Towards climate action and UN sustainable development goals in BRICS economies: do export diversification, fiscal decentralisation and environmental innovation matter?

ABSTRACT

Several nations across the world place a high focus on achieving carbon reduction objectives. Climate change is the most catastrophic result of human activity. Eco-innovation, export diversification, and fiscal decentralization are all viable approaches for resolving environmental concerns and achieving environmental sustainability goals. These tactics could help countries and levels of government pursue what they consider to be sustainable development. This research assesses the combined impact of export diversification, green technical innovation, and fiscal decentralization in order to accomplish the environmental sustainability goals of the BRICS countries from 1970 to 2020. The long-run dynamic equilibrium between the chosen variables is explored using the augmented mean group (AMG) approach. The results show that while the use of green technology and renewable energy improves the environment, ecological harm is aggravated by export diversification, fiscal decentralization, and economic growth. The BRICS nations should exercise caution while implementing export diversification and fiscal decentralization programs.

KEYWORDS:

• Environmental quality
• Eco-innovation
• Renewable energy
• Fiscal decentralisation
• Export diversification
• BRICS economies

1. Introduction

Protecting the environment and ecosystem has gained importance over the past few decades and is now a major topic of discussion in the fields of politics, social issues, and the economy (Sharma et al. 2021; Udeagha and Breitenbach 2023a, 2023b). Increased carbon dioxide (CO₂) emissions have occurred because of increased energy use by the industrial and consumer sectors. The usage of non-renewable energy sources is one of the main causes of CO₂ emissions. The greenhouse effect and global warming have been brought on by these emanations. Rainstorms, extreme heat, typhoons, agriculture damage, floods, heat damage, water deficit, and an increase in oxygen concentration are all examples of how CO₂ emissions have negatively impacted weather events. By their detrimental impacts on livestock, crops, and natural resources in both industrialised and emerging countries, excessive greenhouse gas emissions have hampered sustainable development (Sharma and Kautish 2021; Udeagha and Muchapondwa 2022a).

As a solution to environmental problems, some experts suggest that innovation in green and sustainable technologies is one of the most effective ways to decrease CO₂ emissions (Udeagha and Muchapondwa 2023a, b; Sharma and Kautish 2020a, b). Innovation is what drives economic growth in both prosperous and emerging economies. The extreme climate change increases public understanding of low-carbon development, and there is a pressing need for economic growth that produces low carbon emissions through cutting-edge technologies. As is commonly accepted, the fact that economic expansion mostly relies on fossil fuel energy is the reason why it raises CO₂ emissions. Technology advancement may aid in the transition from fossil fuel energy to sustainable energy sources, which can further support economic growth while reducing negative impacts on the environment. The Environmental Kuznets Curve (EKC) in economic growth demonstrates how the usage of fossil fuels may lead to environmental contamination, and how this can result in an inverted U-shape curve (Sharma and Kautish 2019; Udeagha and Breitenbach 2023e) However, the most recent empirical evidence demonstrates that technology advancement-driven economic expansion may reduce CO₂ emissions. Researchers have also shown that innovation can increase CO₂ emissions. Our investigation in this work is also driven by the explosion of research on the connection between CO₂ emissions and environmental innovation.

Fiscal decentralisation has gained popularity in recent years as a means of ensuring sustainable development in both industrialised and developing nations while also addressing environmental issues (Cheng et al. 2021, Liu et al., 2022; Li et al. 2021). Following the ‘race to the top approach’ of implementing environmentally friendly practices, it is seen as an important tool for enhancing environmental quality. However, it is connected to ‘race to the bottom’ strategies, when local governments purposefully relax investment rules to draw in international firms that exacerbate environmental degradation, which cannot be ignored. Experts from different nations have investigated the impact of fiscal decentralisation on CO₂ emissions, and all of their findings can be divided into three categories: those that claim fiscal decentralisation can increase CO₂ emissions (Xu 2022), those that affirm fiscal decentralisation can reduce CO₂ emissions (Xiao-Sheng et al. 2022), and those that purport the impact of fiscal decentralisation on CO₂ emissions is uncertain (Xia et al. 2021). Therefore, it is apparent that there is still a lack of agreement among researchers about the impact of fiscal decentralisation on CO₂ emissions, which is what drove us to carry out this research.

The need for higher living standards drives economies to develop at rapid rates by shifting from an agricultural to an industrialised economy, which has a negative effect on environmental quality (Udeagha and Ngepah 2022b, c; Sharma et al. 2018). Both local and foreign economic considerations may have an impact on environmental quality. In order to reduce their systems’ susceptibility to the macroeconomic shocks that affect open economies, both emerging and industrialised economies employ a variety of approaches and programs on the influence of foreign trade in shaping sustainable environment. Accordingly, the International Monetary Fund (Laeven and Valencia 2020) and the World Bank (2019) encourage industrialised and emerging economies to implement export diversification and trade openness policies in order to reduce reliance on certain categories of export industries and ensure long-term income. By increasing market volume and expanding the range of internationally traded products and services, opening-up and diversification strategies would, on the one hand, enhance the economic prospects of the nations; yet, on the other hand, they may have a negative impact on environmental sustainability. The fundamental cause is that a nation’s export growth is dependent on industrial activity, which necessitates aggressive usage of conventional energy sources and, unless the proportion of renewables in the energy structure is improved, results in pollutant emissions like carbon dioxide (CO₂), which worsen environmental quality. Both industrialised and emerging economies might see a variety of effects from such measures since industrialised nations export finished products predominantly, whilst emerging economies sell primary commodities. This situation provides opportunities to pose some questions. For instance, are export diversification and opening up beneficial for the long-term sustainability of global economies’ ecological processes?

The BRICS region is selected for this study for the following reasons. First, the region is the largest of the middle-income countries and, as a group, account for more than a fifth of the world economy. Moreover, the BRICS countries have made significant progress in economic growth in recent decades. According to New Development Bank (2017), in 2016, the group’s combined economic output rose to about 22% of global GDP, compared with 11% in 2005. As of today, the BRICS combined GDP (based on Purchasing Power Parity) is greater than the G7. Thus, along with rapid industrialisation, BRICS countries continue to be an important driving force of the world economy. Second, due to fast economic progress in the BRICS countries and considering their high population, their increase in energy consumption is inevitable. The source of about 40% of the world’s energy consumption is the BRICS countries, and a large part of the global CO₂ emission from this consumption are their responsibility.

Third, in 2012 alone, the BRICS states were responsible for nearly 40.6% of global GHGs (World Bank 2021). Khattak et al. (2022) reported that these economies were among the top seven CO₂-emanating states in 2014, attributing to high industrial production, fossil fuel use, and exports in these economies. As seen in Figure 1, although the BRICS countries emitted less CO₂ emissions (27.66%) than the middle-income (38.09%) and upper-middle-income (30.51%) economies, these countries emitted more emissions than low income (0.36%), lower-middle-income (7.59%), and higher-income (23.48%) economies. It is essential to understand the sources of CO₂ emissions across the BRICS states because most human caused-CO₂ emissions come from the burning of fossil fuels, e.g. hydrocarbon gas liquids, natural gas, petroleum, and coal. As seen in Figure 2, CO₂ emissions from the consumption of gaseous, liquid, and solid fuels in Russia (49.93%), Brazil (63.99$), and South Africa (84.66%) were higher than other member states in 2016, respectively. Solid and liquid fuels appear to be the primary sources of CO₂ emissions in the BRICS economies. Lastly, the industrial expansion supported by the liberal trade policies and financial deepening encouraged firms to produce at a large scale for domestic and international markets (UNIDO 2019). As a result, the combined trade share of the BRICS region in the world trade reached 17% in 2013–2014, which in 2004 was merely 9.9%. Particularly, the exports from BRICS nations registered an annual growth of 15.9%, whereas the annual export growth rate across all nations increased by 8.8% during 2003–2004 to 2013–2014 (Export Import Bank of India 2014). The escalation of exports unveiled new opportunities for BRICS nations, and these nations witnessed an increased demand for both traditional and new exportable items, which are named as intensive and extensive export margins, respectively in the literature (Shahzad et al. 2021). BRICS countries show higher merchandise exports (Figure 3) compared to the merchandise import. China has the highest contribution towards merchandise export and import, followed by the Russian Federation and Brazil. Whereas for the service import and export, China remains a crucial player, followed by India. Figure 3 endorsed that the bloc of BRICS counties is primarily a merchandise export-oriented trade bloc. Therefore, their trade extensively relies on high energy resources, thus causing high CO₂ emissions.

Figure 1. CO2 emissions in the BRICS nations and other regions (% of global emissions).

Source: World Bank (2017)

Figure 2. CO2 emissions by different sources in BRICS countries (%) in 2016.

SFC solid-fuel consumption, LFC liquid-fuel consumption, GFC gaseous fuel consumption.

Figure 3. Total Merchandise and service trade of BRICS countries.

Source: UNCTAD, country-wise trade profile https://unctadstat.unctad.org/CountryProfile/GeneralProfile/en-GB/076/index.html.

Source: World Bank (2017)

The linkages of fiscal decentralisation and export diversification with CO₂ emissions attained great traction from worldwide researchers. To this end, several earlier studies examined the relationship between fiscal decentralisation and CO₂ emissions (Su et al. 2021; Du and Sun 2021; Jain et al. 2021; Xu 2022). Similarly, a number of those works explored the impact of export diversification on CO₂ emissions (Li et al. 2021; Saboori et al. 2022; Zafar et al. 2022). However, the combined role of export diversification, fiscal decentralisation and environmental innovation on CO₂ emissions reduction has been largely neglected. The research gaps that have been found in previous literature are as follows: First, no research has been done in the BRICS economies to jointly examine how fiscal decentralisation, export diversification, and environmental innovation affect CO₂ emissions, and clarify the exact processes through which these associations operate. Second, there is disagreement among previous studies on the relationship between these factors. This is mostly because few studies have utilised the relevant model for this link, such as the endogenous growth model, and instead have utilised a range of models to study the relationships. Third, a number of research has been conducted to identify the main factors contributing to environmental deterioration. But unlike international trade, economic growth, and renewable energy, the political structure of a country has an indirect influence on environmental quality that is difficult to evaluate. For this reason, studies typically disregard that. In this study, we focus on the BRICS economies to address the related questions below: How may fiscal decentralisation contribute to improve environmental quality? What connection exists between export diversification and CO₂ emissions? How may environmental innovation bring about environmental sustainability? These are all linked questions that require in-depth examination of a sizeable sample over a long period of time. When formulating strategies to attain environmental sustainability, it is crucial to comprehend the connections between export diversification, fiscal decentralisation, environmental innovation, and environmental quality. Fourth, past studies ignored a lot of concerns by focusing on the links between export diversification, trade openness, renewable energy, and CO₂ in parts. In contrast to previous studies, this work examines how the BRICS nations might achieve carbon neutrality target through the interrelated roles of export diversification, fiscal decentralisation, and environmental-related technological innovation. In order to enhance the energy-growth-environment nexus for the BRICS nations, this study adds export diversification for the first time to the body of current literature. Given the following factors, such inclusion is crucial: (i) Increased export diversification broadens the range of exportable goods and the pool of trading partners, which has an impact on overall trade and, consequently, trade openness. (ii) Export diversification increases industrial energy consumption, but it also fundamentally depends on the structure of industrial energy. In this sense, a greater proportion of renewable electricity production in the development of exportable goods might alter the relationship between export diversification and the environment.

This study places high importance on addressing the above-mentioned gaps in light of the broad prominence of earlier endeavours and the several important areas of knowledge that are ignored. The main contributions of this paper are summarised as follows. First, the selection of BRICS nations is hinged on the fact that this group has been central to the global debates and negotiations on environmental sustainability. A primary reason for this growing influence is that many firms in the developed economies have moved and transferred their production units and technologies to BRICS economies due to strict environmental regulation in developed economies and low production costs in emerging markets. As a result of such industrial migration, the levels of CO₂ emissions have increased manifolds in these five economies such that the BRICS nations’ CO₂ emissions are more than other developing economies. For these reasons, this study aims to provide valuable policy implications to the region’s emissions reduction targets, and at the same time, it offers a significant lesson for the global outlook. Second, the association between environmental quality and some determinants such as export diversification, fiscal decentralisation, and environment-related technological innovation has gotten a lot of attention from scholars all around the world. Previous research looked into the link between environmental quality and fiscal decentralisation (Khan et al. 2020). A few of those studies also investigated the influence of export diversification on environmental quality (Wang et al. 2020). However, the combined impact of fiscal decentralisation, export diversification, and environment-related technological innovation on environmental quality has been generally overlooked. Our study also intends to fill this void by looking at the influence of export diversification, fiscal decentralisation, and environment-related technological innovation on environmental quality in the BRICS nations from 1970 to 2020, in the context of GDP, and renewable energy use. Modelling fiscal decentralisation, export diversification, and environment-related technological innovation simultaneously increases our knowledge of how they interact with environmental quality, and this is an important topic for industrialised countries like BRICS nations dealing with environmental issues. In specific language, this research adds to the body of knowledge by simultaneously considering the environmental effects of fiscal decentralisation, export diversification, and environment-related technological innovation. Third, prior empirical research ignored the critical issues of slope heterogeneity and cross-sectional dependence in panel data, resulting in inconsistencies in estimation results. This research uses newly developed econometric approaches that can handle both issues; thereby considering the sophisticated panel data techniques that are robust to the mentioned econometric challenges and addressing critical information gaps. Lastly, since the BRICS countries are working hard to find a sustainable solution to improve the quality of their environment and have made substantial progress towards achieving ecological sustainability in the region, the study intends to offer useful policy recommendations on the importance of fiscal decentralisation, export diversification, and environment-related technological innovation in fostering environmental quality. Moreover, the region has made commitments to cut CO₂ emissions in half by 2030 and reach net-zero emissions by 2050. The region has been continuously investing in R&D in order to meet such lofty goals. From 2016 to 2017, the BRICS countries’ R&D intensity went from 2.34% to 2.37%, indicating that they are increasingly investing in R&D. Furthermore, their real R&D spending increased by 3.8%. More than 70% of all BRICS nations were represented by commercial companies. Similarly, government R&D expenditure was 1.3%, whereas private R&D expenditure was over 28% (BRICS, 2019). As a result, this research is very essential for the BRICS countries under consideration. Thus, our study will serve as a policy document for BRICS government officials and other blocs in the drive for green energy targets amidst economic growth paths.

The rest of the research is structured as follows: Section 2 reviews previous research. Data, sources, and methods are presented in Section 3. Section 4 presents empirical findings and discussion. Section 5 concludes the work and presents policy recommendations.

2. Literature review

This portion is divided into three parts: (i) the relationship between environmental innovation and CO₂ emissions; (ii) the relationship between fiscal decentralisation and CO₂ emissions; (iii) the relationship between export diversification and CO₂ emissions.

2.1. Environmental innovation-CO₂ emissions relationship

Many earlier studies that attempted to quantify the environmental effects of green innovation focused on the relationship between ecologically friendly technical innovation (ERTI) and CO₂ emissions. Improvements in ERTI could minimise CO₂ emissions without lowering the quality of the product (Requate and Unold 2003; Udeagha and Ngepah 2019) because they can assist business owners in lowering their energy needs and environmental problem (Santra 2017). Therefore, to improve the efficiency of energy investment products, businesses and government institutions are making significant investments in ERTI (Ahmad et al., 2019; 2020). The relationship between innovation and CO₂ emissions has been the subject of countless investigations by researchers from various nations. Using pooled-regression modelling, Santra (2017) concluded that ERTI helps businesses reduce CO₂ emissions in the BRICS economies. Santra (2017)’s key focus was the utilisation of CO₂ efficiency as determined by energy-related CO₂ productive capacity, or production per unit of CO₂ emissions. By utilising CO₂ efficiency as determined by energy-related CO₂ yield, or output per unit of CO₂ emissions, Santra (2017) significantly contributed to the growing literature. The research does not, however, examine the combined effects of export diversification, environmentally technological innovation, and fiscal decentralisation on the environment. Likewise, Ahmad and Zheng (2021), who focused on the cyclic and asymmetric consequences of innovation in green technologies, made a substantial addition to the research. They found a direct, positive correlation between CO₂ emissions during the recession and negative shocks to the development of green technologies. During a period of economic recovery, CO₂ emissions are decreased by positive shocks to environmental technology innovation. There was a countercyclical relationship between innovation shocks in environmentally friendly technologies and CO₂ emissions throughout business cycles. Also, Chen et al. (2020) highlighted the beneficial role of ERTI in CO₂ emissions mitigation more recently. The work by Chen et al. (2020) expands our understanding of how environmentally friendly and technological breakthroughs impact China’s transportation-related CO₂ emissions. ERTI helps to minimise CO₂ emissions in the case of the G−6 economies, according to Dauda et al. (2019), who utilised fully modified ordinary least squares (FMOLS) technique. By examining how innovation aids in cutting emissions at diverse geographical settings, Dauda et al. (2019) enriched the existing literature.

In a different work done for China, Hao et al. (2020) made the case that advancement of technology can reduce emissions of smoke dust and sulphur dioxide while raising contaminant oxygen intake. By using wavelet statistical methods, Adebayo and Kirikkaleli (2021) provided a fresh viewpoint on the relationship between CO₂ emissions and GDP growth, renewable energy, technological innovation, and globalisation in Japan. The study used datasets from the 1990 Q1 through the 2015 Q4 quarters using a variety of wavelet methods. Their empirical findings provided evidence of the relationship – in both time and frequency the usage of renewable energy, economic expansion, technical advancement, globalisation, and CO₂ emissions. The authors discovered that while the use of renewable energy reduces CO₂ emissions in the short and medium term, globalisation, GDP growth, and technical innovation all raise CO₂ emissions in Japan. Their conclusion highlights how important it is to put policies into place that are well-coordinated by the decision-makers in order to stop the serious environmental deterioration in Japan. Similarly, Khan et al. (2020) investigated the importance of a conceivable carbon emissions quantitative assessment for designing a suitable climate policy to deal with global pollution. Long-term consumption-based carbon emissions rose due to income and imports.

Furthermore, innovations in green technology are seen as efficient means of achieving a balance between economic development and environmental sustainability. However, there is still a dearth of empirical data, particularly in developing nations, on the relationship between green technological development and CO₂ emissions. Lin and Ma (2022) investigated how the urban innovation environment affected the impact of green technology advances on CO₂ emissions using panel data on 264 prefecture-level Chinese cities from 2006 to 2017. The empirical findings showed that changes in green technology have a variety of effects in various kinds of cities. While the impact is minimal in Chinese cities prior to 2010, green technological development can help reduce CO₂ emissions after 2010. Secondly, through improving industrial structure, green technological improvement can indirectly lower CO₂ emissions. Thirdly, government spending cannot considerably affect the marginal effect of green technology when the urban innovation environment is considered. The minor reduction in CO₂ emissions caused by advances in green technology, however, is only noticeable if a city’s human capital level reaches a particular point. Cities with higher levels of human capital have a greater impact on reducing carbon emissions. Their findings offered crucial illumination for realising the coherence and unity of the economic transformation to innovation-driven, environmentally friendly, and low-carbon development. The study by Lin and Ma (2022) focused on how the urban innovation environment affected the impact of green technology advances on CO₂ emissions. Similar to this, Abid et al. (2022) examined the impacts of technological development, economic growth, foreign direct investment, energy use, and urbanisation on carbon emission in G8 member countries using data from 1990 to 2019. Their results indicated a significant cross-sectional dependency among the panel nations. The FMLOS estimator identified a statistically significant long-run and negative association between foreign direct investment, financial development, and technical innovation in G8 countries. Long-term bidirectional causal linkages have been discovered to exist between economic growth, financial development, urbanisation, trade openness, CO₂ emissions, and energy consumption; however, there is only a one-way causal relationship between carbon emissions and foreign direct investment.

2.2. Fiscal decentralisation-CO₂ emissions relationship

Various researchers have tried to investigate the connection between fiscal decentralisation and carbon emissions, but without promising outcomes (Batterbury and Fernando 2006; Fell and Kaffine 2014). According to the first body of literature, fiscal decentralisation is one of the elements contributing to an increase in pollutant emissions. Fiscal decentralisation in China, according to Song et al. (2018), prioritises the distribution of government revenues while ignoring the expenditure side of government. In order to receive fiscal transfers from national authorities for development policies, the municipal authorities in China adopted a negative approach towards environmental practices and environmental expenses, according to the study. As a result, the major emphasis of state spending is shifted towards the infrastructural development. Sigman (2014) emphasised that because of growing fiscal decentralisation, pollution emissions are associated with negative externalities and government free-riding lowers environmental quality. The study by Fell and Kaffine (2014) also emphasised the unfavourable externality of environmental contamination. Due to regional political rivalry, municipal authorities exhibit protective tendencies and engage in free riding, which increases the amount of pollutant emissions. The Chinese government has begun to take the attribute of environmental governance into account when evaluating municipalities, according to Sun et al. (2020). Nevertheless, because of variances in the degrees of regional economic development, municipalities’ practices connected to environmental governance vary (Tian and Wang 2021). In advanced economies, ecological management is seen as a municipal strategic advantage; as a result, they steadily raise their environmental requirements, expand their environmental contribution, and encourage the environmental climate’s improvement. As a result, in advanced economies, ecological products are given consideration for varied competitiveness (Konisky 2007; Anser et al. 2020). In exchange for industrial transmissions from affluent nations and in expectation of greater transfers for environmental legislation from the federal government, emerging markets decrease their environmental requirements. Consequently, economic progress in poor countries is viewed as a ‘race to the bottom’ (Liu et al. 2020). Thus, under a system where finances are decentralised, municipal authorities prioritise implementing budget-friendly initiatives and environmental safeguards. Different kinds of environmental governance, such as ‘diversified rivalry’ and ‘bottom-by-bottom competition,’ arise as a consequence of growing economic development gaps (Zhang et al. 2017).

According to the second body of literature, fiscal decentralisation lowers pollutant emissions. The second thread is based on the idea of ‘race to the top,’ which contends that more fiscal decentralisation lowers CO₂ emissions while simultaneously improving environmental quality and enabling resource allocation that is done effectively (Millimet 2003). Additionally, the nimbyism effect spurs local governments to tighten environmental laws (Levinson 2003). Du and Sun (2021) claimed that the national authorities actively support fiscal decentralisation because the provincial governments should achieve targeted economic growth and environmental preservation has policy repercussions. Khan et al. (2021) examined the impact of fiscal decentralisation on CO₂ emissions in the BRICS nations. They discovered that positive fiscal decentralisation shocks led to decreases in environmental degradation. Su et al. (2021) looked at the interactions between political risk, fiscal decentralisation, R&D spending, and the use of renewable energy in the BRICS nations. They discovered that fiscal decentralisation decreased non-renewable energy consumption while increasing the use of renewable energy.

2.3. Export diversification-CO₂ emissions nexus

Manufacturers often need a large amount of energy and depend largely on fossil fuel energy use, which causes ecological threats. Regarding this, Barrows and Ollivier (2021) stated that increasing global demand causes local production to increase and export rates to rise, which in turn increases CO₂ emissions. Hence, export diversification is defined as ‘broadening the variety of items that a nation sells’ in order to maintain steady economic development while maintaining trade openness (Dennis and Shepherd 2011). Based on theoretical and empirical considerations, the relationship between export diversification and environmental quality may be categorised. Theoretically, nations with weak trade openness regulations are more likely to experience two comparative advantage-related consequences of environmental pollution, namely (i) factor endowment and (ii) pollution policy. Due to the factor endowment, high-income countries with plenty of capital gain from exporting noxious commodities (Mania and Rieber 2019; Ahmed et al. 2021). Numerous researchers agree with this viewpoint, which states that production and trading items must be developed in accordance with their adaptable strength. It is also compatible with the Heckscher-Ohline doctrine (Gozgor and Can 2016; Ahmed and Le 2021). Mania (2020) notes that export diversification reduces environmental quality over time by raising CO₂ emissions.

On the empirical front, the following categories might be used to summarise the numerous investigations that have thoroughly explored the link between export diversification and CO₂ emissions. According to the first group of research, export diversification reduced CO₂. For instance, in a panel of 22 top remittance-receiving nations over the period 1986–2017, Zafar et al. (2022) investigated the association between remittances, export diversification, education, and CO₂ emissions while adjusting for renewable energy and economic development. The study used generalised quantile regression, Westerlund and Edgerton cointegration approach with structural breaks, Cup-FM and CUP-BC long-run estimation approaches, and second-generation unit root techniques in its econometric methodology. The results showed that remittances, which have a negative impact on emissions, assisted in mitigating environmental deterioration. The diversification of exports also helped to lower CO₂ emissions, and so did the use of renewable energy. Saboori et al. (2022) found that export diversification has lessened the environmental degradation in Oman. Energy was shown to have a positive influence on both environmental indices, albeit the impact on CO₂ emissions was less significant. In Oman, authors suggested that it is highly advised to diversify exports by focusing on high-tech, environmentally friendly, and energy-efficient enterprises. In a group of developed countries, Apergis et al. (2018) showed that the export structure of developed countries reduced CO₂ emissions. In a related study, Liu et al. (2018) demonstrated how diversifying export items lowered emissions in China, Japan, and Korea. Similar to this, Liu et al. (2019) demonstrated that diversifying export markets and products aided in emission reduction for a panel of 125 nations. Also, Hu et al. (2020) found that import product diversification only decreases emissions for groups of rich countries, whereas it increases CO₂ emissions in poor countries. Comparatively, Li et al. (2021) found that export diversification reduces China’s emissions levels. According to Sharma et al. (2021), BRICS countries’ air quality improves when export product variety is reduced.

The second category of study found that export diversification increased CO₂ emissions. For instance, using panel data from 37 OECD nations and an AMG estimator, Iqbal et al. (2021) assessed the impact of export diversification and fiscal decentralisation on carbon emissions. They revealed that when export diversification increased in the investigated nations, carbon emissions increased as well. In recent literature, researchers have tried to investigate the compositional impact of international trade on environmental degradation by utilising the variety and/or concentration of export and import products as a proxy. In a case study involving Turkey between the years of 1971 and 2010, Gozgor and Can (2016) looked at the link between export diversification and CO₂ emissions. Their results show that export diversification increases emissions. The major focus of Gozgor and Can (2016) was to estimate the short- and long-term environmental Kuznets curve for Turkey from 1971 to 2010. In 84 developing countries, Can et al. (2020) found that export diversification, intensive, and extensive margins increase CO₂. Can et al. (2021) showed in a case study for 10 newly industrialised countries that diversifying export products increases energy and CO₂ emissions. Similarly, Wang et al. (2020) demonstrated that the export mix of these nation groupings boosted emissions in a case study for the G−7 countries. In a similar vein, Mania (2020) found that while export product concentration reduces emissions in the developed country group, export diversification increases emissions in poor countries. While Mania (2020) centred on the impact of export diversification on CO₂ emissions in the framework of an environmental Kuznets curve theory, Wang et al. (2020) concentrated on the contribution of eco-innovation and export diversification in supporting environmental sustainability. In a different research, Khan et al. (2021) used a case study of countries in the Regional Comprehensive Economic Partnership to assess the effects of export diversification and the composite risk index on CO₂ from 1987 to 2017. According to the research, export diversification is a significant factor that raises emissions.

Several researchers now use the Theil index to gauge the variety of export products. Being a concentration index, the Theil index is noteworthy. Its lower original values (before conversion) imply a varied export basket, whilst its higher values denote a concentrated export mix. In more specific terms, this index’s lower values correspond to more diversification, and vice versa. International trade literature states that whereas rich nations are in a stage of concentration, emerging countries are in the process of diversification. According to this theoretical perspective, the Theil index’s initial value’s negative sign represents the beneficial effect of ‘export diversification’ on the dependent variable (vice versa). Theil index’s initial value should be regarded as having a negative sign for developed nations since this indicates that ‘export concentration’ has a negative influence on the dependent variables (and vice versa) (Sheikh and Hassan 2021). Additionally, as both nation groups are at different levels of diversification and concentration, emerging and developed countries should be divided for the panel research; otherwise, the interpretations of the results and suggested policies would be contradictory (Can et al. 2021). Regarding the link between export diversification and environmental quality in this context, there are problems with interpretation. For example, from 1986 to 2017, Zafar et al. (2022) examined the effect of export diversification on CO₂ in a panel of 22 top remittance-receiving nations, encompassing both developed and developing countries. Their findings show that diversifying exports reduces CO₂. However, in their analysis, the country group included certain industrialised nations like Belgium, France, Germany, Italy, and the USA that are at a stage of concentration and create complex goods. Additionally, their sample included developing nations like Mexico, Nigeria, and Lebanon that are in the process of diversifying their economies. This suggests that they chose countries with different phases of export concentration and diversification for single research, and in that case, it would be challenging to determine the overall effect of export diversification on CO₂.

Similar to this, Shahzad et al. (2021) investigated how export product diversification affected CO₂ emissions in both industrialised and developing nations. They interpreted the Theil index’s negative value as a decline in CO₂ for both sets of countries. Theil index’s negative value, however, is thought to show the growing effect of export diversification on CO₂ for developing nations. In contrast, the fact that this indicator is negative for developed nations suggests that export concentration reduces CO₂. Iqbal et al. (2021) looked at the effect of export diversification on CO₂ in 37 OECD countries in different research. They concluded that diversifying exports increases CO₂. However, both developed and developing countries are included in the OECD. Therefore, it is challenging to comprehend the consequences of export diversification without taking sub-panels in terms of development levels as development level might affect the impact of export diversification or concentration. The effect of export product diversification on CO₂ in China from 1989 to 2019 was examined by Li et al. (2021). Since they did not employ the inverse version of this index, the findings indicated a negative coefficient for the Theil index, which suggests that diversification has a growing influence on CO₂. Once more, the researchers anchored their policy recommendations on the Theil index’s negative sign and urged export diversification to help achieve sustainability goals. Since greater value of the Theil index indicates less diversification and vice versa, this kind of conclusions and the accompanying policy considerations run counter to the theoretical argument, as was previously established.

In conclusion, there is a lot of controversy in the earlier literature. The majority of earlier research focused on time series data and concluded that export diversification had both positive and negative effects on environmental deterioration. To avoid interpretation issues, the Theil index was not transformed into its inverse form in many earlier investigations. The literature review also reveals that the majority of research explore the relationship between export diversification and the environment by employing an ecological footprint as a stand-in for environmental damage. In the context of the BRICS countries, which have significant economies and large populations, there are no empirical research which jointly looked at how fiscal decentralisation, export diversification, and environment-related technological innovation affected CO₂ emissions in the BRICS countries between 1970 and 2020. For improved results interpretation, the Theil index was also transformed into its inverted version.

3. Theoretical underpinning, data, and methods

Figure 4 shows the methodological roadmap of the research.

Figure 4. Methodological roadmap of the study.

3.1. Theoretical underpinning

The conceptual foundation of this investigation is discussed in this part. Race to the top hypothesis, race to the bottom principle, and ecological modernisation theory all serve as the conceptual backbone for this research (EMT). One way that the race to the top hypothesis relates to local governments’ ability to control ecologically risky firms and compel them outside of their purview is through a decentralised fiscal framework. Given that different areas have unique local hurdles, the regional authorities may promote a race to the top among the local firms on this ground. This is because they have the informational resources to adequately tackle local problems (Udeagha and Muchapondwa 2022b). By improving environmental health and ecological balance, fiscal decentralisation (FDST) is predicted to lower CO₂ emissions in this situation, i.e. $\left({\delta }_1=\frac{\alpha C{O}_2}{\alpha FDST}<0\right)$. The race to the bottom argument, on the other hand, contends that fiscal decentralisation damages environmental health since municipal authorities are induced to implement unduly permissive environmental laws to draw mobile investments. Municipal authorities are under pressure to reduce pollution standards in a ‘race to the bottom’ as they strive for competition and rapid industrialisation with other municipalities and regions that possess more ‘investment-friendly’ standards. Therefore, it is projected that FDST will result in higher CO₂ emissions, i.e. $\left({\delta }_1=\frac{\alpha C{O}_2}{\alpha FDST}<0\right)$.

By splitting export diversification into export product and export market diversifications, the relationship between export diversification and environmental damage might be further explored. Vertical and horizontal diversifications are the two types of export product diversification. A broader ‘basket’ and more emissions are the results of horizontal export diversification, which promotes a greater variety of exporting routes with many different goods. As opposed to horizontal diversification, vertical diversification incorporates innovative thinking and quality while also reducing emissions through the ‘learning by doing impact’ and encouraging the technology effect. Additionally, as a nation broadens its export opportunities, it trades with several nations, necessitating the export of bulky goods and requiring the use of higher emission inputs like transportation and energy derived from fossil fuels. Therefore, if export diversification (EDI) is not environmentally favourable, it is projected that CO₂ emissions would grow, i.e.$\left({\delta }_2=\frac{\alpha C{O}_2}{\alpha EDI}>0\right)$; otherwise, i.e. $\left({\delta }_2=\frac{\alpha C{O}_2}{\alpha EDI}<0\right)$ if it is eco-friendly.

Net exports, public expenditure, and investment are just a few of the economic factors that the gross domestic product (GDP) tracks (Udeagha and Ngepah 2021). Domestic expenditure accounts for a sizeable component of GDP. Therefore, increased residential usage may result in a significant rise in CO₂ emissions. Therefore, it becomes sense to infer that as the BRICS economies’ income increase, emissions have been imported through trade and consumption. Additionally, it is thought that GDP-related demand shifts will increase CO₂ emissions. The EKC hypothesis states that as output rose because resources based on fossil fuels were more readily available, pollution levels in the nation would initially rise. However, if a nation achieves a certain level of development, income is predicted to have an effect on lowering CO₂ emissions. As a result, it is anticipated that GDP will cause CO₂ emissions to rise, i.e. $\left({\delta }_3=\frac{\alpha C{O}_2}{\alpha GDP}>0\right)$.

The ecological modernization theory (EMT) suggests that increasing resource efficiency (renewable energy) through technological innovation might aid in reducing the environmental risks brought on by economic expansion. We establish relationships between the development of technology and CO₂ emissions in light of the aforementioned assumptions (Udeagha and Ngepah 2022a). Through technological advancement, energy usage can be reduced, energy efficiency can be raised, and the environment can be enhanced. Advanced technologies must be used in order to promote sustainable manufacturing and enhance environmental sustainability. This study looks at how technology advancement affects ecological integrity since it has the potential to make renewable energy sources like solar and wind less wasteful and more ecologically friendly. Modern energy-intensive industrial machinery is being replaced with greener, more efficient alternatives, which eases the burden on the environment and the economy. As a result, environmental technology innovation (ERTI) is anticipated to reduce CO₂ emissions, i.e. $\left({\delta }_4=\frac{\alpha C{O}_2}{\alpha ERTI}<0\right)$.

The use of environmentally friendly technological advancements and renewable energy sources is anticipated to cut CO₂ emissions in a variety of ways. First off, the use of renewable energy may help developing nations diversify their energy sources, reduce costs, and offer a more dependable energy supply. Second, by diversifying and ensuring a consistent supply of energy, government agencies may enhance energy security, lessen reliance on foreign oil, and minimize gasoline leaks (Udeagha and Ngepah 2020). Since renewable energy either doesn’t generate any carbon pollution at all or produces very little of it, it is expected that utilizing it will cut CO₂ emissions. As a result, it is anticipated that adopting renewable energy will lower CO₂ emissions, i.e. $\left({\delta }_5=\frac{\alpha C{O}_2}{\alpha REC}<0\right)$.

3.2. Data and description of variables

This research includes panel data from the BRICS nations (Brazil, Russia, India, China, and South Africa) from 1970 to 2020. This research considers both the fiscal decentralisation index (FDST) and the ratio of revenue/expenditure to general government revenue/expenditure. Su et al. (2021) claim that the construction of the index is crucial because it addresses the decentralisation of revenue (Rd) and expenditure (Ed):

(1) $R_d=\frac{Thebudgetaryexpenditureofthelocalgovernment}{Thebudgetaryexpenditureofthecentralgovernment}$(1)

(2) $E_d=\frac{Taxrevenueofthelocalgovernment}{Taxrevenueofthecentralgovernment}$(2)

The statements that follow provide an explanation of the data that was obtained for our key variables. Fiscal decentralisation index (FDST) and export diversification index (EDI) are taken from the International Monetary Fund (IMF) database. Data on renewable energy consumption (REC) measured in % of total energy consumption, ERTI (environmental-related technological innovation (ERTI) as a percentage of overall technological innovation, GDP in constant (base = 2010) US dollars, carbon dioxide (CO₂) emissions in metric tons are all provided by the World Bank World Development Indicators.

3.3. Model construction

Given the theoretical underpinning linking environmental quality to each of the aforementioned factors, this study makes use of the following basic framework:

(3) $C{O}_{2it}=f\left(FDS{T}_{it},ED{I}_{it},GD{P}_{it,}ERT{I}_{it},RE{C}_{it}\right)$(3)

Equation (3)(3) $C{O}_{2it}=f\left(FDS{T}_{it},ED{I}_{it},GD{P}_{it,}ERT{I}_{it},RE{C}_{it}\right)$(3) is changed to the linear connection shown below:

(4) $C{O}_{2it}={\delta }_0+{\delta }_{1}FDS{T}_{it}+{\delta }_{2}ED{I}_{it}+{\delta }_{3}GD{P}_{it}+{\delta }_{4}ERT{I}_{it}+{\delta }_{5}RE{C}_{it}+{\epsilon }_{it}$(4)

Where i stands for cross-sectional dimension (BRICS nations), and ‘t’ denotes the panel data’s time horizon, spanning from 1970 to 2020. The ${\delta }_s$ stands for estimates to be computed, and coefficients, and ${\epsilon }_{it}$ represents for stochastic error term.

Fiscal decentralisation (FDST) is anticipated by Equation (4)(4) $C{O}_{2it}={\delta }_0+{\delta }_{1}FDS{T}_{it}+{\delta }_{2}ED{I}_{it}+{\delta }_{3}GD{P}_{it}+{\delta }_{4}ERT{I}_{it}+{\delta }_{5}RE{C}_{it}+{\epsilon }_{it}$(4) to reduce pollution in the ‘race to the top’ pathway, that is, ${\delta }_1<0$, and conversely is projected for the ‘race to the bottom’ approach. Contingent to the kind of export diversification under consideration, the export diversification (EDI) is likely to be connected positively $({\delta }_2>0)$ or negatively (${\delta }_2>0$) to pollution. The influence of GDP on pollution may hence be either beneficial $({\delta }_3>0)$ or negative$({\delta }_3<0)$. Environmental technology innovation (ERTI) is anticipated to have a negative $({\delta }_4<0)$ influence on pollution if it is ecologically advantageous and safe; but, it may even increase pollution, that is$({\delta }_4>0)$. Lastly, the use of renewable energy (REC) is anticipated to have a negative effect $({\delta }_5<0)$ on pollution by displacing fossil fuel-based sources of energy.

3.4. Econometric estimation methods

3.4.1. Slope coefficients homogeneity and cross-section dependence test

Before the fulfilment of the econometric investigation in this work, various procedures have to be taken (Ahmad et al., 2019). First, there are always interdependence effects across regions within a territory as a result of shared impacts or integration of the political system and economic growth. Therefore, when employing panel data, cross-section dependence (CSD) should be accounted for to prevent estimation prejudice. In order to determine if the variable series has the CSD issue, this work employs the CSD-test suggested by Pesaran (2007). Additionally, three tests are performed to determine whether CSD is present in panel data models with fixed or random effects. Second, to confirm the existence of the heterogeneous slopes in the model, this research employs a slope homogeneity (SCH) test that was developed by Pesaran and Yamagata (2008) that is robust to the CSD. As a result, the following are the exact requirements for SCH:

(5) ${\tilde{\mathrm{\Delta }}}_{SCH}={\left(N\right)}^{\frac{1}{2}}{\left(2k\right)}^{-\frac{1}{2}}\left(\frac{1}{S}\tilde{S}-k\right)$(5)

(6) ${\tilde{\mathrm{\Delta }}}_{ASCH}={\left(N\right)}^{\frac{1}{2}}{\left(\frac{2k\left(T-k-1\right)}{T+1}\right)}^{-\frac{1}{2}}\left(\frac{1}{S}\tilde{S}-k\right)$(6)

Absolute and adjusted slope coefficient homogeneity are denoted as ${\tilde{\mathrm{\Delta }}}_{SCH}$ and ${\tilde{\mathrm{\Delta }}}_{ASCH}$, respectively. In the context of different shocks that lead to global interconnection, such as the 2007–2008 financial crisis, the 1997 Asian financial crisis, the 2008 COVID−19 pandemic, and the oil price crisis, Pesaran (2004)‘s test is employed to seek for CSD. As contrast to Levin et al. (2002) and Im et al (2001)‘s first-generation approaches, it also incorporates Pesaran (2007)‘s unit root testing methodology. Pesaran (2007) suggests the cross-section augmentation IMPS (CIPS) test, which yields reliable results even in the presence of cross-section dependence and different slope coefficients. The generalised CIPS test procedure is thus presented as follows:

(7) $\mathrm{\Delta }{Y}_{it}={\pi }_i+{\pi }_i{Y}_{it-1}+{\pi }_i{\bar{X}}_{t-1}+\sum _{\mathrm{t}=0}^P{\pi }_{it}\mathrm{\Delta }{\bar{Y}}_{t-1}+\sum _{\mathrm{I}=}^P{\pi }_{it}\mathrm{\Delta }{Y}_{it-\mathrm{I}}+{\epsilon }_{it}$(7)

First-differenced terms and lag averages are indicated in Equation (7)(7) $\mathrm{\Delta }{Y}_{it}={\pi }_i+{\pi }_i{Y}_{it-1}+{\pi }_i{\bar{X}}_{t-1}+\sum _{\mathrm{t}=0}^P{\pi }_{it}\mathrm{\Delta }{\bar{Y}}_{t-1}+\sum _{\mathrm{I}=}^P{\pi }_{it}\mathrm{\Delta }{Y}_{it-\mathrm{I}}+{\epsilon }_{it}$(7) . The CIPS statistics for the examination are as follows:

(8) $\widehat{CSIMPS}=\frac{1}{N}\sum _{i=1}^{n}CAD{F}_i$(8)

Cross-sectionally augmented Dickey-Fuller (CADF) is used with Equation 8(8) $\widehat{CSIMPS}=\frac{1}{N}\sum _{i=1}^{n}CAD{F}_i$(8) to represent the Equation (7)(7) $\mathrm{\Delta }{Y}_{it}={\pi }_i+{\pi }_i{Y}_{it-1}+{\pi }_i{\bar{X}}_{t-1}+\sum _{\mathrm{t}=0}^P{\pi }_{it}\mathrm{\Delta }{\bar{Y}}_{t-1}+\sum _{\mathrm{I}=}^P{\pi }_{it}\mathrm{\Delta }{Y}_{it-\mathrm{I}}+{\epsilon }_{it}$(7) . While ‘stationarity’ is promoted by the alternative hypothesis, ‘non-stationarity’ is supported by the null hypothesis.

3.4.2. Westerlund co-integration test (WCT)

After confirming the existence of non-stationarity, the issue of spurious regression is investigated using two distinct panel co-integration tests. First, Pedroni divided the statistics into group dimension and panel dimension in econometric technique, extending the residual-based approach for a panel data scenario. Second, when CSD is present across nations and across factors, the Westerlund-based panel co-integration strategy has additional strength. The basic concept is to test for the lack of co-integration by assessing if error correction exists for each panel member individually or for the entire panel. The weighted average of the individually calculated error corrections and their t-ratios, respectively, is the starting point for the $G_{\alpha }$and $G_t$ test statistics. Hence, rejection of H₀ should be seen as proof that at least one of the cross-sectional units is co-integrated. Rejection of H₀ should be taken as confirmation of co-integration for the panel as a whole because the $P_{\alpha }$ and $P_t$ test statistics pool data across all cross-sectional units. An error-correcting testing approach is used to determine if the variables in the investigation have long-term cointegration (Westerlund 2007). This approach is more dependable than those used by Kao et al. (1999) and Pedroni (2015) for cointegration testing, which are appropriate when errors are cross-sectionally independent and slope coefficients are different (Khan et al. 2020). For this kind of test, the guiding equation is the following:

(9) $G_t=\frac{1}{N}\sum _{i=1}^N\frac{{\stackrel{\mathrm{\prime }}{\alpha }}_i}{SE({\stackrel{\mathrm{\prime }}{\alpha }}_i)}$(9)

(10) $G_{\alpha }=\frac{1}{N}\sum _{i=1}^N\frac{T{\stackrel{\mathrm{\prime }}{\alpha }}_i}{{\stackrel{\mathrm{\prime }}{\alpha }}_i\left(1\right)}$(10)

(11) $P_t=\sum _{i=1}^N\frac{\stackrel{\mathrm{\prime }}{\alpha }}{SE\left(\stackrel{\mathrm{\prime }}{\alpha }\right)}$(11)

(12) $P_{\alpha }=T\stackrel{\text{bprime}}{\alpha }$(12)

3.4.3. The augmented mean group (AMG)

The impacts of the factors highlighted on CO₂ emissions are investigated using Bond and Eberhardt (2013)’s AMG estimator. Heterogeneity and CSD, the two main problems with panel data, are addressed by the AMG (Can et al. 2020). Additionally, it is a helpful tool that may display values of parameters that are country-specific. The Pesaran (2006)’s Common Correlated Effect Mean Group approach is substituted by the AMG framework, whose fundamental functioning model is described as follows:

(13) $\mathrm{\Delta }{Y}_{it}={\pi }_{1i}+{\delta }_i\mathrm{\Delta }{Z}_{it}+{\theta }_{i}UG{F}_i+\sum _{t=2}^T{\sigma }_t{D}_t+{\epsilon }_{it}$(13)

The AMG econometric approach is attained by first obtaining the projected value of ${\tilde{\delta }}_i$ from Equation (13)(13) $\mathrm{\Delta }{Y}_{it}={\pi }_{1i}+{\delta }_i\mathrm{\Delta }{Z}_{it}+{\theta }_{i}UG{F}_i+\sum _{t=2}^T{\sigma }_t{D}_t+{\epsilon }_{it}$(13) , which is specified as:

(14) $\widetilde{AMG}=\frac{1}{N}\sum _{i=1}^N{\tilde{\delta }}_i$(14)

3.4.4. Robustness check

The slope coefficients of interest in panel data models are believed to be uniform across separate groups in many scholarly investigations. The SURE (seemingly unrelated regression equation) framework of Zellner (1962) may be used to test the slope homogeneity hypothesis when the cross-section dimension (N) is relatively small and the time series dimension of the panel (T) large. This approach is especially appealing since it immediately addresses the potential for cross- sectional error correlations and dynamics when N is sufficiently small, and T is high enough. Because T is substantially greater than N in our research, we employed the SURE framework to check for robustness and to compare our findings with those of Pesaran and Yamagata (2008). In accordance with this analysis, we anticipate the probability of some efficiency in the event of contemporaneously linked disturbances involving several regressors across the testing equation.

3.4.4.1. Testing cross-equation restrictions (the hypothesis of slope homogeneity) after sure estimation

The SURE framework, which is particularly appealing since it also automatically handles with the potential for cross-sectional error correlations and dynamics, is used to test the slope homogeneity hypothesis. We use the procedure indicated below to test the slope of homogeneity hypothesis. The essential null hypothesis might be expressed in terms of the coefficients in Equation (4)(4) $C{O}_{2it}={\delta }_0+{\delta }_{1}FDS{T}_{it}+{\delta }_{2}ED{I}_{it}+{\delta }_{3}GD{P}_{it}+{\delta }_{4}ERT{I}_{it}+{\delta }_{5}RE{C}_{it}+{\epsilon }_{it}$(4) as follows:

(15) $H_0:A_2=B_2=C_2=D_2=E_2$(15)

refers to ${\delta }_1$ in Equation (3)(3) $C{O}_{2it}=f\left(FDS{T}_{it},ED{I}_{it},GD{P}_{it,}ERT{I}_{it},RE{C}_{it}\right)$(3) for $i=m_1,m_2,m_3,m_4,m_5$

Against the alternative hypothesis:

(16) $H_1:A_2\ne B_2\ne C_2\ne D_2\ne E_2$(16)

refers to ${\delta }_1$ in Equation (3)(3) $C{O}_{2it}=f\left(FDS{T}_{it},ED{I}_{it},GD{P}_{it,}ERT{I}_{it},RE{C}_{it}\right)$(3) for $i=m_1,m_2,m_3,m_4,m_5$

Since Equation (15)(15) $H_0:A_2=B_2=C_2=D_2=E_2$(15) holds only when $H_0$ is true, we reject $H_0$ if the log-likelihood ratio (LR) statistic for testing these restrictions is well above the 95% critical value of the chi-squared distribution.

4. Empirical results and their discussions

4.1. Summary statistics

The summary statistics is provided in Table 1, which includes the mean, median, maximum, and minimum values of the model’s variables used in the analysis. Fiscal decentralisation (FDST) is second, with a mean value of 7.706, behind gross domestic product (GDP), which has the highest mean value of 8.405. It is worth noting that EDI has the highest standard deviation (1.6538) and CO₂ emissions have the lowest (0.120). EDI is the variable with the largest volatility, according to the statistical evidence, while CO₂ emissions have the lowest volatility. Table 1 summarises the relevant information and characterises the peaks using kurtosis, while the Jarque-Bera diagnostic test confirms the normality of our data. Our data series follow the normal distribution predicted by Jarque-Bera statistics.

Table 1. Descriptive statistics.

Variables	Mean	Median	Maximum	Minimum	Std. Dev	Skewness	Kurtosis	J-B Stat	Probability
CO2	0.264	0.275	0.458	0.091	0.153	0.253	1.616	4.671	0.187
FDST	7.706	7.972	8.963	6.004	0.395	0.514	2.102	4.131	0.130
EDI	4.316	3.703	3.784	2.417	0.848	0.352	2.041	3.442	0.172
GDP	8.405	7.863	9.683	4.893	0.195	0.465	1.448	3.161	0.112
ERTI	5.360	4.203	6.590	2.225	0.771	0.091	1.618	4.403	0.127
REC	6.060	6.537	5.687	2.761	1.305	0.648	2.003	5.746	0.132

Source: Authors’ calculations.

4.2. Slope coefficients homogeneity and cross-section independence test results

Table 2 provides evidence that CSD exists. As a result, CSD is taken into account in this study’s following analysis.

Table 2. SCH and CSD testing methods.

$\mathrm{\Delta }$	p-Values	${\mathrm{\Delta }}^{Adjusted}$	p-Values
−7.931***	0.000	−7.014***	0.000
Ho: Cross-sections are independent
Variables	CD-statistic	p-Value	Mean absolute p
CO2	185.619***	0.000	0.96
FDST	188.710***	0.000	1.00
EDI	190.615***	0.000	1.00
GDP	191.836***	0.000	0.98
ERTI	190.740***	0.000	0.95
REC	195.851***	0.000	1.00

Note: 1%, 5%, and 10% significance levels are captured by the asterisks ***,**,* respectively.

4.3. Order of integration of the respective variables

The stationarity findings are displayed in Table 3. Every variable is I (1). The need to examine for cointegration is therefore justified. Furthermore, we complement the panel unit root test by using a battery of other unit root tests for robustness check. For this purpose, we have used other approaches such as LLC (Levin, Lin & Chu, 1992, 2002) panel unit root, IPS (Im, Pesaran & Shin, 1997, 2003) panel unit root, HT (Harris & Tzavalis, 1999) panel unit root, ADF-Fisher panel unit root, and Breitung panel unit root. The findings shown in Tables 3 and A1 (see Appendix) are in agreement. Both findings demonstrate that any series that is not stationary in levels after the first difference becomes stationary at the first difference in all experiments. The similarity of the outcomes among various panel unit root tests therefore attests to the robustness of the results. The data series appears to be devoid of I(2) variables, according to the evidence.

Table 3. Panel unit root test results.

Variables	Level I(0)		The first difference I(1)
	Intercept	Intercept with trend	Intercept	Intercept with trend
CO2	1.205	1.271	2.851***	2.904***
FDST	1.672	1.504	2.701***	2.624***
EDI	1.164	1.206	2.635***	2.841***
GDP	1.742	1.710	2.724***	2.915***
ERTI	−2.411	−2.583	−5.921***	−5.980***
REC	1.304	1.721	2.732***	2.814***

Note: 1%, 5%, and 10% significance levels are captured by the asterisks ***,**,* respectively.

4.4. Cointegration test results

The chosen variables’ long-term relationships are demonstrated by Table 4. At 1% level, G_t and P_t are both significant. As a result, we cannot reject the possibility that the variables are related in a cointegrating manner. We used the AMG estimator in the context of cointegration to learn more about the possible interactions between the variables (see Table 5).

Table 4. Westerlund ECM panel cointegration testing method.

Statistic	Value	Z-statistic	p-Value
$G_t$	−5.815	−26.015	0.000***
$G_{\alpha }$	−16.825	−5.014	0.000***
$P_t$	−27.416	−14.620	0.000***
$P_{\alpha }$	−16.061	−8.264	0.000***

Note: 1% significance level is captured by the asterisk ***.

Table 5. Augmented mean group (AMG) results.

Variables	Coefficients	Std. Err.	p-Value
FDST	0.104***	0.000183	0.000
EDI	0.021***	0.000014	0.000
GDP	0.310***	0.00084	0.000
ERTI	−0.025***	0.00271	0.000
REC	−0.150**	0.00057	0.016
Intercept	0.201*	0.004183	0.081

Note: 1%, 5%, and 10% significance levels are captured by the asterisks ***,**,* respectively.

4.5. The augmented mean group (AMG) results and their discussions

The outcomes from the AMG estimate approach are shown in Table 5. It is discovered that the estimated fiscal decentralisation coefficient (FDST) is statistically significant and positive over the long run. This data indicates that fiscal decentralisation has a long-term, positive, and large impact on CO₂ emissions. In the BRICS countries, it degrades the environment and raises CO₂ emissions. At a 1% significant level, a 1% rise in fiscal decentralisation is linked to a 0.104% increase in CO₂ emissions. We discovered evidence of the ‘Race to Bottom’ idea among the BRICS nations, where governments favour the reduction of environmental regulatory standards, leading to a loss in environmental sustainability. Fiscal decentralisation reduces environmental quality because the BRICS nations are compelled to have incredibly lenient environmental standards in order to draw mobile capital. Due to rivalry for industrial growth, the BRICS governments are under pressure to slash environmental rules in a ‘race to the bottom.’ This is done to stop potential businesses from moving to neighbouring countries with more ‘investment-friendly’ legislation. The BRICS nations, however, have been influential in global talks and accords on environmental sustainability. A significant contributor to this growing impact is the fact that many companies in developed economies have moved their manufacturing facilities and technological capabilities to the BRICS nations as a result of strict environmental regulations in developed economies and low production costs in developing markets. Due to this industrial migration, the CO₂ emission levels of these five economies have increased dramatically, placing the BRICS countries’ CO₂ emissions above those of other developing economies. Furthermore, regional variations in environmental legislation enhance the chance of polluters acting opportunistically when local governments in BRICS countries scale back environmental oversight in the sake of quick economic growth. Consequently, the difficulty of emission control is increased by the changing of pollutant sources. In addition, local economic development and developmental phases have an impact on regional fiscal capacity in the BRICS nations. The investment variation in regional environmental governance can be attributed, in part, to regional differences in fiscal capability. Environmentally damaging industries will migrate to areas with lax regulations as a result of the government’s differentiation of environmental investment, and the absence of pollution eradication would further limit their capacity to restore contaminated areas. Fiscal decentralisation makes it harder for the BRICS nations to enforce regulations. Due to fiscal decentralisation, local governments have reduced environmental regulations in an effort to achieve rapid economic growth, which has considerably increased CO₂ emissions. Our empirical findings are consistent with those of Xiao-Sheng et al. (2022), who showed how fiscal decentralisation in Chinese cities has a detrimental impact on the environment. Fiscal decentralisation significantly increases CO₂ emissions both inside and outside of the area, according to research by Xia et al. (2021), who drew a similar conclusion. Furthermore, Lin and Zhou (2021) discovered that fiscal decentralisation had a detrimental impact on environmental performance in China’s economically developed and eastern areas. Similarly, Iqbal et al. (2021), who examined the roles of fiscal decentralisation, environmental innovation, and export diversification in achieving the carbon reduction objective for 37 OECD countries, discovered that fiscal decentralisation and export diversification reduce environmental quality while environmental innovation helps to increase environmental sustainability. The results, however, go against those of Xu (2022), who looked at the connection between fiscal decentralisation and environmental management effectiveness in China and found that it significantly promotes environmental governance and efficiency, which lowers the rate of environmental deterioration in the nation. The main cause of this positive driving influence is local governments’ support of technology innovation efforts. Decentralising expenses and revenues would provide local governments’ access to information advantages and other aspects, improving the effectiveness of local governance. This empirical observation was supported by Tufail et al. (2021), who showed that fiscal decentralisation improved environmental quality by reducing CO₂ emissions in seven highly fiscally decentralised OECD countries. This empirical evidence also supports Rani (2021)’s results that fiscal decentralization helped Asian nations reduce pollution. In this context, the authors stated that in order to achieve ecological sustainability in the Asian block under study, fiscal decentralization might be expanded through human capital and green innovation. An analogous finding was reached for nine Asian nations by Jain et al. (2021). Fiscal decentralization encourages ‘eco-friendly economic growth’ through local government, which indirectly improves the environment. Environmentally friendly technologies are strengthened in the sector while industrial innovation is encouraged. This result concurs with Du and Sun (2021)’s observation that fiscal decentralisation enhanced ecological integrity via technology advancements with an ecological focus. Su et al. (2021)‘s observation that fiscal decentralisation might speed up trading and investment ventures by enhancing ecological integrity lends more credence to this conclusion.

Export diversification is shown to increase environmental pollution in the BRICS countries, as indicated by the estimated long-run coefficient on export diversification (EDI), which is found to be positive and statistically significant. The strength of the coefficient indicates that a 1% increase in export diversification corresponds to a 0.021% increase in CO₂ emissions. This quantitative result may be explained by the fact that the socioeconomic success of the BRICS countries is based on the specialisation and diversity of their products as well as their trade system. Export diversification may effectively promote sustainable growth while unavoidably contributing to climate change. Additionally, it calls for significant increases in income, marketing, and transportation. Particularly during the 1990s, the BRICS countries have relied on globalisation and industrial development initiatives as stimulants for eradicating their socioeconomic downturns. The governments of the BRICS countries introduced liberal industrial and trade policies in an effort to increase domestic output and, therefore, job possibilities. Governments also tried to foster a favourable environment for potential investors by implementing the digital system for banking and stock market services. As a result, these countries noticed an increase in the number of registered enterprises. The already-existing businesses expanded their capacity and broadened their product lines at the same time. Large-scale production for both local and foreign markets was fostered by the industrial boom backed by liberal trade policies and financial deepening. As a consequence, these five countries’ aggregate trade share in the world trade increased to 17% in 2013–2014 from just 9.9% in 2004. Specifically, the exports from the BRICS countries had an annual rise of 15.9%, while the global annual export growth rate rose by 8.8% from 2003–2004 to 2013–2014 (Iqbal et al. (2021). The expansion of exports opened up new prospects for the BRICS countries, who also saw a surge in demand for both conventional and novel exportable goods, which are referred to in the literature as intensive and extensive export margins, respectively (Shahzad et al. 2021). However, industrial facilities may have boosted their use of energy solutions and other production elements as a result of the expanding demand in domestic and worldwide markets. Given that the BRICS countries mainly depend on fossil fuels for their energy needs, the burning of new energy sources may have increased pollution levels in these countries. Due to the BRICS countries’ promotion of export market diversification, pollution levels in the regional bloc have significantly increased. The BRICS economies in this situation focus on creating new products while keeping several sites for the same exporting stream. Diversification places a strain on exporting routes that are increasingly diverse, leading to a rise in emissions as a result of rising demand and intensive energy consumption. Our results are in line with those of Iqbal et al. (2021), who examined the effects of fiscal decentralisation and export diversification on carbon emissions in the context of 37 OECD countries and discovered that as export diversification grew, carbon emissions also did so. In a similar line, Can et al. (2021) showed that diversifying export items increases energy and CO₂ emissions in a case study for 10 newly industrialised nations. Khan et al. (2021)‘s analysis of the impacts of export diversification and the composite risk index on CO₂ emissions using a case study of nations in the Regional Comprehensive Economic Partnership offers more evidence of the validity of our findings. In a case study including the G−7 nations, Wang et al. (2020) also showed that the export mix increased emissions. Similarly, Mania (2020) discovered that export diversification increased emissions in developing countries whereas export product concentration decreased emissions in the developed nation group. Our results, however, disagree with those of Zafar et al. (2022), who examined the relationship between remittances, export diversification, education, and CO₂ emissions while controlling for renewable energy and economic growth and discovered that export diversification, which has a negative impact on emissions, helped to mitigate environmental deterioration. Similar findings were made by Saboori et al. (2022), who discovered that export diversification had slowed down Oman’s environmental degradation. Additionally, Hu et al. (2020) demonstrated that diversifying import products only reduces emissions for groups of affluent nations while raising CO₂ in poor countries. Comparatively, Li et al. (2021) revealed that diversifying China’s exports lowers emissions. Sharma et al. (2021) asserted that reducing the diversity of export products enhances the air quality in BRICS nations.

The calculated coefficient on environment-related technological innovation (ERTI), which indicates that a 1% increase in ERTI results in a 0.025% decrease in CO₂ emissions at a 1% level of significance, is shown to be statistically significant and negative over the long run. This empirical research implies that the BRICS area, which has enacted a number of legislative measures meant to stimulate environmental-related technical innovation in pursuit of ecological integrity, may gain from green technological innovation. Ecologically sound technical advancements in the BRICS nations encourage efficient energy usage, give people cheaper access to renewable energy sources, and improve environmental quality. Ecological technological progress helps to lower CO₂ emissions in the BRICS nations by increasing energy efficiency through a range of methods, including that of altering the share of renewables, optimising energy-efficient production systems, and using end-of-pipe technologies. More crucially, the key forces for advances in the ecological sustainability of the region are the high levels of R&D spending in the BRICS countries and technological breakthroughs. The region has put in place a number of initiatives to boost government engagement in R&D, allowing it to progressively transition its industrial operations from very energy-intensive coal-powered systems to highly energy-efficient pathways driven by technological developments. Some of these institutional measures that encourage innovativeness have considerably increased the zone’s ability to reduce environmental damage. Our results are in line with those of Khattak and Ahmad (2022a), who discovered that CO₂ emissions were decreased during the boom times as a result of positive shocks to innovation in green and sustainable technologies in OECD economies. Similar findings were made by Khattak et al. (2022), who discovered that positive shocks to innovation in environmentally friendly and sustainable technologies reduce CO₂ emissions during economic booms in G−7 economies. Similar findings have been reported by Xin et al. (2021), who showed that positive shocks in environmental-related technologies during the expansion phase resulted in a drop in CO₂ emissions in the United States from 1990Q1 to 2016Q4. Furthermore, Ahmad and Zheng (2021) showed how CO₂ emissions were reduced in the BRICS economies during the economic upturn between 1990 (Q1) and 2016 (Q4) due to positive shocks to innovation in environmental-related technologies. Also, our findings are consistent with those of Erdogan (2021) and Guo et al. (2021), who discovered that technological advancements foster an environment that encourages a decrease in energy consumption, an increase in energy efficiency, and ultimately a decrease in greenhouse gas emissions for China and the BRICS countries, respectively. These results are further supported by Anser et al. (2021) for EU countries. Our findings, however, are not consistent with those made by Dauda et al. (2021), who contend that the technological progress in Sub-Saharan African nations compromises environmental health. Usman and Hammar (2021) found equivalent results for Asian countries.

It is noted that the computed coefficient on consumption of renewable energy (REC) is statistically significant and negative over the long run. A 1% increase in REC is highly connected with a 0.150% decline in CO₂ emissions at a 1% level of significance in the BRICS economies. This provides an illustration of how employing renewable energy enhances environmental quality. The substantial investments made in renewable energy by the BRICS countries are already paying off. As a consequence, they are currently on the path to achieving the SDGs by 2030. The region uses techniques for reducing pollution to combat global warming, such as an ecologically friendly transportation sector, a diverse energy mix, and a dependable energy infrastructure. The most efficient way to solve energy production and global warming is to promote the growth of renewable energy sources. Development of renewable energy is now a key plan for the future, according to the BRICS countries. The zone is one of the top producers and consumers of energy in the world, and since it plays a key role in the international economy, developing and implementing renewable energy regulations is particularly notable. Our findings are congruent with those of Weimin et al. (2022), who discovered that REC improves ecological quality by reducing CO₂ emissions in developing economies. In a similar vein, from 1990 to 2018, Khattak and Ahmad (2022b) demonstrated that REC reduces CO₂ emissions among the BRICS states. Moreover, You et al. (2022) revealed that REC helps the United States reduce CO₂ emissions over time, from 1990Q1 to 2018Q4. Our findings are further supported by Baye et al. (2021), who found that REC improves green environment in 32 Sub-Saharan African nations. However, the findings conflict with those of Bölük and Mert (2014), who found that REC reduces ecological sustainability in 16 EU member states. Similar findings were made by Pata (2018), who discovered that REC worsens Turkey’s environmental protection.

Given that GDP has a positive long-run coefficient, this means that GDP has a negative influence on CO₂ emissions. At a 1% significance level, a 1% increase in GDP growth causes a 0.310% increase in CO₂ emissions. The findings therefore confirmed the idea that environmental degradation has been a side effect of economic growth in the BRICS countries. The bulk of the BRICS countries are still in the development stage, as demonstrated by this supporting evidence. In the early phases of their development, nations frequently pay little attention to the state of their environment. The objective is to maintain economic growth despite deteriorating environmental conditions. These findings concur with those reached by Udeagha and Breitenbach (2021), Udeagha and Ngepah (2022b), Ahmed et al. (2020), Khan et al. (2020), Nathaniel et al. (2020), for SADC, South Africa, China, Pakistan, and MENA, respectively, although they are different from those of Ulucak and Ulucak (2020) for BRICS nations, Ahmed et al. (2020) for BRICS nations, and Liu et al. (2020) for G7 economies. The zone that was investigated or the estimation methodology used may be the cause of the results’ inconsistencies.

Finally, it is crucial to identify the factors that are causal as well as how they affect the outcome of policies intended to promote economic growth and environmental preservation. The Dumitrescu and Hurlin (2012) approach is used to evaluate the suggested causal relationships between the variables in this study. Table 6 depicts a one-way causal link between CO₂ emissions and REC, ERTI, GDP, EDI, and FDST. It demonstrates that any short-term policy change in the aforementioned variables will have a long-term effect on the pollution levels in the BRICS countries.

Table 6. Dumitrescu – Hurlin causality testing method for heterogeneous panel.

Causality direction	W-statistics	Z-statistics	p-Value
FDST → CO2	4.01528***	12.0863	0.000
EDI → CO2	1.60142***	4.91524	0.000
GDP → CO2	0.09143***	−5.20517	0.000
ERTI → CO2	0.26041***	−4.70146	0.000
REC → CO2	5.81510***	15.16051	0.000

Note: 1%, 5%, and 10% significance levels are captured by the asterisks ***,**,* respectively.

The right-pointing arrow denotes unidirectional causation between independent and dependent variables. There is no such thing as bidirectional causation, or the flow of information from dependent to independent factors.

4.6. Robustness check

The SURE framework is employed as part of a robustness assessment since N is quite small and T is suitably large (i.e. T is substantially greater than N) in our present investigation. The results are shown in Table A2 (see Appendix).¹. We found evidence of little to no change in the estimated coefficients when the findings of the two approaches (AMG and SURE) were compared, notably with regard to their signs and magnitudes. While keeping their signs, the majority of the variables are statistically significant, though in some cases, their magnitudes change slightly from one another. Since Equation (15)(15) $H_0:A_2=B_2=C_2=D_2=E_2$(15) holds only when $H_0$ is true, we reject $H_0$ since the LR statistics for testing these restrictions is 25.152, which is well above the 95% critical value of the chi-squared distribution with 4 degrees of freedom, and we therefore strongly reject the slope of homogeneity hypothesis. Our finding is consistent with the previous results obtained using Pesaran and Yamagata (2008).

5. Conclusion and policy implications

5.1. Conclusion

Policy leaders and environmental groups have been very concerned about environmental sustainability as the globe struggles with a significant global warming catastrophe brought on by an increase in greenhouse gases. Climate action has received attention from the UN’s SDGs. CO₂ emissions have increased because of the BRICS countries’ rapid economic progress. Despite their commitment to keeping global warming to 1.5 degrees Celsius and stepping up efforts to reduce carbon emissions as they endorsed the Glasgow Climate Pact (COP26 agreement) in 2021, the BRICS economies face a number of significant challenges in achieving environmental sustainability and the SDGs. The BRICS countries are key rising markets in terms of geo-economics, fast economic growth, and involvement in global markets. The BRICS economies had significant economic expansion, which led to 13.985 billion tonnes of CO₂ emissions per person, or 41.8% of world CO₂ emissions. South Africa is first and sixth in Africa in terms of coal and lignite consumption, respectively, and second internationally in terms of CO₂ intensity. Many have questioned whether these growing markets are capable of guiding the earth towards a cleaner atmosphere in light of the alarming rises in CO₂ levels in 2007 in Russia (6%), India (5%), China (16%), and Brazil (1.15%). The BRICS nations actually have a lot to overcome before they can reduce CO₂ emissions and achieve sustainable economic development.

In order to better understand the relationships between fiscal decentralisation, export diversification, environmentally technological innovation, and CO₂ emissions in the BRICS nations from 1970 to 2020, it was necessary to account for economic development and renewable energy sources. To produce short- and long-run estimation findings that were resistant to cross-sectional dependency and slope heterogeneity, several advanced econometric approaches were used. The Pesaran (2007) unit root, Pesaran and Yamagata (2008) slope homogeneity, and Pesaran and Yamagata (2008) slope heterogeneity were used to analyse cross-section dependency. Following that, a long-run cointegrating relationship between the variables under investigation was assessed using Westerlund’s ECM panel cointegration from 2008. The long-run studies were carried out using the augmented mean group (AMG) estimate method. The study’s findings demonstrated that while using environmentally friendly technologies and renewable energy sources enhances environmental sustainability, fiscal decentralisation, export diversification, and economic growth have significantly accelerated environmental degradation.

5.2. Policy implications

The results mentioned above are crucial for the national strategy as they illustrated how diverse economic and ecological issues should work together. The following suggestions for policymaking are thus given in accordance with the outcomes of our investigation:

Firstly, according to the empirical evidence, export diversification initiatives are a key predictor of pollutant emissions and environmental degradation. In this perspective, it is essential to increase the inventive, complex, and knowledge-based commodities in the export composition of the BRICS countries because they are typically thought to be less harmful to the environment. The environment will benefit if the BRICS countries include new eco-friendly items in their export portfolio (Udeagha and Ngepah 2023). Therefore, BRICS authorities ought to offer some incentive schemes to the businesses that produce green goods. Additionally, the partnership between the institution and the corporate sector may aid in the growth of new goods with value-added. The BRICS’ policymakers should concentrate on export diversification-related economic plans, supported by proactive environmental regulations. Moreover, from a policy perspective, the BRICS nations’ policymakers and climatologists should consider total product sophistication, systemic transformation, and economic variability as a strategy to transform the development of the region into one that is environmentally friendly (Udeagha and Ngepah 2022d).

Secondly, in a bid to enhance and facilitate sustainable economy, authorities should urgently focus on optimising the ecological influence of green innovations considering that environmentally technological advancement (green innovation) is ecologically friendly in the BRICS countries. Additionally, the BRICS government should fund initiatives like the adoption of green practices and make a concerted effort to amend all relevant laws in order to foster environmental advances and related technologies (Udeagha and Breitenbach 2023b). In order to promote sustainable growth and address ecological and sustainability issues, it is necessary to implement green policies, enhance the environment, and adopt technology-friendly laws. Controlling the risk factors associated with new innovations and technology breakthroughs will be achievable when more and more responsible technical infrastructures and innovation are established with the incorporation of green strategies. It is also essential for authorities to have a set of standards when choosing environmentally friendly standards for technologies that might improve environmental sustainability (Udeagha and Muchapondwa 2023b). To support environmental advancements, a market architecture that enables companies to share cutting-edge technology and advantages while creating considerable synergies must be created (Udeagha and Ngepah 2021). Authorities can also encourage investments in green technologies and green alternative fuels for greener production. This is because export diversification has a negative impact on CO₂ emissions. Additionally, a sophisticated industrial structure’s desirable transition for fewer pollutant emissions is a problem for authorities in the BRICS nations. The empirical results point to the immediate need to reform pollution regulations, encourage ecologically responsible initiatives, lower taxes for particular industries, and increase the capacity of relevant technology to be absorbed, which will boost the formation of green commodities and enterprises (Udeagha and Ngepah 2022e)

Thirdly, the carbon trading scheme should also be put in place at the fundamental base of regional and local governance. Similar to a two-edged sword, this will function as a very appropriate instrument to promote decentralisation of fiscal income while simultaneously promoting environmental integrity and increasing tax revenue (Udeagha and Ngepah 2022c). A crucial shift away from policy frameworks that encourage rapid economic growth with high emissions to those that foster minimal carbon emissions, especially low carbon economic growth strategies to aid sustainable development (Udeagha and Breitenbach 2023d).

Finally, there is no doubting the importance of renewable energy sources in preserving the environment. In an attempt to switch from conventional energy production techniques and improve and enhance the technologies used to create renewable power, it is suggested to expand green development. The promotion of nuclear, biomass, and solar thermal energy solutions needs to get more attention (Udeagha and Breitenbach 2023c).

5.3. Limitations and potential future study areas

Although the current investigation yielded solid empirical results in the case of the BRICS nations, the research includes a number of limitations that might be taken into consideration in a subsequent investigation. One of the main drawbacks of the investigation is the limited availability of the data outside of the event window, which restricts the range of the time series analysis performed. Future studies may focus on figuring out the ideal level of fiscal decentralisation to fulfill the SDGs’ key aim of economic development while also achieving sustainable environmental goals. Future studies may concentrate on other regional economies in development, such as the G−7 or the OECD, using different panel econometric techniques or micro-disaggregated relevant data.

Also, the combined effect of fiscal decentralisation, export diversification, and environment-related technological innovation on pollution are thoroughly examined in this study, which also broadens the possibilities for related future research. Future studies should use a comprehensive environmental metric in place of carbon emissions to provide more pertinent results. Furthermore, considering nonlinear linkages may produce an intriguing synergy among fiscal decentralisation, export diversification, environment-related technological innovation, and environmental quality. The significance of the study can also be increased by including interaction terms between fiscal decentralisation and export diversification. Comparative analysis of outcomes based on investigated components in various global locations may provide a more in-depth understanding of the findings. Future research should also examine the moderating role of green technological innovation in reducing the detrimental effects of export diversification and fiscal decentralisation on ecosystem in developed and developing nations.

Furthermore, additional growth-related elements that were not taken into account in this study, which include institutional quality, natural resources rent, human capital, foreign direct investment, and trade openness, could also be considered by the future research (Udeagha and Breitenbach 2023a). However, CO₂ emissions were utilised in this study as a gauge of ecological harm. Additional research is required to determine whether consumption-based carbon emissions, and other metrics of carbon footprints, such as nitrous oxide, sulphur dioxide, hydrogen, monoxide, surface hazardous gases, airborne pollutants, and other shallow regional climate shocks, are more reliable indicators of environmental destruction in the BRICS countries. The present research uses CO₂ emissions as a representation of habitat destruction even though they are not the primary driver of ecological degradation. Future research should study this connection by using other environmental degradation measurements, such industrial effluents, and hazardous toxins, in the context of the BRICS countries (Ngepah and Udeagha 2018).

Last but not least, future research may compare country-specific results to global panel outcomes utilising even more complex methodologies by merging time series data with panel data techniques. This can help shed light on the current evidence by offering a thorough comparison with the results of this investigation. The investigation’s focus on a single regional grouping is another serious flaw. Therefore, future research in the global panel setup and other parts of the world should focus on how environmental technology innovation, export diversification, and fiscal decentralisation affect CO₂ emissions. The complementarities among these variables in affecting CO₂ emissions in the BRICS nations and a wider number of country groups like the G7 economies could also be examined in the future to expand this analysis and draw more broad conclusions about the intriguing and useful interactions among these variables (Ngepah and Udeagha 2019).

5.4. Contributions

Maxwell Chukwudi Udeagha and Nicholas Ngepah conceptualised the study idea, drafted the paper, collected data, analysed data, wrote the introduction section, organised the literature review, drafted the methodology section, interpreted the results, and provided the discussions, concluded the study with policy implications and organised the reference list.

Availability of data and materials

The data relevant to this research is publicly available from the World Development Indicators or obtained from the authors by making a reasonable request.

Declarations

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This study was not funded by any organisation.

Notes on contributors

Maxwell Chukwudi Udeagha

Maxwell Chukwudi Udeagha holds a PhD in Economics from the University of Johannesburg, South Africa. He is a senior researcher and an editorial board member of several Journals including American Journal of Environmental Sciences and Engineering, Economics Bulletin, Global Economics Science, Journal of Contemporary African Studies, Journal of Economics, and International Business Management. He has research interest in emerging market economies, development economics, international trade, African environment, regional economic integration, applied economics and applied econometrics. He has published widely in a number of reputable journals including Journal of African Business; Journal of Economic Integration; Economic change and Restructuring; International Journal of Urban Sciences; Environmental Science and Pollution Research; African Review of Economics and Finance.

Nicholas Ngepah

Nicholas Ngepah holds a PhD in Economics from the University of Cape Town. He is a Professor of Economics at the University of Johannesburg. He is an expert in quantitative and qualitative research techniques, economic development and policy impact assessments, including spatial econometrics. He has wide range of experience on development issues, with core expertise in poverty, inequality, labour market dynamics and inclusive economic growth with related policies like agriculture, health issues, gender, climate change, trade, industrialisation etc. Professor Ngepah has undertaken studies for organisations like the World Bank, UK Overseas Development institute, African Economic Research Consortium, Council for Scientific and Industrial Research, Oxfam and the South African Government. His most recent works include: re-examining the growth, poverty and inequality relationships in South Africa in the context of covid-19; and climate change and gender inequality in the labour market. His work on the SDGs in Africa formed part of the bases of the policies debated in the context of the first 1000 days of the SDGs implementation under Development Progress. He has undertaken various projects involving large multi-country surveys and data analyses in Southern Africa. In his role as Professor of Economics, he lectures microeconomics and micro econometrics related subjects and supervises PhDs and Masters Research students on various economic development-relevant topics. He has published widely and in high impact journals. He is the Founder of the African Institute for Inclusive Growth, a research and policy think tank focusing on inclusive development.

Notes

1. The SURE framework requires that the time series have to be stationary to run the regression. Since this condition is fulfilled in this analysis, it is necessary to refer to it in Table A2.

Appendices

Table A1. Panel unit root tests.

Variable	LLC	IPS	HT	ADF-Fisher	Breitung	Test results
CO2	−7.6503***	−2.0264**	−1.7533**	−1.0429	−2.0899**	Stationary
D(CO2)	−12.3931	−7.2226***	−0.0231***	−24.2853***	−16.8466***	Stationary
FDST	−6.7368***	−2.4619***	−5.5440***	−1.8027**	−2.3595***	Stationary
D(FDST)	−13.8490***	−8.4085***	−0.1489***	−23.5859***	−20.6677***	Stationary
EDI	−8.6001***	−2.7975***	−7.6625***	−3.7179***	−2.8924***	Stationary
D(EDI)	−17.4283***	−8.8312***	−0.2131***	−21.9637***	−24.2062***	Stationary
GDP	−7.8499**	−2.3727***	−4.4849***	−3.3717***	−4.3268***	Stationary
D(GDP)	−16.8221***	−8.4277***	−0.2141***	−22.7770***	−20.0133***	Stationary
ERTI	−7.6268**	−3.5617	−3.5780	−3.0547***	−2.1853**	Stationary
D(ERTI)	−9.1756***	−5.2815***	−0.1431***	−24.0598***	−14.7119***	Stationary
REC	−2.9714	−0.9664	−0.5526	−1.5829*	−3.2613***	Unit root
D(REC)	−14.8187***	−7.6259***	−0.0892***	−26.1347***	−21.7547***	Stationary

Source: Authors’ calculations.

Note: (i) LLC denotes the Levin, Lin & Chu (1992, 2002) panel unit root; IPS represents the Im, Pesaran & Shin (1997, 2003) panel unit root; HT means the Harris & Tzavalis (1999) panel unit root. (ii) *, ** and *** denote statistical significance at 10%, 5% and 1% levels respectively.

Table A2. Seemingly Unrelated Regression Equation (SURE) results.

Variables	Coefficients	Std. Err.	t-statistics
FDST	0.137*	0.305	1.98
EDI	0.037**	1.061	2.64
GDP	0.310***	0.163	3.92
ERTI	−0.025*	0.225	−1.99
REC	−0.150***	0.037	−4.31
Intercept	0.229	0.743	0.384
R^2	0.878
Equation log-likelihood	−14.619
Testing for the hypothesis of slope homogeneity
List of restrictions for the Wald test
$H_0:A_2=B_2=C_2=D_2=E_2$
CHSQ (2) = 25.152 (0.000)

Note: 1%, 5%, and 10% significance levels are captured by the asterisks ***,**,* respectively.