Determinants of material footprint in BRICS countries: an empirical analysis

This paper explores the relationship between renewable energy consumption, urbanization, human capital, trade, natural resources, and material footprint for BRICS countries from 1990 to 2016. We apply the cross-sectional dependency test to check the correlation among the cross-section. Then, we use the second-generation panel test like CADF and CIPS to check the stationary in the series. After that, we go for the panel cointegration test, i.e., Pedroni and Westerlund panel cointegration, to know the long-run relationship of the variables. The test results reject the null hypothesis of no cointegration among the variables and accept cointegration. The long-run results indicate that economic growth, natural resources, renewable energy, and urbanization have reduced the environmental quality for BRICS countries in case of material footprint employed to measure environmental degradation. However, foreign trade and human capital improve environmental quality. Based on the empirical results, the study recommended some important policy suggestions to achieve sustainable development in BRICS countries.


Introduction
Climate is changing at an unprecedented rate in the world. It poses a severe threat to humanity, natural life, and the global sustainable environment. The leading cause of climate change is the increase in greenhouse gases (hereafter, GHGs) into the atmosphere. Human activities, such as the production and consumption of non-renewable energy and forest clearing, are responsible for the GHGs, which leads to global warming. The repercussion of global warming is gigantic both for humans as well as for the ecosystem. Rising sea levels, intense floods, drought, melting ice, and massive loss of animal and plant species are the most visible global warming impact of the United Nations Framework Convention on Climate Change (2019). The energy sector is the largest emitter (around 68%) of the global GHGs, and coal accounts for 30% of the GHGs. Even though the world energy demand grew by 2.3% in 2018, it rose in the global carbon dioxide (hereafter, CO 2 ) emission to 1.7% (International Energy Outlook 2018).
Until the emergence of the global financial crisis in 2008, the world witnessed enormous economic growth from 1990 to 2008. Industrializing economies in Asia converged towards high-income developed economies as India's real gross domestic product per capita grew by 115%, and China grew by 219% Pothen and Welsch (2019). Industrialization always signifies the significant increase in energy use and extraction of non-renewable and renewable materials , which leads to emissions of GHGs. Therefore, the world demand for materials during the same years rose rapidly. The extraction of minerals, biomass, and fossil fuels increased to 69.7 billion tons in 2008 from 37.2 billion tons in 1990. Due to globalization and the free flow of international goods, material consumption has grown tremendously. Ansari et al. (2020) analyzed the environment-growth nexus by taking into account the material footprint and ecological footprint as a holistic measure of human pressure on the environment for Asian countries. Their finding supports the existence of the environmental Kuznets curve (EKC), which means that environmental quality degrades during the initial phase of economic growth due to increased material consumption. After a certain level of economic development, environmental quality improves. Pothen and Welsch (2019) also examine the relationship between economic growth and material use for 144 countries between 1990 and 2008 but did not find evidence of decoupling material used for economic activity.
The rapid industrialization linked with significant increases in material consumption poses a severe threat to the environment, such as soil, water, air pollution, loss of biodiversity, and emissions of GHGs. According to IEA (2010), the growth rate in GHGs increased hugely from the emerging economies in Asia.
For this reason, Wiedemann et al. (2015) developed a consumption-based indicator of natural resources, which is called material footprint. It measures the pressure on the natural resources and the material demand from the higherincome countries. They indicate it as another measure to examine cross-country sustainability apart from ecological footprint.
Most studies in the existing literature employ CO 2 emissions to measure environmental pollution. CO 2 emissions as a sole measure of environmental degradation are not sufficient because they do not include other significant pollutants contributing to environmental degradation (Wackernagel and Rees 1998;Al-Mulali et al. 2015;Ulucak and Bilgili 2018). CO 2 emissions are not a comprehensive measure of environmental hazards because hazards are limited to the atmosphere (Nathaniel et al. 2020b). Therefore, there has been a universal call for a more comprehensive indicator of environmental pollution. In this study, we use material footprint to measure environmental sustainability, which is a broader measure than CO 2 emissions solely as an indicator for environmental pollution (Ansari et al. 2020c;Södersten et al. 2020).
The current study focuses on the major emerging economies globally, i.e., Brazil, Russia, India, and China and South Africa (hereafter, BRICS) countries for the following reasons: BRICS countries have become a significant player in global economic development. BRICS countries account for fortytwo percent (42%) of the world population, fifteen percent (15%) of the world trade, and forty percent (40%) of currency reserve. BRICS countries also contribute twenty-five percent (25%) of the global GDP (Siddiqui 2016;Ahmed 2017). The average annual growth rate of BRICS countries is around 6.5 percent (World Development Indicator 2017). It is also projected that China and India will become the 1 st and 2 nd largest economies globally.
In contrast, Russia and Brazil will become the 5 th and 6 th largest economies behind Japan by 2050 (Siddiqui 2016). Over the past decades, this considerable economic growth in this region has led to several resource consumption and environmental issues. In 2013, BRICS countries emitted more than forty percent (40%) of global CO 2 emissions (Liu et al. 2017). There has always been a trade-off between economic growth and environmental quality. Therefore, to better understand the dynamics of the environment and economic development and reverse the current trend in the BRICS countries, robust policies are required.
Due to rapid industrialization and urbanization, it has led to considerable economic improvement over the past decades (Nathaniel and Khan 2020). Still, these are the factors responsible for environmental degradation. Urbanization, solely, has caused the seventy percent (70%) increase in CO 2 emissions, and it will continue to rise (around 76%) by 2030 International Energy Outlook (2019). Another important factor responsible for environmental degradation is international trade. Though globalization has brought the world closer, this has increased in the traded goods and material. Therefore, these traded goods and materials have a social, economic, and environmental impact on the world.
According to the World Development Indicator (2018), around seventy percent (70%) of the world energy demand is attained by non-renewable energy. BRICS economies also depend heavily on non-renewable energy; in 2013, it accounted for more than thirty-five percent (35%) of the global total. Global metal ore extraction tripled to 7.4 billion tons from 1970 to 2010, out of which fifty-four percent (54%) is used by BRICS (Tian et al. 2020). China and India are expected to use more natural resources in the future. Therefore, in the future, all this poses significant environmental challenges for the BRICS region. It makes the BRICS region an attractive case study because it is at a crossroads in new natural and ecological resource management (Nathaniel et al. 2020).
Against this background, this study analyzed the effect of economic growth, renewable energy, human capital, urbanization, and trade on the material footprint. This study enriches the existing literature in the following way: (i) the majority of previous studies in the context of BRICS have used CO 2 emissions and ecological footprint as an indicator of environmental sustainability; this is the first study to the best of our knowledge which has used material footprint as an indicator of the sustainability in the BRICS region. (ii) We have also included human capital (based on school enrollment) in our model because education plays a vital role in improving environmental quality. (iii) We have employed the advanced econometrics technique that gives robust estimates in the presence of endogeneity, cross-sectional dependence, and serial correlation.
Our results suggest that economic growth increases the material footprint in the BRICS region. Our findings are in line with Wiebe et al. (2012), Pothen and Welsch (2019), and Ansari et al. (2020c), which also finds that economic growth leads to a substantial increase in material extraction and which leads to deterioration of the environmental quality. We also find that natural resources and urbanization show a positive and significant relation with material footprint. The previous studies (Martínez-Zarzoso and Maruotti 2011;Shahbaz et al. 2016;Adams and Nsiah 2019;Bekun et al. 2019) find similar findings. We also seek to examine the possible effect of human capital on material footprint and see that it helps conserve the environment. Next, we find an inverse relationship between foreign trade and material footprint. Our findings are in line with the seminal paper findings by Wiedemann et al. (2015), which shows that the fastest-growing countries India and China have attained a relative decoupling for both material footprint and domestic material consumption while South Africa has achieved absolute decoupling. This study will help better understand the BRICS country scenario because of their growing energy demand, their contribution to GHGs, and their commitment to environmental sustainability and conservation.
The rest of the study is organized as follows. The "Review of literature" section is devoted to the review of the literature. The "Data and econometrics methods" section discusses the data and econometric methodology. The "Result discussions" section provides the results. The "Conclusion and policy implication" section concludes with a particular policy recommendation.

Review of literature
Extensive empirical studies have been carried out to investigate the underlying forces responsible for environmental degradation. Although CO 2 emissions are the leading cause of climate change, most research used this to measure environmental pollution. However, it has been blamed for using it as the primary proxy for environmental emissions as it ignores other significant contaminants that contribute to environmental degradation (Al-Mulali et al. 2015). Material consumption has also improved over the year and has been a considerable resource quality measure. Due to globalization and the free flow of international goods, material consumption has increased tremendously. For that reason, Wiedmann et al. (2015) developed a consumption-based indicator of natural resources called material footprint (MF). Hence, we are considering MF as a proxy of environmental quality. This paper investigates the effect of renewable energy, human capital, urbanization, and trade on the MF.

Renewable energy, economic growth, and environmental quality
The number of studies have analyzed the relationship among renewable energy, economic growth, and environmental quality in panel as well as time series technique like (Ozturk and Acaravci 2010;Nassani et al. 2017;Destek and Sarkodie 2019;Baloch and Wang 2019;Khan et al. 2018;Ahmad et al. 2020). Some studies also investigate the environmental Kuznets curve (EKC) hypothesis like Zhao et al. (2016), Ozatac et al. (2017), Pata (2018), Dogan and Turkekul (2016), Lin and Zhu (2018), Salim et al. (2019), and Ansari et al. (2019Ansari et al. ( , 2020a. A recent study by Ansari et al. (2020c) examined the EKC in the Asian sub-region and found that in regions like Central and East Asian countries, there is an existence of EKC hypothesis, and regions like West, South, and Southeast Asia does not validate the EKC hypothesis. Nathaniel et al. (2020c) explore the relationship between renewable energy consumption, urbanization, and environmental degradation. They have found that financial development, economic growth, and urbanization enhance environmental degradation in MENA regions. Few studies also discussed the determinants of environmental degradation with the macroeconomics variables like financial development, trade, and economic growth, which are Beckerman (1992) Mahmood et al. (2020), and Abdouli and Hammami (2020). The studies found that the determinants like urbanization, renewable energy, and economic growth reduce environmental degradation (Martínez-Zarzoso and Maruotti 2011;Zhang et al. 2015;Aye and Edoja 2017;Sahoo and Sahoo 2020a, b). Ansari et al. (2020c) examined the relationship between the energygrowth-environment relationship with top 10 carbon dioxide emitter countries. They found that the EKC hypothesis is not valid for all countries except the USA.
Another strand of literature explores the relationship between renewable energy consumption and economic growth. Some studies found a positive relationship between renewable energy consumption and economic development (Leitão 2014;Soava et al. 2018;Ntanos et al. 2018). However, a study by Menegaki (2011) in 28 EU countries using panel regression analysis found that renewable energy does not significantly affect economic growth in the study period. Lean and Smyth (2014) used disaggregated energy consumption data like petrol and diesel. The results demonstrate that diesel and petrol are driving factors for economic growth. Apergis and Danuletiu (2014) empirically examined the relationship between non-renewable energy, renewable energy consumption, and economic development. They determine the positive relationship between renewable energy and economic growth and the causality of renewable energy to real GDP.
The researchers also investigate the role of energy efficiency for mitigating environmental quality in firm-level and country-level data case and suggest that with improved technologies and cross-border spending in performance improvement programs, materials and energy efficiency should be enhanced (Haider and Ganaie 2017;Bhat 2018, 2019;Haider and Mishra 2019). The renewable energy sector transition is increasing nowadays to protect the environment. In this scenario, material technology plays an essential role in meeting the growing demand for energy in the economy (Starr 2006). Giurco et al. (2019) suggest that with the fast green energy growth and the transport system electrification needed in the 1.5°C scenarios, the demands for resources, especially cobalt and lithium, are also growing significantly. The OECD (2015) reported that it needs aggressive policies to address these challenges scale to promote a significant increase in resource usage, mainly through technological progress and innovation. The push for better resource quality will generate new goods, markets, and job prospects. Karakaya et al. (2020) state that "there are many techniques that are deemed effective in increasing material quality, i.e., using fewer for material design, repair, reuse, and recycling." Natural resources, human capital, urbanization, and environmental quality Researchers and policymakers have used different indicators to investigate environmental quality. Some of the studies have used CO 2 emissions as an indicator for measuring environmental quality (Shahbaz et al. 2020a;Nathaniel and Iheonu 2019;Shahbaz et al. 2020b;Mahmoodi 2017;Pata 2018). A study by Adams and Nsiah (2019) explores the linkage between renewable energy consumption, urbanization, and CO 2 emissions in Sub-Saharan countries. They have applied FMOLS and GMM techniques to analyze the results. They found that renewable energy consumption positively affects CO 2 emissions; however, urbanization reduces it. A similar study by Menyah and Wolde-Rufael (2010) argued that "green energy use has not achieved a stage where it will make a meaningful contribution to reducing pollution." However, Bilgili et al. (2016) examined the relationship between renewable energy consumption, economic growth, and CO 2 emissions in OECD countries. They found that renewable energy consumption has reduced CO 2 emissions in the study period. They argue the need for appropriate environmental policies should be implemented, and residents should meet industrial energy demand through renewable energy production. While some of the studies are neutral about the relationship between renewable energy consumption and CO 2 emissions, Pata (2018) empirically examined Turkey's case using ARDL and cointegration technique. They found that economic growth has a maximum effect on environmental degradation, followed by financial development and urbanization. Nevertheless, renewable energy consumption and hydropower energy consumption do not affect environmental degradation.
On the opposite, a much greater understanding of the connection between natural resources, economic growth, and environmental quality helps decision-makers and government officials minimize environmental emissions and stimulate growth in renewable energy sectors. A recent study by Ulucak and Khan (2020) examined the relationship between renewable energy consumption, economic development, and ecological footprint (EF) in the BRICS economies. They argued that urbanization increases transport and industrialization demand and, in particular, increases the energy consumption of fossil fuel and the EF. Similarly, some of the papers which discussed the linkages between renewable energy, natural resources, urbanization, and environmental quality in BRICS economies are Sebri and Ben-Salha (2014) From the above empirical literature, we can conclude that there are mixed results regarding the linkage between renewable energy consumption, urbanization, and environmental degradation in different countries. It may be due to their economic structure and environmental policies towards controlling environmental pollution. However, various studies have taken various environmental indicators like carbon dioxide emissions and ecological footprint as a proxy for measuring environmental quality. This paper adopts material footprint (MF) as recommended in previous research papers (Giljum et al. 2015;Berrill et al. 2020;Ansari et al. 2020b;Karakaya et al. 2020). In light of the above factors, the goal is to explore the determinants of MF in five BRICS countries by using advanced panel techniques, including renewable energy, urbanization, and natural resources. This study period is constrained from 1990 to 2016 for BRICS countries as per the data limitation. The study contributes to the existing literature by looking at MF instead of carbon emissions because BRICS countries face challenges relating to high material extraction to meet the growing demand.

Data
The study has covered the sample period from 1990 to 2016 for BRICS (Brazil, Russia, India, China, and South Africa) to explore the relationship between renewable energy, urbanization, human capital, and material footprint. All variables were transformed into logarithmic form to reduce the dataset fluctuation (Villanthenkodath and Mushtaq 2021;Villanthenkodath and Arakkal 2020;Sahoo and Sahoo 2020a;Sahoo and Sahoo 2019). This study following (Ansari et al. 2020b;Nathaniel et al. 2020b) the specification of the model is as follows: where MF is the material footprint, GDP is gross domestic product per capita, HC is human capital, NR is natural resources, TR is trade, REN is renewable energy, and URB is urbanization. More specifically, global raw materials have simply been extracted to facilitate the sale to other countries of goods and services. As BRICS is an emerging country, people are mining and exploiting growing quantities of earth natural resources. Overconsumption in developing countries of natural resources and the environmental implications of emerging countries have been serious, for example, deforestation, water scarcity, and climate change. It is necessary to estimate the material extraction in emerging countries and its consequence on the environment. Some studies that also highlighted domestic material consumption (Dittrich et al. 2012). Wang et al. (2012) state that the recent use of the resource investigated driving factors in China, the wealth factor most contributed to rising the direct input of materials. A material footprint is a special form of environmental footprint which assesses natural resources (biomass, fossil fuels, metal ores, and non-metal ores) measured in units of a tonne. In the above equation, i and t represent country and time, respectively. The above functional form can be written as a log-linear model: From the above Eq.
(2), β 1i ..... β 6i symbolizes the elasticity of material footprint, gross domestic product per capita, human capital, natural resources, trade, renewable energy, and urbanization, and ε it is the error term for BRICS economies. The measurement and sources of the variables are presented in Table 1.

Econometrics methods
We have to look at the fact that either the data we obtain has the characteristics of cross-sectional dependency or independence. We selected the cross-sectional dependency test of Pesaran (2004) for this reason. The null hypothesis of the CD test is variables are cross-sectionally independent, but the alternative hypothesis is they are cross-sectionally dependent.
It is significant that when cross-sectional dependence occurs between cross-sections, then unit root tests of the second generation are more suitable than unit root tests of firstgeneration (LLC and ADF). The analysis thus applies the unit root tests for each panel time-series data to the CADF and CIPS techniques along with LLC. Both tests have a similar implementation process, except for the cross-sectional average of CADF test is CIPS. Based on the Dickey-Fuller augmented approach (ADF), the model of panel unit root tests follows: The w it represents variables, ρ is deterministic components, δ is level of significance, and ε is error term in the model. Pesaran (2007) has developed the second-generation unit of root testing of cross-sectional ADF (i.e., CADF) and crosssectional IPS (i.e., CIPS). The cross-sectional IPS (CIPS) equation is the following: where CADF i is the cross-sectional augmented Dickey-Fuller test, and N is the number of observations. After getting the CD and root test results, we applied the second-generation panel cointegration test and the Pedroni cointegration test. Pedroni (1999Pedroni ( , 2004 developed seven different test statistics using the results obtained from panel cointegration regression to test the null hypothesis. Test statistics are calculated by the following Eqs. (6) and (7) The Westerlund (2007) cointegration considers crosssectional dependency. The significant advantage of Westerlund's cointegration test is that it produces accurate cross-sectional dependence results; it can be used in a small sample. The next step is to know the long-run relationship between renewable energy, natural resources, human capital, trade, urbanization, and material footprint. We apply a fully modified ordinary least square (FMOLS) to check the variables' long-run elasticity. Mathematically, the FMOLS model is being written as follows: where , X * i is the average of X t , and ΔX it − k is differential term of X. The results of the FMOLS model are presented in Table 9. It is essential to know the causality of the stated variables in this study. The Granger (1969) non-causality test is extended by Dumitrescu and Hurlin (2012), and the new approach proposed for the checking of the causal direction between the explanatory variables.

Result discussions
Descriptive statistics and trend analysis Table 2 highlights the summary statistics for the panel, in which the highest average mean for the material footprint (21.49) followed by economic growth (8.32), human capital (4.31), urbanization (3.96), trade openness (3.64), natural resource rent (1.48), and renewable energy consumption (0.992), respectively. The median value is almost close to the mean for all the variables. In terms of variances, material footprint (1.20%) has the high variance followed by economic growth (0.95%), natural resource rent (0.71%), renewable energy consumption (0.67%), urbanization (0.40%), and human capital (0.31%), respectively.
In Table 3, the emerging country-wise mean value of each variable is reported during 1990-2016. It is evident from the result that the highest material footprint use was reported in China (23.29). India stands at the second position (22.03), and South Africa reported the least material footprint use. However, the emerging economy average material footprint use stands at 21.49. In terms of per capita GDP, Brazil (9.163) and Russia (9.039) are the top two, followed by South Africa (8.77) and China (7.76). India stands at last in per capita income among these emerging economies. It indicates that countries differ in terms of economic growth. The highest human capital is in Brazil (4.51), and the lowest human capital is in India (3.97). Russia stands at the top in natural resource rent (2.49), whereas the emerging economics average stands 1.48. The trade openness of each emerging economy is almost equal to the average, i.e., the economies follow similar trade openness practices. In terms of renewable energy consumption, these emerging countries have a different pattern. China is the top renewable energy-consuming country, whereas South Africa is the least renewable energy-consuming country among these emerging countries. The top two urbanization positions go to Brazil and South Africa, whereas the least urbanized country among the emerging country is India. Figure 1 shows the pattern of the employed variables during 1990-2016 for the BRICS countries along with global economic variations. Material footprint follows a similar pattern for all the nations. It is also evident that per capita GDP is higher for Russia and Brazil. A rising trend for human capital is visible for all the emerging countries. Similarly, renewable energy consumption shows an upward trend for all countries. In the case of renewable energy consumption in Russia, a In contrast, natural resource rent shows a declining trend during the period for all the panel units. Openness and urbanization of these countries are also increasing during the study period. From the plot, it unveils that the urbanization process is more rapid in China. Moreover, an increasing trend of urbanization has established for all the panel countries. Table 4 delineates the correlations among material footprint, economic growth, human capital, natural resource rent, trade openness, renewable energy consumption, and urbanization during 1990-2016. The result indicates that material footprint negatively correlates with economic growth, human capital, natural resource rent, trade openness, and urbanization, but it has a positive relation which was exhibited with renewable energy consumption. Economic growth indicates a positive relationship with natural resource rent, trade openness, renewable energy consumption, and urbanization. It means that economic growth can push the variable positively. Trade openness and urbanization are positively influenced by natural resource rent, whereas renewable energy consumption negatively impacts natural resource rent. A positive relationship between trade openness and urbanization has been observed against a negative association between renewable energy consumption and trade openness. Again, a positive relation between urbanization and renewable consumption has been revealed. In sum, all the variables except renewable energy consumption negatively correlated with our dependent variable, i.e., material footprint.

Panel unit root tests results
The results in Table 5 demonstrate the empirical analysis of the integration order of the used variables. In the study, choosing the appropriate panel data econometric models sequentially, the first step is to determine the integration order of variables. The first step is critical because it gives directions about the series under consideration. To attain this goal in the study, authors employed the Levin, Lin, and Chu (LLC) unit root proposed by Levin et al. (2002). This unit root test works within the assumptions related to the general process of the unit root test. Hence, it has the null hypothesis of a non-stationary or unit root against the stationarity assumption associated with the alternative hypothesis. In the unit root testing process, the lag length section is important to overcome the inconsistent result; therefore, we have employed the Schwarz Information Criterion (SIC) in the LLC. The evolved results from LLC  indicate that the null hypothesis has to accept for material footprint, economic growth, human capital, natural resource rent, trade openness, renewable energy consumption, and urbanization at the levels of the variables. In contrast, after taking the first difference of the variables, the results of LLC firmly reject the null hypothesis in the case of material footprint, economic growth, human capital, natural resource rent, trade openness, renewable energy consumption, and urbanization. All the variables except urbanization are statistically significant at a 1% level, but urbanization shows the statistical significance at 5% in the rejection of the null hypothesis. Therefore, it highlights that the variables are stationary at the first difference or variables are I (1) series. Test for cross-sectional dependence and secondgeneration unit root tests Table 6 reports the result of cross-sectional dependence proposed by Pesaran (2004). The alternative hypothesis of cross-sectional dependence has unanimously accepted by providing evidence against the null hypothesis regarding CD test statistics and its statistical significance. On the backdrop of cross-sectional dependence, we have to use those methods to consider the cross-sectional dependence to overcome the biased estimated results. The motive behind employing the second-generation panel unit root test is the presence of cross-sectional dependence in the data. Table 7 delineates the results of the CIPS and CADF panel unit root test. The outcome reveals that all the variables are non-stationary at their levels by accepting the null hypothesis. However, the variables follow the mean-reverting process in their first difference. The finding implies that the series under consideration is integrated at I (1). The findings evolved from the second-generation unit root test also support the first-generation unit root, which does not consider the cross-sectional dependence. Table 8 reports the results of the Pedroni panel cointegration test proposed by Pedroni (1999Pedroni ( , 2004. The Pedroni cointegration relies on the seven test statistics with two dimensions, i.e., within-dimension and between-dimension. In within-dimension, it uses Panel v-Statistic, Panel rho-Statistic, and Panel PP-Statistic for the cointegration analysis. Similarly, in-between dimension relies on Group rho-Statistic, Group PP-Statistic, and Group ADF-Statistic for the cointegration. The result reveals that four out of seven statistics, i.e., Panel PPstatistic and Panel ADF-statistic (within-dimension) and Group PP-statistic and Group ADF-statistic (between-dimension), affirm the long-run association (cointegration) among the considered variables in the study. Hence, the Pedroni cointegration test-based result concluded that the variables like the material footprint, economic growth, human capital, natural resource rent, trade openness, renewable energy consumption, and urbanization are in long-run cointegration equilibrium relation dynamics of the variables over the study period. Table 9 reports the Westerlund panel cointegration test results. The test belongs to the second-generation category and uses four error-correction-based panel cointegration tests.   Moreover, it depends on the structural dynamics. The test takes no cointegration as its null hypothesis. The results reveal that out of four tests, two (i.e., Gt and Pt) have rejected the null hypothesis at a 5% and 1% level of significance, respectively. The result is possible to infer that the series are expected to move together in the long-run cointegration existence. The mixed significance of the second-generation cointegration outcomes is consistent with the cointegration finding observed by Usman et al. (2020) for the panel of Africa, Asia, and America, where they observed some test statistics are not significant, but the study concluded in favor of long-run cointegration of the series. Further, this finding supports a Chinese region-based panel study by Liu (2013), where a mixed significance has revealed for the central region from the second-generation cointegration. Therefore, it is possible to argue that if any of the second-generation cointegration statistics is significant, then the series is the cointegrating series.

Long-run elasticity of panel data estimates
Results of long-run elasticity in the BRICS economies are displayed in Table 10. Fully modified ordinary least square (FMOLS) has been employed to estimate the long-run elasticity. The result revealed that the 1% increase in economic growth leads to a positive and significant increase in material footprint by 0.609%. This result is not surprising because economic growth is the outcome of production, which requires various kinds of raw material as an input; hence, it degrades the environment by consuming the resources. This finding corroborates the need to feed the increasing population, economic growth is necessary, but it put pressure on natural resources. Hence, sustainability in natural resource use is vital in emerging economies. There are some kinds of literature that found similar findings like Khan et al. (2018), Zhengge (2008), Jalil and Feridun (2011), Abdouli and Hammami (2020), and Ali et al. (2019).
In contrast, an increase in human capital accumulation helps reduce the material footprint significantly. More precisely, a 1% increase in the accumulation of human capital reduces the material footprint at a rate of −0.392%. It may be because people equipped with education are cautious about their consumption. Moreover, in schooling, they may attain environmental education amid rising climate change across the globe. Therefore, human capital accumulation helps improve environmental quality by preserving the material footprint. This finding is in line with the long-run coefficient by Nathaniel et al. (2020a) for Latin American and Caribbean countries in a CO 2 emissionbased study.
In the case of natural resource rent, the result reveals that a 1% increase in natural resource rent leads to a 0.310% increase in material footprint. The finding is desirable since it indicates that natural resources' exploration pushes economic growth by earning the revenue from exploration, which needs the  As far as the foreign trade is concerned, there exists a statistically significant inverse relationship with the material footprint. This finding is laudable, and the potential interpretation may be that national environmental policy stringency is related to the trade liberalization policies in the era of global warming and climate change. Hence, the trade liberalization policies can be classified into three categories, i.e., the scale effect, the composition effect, and the technique effect in line with Grossman and Krueger (1991). In trade liberalization policies, countries with a comparative advantage in dirty industries may be pushing technique effects over the other two effects, or countries with clean industries may be promoting technique together with the other two effects.
In terms of renewable energy consumption, there is a significant and positive relation with material footprint. Possible implication associated with this result could be that the renewable energy technologies that use renewable energy flows like solar and wind energy may be depleting the mineral resources in these countries. Another possible explanation may be that the resource is also depleting in renewable energy generation, thereby adversely affecting the environmental quality by increasing the material footprint. Our results support the findings of Adams and Nsiah (2019).
The result of urbanization also shows a positive and significant relation with material footprint. Since urbanization is a crucial driver that speeds up depletion, urbanization contributes to natural resource consumption. Suppose people move to urban areas for their settlement. In that case, there is a chance of deforestation, more energy consumption from both industrial and household levels, excess use of water, and so on. The previous studies like Martínez-Zarzoso and Maruotti (2011) for developing economies, Shahbaz et al. (2016) in the case of Malaysia, and Ali et al. (2019) for Pakistan are also similar to our findings. Figure 2 depicts the actual, fitted, and residual graph of the FMOLS model.

Heterogeneous panel causality test
After validating the long-run relationship among variables, we examined the variables' causality using the heterogeneous panel non-causality test by Dumitrescu and Hurlin (2012). Table 11 demonstrates the empirical results of the short-run panel causality test among the material footprint, economic growth, natural resource rent, renewable energy consumption, urbanization, and human capital. The pairwise heterogeneous panel causality test requires the stationarity of the variables under consideration. The study has transformed the variables at their first difference to obtain the objective of stationarity of the variables. The findings reveal a unidirectional causality from economic growth to the martial footprint. Similarly, a one-way causality has been established from material footprint to human capital, natural resource rent, and renewable energy consumption to material footprint. However, a bidirectional causality has been established between trade openness and material footprint. A similar conclusion has been reached between urbanization and material footprint. These findings from the panel causality tests show the importance of explanatory variables on the multivariate function of the material footprint, and its directions are valuable for the policy formulation.

Conclusion and policy implication
This paper aims to investigate the relationship between renewable energy consumption, urbanization, human capital, trade, natural resources, and material footprint for a balanced panel five BRICS countries over the years from 1990 to 2016. In this paper, we apply the cross-sectional dependency test to check the correlation among the cross-section. Then, as the first-generation panel unit root test is not appropriate, we use second-generation panel tests like CADF and CIPS to check  The long-run results revealed there is a positive relationship between economic growth and material footprint in BRICS countries. This finding is not unexpected, since the production is a pre-condition for economic growth, which needs different raw materials as an input, which degrades the environment through consumption of energy. In contrast, an increase in human capital helps reduce the material footprint significantly. It may be because people equipped with education may be cautious about their consumption. They will also gain environmental education during schooling in the sense of growing global climate change. As far as the foreign trade is concerned, there exists a statistically significant inverse relationship with the material footprint. It is praiseworthy, and the likely explanation may be the national environmental policy strictness in the age of global warming and climate change applied to trade liberalization policies. In terms of renewable energy consumption, there is a significant and positive relation with material footprint. Possible implication associated with this result could be that the renewable energy technologies that use renewable energy flows like solar and wind energy may be depleting the mineral resources in these countries. The result of urbanization also shows a positive and significant relation with material footprint. Since urbanization is a crucial driver that speeds up depletion, urbanization contributes to an increase in natural resource consumption. The results of causality revealed that there is bidirectional causality between trade and urbanization with material footprint.
In the era of climate change and global warming, economic development should be sustainable ; MK   Villanthenkodath and Mahalik 2020). Hence, the emerging economies have to reduce the overexploitation of the resources that leads to environmental consequences as a part of economic activities. Therefore, in the emerging economies, policies must be aim to revert the resource consumption pattern to address climate change. The growth with overexploitation of natural resources is not sustainable for the long run; hence, the economic agents should be cautious about the resource depletion and subsequent environmental degradation while performing the production process. The measurement of economic growth should consider both GDP and stock of wealth consisting the natural resources. Making economic growth more sustainable, there is a need for policies like introducing a price mechanism that consists of environmental degradation due to the individual or private activities. Further, deployment of new technology to mitigate environmental degradation due to economic grounds enhances sustainable development investments. The above-stated policies may be used as a catalyst for mitigating the climate change emerging from the consumption of resources, but the extent of the implementation depends on the existing institutional and social settings. Moreover, enhancing energy efficiency in all the areas of economic activities thereby reduces the use of resources like fossil fuels. Therefore, introducing new technology to improve efficiency in renewable energy use, doing so, the countries can reduce pollution and power loss. If the countries do not have enough access to clean financing technology, they can receive foreign aid to mitigate environmental pollution from energy consumption. It is important to note that new technology has to be introduced from the production and exploration to the final energy consumption. However, energy efficiency is attainable only through energy conversion measures. The study also suggests a clear-cut delineation of the regions and countries with a need for clean energy technology. Therefore, at the global level, priority in the technology distribution has to be done accordingly.
The study also advocates that the rent accrued from the natural resources has to be spent on spreading the information related to environmental protection and ecological sustainability. Moreover, the same fund has to be used for sustainable development projects in these countries. Doing so, resource rent has to be collected to follow the principle of equity and sustainability. The distribution of such funds should be used to promote the eco-friendly projects or environmental conservation programs in these countries; then, there is a possibility of future environmental sustainability. The policy related to urban sustainability is highly recommended in the context of environmental degradation. Resource efficiency is an important ingredient in urban sustainability; therefore, it has to be promoted in different temporal and spatial scales. Further, the study recommends an approach to urbanization that depends on environmental management and economic and social development at the national and international levels to mitigate the pollution. Hence, policies have to be formulated with all the stakeholders of the urbanization such as people and local and national governments.
To protect the material footprint, the role of human capital is pivotal; hence, following a well-designed policy for education with a focus on environmental conservation would help the youngsters to understand the impotence quality environment. The inclusion of both theory and application of environmental conservation in the curriculum helps the youngsters think broadly and practically about environmental degradation. Since trade openness shows a negative link with the material footprint, we suggest participation in international trade protects the environment by reducing the material footprint. Hence, the further implementation of new technology in the export sector is inevitable. Also, import quality has to be assured in line with environmental standards so that the countries will get the gains from trade along with environmental sustainability.
Data availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethical approval Not applicable Consent to participate Not applicable

Consent to publish Not applicable
Conflict of interest The authors declare no competing interests.