Finally, countries that attract a large percentage of global tourists should work on improving their economic structures and consumption behavior so environmental degradation caused by EG is minimized.

This study contributes to existing literature on the effect of TOR, GF, CO₂ emissions FDI on EG in leading economies. It also extends prior research by integrating the tourism-led-growth hypothesis, green growth theory and the endogenous growth model into a unified framework, thereby providing a stronger theoretical foundation for explaining growth environment dynamics.

Though the study presents useful empirical insight, it has some constraints that must be addressed. First, it focuses on nine tourist economies due to the data availability for the given timeframe, which could influence the findings. Second, although the study has utilized a sophisticated econometric framework to advance the estimates in quantile form, further research may apply CS-ARDL, AMG and CCEMG models to analyze the long-run data. Furthermore, the sample can be enlarged by considering more countries and recent data to take a more extensive global overview of the association among the parameters. Prospective scholars can also consider a country-focused analysis to investigate the heterogeneous effect at a more disaggregated level, such as a state or city. Finally, further studies can be performed by employing time-varying parameter models with structural breaks to better capture the effects of dynamic heterogeneity and transition over time across economies in the sample.