Environmental impact of financial technology across quantiles: the importance of natural resources in ASEAN countries

The Association of Southeast Asian Nations (ASEAN) plays a pivotal role in the dynamics of the global economy. ASEAN countries make substantial contributions to economic growth, promote trade, attract investment capital and serve as critical links in the global value chain network (Wang et al., 2024a, b). However, accompanying this economic expansion is an environmental trade-off, as suggested by the Environmental Kuznets Curve (EKC) hypothesis (Grossman and Krueger, 1995). In recent years, ASEAN have experienced a significant increase in natural disasters and extreme weather events and is among the most affected by the adverse environmental consequences of global warming (Ding and Beh, 2022).

According to the ASEAN SDG Snapshot Report ^[1], ASEAN is considered one of the most vulnerable regions to climate-related disasters and weather variability. Statistics show that, on average, 2,546.8 individuals per 100,000 population were reported dead, missing, or directly affected by climate-related disasters, an alarming increase compared to 554 cases per 100,000 in 2016. Based on data from the SDG Score ^[2], ASEAN countries have made notable progress in achieving several United Nations Sustainable Development Goals (SDGs), particularly in poverty reduction and human capital development (Tang et al., 2024). However, performance on environmental goals, specifically SDG 13 (Climate action) and SDG 12 (Responsible consumption and production), remains relatively weak. Most ASEAN countries, except for Singapore, are still in the industrialization phase and rely heavily on finite natural resources (including coal, oil, forest land and natural gas), following a predominantly linear economic model (Murshed and Rahman, 2024). These countries have yet to implement effective interventions to address the depletion of natural resources, minerals and fossil fuels driven by excessive economic exploitation (Tang et al., 2022; Le et al., 2026).

Concurrently, Fintech, the intersection of technology and finance, has garnered significant attention from public sector policymakers as well as being a catalyst for engagement among the business community and academic researchers (Dong et al., 2024; Okere et al., 2025a). Fintech has revolutionized the operations of financial, banking and monetary systems through digital transformation over the past decade, contributing to the streamlining of business processes and the expansion of capital mobilization. The automation and digitization of financial services through Fintech have had widespread impacts across various sectors and dimensions. Smart technological solutions such as mobile payments, e-banking, blockchain technology, mobile money, the Internet of Things (IoT) and peer-to-peer (P2P) lending are encompassed within the scope of Fintech. Moreover, Fintech represents a wide range of financial activities, including payments, financial advisory, project financing, deposits, lending and insurance services (Stulz, 2019).

In the ASEAN context, Fintech has effectively contributed to economic growth and financial inclusion. It has presented both opportunities and challenges for traditional financial institutions through digital transformation via technologies such as artificial intelligence (AI), blockchains and cryptocurrencies (Barroso and Laborda, 2022; Okere et al., 2024). The digitalization of payment and remittance services has streamlined processes, reduced costs, and enhanced efficiency for businesses as well as financial and banking institutions (Scardovi, 2017). Fintech has also gained recognition for its potential to democratize financial services and promote financial inclusion, while mitigating negative environmental impacts through increased savings derived from technological efficiency (Thomas and Hedrick-Wong, 2019). Empirical studies indicate that the adoption of advanced technologies, particularly in the financial and banking sector, holds the potential to reduce CO₂ emissions in the financial industry by enabling online transactions, optimizing traditional processes and conserving resources and materials (Zhang and Liu, 2022). As a result, Fintech is regarded as a transformative tool that not only sustains economic growth and productivity but also has the capacity to address environmental concerns (Tamasiga et al., 2022; Okere et al., 2024, 2026).

Given the urgency stated, this study aims to investigate the environmental impacts of Fintech as well as the significance of natural resources. Despite a growing body of literature, several important empirical gaps remain. First, existing studies predominantly focus on the direct relationship between Fintech and environmental outcomes, often overlooking the underlying transmission mechanisms (particularly the role of natural resource utilization), thereby risking omitted variable bias. Second, while evidence on the Fintech–environment nexus is gradually emerging, there is a notable scarcity of research within the ASEAN context, where structural characteristics, institutional quality, and resource dependence may lead to markedly different dynamics. Moreover, prior works tend to rely on single proxies such as CO₂ emissions, which may not fully capture the multidimensional nature of environmental sustainability. To address these limitations, this study provides a more comprehensive empirical framework by incorporating both environmental degradation and environmental quality indicators, while explicitly modeling the mediating channel of natural resources. Furthermore, Fintech has the potential to reduce information asymmetry and transaction costs, thereby improving the allocation of financial resources in economic activity. This study will revisit the EKC hypothesis to revisit the environmental aspects of the country-level development cycle when being controlled by Fintech. In these realms, the study seeks to answer these research questions:

The academic contributions of this research are manifold. Firstly, instead of focusing solely on environmental degradation (proxied by CO2 emissions) as previous studies have (Muganyi et al., 2021; Tao et al., 2022; Udeagha and Muchapondwa, 2023), this study evaluates the impact of Fintech on both environmental degradation (proxied by the ecological footprint) and environmental quality (proxied by the Sustainable Development Index of Hickel (2020) in measuring the ecological efficiency of human development). This dual approach allows for a more comprehensive and multidimensional assessment and enables testing for the consistency of Fintech's environmental effects. Moreover, using these dependent variables provides a better foundation to explain the role of natural resources in moderating Fintech's environmental impact. Secondly, this is the first study to investigate the mediating role of natural resources in the pathway through which Fintech influences the environment. The most recent study by Dong et al. (2024) found that Fintech contributes to reducing CO₂ emissions and enhances resource rent (i.e. the value of natural resource extraction). However, no existing research has empirically examined the mechanism and extent to which natural resources mediate the environmental effects of Fintech. Thirdly, the study seeks to test the EKC hypothesis within the context of ASEAN countries by incorporating the presence and development of Fintech, using panel data from 2000 to 2021. Additionally, the study applies the robust Method of Moments Quantile Regression (MM-QR) estimation of Machado and Santos Silva (2019), which allows for the estimation of heterogeneous effects of Fintech across the conditional distribution of environmental indicators. MM-QR is particularly suitable for addressing issues of non-normality, autocorrelation and heteroskedasticity.

The remaining article is structured as follows. Section 2 provides a review of the relevant literature and hypothesis development of the Fintech-environment nexus. Section 3 outlines the research methodology and data. Section 4 presents the empirical results and discussion. Section 5 concludes with key findings and policy implications.

The financial and banking sector is currently undergoing a significant transformation in a positive direction, largely driven by the explosive growth of Fintech (Chen et al., 2021; Okere et al., 2025a). The term Fintech refers to technological advancements and intelligent solutions within the financial sector aimed at enhancing access to financial services and products (Puschmann, 2017; Mention, 2019). Fintech holds substantial potential in promoting environmentally friendly investment opportunities and mitigating the adverse impacts of climate change (Muganyi et al., 2021). For instance, blockchain and smart contracts, two foundational technologies of Fintech, offer promising capabilities for ensuring transparent monitoring and evaluation of environmental impacts (Parmentola et al., 2021; Macchiavello and Siri, 2022). These solutions contribute to expanding the accessibility of financial markets and facilitate the acceleration of investment decision-making processes for both enterprises and investors (Alt et al., 2018). Fintech plays a pivotal role in redirecting capital flows toward sustainable projects, thereby encouraging the use of renewable energy and enhancing resource productivity and the effectiveness of existing initiatives. Technological advancements in the financial and banking sector also help to reduce market barriers through digital transformation, thereby accelerating the implementation of sustainable finance and investment projects (Bian et al., 2023; Samour et al., 2024).

Recent studies have provided empirical evidence supporting the relationship between Fintech and environmental outcomes. Fintech has been shown to enhance the efficiency of financial operations and even influence consumption behavior across multiple levels, including the macro level (i.e. public procurement and policymaking) and the micro level (i.e. enterprises, investors and households) (Karim et al., 2022). Muganyi et al. (2021) found that Fintech contributes to a reduction in commercial carbon emissions in China. Using the difference-in-differences (DID) methodology, Sreenu (2022) demonstrated that Fintech solutions and green finance initiatives in India have the potential to mitigate carbon emissions. Similarly, through a comprehensive systematic review, Chueca Vergara and Ferruz Agudo (2021) concluded that Fintech facilitates environmentally friendly investment practices, thereby indirectly contributing to improvements in environmental quality. Tao et al. (2022) further explored the nexus between Fintech development and greenhouse gas (GHG) emissions, particularly carbon dioxide (CO₂), and concluded that the advancement of Fintech plays a positive role in accelerating the transition toward a global low-carbon economy.

Using the generalized method of moments (GMMs) and two-stage least squares regression techniques, Croutzet and Dabbous (2021) concluded that the economies of the Organisation for Economic Co-operation and Development (OECD) are progressively transitioning toward sustainable energy use in parallel with the development of Fintech strategies. The authors also observed that OECD countries possess significant potential to leverage Fintech in ensuring energy security amid geopolitical shocks. These findings align with those of Lobato et al. (2021), who asserted that investment, savings, and expenditure through Fintech mechanisms in the clean energy sector can contribute to improving environmental quality. Within the European context, Muhammad et al. (2022), employing Method of Moments Quantile Regression (MM-QR) and Instrumental Variables-GMM (IV-GMM), found that the Fintech industry, emerging as part of the Fourth Industrial Revolution, has a positive and enhancing effect on environmental performance.

Recently, Qin et al. (2023) found that Fintech and its various dimensions, including financial outreach, financial depth and financial digitalization, have a positive impact on environmental outcomes in China. In the Chinese context over the 2000–2020 period, Liu et al. (2024), using the quantile autoregressive distributed lag model, revealed that Fintech contributes to achieving green growth targets, while the continued exploitation of natural resources exerts adverse effects on ecological balance. A similar pattern has been observed in BRICS countries, where Udeagha and Muchapondwa (2023) demonstrated that Fintech exerts a favorable impact on environmental quality, whereas natural resource rents have a detrimental effect. Within the top six global manufacturing economies, Li et al. (2024), employing the GMM-panel vector autoregressive (GMM-PVAR) method, discovered that improvements in financial efficiency can reduce carbon emissions.

At the global level, Mertzanis (2023) provided evidence that Fintech has a positive, albeit modest, influence on both environmental and social performance. Notably, the effects of Fintech do not exhibit a nonlinear trend but instead show a lagged effect, meaning that investments in Fintech tend to yield more substantial environmental benefits over the long term. Further support for Fintech's role as a crucial tool in minimizing CO₂ emissions and achieving carbon neutrality has been found in various empirical studies, including Zhang et al. (2024) in Asian countries, Udeagha and Ngepah (2023) in BRICS economies, and Uddin et al. (2024) in the G20 context. In the ASEAN region during the 1990–2021 period, Tang et al. (2024) affirmed that Fintech can serve as a catalyst for promoting green investments and the development of sustainable projects. However, the continued reliance on and exploitation of natural resources may pose significant environmental threats to ASEAN countries. Based on the aforementioned literature review, this study proposes the following research hypotheses H1a and H1b:

The growing concern over ecological sustainability has rendered resource consumption a critical global issue (Ahakwa et al., 2023; Nwani et al., 2024; Okere et al., 2025b) and rapid economic expansion relies heavily on excessive exploitation of natural resources, significantly contributing to the imbalance between biodiversity and ecological footprint, and thereby undermining environmental sustainability. Both researchers and policymakers are currently striving to increase the availability of financial capital while ensuring a country's economic growth remains unaffected (Saqib and Shahzad, 2024). Natural resources rents refer to the intrinsic value of natural resources and represent the difference between actual revenue and the cost incurred from resource extraction used in the production process to generate income (Okere et al., 2024, 2026). In other words, resource rent is the monetary value paid for the use of a unit of natural resource and an excessive increase in the value of resource rent implies intensive exploitation of natural resources.

Danish et al. (2023) conducted a study in the United States using the ARDL method and found that natural resources rents act as an adverse factor in achieving sustainability as the financial attractiveness associated with resource exploitation leads to the over-extraction of fossil fuels, ultimately resulting in a significant increase in CO₂ emissions. Similar results were observed in South Africa, where Joshua and Bekun (2020) identified natural resource rents as one of the root causes of CO₂ emissions in the country. Mahmood and Furqan (2020) examined the relationship between resource extraction and GHG emissions in the Gulf Cooperation Council region using a spatial Durbin model, concluding that natural resource exploitation contributes to increased pollution levels in the area. Hassan et al. (2018) stated that both economic growth and natural resource exploitation have negative effects, increasing CO₂ emissions in Pakistan. According to Sun et al. (2020), excessive increases in resource rents lead to higher energy consumption, which ultimately causes pollution. These authors proposed the use of renewable energy to achieve environmental sustainability goals, while Dong et al. (2024), in a study of Asian countries, pointed out that Fintech can reduce CO₂ emissions and natural resource rent (i.e. reduce resource exploitation).

The literature suggests that an increase in resource rents may either be a barrier to carbon neutrality goals or potentially an opportunity for sustainable development when resource extraction is maintained at a moderate level. Notably, no research has yet deeply examined, within the ASEAN countries, the mechanism and the extent of the mediating effect of Fintech on the environment through either increasing or decreasing the actual exploitation of natural resources. Therefore, this study proposes Hypotheses H2 and H3:

Numerous studies have demonstrated that economic growth is a primary cause of environmental degradation (Stern et al., 1996; Alvarado and Toledo, 2016; Nuţă et al., 2024) as the trade-off between environmental quality and economic growth objectives is inevitable, especially in countries undergoing industrialization processes, such as ASEAN nations. However, the EKC hypothesis proposed by Grossman and Krueger (1995) suggests that the impact of economic growth follows an inverted U-shaped relationship, and that, specifically during the early stages of economic development, increasing economic activities lead to higher levels of environmental pollution. Nevertheless, once the economy reaches a certain threshold of growth, environmental pollution begins to decline, as the economy then places greater emphasis on sustainability aspects, welfare and cleaner production processes.

Numerous studies within ASEAN countries have confirmed the EKC hypothesis regarding the inverted U-shaped relationship between economic growth and environmental degradation (Amin et al., 2022; Kostakis, 2024; Adha et al., 2024; Yusma et al., 2024). Moreover, contemporary researchers have increasingly sought to validate the EKC hypothesis in various contexts, such as circular economy (Nguyen et al., 2024), globalization (Shahbaz et al., 2016), economic policy uncertainty (Udeagha and Muchapondwa, 2022) and digital transformation (Ullah et al., 2024). Through a comprehensive literature review, this study observes that prior EKC validation efforts predominantly focus on the environmental degradation aspect but pay limited attention to environmental sustainability. Specifically, economic growth tends to deteriorate environmental quality until a certain turning point, after which further economic growth contributes to environmental improvement. Furthermore, the potential nonlinear relationship between economic growth and the environment in the context of Fintech development remains underexplored. Therefore, the study proposes the following hypotheses H4a and H4b. Figure 1 visualizes the research framework employed in this study.

An appropriate model is crucial for accurately assessing the environmental impact of Fintech and preventing estimation bias caused by omitted variables. Therefore, this study employs the foundational Stochastic Impacts by Regression on Population, Affluence and Technology (STIRPAT) model developed by Dietz and Rosa (1997) and York et al. (2003), as represented in Equation (1). Recent studies, such as Shu et al. (2024), have also applied the STIRPAT framework to investigate the environmental effects of Fintech.

The article employs the ecological footprint (ecf) index to represent the environmental degradation, instead of CO₂ emissions or GHG emissions (Tang et al., 2024; Shu et al., 2024) with the ecological footprint providing a more comprehensive and holistic view of human impact on the environment. While CO₂ emissions only measure the amount of carbon released and mainly focus on climate change issues, the ecological footprint accounts for the total natural resources required to sustain human activities, including arable land, water, energy, and waste assimilation. This allows for a fuller reflection of environmental degradation across multiple dimensions, from resource consumption to the depletion of marine life, forests and ecosystems. On the other hand, the study utilizes Hickel's (2020) Sustainable Development Index (sdi) to represent the environmental quality to measure the ecological efficiency of human development. In essence, the research model includes two variables, namely the ecological footprint (ecf) and sustainable development index (sdi), to represent the environmental factor (I) in the STIRPAT model.

The main explanatory variable is Fintech, capturing the technology (T) aspect, in addition to renewable energy (rne) (Shu et al., 2024). For the population (P) factors, we employ urbanization (urb) and unemployment rate (une) as representative variables (Ortiz et al., 2022; Nuţă et al., 2024), while affluence (A) is commonly represented by gross domestic product (GDP) (Ullah et al., 2024). Due to the extensibility of the STIRPAT model, this research incorporates natural resources rents (nrr), which is considered a novel addition as current studies tend to include more factors to better clarify the determinants of the environment (Gani, 2021). Drawn from the EKC, the study introduces the squared term of GDP to examine the nonlinear mechanism in the economic–environment relationship under the Fintech development context, as defined by Equations (2) and (3).

In which i = 1, 2, 3, …, N denotes the observations for country i in year t = 2000, 2001, 2002, … 2021. $β$ denotes the regression coefficients estimated using linear regression methods. ε is the stochastic error term (constant) in the research model.

The study uses a sample of ten ASEAN countries (including Brunei, Cambodia, Laos, Indonesia, Malaysia, Myanmar, Philippines, Singapore, Thailand and Vietnam) for the period from 2000 to 2021 with the secondary research data collected from official databases. Table 1 summarizes information (name, symbol and measurement) of the investigated variables with natural logarithm transformation applied to variables with original units measured outside of index and percentages. The purpose of this transformation is to ensure that the interpretation of the results reflects how a 1% increase in the independent variable leads to a percentage change in the dependent variable. Specifically, the variable converted to a natural logarithm scale is GDP, which is measured in USD. The remaining variables, measured by indexes or percentages, will not be logarithmically transformed.

The dependent variables include the ecological footprint (ecf) and the ecological efficiency (sdi). These two variables represent different environmental aspects, namely, environmental degradation and environmental quality and, therefore, the expected effects of independent variables on these two dependent variables are expected to be opposite. The variable ecf data is collected from the Global Footprint Network, while sdi is aggregated from Hickel's Sustainable Development Index (sdi) (2020).

The main independent variable is Fintech. Currently, most studies measure the Fintech variable using a composite index constructed by principal component analysis from financial development indicators of the International Monetary Fund, Internet usage and mobile phone subscriptions from the World Bank data. Figure 2 illustrates the distribution of Fintech development levels across ASEAN countries. It can be observed that Singapore, Thailand, Malaysia and Vietnam have higher levels of Fintech development compared to other countries in the region.

The second independent variable is natural resources rents, measured as a percentage of GDP from total rental values of natural resources (including oil, gas, coal, minerals and forests). The data on natural resource variables are collected from the World Bank database and other variables serve as control variables in the research model, including unemployment rate (percentage of total labor force), urbanization rate (percentage of urban population over total population), GDP (USD at current exchange rates) and renewable energy (percentage of total energy use). Data for all control variables are sourced from the World Bank.

Based on discussions from previous studies, this research hypothesizes that Fintech has a positive impact on the ASEAN environment through its intermediary effect on reducing or enhancing the exploitation of natural resources. To test this hypothesis (H3 as proposed), the study applies Baron and Kenny's (1986) mediation testing procedure, as denoted by Equations (4)–(6):

Where in, natural resources rents (nrr) serve as the mediating variable M in the causal pathway from the independent variable X – Fintech to the dependent variables Y – ecological footprint (ecf) and ecological efficiency (sdi). First and foremost, we estimate the total effect of Fintech on environmental degradation and environmental quality, Equation (4). Subsequently, we estimate the effect of Fintech on natural resources rents as in Equation (5). Ultimately, as Equation (6), we regress Fintech on the environment, controlling for the presence of natural resources to examine changes in Fintech's environmental impact due to mediation through the enhancement or depletion of resource rents.

Before regression, the article performs various diagnostic tests to check for potential model deficiencies, including non-normal distribution, heteroskedasticity and autocorrelation. If the research data exhibit heteroskedasticity and autocorrelation, the panel-corrected standard error (PCSE) estimator is deployed to address these issues in the model. The rationale for using PCSE is that the dataset has a panel structure where the cross-sectional dimension N (number of countries) is smaller than the time dimension T (number of years). In addition, we explicitly account for potential endogeneity between Fintech and environmental outcomes. Such endogeneity may arise from reverse causality (i.e. environmental degradation or stricter environmental regulations may stimulate the adoption of Fintech solutions), omitted variable bias (i.e. institutional quality, technological readiness, or financial development affecting both Fintech and environmental performance), and measurement-related issues. When the data also exhibit non-normal distribution, heterogeneity and endogeneity, we further adopt the Method of Moments Quantile Regression (MM-QR) of Machado and Santos Silva (2019). The method is currently regarded as the most efficient approach, combining quantile regression with the GMM to simultaneously correct for non-normal distribution and endogeneity. Additionally, this estimator provides insightful findings on the magnitude of Fintech's environmental impact across quantiles.

Table 2 presents the descriptive statistics, consisting of 220 observations, covering 10 ASEAN countries from 2000 to 2021. The first dependent variable, ecological footprint (ecf), captures the environmental degradation dimension and the mean value of ecf is 2.7774 (measured in global hectares per capita), with a standard deviation of 2.1720, indicating significant variation in environmental degradation levels across countries. The minimum value is 0.8410 and the maximum is 8.8258. The skewness is 1.1515, suggesting a right-skewed distribution, while the kurtosis is 2.5629, indicating a distribution moderately close to normal.

The second dependent variable, ecological efficiency (sdi), represents the environmental quality dimension. The variable sdi averages 0.5922 with a standard deviation of 0.1873, showing less variation, and the values range from 0.1210 to 0.8130. The skewness of −1.2171 indicates a left-skewed distribution, while the kurtosis of 3.4356 suggests a slightly higher and sharper peak compared to a normal distribution. The main explanatory variable is Fintech development (fin) and the variable has a mean close to zero (3.53e−10) and a standard deviation of 1.5157, with values ranging from −2.4953 to 2.9469. The skewness is 0.0804, which is very close to zero, indicating a relatively symmetric distribution and the kurtosis is 1.9534, which is lower than 3, implying a flatter and more spread-out distribution than the normal distribution.

Before conducting multivariate regression, the research employs several diagnostic tests to examine the presence of potential econometric issues in the model and dataset across ASEAN countries. Table 3 presents the results of normality and endogeneity tests for the research variables. The Jarque–Bera test is used to assess the normality of the data distribution and the results reveal that none of the variables follows a normal distribution, indicating a serious issue that must be addressed. This is because conventional estimation methods typically rely on the assumption that the variables are normally distributed and use the mean to estimate regression coefficients. When the data deviates from normality and contains outliers, traditional regression results may be biased and unreliable. Figure 3 illustrates the data distributions of the two dependent variables, ecf and sdi. The data points for both variables do not align along a straight line, further confirming non-normality.

In addition, we present the results of diagnostic tests for other econometric issues, including slope heterogeneity based on the test proposed by Pesaran and Yamagata (2008), heteroskedasticity using the Breusch–Pagan test, and autocorrelation using the Wooldridge test. The results indicate that both research models suffer from significant econometric problems, including slope heterogeneity (due to non-normal distribution), heteroskedasticity and autocorrelation.

To address these issues, the author initially applies the PCSE estimation technique to correct for heteroskedasticity and autocorrelation. The PCSE estimator is preferred because it is suitable for panel datasets where the number of cross-sectional units (N) is smaller than the number of time periods (T). To complement the PCSE results, the article employs the method of moments quantile regression (MMQR) proposed by Machado and Santos Silva (2019) as this approach combines quantile regression to handle non-normality and slope heterogeneity with GMM techniques to address potential endogeneity. Additionally, MMQR yields more robust estimates when the dataset exhibits heteroskedasticity and autocorrelation.

Tables 4 and 5, respectively, present the direct effects of Fintech on ecological footprint (ecf) and ecological efficiency (sdi) of ASEAN countries, using the PCSE and MM-QR estimation methods. The direct effect refers to the impact without the mediation of natural resource rents (nrr).

Table 9 reveals that Fintech (fin) exerts an inverse effect on ecological footprint (a proxy for environmental degradation) under both estimation methods. Specifically, a 1% increase in the level of Fintech development in ASEAN countries leads to a 0.4411% reduction in environmental degradation (using the PCSE estimator), and a reduction ranging from 0.4264% to 0.4569% (using the MM-QR estimator), all statistically significant at the 1% level. An interesting finding is that the coefficients under the MM-QR method decline gradually from the tenth to the 90th quantile, suggesting that the environmental mitigation effect of Fintech is stronger in countries with lower levels of pollution. A plausible explanation is that in less polluted countries, the adoption of Fintech may facilitate more efficient resource utilization, improve operational effectiveness, and reduce pollution-generating activities (Hopalı et al., 2022). In contrast, in more polluted countries, the impact of Fintech may be constrained by structural factors such as the dominance of polluting industries, economic systems or outdated infrastructure (Sunanda and Joseph, 2023).

Similarly, Table 10 indicates that Fintech has a proportional effect on improving ecological efficiency (a measure of environmental quality) for both estimators. This finding aligns with expectations, as the effects of Fintech (fin) are expected to be opposite in the cases of environmental degradation and environmental sustainability. Specifically, in ASEAN countries, a 1% increase in Fintech development leads to a 0.0152% improvement in environmental quality under the PCSE method, and an increase ranging from 0.0142% to 0.0238% across the 30th to 90th quantiles under the MM-QR method. A noteworthy point is that the regression coefficients in the MM-QR method increase gradually from the 30th to the 90th quantile, implying that the beneficial effect of Fintech on environmental quality becomes more pronounced and significant in countries with higher environmental quality. This can be attributed to the fact that these countries likely possess more supportive infrastructure and policy frameworks for the implementation of Fintech, thereby enhancing the effectiveness of Fintech in contributing to sustainable solutions (Almansour, 2023; Le et al., 2026).

Conversely, in countries with very low environmental quality (i.e. 10th and 20th quantiles), Fintech appears to have no significant impact on environmental quality. One plausible explanation is that these countries face more severe and structurally complex environmental challenges, where the marginal contribution of Fintech alone is insufficient to generate measurable improvements. In such contexts, governments may need to complement Fintech initiatives with additional policy measures, such as enhancing environmental regulatory frameworks or investing in green technologies, to achieve more meaningful environmental outcomes. These findings highlight the conditional nature of Fintech's environmental impact and imply that policy effectiveness depends on the broader socio-economic and institutional context. They are also consistent with the results of Saqib and Shahzad (2024), who employed the load capacity factor as a proxy for environmental quality. Therefore, the evidence provides partial support for Hypothesis H1b, particularly in middle- and higher-quantile contexts rather than universally across all countries.

Subsequently, to examine the mediating role of natural resource rents (nrr) in the environmental impact of Fintech, the study has to assess the effect of Fintech on natural resource rents and ensure that the coefficient of Fintech is statistically significant. The results in Table 6 indicate that Fintech has the potential to reduce natural resource rents in ASEAN countries. Specifically, at the 1% level of statistical significance, a 1% increase in Fintech development leads to an average reduction of approximately 3.4351% in natural resource rents under the PCSE method, and a reduction ranging from 1.7024% to 5.4273% across the tenth to 90th percentiles under the MM-QR. Figure 4 depicts the quantile distribution effects of variables on natural resources.

These findings imply that the impact of Fintech in reducing natural resource exploitation is more pronounced in countries with higher levels of resource rents. ASEAN countries that heavily rely on natural resource exploitation are likely to experience more substantial benefits from Fintech in terms of reducing dependence on resource rents. The digital transformation and promotion of technological solutions in the financial and banking sector can contribute to lowering resource exploitation by improving resource-use efficiency and fostering the development of sustainable business models (Delina, 2023). Fintech can accelerate the implementation of environmentally friendly solutions such as green finance, sustainable investment, and digital financial services that enhance the capacity to manage and utilize resources more efficiently. This finding may encourage the sustainable use of natural resources and reduce reliance on extractive activities (Dong et al., 2024).

Moreover, the study identifies both linear (from the tenth to the 30th percentiles) and nonlinear (from the 40th to the 90th percentiles) effects of economic growth on natural resource rents. For countries with low levels of resource extraction, economic growth positively impacts the value of natural resource rents. This can be explained by the fact that initial economic development often relies heavily on natural resource exploitation to drive production and GDP growth. At this stage, resource extraction plays a crucial role in supplying raw materials and energy inputs necessary for the industrial and economic development of the country. Therefore, as countries begin to experience economic growth, the demand for natural resources rises, leading to increased resource extraction. For ASEAN countries with moderate to high levels of resource extraction, the effect of economic growth on natural resource rents is initially positive, meaning economic growth tends to increase the level of resource exploitation. However, when the GDP per capita reaches a certain turning point, economic growth starts to have a negative impact, reducing natural resource extraction. This turning point is estimated to fall within the range of GDP per capita between USD 100,624 and USD 931,567.

The MM-QR regression results indicate that the more a country depends on natural resources, the higher the GDP threshold required to reach the turning point where economic growth begins to reduce resource extraction. This implies that countries with a high level of resource dependency will take longer to achieve a stage of sustainable and resource-friendly development. In other words, ASEAN countries with intensive natural resource use must undergo a longer and more challenging transition to shift toward a resource-efficient economy, as they have historically relied on natural resources as a key driver of economic growth. This delay may pose long-term environmental and sustainability challenges if appropriate policies and strategies are not implemented to facilitate the transition toward less resource-dependent economic sectors.

Ultimately, the study carried out a regression analysis, incorporating the natural resources rents (nrr) to clarify the impact path on the environmental impact of Fintech development. Table 7 presents the mediating role of natural resources in the nexus between Fintech and ecological footprint. It is evident that the natural resource rents (nrr) have a positive effect, meaning they increase the ecological footprint in ASEAN countries, which implies that, as the natural resource rents rise, the level of environmental degradation also worsens. Holding other factors constant, a 1% increase in natural resource rents (relative to GDP) corresponds to an increase in environmental degradation ranging from 0.0347% to 0.0564%. This clearly indicates that natural resource extraction in ASEAN countries contributes to increasing environmental pressures, leading to ecosystem degradation.

Table 7 and Figure 5 illustrate that the negative impact of natural resources rents diminishes in countries with higher levels of environmental pollution. This can be explained by the fact that countries with higher pollution levels have likely undergone phases of excessive resource exploitation, and the marginal negative effect of additional resource extraction may be less pronounced compared to countries with lower pollution levels. Another plausible reason is that countries with higher pollution levels often face international community pressure, environmental regulations or have reached a saturation point in resource extraction, making further impacts from resource exploitation less significant. The findings are consistent with Ullah et al. (2021) and Alvarado et al. (2021).

Moreover, in Table 7, the coefficients of Fintech indicate an inverse effect, meaning that Fintech contributes to reducing environmental degradation. However, the magnitude of this effect changes when the mediating role of natural resources is introduced, specifically the extent of the mediation effect that has been calculated and presented in Table 8. Fintech exerts an inverse impact on environmental degradation by positively reducing the resource rent value. In other words, the development of Fintech effectively mitigates the ecological footprint (ecf) in ASEAN countries through mechanisms that enhance resource use efficiency and conserve resource capital, thereby reducing the exploitation of natural resources. Fintech not only plays a role in improving financial efficiency but also supports more sustainable use and management of natural resources. As ASEAN countries adopt and develop Fintech, resource management becomes more effective, leading to a reduction in unnecessary resource extraction and consumption. Another point to note is that the mediating effect of natural resources decreases (from 49.726% to 20.802%) as the level of environmental pollution in a country increases. This can be explained by the fact that countries with higher pollution levels have often already exploited their natural resources to the maximum extent or have undergone significant resource degradation, making the impact of Fintech through the mechanism of reducing resource extraction less effective. Therefore, the study validates Hypothesis H3.

The study continues to test H3 concerning the environmental quality dimension, with the dependent variable being ecological efficiency (sdi). The mediating role of natural resources in the impact of Fintech on environmental quality is presented in Table 9 and Figure 6. Specifically, an increase in the natural resources rents has a positive effect on environmental quality, although the estimated coefficients range only between 0.0017 and 0.0028 (from the median to higher quantiles), which indicates that the increase in natural resource extraction significantly affects only countries with moderate to high environmental quality. The reason may lie in the fact that these countries possess better infrastructure and environmental management systems, allowing them to utilize natural resource inputs more efficiently. The development of Fintech can act as a facilitating factor, helping to optimize the extraction and management processes of natural resources, thereby contributing to the improvement of environmental quality. Conversely, in countries with low environmental quality, the impact of natural resource extraction is not clearly evident, which suggests these countries may lack effective environmental management policies or the capacity to utilize Fintech appropriately to optimize resource value. Moreover, resource extraction without adequate consideration of environmental quality may lead to adverse consequences, such as environmental pollution and biodiversity loss (Niu et al., 2024). Therefore, this study further confirms Hypothesis H2. While increased natural resource extraction or growing resource scarcity may exacerbate environmental degradation, it may also simultaneously contribute to improvements in ecological efficiency.

Table 10 presents the extent of the mediating effect of natural resources in the relationship between Fintech and environmental quality (ecological efficiency). The mediating role of natural resources in the impact of Fintech on environmental quality occurs only in countries with medium to high ecological quality (from the 40th to the 90th percentile). This finding suggests that, in countries with better environmental quality, Fintech can be applied more effectively, leading to optimized natural resource management and improvements in environmental quality. For example, when natural resources are well managed through Fintech technologies and services, they can promote sustainable behaviors, thereby generating positive environmental outcomes.

Moreover, the magnitude of the mediating effect gradually increases from 39.437% to 64.706% across the quantiles, providing strong evidence that natural resources act as a key transmission channel through which Fintech influences environmental quality and this pattern indicates that the impact of Fintech is not only direct but also operates indirectly by shaping how natural resources are utilized. Specifically, Fintech enhances the efficiency of resource allocation by reducing information asymmetry and transaction costs, thereby facilitating investment flows into activities related to resource extraction, processing, and energy use (Okere et al., 2025b). Through this mechanism, natural resources become an important conduit linking Fintech development to environmental outcomes. When resource use is directed toward cleaner production processes, the indirect effect of Fintech contributes to environmental improvement. The increasing magnitude of the mediating effect across quantiles suggests that, as environmental quality improves, the role of natural resources in transmitting the benefits of Fintech becomes more pronounced. In such contexts, resource utilization is more aligned with sustainability-oriented activities, allowing Fintech to amplify its positive environmental impact through this channel. Conversely, in lower quantiles where environmental quality is weaker, the mediating role of natural resources is less effective. In these cases, Fintech-driven financial flows may not sufficiently redirect resource use toward environmentally friendly activities, limiting their indirect contribution to environmental quality. Accordingly, the results provide robust empirical support for the mediating role of natural resources and strongly validate Hypothesis H3.

Concerning the EKC hypothesis in ASEAN countries, the study observes that the traditional inverted U-shaped EKC does not appear to hold in the context of Fintech development. The MM-QR and PCSE estimation, when using ecological footprint (ecf) as the dependent variable (in Table 7), reveal that GDP and the square of GDP have statistically significant negative and positive effects, respectively, across all percentiles. This finding suggests that, under the influence of Fintech, the EKC hypothesis may undergo a structural shift. Specifically, economic growth initially reduces ecological degradation; however, beyond a certain threshold of GDP per capita, further growth tends to intensify environmental degradation and pollution. The turning points are calculated and presented in Table 8, ranging from USD 167,503 to USD 260,442 – levels below the average GDP per capita in the research sample. Many ASEAN countries still have the potential to achieve economic growth without triggering excessive environmental degradation and, furthermore, the threshold at which GDP growth begins to worsen environmental conditions increases in countries with more severe environmental degradation and where the role of natural resources diminishes. Fintech can offer numerous opportunities for investment in green and sustainable solutions; however, it may also increase environmental pressure if economic activities are expanded without adequate environmental safeguards. Therefore, Fintech must be accompanied by robust environmental policies to ensure that economic growth does not come at the expense of environmental quality. As a result, the study rejects Hypothesis H4a regarding the conventional inverted U-shaped EKC and concludes that, in the context of Fintech, the EKC follows a U-shape, with a turning point at a relatively low level of GDP per capita.

When ecological efficiency (sdi) is used as the dependent variable (in Table 9), the linear relationship between economic growth and environmental quality under Fintech development contradicts the traditional EKC hypothesis. Specifically, GDP exerts a positive impact on environmental quality, while the squared term of GDP has a negative effect, and this indicates that in initial stages, GDP growth may enhance environmental quality through the adoption of sustainable Fintech such as digital payments, accessible online lending and investments in renewable energy. However, once a certain GDP threshold is reached, the positive impact diminishes, leading to a decline in environmental quality and this result reflects that unsustainable development may occur when economic growth exceeds the environmental management capacity of countries. Table 10 shows that the threshold at which GDP growth transitions from positively to negatively affecting environmental quality ranges from USD 981,190 to USD 1,277,444. A key insight is that countries with higher environmental quality can sustain the environmentally benign phase of Fintech development longer than countries with lower environmental quality, which highlights that effective environmental policies, in tandem with sustainable Fintech development, can enable countries to maintain economic progress without harming the environment. As a result, we reject Hypothesis H4b as the EKC for ecological efficiency follows an inverted U-shape.

This study provides a fresh examination of the interrelationship between Fintech, economic growth and the environment (including environmental degradation and environmental quality) across ASEAN countries during the 2000–2021 period. By employing advanced quantitative methodologies such as PCSE and MM-QR, the study offers robust empirical evidence on the role of Fintech in influencing environmental outcomes. Initially, Fintech exerts a significantly advantageous impact on the environment, notably through the reduction of ecological footprints and the enhancement of eco-efficiency. These results underscore the potential contribution of Fintech to sustainable development objectives, particularly when accompanied by effective natural resource management policies. Moreover, the mediating role of natural resources in the Fintech–environment nexus is empirically validated, suggesting that efficient governance and utilization of natural capital are essential for amplifying the environmental benefits of digital financial development. Specifically, the mediating effects range from a decrease of 20.802% to 49.726% for environmental degradation and an increase from 39.437% to 64.706% for environmental quality across different quantiles. Importantly, the empirical results deviate from the traditional EKC hypothesis, indicating a nonlinear U-shaped curve with ecological footprint and an inverted U-shaped curve with ecological efficiency. While economic growth initially contributes to environmental improvement, surpassing a certain GDP level reverses this trend, exacerbating ecological deterioration. For instance, ecological footprints decline with early-stage GDP growth but subsequently increase after crossing a critical income threshold. However, in contrast, eco-efficiency initially rises with GDP but diminishes beyond a relatively low GDP level. These dynamics suggest that digital transformation through Fintech necessitates nuanced and adaptive policy responses to ensure a sustainable balance between economic expansion and environmental preservation.

Generally, the study sheds light on the complex interactions between Fintech, natural resources and environmental sustainability in the ASEAN context. The findings highlight the imperative of designing forward-looking, integrated policy strategies that align digital financial innovation with environmental stewardship. Firstly, ASEAN policymakers should foster a supportive legal and institutional framework to encourage Fintech development as a tool for achieving environmental and sustainability goals. This includes integrating Fintech into natural resource management, enhancing transparency in environmental monitoring and incentivizing green Fintech projects through regional cooperation and knowledge sharing. Secondly, to amplify the environmental benefits of Fintech, policymakers must pair Fintech advancement with effective natural resource governance, which involves clear regulatory frameworks, tax incentives, infrastructure support and the use of digital technologies for real-time monitoring and sustainable resource utilization. Finally, given the nonlinear relationship between economic growth and environmental quality, ASEAN countries should strategically deploy Fintech to boost eco-efficiency in early growth stages, while implementing targeted fiscal, credit, and green investment policies beyond certain GDP thresholds to mitigate environmental degradation and ensure sustainable development.

This study is not without limitations. Firstly, this study primarily focuses on the impact of Fintech on environmental quality, without fully accounting for other influential factors such as environmental policies, green technology investment or broader macroeconomic variables. Secondly, future research could extend the analysis to a global scale to examine the potential convergence of Fintech development. This direction may uncover mechanisms for optimizing economic–environmental policies based on convergence patterns and cooperation among countries with similar Fintech maturity levels. Finally, beyond natural resources, various mediating factors such as government regulation, technological infrastructure, the development of the digital economy and public environmental awareness may also shape the relationship between Fintech and environmental quality and warrant deeper investigation.