Journal of Cleaner Production 233 (2019) 1240e1249

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Taking the bad with the good: The nexus between tourism and
environmental degradation in the lower middle-income Southeast
Asian economies
Fayyaz Ahmad a, Muhammad Umar Draz b, *, Lijuan Su a, Abdul Rauf c
a
School of Economics, Lanzhou University, Lanzhou, 730000, Gansu, PR China
b
Department of Management and Humanities, Universiti Teknologi PETRONAS, 32610, Seri Iskandar, Perak, Malaysia
c
School of Economics and Management, Southeast University, Nanjing, 211189, Jiangsu, PR China

a r t i c l e i n f o a b s t r a c t

Article history: The contribution of tourism to economic development is an open secret but it is a double-edged sword.
Received 13 June 2018 The opportunity cost for this significant share is environmental degradation without required measures
Received in revised form to protect the environment. Using the Fully Modified Ordinary Least Squared approach from 1995 to
27 May 2019
2014, this study aims to explore the nexus between tourism and environmental pollution for three lower
Accepted 13 June 2019
Available online 14 June 2019
middle-income Southeast Asian economies: Indonesia, the Philippines and Vietnam. This paper uses
carbon emissions as a proxy for environmental pollution against tourist arrivals with a few control
Handling Editor: Dr Sandra Caeiro variables for analysis. The Zivot-Andrews unit root test is applied to deal with structural breaks in data
and the Gregory-Hansen test for robustness. The results confirm a negative impact of tourism on the
JEL classification: environment for Indonesia and the Philippines; however, tourism improves the environmental quality of
L83 Vietnam. This implies that the relationship varies for different countries in the same region, depending
L88 on the country-specific characteristics and corresponding policies to protect the environment. The
O53 impact of governmental policies also differs for high and lower middle-income countries as one size
Q53
cannot fit all. This study provides a comprehensive milieu of the impact of tourism on the environment.
Q58
The identified dominant factors can guide Southeast Asian and other developing countries’ governments
Keywords: at all levels in systematically formulating policies; by using these policies, carbon emissions from tourism
Carbon emissions can be reduced efficiently, resulting in sustainable development in the region.
Energy use © 2019 Elsevier Ltd. All rights reserved.
Environmental pollution
Southeast Asian economies
Tourism
Tourist arrivals

1. Introduction creation is significant, i.e. one fifth of the total employment

worldwide during the last decade. Consumer spending was on the
Tourism is a hot topic of research in recent literature and its rise in 2017 and the direct growth of this sector in that particular
contribution to the economy is significant (WTO, 1980). This year reached 4.6% (WTTC, 2018a). This recent surge is robust,
contribution has its direct and indirect impact on the economy (Lee particularly for Asian economies. By virtue of its promising growth,
and Chang, 2008). Tourist arrivals can play an important role in the tourism is capable of creating 100 million jobs in the coming
improvement of various sectors such as the economic, social and decade.
infrastructural status of a host country. Tourism also accelerates Recent statistics from the World Travel and Tourism Council
international trade, mobilizes domestic resources and fills the (WTTC) show that the total contribution of the Travel and Tourism
vacuum of unemployment. In fact, the share of tourism in job (T&T) industry to Gross Domestic Product (GDP) is $8272.3 billion
in 2017 and is expected to rise from 10.4% to 11.7% of GDP by 2028.
This will turn the total economic share of T&T into $12450.1 billion
* Corresponding author. after a decade. The direct share of this sector in GDP is $2570.1
E-mail addresses: fayyaz@lzu.edu.cn (F. Ahmad), umar.draz@utp.edu.my billion in 2017, and by the year 2028 this sum is expected to be
(M.U. Draz), sulj@lzu.edu.cn (L. Su), abdulrauf@seu.edu.cn (A. Rauf).

https://doi.org/10.1016/j.jclepro.2019.06.138
0959-6526/© 2019 Elsevier Ltd. All rights reserved.
 F. Ahmad et al. / Journal of Cleaner Production 233 (2019) 1240e1249 1241

$3890.0 billion. Furthermore, in 2017, T&T created 313.22 million adverse effects on the environment of the host country. The annual
jobs i.e. 9.9% of total employment. These numbers will be increased tourist arrivals to Southeast Asia are estimated to be around 125.78
to 413.56 million i.e. 11.6% of total employment by 2028. T&T also million and the numbers are expected to grow to 208.77 million by
has a significant impact on exports and investment worldwide. The 2028 (WTTC.Southeast Asia, 2018); therefore, it will be crucial for
industry generated a total sum of $1494.2 billion visitors’ exports in the Southeast Asian countries to be proactive in handling the
2017. This sum is forecasted to grow from 6.5% to 6.9% of total ex- adverse effects of tourist arrivals.
ports in the decade ahead. Likewise, the investment share of T&T Generally speaking, the tourist overflow increases waste and
industry will rise from 4.5% to 5.1% of total investment in the next decreases available natural resources in those places which already
decade (WTTC, 2018a). face the problem of scarcity. The excessive water use and increasing
Southeast Asia is among the fastest growing regions in terms of waste at natural sites may cause soil erosion, increase air, water and
the T&T sector. The recent figures from WTTC indicate that it is in land pollution and may eventually destroy the original sources of
the top three out of thirteen regions; 12% of aggregate GDP comes attraction for tourists. The more alarming fact is that tourism is a
from this industry and it is expected to grow to 13% in next ten significant contributor in global CO2 emissions, mainly due to
years. T&T is a substantial contributor of employment, providing 4% transportation, use of electricity and housing facilities for tourists.
of total jobs and is expected to grow by 3% per annum during the Therefore, the estimated potential of the T&T industry demands an
next decade. Additionally, the impact of T&T on exports and in- urgent shift of domestic and global policy measures to protect and
vestment is also significant and in 2018 it was likely to rise by 6% improve the environment. The eco-friendly attempts from the
and 5.4% respectively. By the next decade, both figures will grow at government may provide the best alternative to traditional tourism
5.4% and 6% per annum. Tourism is a flourishing industry in and eventually might lead to a positive environmental impact of
Southeast Asia and makes a major contribution to the development tourism on the host nation (Zhong et al., 2011; Ahmad et al., 2018).
of high, upper middle and lower middle-income countries in the Taking into account the remarkable contribution of the T&T
region (WTTC.Southeast Asia, 2018). industry in Southeast Asia, it is clear that this region is an attractive
Considering this division, Indonesia is an attractive destination destination for tourists. As well as the high and upper middle-
for tourists worldwide and the share of T&T is identical with overall income countries such as Singapore, Malaysia and Thailand, the
growth. The key facts from the WTTC show that Indonesia stands lower middle-income countries of this region are also receiving
23rd in the world in terms of absolute growth in T&T and secures surges of tourists from all over the world. The numbers are growing
7th position for long term growth. Moreover, the sector currently at a decent rate for Indonesia, the Philippines and Vietnam. If other
contributes 5.8% of total GDP and this share will rise to 6.6% in the factors remain constant, the upsurge of tourists will result in high
next decade. In terms of total employment, 10% of total jobs in CO2 emissions and eventually affect the environment. In fact, this
Indonesia are related to the T&T industry (WTTC, 2018b). The relationship is tricky as, on one hand, the positive contribution of
Philippines is also counted among the lower middle-income this industry towards the economy and environment cannot be
countries of this region and is ranked 18th in terms of T&T overlooked and, on the other hand, the possibility of environmental
contribution to GDP. The current contribution of tourism to GDP in degradation is also high with tourism and tourist activities
the Philippines is 21.1%; with the annual growth of 5.8%, this share (Fernandez et al., 2019). Therefore, to explore the exact impact of
will reach 22.4% after ten years. At present, the country stands in tourism on the environment via quantitative analysis and logical
6th place in terms of T&T contribution to employment. Further- reasoning, a detailed study is indispensible for the aforementioned
more, the possibility of growth in tourist exports and investment is lower middle-income economies. Thus, the primary objective of
significant in the coming years (WTTC, 2018c). this paper is to explore the possible link between tourism and
The contribution of T&T to Vietnam's economy is worthy of environmental pollution for three Southeast Asian countries,
discussion. Vietnam is listed as a lower middle-income country, namely Indonesia, the Philippines and Vietnam from 1995 to 2014.
ranked 47th in terms of tourism's contribution to the economy, and In a recent study, Azam et al. (2018) explored this issue by taking
is included in top ten nations for long term growth prospects in high and upper middle-income countries in Southeast Asia i.e.
tourism. The current share of tourism in GDP is nearly 10% and the Singapore, Malaysia and Thailand. Following their rationale, this
industry is expected to grow at a similar rate during the next paper takes three lower middle-income countries and uses the
decade; the total contribution to employment is estimated to rise longest available data range for analysis. The importance of tourism
from 7.6% to 8% for the same time span. Tourism driven investment for the selected countries is justified by the recent economic con-
and exports in Vietnam were 8.5% and 4% respectively in 2017; this tributions i.e. 5.8%, 10% and 21.1%, made by tourism in Indonesia,
share is forecasted to grow with excellent rates in the coming Vietnam and the Philippines respectively. Moreover, the upsurge of
decade (WTTC, 2018d). tourists in the abovementioned countries demands an analytical
The sustainable development goals (SDGs) of the United Nations review of the connection between tourist arrivals and environ-
elaborate that the lines of economic and social expansion are mental degradation. Based on this argument, this study is expected
associated with a robust ecological basis. SDGs fourteen and fifteen to provide deep insights into the link between tourism and envi-
describe the importance of a strong and clean eco-system for sus- ronmental pollution through a comparison of the Southeast Asian
tainable development. Similarly, the emphasis on controlling countries with different income levels. Furthermore, the logical
climate change and protecting the environment with advanced difference of opinion can be helpful for readers and policy makers
industrial setup and environmentally friendly processes are the to better understand the relationship and the possible factors
contemporary measures towards cleaner production. Furthermore, driving it. Thus, this work will provide concrete findings about the
the promotion of tourism as an alternative growth sector for green welfare and environmental effects of tourism.
development to decrease carbon (CO2) emissions is the essential A review of recent studies is presented in section 2 and section 3
agenda of these goals (UN, 2015). Conversely, there is a dark side of offers the theoretical background of this study. Section 4 deals with
this remarkable economic growth. The concept of tradeoff implies data, variables and methodology and sections 5 and 6 explain the
that there always exists a compromise among various available findings and offer some discussion. Finally, the concluding remarks
choices. This fast growing industry may also have significant and detailed policy implications are given in section 7.
1242 F. Ahmad et al. / Journal of Cleaner Production 233 (2019) 1240e1249

2. Literature review et al. (2016) confirm that tourism has an adverse impact on envi-
ronment for the USA; furthermore, tourism is also associated with
2.1. Tourism and environment in developed and less developed waste generation that negatively affects the environment. Exam-
economies ining the policy measures for Greece, Michailidou et al. (2016)
conclude that rational use of energy and water management is
Tourism and its economic and social impact is a relatively under- necessary to control the negative effects of tourism on climate.
researched area. Although the economic impact of tourism is well In addition to the abovementioned studies, it is also argued in
established in the existing literature, studies about its environ- the recent literature that tourism and pollution have a feedback
mental effects are rather scarce; specifically, studies about lower- relationship and that pollution levels may affect tourist arrivals. The
middle and lower-income economies are very limited. To find out findings of Tugcu and Topcu (2018) from a panel ARDL approach
the link between tourism and CO2 emissions, Leo
n et al. (2014) use show a mixed impact of various emissions on tourist receipts for
the Stochastic Impacts by Regression on Population, Affluence and ten major tourist destinations. Similarly, assessing the link between
Technology (STIRPAT) analysis on panel data from developed and tourism growth and environmental quality, Sghaier et al. (2018)
Less Developed Countries (LDCs); the results show a substantial draw several conclusions for Morocco, Tunisia and Egypt: the
impact of tourism on CO2 emissions for both panels, however, this long-term association between tourism and GDP proves true at
relationship is more significant for advanced economies due to high different significance levels; tourism and environmental connec-
CO2 emissions during the production and consumption stages of tion varies, for instance, the impact is negative for Egypt, and
the T&T industry. Similarly, taking the transport sector into positive for Tunisia; additionally, Environmental Kuznets Curves
consideration, Mulali et al. (2015) use data from 48 top tourist (EKC) hypothesis is also valid for the three economies studied.
destinations to examine the impact of tourism on environment. The
results indicate that tourist arrivals significantly increase CO2 2.2. Tourism and environment in Asia-Pacific and South Asian
emissions in all regions except Europe; the authors are of the view economies
that strict environment-friendly policies are essential in non-
European countries. Furthermore, Dogan et al. (2017) investigated The available literature presents interesting findings regarding
the relationship between energy, trade, GDP and CO2 emissions for the Asian-Pacific and South Asian economies. Investigating the link
OECD countries; the authors confirmed through various econo- between tourism, growth and pollution, Zhang and Gao (2016)
metric approaches that energy use and tourism increase CO2 discover significant differences among various regions of China;
emissions while trade improves environmental quality; likewise, contrary to the general perception, the link between CO2 emissions
the causal links among these variables also support the association. and tourism is negative in China's eastern region and a causal link
Apart from CO2 emissions, tourism also causes increase in solid among these variables exists in the long run. On the other hand,
waste; Using panel data from EU countries, Arbulú et al. (2015) find testing the EKC hypothesis for selected Asia-Pacific countries,
a significant relationship between tourism and waste generation. Shakouri et al. (2017) find a positive long-term connection between
The aforementioned studies reveal that tourism causes more tourist arrivals and CO2 emissions; moreover, there exists a unidi-
environmental degradation in developed countries compared to rectional causality among energy consumption, tourist arrivals and
the LDCs. CO2 emissions. Likewise, Sharif et al. (2017) found long-term as-
The existing literature also shows that economic development sociations between tourist arrivals and CO2 emissions for Pakistan
contributes to environmental degradation. For instance, in case of and unidirectional causality running from tourist arrivals to CO2
advanced and developing countries of various regions, Zaman et al. emissions. However, Gamage et al. (2017) investigate the long run
(2016) validate the CO2 emissions and tourism relationship; the association among income, energy consumption, CO2 emissions
authors indicate that energy use and economic development also and tourism development for Sri Lanka; their results suggest that
contribute to CO2 emissions; furthermore, investment, growth and these variables are linked in the long run, but the EKC hypothesis
health sector development promote tourism. Likewise, studying does not hold true; furthermore, energy consumption has an
tourism, CO2 emissions and GDP growth nexus, Paramati et al. adverse effect on environment in both the short run and long run,
(2017) conclude that tourism enhances economic development in while tourism development increases environmental pollution in
the European Union (EU) countries; however, the role of tourism in the long run only. Analyzing the sample of 20 Asia-Pacific Economic
environmental degradation varies among different EU regions; it Cooperation (APEC) countries, Wu et al. (2018) propose that energy
may improve the climate or cause pollution depending on mea- use for economic purposes has significant impact on environment.
sures to encourage sustainable tourism. Using the VECM and cau- Inspecting the five western provinces of China, Ahmad et al.
sality approaches for the top ten worldwide tourist destinations, (2018) establish that tourism has negative effect on the environ-
Jebli and Hadhri (2018) find that tourism improves environmental ment in Gansu, Shanxi, Qinghai and Ningxia, yet tourism devel-
conditions, as there is a negative relationship between tourism and opment improves the environmental situation in Xinjiang;
CO2 emissions; however, this link is positive for GDP growth and nevertheless, the negative impacts of economic growth and energy
CO2 emissions nexus indicating that economic development causes consumption are more significant than tourism on CO2 emissions in
pollution. the long run. Likewise, Rauf et al. (2018) find that the increased
Besides developed regions and LDCs, the existing literature also energy consumption promotes rapid economic growth and ur-
presents some interesting results about individual economies. banization increases environmental pollution in the Belt and Road
Examining the long-term equilibrium of tourism development, countries; however, trade openness and focus on alternative
growth and environment in Turkey, Katircioglu (2014) reveals that growth channels reduce CO2 emissions. Combining decoupling
the use of energy increases due to tourist arrivals, which not only analysis with EKC, Jiang et al. (2019) cite that CO2 emissions control
contributes to economic development but also affects the climate of is an imperative task for China; the decoupling contribution varies
the country. Ng et al. (2015) examine the tourist-led CO2 emissions significantly, and the role of the household and service sector is
from electricity, heating and transport sectors for Malaysia; the important in influencing overall CO2 emissions in China.
authors apply a bound test and the Vector Error Correction Model Apart from CO2 emissions, the existing literature also focuses on
(VECM) causality approach confirming a long-term association environmental sustainability. For instance, Fernandez et al. (2019)
among the variables studied. Using the Wavelet approach, Raza points out that on the one hand, tourism expansion adds to
 F. Ahmad et al. / Journal of Cleaner Production 233 (2019) 1240e1249 1243

environmental pollution and on the other hand, some particular might damage the natural environment and its attractiveness
factors are related to environmental sustainability; thus, the asso- (Jiang, 1996). Waste augmentation can turn a stunning place into a
ciation between tourism and environment is bidirectional. Simi- dump. Additionally, tourism significantly increases noise pollution;
larly, in case of developing countries, Goffi et al. (2019) argue that this includes physical noise and the use of transportation vehicles
sustainability improves the competitiveness of a tourist destina- (Zhong et al., 2011).
tion; furthermore, tourism is associated with increased environ- Fig. 1 shows the number of arrivals from 1995 to 2016 for three
mental and socioeconomic issues, and cleaner production in countries under consideration. Indonesia (IDN) is the top recipient
tourism related industries can help to improve environmental entertaining more than 10 million tourists from around the world,
quality. Vietnam (VNM) secures second position and the Philippines (PHL)
stands at third. Overall, these three countries are attractive desti-
2.3. Tourism and environment in Southeast Asian economies nations for tourists and surges of tourists provide grounds for a
significant impact of the T&T industry on these countries and
The current literature on tourism and environmental degrada- motivate our empirical investigation.
tion present limited findings about the Southeast Asian economies.
Using panel data from selected Southeast Asian countries, Jahromi 4. Data, variables and methodology
et al. (2017) confirm the existence of EKC and a significant impact of
tourism on environmental pollution, whereas energy consumption To explore the aforementioned relationship, the major variables
and economic development are the significant contributors of CO2 are: tourist arrivals (TR) in countries under consideration, and
emissions. However, taking three Southeast Asian economies into following the literature, CO2 emissions (per capita in metric tons),
account, Azam et al. (2018) provide mixed evidence regarding denoted by CO2, are used as a proxy variable for environmental
impact of tourism on pollution in Malaysia, Thailand and degradation. The CO2 includes both solid and liquid fuel emissions,
Singapore; this link is positive for Malaysia and negative for other and gas flaring. Additionally, GDP per capita and energy use (EU)
two economies. Furthermore, Volpi and Paulino (2018) establish are employed as control variables. This paper uses annual data for
that the servicing sector of tourism generally has a negative impact all variables from 1995 to 2014 in United States Dollars (USD). The
on the environment; the physical facilities including hotel units, data source for all variables is the World Development Indicators
bathrooms and laundry rooms have significant adverse impacts on 2017 for Indonesia, Vietnam and the Philippines.
the environment. To test the stationary properties of data, the Zivot and Andrews
The existing literature validates the view that the T&T industry (2002) test (Z&A test) is applied. This method is capable of dealing
and improved environmental quality play a strong role in the with structural breaks in data at various points. Perron (1989)
economic progress of a country. The available studies also highlight claimed that in the case of structural breaks in data the tradi-
the threats of increase in environmental degradation due to tourist tional methods might produce vague conclusions about stationary
upsurges and eventual CO2 emissions. At present, the evidence is structures. Alternatively, the break points are treated as unknown
mixed, and most studies take into account the advanced economies in this test which is an advanced form of the Perron test. Therefore,
with great numbers of visitors every year. Likewise, researchers this test generates precise estimations for series with breaks. The
mostly employ panel data to analyze the association between mathematical form of this test is described as follows:
tourism and environmental degradation. Thus, the current litera-
ture is lacking in policy implications for the lower-middle income X
k
Southeast Asian economies. The aim of this study is to offer a DXt ¼ b þ bxt1 þ ct þ bDTt þ dj DXtj þ mt (1)
j¼1
thorough analysis of three lower middle-income countries in
Southeast Asia namely: Indonesia, the Philippines and Vietnam.
This study will explore the country-specific characteristics that may X
k
influence the linkage of tourism with environmental pollution for DXt ¼ c þ cxt1 þ ct þ dDUt þ dDTt þ dj DXtj þ mt (2)
the lower middle-income level countries under consideration. j¼1

3. Theoretical background of the study DUt denotes dummy variables used for mean shift with time breaks
for a particular point; whereas DTt is used for time breaks in the
Tourism is considered an imperative growth factor because series. For unit root break dates, the null hypothesis states that the
modern economic systems have been concentrating on sustainable series has a unit root with an unknown structural break point or
development. The main purpose of focusing on sustainability and c ¼ 0. The other case states the series is stationary where c < 0. This
eco-efficiency is to promote the idea of a green economy. Tourism test reflects all possible break points and evaluates them succes-
offers several benefits to a domestic economy including job crea- sively. It does not include the end sample points during break point
tion, improvement of living standards and promoting native cul- selection.
ture. However, this remarkable contribution is coupled with certain To test the co-integration among all variables, this study applies
environmental costs because this rapid development significantly the Gregory-Hansen approach which is an extension of the avail-
adds to energy depletion and deteriorates the environment with able conventional tests. This test uses a general hypothesis of no co-
increased CO2 emissions (Xie and Zheng, 2001). Pigram (1980) ar- integration and is effective in case of possible regime shift. This
gues that the association between tourism and environment could method can detect the link among variables in the presence of a
be defined as substantially negative, marginally negative and pos- break in intercept and slope coefficients. For such cases, the con-
itive. The connection automatically differs after attaining the ventional Augmented Dickey-Fuller (ADF) test is not a smart choice
threshold level. Considering the country-specific factors, the elas- (Gregory and Hansen, 1996a; 1996b). The three different models
ticity between these two variables explains the final effect of with various assuptions are: level shift, level shift with trend and
tourism on environmental pollution. Tourism has both biophysical regime shift. The general mathematical forms of three models,
and socio-cultural environmental effects. For instance, tourism respectively, are as follows:
causes atmospheric pollution due to emissions of smoke, sulfur
Yt ¼ m1 þ m2 ftk þ b1 t þ a1 Xt þ εt (3)
dioxide, nitrogen oxides and other harmful gases. Tourist activities
1244 F. Ahmad et al. / Journal of Cleaner Production 233 (2019) 1240e1249

15000000

10000000

PHL
5000000
VNM
0 IDN

Fig. 1. Tourist arrivals in the Philippines, Vietnam and Indonesia.

with zero mean and a fixed positive-definite covariance matrix. The

Yt ¼ m1 þ m2 ftk þ b1 t þ a1 Xt þ a2 Xt ftk þ εt (4) FMOLS estimation can be retrieved in two stages. The dependent
variable is reformed for the long-run T interdependence and the
Yt ¼ m1 þ m2 ftk þ b1 t þ b2 tftk þ a1 Xt þ a2 Xt ftk þ εt (5) residuals of the OLS estimator are also calculated. This will lead us
to the following equation:
This test framework is used to determine the possible breaks
and the break dates in the data. The test uses the highest absolute 0 1
_
ADF test value for break selection. The calculated value is compared m t A; t ¼ 2; 3:::::n
xt ¼ @ _ (8)
against the critical value to decide the status of a series. In the nt
econometric model, Y is the dependent and X is an independent
_ _ _
variable and k represents the break date in a data series. In this equation, n t ¼ DXt  m for t ¼ 2, 3 …. n and m t ¼
P
This study utilizes the Fully Modified Ordinary Least Squared ðn  1Þ1 nt¼2 DXt . Whereas, the long-term variance for xt can be
(FMOLS) method developed by Phillips and Hansen (1990) to derived as follows:
investigate the impact of tourism on the three lower middle-
0 1
income Southeast Asian economies, i.e. Indonesia, the Philippines
and Vietnam. The general form of regression after including all B_ _ _ _ C
X B U11 x1U11 U221 xkU21 C
variables can be described as follows: _ _ _ B 1 C
U¼ þLþL ¼B
B_
0 C
C (9)
_ _ _
B U C
CO2t ¼ b0 þ b1 TRt þ b2 GDPt þ b3 GDP þ b4 EUt þ εt
2
(6) @ U11 x1 21 U221 x1U22 A
1

where εt indicates an error term and b0 refers to the intercept. The

In_ this equa-
CO2 emission is the dependent variable and GDP per capita, energy P Pn _ _ _ Pm _ _ Pns _ ~
t¼2 x t x t ; L ¼ s¼1 wðs; mÞ G s ; G s ¼ n t¼1 x t x
tion, 1
¼ n1 1
and tourism are used as independent variables. The GDP2 indicates
the square of the GDP per capita to evaluate the EKC hypothesis. The whereas, wðs; mÞ is the lag window with horizon m.
FMOLS is an advanced version of Ordinary Least Squares OLS to Let us suppose:
provide more specific results and efficiency in several aspects. The 0 1
X
_ _ _
corrections in the original OLS can be used to determine the _ _ L11 L12
D¼ ¼L¼ @_ _
A (10)
important empirical effects of this new version. The FMOLS uses the
L21 L22
standard Wald test based on an asymptotic Chi-square statistical
interpretation. Generally, this method takes endogeneity and serial
correlation into consideration. It provides more options for in- _ _ _
~ 1 U
Ζ ¼ L21  L22 U ~ (11)
vestigators to find the differences between both techniques as it 22 21

offers impartial estimators of co-integrating regressions in a single

equation. Furthermore, this method is asymptotically balanced and _* 1 _
Yt ~ U
¼ Yt  U ~ vt (12)
suitable in the existence of mixed normal asymptotic. The detailed 12 t
discussion starts from a simple regression:
2 3
0
Yt ¼ b0 þ bt Xt þ mt ; t ¼ 1; :::n (7) 6 1 xk0 7
ðk þ 1Þxk ¼ 4 lk 5 (13)
lk
In this equation, the independent variables are of order I(1) and k xk
not co-integrated. Therefore, the stationary process for indepen-
dent variables is needed to separate the vector of drift from a sta- The second stage of FMOLS for b calculations can be expressed
tionary variable. This method assumes a strict stationary procedure as follows:
 F. Ahmad et al. / Journal of Cleaner Production 233 (2019) 1240e1249 1245

Table 2
_*  _* _
 FMOLS estimations.
0 1
b ¼ ðw wÞ w0 y  nD z (14)
0 Variables Indonesia Philippines Vietnam
_* _* _* _*
In this equation, y ¼ ð y 1 ; y 2 ::::: y n Þ ; w ¼ ðtn ; XÞ and tn ¼ EU 0.012** (0.23) 0.002* (11.12) 0.002* (5.75)
0
ð1; 1:::1Þ . GDP 0.002** (1.51) 0.003** (1.52) 0.004* (7.83)
GDP2 0.005** (-0.88) 0.97** (0.073) 0.005* (-5.89)
TR 2.61** (0.19) 8.69** (0.32) 3.28* (-2.53)
5. Results and discussion Const. 0.66** (-0.36) 0.54* (-2.89) 0.54* (-13.31)
R2 0.74 0.85 0.98
Adjusted R2 0.67 0.80 0.97
To test the stationary properties of the data, this paper applies
the Z&A test. This method is designed to deal with structural breaks Notes.
in the data. The concept of stationary is important in regression 1. The “*” and “**” represent the level of significance at 5% and 1% respectively.
2. The values in parentheses are the t-statistics of coefficients.
analyses as a non-stationary series leads to misleading results 3. CO2 is the dependent variable.
commonly known as spurious regression. The transformation of
data is the remedy to solve this problem (Granger and Newbold,
1974). To solve the non-stationary issue, this study uses the dif- to CO2 emissions and causes environmental degradation in the
ferencing option for the variables with unit root problem at level. country.
The results of the unit root test are presented in Table 1 for the three For example, the famous tourist destinations such as Bali in
selected countries. Indonesia are immensely polluted and even the tourism industry
These test results shown in Table 1 indicate a mixed trend and marine life are both in danger. Indonesia stands second among
among the variables for all three economies. For instance, CO2 the marine waste producers after China. The most crowded tourist
emissions, energy use and tourism are not stationary at all levels for places, e.g. Kuta in Bali, are regularly covered with solid waste
Indonesia, but GDP per capita is stationary at all levels. For the including garbage, plastic bags and other tourist castoffs (Oliphant,
Philippines, all variables show the unit root at all levels but become 2017; Coyle, 2019). The junk piles foster CO2 emissions and pollute
stationary at the 1st difference. The response of variables is also the atmosphere. Thus, tourism related activities increase the
mixed for Vietnam, showing all other variables are stationary at all pollution level in Indonesia. Volpi and Paulino (2018) and Goffi et al.
levels except energy use. However, all variables are free from unit (2019) also suggest that tourist services and activities produce CO2
root problem after taking the 1st difference with different break- and contaminate the environment. The GDP2 indicates the square
points for both levels. of the GDP per capita to evaluate the EKC hypothesis. Therefore,
destination sustainability and eco-friendly policies are necessary to
5.1. FMOLS approach improve the environment. Furthermore, the coefficient of GDP2 is
negatively linked with CO2. This connection confirms the validity of
The presence of unit root in the data and its removal provides EKC hypothesis in Indonesia.
the basic platform to apply the FMOLS. For this purpose, CO2 is the The results show that energy use is the most significant
dependent variable including few independent variables to confirm contributor of CO2 emissions in the Philippines, whereas, the share
the impact of tourism on environmental degradation for the of GDP and tourism is relatively low but still positive. The coeffi-
selected lower middle-income Southeast Asian economies. The cient of GDP2 is also positive and does not provide evidence for EKC
FMOLS estimations are reported in Table 2. The results show that hypothesis in the Philippines. The Philippines is a holiday paradise
economic development is a significant contributor of CO2 emissions and attracts thousands of domestic and international tourists on a
as the GDP coefficient is positive and significant for Indonesia. The daily basis due to its beautiful natural beaches. One of the major
association between energy use and CO2 is also positive and sig- reasons for environmental degradation is the less effective waste
nificant. The coefficient of tourism indicates a positive association management policies (McKinsey & Company and Ocean
with CO2 showing that tourist arrivals significantly pollute the Conservancy, 2015; Palafox Jr., 2018). Despite the abundance of
environment in Indonesia. Therefore, the results indicate that both forests, plants and other natural factors to reduce the CO2
GDP and T&T are the substantial consumers of energy which leads augmentation, corruption and a ‘do nothing’ approach are common
features in the LDCs, resulting in a polluted environment. All these
elements lead to an adverse impact of tourism on the environment.
Table 1 Overall, the results suggest that both industrial and service sectors,
Unit root test with structural breaks. including tourism, foster CO2 emissions and adversely affect the
Countries Variables Z&A test for level Z&A test for 1st environment. If other factors remain constant, the share of indus-
difference trial sector emissions outweighs those of the service sectors. Mulali
T-Stat Break T-Stat Break et al. (2015) also suggest that strict environment-friendly policies
are essential to reduce CO2 emissions from tourism in non-
Indonesia CO2 3.66 2007 7.71 2011
TR 4.04 2008 5.33 2003 European countries. Similarly, Arbulú et al. (2015) confirm a sig-
GDP 6.63* 2010 6.30 2004 nificant relationship between tourism and waste generation.
EU 3.71 2005 5.88 2002 Keeping other factors constant, the scenario is different for
Philippines CO2 3.88 2008 5.92 2010
Vietnam, where the coefficient of energy use is positive and sig-
TR 3.58 2002 5.39 2000
GDP 2.95 2003 4.96 1998
nificant, illustrating a direct significant link between environment
EU 3.55 2011 6.30 2011 and energy use. The significant positive coefficient of GDP turns
Vietnam CO2 4.96* 2000 5.58 1999 negative and remains significant for GDP2. This association vali-
TR 4.32** 2010 7.22 2010 dates the existence of the EKC hypothesis in Vietnam. Conversely,
GDP 4.82* 2002 6.63 1998
the tourism coefficient is negatively associated with CO2. The sig-
EU 2.51 2000 5.99 2007
nificant link implies that tourism improves the environmental
Notes.
conditions in Vietnam. Thus, the FMOLS estimations clearly indi-
1. “*” and “**” indicate that variables are stationary at 5% and 10% respectively.
2. The level of significance is 5%.
cate that energy use, mainly in the industrial sector, has adverse
1246 F. Ahmad et al. / Journal of Cleaner Production 233 (2019) 1240e1249

impacts on the environment; as both energy use and GDP are among variables. The years of break for the Philippines are 2008
significantly associated with CO2 emissions. Furthermore, tourism and 2009. The link proves true in all scenarios as the test statistics
improves environmental quality through its role in reducing CO2 are higher than the critical values.
emissions. The major reason for this negative impact is the coal- The estimated results of Vietnam are shown in Table 5. Like the
backed energy production in Vietnam. The consumption of coal is other two economies, the connection between CO2 emissions and
rapidly deteriorating the air quality in Vietnam and industrializa- tourism is consistent in the long run. The calculated t-statistics are
tion is adding fuel to the fire. But the government is launching higher than critical values in absolute terms. This postulates that
various policies to improve environmental quality (Baker, 2018). variables share a co-integration relation in the long run. The breaks
Vietnam follows a more proactive approach than other lower occur mainly in 2006 in the case of Vietnam.
middle-income Southeast Asian countries to protect the environ-
ment. For example, the Vietnamese government passed a separate
bill introducing environmental regulations and penalties for 6. Discussion
violating the rules (Hoang et al., 2017). Similarly, the government is
working in collaboration with Japan to learn the recent de- The present literature on the tourism and environmental
velopments in waste management; both countries have signed pollution nexus presents mixed evidence. It may differ from
several memoranda to increase cooperation. Likewise, Vietnam is country to country depending on the domestic characteristics of a
one of the first countries to introduce an action plan for the Paris specific economy. One of the recent studies on this relationship for
agreement for environmental quality. However, the country needs Southeast Asian countries, including Malaysia, Thailand and
constant efforts to reduce the negative impact of energy con- Singapore, by Azam et al. (2018) argues that flora and fauna is the
sumption and must use its T&T industry to protect the environment main driving force for the positive impact of tourism on Singapore's
in the long run. environment; since their study also finds mixed responses from
These findings are similar with Sghaier et al. (2018) for Morocco, three countries, a few aspects of this complex relationship are still
Tunisia and Egypt validating a long-term association between uncovered. Thus, a detailed discussion to elaborate the findings can
tourism and GDP at different significance levels. Moreover, tourism motivate further studies on this issue for different regions.
and environmental connection varies; such as the impact is nega- Firstly, the Philippines and Indonesia secured 18th and 23rd
tive for Egypt, and positive for Tunisia. Also, the findings support position in the world in terms of the absolute growth of the T&T
the EKC hypothesis. Similarly, Azam et al. (2018) provide mixed industry. This industry is contributing a significant share of GDP for
evidence regarding tourism's impact on pollution in Malaysia, the both economies (WTTC, 2018b; WTTC, 2018c). The statistics
Thailand and Singapore. This link is positive for Malaysia and indicate that both countries have sufficient natural resources in the
negative for the other two economies. In the light of these results, form of forests and plantations. For instance, the World Bank forest
Indonesia and the Philippines need to pay more attention to the data states that the physical planting and forest areas for the
environmental issues. Improvements are required in both service Philippines and Indonesia are 24% and 46% respectively. This out-
and industrial sectors to reduce the levels of CO2 emissions. weighs Singapore in terms of greenery, as the forest area in
Singapore is only 3% of its total land (World Bank, 2018). This im-
plies that flora and fauna are not the only factors that cause re-
5.2. Cross check: the Co-integration approach ductions in CO2 emissions.
Secondly, the number of tourists visiting every year is an
To deal with the potential structural breaks, this study uses the important factor to consider. It is a great challenge for the LDCs to
Gregory-Hansen co-integration test for three countries. This entertain a large number of tourists and provide them
method tests the long-term relationships using three cases i.e. environment-friendly services. Therefore, it is likely that tourist
change in levels, change in trend and change in regime. The results arrivals will negatively affect the environment unless governments
for Indonesia are presented in Table 3 indicating a long-term co- introduce effective measures to control it (Ng et al., 2015; Shakouri
integration relationship among variables; the test values are sig- et al., 2017). According to the World Bank tourist arrivals data,
nificant at the given level of significance for Indonesia, illustrating Indonesia, Vietnam and the Philippines received 11.51 million,
the long-term association between CO2 and tourism. The major 10.01 million and 5.96 million tourists respectively in 2016. The
break date is 2009 in three cases for Indonesia. surges of tourist arrivals in remote places increase energy use.
The results for the Philippines are reported in Table 4. The values Transportation fosters CO2 emissions through fuel consumption
of ADF and Zt are consistent confirming the long run co-integration and low quality transport carriers further add to the environmental

Table 3
Co-integration test for Indonesia.

Gregory-Hansen Test for Co-integration with Regime Shifts: Change in Level

Test Statistic Breakpoint Date Asymptotic Critical Values

1% 5% 10%

ADF 6.19 14 2008 5.77 5.28 5.02

Zt 5.79 15 2009 5.77 5.28 5.02
Za 18.86 15 2009 63.64 53.58 48.65
Change in Level and Trend
ADF 6.67 8 2002 6.05 5.57 5.33
Zt 5.83 15 2009 6.05 5.57 5.33
Za 19.08 15 2009 70.27 59.76 54.94
Change in Regime
ADF 7.42 10 2004 6.51 6.00 5.75
Zt 6.18 15 2009 6.51 6.00 5.75
Za 28.09 15 2009 80.15 68.94 63.42
 F. Ahmad et al. / Journal of Cleaner Production 233 (2019) 1240e1249 1247

Table 4
Co-integration test for the Philippines.

Gregory-Hansen Test for Co-integration with Regime Shifts: Change in Level

Test Statistic Breakpoint Date Asymptotic Critical Values

1% 5% 10%

ADF 6.05 15 2009 5.77 5.28 5.02

Zt 5.81 14 2008 5.77 5.28 5.02
Za 22.76 14 2008 63.64 53.58 48.65
Change in Level and Trend
ADF 6.40 14 2008 6.05 5.57 5.33
Zt 5.62 14 2008 6.05 5.57 5.33
Za 22.66 14 2008 70.27 59.76 54.94
Change in Regime
ADF 6.54 12 2007 6.51 6.00 5.75
Zt 6.32 14 2008 6.51 6.00 5.75
Za 23.45 14 2008 80.15 68.94 63.42

Table 5
Co-integration test for Vietnam.

Gregory-Hansen Test for Co-integration with Regime Shifts: Change in Level

Test Statistic Breakpoint Date Asymptotic Critical Values

1% 5% 10%

ADF 6.99 12 2006 5.77 5.28 5.02

Zt 5.97 12 2006 5.77 5.28 5.02
Za 23.54 12 2006 63.64 53.58 48.65
Change in Level and Trend
ADF 7.18 12 2006 6.05 5.57 5.33
Zt 6.02 12 2006 6.05 5.57 5.33
Za 23.52 12 2006 70.27 59.76 54.94
Change in Regime
ADF 7.24 10 2002 6.51 6.00 5.75
Zt 6.11 12 2006 6.51 6.00 5.75
Za 25.16 12 2006 80.15 68.94 63.42

degradation process. Similarly, the construction industry in the environment are a double-edged sword and their nexus does not
LDCs does not follow international construction standards. The merely depend upon the flora and fauna of a country. There are
residential and infrastructural facilities for tourists significantly certain factors that can influence this association and it can go
contribute towards CO2 emissions (Dogan et al. 2017). On the same either way depending upon a region, country or a specific
note, the tourist activities generate different kinds of waste destination.
including littering, water waste, depletion of natural resources and
damage to the natural sites, causing environmental degradation
7. Conclusions
especially in the LDCs (Michailidou et al., 2016).
On the contrary, tourist arrivals can play an essential role in CO2
At present, tourism is among the key sectors of economy for
reduction. For instance, natural, beautiful and clean destinations
sustainable development. Besides economic benefits, tourism can
are the first priority of tourists. This motivates domestic govern-
promote cultural heritage and brings the masses together. This may
ments and the masses to maintain the beauty and eco-friendly
help to improve the image of a country and bring harmony among
environment of not only the tourist resorts but the overall
people from different parts of the world. Besides its remarkable
improvement of their city atmosphere. For example, plantations in
contribution to economy and society, this industry has a significant
cities, environmental awareness campaigns, usage of environ-
impact on the environmental quality of the host country. Therefore,
mentally friendly transport and construction of green buildings for
this paper aimed to investigate the link between tourism and
tourists can improve the overall environmental quality (Ahmad
environmental degradation taking the annual tourist arrivals and
et al., 2018). However, the industrial and household sectors of
CO2 emissions as main variables for three lower middle-income
developing countries cause serious threats to their environment. A
countries of Southeast Asia namely Indonesia, the Philippines and
substantial portion of energy is utilized in industries to keep the
Vietnam. This paper used the annual data from 1995 to 2014.
momentum of rapid economic growth of the developing nations.
The Zivot-Andrews unit root test was applied to deal with the
This, in turn, significantly increases smoke, dust and various in-
stationary properties and structural breaks in the data. The test
dustrial wastes, which are the main sources of CO2 augmentation.
results indicate a mixed trend among variables for all three coun-
Likewise, underprivileged households use kerosene and wood
tries. CO2 emissions, energy use and tourism are not stationary at
as the main energy source, resulting in high CO2 emissions and
all levels for Indonesia but GDP per capita is stationary at all levels.
environmental pollution (Wu et al., 2019). As the major planting
All variables show the unit root at all levels but become stationary
areas are generally located in the remote places rather than cities,
at the 1st difference for the Philippines. The response of variables is
all these factors can pollute the environment in the countries with
also mixed for Vietnam showing all other variables as stationary at
huge forests, including the Philippines and Indonesia. Therefore,
all levels except energy use. Nevertheless, all variables become
the above given discussion confirms that tourism and the
stationary after taking the 1st difference with different breakpoints
1248 F. Ahmad et al. / Journal of Cleaner Production 233 (2019) 1240e1249

for both levels. governments specifically and other lower income countries in
The aforementioned unit root test result fulfills the preliminary general should introduce strict regulations for waste management.
condition for the FMOLS used for further analysis. The FMOLS re- The governments should encourage waste segregation and provide
sults for Indonesia show that economic development significantly funds and technical assistance to individuals and companies pro-
contributes to CO2 emissions. The relationship between energy use moting this culture. Likewise, governments should relocate
and CO2 emissions is also positive and significant. The coefficient of beverage manufacturing near clean water sources. This will auto-
tourism shows that tourist arrivals significantly pollute the envi- matically help to reduce pollution as the water is a major source of
ronment. Additionally, the EKC hypothesis is valid in the case of their business.
Indonesia. The estimations for the Philippines suggest that both Likewise, the government should encourage the farm to market
industrial and service sectors, including tourism, foster CO2 emis- concept to discourage water waste, the use of packed food, plastic
sions and adversely affect the environment. Holding other factors bags and to promote the use of eco-friendly bags and recycling
constant, tourism adds to environmental pollution and the share of water to reduce pollution at tourist destinations. Similarly, the
the industrial sector outweighs that of the service sector. The pre- government should engage the local community to promote
sent study finds no proof for the validity of EKC hypothesis for the tourism and environmental protection; this includes the environ-
Philippines. mental protection workshops to train and motivate locals. The job
The main findings for Vietnam indicate a negative association opportunities for natives as guides will work both ways: firstly, the
between tourism and CO2 emissions, implying that tourism im- reduction of poverty through job creation; secondly, the provision
proves environmental conditions. The coefficient of energy use is of manpower to encourage environmentally friendly tourism.
positive and significant, illustrating the direct significant link be- Similarly, taking Singapore as an example, the government of lower
tween environment and energy use. The results also validate the middle-income countries should offer tax incentives to encourage
existence of the EKC hypothesis. Therefore, as both energy use and plantation in cities. As the physical forest areas are far from cities, it
GDP are significantly associated with CO2 emissions, it concludes will help to enhance environmental quality in cities. Furthermore,
that energy use mainly in the industrial sector has an adverse incentives should be offered for reporting any violations.
impact on environment. Besides, tourism improves environmental In addition, the governments of Southeast Asian countries and
quality in Vietnam through its role in reducing CO2 emissions. The other developing regions should act proactively to design policies
Gregory-Hansen test is used to check the robustness of the esti- for their industrial sectors. The economic development of a country
mations. The test statistics confirm a long-term co-integration cannot be compromised but the implementation of eco-friendly
among all variables. Thus, the relationship between tourism and methods should be encouraged. For this, the governments should
environmental degradation varies for Indonesia, the Philippines promote the use of low-carbon technology. Monetary incentives via
and Vietnam. Also, these findings are robust using different esti- tax reductions should be offered to those industries which are
mation tests. promoting low-carbon energy production, building of infrastruc-
The findings of present study reveal several important aspects of ture and eco-friendly transport systems to enhance sustainable
tourism and pollution relationships for lower middle-income growth with the help of tourism. Finally, the Southeast Asian gov-
countries. The relationship is complex and needs to be handled ernments should work in collaboration with their Asian counter-
with immense caution. The natural environment and scenic spots parts and countries in other regions. This includes policy and
are attractions for tourists but strict regulations are required to technical assistance from the developed countries to promote the
create a positive impact for planting and natural greenery in the T&T industry without degrading the environment. Last but not
forests in Southeast Asian economies. At present, the negative least, the selected countries of this study and other lower middle-
impact from tourism and energy use in several sectors outweighs income countries should focus on eco-friendly energy sources to
the positive contribution of natural plantation in Indonesia and the make their industrial sectors environment-friendly without cutting
Philippines, confirming the crucial role of government policies to down its share in economic development.
promote green and environment-friendly tourism in the LDCs. Finally, the findings and abovementioned discussions confirm
Therefore, governments in Southeast Asia should take necessary that the impact of tourism on the environment depends on several
action urgently to protect their environment. For instance, the factors and varies among different countries in the same region.
tourism industry produces tons of waste and this waste is Taking this point into account, future studies can, for instance,
dangerous for humans and marine life. The junk piles foster CO2 explore the link for lower middle-income countries of Asia and
emissions and pollute the atmosphere. The plastic consumed by Africa. The contribution of natural and industrial factors and their
marine life affects humans as the seafood is a major content of impact on this link in various regions and particular countries will
native cuisines in Southeast Asia. Moreover, this huge waste in itself enrich the knowledge of academics and policy makers. Further-
is a negative omen for the T&T industry. It is likely that tourists more, various factors influencing this association in lower middle-
from all over the world will lose their interest gradually and select income countries will help to design a combined policy to promote
clean and healthy destinations instead. Therefore, active interven- social, economic and environmental sustainability. This will ulti-
tion from government to reduce the waste and control the corre- mately help in promoting socioeconomic development in the T&T
sponding CO2 emissions is vital and urgent in Indonesia. On the oriented economies.
same note, effective waste management policies are crucial for the
Philippines to reduce the CO2 emissions from tourism.
References
The industrial sector of Vietnam is the leading contributor of
environmental degradation, but the country owns substantial en- Ahmad, F., Draz, M.U., Su, L., Ozturk, I., Rauf, A., 2018. Tourism and environmental
ergy alternatives. The central and southern areas are enriched with pollution: evidence from the one Belt one Road provinces of western China.
Sustainability 10 (10), 3580.
solar energy production to minimize the pollution. Furthermore,
Arbulú, I., Lozano, J., Maquieira, J.R., 2015. Tourism and solid waste generation in
wind energy and biomass plants are the best options to reduce CO2 Europe: a panel data assessment of the environmental Kuznets curve. Waste
emissions from the industrial sector. Additionally, improvements in Manag. 46, 628e636.
the tourist sector will help to reduce the adverse effects and help Azam, M., Alam, M.M., Hafeez, M.H., 2018. Effect of tourism on environmental
pollution: further evidence from Malaysia, Singapore and Thailand. J. Clean.
the economy in general. Taking the similar nature of environmental Prod. 190, 330e338.
degradation into consideration, the Philippines and Indonesian Baker, J., 2018. This Clean Energy Champion Is Out to Break Vietnam's Coal Habit.
 F. Ahmad et al. / Journal of Cleaner Production 233 (2019) 1240e1249 1249

May 21. Retrieved May 17, 2019, from Forbes. https://www.forbes.com/sites/ rising-tide-plastic-buries-beaches/.
jillbaker/2018/05/21/this-clean-energy-champion-is-out-to-break-vietnams- Palafox Jr., A.F., 2018. Waste Management: Saving The Philippine Islands, Tourism
coal-habit/#466d9359476b. and the Environment. March 8. Retrieved May 17, 2019, from The Manila Times.
Coyle, H., 2019. Worrying Footage Shows Bali's Famous Kuta Beach in Bali Swamped https://www.manilatimes.net/waste-management-saving-the-philippine-
by Plastic Waste. February 20. Retrieved May 17, 2019, from Mirror. https:// islands-tourism-and-the-environment/384758/.
www.mirror.co.uk/news/world-news/worrying-footage-shows-balis-famous- Paramati, S.R., Shahbaz, M., Alam, M.S., 2017. Does tourism degrade environmental
14027714. quality? A comparative study of eastern and western European Union. Trans-
Dogan, E., Seker, F., Bulbul, S., 2017. Investigating the impacts of energy consump- port. Res. Part D 50, 1e13.
tion, real GDP, tourism and trade on CO2 emissions by accounting for cross- Perron, P., 1989. Testing for a Unit Root in a Time Series with a Changing Mean.
sectional dependence: a panel study of OECD countries. Curr. Issues Tourism Department of Economics e Econometric Research Program, Princeton. Paper
20 (16), 1701e1719. 347.
Fernandez, J.I., García, P.J., Pulido, J.A., 2019. Does environmental sustainability Phillips, P.C., Hansen, B.E., 1990. Statistical inference in instrumental variables
contribute to tourism growth? An analysis at the country level. J. Clean. Prod. regression with I(1) processes. Rev. Econ. Stud. 57 (1), 99e125.
213, 309e319. Pigram, J., 1980. Environmental implications of tourism development. Ann. Tourism
Gamage, S.K.N., Kuruppuge, R.H., Haq, I.U., 2017. Energy consumption, tourism Res. 7 (4), 554e583.
development, and environmental degradation in Sri Lanka. Energy Sources B Rauf, A., Liu, X., Amin, W., Ozturk, I., Rehman, O.U., Sarwar, S., 2018. Energy and
Energy Econ. Plan. Policy 12 (10), 910e916. ecological sustainability: challenges and panoramas in Belt and Road initiative
Goffi, G., Cucculelli, M., Masiero, L., 2019. Fostering tourism destination competi- countries. Sustainability 10 (8), 2743.
tiveness in developing countries: the role of sustainability. J. Clean. Prod. 209, Raza, S.A., Sharif, A., Wong, W.K., Karim, M.Z., 2016. Tourism development and
101e115. environmental degradation in the United States: evidence from wavelet-based
Granger, C., Newbold, P., 1974. Spurious regressions in econometrics. J. Econom. 2 analysis. Curr. Issues Tourism 20 (16), 1768e1790.
(2), 111e120. Sghaier, A., Guizani, A., Jabeur, S.B., Nurunnabi, M., 2018. Tourism development,
Gregory, A.W., Hansen, B.E., 1996a. Residual-based tests for cointegration in models energy consumption and environmental quality in Tunisia, Egypt and Morocco:
with regime shifts. J. Econom. 70 (1), 99e126. a trivariate analysis. Geojournal 1e17.
Gregory, A.W., Hansen, B.E., 1996b. Tests for cointegration in models with regime Shakouri, B., Yazdi, S.K., Ghorchebigi, E., 2017. Does tourism development promote
and trend shifts. Oxf. Bull. Econ. Stat. 58 (3), 555e560. CO2 emissions? Anatolia Int. J. Tourism Hospit. Res. 28 (3), 444e452.
Hoang, T.A., Chu, N.X., Tran, T.V., 2017. The environmental pollution in Vietnam: Sharif, A., Afshan, S., Nisha, N., 2017. Impact of tourism on CO2 emission: evidence
source, impact and remedies. Int. J. Sci. Technol. Res. 6 (2), 249e253. from Pakistan. Asia Pac. J. Tourism Res. 22 (4), 408e421.
Jahromi, R.S., Othman, M.S., Law, S.H., Ismail, N.W., 2017. Tourism and CO2 emissions Tugcu, C.T., Topcu, M., 2018. The impact of carbon dioxide (CO2) emissions on
nexus in Southeast Asia: new evidence from panel estimation. Environ. Dev. tourism:does the source of emission matter? Theor. Appl. Econ. 125e136.
Sustain. 19 (4), 1407e1423. UN, 2015. Transforming Our World: the 2030 Agenda for Sustainable.
Jebli, M.B., Hadhri, W., 2018. The dynamic causal links between CO2 emissions from Volpi, Y.D., Paulino, S.R., 2018. The sustainability of services: considerations on the
transport, real GDP, energy use and international tourism. Int. J. Sustain. Dev. materiality of accommodation services from the concept of life cycle thinking.
World Ecol. 25 (6), 568e577. J. Clean. Prod. 192, 327e334.
Jiang, J.J., Ye, B., Zhou, N., Zhang, X.L., 2019. Decoupling analysis and environmental World Bank, 2018. Forest Area (% of Land). World Bank.
Kuznets curve modelling of provincial-level CO2 emissions and economic WTO, 1980. Manila Declaration on World Tourism. World Tourism Organization,
growth in China: a case study. J. Clean. Prod. 212, 1242e1255. Madrid.
Jiang, W., 1996. Study on the effect tof tourism on the ecological environment of WTTC, 2018a. Travel and Tourism Economic Impact. World Travel and Tourism
Mountain Emeiand protective strategies. Chin. J. Environ. Sci. (Beijing) 17, Council.
48e51. WTTC, 2018b. Travel and Tourism Economic Impact of 2018 for Indonesia. WTTC.
Katircioglu, S.T., 2014. International Tourism,Energy Consumption,and environ- WTTC, 2018c. Travel and Tourism Economic Impact of 2018 for the Philippines.
mental pollution: the case of Turkey. Renew. Sustain. Energy Rev. 36, 180e187. WTTC.
Lee, C., Chang, C., 2008. Tourism development and economic growth: a closer look WTTC, 2018d. Travel and Tourism Economic Impact of 2018 for Vietnam. WTTC.
at panels. Tourism Manag. 29, 180e192. WTTC.Southeast Asia, 2018. Travel and Tourism Economic Impact. World Travel and
Leo
n, C.J., Arana, J.E., Alem
an, A.H., 2014. CO2 emissions and tourism in developed Tourism Council.
and less developed countries. Appl. Econ. Lett. 21 (16), 1169e1173. Wu, S., Lei, Y., Li, S., 2019. CO2 emissions from household consumption at the
McKinsey & Company and Ocean Conservancy, 2015. Stemming the Tide: Land- provincial level and interprovincial transfer in China. J. Clean. Prod. 210,
Based Strategies for a Plastic-free Ocean. Ocean Conservancy and McKinsey 93e104.
Center for Business and Environment, Chicago, United States. Wu, T.-H., Chen, Y.S., Shang, W., Wu, J.T., 2018. Measuring energy use and CO2
Michailidou, A.V., Vlachokostas, C., Moussiopoulos, N., 2016. Interactions between emission performances for APEC economies. J. Clean. Prod. 183, 590e601.
climate change and the tourism sector: multiple-criteria decision analysis to Xie, C., Zheng, X., 2001. An analysis of the destruction to the tourist ecosystem by
assess mitigation and adaptation options in tourism areas. Tourism Manag. 55, holiday tour peaks and its counter measure. Tour. Sci. 3, 33e35.
1e12. Zaman, K., Shahbaz, M., Loganathan, N., Raza, A.S., 2016. Tourism development,
Mulali, U.A., Fereidouni, H.G., Mohammed, A.H., 2015. The effect of tourism arrival energy consumption and environmental Kuznets curve: trivariate analysis in
on CO2 emissions from transportation sector. Anatolia Int. J. Tourism Hospit. the panel of developed and developing countries. Tourism Manag. 54, 275e283.
Res. 26 (2), 230e243. Zhang, L., Gao, J., 2016. Exploring the effects of international tourism on China's
Ng, T.H., Lye, C.T., Lim, Y.S., 2015. A decomposition analysis of CO2 emissions: evi- economic growth, energy consumption and environmental pollution: evidence
dence from Malaysia's tourism industry. Int. J. Sustain. Dev. World Ecol. 23 (3), from a regional panel analysis. Renew. Sustain. Energy Rev. 53, 225e234.
266e277. Zhong, L., Deng, J., Song, Z., Ding, P., 2011. Research on environmental impacts of
Oliphant, R., 2017. Bali Declares Rubbish Emergency as Rising Tide of Plastic Buries tourism in China: progress and prospect. J. Environ. Manag. 92 (11), 2972e2983.
Beaches. December 28. Retrieved May 16, 2019, from The Telegraph. https:// Zivot, E., Andrews, D.W., 2002. Further evidence on the great crash, the oil- price
www.telegraph.co.uk/news/2017/12/28/bali-declares-rubbish-emergency- shock, and the unit-root hypothesis. J. Bus. Econ. Stat. 20 (1), 25e44.