The Relationship Between Energy and Climate Action in the Context of Sustainable Development Goals: An Empirical Analysis of BRICS–T Economies

Abstract

There is a lack of empirical studies investigating the individual and combined effects of environmental policy stringency, energy transition, and green technologies on greenhouse gas emissions in the context of BRICS–T countries. To address this gap in the literature, the article presents empirical evidence from panel quantile, Driscoll–Kraay, and Ridge regression models for examining energy and climate action within the framework of Sustainable Development Goals 7 and 13 in BRICS–T economies during the period 1995–2020. The main findings obtained from the analyses show that environmental policy stringency, as well as the combined effect of environmental policy stringency with green technology, reduce ecological deformation. On the other hand, energy transition, green technology, primary energy consumption, and the combined effect of energy transition and environmental policy stringency have been shown to increase emissions. Dumitrescu–Hurlin Granger causality findings indicate that all variables exhibit two–way causality relationships reflecting the feedback effect. The results highlight that countries should focus on implementing stricter environmental regulations, promoting green innovation, adopting comprehensive fiscal and environmental policies, accelerating the transition from conventional to clean energy, and strengthening policy measures to achieve long-term ecological goals.

1. Introduction

The phenomenon of sustainable development is a concept that concerns not only economists but also environmentalists, geologists, ecologists, geographers, and even physicists [1]. In this context, it is important to evaluate sustainable development, which requires urgent action plans for humanity, particularly in light of the growing importance of the relationship between energy and climate action.

Making development sustainable, the earth’s resources, such as water, soil, energy, and the diversity of life, the atmosphere, and the climate system, must be used in a way that will allow future generations to live and develop [2]. As can be understood from this definition, the natural environment is of critical value for sustainable development.

The natural environment is reflected in the content of the Sustainable Development Goals (SDGs). For example, the goals directly linked to environmental quality are SDG 6 (clean water and sanitation), SDG 13 (climate action), SDG 14 (life below water), and SDG 15 (soil and biodiversity). In contrast, others are indirectly linked to the environment. Some researchers believe that environmental variables are inherent in almost all SDGs and argue that there is at least one ecological indicator for each goal [3].

Economic, social, and environmental policies are the main elements showing the relationship between global climate change and sustainable development [4]. For long-term survival and sustainable development, businesses need to not only achieve short-term business goals by increasing operational efficiency, but also actively undertake environmental and social responsibilities to thrive in competitive and expanding business environments [5]. The climate issue is a critical issue affecting both the sustainable development process and humanity. On the other hand, energy and climate issues, perhaps the most important agenda items today, have been included in this study due to their importance for future generations as well as sustainable development, and an empirical analysis has been conducted on BRICS–T countries.

The other sections of the study, after the introduction, proceed as follows. In the Section 2, the theoretical framework of the study is explained, and its place and importance in the literature are discussed. In the Section 3, the materials and methods are included. In the Section 4, the results obtained from the econometric analysis are reported. The Section 5, constitutes the discussion section based on the findings. The study concludes with a section that examines the relationship between energy and climate action in relation to the SDGs through empirical analysis and provides recommendations that guide future studies.

2. Theoretical Framework and Literature Review

2.1. Nexus Between Environmental Policy Stringency and Climate Action

Progressive decarbonization efforts require a regular tightening of environmental standards. To monitor these policy efforts, the environmental policy stringency (EPS) index developed by Botta and Koźluk [6] has been on the agenda of OECD countries. EPS, a composite index based on the stringency of three subcomponents and 13 environmental policy instruments that monitor climate and air pollution reduction policies, is an indicator that calculates how stringent and comprehensive a country’s environmental policies are. It consists of three equally weighted sub-indices that group market-based policies, non-market policies, and technology support, respectively. It measures the stringency of individual and aggregate environmental policies, considering both the explicit and implicit costs associated with environmentally damaging situations. It helps evaluate and compare the effectiveness of countries’ environmental protection efforts and is also used to evaluate different types of policies. The index, calculated for 40 countries, is scored between 0 and 6, and higher index values indicate stricter implementation of environmental policies. The increase in value explains the higher costs that must be incurred for a unit of pollution. While stricter environmental regulations may bring compliance costs, they may also encourage environmental innovation, and the benefits of these innovations may exceed the costs [7,8,9,10].

EPS, as an explanatory variable, reflects the effectiveness of decision-makers in the field of environmental protection [7]. It is observed that most studies examining the relationship between the index and climate action primarily focus on carbon dioxide (CO₂) emissions [7,8,10,11,12,13,14,15]. For example, Ahmed [11] found that increases in EPS and green technology (GT) in 20 OECD countries reduce emissions and accelerate sustainable development in the long run. However, it was reported that EPS increases environmental degradation in the short run, while GT is statistically insignificant. Ahmed and Ahmed [12] estimated the impact of EPS and economic activity on CO₂ emissions in China using data for the period 1990–2012. The adjusted gray model simulation findings showed that EPS reduced the emission intensity but increased gross domestic product (GDP). Anwar and Sharif [13] empirically examined the impact of EPS, technological innovations, natural resource rent, and renewable energy consumption on emissions for the period 1996–2019 in six developing countries. PQR findings reported that the interaction of EPS, technological innovations, renewable energy consumption, EPS and natural resource rent reduced emissions. Bhowmik et al. [14] investigated the linkages between EPS, natural resources, technological innovations, and emissions in G–7 countries using data for 1990–2020. Panel Quantile Regression (PQR) findings confirmed that natural resources increased emissions, while the synergy of technological innovations, natural resources, and EPS decreased them. Durgun [7] tested the effects of EPS, growth, renewable energy, and foreign direct investment on the carbon index between 1990 and 2020 in the case of Turkey. The analysis findings using Fourier-based approaches revealed that growth, renewable energy consumption, and foreign direct investment increased environmental degradation, while the effect of EPS was insignificant. Frohm et al. [8] provided empirical evidence on the short–term and long-term sectoral effects of EPS on emissions in 30 OECD countries. The regression findings indicated that stricter environmental policies were associated with lower emissions during the period 2000–2014, and that this effect increased over time, differing across sectors depending on their fossil fuel intensity at the same time. Liu et al. [15] examined the dynamic relationship between EPS and emissions at the scale of the most polluted countries in the Asia Pacific Region in the period 1991–2021. Nonlinear Auto-Regressive Distributed Lag (ARDL) findings explained that a positive shock in EPS and an increase in renewable energy consumption reduce environmental degradation, while a negative shock in EPS has a significant positive effect on emissions in the short and long term. Olasehinde-Williams and Akadiri [10] analyzed the relationship between EPS and carbon leakage for the EU–20 countries during the period 1995–2020. Ordinary Least Squares (OLS) findings supported that EPS causes more carbon leakage by increasing the amount of foreign carbon emissions. Causality findings also confirmed that the predictive powers of country–specific responses between EPS and carbon leakage are common among EU countries.

For the effects of EPS on greenhouse gas (GHG) emissions, studies by [16,17,18] are among the literature examples. In this context, Ahmed et al. [16] evaluated the effects of EPS, GDP, urbanization, and energy consumption on emissions at the Pakistan scale using the ARDL method. The findings suggest that urbanization has positive effects on emissions in the short and long term, while energy consumption has negative effects, and EPS has a significant effect in reducing emissions. Chen et al. [17] examined the effects of environmental tax and EPS shocks on reducing pollution emissions in China during the period 1993–2019. The long-term findings of nonlinear ARDL estimation techniques indicate that positive shocks in EPS lead to reductions in CO₂ and GHG emissions. Kartal et al. [18] analyzed the cases of Finland and Sweden to reveal the effect of environmental measures on sectoral GHG emissions in the period 1991/Q1–2020/Q4. Quantile regression and Granger causality findings showed that EPS reduced emissions from fuel exploitation, industrial combustion, and energy industry sectors at lower and middle quantiles; on the other hand, it led to an increase in emissions from agriculture and construction sectors at higher percentiles, supporting that EPS has a causal effect on sector-specific emissions at different percentiles.

In some studies examining the connection between the EPS index and climate action outside the axis of CO₂ and GHG emissions, it is observed that various parameters such as ecological footprint, particulate matter, sulfur dioxide, volatile organic compounds, and nitrogen dioxide are used [e.g., [19,20,21,22]].

The following hypothesis was formed based on the theoretical basis and empirical literature evidence:

Hypothesis 1. 

EPS improves environmental quality through technology support, market, and non-market-based policies.

2.2. Nexus Between Energy and Climate Action

The increase in the world’s surface temperatures due to the increase in GHG emissions in the atmosphere is the main cause of global warming and climate change. Energy production and consumption directly affect this process. The burning of fossil resources, which causes the release of large amounts of GHG into the atmosphere, accelerates global warming. Renewable energy sources have a critical role in reducing the effects of global warming since they have much lower emissions than fossil fuels. However, more than ¾ of the world’s energy needs are met by the burning of fossil fuels. Despite the increasing use of renewable energy sources, coal, oil, and natural gas continue to be the world’s primary energy sources [23,24].

A comprehensive literature review empirically demonstrates the negative impacts of fossil fuel energy sources, particularly coal and oil, on the ecosystem. For example, Li et al. [25] followed different panel estimator methodologies in their study examining BRICS countries using 1990–2019 data. Estimation findings supported that non-renewable energy, GDP, and government expenditures intensify CO₂ emissions, while EPS, tax revenues, and renewable energy reduce them. At the same time, causality findings showed that renewable energy, EPS, government expenditures, and tax revenues can predict emissions. Wolde-Rufaela and Weldemeske [26] examined the relationship between EPS, renewable energy, fossil energy, oil prices, income, and emissions in BRIICTS countries during the period 1993–2014. Pooled Mean Group ARDL findings indicated that there is an inverted U-shaped relationship between EPS and emissions: renewable energy consumption is negatively related to emissions, while GDP, fossil energy consumption, and oil prices are positively and significantly related.

Historically, the first energy transition occurred during the Industrial Revolution, from wood and other biomass to coal and then to oil and natural gas [27]. The continuous increase in environmental degradation has led researchers, decision-makers, and policymakers to focus their attention on energy transition (ET). ET represents a significant structural shift in energy supply and demand within an energy system. Today, the transition to sustainable energy is ongoing to slow down climate change. Since most sustainable energy is renewable, ET is also referred to as the renewable energy transition. The current transition aims to rapidly and sustainably reduce GHG emissions from energy, mostly by phasing out fossil fuels and switching as many processes as possible to run on low–carbon electricity [28].

For the impact of energy transition on CO₂ emissions, Abbasi et al. [23] provided empirical evidence on the impact of ET, EPS, GT, and primary energy consumption (PEC) on emissions in 26 OECD countries between 1995 and 2019. Method of Moments Quantile Regression (MMQR) findings indicated that the interaction between ET and (EPS*GT) increased environmental quality in all quantiles, while the interaction between EPS, PEC, and (EPS*ET) decreased it. However, causality findings showed that GT, ET, EPS, (EPS*GT), and PEC had a two-way linkage with emissions, while (EPS*ET) had a one-way linkage. Albulescu et al. [29] examined how ET, 11 EPS components, and energy use affected emissions for 32 OECD countries during the period 1990–2015. PQR findings indicated that increases in ET and EPS had a negative impact on emissions. It has also been shown that energy use contributes to environmental degradation. Güler and Özarslan Doğan [30] analyzed the effects of renewable energy, innovation, and EPS on emission reduction in G–7 countries from 1994 to 2019. Driscoll–Kraay findings showed that renewable energy consumption and EPS reduce emissions, but GT, GDP, energy use, and human capital increase. Wang et al. [31], who examined BRICS economies in the 1990–2019 period, explained the effects of the transition to renewable energy and EPS on environmental quality. ARDL findings reported that renewable energy and EPS are important in preventing emissions in the short and long term.

Based on the theoretical scope and empirical literature, the following hypotheses were established:

Hypothesis 2. 

The transition process from conventional energy to renewable energy reduces environmental degradation by reducing GHG emissions.

Hypothesis 2.a. 

The combined effect of EPS and ET has a more positive impact on environmental degradation. Considering the variables together contributes to the assessment of emissions in light of strict environmental regulations and progress towards a sustainable energy transition. The variables can be used as a proxy for renewable energy.

Hypothesis 3. 

PEC negatively affects environmental quality by increasing environmental externalities.

2.3. Nexus Between Green Technology and Climate Action

It has been observed that technological innovations have recently received more attention in the climate change literature. The literature review supports the idea that GT has a strong connection with emissions. In this context, Ağan [32] investigated the potential role of EPS and emissions in mitigating climate change in EU countries using 2000–2021 data for CO₂ emissions. The findings explained that EPS, GDP, and emissions showed positive effects on GT at higher quantile values. At the same time, it was observed that there was a bidirectional causal relationship between GT and EPS. Kurt [33] evaluated the possible nonlinear effects of GT, EPS, and market-based policy stringency on emissions in 16 European countries for the period 2005–2019. In the threshold regression findings, it was observed that there was no threshold effect for EPS in the relationship between GDP and carbon emissions, but there was a threshold effect when urban population, service sector, and GT variables were used. In this context, a high EPS regime reduces emissions, whereas GT, urban population, GDP, and service sector share increase them. A 1% increase in the GT variable tends to increase emissions by approximately 0.08%. Li et al. [34] examined the role of GT, EPS, and solar energy for the purpose of reducing emissions in OECD countries in the period of 2001–2018. The main findings confirmed that these variables are positive and play important roles in ensuring environmental sustainability in the long term. Ouni et al. [35] investigated the role of GT, green finance, green energy, economic growth, digital economy, and urbanization on CO₂ emissions from the transportation sector in G20 economies in the period of 2002–2022. PQR findings showed that green energy and GDP reduce emissions in the lower quantiles, while green finance, green energy, and urbanization increase environmental quality in the upper quantiles. GT is associated with higher emission rates in all quantiles. Udeagha and Ngepah [36] examined the integrated effects of GT, renewable energy, GDP, compound risk, and EPS on emissions in BRICS countries during the period 1960–2020 using the cross-sectional ARDL framework. The findings proved that GT, renewable energy, and EPS reduced emissions, but compound risk and GDP increased them. Xie et al. [37] analyzed the empirical link between emissions and disaggregated trade, GT, and EPS in OECD countries during the period 1990–2020. According to the MMQR findings, it was determined that GT, exports, imports, and EPS made significant contributions to explaining the variance in emissions. Notably, EPS had the greatest impact, enhancing the benefits of green technologies by utilizing environmentally friendly practices, while imports had a negative effect on environmental quality.

For the effect of GT on GHG emissions, Fatima et al. [38] evaluated the moderating effect of EPS on the relationship between renewable energy and emissions in 36 OECD countries during the period 1990–2020. MMQR findings showed that GT, green innovation, and (EPS*ET) interaction contributed to reducing emissions in all quantiles. Ghazouani et al. [39] empirically demonstrated the role of GT, clean energy use, environmental policies, and regulations on emissions in nine European countries during the period 1994–2018. Fully Modified OLS, dynamic OLS, and PQR findings supported that GT, environmental taxes, and clean energy sources reduce the overall pollution flow.

The following hypotheses were formed within the scope of the theoretical framework and empirical literature.

Hypothesis 4. 

GT reduces GHG emissions by reducing carbon emissions and contributing to carbon neutrality.

Hypothesis 4.a. 

The combined effect of EPS and GT creates more positive effects on environmental degradation. The combined impact of environmental regulations and technological innovations provides insights into innovations focused on climate change mitigation, adaptation, and sustainability. The impact can be attributed to the use of renewable energy.

2.4. Literature Gap

Existing literature is lacking in determining the effects of stringent environmental laws on energy transition and green technology adoption in BRICS–T economies. Although some countries have made progress in environmental regulations, the general framework for the countries in the sample set is insufficient to address larger environmental challenges. Data on GHG emissions for the countries in the sample set in 2023 report that the total emissions of six countries account for more than half of global emissions [40]. The BRICS–T group consists of emerging market economies seeking to increase their influence globally beyond their significant share of global emissions. To this end, the bloc is developing various strategies to advocate for greater representation in international organizations, coordinate the diplomatic and economic policies of its members, create an alternative financial system, and reduce dependence on the US dollar. Furthermore, the countries’ economic performance exhibits a significant positive divergence compared to the rest of the world [41,42]. Therefore, the group’s increasing global and regional power also brings about potential collaborations on the climate agenda. For these reasons, BRICS–T countries were selected as the sample for this study.

In addition to the importance of the sample group, the literature review shows that the majority of studies are based on OECD countries [8,11,21,22,23,29,34,37,38], making it necessary to make an evaluation specifically for non-OECD countries. In this respect, the empirical contribution of the current study is valuable. The absence of a study specifically for BRICS–T, as well as the absence of any study in the literature on the possible impact of the combined effect of (EPS*ET) and (EPS*GT) for SDG 7 and SDG 13 on GHG emissions, reveals the original contribution of this article. At the same time, the literature review showed that mostly CO₂ emissions [7,8,10,11,12,13,14,15,23,25,26,29,30,31] are mostly used as the outcome variable, whereas GHG emissions, which are a broader scale indicator compared to carbon emissions, need to be examined more. Depending on the reasons stated, the article aims to contribute to the literature from different perspectives.

3. Materials and Methods

3.1. Materials

This research aims to provide empirical evidence on energy and climate action in the BRICS–T countries within the framework of SDG 7 and SDG 13 in the period 1995–2020. In line with this goal, the model of the research is established as follows, with reference to the study conducted by Abbasi et al. [23]:

$${GHG}_{it}={\beta }_0+{{\beta }_{1}EPS}_{it}+{{\beta }_{2}ET}_{it}+{{\beta }_{3}GT}_{it}+{{\beta }_{4}PEC}_{it}+{{\beta }_5({EPS}^{*}ET)}_{it}+{{\beta }_6({EPS}^{*}GT)}_{it}+{\xi }_{it}$$

(1)

The variables in the model are GHG, greenhouse gas emissions (t CO₂e/capita); EPS, environmental policy stringency index; ET, energy transition (% of total renewable electricity output); GT, green technology (number of patents for environment–related technologies); PEC, primary energy consumption (gigajoules per capita). The variables (EPS*ET) and (EPS*GT) in the equation were included in the model with the assumption that the combined effects of environmental policy stringency with energy transition and environmental policy stringency with green energy practices would have a greater impact on emissions. Annual secondary data for variables were compiled from the official websites of the World Bank [40] for GHG and ET, OECD [43] for EPS and GT, and British Petroleum [44] for PEC. β₀, constant coefficient; β₁, β₂, β₃, β₄, β₅, and β₆ are parameter coefficients; ξ, error term. The subscript i represents the countries of Brazil, Russia, India, China, South Africa, and Turkey, which represent the cross-sectional units. The subscript t indicates the time course between 1995 and 2020. In this context, it is seen that the model has a balanced panel structure consisting of 156 (6 × 26) observation numbers.

The parameter coefficients of the model variables shown in Equation (1) are expected to be β₁, β₂, β₃, β_5, and β₆ < 0, β₄ > 0 [e.g., [16,17,23,29,30,31,34,36,37,38,39]]. In this context, the a priori expectation for the independent variables is that increases in the levels of environmental policy stringency, energy transition, and green technology, as well as decreases in primary energy consumption, will positively affect climate action and reduce GHG emissions.

Figure 1 shows the descriptive graph of the datasets included in the model through box plots. An overview of the variables selected within the scope of the study is provided. Thus, the impact of energy and green solutions in mitigating climate change among BRICS–T countries in the period 1995–2020 is evaluated.

3.2. Methods

Figure 2 presents the framework of the methodological path that forms the basis of this study. In this context, the Variance Inflation Factor (VIF) criterion was used to determine the multicollinearity problem within the scope of pre–tests. Then, Shapiro–Wilk W and Skewness/Kurtosis normality tests were performed to determine whether the error terms provided the normal distribution condition. In the second stage, cross-sectional dependence and slope homogeneity tests were applied to decide the type of unit root tests. In the next stage, the existence of a long-run relationship between variables was determined using the Westerlund [45] cointegration test, which eliminates slope heterogeneity and cross-sectional dependence issues. In the third stage, Bootstrap PQR analysis was performed to determine the long-run relationships between parameter coefficients, as recommended when the error terms in the model are not normally distributed. In the fourth stage, the analysis continued with Driscoll and Kraay [46] to test the robustness of the results in cases of heteroscedasticity, autocorrelation, and cross-sectional dependence, and with Ridge regression estimators resistant to multicollinearity. In the last stage, Dumitrescu and Hurlin [47]. Granger causality analysis recommended for heterogeneous panels was performed.

4. Econometric Results

4.1. Preliminary Tests

The problem of multicollinearity, one of the important assumptions of linear regression models, explains the deviation from the assumption that there is no relationship between the independent variables. To test the hypothesis in question, the existence of a multicollinearity problem between the independent variables was tested using the VIF criterion. The findings are given in

Table 1. VIF test for independent variables.

Variable	VIF	1/VIF
EPS	1.45	0.691685
ET	1.11	0.899032
GT	1.37	0.731374
PEC	1.10	0.907216
Mean VIF	1.26

. Upon examining the table, it is evident that the average VIF criterion value is 1.26. In this context, it is stated that there is no multicollinearity problem between the independent variables since the average VIF criterion value is less than 5 for each variable.

This section may be divided by subheadings. It should provide a concise and precise description of the experimental results, their interpretation, as well as the experimental conclusions that can be drawn.

Shapiro–Wilk W and Skewness/Kurtosis normality tests were performed to determine whether the normal distribution assumption was met in the established model. The normal distribution test findings on the error terms are summarized in

Table 2. Normality tests on error terms.

Skewness/Kurtosis	Pr (Skewness)	Pr (Kurtosis)	Adj chi2(2)	Prob > chi2
	0.0001	0.7196	12.80	0.0017 **
Shapiro–Wilk	W	V	z	Prob > z
	0.90896	10.956	5.438	0.0000 *
Shapiro–Francia	W′	V′	z	Prob > z
	0.91313	11.454	4.961	0.0000 *

The symbols * and ** are 1% and 5% probability values, respectively.

. The findings explain that the H0 hypothesis was rejected at the 5% significance level and that the normal distribution assumption was not met in the established model. The PQR model does not require the variables to comply with the normal distribution assumption.

Quantile–quantile normality graphs were drawn to examine the normality status of the variables. Figure 3 visualizes that the entire dataset deviates significantly from the normal distribution line. The findings indicate that the variables are not normally distributed, and the PQR approach is the appropriate approach for estimation.

4.2. Cross–Sectional Dependence, Slope Homogeneity and Unit Root Tests

In this study, the cross-sectional dependence (CD) test, which was developed by Breusch and Pagan [48] and suggested for the case of T > N (26 > 6), is used. Breusch and Pagan [48] recommend the Lagrange multiplier (LM) test statistic to test zero cross–equation error correlations. However, the fact that the constant and slope parameters are homogeneous or heterogeneous according to the units causes the cointegration and causality estimators to differ [49]. In this context, the Swamy’s S test was applied for the slope homogeneity of the parameters.

Table 3. Cross-sectional dependence and slope homogeneity tests.

Variables		CDLM	Slope Homogeneity
	LM–Test	p-Value	chi2(5)	Prob > chi2
GHG	217.4651	0.0000 *	6820.13	0.0000 *
EPS	53.07	0.0000 *	101.51	0.0000 *
ET	114.3	0.0000 *	6179.15	0.0000 *
GT	36.71	0.0000 *	116.47	0.0000 *
PEC	43.73	0.0000 *	7236.30	0.0000 *
(EPS*ET)	86.62	0.0000 *	345.63	0.0000 *
(EPS*GT)	102.8	0.0000 *	69.93	0.0000 *

The symbol * is 1% probability value.

reports the CD_LM and slope homogeneity findings. The findings support the evidence of panel cross-sectional dependence at a 99% critical value (p < 0.01). In this context, the existence of a correlation between the units is confirmed. However, the slope homogeneity shows that the H0 hypothesis is rejected at the 1% significance level, and the slope heterogeneity is valid. Therefore, it is found that the parameters are not homogeneous and vary from unit to unit.

The type of unit root tests was determined in line with the 2nd generation panel unit root tests recommended in the presence of cross-sectional dependence. In this direction, the preferred Multivariate Augmented Dickey–Fuller (MADF), Cross-sectional Augmented Dickey–Fuller (CADF) and Cross-sectional Im–Pesaran–Shin (CIPS) unit root test findings are given in

Table 4. MADF, CADF, and CIPS unit–root tests.

Variables	MADF		CADF		CIPS		Stationarity
	Level	Δ	Level	Δ	Level	Δ
GHG	11.381	83.757 **	2.419 ** (0.048)	2.971 * (0.001)	−2.120	−3.208 *	I(1)
EPS	15.230	156.948 **	3.171 * (0.000)	3.351 * (0.000)	−2.316 ***	−4.265 *	I(0)
ET	23.519	193.230 **	0.668 (0.997)	3.206 * (0.000)	−1.276	−5.126 *	I(1)
GT	196.577	199.471 **	0.840 (0.990)	3.589 * (0.000)	−1.102	−4.680 *	I(1)
PEC	20.248	129.056 **	2.020 (0.256)	2.727 ** (0.007)	−2.454 **	−3.800 *	I(1)
(EPS*ET)	24.127	148.230 **	2.345 *** (0.070)	3.219 * (0.000)	−1.886	−4.193 *	I(1)

The symbols *, **, and *** are 1%, 5%, and 10% probability values, respectively.

. The findings show that the variables GHG, ET, GT, PEC, (EPS*ET), and (EPS*GT) are stationary at level I(1), and the EPS variable is stationary at level I(0) and does not contain a unit root.

4.3. Long-Run Relationship

Westerlund [45] cointegration test, which eliminates slope heterogeneity and cross-sectional dependence problems, was applied to investigate the existence of a possible long-run relationship between variables. According to the findings summarized in

Table 5. Long-run Westerlund cointegration test.

Statistic	Value	z-Value	p-Value	Robust p-Value
Gt	−6.990	−12.630	0.000 *	0.000 *
Ga	−11.377	−1.398	0.081 **	0.010 *
Pt	−12.750	−7.267	0.000 *	0.000 *
Pa	−14.908	−4.126	0.000 *	0.000 *

The symbols * and ** are 1% and 10% probability values, respectively.

, the H0 hypothesis, which states that there is no cointegration relationship between variables, is rejected, and the existence of a long-run cointegration relationship between variables is supported.

Quantile regression is a method used to estimate functional relationships between variables for all parts of the probability distribution. It provides information about the entire conditional distribution of the dependent variable with respect to the independent variable. In this way, when the error terms are not normally distributed, the estimates obtained from quantile regression are more effective than the estimates obtained from the OLS method. At the same time, since it is resistant to heterogeneity in the dataset, it provides more consistent and reliable results than traditional econometric approaches. In addition, the emergence of different results at different quantile values is interpreted as the dependent variable responding differently to changes in the independent variable to varying points of the conditional distribution [25,50].

Table 6. Bootstrap panel quantile regression test.

Variables	Lower Quantile			Middle Quantile			Higher Quantile
	q–10	q–20	q–30	q–40	q–50	q–60	q–70	q–80	q–90
EPS	−0.4775 *	−0.3428 **	−0.3556 *	−0.2774	−0.4573 *	−0.4974 *	−0.5427	−0.2548	−0.4468
ET	−0.0002	0.0046	0.0044 ***	0.0042	−0.0013	−0.0032	−0.0085	−0.0037	−0.0064
GT	0.0018 *	0.0017 *	0.0017 *	0.0015 *	0.0013 *	0.0012 **	0.0005	0.0001	−0.0003
PEC	0.0729 *	0.0736 *	0.0739 *	0.0738 *	0.0723 *	0.0719 *	0.0832 *	0.0975 *	0.0990 *
(EPS*ET)	0.0112 **	0.0055	0.0070 **	0.0053	0.0077 **	0.0093 **	0.0019	−0.0078	−0.0070
(EPS*GT)	−0.0005 *	−0.0005 *	−0.0005 **	−0.0004 *	−0.0003 *	−0.0003 ***	−0.0001	0.0000	0.0002
C	1.0457 *	0.9401 *	0.9603 *	1.0728 *	1.649 *	1.8297 *	2.0545 ***	1.2749	1.7141
Pseudo R2	0.87	0.86	0.85	0.84	0.83	0.83	0.81	0.82	0.83

The symbols *, **, and *** are 1%, 5%, and 10% probability values, respectively.

shows the findings of the bootstrap PQR models calculated for the low (10th–30th), medium (40th–60th), and high (70th–90th) ranges. When the models are considered, it is seen that the PEC variable is statistically significant and positive at the 1% significance level in all quantiles. In this context, primary energy consumption suppresses global warming and climate change by increasing GHG. The variable that affects emissions the least in the same direction (30th) is the ET variable. The findings obtained provide information indicating that primary energy consumption has the most significant impact on environmental quality, while energy transition has the least significant impact. However, the EPS variable reduces emissions in line with expectations at low quantiles, as well as at 50th and 60th quantiles. The GT variable causes environmental deterioration at all low and medium quantiles. The (EPS*ET) variable increases global warming at the 10th, 30th, 50th, and 60th quantile values. At the same time, the (EPS*GT) variable provides environmental improvement at all low and medium quantities. The general evaluation in terms of quantile values supports that all variables are statistically significant at the 30th quantile. The values with the smallest statistical significance levels are the 80th and 90th quantiles. In this context, it is seen that higher GHG emissions are associated with increased climate change mitigation efforts at lower quantiles. It is also noteworthy that the R² values indicating the explanatory power of the established models decrease at the 10th–70th quantiles and increase at the 80th–90th quantiles.

4.4. Robustness Checks

In order to test the robustness of the results in the assumptions of possible heteroscedasticity, autocorrelation, and cross-sectional dependence in the model, the model was re-estimated using the Driscoll–Kraay robust estimator. In addition, the Hausman test was performed for the preference processes of the fixed effect model or random effect model for the relationship between the unobservable effect and the independent variables, and the findings are shown in

Table 7. Driscoll–Kraay and Ridge regression model robustness tests.

Variables	Driscoll–Kraay Model		Ridge Regression Model
	Coef.	p > |t|	Coef.	p > |t|
EPS	−0.1257298 ***	0.091	−0.5144 *	0.010
ET	0.0071383 ***	0.051	−0.0112 **	0.013
GT	0.0004587 ***	0.052	0.0007 ***	0.067
PEC	0.0818708 *	0.000	0.0732 *	0.000
(EPS*ET)	−0.0029816	0.235	0.0034	0.600
(EPS*GT)	−0.0001412 **	0.038	−0.0002	0.247
C	0.7676726 **	0.037	2.3925 *	0.000
R2	0.96		0.96
Hausman Test	chi2(6) = 26.18 Prob > chi2 = 0.0002

The symbols *, **, and *** are 1%, 5%, and 10% probability values, respectively.

. The findings obtained explain that the H0 hypothesis was rejected at the 5% significance level and that the random effects estimator was inconsistent, and the fixed effects estimator was consistent. When the Driscoll–Kraay robust findings of the fixed effects estimator are examined, it is seen that the EPS and (EPS*GT) variables improve environmental quality, while the ET, GT, and PEC variables deteriorate it, consistent with the findings obtained from the PQR estimator. Ridge regression analysis was performed against the possible multicollinearity problem in the model due to the (EPS*ET) variables derived from the EPS, ET variables, and (EPS*GT) variables derived from the EPS and GT variables. When the findings are evaluated together with the PQR findings, it is indicated that the EPS, ET, GT, and PEC variables determine environmental degradation. In this context, it is observed that the EPS variable remains the most significant factor affecting air quality. At the same time, the robustness of the PQR findings is also supported.

4.5. Granger Causality Test

Table 8. Dumitrescu–Hurlin panel Granger causality test.

H0 Hypothesis	Wald Stat.	z-Val.	p-Val.	H0 Hypothesis	Wald Stat.	z-Val.	p-Val.
GHG ≠ EPS	9.8218	2.7024	0.0069 *	GHG ≠ PEC	2.8474	3.1998	0.0014 *
EPS ≠ GHG	10.9953	4.6440	0.0000 *	PEC ≠ GHG	3.6275	1.9933	0.0462 **
GHG ≠ ET	2.2616	2.1851	0.0289 **	GHG ≠ (EPS*ET)	4.4070	5.9010	0.0000 *
ET ≠ GHG	15.9898	7.0639	0.0000 *	(EPS*ET) ≠ GHG	10.3513	3.0769	0.0021 *
GHG ≠ GT	10.5627	3.2263	0.0013 *	GHG ≠ (EPS*GT)	6.3159	5.2859	0.0000 *
GT ≠ GHG	9.8287	2.7073	0.0068 *	(EPS*GT) ≠ GHG	18.7109	8.9879	0.0000 *

The symbols * and ** are 1% and 5% probability values, respectively.

explains the Dumitrescu–Hurlin panel causality test findings. According to the table, it is reported that there is a two–way causality relationship between GHG emissions and EPS, ET, GT, PEC, (EPS*ET), and (EPS*GT) variables that affect each other. The findings obtained are consistent with the findings of the bootstrap PQR estimator in Table 6.

5. Discussion

This article evaluates the impact of environmental policy stringency, energy transition, green technology, and primary energy consumption on GHG emissions for the SDG 7 and SDG 13 goals of BRICS–T economies. Renewable energy is integrated into the model through the interactions of energy transition with stringent environmental policies and green technology.

PQR, Driscoll–Kraay, and Ridge regression findings explain that the EPS variable positively affects climate action in line with expectations. In this context, the hypothesis ‘H1: EPS improves environmental quality through technology support, market and non-market-based policies’ is accepted. Therefore, the sound EPS policies implemented in BRICS–T countries have helped to increase the positive impacts on environmental improvement. The findings are consistent with the empirical literature findings of studies conducted on OECD countries [8,11,21,22,29,34,37] and at the global level [12,13,14,15,16,17]. In the same direction, the PQR and Driscoll–Kraay findings accepted the hypothesis ‘H4.a: The combined effect of EPS and GT has a more positive effect on environmental degradation’, which states that the (EPS*GT) variable reduces emissions in the long run and accelerates sustainable development. The findings are similar to the findings of the studies conducted by Abbasi et al. [23] and Fatima et al. [38] on the scale of OECD countries. The evaluation of the findings together points to the transition of renewable energy through strict environmental regulations and green technology in BRICS–T economies. It is stated that the implementation of strict environmental policies encourages the adoption of green energy to reduce emissions and mitigate the negative effects of climate change. However, it is reported that the individual effect of the EPS variable on GHG reduction is greater than the combined effect of the (EPS*GT) variable.

The findings that the ET and GT variables worsen environmental degradation in PQR and Driscoll–Kraay analyses led to the rejection of the hypotheses ‘H2: The transition process from conventional energy to renewable energy reduces environmental degradation by reducing GHG emissions’ and ‘H4: GT reduces GHG emissions by reducing carbon emissions and contributing to carbon neutrality’. The PQR analysis indicated that the (EPS*ET) variable increases the level of emissions, indicating that the hypothesis ‘H2.a: The combined effect of EPS and ET has a more positive impact on environmental degradation’ is not confirmed. One reason for the results for these three hypotheses is the high share of conventional energy in the energy supply of BRICS–T economies. The findings are similar to the results of the studies conducted by Abbasi et al. [23] for OECD countries, Kurt [33] for 16 European countries, and Ouni et al. [35] for the G20 countries in the literature selection. However, it was observed that the effect of the relevant variables on emissions was quite limited. Therefore, it is emphasized that ET, GT, and (EPS*ET) are not sufficient for improvements in environmental sustainability. The reasons for the increase in the specified variables and the tendency to increase emissions are considered insufficient investments in low–carbon technologies, the fact that technological innovations are still in the development stage [33] and the low share of renewable energy in the energy supply. It is currently assumed that innovative technologies increase carbon emissions because they are produced with established resources. Theoretically, the potential for environmentally based technological innovations to turn into an adverse effect in the future, with the spillover effect of technologies, can be expected [30]. The heterogeneous effect of environmental technologies may be due to the different levels of adoption of environmental technologies in the sample countries [38].

In line with the findings obtained from all models indicating that PEC negatively affects climate change, the hypothesis ‘H3: PEC negatively affects environmental quality by increasing environmental externalities’ was accepted. The fact that primary energy consumption has a strong tendency to increase emissions has similarities with the literature studies conducted by Abbasi et al. [23] and Frohm et al. [8] in OECD countries with Wolde-Rufaela and Weldemeske [26] in BRIICTS countries. The increasing effect of PEC on GHG emissions is related to the resource composition of primary energy in the countries in question [7]. The BRICS–T group consumes approximately 40% of total energy globally. Of these countries, Brazil consumes the highest amount of electricity from renewable energy, boasting one of the cleanest energy matrices in the world. Furthermore, approximately 89% of the country’s total electricity supply comes from renewable energy sources. Conversely, China and South Africa have the lowest rates of renewable energy use [51]. Consequently, each country in the sample is considered to have its own unique qualifications. Therefore, different policies are needed for improving and accelerating the energy transition. The characteristics of the countries in the sample with different income levels (Brazil, China, South Africa, and Turkey are upper–middle income; Russia, high-income; India, lower–middle-income) [52] are seen as another reason for the results obtained. Especially in middle-income economies with higher economic growth rates and greater energy demand, governments need to implement a variety of supportive policies to facilitate access to electricity from clean sources, aligning with the SDGs.

Limitations

One of the main goals of developing environmental policies is to achieve sustainable and traceable outcomes, such as a lower carbon economy and GHG emissions. In this context, it is observed that the EPS index measurement is frequently employed by researchers to assess empirical outcomes related to climate action. However, the fact that the index covers 40 countries (34 OECD; 6 non-OECD) and the dataset starts from 1990 are the most important limitations of the research. Another limitation is that the temporal path of the EPS dataset is currently up to 2020. Since this study focuses on BRICS-T economies, the results obtained on energy and climate action cannot be generalized to different country groups. Furthermore, differences in economic development levels among countries within the BRICS–T group, which are considered emerging market economies due to their economic performance, are also considered a research limitation. Another limitation is the selection of explanatory variables included in the model when examining environmental sustainability within the relevant sample group.

6. Conclusions

This paper aims to guide the development of policies that promote strict environmental policies, energy transition, and green technology in BRICS–T economies and assess their impact on sustainable development. Insights highlight the need to focus on promoting green innovation, adopting comprehensive fiscal and environmental policies, regulating emissions through carbon pricing, prioritizing renewable energy, and strengthening policy measures to achieve long-term ecological goals, thereby enhancing sustainability and reducing emissions.

The findings highlight that efforts to move toward a more sustainable ecological trajectory in BRICS–T countries are insufficient. It points to the need for policymakers to reorganize [37] and critically evaluate existing environmental policies, implement necessary reforms that balance stricter implementation of environmental policies with financial incentives to reduce the costs of the clean energy transition. This necessitates the development of more effective and comprehensive strategies to promote environmental sustainability. The fact that the effect of the EPS index on GHG emissions is greater than other variables should be taken into account, especially in the design of environmental policies. Despite this positive result, primary energy use needs to be minimized. It is suggested that environmental regulations related to energy should be reviewed. The mentioned measure is especially important, considering that most of the emissions are produced by the energy sector [29].

The BRICS–T group, characterized as emerging market economies within the scope of developing countries, has focused on a pro–growth agenda and paid little attention to environmental sustainability. Economic development is significantly dependent on energy use, particularly as it drives growth and industrialization. The large disparity between renewable and conventional energy sources in the energy mix (except for Brazil) disadvantages clean energy. Furthermore, energy demand has increased in recent years due to several factors, such as rapid population growth, urbanization, and industrialization. Therefore, growth, which is a primary goal for countries, highlights the negative impacts of the linkages between energy and the environment. Consequently, among the economies in the BRICS–T group, China, India, Turkey, and Russia are among the world’s ten most polluting countries and are classified as high-polluting economies [51].

The high installation costs of renewable energy sources, combined with the requirement for specialized technological infrastructure, can negatively impact economic growth. However, policymakers in BRICS–T countries should increase investment in clean and green energy sources with a long-term perspective, despite the initial negative impacts. This will enable them to approach sustainable development goals in the long term by reducing dependence on imported energy and addressing environmental degradation. Furthermore, decision-makers should develop and implement cost-cutting policies. In this context, providing cost–cutting incentives through tax deductions and subsidies for private sector R&D activities could be considered as alternatives. At the same time, efforts should be made to address technological infrastructure deficiencies by increasing investments in R&D activities in the public sector [53].

In light of the results, sample policy scenarios show that achieving emission reductions consistent with net zero targets by BRICS–T economies requires implementing stricter environmental policies [8,21] supported by technological developments and clean energy initiatives. It underlines the need to increase green technology innovations, reduce dependence on non-renewable energy imports, and accelerate energy transition measures to reduce GHG emissions. It emphasizes the need for integrated, dynamic, and adaptable environmental policies in line with SDG 7 and SDG 13. However, these policies should be complemented by environmental awareness strategies. Policy implications may enable these economies to advance their energy transitions while also accelerating their sustainability goals.

Based on the contributions and limitations of this study, future research directions are recommended to primarily address the integrated relationships between environmental policy stringency and energy transition, and between environmental policy stringency and green technology, at a micro-level. Therefore, future research could provide a more detailed perspective on countries within the context of the Sustainable Development Goals. Furthermore, conducting similar studies across different regions could allow for the extension of the results to a broader context. Furthermore, using different proxy variables to represent energy transition and green technology could increase the clarity of the results on energy and climate action.