The contagion effect of COVID-19-induced uncertainty on US tourism sector: Evidence from time-varying granger causality test

Abstract

This study investigates the effect of COVID-19-induced uncertainty on the overall stock market and the stock performance of the tourism and hospitality industry and its subsectors utilizing a novel time-varying robust Granger causality test. The results show that the COVID-19 pandemic–induced uncertainty has an adverse impact on the overall economy, tourism and hospitality industry, and subsectors of tourism and hospitality. However, the impact of COVID-19 pandemic–induced uncertainty is more significant in the tourism and hospitality industry and its subsectors. In particular, hotels sector has experienced the largest impact from the COVID-19 pandemic, followed by restaurants and airline sectors, respectively. Research and practical implications are discussed.

JEL Classification: C1, C5, F2, G11, G17, G41

Keywords

Pandemic COVID-19 uncertainty tourism time-varying causality

Introduction

A rapid spike in new cases of COVID-19 outside of China has fueled concerns that the coronavirus could turn into a pandemic with more extensive and significant economic effects than initially predicted. The pandemic has led to a widespread introduction of restrictions by government authorities, including curfews, sector specific closures, travel restrictions, and other measures worldwide in an attempt to control and mitigate the spread of the coronavirus in early 2020. Although many, if not all, sectors have been adversely affected by this pandemic, tourism and hospitality industry has been one of the hardest impacted industries due to almost a cessation of international and domestic travel. The tourism and hospitality industry accounts for approximately 7.8 million jobs and 3% of the GDP in the United States (SelectUSA, 2017). However, the lack of demand for travel has led to closures in subsectors of tourism and hospitality industry, such as travel agencies, airlines, cruises, hotels, and restaurants.

Investors of tourism and hospitality related companies liquidated their equity-holdings so radically that some companies have lost 80% of their market value in a significantly short period due to the huge uncertainty related to short- and long-term potential impact of the COVID-19 pandemic on the tourism and hospitality industry (Kaczmarek et al., 2021). However, although a number of studies have investigated the effect of COVID-19 pandemic on the tourism and hospitality industry (see Ozdemir et al., 2021), the extent to which COVID-19 pandemic–induced uncertainty affected the tourism and hospitality industry in general and its subsectors, in particular, is not clear.

A vast majority of empirical papers have analyzed the effects of various sources of uncertainty on tourism and hospitality industry, such as uncertainty surrounding the presidential elections (Das et al., 2020; Demiralay, 2020), uncertainties related to economic policy (Demir and Gozgor, 2018; Balli et al., 2018), worldwide economic uncertainty (Karabulut et al., 2020), and geopolitical risks (Tiwari et al., 2019). In a recent study, Madanoglu and Ozdemir (2018) showed that economic policy uncertainty leads to deterioration of hotel operating performance in the United States. In the context of the COVID-19 pandemic, Sharma and Nicolau (2020) investigated the stock market performance of tourism and hospitality industry specific Dow Jones industry indices using the daily COVID-19 infection and fatality data with a sample ending in April 2020. Kaczmarek et al. (2021) examined how the effect of the US government response to COVID-19 on stock performance tourism and hospitality companies for period between January and April 2020.

While these former studies showed the negative effects of COVID-19 pandemic on tourism and hospitality industry, the study periods do not go beyond April 2020, the analyses are limited to sector specific effects, and the empirical models do not account for essential factors, such as the measure of uncertainty. Therefore, the implications of the extant studies are naturally limited to specified time period and that is likely to be inconclusive due to exclusion of essential uncertainty measures that contributes to the explanation of the effect of COVID-19 pandemic on tourism and hospitality industry.

Accordingly, the purpose of this study is to investigate the effect of COVID-19-induced uncertainty on the overall economy and on the tourism and hospitality industry. In particular, we analyze the extent to which COVID-19-induced uncertainty affects the overall stock market and the stock performance of tourism and hospitality industry and its subsectors, including airline, hotel, and restaurant and bars sectors utilizing a novel time-varying robust Granger causality (TVR-GC) test for the period between January 2019 and October 2021. To pursue this research objective, our study seeks to answer the following research questions: To what extent does the COVID-19 pandemic–induced uncertainty affect the overall economy? To what extent does the COVID-19 pandemic–induced uncertainty affect the tourism and hospitality industry? To what extent does the effect of COVID-19 pandemic–induced uncertainty on overall economy and tourism and hospitality industry diverge? To what extent does the COVID-19 pandemic–induced uncertainty affect the subsectors of tourism and hospitality industry, such as hotel, restaurant, and airline sectors?

In so doing, this study attempts to contribute to the extant literature on the relationship between COVID-19 pandemic and tourism and hospitality industry. We utilize a recently constructed COVID-19 pandemic–induced uncertainty index of Baker et al. (2020) that is developed based on the daily counts of newspaper articles containing keywords such as virus, epidemic, disease, pandemic, flu, coronavirus, and so on. That is, we utilize the news-based COVID-19 uncertainty index measure, which allows to eliminate the erroneous inferences arising from inconsistent COVID-19-related data reports. Second, the sample of this study covers an extended time period (January 2019–October 2021). This time frame allows to capture the effect of the second wave of COVID-19 infections on over, all economy and tourism and hospitality industry, and eliminate the biased parameter estimates encountered in small-samples. Furthermore, we add a new dimension to the extant literature by comparing the effect of COVID-19 pandemic–induced uncertainty on overall economy, tourism and hospitality industry and subsectors tourism and hospitality industry, including airline, hotel, and restaurant and bar sectors. The findings are expected to provide valuable information that helps forecast the future stock returns and volatility of the tourism subsectors including airline, hotel, and restaurant and bar sectors.

Literature Review

Tourism is the world’s second largest industry constituting 10.3% of world’s GDP (WTTC, 2020). This giant sector is extremely fragile and sensitive to sudden shocks in the economy. The last two decades have shown that any type of domestic and global shocks may drastically impact the tourism industry in the short-run and distort the regular functioning of supply and demand mechanism. Ranging from natural disasters to human-made crises, these events urge an immediate fear, stimulate cautious behavior and re-shape the consumption patterns of consumers. Given that, tourism firms with their easily given-up products under uncertainty are prone to numerous challenges. On the one hand, there are the operational challenges to be able to continue operations and maintain a steady demand while being subject to state and federal restrictions. On the other hand, there is the market reaction to tourism stocks fueled by the investor sentiment, which stemmed from the potentially sustained uncertainty that may destroy firm value over a long-horizon. There is substantial evidence in the extant tourism literature that highlights the sectoral, operational, and market outcomes of various shocks in the tourism sector. A major source of instability in the tourism sector is the terror attacks and threats. For instance, 9/11 terror attacks led to a disruptive decline in the travel and tourism demand due to fear of recursive attacks (Arana and Leon, 2008; Kosova and Enz, 2012). Indeed, the operational statistics reveal the severity of the attacks on the US hotel industry. In the first month following the attacks, hotel bookings declined dramatically (Goodrich, 2001). Kosova and Enz (2012) reported that top-end luxury market had seen about 7.2% decline in occupancy, whereas the economy market only experienced 1.4% decrease due to 9/11 terror attacks. Similar effects have also been reported for ADR (average daily rate) and RevPAR (revenue per available room).

Financial crises are another source of uncertainty for tourism sector and the consequences of financial crises may be dire for firms in this sector. The most recent financial crisis, 2007/09, led to significant drop in international tourism demand. United Nations World Tourism Organization (UNWTO, 2011) reported that negative effects of the financial crisis lasted 15 consecutive months and peaked in March 2009 with a 12% decline in international tourist arrivals. During 2009, international tourist arrivals was down 4% globally with a 6% decline in tourism receipts (UNWTO, 2010). US hotel industry suffered significantly and experienced sizeable declines in ADR and RevPAR due to the adversities of the financial crisis (Kosova and Enz, 2012).

Spread of infectious diseases is another source of instability for tourism demand. Previous pandemics have clearly demonstrated that world tourism activity is severely affected by infectious diseases. For instance, severe acute respiratory syndrome (SARS) in 2003 mostly affected mainland China, Hong Kong, Taiwan, and Canada (Gossling et al., 2020) and resulted in nearly 10% decline in the hotel occupancy of the affected countries (Oaten et al., 2015). In May 2003, occupancy rate was as low as 18% in China (Hospitalitynet, 2020). Likewise, H1N1 pandemic, swine flu as most commonly known, caused a significant decline in the tourism demand for Mexico, where the disease first appeared, and resulted in $2.8 billion loss in tourism receipts for the country (Russy and Smith, 2013).

The negative effect of crises on tourism sector is quite evident and highlights the fragile nature of the sector. The sector is very sensitive to external shocks and gets hit in the most severe way at times of economic, political, and social turbulences. These effects are immediately apparent in the form of weakening demand and declining revenue. Beyond such effects, crises and resulting uncertainties also have financial market implications for publicly traded tourism companies. Increasing risk perception in the financial markets due to sustained economic slowdown in tourism firms’ economic activity and earning capacity leads to an expected market reaction. Although the literature that particularly scrutinizes the financial market reaction to crises in the tourism sector is scant, there is some notable evidence that stock values of tourism companies are affected from such non-macro effects. Zopiatis et al. (2019) have studied the response of five large hospitality and tourism stock indices from different regions of the world (i.e., FTSE Travel and Leisure World, FTSE Travel and Leisure Asia Pacific, FTSE Travel and Leisure Australia, FTSE Travel and Leisure America, and FTSE Travel and Leisure Europe) to non-macro effects such as natural disasters, terrorism, and war conflicts. Their analyses reveal that such shocks negatively affect the hospitality and tourism stock returns, particularly on the days following the event day, and the volatility of these returns increases considerably in the aftermath of the events.

From a firm-level perspective, Chen et al. (2005) studied returns of listed Taiwanese hotel companies following major non-macro events (i.e., presidential elections in Taiwan, 1999 earthquake, SARS pandemic, 2000 Sydney Olympics, 2002 Japan/Korea World Cup tournament, Asian economic crisis of 1997/1 998, the Iraqi War in 2003, and the 9/11 terrorist attacks in the United States) and found that all non-macro events, except presidential elections, had a negative effect on hotel returns with the largest effect −25.9% being reported by the SARS outbreak. In a follow-up study conducted in Singapore, Chiang et al., (2014) reported similar results to that of Chen et al. (2005), that is, 9/11 terrorist attacks, Iraqi War, and the SARS pandemic had a negative effect on the hotel stock returns.

In late 2019, a new type of coronavirus was identified in Wuhan region of China. This highly infectious coronavirus (COVID-19) quickly made its way to other countries. On 11 March 2020, World Health Organization (WHO) termed the novel coronavirus, COVID-19, as a global pandemic (WHO, 2020). As of 24 January 2021, COVID-19 cases have exceeded 99.6 million globally and over 2.1 million people have lost their lives (Worldometers, 2020). The United States has been one of the hardest hit countries so far with over 25 million cases and 428,770 COVID-19-related deaths. Government-ordered lockdowns, stay-at-home orders, self- or mandatory-quarantine, and travel restrictions have halted national economy and slowed down global travel and tourism activity (Ozdemir et al., 2022). The impact of the COVID-19 pandemic on the global tourism sector has been unprecedented so far with airplanes on ground, hotels closed fully or partially, and restaurants operating with capacity restrictions. The US tourism sector was hit particularly strongly by the pandemic. The impact of COVID-19 on the travel and tourism sector has been nine times that of 9/11 (AHLA, 2020). As of September 2020, international tourist arrivals and tourism receipts of the United States were down by 72% and 58% year-over-year, respectively (UNWTO, 2020).

The US travel economy has accumulated over $528 billion losses in travel spending between the beginning of March and second week of December 2020 (U.S. Travel Association, 2020). All industries of the tourism sector have suffered significant losses. US hotel industry has also experienced a significantly adverse impact with annual room occupancy falling to 44% in 2020, which corresponds to a decline of 458 million rooms occupied compared to 2019 (AHLA, 2020). In parallel to decreases in occupancy, annual room revenue fell by nearly 50% across the United States settling $84.6 billion (AHLA, 2020). The pandemic also had a dramatic effect in the hotel employment with job losses reaching to 670,000 (AHLA, 2020). Restaurant industry has also been affected dramatically from the COVID-19 restrictions. According to National Restaurant Association (NRA), restaurant sales in the United States was down by 47% during the first 3 weeks of March (NRA, 2020a), which was a period of immense fear and uncertainty due to the pandemic. NRA (2020b) also reported that overall national restaurant revenues were down $240 billion from expected levels in 2020. Similarly, restaurants, bars, and other food and beverage establishments have also laid of millions of employees due to partial and full shutdowns in the early months of the pandemic. The Bureau of Labor Statistics (NRA, 2020c) statistics show that 633,200 and 5,442,800 restaurant employees were laid off in March and April, respectively.

The US tourism and hospitality industry has been suffering significant economic losses since the surge of the pandemic in early 2020. The operational and financial losses are apparent and hurting companies in all aspects. Moreover, the future earnings potential of the tourism and hospitality companies is questionable, which adds to the financial challenges of the operators and make the investment volatile in the sector. Thus, this study sets out to examine the financial markets implications of COVID-19 for tourism and hospitality firms using Baker et al.’s (2020) infectious disease equity market volatility tracker (EMVID) as the measure of uncertainty.

Sample and Data

The sample period for causality analysis is chosen from 2 January 2019 through 15 October 2021, including the coronavirus outbreaks, the March period when the COVID-19 pandemic began spreading worldwide, and the new mutation of COVID-19 called as Delta variant.¹ All series are downloaded from the Bloomberg terminal and made stationary via log-level or log-difference techniques where relevant.

The data includes the price of overall stock market indices, including Dow Jones Travel & Leisure index (hereinafter tourism and hospitality industry), S&P 500 index, and Dow Jones industrial average index. Tourism and hospitality industry index is calculated based on the market capitalization-weighted index of the publicly traded 43 tourism companies in US operating in different tourism-related sectors. Hence, this index is a significant proxy for how the tourism and hospitality industry is performing. We further included the stock market index of tourism and hospitality subsectors, including Dow Jones US Restaurants & Bars Index, Dow Jones US Airlines Index, and Dow Jones US Hotels Index. Daily returns are computed as: $r_{t, j} = (ln P_{t} - ln P_{t - 1})$ where P_t is the daily closing value of stock price of the corresponding tourism sub-sector j. We define the volatility of the stock price index as the absolute return: V_tj = | ln P_t − ln P_t−1|.

To measure uncertainty related to infectious disease outbreaks, we use the daily Infectious Disease Equity Market Volatility Tracker (EMVID) constructed by Baker et al. (2020). The EMVID index counts the number of articles including infectious disease related keywords such as pandemic, virus, flu, disease, coronavirus and so on (the complete list of keywords can be accessible from: https://www.policyuncertainty.com/infectious_EMV.html by searching approximately 3000 US Newspapers. Then, EMVID index is normalized by the total numbers of articles for a given day. Since the COVID-19 pandemic has resulted in significant disruptions in global macroeconomic conditions, we also collect a set of control variables, including CBOE Volatility Index (VIX), economic policy uncertainty index (EPU) of Baker et al. (2016), and Aruoba-Diebold-Scotti Business Conditions Index (ADS) of Aruoba et al. (2009). The VIX index is a well-known measurement for a proxy of market expectations of near-term volatility in the stock market. To construct the EPU index, Baker et al. (2016) counts a string of words related to economic and policy uncertainty by searching news in 10 large US newspapers. The ADS index is a high-frequency measure of economic activity based on macroeconomic indicators driving the economic growth. The detailed list of indicators is available in Aruoba et al. (2009). We present the descriptive statistics of the sample variables in Table 1.

Table 1.

Descriptive statistics.

	Mean	Median	Max	Min	Std. Dev.	Skewness	Kurtosis	Observ.
DJ US Airlines (% return)	−0.03	0.00	18.80	−22.71	3.14	−0.44	14.23	727
DJ US Hotels (% return)	0.07	0.03	14.13	−15.81	2.35	−0.25	11.77	727
DJ US Restaurants & Bars (% return)	0.06	0.11	15.57	−17.04	1.64	−0.52	35.40	727
DJ Travel & Leisure (% return)	0.04	0.05	14.32	−15.48	1.90	−0.91	20.26	727
DJ industrial average (% return)	0.06	0.07	10.76	−13.84	1.50	−1.16	24.49	727
S&P 500 (% return)	0.08	0.10	8.97	−12.77	1.43	−1.15	21.94	727
COVID-induced uncertainty (level)	10.97	8.39	68.37	0.00	12.11	1.44	5.49	727
USEPU (log-level)	4.97	4.89	6.69	2.99	0.65	0.24	2.63	727
VIX (log-level)	3.00	2.93	4.42	2.45	0.36	1.01	4.22	727
ADS (level)	−0.51	0.02	9.08	−26.65	5.42	−3.07	14.67	727

Methodology

To investigate the influence of daily uncertainty surrounding infectious disease (EMVID) on stock price return and volatility of the tourism and hospitality industry and its subsectors, the time-varying robust Granger causality method (TVR-GC) of Rossi and Wang (2019) is implemented. The main advantage of TVR-GC method is that it is more efficient than the conventional Granger causality test in the presence of instabilities and it can also be used to identify the periods when Granger causality occurs or breaks down in the data.² Furthermore, in our particular case, which includes the periods of COVID-19 resulting in a destabilizing effect on tourism and hospitality industry, the time-varying robust Granger causality method allows the investigation of time-varying causal relationships over time and hence provides a more appropriate estimation of the relationship than a constant parameter Granger causality method.

In particular, we consider a VAR model with time-varying parameters as follows

y_{t} = Φ_{1, t} y_{t - 1} + Φ_{2, t} y_{t - 2} + \dots + Φ_{p, t} y_{t - p} + ε_{t}

(1)

where

y_{t} = {[y_{1, t}, y_{2, t}, \dots, y_{n, t}]}^{'}

is an n×1 vector, Φ_jt, j = 1, …, p are functions of time-varying coefficient matrices, and ϵ_t denotes idiosyncratic shocks which are assumed to be heteroscedastic and serially correlated. In our case, endogenous variables vector y_t in the VAR model includes alternatively returns or volatility of the stock price for a given tourism sectors including airlines, hotels, and restaurants and bars, and stock price of overall markets including S&P 500 index, Dow Jones industrial average and tourism and hospitality industry. We also employ the EMVID, EPU, VIX, and ADS as control variables to account for overall economic and policy uncertainties in the model.

Besides, we implement a direct multi-step VAR-LP forecasting model with time-varying parameters to investigate the out-of-sample forecasting ability of EMVID index.³ By iterating equation (1), y_t+h can be projected onto the linear space generated by ${(y_{t - 1}, y_{t - 2}, \dots, y_{t - p})}^{'}$ using the following equation

y_{t + h} = Φ_{1, t} y_{t - 1} + Φ_{2, t} y_{t - 2} + \dots + Φ_{p, t} y_{t - p} + ε_{t + h}

(2)

Given that θ_t is appropriate subset of $(Φ_{1, t}, Φ_{2, t}, \dots, Φ_{p, t})$ , we test the null hypothesis that the lags of EMVID index does not Granger cause the return or volatility of the stock prices for a given tourism sectors and overall stock markets

H_{0} : θ_{t} = 0, \forall t = 1,2 \dots T

(3)

Rossi (2005) suggests several test statistics by taking into account the possibility of parameter instabilities. More generally, considering that β_t may change at some unknown time point, let ${\hat{β}}_{1 τ}$ and ${\hat{β}}_{2 τ}$ represent the coefficient estimates after and before the possible structural break where τ: β_t = β₁ × 1 (t ≤ τ) + β₂ × 1 (t > τ). In this regard, the robust test can be constructed on two components ( ${\hat{β}}_{1 τ} - {\hat{β}}_{2 τ}$ and $(τ / T) {\hat{β}}_{1 τ} + {1 - (τ / T)} {\hat{β}}_{2 τ}$ ) to test the null hypothesis of no Granger causality for each time period (i.e., H₀: β_t = β = 0). While a test on whether the first component is zero captures cases where the parameter changes, a test on the second component captures the cases where the parameter β_t is constant and different from zero. In this way, it is possible to detect cases where the explanatory variable Granger causes the dependent variable so that the mean of the estimates is zero but the parameter changes. In the appendix, we provide additional technical details to further motivate the using of time-varying robust Granger causality test. In doing so, Rossi (2005) reports mean Wald (MeanW) test statistics.⁴ The lag length of the VAR model is selected one based on the Schwarz Information Criterion (SIC). Moreover, we choose the standard trimming parameter 0.10 in an effort to cover as much of data. Based on the extant structural break literature, the potential break dates are usually trimmed to omit the beginning and end of the sample period.

Results

Table 2 shows the results obtained from the novel time-varying robust Granger causality (TVR-GC) test and constant parameter Granger causality test that examine the effect of COVID-19 pandemic–induced uncertainty on the overall economy, tourism and hospitality industry and subsectors of tourism and hospitality industry in terms of stock returns. The greater the test statistics reported in Table 2 implies a greater significant effect of COVID-19 pandemic–induced uncertainty. The results based on the standard constant parameter Granger causality test

(χ_{q}^{2})

show that COVID-19 pandemic–induced uncertainty does not have an effect on the overall economy as measured by S&P 500 index and Dow Jones industrial index. However, utilizing a novel time-varying robust Granger causality test (MeanW), we observe the adverse effect of COVID-19 pandemic–induced uncertainty not only on the overall economy but also on the tourism and hospitality industry and its subsectors controlling for macroeconomic conditions and economic policy uncertainty. Similarly, the null hypothesis of no Granger causality between COVID-19 and hospitality and its subsectors (Airlines and Hotels) is rejected as 5% significance level using the constant and time-varying parameter Granger causality tests. However, while COVID-19-induced uncertainty does not Granger cause restaurants and bars sector based on traditional Granger causality test, the null hypothesis is strongly rejected in case of time-varying parameter Granger causality test.

Table 2.

Time-varying parameter Granger causality tests—stock price returns.

Test statistics
	$χ_{q}^{2}$	MeanW
S&P 500	1.33	72.47***
	(0.24)	(0.00)
Dow Jones Industrial	1.41	61.64***
	(0.23)	(0.00)
Tourism and hospitality	8.42***	259.54***
	(0.00)	(0.00)
Airlines	1.56**	86.55***
	(0.21)	(0.00)
Hotels	9.51***	177.35***
	(0.00)	(0.00)
Restaurants and bars	3.21*	110.08***
	(0.07)	(0.00)

Note: In the first column, $χ_{q}^{2}$ shows the chi-square statistic of constant parameter Granger causality test where the lag length q is selected based on SIC. In the second column, mean Wald statistics is reported based on time-varying robust Granger causality test of Rossi and Wang (2019). The corresponding p-values are given in parenthesis. The null hypothesis is that COVID-induced uncertainty index does not Granger cause the return or volatility of the stock prices for a given tourism sectors. ***, **, and * denote 1%, 5%, and 10% statistical significance levels, respectively.

This outcome suggests that the novel time-varying robust Granger causality (TVR-GC) test is more informative compared to the standard constant parameter Granger causality test content of the EMVID index regarding the in-sample prediction of stock market returns of the tourism and hospitality industry. These findings are to be expected because the underlying stationarity and homoscedastic idiosyncratic shocks assumption of constant parameter Granger causality test may lead to flawed inferences since it is unlikely that these assumptions continue to hold in the presence of instabilities. However, TVR-GC is specially designed to capture the instabilities (structural changes) in the parameters arising from the shifts in sign and magnitude across time, which are highly expected to occur in parameters after the onset of the COVID-19 pandemic. Therefore, the TVP-GC method provides a valid inference by allowing time-varying parameters and heteroskedastic idiosyncratic shocks.

The effect of COVID-19 pandemic–induced uncertainty is more significant for the tourism and hospitality industry compared with the overall economy (test statistics are 259.5 vs 61.6). These results suggest that the COVID-19 pandemic–induced uncertainty has more significant impact on the tourism and hospitality industry compared to that of the overall economy. A closer look at the second column of Table 2 shows a consensus among findings confirming that the effect of COVID-19 pandemic–induced uncertainty on the tourism and hospitality industry and its subsectors when the TVR-GC test is employed. However, the degree of impact of the COVID-19 pandemic–induced uncertainty is not uniform across subsectors of the tourism and hospitality industry. The results show that hotels sector is the most significantly affected by the COVID-19 pandemic–induced uncertainty (MeanW=177.3), which is followed by restaurants and bars sector (MeanW = 110.1) and airlines industry (MeanW = 86.6), respectively. Although the MeanW statistics are significant in S&P 500 index (MeanW = 72.5) and Dow Jones Industrial index (MeanW=61.6), which represent the overall economy, the value of MeanW statistics is lower than those of the tourism and hospitality industry and its subsectors, indicating that the tourism and hospitality industry and its subsector is more significantly affected by the COVID-19-induced uncertainty compared with the overall economy. In Tables A6–A12 of the appendix, we report the coefficient estimates of VAR models. As expected, the estimated coefficients of COVID-19 uncertainty are negative for all sectors. However, the coefficients are statistically significant and higher in magnitude for tourism and hospitality and hotels sectors.

Table 3 shows the results obtained from the novel time-varying robust Granger causality (TVR-GC) and constant parameter Granger causality tests that examine the effect of COVID-19 pandemic–induced uncertainty on the overall economy, tourism and hospitality industry and subsectors of tourism and hospitality industry in terms of stock price volatility. The null hypothesis of no Granger causality between COVID-19 and the corresponding sectors is rejected as 10% significance level using the constant and time-varying parameter Granger causality tests except for the airlines sector. Put differently, the Granger causality tests results show different outcomes in case of airlines sector, indicating that, similar to the initial analysis, the constant parameter Granger causality test may yield inconsistent and unreliable estimates. This outcome again highlights the importance of using the time-varying parameter Granger causality test in the presence of instabilities, implying that the TVR-GC test is a more robust test than the classical Granger causality approach and provides correct inferences about causality. Therefore, we rely on the results generated utilizing the novel time-varying robust Granger causality (TVR-GC) test, which appear in column 2 of Table 3.

Table 3.

Time-varying parameter Granger causality tests—stock price volatility.

Test statistics
	$χ_{q}^{2}$	MeanW
S&P 500	3.51*	238.91***
	(0.06)	(0.00)
Dow Jones Industrial	5.45**	254.41***
	(0.02)	(0.00)
Tourism and hospitality	6.34**	134.29***
	(0.02)	(0.00)
Airlines	1.45	114.34***
	(0.23)	(0.00)
Hotels	3.36*	111.71***
	(0.07)	(0.00)
Restaurants and bars	14.91***	194.14***
	(0.00)	(0.00))

Accordingly, the results show that the COVID-19 pandemic–induced uncertainty has adverse impact on the overall economy, tourism and hospitality industry and subsectors of tourism and hospitality industry in terms of stock price volatility. The findings suggest that the effect of COVID-19 pandemic–induced uncertainty is again significant for the tourism and hospitality industry, but its magnitude is lower compared with the overall economy (134.2 vs 254.4). On the other hand, if we replicate the results for after the COVID-19 outbreak period, the effect is more significant for the tourism and hospitality industry than the overall economy.⁵ The reason might be that tourism and hospitality industry has been one of the hardest impacted industry due to almost a cessation of international, domestic travel and tourism-related activities due to outbreak of COVID-19. Although the effect of COVID-19 pandemic–induced uncertainty on the tourism and hospitality industry is substantially evident, the degree of impact of the COVID-19 pandemic–induced uncertainty is not uniform across subsectors of the tourism and hospitality industry. The results show that restaurants and bars sector is the most significantly affected by the COVID-19 pandemic–induced uncertainty (MeanW = 194.1), which is followed by airlines sector (MeanW = 114.3) and hotels sector (MeanW = 111.7), respectively. These findings suggest that the tourism and hospitality industry and its subsector is significantly affected by the COVID-19–induced uncertainty in line with the overall economy in terms of stock price volatility. As a robustness check, we estimate the volatilities using GARCH(1,1) instead of taking the absolute values of the daily stock returns and re-estimate the models using the GARCH based volatility measure. The results reported in the Table A7 of the appendix show that our results are robust to a different measure of stock return volatility.

In addition to the findings from the analyses reported in Tables 2 and 3, we further present a detailed picture of timeline of the effects of COVID-19 pandemic–induced uncertainty on the overall economy, tourism and hospitality industry and its subsectors in Figures 1 and 2. Put differently, these figures show the whole sequence of the impact of the COVID-19 pandemic–induced uncertainty on the overall economy, tourism and hospitality industry and its subsectors over the study period. The red-dotted line represents the critical values corresponding to 95% confidence level. The Wald test statistic is above the red-dotted line, suggesting that return and volatility in overall stock market and tourism and hospitality industry is induced by COVID-19 pandemic uncertainty. These results further confirm the effects of COVID-19 pandemic–induced uncertainty on the overall economy, tourism and hospitality industry and its subsectors. That is, the COVID-19 pandemic–induced uncertainty leads to changes in returns (see Figure 1) and volatility (see Figure 2) of the overall stock market and tourism and hospitality industry. In particular, the significance of time-varying effect is evident in March 2020 for the entire economy. That is, all sectors of the economy appear to have been impacted by the COVID-19 pandemic–induced uncertainty in March 2020. Furthermore, concerns around the impact of the more transmissible delta variant of COVID-19 lead to retard in the intensity of re-opening in the United States, and these, in turn, have caused sell-off in the sectors that are exposed to COVID-19 risks. As shown in Figure 1, the effect is more pronounced in the tourism and hospitality sector.

Figure 1.

Time-varying Wald test statistics—stock price returns. Notes: The blue line depicts the time-varying Wald statistics with VAR(1) under SIC, testing whether EMVID Granger causes the stock price returns for a given tourism subsectors. The dotted red line shows the critical test statistics.

Figure 2.

Time-varying Wald test statistics—stock price volatility. Notes: The blue line depicts the time-varying Wald statistics with VAR(1) under SIC, testing whether EMVID Granger causes the stock price volatility for a given tourism subsectors. The dotted red line shows the critical test statistics.

Certainly, the effect of the COVID-19 pandemic–induced uncertainty on tourism and hospitality is not uniform across its subsectors. A closer look at the Figures 1 and 2 reveals that the effect of COVID-19 pandemic–induced uncertainty on hotel sector is more significant in magnitude followed by the airline sector, respectively, compared to that of overall economy (Dow Jones index). While people still had to buy food through online ordering from restaurants, the tourism and hospitality and airline sectors were forced to shut down either through government regulations and restrictions or simply because people postponed their travel to avoid crowds during the pandemic.

Although the effect of COVID-19 pandemic–induced uncertainty on the overall economy, as measured by S&P 500 index and Dow Jones industrial index, decreases substantially after March 2020 and approaches the insignificant level threshold, the COVID-19 pandemic–induced uncertainty continues to significantly affect the tourism and hospitality industry and its subsectors throughout 2020. Furthermore, we observe a slight spike in Wald test statistics in Figure 1 around September 2020 that was led by the COVID-19 pandemic–induced uncertainty. The timing of these surges coincides with the second wave of the COVID-19 pandemic. However, the observed impact of the COVID-19 pandemic–induced uncertainty on tourism and hospitality industry and its subsectors appears to be more significant. This outcome shows that the COVID-19 pandemic–induced uncertainty is perceived to be more significant effect for tourism and hospitality industry than that of the overall economy.

Furthermore, our estimations also cover robust Granger causality tests based on direct multi-step VAR-LP (VAR – Local projections) forecasting over short (1-day), medium (1-week), and long (1-month) horizons to explain whether the COVID-19 pandemic–induced uncertainty has an informative value in predicting the future stock returns and volatility of the overall economy and tourism and hospitality industry and its subsectors. The motivation of using the VAR-LP model is that it allows heteroskedastic and serially correlated idiosyncratic shocks in the VAR model to capture unstable predictive ability during the highly volatile episodes (Jordà, 2005). Empirical results reported in Table 4 confirm that the COVID-19 pandemic–induced uncertainty has an informative value in predicting the future stock returns for all prediction horizon (h=1, 5, 22). The COVID-19 pandemic–induced uncertainty has a stronger informative value in predicting the future stock returns of the hotels sector for the short horizon (MeanW=283.3). The informative value of the COVID-19 pandemic–induced uncertainty in predicting the future stock returns in the mid- and long-term horizons is the strongest for the tourism and hospitality and restaurant and bars sectors, respectively (MeanW=102.4 for mid-term and MeanW=51.4 for long term).

Table 4.

Robust Granger causality tests in the direct multi-step VAR-LP forecasting model—stock price returns.

MeanW test statistics
	1-day	1-week	1-month
S&P 500	159.1***	68.3***	17.8***
	(0.00)	(0.00)	(0.00)
Dow Jones Industrial	173.3***	85.6***	23.0***
	(0.00)	(0.00)	(0.00)
Tourism and hospitality	192.2***	102.4***	26.6***
	(0.00)	(0.00)	(0.00)
Airlines	175.8***	67.4***	44.8***
	(0.00)	(0.00)	(0.00)
Hotels	283.3***	79.6***	40.1***
	(0.00)	(0.00)	(0.00)
Restaurants and bars	105.7***	64.7***	51.4***
	(0.00)	(0.00)	(0.00)

Note: Entries correspond to mean Wald statistics based on time-varying robust Granger causality test of Rossi and Wang (2019) using the VAR-LP forecasting model. The corresponding p-values are given in parenthesis. We assume heteroskedastic and serially correlated idiosyncratic shocks. ***, **, and * denote 1%, 5%, and 10% statistical significance levels, respectively.

Table 5 shows that the COVID-19 pandemic–induced uncertainty has an informative value in predicting the future stock price volatility for all prediction horizon (h=1, 5, 22). The COVID-19 pandemic–induced uncertainty has a stronger informative value in predicting the future stock returns of the restaurants and bars sector in the short term (MeanW=163.4). The informative value of the COVID-19 pandemic–induced uncertainty in predicting the future stock returns in the mid- and long-term horizons is the strongest for the hotel sector (MeanW=320.5 for mid-term and MeanW=228.1 for long-term).

Table 5.

Robust Granger causality tests in the direct multi-step VAR-LP forecasting model—stock price volatility.

MeanW test statistics
	1-day	1-week	1-month
S&P 500	126.4***	181.8***	213.2***
	(0.00)	(0.00)	(0.00)
Dow Jones Industrial	150.9***	207.5***	140.0***
	(0.00)	(0.00)	(0.00)
Tourism and hospitality	100.2***	146.8***	182.1***
	(0.00)	(0.00)	(0.00)
Airlines	148.4***	171.4***	131.1***
	(0.00)	(0.00)	(0.00)
Hotels	101.5***	320.5***	228.1***
	(0.00)	(0.00)	(0.00)
Restaurants and bars	163.4***	153.8***	213.7***
	(0.00)	(0.00)	(0.00)

Overall, the findings from the Granger causality tests based on direct multi-step VAR-LP forecasting over short (1-day), medium (1-week), and long (1-month) horizons suggest that the COVID-19 pandemic–induced uncertainty has an informative value in predicting the future stock returns and volatility of the overall economy and tourism and hospitality industry and its subsectors. Accordingly, the EMVID index that captures the COVID-19 pandemic–induced uncertainty is an effective factor that allows predicting the stock market returns and volatility of the overall economy, tourism and hospitality industry and its subsectors.

Discussion and Conclusion

The unprecedented nature of the COVID-19 pandemic and difficulty to predict its potential impact on society and economies around the world has led authorities to impose strict restrictions, such as travel restrictions, stay-at-home orders, curfews, mandatory closures, and so on. While these restrictions were implemented for the sake of public health, the economic impact of these restrictions has been severe. Although many sectors of the economy have experienced adverse effects due to the COVID-19 pandemic and restrictions associated with this pandemic, tourism and hospitality industry has been one of the most affected industry, as domestic and international travel was substantially restricted, restaurants and bars were mostly required to shut down their operations, and hotels were not able to sustain their operations with little or no travel demand.

In this study, we examined the effect of COVID-19 pandemic–induced uncertainty on the overall economy and on the tourism and hospitality industry. The results show that the COVID-19 pandemic–induced uncertainty has significant adverse impact on the overall economy, tourism and hospitality industry and subsectors of tourism and hospitality industry in terms of both stock returns and volatility. However, the impact of COVID-19 pandemic–induced uncertainty is more significant in tourism and hospitality industry and its subsectors both in terms of stock returns. A substantial negative global demand shock from the proliferation of COVID-19 pandemic globally appears to have resulted in a significantly and substantially adverse impact to tourism and hospitality industry compared to overall economy in general.

The results from the analyses present further evidence that the effect of COVID-19 pandemic–induced uncertainty is not uniform across subsectors of tourism and hospitality industry in the United States. In particular, the US hotel sector has experienced the largest significant impact from the COVID-19 pandemic, followed by restaurants and bars and airline sectors, respectively in the United States. Certainly, the initial response of consumers to the outbreak of the COVID-19 pandemic was to halt going out to crowded places, such as restaurants and bars, which has led to a sudden decrease in consumer demand for restaurant an bars sector. This impact was further intensified with government restrictions and limitations. Likewise, leisure and business travel have almost come to an end during early periods of the COVID-19 pandemic. While leisure travelers have postponed or canceled their travel plans and hotel stays, businesses have quickly shifted to conduct their operations and meetings through virtual communication tools. This sudden change in consumer behavior caused demand for airline and hotels to drop significantly. Also, the government’s strategy of travel restrictions, entry bans, and border control measures to mitigate and the adverse effect of the COVID-19 pandemic on public health has caused a drastic decline in demand for airlines.

Furthermore, the effect of COVID-19 pandemic–induced uncertainty on overall economy and tourism and hospitality industry is elevated in early March, the timing of which coincides with the announcement of the World Health Organization’s recognition of the COVID-19 as a “Public Health Emergency of International Concern” and on the declaration of it a Pandemic in 11 March 2020. In particular, a sharp increase in adverse effects of COVID-19 pandemic–induced uncertainty on restaurants and bars sector is observed during this time period, which suggest that a shift in market sentiment toward the restaurants and bars sector has been observed following the announcement of a new series of coronavirus restrictions, including the shutdown of restaurants, bars, pubs, and other hospitality businesses in the United States. This results in a sudden drop in the sales of restaurants and bars and puts high pressure on companies operating in this sector.

Although the COVID-19 pandemic–induced uncertainty has had a significantly substantial effect of tourism and hospitality industry during the initial stages of the pandemic, the adverse effects of COVID-19 pandemic–induced uncertainty have started to decrease in the following months, which led to less strict regulations and restrictions regarding the COVID-19 pandemic. Accordingly, the easing of lock-downs and other policy measures has led to hopes of potential recovery in mid to late summer of 2020 and that the effect of COVID-19 pandemic–induced uncertainty on overall economy and tourism and hospitality industry has started to diminish in June 2020. However, these hopes have phased out with further spike in the spread of the COVID-19 cases and related hospitalizations and deaths. This is also evident in our analysis that the adverse effect of COVID-19 pandemic–induced uncertainty on overall economy and tourism and hospitality industry has intensified in early June 2021, which is considered to be the time of the delta variant of COVID-19.

Research and Practical Implications

The findings have important theoretical and practical implications. Theoretically, this study contributes to the emerging body of literature on the implications of the COVID-19 pandemic on the overall economy in general and tourism and hospitality industry in particular. Although the implications of various sources of uncertainty (e.g., economic uncertainty, financial crisis, and elections) have been previously examined, the effect of COVID-19 pandemic–induced uncertainty on tourism and hospitality industry has not been widely examined with the exception of studies of Sharma and Nicolau (2020) and Kaczmarek et al. (2021). However, the study period in these studies do not extend beyond the month of April 2020, the analyses are limited to sector specific effects, and the empirical models do not account for essential factors, such as the measure of uncertainty.

Furthermore, the previous studies merely focused on the tourism and hospitality industry when analyzing the implications of the COVID-19 pandemic. While this approach shows the effect of COVID-19 pandemic on tourism and hospitality industry during the study period, it does not show the extent to which the effect of COVID-19 pandemic–induced uncertainty on overall economy and tourism and hospitality industry diverge. Our study provided evidence that the effect of COVID-19 pandemic on tourism and hospitality industry tourism is more significant in magnitude compared to that of the overall economy. That is, our findings suggest that tourism and hospitality industry is more vulnerable to pandemic-related uncertainties than other sectors of the economy.

From a practical perspective, the findings from this study shows that while all subsectors of the tourism and hospitality industry have been affected by the COVID-19 pandemic, our findings provide evidence for heterogeneous effects of COVID-19 pandemic–induced uncertainty on tourism subsectors. Specifically, the hotel sector experienced the largest adverse impact from the COVID-19 pandemic. Our findings therefore suggest that policymakers may design stimulus and rescue packages based on subsectors’ sensitivity to COVID-19 pandemic–induced uncertainty to alleviate the possible contagion effects of the tourism and hospitality sectors to other industries. That is, policymakers should appropriate their aids for the firms and sectors based on the degree of impacts experienced by the pandemic instead of an umbrella package that serve all sectors the same. Dogru and Bulut (2018) argued that tourism can be “an engine for economic recovery.” However, this argument cannot be achieved without a strategic plan that focuses on post-pandemic recovery plans and implementing pandemic preparedness plans in the event of yet another pandemic. This may also require a rethinking of the whole operations and design of the tourism and hospitality industry. Certainly, this will be an innovation-based approach that industry stakeholders need to develop, and while some incumbents may resist to changes, others will adopt and survive and may even thrive.

New companies and products and services are likely to emerge as a result of the COVID-19 pandemic because some consumers are likely to change their behavior permanently and this may lead to new product developments to serve the needs of guests or travelers in a post-pandemic era. Also, the informative nature of COVID uncertainty for out-of-sample forecasting of stock price movements provides new insights for market participants to form hedging strategies for mitigating downside risk in the tourism sectors. For instance, when the VAR-LP forecasting model captures the Granger causality between COVID-19–induced uncertainty and 1-week ahead tourism and hospitality sector stock returns, investors can use this signal as an early indicator to adjust their exposure to tourism sectors in a timely manner.

Limitations and Recommendations for Future Research

While the importance of considering time-varying parameters Granger causality test to draw correct inference about the effect of COVID-19–induced uncertainty on the tourism and hospitality sector is highlighted by our work, in the process of making portfolio decisions associated with tourism sectors, market participants would need to incorporate the role of economic uncertainty. Considering the leading role of the tourism and hospitality industry for the macro-economy, policymakers may develop sector specific policies that help to support the tourism and hospitality industry. Certainly, fiscal policies can mitigate the adverse effects of COVID-19 uncertainty on tourism and hospitality industry. Although it may be difficult to combine the impact of these policies in an econometric framework due to their unscheduled natures, future research should investigate the expansionary monetary policy and possibly fiscal policy to revive the tourism and hospitality industry. Also, our analyses are limited to the firms in the United States, and thus, they may not be generalizable beyond this context. Future studies are necessary to examine this issue in different country settings.

Although the effect of COVID-19 pandemic–induced uncertainty on the tourism and hospitality industry is examined in this study in the context of stock markets, the company level impact and the effect of COVID-19 pandemic–induced uncertainty on the operational aspects of the tourism and hospitality firms have not been analyzed in this study. Indeed, the implications of the COVID-19 pandemic needs to be further investigated at firm level both from companies’ profitability standpoint and also from employment perspective. While the adverse effect of COVID-19 pandemic–induced uncertainty on the tourism and hospitality industry is evident based on the findings from this study, the extent to which COVID-19 pandemic affected the operational performance of tourism and hospitality firms and employment in the tourism and hospitality industry need to be further investigated. Also, as part of future research, it would be interesting to extend our analysis by looking at the effects of COVID-19 pandemic–induced uncertainty on the tourism and hospitality industry of other developed and emerging economies.

Furthermore, the COVID-19–induced uncertainty index used in our study is based on counting the pandemic-related words in the news, and hence, it does not indicate a sentiment about the content of the information. For this reason, the effects of positive and negative pandemic-related news on tourism and hospitality and its subsectors may vary according to the content of the news. Therefore, extending our study by implementing causality analysis on the positive and negative pandemic-related news sentiment obtained using the advanced text-mining tools would be an interesting research question.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

ORCID iDs

Oguzhan Cepni

Tarik Dogru

Ozgur Ozdemir

Notes

Author biographies

Oguzhan Cepni is a Ph.D. Fellow at Copenhagen Business School (CBS). Before joining CBS, He was an Economist at the Central Bank of Turkey with more than six years of experience. His primary research interests include applied macroeconomics, empirical asset pricing, and forecasting.

Tarik Dogru is an assistant professor of hospitality management in the Dedman College of Hospitality at Florida State University. He earned a doctorate degree in hospitality management from University of South Carolina. Prior to joining Florida State University, Dr. Dogru worked at Boston University as an assistant professor for two years. His research interests include sharing economy, corporate finance, franchising, hotel investments, tourism economics, climate change, and blockchain technology.

Ozgur Ozdemir is an assistant professor in the William F. Harrah College of Hospitality at University of Nevada, Las Vegas. He obtained his Ph.D in Hospitality Management with a concentration in financial management from the Pennsylvania State University. Dr. Ozdemir's main research areas include corporate finance, capital markets, agency issues, initial public offerings (IPOs), and corporate governance in the hospitality industry. His work has been published in hospitality, tourism and business journals including Tourism Management, International Journal of Hospitality Management, International Journal of Contemporary Hospitality Management, Journal of Hospitality & Tourism Research, Hospitality Financial Management, and International Business Review.

Appendix

References

AHLA (2020) State of the hotel industry 2021. https://www.ahla.com/sites/default/files2021StateoftheHotelIndustry.pdf

Araña

León

(2008) The impact of terrorism on tourism demand. Annals of Tourism Research 35(2): 299–315.

Aruoba

Diebold

Scotti

(2009) Real-time measurement of business conditions. Journal of Business & Economic Statistics 27: 417–427.

Baker

Bloom

Davis

(2016) Measuring economic policy uncertainty*. The Quarterly Journal of Economics 131: 1593–1636.

Baker

Bloom

Davis

, et al. (2020) The unprecedented stock market impact of COVID-19. In: Technical Report. National Bureau of Economic Research.

Balli

Shahzad

SJH

Salah Uddin

(2018) A tale of two shocks: what do we learn from the impacts of economic policy uncertainties on tourism? Tourism Management 68: 470–475.

Chen

M-H

Kim

(2005) The impact of macroeconomic and non-macroeconomic forces on hotel stock returns. International Journal of Hospitality Management 24(2): 243–258.

Chiang Leong

Hui

T-K

(2014) Macroeconomic and non-macroeconomic variables linking to singapore hotel stock returns. Advances in Hospitality and Leisure 10: 21–36.

Das

Dutta

Bhadra

, et al. (2020). Role of presidential uncertainties on the hotel industry. Annals of Tourism Research 81(C). DOI: 10.1016/j.annals.2019.102762.

10.

Demir

Gözgör

(2018) Does economic policy uncertainty affect tourism? Annals of Tourism Research 69(C): 15–17.

11.

Demiralay

. (2020). Political uncertainty and the us tourism index returns. Annals of Tourism Research. 84(C).

12.

Dogru

Bulut

(2018) Is tourism an engine for economic recovery? Theory and empirical evidence. Tourism Management 67: 425–434.

13.

Goodrich

(2001) Attack on America: a record of the immediate impacts and reactions in the USA travel and tourism industry. Tourism Management 23(6): 573–580.

14.

Gössling

Scott

Hall

(2020) Pandemics, tourism and global change: a rapid assessment of COVID-19. Journal of Sustainable Tourism 29(1): 1–20.

15.

Hospitalitynet (2020) China hotel performance recovered quickly following SARS. https://www.hospitalitynet.org/news/4097 010.html

16.

Jordà

(2005) Estimation and inference of impulse responses by local projections. American Economic Review 95(1): 161–182.

17.

Kaczmarek

Perez

Demir

, et al. (2021) How to survive a pandemic: the corporate resiliency of travel and leisure companies to the COVID-19 outbreak. Tourism Management 84: 104 281

18.

Karabulut

Bilgin

Demir

, et al. (2020) How pandemics affect tourism: International evidence. Annals of Tourism Research 84: 102991.

19.

Kosová

Enz

(2012) The terrorist attacks of 9/11 and the financial crisis of 2008. Cornell Hospitality Quarterly 53(4): 308–325.

20.

Madanoglu

Ozdemir

(2018) Economic policy uncertainty and hotel operating performance. Tourism Management 71: 443.

21.

National Restaurant Association . (2020a). New research details early impact of coronavirus pandemic on restaurant industry https://restaurant.org/articles/news/study-details-impact-of-coronavirus-on-restaurants

22.

National Restaurant Association (2020b). Retrieved from https://restaurant.org/articles/news/restaurant-sales-fell-to-lowest-level-since-june

23.

Natioanl Restaurant Association (2020c). Retrieved from https://www.restaurant.org/articles/news/restaurant-employment-recovery-is-in-d

24.

Oaten

Le Quesne

Segal

(2015) Adapting to uncertainty- the global hotel industry. The Travel & Tourism Competitiveness Report 2015. Retrieved from http://www3.weforum.org/docs/TT15/WEF_TTCR_Chapter1.2_2015.pdf

25.

Ozdemir

Dogru

Kizildag

, et al. (2021) Quantifying the economic impact of COVID-19 on the US hotel industry: examination of hotel segments and operational structures. Tourism Management Perspectives 39, 100864.

26.

Ozdemir

Kizildag

Dogru

, et al. (2022) Measuring the effect of infectious disease-induced uncertainty on hotel room demand: a longitudinal analysis of US hotel industry. International Journal of Hospitality Management 103, 103189.

27.

Rossi

(2005) Optimal tests for nested model selection with underlying parameter instability. Econometric Theory 21: 962–990.

28.

Rossi

Wang

(2019) Vector autoregressive-based granger causality test in the presence of instabilities. The Stata Journal: Promoting Communications on Statistics and Stata 19(4): 883–899.

29.

Russy

Smith

(2013) The economic impact of H1N1 on Mexico’s tourist and pork sectors. Health Economics 22(7): 824–834.

30.

SelectUSA (2017) Travel, tourism and hospitality spotlight. Retrieved from https://travel.trade.gov/outreachpages/download_data_table/Fast_Facts_2017.pdf

31.

Sharma

Nicolau

(2020) An open market valuation of the effects of COVID-19 on the travel and tourism industry. Annals of Tourism Research 83: 102990.

32.

Tiwari

Das

Dutta

(2019) Geopolitical risk, economic policy uncertainty and tourist arrivals: evidence from a developing country. Tourism Management 75: 323–327.

33.

UNWTO. (2020) International tourism and Covid-19. UNWTO tourism dashboard. Retrieved from https://www.unwto.org/international-tourism-and-covid-19

34.

UNWTO. (2011) Economic crisis, international tourism decline and its impact on the poor. Retrieved from https://assets.publishing.service.gov.uk/media/57a08ac040f0b652dd0008a0/UNWTO_29Nov11.pdf

35.

UNWTO. (2010). World tourism barometer 8(1), January 2010.

36.

U.S. Travel Association . (2020). Weekly coronavirus impact on travel expenditures in the U.S. Retrieved from ustravel.org/sites/default/files/media_root/document/Coronavirus_weeklyimpacts_12.17.20.pdf

37.

Worldometers (2020). Retrieved from https://www.worldometers.info/coronavirus/

38.

World Health Organization (2020) Timeline: WHO’s COVID-19 reponse. Retrieved from https://www.who.int/emergencies/diseases/novel-coronavirus-2019/interactive-timeline#event-71

39.

WTTC. (2020) Economic impact reports. Retrieved from https://wttc.org/Research/Economic-Impact

40.

Zopiatis

Savva

Lambertides

, et al. (2019) Tourism stocks in times of crisis: an econometric investigation of unexpected nonmacroeconomic factors. Journal of Travel Research 58(3): 459–479.