Assessing the role of Environmental Legislation in Mitigating Climate Risk in GCC: A quantile ARDL approach

Assessing the role of Environmental Legislation in Mitigating Climate Risk in GCC: A quantile ARDL approach

Assessing the role of Environmental Legislation in Mitigating Climate Risk in GCC: A quantile ARDL approach

28 views
Categories: Articles
Authors: Mohamed Sami Ben Ali, Alanoud Al-Maadid, Kamal Si Mohammed
College of Business & Economics, Qatar University, Doha, Qatar
Received 6 March 2025, Revised 30 August 2025, Accepted 1 October 2025, Available online 2 October 2025, Version of Record 9 October 2025.

Climate risk manifested through rising temperatures, extreme weather events, and ecological degradation poses a growing threat to environmental and economic stability, particularly in resource-dependent regions such as the Gulf Cooperation Council (GCC). This paper aims to empirically examine the short- and long-term effects of environmental legislation on climate risk, considering the roles of eco-friendly regulations, Information and Communication Technology (ICT), and domestic extraction in the GCC countries. Using the panel Quantile ARDL technique, the findings suggest that climate legislation effectively mitigates climate risk. This effect is notably increased at the upper quantile in the long run. A similar effect of ICT indicates its vital role in managing climate risk. On the contrary, GDP significantly exacerbates climate risk, especially at lower and median quantiles in the short term. The results demonstrate the uni-directional effect of climate legislation and ICT on climate risk, illustrating the effectiveness of ICT and regulatory measures in addressing climate challenges. These findings underscore the significance of the legislative framework in mitigating climate risk, underscoring the need for new policy interventions to address climate vulnerability.

1. Introduction

The most pressing unresolved issue of our time is that we are causing increasing environmental harm through our manufactured interventions, compared to our efforts to curb atmospheric emissions. The mismatch in the consumption of ecosystem services compared to our cumulative efforts to restore them, exorbitant consumption of fossil-based hydrocarbons, deforestation, building embankments to alter river flows, and unwillingness to pay for the restoration of ecological balance are some of the critical factors that contribute to the ongoing climate risk globally (Mathews et al., 2025Welsch, 2025). However, desperate attempts have been made to restore environmental sustainability through various international treaties (i.e., Rio Climate Summit, Glasgow Climate Pact, etc.) and the platform of the Intergovernmental Panel on Climate Change (IPCC) of the United Nations. The Conference of Parties (COPs), a wing of the United Nations climate change forum, is also vocal about the global adverse effects of climate change. In this context, climate risk refers to the potential for destructive impacts from climate change, including physical risks, such as damage from flooding, hurricanes, and drought, as well as transition risks, including policy changes and technological shifts (van Zanten and Putintseva, 2025Ben Ali & Al-Maadid, 2025). These risks can have a severe impact on financial stability, infrastructure, food security, and public health, necessitating robust policy frameworks to mitigate their effects (Lee et al., 2024Sun et al., 2022). The need for climate legislation, regulations, and policies that can reduce emissions and promote sustainable practices has been one of the most widely recognized aspects of the global climate change narrative. However, the degree to which such legislation translates into tangible action and can reduce climate risk is still insufficiently understood. The interplay between climate legislation and environmental impacts lies at the heart of national and international efforts to mitigate greenhouse gases (GHGs) and promote sustainability. This study examines the impact of environmental legislation on climate risk, providing valuable insights into the multifaceted nature of climate governance and its effectiveness.
Climate legislation encompasses the various laws and regulations that aim to address the complex problems of climate change, including emission reduction mandates, renewable energy targets, carbon pricing mechanisms, and environmental policies that embed sustainability in economic and social governance. The link between the ecological legislation enacted and its impact on climate risk is unclear. Critical determinants include the strictness of the laws, the nature of their enforcement, and the social and economic context of the territory (Trotter et al., 2022). In addition, the origins of rising human-induced and natural climate risks introduce additional complexity to this relationship.
Recent research highlights the range and depth of legislative measures enacted worldwide to mitigate environmental damage and promote renewable energy (Al-Maadid et al., 2025bKorkut Pata et al., 2025Radulescu et al., 2025). Chidiogo Uzoamaka Akpuokwe et al. (2024) conducted a systematic review of global legislative actions related to clear emissions reduction targets, renewable energy targets, and carbon pricing mechanisms, which are critical. They concluded that international cooperation and the effective transfer of knowledge are crucial in enhancing the multidimensional aspects of climate change legislation. In a similar effort, Liu and Feng (2023) demonstrate that the differences in promoting renewable energy primarily lie in whether national energy policy focused on such promotion was enacted through legislation. They observe that legislative efforts tend to have a more substantial impact in higher-income countries. Political institutions play a crucial role in designing and implementing climate and clean energy policies (Dent, 2018). Omri and Ben Jabeur (2024) and Omri and Boubaker (2024) argue that strong political institutions are crucial for enforcing the enactment of climate change legislation, which is essential for the transition to renewables and reducing climate risk.
Environmental legislation has played a crucial role in addressing the challenges of climate change and promoting the development of renewable energy. The UK Climate Change Act of 2008 is a model of framework legislation that has led to similar initiatives in other countries, such as Ireland and Mexico, and was an essential step toward demonstrating the necessity of national action within a global framework (Averchenkova et al., 2021Dobush et al., 2022). According to Torney (2017), the UK environmental legislation has significantly influenced other countries in developing climate regime policies. In this context, Eskander et al. (2021) examined how UK national legislation is increasingly aligning with a globally recognized climate crisis, such that almost every country now has some form of climate legislation. Puri et al. (2023) highlighted the role of regulating contaminants and their emerging regulation in both developed and developing economies in achieving water sustainability and reducing human health risks from environmental pollutants. Moreover, Eskander and Fankhauser (2023) studied the impact of national climate legislation on trade-related carbon emissions. They found that the effect of such laws on domestic carbon emissions is significantly negative, suggesting that legislative measures do not induce higher international emissions, an effect often associated with unilateralist climate policy actions. China recently enacted the Environmental Protection Law (EPL) (Zhang et al., 2016). However, despite passing this law, many implementation challenges remain, indicating that the effectiveness of environmental governance requires more than legislation and necessitates vigorous enforcement and accountability mechanisms. Zhang et al. (2019) investigated the impact of environmental regulation on pollution in China, examining the synergistic effects of industrial agglomeration through Bayesian posterior probability and optimal model structure across 30 provinces from 2003 to 2016. They found that environmental regulations reduced pollution overall in the eastern, central, and northeastern areas but had the opposite effect in the western region, where pollution increased. However, industrial agglomeration, combined with environmental regulation, worsened pollution in most areas but improved it in the northeast. Similarly, Zhou et al. (2023) examined how urban environmental legislation in China has significantly improved local green total factor energy efficiency, influenced by industrial upgrading and the innovation effect, while Chen et al. (2023) highlighted the importance of local government actions aimed at achieving broader environmental objectives. The policies outlined in Canada exemplify the integration of environmental and legislative strategies to promote forest and landscape restoration. Mansuy et al. (2022) discussed policy and legislative support for Forest Landscape Restoration (FLR) in Canada, highlighting how legislation promotes sustainable land management that addresses the challenges of ecological degradation, climate change, and biodiversity loss in Canada. Papu-Zamxaka et al. (2010) highlighted the gap between the enactment of climate legislation and tangible environmental outcomes in South Africa, emphasizing the need for legislation and regulation to be effectively enforced. They noted that contaminants such as mercury have persisted in the environment despite laws banning their use.
The present study examines this effect on the Gulf Cooperation Council (GCC) countries for several compelling reasons. Firstly, the GCC economies heavily depend on oil and gas exports, and studying their climate risks can provide valuable insights into global energy strategies. Secondly, the extreme climatic conditions in these countries, such as severe heat and aridity, underscore the need to understand and mitigate climate risks. Given their recent policy pushes for economic diversification and carbon footprint reduction, GCC countries have also presented a valuable testing ground for the impact of such legislation. Third, the legislative measures taken by the GCC can significantly impact global oil markets, providing incentives for energy prices and investment trends that extend far beyond the GCC region, making the study of climate legislation relevant for international stakeholders and international energy firms. Fourth, due to its significant financial resources, the area could serve as a model for the adoption of a green ICT frontier, presenting an opportunity to consider how legal environments promote the transition toward green technologies. Finally, the unprecedented rapid urbanization of GCC countries presents unique challenges and opportunities for sustainable urban and land planning, urging us to explore how climate legislation can inform urban adaptation strategies to mitigate climate risks.
Fig. 1 illustrates the regional distribution of climate risk across the GCC, plotting climate risk index values between 55.12 and 119.26. The darkest shade of blue is concentrated over Saudi Arabia, indicating the region with the most climate risk. In contrast, the lighter shades surrounding the UAE, Oman, and Kuwait indicate relatively lower-risk areas. These significant differences in the spatial variability of climate risk indicate that central regions are more vulnerable to climate-related issues. However, Fig. 2 shows the distribution of climate legislation across the GCC from 2005 to 2024. The darker blue corresponds to more stringent and stronger climate laws, with Saudi Arabia standing out as the only country with a brighter blue. This implies that for Saudi Arabia, climate policies have become increasingly stringent, which may result from the country's sustainability commitments made as part of its Vision 2030 diversification goals. Similarly, the regions marked in relatively lighter shades, including the UAE, Oman, Qatar, Kuwait, and Bahrain, indicate more recent progress in adopting substantial climate legislation.
Fig. 1
  1. Download: Download high-res image (208KB)
  2. Download: Download full-size image

Fig. 1. Annual data of climate risk index in GCC countries during the period of estimation (2004-2023).

Source: Germanwatch dataset.
Fig. 2
  1. Download: Download high-res image (154KB)
  2. Download: Download full-size image

Fig. 2. Annual data of climate legislation in GCC countries during the period of estimation (2004-2023).

Source:ECOLEX Environmental Law Database (FAO).
To the best of our knowledge, this is the first study to assess the effect of climate legislation on climate risk, making essential contributions to both the academic literature and practical efforts to implement climate action. This paves the way academically for exploring how practical legal tools are used to address climate impacts, providing an extensive basis for future study. Secondly, the study advances the academic debates on the relationship between climate change and governance, contributing empirical evidence on the link between droughts, extreme weather, legislation, and eco-friendly policy changes. In addition, this study helps raise public awareness of the importance of legal measures in action against climate change, which can boost public support and community engagement in responding to new regulations. Thirdly, this research contributes to the global discussion on environmental justice by demonstrating how legal action can mitigate climate risks, fostering fairness and equality in ecological impacts. As the first GCC-relevant study of its kind, this work provides a key point of reference to support global action in understanding and optimizing legislative management of climate risk. Fourth, this study makes a substantial contribution to achieving four UN Sustainable Development Goals related to climate risk issues, providing a holistic assessment context that aligns with global efforts for sustainable development. The research has a direct relationship with SDG 13 (Climate Action), as it assesses the effectiveness of climate legislation and how different climate legislation impacts risk, providing crucial insights into urgent steps to mitigate climate change. The results further contribute to achieving SDG 11 (Sustainable Cities and Communities) by harnessing the transformative potential of incorporating climate risk management elements into urban planning for sustainable and resilient city development. Moreover, the study promotes SDG 9 (Industry, Innovation, and Infrastructure) due to the role of technological development in climate risk mitigation, encouraging sustainable industry practices and innovation. It supports SDG 7 (Affordable and Clean Energy) because it emphasizes how ICT will drive the development of renewable energy solutions, a crucial step toward the energy systems we desire. Domestic material consumption impacts are relevant to SDG 12 (Responsible Consumption and Production) and contribute to policies that promote sustainable consumption patterns and efficient resource use. This study is structured as follows. Section 2 describes the methods used for the data collection. Section 3 describes the empirical results. Lastly, Section 4 presents conclusions and policy implications.

2. Data and Methodology

2.1. Data

This study examines the effects of Climate legislation (LEGS), Information and Communication Technology (ICT), economic growth (GDP), Domestic Extraction Used (DMC), and Urbanization (URB) on the Climate Risk Index (CRI). This study selected GCC countries: the United Arab Emirates, Qatar, Bahrain, Oman, Kuwait, and Saudi Arabia. Due to data constraints, the study was conducted from 2005 to 2024 using an unbalanced panel with a total of 110 observations. Table 1 provides a brief account, including abbreviations and summary statistics, for these variables. Fig. 3 depicts the data trend. Following the logic of this research, the model was developed as follows:

Table 1. Table of Variables.

VariableVariableExplanationEmpty CellEmpty Cell
Climate risk indexCRIIndex (Ln)Germanwatchhttps://www.germanwatch.org/en/cri
Climate legislationsCLGSNumber of laws enactedECOLEX Environmental Law Database (by FAO)https://www.ecolex.org
Domestic extraction usedDMCMillions, Tonnes (Ln)OECD.Stat (Organisation for Economic Cooperation and Development)https://stats.oecd.org
GDP per capitaGDPGDP Per capita (Ln)World Development Indicators (World Bank)https://databank.worldbank.org/source/world-development-indicators
UrbanizationURBUrban Population to Total Population (%)World Development Indicatorshttps://databank.worldbank.org/source/world-development-indicators
Information and Communication TechnologyICTMobile cellular subscriptions (per 100 people/Ln)World Development Indicatorshttps://databank.worldbank.org/source/world-development-indicators
Climate regulationREGNumber of legislationsECOLEX Environmental Law Database (FAO)https://www.ecolex.org
Fig. 3
  1. Download: Download high-res image (869KB)
  2. Download: Download full-size image
Fig. 3
Where  is the slop intercept,  Reveals slope coefficients of explanatory variables. While  is the white noise error term.
This study utilizes the Climate Risk Index introduced by Germanwatch (Huang et al., 2018Ozkan et al., 2022). The index measures annual and long-term climate risks resulting from weather-related disasters, such as storms and flooding, basing its score on four key indicators: total and per capita deaths, economic losses in US dollars, and losses as a percentage of GDP. To enable empirical analysis, annual and long-term indices are inverted, with lower induced risk scores corresponding to higher risk levels (Ozkan et al., 2022Li et al., 2024). The index measures annual and long-term climate risks resulting from weather-related disasters, such as storms and flooding, basing its score on four key indicators: total and per capita deaths, economic losses in US dollars, and losses as a percentage of GDP. To enable empirical analysis, annual and long-term indices are inverted, with lower induced risk scores corresponding to higher risk levels (Ozkan et al., 2022). The first explanatory variable is the climate legislation, which captures the laws and policies a country has enacted to combat climate change, focusing on reducing emissions and promoting sustainability. To ensure the robustness of our analysis, the climate regulations used were sourced from the FAO's FAOLEX database. While this study primarily uses the quantity of enacted legislation as a proxy for climate governance, we acknowledge that the effectiveness of legislation also depends on its strength, enforcement, and institutional quality. Although detailed legislative strength indices are not uniformly available across GCC countries for the study period, the use of the number of enacted laws provides a consistent and comparable proxy across countries. This limitation is mitigated by complementing the quantitative measure with robust econometric modeling and robustness tests, which help capture the aggregate effect of legislative frameworks on climate risk. The third variable is domestic extraction, which refers to the natural resources extracted from a country's environmental stocks, including minerals, fossil fuels, and biomass (Li et al., 2025Bakkar et al., 2024Mohammed et al., 2025Aghaloo and Sharifi, 2024). The next variables of the study are GDP and URB. The URB variable is measured by the ratio of urban population to total population, and the data is compiled from the World Bank. Urban areas have more developed infrastructure and energy supply distribution systems than rural areas, which can impact the energy transition (Maleknia and Enescu, 2025Xu et al., 2024). Extensive studies employed GDP as a key variable when examining its effect on climate risk, given its significant impact on the economic landscape of GCC countries (Byrne and Vitenu-Sackey, 2024Ulussever et al., 2024Zenios, 2024Al-maadid et al., 2025aBen Ali & Lechman, 2024). The GDP figures were obtained from the World Bank. Another variable in this study is ICT, measured by cellular phone subscriptions, as per data from the WDI. This measure indicates ICT and progress, allowing for an assessment of how technological advances contribute to mitigating climate risk in the GCC countries (Ulussever et al., 2024Adebayo et al., 2025).

2.2. Methodology

The empirical strategy consists of several panel diagnostic tests to ensure robust estimation, followed by quantile-based dynamic modeling. We employ Cross-Sectional Dependence(CSD) and slope heterogeneity tests to validate the panel's structure. Stationarity and cointegration tests confirm that our series is suitable for long-run modeling using the QARDL framework. These steps ensure methodological rigor and justify the model selection.

2.3. CSD Test

To address the study objectives, the CSD test is used to identify disparities between GCC countries, employing Breusch and Pagan (1980) and Pesaran (20042006) tests. For an unbalanced panel, the Pesaran CD test can be applied to assess the degree of CSD. One simple test for CSD in panel data is the Breusch-Pagan Lagrange multiplier (LM) test. A different approach is presented in the Pesaran-scaled LM test, which is more robust, allowing for CSD in an unbalanced panel (Xie et al., 2024). However, an appropriate test is needed for long-time dimensions and CSD to ensure the accuracy of panel data analysis. Also, the CSD test is valid for mixed-level (0) and level (1) data. Equation 2 summarizes the CD-test statistic of Pesaran (2004):

2.4. Slope Heterogeneity Test

This test refers to slope heterogeneity (SH), which any panel data study should accommodate, meaning that the slope differs among panels. We can achieve this by applying Blomquist and Westerlund (2013) and Pesaran et al. (2008). This last approach is a procedure based on a fixed-effects model that tests for slope heterogeneity across individuals by examining the relationship between the slope estimates for each subject and the means of those estimates for all subjects. The Blomquist and Westerlund (2013) approach compares the estimated slope coefficients applicable to fixed and random techniques. The panel results rely on the accuracy of the tests in the panel data estimates. The dynamic effect of SH issue can be shown by equation 3:
 and  divulge the adjusted delta tide and delta tide, respectively.

2.5. Causality Tests

This article employs the Granger Non-Causality Test (GNCT) method (Xiao et al., 2022), a generalization of the Dumitrescu and Hurlin (2012) causality method. This approach also generalizes the Granger (1969) test for time series causality. It effectively removes scale errors and corrects for parameter bias, to such an extent that in large datasets, such as those of the EU countries, it becomes highly effective. This, in turn, improves the accuracy of correlations, thus refining inferences that can be drawn from panel regression. Additionally, panel data models can help make more accurate inferences by controlling for potential omitted variable bias and enhancing the quality and quantity of data (Perone, 2024Wang et al., 2023). The equations for GNCT can be expressed as :Where
Where n is the lags,  present error terms, and 1 to 4 enote the parameters to be estimated. This section outlines the coefficients that need to be estimated. If the coefficients on lagged values of variable N are set equal to 0, then M does not Granger cause N. The DH causality equation is as follows:
Finally, we can reformulate the causality of (Xiao et al., 2022) as follows:
In the context where Z and Z' are observed over N cities and T periods, we define β(qi) as the Granger causation parameter, where h(i,t) is the matrix of dimensions (t*c).

2.6. Unit-root Test

The unit root test determines if the panel series is stationary. Unit-root testing can be performed through various statistical tests such as the CIPS test (Pesaran, 2007) for second-generation panels (Hadri, 2000Im et al., 2003Levin et al., 2002Maddala and Wu, 1999Pesaran et al., 1999) for the first generations. It is essential to apply appropriate tests when analyzing time series data to obtain results that can be considered reliable and valid. The Im, Pesaran, and Shin W-stat test is a panel data model that allows for CSD. In contrast, the CIPS test is straightforward, linear regression-based, and robust against serial correlation.
The CIPS test (Pesaran, 2007) has the following asymptotic in equation 8

2.7. Cointegration Test

Several tests, such as the (Kao, 1999Pedroni, 1999Westerlund, 2007), are available to conduct panel cointegration data. We conduct the Westerlund test, which is more advantageous for long-run and short-run variation relationships and non-stationary series. The Kao test, on the other hand, follows a residual-based detection of unit roots in the regression residuals, identifying trends in the data. The ADF test is extended to the Pedroni test, which considers unit roots at both levels and first differences. The application of several tests verifies the validity.
Asymptotic (Kao, 1999) can be written in equation -9.
Where equation (Pedroni, 1999) in equation-10:inally, the cointegration of (Westerlund, 2007) test can be written as follows:

2.8. Panel Quantile ARDL

The model proposed by Sim and Zhou (2015) has been widely used in recent literature to evaluate the impacts of quantile-dependent series and assess the effect of explanatory variables at different market conditions. The Panel Quantile Autoregressive Distributed Lag (Panel Quantile ARDL) model is a sophisticated econometric technique that integrates quantile regression with panel data analysis, examining dynamic relationships between variables across multiple quantiles. This unique model allows you to analyze how explanatory variables affect different parts of the dependent distribution (the median, lower, and upper quantiles providing analysis on short-run dynamics as well as long-run equilibria (Radulescu et al., 2025Ul-Durar et al., 2025). This capability of PQARDL, which helps extract hidden relationships in data that mean-based analyses may miss, is beneficial for researchers and policymakers seeking to understand how effects might differ under different economic conditions. (Arshed et al., 2022)

3. Results and Discussion

3.1. Descriptive statistics

This section presents the descriptive statistics for the variables CRI, DMC, GDP, LEGS, REG, ICT, and URB. The skewness, kurtosis, and the Jarque-Bera normality tests shown in Table 2 suggest that most variables deviate from normality. More specifically, the distributions of CRI and DMC are relatively symmetric, as their means and medians are closer together. However, CRI is slightly skewed to the left, and DMC is slightly skewed to the right. GDP and URB also display right skewness, with their means greater than their medians; however, URB's skewness, which is close to zero, indicates a more symmetric distribution than GDP. LEGS and REG variables exhibit a considerable difference between the means and medians, with high standard deviations, indicating a significant dispersion and high rightward skewness. That indicates significant skewness and long-tail behavior. ICT appears symmetric around its mean but shows a surprising negative skewness with a high kurtosis, suggesting a more peaked distribution. In summary, the association between the diverse characteristics of the different variables is evident from the statistics presented, and further statistical analysis using non-parametric methods may reveal different treatments for these variables.

Table 2. Descriptive Statistics Information.

Empty CellCRIDMCGDPLEGSREGICTURB
Mean4.4625.31810.3202.2913.5914.90989.182
Median4.4735.29710.0981.0002.0004.94288.075
Maximum4.7437.10711.20510.00020.0005.400100.000
Minimum3.9943.1559.7450.0000.0003.91072.400
Std. Dev.0.2061.0780.4522.2724.1560.3127.963
Skewness-0.532-0.2510.5351.1271.682-1.2050.054
Kurtosis2.4412.7061.8584.0305.6784.6721.963
Jarque-Bera6.6154.54811.22828.14984.77439.4514.985
Probability0.0370.0870.0040.0000.0000.0000.083

3.2. Correlations results

In this section, we analyze the correlation between CRI and DMC, GDP, ICT, LEGS, REG, and URB (See Table 3). DMC correlated weakly and positively with CRI, as indicated by a correlation coefficient of approximately 0.02. The negative sign in front of GDP, approximately -.05, suggests that increased economic output may slightly dampen climate risk. In addition, ICT has a positive correlation with CRI (-0.48), suggesting that advances in technological capacity could also mitigate climate risk. We observe a weak negative correlation between CRI and LEGS, with a correlation coefficient of -0.04. REG strongly correlates negatively with CRI (-0.08), suggesting moderate regulations and legislation.

Table 3. Correlation matrix.

Image, table 3
Although the correlation between ICT and CRI was moderately high, similar to the correlation between ICT and climate risk prior to estimating the QARDL results, we ensured the reliability of the estimated coefficients by mitigating concerns regarding multicollinearity through a Variance Inflation Factor (VIF) diagnostic test. The results, reported in Table 4, confirm the absence of multicollinearity among the explanatory variables. All VIF values fall well below the conventional threshold of 10. This indicates a low level of intercorrelation among the regressors, reinforcing the robustness of the model's specification and supporting the statistical validity of the inference drawn from the quantile-based estimations.

Table 4. The result of VIF.

VariableVIF1/VIF
GDP1.910.524
LEGS2.120.472
URB1.640.610
TEC1.480.672
DMC1.520.658
CRI2.130.496
REG1.860.538
Mean1.81

3.3. Main Results

We use the Levin, Lin, and Chu test, a non-parametric method that employs the empirical distribution function. Various tests (Breusch & Pagan, 1980Pesaran, 2004Pesaran, 2006) confirm the presence of CSD through countries' interdependence with each other. Under the null hypothesis of the CD test, there is no CSD among the integrated countries, while the alternative hypothesis is the presence of CSD. The results in Table 5 indicate the presence of CSD in the residuals, suggesting that nations are interdependent. Such requirements necessitate second-generation tests, which are necessary to capture CSD accurately and provide unbiased estimates.

Table 5. CSD Test.

TestStatisticP-value
B P L M227.2750.00
Pesaran L M38.7560.000
Pesaran C D15.0400.000
.

3.4. Test of slope heterogeneity

Table 6 presents the results of the slope heterogeneity (SH), by applying (Blomquist and Westerlund, 2013Pesaran et al., 2008). Based on a fixed-effects model, the findings confirm the presence of slope heterogeneity across individuals, rejecting the null hypothesis and accepting the alternative hypothesis.

Table 6. The SH findings.

Empty CellDeltap-value
Pesaran, Yamagata. 20084.5220.000
adj. 5.8020.000
Blomquist, Westerlund. 2013.1.930.054
adj. 2.47520.0013
Note: H0: slope coefficients are homogenous
Next, we examine the unit root status of the surveyed variables to mitigate potential biases, inconsistencies, and inaccuracies that may arise from spurious regression. The estimates of the unit root test are demonstrated in Table 7. The 2nd generation CIPS stationarity check test, as proposed by (Pesaran, 2007), has been employed. Additionally, we utilize the approach suggested by Bai and Ng (2004) to reassess unit roots, reinforcing the credibility of the CIPS technique. The findings from both tests indicate that the surveyed variables have no unit root issue, specifically after the first difference for the legislation and urbanization variables, and are stationary at the level for the rest of the variables. These mixed results on the unit root status were conducted to test the Westerlund (2007) cointegration, a second-generation bootstrapped cointegration test.

Table 7. Stationary Tests.

Bai & Ng test
VariablesLevel1st Diff.
CRI-3.166**
(0.02)
/
DMC-2.083
(0.03)**
/
GDP-1.914**
(0.055)
/
LEG1.229
(0.218)
-3.680***
(<0.01)
REG-2.221***
(0.09)
/
URB
ICT
-0.875
(≥.10)
-4.560***
(<0.01)
-4.559***
(0.01)/
CIPS unit root test
CRI3.17*
(<0.01)
/
DMC-2.431**
(<0.05)
-/
GDP-2.627*
(0.01)
/
LEG2.13
(≥.10)
-2.971***
REG-2.634**
(<0.05)
/
URB
TEC
-0.870
(≥.10)
-3.013***
(<0.01)
-3.993***
(<0.01)/
Note: Values in parentheses represent standard errors; *** and** show 1% and 5% significance levels, respectively.
Table 8 presents the results of the Westerlund cointegration test across two models, examining whether the variables in each model share a long-term equilibrium relationship. The test statistics are highly significant, indicating strong evidence of cointegration. The significant values across all test statistics further support the results, confirming robust long-term associations in most models, implying that the variables are moving together in equilibrium over time. The test statistic confirms that the response variable and its predictors exhibit a long-term co-movement in the studied territories, which can be used to estimate both long-term and short-term effects, such as those in PQARDL.

Table 8. Cointegration Test Westerlund (2007).

ModelsSTP-Value
Model 1-1.4166*- 0.0783*
Model 2-1.4222***-0.0775*
Note; Model 1 presents the estimation of equation one, while the second Model 2 presents the robustness test.

3.5. Main results from the QARDL model

Table 9 presents the results of the effect of GDP, climate legislation, urbanization, technologies, and domestic extraction of energy and natural resources on climate risk in GCC countries using panel QARDL. While examining the short-term impacts of distinct factors on the climate vulnerability index in GCC nations across varying quantiles, several notable patterns materialize, particularly when considering the significance levels for most effects. GDP generally shows a positive correlation with climate risk across most quantiles, suggesting that as the economy expands in the GCC region, it may contribute to increasing climate risk. This influence is substantial at the decreased and median quantiles (5th, 25th, and 50th) but diminishes and becomes inconsequential at the highest quantile (95th). For the 5th and 95th percentiles, a 1% increase in GDP translates into 0.33% and 0.35% increases in climate risk. This trend continues for the 25th quantile, although the impact is slightly higher than 0.335%, then decreases relative to the 5th and 75th quantiles, to approximately 0.25 and 0.23, respectively. This outcome aligns with prior studies (Huang et al., 2018Ozkan et al., 2022), confirming that increased GDP contributes to the acceleration of climate risk. Climate legislation consistently displays a mitigating influence on climate risk at lower and medium quantiles (5th, 25th, and 50th), with the impact remaining substantial but decreasing somewhat as we move from the 5th to the 50th quantile. This propensity indicates that statutory steps are effectively reducing climate vulnerability at these levels (Eskander et al., 2021). However, the influence becomes inconsequential at the highest quantile, proposing possible limitations in the effectiveness of prevailing statutes at the extreme upper quantile. Urbanization presents a nuanced impact, which somewhat elevates climate vulnerability at the lowest quantile but becomes substantially more impactful at the 25th quantile. Intriguingly, at the 75th and 95th percentiles, the influence of urbanization on climate vulnerability is negative, albeit minimal, suggesting a complex relationship that may involve factors such as advanced urban planning at higher levels of urban development. ICT plays a vital role in climate risk mitigation in the GCC, and even a marginal increase in ICT of about 1% can mitigate climate risk by around 0.23% on average for each quantile in the short term. Finally, DMC consistently displays a negative relationship with climate vulnerability across all quantiles, substantially decreasing it in the short run. This suggests practical substance usage and consumption patterns that mitigate climate vulnerability, indicating that sustainable practices in substance utilization are beneficial across all levels of economic activity. These results confirm prior studies (Li et al., 2025) that indicate the effect of DMC on accelerating climate risk.

Table 9. QARDL Results.

QuantileGDPLEGSURBICTDMCECM
5th0.332*-0.08*0.003*-0.231*-0.055*0.72*
Std. err0.0831-0.0070.00090.056*0.0186
25th0.335*-0.07*0.0196*-0.220*-0.087*0.72*
Std. err0.0810.000.0070.025-0.079
50th0.255*-0.06*0.002-0.210*-0.001*0.74*
Std. err0.0610.0010.0020.1900.017
75th0.215-0.08*-0.006-0.185-0.0650.73*
Std. err0.023*0.0030.002*0.05*0.0069*
95th0.351-0.07-0.004-0.234-0.0360.75*
Std. err0.011*0.001*0.001*0.001*0.0065*
GDPLEGSURBICTDMCECM
5th-0.0025-0.06-0.0009-0.3891.7030.79*
Std. err0.00630.013*0.00150.016*0.006*
25th-0.003-0.08*-0.001-0.38*0.1650.77*
Std. err0.0040.02*0.0050.009*0.004*
50th-0.004-0.08-0.0001-0.3360.170*0.77*
Std. err0.00420.05*0.001*0.009*0.010
75th-0.014-0.1-0.000-0.3690.1600.76*
Std. err0.002*0.004*0.001*0.006*0.002*
95th0.001-0.12*-0.009-0.3720.1600.75*
Std. err0.0008*0.0010.000*0.002*0.0009*
Note: ***,**, and * denote 10, 5, and 1 statistical significance levels.
In analyzing the long-term effects of climate vulnerability determinants in GCC countries across quantiles, we observe that the relationships differ from those observed in the short term. Across the quantiles, GDP appears to have primarily no impact on climate risk, indicating that economic growth and practices to mitigate environmental impacts may be consistent in the long run. LEGS's mitigatory effect significantly increases in all quantiles, with the 75th (-0.1) and 95th (-0.12) quantiles exhibiting significant effects compared to the short run. This means that legislative measures will be more effective in decreasing climate risk at increasingly higher quantiles, underscoring the importance of sound legal systems for long-term climate risk mitigation. The effect of urbanization on climate risk is minimal for the lower quantiles and turns slightly negative in the 95th quantile. This implies that as Urbanization in the GCC achieves a certain level of development, it may be accompanied by better planning and infrastructure that help mitigate climate vulnerability (Maleknia and Enescu, 2025Xu et al., 2024). ICT has a uniform and statistically significant negative effect across all quantiles, validating its importance as an integral factor in climate risk reduction. This finding aligns with the study by Ulussever et al. (2024), which found that ICT has a positive effect on mitigating climate change and environmental degradation. It also showcases the long-term potential of scaling up technological solutions to reduce environmental harm. DMC exhibits a positive impact across all quantiles regarding climate risk; interestingly, this effect is the highest for the 5th quantile.

3.6. Robustness test

As a robustness check of the previous results on the effect of climate legislation in GCC countries, we suggest substituting the term' climate legislation' with 'climate regulation' in the analysis (See Table 10). This substitution aims to determine whether a broader definition of regulatory action would yield different effects than a narrowly defined legislative action. For the short-run analysis, we posit that climate regulation may be slightly less effective at the lowest quantiles of the distribution. At higher quantiles (25th and 50th), we find effectiveness adjustments for each group that reflect uniform and positive impacts. For the upper tail, significantly at the 75th and 95th quantiles, we predict that the regulation's overall more substantial negative effect will underscore its essential role in designing activities to mitigate climate risk. For the long-term analysis, the estimated changes for the lowest quantiles suggest that better regulation may be implemented over time, possibly indicating that some of the effects of comprehensive regulatory approaches compound in the long run. The consistent median impacts and even slight improvements at the higher quantiles demonstrate that well-designed regulations may be a viable alternative to more legislatively driven options.

Table 10. QARDL Results.

QuantileGDPREGURBICTDMCECM
Short term
5th0.140*-0.03*0.03*-0.433*-0.057*0.58*
Std. err0.009-0.0040.0050.015*0.0127
25th0.154*-0.04*0.024*-0379-0.096*.59*
Std. err0.0160.0050.0110.103-0.0268
50th0.087*-0.07*0.037-0.391-0.265*0.59*
Std. err0.0300.00050.0060.07990.027
75th0.16-0.05*-0.021-0.405-0.0660.57*
Std. err0.024*0.0040.013*0.08*0.014*
95th0.177-0.05-0.03-0.379-0.0930.58*
Std. err0.009*0.002*0.002*0.002*0.004*
GDPLEGSURBTECDMCECM
5th-0.03-0.068-0.0070.210.050.65*
Std. err0.0080.0559*0.0010.001*0.003*
25th-0.026-0.087*-0.0030.217*0.0010.67*
Std. err0.0050.0006*0.00070.014*0.002*
50th-0.006-0.09-0.0080.2360.012*0.67*
Std. err0.00110.05*0.001*0.019*0.003
75th-0.008-0.1-0.0090.2140.050.68*
Std. err0.002*0.004*0.00010.001*0.002*
95th0.006-0.11*-0.0030.2080.050.68*
Std. err0.0009*0.0010.000*0.001*0.0006*
Note: ***,**, and * denote 10, 5, and 1 statistical significance levels.
Apart from quantifying the effects of the climate regulatory framework, examining the robustness of the impacts of other determinants, such as GDP, Urbanization, ICT, and DMC, in both the short and long run is crucial. The short-run analysis indicates that GDP has a positive impact on climate risk in most quantiles, suggesting that GDP may contribute to climate risk. This trend is particularly pronounced at lower and median quantiles. Urbanization has a differential impact, elevating climate risk minimally in the lowest quantiles and at the 25th quantile, where it becomes more pronounced. At higher quantiles, its influence is reversed by a tiny amount, indicating that more sophisticated urban planning could help alleviate climate risk at those extremes. ICT is a key determinant of climate risk at all quantiles, with small increases in ICT implementation resulting in significant reductions in climate risk. DMC has a consistently negative association with climate vulnerability, suggesting that current material use patterns may be exposing countries to growing climate risks. The long-term analysis confirms the initial climate legislation analysis, indicating that sustainable practices that limit environmental impacts can naturally align with economic growth. The coverage of urbanization is negligible across lower quantiles and becomes slightly negative at the highest quantile. The ICT variable exhibits a consistent and significantly negative effect across all quantiles, underscoring the importance of ICT in mitigating climate risk in the long term. The more interesting finding regarding DMC is that it has a positive and significant effect in all quantiles. In contrast, the most potent effect is observed in the lowest quantile, which may indicate the importance of long-term development in sustaining material consumption.
The further robustness check, conducted using bootstrapping quantile regression (BQR) with surface temperature change (STC) replacing the CRI, confirms the stability of the long-run results obtained from the PQARDL model. The BQR estimates reveal that GDP maintains a positive effect on climate risk across most quantiles. However, the magnitude slightly diminishes toward the upper quantiles, reflecting the persistent warming pressures associated with economic expansion. Climate legislation consistently shows a negative relationship with STC, with the mitigating impact strengthening at higher quantiles, underscoring the effectiveness of regulatory measures under heightened risk conditions. The effect of urbanization remains mixed, with a slightly negative impact at lower quantiles but approaching neutrality at higher levels, suggesting that advanced urban infrastructure and planning may help buffer climate vulnerabilities. Similarly, ICT demonstrates a robust and increasingly negative impact across all quantiles, highlighting its critical role in mitigating surface temperature increases, particularly under extreme conditions. Conversely, DMC exhibits a positive relationship throughout the distribution, with its effect intensifying at the upper quantiles, suggesting that unsustainable material use continues to exacerbate climate vulnerabilities. Overall, these results corroborate the long-run dynamics of the primary analysis, reinforcing the robustness and reliability of the findings for guiding policy interventions in GCC countries.

3.7. Causality findings

This section presents the Granger Non-Causality results, providing an insightful perspective on the directional effects that drive the climate risk index, as shown in Table 11. First, we find strong uni-directional causality from the GDP to the climate risk index, indicating that the predictability of GDP changes affects climate risk, and economic expansion affects climate conditions. In contrast, climate risk is not predictive of changes in GDP, indicating that climate conditions do not significantly influence economic outcomes. With climate regulation, the predictability between the action and outcomes is uni-directional, from LEGS to CRI, which explains the workability of climate regulation in addressing climate challenges. Similarly, there is a bi-directional trend from DMC to climate risk, while changes in CRI do not significantly predict changes in urbanization and vice versa. ICT provides high uni-directional causality to climate risk, and ICT can predict a substantial decrease in climate risk. The importance of ICT in climate risk mitigation cannot be overstated. Nevertheless, climate risk can hardly trigger technological progress, which suggests that climate risk levels do not directly incentivize technological innovations.

Table 11. Granger non-causality Test results.

Null Hypothesis:F-StatisticStd. err.P-ValueCausalityDecisions
GDP  CRI0.0460.0230.04YesUni-directional
CRI GDP0.0270.1610.865No
LEGS 0.0170.0070.022YesUni-directional
CRI 0.00160.0060.814NO
 CRI0.5911.040.57No
No association
CRI URB0.02790.0520.59No
DMC CRI0.0740.0390.061YesBi-directional
CRI DMC0.4910.1090.003No
TEC  CRI0.4390.05590.000YesUni-directional
CRI TEC0.02850.0470.0556No

4. Conclusion

The study examined the impact of climate legislation on climate risk, taking into account the roles of GDP, Urbanization, ICT, and DMC in the context of GCC countries, using the Panel QARDL model and Granger Non-Causality Tests. In particular, GDP tends to exacerbate climate risk in almost all quantiles, indicating that economic growth often comes with an environmental cost. Unlike climate legislation and ICT, which effectively mitigate climate risk, significant climate legislation and ICT are recognized as key components of ecological governance. Urbanization appears to have more nuanced implications for climate risk, with its effect not being assertive, suggesting that its current scale in the GCC cannot significantly impact climate risk. Lastly, DMC is often associated with increased climate risk, where current consumption behaviors may be unsustainable and lead to further environmental degradation. The findings present numerous policy implications that could make a significant difference on the ground, considering the availability of resources in the GCC countries. Policymakers can drive significant improvements in legislation and regulation, especially for industries such as crude oil and gas, construction, and transportation, by establishing strict emissions standards and meeting renewable energy needs. The findings of this study reveal that an increase in the quantity of enacted environmental legislation is associated with a significant reduction in climate risk across the GCC countries. This suggests that expanding legislative frameworks remains a critical pathway for mitigating climate vulnerabilities in the region. Accordingly, the key policy implication is the need to intensify the development and enactment of climate-related laws and regulations, ensuring that environmental governance remains dynamic and responsive to evolving challenges. By consistently expanding the legislative base, GCC countries can reinforce their institutional commitment to addressing climate risks and supporting the transition toward more sustainable and resilient economic systems.
The study items have several limitations that impact the interpretation of the results. First and foremost is the availability of data before 2005. Explanatory variables, such as global oil prices and international trading, might not have been fully accounted for in determining climate risk. In this article, certain aspects, such as the assumption of uniform policy implementation within widely variable governance and administrative contexts, may be limiting. Another important consideration is that the effectiveness of climate legislation may depend heavily on the strength of governance institutions. In contexts with strong institutional frameworks, environmental laws are more likely to be implemented, monitored, and enforced rigorously. However, in environments with weaker governance, even well-intentioned regulations may not produce significant outcomes. Although this study does not explicitly model governance quality, we acknowledge its mediating role and recommend that future research incorporate governance indicators, such as regulatory quality, corruption control, or the rule of law, into the empirical framework to better isolate conditional policy effectiveness. While the evidence in this study is quite robust regarding the impacts of environmental legislation in the GCC and the mitigation of climate risk, future research will need to generalize its findings across other oil-dependent or emerging markets with similar structural features. Furthermore, this study does not explicitly incorporate external global drivers such as oil price volatility, foreign direct investment (FDI), or participation in international climate treaties (e.g., the Paris Agreement), all of which are highly relevant to climate policy formation and environmental governance in the GCC region. Given the GCC's dependence on hydrocarbon markets and global capital flows, these factors could significantly moderate or amplify the effectiveness of domestic legislation. We recommend that future studies incorporate global energy shocks, capital mobility, and treaty participation indicators to provide a more comprehensive understanding of climate risk governance under global interdependence. Finally, this study applies a panel framework across the GCC countries, which assumes a degree of policy uniformity. However, this assumption may oversimplify the substantial heterogeneity that exists among GCC members in terms of enforcement strength, institutional capacity, political will, and regulatory effectiveness. For instance, Saudi Arabia’s Vision 2030 roadmap differs significantly from the UAE’s Net Zero Strategy or Qatar’s ESG initiatives. Future studies should explore country-level fixed effects, spatial interactions, or heterogeneous panel techniques to capture this variation better and produce more granular policy insights. Moreover, future research might analyze the interplay with institutional quality, enforcement mechanisms, and global cooperative agreements. Including firm- or city-level information may also help shed light on how environmental policies affect behavior and outcomes at the micro level. Including additional variables, such as oil and FDI, and dynamic simulations, as well as structural modeling approaches, could be used to assess the long-term transition pathways of alternative policy scenarios.

Funding

Research reported in this publication was supported by the Qatar Research Development and Innovation Council under grant #ARG01-0508-230093. The content is solely the responsibility of the authors and does not necessarily represent the official views of Qatar Research Development and Innovation Council.

CRediT authorship contribution statement

Mohamed Sami Ben Ali: Validation, Supervision, Resources, Project administration, Methodology, Funding acquisition. Alanoud Al-Maadid: Writing – review & editing, Validation, Supervision, Project administration, Funding acquisition. Kamal Si Mohammed: Writing – original draft, Software, Formal analysis, Data curation.
Keywords:
Climate change