The Race to Zero Emissions in MINT Economies: Can Economic Growth, Renewable Energy and Disintegrated Trade Be the Path to Carbon Neutrality?

Abstract

The current paper evaluates the role of disintegrated trade, financial development, and renewable energy on consumption-based carbon emissions (CCO₂) in MINT nations between 1990Q1 and 2019Q4. This paper utilizes the novel Bootstrap Fourier Granger causality in quantiles (BFGC-Q) to evaluate this connection. This approach produces tail-causal and asymmetric causal connections between the indicators within the Fourier approximation, contrary to the Toda–Yamamoto causality and other conventional Granger tests. The outcomes uncover a unidirectional causality from economic growth and renewable energy to CCO₂ emissions in each MINT nation. Moreover, unidirectional causality emerged from financial development to CCO₂ for Indonesia, Nigeria, and Turkey. Moreover, exports have predictive power over CCO₂ in Indonesia, Turkey, and Mexico, while imports only have predictive power over CCO₂ emissions in Turkey. Lastly, financial development causes CCO₂ in Indonesia, Nigeria, and Mexico. In summary, green energy and exports are essential factors that decrease CCO₂ emissions and therefore decrease ecological deterioration in Mexico, Indonesia, and Turkey. On the flip side, imports only trigger CCO₂ emissions in Turkey and Mexico. Lastly, the financial development effect on CCO₂ emissions is positive in Mexico, Indonesia, and Nigeria, while an insignificant impact is found in Turkey. Based on these findings, policy ramifications are initiated.

1. Introduction

The difficulties of environmental degradation and climate change have rapidly surfaced in recent years, posing serious concerns for the international community’s and policymakers’ pursuit of sustainable development. The global economy has entered a phase of fast growth following the industrial revolution, and the wealth disparity has been growing [1,2]. Ecological contamination is a problem that arises concurrently with economic growth and poses a danger to human life. Huge industrial waste, intensive use of natural resources, and the usage of energy based on fossil fuels are the main causes of these problems [3,4]. Various nations have established the targets for carbon emission peaks and reductions at the recent summits (COP21, COP26) on ecological regulation and climate change in order to attain net zero emissions and achieve harmonized sustainable environment and growth. In order to accomplish zero emissions in the next decades, global leaders are working to put regulations/policies in place that will result in net-zero emissions.

Trade economists are the first to evaluate the issue of ecological deterioration [5]. These scholars offer a fundamental basis for comprehending how trade and the environment are correlated. One of the CO₂ emissions drivers is international trade [6,7]. On the one hand, international trade has augmented the flow of services and goods, thus boosting economic operations. Nevertheless, it has unfavorable effects on the ecosystem. According to [1], international trade allows nations to transfer their polluting sectors to other nations, contributing to environmental deterioration. However, trade raises nations’ economic levels, which may be utilized to slow down ecological damage in its latter phases [5]. Nations worldwide are reallocating their resources to effective initiatives and implementing eco-friendly technologies to harmonize the relationship between CO₂ emissions and trade.

Overusing energy puts a lot of strain on the ecosystem [8]. Nevertheless, green energy (such as solar, biomass, wind, hydro, and geothermal) results in less CO₂ than using fossil fuels, which are thought to be the primary cause of global warming and CO₂ emissions [9]. Therefore, renewable energy sources are one of the most vital strategies to curb CO₂ [10,11]. After the well-known COP 21 and Kyoto Protocol in 2005, most advanced nations embraced renewable energy sources as a propelling tactic to attain a target of low GHG emissions. Various studies have incorporated renewable energy (REC) as a significant variable in the CO₂ emissions framework due to its significance in reducing CO₂ emissions [12,13].

Financial development (FD) also contributes significantly to a nation’s growth. A robust and enhanced financial sector boosts the financial system’s effectiveness while promoting economic development and growth [14]. Although energy is the main driver of economic expansion, it also has an unavoidable impact on the environment. As a result, the flow of financial resources is correlated with the need for energy. More funds are required for manufacturing to increase energy efficiency and deploy superior technologies to promote economic growth. Due to this, financial development has increasingly been a significant factor in economic growth [15,16,17]. Financial development boosts the economy, but it also has disadvantages since it may have an adverse effect on the environment and deplete natural resources in many ways. Particularly, the growth of finance pushes people to purchase more homes, machines, cars, and gadgets, intensifying the increasing need for energy [18,19].

In light of the preceding debate, this paper aims to inspect the impact of disintegrated trade, renewable energy, and financial development on consumption-based carbon emissions (CCO₂) in MINT nations. Brazil, Russia, India, China, and South Africa, together known as the BRICS nations, are powerful emerging blocs that have drawn significant attention recently. Moreover, Ref. [20] also acknowledged additional emerging markets in 2013, such as Mexico, Indonesia, Nigeria, and Turkey (MINT). The MINT nations account for between 1 and 2 percent of the global economy and have a good chance of surpassing other nations’ economies in terms of economic size and technological advancement in the next decades. Although the USA likewise anticipates a 5% growth rate for each MINT country globally, Gold Sachs predicts consistent, steady growth for MINT nations [21]. Now the question arises, does developing nations like MINT economies uphold sustainable development with financial development and disintegrated trade without damaging the environment? The present investigation is carried out to provide an answer to this question.

This paper adds to the ongoing literature in three distinct ways: Firstly, the present investigation considered the impact of disintegrated trade by evaluating the role of imports and exports on CCO₂ emissions. Secondly, unlike prior studies such as [8,22,23,24,25] that used CO₂ emissions to gauge environmental degradation, the current investigation used CCO₂ emissions to measure ecological deterioration. As stated by [26], the CCO₂ is a comprehensive measure of ecological damage because it helps differentiate emissions produced in one country and consumed in another. Thus, emissions from imports and exports are taken into consideration when using this metric. Thirdly, the research employed BFGC-Q, initiated by [27], for the MINT nations between 1990Q1 and 2019Q4. This approach produces tail-causal and asymmetric causal connections between the indicators within the Fourier approximation, contrary to the Toda–Yamamoto causality and other conventional Granger tests.

A synopsis of relevant investigations is presented in the next part, and then Section 3 contains the data and methodology. In Section 4, study results and analyses are reported, and Section 5 brings the research to a close.

2. Literature Review

2.1. Synopsis of Studies between Environmental Quality and Financial Development, and Renewable Energy

Studies on the nexus between ecological quality (EQ) and financial development and green energy have been conducted in the empirical literature. For instance, Ref. [28] inspect the role of green energy (GRN)and export diversification on Indonesia’s ecological quality (EQ) using the novel Fourier quantile causality method from 1965Q to 2014Q4. The findings indicate that there is one-way causation from fossil fuel to EQ at all quantiles. Still, the causes of EQ at the intermediate and higher quantiles include income, green energy, and export diversification. The EQ is most significantly raised by green energy and export diversification. In contrast, a rise in real growth and the use of fossil fuels lower EQ. Moreover, Ref. [29] revisits the nexus between green energy, financial development, and EQ towards attaining sustainable development in China. The research assesses updated time series data for China between 1988 and 2018 and employs cutting-edge econometric methods, including the Maki cointegration and frequency domain causality test. The empirical finding demonstrates that EQ is enhanced by increased financial development and using REC. Income, on the other hand, lowers EQ. Additionally, the 2008 structural break year and financial development raise EQ. The strong correlation between financial development and EQ confirms the school of belief that relates financial development with sustainability.

Using data from 1980 to 2018, Ref. [30] inspected the nexus between renewable energy, financial development, and EQ in selected Asian nations using panel methods between 1990 and 2014. The empirical study indicates that while economic expansion and financial development lower EQ, renewable energy helps raise it, while agriculture has less influence. The findings propose that all regressors can forecast EQ in the chosen countries, and the causality between the variables is tested using the variance decomposition and impulse response function approaches. Likewise, Ref. [31] inspected the environmental effects of financial development and REC using Driscoll–Kraay Panel Corrected Estimators for 16 developing nations between 2000 and 2018. The findings show that REC and financial development strengthen EQ. The developing nations have already passed the EKC tipping threshold for internet usage, wherein EQ rises as internet penetration increases. Furthermore, robustness testing using bootstrapped panel-quantile regression also supports the notion that financial development and REC promote EQ in each quantile.

Likewise, Ref. [32] evaluates how renewable energy and financial development promote EQ using global data between 1990 and 2018. The research used the estimators’ DOLS, CCR, and FMOLS to assess the nexus. The long-term interrelationship between the indicators is supported by empirical research. Their findings also discover that worldwide economic expansion lowers EQ globally, whereas financial development and green energy consumption have a long-term significantly favorable impact on EQ. Ecological issues in the era of industrialization were evaluated by [33] by incorporating financial development and REC as control variables. The study used panel data from NICs for the years 1990 to 2019. The study used panel data from NICs for the years 1990 to 2019. The augmented mean group (AMG) results indicate that EQ in these nations is significantly impacted by financial development. On the other hand, using renewable energy greatly raises EQ over time. Additionally, these findings are in line with long-term and disaggregated level estimates. The panel causality test findings also found a unidirectional causation relationship from REC to EQ. Additionally, it was shown that EQ and financial development had a reciprocal causal interrelationship.

Using data from 1960Q1 to 2019Q4, Ref. [34] evaluated the environmental cost of economic progress, financial development, and renewable energy in Pakistan using nonparametric causality-in-quantiles techniques. The research indicated that EQ strongly correlates with financial development and REC, showing asymmetric prediction over ecological dispersion. Additionally, there is a connection between financial development and EQ at higher quantiles.

2.2. Synopsis of Studies between Environmental Quality and Exports and Imports

Mahmood [35] used spatial regression analysis to assess the effect of trade (imports and exports) on EQ in GCC nations, utilizing data from a period between 1990 and 2019. Exports have positive spillovers, direct and total impacts on EQ, and negative direct effects on EQ. The fact that exports negatively impact EQ indicates that exports are lowering EQ in domestic economies. On the other hand, the positive direct impact of exports on EQ demonstrates how exports enhance EQ in domestic economies.

Hasano et al. [36] assessed the impact of international trade on EQ in oil-exporting nations using both consumption and territory emissions accounting. The error correction and cointegration models indicate that imports and exports have significant statistical effects of opposing signs on EQ in both the short and long-term, and that the consequences of alterations in the trade-CCO₂ connection will be entirely assimilated in three years. Nevertheless, regarding territory-based CO₂ emissions, imports and exports are statistically insignificant. Moreover, Ref. [26] inspected the role of international trade in G7 countries on consumption-based carbon emissions using second-generation approaches. The findings support a consistent long-term connection between CO₂ emissions and trade. In the long term, exports reduce CCO₂, whereas imports increase it. The outcomes are also supported by the CCEMG and AMG methodologies. Based on the findings of the Granger causality test conducted by [37], it is said that any policy that targets imports and exports has a considerable impact on CCO₂ emissions.

Similarly, Ref. [38] evaluated the theoretical framework for the effect of trade (exports and imports) on CCO₂. The data from the BRICS nations for 1990 to 2017 are then used to evaluate this connection. The research also considers the panel data’s integration, cointegration, heterogeneity aspects, and cross-country interdependence, resulting in reliable findings and well-founded policy recommendations. According to their findings, export size contributes to EQ growth, whereas import size dampens EQ. Furthermore, Ref. [39] using the BRICS nations evaluates the nexus between trade and CCO₂ using data from 1990 to 2018. The study evaluates these interactions using the AMD and CCEMG causality methodologies. The study results show that in the BRICS economies, exports reduce CCO₂, however, imports increase CCO₂. Furthermore, all the parameters can predict CCO₂ emissions according to the panel causality results. To achieve carbon neutrality for the G7 nations, Ref. [7] evaluated the disintegrated trade effect on EQ using data from 1990 to 2018. The results suggested that exports and imports are factors of CCO₂ in the G7 nations. In addition, exports curb CCO₂ while imports upsurge CCO₂ emissions.

In a variety of ways, this research contributes to the expanding body of scholarship on ecological deterioration. Firstly, the analysis is new because it uses the newly formed CCO₂ emissions, which determine emissions based on domestic fossil fuel usage plus incorporated emissions from net exports (export minus import). For the purpose of developing an effective climate strategy to address ecological issues, a precise assessment of CO₂ emissions is crucial. Following the Paris climate summit (COP, 21), it is possible to propose a pertinent climate policy response based on trade-adjusted data on CO₂ emissions. Secondly, the research employed BFGC-Q initiated by [27], for the MINT nations between 1990Q1 and 2019Q4. This approach produces tail-causal and asymmetric causal connections between the indicators within the Fourier approximation, in contrast to the Toda–Yamamoto causality and other conventional Granger tests.

3. Data and Methodology

3.1. Data

The current research evaluates the causal/interrelationship between CCO₂ emissions and imports, green energy, exports, financial development, and economic growth in the MINT nations. The study used data from 1990 to 2019 to assess the interrelationship. The dependent variable is CCO₂ while imports, renewable energy, financial development, exports, and GDP are the regressors. To minimize issues with small observations, all the yearly frequency data are adjusted to logarithmic values utilizing the quadratic match-sum approach and then normalized to quarterly frequencies. It is favored over other interpolation approaches because it takes seasonality into account by minimizing data changes when it switches from low to high frequency [4,25,40]. Statistical descriptions for quarterly data are provided in

Table 1. Descriptive Statistics.

	Mexico
	CCO2	FD	REC	GDP	IMP	EXP
Mean	448.20	0.3344	10.949	8747.9	28.249	26.666
Median	469.80	0.3282	10.228	8820.9	27.674	26.142
Maximum	547.74	0.4008	15.187	9954.3	41.454	39.410
Minimum	312.54	0.2521	8.9487	7343.5	15.156	11.459
Std. Dev.	71.262	0.0468	1.6589	735.07	6.6390	7.0045
Skewness	−0.4887	−0.1151	0.5316	−0.2295	0.1749	−0.1450
Kurtosis	1.8650	1.5927	1.8958	2.0920	2.4342	2.6829
Jarque–Bera	11.218	10.167	11.748	5.1754	2.2127	0.9236
Probability	0.0036	0.0061	0.0028	0.0751	0.3307	0.6301
	Indonesia
Mean	357.44	0.3122	2423.7	25.676	40.485	28.481
Median	317.97	0.3075	2159.1	25.309	41.515	26.742
Maximum	693.81	0.4001	3931.9	44.226	58.833	54.776
Minimum	132.54	0.2383	1462.1	16.601	18.943	16.979
Std. Dev.	159.57	0.0415	691.89	4.6656	10.914	7.1238
Skewness	0.4235	−0.0019	0.6322	1.5145	−0.1826	1.4132
Kurtosis	1.8527	2.0246	2.1623	7.7038	2.1152	5.8497
Jarque–Bera	10.169	4.7569	11.501	156.50	4.5811	80.553
Probability	0.0061	0.0926	0.0031	0.0000	0.1012	0.0000

and

Table 2. Descriptive Statistics.

	Nigeria
	CCO2	FD	GDP	IMP	REC	EXP
Mean	76.929	0.1898	1969.8	15.379	85.064	21.893
Median	80.407	0.1893	1916.9	14.378	85.168	22.052
Maximum	131.12	0.2739	2705.1	23.428	88.842	37.157
Minimum	33.529	0.1167	1411.3	8.595	80.541	8.8927
Std. Dev.	33.004	0.0352	470.43	3.9310	2.2691	6.4406
Skewness	0.1886	−0.0619	0.1982	0.4743	−0.2275	0.0526
Kurtosis	1.6725	2.6520	1.4011	2.1677	1.9715	2.6046
Jarque–Bera	9.5224	0.6819	13.568	7.9636	6.3246	0.8368
Probability	0.0085	0.7110	0.0011	0.0186	0.0423	0.6580
	Turkey
Mean	306.00	0.4006	7941.7	24.680	17.005	22.299
Median	288.51	0.4053	7475.6	25.105	16.010	22.840
Maximum	445.09	0.5311	12022.	31.515	24.712	32.760
Minimum	206.13	0.1920	5286.7	16.568	11.208	12.629
Std. Dev.	82.533	0.1010	2179.5	4.2417	4.4882	4.4437
Skewness	0.2293	−0.4415	0.5619	−0.2954	0.4453	−0.1846
Kurtosis	1.4464	2.0872	1.9781	2.1182	1.7070	3.5362
Jarque–Bera	13.120	8.0644	11.536	5.6326	12.325	2.1200
Probability	0.0014	0.0177	0.0031	0.0598	0.0021	0.3464

. The six variables in each MINT country are not distributed normally, as shown by the Jarque–Bera test probability values. It is preferable to utilize median-based quantile causality tests for non-normally distributed series rather than mean-based conventional Granger causality tests [25,41]. Premised on this, we use the Fourier causality test to assess the factors affecting CCO₂ emissions. Figure 1 presents the flow of the study.

3.2. Theoretical Framework

This section explains the theoretical procedure through which imports and exports, economic growth, financial development, and renewable energy impact CCO₂ emissions. CCO₂ emissions encompass both household and government final domestic consumption demand, gross fixed capital formation, inventory changes, and purchases made overseas by residents [9,42]. This indicator is trade-adjusted, covers the entire carbon chain, and aids in identifying the production of carbon emissions in one nation and their absorption in another [7,43,44]. As a result, the impact of international trade in this research is calculated by separating imports and exports. According to the theory, increased exports give more products and services to recipient nations to consume while leaving less for local consumption. Exports include services and goods produced in one nation and used in another. As a result, the receiving nation’s CO₂ from exports must be emitted. Thus, EXP is anticipated to decrease CCO₂ emissions, i.e., ${\mathsf{\beta }}_2=\left(\frac{{\mathsf{\theta }\mathrm{CCO}}_2}{\mathsf{\theta }\mathrm{EXP}}<0\right)$.

On the other hand, imports encompass services and goods manufactured by a foreign nation and used locally, and must release CO₂ domestically. It is projected that boosting exports will cut CCO₂ emissions in the host nation, whereas expanding imports will boost CCO₂ emissions in the recipient state. Aside from imports and exports, carbon emissions from the process of production are retained in the host nation [36,38,45,46]. Theoretically, an increase in imports is associated with increased consumption because it is regarded as one of the essential parts of any nation’s overall consumption level, which is particularly true in the case of MINT nations. The MINT economies are emerging economies, and their imports include a significant share of intermediate and final services and goods consumed by the host economies. Several studies, such as [35,38,47], have already noticed this behavior. Thus, REC is anticipated to decrease CCO₂ emissions, i.e., ${\mathsf{\beta }}_2=\left(\frac{{\mathsf{\theta }\mathrm{CCO}}_2}{\mathsf{\theta }\mathrm{IMP}}>0\right)$.

Likewise, GDP is a gauge of the economy’s health and includes several parts, such as consumption, investment, government spending, and net exports. Since consumption accounts for the majority of GDP, increased consumption is positively related to CCO₂ emissions [48,49]. Thus, GDP is anticipated to increase CCO₂ emissions, i.e., ${\mathsf{\beta }}_3=\left(\frac{{\mathsf{\theta }\mathrm{CCO}}_2}{\mathsf{\theta }\mathrm{GDP}}>0\right)$. The theoretical foundation for the renewable energy consumption and CCO₂ emissions negative connection is that renewable energy technologies use sustainable and greener energy sources that meet future and current demands [50,51]. Based on the above principles, renewable energy usage is predicted to reduce CO₂ emissions. Thus, REC is anticipated to decrease CCO₂ emissions, i.e., ${\mathsf{\beta }}_5=\left(\frac{{\mathsf{\theta }\mathrm{CCO}}_2}{\mathsf{\theta }\mathrm{REC}}<0\right)$.

A stable financial market has the potential to support sustainable energy, which would be advantageous for the ecosystem. Likewise, some investigations contend that the stock market will help to preserve the ecosystem by increasing financial access, expanding financial networks, mobilizing the capital needed to invest in eco-friendly infrastructure and lowering manufacturing costs. According to some analyses, financial development may attract FDI and spur innovative research to enhance the ecosystem. As per [52], financial development may facilitate investment in energy conservation technologies to increase ecological integrity. On the other hand, some studies have cautioned that higher financial development may lead to more CO₂ [25,53]. According to [54], a stable financial system can encourage investment but also damage the environment by increasing energy use. Thus, financial development is anticipated to decrease CCO₂ emissions, i.e., ${\mathsf{\beta }}_5=\left(\frac{{\mathsf{\theta }\mathrm{CCO}}_2}{\mathsf{\theta }\mathrm{FD}}<0\right)$ or increase CCO₂ if not eco-friendly, i.e., ${\mathsf{\beta }}_5=\left(\frac{{\mathsf{\theta }\mathrm{CCO}}_2}{\mathsf{\theta }\mathrm{FD}}>0\right)$.

3.3. Methodology

Nonlinearities and structural break (s) are not considered by the traditional [55] causality test. Moreover, Ref. [56] improved the vector autoregression (VAR) model by including Fourier approximations to avoid causality analysis by ignoring structural breaks. This allowed for the inclusion of smooth structural break(s) in the causality analysis. Nevertheless, the method in [56] does not guard against information loss over the long term. As a result, the [57] causality test was updated by [58] to include the Fourier approximation to safeguard against long-term information loss and consider smooth structural modifications. In this approach, termed Fourier-TY, the technique of [59] is utilized as shown in Equation (1).

$$\mathsf{\alpha }\left(\mathrm{t}\right){=\mathsf{\alpha }}_0{ + \mathsf{\gamma }}_1\sin \left(\frac{2\mathsf{\pi }\mathrm{kt}}{\mathrm{T}}\right){+\mathsf{\gamma }}_2\cos \left(\frac{2\mathsf{\pi }\mathrm{kt}}{\mathrm{T}}\right)$$

(1)

where sin and cos represent sine and cosine, optimal frequency is shown by k; the trend is depicted by t, the observation number is shown by T, and $\mathsf{\pi }$ is roughly equal to 3.145. The TY causality test in Equation (2) is replaced by α(t) in this Fourier approximation.

$$y_t=\alpha \left(t\right)+{\delta }_1{y}_{t-1}+\dots +{\delta }_{j+pmax}{y}_{t-\left(j+pmax\right)}+e_t$$

(2)

where the time intercept is denoted by $\alpha \left(t\right)$ time-dependent intercept, the optimal lag is denoted by j, the maximum integration order of variables is denoted by pmax, and the error term is shown by $e_t$. The presumption that the constant term does not shift with time is relaxed by [58] by replacing the Fourier approximation in Equation (1) for α(t). As a result, the TY causality test considers smooth structural transitions with an undetermined structure, date, and number, as depicted in Equation (3).

$${\mathrm{y}}_{\mathrm{t}}{=\mathsf{\alpha }}_0{+\mathsf{\gamma }}_1\sin \left(\frac{2\mathsf{\pi }\mathrm{kt}}{\mathrm{T}}\right){+\mathsf{\gamma }}_2\cos \left(\frac{2\mathsf{\pi }\mathrm{kt}}{\mathrm{T}}\right)+{\delta }_1{\mathrm{y}}_{\mathrm{t}-1}+\dots +{\delta }_{\mathrm{j}+\mathrm{pmax}}{\mathrm{y}}_{\mathrm{t}-\left(\mathrm{j}+\mathrm{pmax}\right)}{+\mathrm{u}}_{\mathrm{t}}$$

(3)

In Equation (3), the cos and sin significance are evaluated with an F-test to help ascertain whether their coefficients are equal to 0 (${\mathsf{\gamma }}_1$ = ${\mathsf{\gamma }}_2$ = 0). It is suitable to employ the Fourier approximation if the coefficients differ from 0. Therefore, the causality interrelationships between indicators can be evaluated as ${\delta }_1=\dots {\delta }_{\mathrm{j}}=0$.

While the FTY causality test initiated by [58] has several benefits, the conventional least squares approach is ineffective when the series is distributed normally and has a non-linear form. Hence, based on quantile regression, [27] recommended using the Fourier TY causality test. This novel method, termed “Bootstrap Fourier Granger causality in quantiles” (BFGC-Q), is shown below in Equation (4).

$${\mathrm{y}}_{\mathrm{t}}(\mathsf{\tau }│{\mathrm{Z})=\mathsf{\alpha }}_0\left(\mathsf{\tau }\right)+{\displaystyle \sum }_{\mathrm{k}=1}^{\mathrm{n}}{\mathsf{\mu }}_1\left(\mathsf{\tau }\right)\sin \left(\frac{2\mathsf{\pi }\mathrm{k}*\mathrm{t}}{\mathrm{T}}\right)+{\displaystyle \sum }_{\mathrm{k}=1}^{\mathrm{n}}{\mathsf{\mu }}_2\left(\mathsf{\tau }\right)\cos \left(\frac{2\mathsf{\pi }\mathrm{k}*\mathrm{t}}{\mathrm{T}}\right)+{\delta }_1{\left(\mathsf{\tau }\right)\mathrm{y}}_{\mathrm{t}-1}+\dots +{\delta }_{\mathrm{j}*+\mathrm{pmax}}{\left(\mathsf{\tau }\right)\mathrm{y}}_{\mathrm{t}-\left(\mathrm{j}*+\mathrm{pmax}\right)}{+\mathrm{v}}_{\mathrm{t}}$$

(4)

where $\mathrm{k}*$ and $\mathrm{j}*$ are the optimal frequency and lag length, respectively, τ and Z represent a specific quantile and covariate matrix. The following can be used to test the null of non-causality in various quantiles by estimating Equations (5) and (6):

$$H_0:{\widehat{\delta }}_1\left(\mathsf{\tau }\right)={\widehat{\delta }}_2\left(\mathsf{\tau }\right)\dots {\widehat{\delta }}_{\mathrm{j}*}=0, \forall \mathsf{\tau }\in \left(0,1\right)$$

(5)

$$Wald=\left[T\left({\left(\widehat{\delta }\left(\mathsf{\tau }\right)\right)}^{\prime }\right){\left(\widehat{\Omega }\left(\mathsf{\tau }\right)\right)}^{-1}\left(\widehat{\delta }\left(\mathsf{\tau }\right)\right)\right]/\mathsf{\tau }(1-\mathsf{\tau })$$

(6)

Following that, the BFGC-Q causality test’s Wald statistics are computed. The critical values acquired from the bootstrap simulations are then contrasted with the Wald statistics gathered using Equation (6). The occurrence of causation can be determined if the Wald statistic for the relevant quantile is higher than the threshold value.

4. Empirical Results

4.1. Stationarity Test Results

In this paper, we employ the BFGC-Q method to test the causality between CCO₂ emissions and IMP, EXP, FD, GDP, and REC in the MINT nations, utilizing data between 1990Q1 and 2019Q4. The maximum integration order of the series is verified in the first phase of the investigation using conventional unit root tests, and the findings are presented in

Table 3. Results of unit root tests.

	Mexico		Indonesia		Nigeria		Turkiye
	ADF
Variables	T-stat.	Lag	T-stat.	Lag	T-stat.	Lag	T-stat.	Lag
$\Delta$LnCCO2	−5.583 *	8	−3.748 ***	8	−3.249 ***	8	−3.995 **	5
$\Delta$LnGDP	−3.454 ***	7	−3.497 **	5	−3.350 **	7	−3.683 ***	8
$\Delta$LnREC	−3.692 **	11	−3.401 ***	8	−4.385 *	9	−3.275 ***	8
$\Delta$LnFD	−6.855 *	4	−3.612 ***	8	−3.224 ***	8	−4.639 *	5
$\Delta$LnEXP	−4.714 *	7	−6.840 *	7	−3.669 **	12	−3.683 **	8
$\Delta$LnIMP	−10.783 *	9	−4.396 *	8	−3.616 **	12	−3.976 **	8
	ERS
$\Delta$LnCCO2	−5.747 *	6	−3.154 **	12	−3.273 ***	8	−3.153 ***	4
$\Delta$LnGDP	−4.452 *	10	−3.396 ***	5	−3.284 **	6	−3.691 *	8
$\Delta$LnREC	−2.858 ***	8	−2.770 ***	12	−3.190 **	8	−2.858 **	7
$\Delta$LnFD	−4.139 *	4	−2.832 ***	8	−2.877 ***	8	−2.919 ***	4
$\Delta$LnEXP	−2.975 ***	5	−6.846 *	7	−3.089 ***	12	−3.691 **	8
$\Delta$LnIMP	−2.926 ***	12	−4.124 *	8	−2.935 **	12	−3.330 **	8

Note: *, ** and *** show the rejection of the null of unit root at 1%, 5% and 10% significance levels, respectively.

. The ERS and ADF unit root test findings uncover that all the indicators are stationary at the first difference (I(1)).

4.2. Cointegration

The current study proceeded by testing the cointegration between CCO₂ emissions and the regressors. In doing so, we used the bounds test with the results presented in

Table 4. Bounds Test Results.

Countries	Models	F-Statistics	Lag Selection	Cointegration
Mexico	LnCCO2 = f(LnGDP, LnIMP, LnEXP, LnREC, LnFD)	8.937 *	1, 2, 1, 1, 0, 0	Yes
Indonesia	LnCCO2 = f(LnGDP, LnIMP, LnEXP, LnREC, LnFD)	5.971 *	1, 1, 1, 0, 2, 1	Yes
Nigeria	LnCCO2 = f(LnGDP, LnIMP, LnEXP, LnREC, LnFD)	5.836 *	1, 0, 0, 2, 2, 1	Yes
Turkey	LnCCO2 = f(LnGDP, LnIMP, LnEXP, LnREC, LnFD)	7.530 *	1, 2, 1, 1, 0, 0	Yes

Note: * depicts a significance level of 1%.

. The outcomes disclose evidence of cointegration among the variables in each country.

4.3. Fourier Test Results

In the second phase of the research, we analyze the Fourier terms’ significance by utilizing the F-test (see

Table 5. Results of F-test.

Models	Mexico	Indonesia	Nigeria	Turkey
Optimum Frequency	0.8	2.2	0.8	2.0
Optimum Lag	6	6	6	6
F-statistics for Fourier expansion	12.968 *	14.858 *	8.474 *	10.968 *
10% CV	10.211	12.032	6.184	7.211
5% CV	10.882	12.772	6.846	8.204
1% CV	11.317	13.460	7.503	8.995

Notes: * signify 1% levels of significance respectively. The optimal Frequency (k*) and optimal lag lengths (p*) were selected based on AIC.

), after the indicators’ order of integration is affirmed. The results uncover that the Ho hypothesis of the absence of Fourier parts, i.e., γ1 = γ2 = 0, is dismissed at a significance level of 5% in the MINT nations.

4.4. Fourier Quantile Causality Results

In the final phase of the investigation, we apply the BFGC-Q test to examine the causal effects of imports, economic growth, exports, financial development, and renewable energy on consumption-based carbon emissions (CCO₂). The outcomes of the BFGC-Q causality test for the MINT nations are shown in

Table 6. Bootstrap Quantile Causality test (Mexico).

H0: LnGDP↛LnCCO2
quantile	Wald stat.	10% CV	5% CV	1% CV
0.1	31.978	63.388	67.092	81.577
0.3	26.653	51.287	52.845	61.023
0.5	25.818	51.568	56.829	67.076
0.7	17.751	64.849	73.141	101.56
0.9	87.062*** (+)	82.583	87.108	117.63
H0: LnREC↛LnCCO2
0.1	9.0119	16.482	18.31736	23.171
0.3	29.125* (−)	13.597	15.20117	21.076
0.5	15.323** (−)	11.576	14.42830	15.608
0.7	4.7447	14.835	16.14676	21.377
0.9	1.9978	27.162	28.40824	34.811
H0: LnEXP↛LnCCO2
0.1	8.3898	28.373	32.80566	55.1
0.3	7.6605	25.010	32.53499	36.648
0.5	19.1422*** (−)	18.639	22.62982	24.484
0.7	28.5644*** (−)	27.881	33.61882	41.403
0.9	11.895	32.187	34.80926	46.525
H0: LnIMP↛LnCCO2
0.1	13.146	20.018	21.44194	26.673
0.3	9.0842 *** (+)	8.7170	10.36571	15.382
0.5	9.8575*** (+)	7.5424	9.569465	10.425
0.7	4.0832	8.2339	9.546957	14.768
0.9	13.324	15.074	17.28618	26.622
H0: LnFD↛LnCCO2
0.1	12.64739	24.178	29.077	38.807
0.3	13.80425*** (+)	12.672	13.942	19.690
0.5	6.280467	9.9456	13.971	16.627
0.7	2.005601	12.055	13.820	17.593
0.9	2.778037	17.914	18.573	23.587

Note: ***, ** and * denote the dismissal of the null of no-causality at and 10%, 5% and 1% significance levels, respectively. CV represent critical values. (+) and (−) illustrate positive and negative effects.

,

Table 7. Bootstrap Quantile Causality test (Indonesia).

H0: LnGDP↛LnCCO2
quantile	Wald stat.	10% CV	5% CV	1% CV
0.1	155.016*** (+)	152.07	169.6371	182.03
0.3	54.3869	115.58	150.2648	163.12
0.5	53.3321	104.68	112.4484	136.03
0.7	64.3851	142.37	159.7852	170.04
0.9	76.1747	222.61	241.0449	316.92
H0: LnREC↛LnCCO2
0.1	46.606** (−)	41.780	46.295	113.77
0.3	29.447*** (−)	28.860	31.215	43.643
0.5	9.5617	22.751	25.283	50.833
0.7	4.6260	26.136	29.177	64.730
0.9	12.265	39.507	51.031	75.474
H0: LnEXP↛LnCCO2
0.1	8.7534	173.89	182.84	218.72
0.3	4.7759	149.94	163.65	183.70
0.5	4.8257	136.53	157.20	167.85
0.7	2.0517	146.86	166.26	205.21
0.9	6.7543	175.68	206.35	253.02
H0: LnIMP↛LnCCO2
0.1	8.5628	177.87	187.94	211.6
0.3	1.9128	74.241	100.72	148.80
0.5	5.5059	37.284	38.902	72.992
0.7	1.9302	32.128	46.606	51.729
0.9	5.085	47.377	58.787	72.820
H0: LnFD↛LnCCO2
0.1	102.967* (+)	56.141	67.663	97.104
0.3	40.66** (+)	25.050	27.437	42.239
0.5	9.0807	14.700	19.357	35.294
0.7	7.4778	19.663	25.182	30.660
0.9	36.756	45.243	50.380	73.003

Note: ***, ** and * denote the dismissal of the null of no-causality at and 10%, 5%, and 1% significance levels, respectively. CV represents critical values. (+) and (−) illustrate positive and negative effects.

,

Table 8. Bootstrap Quantile Causality test (Nigeria).

H0: LnGDP↛LnCCO2
quantile	Wald stat.	10% CV	5% CV	1% CV
0.1	27.220	43.719	49.391	57.726
0.3	45.567** (+)	32.852	37.150	47.866
0.5	32.972*** (+)	31.719	33.170	46.260
0.7	48.859*** (+)	35.334	41.658	51.812
0.9	15.253	42.8749	43.440	50.842
H0: LnREC↛LnCCO2
0.1	19.316*** (−)	18.620	20.738	26.566
0.3	17.158*** (−)	15.859	22.644	24.593
0.5	8.8579	17.022	20.349	33.701
0.7	10.757	25.134	30.30	61.526
0.9	14.192	38.486	45.18	59.555
H0: LnEXP↛LnCCO2
0.1	24.551	31.168	33.907	56.850
0.3	12.032	17.394	21.465	29.676
0.5	5.7322	12.492	13.865	16.700
0.7	4.7990	11.424	14.101	17.905
0.9	29.685*** (−)	24.913	29.192	33.831
H0: LnIMP↛LnCCO2
0.1	2.4349	11.527	13.534	23.828
0.3	3.5846	7.1759	8.6019	9.5594
0.5	3.5608	6.0437	8.5011	11.572
0.7	5.5401	7.2298	9.7265	10.753
0.9	4.3756	18.724	19.768	28.575
H0: LnFD↛LnCCO2
0.1	10.564	24.939	30.925	32.743
0.3	5.6160	15.779	22.195	35.045
0.5	19.540** (+)	11.978	15.302	21.349
0.7	13.008	16.530	20.794	22.502
0.9	19.834	27.332	29.549	45.254

Note: ***, and ** denote the dismissal of the null of no-causality at and 10%, and 5% significance levels, respectively. CV represent critical values. (+) and (−) illustrate positive and negative effects.

and

Table 9. Bootstrap Quantile Causality test (Turkey).

H0: LnGDP↛LnCCO2
quantile	Wald stat.	10% CV	5% CV	1% CV
0.1	36.428	66.243	80.359	86.919
0.3	28.057	49.487	60.666	63.138
0.5	18.002	40.485	46.294	56.863
0.7	44.586*** (+)	44.487	49.658	53.456
0.9	18.721	47.58500	57.543	73.676
H0: LnREC↛LnCCO2
0.1	10.197	39.429	53.859	67.185
0.3	5.2158	25.054	27.767	33.596
0.5	31.358* (−)	19.444	20.862	30.186
0.7	25.154** (−)	20.586	24.542	36.920
0.9	5.9246	38.479	42.183	54.410
H0: LnEXP↛LnCCO2
0.1	5.2356	41.212	46.597	60.857
0.3	72.227** (−)	34.444	36.143	47.222
0.5	30.658*** (−)	30.024	31.161	50.435
0.7	43.076** (−)	32.715	36.800	43.846
0.9	125.51	43.520	46.856	73.936
H0: LnIMP↛LnCCO2
0.1	6.6256	26.208	45.458	93.274
0.3	12.670	17.397	17.999	40.338
0.5	24.698* (+)	16.254	20.574	21.229
0.7	14.644	19.519	25.487	31.232
0.9	22.369	36.157	44.813	80.656
H0: LnFD↛LnCCO2
0.1	19.485	23.438	25.57	34.017
0.3	4.3859	14.253	18.2903	22.908
0.5	10.468	13.323	15.213	18.944
0.7	10.041	16.003	16.681	19.989
0.9	8.4628	27.354	34.427	47.014

Note: ***, ** and * denote the dismissal of the null of no-causality at and 10%, 5% and 1% significance levels, respectively. CV represent critical values. (+) and (−) illustrate positive and negative effects.

.

The outcomes of the BFGC-Q causality test for Mexico are depicted in Table 5. The outcomes reveal unidirectional causality from economic growth to CCO2 emissions in the higher quantiles (0.90). Likewise, in the lower (0.30) and middle (0.50) quantiles, unidirectional causality from renewable energy to CCO₂ emissions surfaced. Furthermore, a unidirectional causality emerged in the middle (0.50) and higher (0.70) quantiles from exports to CCO₂ emissions. Moreover, in the lower (0.30) and middle (0.50) quantiles, unidirectional causality from imports to CCO₂ emissions emerged. In the middle quantile (0.30), financial development Granger cause CCO₂ emissions. These results disclose that the interrelationship between financial development, imports, economic growth, exports, and renewable energy is sensitive to quantiles.

Regarding the sign of the effect, economic growth impacts CCO₂ positively in Mexico. This result is anticipated, given that Mexico is a developing nation, and initiatives towards economic expansion are often favored at the expense of the ecosystem. A similar result in the case of Mexico is documented by [60,61]. Furthermore, we observe the negative effect of clean energy on CCO₂, which is as expected. This shows that renewable energy use in Mexico contributes to a significant reduction in CCO₂ emissions. This outcome is as anticipated given the recent development in Mexico’s renewable energy. According to the Mexican government’s energy growth plan, 328,597.98 GWh of electricity were produced in Mexico in 2021, with 29.5% of that energy coming from renewable sources, including efficient cogeneration, solar photovoltaic, wind, biofuel, geothermal, nuclear power, and hydroelectric. Wu et al. [8,62,63] reported similar results. Moreover, the effect of imports on CCO₂ is positive while the effect of exports on CCO₂ is negative, which corroborates the theoretical framework. The results also disclosed that financial development impacts CCO₂ positively, suggesting that an upsurge in financial development triggers the intensification of CCO₂. The studies [64,65] documented similar findings. Figure 2 portrays the summary of findings for Indonesia.

Table 7 presents the causal/interrelationship between CCO₂ and the regressors in Indonesia. We fail to accept the Ho hypothesis of no causality from GDP to CCO₂ emissions in each quantile. This finding shows that GDP has predictive power over CCO₂ in each quantile. Furthermore, in the lower quantile (0.1–0.30), renewable energy has predictive power over CCO₂ emissions, which is in line with the studies of [2,22,66]. Surprisingly, exports and imports do not have predictive power over CCO₂ in each quantile. These outcomes contradict the studies [31,67]. At the lower tails (0.10–0.30), we find causality from financial development to CCO₂, suggesting that financial development has predictive power over CCO₂.

Regarding the sign of the interrelationship, we found a positive effect of economic growth on CCO₂. Similar to Mexico, Indonesia is a developing nation where priority is given to constant economic expansion while neglecting ecological sustainability. For instance, between 1999 and 2019, Indonesia witnessed a 115% increase in GDP (World Bank, 2022). This growth is accompanied by a 76% increase in CO₂ emissions per capita (World Bank, 2022, https://data.worldbank.org/country/indonesia, assessed on 5 January 2022). Furthermore, the decreasing effect of renewable energy on CCO₂ suggests that the intensification of green energy upsurges ecological quality in Indonesia. The investment in renewable energy in Indonesia is responsible for this favorable impact of renewable energy on ecological quality. For instance, as of April 2021, Indonesia’s energy mix had 13.83% renewable energy, with hydropower accounting for 7.9%, geothermal for 5.6%, and other renewable energy providing 0.33% (https://www.ashurst.com/en/news-and-insights/legal-updates/indonesia-renewable-energy-laws-and-regulations-2022/, assessed on 4 October 2022). Indonesia contributes approximately 12% of the country’s renewable energy. Indonesia can only achieve this goal by shifting energy investment toward renewable resources. Shahbaz et al. [9,68,69] documented similar findings. Similar to Mexico, imports and exports do not significantly influence CCO₂ emissions, which is in line with the studies of [64], who found an insignificant connection between CO₂ and financial development in Malaysia. Figure 3 portrays the summary of findings for Nigeria.

Table 8 presents the causal/interrelationship between CCO₂ and the regressors in Nigeria. In the lower (0.3), middle (0.50), and higher (0.70) quantiles, a unidirectional causality emerged from GDP to CCO₂ emissions, which is similar to the results obtained for Mexico and Indonesia. Furthermore, in the lower (0.1–0.30) tails, renewable energy Granger cause CCO₂ in Nigeria, demonstrating the predictive power of renewable energy over CCO₂. Similar to Mexico, exports have predictive power over CCO₂ emissions in the extreme higher (0.90) quantile.

The sign of the relationship shows that economic growth upsurges CCO₂ which is expected given that Nigeria is an emerging nation. Emerging nations such as Nigeria are pro-growth, which implies that they are pro-growth in their policies. Little or no attention is given to their ecosystem. The studies [1,61] documented similar results. Likewise, the negative effect of green energy is observed, demonstrating that clean energy boosts ecological integrity in Nigeria. The findings of [29,33] comply with this finding. Similarly, exports boost ecological quality as shown by the negative effect of exports on CCO₂ which is in line with the study of [26]. Lastly, financial development contributes to the devastation of the ecosystem, as shown by the positive effect of financial development on CCO₂. The studies [29,34] documented similar findings. Figure 4 portrays the summary of results for Turkey.

Table 9 presents the causal/interrelationship between CCO₂ and the regressors in Turkey. In the higher (0.70) quantile, a unidirectional causality surfaced from GDP to CCO₂ emissions, similar to the results obtained for Mexico, Nigeria, and Indonesia. In the middle (0.5) and higher (0.70) tails, renewable energy has predictive power over CCO₂. Moreover, we observe a unidirectional causality from exports to CCO₂ in the middle (0.50) and higher (0.70) quantiles. Likewise, in the middle (0.50) quantile of imports, Granger causes CCO₂ emissions.

Regarding the effect, economic growth impacts CCO₂ positively, which is anticipated given that Turkey is an emerging nation. Emerging nations such as Turkey need to improve the standard of living of their citizens. As a result, they always prefer increasing their GDP while paying less attention to environmental sustainability. For instance, Turkey witnessed 92% economic growth between 1999 and 2019. This growth is accompanied by a 54% increase in CO₂ emissions per capita [70]. Prior studies [25,63,71] reported similar results. Moreover, the effect of clean energy on CCO₂ is negative, as expected, demonstrating that renewable energy boosts the integrity of the environment in Turkey. Over the previous five years, Turkey’s renewable energy capacity increased by 50%. The year 2019 saw Turkey add the 5th highest amount of new renewable capacity in Europe and the 15th highest globally. Given its abundant resource endowment, Turkey, according to the IEA research, may attain even higher growth in renewables, particularly wind, solar, and geothermal energy. Its robust prospect for expanding renewable energy sources applies to the heating industry and power generation. Importantly, Turkey employs barely 15% of its onshore wind capacity and an estimated 3% of its solar potential. With expenditures reaching about USD 7 billion, Turkey built the highest renewable capacity in a single year in 2020, at around 4800 megawatts (MW) (https://www.iea.org/news/turkey-s-success-in-renewables-is-helping-diversify-its-energy-mix-and-increase-its-energy-security, assessed on 10 September 2022). This outcome aligns with the studies of [72,73] for Turkey; however, the study of [74] contradicts this finding.

As expected, the effect of exports on CCO₂ is negative, demonstrating that intensification in Turkey’s exports boosts ecological quality. A similar result is documented by the studies [1,75]. Furthermore, Turkey’s imports contribute to ecological quality decrease, as shown by the negative sign. The studies [4,71] also reported similar results. Lastly, an insignificant nexus exists between financial development and CCO₂ emissions, which is anticipated given that Turkey’s financial system is in the initial phase. At this phase, financial development is expected not to boost EQ. The research of [64] also documented similar results. Figure 5 presents the summary of results.

5. Conclusions and Policy Ramifications

5.1. Conclusions

The MINT (Mexico, Indonesia, Nigeria, and Turkey) nations are among the top energy consumers and emitters of CO₂ emissions. Notwithstanding the well-known Kyoto Protocol and Paris Accord, the globe’s temperature is rising, and CO₂ emissions are at an all-time high. This has prompted scholars to look into the undiscovered factors that influence CCO₂ emissions. In the literature, energy trade and consumption are well-known major contributors to CO₂ emissions. Nonetheless, renewable energy is among the most effective strategies to reduce CO₂ emissions. Therefore, to promote sustainable development, nations all over the globe are choosing eco-friendly strategies. The study utilizes BFGC-Q for the MINT nations between 1990Q1 and 2019Q4. This approach produces tail-causal and asymmetric causal connections between the indicators within the Fourier approximation, in contrast to the Toda–Yamamoto causality and other conventional Granger tests. The outcomes uncover a unidirectional causality from economic growth and renewable energy to CCO₂ emissions in each MINT nation.

5.2. Policy Ramifications

This paper’s conclusions suggest that domestic consumption levels should be prioritized, particularly in those more energy-intensive sectors contributing to rising carbon emissions, to lessen the impact of imports and economic expansion on CCO₂. Initiatives that do not impede trade and simply focus on reducing carbon emissions should be used to curb emissions-oriented imports. Since transportation and production machinery make up most of these nations’ imports, these nations should prioritize acquiring eco-friendly manufacturing equipment, which would lessen the impact of import emissions and the externality effect brought on by exports via trade. The role of the governmental initiative to completely assimilate it will be realized via international trade and CCO₂ emissions initiatives. Moreover, taxing imported products that produce a lot of emissions would raise funds, tighten ecological rules and reduce import emissions. However, such a policy ramification might not be ideal.

Secondly, using renewable energy drastically reduces CCO₂ emissions in the MINT nation. So, in terms of energy consumption, non-renewable energy or fossil fuel should be reduced, and green energy should be given priority to lower CCO₂ emissions. In this context, additional funding is required to expand the sources of clean energy through supporting wind, hydro, and solar energy, as well as by encouraging and providing incentives for the general public to use energy-efficient appliances and technologies. For this, developing and implementing an appropriate energy policy is necessary. Thirdly, CO₂ emissions rise as the economy expands. Thus, a major factor in reducing CO₂ emissions is the execution of inclusive economic development, growth, and initiatives that do not affect the environment. Moreover, sustainable development will be ensured through green technology implementation, green growth, green urbanization, and green industrialization. Fourth, export quality reduces CCO₂ emissions. Therefore, emphasizing cleaner, more effective, and eco-friendly industrial practices for producing goods promises to reduce CO₂ emissions. In light of this, extensive and broad-based policy initiatives focused on improving export quality will be useful for enhancing ecological integrity in these nations without compromising the intended economic expansion.