Original Research Article

Evaluation Review
2022, Vol. 0(0) 1–30
Analyzing the Nexus © The Author(s) 2022
Article reuse guidelines:
Between Geopolitical sagepub.com/journals-permissions
DOI: 10.1177/0193841X221085355
journals.sagepub.com/home/erx
Risk, Policy Uncertainty,
and Tourist Arrivals:
Evidence From the
United States

Umer Shahzad1 , Muhammad Ramzan2,3 ,

Muhammad Ibrahim Shah4,5, Buhari Doğan6, and
Ahdi Noomen Ajmi7,8

Abstract
This study attempts to explore the causal linkage of the COVID-19 pandemic,
economic policy uncertainty, geopolitical risk, and tourism arrivals in the
United States taking data from January to November 2020. In order to analyze
the above relationship, this study uses a novel time-varying granger causality

1
School of Statistics and Applied Mathematics, Anhui University of Finance and Economics,
Bengbu, China
2
School of International Trade and Economics, Shandong University of Finance and Economics,
250014, Jinan, Shandong, China
3
Faculty of Management Sciences, department of Commerce, University of Sialkot, Punjab,
Pakistan
4
Resource Economics and Environmental Sociology (REES), University of Alberta, Edmonton,
Canada
5
Alma Mater Department of Economics, University of Dhaka, Dhaka, Bangladesh
6
Department of Economics, Suleyman Demirel University, Isparta, Turkey
7
Department of Business Administration, College of Science and Humanities in Slayel, Prince
Sattam Bin Abdulaziz University, Saudi Arabia
8
ESC de Tunis, Manouba University, Manouba, Tunisia

Corresponding Author:
Umer Shahzad, School of Statistics and Applied Mathematics, Anhui University of Finance and
Economics, 243 Hong Ye Lu, Bengshan District, Bengbu, Anhui 233030, China.
Email: Shehzad-umer@hotmail.com; umer@aufe.edu.cn
2 Evaluation Review 0(0)

test developed by Shi et al. (2018), which incorporates its three causality
algorithms such as forward recursive causality, rolling causality, and recursive
evolving causality. The findings from forward recursive causality could not
confirm any significant causal relationship between COVID-19 and tourism,
geopolitical risk (GPR) and tourism, economic policy uncertainty and tourism,
and geopolitical risk and COVID-19 but found causality between economic
policy uncertainty and COVID-19. The rolling window causality reported
bidirectional causality between COVID-19 and tourism and unidirectional
causality running from tourism to geopolitical risk. However, the recursive
evolving causality identified a significant bidirectional causal relationship
between all the variables. Based on the findings, policy implications for the
tourism sector are provided.

Keywords
Time-varying causality, Geopolitical risk, economic policy uncertainty, tourist
arrivals, COVID-19, USA

Research Highlights
· Causality among geopolitical risk, policy uncertainty, COVID-19, and
tourism is investigated for the USA.
· Time-varying causality is used for empirical analysis.
· COVID-19 daily new cases granger cause tourism growth.
· Policy Uncertainty and geopolitical risk significantly influence tourism
arrivals.

Introduction and Contribution

In this modern era, the tourism sector is a crucial component for the economic
advancement of both developed and developing nations due to its unprec-
edented contribution in increasing foreign reserves by tourist income and
providing employment opportunities for domestic citizens (Demir, Gozgor, &
Paramati, 2020). Meanwhile, the tourism industry contributes to socioeco-
nomic development by generating tax revenue and reducing poverty and
contributing to socio-cultural development by promoting peace and human
prosperity (Alam & Paramati, 2016; Manzoor, Wei, & Asif, 2019). Therefore,
tourism-related revenue inflows are a significant part of each economy. The
tourist’s arrivals to the host country stimulate foreign investment through
various channels, including currency earnings, captivating foreign invest-
ments, jobs creation, and tax-related revenues. Thus, each country must
promote and sustain a favorable tourism landscape to attract tourists from all
Shahzad et al. 3

over the world. In terms of statistics, the tourism industry ranks third after the
IT and financial sectors, with an annual growth rate of 3.5% and 10.3% to
global GDP in 2019. Over the last 5 years, the sector has generated 1/10th of
total employment and 1/4th of net jobs, with 330 million jobs created only in
2019. Furthermore, the industry accounts for 4.4% of total national invest-
ments, with an amount of US$ 941 billion, and this ratio is projected to rise to
5.0% by 2029 (World Travel and Tourism Council, 2020).
Having said this, it is also true that tourism is a highly sensitive industry,
affecting the decision-making process and the perspectives of key stake-
holders. Tourism stakeholders, like most investors, are likely to invest more
when industry forecasts indicate a low level of risk or uncertainty (Akadiri,
Eluwole, Akadiri, and Avci, 2020).Whereas the degree of uncertainty about
potential development cannot be constant, practitioners frequently strive to
keep it to a minimum. It is logical to capitulate that event such as terrorism and
political instability, which results in policy uncertainty, influences a country’s
tourism earnings, as tourists, instinctively seek out sites with an established
safety and security track record (Tiwari, Das, & Dutta, 2019). Consequently,
tourism reacts to geopolitical events, and as it changes and evolves, it adapts to
a wider political environment. In essence, the economy, tourism, and other
business activities are all affected significantly by the various local and foreign
political environments (Antonakakis et al., 2017). In this sense, geopolitical
tensions and associated events such as elections trigger political uncertainty
and frictions, leading to unexpected consequences or events resulting in
economic policy uncertainty (Khalfaoui et al., 2022). Thus, political and
policy uncertainty can significantly affect tourism imports, overnight stays
numbers, tourist arrivals, and other tourism-related indicators (Demir, Gozgor,
& Paramati, 2019; Hailemariam & Ivanovski, 2021a).
On the other side, the COVID-19 pandemic has created global health,
social, and economic emergency that has never been seen before. In the
meantime, the most affected sector by the COVID-19 pandemic is the tourism
industry due to border shutdown, travel bans, and decline in market demand,
which has ravaged the entire world and destroyed political and economic
structures (Sikarwar, 2021; Sharma et al., 2022). In 2020, an estimated US$
4.5 trillion massive loss had been suffered by the world tourism sector and
caused its contribution to global GDP to plummet by 49.1%, which is about
$4.7 trillion. The pandemic also has snatched the 62 million jobs in just 1 year
(The World Travel and Tourism Council, 2021).
In comparison to the rest of the world, the spread of the novel COVID-19
outbreak is causing significant economic shocks in the United States, and
these challenges are probable to provoke a long-term economic slowdown. In
2019, the United States led in terms of the highest GDP contribution from the
travel and tourism sector, with 1839 billion dollars (World Travel and Tourism
Council, 2020). However, the COVID-19 pandemic continues to spread
4 Evaluation Review 0(0)

rapidly in the United States, having drastic effects on the country’s economic
policy instability and hurting the tourism industry, with jobs falling by 22.4
million in March and April 2020 only, while unemployment increased by a
smaller threshold of 15.9 million (ILO-OECD, 2020). The United States’
vulnerability to macroeconomic turmoil events could stifle tourist inflows,
slowing the pace of national economic growth.
In recent years, the USA as a global destination for tourism has been losing
appeal due to several problems, including the pandemic. According to
Majcher (2021), in this country, in 2018, the overseas travelers’ share reduced
from 13.7% to 11.7%. According to the US travel association, this would
further lead to a decline of 10.4% within 2023. From 2015, international
visitors’ number declined by 2% in 2016. Several reasons include rising hate
crime, mass shootings, racism, ban on several countries during the Trump era,
and hostility toward foreigners. These all lead to uncertainty and risk in the
economy, hurting the overall economy and its service sector, of which tourism
is a major product. Therefore, given this above scenario, it is pertinent to delve
into the USA case to examine the nexus between geopolitical risk, uncertainty,
and the tourism sector of the country. Policymakers and other stakeholders
emphasize examining the risk factors associated with the tourism sector since
it has such a crucial role in the economy (Hailemariam and Ivanovski, 2021b).
This is one of the major reasons, that authors select the USA as a case study.
The previous studies have focused on Economic Policy Uncertainty (EPU)
and/or Geopolitical Risks (GEO) for investigating their role in tourism in-
flows. In contrast, the effect of COVID-19 has been rarely studied in the
tourism literature. Thus, this novel study investigates the causal effect between
EPU, GEO, COVID-19, and tourist arrivals in the United States. Furthermore,
the study also explores the time-varying causality between GEO, EPU, and
COVID-19. To achieve this, the time-varying causality technique initiated by
Shi et al. (2018), a novel version of the Granger causality approach, was
employed. To the best of author’s knowledge, this is the first work in the
tourism sector to investigate the time-varying causality between these vari-
ables using daily frequency data from January 22, 2020, to November 22,
2020, as well as to study the pandemic data for geopolitical risk and un-
certainty. Thus, the current work adds to the current literature in several ways.
First, the studies on the role of the EPU in tourist arrivals are generally
available, while evidence of the role of the GEO in tourism is scarce in
previous studies. The EPU index is constructed based on uncertainty re-
garding trade, fiscal, monetary, healthcare, national security, and other related
policies. On the other hand, GEO index comprises information on several
critical events such as diplomatic discrepancies, transnational political dis-
putes, and war-like events that seem to directly impact tourism and travel
(Tiwari, Das, & Dutta, 2019).
Shahzad et al. 5

The underlying factor may be that economic actors encourage typical risk-
aversion behaviors. Thus, by illustrating a comparative evaluation between
EPU and GEO about tourist arrivals in the United States, it is significant to
scrutinize the source of uncertainty to which tourists are more susceptible.
Considering this fact, the study provides practitioners and policymakers with
real-time connectivity among these factors, allowing them to formulate
flexible policies to thrive in the tourism sector. Second, to the best of our
knowledge, based on the available literature, no research has been conducted
on the implications of COVID-19 for tourism for the USA, accounting for
these factors. As previously mentioned, this epidemic halted cross-border
travel due to travel bans, which obliterated policies and stranded tourist travel
worldwide. Hence, this research is unique in the tourism literature. Last, it has
been noted that previous studies utilized classical linear or nonlinear Granger
causality to explore the impact of EPU and GEO on tourist arrivals, which
only depict the causality direction. However, real-time causalities among
factors do not remain constant over time; they change. Therefore, this is the
first attempt to use a novel time-varying Granger causality method to assess
the time-varying causal linkage from EPU, GEO, and the COVID-19 to tourist
arrivals in the United States, as well as time-varying causality between
COVID and TA, TA and EPU, TA and GPR, GPR and COVID, and EPU and
COVID.
The remainder of this paper’s contents is arranged as follows. The review
of relevant literature and their respective inferences are discussed in the
section Literature Review. In the section Data and Methodology, we go
through the details of the materials and methods used in this research. The
details about the findings and their discussion are given in the section Em-
pirical Results and Discussion. The concluding remarks are presented in the
subsequent section.

Literature Review
It is generally accepted that people, by nature, are risk-averse in both cases,
whether it is related to investment or personal pleasure through travel and tour
(Demir et al., 2020). The EPU contains information on the level of uncertainty
underlying all a country’s major macroeconomic policies, such as national
security, medical care, monetary, fiscal, trade, and other linked policies. There
has been a surge in interest in the literature since Baker et al. (2016) introduced
the EPU. It was commonly used as a proxy for risk or uncertainty. Although
many studies measure the effect of uncertainty on economic development,
some also examine its effects on tourist demand. For instance, Dragouni et al.
(2016) was the first to present a study on the effect of the economic policy
uncertainty index on tourism demand, which used data from 1996 to 2013.
The EPU index was used as a proxy for mood and sentiment in this study, and
6 Evaluation Review 0(0)

it was hypothesized that sentiment and mood have a time- and event-
dependent effect on tourism demand. The study discovered a significant
spillover effect of EPU on tourism when there is high uncertainty, whereas
there is no spillover effect when there is low uncertainty.
More recently, Gozgor and Ongan (2016) investigated the dynamic impact
of EPU on tourism demand in the United States using quarterly data from
1998 to 2015. The study found that when policy uncertainty was high, tourism
demand dropped significantly in the long run, utilizing Maki Cointegration,
Dynamic ordinary least square (DOLS), error correction model (ECM), and
Granger causality test. By utilizing the same methodology, Ongan and Gozgor
(2018) reported that a 4.7% drop in Japanese tourist travel to the United States
occurs in the long run due to a 1% rise in the degree of economic uncertainty.
Balli et al. (2018) inspected the impact of EPU on tourism demand in selected
sex countries for the period of January 1997 to August 2017. By using partial
and multiple wavelets approaches, it is found that the impact of EPU on
tourism demand varies by country, and this effect is at its peak and has a severe
effect on tourism during times of high uncertainty. Tiwari et al. (2019) also
employed a wavelet approach in India and found that EPU led to significant
negativity for tourism arrivals. Sharma (2019) established that EPU has a
negative asymmetric effect on tourism in India by employing the NARDL
approach. Furthermore, Chen et al. (2020) used the Markov regime-switching
test to investigate the impact of EPU on hotel room demand in China, Japan,
and Taiwan in a recent study. This study’s findings revealed a significant
decrease in demand for rooms in China and Japan due to a lack of certainty.
The Geopolitical risks consist of information about the uncertainty as-
sociated with tensions between states, terrorism, and wars that disrupt foreign
relations’ healthy and peaceful path (Jiang et al., 2020). Balcilar et al. (2018)
used a nonlinear causality approach to investigate the first-time GEO index
and assess its effect on the stock markets of BRICS countries. The research
revealed that the GEO index is the most significant determinant of investment
in all countries studied. However, a rare number of researchers have used the
GEO index as a proxy for political uncertainty to determine its effect on
tourism demand, with the majority of them finding an inverse relationship.
However, the intensity of the linkage between GEO and tourist arrivals varies
by country.
For instance, Demir, Gozgor, and Paramati (2019) investigated the impact
of political risk on inbound tourism in selected 18 countries throughout 1995–
2016. Using the LSDV model, it is found that GEO risk causes a significant
decrease in inbound tourism. In Turkey, Akadiri, Eluwole, Akadiri, and Avci
(2020) used Toda-Yamamoto causality to examine the causal relationship
between geopolitical risk, economic development, and tourism for quarterly
data from 1985Q1 to 2018Q4. This study established that political risk has a
unidirectional causal relationship with tourism and significantly reduces
Shahzad et al. 7

tourism in both the long and short term. Likewise, Tiwari, Das, and Dutta
(2019) employed wavelet approaches to infer that the GOE has long-term
implications for Indian tourism. Demir et al. (2020) discovered an asymmetric
relationship between political uncertainty and tourist arrivals in Turkey from
January 1990 to December 2018. In the short run, the results of the NARDL
approach revealed an asymmetric negative relationship between these two
factors.
Moreover, Hailemariam and Ivanovski (2021a) also revealed that political
instability negatively impacts tourism in the United States. Furthermore, Lee
et al. (2021) identified a bidirectional causality between tourist arrivals and
geopolitical risk in the United States between April and November 2020. For
the case of five European countries, Balsalobre-Lorente et al. (2021) men-
tioned that tourism and economic growth have direct and moderating impacts
on environmental pollution. Sharma, Thomas, and Paul (2021) reported
detailed review on revival of tourism industry after the pandemic. The authors
mention the need for novel implications to revive the tourism sector and social
well-being of people. In another study, Sharma, Kumar, Jain, Yadav, and
Srivastava (2021) reported the environmental, social, and economic reper-
cussions of COVID-19, which include the tourism sector.
Xuefeng et al. (2021) recently examined the dependence between several
variables in the USA. The authors examined several variables in their re-
search, such as tourism, the COVID pandemic, oil price, and carbon emission.
They used daily data for COVID-19 and applied the Wavelet approach. From
their results, it was revealed that tourism and COVID-19 had an anti-cyclic
relationship, and that tourism was led by COVID-19 with negative
c-movements. Utilizing a similar Wavelet framework, Yan et al. (2022) ex-
amined the linkage between tourism, COVID, and air quality in Hawaii. In
phase coherence was discovered between COVID-19 case and tourists visit
from the Wavelet Coherence analysis. The wavelet causality further informed
that there is bidirectional causality between tourism and COVID-19 in this
state. More recently, Pata and Balsalobre-Lorente (2022) argued that tourism
and economic growth significantly influence load factor capacity in Turkey.
Zhang et al. (2022) have examined the relationship between inbound
tourism and uncertainty, taking China as a case study. The uncertainty was
measured by domestic and global uncertainty, and they took monthly data
from 2000 to 2018. A novel method such as the time-varying parameter vector
autoregression (TVP-VAR) model was applied in their study. The result from
this model demonstrated that uncertainty affects tourism differently, and it
varies with time. With a rise in the lag period, this effect also gets weakened.
The authors also found that the effect of tourism on economic policy un-
certainty was lower than that of geopolitical risk in the 2008 crisis while the 9/
11 attack had a substantially higher impact on tourism.
8 Evaluation Review 0(0)

In another study, Lu et al. (2022) examined the impact of COVID-19 on the

US’s airline industry. In contrast to previous literature studies, the authors
investigated the risk perception on the stock market of the airline industry. The
result revealed this regardless of the type of airline; this pandemic caused a
disastrous impact on the airline industry. However, as risk perception de-
creased gradually, the impact of COVID-19 also decreased owing to the
measures taken by this industry.
Lee and Chen (2021) utilized over 100 countries to see the role of eco-
nomic, financial, and political stability on tourism sector returns, revenue, and
arrivals. They utilized method of moment quantile regression approach and
took annual data from 2006 to 2017. Different risks were found to impact
tourism development differently. But tourism develops with a greater level of
stability in the country. In another study of Lee and Chen (2020), the authors
examined the impact of COVID-19 on travel and leisure industry returns. In
this study, they took 65 countries and daily data from the beginning of the
pandemic to May 2020. Comparing how deaths and cases due to COVID
affect the returns, they found that death has a more adverse impact. Also, the
authors discovered a correlation which is V-shaped between returns and
recovery case of COVID-19.
Hailemariam and Ivanovski (2021b) applied the SVAR framework to study
the linkage between economic policy uncertainty and tourism net export in the
USA. They considered monthly data from the beginning of 1999 to October
2020. From the SVAR result, it was detected that policy uncertainty adversely
affects the net export of the USA. In the long term, global economic policy
uncertainty induces more than 26 percent variation of tourism net export.

Research Gap
Tourists and tourism-related activities require interactions between organi-
zations and individuals from various countries and frequent travel from one
place to another; however, the COVID-19 pandemic halted all of these ac-
tivities due to global border closures and travel bans. The tourism sector has
been the hardest hit by the outbreak in this regard (Polyzos, Samitas, &
Spyridou, 2020). To the best of our knowledge, only a few studies in the
tourism literature in which scholars simply qualitatively demonstrate the
COVID-19 implications for the tourism industry. For example, Gallego and
Font (2020) exposed that from May to September 2020, international flight
travel in America and Europe fell by 30%, and travel in Asia plunged by 50%.
To combat crises like COVID-19, Wen et al. (2020) proposed that inter-
disciplinary research collaboration is necessary. Fotiadis and Huan (2021)
recently predicted that international tourist arrivals are expected to decrease by
30.8%–76.3% until June 2021. Albulescu (2020) explored that the rising
Shahzad et al. 9

number of COVID cases would cause unpredictable delays in the institutions’

activities, resulting in increased policy uncertainty.
Furthermore, the fear of catching COVID-19 in the target country is also
causing a high level of uncertainty (Uzuner et al., 2020). Hailemariam and
Ivanovski (2021b) used the SVAR framework in their study and detected
structural breaks in their study during 2012. However, they utilized data up to
2020, yet no structural break for oil price shock, geopolitical risk, and un-
certainty associated with COVID-19 was discovered in their study. Therefore,
this paper intends to use novel methods which can capture multiple structural
breaks, thus identifying the shocks at different periods of the study period.
However, it appears that the effect of COVID-19 on tourism has yet to be
fully explored, requiring the use of real-time data. In 2022 still, tackling the
impacts of COVID-19 in every economy is a critical challenge to consider
(Shahzad et al., 2020). The data used for variables in this study is daily data,
which may assist in the discovery of a real-time dynamic causal linkage
between EPU, GEO, COVID, and tourism arrivals in the United States. The
underlying reason for using daily data is that the literature suggests that the
interaction between COVID-19, political and economic uncertainty, and
tourist arrivals does not remain constant over time. Thus, the approach used in
this research, known as time-varying causality, is novel. It provides real-time
causal connectivity between proposed factors to the reader, including poli-
cymakers and practitioners.

Data and Methodology

Data Specification. This study aims to see the causal linkage of the COVID-19
pandemic, economic policy uncertainty, geopolitical risk, and tourism arrivals
in the USA. The study period is from January 22, 2020, to November 22,
2020. According to Lu et al. (2022), the pandemic caused disastrous impacts
on tourism industry, especially the airline industry, which is vital for tourism.
But with time as risk perception regarding tourism decreased, the impact of
COVID-19 on this industry also gradually weakened. Based on this, we have
taken the initial period of COVID-19 data since this had disastrous impli-
cations for the tourism industry. Following Drake (2020),1 the first wave of
COVID-19 is defined as from the beginning to May 31, the second wave is
from June to August, and the third wave started from September.
Here, we utilize daily data from three different sources. For example, the
daily data for COVID-19 new cases come from the European Union database,
whereas the data for tourist arrivals come from the International Trade Ad-
ministration database. The economic policy uncertainty is based on three
components and collected from the Policy Uncertainty database. The first
component involves policy-related economic uncertainty in the newspaper
coverage. The second component involves federal tax code provisions that are
10 Evaluation Review 0(0)

to be expired in future years. The third component is the disagreement among

the economic forecasters used as a proxy for uncertainty.
On the other hand, the geopolitical risk index was developed by Caldara
and Iacoviello (2018) and based on the newspaper coverage of geopolitical
tensions such as the 9/11, 2003 Iraq invasion. The authors define the geo-
political risk index as an index of the risks associated with terrorists’ attacks,
wars, and tension between the states, influencing the normal course of in-
ternational relations. However, tourist arrivals, economic policy uncertainty,
and geopolitical risk data are monthly data. Hence, following the seminal
paper by Shahbaz et al. (2017), the quadratic match sum method is used to
convert the monthly data daily. One major advantage of this method is that it
can take care of the seasonality problem existing in the data and minimize the
point to point variation in the data (Zichu and Masih, 2020). The detailed data
description and their sources are provided in Table 1.

Time-Varying Causality Test

This study employs a time-varying causality test to examine the relationship
between COVID, Tourism, Economic Policy Uncertainty, and Geopolitical
Risk. The reason for choosing the time-varying causality method is because
our variables contain daily data, and therefore causality among these variables
will not remain constant over time. The causality among the variables may
show dynamic character (Hammoudeh et al., 2020). Moreover, time-varying
causality can capture multiple structural breaks in the data with having
possible shifts in parameters in specific periods (Balcilar et al., 2019). For the
granger causality method with variation in time, several techniques exist, such
as the forward expanding window test (Thoma, 1994), rolling window
Granger causality test (Swanson, 1998), and recursive rolling Granger cau-
sality (Shi et al., 2020, 2018). This study uses the most recent time-varying
causality developed by (Shi et al. (2020, 2018) with its three causality

Table 1. Description of the Data.

Abbreviation Variable Source Reference

COVID Daily COVID-19 European Union Database Sharif et al. (2020)

new cases
TA Tourists arrivals International Trade Lee et al. (2021)
Administration database
EPU Economic policy Policy Uncertainty Akadiri, Alola, and
uncertainty index database Uzuner. (2020)
GPR Geopolitical risk Policy Uncertainty Lee et al. (2021)
index database
Shahzad et al. 11

algorithms as forwarding recursive causality, rolling causality, and recursive

evolving causality.
According to Hurn et al. (2016), rolling window Granger causality is
superior to the other two tests because it has a high detection rate. However,
the rolling window also possesses the highest rate of false detection. Com-
paratively, recursive rolling Granger causality has a lower false detection rate,
but it is at the cost of achieving the highest successful detection rate. It also has
a balanced performance relative to the other two tests. On the other hand, the
forward expanding window version test has the lowest performance relative to
the other two tests.
The corrective bootstrap algorithm is described in several steps as follows:
In the first step, a VAR (1) model is estimated with no granger causality as the
null hypothesis. In the second step, a bootstrap sample is calculated as follows
! ! ! !
x1t ô11 0 x1t1 ε1t
¼ þ
x2t ô12 x22 x2t1 ε2t

where x1t and x1t are the variables of interest in causality relation, ε1t as well as
ε2t denote residuals of VAR(1).
Now, in the third step, the statistic sequence for three tests is developed. For
example, Thoma (1994) is based on the procedure below
 
b
M1,t ¼ maxt 2½τ0þτb1 W1,t
b

Swanson (1998) test is based on the following procedure

 
b
Mtτ0þ1,t ¼ maxt 2½τ0þτb1 Wtτ0þ1,t
b

finally, the test based on the recursive rolling window of Shi et al., (2018) is
described as follows
 
SMtb ðτ 0 Þ ¼ maxt 2½τ0þτb1 SMtb ðτ 0 Þ

In the step 4, for B=1,….,499, steps 2 and 3 are repeated.

Finally, in the fifth step, provided by 95% percentile, forward, rolling, and re-
cursive processes are given by fM1,t b b
, Mtτ0þ1,t and SMtb ðτ 0 Þ gBb=1 respectively.

Empirical Results and Discussion

Preliminary Analysis
Shi et al. (2018) suggested that the time-varying causality procedure needs to
pre-filter the data, but it requires the maximum order of integration for the
12 Evaluation Review 0(0)

VAR. Therefore, to check the static properties of our variables, we have

employed the Augmented Dickey-Fuller test (ADF) and Phillips–Perron test
(PP). Table 2 presents the ADF and PP unit root tests where we find that the
maximum order of integration is I (2).

Time-Varying Causality Result

Following Shi et al. (2018), we implement the time-varying Granger causality
analysis based upon a lag augmented VAR model with d=2. This methodology
helps us figure out the origination and termination dates in the causal as-
sociation among different variables by modeling the time-varying granger
causal relationship (Çağlı, 2019). Here, we implement the causal relationship
among the variables using the recursive evolving procedure of Hurn et al.
(2016), forward expanding window method of Thoma (1994), and rolling
window procedure of Swanson (1998). Figures 1 to 5 show the test statistics
sequence and the 5% critical value sequence. The null hypothesis of no
Granger causality between the two-time series variable can be rejected if the
critical value sequence falls below the test statistics sequence.
Let us first start with Figure 1, where time-varying causality between
COVID and tourism arrivals is presented. From both the forward recursive
and rolling causality, the null hypothesis of no Granger causality cannot be
rejected since the test statistic does not exceed the critical value statistic over
the sample period. The recursive evolving causality procedure detects several
episodes of Granger causality running from COVID to tourism arrivals. The
first is detected on 28 June, which lasts for 1 month up to 28 July 2020 (second
wave). The second episode lasts from September 02 to September 27, 2020
(third wave). The final episode between these two variables is detected on

Table 2. Unit Root Test Results.

COVID EPU GPR TA

ADF
Level 0.3716 0.4618 0.0194b 0.3322
First diff. 0.1433 0.0000a - 0.4996
Second diff. 0.0000a - - 0.0000a
PP
Level 0.2511 0.0000a 0.0000a 0.8719
First diff. 0.0000a - - 0.4781
Second diff. - - - 0.0000a
Order of integration I (1)/I (2) I (1)/I (0) I (0) I (2)

Notes: Figures denote p-values. a and b indicate the rejection of the null hypothesis at the 1% and
5% levels, respectively.
Shahzad et al. 13

Figure 1. Time-varying causality between COVID and TA.

October 09, which lasts up to November 01, 2020 (third wave). Thus, only the
recursive evolving causality between the three procedures detects three
significant episodes of Granger causality running from COVID to tourism
arrivals. Such finding indicates that COVID granger causes tourists’ arrivals
from the end of June 2020 to July 2020 (second wave), from September to 27
September and finally from October 09 to the first of November 2020 (third
wave). USA’s travelling industry, especially the airline industry, was sig-
nificantly affected by the pandemic, as found in the work of Lu et al. (2022) for
the USA. This result is supported by several previous findings such as that of
Xuefeng et al. (2021) where the authors found anti-cyclic relationship
14 Evaluation Review 0(0)

Figure 2. Time-varying causality between EPU and TA.

between tourism and COVID with COVID leading tourism and Yan et al.
(2022) where the authors found causality from COVID to tourism.
These dates fit with the USA restrictions put in place for international’s
tourists. For example, at the end of June 2020, due to the additional COVID-
19 hotspots emergence worldwide, the USA extended travel restrictions with
its neighboring countries, Canada and Mexico, which was initially issued in
March 2020. As the USA experienced a surge in COVID-19 cases in almost
half of the states, the travel ban with Canada and Mexico was further extended
by the department of homeland security on July 20, 2020 (Tate et al., 2020). In
September, both the Nebraska and Oklahoma states of the USA imposed
Shahzad et al. 15

Figure 3. Time-varying causality between GPR and TA.

international and domestic travel restrictions, respectively, which could have

affected the tourist arrivals in these states. In addition, some restrictions were
extended till November to both international and domestic travel (Georgetown
University, 2020). The finding is justified because the USA travel sector
experienced a 42% yearly decline in 2020 compared to 2019. Specifically,
from March 2020 to the end of 2020, the USA’s travel sector loss due to
COVID-19 amounted to $492 billion, equivalent to a $1.6 billion daily loss for
the previous 10 months (US Travel Association, 2021).
While examining the time-varying causality from TA to COVID, similar to
causality from COVID to TA, only recursive evolving causality reveals a
significant episode from 11 to 20 July 2020 which suggests that during this
period, tourism arrivals affected COVID incidence in the USA. A survey of
16 Evaluation Review 0(0)

Figure 4. Time-varying causality between GPR and COVID.

American travelers by Destination Analytics (2021), which surveys travelers

about their perception, behavior, and feelings about traveling in the pandemic,
showed that people’s perception about the safety of traveling increased
significantly during June–July 2020. Therefore, people’s perception about the
security to travel during this period might have affected the COVID situation
in this country at this time of the year. This confirms the finding of Yan et al.
(2022). The authors found from their Wavelet causality result that tourism
affects COVID-19.
When uncertainty arises in an economy, tourists may cancel or delay travel
plans due to security and safety concerns (Demir and Gozgor, 2018).
Moreover, COVID-19 fear is currently imposing a lot of uncertainty regarding
Shahzad et al. 17

Figure 5. Time-varying causality between EPU and COVID.

being infected with COVID in the destination country (Uzuner et al., 2020).
Therefore, we analyze the relationship between economic policy uncertainty
and tourism for the USA during the COVID period. In Figure 2, forward
recursive causality from EPA to TA arrival is provided, failing to reject the null
hypothesis of no causality. However, both the rolling and recursive evolving
causality detect the significant causal relationship between EPU and TA. The
causality can be found from the rolling causality on 27 June, 15 August, and
21–22 August. Then from August 23 to 13 November (third wave), there is no
causality found between these two variables. Still, from 14 November to 15
November, and between 21 and 22 November (third wave), significant
causality between these two variables can be detected. The causality between
EPU and TA is detected from the recursive evolving technique from 27 June to
19 July (second wave), for 5 days between 18 and 22 August and 9 days
between 14 November and 22 November 2020 (third wave). This part of the
18 Evaluation Review 0(0)

result is in line with Dragouni et al. (2016), who found that economic policy
uncertainty has a significant spillover effect on the tourism sector in the event
of high uncertainty. Since the period of this study involves a highly uncertain
period of COVID-19, the finding of the unidirectional causality from EPU to
TA is reasonable. For the USA, Gozgor and Ongan (2016) also found that
economic policy uncertainty reduces the tourism spending during 1998–2015.
While looking at the causality from TA to EPU, rolling and recursive evolving
causality shows causality episodes of 5 and 6 days, respectively, during the
middle of July 2020 (second wave). This segment of the result extends the
finding of Akadiri et al, (2020), who showed that economic policy uncertainty
is associated with international tourist arrivals in 7 countries. Furthermore,
they confirmed the bidirectional causality between tourists’ arrival and
economic policy uncertainty which further supports our results. This is
consistent with the finding of Zhang et al. (2022) as well. In their study, the
authors measured the association between inbound tourism and uncertainty in
China using TVP-VAR model. The authors found that this association differs
with time, but it is certain that economic uncertainty due to domestic and
global issues definitely affects tourism industry in a significant way.
Overall, the causality between economic policy uncertainty and tourism
can be observed mostly in June–July–August and November. These dates
coincide with the wildfires which started in the Western part of the USA. The
wildfires started in July in Washington state and were further extended to
California, which saw a record-breaking wildfire season in its history (Yan
et al., 2020). Other US states such as Oregon, Colorado, New Mexico,
Arizona, Utah, and Nevada also experienced several wildfires in 2020. These
have presented the USA with significant uncertainty, which has further af-
fected the tourists’ arrivals in these states.
Let us now see the time-varying causality from Geopolitical risk to tourism.
The finding from forwarding recursive causality does not detect any sig-
nificant causality for the whole study period between geopolitical risk and
tourism. The rolling causality reveals that the geopolitical risk granger causes
tourist arrivals for 9 days between 5 April and 13 April 2020 (first wave).
Another significant causality is observed at the date of 22 April 2020. On the
other hand, recursive evolving causality shows almost the similar impact like
rolling causality but with causality running 1 day further to 14 April 2020.
Another interesting finding from the recursive evolving causality procedure is
that geopolitical risk granger causes tourist arrivals for 5 days from 17 April to
22 April 2020 (first wave). The finding emphasizes the role of geopolitical
stability in bringing tourists’ revenues into the economy and is validated by
several previous studies. The study is congruent with Hailemariam and
Ivanovski’s (2021a) findings, who analyzed geopolitical and tourism link-
age for the USA and showed that geopolitical risk negatively affects the
tourism service exports in this country. The finding also reflects the estimation
Shahzad et al. 19

result of Akadiri et al. (2020), where the authors discovered a unidirectional

causality from geopolitical risk to tourism. Moreover, Lee et al. (2021) found
that geopolitical risk will have a greater negative impact on tourism in times of
pandemic outbreaks than the normal times. Lee and Chen (2021) also found
that if a country has a higher level of stability, it might lead to higher tourist
visitors in the country. According to Demir, Gozgor, and Paramati (2019),
political risk causes a significant decrease in inbound tourism. All these
studies, therefore, lend support to our findings.
Following this, the causality from TA to GPR is tested via three causality
procedures. While forward recursive and rolling causality do not identify any
significant episode, the recursive evolving causality suggests a significant
causality episode for 12 days between 11 and 22 November 2020 (third wave).
The finding confirms the result of Lee et al. (2021), where the authors
demonstrated causality from tourists’ arrivals to geopolitical risk for the five
countries of Europe and Asia. The bidirectional causality above can be seen in
April and November when tourists’ arrivals and geopolitical risk affected each
other. The geopolitical risk of November month coincides with the event of
the USA presidential election of 2020.
In addition to the impacts between tourism and COVID, it is also important
to analyze the causal relationship between GPR and COVID-19 and EPU and
COVID-19. The tourists are considered rational consumers because, before
traveling, they always consider the cost-benefit of traveling where cost in-
volves the risk factors they may get exposed to, and the benefits include their
satisfaction level (Lee et al., 2021). On the other hand, COVID-19 is also
considered the major geopolitical shock for all the world countries, which will
certainly affect almost every sector of the economy, including tourism (Sharif
et al., 2020).
Now, Figure 4 shows the causality between GPR and COVID. For GPR to
COVID or COVID to GPR, the forward recursive causality fails to detect any
significant causality for our study period. Rolling window causality detects a
significant episode of only 1 day on 10 May, 2020 while the Recursive
evolving causality identifies several episodes for GPR to COVID causality.
The first episode is detected on May 10, 2020 (first wave), which is similar to
what has been found from rolling window causality. But recursive evolving
causality also identifies causality on 6 June 2020, from 26 June to July 1 for
6 days, from July 3 to July 9 for 7 days, from 23 August to 22 October 2020 for
61 days (second and third waves). The finding is consistent with Lee et al.
(2021), who found that pandemic outbreaks can exacerbate outbreaks of
geopolitical risk on tourism.
Additionally, the USA, especially Minneapolis city, experienced signifi-
cant local unrest during June–October, 2020 as the protests took place due to
the murder of George Floyd on May 25, 2020, by the Minneapolis Police
20 Evaluation Review 0(0)

Department. (Laurie, 2020; Lauritsen, 2020). These political protests might

have affected the COVID incidence in this country.
Now, while examining causality from COVID to GPR, literature has
identified several mechanisms. For instance, COVID-19 has led to huge
unemployment across the globe, especially in the USA economy. Such facts
argue that unemployment further leads to a decline in people’s income, and as
a result, the protests and violence may increase, implying a surge of geo-
political risk. But on the other hand, travel restrictions may also indicate a
lower probability of terrorism-related activities, suggesting lower geopolitical
risk during COVID-19 (Wang et al., 2020). Therefore, there might be two
effects occurring simultaneously, which can be found from our result. Looking
at the causal impact from COVID to GPR, the result reveals several significant
episodes from the rolling and recursive evolving causality methods. The
rolling causality detects causality on September 6, November 2 and 7 and
from November 11 to 14 for 4 days in 2020.
On the other hand, the recursive evolving causality suggests causality from
June 4 to 6 and June 9 to 11 for 3 days, from September 6 to October 1 for
26 days, on November 2, from November 11 to 15 for 5 days, November 18 to
19 and November 21 to 22 for 2 days, during the year 2020 (second and third
waves). The significant causality is detected during the October–November
months of 2020. This again can be attributed to the US presidential election of
2020, a major political event in the world for several decades. The outcome is
consistent with Sharif et al. (2020), who found that an increase in COVID-19
cases, which has been exponential so far, is associated with a substantial rise in
the geopolitical risk of the USA. The authors note that this substantial increase
in geopolitical risk has been mainly due to the USA’s COVID-19 and oil price
shock.
Rajendran (2021) noted that the pandemic outbreak escalated tensions
around social injustice, health care access, and economic inequality. There-
fore, it is very much a geopolitical issue. Moreover, geopolitical risk has been
multiyear high due to the COVID-19 outbreak. But our result can also be
explained by the fact that the higher fatality rate of COVID-19 brings more
lockdown and restrictions within the boundary of an economy, reducing
geopolitical risk (Wang et al., 2020). Therefore, from the causality result, it
might be said that COVID might have increased geopolitical risks in the USA.
Still, at the same time, restrictions and lockdown imposed due to this virus
might have reduced some geopolitical events which could have taken place
had there been no lockdown.
Finally, the causal association between COVID and EPU is presented in
Figure 5. While the forward recursive causality detects a significant episode
only on April 6, 2020, from EPU to COVID, the other two causality tech-
niques identify several more causality episodes in the April, July, August–
September, and November (first, second, and third waves). For instance,
Shahzad et al. 21

causality during July 24, from July 27 to August 6 for 11 days are detected by
rolling causality and during July 24, July 27 to August 3 and August 5 to 12
for 8 days, September 7 to 26 for 20 days and November 13 to 14 for 2 days by
the recursive evolving causality. During these turbulent times of pandemic
outbreak, the economic policies have been highly uncertain, which further
exacerbated the pandemic situation as the uncertain economic policies af-
fected all the market participants and distorted the vision of the country’s
economy. Historically, economic policy uncertainty is associated with adverse
effects such as lower economic performance and higher unemployment. This
can be seen from the USA where the pandemic has led to the stoppage of
production, and as a result of uncertainty, many workers were laid off, and
unemployment even rose 14.7% in April 2020 (Al-Thaqeb et al., 2020).
Moreover, higher COVID cases and the fatality rate can lead to delay reactions
of the authorities involved, suggesting an increase in policy uncertainty
(Albulescu, 2020). This demonstrates that a lack of capacity to coordinate and
cooperate during the outbreak can further exacerbate the already existing
situation.
Now, for causality from COVID to EPU, unlike other causal associations
detected before, all the three causality techniques identify significant causality
episodes during our study period. From all three tests, 3 days causality from
April 2 to 4 is confirmed (first wave). The causality from April 7 to 24 for
18 days is confirmed by forwarding recursive and rolling causality, while
recursive evolving causality detects the causality for 17 days during almost the
same period. The causality from April 28 to May 1, June 28 to July 9 for
12 days, July 19 to 20 for 2 days, and August 28 to November 22 for 87 days is
detected by Forward recursive causality (first, second, and third waves).
Moreover, rolling causality confirms the COVID to EPU causality during
April 26 and 29, August 30, September 6, September 8 to 23 for 16 days,
September 26 to 27 for 2 days, September 29 to October 7 for 9 days, October
10 to 19 for 9 days, and during 26 October 2020 (first, second, and third
waves). However, from recursive evolving causality procedure, apart from
April 2 to 4, 8 to 24, 26 and 29–30, causality is detected from June 28 to 29 for
2 days, June 3 to 5 for 3 days, and finally from September 6 to November 16,
2020, for 72 days (first, second, and third waves). This segment of the
outcome can be justified because announcements of COVID cases and deaths
each day which increased in the USA substantially during the above periods
might have also increased the economic policy uncertainty of the country
(Albulescu, 2020). For example, in July 2020, as the previous round of
COVID aid was running out in the USA and the next round was about to be
declared, lots of uncertainty and disagreements related to several issues such
as aid for the unemployment and school funding existed which created policy
uncertainty in this country (ABC, 2020). The outcome is also in line with
22 Evaluation Review 0(0)

Sharif et al. (2020), who found a profound impact of COVID-19 on economic

policy uncertainty.

Concluding Remarks and Implications

In the 21st century, tourism is one of the fastest and largest growing sectors
worldwide, which substantially contributes to a nation’s GDP through the
increase in employment generation and governmental revenues. However, this
sector witnessed a major blow with the onset of deadly COVID-19 in late
2019. This pandemic adversely affects all the economic sectors, supply chain,
education, social life, and even the governance system of countries and
destroys many lives and livelihoods. Hit hard by the pandemic, the United
States of America is also going through different geopolitical risks and policy
uncertainty risks simultaneously, and there lies the contribution of this study.
This paper aims to investigate whether the geopolitical risk, policy un-
certainty, and COVID-19 influence the daily tourist arrivals in the USA.
Further, the study examines the time-varying forward recursive, rolling, and
recursive evolving causality between the key variables of interest. According
to the economic reports, tourism contributes to the $1.6 trillion in GDP,
supporting 7.8 million US jobs. Such facts demonstrate that tourism is the
largest and single source of revenue of capital inflows for several states of the
United States. For the empirical examination, the study employs novel
econometric techniques such as time-varying causality with forward and
recursive and recursive evolving causality on the daily data of variables
covering the period from January 22, 2020, to November 22, 2020.
The empirical results indicate some interesting findings. Firstly, the cau-
sality estimates demonstrate that COVID-19 daily cases granger cause the
tourist arrivals. Such finding might be due to the traveling restrictions for
international tourists in the United States from June 2020, to November 2020.
During this period, the USA witnessed a sudden surge in COVID-19 cases in
half of the states, the department of homeland security imposed a ban on
tourism from Canada, Mexico, and other countries. Concerning the causality
from tourism toward COVID-19, the causality results indicated recursive
evolving causality in the period of July 2020. Secondly, the empirical results
of recursive causality indicate significant variations between the tourist ar-
rivals, geopolitical risk, COVID-19 daily cases, and policy uncertainty. The
empirical findings conclude that there is unidirectional causality from eco-
nomic policy uncertainty toward the tourist arrivals during the high uncertain
period. It is justified that during the high uncertain times (in a pandemic)
people might think to delay or cancel the travelling plans. Further, different
states in the United States have different traveling regulations. The causality
between policy uncertainty and tourism was found in June to August and
November, which might be due to the wildfire in the western part of the USA.
Shahzad et al. 23

Further, the empirical findings for geopolitical risk mention that geopo-
litical stability is an important factor for tourist revenues in the economy. Such
a finding is very interesting and eye-catching, it allows us to conclude that
geopolitical risk and stability is critical factor for tourism growth. The findings
further highlight bidirectional causality between tourism and geopolitical risk.
In addition, the empirical results highlight bidirectional causality among
geopolitical risk with tourist arrivals and geopolitical risk to COVID-19 daily
new cases. Lastly, the study concludes that COVID-19 might exaggerate the
uncertainty and adversely affect tourism growth.
The findings of this paper allow us to draw some fruitful implications and
innovative conclusions for the tourist destinations in the United States. The
governments and policymakers should strive to forecast the adverse impacts
of the tourism industry on the magnitude of COVID-19. The closure of
international borders for a certain time might be an effective policy to control
the pandemic. Further, the countries might introduce some contact tracing and
artificial intelligence-related software to reduce and trace the cases of COVID-
19. Such policies would adversely affect the revenues from tourism in the
short term; however, it would contain the infection in limited areas and contain
from spreading exponentially.
Several challenges severely hurt US tourism industry during Trump’s
administration as we have already mentioned before. But President Biden has
taken different steps to revive the tourism industry. For example, in January
2021, the ban on Muslim majority countries was lifted up. But many chal-
lenges lie ahead of the Biden administration in recovering the tourism sector
completely out of the risk uncertainty associated with COVID-19. Also,
problems like hate crime and racism need to be tackled all across the USA,
which induces economic uncertainty (Majcher, 2021). To revive the tourism
industry, the airline and transport sector should adopt new preventive strict
measures in line with public health safety regulations of the USA. Also, a
flexible booking system and travel warning may reduce the uncertainty as-
sociated with COVID-19 in the tourism industry. Avoiding misinfodemics is
also necessary during the pandemic, which can negatively influence tourists’
perspectives (Pandey et al., 2021).
Further, the innovative policies regarding geopolitical risk, pandemic, and
tourism might be useful to revive the tourism industry and other economic
sectors with the improvements on pandemic situation. The future research can
be conducted on the major tourist destinations in the world, role of vaccination
for tourism, and production sectors. Such interesting research might be fruitful
for gaining innovative policies regarding COVID-19, industry, and tourism
sector.
24 Evaluation Review 0(0)

Acknowledgments
We would like to extend our gratitude toward our employers for providing us the
infrastructure and resources to work in this research. No funding was received for this
work.

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) received no financial support for the research, authorship, and/or
publication of this article.

Data Availability
Data are available upon request from the corresponding author.

ORCID iDs
Umer Shahzad  https://orcid.org/0000-0002-7010-4054
Muhammad Ramzan  https://orcid.org/0000-0001-7803-7960

Note
1. https://www.forbes.com/sites/johndrake/2020/12/07/the-real-cause-of-americas-
third-wave-of-covid-19/?sh=2f32299812fd

References
ABC, (2020). Next COVID-19 relief bill: Bipartisan support for 2nd stimulus check,
but disagreements on unemployment aid, school funding, more.
Akadiri, S. S., Alola, A. A., & Uzuner, G. (2020). Economic policy uncertainty and
tourism: evidence from the heterogeneous panel. Current Issues in Tourism. 23,
2507–2514. https://doi.org/10.1080/13683500.2019.1687662
Akadiri, S. S., Eluwole, K. K., Akadiri, A. C., & Avci, T., (2020). Does causality
between geopolitical risk, tourism and economic growth matter? Evidence from
Turkey. Journal of Hospitality and Tourism Management. 43, 273–277. https://
doi.org/10.1016/j.jhtm.2019.09.002
Al-Thaqeb, S. A., Algharabali, B. G., & Alabdulghafour, K. T. (2020). The pandemic
and economic policy uncertainty. International Journal of Finance & Economics.
1–11. https://doi.org/10.1002/ijfe.2298
Alam, M. S., & Paramati, S. R. (2016). The impact of tourism on income inequality in
developing economies: Does Kuznets curve hypothesis exist? Annals of Tourism
Research, 61, 111–126. https://doi.org/10.1016/j.annals.2016.09.008
Shahzad et al. 25

Albulescu, C. (2020). Do COVID-19 and Crude Oil Prices Drive the US Economic
Policy Uncertainty? SSRN Electronic Journal, 25–26. https://doi.org/10.2139/
ssrn.3555192
Antonakakis, N., Gupta, R., Kollias, C., & Papadamou, S. (2017). Geopolitical risks
and the oil-stock nexus over 1899-2016. Finance Research Letters, 23, 165–173.
https://doi.org/10.1016/j.frl.2017.07.017
Baker, S. R., Bloom, N., & Davis, S. J. (2016). Measuring economic policy un-
certainty*. The quarterly Journal of Economics, 131(4), 1593–1636. https://doi.
org/10.1093/qje/qjw024
Balcilar, M., Bonato, M., Demirer, R., & Gupta, R. (2018). Geopolitical risks and stock
market dynamics of the BRICS. Economic Systems, 42(2), 295–306. https://doi.
org/10.1016/j.ecosys.2017.05.008
Balcilar, M., Ozdemir, Z. A., & Shahbaz, M. (2019). On the time-varying links be-
tween oil and gold: New insights from the rolling and recursive rolling ap-
proaches. International Journal of Finance & Economics. 24, 1047–1065. https://
doi.org/10.1002/ijfe.1704
Balli, F., Shahzad, S. J. H., & Salah Uddin, G. (2018). A tale of two shocks: What do
we learn from the impacts of economic policy uncertainties on tourism? Tourism
Management, 68, 470–475. https://doi.org/10.1016/j.tourman.2018.04.008
Balsalobre-Lorente, D., Driha, O. M., Leitão, N. C., & Murshed, M. (2021). The
carbon dioxide neutralizing effect of energy innovation on international tourism in
EU-5 countries under the prism of the EKC hypothesis. Journal of Environmental
Management, 298, 113513. https://doi.org/10.1016/j.jenvman.2021.113513
Çağlı, E. Ç. (2019). The causality between consumer confidence index and stock
returns: Evidence from recursive evolving granger causality test. J. Yasar Univ.
14, 164–172.
Caldara, D., & Iacoviello, M. (2018). Measuring geopolitical risk, international finance
discussion paper. https://doi.org/10.17016/ifdp.2018.1222
Chen, C.-M., Hua, K.-T., Chyou, J.-T., & Tai, C.-C. (2020). The effect of economic
policy uncertainty on hotel room demand-evidence from Mainland Chinese and
Japanese tourists in Taiwan. Current Issues in Tourism, 23(12), 1443–1448.
https://doi.org/10.1080/13683500.2018.1556621
Demir, E., Gozgor, G., & Paramati, S. R. (2019). Do geopolitical risks matter for
inbound tourism? Eurasian Business Review, 9, 183–191. https://doi.org/10.
1007/s40821-019-00118-9
Demir, E., Gozgor, G., & Paramati, S. R. (2020). To what extend economic uncertainty
effects tourism investments? Evidence from OECD and non-OECD economies.
Tourism Management Perspectives, 36, 100758. https://doi.org/10.1016/j.tmp.
2020.100758
Demir, E., & Gozgor, G. (2018). Does economic policy uncertainty affect Tourism?
Annals of Tourism Research, 69, 15–17. https://doi.org/10.1016/j.annals.2017.12.
005
26 Evaluation Review 0(0)

Demir, E., Simonyan, S., Chen, M.-H., & Marco Lau, C. K. (2020). Asymmetric
effects of geopolitical risks on Turkey’s tourist arrivals. Journal of Hospitality and
Tourism Management, 45, 23–26. https://doi.org/10.1016/j.jhtm.2020.04.006
Dragouni, M., Filis, G., Gavriilidis, K., & Santamaria, D. (2016). Sentiment, mood and
outbound tourism demand. Annals of Tourism Research 60, 80–96. https://doi.
org/10.1016/j.annals.2016.06.004
Drake, J. (2020). The real cause Of America’s third wave of Covid-19. https://www.
forbes.com/sites/johndrake/2020/12/07/the-real-cause-of-americas-third-wave-of-
covid-19/?sh=2f32299812fd
Fotiadis, A., Polyzos, S., & Huan, T.-C. T. C. (2021). The good, the bad and the ugly on
COVID-19 tourism recovery. Annals of Tourism Research, 87, 103117. https://
doi.org/10.1016/j.annals.2020.103117
Gallego, I., & Font, X. (2020). Changes in air passenger demand as a result of the
COVID-19 crisis: using Big Data to inform tourism policy. Journal of Sustainable
Tourism, 1–20. https://doi.org/10.1080/09669582.2020.1773476
Georgetown University (2020). The covid analysis and mapping of policies [WWW
Document]. URL https://www.covidlocal.org/amp
Gozgor, G., & Ongan, S. (2016). Economic policy uncertainty and tourism demand:
empirical evidence from the USA. International Journal of Tourism Research,
19(1), 99–106. https://doi.org/10.1002/jtr.2089
Hailemariam, A., & Ivanovski, K. (2021a). The impact of geopolitical risk on tourism.
Current Issues in Tourism, 24, 3134, 3140. https://doi.org/10.1080/13683500.
2021.1876644
Hailemariam, A., & Ivanovski, K. (2021b). The effect of economic policy uncertainty
on US tourism net exports. Tourism Economics, 13548166211025334.
Hammoudeh, S., Ajmi, A.N., & Mokni, K. (2020). Relationship between green bonds
and financial and environmental variables: A novel time-varying causality. En-
ergy Economics, 92. https://doi.org/10.1016/j.eneco.2020.104941
Hurn, S., Phillips, P.C.B., & Shi, S.-P. (2016). Change Detection and the Causal Impact
of the Yield Curve. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.
2881954
ILO-OECD. (2020). The impact of the COVID-19 pandemic on jobs and incomes in
G20 economies. Saudi Arabia’s G20 Presidency.
Jiang, Y., Tian, G., Wu, Y., & Mo, B. (2020). Impacts of geopolitical risks and
economic policy uncertainty on Chinese tourism-listed company stock. Inter-
national Journal of Finance & Economics, 27, 320–333. https://doi.org/10.1002/
ijfe.2155
Khalfaoui, R., Solarin, S. A., Al-Qadasi, A., & Ben Jabeur, S. (2022). Dynamic
causality interplay from COVID-19 pandemic to oil price, stock market, and
economic policy uncertainty: evidence from oil-importing and oil-exporting
countries. Annals of Operations Research. https://doi.org/10.1007/s10479-021-
04446-w
Shahzad et al. 27

Laurie, J. (2020). Weeks Later, 500 People Still Face Charges for Peacefully Protesting
in Minneapolis.
Lauritsen, J. (2020). I Believe In Justice: Hundreds Of Protesters Gather Outside
George Floyd Pretrial Hearing.
Lee, C. C., & Chen, M. P. (2020). The impact of COVID-19 on the travel and leisure
industry returns: Some international evidence. Tourism Economics,
1354816620971981.
Lee, C.-C., & Chen, M.-P. (2021). Do country risks matter for tourism development?
International evidence. Journal of Travel Research, 60(7), 1445–1468.
Lee, C. C., Olasehinde-Williams, G., & Akadiri, S. S. S., 2021. Geopolitical risk and
tourism: Evidence from dynamic heterogeneous panel models. International
Journal of Tourism Research, 23, 26–38. https://doi.org/10.1002/jtr.2389
Lu, Z., Zhu, L., Li, Z., Liang, X., & Zhang, Y. (2022). The effects of passenger risk
perception during the COVID-19 pandemic on airline industry: Evidence from the
United States stock market. Frontiers in Psychology, 12, 795940.
Majcher, K. (2021). The US is slipping as a tourist destination, and more than Covid is
to blame
Manzoor, F., Wei, L., & Asif, M. (2019). The contribution of sustainable tourism to
economic growth and employment in Pakistan. International Journal of Envi-
ronmental Research and Public Health, 16(19). https://doi.org/10.3390/
ijerph16193785
Ongan, S., & Gozgor, G. (2018). Tourism demand analysis: The impact of the eco-
nomic policy uncertainty on the arrival of Japanese tourists to the USA. Inter-
national Journal of Tourism Research, 20(3), 308–316. https://doi.org/10.1002/
jtr.2182
Pandey, V., Talan, A., Mahendru, M., & Shahzad, U. (2021). Studying the psychology
of coping negative emotions during COVID-19: A quantitative analysis from
India. Environmental Science and Pollution Research, 1–18.
Pata, U. K., & Balsalobre-Lorente, D. (2022). Exploring the impact of tourism and
energy consumption on the load capacity factor in Turkey: A novel dynamic
ARDL approach. Environmental Science and Pollution Research, 29(9),
13491–13503. https://doi.org/10.1007/s11356-021-16675-4
Polyzos, S., Samitas, A., & Spyridou, A. E. (2020). Tourism demand and the COVID-
19 pandemic: An LSTM approach. Tourism Recreation Research, 46(2),
175–187. https://doi.org/10.1080/02508281.2020.1777053
Rajendran, S. (2021). Geopolitical risks arising from COVID-19 pandemic.
Shahbaz, M., Hoang, T. H. V., Mahalik, M. K., & Roubaud, D. (2017). Energy
consumption, financial development and economic growth in India: New evi-
dence from a nonlinear and asymmetric analysis. Energy Economics, 63,
199–212. https://doi.org/10.1016/j.eneco.2017.01.023
Shahzad, F., Shahzad, U., Fareed, Z., Iqbal, N., Hashmi, S. H., & Ahmad, F. (2020).
Asymmetric nexus between temperature and COVID-19 in the top ten affected
28 Evaluation Review 0(0)

provinces of China: A current application of quantile-on-quantile approach.

Science of The Total Environment, 736, 139115.
Sharif, A., Aloui, C., & Yarovaya, L. (2020). COVID-19 pandemic, oil prices, stock
market, geopolitical risk and policy uncertainty nexus in the US economy: Fresh
evidence from the wavelet-based approach. International Review of Finance
Analysis, 70, 101496. https://doi.org/10.1016/j.irfa.2020.101496
Sharma, C. (2019). Testing the asymmetric effects of the economic policy uncertainty
on the tourism demand in India. Tourism Economics, 1–7. https://doi.org/10.1177/
1354816619894080
Sharma, G. D., Thomas, A., & Paul, J. (2021). Reviving tourism industry post-COVID-
19: A resilience-based framework. Tourism Management Perspectives, 37,
100786. https://doi.org/10.1016/j.tmp.2020.100786
Sharma, G. D., Kumar, A., Jain, M., Yadav, A., & Srivastava, M. (2021). COVID-19
and environmental concerns: A rapid review. Renewable and Sustainable Energy
Reviews, 148, 111239. https://doi.org/10.1016/j.rser.2021.111239
Sharma, G. D., Sarker, T., Rao, A., Talan, G., & Jain, M. (2022). Revisting con-
ventional and green finance spillover in post-COVID world: Evidence from
robust econometric models. Global Finance Journal, 51, 100691. https://doi.org/
10.1016/j.gfj.2021.100691
Shi, S., Hurn, S., & Phillips, P. C. (2020). Causal change detection in possibly in-
tegrated systems: Revisiting the money-income relationship. Journal of Financial
Econometrics, 18(1), 158–180.
Shi, S., Phillips, P. C., & Hurn, S. (2018). Change detection and the causal impact of the
yield curve. Journal of Time Series Analysis, 39(6), 966–987. https://doi.org/10.
1111/jtsa.12427
Sikarwar, E. (2021). Time-varying foreign currency risk of world tourism industry:
Effects of COVID-19. Current Issues in Tourism, 24(7), 887–891. https://doi.org/
10.1080/13683500.2020.1868413
Swanson, N. R. (1998). Money and output viewed through a rolling window. Journal
of Monetary Economics. 41, 455–474. https://doi.org/10.1016/s0304-3932(98)
00005-1
Tate, C., Deerwester, J., & Hines, M. (2020). US-Canada, Mexico border closures
extended into late August as American COVID-19 cases surge.
The World Travel and Tourism Council. (2021). WTTC research reveals global Travel
& Tourism sector suffered a loss of almost US$4.5 trillion in 2020 due to the
impact of COVID-19. WTTC.
Thoma, M. A. (1994). Subsample instability and asymmetries in money-income
causality. Journal of Econometrics, 64(1), 279–306. https://doi.org/10.1016/
0304-4076(94)90066-3
Tiwari, A. K., Das, D., & Dutta, A. (2019). Geopolitical risk, economic policy un-
certainty and tourist arrivals: Evidence from a developing country. Tourism
Management, 75, 323–327. https://doi.org/10.1016/j.tourman.2019.06.002
Shahzad et al. 29

US Travel Association (2021). Weekly Coronavirus Impact On Travel Expenditures In

The U.S. Washington.
Uzuner, G., Akadiri, S. S., & Alola, A. A. (2020). Tourist arrivals in four major
economies: Another side of economic policy uncertainty and fear. Environmental
Science and Pollution Research 27(23), 29659–29665. https://doi.org/10.1007/
s11356-020-09219-9
Wang, L., Li, C., Chen, X., & Zhu, L. (2020). Causal Relationship Between the Spread
of the COVID-19 and Geopolitical Risks in Emerging Economies. Frontiers in
Public Health. 8, 1–5. https://doi.org/10.3389/fpubh.2020.626055
Wen, J., Wang, W., Kozak, M., Liu, X., & Hou, H. (2020). Many brains are better than
one: The importance of interdisciplinary studies on COVID-19 in and beyond
tourism. Tourism Recreation Research, 46(2), 310–313. https://doi.org/10.1080/
02508281.2020.1761120
World Travel and Tourism Council. (2020). Travel & Tourism: Global Economic
Impact & Trends 2020. WTTC.
Xuefeng, Z., Razzaq, A., Gokmenoglu, K. K., & Rehman, F. U. (2021). Time varying
interdependency between COVID-19, tourism market, oil prices, and sustainable
climate in United States: Evidence from advance wavelet coherence approach.
Economic Research-Ekonomska Istraživanja, 1–23.
Yan, H. M., Cheri Moshtaghian, A., & Vercammen, P. & (2020). California sets new
record for land torched by wildfires as 224 people escape by air from a “hellish”
inferno. CNN.
Yan, Y., Shah, M. I., Sharma, G. D., Chopra, R., Fareed, Z., & Shahzad, U. (2022). Can
tourism sustain itself through the pandemic: Nexus between tourism, COVID-19
cases and air quality spread in the ‘Pineapple State’Hawaii. Current Issues in
Tourism, 25(3), 421–440.
Zhang, H., Jiang, Z., Gao, W., & Yang, C. (2022). Time-varying impact of economic
policy uncertainty and geopolitical risk on tourist arrivals: Evidence from a
developing country. Tourism Management Perspectives, 41, 100928.
Zichu, J., & Masih, M. (2020). Nexus of infrastructure investment, economic growth
and domestic credit level: Evidence from China based on nonlinear ARDL
approach.

Author Biographies
Dr Umer Shahzad currently works as Assistant professor in Anhui Uni-
versity of Finance and Economics. Dr. Umer’s current research is focused on
green innovation, energy finance, trade aspects, economic complexity, and
ecological issues. Dr. Umer’s research has been published in several top-
notch Journals such as; Current Issues in Tourism, Technological Forecasting
& Social Change, IEEE Transactions on Engineering Management, Re-
newable & Sustainable Energy Reviews, Energy & Buildings among others;
focused on environmental aspects of energy, climate change, tourism and
30 Evaluation Review 0(0)

international trade. He is also serving as guest editor and review editor in

several Journals.
Muhammad Ibrahim Shah is currently pursuing his MSc in Agricultural and
Resource Economics at the University of Alberta, Canada. He has completed
both his bachelor’s and master’s degree in Economics from University of
Dhaka. His research interests include Applied Macroeconomics, Environ-
mental economics, Energy Economics, sustainable development etc.
Muhammad Ramzan is pursuing his PhD in international trade at Shandong
University of Finance and Economics, Shandong, Jinan, China. Since 2013,
he has worked at several universities in Pakistan. His research interests include
stock markets, governance, sustainable development, financial economics,
energy economics, environmental economics, and agricultural economics.
Buhari Doğan is working as PhD in the Faculty of Economics and Ad-
ministrative Sciences in Suleyman Demirel University. Then, he holds BA
from Kocaeli University, Department of Economics of the Faculty of Eco-
nomics and Administrative Sciences. He completed his MA in Department of
Econometrics at Suleyman Demirel University, and then he completed his
Ph.D. education in Department of Economics at the same university. He writes
on issues of energy, environmental economic growth, and globalization.
Prof. Ahdi Noomen Ajmi holds a PhD in Mathematical Economics and
Econometrics from the University of the Mediterranean (France) and obtains
his HDR (Habilitation for Supervising Doctoral Research) in Economics from
University of Manouba (Tunisia). He joined Prince Sattam bin Abdulaziz
University (KSA) as Professor of Economics. Previously, he served as As-
sistant Professor in Quantitative Methods at ESC of Tunis (Tunisia) and
Teaching Assistant at High Institute of Management of Tunis. His principal
research areas concern: time series analysis, financial econometrics, long
Memory processes, causality analysis, energy economics and volatility
modeling.