Impacts of Terrorist Events on Tourism Development: Evidence from Asia

Abstract

To examine terrorism impacts on international tourism industry in Asia, we collected data from 46 countries from 1995 to 2016, with a total of 1,012 samples. Our results showed that terrorism had a more consistent negative impact on international tourism revenue (ITR) than on international tourist arrivals. While the frequency of terrorism had a significant positive impact on ITR, when controlling for terrorist events which did not cause fatalities, such positive impact changed to negative. A further finding was that the Asian tourism market greatly developed following the 911 event, even though ITR decreased in Muslim countries with high risk of terrorist attacks. The current study makes a contribution to the understanding of terrorism features which may prove useful to strengthening antiterrorism policy in the tourism sector.

Keywords

tourism terrorism frequency severity Islam 911 event

Introduction

Tourism is highly sensitive to internal and external shocks as diverse as economic downturns, natural disasters, epidemic disease and terrorist events (Chesney et al., 2011; Schmude et al., 2018; Sönmez et al., 1999). Among these shocks, terrorism has evolved into a major global concern for tourism industry (Liu & Pratt, 2017; Llorca-Vivero, 2008). Many previous studies have examined the impacts of terrorism on tourism development in European countries and the United States (W. Enders & Sandler, 1991; Pennington-Gray et al., 2014; Samitas et al., 2018), but it is only recently that scholars have started paying attention to the terrorism-tourism relationship in Asian countries (Bassil et al., 2019; Feridun & Sezgin, 2008).

The Asian tourism market has become one of the largest and fastest-growing economic sectors in the world. In this market in 2016, international tourist arrivals (ITA) reached 346.4 million (28% of the global market share), and revenue from international tourists reached US$ 377.5 billion (31% of the global market share; UNWTO, 2018). This occurred despite the fact that Asia is the most vulnerable region to terrorism in the world, represented by the evidence that five Asian countries—Iraq, Afghanistan, India, Pakistan, and the Philippines—were ranked among the countries most severely affected by terrorism in 2016 (Global Terrorism Database [GTD], 2016).

Given the scarce literature on the study of the terrorism-tourism relationship in Asia, the present study collected panel data on 46 countries in the period of 1995 to 2016 to empirically examine the extent to which terrorist events have directly impacted on the development of the international tourism industry. This article contributes to the literature by connecting three lines of inquiry: studies differentiating between the impacts of the frequency and severity of terrorist events on the international tourism industry, studies comparing the impacts of terrorism on tourism between different groups (Muslim and non-Muslim countries, low-risk and high-risk countries in terms of terrorist attacks, and four geographically divided regions), and studies examining the broad influence of mass media report and cultural distance on tourism industry.

Literature Review and Hypotheses

Attributes of Terrorism

Much of the existing literature in this field contends that terrorist behavior could be seen as rational (Conrad & Greene, 2015). This consideration is based on the assumption that terrorist organizations hold internally consistent sets of beliefs, values, and imaginations of the environment (Conrad & Greene, 2015; Crenshaw, 1981). Terrorist organizations select the strategies and tactics that can offer them the highest expected utility and, as such, convince increasing numbers of people to participate, or at least sympathize with their behavior (Conrad & Greene, 2015). As rational actors, terrorists deploy their extranormal violence or its threatened means on their target audience to achieve a political objective through intimidation or fear (W. Enders & Sandler, 1991). Through attacking civilians, the military, police or government personnel, a terrorist organization can proclaim that the government is unable to protect the population, and can propagate the notion that the organization is more capable and credible (Conrad & Greene, 2015). Hence, terrorist acts not only aim to incur immediate and direct casualties, but target at a larger audience in order to succeed in social, political, or religious causes through violence or the threat of violence (Ivianski, 1987; Samitas et al., 2018).

In order to describe the specific attributes of terrorism, scholars have developed various measures with regard to terrorist acts and results (Conrad & Greene, 2015; Pizam, 1999; Walter Enders et al., 1992). For instance, based on an extensive literature review of violent events, Pizam (1999) created a typology identifying five attributes of violent acts at tourist destinations; namely, motive, victim, location, severity, and frequency. Other researchers have suggested that terrorist acts have differential effects on the society depending on the severity of the event and the frequency of occurrence (Conrad & Greene, 2015; Pizam & Fleischer, 2002). Pizam and Fleischer (2002) claim that terrorist acts resulting in mass destruction of life and property, followed by loss of life and bodily harm, have the strongest effect on tourism demand, and that acts occurring more frequently will exert a more adverse effect on tourism demand than those occurring less frequently. Furthermore, the severity of terrorist attacks can be used to measure the competitive capabilities of terrorist organizations. Conrad and Greene (2015) focused on two pieces of information regarding terrorist attacks: the level of severity, based on the type of target, and the level of severity of the methods used in the attack. These authors found that the severity of terrorist attacks based on both the type of target and the type of method employed increased dramatically as organizations faced greater levels of competition from other terrorist groups.

Empirical studies usually employ monthly or yearly data of terrorism incidents (Buigut & Amendah, 2016; Samitas et al., 2018), or data of the number of terrorist attacks (Afonso-Rodriguez & Santana-Gallego, 2018; Blomberg et al., 2004) to explore the effects of terrorist events on tourism. Most of these previous studies conclude that terrorism has detrimental effects on tourism industry. In line with this, and from the perspective of statistical analysis, the common attributes of terrorism as demonstrated by previous literature can be classified as follows: the number of terrorist events (NTE) and the number of casualties (fatalities or injuries) caused by terrorism. Therefore, in the present study, we define terrorism frequency as the NTE, and terrorism severity as the number of casualties caused by terrorist events.

Terrorism Damages to Tourism

Terrorism acts and their negative effects on tourism demand have been studied since 1980s, when time series data of tourism receipts were made available and accessible to researchers (W. Enders & Sandler, 1991). For a terrorist group, tourism destinations are easy to infiltrate, acts against tourists provide guaranteed international media coverage, and the impacts of an attack can be massive (Pizam, 1999; Sönmez et al., 1999). The risk of terrorism tends to intimidate tourists away from traveling more severely than other disasters (W. Enders & Sandler, 1991; Sönmez et al., 1999). Substantial declines in tourist arrivals due to terrorism have occurred in Greece (Samitas et al., 2018), the United States (Lepp & Gibson, 2003), Tunisia and Egypt (Perles-Ribes et al., 2018), and Kenya (Buigut & Amendah, 2016). As an example, security concerns in Kenya have witnessed a decline in visitor arrivals from a peak of 1.8 million in 2011 to approximately 1.5 million in 2013, whereby a 1% increase in conflict fatalities significantly reduced tourist arrivals by around 0.13% (Buigut & Amendah, 2016). Taken together, these findings reveal a direct and unidirectional negative impact of terrorist events on the number of domestic or international tourists.

This leads to the generation of three testable hypotheses involving a series of tests of the underlying logic of the detrimental effects of terrorism on tourism demand, namely, that more terrorist events will cause the tourism industry to suffer greater loss of tourists.

Hypothesis 1-1: The NTE has a negative impact on ITA.

Null Hypothesis, H_0-1-1: The NTE has no impact on ITA.

Alternative Hypothesis, H_a-1-1: The NTE has an impact on ITA.

Hypothesis 1-2: The number of fatalities caused by terrorist events has a negative impact on ITA.

Null Hypothesis, H_0-1-2: The number of fatalities caused by terrorist events has no impact on ITA.

Alternative Hypothesis, H_a-1-2: The number of fatalities caused by terrorist events has an impact on ITA.

Hypothesis 1-3: The number of injuries caused by terrorist events has a negative impact on ITA.

Null hypothesis, H_0-1-3: The number of injuries caused by terrorist events has no impact on ITA.

Alternative hypothesis, H_a-1-3: The number of injuries caused by terrorist events has an impact on ITA.

Regarding the economic impact of terrorist events on the performance of the tourism industry, previous studies mainly reflect a speculative negative impact (Buigut & Amendah, 2016; Drakos & Kutan, 2003; W. Enders & Sandler, 1991). For instance, using data for three Mediterranean countries, Drakos and Kutan (2003) found that the effect of domestic terrorism on the Greek market share represented an estimated loss of 3.64%, on that of Israel a loss of 0.67%, and on that of Turkey a loss of 5.21%, assuming that each country’s market share depends on a set of state variables (i.e., infrastructure, level of services, and weather). Buigut and Amendah (2016) used the declining numbers of foreign tourists due to terrorism in Kenya to estimate tourism revenue loss, stating that at the earning rate of 62,666 Kenya Shillings per visitor, an increase of one fatality translated into a reduction in annual tourist earnings of about 157.1 million Kenya Shillings per year.

Regarding the direct impact of terrorism on tourism revenue, very few studies have carried out an empirical analysis on this issue. While Arunatilake et al. (2001) argue that the economic costs of civil wars in Sri Lanka since 1983 may be at least equivalent to twice the country’s 1996 gross domestic products (GDP), they merely take conceptual and methodological problems into consideration. While Muckley (2010) estimates that a fatality as a result of terrorism imputes a minimum economic cost of £0.26 million pounds sterling on real foreign tourist revenue, he only discusses the case of Northern Ireland, rather than a broader region. Given the scant literature on the direct impact of terrorist events on international tourism revenue (ITR) in Asia, an empirical analysis on this issue is necessary to fill this gap. Hence, the current study proposes the following three hypotheses to address such research gap:

Hypothesis 2-1: The NTE has a negative impact on ITR.

Null Hypothesis, H_0-2-1: The NTE has no impact on ITR.

Alternative Hypothesis, H_a-2-1: The NTE has an impact on ITR.

Hypothesis 2-2: The number of fatalities caused by terrorist events has a negative impact on ITR.

Null Hypothesis, H_0-2-2: The number of fatalities caused by terrorist events has no impact on ITR.

Alternative Hypothesis, H_a-2-2: The number of fatalities caused by terrorist events has an impact on ITR.

Hypothesis 2-3: The number of injuries caused by terrorist events has a negative impact on ITR.

Null hypothesis, H_0-2-3: The number of injuries caused by terrorist events has no impact on ITR.

Alternative hypothesis, H_a-2-3: The number of injuries caused by terrorist events has an impact on ITR.

Religion, Media Report, and Cultural Distance

Scholars believe that terrorists might attack tourists to achieve ideological objectives, specifically in the name of religion or clashing cultures (Sönmez et al., 1999). When tourism represents a threat to well-established traditions, societal norms, value systems and religious convictions, the intention to protect these sacred beliefs can, regrettably, manifest itself through terrorist behavior (Sönmez et al., 1999). Specifically, in recent years, a recurring phenomenon is that, religion has been used as a fig leaf by violence (Rink & Sharma, 2018). One extreme terrorist group is the Islamic State in Iraq and Syria (ISIS), which aims to develop a state based on Islamic laws (Y. Hoffman, 2018). Hence, the religion of Islam has frequently been reported by media when terrorist events occur (Khan & Estrada, 2016).

There are some disagreements as to whether Islam itself, as a religion, encourages violent attacks on tourists. From the perspective of media reports, studies show that mass media tends to describe Islam by using negative connotations, and to depict Islam as a religion of strict taboos (Aziz, 1995). These reports may cause the viewer’s or reader’s mistrust and suspicion toward Islam. As terrorist organizations intentionally use media coverage to send their message and disperse fear among a broader audience (Conrad & Greene, 2015), terrorist attacks with a decidedly religious dimension could cause great harm to the tourism industry in Muslim countries. Nevertheless, in the field of tourism studies, very few scholars have examined the factor of Islam, as a religion, with regard to how this might affect tourism demand.

Attention from the mass media report even became measurement criteria to the success of terrorist activities. And hence, a few researchers claim that terrorist attacks would rarely occur in states that restrict the press’s ability to cover attacks (Asal & Hoffman, 2016; A. M. Hoffman et al., 2013). The denial of press freedom squelches reports about terrorist attacks, on one hand, and denies terrorist organizations the publicity they need, on the other hand (Asal & Hoffman, 2016). Using data on transnational terrorism between 1975 and 1995, A. M. Hoffman et al. (2013) find that states which protect press freedom are two times more likely to be targeted by foreign perpetrators than states that restrict it. Similarly, Keels and Kinney (2019) point out that, if the press is allowed to independently cover terrorist attacks, the use of violence by rebel groups could increase. These results indicate that the low degree of press freedom might be the reason why terrorist organizations decline to attack North Korea (A. M. Hoffman et al., 2013). However, there are controversies over the relationship between press freedom and terrorist activity. For instance, using Asal and Rethmeyer’s BAAD1 data on terrorist organizations, Asal et al. (2016) find that the propensity of organizations to launch foreign attacks appears unrelated to press freedom. While an increasing volume of studies paid attention to the predictive power of mass media report on terrorist activities, up to date, very few studies have investigated the impact of terrorist attacks on international tourism industry using press freedom as a control variable.

The impacts of cultural factors on international tourism industry have long been discussed by economists or sociologists (Armstrong et al., 1997; Cohen, 1979). Cultural distance, which implies the extent of cultural differences between destination countries and tourists’ home countries (Ng et al., 2007), is used to evaluate the influence of cultural factors on international tourism development. Hofstede suggests that there are six dimensions of cultural distance: power distance index (PDI), individualism versus collectivism (IDV), masculinity versus femininity (MAS), uncertainty avoidance index (UAI), long-term orientation versus short-term normative orientation (LTO), and indulgence versus restraint (IND; Hofstede, 2005; Hofstede & Bond, 1984). Using Hofstede’s cultural dimensions, Al Qudah et al. (2019) empirically demonstrate that there is an indirect relationship between culture and terrorist financing. Using a combination of historical and psychoanalytic approaches, Meftah (2018) states that the roots of sadistic terrorism crimes are cultural, namely mythology of heroes and revenge. Hence, this current study applies Hofstede’s six cultural dimensions as control variables to examine the impact of terrorist attacks on international tourism industry. Bearing in mind of these previous literature and research gaps, the current study proposes the following conceptual framework (Figure 1) which demonstrates the links between the study variables and analysis perspectives.

Figure 1

Conceptual Framework

Materials and Methods

Data Collection

We collected five types of data. The first type pertained to terrorist events, collected from the GTD. The GTD is an open-source database that includes information on terrorist events around the world from 1970 through to 2017. We selected data for all Asian countries apart from Mongolia and North Korea, due to a lack of related statistics in the GTD, with, ultimately, data collected for 46 countries and regions in Asia. We selected three data items: the NTE, the number of fatalities caused by terrorism (NTF), and the number of injuries caused by terrorism (NTI), all per year in the period of 1970 to 2017. The second type of data related to tourism development and was collected from the World Bank, which has stored data on tourism from 1995 to 2016. We selected two data items: ITA and ITR. The third type of data concerned economic development and was collected from the United Nations. We selected one item: GDP. The fourth type of data related to mass media report and was collected from Reporters without Borders. We selected one item: press freedom index (PFI). The range of scores to assess the state of press freedom in each country are set as follows: <15 points (good); 15.01 to 25 points (fairly good); 25.01 to 35 points (problematic); 35.01 to 55 points (bad); >55.01 points (very bad). The fifth type of data concerned cultural distance according to Hofstede’s six cultural dimensions and was collected from Geert Hofstede. We selected six items: PDI, IDV, MAS, UAI, LTO, and IND. In the end, our data covered 46 countries and regions in Asia in the period of 1995 to 2016. The total number of data samples was 1,012 (46 countries × 22 years). In addition, data characteristics, data source, and data definitions could be seen in online Supplement Table 3.

Measurements

Variables. We set two indexes of tourism development as dependent variables: ITA and ITR. To measure the impact of terrorist events on tourism development, we applied three indexes as explanatory variables: NTE, NTF, and NTI. Considering that the impacts of terrorism on tourism may have a time lag, we adopted three additional explanatory variables of terrorism from the previous year, as follows: NTE-1, NTF-1, and NTI-1.

Regarding control variables, we included the GDP index of economic development and the contribution of ITR to the GDP of each country, measured by ITR/GDP. Also, PFI and those six cultural dimensions (PDI, IDV, MAS, UAI, LTO, and IND) were involved. We then adopted two additional control variables: the time period in which the 911 event occurred, and the religion of Islam. Studies have pointed out that the impact of the 911 event left a legacy on international tourism, and that the mindset of international tourists fundamentally changed following this event (Khan & Estrada, 2016; Liu & Pratt, 2017; Llussa & Tavares, 2011). We constructed a dummy variable of time, with “1” representing “after 911” (the period 2002-2016), and “0” representing “before 911” (the period 1995-2001). We placed the year of 2001 itself into the “before 911” group because tourism statistics showed that the annual peak tourism season is July and August in most Asian countries and the 911 event occurred in September (Rosselló & Sansó, 2017). Similarly, we constructed a dummy variable of religion, with “1” representing “Muslim countries” and “0” representing “non-Muslim countries.” The classification criterion pertaining to Muslim and non-Muslim countries was that countries with a Muslim population of over 50% were categorized as Muslim, according to the 2016 populations of the 46 sampled Asian countries (United Nations Statistics Division, 2019). Thus, 27 countries were counted in the Muslim group, and 19 countries were in the non-Muslim group.¹ The variable of the Islam provided this study with the opportunity to conduct comparative studies in Asia. In addition, in the following section, we intend to further investigate differences of terrorism impacts on tourism from two perspectives: geographical location and terrorism risk.

Sample grouping methods. In order to conduct a comparative study of the impacts of terrorism on tourism in Asia, two independent criteria were employed in the current study to group the samples. First, from the perspective of geographical location, Asia was divided into four regions, as follows: East and Southeast Asia, including 14 countries; South Asia, including 7 countries; West Asia, including 19 countries; and Central Asia, including 6 countries.²

Second, from the perspective of differing degrees of terrorism risk, a distance (D) value was created and Asia was divided into two groups in terms of the possibility of terrorism attacks: the low-risk group and the high-risk group. The D value was measured by the following equation:

D = \sum (X_{i t} - M i n_{i t}) / (M a x_{i t} - M i n_{i t})

(Equation 1)

where X, Min, and Max represent terrorism damages in terms of NTE, NTF and NTI. Specifically, X represents a country’s damage, Max represents the damage done to the worst-affected country, and Min represents the damage done to the least-affected country; i represents a country with a value between 1 and 46, and t represents the time period, with the value ranging from 1995 to 2016. Accordingly, we ensured that the D value ranged from 0 to 1. The closer the value is to 1, the more damage the country suffered from terrorist activities. For each country, three D values were obtained with regard to NTE, NTF, and NTI. As there were no measurement units for the D value, the three D values were summed up and a ranking obtained for the 46 countries. Then, the 46 countries were evenly split, as follows: 23 countries with low D values were classified as the low-risk group, and the other 23 countries were classified as the high-risk group (see online Supplement Table 4).

Methodology. Our methodology involved investigating the impact of terrorist events on tourism development. Thus, the following equation was constructed:

\begin{array}{l} T o u r i s m_{i t} = C + α T e r r o r i s m_{i t} + β E c o n o m i c_{i t} + \\ γ R e p o r t + δ C u l t u r e + ε \end{array}

(Equation 2)

where i represents a country and t represents the time period, the same as in Equation 1. Tourism comprises ITA and ITR. Terrorism includes NTE, NTF, NTI, NTE-1, NTF-1, and NTI-1. Economic consists of GDP and ITR/GDP. Report represents PFI. Culture involves PDI, IDV, MAS, UAI, LTO, and IND. C is a constant term, $α, β, γ, δ$ are estimation coefficients, and ε is a random disturbance term.

Data Analysis

Three types of analysis methods were applied. First, we used the ordinary least square (OLS) regression to explore the potential correlates between variables. Second, we applied the panel data model (PDM) to test the validity of OLS results. Tests of the model assumptions were conducted using the fixed effect model and random effect model. The Hausman test was used to calculate the model fit. Third, the two-stage least square (2SLS) regression was used to explore the impacts of terrorist attacks on tourism development by constructing regression models with endogenous variables. The lagged variables of terrorist events (NTE, NTF, NTI, NTE-1, NTF-1, and NTI-1) were taken as the tool variable. The analysis was processed using Eviews 10, IHS Global Inc.

Results

The present study used a total of 1,012 samples. The Muslim group contained 594 samples and the non-Muslim group 418. The sample breakdown across the four geographical regions was as follows: East and Southeast Asia—308, South Asia—154, West Asia—418, and Central Asia—132. The low-risk and high-risk groups in terms of terrorist attacks both contained 506 samples.

Descriptive Analysis

The total Asia sample. Supplement Table 1 (available online) shows the results of the descriptive analysis of tourism development, terrorist events, GDP, PFI, and six cultural dimensions. Regarding tourism development, the average ITA to the Asian countries sampled was seen to be 4.45 million people per year (SD = 8.48 million), with a maximum of 59.27 million and a minimum of 700 people. The average ITR emerged as US$ 4.16 billion (SD = US$ 7.68 billion) per year, with a maximum of US$ 52.5 billion and a minimum of US$ 1 million. The average GDP per country per year was US$ 329 billion (SD = US$ 1.1 trillion), with a maximum of US$ 11.2 trillion and a minimum of US$ 0.29 billion.

Regarding terrorist events, the average NTE per country was found to be 90 (SD = 324) events per year, with a maximum of 3,926 events. The average NTF emerged as 202 (SD = 890) people per country per year, with the maximum number reaching 13,050 people. The average NTI was seen to be 342 (SD = 1,367) people per country per year, with the maximum number reaching 16,795 people.

Supplement Figure 1 (available online) shows the change trends in terms of tourism, terrorism, and economic indicators. Here, it can be seen that both tourism and economic indicators increased steadily during the period of 1995-2016. Regarding terrorism indicators, while there were sharp declines in several years (e.g., 1998 and 2011), as a whole, the trend was upward-going.

For PFI, the average score in Asia turned out to be 46.83 (SD = 24.04), with the maximum score reaching 140.67. This result indicates that the state of press freedom in Asia is comparatively bad. Regarding those six cultural dimensions, the mean value of PDI is 76.04 (SD = 14.97), the IDV is 31.59 (SD = 10.33), the MAS is 47.07 (SD = 11.01), the UAI is 64.04 (SD = 23.78), the LTO is 48.27 (SD = 21.86), and the IND is 36.72 (SD = 16.17).

Different sample groups. In terms of the comparative results regarding Muslim and non-Muslim countries, Supplement Figures 2 and 3 (available online) show the trend changes in terms of terrorism, tourism and GDP. From these, it can be seen that the terrorism indicators of non-Muslim countries did not increase significantly after 911 event, while they rose sharply in Muslim countries in the same period. Regarding the tourism indicators, these increased rapidly in non-Muslim countries, especially tourism revenue, while their growth slowed down in Muslim countries following the 911 event. For detailed statistics, see online Supplement Table 5.

Regarding the comparative results for the four geographic regions, Supplement Table 1 (available online) demonstrates that East and Southeast Asian countries had the highest average ITA (8.33 million people) and the largest average share of ITR (US$ 8.25 billion), while these same two figures in Central Asian countries were the lowest, with 1.04 million people and US$ 0.235 billion, respectively. In terms of terrorist events, East and Southeast Asian countries were found to have the lowest NTE (at 31), NTF (at 31), and NTI (at 86). South Asian countries had the highest average NTE, at 163. West Asian countries were found to have the largest share of NTF (at 287) and NTI (at 495).

Regarding the comparative results in the low-risk and high-risk groups, Supplement Figures 4 and 5 (available online) show the trend changes in terms of terrorism, tourism and GDP. The comparative results are very similar to those of the Muslim and non-Muslim groups. Specifically, the terrorism indicators in the low-risk group did not increase significantly after 2002, while they rose sharply in the high-risk group in the same period. Regarding the tourism indicators, these increased rapidly in the low-risk group, while their growth slowed down in the high-risk group from 2002. For detailed statistics, see online Supplement Table 5.

Overall, results of these different sample groups demonstrate characteristics of our data, that a number of countries in the sample have relatively similar (small or large) scores in indicators of either tourism or terrorism. Therefore, by splitting those 46 countries into different groups, we could further explore the near-real impacts of terrorism on tourism. The reason is that, huge fluctuations of indicators among the whole sample can be controlled.

Regression Analysis

We first examined the overall characteristics of the impact of terrorism on tourism development in the Asian Continent. Second, we analyzed and conducted three types of comparative studies on the impacts of terrorism on tourism with regard to the following groups: Muslim and non-Muslim groups, the low-risk and high-risk groups, and the four geographically divided regions, respectively.

The Asian Continent sample. ITA and ITR were used as dependent variables. Due to collinearity problems between NTF and NTI, we constructed two analysis models for the explanatory variables. The explanatory variables in Model 1 were NTE, NTF, NTE-1, and NTF-1. In Model 2, the explanatory variables were NTE, NTI, NTE-1, and NTI-1. The analysis methods applied in each model were OLS, PDM, and 2SLS. The results of the impact of terrorism on tourism in Asia are shown in Table 1.

Table 1

Regression Results for the Asian Continent Sample (N = 1,012)

	Model 1 (ITA)			Model 2 (ITA)
	OLS	PDM	2SLS	OLS	PDM	2SLS
NTE	1717.255	1479.715	1916.894	756.6948	1148.629	759.8646
NTF	−404.2161	−370.0091	−490.0035	/	/	/
NTI	/	/	/	184.3513	−94.41589	207.9370
NTE-1	52.38283	477.2697	419.0365	663.8318	508.4433	886.7299
NTF-1	−444.2867	−148.6439	−180.7064	/	/	/
NTI-1	/	/	/	−636.2964	−189.8107	−461.3379
GDP	6.02E−06***	3.45E−06***	3.34E−06***	6.03E−06***	3.50E−06***	3.34E−06***
ITR/GDP	−3896352	12372797***	13802053***	−3921849	11390598***	13756100***
PFI	32691.86***	45689.03***	74351.69***	32024.95***	45061.18***	73190.71***
PDI	87789.89***	67926.16	63981.78	87719.00***	68244.62	65495.85
IDV	−175029***	−110291.7	−145686.1	−172595***	−111137.9	−142458.8
MAS	−101543***	71011.74	81220.28	−100645***	67778.74	80469.34
UAI	12394.22	−6283.612	12163.06	11641.10	−6110.177	11170.08
LTO	−71026.8***	8709.718	17155.04	−70528.2***	6689.231	15895.90
IND	82075.15***	63079.11	89466.69	82225.57***	64210.34	89802.58
_cons	4479472*	−6802645	−9720476	4427147*	−6520473	−9728073
Hausman test	/	random effect	random effect	/	random effect	random effect
R ²	.449153	.300967	.279929	.449365	.303936	.276089
F	59.71124	31.52928	30.67120	59.76254	31.97622	30.40846
	Model 1 (ITR)			Model 2 (ITR)
	OLS	PDM	2SLS	OLS	PDM	2SLS
NTE	5626726**	3268421**	5219036**	3903314*	2878724**	3402686*
NTF	−1399898**	−955737.5**	−1850821**	/	/	/
NTI	/	/	/	−184298.7	−360318.2	−429343.9
NTE-1	1023396	1934652	1131013	1151933	1631270	1492285
NTF-1	−858064.4	−579322.7	−145390.6	/	/	/
NTI-1	/	/	/	−859847.5**	−514358.0**	−546177.4
GDP	0.005653***	0.004080***	0.003961***	0.005672***	0.004085***	0.003962***
ITR/GDP	−1.49E+09	1.64E+10***	1.77E+10***	−1.52E+09	1.63E+10***	1.84E+10***
PFI	−6139246	54640303***	85635520***	−7614353	53437174***	86369550***
PDI	20492656	−1734712	−19836935	19850555	−2762635	−25060828
IDV	−1.30E+08***	−93696689	−1.05E+08	−1.25E+08***	−89666777	−93109575
MAS	−1.23E+08***	7024855	19178443	−1.23E+08***	7705543	19849024
UAI	2283620	11311602	24136731	457005.0	9794486	19762461
LTO	−51979160***	87710.21	9186283	−51699366***	−530421.1	5703317
IND	65589289***	60045109	78393454	65033135***	57873001	70260075
_cons	1.09E+10***	−8.33E+08	−3.00E+09	1.10E+10***	−6.03E+08	−2.30E+09
Hausman test	/	random effect	random effect	/	random effect	random effect
R ²	.479146	.372112	.349429	.474355	.367962	.344968
F	67.36674	43.39957	42.46731	66.08524	42.63368	40.81288

Note: ITA = international tourist arrivals; OLS = ordinary least square; PDM = panel data model; 2SLS = two-stage least square; NTE = number of terrorist events; NTF = number of fatalities caused by terrorism; NTI = number of injuries caused by terrorism; GDP = gross domestic products; ITR = international tourism revenue; PFI = press freedom index; PDI = power distance index; IDV = individualism versus collectivism; MAS = masculinity versus femininity; UAI = uncertainty avoidance index; LTO = long-term orientation versus short-term normative orientation; IND = indulgence versus restraint.

The symbol “/” means that there is no value for the variable in this column.

p < .1. **p < .05. ***p < .01.

The results showed that terrorism had no impact on ITA. Neither terrorism data of the current nor the previous year correlated with ITA. As control variables, GDP, ITR/GDP, PFI, PDI, and IND were significantly positively correlated with ITA, while IDV, MAS, and LTO were significantly negatively correlated with ITA. UAI had no correlation with ITA. These results mean that, countries having high levels of power distance and indulgence tend to attract more international tourists, while countries having low levels of press freedom, individualism, masculinity, and long-term orientation tend to attract more international tourists.

In terms of the impact of terrorism on ITR, the results showed that NTF and NTI-1 had significant negative impacts on ITR, while NTE had a significant positive impact on ITR. For control variables, except for PDI which showed no significant correlation with ITR, the other variables demonstrated similar correlates with ITR as their correlations with ITA.

Therefore, Hypothesis 1-1, that NTE has a negative impact on ITA, could not be verified in our analysis models; together with Hypothesis 1-2, that NTF has a negative impact on ITA; and Hypothesis 1-3, that NTI has a negative impact on ITA. Hypotheses 2-2, that NTF has a negative impact on ITR; and Hypothesis 2-3, that NTI has a negative impact on ITR, were supported. However, Hypothesis 2-1, that NTE has a negative impact on ITR, was rejected.

To explore the reason why NTE was found to have significant positive impacts on ITR, we created the following new subvariables: the NTE that caused fatalities (NTE-f, NTE-1-f), the NTE that caused injury (NTE-i, NTE-1-i), and the NTE that caused neither fatality nor injury (NTE-no, NTE-1-no). We then applied PDM to examine the impacts of these subvariables on ITR.

The results show that NTE-1-f had a significant negative impact on ITR, both NTE-no and NTE-1-i had significant positive impacts on ITR, while the other subvariables had no impacts on ITR (Supplement Table 2, available online). These results indicate that, when controlling for terrorist events which did not cause fatalities, the impact of terrorism frequency on ITR changes from positive to negative. Furthermore, the results demonstrated that the impacts of terrorism frequency which caused fatalities on ITR had a time-lag. Hence, Hypothesis 2-1 was supported.

Influence of 911 and Islam. A total of 594 samples from Muslim countries and 418 samples from non-Muslim countries were used for examining the influence of the 911 event and Islam with regard to the impacts of terrorism on tourism. As the control variables of time and religion were both dummy variables, their interaction term of “time × religion” was also a dummy variable, with “1” representing Muslim countries after 911, and “0” representing Muslim countries before 911 and non-Muslim countries from 1995 through to 2016. Thus, the following analytic equation was extended:

\begin{array}{l} T o u r i s m_{i t} = C + α T e r r o r i s m_{i t} + β E c o n o m i c_{i t} + \\ γ R e p o r t + δ C u l t u r e + θ T i m e_{i t} + ζ \\ R e l i g i o n_{i t} + η T i m e_{i t} \times R e l i g i o n_{i t} + ε \end{array}

(Equation 3)

where $θ, ζ, η$ are estimation coefficients.

We applied PDM to conduct the analysis. As shown in Table 2, for the Asian Continent sample after 911, both ITA and ITR demonstrated significant increase. However, in Muslim countries following 911, ITR significantly decreased. Therefore, the results showed that Muslim countries suffered more in terms of terrorism impacts on ITR than non-Muslim countries. In addition, those six cultural dimensions did not show any impacts on tourism.

Table 2

Influence of 911 and Islam in Terms of Terrorism Impacts on Tourism (N = 1,012)

	PDM (ITA)		PDM (ITR)
NTE	482.2644	548.9483	2357048*	2223470*
NTF	−289.7253	/	−836556.6**	/
NTI	/	−163.3408	/	−380127.6
NTE-1	693.5090	473.7847	1994336	1514450
NTF-1	−208.4008	/	−574801.1	/
NTI-1	/	−149.7650	/	−433441.8*
GDP	3.17E−06***	3.21E−06***	0.003743***	0.003741***
ITR/GDP	7020079**	6407296**	1.27E+10***	1.28E+10***
Religion	173342.4	175963.5	−4.87E+08	−5.24E+08
Time	1961385***	1950498***	2.83E+09***	2.85E+09***
Religion × Time	495617.1	527335.8	−1.15E+09**	−1.14E+09**
PFI	36231.85***	35757.87***	48255790***	47352399***
PDI	63016.49	62858.62	7976802	7126409
IDV	−108635.4	−108679.4	−98203991	−94493331
MAS	81299.02	78732.75	23273907	24357805
UAI	−14089.24	−14259.20	17383766	16219520
LTO	15229.34	13729.61	−8399489	−9154351
IND	69956.78	70341.30	50468329	47961285
_cons	−7860018	−7604039	−2.42E+09	−2.23E+09
Hausman test	random effect	random effect	random effect	random effect
R ²	.362750	.363671	.421857	.418490
F	33.76329	33.89800	43.27883	42.68488

Note: PDM = panel data model; ITA = international tourist arrivals; ITR = international tourism revenue; NTE = number of terrorist events; NTF = number of fatalities caused by terrorism; NTI = number of injuries caused by terrorism; GDP = gross domestic products; PFI = press freedom index; PDI = power distance index; IDV = individualism versus collectivism; MAS = masculinity versus femininity; UAI = uncertainty avoidance index; LTO = long-term orientation versus short-term normative orientation; IND = indulgence versus restraint.

The symbol “/” means that there is no value for the variable in this column.

p < .1. **p < .05. ***p < .01

Low-risk and high-risk groups. Table 3 shows the PDM regression results of the impacts of terrorism on tourism in low-risk and high-risk groups in terms of terrorist attacks. Here, we got almost contrary results between these two groups. Regarding terrorism impacts on tourism, significant impacts were seen in the low-risk group, while no significant impacts were seen in the high-risk group. Regarding cultural proximity impacts on tourism, MAS and LTO had significant positive impacts on tourism in the low-risk group, while such impacts became negative in the high-risk group. UAI had significant negative impacts on tourism in the low-risk group, while no impacts were seen in the high-risk group.

Table 3

Regression Results of Low-Risk and High-Risk Groups

	PDM (ITA)		PDM (ITR)
Low-risk group (n = 506)
NTE	2329.673	4124.465	11869896*	11994704**
NTF	−4172.389	/	−5517336	/
NTI	/	−3068.979*	/	−1819698
NTE-1	−6445.902	−10886.63	−447217.6	−2206588
NTF-1	−4484.188	/	−59693228	/
NTI-1	/	−2910.314*	/	−2577269*
GDP	3.11E−06***	3.07E−06***	0.003495***	0.003417***
ITR/GDP	44880197***	44766765***	2.58E+10***	2.61E+10***
Religion	1751637	1833895	1.84E+09	2.06E+09
Time	2240549***	2429943***	2.52E+09***	2.70E+09***
Religion × Time	657844.6	468979.5	−4.81E+08	−6.50E+08
PFI	24253.61*	24304.65*	25188177**	26417126**
PDI	108348.4	101194.5	18996026	21388168
IDV	96326.19	128419.9	90037226	65509767
MAS	610629.9***	626646.8***	3.91E+08***	3.85E+08***
UAI	−98889.59**	−107072.8**	−68095039*	−62084514*
LTO	152845.8**	159957.1**	1.26E+08**	1.23E+08***
IND	−35928.97	−40394.71	16398674	18848234
_cons	−40992980***	−41822338***	−2.65E+10***	−2.62E+10***
Hausman test	random effect	random effect	random effect	random effect
R ²	.503108	.514317	.643091	.644381
F	29.48932	30.84216	52.47834	52.77437
High-risk group (n = 506)
NTE	380.0723	485.5293	1001071	628453.6
NTF	−313.0541	/	−517799	/
NTI	/	−194.0133	/	−167866.8
NTE-1	779.5138	551.4136	1574501	1333116
NTF-1	−185.1659	/	−334599	/
NTI-1	/	−126.1111	/	−249234.3
GDP	8.88E−06***	8.98E−06***	0.010339***	0.010640***
ITR/GDP	3720100	3917011	1.14E+10***	1.25E+10***
Religion	1362894	1334324	−2.65E+08	−3.45E+08
Time	1280847**	1260364**	2.44E+09***	2.39E+09***
Religion × Time	448125.1	479879.7	−1.37E+09*	−1.37E+09*
PFI	48486.88***	47984.03***	58927107***	58562949***
PDI	256.0082	−462.2268	−34713444	−36080987
IDV	−219180.2	−212087.6	−2.35E+08	−2.22E+08
MAS	−407409.6***	−413808.2***	−4.47E+08***	−4.64E+08***
UAI	127106.7	123622.8	1.48E+08	1.42E+08
LTO	−235213.4***	−237610.0***	−2.77E+08***	−2.85E+08***
IND	72548.83	67336.07	26065409	13684915
_cons	22066330*	22750288*	2.88E+10***	3.05E+10**
Hausman test	random effect	random effect	random effect	random effect
R ²	.318710	.317988	.314169	.308300
F	13.62476	13.57952	13.34174	12.98138

The symbol “/” means that there is no value for the variable in this column.

p < .1. **p < .05. ***p < .01.

Furthermore, Table 3 shows that in the high-risk group, ITR significantly decreased in Muslim countries following the 911 event. When compared with the regression results of the Asian Continent sample (Table 1, where ITR showed a decreasing trend after the 911 event), it was found that the results of a decreasing trend in ITR were actually caused by the data of Muslim countries with a high risk of terrorist attacks. Based on results of Table 3, Hypothesis 1-3, that NTI has a negative impact on ITA, was supported.

Four geographical regions. This section details the results of comparative analyses of the four geographically divided regions. As the correlations between cultural distance and tourism were not consistent in the aforementioned analyses results (Tables 1 to 3), and considering that cultural distance might be very small in countries of each geographical region, we thus excluded the control variable of cultural distance from the following analyses models.

As shown in Table 4, the results demonstrate that significant impacts of terrorism on tourism can only be seen in East and Southeast Asia, and South Asia. The findings pertaining to East and Southeast Asia supported Hypothesis 2-2 (NTF is negatively correlated with ITR). The findings pertaining to South Asia supported Hypothesis 1-1 (NTE is negatively correlated with ITA), Hypothesis 2-1 (NTE is negatively correlated with ITR), and Hypothesis 2-2.

Table 4

Impacts of Terrorism on Tourism in Four Geographical Regions

	PDM (ITA)		PDM (ITR)
East and Southeast Asia (n = 308)
NTE	10940.55	4320.794	17125609*	508206.1
NTF	−11040.63	/	−15116460*	/
NTI	/	689.5688	/	8006882**
NTE-1	14069.69*	8671.093	25820013***	18317063**
NTF-1	−10239.60	/	−14144453	/
NTI-1	/	−648.4250	/	−884764.3
GDP	3.34E−06***	3.11E−06***	0.003975***	0.003683***
ITR/GDP	55565315***	59942345***	7.79E+10***	8.18E+10***
PFI	66287.70***	68303.92***	97514621***	1.01E+08***
_cons	308511.8	−35135.48	−2.94E+09	−3.63E+09*
Hausman test	random effect	fixed effect	random effect	random effect
R ²	.509868	.919934	.518597	.517057
F	42.50228	156.8336	44.01381	43.74316
South Asia (n = 154)
NTE	212.3801	127.5255	853707.8*	253701.1
NTF	61.92223	/	−405390.5	/
NTI	/	4.302714	/	−92065.17
NTE-1	−244.2645	−413.0597**	−500641.8	−817422**
NTF-1	−149.2495	/	−533362.6*	/
NTI-1	/	55.52247	/	−61136.34
GDP	3.14E−06***	3.22E−06***	0.010298***	0.010660***
ITR/GDP	1055101***	1132456***	4.56E+09***	4.55E+09***
PFI	448.6217	656.0089	5546383**	4932065*
_cons	342108.4**	294554.2**	−5.91E+08*	−6.65E+08**
Hausman test	random effect	random effect	random effect	random effect
R ²	.922527	.922139	.968674	.966332
F	236.4519	235.1751	614.0222	569.9391
West Asia (n = 418)
NTE	−265.4779	229.8077	315831.2	350391.9
NTF	−185.2221	/	−262427.5	/
NTI	/	−216.5534	/	−154387.2
NTE-1	−68.95414	128.5452	154545.2	425765.5
NTF-1	76.19292	/	55405.55	/
NTI-1	/	−47.78047	/	−74802.21
GDP	3.12E−05***	3.13E−05***	0.025018***	0.025069***
ITR/GDP	17957413***	18351505***	2.04E+10***	2.06E+10***
PFI	7212.737	7055.890	10437323	10251658
_cons	−784056.5	−782226.6	−1.15E+09***	−1.14E+09***
Hausman test	random effect	random effect	fixed effect	fixed effect
R ²	.588179	.590280	.882879	.883120
F	79.77739	80.47275	112.4700	112.7325
Central Asia (n = 132)
NTE	502.5268	210.2274	151082	63205.75
NTF	−171.1562	/	−50377.48	/
NTI	/	−19.36364	/	−5582.834
NTE-1	−155.8159	−345.0146	1888.912	−108735.6
NTF-1	49.22889	/	−1546.428	/
NTI-1	/	43.87679	/	16120.06
GDP	2.42E−05***	2.42E−05***	0.006657***	0.006661***
ITR/GDP	28144653***	28164209***	5.93E+09***	5.94E+09***
PFI	−4791.596	−4739.683	−1014825	−1014079
_cons	284488	282807.7	33393847	33709518
Hausman test	random effect	random effect	random effect	random effect
R ²	.735647	.735409	.858796	.858147
F	46.91043	46.85315	102.5240	101.9781

The symbol “/” means that there is no value for the variable in this column.

p < .1. **p < .05. ***p < .01

Implications and Limitations

Theoretical Implications

Regarding our descriptive results, we found that the tourism industry in some Asian countries has been seriously affected by terrorism, notably West Asia and South Asia (Supplement Table 1, available online). One explanation for this might be that both regions are close to the ISIS base, the terrorist group occupying territory in eastern Syria and Iraq. This group also aims to extend its boundaries to Jordan and Lebanon (Khan & Estrada, 2016), which has significantly increased political tensions and public security in the neighboring regions of West Asia and South Asia. Therefore, tourists’ safety concerns may significantly have affected tourism development in these two regions. Also, the results of PFI has a consistent positive correlation with tourism, which means countries having low level of press freedom tend to be attractive for international tourists, were in line with previous studies (Asal & Hoffman, 2016; A. M. Hoffman et al., 2013; Perpina et al., 2019). One popular explanation is that panic from international tourists can be limited in states that restrict the press’s ability to cover attacks, and terrorist attacks would disproportionately occur in such states due to low publicity (Keels & Kinney, 2019).

Regarding our regression analysis results, we found that, first, there is significant importance of loss of life in a terrorist event. Despite the results that terrorist events have had a devastating influence on the tourism industry (Hypotheses 1-1, 1-3, 2-1, 2-2, and 2-3), which are consistent with previous reports (Pennington-Gray et al., 2014; Samitas et al., 2018; Wouters & Naert, 2004), the present study makes a contribution to examine different types of terrorism frequency impact on tourism differently. That is, when controlled for NTE which did not cause fatality, the impact of terrorism frequency on tourism would always be negative (Supplement Table 2, available online). Explanations are that, terrorist acts resulting in mass destruction of life and property, followed by loss of life, may have the strongest effect on tourism demand (Pizam & Fleischer, 2002).

Our second finding is that, there exists a different characteristic of terrorism’s influence between ITA and ITR. It is clear that terrorism negatively impacts on ITR, which were easily verified and more than one of our analysis models supported such conclusion (Supplement Table 5, available online). Our results were consistent with previous studies (Buigut & Amendah, 2016; Drakos & Kutan, 2003; W. Enders & Sandler, 1991; Llorca-Vivero, 2008). However, the negative impacts of terrorism on ITA were not robust. None of our analysis models can verify Hypothesis 1-2 (NTF negatively affects ITA). These results seem inconceivable and are not in line with previous studies (Buigut & Amendah, 2016). Explanations might be that, grouping methods used in the analysis models counteract the ITA decline in some countries that were severely attacked by terrorism. That is because similarity was chosen as the primary criteria for dividing those 46 countries into different groups. For instance, counties that enjoy a fast-growing tourism market were divided into the low-risk group, non-Muslim countries, and East and Southeast Asia. In these countries, the strong developing trend in tourism market may be resilient for tourist decline due to terrorism. On the contrary, for countries that have not experienced such fast-increasing tourism market, the impacts of terrorism on ITA may be limited.

Our third finding is that the Islam has no significant correlation with tourism, and the fourth finding is that the 911 event significantly impacted ITR in Muslim countries. For the former result, reasons might be that those who visited Muslim countries may have done so for official purposes (i.e., being in transit, for conferences, or to visit family), or for religious reasons. For example, millions of pilgrims visiting Mecca or Hajj as a ritual, given that pilgrimage is one of the basic principles or five pillars of Islam. And pilgrimage itself is regarded by many scholars as a form of tourism (Aziz, 1995). Similar explanations can be applied to the later result, that most of the tourists who visited Muslim countries after 911 have done so for official purposes or religious reasons, not for tourist shopping which is an important source of revenue at the tourist destination (Albayrak et al., 2016). Furthermore, Muslim countries may face more difficulties in restoring their image of being a safe tourist destination due to the devastating fatalities caused by terrorism. Other Muslim tourist destinations, even if they have good public security, may thus be misleadingly grouped as dangerous places as well. These perceptions might, in turn, cause a vicious circle of tourism depression and moderate tourist consumption in local tourism destinations in Muslim countries.

The fifth finding is that MAS positively impacts on tourism in countries of the low-risk group. This result is contrary to previous literature, which usually states that masculinity negatively influences tourists’ perceptions of the image of a destination (Huang & Cheng, 2012; Wu et al., 2019). However, Asian countries are traditionally deemed as patriarchal societies (Bhopal, 1997), even in developed countries such as Japan and South Korea. Also, the Chinese society is dominated by patriarchal thoughts and customs (Tang, 2008). These traditional social routines might be the reason why societies having high level of masculinity seem to attract international tourists, but the true reasons are indeed high-level tourism services and facilities.

Managerial Implications

Based on these findings, we present the following recommendations for anti-terrorism policy makers and tourism crisis management. First, prevent casualties from happening. Measures could involve strengthening and periodically examining security measures (i.e., security checks, emergency plans and emergency drills) in popular tourist sites, as also supported by Becken and Hughey (2013). Second, appropriately evaluate the sincere value of reporting terrorist attacks. Measures could involve the collaboration between governmental agencies and mass media organizations, and restricting or squelching broadcasting channels operated by terrorists. Third, reduce the probability of causing loss of life in each terrorist attack. Measures could involve providing emergency medical services by establishing emergency medical rescue teams that are equipped with the necessary and advanced medical apparatus around tourist sites.

Limitations and Further Research

Limitations of this study are that we did not consider the unique features of Asian tourism resources, and attacking methods applied by terrorists. As tourism in this region is largely dependent on particular geographic characteristics in tourist destinations (such as beaches, mountains, volcanoes and rivers; Barros et al., 2011), a more detailed grouping methods of Asian countries might generate interesting findings. As the attacking methods in some extent could reveal the competitive capability of terrorist organizations, future studies would benefit from taking these aspects into account, in order to better understand the different impacts of terrorism on tourism across the continent’s different regions and countries. Furthermore, it could be valuable to undertake more country specific as opposed to region specific follow up studies on the terrorism-tourism relationship.

Conclusion

In this current study, we constructed several analysis models to investigate the impacts of terrorism on tourism, by controlling for variables of GDP, media report, cultural distance, the 911 event, and Islam. Overall, most of our hypotheses were clearly supported (Supplement Table 5, available online), meaning that our results were either basically in line with previous studies or verified existing phenomenon empirically. Accordingly, the research contributes to the tourism literature to make it clear about the differences between the impacts of the frequency and severity of terrorist events on tourism development. By comparing data on Muslim and non-Muslim countries in Asia, this study makes a contribution by revealing major safety concerns among global tourists traveling to Muslim countries. In addition, this study provides a perspective from which to consider antiterrorism policy, with the hope of helping to make this more effective in the Asian tourism sector. Our findings imply that antiterrorism policy should recommend that tourism stakeholders directly implement effective and sustainable crisis management strategies. Thus, we provide new information on the extent to which understanding the attributes of terrorism can contribute to terrorism prevention and response, and statistically demonstrate how the impacts of terrorist events on tourism can be reduced.

Supplemental Material

sj-pdf-1-jht-10.1177_1096348020986903 – Supplemental material for Impacts of Terrorist Events on Tourism Development: Evidence from Asia

Supplemental material, sj-pdf-1-jht-10.1177_1096348020986903 for Impacts of Terrorist Events on Tourism Development: Evidence from Asia by Yingying Sun and Mingzhi Luo in Journal of Hospitality & Tourism Research

Footnotes

Authors’ Contributions

Yingying Sun reviewed the literature and wrote the text. Mingzhi Luo proposed the idea and analyzed the data.

Authors’ Note:

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. This work was supported by Fundamental Research Funds for the central Universities (2018hhs-16) and Education Department of Sichuan Province (LYC18-17).

ORCID iD

Mingzhi Luo

Supplemental Material

Supplemental material for this article is available online.

Notes

Yingying Sun (e-mail: sunying@scu.edu.cn) is a lecturer at the School of Public Administration, Sichuan University, Chengdu, China. Mingzhi Luo (e-mail: luomingzhi_123@163.com) is a lecturer at the Tourism School, Sichuan University, Chengdu, China.

References

Afonso-Rodriguez

J. A.

Santana-Gallego

(2018). Is Spain benefiting from the Arab Spring? On the impact of terrorism on a tourist competitor country. Quality & Quantity, 52(3), 1371-1408. https://doi.org/10.1007/s11135-017-0527-2

Al Qudah

Bani-Mustafa

Yamen

. (2019). The mechanism of control of corruption and the rule of law mediating the effect of culture on terrorism financing. Journal of Money Laundering Control, 22(3), 498-514. https://doi.org/10.1108/jmlc-11-2018-0066

Albayrak

Caber

Çömen

(2016). Tourist shopping: The relationships among shopping attributes, shopping value, and behavioral intention. Tourism Management Perspectives, 18(April), 98-106. https://doi.org/10.1016/j.tmp.2016.01.007

Armstrong

R. W.

Mok

F. M.

Chan

(1997). The importance of cross-cultural expectations in the measurement of service quality perceptions in the hotel industry. International Journal of Hospitality Management, 16(2), 181-190. https://doi.org/10.1016/S0278-4319(97)00004-2

Arunatilake

Jayasuriya

Kelegama

(2001). The economic cost of the war in Sri Lanka. World Development, 29(9), 1483-1500. https://doi.org/10.1016/s0305-750x(01)00056-0

Asal

Deloughery

Murdie

(2016). Responding to terrorism? Human rights organization shaming and terrorist attacks. Studies in Conflict & Terrorism, 39(3), 240-259. https://doi.org/10.1080/1057610x.2015.1093887

Asal

Hoffman

A. M.

(2016). Media effects: Do terrorist organizations launch foreign attacks in response to levels of press freedom or press attention? Conflict Management and Peace Science, 33(4), 381-399. https://doi.org/10.1177/0738894215579622

Aziz

(1995). Understanding attacks on tourists in Egypt. Tourism Management, 16(2), 91-95. https://doi.org/10.1016/0261-5177(94)00016-4

Barros

C. P.

Botti

Peypoch

Robinot

Solonandrasana

, & A, G. A. (2011). Performance of French destinations: Tourism attraction perspectives. Tourism Management, 32(1), 141-146. https://doi.org/10.1016/j.tourman.2010.01.015

10.

Bassil

Saleh

A. S.

Anwar

(2019). Terrorism and tourism demand: a case study of Lebanon, Turkey and Israel. Current Issues in Tourism, 22(1), 50-70. https://doi.org/10.1080/13683500.2017.1397609

11.

Becken

Hughey

K. F. D.

(2013). Linking tourism into emergency management structures to enhance disaster risk reduction. Tourism Management, 36(June), 77-85. https://doi.org/10.1016/j.tourman.2012.11.006

12.

Bhopal

(1997). Gender, "race" and patriarchy: A study of South Asian women. Ashgate.

13.

Blomberg

S. B.

Hess

G. D.

Orphanides

(2004). The macroeconomic consequences of terrorism. Journal of Monetary Economics, 51(5), 1007-1032. https://doi.org/10.1016/j.jmoneco.2004.04.001

14.

Buigut

Amendah

D. D.

(2016). Effect of terrorism on demand for tourism in Kenya. Tourism Economics, 22(5), 928-938. https://doi.org/10.5367/te.2015.0467

15.

Chesney

Reshetar

Karaman

(2011). The impact of terrorism on financial markets: An empirical study. Journal of Banking & Finance, 35(2), 253-267. https://doi.org/10.1016/j.jbankfin.2010.07.026

16.

Cohen

(1979). A phenomenology of tourist experiences. Sociology, 13(1), 179-201. https://doi.org/10.1177/003803857901300203

17.

Conrad

Greene

(2015). Competition, differentiation, and the severity of terrorist attacks. Journal of Politics, 77(2), 546-561. https://doi.org/10.1086/680262

18.

Crenshaw

(1981). The causes of terrorism. Comparative Politics, 13(4), 379-399. https://doi.org/10.2307/421717

19.

Drakos

Kutan

A. M.

(2003). Regional effects of terrorism on tourism in three Mediterranean countries. Journal of Conflict Resolution, 47(5), 621-641. https://doi.org/10.1177/0022002703258198

20.

Enders

Sandler

(1991). Causality between transnational terrorism and tourism: The case of Spain. Terrorism, 14(1), 49-58. https://doi.org/10.1080/10576109108435856

21.

Enders

Sandler

Parise

G. F.

(1992). An econometric analysis of the impact of terrorism on tourism. Kyklos, 45(4), 531-554. https://doi.org/10.1111/j.1467-6435.1992.tb02758.x

22.

Feridun

Sezgin

(2008). Regional underdevelopment and terrorism: The case of South Eastern Turkey. Defence and Peace Economics, 19(3), 225-233. https://doi.org/10.1080/10242690801972196

23.

Global Terrorism Database. (2016). The GTD world map: Forty-five years of terrorism. http://www.start.umd.edu/gtd/

24.

Hoffman

A. M.

Shelton

Cleven

(2013). Press freedom, publicity, and the cross-national incidence of transnational terrorism. Political Research Quarterly, 66(4), 896-909. https://doi.org/10.1177/1065912913478840

25.

Hoffman

(2018). Who is afraid of ISIS? ISIS anxiety and its correlates. Stress & Health, 34(1), 84-92. https://doi.org/10.1002/smi.2764

26.

Hofstede

(2005). Culture and organizations: Software of the mind. McGraw-Hill.

27.

Hofstede

Bond

M. H.

(1984). Hofstede’s culture dimensions: An independent validation using Rokeach’s value survey. Journal of Cross-Cultural Psychology, 15(4), 417-433.

28.

Huang

M. H.

Cheng

Z. H.

(2012). The effects of inter-role conflicts on turnover intention among frontline service providers: Does gender matter? Service Industries Journal, 32(3), 367-381. https://doi.org/10.1080/02642069.2010.545391

29.

Ivianski

(1987). The terrorist revolution: Roots of modern terrorism. Journal of Strategic Studies, 10(4), 129-149. https://doi.org/10.1080/01402398708437318

30.

Keels

Kinney

(2019). “Any press is good press?” Rebel political wings, media freedom, and terrorism in civil wars. International Interactions, 45(1), 144-169. https://doi.org/10.1080/03050629.2019.1522309

31.

Khan

Estrada

M. A. R.

(2016). The effects of terrorism on economic performance: The case of Islamic State in Iraq and Syria (ISIS). Quality & Quantity, 50(4), 1645-1661. https://doi.org/10.1007/s11135-015-0226-9

32.

Lepp

Gibson

(2003). Tourist roles, perceived risk and international tourism. Annals of Tourism Research, 30(3), 606-624. https://doi.org/10.1016/s0160-7383(03)00024-0

33.

Liu

Pratt

(2017). Tourism’s vulnerability and resilience to terrorism. Tourism Management, 60(June), 404-417. https://doi.org/10.1016/j.tourman.2017.01.001

34.

Llorca-Vivero

(2008). Terrorism and international tourism: New evidence. Defence and Peace Economics, 19(2), 169-188. https://doi.org/10.1080/10242690701453917

35.

Llussa

Tavares

(2011). The economics of terrorism: A (simple) taxonomy of the literature. Defence and Peace Economics, 22(2), 105-123.

36.

Meftah

J. b. T.

(2018). Roots of sadistic terrorism crimes: Is it Islam or Arab culture? Aggression and Violent Behavior, 42(September-October), 52-60. https://doi.org/10.1016/j.avb.2018.06.003

37.

Muckley

C. B.

(2010). Terrorism, tourism and FDI: Estimating a lower bound on the peace dividend in Northern Ireland. Journal of the Statistical and Social Inquiry Society of Ireland, 40, 116-132. http://www.tara.tcd.ie/handle/2262/62346

38.

S. I.

Lee

J. A.

Soutar

G. N.

(2007). Tourists intention to visit a country: The impact of cultural distance. Tourism Management, 28(6), 1497-1506. https://doi.org/10.1016/j.tourman.2006.11.005

39.

Pennington-Gray

Schroeder

Donohoe

Cahyanto

(2014). Travelers’ perceptions of crisis preparedness certification in the United States. Journal of Travel Research, 53(3), 353-365. https://doi.org/10.1177/0047287513496470

40.

Perles-Ribes

J. F.

Ramon-Rodriguez

A. B.

Moreno-Izquierdo

Torregrosa Marti

M. T.

(2018). Winners and losers in the Arab uprisings: A Mediterranean tourism perspective. Current Issues in Tourism, 21(16), 1810-1829. https://doi.org/10.1080/13683500.2016.1225697

41.

Perpina

Camprubi

Prats

(2019). Destination image versus risk perception. Journal of Hospitality & Tourism Research, 43(1), 3-19. https://doi.org/10.1177/1096348017704497

42.

Pizam

(1999). A comprehensive approach to classifying acts of crime and violence at tourism destinations. Journal of Travel Research, 38(1), 5-12. https://doi.org/10.1177/004728759903800103

43.

Pizam

Fleischer

(2002). Severity versus frequency of acts of terrorism: Which has a larger impact on tourism demand? Journal of Travel Research, 40(3), 337-339. https://doi.org/10.1177/0047287502040003011

44.

Rink

Sharma

(2018). The determinants of religious radicalization: Evidence from Kenya. Journal of Conflict Resolution, 62(6), 1229-1261. https://doi.org/10.1177/0022002716678986

45.

Rosselló

Sansó

(2017). Taxing tourism: The effects of an accommodation tax on turism demand in the Balearic Island (Spain). Cuadernos Económicos de ICE, 93.

46.

Samitas

Asteriou

Polyzos

Kenourgios

(2018). Terrorist incidents and tourism demand: Evidence from Greece. Tourism Management Perspectives, 25(January), 23-28. https://doi.org/10.1016/j.tmp.2017.10.005

47.

Schmude

Zavareh

Schwaiger

K. M.

Karl

(2018). Micro-level assessment of regional and local disaster impacts in tourist destinations. Tourism Geographies, 20(2), 290-308. https://doi.org/10.1080/14616688.2018.1438506

48.

Sönmez

S. F.

Apostolopoulos

Tarlow

(1999). Tourism in crisis: Managing the effects of terrorism. Journal of Travel Research, 38(1), 13-18. https://doi.org/10.1177/004728759903800104

49.

Tang

(2008). Family and gender review. Social Sciences Academic Press. (In Chinese)

50.

United Nations Statistics Division. (2019). Population by religion, sex and urban/rural residence. http://data.un.org/Data.aspx?q=religion&d=POP&f=tableCode%3a28

51.

UNWTO. (2018). UNWTO Annual Report 2017. https://www.e-unwto.org/doi/book/10.18111/9789284419807

52.

Wouters

Naert

(2004). Of arrest warrants, terrorist offences and extradition deals: An appraisal of the EU’s main criminal law measures against terrorism after “ 11 September.” Common Market Law Review, 41(4), 909-935.

53.

H. C.

Chen

X. L.

Cheng

C. C.

(2019). Relationships between experiential cultural distance, experiential relationship quality and experiential future intentions: The case of Mainland Chinese tourists. Journal of Tourism and Cultural Change. Advance online publication. https://doi.org/10.1080/14766825.2019.1679160

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.61 MB