Impact Assessment of Climate Mitigation Finance on Climate Change in South Asia

Abstract

Climate change is considered the greatest threat to human life in the 21st century, bringing economic, social and environmental consequences to the entire world. Environmental scientists also expect disastrous climate changes in the future and emphasize actions for climate change mitigation. The objective of this study was to explore the influence of climate mitigation finance on climate change in the region most vulnerable to climate shock, i.e., South Asia, in the period from 2000 to 2019. The panel autoregressive distributed lag model was used to estimate the influence of climate mitigation finance on climate change. The findings of this study demonstrate that, in the long-run, climate mitigation finance has a significant role in mitigating climate change, while in the short-run, climate mitigation finance has an insignificant effect on climate change. The result also shows that, in the long-run, climate change has a negative causal relation with GDP and globalization, but it has a positive causal relationship with energy consumption. The short-term effects of all independent variables are insignificant. Finally, based on the outcome of this study, several policy measures are recommended in order to mitigate climate change.

1. Introduction

Climate change is one of the burning issues of the current century. The World Economic Forum declared climate change and other environmental issues as the top five risks likely to occur over the next ten years [1]. It is widely acknowledged that climate change has adverse effects on water supply, agricultural productivity and growth, health of human beings and animals, ecology, social and economic sectors [2,3]. Gemenne et al. [4] predicted that the effect of climate change will be greater on poor and emerging economies than the rich economies. Climate change caused by human activities aggravates natural disasters worldwide and people of poor economies bear the consequences of these climate irregularities [3]. The release of greenhouse gases (GHGs) increased quickly after industrialization and the extravagant use of fossil fuels, which consequently caused global warming, change in the composition of atmosphere and increased the average global temperature by 1.1 °C to 1.2 °C [5,6]. Climate change causes adverse weather events and water shortages [7]. In last fifteen years, seventy percent of extreme weather events, i.e., floods and droughts have been water related. This percentage of climate change is expected to increase due to the rapid climate change [8].

Houghton [9] asserted that developing economies have a lower fuel efficiency than developed and industrialized countries. Houghton [9] further stated that the unavailability of modern technology and the proportionally high consumption of biomass and coal lead to the inefficient use of fuels, and this inefficient use of fuels results in high GHGs emissions per unit of energy. These developing countries also burn forest for land clearance and burn non-renewable biomass for cooking purposes [10], which aggravates climate change. The GHGs emissions by developing countries are expected to increase and will be in line with the GHGs emission of developed countries due to high populations and economic growth [11]. In order to prevent human-induced damages to the climate and to work toward the treatment of climate damages, the United Nations Framework Convention on Climate Change (UNFCCC) developed a mechanism for climate finance under which industrialized countries agreed to provide climate finance to developing countries. The aim of the climate finance mechanism is to keep the level of greenhouse gases at a level which is not harmful to the climate and to achieve the desired level of GHGs concentration in sufficient time, so as to allow the ecosystem to adjust naturally to climate changes, to ensure unthreatened food production and to achieve sustainable economic development [6]. Any finance provided by local, national and international entities for supporting adaptation and mitigation actions that address climate change is called climate finance [12]. Several studies support the establishment of the climate finance mechanism, such as the study by [13], which supports the establishment of the climate finance mechanism as it emphasizes the decarbonization of the energy sector and states that GHGs emission can be reduced by a transition to renewable energy. The studies by Owusu and Asumadu-Sarkodie [14] and Mohtasham [15] also support the objective of the UNFCCC, as they advocate the transition to environmentally friendly energy production. They also assert that renewable energy has low production costs and a low environmental impact; therefore, their studies support the establishment of the UNFCCC climate finance mechanism.

Ritchie et al. [16] highlighted many sources of GHGs emission and climate changes. Their report states that energy is the largest contributor to the total emission of GHGs, which accounts for 73.2 percent of total emission. The second largest source of emission highlighted by ref. [16] is the agricultural, forestry and land use (land use for use human food and livestock) which accounts for 18.4 percent of total emission. Direct industrial process and waste emit 5.2 percent and 3.2 percent of GHGs, respectively. The OECD [17] reports that the climate mitigation finance was provided to all GHGs-emitting sectors. The OECD [18] reports that the highest amount of mitigation finance was provided to the highest GHGs-emitting sector, i.e., the energy sector, which is 32 percent of total climate finance, and the transport sector, which received the second largest amount of mitigation finance at 14 percent of total climate finance. The findings of the extant studies support this high allocation of climate finance to the transport sector, such as the European environmental protection report [19], which highlighted that the transport sector is still among the highest of the GHGs-emitting sectors. The study by Costa et al. [13] reveals that conventional internal combustion engine vehicles (ICEVs) emit excessive GHGs and battery electric vehicles (BEVs) are environmentally friendly and emit less GHGs, especially when renewable energy has a high share in the total energy mix of a country. The study by Galati et al. [20] demonstrates that the conventional transport sector emits more GHGs, and BEVs emit less GHGs. The climate mitigation finance was allocated to different sectors.

Table 1. List of recipients’ sectors of climate mitigation finance in South Asia.

Sectors	Sub-Sectors
Energy	Energy education/training; Hydro-electric power plants; Oil-fired electric power plants; Energy policy and administrative management; Nuclear energy electric power plants and nuclear safety; Solar energy for centralized grids; Electric power transmission and distribution (centralized grids); Energy generation, non-renewable sources, unspecified; Electric power transmission and distribution (centralized grids); Biofuel-fired power plants; Energy generation, renewable sources—multiple technologies; Retail gas distribution
Transport and Storage	Transport policy and administrative management; Rail transport; Road transport; Water transport; Education and training in transport and storage
Agricultural, forestry, fishing	Forestry education/training; Forestry development; Forestry policy and administrative management; Agricultural education/training; Agricultural development; Agricultural research; Food crop production
Water and Sanitation	Basic drinking water supply and basic sanitation; Water sector policy and administrative management; Waste management/disposal; River basins development; Water resources conservation; Water supply and sanitation—large systems; Education and training in water supply and sanitation; Water sector policy and administrative management
General Environmental Protection	Environmental policy and administrative management; Biosphere protection; Environmental education/training; Biodiversity
Multisector	Disaster Risk Reduction; Urban development and management

Source: [18].

shows the sectors and sub-sectors which are recipients of climate change mitigation finance.

The South Asian region was selected for this study as this region is considered a large emitter of GHGs and the most vulnerable region to climate shocks [21] According to ref. [22], South Asia is the world’s most vulnerable region to climate shocks because a large portion of South Asian countries are located near coastal areas, and their proximity to seas make them more vulnerable to unpleasant weather events and sea-level rise. The World Bank’s report [23] delineated that the people of this region live in ‘new climate normal’, in which cyclones, intensifying heat waves, floods and droughts test the ability of governments, businesses and people to face climate change. A report by the World Bank [23] further highlighted that in the last two decades, 750 million people or half of all South Asian citizens were affected by climate changes, and that climate-related disasters adversely affect the living standards of 800 million people of the region, who already constitute the world’s most vulnerable and poorest population. Therefore, South Asian countries were provided the climate change mitigation finance along with other developing countries by some developed countries and international institutions, which were mandated to work for environmental protection and the prevention of climate change. The list of donors include Australia, Austria, Belgium, Canada, Denmark, Germany, Finland, Switzerland, Norway, the Asian Development Bank, the Green Climate Fund, the Global Green Growth Institute, the Oak Foundation, the Global Environment Facility General Trust Fund, EU Institutions, and multilateral development banks [17]. The review of previous studies presented in the literature review section of this article explored the adverse effects of climate change on human life and economies and analyzed different factors that cause climate change. International communities made efforts to prevent these detrimental climate changes, and one of these efforts is the UNFCCC establishment of climate finance mechanism and the provision of climate mitigation finance to developing countries. Keeping in view the adverse effects of climate change and the allocation of climate mitigation finance, this study was conducted with the aim to explore the effect of climate mitigation finance on climate change in six countries within South Asia.

This article will make a significant contribution to the literature of determinants of climate changes and to the economics of climate changes. Most of the previous studies have investigated the influence of different economic and non-economic variables on climate change, but none of the extant studies examined the influence of climate mitigation finance on climate change. This is the first study to investigate the impact of climate mitigation finance on climate change in six South Asian countries using balanced panel data. In the previous studies of climate change, only a single indicator of climate change was used as a dependent variable. In this study, two indicators, i.e., CO₂ and temperature, were used to construct a climate change index by a principal component analysis. Therefore, this study will bring an addition to the existing stock of literature about climate change and its determinants.

The remaining study proceeds as follows: the next section reviews previous research studies related to this study. The third section explains the whole methodology used in this study. The fourth section presents the results and discussion of this study in detail. The last section presents the concluding remarks and outlines the policy measures regarding climate change mitigation.

2. Literature Review

The literature review of this study is twofold: first, this study reviews the theoretical literature related to the relationship of climate change with the independent variables of this study. Second, this section presents reviews of empirical studies directly or indirectly related to this study. Reviews of empirical studies are divided into reviews of existing studies on climate change for the South Asian region and the existing studies on climate change for other regions.

The pollution halo hypothesis explains the theoretical link between climate change and climate mitigation finance. The pollution halo hypothesis posits that the transfer of environmentally friendly technologies and the best management practices from developed countries to developing countries reduces CO₂ emissions and improves the environment in host countries [24]. The study by Kılıçarslan [25], which tested the pollution halo hypothesis, states that environmentally friendly technologies transferred through foreign direct investment reduces CO₂ emissions in host countries. Climate mitigation finance, as indicated in Table 1, is the transfer of environmentally friendly technology and best management practices from industrialized countries to poor countries; therefore, climate mitigation finance is expected to decrease CO₂ emissions and mitigate climate changes. The theoretical link between climate change and gross domestic product can be found in the environmental hypothesis of the Kuznets curve. This hypothesis postulates the inverted U hypothesis between environmental degradation and economic growth; in the early stage of GDP growth, environmental degradation increases, and after a certain level of GDP growth, environmental degradation starts to decrease as the government and people become more concerned about environmental damage [26,27]. The work of Grossman and Krueger [27] delineates that globalization can have positive, negative and unknown effects on the environment and climate change. They categorized the effects of globalization as scale, technological and composition. Technological effect refers to the effect of technology imports on the economy and environment of developing countries. The transfer of environmentally friendly technologies improve the environment of the imported country and slows down climate change. Globalization can change the composition of an economy, i.e., one sector can expand and another sector can contract. If the contracting sector generates more pollution than the expanding sector, then the total emission in the economy will decrease, and if the expanding sector generates more pollution, than the total emission in the country will increase. Therefore, both positive and negative relationships are expected between climate change and globalization due to this composition effect of globalization [28]. The scale effect of globalization refers to the increase in economic activities due to the increased inflow of foreign direct investment (FDI). The increase in FDI brings industries to the host countries, and these industries cause GHGs emission which result in environmental degradation and climate changes. The pollution halo and pollution haven hypotheses also postulate the relationship of climate change with globalization. The pollution halo hypothesis postulates a negative relationship of climate change with globalization. The pollution halo hypothesis states that emissions in host countries decrease due to the transfer of clean, environmentally friendly and modern technologies to the host countries [29]. Ref. [29] further states that the use of better management practices by multinational enterprises (MNEs) and foreign firms reduces emissions in the host countries. The pollution haven hypothesis postulates the positive causal relation of climate change with globalization [30]. It states that the relocation of dirty sectors from developed countries to developing countries increases emissions and climate change in developing countries [31]. The extant studies have also deduced the linkage between climate change and energy consumption. The adverse effects of energy consumption on climate change, as delineated by the study of Osobajo et al. [32], are the frequent occurrence of heatwaves, prolonged droughts and sea-level rise. The study by Osobajo et al. [32] further states that energy consumption is considered a key cause of global warming and climate change. Elayouty and Abou-Ali [33] delineate the adverse effects of GHGs on the climate. They state that CO₂ is the primary GHG which is emitted from energy consumption.

Abbasi et al. [34] examined the effect of energy consumption and urbanization on CO₂ emissions for eight South Asian countries over the period 1982–2017. Panel cointegration tests were used for the analysis of data. The findings of their study demonstrate that, in the long-run, cointegration exists between independent variables, i.e., urbanization and energy consumption, and the outcome variables of this research study, i.e., CO₂ emission. The study also found significant and positive effects of energy consumption and urbanization on CO₂ emissions. Siddique et al. [35] used panel cointegration test and Granger causality test to estimate the influence of urbanization and energy consumption on CO₂ emissions in South Asia from 1983 to 2013. The finding of their study shows a long-term relation between CO₂ emissions and energy consumption and urbanization, and the results further demonstrate that increase in both energy consumption and urbanization increases CO₂ emissions. Nasreen et al. [36] examined the effects of financial stability and energy consumption on CO₂ emissions in South Asian countries over the period from 1980 to 2012. They used Bound cointegration test and Granger causality test for data analysis and found a positive causal relation between CO₂ and energy consumption, and a negative causal relation with financial stability. Ahmed et al. [37] used the data of selected South Asian countries to study the four drivers of CO₂, i.e., energy consumption, income, population and trade openness from 1971 to 2013. Panel cointegration tests and a fully modified ordinary least square (FMOLS) technique was used to analyze the data. The findings of the study show that energy consumption, trade openness and population have positive effects on CO₂ emissions, whereas income has a negative effect on CO₂ emissions, and the effect of income on CO₂ emissions is consistent with the theory of the environmental Kuznets curve. Ullah et al. [38] studied the nexus between agricultural ecosystem and CO₂ emissions in Pakistan over the period from 1972 to 2014. Johansen cointegration and ARDL tests were used to analyze the data. The findings show that livestock, fertilizer, cereal production, agricultural machinery and some other crops increase the emission of CO₂ in Pakistan. Khan and Ullah [39] investigated the nexus between globalization and CO₂ emissions in Pakistan over the period 1975–2014. Johansen co-integration, ARDL bound testing approach, and variance decomposition analysis were employed for data analysis, and their findings show the positive relationship between globalization and CO₂ emissions in Pakistan. Khan et al. [40] investigated the effects of energy consumption and economic growth on CO₂ in Pakistan from 1965 to 2015 by using the ARDL model and found positive effects of both energy consumption and economic growth on CO₂.

Several studies also examined the determinants of climate change for other regions, such as the study of [41], which examined the effect of energy consumption on CO₂ emissions in MENA countries from 1990 to 2015. Panel quantile regression was used to analyze the data and the results of this study show that increase in energy consumption increases CO₂ emission. Brini [42] studied the impacts of economic growth, renewable electricity consumption and non-renewable electricity consumption on climate change for 16 selected countries of Africa over the period 1980–2014. PMG model was employed in this study and findings indicate the negative influence of renewable energy on climate change, meaning that renewable energy prevents and mitigates climate change, whereas non-renewable energy and economic growth have positive effects on climate change, i.e., adverse effects on climate. Akhmat et al. [43] used the vector autoregressive technique to examine the nexus between energy consumption and climate change in five regions, i.e., Middle East and North Africa, South Asia, Sub-Saharan Africa, East Asia and the Pacific, and the world aggregate data over the period 1975–2011. The study found a long-term relation between energy consumption and climate change in all selected regions. This study further found that unidirectional causality exists between electric energy consumption and climate variables. Akpan and Akpan [44] used the ARDL approach to investigate the influence of energy consumption on climate change in Nigeria from 1970 to 2008 and found that the rapid growth in energy consumption causes a rapid growth in CO₂ emissions. Liu and Bae [45] employed the ARDL technique to examine the effects of industrialization and urbanization on CO₂ emissions in China from 1970 to 2015, and their findings indicate that urbanization and industrialization cause an increase in CO₂ emissions by 1% and 0.3%, respectively.

Few studies examined the nexus of the climate change indicator, i.e., CO₂ with globalization, such as Bu et al. [46], who examined the effect of globalization on climate change for OECD and non-OECD countries collectively and separately for each group over the period 1990–2009. The instrumental variables method was used in this study. The study found that climate change and globalization have positive causal relation when both groups are considered collectively. The results further show that climate change has a negative causal relation with globalization in the case of OECD countries, and these two variables have a positive association in the case of non-OECD countries. Destek [47] investigated the effects of overall globalization and sub-indices of globalization on CO₂ emissions for CEECs over the period 1995–2015. Three econometrics techniques, i.e., the Durbin–Hausman cointegration, augmented mean group and panel DH causality test were used to analyze the data. The findings of their study show that social globalization, economic globalization and overall globalization have a positive association with CO₂ emissions, whereas political globalization has a negative effect on CO₂ emissions. Huo et al. [48] studied the effects of globalization on carbon emissions in the United Kingdom over the period 1970–2019 by applying the quantile on quantile regression (QQR) approach and wavelet coherence (WC) approach. Their findings show the positive relation of carbon emission with overall globalization, economic globalization and coal consumption. Zafar et al. [49] employed continuously updated fully modified ordinary least square (CUP-FM) and continuously updated bias-corrected (CUP-BC) approaches to examine the impact of globalization on CO₂ emissions in selected OECD countries from 1990 to 2014, and their results show the negative effect of globalization on CO₂. Lenz and Fajdetic [50] used the panel cointegration test and Granger causality test for investigating the impact of economic globalization on climate change for a panel of 26 countries from 2000 to 2019, and all these countries were divided into two groups, i.e., above average GDP per capita and below average GDP per capita. They found that trade has a positive association with climate change, meaning that economic globalization increases GHGs emissions, whereas the effect of financial globalization is weak and increases emission of GHGs only in below average GDP per capita countries.

The above reviews of extant studies elaborate the different determinants of climate changes, such as energy consumption, GDP, globalization, FDI, industrialization and urbanization, but none of these studies analyzed the role of climate mitigation finance as a determinant of climate change. The novelty of this study is to investigate the influence of climate mitigation finance on climate change. The second novel aspect of this study is the construction of climate change index from two indicators of climate change, i.e., CO₂ emissions and temperature, unlike the extant studies, where a single indicator of climate change, i.e., CO₂, was analyzed.

3. Materials and Methods

The aim of this study is to investigate the effects of climate mitigation finance on climate change for six South Asian countries from 2000 to 2019. The data of climate mitigation finance for Bangladesh and the Maldives were unavailable. Therefore, these countries were not considered in this study. The study of Dumrul et al. [51] defined climate change as the change in the average weather over a long period of time whether caused by human activity or natural variability. There are many indicators of climate change, such as CO₂ emissions, temperature, precipitation, rise in sea level and ocean acidification [52]. Many studies have used CO₂ emissions as an indicator of climate change, and a study by Leon et al. [53] used average temperature as an indicator of climate change. To authenticate climate change by more data, this study constructed a climate change index by using two indicators of climate change, i.e., CO₂ emissions and average temperature. Other indicators of climate change were not considered for the construction of the climate change index due to an unavailability of complete data of other indicators. Nowadays, researchers often work with large datasets and these large datasets are multi-dimensional. Unit and scale of variables also vary, for example, CO₂ emissions and temperature have different measurement units. Moreover, some variables have high volatility and some variables have low volatility [54]. Due to these reasons, these datasets are difficult to interpret. Researchers need methods and techniques for the reduction of dimensions of large datasets without losing information contained by original data and which also increase the interpretability of new variables. Many techniques are used for the said purpose; however, the principal component analysis (PCA) is a statistical technique which is widely used by researchers. PCA decreases the dimensions of data and increases the interpretability of large datasets with minimum information loss [55]. Therefore, PCA was used to construct the climate change index. The first step is to standardize all variables so that all variables have equal weight in the resultant new variable, i.e., index. The reason behind standardizing all variables prior to constructing the index through PCA is the sensitivity of PCA to the variance of original variables. If the variance of one variable is larger than the variance of another variable, then the variable with the larger variance dominates the other and will produce the biased result. Standardizing the variables can solve this problem and can produce a reliable index. PCA method assumes that variables in the transformed matrix should have as low a correlation as possible. Therefore, the second step of PCA is to estimate the covariance between transformed variables. The covariance measures the relationship between two variables. The third step is to calculate the eigenvectors and eigenvalues of the covariance matrix for the identification of the principal components. The eigenvectors of the covariance matrix show the directions of axes which contain the most information or most variance, and thus are known as principal components. The eigenvector having the highest eigenvalue is called the first principal component. The eigenvalues are the coefficients of eigenvectors which provide the amount of variance shown by each principal component. The fourth step is to arrange the eigenvectors in descending order based on their eigenvalue. In this step, all these components are kept and the lesser significant components are discarded. Finally, the feature matrix is multiplied with standardized original data to obtain a new dataset with reduced dimensions or indexes [56,57].

Any effort or activity which prevents or decreases the emission of greenhouse gases is called climate change mitigation. Increasing energy efficiency of existing equipment, replacing non-renewable energy with renewable energy sources, using new technologies, changing management practices, and changing customer behavior are all examples of mitigation [6,58]. Climate mitigation finance refers to the financing of projects which work toward the stabilization of greenhouse gases concentrations to a level that would obstruct any disastrous human interference in the climate system, or projects which strive to limit or decrease the emission of GHGs [12,58]. The data for the dependent variable and independent variables were extracted from different data sources. These sources include OECD [59], KOF [60] the World Bank [61] and USEIA [62] (2021). Climate change is the dependent variable, and the independent variables are climate mitigation finance measured in the current million US dollar, GDP measured at a 2015-constant US dollar, globalization index and total energy consumption measured in QDBTU. The description of the variables employed in this investigation is provided in

Table 2. Description of the study’s variables.

Variables	Abbreviation	Source	Unit
Climate change index	CCI	Author calculation	Index
Climate mitigation finance	CMF	[59]	Million current US dollar
Globalization	GI	[60]	Index
Gross domestic product	GDP	[61]	GDP measured in constant 2015 billion US dollar
Energy consumption	EC	[62]	QDBTU

The below given expression was specified to estimate the effect of climate mitigation finance and other control variables on climate change:

CCI = ƒ(CMF, GI, GDP, EC)

(1)

where CCI denotes climate change index, CMF represents climate mitigation finance, GI represents globalization index, GDP represents gross domestic product and EC represents energy consumption. The variables of the above expression were then converted into log form to obtain a more dependable and efficient estimation, and Equation (1) was then converted to the below log-log model for achieving the objective of this study:

$$lnCC{I}_{it}={\beta }_i+{\beta }_{1it}CM{F}_{it}+{\beta }_{2it}G{I}_{it}+{\beta }_{3it}GD{P}_{it}+{\beta }_{4it}E{C}_{it}+{\epsilon }_{it}$$

(2)

where i denotes the cross-sectional unit or country, t denotes time period β₀, β₁, β₂, β₃, β₄ and ${\epsilon }_{it}$ denotes intercept, slopes or coefficients of each variables and error term, respectively. The reasons for adding these variables in the model and their expected relation with climate change are delineated below:

Climate mitigation finance. Human activities, such as the use of fossil fuels for transportation, electricity production and heating, release large amounts of GHGs, and these GHGs cause severe climate changes. Production and consumption of energy have massive effects on the climate. Energy consumption is the largest source of GHGs release at a global level [63]. Transportation is another major source of climate change. Use of petroleum fuels for transportation results in an excessive release of warming gases such as black carbon, ozone and carbon dioxide, which consequently cause climate changes [64]. Therefore, more climate mitigation finance was allocated to these two sectors, i.e., energy and transportation. A small amount of climate mitigation finance was also allocated to the other sectors such as agricultural, health and other social infrastructure and services. These sectors have more potential for GHGs emission abatement [18]. Therefore, a negative causal relationship is expected between climate mitigation finance and climate change.

Globalization. According to the work of Grossman and Krueger [27], globalization can have positive, negative and unknown effects on the environment and climate change. Climate change has a negative causal relation with globalization due to the technological effect, and it has a positive relation with globalization due to the scale effect, whereas the relation between these two variables is uncertain due to the composition effect. Moreover, the pollution halo hypothesis postulates a negative causal relation of climate change with globalization and the pollution haven hypothesis postulates a positive causal relation of climate change with globalization.

GDP. The environmental Kuznets curve postulates the inverted U-shaped relationship between climate change and GDP. Environmental degradation starts increasing at the initial level stages of GDP growth and then it decreases at a higher level of economic growth [26].

Energy consumption. Several extant studies, such as those by Abbasi et al. [34], Siddique et al. [35], Ahmed et al. [37] and many more, demonstrate that an increase in energy consumption causes an increase in CO₂ emissions and consequently cause climate change. Therefore, a positive causal relationship is expected between energy consumption and climate change.

Table 3. Summary of literature reviews.

Extant Studies on Climate Change for South Asian Region/Countries
Authors	Sample Area	Title	Technique	Source	Findings
[34]	South Asia/1982–2017	Effect of urbanization and energy consumption on CO2	Panel cointegration tests	Environmental Science and Pollution Research	Positive effect of both variables on CO2
[35]	South Asia/1983–2013	Impact of energy consumption and urbanization on CO2	Padroni cointegration and panel Granger causality tests	Journal of South Asian Studies	Positive effect of both variables on CO2
[36]	South Asia/1980–2012	Financial stability, energy consumption and environmental quality	Bound cointegration test and Granger causality test	Renewable and Sustainable Energy Reviews	Negative effect of financial stability and positive effect of energy consumption on CO2
[37]	South Asia/1971–2013	What drives carbon dioxide emissions in the long-run?	Panel cointegration tests and FMOLS	Renewable and Sustainable Energy Reviews	Energy, trade and population have positive and income has negative effect on CO2
[40]	Pakistan/1965 to 2015	Relationship between energy consumption, economic growth and CO2	FMOLS	Financial Innovation	Positive effects of both variables on CO2
[39]	Pakistan/1975–2014	Testing the relationship between globalization and CO2 emissions	Johansen co-integration, ARDL bound testing approach and variance decomposition analysis	Environmental Science and Pollution Research	Positive relation of CO2 emission with globalization
[38]	Pakistan/1972–2014	Does agricultural ecosystem cause environmental pollution?	Johansen cointegration and ARDL tests	Environmental Science and Pollution Research	Positive effects of all crops on CO2
Extant Studies on Climate Change for Other Regions
[41]	MENA countries/ 1990–2015	Analysis of CO2 emissions and energy consumption	Panel quantile regression	Environmental Science and Pollution Research	Positive effect of energy consumption on CO2
[42]	16 African countries/1980–2014	Renewable and non-renewable electricity consumption, economic growth and climate change	PMG	Energy	Negative effect of renewable energy and positive effect of non-renewable energy on climate change
[43]	5 different regions/ 1975–2011	Does energy consumption contribute to climate change?	Vector autoregressive technique	Renewable and Sustainable Energy Reviews	Energy consumption does affect climate change
[44]	Nigeria/1970–2008	Electricity consumption, carbon emissions and economic growth	ARDL	International Journal of Energy Economics and Policy	Rapid growth in energy consumption causes a rapid growth in CO2
[45]	China/1970–2015	Urbanization and industrialization impact of CO2 emissions in China	ARDL	Journal of Cleaner Production 2018	Urbanization and industrialization cause increase in CO2 emissions
[46]	OECD and non-OECD/1990–2009	Globalization and climate change	Instrumental variable method	Journal of Economic Surveys	Positive effect of globalization climate in non-OECD countries and negative effect of globalization on climate change in OECD countries
[47]	CEECs/1995–2015	Investigation on the role of economic, social, and political globalization on environment	Durbin Hausman cointegration, augmented mean group, panel DH causality test	Environmental Science and Pollution Research	Social, economic and overall globalization has positive effect on globalization and political globalization has negative effect on climate change
[48]	United Kingdom/1970–2019	Recent scenario and nexus of globalization to CO2 emissions	Quantile on quantile regression and wavelet coherence approach	Environmental Research	Positive relation of carbon emission with overall globalization, economic globalization and coal consumption
[49]	Selected OECD countries/1990–2014	The impact of globalization and financial development on environmental quality	Continuously updated fully modified ordinary least square (CUP-FM) and continuously updated bias-corrected (CUP-BC) approaches	Environmental Science and Pollution Research	Negative effect of globalization on CO2
[50]	Two groups of countries: countries above average per capita income and countries below average per capita income/2000–2019	Does economic globalization harm climate?	Panel cointegration test and Granger causality test	Energies	Economic globalization increases CO2 in both groups, whereas financial globalization increases CO2 in a group of countries having income below average per capita

presents a summary of the literature reviews which also supplements the reasons for incorporating all independent variables in the model as determinants of climate change.

Before estimating the causal relationship of climate change with climate mitigation finance and other control variables, cross-sectional dependence tests and unit root tests were conducted. Most economic time series are non-stationary or have problem of unit root. Regressing nonstationary variable on another nonstationary variable produces spurious regression. Unit root tests are used to identify unit root problem in a variable. Researchers use first generation panel unit root tests to identify unit root problem in panel data. First generation panel root tests assumes that the cross-sectional units of the panel are independent of each other, while in real-life, cross-sectional units are correlated. To tackle the correlation problem in cross-sectional units, second generation unit root tests were used. Before using second generation panel unit root tests, cross-sectional dependence test was used to examine the correlation between cross-sectional units (countries). The presence of cross-sectional dependency makes estimators consistent, but causes estimator to be inefficient. Therefore, cross-sectional dependency produces biased standard errors. In the case of countries as cross-sectional units, cross-sectional dependency may occur due to their growing financial and economic integration. Pesaran scaled Langrangian multiplier test [65], Pesaran cross-sectional dependence (CSD) test [66] Breush–Pagan Langrangian multiplier [67] and bias-corrected scale Langrangian multiplier [68] test were conducted for checking cross-sectional dependence. CIPS test introduced by [66] was used to check the unit root in the variables of this study. The Hausman test was carried out before the estimation of PMG-ARDL model. The Hausman test was used to determine which is the more appropriate model between PMG and MG. The null hypothesis of the Hausman test is formulated as the PMG-ARDL, stating that it was the appropriate model, while the alternative hypothesis stated that the MG was the appropriate model.

After checking for cross-sectional dependence and unit roots and conducting the Hausman test, pooled mean group/autoregressive distributed lag (PMG-ARDL) model was used to estimate the long-run and short-run relationship between the regressed and regressors of this study. PMG-ARDL is used when all variables are I(0) or I(1) or all variables have a mixed integration of order I(0) and I(1), but before considering the order of integration, panel data must satisfy the principal assumption, i.e., T must be greater than N for the application of PMG [69]. The extant econometrics literature considers the PMG-ARDL as a superior estimation technique. This model resolves the problem of endogeneity and makes possible the hypothesis testing based on estimated long-run coefficients, which is difficult through the Engle–Granger approach [70]. This model also estimates both short-run and long-run coefficients simultaneously. PMG estimators allow short-run coefficients, intercept and error variance to be heterogeneous, whereas the long-run coefficients are constrained to be homogeneous. Long-run equilibrium relationship between variables was found to be the same across groups due to a few reasons, such as the use of common technologies by all groups, solvency problem and arbitrage.

Finally, the following PMG-ARDL model was utilized to achieve the objective of this study:

$$\begin{array}{l}\Delta CC{I}_{it}={\alpha }_i+{\varnothing }_{i}CC{I}_{i,t-1}+{\beta }_{1i}CM{F}_{i,t}+{\beta }_{2,i}G{I}_{i,t}+{\beta }_{3,i}GD{P}_{i,t}+{\beta }_{4,i}E{C}_{i,t}+\\ {\displaystyle \sum }_{j=1}^{p-1}{\lambda }_{ij}\mathsf{\Delta }CC{I}_{j,t-1}+{\displaystyle \sum }_{j=0}^{q-1}{\delta }_{1ij}\Delta CCM{F}_{i,t-j}+{\displaystyle \sum }_{j=0}^{q-1}{\delta }_{2ij}\Delta G{I}_{i,t-j}+\\ {\displaystyle \sum }_{j=0}^{q-1}{\delta }_{3ij}\Delta GD{P}_{i.t-j}+{\displaystyle \sum }_{j=0}^{q-1}{\delta }_{4ij}\Delta E{C}_{i,t-j}+{\epsilon }_{ij}\end{array}$$

(3)

The results of unit root test and Hausman test suggest using PMG model. Therefore, Equation (2) was converted into panel-ARDL/PMG form, which is given in Equation (3). $\Delta CC{I}_{it}$ represents the first difference of CCI and $\mathsf{\Delta }CC{I}_{j,t-1}$ denotes one lag value of $\Delta CCI$. $CC{I}_{i,t-1}$ denotes error correction term, α_i measures country-specific fixed effect, β₁, β₁, β₃ and β₄ are the long-run coefficients, δ₁, δ₂, δ₃ and δ₄ are the short-run coefficients and ∅_I is the parameter of speed of adjustment from short-run disequilibrium to long-run equilibrium.

4. Results and Discussion

This study empirically investigated the impact of climate mitigation finance on climate change.

Table 4. Descriptive statistics of the data.

Variables	Afghanistan	Bhutan	India	Nepal	Pakistan	Sri Lanka
CCI
Mean	1.184	1.804	4.308	1.408	3.081	3.724
SD	0.131	0.572	0.527	0.211	1.41 × 10−1	0.29
Min	1	1.118	3.618	1.225	2.865	3.411
Max	1.431	2.769	5.086	1.897	3.366	4.27
CF
Mean	115.34	20.76	2670.57	92.34	395.97	114.24
SD	144.74	30.63	2733.96	111.31	517.12	132.28
Min	0.062	0.29	6.45	1.61	1.90	0.88
Max	577.24	126.91	10041.88	442.94	1335.89	464.25
GI
Mean	33.5	33.65	57.8	41.217	52.194	57.564
SD	5.395	6.098	5.288	5.305	2.789	1.97
Min	24	25	46	32.693	45.741	52.223
Max	39	42	63	47.481	54.671	59.819
GDP
Mean	13.7	1.48	1570.00	19.97	225.60	60.95
SD	5.4	0.59	603.00	5.12	54.21	19.64
Min	5.95	0.66	801.00	13.43	146.50	35.56
Max	21.12	2.47	2690.00	30.61	324.40	92.19
EC
Mean	0.081	0.047	21.233	0.088	2.556	0.264
SD	0.054	0.019	6.556	0.037	0.542	0.065
Min	0.015	0.017	12.207	0.05	1.811	0.193
Max	0.161	0.076	31.783	0.17	3.434	0.368

Source: [59,60,61,62].

presents the descriptive statistics of the variables employed in this study. The descriptive statistics highlight the key characteristics of the data series. The series used in study include climate change index, climate change mitigation finance, globalization index, GDP and energy consumption. Table 2 presents the descriptive statistics individually for each country. The mean value of the climate change index of India, Sri Lanka, Pakistan, Afghanistan, Bhutan and Nepal are 4.308, 3.724, 3.081, 1.184, 1.804 and 1.408, respectively. India has the highest mean value and Nepal has the lowest mean value of climate change index. The average value of climate mitigation finance of India, Pakistan, Afghanistan, Sri Lanka, Nepal and Bhutan are 2670.57, 395.97, 115.34, 114.24, 92.34 and 20.76 million US dollars, respectively. India has the highest mean value and Bhutan has the lowest mean value of climate mitigation finance. The mean value of globalization index of India, Sri Lanka, Pakistan, Nepal, Bhutan and Afghanistan are 57.8, 57.6, 52.2, 41.2, 33.7 and 33.5, respectively. India has the highest mean value of globalization index and Afghanistan has the lowest mean value of globalization index. The mean value of GDP of India, Pakistan, Sri Lanka, Nepal, Afghanistan and Bhutan are 1570, 225.60, 60.95, 13.7 and 1.48 billion US dollars, respectively. India has the highest mean value and Bhutan has the lowest mean value of GDP. The average energy consumption of India, Pakistan, Sri Lanka, Nepal, Afghanistan and Bhutan are 21.23, 2.56, 0.27, 0.09 and 0.08 QDBTU, respectively.

Table 5. Findings of principal component analysis.

Correlation Matrix
Variables		CO2	Temp
CO2		1.000	0.55
Temp		0.55	1.000
Component analysis
Component	Eigenvalue	Difference	Proportion	Cumulative
Comp1	1.547	1.093	0.773	0.773
Comp2	0.453		0.227	1.000
Principal components (eigenvectors)
Variable	Comp1	Unexplained
CO2	0.71	0.23
Temp	0.71	0.23

shows the findings of principal components analysis. The first section of this table presents the correlation matrix of the sub-indicators of climate change index, i.e., CO₂ emissions and temperature. The second section of Table 3 presents the eigenvalues of each component, the difference between the eigenvalues of two components, the proportion of variation explained in climate change index by each component and the last column presents the cumulative proportion explained by the two components.

Table 6. Findings of cross-sectional dependence tests.

Variables	Breusch–Pagan		Pesaran Scaled		Bias-Corrected Scaled		Pesaran CD
	Statistics	p-Value	Stat	p-Value	Stat	p-Value	stat	p-Value
CCI	177.64 ***	<0.01	29.70 ***	<0.01	29.54 ***	<0.01	13.06 ***	<0.01
CF	108.37 ***	<0.01	17.05 ***	<0.01	16.89 ***	<0.01	9.92 ***	<0.01
EC	243.57 ***	<0.01	41.73 ***	<0.01	41.57 ***	<0.01	15.57 ***	<0.01
GDP	291.46 ***	<0.01	50.47 ***	<0.01	50.32 ***	<0.01	17.07 ***	<0.01
GI	242.34 ***	<0.01	41.51 ***	<0.01	41.35 ***	<0.01	15.5 ***	<0.01

*** denotes significant at 1% level of significance.

presents the outcome of four cross-sectional dependence tests. The results of all four tests demonstrate that there is cross-sectional dependence among the groups/countries selected for this study. The null hypothesis of cross-sectional independence was rejected in favor of the alternative hypothesis of cross-sectional dependence as the p-values of each test for all variables are less than 0.01.

After checking the cross-sectional dependence, the problem of unit root was checked in the data series. While investigating stationarity/unit root in the panel data, the presence of cross-sectional dependency necessitates the use of a unit root test that is applicable in the presence of cross-sectional dependence. First generation unit root tests are not applicable in the presence of cross-sectional dependence. Therefore, unit root tests of second generation, i.e., CIPS, was used to check the unit root in each series.

Table 7. Findings of a unit root test.

Variables	CIPS
	Level		1st Difference
	Constant	Constant and Trend	Constant	Constant and Trend
CCI	−1.71	−2.00	−4.56 ***	−4.57 ***
CF	−3.27 ***	−3.32 ***	-	-
GI	−2.78 ***	−3.09 ***	-	-
GDP	−1.61	−2.85 *	−3.43 ***	−3.36 ***
EC	−1.30	−2.46	−3.65 ***	−3.62 ***

*** and * represents level of significance at 1% and 10%, respectively.

presents the results of a unit root test. The test results show that the CCI, CCMF and GI are stationary at level and have no unit-root problems, while the variables, i.e., GDP and EC, are non-stationary at level and have unit root. GDP and EC become stationary at the first difference. The results of the unit root test show that there is a mix order of integration, i.e., I(0) and I(1). Therefore, the PMG-ARDL model was used in this study as this model is applicable when some series are of I(0) and some series are of I(1).

Finally, PMG-ARDL model was estimated to achieve the objective of this study.

Table 8. Findings of PMG-ARDL model.

Variables	Coefficient	St. Error		T-Statistics	p-Value
Long-run coefficients
LnCF	−0.01 **	0.003		−2.55	<0.05
LnGI	−0.25 **	0.12		−2.06	<0.05
LnGDP	−0.23 ***	0.04		−5.75	<0.01
LnEC	0.66 ***	0.05		12.29	<0.01
Short-run coefficients
COINTEQ01	−0.36 *	0.21		−1.73	<0.1
D(LCF)	−0.001	0.004		−0.11	0.91
D(LGI)	0.040	0.13		0.30	0.76
D(LGDP)	0.20	0.12		1.59	0.12
D(LEC)	0.09	0.13		0.68	0.50
Hausman Test
Chi-square test value			3.46
p-value			0.48

***, ** and * denotes level of significance at 1%, 5% and 10%, respectively.

presents the results of both long-run and short -run coefficients and coefficient speed of adjustment. The results show that the long-run coefficients are significant at one percent and five percent level of significance, while short-run coefficients are insignificant. In the long-run, the main independent variable of this study, climate mitigation finance, has a significant and negative causal relationship with climate change. This negative relationship is due to the allocation of climate mitigation finance to the use of innovative technology, renewable energy use and using old equipment more efficiently. This result was obtained as expected. The climate mitigation finance is similar to a transfer of environmentally friendly capital and technology from developed countries to developing countries, as it is provided to South Asian countries by industrialized countries for the prevention of GHGs emissions; therefore, this result is corroborated by the pollution halo hypothesis, which states that emission in host countries decreases due to transfer of clean, environmentally friendly and modern technologies to the host countries, as well as the use of better management practices by MNEs and foreign firms [29]. This result is supported by the study of [71], whose authors discussed channels through which climate mitigation finance reduces GHGs emissions in the energy sector. The study highlighted the cost-effective technical options for the abatement of GHGs in the energy sector, which are classified as a transition to a low carbon intensive option and an enhancement of energy efficiency. Transition to low carbon intensive fuels means replacing energy of higher carbon content per unit with energy of low carbon content per unit. Energy efficiency can be improved by both supply and demand. Supply of energy efficiency is improved by improving the equipment of energy generation and improving channels of energy transmission and distribution. The demand side of energy efficiency is improved by using energy-efficient items such electric motors, cooking stoves, lamps and vehicles. Table 1 shows that climate mitigation finance was allocated to these two options and therefore, climate mitigation finance has a negative relation with climate change. Climate mitigation finance was also allocated to other sectors and this negative relation of climate change with climate mitigation finance is also due to a decline in emission by those sectors after their utilization of the climate mitigation finance.

The coefficient of the globalization index is also significant and negative, which demonstrates that globalization mitigates climate changes. This negative relationship of climate change with globalization occurred due to the technological effect of globalization. Technological effect of globalization occurs due to the imports of climate and environmentally friendly technology which mitigates climate change. The negative causal relation of climate change with globalization is supported by the pollution halo hypothesis. Bu et al. [45] also investigated the effect of globalization on climate change and found the same result in the case of OECD countries. This result is also consistent with the findings of Zaidi et al. [72] and You and Lv [73] who found a negative causal relation of CO₂ with globalization. In the long-run, increase in GDP mitigates climate change. The results further show that the long-run coefficient of GDP is negative and significant, which means an increase in GDP mitigates climate change. This finding is corroborated by the second stage of environmental Kuznets curve (EKC) hypothesis. EKC hypothesis states that in the initial stage of economic growth (increase in GDP), GHGs emissions increase, which means that an increase in GDP causes climate change, and after a certain threshold level, economic growth causes a decrease in GHGs emissions, which in turn means that an increase in GDP mitigates climate change. This outcome is consistent with the findings of Polloni-Silva et al. [28], who found a negative and significant effect of GDP on CO₂ in the long-run. This outcome is also consistent with the findings of Bu et al. [46], who found a negative long-run relationship of climate change with GDP. Climate change has a positive and significant causal relationship with energy consumption. The energy generated from fossil fuels causes excessive use of GHGs emissions, resulting in climate change. The same result was also found by Ozturk [74], who investigated the effect of energy use and air quality on climate change. Abbasi et al. [34], Siddique et al. [35] and Ahmed et al. [37] came to similar conclusions.

In the short-run, the sign of all coefficients appeared as expected. The climate mitigation finance mitigates climate change, but the effect is insignificant. In the short-run, climate change has a positive association with globalization and this positive association is due to the scale effect of globalization. The reasons behind this positive short-run relation of climate change with globalization were explained by Zafar et al. [49], who asserted that developing countries need more inflow of FDI and, therefore, retain their environmental regulation relax. In the short-run, GDP has a positive effect on climate change and this finding is in line with the first stage of Kuznets curve, which states that as a country becomes more industrialized and developed, its environment starts to degrade due to an excessive use of natural resources, more generation of pollution, inefficient use of technology and an unawareness about the environmental consequences of economic growth [75]. This outcome is also corroborated by the findings of Polloni-Silva et al. [28]. The short-run relation between climate change and energy consumption is also positive. The coefficient of the ECM term is negative and significant, which indicates the long-run adjustment, i.e., movement toward long-run equilibrium from disequilibrium. Table 5 also displays the result of the Hausman test and the result shows that the PMG-ARDL is the appropriate model.

5. Conclusions

The core objective of this study is to explore the influence of climate mitigation finance on climate change in six countries of South Asia using balanced panel data over the period from 2000 to 2019. The reason behind selecting South Asia for this study is the report made by the World Bank [23], which highlighted the high vulnerability of this region to climate shocks. This study was conducted for six countries within South Asia, and two countries, i.e., Bangladesh and the Maldives, were excluded due to the unavailability of their data of a key variable, i.e., climate mitigation finance. The findings of this study demonstrate that climate mitigation finance plays a significant role in mitigating climate change in both the short-run and long-run. This effect is corroborated by the pollution halo effect hypothesis. The long-run effect of globalization is also corroborated by the pollution halo hypothesis as the findings show that globalization mitigates climate change in the long-run, and the short-run effect of globalization on climate change is corroborated by the pollution haven hypothesis as the findings show that globalization has a positive effect on climate change in the short-run. In the short-run, GDP growth has a positive effect on climate change, indicating that an increase in GDP causes climate change, and this short-run result is consistent with the first stage of EKC. The long-run relationship of climate change with GDP is consistent with the second stage of EKC as the findings demonstrate that an increase in GDP mitigates climate finance in the long-run. Energy consumption has a positive causal relation with climate change in both the short-run and long-run. Finally, this study makes some suggestions for the future course of action regarding the climate change mitigation policy for the whole world in general, and the countries of South Asia in particular. This study found that climate mitigation finance has a significant role in mitigating climate change, which indicates the effective utilization of climate mitigation finance; therefore, more finance should be provided to the initial recipients of climate mitigation finance, i.e., the energy sector, transport and storage, agricultural, forestry, water and sanitation and general environmental protection. Developed countries should continue the provision of climate mitigation finance to developing countries, especially to South Asian countries, as committed under the UNFCCC. The study also concludes that globalization mitigates climate change and the reason behind this result, as identified by the extant literature, is the transfer of renewable energy technology and an adaptation of the energy management practices of developed countries. Therefore, governments of South Asian countries are suggested to import renewable energy technology and replace energy generated by fossil fuels with renewable energy, as the findings of Brini [42] demonstrate that CO₂ emissions has a negative relation with renewable energy.

This study has two limitations and also identified a direction for future research. The first limitation is that this study was conducted for six countries within South Asia and two countries, i.e., Bangladesh and the Maldives, were excluded due to the unavailability of their data of a key variable, i.e., climate mitigation finance. Second, due to the unavailability of data of other variables of the study beyond 2019, the sample period was kept limited to 2019. Finally, this study provides a path to future research: to investigate the effect of climate mitigation finance on climate change in other recipient countries of climate mitigation finance. This study also suggests conducting a study for examining the role of climate adaptation finance in the reduction of the harmful effects of climate change.