Environmental Science and Pollution Research

https://doi.org/10.1007/s11356-020-09464-y

RESEARCH ARTICLE

FDI and environmental degradation: the role of political institutions

in South Asian countries
Samina Sabir 1 & Unbreen Qayyum 2 & Tariq Majeed 3

Received: 1 April 2020 / Accepted: 26 May 2020

# Springer-Verlag GmbH Germany, part of Springer Nature 2020

Abstract
This study examines the impact of foreign direct investment (FDI) on environmental degradation by keeping in view the role of
institutional quality (IQ) in the South Asian region. Second-generational unit root test is used to address the issue of cross-
sectional dependency. Mixed order of integration enabled us to use panel autoregressive distributed lag (ARDL) method to
investigate short-run and long-run association. Moreover, this study also employs Granger causality test to explore causalities
among the selected variables. Empirical results reveal that FDI has a positive and statistically significant effect on environmental
degradation. Institutional quality is measured as rule of law has insignificant impact on ecological footprint both in short run as
well as long run. However, government stability decreases environmental degradation while corruption enhances environmental
hazards tremendously. This study also supports environmental Kuznets curve (EKC) for the South Asian countries. This study
emphasized the role of political institutions in environmental degradation. South Asian countries must focus on the implemen-
tation of environmental laws and regulations by improving the quality of institutions to achieve the goal of sustainable
development.

Keywords FDI . Institutions . Income level . Ecological footprint

Introduction the increase of CO2 emission. More than three quarter of the
greenhouse gas (GHG) emission is the worst outcome of in-
Environmental pollution is one of the most pertinent issues dustrialization and fossil fuel combustion during the period of
widely debated in various policy circles across the globe. 1970–2010. Environmental economists proposed an inverted
Carbon emission is considered as a major anthropogenic con- U-shaped relationship between environmental degradation
tributor in environmental degradation; according to and economic growth, which is also known as environmental
Intergovernmental Panel on Climate Change (IPCC) (2014), Kuznets curve (EKC). At early stages of economic develop-
unplanned economic and population growth owe greatly to ment, environmental degradation increases due to unsustain-
able production processes, but once country achieves a certain
Responsible editor: Nicholas Apergis threshold of income, environmental degradation declines. The
decline in environmental pollution at the last stage of EKC
* Samina Sabir depicts the adoption of cleaner production technology and a
[email protected]
major shift from industrial to services sector (Ozturk and
Acaravci 2010).
Unbreen Qayyum
[email protected] The EKC hypothesis has been extensively tested by
looking at various economic factors those are assumed to be
Tariq Majeed
[email protected]
the key in economic development and contaminating environ-
ment, i.e., international trade, financial development, industri-
1
Kashmir Institute of Economics, University of Azad Jammu & alization, and foreign direct investment (Gökmenoğlu and
Kashmir, Muzaffarabad, Pakistan Taspinar 2016; Tsurumi and Managi 2010; Stern 2003;
2
Pakistan Institute of Development Economics, Islamabad, Pakistan Panayotou 1993). Among others, we cannot negate the role
3
School of Economics, Quaid-i-Azam University,
of foreign direct investment (FDI) inflows in economic
Islamabad, Pakistan growth of developing countries. Most of these countries take
 Environ Sci Pollut Res

various policy measures such as trade liberalization and de- easily be endorsed when environmental rules and regulations
regulation of financial sector to attract FDI inflows to boost are cautiously designed and truly implemented. Lack of law
economic development. FDI creates job opportunities for and order, corruption, and government instability may work in
skilled and unskilled workers and encourages entrepreneur- favor of those polluting firms who are unable to comply with
ship and competitiveness; these are considered as the engines CO2 emission regulations. Thus, it is very important to exam-
of economic growth for most of under-developed countries ine the level of institutional quality while considering the im-
(Hermes and Lensink 2003; Batten and Vo 2009; Reiter and pact of FDI on environment especially in the case of develop-
Steensma 2010; Lee 2013; Gökmenoğlu and Taspinar 2016; ing countries. South Asian region is the highest recipient of
Sabir et al. 2019). An increase in FDI facilitates economic FDI that has resulted in CO2 and other greenhouse gas emis-
growth process not only through capital formation but it also sions along with deforestation. Therefore, the impact of FDI
increases productivity caused by the transfer of technology, and political institutions on environmental degradation needs
managerial skills, and knowledge to modernize economy to be investigated for South Asian region.
through technology and innovations (Sabir et al. 2019; Lee To our knowledge, no one has investigated the impact of
2013). FDI on ecological footprints by taking into account the quality
Along with several positive contributions towards the eco- of political institutions of recipient countries. This study is
nomic development, we find contradicting viewpoints regard- primarily geared at investigating impact of FDI and political
ing the role of FDI in environmental degradation of numerous institutions on ecological footprints of selected South Asian
countries. One strand of literature argues that FDI reduces countries. We contributed in the literature in two ways: first by
CO2 emission by carrying cleaner technology to facilitate pro- examining the role of political institution in shaping the rela-
duction process through the investment in research and devel- tionship between FDI and environmental degradation.
opment sector (Stretesky and Lynch 2009; Jalil and Feridun Second, this study focused on ecological footprints for
2011). While others deny this claim by stating that FDI causes South Asian countries. We preferred ecological footprint as
air pollution through the import of polluting technology. They an indicator of environmental degradation because it is a dy-
support the idea that FDI spurs economic growth by using namic measure of environmental quality than CO2 emissions.
methods that are not ecofriendly (Sabir and Gorus 2019). As Numerous studies have used it as an indicator of environmen-
the low-income countries have contributed less towards global tal degradation (Hassan et al. 2019; Ulucak and Bilgili 2018;
environmental degradation, the international law provides re- Ozturk et al. 2016; Mrabet and Alsamara 2017). As mentioned
laxation in terms of utilizing cleaner technologies for produc- before, the momentous surge of FDI in South Asian countries
tion or the amount of greenhouse gas emissions. This provides and high rate of environmental degradation over last two de-
an additional incentive to foreign investors to increase FDI cades has motivated us to select India, Pakistan, Bangladesh,
inflows into the countries with relatively less strict environ- and Sri Lanka in our sample.
mental regulations, less stringent environmental tax policies, The plan of the rest of study is as follows: “Literature re-
and lower standards for the adoption of cleaner technology view” section provides literature review. “Methodology and
(Seker et al. 2015). Fossil fuel is the main source of energy data” section describes data and methodology while
production in developing countries because cost of the renew- “Empirical results and discussion” section presents empirical
able energy sources is not affordable. Fossil fuel consumption result and discussion. Finally, “Conclusion” section concludes
increases the emission of greenhouse gases; especially CO2 is the study.
a serious threat to the environment we live in. Due to large
FDI inflows, these places are becoming pollution havens and
environmental degradation has increased enormously in last Literature review
few decades (Zhang and Zhou 2016).
The polluting firms maximize profits by investing in coun- The relationship between GDP per capita and environmental
tries where inadequate and patchy environmental regulations degradation has been studied extensively in the last couple of
are in practice. Therefore, this provides a strong rationale for decades. Numerous researchers have investigated the EKC
the existence of strong association between political institu- using time series and panel data (Grossman and Krueger
tions and environmental degradation. It is believed that poor 1991; Jalil and Mahmud 2009; Jaunky 2011; Apergis and
quality of political institutions has a negative impact on envi- Ozturk 2015; Bilgili et al. 2016; Sabir and Gorus 2019).
ronmental regulations by introducing bias in shaping, EKC indicates the scale, composition, and technical effects.
adopting, and implementing stringent environmental laws The scale effect shows how an increase in manufacturing de-
(Welsch 2004). Thus, the level of government stability, de- teriorates environmental quality through carbon and other gas
mocracy, rule of law, and control of corruption in a country emissions in the first stage of the EKC curve. At the second
may have direct or indirect influence on the quality of envi- stage of development, there occurs sectoral shift from
ronment (Abid 2016). CO2 emission control techniques can manufacturing to heavy industries that lead to the further
Environ Sci Pollut Res

environmental degradation. At later stage, sectoral shift en- political institutions while examining the GDP-CO2 relation-
courages adoption of cleaner technologies for the production ship. We have not found any study on South Asian countries
of goods and services, resulting in less pollution and damage that has examined the impact of FDI and political institutions
to environment (Álvarez-Herránz et al. 2017; Balsalobre et al. on environmental degradation. Therefore, this study contrib-
2018; Shahbaz et al. 2019). In the literature, we can find dif- utes in the literature by considering the role of institution and
ferent economic indicators such as energy consumption, trade FDI on environmental degradation by specifically targeting
openness, financial development, gross fixed capital forma- the South Asian region.
tion, population density, labor, urbanization, and technologi-
cal development to test the EKC (see the studies of Suri and
Chapman (1998), Soytas and Sari (2003), Dinda (2004),
Soytas and Sari (2009), Al-Mulali and Ozturk (2015), Baek Methodology and data
(2016), Javid and Sharif (2016), Shahbaz and Sinha (2019)
and many others). To estimate the impact of foreign direct investment and polit-
Many researchers have investigated the impact of foreign ical institutions on environmental degradation for South Asian
direct investment on environmental quality (Nguyen and countries, this study specifies the following econometrics re-
Nguyen 2007; Anwar and Nguyen 2010; Azman-Saini et al. gression:
2010; Lau et al. 2014; Omri et al. 2014; Abdouli et al. 2018;
lnEFit ¼ β° þ β1 lnyit þ β2 lny2it þ β3 lnECONit
Liu and Lin 2019). These studies warn that FDI brings pollu-
tion causing technology that deteriorates the environmental þ β4 lnFDIit þ β5 lnInstit þ eit ð1Þ
quality. This phenomenon is referred to as pollution haven
hypothesis (PHH) (Ajide and Adeniyi 2010; Lan et al. 2012; where i reflects South Asian countries, t is the time period
Lee 2013; Lau et al. 2014; Tang and Tan 2015; Baek 2016). chosen for this study that lies between 1984 to 2019, EFit is the
On the other side, various scholars also highlight that FDI ecological footprint used as a measure of environmental deg-
increases environmental quality by importing clean technolo- radation, yit is the real GDP per capita, y2it is the square of real
gy in host country also known as pollution halo hypothesis GDP per capita, ECON is the energy use in kg of oil equiva-
(Tamazian et al. 2009; Al-Mulali and Tang 2013; Hao and Liu lent per capita, FDIit is the foreign direct investment as a
2015; Mert and Bölük 2016). Aliyu (2005) and Shaari et al. percentage of GDP, Inst is the political institutional variables
(2014) rejected both of the hypotheses, while several other such as law and order, government stability and corruption,
studies advocate inconclusive results (Chandran and Tang and eit is the error term. All variables mentioned in Eq. (1) are
2013; Yıldırım 2014; Shahbaz et al. 2015; Keho 2015). expressed in natural logarithm.
The use of ecological footprint as a measure of environ- Data of GDP per capita (constant $2011 US), energy use
mental degradation is reported by several studies (Wang et al. per capita in kg of oil equivalent, and foreign direct inflows
2013; Uddin et al. 2017; Mrabet et al. 2017). They carried out are retrieved from the World Development Indicators (WDI)
research on the positive relationship between GDP and eco- for the time span of 1984–2019 while data of the political
logical footprints and found it significant. Moreover, environ- institutions such as rule of law, government stability, and cor-
ment also deteriorates in the form of ecological footprint due ruption are taken from International Countries Risk Guide
to trade liberalization, financial development, urbanization, (ICRG), and data of ecological footprint per capita is collected
and energy consumption (Charfeddine 2017). The research from the Global Footprint Network. Ecological footprint var-
by the Figge et al. (2017) highlights positive impact of glob- iable consists of six variables, i.e., built-up land, crop land,
alization on environmental quality for 171 countries through CO2 emission, fishing, forest, and grazing. Ecological foot-
multivariate regression. Social globalization is affecting envi- print variable is constructed using national footprint account-
ronment negatively while political globalization has no effects ing. To verify and support EKC, we expect β1 > 0 and β2 < 0
on ecological footprints. However, contrary to this, globaliza- that confirms inverse U-shaped relationship between GDP per
tion posits positive impact on ecological footprints in South capita and environmental quality in terms of ecological foot-
Asian countries. Additionally, per capita income and natural prints. This implies that initially increase in GDP per capita
resources deteriorate ecological footprint significantly in increases environmental degradation but after achieving cer-
South Asian countries hence degrades environmental quality. tain threshold level of GDP per capita, environmental quality
The existing literature ignored the role of political institu- improves. If the coefficient of square of real GDP per capita is
tions in attracting FDI inflows and thereby its influence on insignificant, then there exists a linear and monotonically in-
environmental quality. There is limited research to examine creasing relationship between GDP per capita and environ-
the impact of institutions on environmental quality (Leitao mental quality captured with ecological footprints. The coef-
2010; Tamazian and Rao 2010; Buitenzorgy and Mol 2011). ficient of FDI (β4) is expected to be positive because foreign
A recent study by Laegreid and Povitkina (2018) has included investors bring technologies that are causing environmental
 Environ Sci Pollut Res

degradation in the South Asian countries. Moreover, coeffi- q q

cient β4 can be positive or negative subject to the level of ΔlnEFit¼ φi ECTit−1 þ ∑ γ 1i Δlnyit− j þ ∑ γ 2i Δlny2it− j ð3Þ
j¼0 j¼0
institutional development of the country. q q
þ ∑ β 3i ΔlnECONit þ ∑ γ 4i ΔlnFDIit− j
This study relies on the autoregressive distributed lag j¼0
q
j¼0
p
(ARDL) method of Pesaran et al. (1999) to estimate þ ∑ γ 5i ΔlnInstit− j þ ∑ γ 6i ΔlnEFit− j þ μi þ uit Where ECTit ¼ lnEFit
j¼0 j¼1
short-run and long-run relationship between GDP per q q q q
−β° − ∑ β1i lnyit− j − ∑ β 2i lny2it− j − ∑ β 3i lnECONit − ∑ β4i lnFDIit− j
capita, FDI, energy use, institutions, and environmental j¼0 j¼0 j¼0 j¼0
q
degradation. It is pertinent to mention that cross- − ∑ β 5i lnInstit− j
j¼0
sectional dependence exists among selected South
Asian countries owing to globalization and trade open- However, φi is coefficient of speed of adjustment towards
ness. Moreover, omitted variable bias, contemporaneous the stable or long-run equilibrium point, and this coefficient
correlation between the residuals, and spatial spillover carries negative sign to assure the convergence towards long-
effects are the main determinants of cross-sectional de- run equilibrium. This study also applies diagnostic test for
pendence (Pedroni 2001). In order to evade the problem reliability and accuracy of the results. This study uses
of spurious correlation between the variables, it is nec- Granger causality test to check out short-run causality among
essary to examine the stationarity of the variables the variables.
(Munir and Riaz 2020). In the presence of cross-
sectional dependency problem, the first-generational unit
root tests are inappropriate to examine the stationarity
of the variables because this leads to large biases in the Empirical results and discussion
estimated results. Additionally, cointegration tests do not
provide us good estimates in the presence of cross- Globalization played very important role in inter-connecting
sectional dependency. Therefore, to fix this problem, the countries across the world through the trade of goods and
study uses cross-sectional dependence test to identify services. Therefore considering the importance of globaliza-
the cross-sectional dependencies in the selected vari- tion and its effects on economies, this study examines the
ables before employing unit root test. We employ cross-sectional dependency among selected South Asian
second-generational unit root test of Pesaran et al. countries using Pesaran dependency test. The results reported
(2001) to check the stationarity of the variable in the in Table 1 indicate the problem of cross-sectional dependency
presence of cross-sectional correlations. Pesaran and in all three models. It implies that shock occurs in any South
Smith (1995) and Pesaran et al. (1999) introduced Asian country can spillover to rest of the selected countries.
pooled mean group (PMG) method to estimate dynamic Moreover, the study also investigates the cross-sectional de-
panel data model. This technique is otherwise known as pendence in each variable, and results are reported in Table 2.
panel ARDL. Panel ARDL technique is used to find out The findings indicate that all variables have the problem of
short-run and long-run relationship between FDI, politi- cross-sectional dependence. This prevents us from using the
cal institutions, and ecological footprints. We specify first-generational unit root test in the presence of cross-
the following ARDL equation to estimate the long-run sectional dependency problem.
relationship between the chosen variables. Thus, the study is relying on second-generational unit root
test of Pesaran et al. (2001) to determine stationarity of vari-
q q ables. The CIPS and CADF statistics are reported in Table 3. It
lnEFit ¼ β° þ ∑ β1i lnyit− j þ ∑ β2i lny2it− j is found that natural logarithm of ecological footprints (lnef),
j¼0 j¼0
natural logarithm of GDP per capita (lnGDP), energy con-
q q
þ ∑ β3i lnECONit þ ∑ β4i lnFDIit− j sumption (lnECON), law and order (LO), and corruption
j¼0 j¼0 (COR) have unit root problem at level, but these variables
q p are stationary at first difference. Therefore, these variables
þ ∑ β5i lnInstit− j þ ∑ β6i lnEFit− j þ μi þ uit ð2Þ are said to be I(1). However, square of lnGDP, lnFDI, and
j¼0 j¼1
government stability (GS) are stationary at level, and are
I(0). This implies that all variables have mix order of integra-
where γ1i is the unknown parameter of lagged dependent tion I(1) and I(0). Thus panel ARDL technique is used to
variable (lnEFit − j), while β1i, β2i, β3i, β4i, β5i, and β6i are the estimate the long-run and short-run cointegrating relationship
unknown parameters of variable of interest for instance GDP among the variables using Bound test method, and the results
per capita, square of GDP per capita, energy use per capita, are presented in Table 4. Maximum lags are determined by
FDI, and political institutions, respectively. We transform and Schwartz information criteria (SIC). Table 4 shows that
reparametrize Eq. (2) into error correction model as follows: lnGDP per capita has positive and significant impacts on
Environ Sci Pollut Res

Table 1 Cross-sectional dependency test results for each model

Models Pesaran statistics with p values Average value of off-diagonal elements

1. ln(EF) = f [ln(GDP), (lnGDP)2, ln(ECON), ln(FDI), LO] 22.694 (0.001) 0.211

2. ln(EF) = f [ln(GDP), (lnGDP)2, ln(ECON), ln(FDI), (GS)] 23.664 (0.001) 0.244
3. ln(EF) = f [ln(GDP), (lnGDP)2, ln(ECON), ln(FDI), COR] 23.698 (0.001) 0.242

Author’s own calculation

environmental degradation in the short run (SR) and long run heavy carbon emissions and other hazardous gases. Further,
(LR) whereas square of lnGDP has negative and statistically FDI inflicts environmental damages owing to the use of old
significant impact on environmental degradation in both LR methods of production.
and SR. FDI affects environmental degradation (lnef) positively in
This allows us to conclude that EKC holds in the South both SR and LR. Government stability has positive and sta-
Asian countries. This finding is consistent with study of Sabir tistically significant impact on environmental degradation in
and Gorus (2019), respectively. These results are analogous the LR, but it has insignificant effect in the SR. It is indeed
with previous findings of different studies (Ozturk and logical that government stability does impact environment
Acaravci 2010; Saboori et al. 2012; Shahbaz et al. 2015; with lags due to effective implementation of policies. If there
Javid and Sharif 2016). Energy consumption has positive exists government instability, then reversal of policies occurs
and statistically significant impact on the environmental dete- that impacts the economy badly.
rioration in South Asian countries both in SR and LR. This Further, we replace the government stability with corrup-
implies that environmental quality decreases with the rise in tion to evaluate its impact on environmental quality by taking
energy consumption. This finding is akin with Heidari et al. other variables as fixed. The results are reported in the column
(2015), Jalil and Feridun (2011), and Ozturk and Acaravci 4 of Table 4. Our results reflect the harmony and coherence
(2013). There is a need to swiftly switch on the renewable with our previous results. The table shows that EKC holds
energy sources to enhance energy efficiency in these with the addition of new variable, i.e., corruption both in the
countries. LR and SR, respectively. Energy consumption and FDI are
FDI has positive and statistically significant effect on envi- causing environmental degradation in the South Asian
ronmental deterioration (EF). This region is the greater recip- countries in the short run and long run. As Dasgupta and De
ient of FDI particularly in the energy and infrastructure pro- Cian (2016) mention that corruption leads to environmental
jects. FDI brings technology that damages environment due to degradation in developing countries owing to sub-optimal use
of natural resources and inefficiencies. Bureaucrats are always
susceptible to accept bribes and allowing activities that are
Table 2 Cross-sectional dependency test results for each series damaging the environment. Thus corruption directly or indi-
Variables Pesaran CD Conclusion rectly impacts environment harmfully. ECM carries negative
sign in all estimated models that proved the long-run relation-
ln(EF) 56.230*** Cross-sectional dependence exists. ship among chosen variables.
(0.000)
We have estimated the cointegrating relationship between
ln(GDP) 57.960*** Cross-sectional dependence exists.
FDI, institutions, and ecological footprint for each South
(0.000)
Asian country by employing ARDL method. Each selected
ln(GDP)2 58.104*** Cross-sectional dependence exists.
(0.000) country satisfies the basic diagnostic tests of autocorrelation,
ln(ECON) 47.511*** Cross-sectional dependence exists. homoscedasticity, functional form, and normality. The results
(0.000) of diagnostic tests are reported in Table 5. Additionally, re-
ln(FDI) 49.232* Cross-sectional dependence exists. sults of SR dynamics are provided in Table 6 for each selected
(0.080) South Asian country. Results approved the existence of LR
LO 38.910*** Cross-sectional dependence exists. cointegrating association between FDI, energy consumption,
(0.000) institutions, and environmental degradation in terms of eco-
GS 31.044**** Cross-sectional dependence exists. logical footprints. EKC holds for Bangladesh, Pakistan, and
(0.000)
Sri Lanka while U-shaped relationship holds for India.
COR 22.922*** Cross-sectional dependence exists.
(0.000)
Law and order improves environmental quality in the SR in
Bangladesh, India, Pakistan, and Sri Lanka. While govern-
** and *** indicate 1 and 10% level of significance ment stability and corruption have deteriorates environment
 Environ Sci Pollut Res

Table 3 Panel unit root test results

CIPS CADF

Variables Lags Constant Difference Constant Difference

ln(EF) 1 − 1.979 − 6.186*** − 1.675 − 3.482*** I(1)

2 − 1.980 − 6.316*** − 1.881 − 2.494* I(1)
ln(GDP) 1 − 2.174 − 4.499*** − 2.167 − 3.472*** I(1)
2 − 2.291 − 4.499*** − 2.175 − 2.975*** I(1)
ln(GDP)2 1 − 4.508*** − 3.185*** I(0)
2 − 4.508*** − 3.215*** I(0)
ln(ECON) 1 − 2.075 − 5.112*** 2.206 − 3.372*** I(1)
2 − 2.244 − 5.112*** 2.182 − 2.770*** I(1)
ln(FDI) 1 − 3.480 − 5.680*** − 3.883 − 4.748*** I(1)
2 − 3.564 − 5.603*** − 3.166 − 3.841*** I(1)
LO 1 − 2.634 − 5.433*** − 2.589 − 4.826*** I(1)
2 − 2.501 − 5.433*** − 2.435 − 3.361*** I(1)
GS 1 − 5.814*** − 2.903*** I(0)
2 − 5.814*** − 2.919*** I(0)
COR 1 − 1.886 − 4.899** − 1.888 − 3.522*** I(1)
2 − 1.900 − 4.553*** − 1.840 − 3.976*** I(1)

Author’s own calculations

**
The level of significance at 5%
***The level of significance at 1%

of all selected South Asian countries. This helps us to draw the Conclusion
conclusion that weak political institutions are solely responsi-
ble for the environmental damages in the South Asian coun- This study has investigated the impact of FDI on envi-
tries. Other two variables of interest are energy consumption ronmental degradation by taking into account the insti-
and FDI which have damaged the environment in the South tutional quality. We have controlled the GDP per capita,
Asian countries. ECM carries negative sign in all estimated square of GDP per capita, and energy consumption to
models that proved the long-run relationship among chosen understand the environmental degradation in four South
variables. We have estimated the cointegrating relationship Asian countries (India, Pakistan, Sri-Lanka, Bangladesh)
between FDI, institutions, and ecological footprint for each between the time span of 1984–2019. This study has
South Asian country individually employing ARDL method. used ecological footprint as a unique proxy for environ-
Finally, we estimated the casual relationship between mental degradation. To address cross-sectional depen-
political institutions, FDI, and ecological footprints using dency issue, second-generational unit root test of
heterogeneous panel causality process. The results are Pesaran et al. (2001) is used to find out the order of
displayed in Table 7. There exists a bidirectional causality integration of selected variables. We found that vari-
between GDP per capita and environmental degradation ables have mixed order of integration; therefore, panel
which confirms that economic development coupled with ARDL is applied to explore long-run relationship be-
industrialization enhances natural resource depletion and tween FDI, political institutions, and ecological foot-
reduces the bio-capacity of the environment while esca- prints. The results confirm the presence of environmen-
lating ecological footprints. Moreover, bidirectional cau- tal Kuznets curve (EKC) in short run as well as long
sality is observed between FDI and ecological footprints. run. This implies that at the initial level of development,
This also ensures that foreign direct investment has posi- ecology gets ruined owing to the emission of carbon
tive impacts on environmental degradation in terms of EF. dioxide, sulfur oxides, and other gases in the environ-
However, uni-directional causality has been seen from ment, but after achieving the threshold level of income,
energy consumption to EF, EF to law and order, govern- environmental quality improves due to the adoption of
ment stability to ecological footprints, and corruption to cleaner energy and efficient infrastructure technologies.
ecological footprints. Moreover, it is witnessed that energy consumption and
Environ Sci Pollut Res

Table 4 Panel ARDL estimation Table 5 Diagnostic tests

Variables Model 1 Model 2 Model 3 Model Model Model Model Model Model
1 2 3 1 2 3
ln(GDP) 1.136** 1.425** 1.286** Bangladesh India
(0.025) (0.042) (0.035) Serial correlation 0.171 0.280 0.210 0.130 0.045 1.287
ln(GDP)2 − 0.078** − 0.096** − 0.091*** (0.603) (0.711) (0.621) (0.890) (0.950) (0.311)
(0.046) (0.024) (0.015) Functional form 0.901 2.450 2.860 2.563 2.630 2.275
ln(ECON) 1.203*** 1.240*** 1.290*** (0.334) (0.174) (0.221) (0.815) (0.217) (0.250)
(0.000) (0.000) (0.000) Normality 0.021 0.187 0.010 1.567 0.360 0.388
ln(FDI) 0.031*** 0.030*** 0.033*** (0.985) (0.967) (0.995) (0.457) (0.229) (0.690)
(0.000) (0.000) (0.000) Heteroscedasticity 0.215 0.249 0.138 0.203 2.147 0.210
LO − 0.004 (0.637) (0.567) (0.550) (0.367) (0.361) (0.850)
(0.760) Pakistan Sri-Lanka
GS − 0.030** Serial correlation 4.218 2.891 1.650 0.167 0.189 0.122
(0.025) (0.623) (0.664) (0.373) (0.856) (0.814) (0.507)
COR 0.013*** Functional form 0.536 0.217 0.255 0.113 0.102 0.142
(0.012) (0.460) (0.509) (0.885) (0.760) (0.718) (0.638)
Constant 2.631** 1.995*** 2.227* Normality 1.242 1.247 2.661 1.229 0.891 1.185
(0.037) (0.052) (0.053) (0.414) (0.205) (0.302) (0.537) (0.704) (0.691)
Threshold level of income 7.770 8.661 6.621 Heteroscedasticity 0.260 0.203 0.766 0.657 1.351 0.220
Short-run ECM (0.259) (0.536) (0.706) (0.315) (0.395) (0.371)
ECT − 0.337** − 0.308** − 0.318**
(0.020) (0.022) (0.032) Author’s own calculations
Δln(GDP) 1.048** 0.644** 0.971*
(0.010) (0.048) (0.083)
Δln(GDP)2 − 0.074** − 0.050* − 0.071**
institutional quality on environmental degradation for
(0.025) (0.067) (0.039)
each selected country separately. This study reports EKC
Δln(ECON) 0.010* 0.064* 0.053*
for Bangladesh, Pakistan, and Sri Lanka while U-shaped
(.095) (0.067) (0.074)
relationship between GDP and EF follows in India. In
Δln(FDI) 0.001* 0.001* 0.031*
addition, energy consumption and FDI are tarnishing en-
(0.085) (0.086) (0.055)
vironmental quality in all selected South Asian countries.
ΔLO − 0.010
Further, law and order has negative impact on environ-
(0.599)
mental degradation in India and Sri Lanka while it is
ΔGS − 0.004
insignificant for the rest of two countries. Government
(0.390)
stability has negative effects on environmental quality
ΔCOR 0.010*
while corruption increases environmental degradation in
(0.056)
all the South Asian countries.
Author’s own calculations. ***, **, and * represents 1%, 5%, and 10% This study has also examined the causality relation-
level of significance, respectively ship between variables using Granger causality test. We
have observed the bidirectional causality between GDP
and ecological footprints. Bidirectional causality has
FDI deteriorate environmental quality in selected South been noticed from FDI to ecological footprint and from
Asian countries. FDI increases in the South Asian coun- ecological footprint to FDI. One-way causality is found
tries at the cost of contaminating environmental quality from energy consumption to EF, ecological footprints to
both in short run and long run. Unfortunately, institu- law and order, GS to EF, and corruption to EF. In
tional quality measured as law and order has no signif- short, FDI is important for South Asian countries to
icant impact on environment in short run and long run. enhance socioeconomic development of the country at
However, government stability reduces environmental the same time it impacts the environmental quality.
degradation in the South Asian countries while corrup- There is a dire need to implement and regulate environ-
tion triggers environmental degradation. mental laws in South Asia by moving towards clean
Furthermore, this study has investigated the impacts of and environment-friendly technologies for production
GDP per capita, energy consumption, FDI, and purposes. Technologies bring through FDI should be
 Environ Sci Pollut Res

Table 6 Short-run dynamics

Bangladesh Pakistan
Model 1 Model 2 Model 3 Model 1 Model 2 Model 3
ECT − 0.492*** − 0.367*** − 0.450*** ECT − 0.107*** − 0.103*** − 0.110***
(0.001) (0.005) (0.001) (0.003) (0.01) (0.010)
Δln(GDP) 0.685* 1.172* 0.475* Δln(GDP) 1.446* 0.070 0.204*
(0.081) (0.070) (0.087) (0.070) (0.099) (0.091)
Δln(GDP) 2
− 0.043* − 0.072* − 0.031* Δln(GDP)2 − 0.068* − 0.014* − 0.005*
(0.082) (0.071) (0.081) (0.083) (0.097) (0.098)
Δln(ECON) 0.122 0.082 0.154 Δln(ECON) 0.191* 0.121 0.144
(0.029) (0.070) (0.046) (0.055) (0.073) (0.099)
Δln(FDI) 0.006** 0.004** 0.010** Δln(FDI) 0.014* 0.011** 0.013*
(0.025) (0.048) (0.025) (0.031) (0.052) (0.038)
ΔLO − 0.025** ΔLO − 0.033**
(0.048) (0.019)
ΔGS 0.011* ΔGS 0.017*
(0.072) (0.020)
ΔCOR 0.004* ΔCOR 0.011*
(0.081) (0.058)
India Sri Lanka
Model 1 Model 2 Model 3 Model 1 Model 2 Model 3
ECT − 0.668*** − 0.662*** − 0.101** ECT − 0.083** − 0.100*** − 0.490***
(0.000) (0.000) (0.046) (0.017) (0.053) (0.036)
Δln(GDP) − 4.719** − 3.102 − 0.204** Δln(GDP) 0.235* 0.718* 1.551*
(0.080) (0.014) (0.015) (0.095) (0.082) (0.068)
Δln(GDP)2 0.288** 0.194** 0.005** Δln(GDP)2 − 0.087* − 0.061* − 0.064*
(0.027) (0.013) (0.025) (0.081) (0.076) (0.076)
Δln(ECON) 0.363*** 0.475 0.144 Δln(ECON) 0.255 0.180 0.254
(0.014) (0.046) (0.067) (0.013) (0.028) (0.015)
Δln(FDI) 0.010** 0.010* 0.005* Δln(FDI) 0.004** 0.010* 0.002*
(0.012) (0.029) (0.082) (0.078) (0.075) (0.070)
ΔLO 0.007** ΔLO − 0.023*
(0.041) (0.085)
ΔGS 0.002* ΔGS 0.017**
(0.062) (0.020)
ΔCOR 0.018** ΔCOR 0.011*
(0.043) (0.075)

***, **, and * show the level of significance at 1%, 5%, and 10%, respectively

monitored to improve the quality of environment. Last could maintain environmental quality to save human
but not least, there is need to improve the institutional lives and ecology.
quality to articulate and implement the policies that

Table 7 Granger Causality test

Null hypothesis F statistic P value Causality

ln(GDP) does not Granger cause ln(EF) 6.610 0.002 Yes

ln(EF) does not Granger cause ln(GDP) 6.215 0.025 Yes
ln(ECON) does not Granger cause ln(EF) 1.489 0.030 Yes
ln(EF) does not Granger cause ln(ECON) 4.280 0.216 No
ln(FDI) does not Granger cause ln(EF) 5.820 0.002 Yes
ln(EF) does not Granger cause ln(FDI) 6.861 0.000 Yes
LO does not Granger cause ln(EF) 0.817 0.444 No
ln(EF) does not Granger cause LO 2.334 0.101 Yes
GS does not Granger cause ln(EF) 4.944 0.091 Yes
ln(EF) does not Granger cause GS 2.164 0.576 No
COR does not Granger cause ln(EF) 1.880 0.056 Yes
ln(EF) does not Granger cause COR 1.254 0.268 No
Environ Sci Pollut Res

Author contributions Samina Sabir has major contribution in all the sec- Dasgupta S, De Cian E (2016) Institutions and the environment: existing
tions of this article while Unbreen Qayyum and Tariq Majeed assisted in evidence and future directions. In: MITP: Mitigation, Innovation
refining research idea by working over literature review and and Transformation Pathways 240747. Eni Enrico Mattei (FEEM),
methodology. Fondazione
Dinda S (2004) Environmental Kuznets curve hypothesis: a survey. Ecol
Econ 49(4):431–455
Compliance with ethical standards Figge L, Oebels K, Offermans A (2017) The effects of globalization on
ecological footprints: an empirical analysis. Environ Dev Sustain
Conflict of interest The authors declare that they have no competing 19:863–876
interests. Gökmenoğlu K, Taspinar N (2016) The relationship between CO2 emis-
sions, energy consumption, economic growth and FDI: the case of
Turkey. J Int Trade Econ Dev 25(5):706–723
Grossman GM, Krueger AB (1991) Environmental impacts of a north
References American free trade agreement. NBER, Working Paper Series No
3914, 1–57
Abdouli M, Kamoun O, Hamdi B (2018) The impact of economic Hao Y, Liu YM (2015) Has the development of FDI and foreign trade
growth, population density, and FDI inflows on CO2 emissions in contributed to China’s CO2 emissions? An empirical study with
BRICTS countries: does the Kuznets curve exist? Empir Econ provincial panel data. Nat Hazards 76(2):1079–1091
54(4):1717–1742 Hassan ST, Xia E, Khan NH, Shah SMA (2019) Economic growth,
Abid M (2016) Impact of economic, financial, and institutional factors on natural resources, and ecological footprint: evidence from
CO2 emissions: evidence from sub-Saharan Africa economies. Util Pakistan. Environ Sci Pollut Res 26(3):2929–2938
Policy 41:85–94 Heidari H, Turan KS, Saeidpour L (2015) Economic growth, CO2 emis-
Ajide KB, Adeniyi O (2010) FDI and the environment in developing sions, and energy consumption in the five ASEAN countries. Int J
economies: evidence from Nigeria. Environ Res J 4(4):291–297 Electr Power Energy Syst 64:785–791
Aliyu MA (2005) Foreign direct investment and the environment: pollu- Hermes N, Lensink R (2003) Foreign direct investment, financial devel-
tion haven hypothesis revisited. In: Eight Annual Conference on opment and economic growth. J Dev Stud 40(1):142–163
Global Economic Analysis, Lübeck, Germany IPCC (Intergovernmental Panel on Climate Change) (2014) Climate
Al-Mulali U, Ozturk I (2015) The effect of energy consumption, urban- change 2014: the physical science basis. In: Stocker TF et al (eds)
ization, trade openness, industrial output, and the political stability Contribution of working group I to the sixth assessment report of the
on the environmental degradation in the MENA (Middle East and IPCC. IPCC, Geneva
North African) region. Energy 84:382–389 Jalil A, Feridun M (2011) The impact of growth, energy and financial
Al-Mulali U, Tang CF (2013) Investigating the validity of pollution ha- development on the environment in China: a cointegration analysis.
ven hypothesis in the gulf cooperation council (GCC) countries. Energy Econ 33(2):284–291
Energy Policy 60:813–819 Jalil A, Mahmud SF (2009) Environment Kuznets curve for CO2 emis-
Álvarez-Herránz A, Balsalobre D, Cantos JM, Shahbaz M (2017) Energy sions: a cointegration analysis for China. Energy Policy 37(12):
innovations-GHG emissions nexus: fresh empirical evidence from 5167–5172
OECD countries. Energy Policy 101:90–100 Jaunky VC (2011) The CO2 emissions-income nexus: evidence from rich
Anwar S, Nguyen LP (2010) Foreign direct investment and economic countries. Energy Policy 39:1228–1240
growth in Vietnam. Asia Pac Bus Rev 16(1–2):183–202 Javid M, Sharif F (2016) Environmental Kuznets curve and financial
Apergis N, Ozturk I (2015) Testing environmental Kuznets curve hypoth- development in Pakistan. Renew Sustain Energy Rev 54:406–414
esis in Asian countries. Ecol Indic 52:16–22 Keho Y (2015) Is foreign direct investment good or bad for the environ-
Azman-Saini WNW, Law SH, Ahmad AH (2010) FDI and economic ment? Times series evidence from ECOWAS countries. Econ Bull
growth: new evidence on the role of financial markets. Econ Lett 35(3):1916–1927
107(2):211–213 Laegreid OM, Povitkina M (2018) Do political institutions moderate the
Baek J (2016) A new look at the FDI–income–energy–environment nex- GDP-CO2 relationship? Ecol Econ 145:441–450
us: dynamic panel data analysis of ASEAN. Energy Policy 91:22– Lan J, Kakinaka M, Huang X (2012) Foreign direct investment, human
27 capital and environmental pollution in China. Environ Resour Econ
Balsalobre D, Shahbaz M, Roubaud D, Farhani S (2018) How economic 51(2):255–275
growth, renewable electricity and natural resources contribute to Lau LS, Choong CK, Eng YK (2014) Investigation of the environmental
CO2 emissions? Energy Policy 113:356–367 Kuznets curve for carbon emissions in Malaysia: do foreign direct
Batten JA, Vo XV (2009) An analysis of the relationship between foreign investment and trade matter? Energy Policy 68:490–497
direct investment and economic growth. Appl Econ 41(13):1621– Lee JW (2013) The contribution of foreign direct investment to clean
1641 energy use, carbon emissions and economic growth. Energy
Bilgili F, Koçak E, Bulut Ü (2016) The dynamic impact of renewable Policy 55:483–489
energy consumption on CO2 emissions: a revisited environmental Leitao A (2010) Corruption and the environmental Kuznets curve: em-
Kuznets curve approach. Renew Sustain Energy Rev 54:838–845 pirical evidence for sulfur. Ecol Econ 69(11):2191–2201
Buitenzorgy M, Mol AP (2011) Does democracy lead to a better envi- Liu K, Lin B (2019) Research on influencing factors of environmental
ronment? Deforestation and the democratic transition peak. Environ pollution in China: a spatial econometric analysis. J Clean Prod 206:
Resour Econ 48(1):59–70 356–364
Chandran VGR, Tang CF (2013) The impacts of transport energy con- Mert M, Bölük G (2016) Do foreign direct investment and renewable
sumption, foreign direct investment and income on CO2 emissions energy consumption affect the CO2 emissions? New evidence from
in ASEAN-5 economies. Renew Sustain Energy Rev 24:445–453 a panel ARDL approach to Kyoto Annex countries. Environ Sci
Charfeddine L (2017) The impact of energy consumption and economic Pollut Res 23(21):21669–21681
development on ecological footprint and CO2 emissions: evidence Mrabet Z, Alsamara M (2017) Testing the Kuznets curve hypothesis for
from a Markov switching equilibrium correction model. Energy Qatar: A comparison between carbon dioxide and ecological foot-
Econ 65:355–374 print. Renew Sust Energ Rev 70:1366–1375
 Environ Sci Pollut Res

Mrabet Z, AlSamara M, Hezam JS (2017) The impact of economic de- Shahbaz M, Sinha A (2019) Environmental Kuznets curve for CO2 emis-
velopment on environmental degradation in Qatar. Environ Ecol sions: a literature survey. J Econ Stud 46(1):106–168
Stat 24:7–38 Shahbaz M, Mallick H, Mahalik MK, Loganathan N (2015) Does glob-
Munir K, Riaz N (2020) Asymetric impact of energy consumption on alization impede environmental quality in India? Ecol Indic 52:379–
environmental degradation: evidence from Australia, China, and 393
USA. Environ Sci Pollut Res 27:11749–11759. https://doi.org/10. Shahbaz M, Balsalobre-Lorente D, Sinha A (2019) Foreign direct invest-
1007/s11356-020-07777-6 ment-CO2 emissions nexus in Middle East and North African coun-
Nguyen AN, Nguyen T (2007) Foreign direct investment in Vietnam: an tries: importance of biomass energy consumption. J Clean Prod 217:
overview and analysis of the determination of spatial distribution. 603–614
Working Paper 1921, Development and Polices Research Center, Soytas U, Sari R (2003) Energy consumption and GDP: causality rela-
Hanoi tionship in G-7 countries and emerging markets. Energy Econ 25:
Omri A, Nguyen DK, Rault C (2014) Causal interactions between CO2 33–37
emissions, FDI, and economic growth: evidence from dynamic Soytas U, Sari R (2009) Energy consumption, economic growth and
simultaneous-equation models. Econ Model 42:382–389 carbon emissions: challenges faced by an EU candidate member.
Ozturk I, Acaravci A (2010) CO2 emissions, energy consumption and Ecol Econ 68:1667–1675
economic growth in Turkey. Renew Sust Energ Rev 14(9):3220– Stern DI (2003) The rise and fall of the environmental Kuznets curve the
3225 rise and fall of the environmental Kuznets curve, Working Papers in
Ozturk I, Acaravci A (2013) The long-run and causal analysis of energy, Economics. New York
growth, openness and financial development on carbon missions in Stretesky PB, Lynch MJ (2009) A cross-national study of the association
Turkey. Energy Econ 36:262–267 between per capita carbon dioxide emissions and exports to the
Ozturk I, Al-Mulali U, Saboori B (2016) Investigating the environmental United States. Soc Sci Res 38(1):239–250
Kuznets curve hypothesis: the role of tourism and ecological foot- Suri V, Chapman D (1998) Economic growth, trade and the environment:
print. Environ Sci Pollut Res 23:1916–1928 implications for the environmental Kuznets curve. Ecol Econ 25:
Panayotou T (1993) Empirical tests and policy analysis of environmental 195e208
degradation at different stages of economic development, working Tamazian A, Rao BB (2010) Do economic, financial and institutional
paper WP238, Technology and Employment Program, Geneva: developments matter for environmental degradation? Evidence from
International Labor Office transitional economies. Energy Econ 32(1):137–145
Pedroni P (2001) Fully modified OLS for heterogeneous cointegrated Tamazian A, Chousa JP, Vadlamannati KC (2009) Does higher economic
panels. In: Baltagi BH, Fomby TB, Hill RC (eds) Nonstationary and financial development lead to environmental degradation: evi-
Panels, Panel Cointegration, and Dynamic Panels; Advances in dence from BRIC countries? Energy Policy 37(1):246–253
Econometrics, vol 15. Emerald Group Publishing Limited, Tang CF, Tan BW (2015) The impact of energy consumption, income
Bingley, pp 93–130 and foreign direct investment on carbon dioxide emissions in
Pesaran MH, Smith R (1995) Estimating long-run relationships from Vietnam. Energy 79:447–454
dynamic heterogeneous panels. J Econ 68(1):79–113 Tsurumi T, Managi S (2010) Decomposition of the environmental
Pesaran MH, Shin Y, Smith RP (1999) Pooled mean group estimation of Kuznets curve: scale, technique, and composition effects. Environ
dynamic heterogeneous panels. J Am Stat Assoc 94(446):621–634 Econ Policy Stud 11(1):19–36
Pesaran MH, Shin Y, Smith RJ (2001) Bounds test approaches to the Uddin GA, Salahuddin M, Alam K, Gow J (2017) Ecological footprint
analysis of level relationships. J Appl Econ 16:289–332 and real income: panel data evidence from the 27 highest emitting
Reiter SL, Steensma HK (2010) Human development and foreign direct countries. Ecol Indic 77:166–175
investment in developing countries: the influence of FDI policy and Ulucak R, Bilgili F (2018) A reinvestigation of EKC model by ecological
corruption. World Dev 38(12):1678–1691 footprint measurement for high, middle and low income countries. J
Sabir S, Gorus MS (2019) The impact of globalization on ecological Clean Prod 188:144-157
Footprint: Empirical Evidence from the South Asian Countries. Wang Y, Kang L, Wu X, Xiao Y (2013) Estimating the environmental
Environ Sci Pollut Res. https://doi.org/10.1007/s11356-019- Kuznets curve for ecological footprint at the global level: a spatial
06458-3 econometric approach. Ecol Indic 34:15–21
Sabir S, Rafique A, Abbas K (2019) Institutions and FDI: evidence from Welsch H (2004) Corruption, growth, and the environment: a cross-
developed and developing countries. Financ Innov 5(8):1–20 country analysis. Environ Dev Econ 9:663–693
Saboori B, Sulaiman J, Mohd S (2012) Economic growth and CO2 emis- Yıldırım E (2014) Energy use, CO2 emission and foreign direct invest-
sions in Malaysia: a cointegration analysis of the environmental ment: is there any inconsistence between causal relations? Front
Kuznets curve. Energy Policy 51:184–191 Energy 8(3):269–278
Seker F, Ertugrul HM, Cetin M (2015) The impact of foreign direct Zhang C, Zhou X (2016) Does foreign direct investment lead to lower
investment on environmental quality: a bounds testing and causality CO2 emissions? Evidence from a regional analysis in China. Renew
analysis for Turkey. Renew Sustain Energy Rev 52:347–356 Sustain Energy Rev 58:943–951
Shaari MS, Hussain NE, Abdullah H, Kamil S (2014) Relationship
among foreign direct investment, economic growth and CO2 emis- Publisher’s note Springer Nature remains neutral with regard to jurisdic-
sion: a panel data analysis. Int J Energy Econ Policy 4(4):706–715 tional claims in published maps and institutional affiliations.