The Impact of Social Inclusion and Financial Development on CO₂ Emissions: Panel Analysis from Developing Countries

Abstract

The intricate interplay between the environment and the economy entails numerous multifaceted factors that require thorough investigation. Civic activism, intergroup cohesion, and gender equality are among the pertinent factors that hold the potential to significantly impact CO₂ emissions in developing economies. However, these variables have not been explored to the extent that their importance warrants, leaving much to be studied and understood about their complex relationships with carbon emissions. Currently, developing nations find themselves more vulnerable and exposed to a plethora of environmental issues. In response to this pressing matter, the focus of this study is to expound upon the impact of various factors on the environment. To achieve this aim, this study utilizes annual data from 46 developing countries, spanning the extensive period from 1990 to 2014. Using the generalized method of moments and empirical Bayes methods, this study’s results emphasize the significant impact that civic activism, gender equality, intergroup cohesion, and financial development can have on increasing CO₂ emissions. However, civic activism reduces CO₂ emissions. These findings highlight the crucial importance of adopting a comprehensive approach that accounts for both economic and social cohesion indicators when tackling environmental challenges.

1. Introduction

The idea of diversity has become a fundamental principle for achieving sustainable development and economic growth in countries. Previous research has suggested a connection between polarization and fractionalization among different ethnic groups, and how this relates to the provision of public goods, quality of institutions, and sustainable development. There is a relationship between civil conflict and other socio-economic problems and the level of diversity or lack of social cohesion within a particular population. There has been a lot of literature exploring the determinants of environmental degradation but very few studies can be found on social concerns affecting environmental degradation. In developing nations, the private sector’s focus on meeting the preferences of different ethnic groups can lead to social and economic issues. This is because the private sector tends to prioritize the provision of goods to certain groups at the expense of others, disregarding the broader social impact of their actions. As a result, differences in preferences between ethnic groups and a lack of social cohesion can lead to conflict and other negative outcomes. There appears to be a significant gap in the literature about the impact of ethnic diversity, cultural fragmentation, and social cohesion on socio-economic issues, particularly in the context of developing nations. This lack of research is particularly concerning given the role these factors can play in contributing to environmental degradation.

Environmental degradation due to carbon emissions has been less likely discussed in association with social cohesion and ethnic diversity. The major factors that have impacted diversity have impeded economic development, spanning the entire course of human civilization from its origins to the present day, resulting in a persistent lack of progress in terms of social and economic prosperity. [1] found that the migratory distance to indigenous settlements negatively affected genetic diversity, resulting in a U-shaped impact on development outcomes, which shows the trade-off between the benefits and effects of diversity on productivity.

The past literature ensures that climate change has been the most provocative environmental issue in our era and has attracted the attention of international institutions, policymakers, and researchers, due to its impact on developing and some developed countries. If prompt actions are not taken to mitigate it, then it is an approximation that emissions of CO₂ will double or triple by 2050 [2]. The previous literature suggested that CO₂ emissions are the key factor in climatic changes and are responsible for raising the global temperature. The global emissions of CO₂ have risen from 2 billion tons of CO₂ in 1900 to over 36 billion tons in 2015. CO₂ emissions were further multiplied by 2.7% and 0.6% in 2018 and 2019, respectively, whereas the global average CO₂ emissions were 4.8 tons per person in 2017. The Earth’s temperature is increasing sharply due to increased CO₂ emissions in the atmosphere. From 1961–1990, the temperature rose by 0.7 °C, while it was 4 °C colder in 1850. This is primarily caused by the accumulation of greenhouse gases in the atmosphere from human activities such as burning fossil fuels and deforestation.

The Earth’s temperature is progressively increasing, and this fact is widely endorsed by the academic community due to the overwhelming evidence of the human impact on climate change [3]. The issue of climate change is associated with many social aspects, such as the role of economic motivation [4,5], and the degree to which some financial components such as populace, urbanization, and exchange may influence contamination [6,7]. Issues related to urbanization and population growth can lead to a lack of social cohesion. This can be attributed to various factors such as a lack of civic engagement, intergroup conflict, and gender inequality. However, the existing literature has not yet established a clear understanding of how social cohesion affects global change. Some scholars, such as [5], suggest that global climate change may be the result of common challenges. Others argue that social cohesion can lead to increased stability in preferences and higher levels of social trust, which can ultimately result in lower levels of diversity or disintegration [1].

Many populated developing countries are facing the problem of a lack of social cohesion due to the existence of several ethnic groups ultimately responsible for high carbon emissions. Developed countries have overcome these issues by ensuring cohesiveness in society and uniform preferences, thereby controlling environmental damage. Still-developing countries have found this social issue to be one of the major factors in creating more carbon emissions.

Diversity and cohesiveness have an inverse relationship in wealthy economies, as confirmed by the literature. Reducing diversity can facilitate coordination in addressing the difficult issue of carbon emissions. This can particularly aid in efforts to mitigate the global effects of these pollutants on climate change. When examining this relationship, it is crucial to identify the underlying reasons for the concentration of these pollutants in the atmosphere. This can be challenging as the ultimate common determinant may not be as visible as other factors, further complicating coordination efforts. The opportunity cost of widening such problems may be huge. Particularly for poor countries, it involves foregoing or employing resources that could be used for other social programs to be implemented in the short run. Low social cohesion can hinder efforts to address carbon emissions, particularly for countries facing significant opportunity costs in foregoing economic growth. Therefore, the present study explored the determinants of environmental concerns to social issues.

Environmental problems vary between rich and poor nations, leading to the development of two hypotheses that suggest responses to environmental issues may differ depending on a country’s level of prosperity [4,5,8]. Social scientists and economists agree that diversity can affect trust and preferences, making it challenging for non-cohesive nations to reach agreements when tackling environmental issues. This highlights the importance of approaches that focus on managing social disruptions and promoting social capital as part of a broader strategy aimed at addressing the human impact on climate change.

Economic problems are closely linked with social issues, particularly regarding the impacts of financial development on carbon emissions. There are two primary theoretical perspectives on this topic. Financial development has positive and negative impacts on CO₂ emissions. The importance of these two theoretical views is that they expect a linear relationship between financial development and carbon emissions [9]. For the development of an economy, the financial sectors enable the people and firms to execute their savings and investment choices [10], which permits nations to deal with their restricted assets effectively and helps to increase investment projects which make less expensive capital, apportioning assets to beneficial channels, enhancing the exchange and urging firms to use the most recent innovation to uplift the level of domestic productivity [11].

Although financial development plays an important role in economic growth, it also boosts the level of energy consumption [12]. So, in the long run, CO₂ emissions will rise [13]. For instance, financial development assists with lessening the cost of acquiring products which permits firms to purchase new machinery and new plants, which expand energy utilization and CO₂ emissions. Essentially, financial development makes it simpler for customers to purchase extravagant things like houses, vehicles, and climate control systems that consume energy and add to CO₂ emissions.

Whether financial development is a prominent factor in environmental enrichment or deterioration or has no impact on the environment is a debatable matter. However, the literature still has no consent on this matter. In addition, there is no one perception of the impact because some studies showed that there is a negative and minimal impact on financial development and CO₂ emissions. In the existing literature, three alternative hypotheses have been developed. Some researchers found that financial development encourages CO₂ emissions [14,15,16,17]. Some researchers have investigated that there is quite a cut in CO₂ emissions with financial development [18].

Urbanization has an important role in changes in social and economic issues, prosperity, and economic development. However, most of the economic activities are concentrated in cities. Urban inhabitants contribute approximately 66% to global energy consumption, generating over 70% of CO₂ emissions. The proliferation of urban denizens has engendered a surging demand for an array of indispensable necessities, including but not limited to food, housing, land utilization, public utilities, residential home appliances, and urban transportation. This mounting clamor for sustenance and amenities has exerted unprecedented pressure on the energy grid, necessitating a corresponding augmentation in the energy production and distribution infrastructure. As such, the incessant expansion of the urban populace has become a formidable challenge for policymakers and urban planners who must grapple with the ramifications of the energy consumption that is intimately linked to the urban population’s growth trajectory [10]. The management of rural–urban migration, while intended to alleviate some of the social and economic challenges faced by rural populations, can also exacerbate a host of pressing issues, including the removal of excessive amounts of waste materials, the unsustainable utilization of energy resources, water scarcity, and inadequate sanitation facilities. These issues, left unaddressed, can have significant detrimental impacts on both human health and the environment at large [19]. The primary objective of the present study is to investigate the role of social cohesion practices in the context of developing countries. Developing countries are typically classified into two categories—low-income countries and middle-income countries. The selection of countries for this study is based on their gross national income (GNI) per capita, which has been sourced from the World Development Indicators (WDI) using the Atlas method. By using pooled data, this study may examine the impact of social cohesion on CO₂ emissions in developing countries for the period from 1990 to 2010, to see whether social cohesion has a positive or negative impact on CO₂ emissions. Additionally, this study will explore the impact of financial development on CO₂ emissions in developing countries.

The previous literature has largely neglected to examine the relationship between CO₂ emissions and social cohesion in the context of developing countries. Ashraf and Galor [1] utilized the measure of population diversity to assess the level of coordination and cooperation within a country. Their findings revealed a robust positive impact of the lack of cohesion on carbon emissions in the case of wealthy nations. Diego Ramos Toro [20] demonstrated that population diversity and its negative impact on social cohesion in sufficiently affluent countries have a robust and causal effect on CO₂ emissions, resulting in a positive relationship between these variables.

In the case of developing countries, the literature hardly presents any nexus of social cohesion, financial development, and CO₂ emissions. Since the topic is hardly explored, very few studies are available that provide contradictory results in comparison with rich countries. Given the lack of clarity and conflicting results regarding the relationship between financial development and CO₂ emissions in developing countries, the present study aims to investigate this nexus. Specifically, this study will focus on exploring the impact of financial development on CO₂ emissions in developing countries.

2. Literature Review

The literature review section provides an overview of the existing research on the relationship between cities, social cohesion, and the environment. It discusses how social cohesion extends beyond social factors and affects socio-environmental cohesion, particularly in the context of urbanization. The studies have explored the various perspectives on the interplay between urban communities, social cohesion, and the environment, including urban sustainability, equality, and justice considerations. It also emphasizes the need for further research to gain a comprehensive understanding and facilitate effective political action in this area. Furthermore, the review acknowledges the importance of addressing environmental issues while considering the potential impact on social cohesion.

We will begin by exploring the seminal work addresses three misconceptions regarding population growth, environmental deterioration, and resource depletion. It emphasizes the significant relationship between the size and growth rate of the human population and the escalating ecological problems faced by society. By considering the broader concept of environmental deterioration beyond mere pollution, the authors shed light on the importance of population factors in generating environmental disruption. The review will delve into the various dimensions of environmental problems, highlighting both direct assaults on human welfare and indirect effects on natural systems.

Ref. [3] examined the relationship between cities, social cohesion, and the environment. They argue that the negative impacts of social cohesion extend beyond social factors and also affect socio-environmental cohesion, particularly through the processes of urbanization and the enabling and disabling conditions it creates. This study explores three contemporary perspectives on the relationship between urban communities, social cohesion, and the environment. It begins with a fundamental assessment of the concept of urban sustainability and subsequently considers two approaches that emphasize issues of equality and justice in the urban political environment. Finally, this study identifies four areas of research that urban scholars must investigate to gain a more comprehensive understanding and take more effective political action on the nexus between cities, social cohesion, and the environment. Meanwhile, expounding on economic incentives, social preference, and constructed different dimensions showed social cohesion is determined by collective problems, i.e., comprehended in these documents as a category capturing social coordination and cooperation at a national and sub-national level. Thus, it is important to recognize that addressing the environmental issue of carbon emissions cannot rely solely on incentive-based solutions. Such measures may fail to account for the potentially negative impact on social cohesion and could exacerbate existing challenges related to cohesion [21].

Genetic diversity and the origins of cultural fragmentation in the study of [1] showed that diversity has an effect on the prosperity and stability of countries and did not expound on the irregular distribution of ethnic and cultural fragmentation across nations, showing that biographical forces have a strong relation with development. This research empirically represented that genetic diversity, which is determined during the past “out of Africa” migration of humans, is an implicit cause of different existing discoveries of ethnic linguistic heterogeneity. Refs. [4,7] have identified a potential solution to address these issues by aligning incentives appropriately. To personalize the social costs of carbon-removal decisions, the literature explores various policy avenues including implementing a system of tariffs and sanctions at an international level. Ref. [20] investigated the relationship between social cohesion and CO₂. Their study revealed that population diversity and its negative impact on social cohesion had a significant causal and positive effect on CO₂ emissions in wealthy countries.

Examining geocoded data on fossil fuel emissions has also shown substantial results. The impact of diversity was controlled through its effects on mistrust and heterogeneity in preferences, revealing the importance of social dimensions in reducing human carbon footprints. Ref. [22] found a relationship between gender equality and CO₂ emissions. This study addressed the question of how gender equality is related to CO₂ emissions. They hypothesized that these are negatively correlated. The present study employs a theoretical framework based on demand theory and uses a two-way fixed-effects regression analysis on panel data for 139 countries over the period from 1995 to 2014. The results showed a significant negative relation between CO₂ emissions and gender equality in developing countries. Ref. [23] investigated how gender equality influences climate change and the multidimensionality of gender equality effects on the CIWB (a ratio measuring the amount of CO₂ emitted per unit of life expectancy at birth). This study used panel data from 70 countries from between 1995 and 2013. They found that the increase in the percentage of expected years of education and women in parliament reduced the CIWB across all the countries. While the increase in the percentage of women in the labor force raised the CIWB.

The impact of macroeconomic and financial development on carbon emissions in Pakistan for annual time-series data from 1982 to 2018 was empirically assessed. By using the ARDL method, this study revealed that the growth of the stock market, foreign direct investment (measured by the financial development index), economic growth, and consumption of oil had a positive impact on CO₂ emissions, while domestic credit had a negative effect on CO₂ emissions in both the short and long run [24].

Ref. [25] conducted a study to investigate the relationship between financial development and carbon emissions using the system generalized method of moments and data from 155 countries. The results of the study showed that financial development had a significant positive effect on CO₂ emissions at a global level, but this effect was insignificant in developed countries.

To the financial development end, another study by [26] conducted a study on the relationship between financial development, economic growth, and carbon dioxide emissions. Their findings suggest that financial development is a significant factor in driving economic growth and can also have an impact on the level of carbon dioxide emissions in the environment. This study examined the causal relationship between carbon dioxide (CO₂) emissions and financial development in both D8 and G8 nations from 1999 to 2013, using the PMG-panel ARDL technique. The empirical results indicate that financial development has a significant and positive impact on CO₂ emissions, both in the panel and in the long run, at a statistical level of 1%. Ref. [2] conducted a study on the impact of urbanization on carbon dioxide emissions in developing countries. In this study, population was used as a predictor variable in the model, instead of assuming a unitary elasticity of emission with respect to population growth. Data were collected from 1975 to 2005 for various groups of nations classified according to their income levels. The findings indicate that the impact of urbanization on CO₂ emissions varies for low- and lower-middle-income countries. Less-developed countries had a higher impact on CO₂ emissions, while upper-middle-income and developed countries showed a negative impact.

Ref. [2] conducted a study on the social and economic factors that contribute to environmental performance. The study aimed to explore the impact of political, social, and economic indicators on environmental performance, using annual data from 163 developed and developing countries from 1996 to 2016. The findings of the study indicate that institutional quality, political freedom, and ethnic diversity play a crucial role in reducing CO₂ emissions, while GDP growth, energy consumption, and financial development intensify industrialization and urbanization, leading to higher levels of CO₂ emissions. The panel ARDL approach was used to determine the long-run relationship among variables, and the order of integration was determined through Pedroni’s cointegration test.

An investigation on the moderation impact of financial development on the environmental Kuznets curve (EKC) used the significant pollutants of CO₂, PM_2.5, and CH₄. This study utilized panel data from 115 countries, covering the period from 1990 to 2016, and divided the countries into different income groups. The empirical results showed that the interaction effect of financial development on the relationship between squared GDP and carbon emissions became negative for all income groups. Furthermore, in the high- and middle-income groups, financial development had a negative impact on the relationship between GDP and both CH2 and PM2.5 emissions. This result suggested that financial development can reduce environmental degradation without adverse effects on growth [27]. Ref. [12] conducted a study on the nonlinear relationship between CO₂ emissions and urbanization in Indonesia, covering the period from 1971 to 2014. The study detected an inverted U-shaped relationship between carbon emissions and urbanization. The ARDL approach and the Toda and Yamamoto causality test were applied to determine the cointegration and causality between the variables. The findings indicate that urbanization increases carbon emissions, but after reaching a certain level, it negatively impacts emissions. Energy intensity and economic growth were found to increase CO₂ emissions, while trade openness had no significant impact on carbon emissions.

The impact of urbanization and economic growth on CO₂ emissions is explained by [28]. This study analyzed the major determinants of CO₂ emissions in nine Far East nations during the period from 1980 to 2017, using a fixed-effects approach. The results revealed that GDP, trade openness [29] and urbanization were significantly correlated with CO₂ emissions in the selected countries.

Ref. [30] conducted a study on the impact of urbanization on CO₂ emissions in developing countries from 1975 to 2003. The study used a semi-parametric mixture model that accounts for unknown distributional shapes and endogenously classifies countries into homogeneous groups. Three groups of countries were considered in the study. For the first and third groups of countries, the elasticity of emissions to urbanization is negative for high urbanization levels and positive for low levels. In the second group, urbanization was found to be not statistically significant. The results indicated an inverted U-shaped relationship between urbanization and carbon dioxide emissions.

Another study by [31] investigated the relationship between social cohesion and economic growth using panel data from a large number of countries. Three different indices—membership of clubs, voluntary associations of clubs, and intergroup cohesion—were used as proxies for social cohesion. The study also considered institutional quality, rule of law, inequality, safety networks, and democratic rights as indicators of social cohesion. The data covered the period post-1970, and econometric techniques were employed to analyze the relationship between social cohesion and economic growth. The study found that the division in society was a cause of quickly decreasing growth for some nations. The results indicate that political and social aspects cannot be excluded from mainstream economics when considering a country’s growth.

Energy resources and financial development are responsible for environmental degradation and economic growth [32]. This study analyzed the panel data of E-7 countries for the period from 2000 to 2020, using OLS, PQR, and CCEMG techniques. The study found that non-renewable energy resources and financial development may lead to environmental degradation. In another study by [33] the nexus between financial development, institutional quality, technological innovation, and environmental quality was explored, and the results revealed that financial development contributes to the selected carbon emission dimensions in OECD countries.

3. Theoretical Framework

The main objective of this document is to identify the factors that contribute to CO₂ emissions. The hypothesis is that the absence or lack of social cohesion within a country leads to a higher intensity of emissions, due to a lack of coordination and cooperation among community members. Therefore, it is reasonable to suggest that the inability to develop a common strategy against environmental threats due to a lack of cohesion creates a positive relationship between CO₂ emissions and a lack of cohesion. However, this relation seems to be strengthened via the mediating variable of a country’s wealth as it hampers the ability of countries to transition into less polluting economic activity [34].

This suggests that the relationship between social fragmentation and carbon emissions should be examined empirically, as an increase in income coupled with a lack of social cohesion can lead to higher levels of CO₂ emissions. This highlights the significant role of increasing wealth in determining carbon emissions, as discussed by [20] and suggests that a lack of social cohesion is a key factor in determining carbon emissions, and this link was first revealed by [5] who identified it as a major contributor to global climate change. A cohesive and less fragmented society is associated with greater homogeneity in preferences and higher levels of social trust, as demonstrated by research by [1].

The lack of social cohesion can have a significant negative impact on efforts to address emissions issues in developed economies. However, it may not play as significant a role in nations that face significant opportunity costs by forgoing economic gains in their efforts to address emissions [1] as it is less likely for the population of non-cohesive countries to attain an agreement when tackling the environmental challenges. This study suggests that there is a strong relationship between a lack of social cohesion and higher levels of CO₂ emissions in sufficiently wealthy economies. This means that the impact of social cohesion on carbon emissions becomes more severe as income increases. This has been supported by previous research by [34].

In developing countries, economic growth often leads to increased industrialization, consumption, and production activities. As societies become more cohesive and different groups come closer to each other, per capita income increases, which may affect environmental quality [35]. CO₂ emissions tend to increase when social cohesion increases in developing countries, while in a few wealthy nations, a lack of social cohesion can lead to an increase in CO₂ emissions. The underlying hypothesis of higher levels of carbon emissions associated with a lack or absence of cohesion has been less explored. Considering the studies of [1,5] for the concerned variables of social cohesion and CO₂ emissions, the present study proposed the additional variables of urban population and financial development as followed by [24,36,37,38,39]. Ref. [18] explored the effect of financial development on CO₂ emissions and found that there may be a gradual reduction in CO₂ emissions with financial development.

Urbanization is a process that occurs as a result of infrastructure development, industrialization, and the availability of social services, economic opportunities, and markets in urban areas. This encourages people from remote areas to migrate to cities in search of better opportunities. The extent and density of urban areas determine the degree of urbanization [40]. Urbanization offers opportunities and amenities for a more contemporary lifestyle, which is linked to environmental degradation. Many researchers have discovered a substantial impact of urbanization on environmental pollution, increasing carbon emissions [2,41,42].

The present study follows the model of [1,5] by including an additional variable of financial development along with social concern. Further divided ethnic diversity as the cause of the lack of social cohesion affecting CO₂ emissions. This study focuses on different types of ethnic issues such as religious and linguistic ethnicity as a problem of a lack of cohesion, leading to the problem of CO₂ emissions.

$$\begin{gathered} \ln \left({CO2pc}_{it}\right)=a+\beta \left({Cohesion}_{it}\times \ln \left({GDPpc}_{it}\right)\right)+\delta {Cohesion}_{it} \\ +\pi \ln \left({GDPpc}_{it}\right)+e_{it} \end{gathered}$$

(1)

The current study examines the impact of missing social cohesion in society, specifically through factors such as a lack of intergroup cohesion, civic activism, and gender inequality, on the issue of high carbon emissions. In addition, this study includes the variable of financial development to explore the economic and social aspects of the issue.

4. Data and Methodology

4.1. Data Description

Based on the discussion above about the relationship between social cohesion and carbon emissions (CO₂), this study aims to explore this ambiguous association by including both social and economic variables. Ref. [1] pioneering study highlighted this relationship by including social cohesion and GDP, while [5,20] followed this model by using proxies of social cohesion such as ethnic fractionalization, ethnolinguistic fractionalization, religious fractionalization, cultural divisiveness, and population diversity. This study provides evidence that different measures of cohesion, as income increases, reveal a significant relationship with carbon emissions at a country level. The present study focuses on new dimensions of CO₂ emissions in the context of the social domain and identifies social and economic determinants of CO₂ emissions.

CO₂ emission is measured in metric tons per capita as a dependent variable and the independent variables are civic activism (CA), intergroup cohesion (IC), gender equality (GE), financial development (FD), urbanization (UP), and gross domestic product (GDP) per capita. Civic activism (CA), intergroup cohesion (IC), and gender equality (GE) are used to determine social cohesion [20]. The literature supports many factors that are associated with carbon emissions, but as discussed earlier, a theoretical framework mentioning aforesaid variables in this study led to the development of the model including financial development, urbanization, and economic growth as follows [24,36,37,38,39]. Furthermore, social cohesion is hypothesized as a distinguished variable to affect CO₂ emissions as introduced by [5] and then empirically introduced by [1]. Social cohesion includes intergroup cohesion, civic activism, and gender equality. These indices are used because of the availability of data for the selected countries in these studies.

The relationship between social cohesion and carbon emissions can be explained by examining the impact of civic activism, intergroup cohesion, and gender equality on cohesion. This has been explored by economists and social scientists through theoretical reasoning, empirical evidence, and data collection.

Gender equality is a key aspect of social cohesion that ensures women have the same rights and opportunities as men. Despite the assumption that gender equality and CO₂ emissions are related, the literature does not provide a clear link. However, demand theory supports a negative correlation between the two, and research suggests that including women in the labor force may improve environmental efficiency. Men consume more energy than women, leading to a lower release of carbon dioxide in their daily lives.

Civic activism measures the level of social practices, promotes critical engagement, and is more active with political regimes. These include access to participation in media, civic association, and the means to participate in civic activities such as peaceful affirmation or appeal. Civic activism ensures the transparent and inclusive functioning of public institutions. Intergroup cohesion refers to collaboration among diverse groups in a society. A lack of cohesion can lead to conflicts such as discrimination, violence, and guerrilla activities between groups. In a classification of activism, there is no direct effect on greenhouse gas emissions, rather they work to force political and economic actors to change policies and behaviors in a way that will lead to a decrease in emissions. There are indirect goals, and these kinds of engagements force power regulators, businesses, and policymakers to change their choices and intensify their attempts to reduce greenhouse emissions. Ref. [43]. In summary, social cohesion is measured by variables such as civic activism, intergroup cohesion, and gender equality, which play important roles in the model.

This study explains the empirical evidence of developing countries of two types, i.e., low-income countries and middle-income countries. Countries are selected based on GNI per capita which has been taken from WDI (Atlas method, 2020), where low-income countries are those with $1035 or less, and lower-middle-income countries are those with $1036 to $4045. While collecting the data for the study, we selected forty-six countries whose data were available at a maximum (a combination of lower-income and middle-income countries). Data were taken from the World Development Indicators (WDI) from 1990 to 2014, depending on the availability of data. To investigate the relationship between social cohesion and CO₂ emissions, three indices of social cohesion are collected which are intergroup cohesion, gender equality, civic activism, and the number of other indices that have been considered in the model are urbanization, GDP growth, and financial development.

The variables of this study include CO₂ emissions, GDP growth, financial development index, social cohesion, and urbanization. CO₂ emission is taken in metric tons and these data are collected from the World Development Indicators (WDI). The annual growth rate of GDP per capita is primarily based on consistent local currency. Aggregates are totally based on constant $2010. Gross domestic product per capita product was divided with the population of midyear. GDP is taken in percentages and collected from the World Development Indicators (WDI).

The financial development index is created by following a three-step approach described in the literature, which reduces multidimensional data into a single measure. The variables are (i) normalization of variables; (ii) aggregation of normalized variables into the sub-indices representing a particular functional dimension; and (iii) aggregation of the sub-indices into the final index. Nine indices were created to evaluate the development of financial systems across countries at varying levels of abstraction. Data on financial development were collected from the International Monetary Fund (IMF).

This study uses three proxies to measure social cohesion which are gender equality, intergroup cohesion, and civic activism. These data have been taken from the ISD (Indices of Social Development) by the International Institute of Social Science (2012).

4.2. Two-Step GMM

Using a two-step GMM model, researchers can save data from additional data loss. So, in the case of a balanced pooled data set, for the involved coefficient, the two-step GMM model yields more consistent and efficient estimations [44]. A main feature of the generalized method of moments is to permit more moment conditions than there are parameters to be estimated and parameters to be over-identified. Ordinary least squares on the first differenced in a dynamic panel because of inconsistent parameter estimates, so this opens up the possibility of using instrumental variable estimations using the lagged variables as instruments [45]. The use of internally generating instruments is not allowed in ordinary least-square models and the endogeneity problem is solved by dynamic GMM. The variables from regression that are not correlated with the error term (including lagged variables) can probably be used as valid instruments [46,47,48,49].

The GMM model is commonly used to capture the dynamic of the data, especially when the model is facing the problem of endogeneity. Ref. [50] proposed GMM based on the MM method and encountered the problem of heteroscedasticity and simultaneity bias.

Suppose the static regression model is:

$$y_{it}=a_i+x_{it}^{\prime }\theta +u_{it}$$

(2)

where i = 1, 2, 3, …, N and t = 1, 2, 3,…, T

The $y$ is the independent variable, the $x$ denotes the vector of regressors, and $\epsilon$ is the error term. The $\delta$ is the vector of coefficients of regressors. The dynamic equation of the system GMM model is:

$$y_{it}=a_i+x_{it}^{\prime }\theta +\pi y_{it-1}+u_{it}$$

(3)

Now, take the difference to eliminate the unobserved heterogeneity or individual-specific effect, then the equation:

$$∆y_{it}=∆x_{it}^{\prime }\theta +\pi ∆y_{it-1}+∆u_{it}$$

(4)

This $∆$ shows the difference between the current value and the previous value. Due to the difference, the intercept will be eliminated. Let our model run into a simultaneity problem. The $I$ is the vector of instruments of dimension $(1\times H)$ and for $x(H\ge K)$ which accomplishes the conditions of orthogonality:

$$E\left(∆y_{it-1},∆u_{it}\right)=0$$

(5)

$$E\left(I_{it},∆u_{it}\right)=0$$

(6)

The Equations (5) and (6) show that the instruments and lag value of the dependent variables are independent of the error term. The Equation (4) can be rewritten in generalized form as:

$$∆y=∆\gamma G+∆u$$

(7)

The efficient GMM estimator can be derived from:

$${\widehat{\beta }}_{GMM}=[∆G^{\prime }I(I^{\prime }\omega I{)}^{-1}{I}^{\prime }∆G{]}^{-1}∆G^{\prime }I(I^{\prime }\omega I{)}^{-1}{I}^{\prime }∆y$$

(8)

where I is the instrument matrix and G is the matrix of the independent variables.

4.3. Empirical Bayesian Method (EBM)

The EBM is a nonparametric estimator used as an alternative to conventional quantitative procedures for empirical analysis. The conventional approach neglects the past information, but the empirical Bayes is based on the prior and posterior information. The EBM method uses the prior information in the model which makes it more flexible and reliable. The EBM is good at controlling heterogeneity [51]. The Bayesian estimator is useful when the regressors are endogenous [52]. The EBM is better equipped for data modeling and sensitive to the selection of specifications of the prior distribution [53].

Let us have a regression model of Equation (1) $y_{it}=a_i+x_{it}^T\theta +u_{it}$.

Then, the vector of parameter of pooled regression:

$$\widehat{\theta }=(x_{it}^{\prime }{x}_{it}{)}^{-1}{x_{it}}^{\prime }{y}_{it}$$

(9)

The variance–covariance matrix $COV\left(\widehat{\theta }\right)=\widehat{Q}={\widehat{\sigma }}^2(x_{it}^{\prime }{x}_{it}{)}^{-1}$. The high value of $\widehat{Q}$ means the low precision of parameters. The $(\widehat{Q}{)}^{-1}$ is the indicator of the precision of the $\widehat{\theta }$. In the second step, we estimate the precision-weighted average $V=\left(\right(\widehat{Q}{)}^{-1}{)}^{-1}\times \left[\right(\widehat{Q}{)}^{-1}\widehat{\theta }]$. The weight is assigned to each parameter according to the precision. The third step is to estimate the Bayesian estimates:

$${\widehat{\theta }}^{EB}=\left(\right(\widehat{Q}{)}^{-1}{)}^{-1}\times \left[\right(\widehat{Q}{)}^{-1}\widehat{\theta }+\left(\widehat{Q}{)}^{-1}V\right]$$

(10)

The EBM estimates are more accurate and efficient [54,55].

The results of the two-step system GMM analysis are presented in

Table 1. Two-step system GMM model.

Variables	Model 1	Model 2	Model 3	Model 4
CO2 (−1)	0.9651 *** (15.533)	0.9234 ** (2.0269)	0.5191 *** (5.4420)	0.6570 ** (2.7343)
UP	0.6443 *** (3.4789)	0.4790 *** 4.8914	0.5989 ** (2.8225)	0.5657 ** (2.2935)
FD	0.3111 *** (6.5305)	0.7658 ** 2.6299	0.5215 *** (5.8590)	0.3042 *** (3.8619)
IC	0.1964 ** (2.1204)	----	----	0.1237 *** (4.1925)
CA	----	−0.4631 *** −7.8025	----	−0.4094 ** (−1.9943)
GE	----		0.6291 *** (8.6740)	0.4781 *** (14.050)
GDP	0.6238 ** (2.6071)	0.0621 *** 4.0830	0.3159 * (1.7324)	0.1888 *** (3.0474)
Wald test Joint Significance	253.21 *** [0.0000]	189.33 *** [0.0000]	641.39 *** [0.0000]	346.42 *** [0.0000]
Arellano–Bond AR(2)	0.8351 [0.4612]	1.7632 [0.1824]	1.4932 [0.9463]	1.8768 [0.7713]
Hansen J-test	0.6776 [0.5220]	1.2169 [0.1287]	2.0023 [0.2111]	1.3684 [0.6532]

***, **, and * show the level of significance at 1%, 5%, and 10%, respectively. The t-values are in parenthesis and the p-values are in square brackets.

. It shows that there is a significant impact of the lag value of CO₂ on the current value of CO₂ emissions. This suggests that if greenhouse gases are not controlled, their effects will continue in the future. Urbanization, which is a major determinant of CO₂ emissions from the demand-side economy, has a significant effect on CO₂ emissions. The magnitude of its impact is considerable, with a one percent increase in urbanization leading to a 0.5 percent increase in CO₂ emissions. The United Nations report predicts that by 2050, more than 68% of the population in developing countries will be living in urban areas, which is concerning if carbon storage, utilization, and capture measures are not put in place. Urbanization leads to an increase in energy demand due to the availability of basic infrastructure, which in turn contributes to the problem of CO₂ emissions [19,41,56,57]. Financial development is also a threat to CO₂ emissions which is significant and has a strong magnitude. Financial development is a significant contributor to CO₂ emissions, especially in developing countries where it funds various economic activities. In the early stages of development, these activities generate CO₂ emissions, but as the country matures, it can reduce its carbon emissions. Intergroup cohesion, measured by the indices of social development, is positively related to CO₂ emissions. An increase in intergroup cohesion can lead to a rise in CO₂ emissions, with a coefficient of 0.12% indicating that a 1% increase in intergroup cohesion can cause a 0.12% increase in CO₂ emissions. Intergroup cohesion is measured using data on intergroup disparities, feelings of distrust among different groups, and incidents of discrimination, violence, assassination, and terrorism. A lack of intergroup cohesion can lead to these negative activities increasing in society. To further measure intergroup cohesion, ISD used the data on incidents of violence, incidents of assassination, and incidents of terrorist attacks. Therefore, when all these activities increase in a society, then intergroup cohesion may be absent in a society. Therefore, previous studies such as [1] have confirmed that the absence of intergroup cohesion leads to an increase in CO₂ emissions.

The study found that civic activism has a significant and negative impact on carbon emissions. This suggests that civic activism plays an important role in reducing CO₂ emissions, with a coefficient value of −0.40 which is significant at the one percent level of significance. These findings confirm previous research, which also showed a negative relationship between civic activism and CO₂ emissions. When civic activism increases in an economy, businesses, regulators, and policymakers are motivated to change their behavior and put more effort into reducing greenhouse gas emissions [58,59,60]. The coefficient sign of gender equality is positive which indicates the positive relation with CO₂ emissions. However, our results contradict the previous study and show a negative relation between gender equality and CO₂ emissions based on demand theory [22]. One possible explanation for the positive relationship between gender inequality and CO₂ emissions could be the absence or lack of gender equality in developing countries. High gender inequality could lead to a positive association between CO₂ emissions and the underlying variable. Another factor affecting CO₂ emissions is GDP, which shows a positive and significant contribution to developing economies. This is because increased economic activity in the initial stage of development may have a positive impact on CO₂ emissions, which continues until the stage of development is reached.

In

Table 2. Empirical Bayes method.

Variables	Model 1	Model 2	Model 3	Model 4
CO2 (−1)	0.6180 *** (6.8067)	0.5414 * (1.8280)	0.4842 *** (7.6255)	0.5209 ** (2.0433)
UP	0.4328 ** (2.8376)	0.3326 *** (4.4380)	0.2747 ** (2.5145)	0.4792 *** (3.3037)
FD	0.2726 *** (5.0178)	0.2194 * (1.9183)	0.1865 ** (2.1283)	0.2290 ** (2.0844)
IC	0.1633 * (1.9392)	----	----	0.1448 ** (2.2414)
CA	----	−0.4307 *** (−4.8717)	----	−0.5186 ** (−1.9651)
GE	----	----	0.5684 *** (5.7311)	0.6830 *** (6.6533)
GDP	0.7554 *** (3.1483)	0.6672 *** (16.763)	0.8521 *** (6.1689)	0.2591 ** (2.7399)

***, **, and * show the level of significance at 1%, 5%, and 10%, respectively. The t-values are in parenthesis and the p-values are in square brackets.

, this study confirms the results of the GMM methodology by using the empirical Bayes method for the cross-validity of the above results through the second approach of the Bayes method as it is appropriate for a small sample.

Table 2 supports and reinforces the findings of Table 1, with different values of the same parameters but with similar signs and significance. The results obtained are comparable to those in Table 1, indicating the consistency of the results. This is further supported by

Table 3. Comparison of coefficients of GMM and EBM procedures.

	Model 1		Model 2		Model 3		Model 4
	GMM	EBM	GMM	EBM	GMM	EBM	GMM	EBM
CO2 (−1)	0.9651	0.6180	0.9234	0.5414	0.5191	0.4842	0.6570	0.5209
UP	0.6443	0.4328	0.4790	0.3326	0.5989	0.2747	0.5657	0.4792
FD	0.3111	0.2726	0.7658	0.2194	0.5215	0.1865	0.3042	0.2290
IC	0.1964	0.1633					0.1237	0.1448
CA			−0.4631	−0.4307			−0.4094	−0.5186
GE					0.6291	0.5684	0.4781	0.6830
GDP	0.6238	0.7554	0.0621	0.6672	0.3159	0.8521	0.1888	0.2591

, where all the results are summarized in one table for comparison purposes. Model 4 in particular confirms the observations made earlier about the parameters.

The graphical representation of Model 1 is presented in Figure 1. It can be seen that the graphical representation of Model 1 shows the behavior of the two methodologies—empirical Bayes method (EBM) and GMM. The orange line in the figure represents the impact of GDP on CO₂ emissions in the EBM methodology, which is corrected in terms of magnitude compared to the blue line representing the GMM methodology. It can be observed that the impact of GDP on CO₂ emissions is higher in the EBM methodology, especially in developing countries. This provides further evidence of the validity of the environmental Kuznets curve (EKC) hypothesis at the initial stage.

Figure 2 illustrates the validity of the environmental Kuznets curve (EKC) and the larger impact of GDP on CO₂ emissions in the empirical Bayes method (EBM) as compared to GMM. In Model 2, the impacts of magnitude are higher than in Model 1, and the correction in EBM with a steeper orange slope indicates not only a larger impact of GDP in EBM but also that as GDP increases, CO₂ emissions increase.

Figure 3 shows the results of Model 3, where the variable of gender equality is included. In GMM, gender equality has a larger impact on CO₂ emissions compared to EBM, where its impact is reduced. This confirms the larger impact of GDP on CO₂ emissions in the EBM method. Finally, Model 4, which is the core model of the study, shows the highest impact of both gender equality and GDP on CO₂ emissions, as depicted in Figure 4.

5. Conclusions

This study investigates the relationship between carbon dioxide emissions, social cohesion, financial development, and other control variables including urbanization, GDP per capita, intergroup cohesion, civic activism, and gender equality, for a panel of 46 developing countries from 1990 to 2014. This study uses a two-step generalized method of moments (GMM) to estimate the relationships, and the results show that financial development, intergroup cohesion, and civic activism play significant roles in reducing carbon dioxide emissions. However, economic growth, urbanization, and gender equality have a positive and significant impact.

This study uses three indices of social cohesion, i.e., civic activism, intergroup cohesion, and gender equality, and finds that when social cohesion increases, carbon dioxide emissions decrease.

In conclusion, this study highlights the significant impact of lagged CO₂ values and urbanization on current CO₂ emissions. Financial development, intergroup cohesion, and gender inequality also contribute to emissions. Civic activism plays a crucial role in reducing CO₂ emissions, while GDP positively influences emissions during development stages. These findings emphasize the need for effective carbon reduction measures and sustainable urban planning. Overall, the results of this study suggest that promoting social cohesion and sustainable development practices are essential for mitigating the negative impacts of carbon dioxide emissions.