Revisiting the impact of energy consumption, foreign direct investment, and geopolitical risk on CO2 emissions: Comparing developed and developing countries

A growing body of literature probes the impact of geopolitical risk (GPR) on CO₂ emissions. However, no study compares the findings in the case of developed and developing countries. Hence, this study aims to probe the impact of GPR on CO₂ emissions for selected developed and developing countries while controlling for energy consumption, foreign direct investment, and economic growth. For this purpose, we make use of a panel dataset covering the period 1990–2020. In the long-run, we report that the Environmental Kuznets Curve hypothesis exists for developing countries. Next, the pollution haven hypothesis is validated for the developed countries in the long-run. Also, GPR escalates emissions for developed and developing countries in the long-run. In the short-run, the Environmental Kuznets Curve and pollution haven hypothesis are found invalid. Moreover, in the short-run, GPR impedes emissions in both developed and developing countries. Further, energy consumption upsurges emissions across all samples (i.e., either developed or developing countries) in either its short- or long-run. The heterogeneous findings across the long- and short-run, for developed and developing countries, propose to formulate unalike policies for countries with different levels of income.

1 Introduction

Environmental degradation has witnessed an unprecedented upsurge, especially after the recent episodes of the industrial revolution. Extreme weather events, a hike in average global temperature, and climate change are the key concerns that are triggered by environmental degradation. Moreover, environmental deterioration affects the decisions of economic agents related to buying and selling of goods and services (Hashmi et al., 2022). Not only this, environmental degradation triggers various human diseases. Thus, it is an unavoidable global issue, which needs to be addressed.

The researchers/scientists claim that carbon dioxide emission (henceforth EMS) is a key component of greenhouse gases, which mainly deteriorate the environment (Mirza et al., 2022). Global emissions have been escalating over the years, especially after the 1950’s. This unparalleled hike in emissions has been a concern for the entire world. To curb emissions, several global efforts have been made such as Paris Agreement and COP26. However, the volume of carbon emissions is yet considerably high, calling for research-based endeavors to discern the factors that affect carbon emissions.

Hence, the literature on environmental economics attempts to seek the impact factors of EMS. The prior literature highlights that economic growth (ECG) enhances the EMS (Saboori et al., 2012). Thereafter, energy consumption (ENC) mainly leads to ECG (Ray, 2021a, Ray, 2021b) at the cost of EMS (Tang and Tan, 2015). Similarly, financial development (Shoaib et al., 2020), Energy efficiency (Liu et al., 2022), unemployment (Bhowmik et al., 2022), trade (Dou et al., 2021), information technology (Wen et al., 2022), and policy-related uncertainties (Syed and Bouri, 2021) affect EMS. One line of research notes that geopolitical risk (GPR) is one of the emerging factors that explain EMS. Anser et al. (2021a) propose two channels, linking GPR with EMS. The first channel argues that GPR may impede ECG and ENC, which ultimately mitigate EMS. The second channel notes that GPR may plunge innovations, technological improvement, and R&D. As a result, EMS witness an escalation.

Regarding the literature on the GPR-environment nexus, Anser et al. (2021b) employ the AMG estimator to conclude that GPR deteriorates environmental quality in emerging countries. Using the ARDL approach, a similar conclusion is reported by Adams et al. (2020) in the selected resource-rich countries. Similarly, Husnain et al. (2022) conclude that GPR decreases EMS in E7 countries. Contrarily, using quantile-based methods, Syed et al. (2022) note that GPR enhances EMS at various quantiles in BRICST countries. Using the bootstrap ARDL method, Hashmi et al. (2022) declare that GPR triggers global emissions in the long-run. It is worth noting point that existing studies on the GPR-emissions nexus report ambivalent findings. On top of this, prior studies overlook some key issues, which may lead to unauthentic outcomes. Firstly, no study attempts to explore the GPR-emissions nexus for developed countries. Next, no study compares the GPR-emissions nexus for developed and developing countries. Hence, filling the aforementioned research gaps is imperative to reach an unambiguous conclusion.

Therefore, the study’s objective is to reinvestigate the impact of GPR on EMS in selected developed and developing countries. For this purpose, we choose eight countries from the top 10 carbon emitters, which are further divided into two sets: 1) developed countries; 2) developing countries. We enhance the existing literature through two empirical innovations. First, this is an earliest attempt to investigate the GPR-emissions nexus in the case of developed countries. Second, we compare the GPR-emissions nexus for developed and developing countries for the first time in the literature.

Regarding the beneficiaries of this study, the present study will help the policymakers to introduce policies to achieve carbon neutrality through managing foreign direct investment, energy demand, geopolitics, and economic growth. Not only this, the current study will assist researchers to understand the dynamic nexus among the considered variables across developed and developing countries. Finally, this study will help to resolve the dilemma of ambivalent findings on the GPR-EMS nexus.

The remainder of this study is reported as follows. Section 2 provides the literature survey while methods are discussed in section 3. We note the empirical findings in section 4 whereas the conclusion and policy implications are presented in section 5.

2 Review of relevant literature

This part of the study provides a brief review of existing literature on the impact factors of EMS. There exists well-established literature that argues that ECG is the most fundamental driver of EMS. To model the link between ECG and EMS, the Environmental Kuznets Curve (EKC) framework has repeatedly been adopted. The EKC points out that there is an inverse U-shaped association between ECG and EMS. For the Central American economies, Apergis and Payne (2009) noted that ENC causes EMS. Moreover, the authors validated the EKC hypothesis. Similar findings are reported by Jalil and Mahmud (2009) for China, wherein the authors validated the EKC hypothesis. Seker et al. (2015) also confirmed the existence of the EKC hypothesis for Turkey. Besides, the study notes that ENC and foreign direct investment (henceforth FDIN) are the key drivers of EMS. For Tunisia, Shahbaz et al. (2014) validate the EKC hypothesis. Moreover, the authors report that ENC and trade cause EMS. Pao and Tsai (2011) noted that the EKC and pollution haven hypothesis exist in the case of BRIC countries. Hitam and Borhan (2012) validated the EKC hypothesis in consort with the pollution halo hypothesis for Malaysia. Yu et al. (2020) noted that the structure and intensity of energy affect energy-related emissions in China. Similar findings are reported by Yu et al. (2021) and Jiang et al. (2022a), Jiang et al. (2022b).

Acharya (2009) noted that FDIN increases EMS in India. Contrarily, Zhang and Zhang (2018) reported that FDIN does not affect EMS in China. Salahuddin et al. (2017) revealed that FDIN exerts a positive impact on EMS in the long-run (henceforth LR). Next, in Pakistan, Khan et al. (2019) noted that FDIN, ENC, and trade enhance EMS. Wang (2019) revealed that trade and ENC enhance EMS. Cheng et al. (2019) noted that FDIN and renewables plunge EMS whereas ECG and exports lead to higher EMS. Saud et al. (2018) concluded that, interestingly, natural resources do not exert any impact on EMS in selected developing countries. Ke et al. (2022) noted that ICT reduces EMS. Moreover, the authors validate the presence of the pollution haven hypothesis in selected developing countries. For the MINT countries, Li et al. (2022) reported that green investment, globalization, and innovations affect EMS. Similarly, Yildirim (2014) probes the causality among FDIN, ENC, and EMS for countries with different levels of income. The study validates the pollution haven hypothesis for certain countries while the pollution halo hypothesis is validated for the remaining countries. In the case of BRI countries, Lu et al. (2021) note that FDIN escalates ECN. We can infer that a rise in ECN will upsurge emissions. Recently, Udeagha and Muchapondwa (2022) conclude that economic uncertainty acts as a moderator among ECN, ECG, and EMS in South Africa. Next, Udeagha and Ngepah (2019), Udeagha and Ngepah (2022), and Udeagha and Ngepah (2021a) noted that trade affects both ECG and EMS across the short- and long-run. Similarly, Udeagha and Ngepah (2021b) investigate whether disaggregated energy has a heterogeneous impact on EMS. The findings highlight that renewables mitigate EMS whereas non-renewables escalate it. Contrarily, Udeagha and Breitenbach (2021c) reported that trade escalates environmental quality by mitigating emissions in the selected African countries.

Concerning the GPR-emissions nexus, Anser et al. (2021a) propose two channels, linking GPR with EMS. The first channel notes that GPR may mitigate ECG and ENC, which ultimately plunge EMS. The second channel argues that GPR may decrease innovations, technological improvement, and R&D. As a result, EMS witnesses an escalation. Anser et al. (2021b) employ the AMG estimator to conclude that GPR deteriorates environmental quality in emerging countries. Using the ARDL approach, a similar conclusion is reported by Adams et al. (2020) in the selected resource-rich countries. Similarly, Husnain et al. (2022) conclude that GPR decreases EMS in E7 countries. Contrarily, using quantile-based methods, Syed et al. (2022) note that GPR enhances EMS at various quantiles in BRICST countries. Using the bootstrap ARDL method, Hashmi et al. (2022) declare that GPR triggers global emissions in the long-run. Recently, Zhao et al. (2021) discern whether GPR affects ENC and EMS for BRICS countries using NARDL modeling. The authors report that there exists an asymmetric impact of GPR on EMS. Adebayo et al. (2019) adopted the quantile-based method to discern the GPR-EMS nexus for India. The study reports that GPR escalates EMS at median quantile whereas it plunges EMS at lower and higher quantiles.

The aforementioned literature on the GPR-EMS nexus contains some short comings. Firstly, no study analyses the impact of GPR on EMS for developed countries. Moreover, there does not exist any study that compares the relationship between GPR and EMS in the case of developed and developing countries. Hence, the current study fills these research gaps.

3 Materials and methods

3.1 Model

To attain its objective (i.e., investigating whether GPR affects EMS in developed and developing countries), the current study adopts the environmental Kuznets curve (EKC) model. The EKC model explains that income and environment have an inverted U-shaped association. Recently, various research outlets adopt the EKC model to investigate the drivers of EMS (see, e.g., Hashmi et al., 2022). Next, ENC is perceived as a core driver of EMS (Syed et al., 2022), therefore, we incorporate ENC as a control variable. Finally, we include GPR as a focused variable, and hence the final model yields:

$\mathbf{E}\mathbf{M}\mathbf{S}=\mathbf{f}\left(\mathbf{E}\mathbf{C}\mathbf{G},\mathbf{E}\mathbf{C}\mathbf{G}2,\mathbf{F}\mathbf{D}\mathbf{I}\mathbf{N},\mathbf{E}\mathbf{N}\mathbf{C},\mathbf{G}\mathbf{P}\mathbf{R}\right)(1)$

In Eq. 1, EMS refers to CO₂ emissions, ECG is per capita GDP, ECG2 denotes the square of per capita GDP, ENC represents energy consumption, FDIN is foreign direct investment, and GPR indicates geopolitical risk. If ECG and ECG2 are >0 and <0, we will validate the existence of the EKC. We expect ENC to be > 0, implying that ENC enhances EMS. Since the relevant literature reports ambivalent findings for the GPR-EMS relationship, we do not envisage the sign for GPR.

3.2 Methodology

Since we make use of panel data with T > N, the four-step methodological procedure is followed. In the first stage, we probe the existence of cross-sectional dependence (hereafter CRD) with three tests. It is a point to note that CRD refers to a situation in which a shock to one country (cross-section) transfers to another. The existence of CRD may provide unreliable inferences, therefore, proper handling of the CRD is imperative (Pesaran, 2015).

We test the order of integration, with the help of the unit root test, in the second step. The study adopts the second-generation unit root test (i.e., CIPS test) due to its capability to cover the CRD (Pesaran, 2007). This ability of the CIPS test makes it superior to other first-generation unit root tests, which do not handle the CRD.

After probing the order of integration, we discern co-integration through the Westerlund (2007) test. Unlike conventional co-integration tests (i.e., first-generation panel data methods), The Westerlund test (i.e., the second-generation test) counters the CRD and hence provides robust outcomes (Anwar et al., 2021). The Westerlund test provides four statistics (i.e., Gt, Ga, Pt, and Pa) with the H₀ of no co-integration.

Finally, we adopt the PMG-ARDL approach of Pesaran and Smith (1995), which is further upgraded by Pesaran et al. (1999). This method is viable in the case of mixed ordering of the dataset. However, no variable should be integrated at I (2) to employ the PMG-ARDL model. Moreover, this model renders homogeneous estimates in the LR whereas it renders heterogeneous estimates in the SR.

3.3 Data

To compare the impact of GPR on EMS for developed and developing countries, we make use of panel data for eight countries that are opted from the top ten emitter countries. These eight countries are further segregated into two panels: 1) the first panel contains four developed countries (i.e., the US, Germany, Japan, and Saudi Arabia); 2) the second panel covers the developing countries (i.e., China, Russia, India, and Indonesia). Moreover, for both panels, the data covers the period 1990–2020. We use CO2 emissions (EMS—measured as a metric ton per capita) as a dependent variable, whereas per capita GDP (ECG–measured in constant $2015), foreign direct investment (FDIN- the percentage of GDP), and energy consumption (ECN–measured in oil equivalent per capita) are adopted as control variables. Next, geopolitical risk (GPR-measured as geopolitical risk index) is the core independent variable. The data on EMS, ECG, FDIN, and ENC are gathered from WDI¹. However, the data on GPR is downloaded from policyuncertainty.com. Following the methodology of Baker et al. (2016), Caldara and Iacoviello (2022) propose the GPR index that is measured through the text mining approach. The entire dataset is transformed into logarithmic formation.

The descriptive statistics are presented in Table 1, containing three samples of the panel dataset: 1) full sample; 2) developed countries’ sample; 3) developing countries’ sample. Note that the mean value is the largest for ECG across all samples. Moreover, the standard deviation is the highest for ECG, FDIN, and GPR in the full sample, developed countries’ sample, and developing countries’ sample, respectively. The entire dataset follows non-normal distribution except for GPR (full sample).

TABLE 1

TABLE 1. Descriptive statistics.

4 Empirical findings

We report the empirical outcomes in this part of the study. First of all, we report the findings from the CRD tests. In the panel dataset, CRD is considered as an imperative issue which needs to be properly addressed to report reliable findings. The CRD is transmission of a shock from one cross-section (i.e., country in our case) to another. If a shock transmits from one country to another, it could affect the economic dynamics of the other country. To control for this transmission, we need to adopt the appropriate econometric methods. We adopt three tests to probe the presence of CRD, namely, the Breusch-Pagan LM test, Pesaran CRD test, and Pesaran scaled LM test. The H₀ of these tests is no CRD whereas the H₁ notes vice versa. Table 2 depicts that there exists CRD in all panels. Hence, we confirm that a shock in one country may transmit to another country.

TABLE 2

TABLE 2. CRD test.

Testing the variables’ ordering is an imperative part of empirical analysis to apply an appropriate method, which helps to avoid spurious outcomes. Therefore, we test the order of integration through the CIPS unit root test. The CIPS unit root test is one of the eminent second-generation tests, having the ability to counter the issue of CRD. Therefore, it outperforms the first-generation unit root tests. The CIPS has H₀ that there exists a unit root, implying that data is not stationary. We report the findings from the CIPS test in Table 3.

TABLE 3

TABLE 3. Unit root test.

It can be concluded that all variables are either integrated at I (0) or I (1) in all samples, allowing for adopting the PMG−ARDL approach.

In the third step, we probe for the co−integrating association among the selected variables. This analysis uses Westerlund (2007) test, which is a second−generation method and hence counters the CRD. The findings from the test are noted in Table 4.

TABLE 4

TABLE 4. Testing co−integration.

We can claim that there exists co−integration among the selected variables since we can reject the H₀ for all samples.

Finally, we employ the PMG−ARDL approach to retrieve the short−run (SR) and long−run (LR) estimates. The findings for the developed countries’ samples are presented in Table 5. In the LR, all coefficients are statistically significant. ECG and ECG2 are negative and positive, respectively. This indicates the validity of N−shaped EKC. In particular, a 1% increase in ECG plunges the EMS by 0.84% while a 1% increase in ECG2 upsurges EMS by 0.02%. These findings are also backed by Gyamfi et al. (2021). ENC is 0.92, showing that a 1% upsurge in ENC escalates EMS by 0.92%. This outcome is also reported by Dogan et al. (2017). Next, FDIN is 0.03%, claiming that a 1% increase in foreign direct investment enhances emissions by 0.03%. Thus, we report the validity of the Pollution haven hypothesis for selected developed countries. These results are somehow in line with the findings of Banerjee and Murshed (2020). Finally, GPR is 0.04, indicating that a 1% increase in geopolitical risk escalates emissions by 0.04%. It might be possible that GPR plunges R&D, green investment, and innovation. As a result, EMS witnessed an upsurge over time. Over findings, related to GPR, are similar to the findings of Hashmi et al. (2022).

TABLE 5

TABLE 5. Findings from PMG−ARDL for developed countries.

Regarding the SR results, all coefficients are statistically significant excluding ECG and ECG2. The insignificancy of ECG and ECG2 indicates that the EKC does not exist in the SR. Moreover, ENC is 0.53, implying that a 1% upsurge in ENC leads to 0.53% emissions. FDIN is −0.00, indicating that although foreign direct investment affects EMS but its impact is not profound. Next, GPR is −0.01, showing that a 1% rise in geopolitical risk impedes emissions by 0.01%. It might be possible that GPR mitigates economic growth and energy utilization which in turn exerts detrimental impacts on emissions. This conclusion is backed by Adams et al. (2020). The error correction term (ECT) is negatively significant and <1, implying that any shock will converge and the equilibrium will be achieved. The Jarque−Bera test and ARCH tests are adopted as diagnostics. The findings from these aforementioned tests report that there is neither the issue of non−normality nor heteroskedasticity.

We report the findings from the PMG−ARDL approach, for the developing countries, in Table 6. In the LR, all coefficients are statistically significant. ECG and ECG2 are positive and negative, respectively. This indicates the validity of the EKC hypothesis. In particular, a 1% increase in ECG increases the EMS by 0.52% while a 1% increase in ECG2 impedes EMS by 0.03%. These findings are also backed by Alola and Ozturk (2021). ENC is 1.12, showing that a 1% upsurge in ENC escalates EMS by 1.12%. This outcome is similar to the conclusion of Anser et al. (2021a). Next, FDIN is insignificant, reporting that foreign direct investment does not affect emissions in the LR. Thus, we report the invalidity of the Pollution haven hypothesis for selected developing countries. These results are somehow in line with the findings of Zhang and Zhou (2016). Finally, GPR is 0.03, indicating that a 1% increase in geopolitical risk escalates emissions by 0.03%. It might be possible that GPR plunges R&D, green investment, and innovation. As a result, EMS witnessed an upsurge over time. Over findings, related to GPR, are similar to the findings of Syed et al. (2022).

TABLE 6

TABLE 6. Findings from PMG−ARDL for developing countries.

Regarding the SR results, all coefficients are statistically significant excluding ECG and ECG2. The insignificancy of ECG and ECG2 indicates that the EKC does not exist in the SR. Moreover, ENC is 0.31, indicating that a 1% surge in ENC leads to 0.31% emissions. FDIN is insignificant, indicating that foreign direct investment does not affect EMS. Next, GPR is −0.02, showing that a 1% rise in geopolitical risk impedes emissions by 0.02%. It might be possible that GPR mitigates economic growth and energy utilization which in turn exerts detrimental impacts on emissions. This conclusion is backed by Adams et al. (2020). The error correction term (ECT) is negatively significant and <1, implying that any shock will converge and the equilibrium will be achieved. We employ the Jarque−Bera test and ARCH test as diagnostics. The findings from these aforementioned tests report that there is neither the issue of non−normality nor heteroskedasticity.

We report the findings from the PMG−ARDL approach, for the full sample, in Table 7. In the LR, all coefficients are statistically significant. ECG and ECG2 are positive and negative, respectively. This indicates the validity of the EKC hypothesis. In particular, a 1% increase in ECG increases the EMS by 0.39% while a 1% increase in ECG2 impedes EMS by 0.02%. These findings are also backed by Farooq et al. (2022). ENC is 1.11, showing that a 1% upsurge in ENC escalates EMS by 1.12%. This outcome is similar to the conclusion of Apergis and Payne (2009). Next, FDIN is 0.02, reporting that a 1% surge in foreign direct investment enhances emissions by 0.02%. Thus, we report the validity of the Pollution haven hypothesis. These results are somehow in line with the findings of Guzel and Okumus (2020). Finally, GPR is 0.02, indicating that a 1% increase in geopolitical risk escalates emissions by 0.02%. It might be possible that GPR plunges R&D, green investment, and innovation. As a result, EMS witnessed an upsurge over time. Over findings, related to GPR, are similar to the findings of Anser et al. (2021a).

TABLE 7

TABLE 7. Findings from PMG−ARDL for the full sample.

Regarding the SR results, all coefficients are statistically significant excluding ECG and ECG2. The insignificancy of ECG and ECG2 indicates that the EKC does not exist in the SR. Moreover, ENC is 0.35, indicating that a 1% surge in ENC leads to 0.35% emissions. FDIN is significant, indicating that foreign direct investment affects EMS but the strength of the relationship is negligible. Next, GPR is −0.01, showing that a 1% rise in geopolitical risk impedes emissions by 0.01%. It might be possible that GPR mitigates economic growth and energy utilization which in turn exerts detrimental impacts on emissions. This conclusion is backed by Anser et al. (2021b). The error correction term (ECT) is negatively significant and <1, implying that any shock will converge and the equilibrium will be achieved. We employ the Jarque−Bera test and ARCH test as diagnostics. The findings from these aforementioned tests report that there is neither the issue of non−normality nor heteroskedasticity.

While comparing the findings for all samples, we conclude the following points in the LR: 1) the EKC exists for the full sample and developing countries’ sample; 2) ENC escalates EMS across all samples and the magnitude of the relationship is the strongest for the developing countries; 3) pollution haven hypothesis is validated for the full sample and developed countries; 4) GPR escalates EMS across all samples, however, its impact on EMS is the strongest for the developed countries, which is followed by developing countries. Next, a comparison of findings in the SR notes that: 1) the EKC does not exist in any sample; 2) ENC upsurges emissions in all samples and its impact on EMS is the strongest for developed countries; 3) FDIN does not exert a profound impact on emissions; 4) GPR impedes emissions in all samples, whereas, its impact is the strongest for the developing countries.

5 Conclusion

It is a well−established argument that environmental degradation has been upsurging with a positive growth rate. The Researchers claim that carbon emissions are the key culprit for this critical issue. To combat carbon emissions, it is imperative to explore its impact factors. Therefore, we probe the impact of GPR, ENC, FDIN, and ECG for selected developed and developing countries.

We find that the entire dataset is integrated either at I (0) or I (1). We also confirm the validity of co−integration among the selected variables. The long−run findings confirm that ENC enhances emissions in both developed and developing countries. Next, the EKC is validated in developing countries while the N−shaped EKC is observed in developed countries. We document the existence of the pollution haven hypothesis for developed countries. Next, GPR upsurges emissions both in developed and developing countries. However, the impact of GPR on emissions is relatively strong for developed countries.

In the short−run, the EKC does not exist for developed and developing countries. We also report that the pollution haven hypothesis does not present across selected developed and developing countries. Also, ENC leads to higher emissions. Finally, GPR mitigates emissions in developed and developing countries. While its impact on emissions is relatively profound for developing countries.

While proposing policy recommendations, we suggest that the share of renewables should be increased in order to offset the detrimental impact of fossil fuel energy. The income level (i.e., whether the country is developed or developing) does not matter in the case of the energy−emissions nexus, therefore, both developed and developing countries need to cut the share of non−renewables in their energy mix across the short− and long−run. Next economic growth should be derived through innovation, technological advancement, institutional reforms, and clean energy in order to avoid the adverse environmental impacts of economic growth. To decrease the reliance on energy led−growth, both developed and developing countries should introduce reforms to upsurge energy efficiency which in turn escalates ECG without deteriorating environmental health. The developed countries should reshape the sectors wherein FDIN is being consumed. For this purpose, developed countries have to incentivize green investment and green sectors. To this end, developed countries may set high returns on investment in green investment and green sectors. Moreover, both developed and developing countries should try to shrink geopolitical risk to mitigate emissions in the long−run. For this purpose, countries should sign peace treaties and agreements. Parallel to this, countries should build strong friendly relations with their neighbors and trade partners. Not only this, the international organizations/institutions (e.g., The United Nations, etc.) should play an active role to impede conflicts within and among countries. Further, to offset the adverse environmental impacts of low geopolitical risk, renewables, R&D, and green investment should be promoted, especially in the short run.

Regarding future research directions, the researchers can investigate the asymmetric impact of GPR on emissions. Moreover, quantile−based methods could be adopted to investigate the impact of GPR on emissions. Also, the interaction of GPR with energy and/or economic growth could be used to probe the impact of energy and growth amidst GPR.

Data availability statement

Publicly available datasets were analyzed in this study. This data can be found here: policyuncertainty.com.

Author contributions

All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Abbreviations

LR, long−run; SR, short−run; EMS, CO₂ emissions; ECG, economic growth (GDP); ENC, energy consumption; FDIN, foreign direct investment; GPR, geopolitical risk; EKC, environment Kuznets curve; PMG−ARDL, pooled men group−autoregressive distributed lags; CRD, cross−sectional dependence.

Footnotes

¹World Development Indicators.

References

Acharya, J. F. D. I. (2009). Growth and the environment: Evidence from India on CO2 emission during the last two decades. J. Econ. Dev. 34, 43–58. doi:10.35866/caujed.2009.34.1.003

CrossRef Full Text | Google Scholar

Adams, S., Adedoyin, F., Olaniran, E., and Bekun, F. V. (2020). Energy consumption, economic policy uncertainty, and carbon emissions; causality evidence from resource−rich economies. Econ. Analysis Policy 68, 179–190. doi:10.1016/j.eap.2020.09.012

CrossRef Full Text | Google Scholar

Adebayo, T. S., Akadiri, S. S., Riti, J. S., and Tony Odu, A. (2019). Interaction among geopolitical risk, trade openness, economic growth, carbon emissions and its implication on climate change in India. Energy & Environ., 0958305X221083236. doi:10.1177/0958305X221083236

CrossRef Full Text | Google Scholar

Alola, A. A., and Ozturk, I. (2021). Mirroring risk to investment within the EKC hypothesis in the United States. J. Environ. Manag. 293, 112890. doi:10.1016/j.jenvman.2021.112890

PubMed Abstract | CrossRef Full Text | Google Scholar

Anser, M. K., Syed, Q. R., and Apergis, N. (2021a). Does geopolitical risk escalate CO2 emissions? Evidence from the BRICS countries. Environ. Sci. Pollut. Res. 28 (35), 48011–48021. doi:10.1007/s11356−021−14032−z

CrossRef Full Text | Google Scholar

Anser, M. K., Syed, Q. R., Lean, H. H., Alola, A. A., and Ahmad, M. (2021b). Do economic policy uncertainty and geopolitical risk lead to environmental degradation? Evidence from emerging economies. Sustainability 13 (11), 5866. doi:10.3390/su13115866

CrossRef Full Text | Google Scholar

Anwar, A., Siddique, M., Dogan, E., and Sharif, A. (2021). The moderating role of renewable and non−renewable energy in environment−income nexus for ASEAN countries: Evidence from Method of Moments Quantile Regression. Renew. Energy 164, 956–967. doi:10.1016/j.renene.2020.09.128

CrossRef Full Text | Google Scholar

Apergis, N., and Payne, J. E. (2009). CO2 emissions, energy usage, and output in Central America. Energy Policy 37, 3282–3286. doi:10.1016/j.enpol.2009.03.048

CrossRef Full Text | Google Scholar

Baker, S. R., Bloom, N., and Davis, S. J. (2016). Measuring economic policy uncertainty. Q. J. Econ. 131 (4), 1593–1636. doi:10.1093/qje/qjw024

CrossRef Full Text | Google Scholar

Banerjee, S., and Murshed, M. (2020). Do emissions implied in net export validate the pollution haven conjecture? Analysis of G7 and BRICS countries. Int. J. Sustain. Econ. 12 (3), 297–319. doi:10.1504/ijse.2020.111539

CrossRef Full Text | Google Scholar

Bhowmik, R., Syed, Q. R., Apergis, N., Alola, A. A., and Gai, Z. (2022). Applying a dynamic ARDL approach to the Environmental Phillips Curve (EPC) hypothesis amid monetary, fiscal, and trade policy uncertainty in the USA. Environ. Sci. Pollut. Res. 29 (10), 14914–14928. doi:10.1007/s11356−021−16716−y

CrossRef Full Text | Google Scholar

Caldara, D., and Iacoviello, M. (2022). Measuring geopolitical risk. Am. Econ. Rev. 112 (4), 1194–1225. doi:10.1257/aer.20191823

CrossRef Full Text | Google Scholar

Cheng, C., Ren, X., Wang, Z., and Yan, C. (2019). Heterogeneous impacts of renewable energy and environmental patents on CO2 emission − evidence from the BRIICS. Sci. Total Environ. 668 (2019), 1328–1338. doi:10.1016/j.scitotenv.2019.02.063

PubMed Abstract | CrossRef Full Text | Google Scholar

Dogan, E., Seker, F., and Bulbul, S. (2017). Investigating the impacts of energy consumption, real GDP, tourism and trade on CO2 emissions by accounting for cross−sectional dependence: A panel study of OECD countries. Curr. Issues Tour. 20 (16), 1701–1719. doi:10.1080/13683500.2015.1119103

CrossRef Full Text | Google Scholar

Dou, Y., Zhao, J., Malik, M. N., and Dong, K. (2021). Assessing the impact of trade openness on CO2 emissions: Evidence from China−Japan−rok FTA countries. J. Environ. Manag. 296, 113241. doi:10.1016/j.jenvman.2021.113241

CrossRef Full Text | Google Scholar

Farooq, S., Ozturk, I., Majeed, M. T., and Akram, R. (2022). Globalization and CO2 emissions in the presence of EKC: A global panel data analysis. Gondwana Res. 106, 367–378. doi:10.1016/j.gr.2022.02.002

CrossRef Full Text | Google Scholar

Guzel, A. E., and Okumus, İ. (2020). Revisiting the pollution haven hypothesis in ASEAN−5 countries: New insights from panel data analysis. Environ. Sci. Pollut. Res. 27 (15), 18157–18167. doi:10.1007/s11356−020−08317−y

PubMed Abstract | CrossRef Full Text | Google Scholar

Gyamfi, B. A., Adedoyin, F. F., Bein, M. A., and Bekun, F. V. (2021). Environmental implications of N−shaped environmental Kuznets curve for E7 countries. Environ. Sci. Pollut. Res. 28 (25), 33072–33082. doi:10.1007/s11356−021−12967−x

CrossRef Full Text | Google Scholar

Hashmi, S. M., Bhowmik, R., Inglesi−Lotz, R., and Syed, Q. R. (2022). Investigating the Environmental Kuznets Curve hypothesis amidst geopolitical risk: Global evidence using bootstrap ARDL approach. Environ. Sci. Pollut. Res. 29 (16), 24049–24062. doi:10.1007/s11356−021−17488−1

CrossRef Full Text | Google Scholar

Hitam, M. B., and Borhan, H. B. (2012). FDI, growth and the environment: Impact on quality of life in Malaysia. Procedia − Soc. Behav. Sci. 50, 333–342. doi:10.1016/j.sbspro.2012.08.038

CrossRef Full Text | Google Scholar

Husnain, I. Ul., Syed, Q. R., Bashir, A., and Khan, M. A. (2022). Do geopolitical risk and energy consumption contribute to environmental degradation? Evidence from E7 countries. Environ. Sci. Pollut. Res. 29 (27), 41640–41652. doi:10.1007/s11356−021−17606−z

CrossRef Full Text | Google Scholar

Jalil, A., and Mahmud, S. F. (2009). Environment Kuznets curve for CO2 emissions: A CO−integration analysis for China. Energy Policy 37, 5167–5172. doi:10.1016/j.enpol.2009.07.044

CrossRef Full Text | Google Scholar

Jiang, T., Li, S., Yu, Y., and Peng, Y. (2022a). Energy−related carbon emissions and structural emissions reduction of China’s construction industry: The perspective of input–output analysis. Environ. Sci. Pollut. Res. 29 (26), 39515–39527. doi:10.1007/s11356−021−17604−1

CrossRef Full Text | Google Scholar

Jiang, T., Yu, Y., Jahanger, A., and Balsalobre−Lorente, D. (2022b). Structural emissions reduction of China's power and heating industry under the goal of" double carbon": A perspective from input−output analysis. Sustain. Prod. Consum. 31, 346–356. doi:10.1016/j.spc.2022.03.003

CrossRef Full Text | Google Scholar

Ke, J., Jahanger, A., Yang, B., Usman, M., and Ren, F. (2022). Digitalization, financial development, trade, and carbon emissions; the implication of pollution haven hypothesis during globalization mode. Front. Environ. Sci. 211. doi:10.3389/fenvs.2022.873880

CrossRef Full Text | Google Scholar

Khan, M. K., Teng, J. Z., and Khan, M. I. (2019). Effect of energy consumption and economic growth on carbon dioxide emissions in Pakistan with dynamic ARDL simulations approach. Environ. Sci. Pollut. Res. 26 (23), 23480–23490. doi:10.1007/s11356−019−05640−x

PubMed Abstract | CrossRef Full Text | Google Scholar

Li, S., Yu, Y., Jahanger, A., Usman, M., and Ning, Y. (2022). The impact of green investment, technological innovation, and globalization on CO2 emissions: Evidence from MINT countries. Front. Environ. Sci. 156. doi:10.3389/fenvs.2022.868704

CrossRef Full Text | Google Scholar

Liu, S., Durani, F., Syed, Q. R., Haseeb, M., Shamim, J., and Li, Z. (2022). Exploring the dynamic relationship between energy efficiency, trade, economic growth, and CO2 emissions: Evidence from novel fourier ARDL approach. Front. Environ. Sci. 10, 945091. doi:10.3389/fenvs.2022.945091

CrossRef Full Text | Google Scholar

Lu, J., Imran, M., Haseeb, A., Saud, S., Wu, M., Siddiqui, F., et al. (2021). Nexus between financial development, FDI, globalization, energy consumption and environment: Evidence from BRI countries. Front. Energy Res. 466. doi:10.3389/fenrg.2021.707590

CrossRef Full Text | Google Scholar

Mirza, F. M., Sinha, A., Khan, J. R., Kalugina, O. A., and Zafar, M. W. (2022). Impact of energy efficiency on CO2 emissions: Empirical evidence from developing countries. Gondwana Res. 106, 64–77. doi:10.1016/j.gr.2021.11.017

CrossRef Full Text | Google Scholar

Pao, H., and Tsai, C. M. (2011). Multivariate Granger causality between CO2 emissions, energy consumption, FDI (foreign direct investment), and GDP (gross domestic product): Evidence from a panel of BRIC (Brazil, Russian Federation, India, and China) countries. Energy 36, 685–693. doi:10.1016/j.energy.2010.09.041

CrossRef Full Text | Google Scholar

Pesaran, M. H., and Smith, R. (1995). Estimating long-run relationships from dynamic heterogeneous panels. J. Econom. 68 (1), 79–113. doi:10.1016/0304-4076(94)01644-F

CrossRef Full Text | Google Scholar

Pesaran, M. H., Shin, Y., and Smith, R. P. (1999). Pooled mean group estimation of dynamic heterogeneous panels. J. Am. Stat. Assoc. 94 (446), 621–634. doi:10.2307/2670182

CrossRef Full Text | Google Scholar

Pesaran, M. H. (2007). A simple panel unit root test in the presence of cross‐section dependence. J. Appl. Econ. Chichester. Engl. 22 (2), 265–312. doi:10.1002/jae.951

CrossRef Full Text | Google Scholar

Pesaran, M. H. (2015). Testing weak cross−sectional dependence in large panels. Econ. Rev. 34 (6−10), 1089–1117. doi:10.1080/07474938.2014.956623

CrossRef Full Text | Google Scholar

Ray, M. S. (2021a). Dry anaerobic digestion technique in biogas and its economic applications: Poverty eradication programme. Globsyn Manag. J. (GMJ) Editor. Board 15. 67−103.

Google Scholar

Ray, S. (2021b). How can we learn from our mistakes during COVID−19? Circular economy in India through biogas economics. Econ. Environ. (2), 59–65. doi:10.36683/2306−1758/2021−2−36/59−65

CrossRef Full Text | Google Scholar

Saboori, B., Sulaiman, J., and Mohd, S. (2012). Economic growth and CO2 emissions in Malaysia: A cointegration analysis of the environmental Kuznets curve. Energy Policy 51, 184–191. doi:10.1016/j.enpol.2012.08.065

CrossRef Full Text | Google Scholar

Salahuddin, M., Alam, K., Ozturk, I., and Sohag, K. (2017). The effects of electricity consumption, economic growth, financial development and foreign direct investment on CO2 emissions in Kuwait. Renew. Sustain. Energy Rev. 81, 2002–2010. doi:10.1016/j.rser.2017.06.009

CrossRef Full Text | Google Scholar

Saud, S., Baloch, M. A., and Lodhi, R. N. (2018). The nexus between energy consumption and financial development: Estimating the role of globalization in next−11 countries. Environ. Sci. Pollut. Res. 25, 18651–18661. doi:10.1007/s11356−018−2069−0

CrossRef Full Text | Google Scholar

Seker, F., Ertugrul, H. M., and Cetin, M. (2015). The impact of foreign direct investment on environmental quality: A bounds testing and causality analysis for Turkey. Renew. Sustain. Energy Rev. 52, 347–356. doi:10.1016/j.rser.2015.07.118

CrossRef Full Text | Google Scholar

Shahbaz, M., Khraief, N., Uddin, G. S., and Ozturk, I. (2014). Environmental Kuznets curve in an open economy: A bounds testing and causality analysis for Tunisia. Renew. Sustain. Energy Rev. 34, 325–336. doi:10.1016/j.rser.2014.03.022

CrossRef Full Text | Google Scholar

Shoaib, H. M., Rafique, M. Z., Nadeem, A. M., and Huang, S. (2020). Impact of financial development on CO2 emissions: A comparative analysis of developing countries (D8) and developed countries (G8). Environ. Sci. Pollut. Res. 27 (11), 12461–12475. doi:10.1007/s11356−019−06680−z

PubMed Abstract | CrossRef Full Text | Google Scholar

Syed, Q. R., Bhowmik, R., Adedoyin, F. F., Alola, A. A., and Khalid, N. (2022). Do economic policy uncertainty and geopolitical risk surge CO2 emissions? New insights from panel quantile regression approach. Environ. Sci. Pollut. Res. 29 (19), 27845–27861. doi:10.1007/s11356−021−17707−9

CrossRef Full Text | Google Scholar

Syed, Q. R., and Bouri, E. (2021). Impact of economic policy uncertainty on CO2 emissions in the US: Evidence from bootstrap ARDL approach. J. Public Aff., e2595. doi:10.1002/pa.2595

CrossRef Full Text | Google Scholar

Tang, C. F., and Tan, B. W. (2015). The impact of energy consumption, income and foreign direct investment on carbon dioxide emissions in Vietnam. Energy 79, 447–454. doi:10.1016/j.energy.2014.11.033

CrossRef Full Text | Google Scholar

Udeagha, M. C., and Breitenbach, M. C. (2021c). Estimating the trade−environmental quality relationship in SADC with a dynamic heterogeneous panel model. Afr. Rev. Econ. Finance 13 (1), 113–165.

Google Scholar

Udeagha, M. C., and Muchapondwa, E. (2022). Investigating the moderating role of economic policy uncertainty in environmental Kuznets curve for South Africa: Evidence from the novel dynamic ARDL simulations approach. Environ. Sci. Pollut. Res. Int., 1–39. doi:10.1007/s11356−022−21107−y

CrossRef Full Text | Google Scholar

Udeagha, M. C., and Ngepah, N. (2021b). Disaggregating the environmental effects of renewable and non−renewable energy consumption in South Africa: Fresh evidence from the novel dynamic ARDL simulations approach. Econ. Change Restruct. 55, 1767–1814. doi:10.1007/s10644−021−09368−y

CrossRef Full Text | Google Scholar

Udeagha, M. C., and Ngepah, N. (2022). Does trade openness mitigate the environmental degradation in South Africa? Environ. Sci. Pollut. Res. 29 (13), 19352–19377. doi:10.1007/s11356−021−17193−z

CrossRef Full Text | Google Scholar

Udeagha, M. C., and Ngepah, N. (2019). Revisiting trade and environment nexus in South Africa: Fresh evidence from new measure. Environ. Sci. Pollut. Res. 26 (28), 29283–29306. doi:10.1007/s11356−019−05944−y

PubMed Abstract | CrossRef Full Text | Google Scholar

Udeagha, M. C., and Ngepah, N. (2021a). The asymmetric effect of trade openness on economic growth in South Africa: A nonlinear ARDL approach. Econ. Change Restruct. 54 (2), 491–540. doi:10.1007/s10644−020−09285−6

CrossRef Full Text | Google Scholar

Wang, Z. (2019). Does biomass energy consumption help to control environmental pollution? Evidence from BRICS countries. Sci. total Environ. 670, 1075–1083. doi:10.1016/j.scitotenv.2019.03.268

PubMed Abstract | CrossRef Full Text | Google Scholar

Wen, Y., Shabbir, M. S., Haseeb, M., Kamal, M., Anwar, A., Khan, M. F., et al. (2022). The dynamic effect of information and communication technology and renewable energy on CO2 emission: Fresh evidence from panel quantile regression. Front. Environ. Sci. 1123.

CrossRef Full Text | Google Scholar

Westerlund, J. (2007). Testing for error correction in panel data. Oxf. Bull. Econ. Stat. 69 (6), 709–748. doi:10.1111/j.1468−0084.2007.00477.x

CrossRef Full Text | Google Scholar

Yildirim, E. (2014). Energy use, CO2 emission and foreign direct investment: Is there any inconsistence between causal relations? Front. Energy 8 (3), 269–278. doi:10.1007/s11708−014−0326−6

CrossRef Full Text | Google Scholar

Yu, Y., Jiang, T., Li, S., Li, X., and Gao, D. (2020). Energy−related CO2 emissions and structural emissions’ reduction in China’s agriculture: An input–output perspective. J. Clean. Prod. 276, 124169. doi:10.1016/j.jclepro.2020.124169

CrossRef Full Text | Google Scholar

Yu, Y., Li, S., Sun, H., and Taghizadeh−Hesary, F. (2021). Energy carbon emission reduction of China’s transportation sector: An input–output approach. Econ. Analysis Policy 69, 378–393. doi:10.1016/j.eap.2020.12.014

CrossRef Full Text | Google Scholar

Zhang, C., and Zhou, X. (2016). Does foreign direct investment lead to lower CO₂ emissions? Evidence from a regional analysis in China. Renew. Sustain. Energy Rev. 58, 943–951. doi:10.1016/j.rser.2015.12.226

CrossRef Full Text | Google Scholar

Zhang, Y., and Zhang, S. (2018). The impacts of GDP, trade structure, exchange rate and FDI inflows on China's carbon emissions. Energy Policy 120, 347–353. doi:10.1016/j.enpol.2018.05.056

CrossRef Full Text | Google Scholar

Zhao, W., Zhong, R., Sohail, S., Majeed, M. T., and Ullah, S. (2021). Geopolitical risks, energy consumption, and CO2 emissions in BRICS: An asymmetric analysis. Environ. Sci. Pollut. Res. 28 (29), 39668–39679. doi:10.1007/s11356−021−13505−5

CrossRef Full Text | Google Scholar