Natural resources abundance, economic globalization, and carbon emissions: Advancing sustainable development agenda

School of Economics and Management, Foshan University, Foshan, China School of International Trade and Economics, University of International Business and Economics, Beijing, China Department of Business Administration, ILMA University, Karachi, Pakistan Financial University under the Government of the Russia Federation, Moscow, Russia Department of Economics, University of Lagos, Lagos, Nigeria University of Groningen, Groningen, The Netherlands

prosperity because of the region's extensive petroleum and natural gas reserves (Kiprop, 2019). The domestic abundance of gas and oil resources, high energy consumption, and energy exports to meet global energy demands increase this region's significance and strategic importance. The economic development sustained by consuming massive natural resource reserves, such as oil and gas, has sparked industrialization, urbanization, and unsustainable agriculture activities in the region (Magazzino & Cerulli, 2019).
These conditions have created a severe challenge for the sustainable development of the region. Almost 85% of GHGs emissions in the region are caused by energy consumption and production. The region's emissions level is higher than the global average in nearly all countries (Charfeddine & Mrabet, 2017). Moreover, the regional population is expected to double in the next 40 years (Magazzino, 2019).
It could significantly increase natural resource exploration and intensify regional CO 2 emissions. In addition, rising regional energy demands, vulnerability to highly volatile energy prices, and the region's unique weather distinguish MENA counties from other nations. Despite significant economic development, the countries in this region have not completed the first industrialization stage, which is generally characterized by less sophisticated products requiring high energy consumption (Can & Gozgor, 2017). Therefore, the region still relies on foreign countries for machinery and equipment. These nations also largely depend upon natural resource exports for their economic development. As such, interacting with other nations through economic globalization is crucial for this region.
Therefore, this research explores the connection among natural resource abundance, economic globalization, and CO 2 emissions in the unique context of MENA countries. Natural resources consist of minerals, gas, oil, and forest resources. Balsalobre-Lorente, Shahbaz, Roubaud, and Farhani (2018) posit that natural resources discourage the use of certain high pollutant fossil fuels by decreasing their import and providing a viable option to switch to low pollutant energy resources such as natural gas. Some empirical investigations support this view. For example, Zafar et al. (2019b) indicate that natural resource has curbed environmental damage in the U.S., and similar findings were found in BRICS economies (Danish, Baloch, Mahmood, & Zhang, 2019).
In contrast, Ahmed et al. (2020a) posited that natural resource abundance pollutes the environment as mining activities degrade the environment. A country's tendency to use ample high pollutant lowcost fossil fuels makes it unlikely to capitalize on abundant natural resources' environmental benefit. Similarly, Sarkodie and Adams (2018) concluded that deforestation, mining, and chain saw operations are leading sources of environmental pollution and natural habitat loss. Ahmed et al. (2020b) and Zafar et al. (2020) also support the view that economic development and the resulting industrialization and urbanization kindle natural resources exploration. It ultimately increases environmental degradation. Previous studies have presented different perspectives and significant disagreement; thus, we propose that whether abundant natural resources degrade the environment or increase environmental sustainability depends on the studied area's unique characteristics and natural resources exploration practices.
Curbing the negative environmental consequences of mining activities relies on upgrades to green exploration. Technological development can also nurture sustainable mining practices, which may mitigate the environmental threats associated with different mining stages. MENA countries have not shown an ability to produce green technology; thus, we included economic globalization in the model.
Economic globalization covers the effects of foreign investment and trade. Both these factors are crucial to adopt modern technology. For example, Ahmed et al. (2019b) found that economic globalization can increase technology transfers through foreign trade, stimulating environmental quality. In contrast, in the absence of favorable environmental regulations, trade openness may decrease environmental sustainability through a scale effect. Likewise, foreign investment can pollute the environment as developed countries preferred to invest in developing countries with relaxed environmental regulations (Shahbaz, Nasir, & Roubaud, 2018). Governments in developing economies tend to promote economic development by offering relaxed environmental regulation and this phenomenon, which is known as pollution havens, increases environmental degradation associated with FDI (Shahbaz, Nasreen, Abbas, & Anis, 2015). In contrast, investing in energy-efficient technology can support environmental sustainability (Zhu, Duan, Guo, & Yu, 2016).
Given this background, this study makes the following contributions to the literature. Firstly, it explores the nexus among natural resource abundance, economic globalization, and CO 2 emissions in the unique setting of MENA countries. To our best knowledge, previous studies have not investigated this complicated relationship in the MENA region, even though the region is blessed with natural resources, and economic globalization plays a critical role in the economic development of this region. The countries in the region are dependent on economic globalization not only to export natural resources but also to avail the technical capacity required for natural resource exploration. Secondly, this research employs an advanced panel data estimation method to mitigate the problem of crosssection dependence (CD). Traditional panel data techniques, such as fully modified least squares (FMOLS) and dynamic least squares (DOLS), assume no dependence among panel cross-sections. It means that a shock in one country/section does not impact other countries/cross-sections. However, due to globalization, economies are closely connected socially, politically, and economically. Therefore, this research uses continuously updated fully modified (Cup-FM) and continuously updated bias-corrected (Cup-BC) methods introduced by Bai, Kao, and Ng (2009) to generate robust and reliable findings. In addition to CD, the methodology used in the paper can also solve autocorrelation and endogeneity problems. Besides long-run analysis, the study used Dumitrescu and Hurlin test to check causality between variables for recommending appropriate policy suggestions.
The rest of the paper is arranged as follows. Section 2 of the paper includes a review of the literature. Section 3 explains the empirical modeling, data, and methodology. Section 4 describes the empirical results and discussion. Section 5 presents the conclusions and policy implications.

| LITERATURE REVIEW
Climate change and global warming, along with an increase in the awareness of these problems, have increased the importance of understanding environmental degradation and its elements. This study investigates the role of natural resources and economic globalization concerning environmental quality for MENA countries. We divide the literature into two sections to elaborate on the relationship between study variables. The first section describes the nexus between natural resources and CO 2 emissions, and the second section addresses economic globalization and CO 2 emissions. They applied cross-sectional augmented autoregressive distributed lags (CS-ARDL) techniques, and the outcomes showed that natural resources increased CO 2 emissions while green investment contributed to improved environmental quality. However, applying the Generalized Method of Moments (GMM) method, A. Khan, Chenggang, Hussain, Bano, and Nawaz (2020) used data from Belt & Road Initiative (BRI) countries and found positive linkages between natural resources and CO 2 emissions. In contrast, I. Khan, Hou, and Le (2021) found that natural resources could control CO 2 emissions for the U.S. Wang, Vo, Shahbaz, and Ak (2020) assessed the effect of economic globalization on CO 2 emissions in the 1996-2017 period for the G-7 economies. That study investigated the role of natural resources and financial development plays in influencing CO 2 emissions. The empirical findings from CS-ARDL revealed that economic globalization, natural resources, and financial development lead to rising CO 2 emissions. Umar, Ji, Kirikkaleli, Shahbaz, and Zhou (2020) examined the linkage among CO 2 emissions determinants in China during the period from 1980 to 2017. Results estimated by FMOLS, DOLS, demonstrate that natural resources and economic growth positively affect China's CO 2 emissions, while globalization tends to boost environmental sustainability. The causality results show that natural resources, globalization, and economic growth contribute to CO 2 emission.

| Theoretical framework
This study explored how natural resources and economic globalization affect environmental sustainability in the MENA region. The unsustainable mining and excessive utilization of natural resources can increase environmental deterioration (Ahmed et al., 2020a). However, abundant natural resources may limit fossil energy source imports, improving environmental quality (Zafar, Zaidi, Shahbaz, & Hou, 2019a).
Economic Development in the MENA region relies on non-mineral and mineral natural resources (Charfeddine & Mrabet, 2017).
Economic globalization can affect natural resource extraction practices because trade openness is associated with efficient technology transfer. However, the scale effect of trade and foreign investment in dirty technology can also lead to environmental pollution (Ahmed et al. 2020b). In the context of the MENA region, economic globalization is critical because exporting natural resources to the rest of the world is the primary source of regional income. Worldwide interactions through economic globalization make it possible to fulfill global demand. These countries also largely depend on the other parts of the world to import equipment and machinery required for resource extraction and other needs.
Environmental degradation is tied with economic development because economic development involves utilizing resources to increase economic activities. Producing and consuming resources place stress on the environment and increase waste generation (Ahmed et al., 2019a).
Urbanization can increase housing, transport, and energy demands, stimulating fossil fuel consumption and generating more CO 2 emissions. In contrast, urbanization may alleviate pollution levels by promoting resource efficiency through train and bus-based collective transportation (Ahmed et al. 2019a). Using the arguments above, we constructed the following model to unfold the impact of natural resources and economic globalization on CO 2 emissions.
For the empirical estimation, the model variables are logtransformed so that the sharpness in data is diminished and variables show better distributional properties. Natural logarithmic transformation helps to remove autocorrelation and heteroskedasticity issues from data. Compared to the linear transformation, results derived from log-transformed models are more consistent and efficient. The log-linear form of augmented carbon emissions is as per the following: where φ 1 , φ 2 , φ 3 , φ 3 and φ 5 are the coefficients of economic growth (Y), trade openness (TO), Natural resources (NR), Economic Globalization (EG), and urbanization (UR).where cross-sections are denoted by "i," MENA economies, while "t" is for the time from1980 to 2018. ln is the natural log, "α" represents the intercept term, " φ" are the parameters, and "ε" is the error term.
It is widely believed that an increase in output contributes to environmental deterioration due to the growing demand for energy and resource consumption. The continuous increase in output in MENA economies poses a significant threat to the environment due to unsustainable growth patterns. Hence based on the above argument, economic growth is expected to have a positive effect on CO 2 emission.
Natural resources play a crucial role in reducing environmental degradation. It is considered as one of the pure and cleaner sources of sustainable energy and fulfills the current and future demands from natural resources (Panwar, Kaushik, & Kothari, 2011). Hence, it is predicted that natural resources lessen environmental degradation and are projected to have a negative impact on CO 2 emission.
Trade openness is the ultimate factor in increasing environmental degradation and climate change (Destek & Sinha, 2020). Hence, trade openness is projected to have a positive impact on CO 2 emissions. Economic globalization is another critical aspect that affects environmental quality. Can and Gozgor (2017) argue that economic globalization exerts a negative impact on carbon emissions, and therefore, it is beneficial in decreasing environmental pollution. Based on the argument, economic globalization is expected to have a negative effect on CO 2 emissions.
Furthermore, Neagu (2020) argues that urbanization poses a positive effect on CO 2 emissions. Consequently, urbanization is anticipated to have a positive and negative effect on CO 2 emissions.

| Data
The study applied a panel data analysis for MENA (Algeria, Bahrain, The Pesaran (2004)' CD test is as follow: where T stands for periods; N is the panel data size; and ρ ij is the correlation coefficient. The null hypothesis of the CD test is that there is cross-sectional independence between the cross-sectional units. The alternative hypothesis is that there is cross-sectional dependence between sample economies.

| Unit root tests
The integration order of the variables is examined after the results of the CD tests. First-generation unit root methods, such as Levin-Lin & Chu and I'm, Pesaran, and Shin (IPS), cannot mitigate CD's problem (Lv & Xu, 2018). Therefore, keeping in view the presence of CD, this study used the second generation cross-sectional augmented IPS (CIPS) and the cross-sectional augmented Dickey-Fuller (CADF) unit root tests (Pesaran, 2007). The equation of the test statistic is as follows: where CA t − 1 and ΔCA t− 1 are the averages for the cross-sections. The study elaborates the statistics of the CIPS test as follows:

T A B L E 1 Data description and source
where CDF is the cross-sectional augmented Dickey-Fuller (CADF) in Equation (6).

| Panel cointegration test
Before estimating the long-run parameters, we assessed whether there is cointegration among the underlying variables. The first and second generation's panel cointegration tests cannot jointly address structural breaks and CD (Larsson, Lyhagen, & Löthgren, 2001;McCoskey & Kao, 1998;Pedroni, 2004;Westerlund, 2005Westerlund, , 2007. According to Phillips and Sul (2003), traditional cointegration techniques can yield deceptive and unreliable findings when the model experiences CD and heteroscedasticity. Therefore, this study used Westerlund and Edgerton (2008) whereɸ i represents the estimator of least squares;σ i the SE of ɸ; and SEɸ i À Á represents the SE ofɸ i . The Westerlund and Edgerton (2008) cointegration analysis presumes that the null hypothesis is that there is no cointegration. The alternative hypothesis is that there are long-run relationships between variables. there is no endogeneity, these measures can predict consistent results (Bai et al., 2009).

| Long-run analysis
The Cup-FM estimation method maintains a constant limited model parameter distribution. Using simulations, the parameters are continuously updated (Cup) over time until they converge. This approach assumes that the error term follows the factor model. As described, we formalize the factor model: where; M F = I T − T −2 FF 0 , I T demonstrates the elements; and T 0 S shows the identity matrix. The error term assumes there are common latent factors. Initial estimates are allocated to F. This process is repeated until convergence is achieved.

| Granger causality test
The Cup-FM and Cup-BC results do not indicate the direction of the relationship between the variables, which is vital for developing policy recommendations. As such, this research applied the Granger causality test of Dumitrescu and Hurlin (2012) to observe the causal relationship between underlying variables. This method yields two statistics: W and Z. The W statistics show the test averages while Z represents the standard normal distribution. The model is expressed as: In this expression j signifies the lag length and β j ( j) specifies the autoregressive parameters. globalization, and urbanization with respect to CO 2 emissions. The outcomes also reveal a negative correlation between natural resources and economic globalization. The empirical research starts by examining the CD, followed by the unit root and cointegration analysis. Hence, due to spillover effects, the variables are cross-sectionally dependent. If we had not considered the CD, the outcomes might have been unreliable (Pesaran, 2004). After confirming the CD, the study examined the integrated level of variables. Table 4 indicates the CIPS and CADF test results and indicates that the variables have a mixed integration order. Table 4 indicates that all the variables, except natural resources, are non-stationary at the level and become stationary at the first difference at a 1% critical value of significance.  Westerlund and Edgerton (2008) was also applied to determine each MENA country's breakpoint. This method was first developed by Bai and Perron (1998) to ascertain the structural breakpoint.  1991, 1992, 1993, 1994, 1997, 1998, 2000, 2001, 2003, 2005, and 2006. These breaks influenced both global shocks and local shocks for each selected country. After confirming the presence of the long-term association using the Westerlund and Edgerton (2008) panel cointegration test, we gauged the long-term relationship elasticities applying the Cup-FM and Cup-BC method.  Asian economies (Zafar et al., 2020).

| RESULTS AND DISCUSSION
The findings further reveal a negative association between natural resources and CO 2 emissions. It is because governments use safe and green technologies to extract natural resources, which contributes to reducing the amount of CO 2 emitted into the atmosphere. The MENA countries have large oil and gas reserves, providing permanent standing in the international economy. The World Atlas indicates that MENA countries hold 45% of the world's natural gas reserves and 60% of the world's oil reserves. In 2018, the MENA countries emitted 3.2 billion tons of CO 2 and generated 8.7% of the total GHGs. These findings are closely related to the MENA countries' use of renewable energy, which produces fewer emissions than fossil-fuel sources such as oil. These results differ from other studies' findings focusing on Europe (Bekun et al., 2019) and Pakistan (Danish, Ulucak, & Khan, 2020). In contrast, these results are consistent with a study on five EU countries (Balsalobre-Lorente et al., 2018).

MENA's economic globalization level is also negatively correlated
with environmental quality. It may be because their globalization process is controlled by strong environmental laws that restrict them from degrading the environment. These environmental laws lead to green and efficient environmental-friendly technologies, contributing to improvements in environmental quality. Economic globalization is more than the movement of manufactured products; it also includes the movement of resources, intermediate goods, and technologies.
Multinational corporations can transfer their expertise in green technology to economies having strong environmental standards. These results diverge from the finding of (Saint Akadiri, Adewale Alola, Olasehinde-Williams, & Udom Etokakpan, 2020;Wang et al., 2020).
The coefficient value of trade openness is positive and significantly impacts CO 2 emissions at the 1% critical value. Production in these economies may use outdated technologies that are damaging to the environment, and the environment in these economies can contain significant levels of pollution caused by dirty factories. It means a rise in trade openness can increase CO 2 emissions under weak environmental regulations because of dirty manufacturing's competitive advantage. This study's findings are consistent with (Danish, 2020;Hakimi & Hamdi, 2016) but not consistent with (Gardiner & Hajek, 2020). In the case of urbanization, the coefficient also indicates a positive and significant effect with respect to CO 2 emissions, indicating that urbanization is harmful to environmental quality. It may be because these nations are not using environmentally friendly policies during urbanization. Urbanization influences the physical environment Note: ***Significant value at 1%, **significant value at 5%, *significant value at 10%.
because it changes the human community, people's behaviors, and the need for resources. This outcome is consistent with other studies (Ali, Bakhsh, & Yasin, 2019;Hanif, 2018).

| CONCLUSIONS AND POLICY IMPLICATIONS
Environmental sustainability is a major global problem, and the topic has drawn the attention of researchers and policymakers because of climate change. This research adds to the literature by analyzing the relationships between natural resources, economic globalization, and CO 2 emissions in the setting of rapid urbanization, trade openness, and economic growth for the period 1980-2018 in MENA countries.
Also, the results of the economic globalization variable show a negative and significant link with CO 2 emissions. Governments should continue to invest efforts to control the exchange of goods and services and implement bilateral trade agreements to reduce CO 2 emissions. In turn, it may positively impact environmental quality. It is well-known that economic globalization provides cleaner production technology transfers to underdeveloped countries and creates great awareness toward cleaner business strategies, and enables countries to achieve environmental sustainability and to design a sustainable future. Sustainable development mandates the protection of the environment and natural resources as well as to provide social and eco-