Environmental quality and pollution concerns have become a subject of discussion for economist’s ecologists, policymakers, and researchers over the past two decades. Human demand for natural resources puts strain on the environment, resulting in a slew of environmental concerns such as climate change, soil degradation, pollution, and biodiversity loss (Adebayo and Kirikkaleli, 2021; Soylu et al., 2021). Humanity’s unrestricted exploitation of natural resources produces irreparable harm to the biosphere, which has a detrimental impact on the globe’s sustainable social development and economic goals (Shahbaz et al., 2021). Excessive exploitation and use of natural resources, as well as growing pollution and waste emissions, pose a danger to national economies. Carbon emissions (CO₂) are now causing serious environmental issues such as global warming, climate change, and biodiversity loss. As a result, nations have taken steps to reduce CO₂ at international gatherings, including the Stockholm Conference, the Montreal and Kyoto Protocols, and the Paris Agreement. Notwithstanding all attempts, CO₂ keeps rising globally. The level of economic growth is a critical component influencing environmental deterioration. The environmental and ecological cost of economic progress, in particular, is a source of worry (Ozturk and Acaravci, 2016; Adebola Solarin et al., 2017; Rjoub et al., 2021). The study of Grossman and Krueger (1991) was the first to examine the inverted U-shaped connection between numerous environmental degradation indices and economic development. This connection demonstrates that as the degree of development increases, degradation of the environment rises initially, and then when a specific limit is reached, economic growth lowers the environmental deterioration.

Furthermore, several academics have recently added the economic complexity index as a measure of economic progress in their study (Ahmad et al., 2021; Pata, 2021). Dreher (2006) constructing the economic complexity index (ECI) to calculate a nation’s technology-intensive exports. ECI is an indication of a country’s economic progress in terms of export since this data only covers exported items. ECI is described in the international trade literature as the technological level and understanding of the production process (Abbasi et al., 2021; Pata, 2021). In other words, a nation’s productive output necessitates a high degree of ingrained skills and knowledge (Hidalgo and Hausmann, 2009). ECI varies according to the diversity and sophistication of each nation’s exports (Doğan et al., 2019). On the one hand, as the economy’s complexity grows, so does product diversity, and more output contributes to higher emissions. Moreover, ECI can have a beneficial impact on the quality of the environment since it involves research and development activities as well as the ability to promote eco-friendly goods and clean technology (Neagu, 2020). As Hidalgo and Hausmann (2009) stated, ECI remains at the center of the rationale for the disparity in per capita income between nations. As a result, ECI is closely related to a nation’s per capita income and wellbeing (Abbasi et al., 2021).

Energy is also critical to economic progress and environmental degradation. Economic activities that consume a lot of energy result in increased CO₂ emissions (Adebola Solarin et al., 2017). Excessive usage of fossil fuels in the industrial production process raises CO₂ and inhibits sustainable growth by creating climate change and ecological problems. With the population of the globe growing, the continued usage of fossil fuel sources including oil, coal, and gas resulted in military and geopolitical conflicts, an increase in environmental concerns, and oil price instability (Orhan et al., 2021). Renewable energy sources, as opposed to fossil fuels, are clean, limitless, and ecologically beneficial. Since the massive growth in fossil fuel consumption has resulted in catastrophes and severe ecological harm, renewable energy should be utilized in place of fossil fuels to improve the sustainability of the environment while also providing energy diversity and security (Sarkodie and Strezov, 2019).

Globalization is another element that contributes to environmental deterioration. Globalization influences human demand on the environment by facilitating economic, technical, and cultural progress. (Dreher, 2006) (KOF) index can be used to gauge globalization. This indicator is divided into three complementary components: political, social, and economic. Globalization, which has gained traction during the 1990s, has resulted in significant economic developments such as capital flows and international trade liberalization. Globalization boosts total factor productivity, opens up new investment options, and strengthens financial markets (Kirikkaleli et al., 2021). To sum up, globalization can affect environmental pollution positively or negatively. Most economists believe that globalization has a positive net impact on the degradation of the environment (Dreher, 2006). Since globalization does not give an exact impact (positive or negative), its influence on the degradation of the environment should be investigated (Pata, 2021).

Over the years, several studies have been done to inform the public on the influence of economic complexity, economic growth, globalization, and renewable energy consumption on CO₂ emissions; however, findings mixed. For instance, some studies (Dogan and Seker, 2016; Pata, 2021; Rafique et al., 2021) found CO₂-ECI positive connections while some studies (Can and Gozgor, 2016; Doğan et al., 2019; Boleti et al., 2021) established negative CO₂-ECI association. Moreover, the studies of Koengkan et al. (2020), Pata (2021), and Khan et al. (2019) disclosed CO₂-GLO positive interrelation whereas the studies of He et al. (2021) and Koengkan et al. (2020) confirmed negative CO₂-GLO association. Additionally, the majority of the studies found that economic expansion triggers emissions of CO₂ (Adebayo, 2020; Coelho et al., 2021; Kirikkaleli and Adebayo, 2021).

In this regard, our study attempts to overcome the gap in the literature by focusing on the top seven economic complexity nations (Japan, Switzerland, South Korea, Germany, Singapore, Australia and Czech) (see Figure 1) ¹ which are ranked as developed nations. This research adds to the current literature in the following ways: First, we assess not only the influence of economic complexity but also the influence of REC, economic growth, and globalization on environmental deterioration. Second, this research introduces the interaction term to assess the influence of renewable energy and economic complexity on CO₂ emissions to capture whether globalization among these nations has any implications on renewable energy utilization and economic complexity and, as a result, for CO₂. To the authors’ understanding, this is the first study to explore these associations using the interaction term. Therefore, the present research fills the gap in the ongoing literature. Third, the current study adds by employing a unique CS-ARDL to address the concerns of heterogeneity and CSD in panel data, which have been overlooked in earlier studies. This approach is resistant to misspecification bias endogeneity cross-sectional dependence, nonstationarity, and heterogeneity (Lin et al., 2021). Thirdly, we applied the CCEMG approach as a robustness check.

The remaining sections of this research are compiled as follows: Literature Review present the data and methods. Findings and Discussion illustrates the findings and discussion and the conclusion is presented in Conclusion and Policy Path.

This section of the study presents a summary of the studies conducted regarding the connection between carbon emissions (CO₂) and economic growth (GDP), globalization (GLO), economic complexity (ECI), and renewable energy utilization (REC). Regarding the interrelationship between CO₂ and economic growth, several studies have been conducted; however, mixed findings are reported. For instance, Using G7 countries and data from 1970 to 2015, (Cai et al., 2018) assessed the GDP-CO₂ interrelationship. The investigators applied Granger Causality based on bootstrap ARDL and the study outcomes unveiled that GDP dampens the quality of the environment. Furthermore, there is proof of unidirectional causal linkage from GDP to CO₂. Similarly, Wang and Ye (2017) using the BRICS Nations assessed the GDP-CO₂ interrelationship using PLS and the study outcome disclosed positive GDP-CO₂ interrelationship. Moreover, Ahmad et al. (2021) study on the GDP-CO₂ association in 30 Chinese provinces and cities using the panel method disclosed that an upsurge in GDP mitigates CO₂. Also, the EKC hypothesis is validated. Moreover, the study of Leitão (2021a) in Portugal using FMOLS disclosed that an increase in GDP contributes to the degradation of the environment. Using the ARDL approach, Khan et al. (2019) assessed the connection between GDP and CO₂ from 1965 to 2015 and their outcome disclosed a positive GDP-CO₂ interrelationship. Using Brazil as a case study, Hdom and Fuinhas (2020) investigated the EKC hypothesis using data from 1975 to 2016 and DOLS, FMOLS, and Causality approaches. The study affirmed the EKC hypothesis also, there is unidirectional causality from GDP to CO₂. Furthermore, the study of Salari et al. (2021) in the United States from 1997 to 2016 disclosed positive CO₂-GDP interconnection.

Globalization boosts total factor productivity (TFP) by increasing trade. Foreign direct investment (FDI) and the transfer of sophisticated technologies between industrialized and developing nations stimulate economic growth. Furthermore, globalization creates investment possibilities via FDI and strengthens financial markets via financial deregulation. Certainly, this process improves trade, economic growth, and financial markets as well as the consumption of energy which increases CO₂ emissions. On the association between globalization and CO₂ emissions, several studies have been done with mixed findings. For instance, Muhammad and Khan (2021) research on the globalization-emissions association using 31 developed and 155 developing countries disclosed that in both developed and developing nations an upsurge in globalization triggers emissions levels. Moreover, the study of Leal and Marques (2021) using 23 African countries from 1999 to 2017 disclosed a positive globalization-emissions association. On the contrary, the study of He et al. (2021) in Mexico established a negative globalization-emissions association interrelationship. The negative globalization-emissions association is also supported by the study of Koengkan et al. (2020) for Latin America and Caribbean Countries between 1975 and 2016. Using South Asian countries, the study of Khan et al. (2019) confirmed a negative GLO-CO₂ interrelationship.

Renewable energy sources, as opposed to fossil fuels, are clean, limitless, and ecologically beneficial. Since the massive growth in fossil fuel consumption has resulted in catastrophes and severe ecological harm, renewable energy should be utilized in place of fossil fuels to improve the sustainability of the environment while also providing energy diversity and security (Sarkodie and Strezov, 2019). On this note, several studies have been done on renewable energy use and CO₂ emissions association. For example, using 25 selected African countries, the study of Zoundi (2017) on the REC-CO₂ interrelationship and Panel FMOLS from 1980 to 2012 disclosed negative CO₂-REC association. Moreover, using Portugal, Italy, Greece, and Spain as a case study, the study of Balsalobre-Lorente et al. (2021) established that renewable energy usage helps to abate CO₂ emissions. Likewise, using European nations as a case study, the study of Leitão and Lorente (2020) established that renewable energy utilization plays a significant role in decreasing emissions of CO₂. In India, Kirikkaleli and Adebayo (2021) assessed the CO₂-REC connection using frequency domain causality test and their outcome disclose that an upsurge REC contributes to the degradation of the environment. Similarly, the study of Leitão (2021b) on the REC-GDP connection using FMOLS disclosed that an upsurge in REC mitigates CO₂ emissions. Furthermore, REC can predict CO₂. Moreover, the study of Cherni and Essaber Jouini (2017) on the REC-CO₂ interrelationship unveiled that an upsurge in REC mitigates CO₂ pollution. This outcome is also validated by the studies of Adams and Nsiah (2019) for 28 Sub-Sahara African countries and Haseeb et al. (2018) for BRICS countries.

Furthermore, several academics have recently added the economic complexity index as a measure of economic progress in their study (Ahmad et al., 2021; Pata, 2021). ECI is an indication of a country’s economic progress in terms of export since this data only covers exported items. ECI is described in the international trade literature as the technological level and understanding of the production process (Abbasi et al., 2021; Pata, 2021). In other words, a nation’s productive output necessitates a high degree of ingrained skills and knowledge (Hidalgo and Hausmann, 2009). ECI varies according to the diversity and sophistication of each nation’s exports (Doğan et al., 2019). Several studies have been done on renewable energy use and CO₂ emissions association. For example, using 55 countries and data from 1971 to 2014, the study of Dogan and Seker (2016) established positive ECI-CO₂ interrelationship. Similarly, the study of Neagu (2020) on the ECI-CO₂ interrelationship using European Union countries from 1990 to 2018, and DOLS, FMOLS, and Panel Causality approaches disclosed that an upsurge in ECI triggers CO₂ emissions. Contrarily, the study of Doğan et al. (2019) established negative ECI-CO₂ interrelationship in 28 OECD countries from 1990 to 2014 using Panel ARDL. Likewise, the study of Boleti et al. (2021) using 88 developed and developing countries an data from 2002 to 2012 revealed negative ECI-CO₂ interrelationship. Table 1 presents the summary of discussed studies.

The empirical analyses commenced by taking a look at the descriptive statistics of CO₂, ECI, GLO, REC, and GDP, which is reported in Table 2. GDP (39914.13) has the highest mean value, followed by GLO (79.52386), CO₂ (9.628900), REC (4.936359), and ECI (1.772849). ECI (0.520993) has a more consistent score which is followed by CO₂ (2.381198), REC (5.704039), GLO (8.262817), and GDP (17947.880) as revealed by the standard deviation. Furthermore, the skewness value disclosed that ECI and GLO are negatively skewed while CO₂, GDP, and REC are skewed positively. The kurtosis value uncovered that all the series (GDP, REC, and GLO) are platykurtic (less than 3) whilst ECI and CO₂ are leptokurtic (greater than 3).

The study proceeds by examining the cross-section dependence (CSD) and Slope heterogeneity which are reported in Tables 3, 4 respectively. The CSD results reported in Table 3 unveiled the issue of CSD as revealed by Breusch-Pagan LM, Pesaran scaled LM, Bias-corrected scaled LM, and Pesaran CD tests. Therefore, we fail to accept the null hypothesis. The CSD’s significance originates from the reality that advanced economies are interconnected in today’s globalized world. This indicates that any disturbance in the underlying variables of a nation might extend to other economies. As a result of the spillover effects, the variables are cross-sectionally dependent. Table 4 shows that the panel of the top seven economic complexity countries exhibit varying levels of technological advancement and growth. As a consequence, the findings suggest the occurrence of variation in slope coefficients. The study further applied the CIPS unit root test which is a second-generation test to detect the stationarity features of series and the outcomes are reported in Table 5. The outcomes of the test revealed that all the series are stationary at first difference.

To identify the cointegration between CO₂ and the regressors, we applied the Westerlund cointegration test which is reported in Table 6. The outcomes from the test unveiled that in the long-run, the series are cointegrated. Therefore, the null hypothesis of “no cointegration” is refuted in this study. Therefore, the outcomes uncovered that there is longrun association between CO₂ and REC, GDP, ECI, and GLO.

After the long-run association between CO₂ and ECI, GDP, REC, and GLO has been confirmed, we continued by assessing the long-run and short-run influence of ECI, GDP, REC and GLO via the CS-ARDL. The outcomes of the CS-ARDL are presented in Table 7 and the outcomes disclosed the followings:

The influence of GDP on CO₂ emissions is positive and significant which implies that a 1% upsurge in GDP will trigger CO₂ by 0.3012% in Model-1, 0.5196% in Model-2, and 0.3286 in Model-3 keeping other indicators constant. This implies that an upsurge in growth contributes to emissions in these nations. Similar results are also observed in the short-run. The findings show that in these nations, the scale impact outweighed the composition and technique effects, implying that economic expansion is driving environmental degradation by consuming more energy and producing more pollutants. This further indicates that these countries prioritize economic growth over environmental degradation. As a result, the environmental quality of these nations has degraded in the course of obtaining higher economic expansion. This outcome is in line with the study of Usman et al. (2020) in the United States who established negative and significant CO₂-GDP connections. Furthermore, the study of Abbasi et al. (2021) corroborates this finding. However, this outcome is not consistent with the studies of Coelho et al. (2021) for South Korea, Akinsola et al. (2021) for Indonesia, Zhang et al. (2021) for Malaysia and Su et al. (2021) for Brazil and Yuping et al. 2021) for Argentina.

Moreover, the CO₂-GDP² interrelationship is negative which infers that a 1% upsurge in GDP² mitigates CO₂ by 0.5604% (model-1), 0.7580% (model-2), and 0.5211% (model-3) when other factors are kept constant. The negative CO₂-GDP² interrelationship was also confirmed by the short-run coefficients. As a result, in these countries, there is an inverted U-shaped link between economic expansion and environmental degradation. It demonstrates that after achieving a certain level of income, ecological problems can be mitigated with improved environmental law, technical advancements, sustainable manufacturing, and consumption habits. It also shows that these nation’s present policies are on the correct track, as their economies steadily transition away from polluting sectors and technology and toward green technologies and low-carbon clean industries. The inverted U-shaped growth-CO₂ emissions nexus is verified in the long term by the negative and statistically significant CO₂-GDP² connection. As a result, economic expansion can be considered to damage the environment at first before benefiting it afterward. This result is consistent with the work of Lin and Zhu (2019) for Chinese province analysis and (Kihombo et al., 2021) WAME nations.

Furthermore, the outcomes from the Table disclosed that in the three models, the influence of economic complexity (ECI) on CO₂ is positive and significant. This outcome demonstrates that an upsurge in ECI contributes to the degradation of the ecosystem. Therefore, ECI does not play a vital role in mitigating emissions in the selected countries. The possible explanation for this association is ascribed to the fact that product complexity and structural changes (production activities) are detrimental to the environment. More precisely, the study finds that diversifying export items increases CO₂ emissions. The study outcome aligns with the studies of Abbasi et al. (2021) and Ahmad et al. (2021) who established that ECI harms the quality of the environment Nevertheless, this outcome refutes the finding of and Neagu (2020) (1) who found ECI-CO₂ negative interrelation.

Moreover, in the three models, the renewable energy usage (REN) influence on CO₂ is negative and significant which implies that REN can play a vital role in combating the degradation of the environment. This demonstrates that cleaner and greener energy sources lower emission levels in the atmosphere. These findings corroborate the theoretical expectation that renewable energy is environmentally friendly. The findings show that renewable energy is an effective instrument for achieving environmental and economic sustainability by reducing the negative consequences of human activities, such as land usage and water and commodities utilization. This outcome is anticipated and it concurs with the works of Apergis and Payne (2014) for sub-Saharan Africa nations, Adebayo and Kirikkaleli (2021) for Japan, and Tufail et al. (2021) for highly decentralized economies.

There effect of globalization on CO₂ is negative which implies that 1% upsurge in GLO caused CO₂ to decrease by 0.0278 (Model-1), 0.011 (Model-2), and 0.0223 (Model-3) respectively, keeping other factors constant. This infers that GLO helps in abating degradation of the environment in the top 7 economic complexity nations. Given that the globalization index and CO₂ emissions levels have both increased over the years, this research contradicts the premise that globalization causes higher CO₂ emissions. This research outcome complies with the study of Yuping et al. (2021) for Argentina who established that globalization helps in mitigating emissions levels in Argentina. The study of He et al. (2021) for Mexico between 1990 and 2018 also validates this outcome by establishing a negative connection between globalization and CO₂. Nonetheless, this result is not consistent with the findings of Kirikkaleli et al. (2021) who discovered that the globalization process produces a significant increase in CO₂ emissions due to the widespread use of energy in production and consumption activities in advanced and developing economies.

We examine the combined effects of globalization and renewable energy use to better understand the explanation for such a puzzling outcome. In the context of Model-2, it is observed that renewable energy consumption and globalization jointly mitigate the emissions level in the long run and short-run. Moreover, in model 3, globalization and economic complexity are found to jointly mitigate emissions of CO₂ which implies that globalization plays a vital role in mitigating CO₂ emissions.

The present study applied the CCEMG long-run estimator to check the consistency of the panel quantile regression outcomes in the top seven economic complexity countries (Japan, Germany, South Korea, Singapore, Czech, Austria, and Switzerland). The CS-ARDL estimator has been chastised for imposing a homogeneity constraint on long-run parameters when countries differ in socioeconomic structure and size. therefore, we utilized the Common Correlated Effect Mean Group (CCEMG) initiated (Pesaran, 2006) which allows the parameters to be heterogeneous in the long run as a robustness check. Table 8 presents the outcomes of the CCEMG. The CCEMG findings reinforce the reliability of the CS-ARDL outcomes since these other techniques provided outcomes that were identical with the CS-ARDL results.

The study assesses the causal interrelation between CO₂ and GDP, ECI, GLO, and REC by applying the Dumitrescu and Hurlin panel causality test. The outcomes of the causality test are presented in Table 9 and the outcomes unveiled unidirectional causality from economic complexity to CO₂. Furthermore, there is a feedback causal linkage between CO₂ and GDP which suggests that GDP can predict CO₂ and vice-versa. Moreover, one-way causality was observed from GLO to CO₂ emissions. Lastly, there is proof of unidirectional causal interrelation from REC to CO₂ which illustrates that REC can predict CO₂ in the top seven economic complexity nations.