Sustainable Energy Technologies and Assessments 53 (2022) 102597 Contents lists available at ScienceDirect Sustainable Energy Technologies and Assessments journal homepage: www.elsevier.com/locate/seta Original article Environmental Kuznets Curve hypothesis from lens of economic complexity index for BRICS: Evidence from second generation panel analysis Divine Q. Agozie a, Bright Akwasi Gyamfi b, Festus Victor Bekun c,d,e,*, Ilhan Ozturk f,h,*, Amjad Taha g a University of Ghana, Business School, Dept. of Operations and Management Information Systems, Ghana b Faculty of Economics and Administrative Sciences, Cyprus International University Nicosia, North Cyprus, Via Mersin 10, Turkey c Faculty of Economics Administrative and Social Sciences, Istanbul Gelisim University, Istanbul, Turkey d Adnan Kassar School of Business, Department of Economics, Lebanese American University, Beirut, Lebanon e Faculty of Economics and Commerce, The Superior University Lahore, Pakistan f Faculty of Economics, Administrative and Social Sciences, Nisantasi University, Istanbul, Turkey g Banking and Finance Department, Eastern Mediterranean University, 99628, Famagusta, North Cyprus Via Mersin 10, Turkey h Department of Medical Research, China Medical University Hospital, China Medical University, Taichung, Taiwan A R T I C L E I N F O A B S T R A C T Keywords: The present study contributes to the ongoing discussion on environmental sustainability, energy efficiency for Economic complexity the case of Brazil, Russia, India, China, and South Africa economies by investigating the dynamic connection SDGs regarding foreign direct investment, economic complexity index, renewable energy, natural resources, urbani- Carbon-reduction zation, and CO2 emission for annual frequency data from 1990 to 2019. The present study employs robust Alternative energy Emerging economies econometric techniques including Augmented Mean Group with Fully Modified-Ordinary Least Squares esti- mators as estimation techniques. Empirical outcome shows both inverted U-Shaped and N-Shaped EKC rela- tionship between ECI and CO2 emission. The empirical findings also lend support to the Pollution Haven Hypothesis, which suggest that foreign direct investment influx is a contributing factor to environmental degradation in Brazil, Russia, India, China, and South Africa economies. Furthermore, renewable energy and the interaction between economic complexity index and urbanization is found to have adverse impact on emission while natural resources and urbanization have positive impact on the environment. Finally, the results from the Dumitrecu and Hurlin causality reveals a bi-directional causality between economic complexity and CO2 emis- sion. Similar causality is found between economic complexity index and urbanization and CO2 emission while a one-way causality is seen running from foreign direct investment to CO2 emission over study period. These findings encourage authorities of the investigated countries to offer a broader energy strategy on alternative energies i.e., renewables improve Brazil, Russia, India, China, and South Africa environmental quality. Furthermore, emphasis on economic strategies that foster a healthier manufacturing activity to engender envi- ronmental sustainability without compromise for economic prosperity should be pursued among the examined economies. altered pollution levels compared with economic expansions in many developed economies such as G-7, United states, OECD among others Introduction [61,42]. Owing to fast developments of urban populations and indus- trialization, many economies both developed and emerging have seen Many policy inroads and structural changes have driven the shift much economic growth in the last decade or so [33]. Particularly among from carbon-intensive energy sources toward more efficient energy BRICS nations (Brazil, Russia, India, China, and South Africa), economic sources thus, contributing meaningful reductions in CO2 emissions in expansion and transformations emanating from industrialization are many countries. In lieu of this, advancements in technology, enforced revealed in terms of their GDP (Danish & Wang, 2019). These BRICS regulations for the environment and environmental sustainability have * Corresponding authors. E-mail addresses: dagozie@ug.edu.gh (D.Q. Agozie), brightgyamfi1987@gmail.com (B. Akwasi Gyamfi), fbekun@gelisim.edu.tr (F. Victor Bekun), ilhan.ozturk@ nisantasi.edu.tr (I. Ozturk), amjad.taha@emu.edu.tr (A. Taha). https://doi.org/10.1016/j.seta.2022.102597 Received 16 May 2022; Received in revised form 16 July 2022; Accepted 29 July 2022 Available online 5 August 2022 2213-1388/© 2022 Elsevier Ltd. All rights reserved. D.Q. Agozie et al. S u s t a i n a b l e E n e r g y T e c h n o l o g i e s a n d A s s e s s m e n ts 53 (2022) 102597 whereas an economy increases its complexity, there is greater diversi- Nomenclature fication of production and output expansion, but expanded output levels consequently drive pollution, negative climate consequences and global Abbreviations warming. Despite this evidence, a strand of literature also finds eco- BRICS Brazil, Russia, India, China, and South Africa countries nomic complexity to have the potential to protect environmental qual- ECI Economic complexity index ity. This is because it is highly driven by high innovation and research FDI Foreign direct investment activity that promote the sustenance of eco-friendly technology and AMG Augmented Mean Group products [65]. Thus, as the economic complexity of a country increases PHH pollution haven hypothesis environmental quality is more likely to increase [69,48]. This is tied to EKC Environmental Kuznets Curve the logic that as the economic complexity index increases research and SDG’s Sustainable Development Goals innovation activities increase and green innovation and pro- UN United Nations environmental production methods. Hence the need to understand the EU European nations role of economic complexity index in mitigating environmental poor OEC Observatory of Economic Complexity quality. CO2 Carbon dioxide emission To this end, Grossmann, and Krueger [40] observed three distinct εit Disturbance term effects: scale, composition, and technological effects to describe the EKC ΔYit Change in output relationship. The initial stage of real income generation is consistent β1⋯⋯.β6 coefficients of the regressors with the scale effect, where at this stage there is greater damage done to β1 > 0, β2<0 andβ3 > 0 positive, negative, and positive the environment due to greater consumption of inputs (labour, natural coefficients respectively resources, and capital) which inadvertently adversely impacts the environment. In the composition effect which depicts the second phase of economic expansion, there is the transition toward service sector. This in other words suggests that after attaining a maximum point of pro- economies have shown significant promise economically (Danish & duction, further increases in production exceeds a turning point beyond Wang, 2019), especially with their GDP haven grown from $2187b to which pollution will begin to recede. This explains the inverted U-sha- $16,266b between 1985 and 2016, averaging a 6.5 % growth rate ped curve. At the third phase, where the technological effect is felt, more annually (World Bank, 2017). Experts thus perceive that, at increased eco-friendly production and equipment are employed to produce goods technology adoption, greater economic expansion, and transition from and services, hence lesser harmful effects on the environment which pollutant intensive (industrial engagements) to service economies will promotes environmental quality [69,56]. In conclusion, the EKC vali- drive much reduction in pollution in the environment [29]. However, dation implies that at the beginning of economic expansion, there no the rapid economic growth of BRICS economies can drive greater modification to technology and structure, hence there is poor environ- resource consumption, which subsequently poses significant environ- mental quality; whiles the economy expands steadily modifications to mental concerns [27,72]. Dong et al., [32] contends that the unsus- technology and labour begin to occur, thus increasing the technological tainable utilization of natural resources also presents environmental development and energy efficiency rates that mitigate environmental consequences such as water pollution, deforestation, climate change and degradation. greenhouse gas emissions. Thus, the fast development rate of BRICS Beyond considerations for the Environment Kuznets Curve frame- nations carries with it potential consequential effects like greater CO2 work, the need for realizing sustainable development goals has driven emissions [31]. the need to examine the link between economic complexity and other The economic expansion and pollution nexus has been scrutinized factors such as urbanization, FDI, natural resource use, renewable en- extensively, beginning with Grossman and Kruger (1991) who intro- ergy sources and CO2 emissions among others (Antonakakis, Chatzian- duced the EKC framework, which has been commonly researched by toniou, & Filis, 2017; Neagu,& Teodoru, 2019). To this end, prior studies many scholars. The EKC has been confirmed by several studies both examining the energy-environment connection indicate that economic recent and earlier ones (see, [60,29,74]; Solarin et al., 2017). As such, expansion and energy consumption are crucial to the situation of CO2 conclusions such as renewable energy utilization mitigating environ- emission [19,30]. Industrial expansion through the injection of foreign mental problems such as carbon emissions, or urbanization driving en- or domestic financial injection coupled with growing urban population ergy consumption growth abound. increases demand for energy. These events will drive economic expan- This current study pursues a similar examination, in which it draws sion, but at the expense of environmental safety and quality. The main on the EKC framework to discern the relationship between economic issue common to these crucial determinants of economic expansion is complexity, and other known but crucial antecedents of CO2 emissions. the increased consumption of unclean energy like fossil sources among To begin with, an emergent but crucial issue is the concept of eco- countries [22]. Furthermore, a series of varying reports exists from nomic complexity, which describes the extent of economic expansion several findings to guide different economic and geographical settings. and defines the level of knowledge and skill acquisition for exporting For instance, in addition to renewable energy sources, and foreign products. Economic Complexity is assessed by the Economic Complexity direct Investments in some examinations is identified as a driver of Index (ECI) which measures an economy’s capabilities to export [46]. ecological quality in line for its mitigating effects on CO2 emission. The ECI explains the extent of sophistication in exports and the ability of However, some other scholars have reported otherwise [50,22]; (Bal- an economy to produce several other products [68]. It assesses the salobre-Lorente et al., 2020), thus showing a grave lack of consensus on interrelatedness of economies to produce items estimated to be this issue. Further, in line with the pollution haven hypothesis, inflow of technology-intensive outputs or the volume of viable awareness accrued FDI into domestic pollution-intensive industries in developing econo- to an economy (Charfeddine & Kahia, 2019; [68]. Considering this, it is mies is higher than among developed economies. Thus, FDIs facilitate perceived that modern economic complexity requires greater energy economic expansion mainly among developing countries. In addition, utilization, which subsequently results in greater pollutant emission into urbanization is considered as a composite process, which includes a the environment. Studies have linked energy markets and carbon population, society, an economy, and some transition within space. emissions reinforcing the direct association between expansion in en- Urbanization is globally considered as an issue of concern because it ergy sources and environmental pollution (Neagu, 2019). The extent of carries with it some negative consequences. For instance, the UN in- superiority and sophistication of export commodities of an economy dicates that massive urbanization exerts pressure on natural and artifi- tells the extent of diversity of its economic complexity. However, cial resources. However, the impact of urbanization on the environment 2 D.Q. Agozie et al. S u s t a i n a b l e E n e r g y T e c h n o l o g i e s a n d A s s e s s m e n ts 53 (2022) 102597 as studied in literature reveals inconsistent findings. For example, ur- of ECI and manufacture structures on environmental quality, revealed banization is found to largely expand CO2 emission in nations while it the need to regulate existing economic growth strategies particularly in also increases request for energy in all levels of income in economies low- and middle-income European nations to drive safe economic (high-, middle- and low-income levels) [59], yet Al-Mulali et al., [7] expansion to control CO2 emissions in the EU? Similarly, Trinh et al. observed a nonlinear relationship between urbanization and environ- [66] observed a U-shaped connection involving ECI and environmental mental pollution. Drawing on the discussion, this current research pollution in the EU. While employing a fixed effect estimator, it was broadens the examination among these factors to cover economic found the economic complexity facilitated the depletion of forest but complexity, urbanization, natural resource use, renewable energy, and mitigated waste generation in Brazil [65]. Hence, the study concluded foreign direct investments on environmental quality among BRICS that economic complexity is a source of shock on deforestation and at- member states. mospheric pollution. This study extends the existing literature in the following ways: First, Other studies have linked ecological deficit to economic complexity it is among the first to examine the effects of economic complexity at with an inverted U-shaped curve (e.g., [26]). The implication of the marginal levels which has the possibility to determine the scale effect, inverted U-shaped EKC is that, relative to economic expansion, envi- structure, and technological effects on CO2 emission in the EKC. It stands ronmental quality deteriorates until economic growth attains a appre- among very few studies to have done this assessment, particularly for ciable level of development and income to drive positive impacts on the BRICS nations. Secondly, this research attempt also examines the dy- environment. More practically, this assertion relates to structural namic relationships involving ECI, renewable energy, urbanization, change which suggests that a productive system must first transition natural resource use, FDI and carbon emissions. Here it bridges the gap from a high pollution-intensive system toward low pollution-intensive in literature by shedding light on evidence from BRICS group of nations. one to reduce negative environmental consequences. It introduces the interaction term economic complexity and urban population growth. This is expected to help policy and BRICS countries FDI and quality environment emerging sustainable dimensions and if urban population growth miti- gates the association between CO2 emissions and economic complexity. Studies reveal that foreign and domestic capital inflows can present adverse environmental consequences, particularly in economies with Review of previous studies high-pollution intensive systems. For instance, Murshed et al. [54] re- veals that foreign direct capital inflows have positive environmental In seeking to discern the association between economic complexity consequences. This is because it is perceived that foreign direct in- and environmental quality, there is also the need to understand the role vestments drive the use of modern technology and cleaner energy of other economic factors including economic expansion, urban popu- sources. Contrary, Solarin and Mulali, [63] found FDI inflows to increase lation growth, and renewable energy in ensuring quality environment. adverse effects on the environment in Ghana. In a similar assessment, In this current study economic complexity, renewable energy use, ur- Elheded et al. [38] confirms the pollution Haven hypothesis among banization, natural resource use, and foreign direct investment. Based countries in the MENA region, by observing increased pollution due to on this scope, the review of extant works is divided into sections dis- FDI inflows. cussing prior studies on each of these factors: Despite these negative effects being established by some prior studies, also established differing views on the unique association be- Economic complexity (ECI) and environmental quality tween direct investment inflows and environment quality. Like Adebayo [1], found no significant link between FDI and environmental degra- The framework of an inverted U-shaped curved proposed by Kuznets dation in twenty countries. Destek and Okumus [28] showed the exis- [47] has been used to explain the association between economic tence of a nonlinear relationship between environmental pollution and expansion and its effects on the environment, widely determined as the foreign direct investment inflows among newly industrialized nations. environmental Kuznets Curve (EKC). This framework has been applied Among OECD states, Alshubiri and Elheddad [9] examined the asym- and extended in many studies till date [42,23]. Drawing on the EKC’s metric association between CO2 emission pollution and FDI between analogy, it asserts that in the early stages of attaining economic 1990 and 2015. As such they observed direct inflows increased pollution expansion there is increased automation and production activity, hence effects at the inception stages of FDI injections, thus confirming the EKC economic expansion increases damage to the environment at this stage, proposition. Among ASEAN countries the influence of FDI inflows, and as the economy steadily develops appropriately with required economic expansion and energy utilization on environmental pollution technology the initial environmental quality begins to be restored. This using the panel quantile regression, also confirmed the pollution Halo indicates a turning point in the relationship between environment hypothesis in the region, between 1980 and 2010 [73]. quality and economic expansion. From a comparative assessment view between developed and A vast body of literature exists on the EKC hypothesis relative to developing economies on the connection regarding FDI inflows as well unclean energy sources which contributes about 80 % of energy as CO2 emissions, Adeel-Farooq, Riaz, and Ali [4] found FDI inflows consumed globally and accounts for 75 % of the world’s greenhouse gas increased CO2 emissions greater in developing economies, thus affirm- emissions [18]. Many empirical examinations indicate that increased ing pollution Halo hypothesis. Contrawise, FDI inflows in the developed renewable energy contribute to reducing carbon emission and envi- economies rather increased environmental quality. In like manner, a few ronmental quality [68,69]. This current examination similarly uses the other studies have extended this conclusion to other regions. For EKC framework, to study the connection among CO2 emission and ECI in example, Usman et al., [69] confirms the EKC and the pollution Halo BRICS (Brazil, Russia, India, China, and South Africa) countries, thus it hypothesis for the Asia and Americas, for MINT countries (Mexico, substitutes income with economic complexity. The economic Indonesia, Nigeria, & Turkey) Balsalobre-Lorente et al., [12] and Baloch complexity indicator determines the extent to which an economy pro- et al. [11] for BRICS. duces and exports wide variety of items [46]. In line with this view, it is as if modern economic complexity drives greater demand for energy, Clean energy utilization and CO2 emission which eventually results in the emission of more pollutants into the environment. This is because the more diverse an economy becomes, the Further, this review investigates the interactions between renewable more it increases in sophistication and superiority in exports, hence energy sources and pollutants. In this light, the argument has widely greater economic complexity. This is evident in the conclusions of some been to determine or discover how to sustain quality environments prior studies. For instance, in Dogan et al., [31], who assessed the impact through the exploration of clean energy sources. A vast body of 3 D.Q. Agozie et al. S u s t a i n a b l e E n e r g y T e c h n o l o g i e s a n d A s s e s s m e n ts 53 (2022) 102597 Table 1 Urban population growth and emission Description of variables, Symbols and unit of measurement utilized for the study. Name of Abbreviation Proxy/Scale of Source The migration from less developed locations to urban centers has Indicator Measurement steadily increased over time, and this has also increased pollution con- Carbon Dioxide CO Measured in metric tonnes BP, 2020 sequences on the environment [13]. Considering this, several studies 2 Emissions Per have explored the actual role of urbanization in environmental pollu- Capita tion. In line with these studies like those of McGee and York [51] Economic ECI Nations productive OEC, Economic confirmed the assertion in a reverse observation. In that, a reduction in Complexity composition appearance complexity Index by combining the rankings, 2020 urban population rate contributed to a significant reduction in pollution information on their levels using data from 1960 to 2010. Elahi et al. [37], Chien et al. [25], variety number of Bong et al. [20] all observed this same association between urbanization commodities it exports and environmental pollution. Thus, all these examinations establish a Square of ECI2 Measures the square of OEC, Economic positive relationship between urbanization and environment pollution, Economic Economic Complexity complexity Complexity Index rankings, 2020 thus confirming the EKC hypothesis in the long run. Index Cube of ECI3 Measures the cube of OEC, Economic Gap in the literature Economic Economic Complexity complexity Complexity Index rankings, 2020 Index Throughout this review, it is observed that not much exists on Foreign Direct FDI % of real GDP WDI, 2020 highlighting the effect of economic complexity and these other in- Investment dicators in BRICS nations. Further, the interactions existing on many of Natural NR Total natural resource rent WDI, 2020 these variables have revealed a litany of mixed results suggesting much resources (% of GDP) inconsistency. This these findings do not provide many grounds for Urban UB (% of total population) WDI, 2020 population objective or conclusive focus for policy guidelines, hence a significant gap still existing needing to be addressed. Therefore, this current study Renewable REU Renewable energy BP, 2020 pursues the goal of bridging this gap by focusing on economic Energy consumption (% of total complexity as a new consideration, FDI, clean/renewable energy natural final energy consumption) Interaction term ECI*UP Economic Complexity resource use as means for improving environmental pollution through Index* Urban population reduced CO2 emissions. Source: Authors compilation. Methodology empirical studies, based on long run and short run macro and micro Data sources and variable description assessments of the interactions between renewable energy, economic expansion, and pollutant pollution in the environment. In Nguyen and The available study has established that radical transformation, the Kakinaka [53] the assessment of the link between environmental quality structure of energy mix (non-renewables, and renewable sources, and renewable energy revealed clean energy utilization in low-income among other things such as trade, product complexity, and economic states drives low emission levels, although this association was not complexity all have an impact on the environment (Neagu and Teodoru, consistent with the findings from the developed or high-income econo- 2019; Destek and Sinha, 2020; [61,13]). Specifically, as pertaining to mies. Farhani and Shahbaz [39] earlier positioned that both clean and the findings of the study, systemic modify and economic complexity may unclean energy sources augment environmental pollution. This evidence have a greater impact on environmental quality since they are deemed as drawn from MENA countries using an FMOLS and DOLS estimations to be precise predictors of economic growth, skill- and knowledge-based indicate an inconsistency in the findings on the link between renewable advanced output, and therefore have a greater impact on environmental energy and environmental pollution. Usman et al. [69] while performing quality. An annual data from 1990 to 2019 is utilized for the BRICS an extended analysis on four continents, examined the interaction be- economics for this study. The data for the economic complexity index tween FDI, economic expansion, trade, renewable and nonrenewable (ECI), which measures the aspect of a country’s industrial mix by inte- energy use on environment degradation confirmed that renewable en- grating knowledge on the diversity of products it exports, is acquired ergy utilization improved environmental degradation. This finding as from the OEC database. Furthermore, the ECI is a comprehensive established by Usman et al. is replicated in other studies such as Danish assessment of the production capacity of significant economic cate- et al. (2020), in BRICS nations using the DOLS and FMOLS between 1992 gories, which are typically towns, nations, or territories. ECI is used to and 2016. Further, Danish et al. (2020) determined that renewable en- define the knowledge acquired by a populace and shown in the eco- ergy and urban population growth increases environmental quality in nomic operations that take place in a town, a nation, or a territory, in BRICS countries. Thus, BRICS nations require a transition toward clean specific. As a means of achieving this goal, the ECI describes material energy sources. Similar findings are reported in Alola et al. [8], who available at a site as median understanding of the activities existing in revealed a connection between economic expansion, trade policy, that site and understanding of an economic operation as the basic ability fertility renewable and nonrenewable energy, and ecological footprints of the areas where that economic activity is carried out. The remaining in the EU. coefficients, thus, FDI, natural resource, and urbanization are taken Although these positive effects are reported on renewable energy and from the world development index (WDI) while CO2 emission and the environmental pollution nexus, there is evidence of consequential renewable energy is taken from British Petroleum (BP). As seen in effects of renewable energy utilization on environmental pollution (see Table 1, a brief description of all variables is provided.Table 2.. [1]; Danish et al., 2020; [8,58,70,24]. In conclusion, there many con- clusions supporting the positive effects of renewable energy likewise its Proposed framework negative effects on environmental pollution. This raises the need for further examinations into this association to better discern the direction With the addition of natural resources, this analysis intends to assess of this effect. the legitimacy of an environmental Kuznets Curve (EKC) assumption based on the study of Balsalobre-Lorente et al [13] which was conducted for PIIGS countries. In our study case the focus is on BRICS countries 4 D.Q. Agozie et al. S u s t a i n a b l e E n e r g y T e c h n o l o g i e s a n d A s s e s s m e n ts 53 (2022) 102597 Table 2 Cross-sectional dependency (CD) and Slope Homogeneity (SH) results. Model Pesaran CD Test p-value Pesaran LM Test p-value Breuch-Pagan LM p-value LCO2 18.255* (0.0000) 48.339* (0.0000) 757.142* (0.0000) LECI 33.256* (0.0000) 85.044* (0.0000) 1252.315* (0.0000) FDI 13.245* (0.0000) 45.378* (0.0000) 634.105* (0.0000) LREU 14.385* (0.0000) 50.617* (0.0000) 787.868* (0.0000) LNR 15.3195* (0.0001) 48.298* (0.0000) 756.583* (0.0000) LUB 21.210* (0.0000) 66.844* (0.0000) 1006.787* (0.0000) LECI*LUP 28.334* (0.0000) 77.116* (0.0000) 1145.360* (0.0000) Slope Homogeneity (SH) COEFFICIENT p-value SH (Δ̃ test) 5.723* (0.0010) SH (Δ̃ adj test) 4.123* (0.0030) NOTE: * represents 1% level of significance. while accounting for the pivotal role of economic complexity in an EKC Methodology environment for emerging blocs like BRICS. This analysis converted the dataset of all investigated coefficients, except for FDI, into a natural Cross-section dependence (CD) logarithm structure to minimize multicollinearity concerns, minimize To establish the suitable methodological approach(s) for this inves- the possibility of misfits from the dataset, and conquer the possibilities tigation, we used the cross-section dependency (CD) approach. The of data sharp and normalcy problems. As a result, the connection findings of the CD approach could help us decide whether to utilize first- involving carbon emission, economic complexity, FDI, renewable en- generation or second-generation panel data estimate approaches. The ergy usage, natural resources, as well as urbanization can be stated in research may be biased, inappropriate, and conflicting if the CD evalu- eqn-1 as: ation is not conducted [42]; Odugbesan et al. 2021). We utilised a robustness evaluation utilising three-CD tests: the Pesaran (2007) CD, MODELI : LCO2it = β0 + β1LECIit + β2ECI2it + β3FDIit + β4LREUit Pesaran (2015) scaled LM and Breusch and Pagan [21] approaches, to + β5LNRit + β6LUBit + εit (1) ensure that the aforesaid difficulties do not emerge. The CD test is depicted as follows: where CO 22 emission, ECI, ECI , FDI, REU, NR, UB are coefficients stated √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ above. Moreover, β1⋯⋯.β6 stands for the coefficients of the regressors ( ) ( ) 2T ∑N− 1 ∑N while I and t represent nations and timeframe respectively. CD = p̂ij (5) N(N − 1) Model-2 investigates the connection involving urbanization and i=1 j=i+1 economic complexity, with the expectation that urbanization and eco- Whereas from Equation (3), p̂ij identifies the indicators of the nomic complexity will have a mediating impact in the reduction of CO2 remaining evaluation of ADF regarding the pairwise cross-sectional output. This regulating role is formally represented as below in eqn-2 as: interconnection. N and T are the panel range and model specifically MODELII : LCO2it = β0 + β1LECIit + β2ECI2it + β3FDIit + β4LREUit for the e time and cross-section. + β5LNRit + β6LUBit + β7LECI*LUPit + εit (2) Stationarity approach Model-III analyzes the existence of a cubic link involving the ECI as It is vital to identify stationarity attributes of indicators under well as CO2 emission in the BRICS nations. The following is the pur- investigation before moving to further analysis. Moreover, if there is an poseful pattern of the cubic-shaped EKC assumption, which can be indication of CD, utilization of first-generation unit root test will pro- represented in eqn-3 as: duce outcomes that are misleading [2,44]. Based on this knowledge we MODELIII LCO2 β β LECI β ECI2 β ECI3 β FDI utilize unit root tests that can identify variables stationarity feature : it = 0 + 1 it + 2 it + 3 it + 4 it ε amidst CD. Thus, we utilized 2nd generations stationarity test to identify + β5LREUit + β6LNRit + β7LUBit + it (3) variables of the investigation stationarity attribute. We utilized both Furthermore, Model-IV reflects on the N-shaped link involving eco- CIPS and CADF to catch the order of the coefficients of integration. nomic complexity and CO2 emission, as well as the moderating influence Equation presents the CADF as follows: of urbanization that is incorporated with economic complexity. The ∑p ∑p following is the mathematical representation for Model-IV, which can be ΔYi,t = γi + γiYi,t− 1 + γiXt− 1 + γilΔYt− l + γilΔYi,t− l + εit (6) found in eqn-4 as: l=0 l=1 MODELIV : LCO2it = β0 + β1LECIit + β2ECI2it + β3ECI3it + β4FDIit In Eq. (8), Yt− 1 and ΔYt− l shows the cross-section average. The value + β LREU + β of CIPS is obtained as follow: 5 it 6LNRit + β7LUBit + β8LECI*LUPit + εit (4) Per the N-shaped EKC assumption, the connection linking CO ∑n2 1ĈIPS = CADFi (7) emission and economic complexity is expected to have positive, nega- N i=1 tive, and positive signs (β1 > 0, β2<0 and β3 > 0) respectively. Moreover, FDI is expected to have positive impact on the environment base on the The cross-section augmented Dickey-Fuller technique obtained from existing literatures since it involves dirty funding, conventional mech- Equation (6) is represented by the term CADF in Equation (7). anism, pollution haven companies as well as the use of fossil fuel and other related energies. Again, renewable energy is expected to have Cointegration approach negative impact with the environment whiles both urbanization and If there is a presence of CD, utilisation of first-generation cointe- natural resources are expected to have positive connection with the gration such as Pedroni and Kao cointegration tests will produce environment. misleading outcomes since they do not consider CD. Based on this knowledge, we utilised Westerlund cointegration initiated by West- 5 D.Q. Agozie et al. S u s t a i n a b l e E n e r g y T e c h n o l o g i e s a n d A s s e s s m e n ts 53 (2022) 102597 Table 3 Table 6 Panel CIPS unit root test. FMOLS estimation for sensitivity check. VARIABLES CIPS Variables MODEL I MODEL II MODEL III MODEL IV I (0) I (1) LECI 0.0483* 0.297* 0.102* 0.308* p-value (0.003) (0.001) (0.001) (0.000) C C&T C C&T Decision LECI2 − 0.001* − 2.89E-* − 9.85E** − 5.78E*** LCO2 − 3.221* − 3.130* – – I(0) p-value (0.000) (0.003) (0.031) (0.088) LECI − 1.324 − 2.561 − 3.713* − 3.930* I(1) LECI3 – – 8.62E-* 5.03E-** FDI − 3.706* − 3.993* – – I(0) p-value – – (0.041) (0.011) LREU − 0.509 − 1.735 − 4.008* − 4.643* I(1) FDI 0.007* 0.009* 0.005* 0.008* LNR − 3.193* − 3.494* – – I(0) p-value (0.008) (0.000) (0.033) (0.025) LUB − 3.217* − 3.305* I(0) LREU − 0.138* − 0.081* − 0.139* − 0.085* LECI*LUP − 3.366* − 3.849* I(0) p-value (0.000) (0.001) (0.000) (0.000) LNR 0.304* 0.411* 0.530* 0.366* NOTE: * represents 1 % level of significance, while C = constant and C&T = p-value (0.000) (0.000) (0.000) (0.000) constant and trend. LUP 0.089* 0.5817* 0.0821* 0.505* p-value (0.001) (0.000) (0.003) (0.006) LECI*LUP – − 0.080* – − 0.071* Table 4 p-value – (0.004) – (0.001) Bootstrapped Westerlund [71] cointegration. Shape obtain Inverted U- Inverted U- N-Shape N-Shape Shape Shape Statistics Value Z-value p-value Robust p-value R_sq 0.973 0.974 0.973 0.974 Gτ 3.143* 0.638 (0.080) (0.000) Adj R_sq 0.971 0.972 0.971 0.972 − − Gα 5.369* No. 6 7 7 8 − − 1.235 (0.058) (0.001) Pτ 5.478* 1.356 (0.086) (0.004) regressors − − Pα 6.101* 2.256 (0.010) (0.005) No. of group 5 5 5 5 − − NOTE: * 0.01. NOTE: * p < 0.01, ** p < 0.05, *** p < 0.1 respectively. < Table 5 Table 7 Augmented Mean Group (AMG) analysis. Dumitrescu and Hurlin [34] causality analysis. Variables MODEL I MODEL II MODEL III MODEL IV W-stat. Z-Stat p-value CAUSALITY FLOW LECI 0.138* 0.119** 0.089* 0.104** LECI → LCO2 3.566*** 1.747 (0.080) LECI ↔ LCO2 p-value (0.000) (0.020) (0.008) (0.041) LCO2 → LECI 3.842** 2.130 (0.033) LECI2 0.002* 0.004* 0.001* 0.001* FDI → LCO2 1.756 − 0.765 (0.444) LCO2 → FDI − − − − p-value (0.000) (0.000) (0.002) (0.003) LCO2 → FDI 4.717* 3.345 (0.000) LECI3 – – 0.003* 0.002* LREU → LCO2 2.529 0.307 (0.758) LCO2 → LREU p-value – – (0.003) (0.004) LCO2 → LREU 5.329* 4.195 (3.E-05) FDI 0.010* 0.011* 0.009* 0.009* LNR → LCO2 3.259*** 1.321 (0.086) LCO2 ↔ LNR p-value (0.002) (0.005) (0.003) (0.003) LCO2 → LNR 6.146* 5.330 (1.E-07) LREU 0.025* 0.023* 0.026* 0.028* LUP → LCO2 6.296* 5.538 (3.E-08) LCO2 ↔ LUP − − − − p-value (0.056) (0.086) (0.042) (0.039) LCO2 → LUP 4.021** 2.379 (0.017) LNR 0.604* 0.591* 0.580* 0.596* LECI*UP → LCO2 5.620* 4.599 (4.E-06) LECI*UP ↔ LCO2 p-value (0.000) (0.000) (0.000) (0.000) LCO2 → LECI*UP 3.682*** 1.908 (0.056) LUP 0.073 0.032* 0.072* 0.118*** NOTE: * p < 0.01, ** p < 0.05, *** p < 0.1 respectively. p-value (0.000) (0.004) (0.000) (0.062) LECI*LUP – − 0.048*** – − 0.050*** p-value – (0.088) – (0.066) Shape of EKC Inverted U- Inverted U- N-Shape N-Shape Pα = Tά (11) Shape Shape Wald test 16.62b 15.68b 10.75b 9.16a The test alternative and null hypotheses are “there is cointegration” P-value (0.0121) (0.0382) (0.0456) (0.0027) and “no cointegration” accordingly. No. 6 7 7 8 regressors No. of group 5 5 5 5 Augmented Mean Group (AMG) and Fully Modified-Ordinary Least NOTE: * p < 0.01, ** p < 0.05, *** p < 0.1 respectively. Squares (FMOLS) Techniques erlund [71] to catch the long-run association between energy intensity Therefore, the authors utilized two robust method that are consid- and the regressors. Unlike both Pedroni and Kao cointegration tests, ered to lodge the latter anxiety of this analysis The Augmented Mean Westerlund [71] considers CD. The Equation below presents Westerlund Group (AMG) as the main technique for the analysis while the Fully [71]. Modified-Ordinary Least Squares (FMOLS) is use as sensitivity check. The AMG method offers the unusual capacity to account for cross- 1 ∑N άi sectional dependence as well as slope heterogeneity. It can sustain a Gt = (8) N i 1 SE(̂I¬i) unique path because of the way commonly affected impacts are handled. − For the AMG these impacts represent a single continuous change that 1 ∑N Tά can be compensated for by deducting it from the dependent factor. The G iα = (9) N Î 1 Augmented Mean Group (AMG) heterogenous panel estimator of Eber-i− 1 ¬i( ) hardt and Bond [35] as well as Eberhardt and Teal [36] were utilized in Î the study following the expression in eqn-12: ¬ PT = (10) ∑ SE(̂I T¬) ΔYit = αi+βiΔXit + π D +φ UCF + μ (12) t=1 t t i t it The OLS method of the difference is applied to the AMG technique. 6 D.Q. Agozie et al. S u s t a i n a b l e E n e r g y T e c h n o l o g i e s a n d A s s e s s m e n ts 53 (2022) 102597 Fig. 1. Graphical presentation of empirical analysis. Source: Authors compilation. This is shown in eqn-13 whereas φi symbolises the projected slope pa- shown in Table 4 traces a long run equilibrium relationship between the rameters of Xit coefficient in eqn-12. highlighted variables in the panel analysis. The conclusion was sup- 1∑ ported by the evidence of rejecting the null hypothesis. N AMG = φ (13) N i=1 i Panel estimation techniques Panel causality To identify the causal interrelationship regarding renewable energy Furthermore, this research analyzes the long-run flexibility of the as well as CO2 emission and the regressors (NR, Y, FDI and UB), this explanatory variables that could impact environmental deterioration, research utilizes [34] panel causal technique. We choose the first dif- either negatively or positively. We used the AMG technique (table 5) for ference of all non-stationary indicators because the test is still not all four models to derive long-run elastic in the contexts of the EKC as relevant to non-stationary data. The cross-sectional knowledge is well as PHH frameworks, while FMOLS (table 6) is used as a sensitivity included in this test, which is an extension of the existing Granger or robustness check for this analysis. However, both outcomes show causality formula. Because of its adaptability to varied mixtures of time relatively close outcomes, with slight variations mainly noted in relation periods and cross-sections, this test has much more strength. In addition, to the scales of the assessed variables and their equivalent level of sta- the test is applicable to both heterogeneous and balanced panels. In the tistical significance. All the models confirm the present of EKC and PHH context of CD, the test yields beneficial results. by adding renewable energy, natural resources, and urbanization to affirm if the economic-FDI-environs relationship fits an N-shaped or not. ∑ α p ∑ z j p j i,t = i + β zi,t− j + γ Ti,t− j (14) Form model I and II, the analysis aims to check for the existence of j=1 i j=1 i inverted U-Shaped EKC connection for economic complexity (β1 > 0, The alternative and null hypotheses are “there is causality” and “no β2<0). From both models, the outcomes prove positively and negatively causality” accordingly. significant coefficients confirming the presence of inverted U-Shaped EKC for the BRICS countries. However, from model III and IV the Empirical outcomes and interpretation outcome showsβ1 > 0, β2<0 andβ3 > 0 which affirms the presence of N- Shape EKC connection involving economic complexity and CO2 emis- Based on the findings of the empirical research, individual time se- sion. As a result of this scenario, carbon emission habit is classified ac- ries are first analyzed to determine whether or not there is cross- cording to ECI, where the first phase of economic expansion generating a sectional dependence (CSD). This is done by applying the Breusch- high amount of ecological degradation (β1 > 0). At this point, the Pagan LM test, the Pesaran scaled LM test, and the Pesaran CD tech- economy will be affected by the scale effect, as previously demonstrated niques, all of which can be found in table 2. The cross-sectional link by Adebayo et al. [2] and Sinha and Shahbaz [62]. Upon attaining a demonstrates that the null hypothesis CSD outcome can be rejected at specific income scale where carbon emissions show a declining trend (β2 the one percent level of significance for all the techniques utilized in this < 0, a rise in economic complexity will result in a reduction in ecological study. This implies that the panel unit root analysis must consider the harm. This verifies the composition and technical effect in the BRICS connection among cross-sectional individuals. However, the Pesaran & economies,which is like previous outcomes of [42,43,17,13]. Further- Yamagata, (2008) SH techniques on the other hand produced a 1 % more, after obtaining the lowest possible degree of emissions, economies siginficat level. This indicates that a shock appears to be transmitted to tend to place less emphasis on ecological protection, and the scale other nations within the panel in each of the oil-producing countries in technological obsolescence effect (β3 > 0) begins to manifest itself as a sub-Saharan Africa. The findings proceed to demonstrate that neither result [16,13,57,3,41]. This proofs that, all the four models affirm the multicollinearity nor serial autocorrelation can be found among the existence of inverted U-Shaped whiles model III and IV affirms the ex- datasets under consideration. The results of the CIPS unit root technique istence of N-Shaped EKC of ECI and carbon emission for the BRICS by Pesaran (2007) presented in Table 3 provide evidence in favor of this economics. assumption for the coefficients that were investigated, and Table 4 Moreover, FDI was seen to also have positive impact on the ecology contains the outcomes of the panel cointegration investigation. The CIPS for the BRICS economics. There are other elements that contribute to the outcomes validate that all variables are stationary after difference. appeal of FDI, such as availability to cheaper labour, proximity to the Subsequently, outcome of the Westerlund [71] Cointegration test sector, and less stringent policies in terms of controlling the abuses of overseas investors, that make this outcome more likely. This serves as a 7 D.Q. Agozie et al. S u s t a i n a b l e E n e r g y T e c h n o l o g i e s a n d A s s e s s m e n ts 53 (2022) 102597 reminder that economies in the BRICS are continuing developing in their Causality analysis economic operations and growth while paying little attention to the health of their ecology. This supported the position of some critics of Form the Dumitrescu and Hurlin [34] Granger causal technique foreign direct investment, particularly those concerned with the long- presented in Table 6, it was observed that there is bidirectional causality term viability of underdeveloped nations. This observation lends cred- between ECI and CO2 emission, natural resources and CO2 emission, ibility to the notion of a pollution haven (PHH). This validates the results urbanization, and CO2 emission as well as the interaction between ECI of Sarkodie and Strezov [60],Udemba [67]; Gyamfi et al. [45] and Steve and urbanization with CO2 emission. However, a uni-directional cau- et al [64]. sality is obtained between CO2 emission and renewable energy as well as The relationship between renewable energy and the environment is FDI and CO2 emission. All these findings are similar with the early determined to be negative and statistically significant. In recent years, it findings from table 5 which indicates that, ECI, natural resources has been proven that transitioning to more environmentally friendly unionization and FDI all have positive impact on the environment while (renewable) energy sources result in improved environmental outcomes renewable energy has negative impact on the environment which the regardless of where the economy is located. This is line with the ob- findings are in line with Gyamfi et al [42], Bekun et al [17] and servations of Dong et al. [32], Gyamfi et al. [42] Bamidele et al [14], Balsalobre-Lorente et al [13]. Therefore, to attain a decrease in CO2 Ohajionu et al [55] and Agboola et al [5]. emission from enterprises in the BRICS economies and to aid efforts The outcome acquired from natural resources, on the other hand, is toward green development and healthy economy, stricter laws will be found to have a positive and statistically significant link with ecological required. As a result, strategy and decision-makers should investigate degradation. This verifies the findings of Amed et al. (2020) and Gyamfi additional techniques for raising renewable energy production, thereby et al. [45] who found that natural resources promote pollutants in BRICS boosting green intake, and ensuring that environmental damage is kept economics. These nations have a substantial quantity of revenue that can to the bare minimum through the adoption of modern innovations for be utilized for both export and inland intake. This discovery, however, reducing carbon emissions. Fig. 1 outlines the causality scheme over the lends weight to the impression that, obtained from natural resources sampled period and chosen variables. from these economics has never been lucrative. Extreme dependence on natural resources contributes to the loss of biocapacity, which is the Conclusion and policy direction capacity of living organisms to reproduce [15]. Furthermore, consid- ering the vital consequence of the BRICS economics, the usage and Conclusion growth of agricultural resources encourage deforestation, which in- creases emission. Aside that, several countries make use of their natural To achieve numerous Sustainable Development Goals (17-SDG) ob- resources (coal, petroleum, and natural gas) to acquire their energy jectives, a cleaner and healthier environment is essential. With only a requirements. It has been claimed that the profusion of resources would little less than decade left before the final deadline of 2030, the world’s enable a nation to become more self-sufficient by dropping energy economies must pick up the tempo and invest more resources in the importation and depending on inland energy generation with less finding of better explanations for rising temperatures, pollutants, habitat emission levels [6]. destruction, and climate change to revitalize cultures and economies. Furthermore, we discovered that urbanization exacerbates ecolog- Following this viewpoint, this study investigated the dynamic connec- ical damage in the BRICS economies. Several previous analysis have tion between ECI, FDI, renewable energy usage, natural resources, ur- determined that urbanization has an adverse impact on ecological banization, and CO2 emission for the BRICS countries. The research deterioration, and this study adds to that body of knowledge [52,45,10]. makes use of datasets spanning the years 1990 and 2018, as well as 2nd It is beneficial to the economy to increase the density of towns with generational methods like CIPS, Westerlund cointegration, slope het- limited resources, just as it is beneficial to increase the density of towns erogeneity (SH), CSD, AMG, FMOLS, and causality proposed by Dumi- with abundant resources. As a result, among other things, it increases trescu and Hurlin [34]. The outcome of the CSD as well as SH provide the demand for transportation, housing, and household appliances [49]. evidence in favor of the adoption of approaches from the second gen- It is noteworthy to consider the ecological consequences of these eration. The results of the Westerlund cointegration analysis revealed shifts in economic models in nations where the major task is shifting the existence of a long-run interconnectivity involving the dependent from the core to the tertiary sector. According to the indicator LECI*LUB coefficient of CO2 emissions and the regressors. The outcome from the (ECI interrelated with urbanization) given in Tables 5 and 6, this methods (i.e., AMG, and FMOLS) show relatively close outcomes, with concept is supported by the fact that there is an adverse association slight variations mainly noted in relation to the scales of the assessed involving urban growth as well as economic complexity and environ- variables and their equivalent level of statistical significance. From the mental emission (β8 > 0). From the analysis, the moderation effect will outcomes, the inverted U-Shaped and the N-Shaped EKC was confirmed decrease pollution in the environment by 0.04 and 0.05 in models II and showing a positive, negative and positive connection between ECI, ECI2 IV respectively. In light of this viewpoint, Balsalobre-Lorente et al [13] and ECI3 with CO2 emission respectively. Moreover, the connection shown that the environmental harms caused by an economy are highly between FDI and CO2 emission was positively significant which affirms connected with the combination of products that it exports and manu- the existence of PHH for the BRICS economics. Again, both natural re- factures. It is therefore critical to analyze which areas of the economy sources and urbanization increases emission whiles renewable energy are dominant in a country order to gain a better understanding of the and the interaction involving ECI, and urbanization have negative country’s environmental efficiency. As economic activity accelerates impact on the environment of BRICS economics. Furthermore, from the and countries’ economies become more prosperous, the financial sys- causality analysis, a bidirectional causality between economic tems of these nations shift in the path of the tertiary sector, thereby complexity, natural resources, urbanization as well as the interaction lowering the degree of ecological degradation in the process. Within this between economic complexity and urbanization and the dependent time frame, there is a greater desire for protection, knowledge, health, variable, thus, CO2 emission while a uni-directional causality is obtained and comfort, which enhances concern about the surroundings.Table 7.. involving FDI as well as renewable energy and the dependent variable CO2 emission. 8 D.Q. Agozie et al. S u s t a i n a b l e E n e r g y T e c h n o l o g i e s a n d A s s e s s m e n ts 53 (2022) 102597 Policy directions time spent traveling. The urbanization of the BRICS countries has resulted in the increased usage of energy supplies in both firms and Following the outcomes of the current study, we will create homes, as well as in transportation. Environmental protection should numerous policy recommendations that will be considered. be shifted away from traditional to sustainable energy usage, and renewable energy should be made available to companies and fam- • The results of the EKC theory, which is associated with economic ilies at appropriate and affordable costs. complexity, suggest that authorities in the BRICS countries should • The outcomes of this study should be taken into consideration by encourage the exporting of skill-intensive, knowledge-based, and authorities. With less than a decade remaining to fulfil UN goals, energy-efficient products to ensure a greener livelihood. Within this more serious attempts in discovering more effective ways to accel- spectrum, greater incentives should be provided to enterprises that erate the tempo of ecological deterioration are urgently required to adopt more environmentally acceptable technologies and use reach these goals. It will be necessary for the globe to make even greener forms of energy, and a Pigovian tax must be levied against greater attempts in discovering answers to urbanization problems, traditional businesses. As a result, the BRICS economies may increase making considerable public funding in green energy resources, and the exporting of sophisticated goods and products with substantial implementing comprehensive pollution control regulations. value while simultaneously protecting the environment. It has been noted that several emerging markets (notably the BRICS) have Limitation and future studies turned to relax laws and restrictions to draw higher levels of in- vestment. Nevertheless, the regulatory agencies of these nations The present study is clearly not without flaws since it is the subject of should now strive to enact severe ecological regulations to reverse several different limitations. Economic complexity, foreign direct in- the cumulative environmental destruction. There is a scarcity of vestment, renewable energy utilization, natural resource, and urbani- economic means, institutional abilities (particularly following the zation are not the only variables that influence carbon emission. 2008 financial crisis), and in numerous instances, political enthu- Another disadvantage of this study is the inaccessibility of the data. siasm, which is exacerbated by the campaigning of proficient in- Moreover, variables like, non-renewable energy, technological innova- vestors, which is impeding substantial progress toward an tion etc. can be added to further obtain it impact on the environment. ecologically responsible environmental framework. To achieve this Lastly, the study can be extended to other areas like Sub-Sahara Africa, goal, the authorities, regulators, and the central supervisory board of emerging seven (E7), G7 etc. these nations should regulate cleaner inflows of foreign direct in- vestment (FDI) to promote greener and healthier advanced tech- CRediT authorship contribution statement nologies and support alternative energy supplies throughout the area. Divine Q. Agozie: Validation, Visualization, Data curation, Writing • The goal of increasing renewable energy usage must be achieved in – original draft, Investigation. Bright Akwasi Gyamfi: Conceptualiza- place to bolster and execute the new financing in hygienic and tion, Formal analysis, Methodology. Festus Victor Bekun: Writing – cleaner energy infrastructure. As a result, more funds must be allo- original draft, Supervision. Ilhan Ozturk: Writing – original draft, cated to R&D to investigate newer and more dependable renewable Investigation. Amjad Taha: Validation, Visualization. energy financial advice. The investment in innovative technologies and the avocation of the usage of renewable energy would, generally, enhance the green healthiness of BRICS economies in general. To Declaration of Competing Interest achieve stable economic expansion and the usage of sustainable power along with the BRICS countries, strategies must be imple- The authors declare that they have no known competing financial mented in the areas of monetary flow, technological improvement, interests or personal relationships that could have appeared to influence and long-term economic plans in the BRICS countries. In a similar the work reported in this paper. vein, rules for discovering newly upgraded renewable resources would significantly boost economics, and technological improve- Data availability ments would aid in the production of renewable energy. Besides, carbon emission plans are also important to encourage investment in Data will be made available on request. renewable energy resources. • Taking into consideration urbanization, which is favourably con- Appendix nected with carbon emission, on the other hand, the safeguarding character of the ecology of urbanization is linked with the complexity of economic systems. As a result, healthy urbanization should be the foundation of healthy economic growth, because the positive impact of urbanization will help to reduce carbon emission. Table A1 Authorities and state agencies should take this viewpoint into List of countries. consideration when planning for sustainable development in all el- LIST OF COUNTRIES: Brazil, Russia, India, China and South Africa ements of urbanization, such as urban housing, sewage, and mobility. Policies aimed at dissuading personal vehicle ownership, increasing non-motorized (oil-free) mobility, and promoting mass transit should be maintained and expanded. Furthermore, strategies for consolidated lodgings should be developed to limit the amount of 9 D.Q. Agozie et al. S u s t a i n a b l e E n e r g y T e c h n o l o g i e s a n d A s s e s s m e n ts 53 (2022) 102597 Table A2 Descriptive statistics and correlation matrix analysis. LCO2 LECI FDI LREU LNR LUB LECI*LUP Mean − 0.896 9.193 0.688 2.233 4.416 3.893 36.358 Median − 0.975 9.301 0.738 2.472 4.476 4.001 34.915 Maximum 0.320 10.648 5.635 4.024 4.604 5.542 56.700 Minimum − 2.240 6.768 − 12.942 − 3.802 3.724 1.180 8.895 Std. Dev. 0.591 1.100 1.657 1.190 0.194 0.823 10.610 Skewness 0.045 − 0.148 − 2.244 − 2.003 − 1.589 − 1.006 − 0.278 Kurtosis 2.261 1.422 21.376 8.254 4.742 4.510 2.565 Jarque-Bera 7.098 33.061 4592.171 560.312 168.667 81.253 6.405 LCO2 1 LECI − 0.845 1 FDI 0.012 − 0.034 1 LREU − 0.163 − 0.063 0.207 1 LNR 0.495 − 0.113 − 0.095 − 0.486 1 LUB − 0.465 0.626 0.263 0.175 0.074 1 LECI*LUP − 0.684 0.849 0.179 0.088 − 0.012 0.939 1 References [20] Bong CPC, Lim LY, Lee CT, Klemes JJ, Ho CS, Ho WS. The characterization and treatment of food waste for improvement of biogas production during anaerobic [1] Adebayo TS. Environmental consequences of fossil fuel in Spain amidst renewable digestion: A review. J Clean Prod 2018;172:1545–58. https://doi.org/10.1016/j. energy consumption: a new insights from the wavelet-based Granger causality jclepro.2017.10.199. approach. Int J Sustain Dev World Ecol 2022:1–14. [21] Breusch TS, Pagan AR. The Lagrange multiplier test and its applications to model [2] Adebayo TS, Awosusi AA, Kirikkaleli D, Akinsola GD, Mwamba MN. Can CO 2 specification in econometrics. Rev Econ Stud 1980;47(1):239–53. https://doi.org/ emissions and energy consumption determine the economic performance of South 10.2307/2297111. Korea? A time series analysis. Environ Sci Pollut Res 2021:1–16. [22] Cai X, Che X, Zhu B, Zhao J, Xie R. Will developing countries become pollution [3] Adedoyin FF, Bein MA, Gyamfi BA, Bekun FV. Does agricultural development havens for developed countries? An empirical investigation in the Belt and Road, induce environmental pollution in E7? A myth or reality. Environ Sci Pollut Res J. Clean. Prod. 198 (2018) 624e632, doi:10.1016/j.jclepro.2018.06.291. 2021;28(31):41869–80. [23] Caglar AE, Zafar MW, Bekun FV, Mert M. Determinants of CO2 emissions in the [4] Adeel-Farooq RM, Riaz RM, Ali T. Improving the environment begins at home: BRICS economies: The role of partnerships investment in energy and economic revisiting the links between FDI and environment. Energy 2021;215:119150. complexity. Sustain Energy Technol Assess 2022;51:101907. https://doi.org/10.1016/j.energy.2020.119150. [24] Chen Y, Zhao J, Lai Z, Wang Z, Xia H. Exploring the effects of economic growth, [5] Agboola PO, Hossain M, Gyamfi BA, Bekun FV. Environmental consequences of and renewable and non-renewable energy consumption on China’s CO2 emissions: foreign direct investment influx and conventional energy consumption: evidence evidence from a regional panel analysis. Renew Energy 2019;140:341–53. https:// from dynamic ARDL simulation for Turkey. Environ Sci Pollut Res 2022:1–14. doi.org/10.1016/j.renene.2019.03.058. [6] Ahmed Z, Asghar MM, Malik MN, Nawaz K. Moving towards a sustainable [25] Chien F, Hsu CC, Ozturk I, Sharif A, Sadiq M. The role of renewable energy and environment: the dynamic linkage between natural resources, human capital, urbanization towards greenhouse gas emission in top Asian countries: Evidence urbanization, econmic growth, and ecological footprint in China. Resour Policy from advance panel estimations. Renewable Energy 2022;186:207–16. 2020;67:101677. [26] Dkhili H. Investigating the theory of environmental Kuznets curve (EKC) in MENA [7] Al-Mulali U, Saboori B, Ozturk I. Investigating the environmental Kuznets curve countries. J Knowledge Econ 2022:1–18. hypothesis in Vietnam. Energy Pol 2015;76:123–31. https://doi.org/10.1016/j. [27] Danish R, Ulucak SU, Khan D. Determinants of the ecological footprint: role of enpol.2014.11.019. renewable energy, natural resources, and urbanization. Sustain Cities Soc 2019;54. [8] Alola AA, Bekun FV, Sarkodie SA. Dynamic impact of trade policy, economic https://doi.org/10.1016/j.scs.2019.101996, 101996. growth, fertility rate, renewable and non-renewable energy consumption on [28] Destek MA, Okumus I. Does pollution haven hypothesis hold in newly ecological footprint in Europe. Sci Total Environ 2019;685:702–9. https://doi.org/ industrialized countries? Evidence from ecological footprint. Environ Sci Pollut 10.1016/j.scitotenv.2019.05.139. Control Ser 2019;26(23):23689–95. https://doi.org/10.1007/s11356-019-05614- [9] Alshubiri F, Elheddad M. Foreign finance, economic growth and CO2 emissions z. Nexus in OECD countries, Int. J. Clim. Chang. Strat. Manag, (2019) doi:10.1108/ [29] Destek MA, Sarkodie SA. Investigation of environmental Kuznets curve for IJCCSM- 12-2018-0082. ecological footprint: The role of energy and financial development. Sci Total [10] Appiah M, Gyamfi BA, Adebayo TS, Bekun FV. Do financial development, foreign Environ 2019;650:2483–9. https://doi.org/10.1016/j.scitotenv.2018.10.017. direct investment, and economic growth enhance industrial development? Fresh [30] Dogal E, Seker F, Bulbul S. Investigating the impacts of energy consumption, real evidence from Sub-Sahara African countries. Portuguese Econ J 2022:1–25. GDP, tourism and trade on CO2 emissions by accounting for cross-sectional [11] Baloch MA, Danish, Qiu Y. Does energy innovation play a role in achieving dependence: a panel study of OECD countries. Curr Issues Tourism 2017;20(16): sustainable development goals in BRICS countries? Environ Technol 2022;43(15): 1701–19. https://doi.org/10.1080/13683500.2015.1119103. 2290–9. [31] Dogan D, Saboori BC, M.. Does economic complexity matter for environmental [12] Balsalobre-Lorente D, Gokmenoglu KK, Taspinar N, Cantos-Cantos JM. An degradation? An empirical analysis for different stages of development. Environ Sci approach to the pollution haven and pollution halo hypotheses in MINT countries, Pollut Res 2019;26(31):31900–12. https://doi.org/10.1007/s11356-019-06333-1. Environ. Sci. Pollut. Control Ser. 26(22) (2019) 23010-23026, doi:10.1007/s11356- [32] Dong K, Sun R, Hochman G. Do natural gas and renewable energy consumption 019- 05446-x. lead to less CO2 emission? Empirical evidence from a panel of BRICS countries. [13] Balsalobre-Lorente D, Ibáñez-Luzón L, Usman M, Shahbaz M. The environmental Energy 2017;141:1466–78. Kuznets curve, based on the economic complexity, and the pollution haven [33] Dong Z, Chen W, Wang S. Emission reduction target, complexity and industrial hypothesis in PIIGS countries. Renewable Energy 2021. performance. J Environ Manage 2020;260:110148. [14] Bamidele R, Ozturk I, Gyamfi BA, Bekun FV. Tourism-induced pollution emission [34] Dumitrescu EI, Hurlin C. Testing for Granger non-causality in heterogeneous amidst energy mix: evidence from Nigeria. Environ Sci Pollut Res 2022;29(13): panels. Econ Model 2012;29(4):1450–60. 19752–61. [35] Eberhardt M, Bond S. Cross-section dependence in nonstationary panel models: a [15] Bekun FV, Alola AA, Sarkodie SA. Toward a sustainable environment: Nexus novel estimator. Munich Personal RePEc Archive 2009. http://mpra.ub.uni between CO2 emissions, resource rent, renewable and nonrenewable energy in 16- -muenchen.de/17692/. EU countries. Sci Total Environ 2019;657:1023–9. [36] Eberhardt M, Teal F. Productivity analysis in global manufacturing production. [16] Bekun FV, Alola AA, Gyamfi BA, Ampomah AB. The environmental aspects of Discussion Paper 515, Department of Economics, University of Oxford. (2010) http conventional and clean energy policy in sub-Saharan Africa: is N-shaped ://www.economics.ox.ac.uk/research/WP/pdf/paper515.pdf. hypothesis valid? Environ Sci Pollut Res 2021;28(47):66695–708. [37] Elahi E, Khalid Z, Zhang Z. Understanding farmers’ intention and willingness to [17] Bekun FV, Gyamfi BA, Onifade ST, Agboola MO. Beyond the environmental install renewable energy technology: A solution to reduce the environmental Kuznets Curve in E7 economies: Accounting for the combined impacts of emissions of agriculture. Appl Energy 2022;309:118459. institutional quality and renewables. J Cleaner Prod 2021;127924. [38] Elheddad M, Alfar AJ, Haloub R, Sharma N, Gomes P. The impact of foreign direct [18] Bilgili F, Khan M, Awan A. Is there a gender dimension of the environmental investment (FDI) on renewable and non-renewable energy in Bangladesh: does the Kuznets curve? Evidence from Asian countries. Environ Dev Sustainability 2022: global climate change emergencies required? International Journal of Emergency 1–32. Services (2022). [19] Bolük G, Mert M. Fossil & renewable energy consumption, GHGs (greenhouse [39] Farhani S, Shahbaz M. What role of renewable and non-renewable electricity gases) and economic growth: evidence from a panel of EU (European Union) consumption and output is needed to initially mitigate CO2 emissions in MENA countries. Energy 2014;74:439–46. https://doi.org/10.1016/j. region? Renew Sustain Energy Rev 2014;40:80–90. https://doi.org/10.1016/j. energy.2014.07.008. rser.2014.07.170. 10 D.Q. Agozie et al. S u s t a i n a b l e E n e r g y T e c h n o l o g i e s a n d A s s e s s m e n ts 53 (2022) 102597 [40] Grossman GM, Krueger AB. Environmental Impacts of a North American Free E7 economies: are human capital and urbanization essential components? Resour Trade Agreement (No. W3914), (1991) National Bureau of economic research, http Policy 2021;74:102435. s://www.nber.org/papers/w3914. [58] Pata UK. Renewable energy consumption, urbanization, financial development, [41] Gyamfi BA. Consumption-based carbon emission and foreign direct investment in income, and CO2 emissions in Turkey: testing EKC hypothesis with structural oil- producing Sub-Sahara African countries: the role of natural resources and breaks, J. Clean. Prod. 187 (2018) 770–779, doi:10.1016/j.jclepro.2018.03.236. urbanization. Environ Sci Pollut Res 2022;29(9):13154–66. [59] Poumanyvong P, Kaneko S. Does urbanization lead to less energy use and lower [42] Gyamfi BA, Adebayo TS, Bekun FV, Agyekum EB, Kumar NM, Alhelou HH, et al. CO2 emissions? A cross-country analysis. Ecol Econ 2010;70(2):434–44. https:// Beyond environmental Kuznets curve and policy implications to promote doi.org/10.1016/j.ecolecon.2010.09.029. sustainable development in Mediterranean. Energy Rep 2021;7:6119–29. [60] Sarkodie SA, Strezov V. Effect of foreign direct investments, economic [43] Gyamfi BA, Adedoyin FF, Bein MA, Bekun FV. Environmental implications of N- development and energy consumption on greenhouse gas emissions in developing shaped environmental Kuznets curve for E7 countries. Environ Sci Pollut Res 2021; countries. Sci Total Environ 2019;646:862–71. 28(25):33072–82. [61] Shahzad U, Fareed Z, Shahzad F, Shahzad K. Investigating the nexus between [44] Gyamfi BA, Adedoyin FF, Bein MA, Bekun FV, Agozie DQ. The anthropogenic economic complexity, energy consumption and ecological footprint for the United consequences of energy consumption in E7 economies: juxtaposing roles of States: New insights from quantile methods. J Cleaner Prod 2021;279:123806. renewable, coal, nuclear, oil and gas energy: evidence from panel quantile method. [62] Sinha A, Shahbaz M. Estimation of environmental Kuznets curve for CO2 emission: J Cleaner Prod 2021;295:126373. role of renewable energy generation in India. Renewable Energy 2018;119:703–11. [45] Gyamfi BA, Bein MA, Udemba EN, Bekun FV. Investigating the pollution haven [63] Solarin SA, Al-Mulali U. Influence of foreign direct investment on indicators of hypothesis in oil and non-oil sub-Saharan Africa countries: Evidence from quantile environmental degradation. Environ Sci Pollut Control Ser 2018;25(25):24845–59. regression technique. Resour Policy 2021;73:102119. https://doi.org/10.1007/s11356-018-2562-5. [46] Hidalgo CA, Hausmann R. The building blocks of economic complexity, Proc. Natl. [64] Steve YS, Murad AB, Gyamfi BA, Bekun FV, Uzuner G. Renewable energy Acad. Sci. Unit. States Am. 106(26) (2009) 10570e10575, doi:10.1073/ consumption a panacea for sustainable economic growth: panel causality analysis pnas.0900943106. for African blocs. Int J Green Energy 2022;19(8):847–56. [47] Kuznets S. Economic growth and income inequality. Am Econ Rev 1955;45(1): [65] Swart J, Brinkmann L. Economic complexity and the environment: evidence from 1–28. https://www.jstor.org/stable/1811581. Brazil, in: Universities and Sustainable Communities: Meeting the Goals of the [48] Le HP, Ozturk I. The impacts of globalization, financial development, government Agenda 2030, Springer, Cham, 2020, pp. 3e45, doi:10.1007/978-3-030-30306-8_1. expenditures, and institutional quality on CO2 emissions in the presence of [66] Trinh VQ, Nguyen ATQ, Vo XV. Export quality upgrading and environmental environmental Kuznets curve. Environ Sci Pollut Control Ser 2020;27. https://doi. sustainability: Evidence from the East Asia and Pacific Region. Res Int Bus Finance org/10.1007/s11356-020- 08812-2. 2022;60:101632. [49] Lin B, Du Z. How China׳ s urbanization impacts transport energy consumption in [67] Udemba EN. Mediation of foreign direct investment and agriculture towards the face of income disparity. Renew Sustain Energy Rev 2015;52:1693–701. ecological footprint: a shift from single perspective to a more inclusive perspective [50] Liobikiene G, Butkus M. Scale, composition, and technique effects through which for India. Environ Sci Pollut Res 2020;27(21):26817–34. the economic growth, foreign direct investment, urbanization, and trade affect [68] Usman M, Hammar N. Dynamic relationship between technological innovations, greenhouse gas emissions. Renew Energy 2019;132:1310–22. https://doi.org/ financial development, renewable energy, and ecological footprint: fresh insights 10.1016/j.renene.2018.09.032. based on the STIRPAT model for Asia Pacific Economic Cooperation countries. [51] McGee JA, York R. Asymmetric relationship of urbanization and CO2 emissions in Environ Sci Pollut Control Ser 2021;28(12):15519–36. https://doi.org/10.1007/ less developed countries. PLoS ONE 2018;13(12). https://doi.org/10.1371/ s11356-020-11640-z. journal.pone.0208388. [69] Usman M, Kousar R, Yaseen MR, Makhdum MSA. An empirical nexus between [52] Nathaniel S, Nwodo O, Adediran A, Sharma G, Shah M, Adeleye N. Ecological economic growth, energy utilization, trade policy, and ecological footprint: a footprint, urbanization, and energy consumption in South Africa: including the continent- wise comparison in upper-middle-income countries. Environ Sci Pollut excluded. Environ Sci Pollut Res 2019;26(26):27168–79. Control Ser 2020;27(31). https://doi.org/10.1007/s11356-020-09772-3. [53] Nguyen KH, Kakinaka M. Renewable energy consumption, carbon emissions, and [70] Varjani S, Shahbeig H, Popat K, Patel Z, Vyas S, Shah AV, et al. Sustainable development stages: some evidence from panel cointegration analysis. Renew management of municipal solid waste through waste-to-energy technologies. Energy 2019;132:1049–57. https://doi.org/10.1016/j.renene.2018.08.069. Bioresour Technol 2022;127247. [54] Murshed M, Nurmakhanova M, Al-Tal R, Mahmood H, Elheddad M, Ahmed R. Can [71] Westerlund J. Testing for error correction in panel data. Oxford Bull Econ Stat intra-regional trade, renewable energy use, foreign direct investments, and 2007;69(6):709–48. economic growth mitigate ecological footprints in South Asia? Energy Sources, [72] Wu R, Geng Y, Liu W. Trends of natural resource footprints in the BRIC (Brazil, Part B: Econ, Plann, Policy 2022:1–26. Russia, India and China) countries. J Cleaner Prod 2017;142:775–82. [55] Ohajionu UC, Gyamfi BA, Haseki MI, Bekun FV. Assessing the linkage between [73] Zhu H, Duan L, Guo Y, Yu K. The effects of FDI, economic growth and energy energy consumption, financial development, tourism and environment: evidence consumption on carbon emissions in ASEAN-5: evidence from panel quantile from method of moments quantile regression. Environ Sci Pollut Res 2022;29(20): regression, Econ. Modell. 58 (2016) 237–248, doi:10.1016/j. 30004–18. econmod.2016.05.003. [56] Onafowora OA, Owoye O. Bounds testing approach to analysis of the environment [74] Zoundi Z. CO2 emissions, renewable energy and the Environmental Kuznets Curve, Kuznets curve hypothesis, Energy Econ. 44 (2014) 47e62, doi:10.1016/j. a panel cointegration approach. Renew Sust Energ Rev 2016;72. https://doi.org/ eneco.2014.03.025. 10.1016/j.rser.2016.10.018. [57] Onifade ST, Gyamfi BA, Haouas I, Bekun FV. Re-examining the roles of economic globalization and natural resources consequences on environmental degradation in 11