Does air pollution affect the tourism industry in the USA? Evidence from the quantile autoregressive distributed lagged approach

Abstract

This study scrutinised the influence of air pollution on the USA’s tourism industry along with macroeconomic variables from 1990 to 2020. The analyses were conducted using a novel quantile autoregressive distributed lagged (QARDL) technique to examine the nonlinear association in the long and short run. Additionally, the constancy of the parameters under the short- and long-run estimations was investigated using the Wald test. The QARDL results under different quantiles confirmed that air pollution in terms of carbon dioxide (CO₂) emissions and particulate matter (PM2.5) had a negative association with the arrival of international tourists in the USA. However, the long-run estimation for the nonlinear association between gross domestic product (GDP) and tourists' arrival (TA) and between the real effective exchange rate (REER) and TA was positively significant under different quantiles. Alternatively, the estimation of the short run corroborated that the past and lagged values of TA showed a positive correlation with the current and lagged values of international TA. Moreover, the findings of the study confirmed, through Granger causality, the bidirectional causality between tourism, GDP, haze pollution such as PM2.5 and CO₂ emissions and REER in the USA.

Keywords

air pollution carbon emissions economic growth tourism industry USA

Introduction

A common notion regarding the global tourism industry is that it is not something that can be described through homogeneity but rather through heterogeneity. This is because a growing body of literature has attempted to explain tourism growth by examining both macroeconomic dynamics and the changing climate in emerging and developed economies. With reference to this, significant contributions were made by researchers (e.g. Ahmad et al., 2021; Nawaz et al., 2020; Omran and Kamran, 2018; Scott et al., 2012) while exploring trends in the natural environment, climate change and the tourism industry. Furthermore, the trends in the tourism industry were also investigated in terms of carbon dioxide (CO₂) emissions (Jin et al., 2018), air pollution (Saenz-de-Miera and Rosselló, 2014) and macroeconomic dynamics (Boianovsky, 2020; Wang, 2009). It is also believed that tourists are deeply concerned about the environment, and this affects their decisions about vacations or visits (Lise and Tol, 2002). Environmental factors have a significant effect on the tourism industry from a range of perspectives (Baloch et al., 2021; Liu et al., 2021; Zhang and Zhang, 2021). However, their regional context is minimal in terms of specific geographic areas (Ahmad and Ma, 2021).

Air pollution affects the atmosphere, growth of natural resources and health of living beings (Chien et al., 2021a; Li et al., 2021a). The social and economic activities of humans are adversely affected by air pollution. The tourism industry, which is based on the atmosphere, the environment, natural resources and human services, is influenced by air pollution. Consequently, air pollution has a major impact on the tourism industry and its sustainable growth. The primary concern is that the natural boundaries of the regions are unknown. It is not only a threat to people’s health but also to society and the economy as well (Qiu et al., 2020).

Although the USA is among the top tourist destinations in the world, it has faced various issues in recent years (Mishra et al., 2020). In the last decade, it was observed that the tourism industry in the USA faced a contradictory scenario for various reasons. Additionally, Chien et al. (2021b) and Shair et al. (2021) argued that pressure from the recessionary environment has created an adverse environment in the tourism industry and caused a slump in tourist footfall. However, there is a changing trend in ecotourism which is reflected in lower ecological quality. Furthermore, Mishra et al. (2020) found that the carbon footprint in the USA’s economy was the highest compared to the rest of the world due to the higher dependence of tourism on transportation. Such conditions have raised serious concerns about sustainable tourism in the USA (Lenzen et al., 2018).

Besides the environmental factors such as a higher level of air pollution in the tourism industry, the linkage between tourism and economic dynamics such as growth factors, trade openness and exchange cannot be neglected. In this regard, Danish and Wang (2018) made a significant contribution, claiming that tourism and economic growth go side by side, promoting each other in the era of globalisation. Regarding the USA’s economy, it was found that in 2017, the travel and tourism industry generated an economic output of US$1.6 trillion while supporting more than 7.8 million jobs. Additionally, the travel and tourism industry of the USA accounts for 11% of exports, contributing 2.8% of the gross domestic product (GDP) (Select USA, 2017). These figures present some interesting facts to explore the dynamic relationship between the tourism industry and the economic outlook of the USA.

Considering the study constructs, it is worthwhile to share that the various effects of air pollution in the form of CO₂ emissions, particulate matter (PM2.5) and economic factors such as GDP, trade openness and the real effective exchange rate (REER) on tourists’ arrival are different (Chien et al., 2020; Sun et al., 2020). Therefore, their relationship needs to be examined in the short and long run before suggesting practical policy measures. The quantile autoregressive distributed lagged (QARDL) technique, encouraged by Cho et al. (2015), has received much attention in the recent literature on energy economics and sustainable environments. This provided the core motivation to employ the QARDL method while investigating the short- and long-run estimated associations between the constructs, specifically in the USA context. Mensi et al. (2019) argued that there are various benefits to the QARDL methodology. For instance, it helps predict the correlation between the study constructs both in the long and short run simultaneously with a span of quantiles along with the conditional distribution of the explained variables. Additionally, various levels of quantiles for economic growth, trade openness, REER and air pollution, such as CO₂ emissions and PM2.5, have impacted tourist arrivals (TA) in the USA due to significant variation in the travel and tourism industry in recent years. Therefore, the application of the QARDL process contributes to the existing literature on tourism, the environment and the economy.

Air pollution has long been a global issue. This problem has become serious and chronic since the Industrial Revolution, as the increase in industrial activities disturbed the balance of the atmosphere, increasing the amount of carbon emissions released into the air (Li et al., 2021b; Zhuang et al., 2021). Domestic and economic activities have been the major sources of carbon emissions, which cause air pollution (Chien et al., 2021c; Xiang et al., 2021). This adversely affects the atmosphere we breathe, natural resources and the health of living beings (Chien et al., 2021a). It proves to be a threat to sustainable tourism growth, where the atmosphere, natural resources, living creatures and human resources must be of good quality (Fotiadis et al., 2021). Interestingly, during the COVID-19 pandemic, the emission of one of the most toxic pollutants, CO₂, was minimal because of the lower usage of fossil fuels, as many large industries, including tourism, were shut down to maintain social distancing (Higgins-Desbiolles, 2020). However, shutting down industries and tourism is obviously not a solution for maintaining air quality. Thus, this study suggests ways to improve air quality through the sustainable growth of the tourism industry. Moreover, this study not only analyzes the influence of air pollution on TA and tourism’s growth but also introduces factors such as GDP and REER to control the problems of air pollution in the tourism industry during COVID-19 (Chien et al., 2021c; Ehsanullah et al., 2021).

This paper is organised into different sections. A literature review is presented in section 2, and the research methodology is explained in section 3. The next section covers the study’s findings and presents a discussion, while the last section concludes the study and discusses its implications.

Literature review

Several researchers have added theoretical and empirical evidence to the literature while exploring trends in the tourism industry involving environmental and macroeconomic factors. For example, Churchill et al. (2020) considered G20 economies when exploring the trends in tourism through air pollution. For this purpose, the effects of two key measures of air pollution, emissions of CO₂ and PM2.5, on the arrival of tourists in G20 member states, were explored. Through panel data models, it was confirmed that both CO₂ emissions and PM2.5 adversely affected the arrival of international tourists in targeted economies. Furthermore, the study findings showed that the effect of CO₂ emissions was more pronounced in developed economies, whereas PM2.5 played a stronger role in developing economies. Gulistan et al. (2020) observed a dynamic association between tourism, environmental degradation, economic growth, trade openness and energy in a global sample of 112 countries between 1995 and 2017. Their findings confirmed the presence of the environmental Kuznets curve and the determination of income as a turning point in improving environmental quality. Additionally, their study findings also highlighted that economic growth, energy consumption and tourism had adverse effects on the natural environment.

Mishra et al. (2020) attempted to provide fresh evidence while examining the dynamic association between the arrival of international tourists, transportation services, economic growth and carbon emissions in the economy of the USA. Their study applied partial multiple wavelet coherence techniques to a dataset from 2001 to 2017. A strong correlation among the constructs was observed, which was not equal across the time scales. Jermsittiparsert and Chankoson (2019) investigated the environmental trends in the tourism industry of Thailand using data collected from 2000 to 2014 with yearly observations. This study confirmed that carbon emissions from manufacturing industries and other sectors had a significant impact on Thailand’s tourism industry. Liu et al. (2011) analysed the energy requirements related to carbon emissions by the tourism industry in Western China in 1999–2004. Energy intensity, industry size and expenditure size are believed to play a substantial role in determining carbon emissions by the tourism industry. Deng et al. (2017) expressed their view that although there is growing concern regarding the effect of air pollution on the tourism industry, this topic still needs to be explored in different economies. Considering this gap in the literature, their study analysed the direct, indirect and total effects of air pollution due to the arrival of international tourists in China during 2001–2013. The study findings confirmed that air pollution has a significant and direct negative impact on international tourists visiting China. Other studies such as Dong et al. (2019), Tang et al. (2019), Zhang et al. (2020) and Zhou et al. (2019) also scrutinised the relationship between air pollution and the tourism industry in different economies.

In addition, tourism industry trends were investigated with the help of economic factors. For example, Belloumi (2010) examined the role of REER and GDP in defining international tourists' arrival in Tanzania by using annual data from 1970–2017. Their results confirmed a co-integrating association between tourism and economic growth, as well as a positive association between tourism and economic growth. De Vita (2014) examined the effect of the exchange rate regime on the flow of international tourism. The study applied the system generalised methods of moments estimation to investigate trends in tourism among 27 Organization for Economic Cooperation and Development (OECD) and non-OECD economies during the period 1980–2011. It is believed that multiple exchange rate regimes play a significant role in determining a stable exchange rate to attract more international tourists. Sigala (2020) investigated economic activity, air pollution and tourism. Business activities such as running of technologies, logistics use and manufacturing of products and services increase air pollution. In performing business processes and running technologies, energy resources such as fossil fuels, which spread SO₂ and CO₂ into the air, are used. The increase in air pollution affects the quality of the environment and natural resources, which in turn affects the health of human resources. Tourism, which is based on environmental and natural resources, is significantly affected by air pollution.

According to the literary investigation conducted by Ruan et al. (2020) on the factors affecting the growth of the tourism industry, air pollution has many adverse impacts on the growth of the tourism industry. Pollution affects the quality of the natural environment, which is a source of recreation and amusement for tourists and serves to expand the tourism industry. Transportation is necessary in the tourism industry, but in the case of a contagious disease like COVID-19, restrictions on transportation will be imposed. Thus, air pollution was more severe before the COVID-19 pandemic, resulting in better air quality during COVID-19. This was because many large industries were closed to maintain social distancing and curtail the spread of COVID-19. Thus, researchers have suggested that governments maintain air quality in the COVID-19 era by encouraging sustainable activities and tourism growth (Mostafanezhad, 2020).

Despite various studies confirming the dynamic relationship between air pollution and macroeconomic variables in the tourism industry, evidence from the perspective of the USA is still insufficient. Furthermore, the USA is the world’s second largest emissions producer in terms of CO₂ emissions per capita. This was the primary motivation for selecting this sample. Furthermore, earlier studies mainly focused on traditional methods of investigating the association between the key determinants of the tourism industry. However, the key gap in the application of advanced techniques, such as QARDL, which provides three different levels of the lower, medium and upper quantiles, is yet to be covered. The benefits of applying the QARDL methodology lie in assimilating nonlinearity, asymmetric association and structural breaks, which are not possible when using traditional ordinary least squares (OLS) techniques or simple quantile regression estimation.

Methodology

This study investigated the nonlinear relationship between TA, GDP, air pollution (i.e. CO₂ and PM2.5), trade openness and REER in the USA during 1990–2020, through an innovative QARDL technique. Data were collected from the World Bank Indicators from 1990 to 2020. TA was the dependent variable, measured as tourism receipts, while PM2.5 and carbon emissions were the independent variables. In addition to GDP, REER was the control variable. Although several econometric approaches have been provided in the literature for exploring the relationship between constructs, it was observed that the literature did not show a nonlinear association between the chosen variables. This gap was filled by Cho et al. (2015), who recommended QARDL. QARDL has received much attention in the recent literature on energy economics and sustainable environments. This provided the core motivation to employ the QARDL methodology while observing the short run as well as the long-estimated correlation among said constructs in the USA context. Multiple benefits emerged and were later noted when the QARDL methodology was applied. For instance, it helped predict the association between the study variables in the long and short run simultaneously with a span of quantiles and the conditional distribution of the explained variables. Since 2015, various researchers have utilised this approach and its econometric significance (Godil et al., 2020; Nawaz et al., 2021; Zhan et al., 2021). The traditional model for ARDL can be represented by the following equation

Y_{t} = α + \sum_{i}^{p} β_{1} Y_{t - i} + \sum_{i}^{q} β_{2} X 1_{t - i} + \sum_{i}^{m} β_{3} X 2_{t - i} + \sum_{i}^{n} β_{4} X 3_{t - i} + \sum_{i}^{r} β_{5} X 4_{t - i} \sum_{i}^{s} β_{6} X 5_{t - i} + ε_{t}

(1)

where

ɛ_{t}

= white noise residuals expressed with the help of TA, GDP, trade openness and REER.

q and r = the order of the lags, as observed with the Schwarz-Info criterion.

The study variables (X1–X5) show titles such as GDP, trade openness, REER, CO₂ emissions and PM2.5, measured through natural logs. The term Y indicates TA after determining the traditional level of the ARDL methodology. Equation (2) provides a more specific layout for the study variables considering the QARDL methodology

Q_{Y_{t}} = α (τ) + \sum_{i}^{p} β_{1} (τ) Y_{t - i} + \sum_{i}^{q} β_{2} (τ) X 1_{t - i} + \sum_{i}^{m} β_{3} (τ) X 2_{t - i} + \sum_{i}^{n} β_{4} (τ) X 3_{t - i} + \sum_{i}^{r} β_{5} (τ) X 4_{t - i} + \sum_{i}^{s} β_{6} (τ) X 5_{t - i} ε_{t} (τ)

(2)

where

ε_{t} (τ) = T A_{t} - Q_{T A_{t}} (τ / ε_{t - 1})

and

0 < τ < 1

are the quantiles. To carry out the investigation, a pair of quantiles

τ

belonging to 0.05–0.95 was utilised. However, because a sequential correlation issue occurred in the white noise residual, a modified version of QARDL was used, as follows

Q_{Δ Y_{t}} = α (τ) {ρY}_{t - i} + ϕ_{1} {X 1}_{t - i} + ϕ_{2} {X 2}_{t - i} + ϕ_{3} {X 3}_{t - i} + ϕ_{4} {X 4}_{t - i} + ϕ_{5} {X 5}_{t - i} + \sum_{i}^{p} β_{1} (τ) Y_{t - i} + \sum_{i}^{q} β_{2} (τ) {X 1}_{t - i} + \sum_{i}^{m} β_{3} (τ) {X 2}_{t - i} + \sum_{i}^{n} β_{4} (τ) {X 3}_{t - i} + \sum_{i}^{r} β_{5} (τ) {X 4}_{t - i} + \sum_{i}^{S} β_{6} (τ) {X 5}_{t - i} + ε_{t} (τ)

(3)

Additionally, equation (3) can be modified to derive the model provided for the QARDL framework, as below

Q_{Δ Y_{t}} = α (τ) + ρ (τ) (Y_{t - i} - ω_{1} (τ) {X 1}_{t - i} - ω_{2} (τ) {X 2}_{t - i} - ω_{3} (τ) {X 3}_{t - i} - ω_{4} (τ) {X 4}_{t - i} - ω_{5} (τ) {X 5}_{t - i}) + \sum_{i = 1}^{p - 1} β_{1} (τ) Δ Y_{t - i} + \sum_{i = 0}^{q - 1} β_{2} (τ) Δ {X 1}_{t - i} + \sum_{i = 0}^{m - 1} β_{3} (τ) Δ {X 2}_{t - i} + \sum_{i = 0}^{n - 1} β_{4} (τ) Δ {X 3}_{t - i} + \sum_{i = 0}^{r - 1} β_{5} (τ) Δ {X 4}_{t - i} + \sum_{i = 0}^{s - 1} β_{6} (τ) Δ {X 5}_{t - i} + ε_{t} (τ)

(4)

It was observed that $Δ$ utilised the collective impact of past values of TA on the present values of TA and was measured through $β_{*} = \sum_{i = 1}^{p - 1} β_{1}$ . In addition, the collective effect of the current and past values of Y on the current circumstances of TA was observed through $β_{*} = \sum_{i = 1}^{q - 1} β_{2}$ . Furthermore, the rest of the cumulative short-term influence of the past and current values of GDP and trade openness on the current and lagged values of TA was examined using the same technique. Finally, the speed of the adjustment coefficient was assumed to be significant and negative under the findings of equation (4). Finally, to scrutinise the causality direction between the study variables, we applied the Granger causality test. This is because our study tried to investigate the quantile-causal of TA, economic growth, trade openness, REER, CO₂ emissions and PM2.5 using the Granger causality quantiles method, which was introduced by Troster (2018).

Discussion

Table 1 provides descriptions of the study, divulging all the constructs evaluated via natural log values. The findings predicted that CO₂ emissions have a greater mean and standard deviation, followed by REER and economic growth through GDP. However, all variables showed a positive mean, standard deviation and so on. It was also observed that over the study period, higher CO₂ emissions were recorded in the USA, with a higher deviation trend. The lowest mean and standard deviations were observed for TA. The Jarque–Bera test was employed to examine the normality of the constructs. The results confirmed that the null hypothesis was rejected, as the data were not normally distributed, with the values being significant at the 1% level. This confirms the implications of the QARDL approach in examining trends in the tourism industry through air pollution, economic growth, trade openness and REER.

Table 1.

Descriptive statistics.

Constructs	Mean	Min.	Max.	Std. Dev.	J-B Stats
TA	2.654	0.987	3.010	0.021	17.010***
GDP	5.159	4.159	6.015	0.006	23.003***
TOP	4.321	3.321	5.104	0.015	21.012***
REER	6.753	5.357	7.106	1.102	25.105***
PM2.5	3.789	2.654	4.013	1.010	19.200***
CO₂	7.146	6.149	8.017	0.050	27.017***

Note: ***, ** and * represent a sig. level at 1%, 5% and 10%, respectively.

Source: Authors’ estimation.

Table 2 reports the results for the unit root test as observed via Augmented Dicky Fuller (ADF) and Phillips and Perron tests. These methods of unit root testing help to eliminate the biases in study findings. The findings showed that the statistical values for the ADF and ZA tests were not significant for all constructs. Nonetheless, these variables had negative and significant outcomes at the 1% sig. level under ADF (Δ) and ZA (Δ), as reported in Table 2.

Table 2.

Unit root test.

Constructs	ADF (Level)	ADF (Δ)	ZA (Level)	Break year	ZA (Δ)	Break year
TA	−1.028	−3.984***	−2.115	2016 Q4	−8.011***	2010 Q1
GDP	−0.341	−5.336***	−1.775	2007 Q4	−7.115***	2009 Q2
TOP	−1.372	−4.118***	−1.417	2001 Q1	−6.315***	2015 Q3
REER	0.541	3.112***	0.008	2013 Q4	−10.057***	2001 Q2
PM2.5	−0.984	−5.781***	−1.417	2014 Q1	−7.158***	2016 Q1
CO₂	−0.588	−3.564***	−1.221	2008 Q4	−6.484***	2012 Q1

Note: The values in the table specify the statistical values of the ADF and ZA tests. ***, ** and * represent the 1%, 5% and 10% significance level, respectively.

Source: Authors’ estimation.

The statistics in Table 3 specify that the supposed speed of p* was negative and significant for all study quantiles. This justifies the argument that there is a reversion to long-run equilibrium among variables such as tourist arrival, GDP, population, REER, haze pollution (PM2.5) and CO₂ emissions. Furthermore, an in-depth investigation of the findings in Table 3 confirmed that the highest was at the 50^th quantile, followed by the 60^th. Additionally, the parameters for CO₂ emissions under long-run estimation through the QARDL methodology were negative and significant for all quantiles except the 0.05^th and 0.10^th. This means that there was a significant and negative link between TA and air pollution in the form of CO₂, specifically in the USA. This further justifies the adverse impact of higher carbon emissions on the tourism industry in the USA. In this regard, the findings of this study are similar to those of Jermsittiparsert and Chankoson (2019), who confirmed the negative impact of CO₂ emissions on the tourism industry in Thailand. Similarly, Sharif, Afshan and Nisha (2017) also observed the dynamic linkage between CO₂ emissions and the tourism industry in Pakistan. In addition, the findings under long-run estimation show a positive and significant relationship between TOP and the tourism industry in the USA. This means that higher TA leads to a positive output for the tourism sector, which specifies a good output. The findings were highly significant for the lowest quantile, but insignificant for the highest quantile. Gulistan et al. (2020) investigated the dynamic linkage between tourism and trade openness while observing a global sample. Their study confirmed that trade openness is positively correlated with tourism.

Table 3.

Quantile autoregressive distributed lag with TA.

Quantiles (τ)	Constant	ECM	Long-run estimates				Short-run estimates
Quantiles (τ)	α^∗(τ)	ρ^∗(τ)	β_CO2(τ)	β_TOP(τ)	β_GDP(τ)	β_REER(τ)	φ₁(τ)	ω₀(τ)	λ₀(τ)	θ₀(τ)	θ₁(τ)	δ₀(τ)
0.05	0.200 (0.110)	−0.213*** (−3.813)	−0.090 (−1.060)	0.380*** (3.280)	0.309 (0.090)	0.1365 (0.305)	0.403*** (4.010)	−0.100* (−1.800)	0.011 (1.101)	0.087*** (6.459)	0.015 (0.008)	0.052*** (5.258)
0.10	0.030 (0.201)	−0.212*** (−4.127)	−0.080 (−1.013)	0.393*** (3.639)	0.307 (0.070)	0.135 (0.323)	0.410*** (3.905)	−0.061* (−1.861)	0.030 (1.013)	0.101*** (5.257)	0.615 (0.106)	0.041*** (4.147)
0.20	0.041 (0.020)	−0.219*** (−3.916)	−0.171** (−2.071)	0.358*** (4.785)	0.218 (0.108)	0.146 (0.061)	0.150 (1.052)	−0.079*** (−4.092)	0.021 (1.009)	0.120*** (4.011)	0.056 (0.065)	0.069** (2.009)
0.30	0.302 (0.103)	−0.272** (−2.012)	−0.102** (−1.999)	0.304** (2.047)	0.232 (0.203)	0.210 (0.101)	0.131 (1.121)	−0.062*** (−4.061)	0.031 (1.312)	0.121** (1.990)	0.328 (0.208)	0.057** (2.007)
0.40	0.005 (0.252)	−0.273** (−2.073)	−0.272** (−2.007)	0.319** (2.016)	0.303* (1.703)	0.212 (1.046)	0.203 (1.021)	−0.015 (−1.051)	0.076 (0.063)	0.102 (1.012)	0.409 (0.094)	0.053** (2.061)
0.50	0.104 (0.340)	−0.315** (−1.965)	−0.231*** (−3.030)	0.310 (1.050)	0.386* (1.786)	0.179 (1.101)	0.113 (1.013)	−0.032 (−1.021)	0.004 (1.013)	0.065 (1.103)	0.132 (0.231)	0.027 (0.732)
0.60	0.100 (0.001)	−0.301* (−1.641)	−0.219*** (−4.619)	0.220 (1.001)	0.412** (2.012)	0.206 (1.010)	0.179 (1.289)	−0.023 (−1.060)	0.108 (0.881)	0.078 (1.032)	0.201 (0.170)	0.301 (0.916)
0.70	0.106 (0.006)	−0.314* (−1.741)	−0.273*** (−3.837)	0.225 (1.052)	0.371** (1.997)	0.202 (1.532)	0.100 (1.200)	−0.064** (−2.010)	0.069 (1.206)	0.067 (1.073)	0.104 (0.305)	0.039 (1.037)
0.80	0.101 (0.010)	−0.221 (−1.012)	−0.202*** (−4.020)	0.218 (1.081)	0.423*** (3.236)	0.204** (2.303)	0.201* (1.802)	−0.073** (−2.071)	0.021 (1.301)	0.058 (1.014)	0.382 (1.163)	0.211 (1.021)
0.90	0.123 (0.032)	−0.204 (−1.303)	−0.201*** (−3.061)	0.260 (1.011)	0.382*** (4.081)	0.246** (2.104)	0.310** (2.010)	−0.017 (−1.070)	0.027 (0.827)	0.061 (1.051)	0.678 (1.008)	0.213 (1.201)
0.95	0.017 (0.012)	−0.200 (−1.010)	−0.219*** (−3.019)	0.231 (1.013)	0.351*** (3.011)	0.242*** (3.208)	0.442** (2.710)	−0.079 (−1.021)	0.097 (0.937)	0.057 (1.070)	0.345 (1.050)	0.004 (1.040)

In addition, the long-run estimation in Table 3 confirms the association of economic growth with the USA tourism industry. However, this relationship was not significant for the lower quantiles, but it was statistically significant for the middle and upper quantiles. The highest impact was observed in the 80^th quantile with a coefficient score of 0.423, followed by the 50^th and 90^th quantiles. These findings confirm that higher economic growth is a positive sign of good trends in the tourism industry in the USA. Tugcu (2014) observed causality between economic growth and the tourism industry. However, the study findings under long-run estimation for tourism arrival and its key determinants showed that REER is a positive and significant indicator. Under the lower and medium quantiles, the association between REER and TA was insignificant, but was highly significant for the 80^th, 90^th and 95^th quantiles. Recently, Karimi et al. (2019) examined the relationship between the real exchange rate and TA in the Malaysian economy and observed a significant and positive linkage between them.

When examining the long-run coefficients, the findings under the short-run indicated that past TA values had a positive influence on the current values of TA in the USA. However, this relationship was only significant for the lower quantiles (0.05^th and 0.10^th) and upper quantiles (0.80^th, 0.90^th and 0.95^th). The contemporaneous variation in the value of CO₂ emissions showed a significant and negative influence on TA for the first four quantiles and 0.70^th and 0.80^th quantiles. This shows a short-run association between CO₂ emissions and the arrival of tourists in the USA. Moreover, the study findings under the short-run estimation clearly showed that there was no significant linkage between TOP and TA. However, in terms of GDP, a positive and significant relationship was confirmed, but only for the lower quantiles. Additionally, the findings for the second lagged in REER showed a positive association with current changes in the TA of the USA. However, the relationship was only significant for the lower quantiles (0.05, 0.10, 0.20, 0.30 and 0.40).

After determining the long- and short-run estimations for TA, CO₂ emissions, GDP, trade openness and REER, the next step was the same for the second measure of air pollution (PM2.5). The estimated speed of the adjustment coefficient observed in Table 4 was significant for all quantiles in a negative manner, except for the higher quantiles. This justifies the claim of the presence of a reversion to a long-run equilibrium among constructs such as TA, CO₂ emissions, trade openness, GDP and REER. The value of the coefficient for PM2.5, shown in Table 4, was significant and negative for all quantiles except for the first two. This showed that there was a significant but adverse relationship between them, indicating that a higher haze pollution level, such as PM2.5, significantly and negatively affects the arrival of international tourists in the USA. A similar result was found in a study by Dong et al. (2019), who claimed a negative association between air pollution, such as PM2.5, and TA in China. Chien et al. (2021a) predicted similar findings and stated that environmental pollution, such as PM2.5, had an adverse impact on international TA in the G20 member states.

Table 4.

Quantile autoregressive distributed lag with TA

Quantiles (τ)	Constant	ECM	Long-run estimates				Short-run estimates
Quantiles (τ)	α^∗(τ)	ρ^∗(τ)	β_PM2.5(τ)	β_TOP(τ)	β_GDP(τ)	β_REER(τ)	φ₁(τ)	ω₀(τ)	ω₁(τ)	λ₀(τ)	θ₀(τ)	θ₁(τ)	δ₀(τ)
0.05	0.001 (0.100)	−0.310*** (−5.071)	−0.080 (−0.090)	0.202*** (4.102)	0.109* (1.691)	0.130** (2.003)	0.513*** (3.104)	−0.098** (−2.708)	−0.067 (−1.067)	0.081** (2.062)	0.180*** (4.079)	0.013 (0.201)	0.037*** (3.235)
0.10	0.030 (0.031)	−0.306*** (−4.000)	−0.132 (−0.032)	0.201*** (4.051)	0.126* (1.762)	0.115** (2.021)	0.430*** (4.101)	−0.081*** (−3.671)	−0.037 (−1.067)	0.069** (2.019)	0.075*** (4.057)	0.011 (0.011)	0.048*** (4.480)
0.20	0.067 (0.017)	−0.332*** (−3.321)	−1.202** (−2.009)	0.207*** (5.109)	0.121* (1.832)	0.201** (2.321)	0.450*** (3.065)	−0.081*** (−4.071)	−0.071 (−1.137)	0.043 (1.334)	0.124*** (3.042)	0.023 (0.210)	0.010** (2.010)
0.30	0.020 (0.106)	−0.302** (−2.020)	−1.400** (−2.409)	0.201** (2.010)	0.203* (1.705)	0.112 (1.202)	0.432*** (3.329)	−0.043*** (−4.006)	−0.060 (−0.810)	0.031 (1.213)	0.107** (2.007)	0.098 (0.282)	0.012 (1.014)
0.40	0.019 (0.009)	−0.302** (−2.102)	−1.303** (−2.303)	0.217** (2.001)	0.210* (1.651)	0.129 (1.231)	0.298*** (3.089)	−0.011 (−1.045)	−0.037 (−1.107)	0.031 (1.023)	0.101 (1.102)	0.056 (0.054)	−0.017 (−1.007)
0.50	0.014 (0.051)	−0.258** (−2.085)	−0.654*** (−4.054)	0.203* (1.801)	0.206* (1.661)	0.102 (1.201)	0.326*** (3.039)	−0.025 (−1.025)	−0.055 (−1.015)	0.079 (1.039)	0.116 (0.014)	0.010 (0.012)	−0.061 (−1.006)
0.60	0.010 (0.020)	−0.191* (−1.719)	−0.789*** (−5.089)	0.202* (1.720)	0.206** (2.016)	0.110 (1.312)	0.565** (2.046)	−0.035 (−1.235)	−0.051 (−1.010)	0.054 (0.808)	0.105 (0.005)	0.019 (0.105)	−0.027 (−1.207)
0.70	0.072 (0.211)	−0.175* (−1.857)	−0.987*** (−6.089)	0.150* (1.715)	0.281** (2.090)	0.103 (1.302)	0.516* (1.890)	−0.023 (−1.043)	−0.061 (−0.970)	0.061 (0.930)	0.069 (1.019)	0.042 (0.119)	−0.036 (−1.026)
0.80	0.008 (0.024)	−0.103 (−1.031)	−0.702*** (−4.207)	0.088 (1.508)	0.307*** (4.012)	0.115 (1.157)	0.380* (1.801)	−0.011 (−1.015)	−0.036 (−0.861)	0.078 (0.857)	0.058 (1.208)	0.070 (1.107)	−0.021 (−1.002)
0.90	0.217 (0.203)	−0.207 (−1.006)	−0.403*** (−3.034)	0.091 (1.284)	0.319*** (3.020)	0.110 (1.141)	0.391* (1.902)	−0.018 (−1.080)	−0.036 (−0.739)	0.061 (1.163)	0.061 (1.316)	0.031 (1.061)	−0.017 (−0.737)
0.95	0.110 (0.211)	−0.152 (−1.012)	−0.507*** (−3.705)	0.162 (1.106)	0.310*** (4.101)	0.206 (1.309)	0.329* (1.729)	−0.013 (−1.103)	−0.099 (−0.994)	0.080 (1.108)	0.040 (0.804)	0.048 (1.108)	−0.022 (−0.821)

Note: The t-statistics are in brackets. ***, ** and * represent a sig. level at 1%, 5% and 10%, respectively.

Source: Authors’ estimations.

Moreover, the results revealed that trade openness had a positive and significant impact on TA under the long-run estimation. This relationship was monitored only for the lower to medium quantiles, where the highest impact was observed for the 40^th quantile, with a score of 0.217 at the 5% significance level. Gulistan et al. (2020) investigated the dynamic linkage between trade openness and economic growth in the tourism industry. They observed a positive association between trade openness and the tourism industry in the global market. From Table 4, the long-run estimation between economic growth and the tourism industry of the USA was also predicted, and a positive correlation between GDP and TA in the long run for all quantiles was confirmed. However, this relationship was significant for the higher and lower quantiles. In the previous literature, Gulistan et al. (2020) confirmed that higher economic growth was positively linked with the tourism industry. Meanwhile, the long-run estimation in Table 4 also shows a positive association of REER with the arrival of tourists in the local market of the USA, where this impact was only observed to be significant for the lower-order quantiles.

The results, as can be seen in Table 4, for the short-run estimation also predicted that the prior statistics of TA determined the current level of TA positively and significantly in the USA. However, this was only true in the case of lower-order quantiles, where the highest value was observed in the 0.2^th quantile. In addition, the study findings also explained that the past and lagged values of PM2.5 negatively and significantly determined the current and lagged values of TA in the USA. Similarly, a positive and significant influence of the past and lagged values of REER on the current and lagged values of TA was observed, but only for the lower-order quantiles. After investigating the trends in the tourism industry in the short- and long-run through QARDL estimation, this study employed the Wald test to observe the constancy of the parameters, as reported earlier. The results are presented in Table 5. Under the Wald test estimation, the null hypothesis for the speed of the adjustment parameter was rejected at the 1% significance level. Similarly, the null hypothesis regarding linearity across different quantiles for the long-run parameters of the variables was also rejected with either a 1% or 5% chance of error. Therefore, it was inferred that the long-run parameters between TA, CO₂ emissions and haze pollution such as PM2.5, GDP and REER were agitated in multiple quantiles for the USA. Additionally, the results further confirmed that variables such as CO₂ emissions, PM2.5, GDP and REER had a nonlinear contemporaneous effect on the level of TA in the USA.

Table 5.

Results of the Wald test for the constancy of parameters.

Constructs	Wald-statistics (CO₂ emission as IV)	Wald-statistics (PM 2.5 as IV)
ρ	9.244*** (0.000)	11.228*** (0.000)
β_CO2/_PM2.5	5.135*** (0.000)	8.002*** (0.000)
β_TOP	15.210** (0.000)	7.114*** (0.000)
β_GDP	5.741*** (0.000)	6.366*** (0.000)
β_REER	4.187*** (0.000)	2.574** (0.028)
φ₁	2.954** (0.011)	11.672*** (0.000)
ω₀	3.445*** (0.001)	5.082*** (0.000)
ω₁	—	6.142*** (0.000)
λ₀	3.135*** (0.003)	2.842** (0.021)
θ₀	4.025*** (0.000)	3.741*** (0.000)
θ₁	1.324 (0.269)	0.125 (0.999)
δ₀	1.910 (0.149)	1.852 (0.158)
Cumulative short-term effect
ω*	—	1.184 (0.121)
θ*	0.442 (0.354)	0.374 (0.446)

Note: ***, ** and * represent a 1%, 5% and 10% level of significance, respectively.

Source: Authors estimations.

Finally, Table 6 shows the significance value for the Granger causality test. From all study quantiles, it can be deduced that there was a two-way relationship among the constructs for the state economy. Moreover, the results predicted that the current and past realisation of economic growth in terms of gross domestic growth was a better determinant of the changes in international TA in the USA. At the same time, the empirical outcome confirmed that there was a bilateral association between TA and (1) PM2.5, (2) TOP, (3) GDP, (4) REER and (5) CO₂.

Table 6.

Granger causality in the quantile test results.

Quantiles	ΔTA_t↓ΔPM2.5_t	ΔPM2.5_t↓ΔTA_t	ΔTA_t↓ΔTOP_t	ΔTOP_t↓ΔTA_t	ΔTA_t↓ΔGDP_t	ΔGDP_t↓ΔTA_t	ΔTA_t↓ΔREER_t	ΔREER_t↓ΔTA_t	ΔTA_t↓ΔCO2_t	ΔCO2_t↓ΔTA_t
[0.05–0.95]	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
0.05	0.000	0.000	0.000	0.000	0.010	0.000	0.000	0.000	0.000	0.000
0.10	0.000	0.000	0.000	0.000	0.100	0.000	0.000	0.000	0.000	0.000
0.20	0.000	0.000	0.000	0.000	0.102	0.000	0.000	0.000	0.000	0.000
0.30	0.000	0.000	0.000	0.000	0.012	0.000	0.000	0.000	0.000	0.000
0.40	0.000	0.000	0.000	0.000	0.001	0.000	0.000	0.000	0.000	0.000
0.50	0.000	0.000	0.000	0.000	0.003	0.000	0.000	0.000	0.000	0.000
0.60	0.000	0.000	0.000	0.000	0.013	0.000	0.000	0.000	0.000	0.000
0.70	0.000	0.000	0.000	0.000	0.031	0.000	0.000	0.000	0.000	0.000
0.80	0.000	0.000	0.000	0.000	0.004	0.000	0.000	0.000	0.000	0.000
0.90	0.000	0.000	0.000	0.000	0.040	0.000	0.000	0.000	0.000	0.000
0.95	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000

Source: Author’s estimation.

Conclusion and policy implications

The current study’s core purpose was to investigate the impact of air pollution factors such as CO₂ emissions, haze pollution (PM2.5) and other explanatory variables on the tourism industry in the USA through the QARDL methodology. The main reason for using this methodology was that it specifies how a variety of quantiles for the chosen constructs, such as trade openness, GDP, CO₂ emissions, PM2.5 and REER, affect the arrival of international tourists in the USA; hence, it provided in-depth knowledge regarding the general dependence of the study variables in contrast with older approaches, such as simple quantile regression estimation or OLS. We further investigated the correlation between the study variables using the Granger causality test. The study findings indicated that the parameter of error correction for the lower order and medium quantiles was negative when CO₂ emissions were added among other explanatory variables for the long-run estimation under the QARDL methodology. Similarly, the p* value for the second output of QARDL also showed significant and negative outputs, but not for the higher-order quantiles. This evidence proves the claim about the presence of a reversion to the long-run association between the constructs chosen for this study. Specifically, the results confirmed that CO₂ emissions have an adverse impact on the arrival of international tourists, with a positive and significant impact from GDP, REER and TA in the long run. An in-depth investigation of the study quantiles further revealed that GDP and TA were positively correlated. However, this was only observed for the lower-order and medium quantiles. Additionally, a negative long-run estimation between CO₂ emissions and TA was found for all quantiles except for the first two.

When the QARDL method was applied to explore the dynamic linkage between the constructs along with PM2.5 and TA, the results showed that CO₂ emissions and PM2.5 were linked negatively with TA in the USA. Furthermore, the positive linkage between the past and lagged values of TA and the current and lagged values was also confirmed. The results in different quantiles via Granger causality also confirmed a bidirectional causality between tourist TA, CO₂ emissions, PM2.5, trade openness, GDP and REER. Plausible policy implications can be drawn from this study’s findings. For example, the long-run estimation confirmed a positive association between economic growth and TA in the USA in a positive manner, and this was also proven by the Granger causality test.

The results showed that the government should focus on economic activities that may boost tourist visits in the region. Furthermore, the negative linkage between CO₂ emissions and TA, and between PM2.5 and TA, also indicated a significant need to control air pollution, specifically from traditional energy sources. This may lower the environmental impact and provide more attraction to international tourists in the USA. Additionally, the government should consider the REER movement and its long-term linkage with the TA from the global market. This means that consistent and robust economic and financial policies are required to retain the long-run association between REER and TA over the coming years. Furthermore, it is recommended that strong policies be implemented to protect the natural environment from air pollution, such as CO₂ emissions and PM2.5.

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This paper is partly funded by Van Lang University, Vietnam.

ORCID iD

Arshian Sharif

Author biographies

Dr Fengsheng Chien is an Associate Professor in School of Finance and Accounting, Fuzhou University of International Studies and Trade, Fujian, China. He has published more than 25 research articles in International Journals (Impact Factor & SCOPUS). He has vast experience of teaching, research publications and reviews in reputed journals such as Journal of Environmental Management, Renewable Energy, Technology in Society, International Journal of Hydrogen Energy, Economics Research, Environmental Science and Pollution Research, Singapore Economic Review, Sustainability etc.

Dr YunQian is a Lecturer in School of Finance and Accounting, Fuzhou University of International Studies and Trade, Fujian, China. She has published more than 10 research articles in International Journals (Impact Factor & SCOPUS). She has vast experience of research publications and reviews in reputed journals such as Journal of Environmental Management, Environmental Science and Pollution Research, Singapore Economic Review etc.

Dr Arshian Sharif is an Associate Professor at the Department of Economics and Finance, Sunway University, Malaysia. He has completed his PhD in Financial Analysis and Policy from the Universiti Utara Malaysia in 2021. His teaching and research interests lie in Applied Econometrics, Financial Econometrics, Financial Management Policy, Energy Finance, Energy, Tourism and Environmental Economics. He has vast experience of publications and reviewing in top publishers like Elsevier, Taylor & Francis, Wiley, Springer, Sage, Inderscience etc. He is an editor of Environmental Science and Pollution Research (Springer). Moreover, he also serves as a Guest Editor of different renowned journals such as Economic Research (Taylor & Francis), Energy and Build Environment (Elsevier) and Annals of Financial Economics (World scientific). Moreover, he has published more than 100 research articles in top International Journals with 4600+ citations and an H-Index of 38. Also, he is among the World’s Top 2% Scientists List published by Stanford University (2022). He has received the Best Researcher Award (Deans Award for Academic Excellence) from Othman Yeop Abdullah Graduate School of Business, Universiti Utara Malaysia, for three consecutive years.

Dr Muhammad Sadiq is a Lecturer of Accounting at Taylor’s University, Malaysia. He received his doctorate in Accounting from Universiti Utara Malaysia. His recent publication includes: Co-movement of energy prices and stock market return: environmental wavelet nexus of COVID-19 pandemic from the USA, Europe and China (Chien et al., 2021); The impact of environmental, social and governance index on firm value: evidence from Malaysia (Sadiq et al., 2020); A Study of Interactive Effects of Political Influences and Earnings Management on Organizational Performance (Sadiq et al., 2020); and Do CEO Incentives Mediate the Relationship between Political Influences and Financial Reporting Quality (Sadiq et al., 2020). His research interests include sustainability, CSR, political connections, corporate governance, financial reporting quality and taxation.

Dr Vu Minh Hieu is a Sr Lecturer in Business & Management at Van Lang University, Vietnam. He has published more than 20 research articles in International Journals (Impact Factor & SCOPUS). He has vast experience of research publications and reviews in reputed journals such as Environmental Science and Pollution Research, Singapore Economic Review and Journal of Environmental Management.

References

Ahmad

(2021) How does tourism development affect environmental pollution? Tourism Economics. Online First. DOI: 10.1177/13548166211000480.

Ahmad

Ullah

Arshad

, et al. (2021) Relationship between corporate social responsibility at the micro-level and environmental performance: the mediating role of employee pro-environmental behavior and the moderating role of gender. Sustainable Production Consumption 27(4): 1138–1148.

Baloch

Tan

Kamran

, et al. (2021) A multi-perspective assessment approach of renewable energy production: policy perspective analysis. Environment, Development and Sustainability 24: 1–29. DOI: 10.1007/s10668-021-01524-8 .

Belloumi

(2010) The relationship between tourism receipts, real effective exchange rate and economic growth in Tunisia. International Journal of Tourism Research 12(5): 550–560.

Boianovsky

(2020) Paul Samuelson’s ways to macroeconomic dynamics. The European Journal of the History of Economic Thought 27(4): 606–634.

, et al. (2021a) The role of information and communication technology in encountering environmental degradation: proposing an SDG framework for the BRICS countries. Technology in Society 65: 101–115.

Chien

Sadiq

Kamran

, et al. (2021b) Co-movement of energy prices and stock market return: environmental wavelet nexus of COVID-19 pandemic from the USA, Europe, and China. Environmental Science and Pollution Research 28: 1–15. DOI: 10.1007/s11356-021-12938-2

Chien

Sadiq

Nawaz

, et al. (2021c) A step toward reducing air pollution in top Asian economies: the role of green energy, eco-innovation, and environmental taxes. Journal of Environmental Management 297: 113420. DOI: 10.1016/j.jenvman.2021.113420

, et al. (2020) An evaluation model of quantitative and qualitative fuzzy multi-criteria decision-making approach for hydroelectric plant location selection. Energies 13(11): 1–14.

(2015) Quantile cointegration in the autoregressive distributed-lag modeling framework. Journal of Econometrics 188(1): 281–300.

(2020) Air pollution and tourism: evidence from G20 countries. Journal of Travel Research 61: 223–234.

12.

Danish Wang

(2018) Dynamic relationship between tourism, economic growth, and environmental quality. Journal of Sustainable Tourism 26(11): 1928–1943.

13.

De Vita

(2014) The long-run impact of exchange rate regimes on international tourism flows. Tourism Management 45: 226–233.

14.

Deng

(2017) Evaluating impact of air pollution on China’s inbound tourism industry: a spatial econometric approach. Asia Pacific Journal of Tourism Research 22(7): 771–780.

15.

Dong

, et al. (2019) The impact of air pollution on domestic tourism in China: a spatial econometric analysis. Sustainability 11(15): 4148.

, et al. (2021) How energy insecurity leads to energy poverty? do environmental consideration and climate change concerns matters. Environmental Science and Pollution Research 28: 1–12. DOI: 10.1007/s11356-021-14415-2

(2021) The good, the bad and the ugly on COVID-19 tourism recovery. Annals of Tourism Research 87: 103–117.

, et al. (2020) The dynamic nonlinear influence of ICT, financial development, and institutional quality on CO2 emission in Pakistan: new insights from QARDL approach. Environmental Science Pollution Research 27(19): 24190–24200.

(2020) Dynamic relationship among economic growth, energy, trade openness, tourism, and environmental degradation: fresh global evidence. Environmental Science Pollution Research 27(12): 13477–13487.

20.

Higgins-Desbiolles

(2020) The “war over tourism”: challenges to sustainable tourism in the tourism academy after COVID-19. Journal of Sustainable Tourism 29(4): 551–569. DOI: 10.1080/02508281.2020.1805933

21.

Jermsittiparsert

Chankoson

(2019) Behavior of tourism industry under the situation of environmental threats and carbon emission: time series analysis from Thailand. International Journal of Energy Economics Policy 9(6): 366–372.

22.

Jin

Cheng

, et al. (2018) Self-driving tourism induced carbon emission flows and its determinants in well-developed regions: a case study of Jiangsu Province, China. Journal of Cleaner Production 186: 191–202.

(2019). Asymmetric effects of real exchange rate on inbound tourist arrivals in Malaysia: an analysis of price rigidity. 21(2), 156–164.

, et al. (2018) The carbon footprint of global tourism. Nature Climate Change 8(6): 522–528.

, et al. (2021a) Nexus between energy poverty and energy efficiency: estimating the long-run dynamics. Resources Policy 72: 102063. DOI: 10.1016/j.resourpol.2021.102063

26.

Chien

Kamran

, et al. (2021b) The nexus between COVID-19 fear and stock market volatility. Economic Research-Ekonomska Istraživanja: 1–22. DOI: 10.1080/1331677X.2021.1914125

27.

Lise

Tol

(2002) Impact of climate on tourist demand. Climatic Change 55(4): 429–449.

(2011) The energy requirements and carbon dioxide emissions of tourism industry of Western China: a case of Chengdu city. Renewable Sustainable Energy Reviews 15(6): 2887–2894.

, et al. (2021) Role of tourism development in environmental degradation: a step towards emission reduction. Journal of Environmental Management 303: 114078. DOI: 10.1016/j.jenvman.2021.114078

, et al. (2019) Long-run relationships between US financial credit markets and risk factors: evidence from the quantile ARDL approach. Finance Research Letters 29: 101–110.

, et al. (2020) Dynamic linkages between tourism, transportation, growth and carbon emission in the USA: evidence from partial and multiple wavelet coherence. Current Issues in Tourism 23(21): 2733–2755.

32.

Mostafanezhad

(2020) The materiality of air pollution: urban political ecologies of tourism in Thailand. Tourism Geographies 9: 1–18. DOI: 10.1080/14616688.2020.1801826

, et al. (2020) Trilemma association of energy consumption, carbon emission, and economic growth of BRICS and OECD regions: quantile regression estimation. Environmental Science Pollution Research 28: 1–15.

, et al. (2021) Nexus between green finance and climate change mitigation in N-11 and BRICS countries: empirical estimation through difference in differences (DID) approach. Environmental Science and Pollution Research 28(6): 6504–6519. DOI: 10.1007/s11356-020-10920-y

35.

Omran

Kamran

(2018) Determining the factors attracting the tourists to visit Kedah State, Malaysia. Journal of Environmental Management 9(226): 355–364.

36.

Qiu

Park

, et al. (2020) Social costs of tourism during the COVID-19 pandemic. Annals of Tourism Research 84: 102–114.

(2020) Applying protection motivation theory to understand international tourists’ behavioural intentions under the threat of air pollution: a case of Beijing, China. Current Issues in Tourism 23(16): 2027–2041. DOI: 10.1080/13683500.2020.1743242

, et al. (2020) The impact of environmental, social and governance index on firm value: evidence from Malaysia. International Journal of Energy Economics and Policy 10(5): 555–562.

39.

Saenz-de-Miera

Rosselló

(2014) Modeling tourism impacts on air pollution: the case study of PM10 in Mallorca. Tourism Management 40: 273–281.

(2012) International tourism and climate change. Wiley Interdisciplinary Reviews: Climate Change 3(3): 213–232.

41.

Select USA (2017) Travel, Tourism & Hospitality Spotlight. 2021. Available at: https://www.selectusa.gov/travel-tourism-and-hospitality-industry-united-states#:∼:text=Overview,supporting%207.8%20million%20U.S.%20jobs.&text=That%20same%20year%2C%20U.S.%20travel,percent%20of%20gross%20domestic%20product (accessed 15 March 2021).

42.

Shair

Shaorong

Kamran

, et al. (2021) Assessing the efficiency and total factor productivity growth of the banking industry: do environmental concerns matters? Environmental Science and Pollution Research 28(16): 20822–20838.

43.

Sharif

Afshan

Nisha

(2017) Impact of tourism on CO2 emission: evidence from Pakistan. Asia Pacific Journal of Tourism Research 22(4): 408–421.

44.

Sigala

(2020) Tourism and COVID-19: impacts and implications for advancing and resetting industry and research. Journal of Business Research 117: 312–321.

45.

Sun

Awan

Nawaz

, et al. (2020) Assessing the socio-economic viability of solar commercialization and electrification in South Asian countries. Environment, Development and Sustainability 23: 1–23. DOI: 10.1007/s10668-020-01038-9.

, et al. (2019) Does air pollution decrease inbound tourist arrivals? The case of Beijing. Asia Pacific Journal of Tourism Research 24(6): 597–605.

47.

Troster

(2018) Testing for Granger-causality in quantiles. Econometric Reviews 37(8): 850–866.

48.

Tugcu

(2014) Tourism and economic growth nexus revisited: a panel causality analysis for the case of the Mediterranean region. Tourism Management 42: 207–212.

49.