Financial instability and environmental degradation: Evidence from South Asia

Abstract

This study examines the impact of financial instability (FI) on environmental degradation (ED) along with economic growth (EG), foreign direct investment (FDI), and energy consumption (EC) in five South Asian economies from 1980 to 2021. The study uses a fixed-effect panel model and a two-step system GMM for robust outcomes. The empirical findings demonstrate that FI has a positive and significant effect on ED in South Asian economies. However, the impact of FI on ED varies across South Asian countries. Overall, the impacts of EG, FDI, and EG are positive and significant on ED in South Asia. The effects of EG, FDI, and EG are also heterogeneous for countries in South Asia. However, FDI reduces ED in Pakistan, Bangladesh, and India. Therefore, the study provides several policy recommendations to combat ED in the South Asian region.

Keywords

Financial instability environmental degradation South Asia

Introduction

Since the early twenty-first century, climate changes and global warming have been among the most controversial topics between climate scientists and policymakers.^1,2 It is evident in empirical literature, that one of the primary causes of global warming is increased carbon emissions and other greenhouse gases. In general, these growing carbon emissions and other greenhouse gas emissions harm the environment and have negative economic implications.³ Therefore, researchers, environmentalists, economists, and policymakers have been following to control and reduce these weather causing emissions for the safety of life on Earth. They identified several determinants of these harmful emissions, such as income inequality and economic growth, trade openness, population growth and urbanization, foreign direct investment, energy consumption, financial development, corruption, and institutions.^4–11 The title suggests that the content of the study will focus on two interconnected issues—financial instability and environmental degradation—and will provide evidence to support the link between them in the context of South Asia. The significance of the study lies in its exploration of the relationship between these two issues, and its focus on South Asia, a region with high levels of both financial instability and environmental degradation. By providing evidence for the link between financial instability and environmental degradation in this region, the study can help policymakers and stakeholders better understand the complex interplay between these two issues and develop more effective strategies to address them.

Financial system always remains a significant player to impact the various aspects of the economic-system for the development and growth of economies. Therefore, theoretically financial system may influence ED in the following ways. An organized and developed financial system can afford to invest in environmental friendly projects that reduce ED both immediately and over time; everything else stays the same.¹² Similarly, a financial system may be able to control how much money it lends to companies that increase environmental waste.¹³ In contrast, Sadorsky¹⁴ argued that financial system facilitates the investors to obtain easy loans for the purchase of machinery for use in production, and householders to purchase luxury items, for example, home appliances, air conditioners and traveling vehicles, which contribute to ED.

Although in recent years a plenty of studies conducted in developed and developing countries to explore the various aspects of carbon emissions and macroeconomic variables.^15–19 In response to the claims that countries with developed financial systems are better able to enjoy a comfortable environment than those with developing or newly emerging financial systems.¹³ There are limited studies that explore the relationship between FI and ED. However, the findings remain controversial.^8,9,13,20 Similarly, Nasreen et al.²⁰ have examined the relationship between financial stability and ED in South Asia, which provides a motivation to examine the impact of FI on ED in order to gain more insights into similar regions under current economic conditions. Therefore, to fill this gap, the aim of this study is to investigate the following research questions. How does the FI influence the ED of South Asian economies? Is the impact of FI varying across South Asian countries to influence the ED? Similar to the other studies we will also highlight the impact of FDI, EC and EG on ED in South Asian economies in the present uncertain situations of globe.

Why study the South Asian region? There are several reasons to choose South Asia and current period for this study. Firstly, South Asia account for about 39.49% of Asia's population which is above 24% of the world's population. Secondly, the high population growth rate in South Asia and higher trend of urbanization have direct impact on the ED. Because the higher population and urbanization increase the demand for EC which significantly impacts ED. Third, the increasing trend of ED is above 1.5 metric tons per year in 2019–2020, which is 40% higher than the yearly rate of 0.6 metric tons per year in 1990. Fourth, the study period contains a variety of good and bad phases, which include the Asian crisis 1997–1998, Global financial crisis 2007–2009 and the most recent Covid-19.

Due to the mixed results of previous studies, it is difficult to say whether FI harms the environment or not. In other words, the significance of FI and ED cannot be overstated at this point. Therefore, this research enriches the environmental economics literature in the following manners. In the first place, this study contributes to explore the link between FI and ED, which has limited empirical evidence to the environmental economics literature. Second, this research enriches the environmental economics literature by examining the South Asian economies, which have not been investigated in the previous environmental economics literature. Third, the study adds to the existing literature by unfolding the differential impact of FI on ED along with EG, FDI, and EC across South Asian economies. The nexus between FI and ED are missing in environmental economics literature in South Asian economies for current period to formulate appropriate policy implications to reduce the ED in South Asia, particularly in the present uncertain conditions of the globe. We believe this is the first study to scientifically investigate the influences of FI on ED along with EC, EG and FDI in South Asian economies by using the two-step system GMM framework for the consistent and unbiased estimators. Fourth, the empirical results of our research will improve the present environmental economics literature by offering helpful conclusions and understanding of the relationship between FI and ED in South Asian nations in particular. Finally, our findings are more robust than previous studies because in this study, we use a two-step system GMM approach to deal with endogeneity problem of explanatory variables along with OLS fixed effect regression and panel-corrected standard errors.

The study explores some interesting insights in adding to environmental economics literature. The findings confirm that FI and ED move together in similar direction in South Asia. Therefore, it is noticeable for policymakers to build an appropriate strategy to control the problem of ED in South Asia. Although, a well-established financial system can contribute to more effective decrease in the ED. But the positive relationship between ED and FI indicates that in South Asia financial system prefers to invest in highly paying projects instead of low-paying ecological friendly projects in uncertain financial situations. In line with the previous studies, the impact of EC on ED is positive and statistically significant in South Asian region. However, the role of FDI varies across South Asian economies to impact the ED. The impact of EG is also varying across South Asian countries to influence the ED. The heterogeneity of findings regarding the impact of FI, FDI, EG and EC to influence the ED in South Asia have economic implications for policymakers, environmentalist, and financial analyst to control the climate change in region for the better life of community.

The remaining sections of this study consist of the following: The Section “Literature review and hypothesis development” includes the development of hypotheses based on theoretical and empirical literature. The Section “Study sample, data sources, variable measurements and econometric model” describes the data collection sources, dependent and independent variable measurements, and the econometric model. The Section “Results and discussion” contains results and discussions. The final section provides a conclusion and policy implications.

Literature review and hypothesis development

This segment of the paper examines previous researches on the relationship between FI, EDs, EC, EG, and FDI. This section is divided into four distinct parts in order to provide a concise summary of studies on these interrelationships.

Impact of FI on ED

Khan and Yoon²¹ have uncovered that FI and ED move together in long-run. They collected data for eighty-eight developing countries over the period from 1980 to 2014. Similarly, Shahbaz²² concludes that FI and ED are correlated. He obtained data for Pakistan over the period from 1971 to 2009. Safi et al.²³ have investigated the link between FI and ED. They used data from seven emerging countries from 1995 to 2018. They find short and long-run movement in FI and ED. Yang et al.⁹ have explored the impact of FI on environmental quality for 54 developing economies over the period from 1980 to 2016, by employing penal GMM framework. Empirical results expose that there is a negative and statistically significant relationship between FI and CO₂ emissions. Omri et al.¹² have studied the relationship between EG and CO₂ emissions for a panel data set of MENA region over the period from 1990 to 2011, by using simultaneous equations modeling approach. The study outcomes confirm that unidirectional causality running from financial development to EG in short-run, other things remain stable. Moreover, an organized financial system may decrease the CO₂ emission by investing in ecological friendly projects. On the other hand, financial system may lends easy loans to general public for luxury products, which may increase CO₂ emissions in short-run.¹⁴ Shahbaz et al.⁸ have examined the impact of financial development on CO₂ emissions in South Africa. The study concludes that financial system adds to an economy by funding to buy efficient technology in various sectors to control pollution. In the similar vein,¹³ conclude that financial system impacts environmental quality in short-run. Burki and Tahir²⁴ have examined the causal relationship between financial development and ED in ASEN region over the period from 2001 to 2020 by employing OLS fixed effects, generalized least squares, and two-stage least squares. They conclude that there is one-way causality running from financial development to ED. Nasreen et al.²⁰ have studied the relationship between financial stability and carbon emissions in South Asian economies over the period ranging from 1980 to 2012 by employing multivariate methodology. The study comes to the conclusion that financial stability and ED are caused in Pakistan and Sri Lanka in a single direction. In light of the aforementioned studies, there is a lack of evidence for the association between FI and environmental deterioration in South Asia, particularly in this era. As a result, in order to fill this void, we come to the conclusion that the following hypothesis should be tested.

Hypothesis 1:

An increase in FI and ED in South Asia do have significant relationship with each other.

Impact of EC on ED

One of the most important aspects influencing environmental quality is EC. Several studies have looked into the impact of EC on the ED. Abban et al.²⁵ have examined the association between EC, FDI, economic development, and CO₂ emissions in high, low, middle, and lower-middle-income economies using data from 1995 to 2015. The findings confirm that CO₂ emissions and energy usage are correlated. However, relationship is unidirectional, indicating that causality is running from CO₂ emissions to energy usage. Omri²⁶ has conducted a study in Middle East and North Africa (MENA) region for the period from 1990 to 2011, by using GMM methods. The findings indicate that there is a unidirectional association between EC and CO₂ emissions in MENA region. Lim et al.²⁷ have examined short-run and long-run causal association between oil consumption, economic expansion and CO₂ emissions in the Philippines by employing co-integration, Granger-causality and an error correction model for the period ranging from 1965 to 2012. They conclude that there is a bi-directional relationship between CO₂ emissions and oil consumption in Philippines. Yang et al.⁹ have explored the impact of FI on environmental quality for 54 developing economies over the period from 1980 to 2016, by employing penal GMM framework. Empirical results indicate that there is a positive and statistically significant relationship between EC and CO₂ emissions. Similarly, several other studies investigated the relationship between EC and environmental quality and provide mixed results.^8,13 The evidence regarding the association between EC and ED in South Asia, especially in the current era, is lacking in light of the aforementioned studies. As a result, in order to fill this gap, we arrive at the following hypothesis to test.

Hypothesis 2:

A significant relationship exists between EC and ED in South Asia.

Impact of EG on ED

Several studies have been carried out in recent years to investigate the association between EG and ED and a variety of other factors. As a result, ecological experts have always been fascinated by the relationship between ED and economic expansion. Salahuddin et al.²⁸ have probed the relationship between EG and CO₂ emissions in Kuwait by using the time series data from 1980 to 2013. The study uses autoregressive distributed lag method and their results reveal that EG and CO₂ emissions are positively associated in Kuwait. Omri²⁶ has studied the relationship between CO₂ emissions and economic expansion in Middle East and North Africa region for the period ranging from 1990 to 2011, by employing simultaneous equations models and GMM estimator framework. The study exposes that there is a bidirectional relationship between economic expansion and CO₂ emissions in MENA region. Khan and Yoon²¹ have exposed that FI and EG move together in long-run. They collected data for eighty-eight developing countries over the period from 1980 to 2014. Lim et al.²⁷ have examined short-run and long-run causal association between oil consumption, economic expansion and CO₂ emissions in the Philippines by employing co-integration, Granger-causality and an error correction model for the period ranging from 1965 to 2012. They conclude that there is a unidirectional relationship between CO₂ emissions and economic expansion in Philippines. Yang et al.⁹ have explored the impact of FI on environmental quality for fifty-four developing economies over the period from 1980 to 2016, by employing penal GMM framework. Empirical findings uncover that there is a positive and statistically significant relationship between economic expansion and CO₂ emissions. In light of the aforementioned studies, there is a lack of evidence for the relationship between EG and ED in South Asia, particularly in the present era. In order to address this gap, we will test the following hypothesis.

Hypothesis 3:

An increase in EG leads to an increase in ED in South Asia.

Impact of FDI on ED

The association between FDI and ED has been investigated by various scholars in the environmental economics literature but with mixed findings which is composed of direct, inverse and insignificant impact of FDI on ED. Khan et al.¹⁸ have investigated the relationship between FDI and carbon emissions in 176 economies of the globe. They employed OLS framework, fixed effects and GMM approach to estimate the parameters. They conclude an inverse relationship between FDI and carbon emissions. Khan et al.²⁹ have explored the interrelationship between FDI and carbon emissions by using the data of 190 countries over the period from 1980 to 2018. They employed both static and dynamic approaches and concluded a significant relationship between FDI and carbon emission. They argued that foreign investment has a positive impact on the ED. Zafar et al.³⁰ have investigated the impact of FDI on ED in OECD economies by using the data over the period from 1990 to 2015. They uncovered that there is a positive relationship between FDI and ED. It indicates that an increase in FDI in OECD countries lead to boost the problem of ED. Khan and Yoon²¹ have uncovered that FI and FDI move in opposite direction in long-run. They collected data for eighty-eight developing countries over the period from 1980 to 2014. In light of the aforementioned studies, there is a lack of evidence in South Asia concerning the association between FDI and ED, particularly in this era. As a result, we have decided to investigate the following hypothesis in an attempt to close this gap.

Hypothesis 4:

An increase in FDI has a significant impact on an increase in ED in South Asia.

Study sample, data sources, variable measurements and econometric model

The purpose of this study is to explore the impact of FI on ED (CO₂ emissions) along with FDI, EG, and EC in South Asia by using latest data ranging from 1980 to 2021. There are several reasons to choose South Asia and current period for this study. Firstly, South Asia account for about 39.49% of Asia's population which is above 24% of the world's population. Secondly, the high population growth rate in South Asia and higher trend of urbanization. The higher population and urbanization increase the demand for EC which significantly impacts ED. Third, the increasing trend of ED is above 1.5 metric tons per year in 2019–2020, which is 40% higher than the yearly rate of 0.6 metric tons per year in 1990. Fourth, the study period contains a variety of good and bad phases, which include the Asian crisis 1997–1998, Global financial crisis 2007–2009 and the most recent Covid-19.

The sample selected consists of yearly observations of the following South Asia countries (Pakistan, India, Sri Lanka, Nepal, and Bangladesh) over the period 1980–2021. The data comes primarily from the World Bank's South Asian development indicators (World Bank, 2021). We use CO₂ emissions per metric tons as EDs which is in line with Yang et al.⁹ and FI is measured by using principal component analysis which is consistent with.¹³ Out of other variables EG and FDI are in line with Abban et al.,²⁵ and EC is consistent with Nasreen et al.²⁰ The list of FI four components include LQL, BRM, DCP sector and DCB sectors are in line with.^9,13 The details of proxies summarized in Table 1 with their measurement and sources.

Table 1.

Description of variables.

Variables	Measurements	Sources
Environmental degradation (ED)	Carbon emissions metric tons per capita	WDI
Financial instability (FI)	Aggregate index of financial factors	WDI
Economic growth (EG)	Annual growth rate of gross domestic product	WDI
Foreign direct investment (FDI)	FDI (% of GDP)	WDI
Energy consumptions (EC)	Fossil fuel EC (% of total)	WDI
Liquid liabilities (M3) LQL	Liquid liabilities (% of GDP)	WDI
Broad money (M2) BRM	Broad money (% of GDP)	WDI
Domestic credit to private sector (DCP)	Domestic credit to private sector (% of GDP)	WDI
Domestic credit to banking sector (DCB)	DCB sector (% of GDP)	WDI

Table 1 consists of variables used for FI (LQL, BRM, DCP, and DCB) and along with ED, EG, FDI, and EC.

Financial instability (FI)

In empirical research, Loayza and Rancière³¹ established an indicator of FI by computing the residual standard deviation based on an assessment of the trend of financial development indicators across the study period. This indicates that the index of FI is computed using the standard deviation of the residual of the variable measuring financial development, regressed on its delayed value and trend. Additionally, this indicates that the index of FI measures the degree to which the variable measures financial development. In a similar vein, Eggoh³² applies FI index of cyclical indicators of financial development in his study. This study follows the methods suggested by^9,13,33 to create FI index based on the financial factors. These studies apply principal component analysis which includes a mathematical process to develop an index based on financial indicators by incorporating the issues of high correlation among financial components.

The components to create FI index by using principal component analysis includes DCP sector, DCB sector, LQL and BRM.^9,13 The principal components analysis employs a mathematical methodology to transform a number of correlated variables into a small number of uncorrelated variables known as principal components.³⁴

Table 2 states the principal components analysis results for FI. Table 2 Part-A shows that the first factor's eigen value score is 3.796, while the second, third, and fourth factors’ maximum eigen values are 0.178, 0.253054, and 0.0.002, respectively. The first principal component explains 94.9% of the standard deviation across all variables, while the second, third, and fourth principal components explain 4.5%, 0.06%, and 0.0001% of the overall standard deviation, respectively. Figure 1 shows the scree plot of the principal components analysis based on factor eigen values, which indicates the strength of components that must contribute FI. Table 2 Part-B shows the eigenvectors loadings score of components 1st, 2nd, 3rd, and 4th based on domestic credit to private sector, DCB sector, LQL and BRM similar to.⁹ We use the loading values of component 1st to construct the FI index due to its higher explanatory power and significance than 2nd, 3rd, and 4th components which have negative score for loading.¹³ The final construction of FI index is as follows:

\begin{aligned} FI = (Change in DCP sector \times 0.4999) + (Change in DCB sector \times 0.4965) \\ + (Change in LQL x 0 .5021) + (Change in BRM x 0 .5018) . \end{aligned}

Figure 1.

Scree plot of eigenvalues after principal component analysis.

Table 2.

Principal component analysis for FI.

Principal components/correlation (Rotation: unrotated = principal)
Prat-A: Eigenvalues of matrix
Number	Eigenvalue	Difference	Proportion	Cumulative
First	3.796	3.617	0.949	0.949
Second	0.178	0.153	0.045	0.994
Third	0.0253054	0.0250446	0.0063	0.999
Fourth	0.002	-	0.00001	1.000
Prat-B: Principal components Eigenvectors
Variable	Comp-1^st	Comp-2^nd	Comp-3^rd	Comp-4^th
DCP	0.4999	0.4592	−0.7343	−0.0052
DCB	0.4965	0.5427	0.6774	0.0050
BRM	0.5018	−0.4971	0.0357	−0.7020
LQL	0.5021	−0.4974	0.0255	0.7072

Table 2 presents the results of principal component analysis for the construction of FI index. The components include DCP sector, DCB sector, LQL, and BRM.

Table 3.

Panel unit root analysis.

Level 1(0)
	Levin, Lin & Chu	Im, Pesaran and Shin W-stat	ADF- Fisher	PP-Fisher
CO₂ emissions	7.153	8.052	0.752	0.768
FI	0.847	1.749	8.449	5.099
EG	−3.653***	−5.039***	54.758***	78.084***
FDI	−1.621	−2.298***	25.575***	26.531***
ECs	0.323	−0.059	8.541	9.052
1^st difference 1(1)
CO₂ emissions	−7.577***	−6.011***	58.241***	79.581***
FI	−7.008***	−7.187***	69.510***	92.400***
EG	−15.009***	−15.679***	160.007***	132.659***
FDI	−7.879***	−9.775***	97.686***	136.423***
ECs	−7.082***	−5.425***	50.856***	117.708***

Table 3 contains the results of panel unit root. We use Levin, Lin & Chu, Im, Pesaran and Shin W-stat, ADF-Fisher and PP-Fisher proposition for stationarity check at lever I (0) and at first differences I (I).

The above construction of FI shows that it is an aggregate value of FI measure and coefficient score of selected financial components. Figure 1 is a scree plot of eigenvalues after principal components analysis. In line with Yang, Ali⁹ the scree plot shows the variance explained by the financial factors used in principal component analysis.

Panel unit root

Unit root is used to identify the level of integration of variables which is pre-condition to decide econometric procedure for consistent, unbiased and appropriate estimators. Here we used panel unit root propositions suggested by Levin, Lin & Chu, Im, Pesaran, and Shin W-stat, ADF-Fisher and PP-Fisher. The proposition of Levin, Lin & Chu, Im, Pesaran and Shin W-stat. The results show that some variables are stationary at level (EG, FDI) while some variables are not stationary at level (ED, FI, and EC). However, all the propositions confirm that all the variables are stationary at first difference. The mix order of variables integration allows us to use, panel ordinary least square, GMM procedure, and panel corrected standard errors to obtain consistent, unbiased and appropriate estimators.⁹

The dynamic panel model

The primary aims of this study include investigating the impact of FI on ED along with FDI, EC and EG for five economies of South Asia by using yearly data from 1980 to 2021. For this, we use a two-step system GMM framework and a panel OLS fixed effects model. The orthogonal estimates are used to control parts of the specification in the fixed effects framework. The predicted estimates of the fixed effects setting remove the appropriate means from the cross-sections, the period from the dependent variables, and the exogenous regressors. They then use the specified regression with the discredited data.^9,35 The one of the benefits of fixed effects model is that it reduces the problem of omitted variables, which remain constant over time. On the other end, fixed effects regression remains restricted to deal with first order autocorrelation and heterogeneity in the instruments and residual of model. Therefore, following the previous studies Yang et al.⁹ a generalized method of movement (GMM) is used. We employ a two-step system GMM procedure in predicating the consistent and unbiased parameters for FI to impact ED in South Asian economies. The use of a two-step system GMM is preferable than one-step system GMM framework.³⁶ In addition, GMM deal with the issue of autocorrelation, heteroscedasticity in instrumental variables and in residuals along with the matter of endogeneity in model.³⁷

Therefore, we use the following dynamic model to explore the impact of FI on ED along with EG, FDI, and EC in the South Asian region.

E D_{i, t} = α + β_{1} E D_{i, t - 1} + β_{2} F I_{i, t} + β_{3} E G_{i, t} + β_{4} FD I_{i, t} + β_{5} E C_{i, t} + ε_{i, t}

(1)

In equation (1)

E D_{i, t}

is dependent variable,

E D_{i, t - 1}

show the lagged value of the dependent variable. Where, i represents countries and t shows the time (year). The independent variables include

{F I}_{i, t}

{E G}_{i, t}, {F D I}_{i, t}

and

and {E C}_{i, t}

Arellano-Bond GMM approach

Numerous mathematical issues may arise from the estimation of equation (1) by using simple ordinary least square: Firstly, FI variable assumed endogenous. Due to this fact, causality may run in both sides from FI to ED and vice versa, and these variables may be correlated with the error term of equation (1). Secondly, time-invariant country characteristics refer to fixed effects in cross-sections. The fixed effects remain contained in the error term, which represent the unobserved country-specific effects, $v_{i,}$ and the observation-specific errors, $e_{i, t}$ .

u_{i, t} = v_{i} + e_{i, t}

(2)

Thirdly, the existence of the lagged value

E D_{i, t - 1}

promotes the problem of autocorrelations. To resolve the issue of fixed effects, we tried fixed effects instrumental estimations in the setting of two-stage least square (2SLS). We tried several instruments like trade-openness, imports, exports, and inflation at different times. In the first stage results of 2SLS, we found that the instruments are weak and not appropriate for final analysis. Therefore, we use Arellano and Bond³⁸ GMM approach to proceed further. The use of difference GMM resolves the problem of fixed effects from equation (1).

Δ E D_{i, t} = β_{1} E D_{i, t - 1} + β_{2} F I_{i, t} + β_{3} E G_{i, t} + β_{4} FD I_{i, t} + β_{5} E C_{i, t} + Δ ε

(3)

The above equation is transformed as follows:

Δ E D_{i, t} = α Δ E D_{i, t - 1} + β_{2} Δ F I_{i, t}^{'} + Δ ε

(4)

The transformation of the equation eliminates the fixed country-specific aspect because it does not vary with time. With the use of equation (2) we can express:

u_{i, t} = v_{i} + e_{i, t}

(5)

u_{i, t} - u_{i, t - 1} = (v_{i} - v_{i - 1)} + (e_{i, t} - e_{i, t - 1}) = e_{i, t} - e_{i, t - 1}

(6)

In this case, the first-differenced lagged dependent variable is also used as an instrument with its past information. We follow the recommendation of Arellano and Bond³⁸ to favor the argument in the sense that the use of a one-step method is severely biased. Windmeijer³⁶ recommended two-step robust standard errors instead of a simple two or one-step approach. From this point of view, the system's two-step estimator is more efficient and consistent. In environmental economic literature, various studies apply GMM producers.^3,9

l n E D_{i, t} = ϕ E D_{i, t - 1} + β F I_{i, t}^{'} + (η_{i} + ε_{i, t})

(7)

Therefore, in short, we applied a two-step system GMM approach to test the set hypotheses of this study by using the equation (1). Along with the GMM results, to validate the predictions reported in this study, we used several methods with similar characteristics to verify the results. For example, we also used panel-corrected standard errors because this method accounts for endogeneity, cross-sectional dependency, and heterogeneity.

Results and discussion

Table 4 contains the results of descriptive details which include average, standard deviation, skewness, and kurtosis statistics for each variable. The results uncover that the average value of ED is 0.534 with a standard deviation value of 0.414. The mean value of FI is 0.001 with a standard deviation of 1.948. Similarly, the average value of EG, FDI and EC is 5.188, 0.738, and 45.183 with a standard deviation of value of 1.996, 0.721, and 22.238 respectively. The skewness and kurtosis values are in normal range. The average, standard deviation values of liquid liabilities, broad money, domestic credit to private sector and domestic credit to banking sector are in line with.^9,13

Table 4.

Descriptive analysis.

Variables	Mean	S.D.	Skewness	Kurtosis
ED	0.534	0.414	1.189	4.614
FI	0.001	1.948	1.096	4.238
EG	5.188	1.996	−0.421	4.098
FDI	0.738	0.7.21	1.397	5.414
ECs	45.183	22.238	−0.319	2.052
LQL	45.087	20.636	1.109	4.001
BRM	47.737	18.274	0.772	3.529
DCP sector	29.324	14.803	1.071	4.256
DCB sector	29.166	14.807	1.074	4.255

Source: Authors development by using statistical package and numerical data obtain from WDI.

Table 5 comprises the average trend of ED in South Asian economies (Bangladesh, India, Nepal, Pakistan, and Sir-Lanka). The findings confirm that India is at the top of the South Asian economies with an average value of 1.031 and the second high average value is for Pakistan 0.704 and the lowest value of ED belongs to Bangladesh. The average values of South Asian economies are justified because India is having higher population and higher number of people to use energy, therefore its carbon emission is higher than rest of the economies in the region and so on. Bangladesh is trying to control the carbon emission by investing in green finance in the region.³⁹ The lower value of carbon emissions of Bangladesh is an example for rest of the countries of the region for better environment.⁴⁰

Table 5.

Countries wise average trend of variables in South Asia.

Variables	Bangladesh	India	Nepal	Pakistan	Sri-Lanka
ED	0.245	1.031	0.152	0.704	0.540
FI	−0.559	1.014	0.706	−0.626	−0.536
EG	5.435	6.217	4.526	4.941	4.820
FDI	0.497	0.990	0.178	0.908	1.119
ECs	52.398	60.650	15.034	56.505	41.327
LQL	38.733	58.259	55.578	38.660	34.208
BRM	39.750	58.341	54.121	47.214	39.258
DCP sector	26.537	35.249	33.535	22.643	28.666
DCB sector	26.403	35.209	33.203	22.394	28.578

Source: Authors development by using statistical package and numerical data obtain from WDI.

Table 6 indicates the results of correlation between variables. We report the probability value of correlation coefficients to understand the significance of relationship between variables. The most of the variables, if not all are in relationship and the value of correlation coefficient is not problematic. The positive relationship between ED and FI indicate that both the variables move together in a similar direction, which remains consistent through the statistical analysis.

Table 6.

Pair-wise correlation-based analysis.

Variables	ED	FI	EG	FDI	ECs
ED	1.000
FI	0.486*	1.000
	(0.000)
EG	0.268*	0.237*	1.000
	(0.000)	(0.000)
FDI	0.642	0.371*	0.164*	1.000
	(0.000)	(0.000)	(0.017)
EC	0.459*	0.048	0.199*	0.390*	1.000
	(0.000)	(0.484)	(0.003)	(0.000)

Table 6 indicates the pair-wise correlation among selected variables. The probability is also reported to understand the significance of relationship between variables.

The next phase in analysis is to explore the impact of FI on ED along with EG, FDI, and EC by employing a two-step system GMM framework. In addition, study uses panel regression, and panel corrected standard errors. To follow the results and discussion of study, we will start from the panel regression fixed effects.

Panel regression fixed effects results

Table 7 presents the results for the impact of FI and its factors on the ED of five South Asian countries covering the unique and latest period of 41 years ranging from 1980 to 2021. Therefore, this study is providing the most recent and updated results to formulate an appropriate policy implication to control ED in the region. The panel regression fixed effect results are reported based on the specification of Hausman test and the model fit is good along with its explanatory power which is representing by adjusted R-square. The findings of fixed effect model uncover that domestic credit to private sector, domestic credit to banking sector, liquid liabilities and broad money increase ED in South Asia, which are contradicting the findings of.⁹ However, the findings are corroborated with the results of.²⁰ They uncover an inverse relationship between financial stability and carbon emissions while studying the South Asian economies. The impact of FI is positive and significant on the ED in Pakistan, India, Bangladesh, Nepal, and Sri-Lanka. It is worth noting that how the financial sector is contributing in the environment in South Asia. The contradicting impact of FI on ED in South Asia has implications for policymakers to understand the diversity of regions and economies at their surface level instead of studying the aggregate data across the globe. The findings also support that in South Asian economies financial sector is not financing the ecological friendly projects which leads to increase the ED.

Table 7.

Regression analysis fixed effect model results.

	(1)	(2)	(3)	(4)	(5)
Variables	ED	ED	ED	ED	ED
DCB sector	0.009***
	(0.001)
DCP sector		0.014***
		(0.001)
LQL			0.018***
			(0.001)
BRM				0.029***
				(0.001)
FI					0.076***
					(0.008)
EG	0.105**	0.005	0.033***	0.042	0.003***
	(0.007)	(0.007)	(0.006)	(0.006)	(0.006)
EC	0.001***	−0.001	0.002***	0.042***	0.002*
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
FDI	0.172***	0.174***	0.157***	0.139***	0.157***
	(0.023)	(0.023)	(0.023)	(0.022)	(0.022)
Constant	0.184***	0.183***	0.141***	0.064	0.470***
	(0.052)	(0.053)	(0.052)	(0.053)	(0.052)
Observations	210	210	210	210	210
R-squared	0.565	0.563	0.587	0.610	0.588
Hausman test	0.000	0.000	0.000	0.000	0.000
Number of Id	5	5	5	5	5

Table 7 contains the results for the impact of FI on ED along with financial factors LQL, BRM, DCP sector, and DCB sector), EG, FDI and EC. The parentheses contain robust standard errors and ***indicates 1%, **displays a 5% and *shows a10% level of significance, respectively.

Table 8 displays the results for Pakistan, India, Nepal, Sri Lanka, and Bangladesh employing the panel regression fixed effect model for the impact of FI on ED. The results confirm that relationship between FI and ED is positive and significant in South Asian economies, which is contradicting the findings of.⁹ It indicates that in South Asia FI is positively contributing in ED. The findings suggest that under the situation of FI financial sector prefer to lend at higher rate of return which is not possible in case of lending to ecological friendly projects. However, the role of FI varies across Pakistan, India, Nepal, Sri-Lanka, and Bangladesh to influence the ED. The coefficient of India is 0.315 which indicate that a one percentage increase in FI leads to increase a 3.15% in ED. The second highest rate for change in ED due to FI is for Pakistan. The major reason for these results is the excessive use of fossil energy in Pakistan and India. The findings remain consistent and significant for Nepal, Sri Lanka, and Bangladesh. Our results recommend that South Asian economies should reduce their fossil energy and increase their renewable sources to control the burning issue of environment degradation.

Table 8.

Regression analysis fixed effect model results.

	(1)	(2)	(3)	(4)	(5)
	Bangladesh	India	Nepal	Pakistan	Sri Lanka
Variables	ED	ED	ED	ED	ED
FI	0.066***	0.315***	0.049***	0.180*	0.203***
	(0.006)	(0.031)	(0.007)	(0.100)	(0.013)
EG	−0.001**	−0.003***	0.002	0.001***	0.014*
	(0.004)	(0.013)	(0.003)	(0.012)	(0.008)
EC	−0.002***	−0.005***	−0.004	0.009***	0.002*
	(0.000)	(0.002)	(0.002)	(0.002)	(0.001)
FDI	0.031**	−0.086	0.065	0.008***	0.041
	(0.015)	(0.056)	(0.053)	(0.043)	(0.032)
Constant	0.357***	1.144***	0.151***	0.275	0.455***
	(0.031)	(0.119)	(0.036)	(0.190)	(0.053)
Observations	42	42	42	42	42
R-squared	0.979	0.914	0.875	0.490	0.885
Number of Id	1	1	1	1	1

Table 8 contains the results for the impact of FI on ED along EG, FDI, and EC country-wise (Pakistan, India, Bangladesh, Nepal, and Sri-Lanka). The parentheses contain robust standard errors and ***indicates 1%, **displays a 5% and *shows a 10% level of significance, respectively.

A two-step system GMM results

Table 9 shows the results of a two-step system GMM approach for the impact of FI on ED along with EG, FDI and EC. The post-estimation reveals that there is no problem of over-identification in instruments, and specifications of first and second order autocorrelation are as per the requirement of the two-step system GMM framework. The results uncover that there is a positive and statistically significant association between FI and ED in South Asia, which is contradicting with the results of.⁹ However, our findings are in line with Nasreen et al.²⁰ they reveal a negative relationship between financial stability and carbon emissions in South Asian region. Our study findings are more robust and consistent because we employ the different and advance econometric specification and most updated data in comparisons with.²⁰ It indicates that an increase in FI leads to increase the carbon emissions in South Asian region which is in line with the Hypothesis 1 of the study. However, the sign and significance of the two-step system GMM methods are consistent with the findings of panel regression fixed effects approach. The results express that the people of South Asia do not consider the issue of FI while their EC. The findings explore that under FI situation the financial sector prefer to invest in highly paid projects instead of lower paying ecological friendly projects. However, the findings confirm that the impact of FI to impact ED varies across South Asian economies. The impact of EG on ED varies across South Asian countries. For instance, an increase in EG leads to boost the ED in Pakistan and Sri Lanka which is in line with the Hypothesis 3, whereas the relationship between EG and ED is negative for India, Nepal and Bangladesh which is contradicting with the Hypothesis 3. In line with our Hypothesis 2, the findings confirm that an increase in EC leads to increase the ED except India in South Asian region. The results explain that an increase in FDI enhance the ED in Nepal and Sri Lanka which is consistent with Hypothesis 4, whereas the FDI decreases the ED in Pakistan, India, and Bangladesh which is contradicting with the Hypothesis 4 but support the outcomes of.²¹

Table 9.

GMM estimates.

	(1)	(2)	(3)	(4)	(5)	(6)
	Overall	Bangladesh	India	Nepal	Pakistan	Sri Lanka
Variables	ED	ED	ED	ED	ED	ED
Lagged ED	0.994***	1.029***	0.925***	0.893***	1.009***	0.596*
	(0.020)	(0.164)	(0.053)	(0.086)	(0.042)	(0.304)
FI	0.006***	0.005***	0.040**	0.007***	0.023***	0.071***
	(0.002)	(0.010)	(0.018)	(0.005)	(0.021)	(0.058)
EG	0.001	−0.001***	−0.002***	−0.001	0.005**	0.006***
	(0.001)	(0.002)	(0.003)	(0.001)	(0.003)	(0.006)
EC	0.001**	0.022***	−0.001***	0.016***	0.005***	0.001**
	(0.000)	(0.000)	(0.000)	(0.001)	(0.001)	(0.001)
FDI	0.001***	−0.007***	−0.016***	0.039*	−0.003**	0.006***
	(0.006)	(0.008)	(0.014)	(0.023)	(0.009)	(0.026)
Constant	0.001	0.001	0.141**	0.017	−0.019	0.191
	(0.007)	(0.052)	(0.064)	(0.020)	(0.044)	(0.160)
Observations	205	41	41	41	41	41
Number of Id	5	1	1	1	1	1
AR (1)	0.000	0.011	0.003	0.004	0.030	0.000
AR (2)	0.476	0.481	0.344	0.144	0.848	0.428
Hansen statistics	0.258	0.168	0.786	0.321	0.069	0.516

Table 9 shows the estimation of a two-step system GMM framework to check the impact of FI along with EG, FDI and EC on ED in South Asian countries. We repot the diagnostic for over identification of instruments, first and second order autocorrelation with Hansen statistics, AR (1) and AR (2) respectively. The Hansen value and AR (1) and AR (2) statistics confirm the validity of GMM estimators. ***denotes1%, **indicates5%and*represents10%levelofsignificance, respectively. The robust standard errors are reported in parentheses.

Robustness checks

The basic aim of the research is to identify problems and propose policy recommendations. Therefore, having consistent, appropriate, and unbiased estimators is essential for adequate policy recommendation. Therefore, we try to check the robustness of our results with alternative econometric methods. We first employ difference GMM approach and find consistent estimators (results are not reported to save the space). Like Yang et al.⁹ then we use panel corrected standard error approach to explore the impact of FI on the ED along with EG, FDI, and EC. Table 10 shows that the results of panel corrected standard errors approach are in line with the results of panel regressions fixed effects method and a two-step system GMM method as reported in Tables 8 and 9, respectively. Table 10 indicates that FI has a significant and positive impact on the ED in South Asian countries. It suggests that an increase in FI leads to an increase the ED. Moreover, the sign of EG, FDI, and EC are also consistent with the base line findings to influence the ED in South Asian economies.

Table 10.

panel-corrected standard error (PCSE) estimates.

	(1)	(2)	(3)	(4)	(5)	(6)
	Overall	Bangladesh	India	Nepal	Pakistan	Sri Lanka
Variables	ED	ED	ED	ED	ED	ED
FI	0.064***	0.066***	0.315***	0.049***	0.180*	0.203***
	(0.009)	(0.005)	(0.029)	(0.007)	(0.094)	(0.012)
EG	0.016***	−0.001***	−0.003***	0.002	0.001***	0.014**
	(0.010)	(0.004)	(0.012)	(0.003)	(0.011)	(0.008)
EC	0.005***	−0.002***	−0.005***	−0.004	0.009***	0.002*
	(0.001)	(0.000)	(0.002)	(0.002)	(0.002)	(0.001)
FDI	0.237***	0.031**	−0.086	0.065	0.008**	0.041
	(0.036)	(0.014)	(0.053)	(0.049)	(0.041)	(0.030)
Constant	0.051	0.357***	1.144***	0.151***	0.275	0.455***
	(0.059)	(0.030)	(0.112)	(0.034)	(0.179)	(0.049)
Observations	210	42	42	42	42	42
R-squared	0.556	0.979	0.914	0.875	0.490	0.885
Number of Id	5	1	1	1	1	1

Table 10 contains the results for the impact of FI on ED along EG, FDI, and EC country-wise (Pakistan, India, Bangladesh, Nepal and Sri-Lanka). The parentheses contain robust standard errors and ***indicats1%, **displays a 5% and *shows a10% level of significance, respectively.

Conclusions

The current study explores the association between FI and ED along with EG, FDI, and EC in five developing economies of South Asia from 1980 to 2021. There is limited literature on the relationship between FI and ED, but there is no empirical study that has investigated the said relationship in South Asian economies, for the period covering from 1980 to 2021 and employing panel regression fixed effects, a two-step system GMM approach and panel corrected standard errors procedures. Therefore, the present study has significant contribution in the environmental economics literature, especially for the region which have high rate of carbon emissions.

The results of a two-step system GMM specification uncover a new perspective that an increase in FI has a positive and statistically significant impact on the ED in South Asian economies. The results of FI factors support the positive relationship between FI and EDs in South Asian region. However, the influence of FI on ED varies across South Asian nations. Overall, the impact of EG, FDI, and EC is positive and significant on the ED in South Asia. The impact of EG, FDI, and EC varies across South Asian economies to impact the ED. For example, an increase in FDI leads to decrease the ED in Pakistan, Bangladesh, and India.

The empirical findings of this study have following policy recommendations and suggestions for the concern stakeholders to control the ED in South Asian region in the present uncertain economic situations. First, based on the heterogeneity in aggregate data results and the findings of each country level, we suggest focusing on the individual country to formulate an appropriate strategy to control the ED. The aggregate data base results may mislead policymakers while formulating guidelines to control the environmental degradation. Second, sustainable development and rapid economic growth are top priorities because of South Asia's long-term economic goals. On the other hand, poor environmental quality might make it difficult for the area to achieve its goals for sustainable development. So, as to allow EG to rise without an increase in emissions, South Asia should start public awareness campaigns and carry out the required structural reforms in order to reduce ED. Third, there is a need to encourage the financial sector to invest a maximum fund in ecological friendly projects. Fourth, the government should regulate and advise financial sector to provide subsidies for renewable energy to decrease the use of fossil energy. Fifth, the government should encourage those FDIs which remain beneficial in decreasing the negative impact of ED in the region. Finally, on a priority basis significant steps are required to improve the environment and its sustainability for a long term-period for better life on the planet earth. The focus of this study is limited to South Asian nations, and the future researchers can examine the rest of the developed and developing countries for deeper insights. Furthermore, the model can be extended in future research by adding more variables.

Footnotes

Consent for publication

Not applicable.

Consent to participate

Not applicable.

Data availability

Data is readily available at .

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethics approval

This study follows all ethical practices during writing.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article

ORCID iD

Faisal Abbas

References

Adebayo

Akadiri

Adedapo

, et al. Does interaction between technological innovation and natural resource rent impact environmental degradation in newly industrialized countries? New evidence from method of moments quantile regression. Environ Sci Pollut Res 2022; 29: 3162–3169.

Awosusi

Adebayo

Altuntaş

, et al. The dynamic impact of biomass and natural resources on ecological footprint in BRICS economies: a quantile regression evidence. Energy Rep 2022; 8: 1979–1994.

Ahmad

Tian

X-L

, et al. Chinese Growth and dilemmas: modelling energy consumption, CO₂ emissions and growth in China. Qual Quant 2019; 53: 315–338.

Cho

. Determinants of the downward sloping segment of the EKC in high-income countries: the role of income inequality and institutional arrangement. Cogent Econ Finance 2021; 9: 1954358.

Wang

Zhang

Wang

, et al. How does income inequality influence environmental regulation in the context of corruption? A panel threshold analysis based on Chinese provincial data. Int J Environ Res Public Health 2021; 18: 8050.

Grunewald

Klasen

Martínez-Zarzoso

, et al. The trade-off between income inequality and carbon dioxide emissions. Ecol Econ 2017; 142: 249–256.

Ota

. Economic growth, income inequality and environment: assessing the applicability of the Kuznets hypotheses to Asia. Palgrave Commun 2017; 3: 1–23.

Shahbaz

Tiwari

Nasir

. The effects of financial development, economic growth, coal consumption and trade openness on CO2 emissions in South Africa. Energy Policy 2013; 61: 1452–1459.

Yang

Ali

Nazir

, et al. Financial instability and CO₂ emissions: cross-country evidence. Air Qual Atmos Health 2020; 13: 459–468.

10.

Hashmi

Alam

. Dynamic relationship among environmental regulation, innovation, CO2 emissions, population, and economic growth in OECD countries: a panel investigation. J Cleaner Prod 2019; 231: 1100–1109.

11.

Chen

Xian

Zhou

, et al. Impact of income inequality on CO₂ emissions in G20 countries. J Environ Manag 2020; 271: 110987.

12.

Omri

Daly

Rault

, et al. Financial development, environmental quality, trade and economic growth: what causes what in MENA countries. Energy Econ 2015; 48: 242–252.

13.

Baloch

Meng

Zhang

, et al. Financial instability and CO₂ emissions: the case of Saudi Arabia. Environ Sci Pollut Res 2018; 25: 26030–26045.

14.

Sadorsky

. The impact of financial development on energy consumption in emerging economies. Energy Policy 2010; 38: 2528–2535.

15.

Bhowmik

Syed

Apergis

, et al. Applying a dynamic ARDL approach to the Environmental Phillips Curve (EPC) hypothesis amid monetary, fiscal, and trade policy uncertainty in the USA. Environ Sci Pollut Res 2022; 29: 14914–14928.

16.

Chu

. Determinants of ecological footprint in OCED countries: do environmental-related technologies reduce environmental degradation? Environ Sci Pollut Res 2022; 29: 23779–23793.

17.

Dagar

Khan

Alvarado

, et al. Impact of renewable energy consumption, financial development and natural resources on environmental degradation in OECD countries with dynamic panel data. Environ Sci Pollut Res 2022; 29: 18202–18212.

18.

Khan

Weili

Khan

. Environmental innovation, trade openness and quality institutions: an integrated investigation about environmental sustainability. Environ Dev Sustain 2022; 24: 3832–3862.

19.

Majeed

Mazhar

Samreen

, et al. Economic complexities and environmental degradation: evidence from OECD countries. Environ Dev Sustain 2022; 24: 5846–5866.

20.

Nasreen

Anwar

Ozturk

. Financial stability, energy consumption and environmental quality: evidence from South Asian economies. Renewable Sustainable Energy Rev 2017; 67: 1105–1122.

21.

Khan

Yoon

S-M

. Financial instability and environmental degradation: a panel data investigation. Appl Econ 2021; 53: 6319–6331.

22.

Shahbaz

. Does financial instability increase environmental degradation? Fresh evidence from Pakistan. Econ Model 2013; 33: 537–544.

23.

Safi

Chen

Wahab

, et al. Financial instability and consumption-based carbon emission in E-7 countries: the role of trade and economic growth. Sustain Prod Consum 2021; 27: 383–391.

24.

Burki

Tahir

. Determinants of environmental degradation: evidenced-based insights from ASEAN economies. J Environ Manag 2022; 306: 114506.

25.

Abban

Mensah

. Analysis on the nexus amid CO₂ emissions, energy intensity, economic growth, and foreign direct investment in Belt and Road economies: does the level of income matter? Environ Sci Pollut Res 2020; 27: 11387–11402.

26.

Omri

. CO2 Emissions, energy consumption and economic growth nexus in MENA countries: evidence from simultaneous equations models. Energy Econ 2013; 40: 657–664.

27.

Lim

K-M

Lim

S-Y

Yoo

S-H

. Oil consumption, CO₂ emission, and economic growth: evidence from the Philippines. Sustainability 2014; 6: 967–979.

28.

Salahuddin

Alam

Ozturk

, et al. The effects of electricity consumption, economic growth, financial development and foreign direct investment on CO₂ emissions in Kuwait. Renewable Sustainable Energy Rev 2018; 81: 2002–2010.

29.

Khan

Kim Oanh

, et al. The dynamic interrelationship of environmental factors and foreign direct investment: dynamic panel data analysis and new evidence from the globe. Math Probl Eng 2020; 2020: 1–12.

30.

Zafar

Shahbaz

Sinha

, et al. How renewable energy consumption contribute to environmental quality? The role of education in OECD countries. J Cleaner Prod 2020; 268: 122149.

31.

Loayza

Rancière

. Financial development, financial fragility, and growth. 2004.

32.

Eggoh

. Financial development, financial instability and growth: A reappraisal. Law, Economics and Management Faculty, University of Orleans, 2008.

33.

Klomp

De Haan

. Central bank independence and financial instability. J Financial Stab 2009; 5: 321–338.

34.

Tchamyou

. The role of knowledge economy in African business. J Knowl Econ 2017; 8: 1189–1228.

35.

Baltagi

. Econometric analysis of panel data. Springer, 2008.

36.

Windmeijer

. A finite sample correction for the variance of linear efficient two-step GMM estimators. J Econom 2005; 126: 25–51.

37.

Abbas

Yousaf

Ali

, et al. Bank capital buffer and economic growth: new insights from the US banking sector. J Risk Financ Manag 2021; 14: 142.

38.

Arellano

Bond

. Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Rev Econ Stud 1991; 58: 277–297.

39.

Akomea-Frimpong

Adeabah

Ofosu

, et al. A review of studies on green finance of banks, research gaps and future directions. J Sustain Finance Invest 2021: 1–24.

40.

Sultana

Rahman

Khanam

. Environmental kuznets curve and causal links between environmental degradation and selected socioeconomic indicators in Bangladesh. Environ Dev Sustain 2022; 24: 5426–5450.