Global Economic Policy Uncertainty and Tourism of Fragile Five Countries: Evidence from Time and Frequency Approaches

Abstract

The purpose of this study is to examine the relationship between global economic policy uncertainty (GEPU) and tourism activities in the Fragile Five (F5) countries, namely, Brazil, India, Indonesia, South Africa, and Turkey. By using wavelet transform context structures and the annual data during the period of 1997–2016. The finding shows that the relationship is generally positive but changes over time, displaying low- to high-frequency cycles. Moreover, the timing and frequency change when GEPU co-moves with tourism. It can be recommended that the government maintain the national security and peace protocols.

Keywords

tourism activities global economic policy uncertainty wavelet analysis time and frequency domains

Introduction

There has already been interest in the relationship between global economic policy uncertainty (GEPU) and tourism activities, although the theoretical ground of such relationship is not well defined. The economies of the world have become more interconnected than they have ever been; however, this phenomenon is no doubt a direct result of uncertainty. A shock wave related to any specific economic, social, or policy activities in one country can travel across the globe instantly because of advanced technology, ubiquitous communications, and pervasive media. The uncertainties in economic policies include uncertainty in the decisions of economic policy makers, which influence the decisions on economic units such as consumption, investment, saving, and lending. Accordingly, the whole economy may be negatively influenced by the uncertainty in policies. The tourism sector is experiencing important challenges and changes as a result of the debt crisis. The economy faced a sharp recession in 2008 at the beginning of the financial crisis and, after some growth in 2010, it reentered recession as a consequence of the debt crisis. Tourism is a strategic sector in the economy and a potential driving force of economic growth, but it is vulnerable to economic crises as in other economies. This work investigates the impact of GEPU on tourism in the Fragile Five (F5) countries. Why are the F5 countries interesting? According to Morgan Stanley’s ranking system for emerging market economies, the medium-term risks such as inflation, real exchange rates, industrial metal prices, current account positions, and balance of payments reliance on income inflows are taken into consideration and then the overall external vulnerability of the given country is calculated. The F5 countries are in a situation of uncertainty about the condition of the economy. This may be important for economies that have fragile tourism systems because the estimation of relationships can explain the connection between global economic policy uncertainty, and this will provide signals for policy makers who attempt to sustain the system stability of tourism activities.

Over the past decades, the relationship between Economic Policy Uncertainty (EPU) or GEPU has been confirmed to have an impact on tourism activities (e.g., Alegre, Mateo, and Pou 2013; del Mar Alonso-Almeida and Bremser 2013; Perles-Ribes et al. 2016; Wu and Wu 2019a). Previous studies (e.g., Andraz and Rodrigues 2016; Birău 2014; Bodosca 2015; Brito 2014; Costa, Gomes, and Montenegro 2014) have used traditional analysis methods based on autoregressive models, and linear or nonlinear cointegration techniques, such as the Johansen, fractional, and threshold cointegration tests, which have been frequently applied in the first strand of the literature to focus on whether the EPU and tourism activities are cyclical or anticyclical associations among the series.

This article proposes the use of a wavelet analysis to explore the real relationship between GEPU and tourism activities in both the time and frequency domains. The wavelet analysis has significant superiority over those conventional time-domain methods applied in previous studies. It expands the underlying time series into a time-frequency space where both time- and frequency-varying information of the series can be visualized in a highly intuitive way. By means of wavelet coherence and wavelet phase difference, a simultaneous assessment of how the co-movement and causalities between GEPU and tourism activities vary across frequencies and change over time is further achieved in a time-frequency window. In this way, the high-frequency (short-term) and low-frequency (long-term) relationships between GEPU and tourism activities, as well as possible structural changes and time-variations in such relationships can be clearly observed. The application of wavelet analysis in economics and finance was introduced by Goffe (1994) and Ramsey and Lampart (1998a, 1998b). However, until recent years, it has been performed popularly by focusing on the co-movement between stock markets as well as between energy commodities and the macro-economy (e.g., Aguiar-Conraria and Soares 2011; Graham and Nikkinen 2011; Loh 2013; McCarthy and Orlov 2012; Rua and Nunes 2009; Wu and Wu 2019b).

This work contributes to the existing literature in several important ways. First of all, the wavelet analysis is well utilized in this study to devote special and full attention to dynamic co-movement and causalities between GEPU and tourism activities for countries. Second, we use the distinctive data of GEPU and real international tourism receipts (ITR) of Baker, Bloom, and Davis (2016) over a long period (1995–2016), implying that our findings would be more suggestive for practitioners and policy makers in GEPU and tourism activities. Accordingly, policy makers can use tourism activities as a policy tool for falling regional welfare inequalities. Third, this study applies the international tourist arrivals (ITA) and real GDP (RGDP) per capita as two control variables to reveal the true relationship between GEPU and tourism activities by removing the effect of economic growth on both GEPU and tourism performance. Subsequently, the partial wavelet coherence and wavelet phase difference are also estimated in this study as necessary complements to the commonly used wavelet coherence and wavelet phase difference. Finally, this article performs a simultaneous assessment of tourism co-movement and causal relationships in the time and frequency domains. The results show very substantial time-and-frequency varying features in the co-movement and causality between GEPU and tourism activities. This provides additional and useful implications of investment strategies and forecasting performance of tourism activities for policy makers in GEPU and tourism activities. Besides, the co-movement during the period of GEPU has been identified since the results demonstrate that GEPU and tourism activities are actually more responsive to economic growth fundamentals rather than responding to each other significantly. In the light of worldwide attention to the economic activities in these countries, tourism investigation into the connection among the sequence is not only appropriate but possibly long overdue. All these notable differences play a crucial role in improving the quality of conclusions.

Literature Review

Theoretical Background

There has been growing literature of particular concern about the causal relationship between GEPU and tourism activities and it is important for policy makers to identify the nature of this causal relationship. Tourism is the foremost sector helping the government policy makers cater for these issues by providing foreign exchange that is beneficial for creating provincial employment opportunities that are essential in managing unemployment and encouraging construction, accommodation, transportation, and beverage or food sectors that can bring a rise in tourism activities by providing added value. Moreover, this sector also develops in conjunction with countries by transferring income to developing countries from developed ones. Accordingly, policy makers can use tourism as a policy tool for falling regional welfare inequalities. Based on previous literature, the relationship between GEPU and tourism activities has generally been addressed by two different components in the tourism economic literature (Sharif, Saha, and Loganathan 2017).

GEPU or EPU and its management constitute a popular topic for tourism researchers. Previous studies have confirmed the characteristics of economic crises and actions taken to overcome uncertainty. In terms of the first component, with respect to the studies analyzing the demand (i.e., Song et al. 2012), numerous studies only address demand forecast. European EPU has been seen as unpredictable shocks that should be considered in demand forecasts for a given destination. Several studies focus on modeling tourism demand (see Alegre, Mateo, and Pou 2013; Song et al. 2011), analyzing demand elasticities and how they are affected by crises (demand segments, supply categories, and the deviation effects between destinations depending on their distance from origin markets, price differentials, market share, etc.). In general, focusing only on demand ignores a large part of the effect of EPU or crises on destinations. As for the second component from the supply-side perspective, previous studies have analyzed how a part of the industry responds to crises (i.e., del Mar Alonso-Almeida and Bremser 2013) to draw management lessons for responding to the prospective shocks and EPU that may occur. However, the answers that are valid for one sector may not be valid for an entire tourism destination, which is a more complex entity with many interactions and conflicting interests.

More recently, Perles-Ribes et al. (2016) have reviewed the existing literature on economic crises and tourism activities, dividing the studies into those focusing on aspects of demand that analyze the reactions of industry and those studying how crises affect tourism destinations. Wu and Wu (2019a) provide new insights into the relationship between economic policy uncertainty and tourism activities in Brazil, Russia, India, and China (i.e., the BRIC countries). They use continuous wavelets, and partial and multiple wavelets to analyze the relationship between EPU and tourism using the annual data. The results show that the relationships among the variables evolve with time and frequencies. From the time-domain view, this article shows strong evidence of relationships between these variables. From the frequency-domain view, it shows significant wavelet coherences and strong lead–lag relationship changes over time, displaying low- to high-frequency cycles. Next, Wu and Wu (2019b) aims to examine the link between European EPU and tourism activities in Portugal, Ireland, Italy, Greece, and Spain using wavelet transform context structures. The results indicate that there is a unidirectional causal influence of European EPU on international tourism receipts in the short run and a bidirectional causal influence of European EPU on ITR in European countries in the long run.

This article contributes to the existing literature in three ways. First, to the best of our knowledge, this is the first attempt to explore the causal nexus between GEPU and tourism development in the F5 countries using a wavelet approach. Second, this work reclassifies the frequency on the y-axis into four bands: 1–2-, 2–4-, 4–8-, and 8–12-year frequency bands, corresponding to the short-run, mid-run, and long-run relationships between GEPU and ITR. The researchers also display the partial wavelet coherence and wavelet phase difference after controlling the other two variables (i.e., RGDP and ITA) for the F5 countries. The partial wavelet coherence and wavelet phase-difference approach applied in this study allow the capture of time variations in the causal relations and, therefore, provide a comprehensive and detailed view of the GEPU and tourism development link in the F5 countries. Third, differently from previous studies, a novel GEPU indicator is used in this study. The contributions of this work are twofold: (1) based on the existing literature, this is the first study to use GEPU as a variable to investigate whether GEPU is an important determinant of the performance in the international tourism sector, and (2) not only domestic uncertainty but also international uncertainty can be significant factors.

Tourism Related Literature on EPU

Numerous studies address the impact of EPU or financial crises on tourism regions or countries. For example, Frechtling (1982) and Schulmeister (1979) stressed the resilience of tourism activities during the energy crises of the 1970s and their asymmetric effects on destinations. Sanuy (1983) claimed that not all Spanish areas were equally influenced by the oil crisis, stressing factors such as changes in demand behavior, increases in the price sensitivity of tourists, reductions in long-distance trips, and increases in last-minute reservations. Aguiló, Alegre, and Sard (2005) summarized the process for the case of Spain and pointed out that the opening-up process was triggered by the fall of the Iron Curtain, together with the emergence of new competitors, which culminated in an exhaustion of Spain’s tourism model. Residential tourism constituted a major part in response to the government’s implementation of programs to adapt the tourism supply to the new changes (Vera 1994).

The existing studies have exclusively used the conventional time-domain methods, namely, the aforementioned cointegration techniques and causality tests as well as correlation analysis. This article is different from the studies of Sharif, Saha, and Loganathan (2017) and the other existing literature, concentrating on the time- and frequency-varying features of both the co-movement and causalities between EPU and tourism activities, rather than analyzing their co-movement. Moreover, this article used GEPU based on the study of Baker, Bloom, and Davis (2016). The ITA and RGDP are used as two control variables in an attempt to extract the real relationship between GEPU and ITR. All these notable differences play a crucial role in improving the quality of the conclusions. Namely, for most of the existing literature, the time-variation existing in the relationship between GEPU and tourism has been ignored. Moreover, the frequency-variation in such relationship has certainly not been identified by the time-domain approaches as well. Nevertheless, as a matter of fact, time- and frequency-varying features in such relationship have important practical implications for government policy makers. Time-varying co-movement implies that the GEPU and diversification benefits of tourism activities evolve with time, and thus policy makers should take it into account when examining the co-movement between GEPU and tourism activities. Frequency-varying co-movement suggests that the policy makers with different GEPU horizons should pay more attention to the co-movement at corresponding frequencies so as to establish tourism activities more effectively. The time- and frequency-varying features in the causality can also significantly affect the accuracy of forecasting tourism activities, and hence influence investment benefits of practitioners and regulatory benefits of policy makers. The review of the aforementioned literature comprehends the existence of a unidirectional causal relationship between GEPU and tourism activities in a majority of countries. However, some GEPU has a direct, indirect, and bidirectional causal relationship with tourism activities. This can reflect the need to further evaluate the causality and the reasons behind the existence of such difference. Accordingly, this work aims to perform an investigation into the relationship between GEPU and tourism activities using the extensive framework of continuous wavelet analysis.

Data Collection and Wavelet Theory

The data of the F5 countries¹ were collected from the World Bank in the World Development Indicators (WDI) (i.e., ITA, ITR, and RGDP), and the sample was restricted to these countries—for which the data of the GEPU index were introduced by Baker, Bloom, and Davis (2016). GEPU is one of the influential global risk factors in global market performance (see Arouri et al. 2016; Baker, Bloom, and Davis 2016). The GEPU index is a (PPP-adjusted) GDP-weighted average of national EPU indices for the 20 countries. Note that the national EPU index in each country reflects the relative frequency of own-country newspaper articles including a trio of terms pertaining to the economy (E), policy (P), and uncertainty (U). This study opts for the GEPU index instead of domestic EPU indices in a particular country or region because the use of international tourism as global tourism activities necessitates the choice of a global measure of economic uncertainty such as the GEPU index. This index can be interpreted as the measure of common variations in economic policy uncertainty and macroeconomic data uncertainty across countries. In particular, note that E, P, and U refer to the factors that are constructed using only “own country” variables. The annual data used in this study included the period from 1997 to 2016 for the F5 countries. The sample period was decided purely by data availability on the measure of tourism activities. All variables were used in their natural log form. The data used in this study focused on ITR to evaluate the performance of tourism activities and GEPU to measure the uncertainties in economic policies for each country. Tables 1 through 3 report the summary statistics for the GEPU, ITR, ITA, and RGDP for each country.

Table 1.

A Summary of Statistics of ITR in the F5 Countries.

Countries	Mean	Maximum	Minimum	SD	Skewness	Kurtosis	JB
Brazil	22.04	22.73	20.75	0.62	−0.66	2.03	2.25
India	22.85	23.86	21.78	0.77	−0.17	1.47	2.06
Indonesia	22.63	23.26	22.17	0.36	0.44	1.80	1.84
Africa	22.65	23.14	21.90	0.49	−0.70	1.70	3.05
Turkey	23.60	24.38	22.37	0.62	−0.56	2.01	1.87

Note: The sample period is between 1997 and 2016. ITR = international tourism receipts; F5 = Fragile Five; SD = standard deviation; JB = Jarque Bera.

Table 2.

A Summary of Statistics of ITA in the F5 Countries.

Countries	Mean	Maximum	Minimum	SD	Skewness	Kurtosis	JB
Brazil	15.43	15.70	14.86	0.19	−1.29	5.32	10.04
India	15.35	16.49	14.67	0.60	0.56	2.28	1.47
Indonesia	15.65	16.26	15.31	0.29	0.75	2.23	2.36
Africa	15.83	16.12	15.42	0.22	−0.25	1.72	1.56
Turkey	16.81	17.50	15.75	0.58	−0.39	1.69	1.95

Note: The sample period is between 1997 and 2016. ITA = international tourism arrivals; F5 = Fragile Five; SD = standard deviation; JB = Jarque Bera.

Table 3.

A Summary of Statistics of RGDP in the F5 Countries.

Countries	Mean	Maximum	Minimum	SD	Skewness	Kurtosis	JB
Brazil	8.75	9.49	7.944	0.53	−0.10	1.56	1.75
India	6.72	7.45	6.01	0.54	−0.06	1.39	2.16
Indonesia	7.41	8.21	6.14	0.68	−0.21	1.71	1.52
Africa	8.47	8.98	7.81	0.37	−0.43	1.82	1.78
Turkey	8.87	9.44	8.05	0.50	−0.44	1.61	2.26
A Summary of statistics of GEPU in the F5 countries.
F5	4.62	5.25	4.13	0.30	0.10	2.28	0.46

Note: The sample period is between 1997 and 2016. RGDP = real gross domestic product; F5 = Fragile Five; SD = standard deviation; GEPU = global economic policy uncertainty; JB = Jarque Bera.

For economic uncertainty, the researchers used the GEPU index. This index is constructed from several components. To measure the GEPU index, the researchers proceeded as follows. First of all, the index re-normalized each national EPU index to a means of imputing missing values for Australia, India, Greece, the Netherlands, and Spain using a regression-based method. This step yielded a balanced panel of monthly EPU index values for the 18 countries² (i.e., the US, Canada, Brazil, Chile, the United Kingdom, Germany, Italy, Spain, France, Sweden, Russia, India, China, South Korea, Australia, Ireland, the Netherlands, and Japan). Second, the researchers computed the GEPU index value for each month as the GDP-weighted average of the 18 national EPU index values, using the GDP data from the IMF’s World Economic Outlook Database. Besides, this work used control variables as other important variables, which if omitted might cause the estimated coefficients to be biased. Accordingly, this work adopted ITA and RGDP as control variables respectively. To have a clear picture, Figure 1 plots the GEPU index versus the ITR index across the F5 countries.

Figure 1.

GEPU versus ITR across the F5 countries.

Continuous Wavelet Transform

Wavelet is a tool that allows the determining of dominant modes of variability and how these modes change with time (Torrence and Compo 1998). Traditional mathematical methods, such as Fourier transform, examine the periodicity of phenomena by assuming that they are stationary in time. The wavelet, however, decomposes time series into a time-frequency space and thus is able to extract localized intermittent periodicities (Grinsted, Moore, and Jevrejeva 2004). It is therefore particularly useful in analyzing nonstationary time series. This section briefly explains the main concepts of continuous wavelet transforms. Wavelet analysis originated in the mid-1980s as an alternative to the well-known Fourier analysis. Though Fourier analysis uncovers how relations vary across frequencies using spectral techniques, time-localized information is completely discarded under the Fourier transform. What is more, Fourier analysis is merely suitable for stationary series. In contrast, wavelet analysis conducts the estimation of spectral characteristics of time series as a function of time (Aguiar-Conraria, Azevedo, and Soares 2008). It is therefore able to extract localized information in both the time and frequency domains. Besides, wavelet analysis has a significant advantage over Fourier analysis when the underlying series are nonstationary or locally stationary (Roueff and Von Sachs 2011).

There are often two kinds of wavelet transforms: discrete wavelet transforms (DWT) and continuous wavelet transforms (CWT). The former is useful for noise reduction and data compression, whereas the latter is helpful for feature extraction and data self-similarity detection (Grinsted, Moore, and Jevrejeva 2004; Loh 2013). In this work, the CWT was chosen as a useful tool to decompose the concerned series into wavelets. Interested readers can refer to several studies (e.g., Aguiar-Conraria and Soares 2014; Wu and Wu 2019a, 2019b; Wu et al. 2019) in detail.

Partial Wavelet Coherence and Wavelet Phase Difference

Wavelet coherence allows a three-dimensional analysis, which simultaneously considers the time and frequency components, as well as the strength of correlation between time series (Loh 2013). In this way, both time variation and frequency variation of the correlation between series can be clearly observed in a time–frequency space. Consequently, the wavelet coherence is well utilized in this study as a much better measure of the co-movement between GEPU and tourism activities in comparison to the conventional correlation analysis as well as the dynamic conditional correlation method (Liow 2012; Loh 2013; Zhou 2010).

The partial correlation is one of the tools that can be used in a simple correlation concept. In the wavelet, the researchers can attain this using the partial wavelet coherence. This approach is able to identify the partial wavelet coherence between the two time series y and X₁after eliminating the power of the third time series X₂. Accordingly, the wavelet coherence coefficients between y and X₁, y and X₂, and X₁ and X₂ are written as follows:

R (y, x_{1}) = \frac{S [W (y, x_{1})]}{\sqrt{S [W (y)] \cdot S [W (x_{1})]}}

(3)

R^{2} (y, x_{1}) = R (y, x_{1}) \cdot R {(y, x_{1})}^{*}

(4)

R (y, x_{2}) = \frac{S [W (y, x_{2})]}{\sqrt{S [W (y)] \cdot S [W (x_{2})]}}

(5)

R^{2} (y, x_{2}) = R (y, x_{2}) \cdot R {(y, x_{2})}^{*}

(6)

R (x_{2}, x_{1}) = \frac{S [W (x_{2}, x_{1})]}{\sqrt{S [W (x_{2})] \cdot S [W (x_{1})]}}

(7)

R^{2} (x_{2}, x_{1}) = R (x_{2}, x_{1}) \cdot R {(x_{2}, x_{1})}^{*}

(8)

R P^{2} (y, x_{1}, x_{2}) = \frac{{| R (y, x_{1}) - R (y, x_{2}) \cdot R {(y, x_{1})}^{*} |}^{2}}{{[1 - R (y, x_{2})]}^{2} {[1 - R (x_{2}, x_{1})]}^{2}}

(9)

where ${RP}^{2}$ is the partial wavelet coherence squared.

Accordingly, a low (high) partial wavelet coherence squared revealed at the high (low) wavelet coherence implies that time series X₁ does not have (has) a significant effect on the time series y at that time-frequency space, while time series X₂ dominates (surrenders) the effect on the variance of y. If both ${RP}^{2}$ ( $y, x_{1}, x_{2}$ ) and ${RP}^{2}$ ( $y, x_{2}, x_{1}$ ) still have noteworthy bands, both X₁ and X₂ have a significant effect on y. This work focuses on the wavelet coherency instead of the wavelet cross spectrum for two reasons: (1) the wavelet coherency has the advantage of being normalized by the power spectrum of the two time series; and (2) Maraun and Kurths (2004) propose that the wavelet cross spectrum can display strong peaks even for the realization of independent processes, suggesting the possibility of spurious significance tests.

$X = x (t)$ $Y = y (t)$

According to Bloomfield et al. (2004), the wavelet phase difference characterizes the phase relationship between x(t)and y(t) as follows:

\begin{array}{l} ϕ_{x y} (τ, s) = \tan^{- 1} (\frac{Τ {S (s^{- 1} W_{x y} (τ, s))}}{Ν {S (s^{- 1} W_{x y} (τ, s))}}), \\ with ϕ_{x y} \in [- π, π] \end{array}

(10)

where T and N are the imaginary and real parts of $W_{x y} (τ, s)$ , which are complex given the Morlet type of mother wavelet.

A wavelet phase difference of zero indicates that the two underlying series move together, whereas a wavelet phase difference of π (-π) implies that it moves in the opposite direction. If $φ_{x y} \in (0, π / 2)$ , then the series is in phase (positively co-moved) with x (t) leading y (t). If $φ_{x y} \in (π / 2, π)$ , then the series is out of phase (negatively co-moved) with y (t) leading x (t). If $φ_{x y} \in (- π, - π / 2)$ , then the series is out of phase with x (t) leading y (t). If $φ_{x y} \in (- π / 2, 0)$ , then the series is in phase with x (t) leading y (t). Note that the wavelet phase difference can also be indicative of causality between x (t) and y (t) in both the time and frequency domains. As a consequence, it is significantly superior to the conventional Granger causality test, which assumes that a single causal relationship holds the whole sample period as well as each frequency (Grinsted, Moore, and Jevrejeva 2004; Tiwari, Mutascu, and Andries 2013).

Main Results and Implications

In this section, we discuss the results of the wavelet methods according to the data analysis above. The partial wavelet coherence, together with the partial wavelet phase difference, provides reliable indications of correlation and lead–lag relationships between GEPU and ITR. As shown in Figures 2 through 6, this study reclassifies the frequency on the y-axis into four bands: 1–2-, 2–4-, 4–8-, and 8–12-year frequency bands, corresponding to the short-run, mid-run, and long-run relationships between GEPU and ITR, respectively. The right side in Figures 2 through 6 displays the wavelet phase difference after controlling the other two variables (i.e., RGDP and ITA) for the F5 countries.

Figure 2.

Brazil.

Figure 3.

India.

Figure 4.

Indonesia.

Figure 5.

South Africa.

Figure 6.

Turkey.

From Figures 2 through 6, the significant correlations of partial wavelet coherence between GEPU and ITR (at 5% level of significance) were found in the short-run, mid-run, and long-run co-movement observed for the 1–2-, 2–4-, 4–8-, and 8–12-year cycle. The short-run co-movement observed for the 1–2-year cycle and the partial phase difference limited in a range between zero and ½ π (i.e., the in-phase relationship between GEPU and ITR) revealed that GEPU was positively correlated with ITR and that GEPU led ITR during the period of 2000–2003 (a period close to the technology bubble crisis) and 2012–2017 (a period close to the Euro crisis) (see Figure 2). In the meantime, the partial phase difference limited in a range between –½ π and zero (i.e., the in-phase relationship between GEPU and ITR) indicated that GEPU was positively correlated with ITR and that ITR led GEPU during the period of 2004–2006. In the midterm run for the 2–4-year cycle, the partial phase difference limited in a range between zero and ½ π indicated that GEPU was positively correlated with ITR and that GEPU led ITR during the period of 1999–2000 (a period close to the technology bubble crisis). In addition, the partial phase difference limited in a range between –½ π and ½ π (i.e., the in-phase relationship between GEPU and ITR) displayed that an interaction causal relationship between ITR and GEPU was found during the period of 2000–2003. In terms of the case of India, the short-run co-movement observed for the 1–2-year cycle and the partial phase difference limited in a range between zero and ½ π revealed that GEPU was positively correlated with ITR and that GEPU led ITR during the period of 2007–2009.

In addition, the partial wavelet phase differences in Figure 4 showed that the short-run co-movement was observed for the 1–2-year cycle. As indicated in Figure 3, the partial phase difference limited in a range between zero and ½ π showed that GEPU was positively correlated with ITR and that GEPU led ITR during the periods of 1998–2000 and 2006–2010. Meanwhile, the partial phase difference limited in a range between –½ π and zero (i.e., the in-phase relationship between GEPU and ITR) displayed that GEPU was positively correlated with ITR and that ITR led GEPU during the period of 2002–2006. Subsequently, in the midterm run for the 2–4-year cycle, the partial phase difference limited in a range between zero and ½ π revealed that GEPU was positively correlated with ITR and that GEPU led ITR during the period of 2004–2006 in Indonesia.

The short-run co-movement was observed for the 1–2-year cycle (see Figure 5). As shown in Figure 5, the partial phase difference limited in a range between zero and ½ π displayed that ITR was positively correlated with GEPU and that GEPU led ITR during the period of 2008–2009. At the same time, the partial phase difference limited in a range between –½ π and zero indicated that ITR was positively correlated with GEPU and that ITR led GEPU during the period of 2003–2004. In addition, the partial phase difference limited in a range between –½ π and ½ π revealed that the interaction causal relationship between ITR and GEPU was found during the period of 1998–2002 (a period close to the Asian financial crisis and the Iraq War). In the midterm run for the 2–4-year cycle, the partial phase difference limited in a range between –½ π and zero showed that ITR was positively correlated with GEPU and that ITR led GEPU during the period of 1999–2000 (a period close to the technology bubble crisis).

With regard to the case of Turkey, the result indicated that the short-run co-movement was observed for the 1–2-year cycle. As shown in Figure 5, the partial phase difference limited in a range between zero and ½ π revealed that ITR was positively correlated with GEPU and that GEPU led ITR during the periods of 2001–2002, 2006–2008, 2008–2010, and 2014–2017 (a period close to the Euro crisis). In the midterm run for the 2–4- and 4–8-year cycles, the partial phase difference limited in a range between –½ π and zero (i.e., the in-phase relationship between ITR and GEPU) revealed that ITR was positively correlated with GEPU and that ITR led GEPU during the period of 2006–2007. In addition, the partial phase difference limited in a range between –½ π and ½ π (i.e., the in-phase relationship between ITR and GEPU) displayed that the interaction causal relationship between ITR and GEPU was found during the periods of 2010–2014 (a period close to the Eurozone budget deficit crisis) and 2001–2010. In the long run for the 8–12 year cycle, the partial phase difference limited in a range between ½ π and π (i.e., the in-phase relationship between ITR and GEPU) indicated that ITR was negatively correlated with GEPU and that ITR led GEPU during the period of 2003–2010 (a period close to the financial crisis). In sum, the effects and lead–lag relationships between GEPU and ITA during the sample period across all the frequency bands are shown in Table 4.

Table 4.

The Lead–Lag Relationship between GEPU and ITR.

Countries	Frequency time
	1–2-Year Band		2–4-Year Band		4–8-Year Band		8–12-Year Band
	Leading	Correlation	Leading	Correlation	Leading	Correlation	Leading	Correlation
Brazil
2000–2003	GEPU	Positive
2004–2006	ITR	Positive
2012–2017	GEPU	Positive
1999–2000			GEPU	Positive
2000–2003			GEPU←→ITR	Interaction
India
2007–2009	GEPU	Positive
Indonesia
1998–2000	GEPU	Positive
2002–2006	ITR	Positive
2006–2010	GEPU	Positive
2004–2006			GEPU	Positive
South Africa
1998–2002	GEPU←→ITR	Interaction
2003–2004	ITR	Positive
2008–2009	GEPU	Positive
1999–2000			ITR	Positive
Turkey
2001–2002	GEPU	Positive
2006–2008	GEPU	Positive
2008–2010	GEPU	Positive
2014–2017	GEPU	Positive
2006–2007			ITR	Positive
2010–2014			GEPU←→ITR	Interaction
2001–2010					GEPU←→ITR	Interaction
2003–2010							ITR	Negative

Note: GEPU = global economic policy uncertainty; ITR = international tourism receipts.

Previously, certain shocks such as the outbreak of SARS, the EPU, the US financial crisis, Euro crisis and the Iraq War were considered as potential determinants in the tourism sector performance (Chen 2007). None of the studies focuses on examining the effects of GEPU variables on tourism issues. The significant effects of uncertainty variables have been central in our findings. Accordingly, the researchers consider these as the novelty of this study since they show that the performance in the tourism sector depends on domestic and international GEPU. The contributions of this study are twofold. First of all, based on the existing literature, this is the first study to use GEPU as a variable to investigate whether GEPU is an important determinant of the performance in the tourism sector. Second, not only domestic uncertainty but also international uncertainty can be significant factors. The suggested relationships are essential because macroeconomic variables (i.e., potential alternative determinants) fail to explain the tourism sector returns.

Conclusions and Research Limitations

The wavelet analysis allows us to make a simultaneous assessment of the co-movement and causality between GEPU and ITR in both the time and frequency domains. The empirical results show robust evidence that the co-movement and causality vary across frequencies and evolve with time. In terms of the lead–lag relationship from the time domain, GEPU and ITR display the commonly positive or negative lead–lag relationship over the past decades with an interaction co-movement exception in Germany. Also, from the frequency domain, GEPU and ITR correlate with each other mainly across lower frequencies, whereas GEPU and ITR are correlated with each other over the long term. Besides, with respect to the causal relationship between GEPU and ITR, the researchers find that the causal effects are generally included at lower frequencies from a frequency-domain view. In contrast, from a time-domain view, the time-varying features in the long-run causalities imply the structural changes in GEPU and ITR. These findings provide an overview of GEPU and ITR in the F5 countries over the past decades and have important implications for policy makers and practitioners.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Department of Education of Guangdong Province (Grant Number 2018WTSCX214).

1.

The F5 is a term that was coined in August 2013 by a financial analyst at Morgan Stanley to represent emerging market economies that have become too dependent on unreliable foreign investment to finance their growth ambitions. As capital flew out of emerging markets to developed markets, many of the currencies experienced significant weakness and it became difficult to finance current account deficits. The lack of new investment also made it impossible to finance many growth projects, which contributed to a slowdown in their respective economies. This created a potential issue for certain vulnerable economies.

2. Author Biographies

Tsung-Pao Wu is a Professor at School of Accounting and Finance in Beijing Institute of Technology, Zhuhai, China. He received his PhD in finance from Feng Chia University and has published his works in the Journal of Applied Economics and Business Research, Journal of Sustainable Tourism, Current Issues in Tourism, Journal of China Tourism Research, Journal of Hospitality and Tourism Management, Journal of Travel Research, Applied Economics, International Journal of Tourism Research, Tourism and Hospitality Research, Tourism Analysis, Applied Economics, Applied Economics Letters, Applied Financial Economics, Global Economic Review, Romanian Journal of Economic Forecasting, Empirical Economics, Macroeconomics and Finance in Emerging Market Economies, Economic Modelling, Tourism Planning and Development, Journal of Real Estate Portfolio Management, International Journal of Economic Policy in Emerging Economies, Japan and the World Economy, and International Economic Journal, among others. His research interest focuses on tourism economics. Tsung-Pao Wu is the corresponding author and can be contacted at wutsungpao@yahoo.com.hk.

Hung-Che Wu is an Associate Professor at Business School in Nanfang College of Sun Yat-Sen University, China. He received his PhD in marketing and economics from Lincoln University and has published his works in International Journal of Hospitality Management, Sustainability, Scandinavian Journal of Hospitality and Tourism, Journal of Tourism and Cultural Change, Journal of Sustainable Tourism, Journal of Applied Economics and Business Research, Journal of Retailing and Consumer Services, Journal of Vacation Marketing, Current Issues in Tourism, International Journal of Bank Marketing, Journal of International Food & Agribusiness Marketing, Marketing Intelligence & Planning, Journal of Destination Marketing and Management, Tourism Management, Journal of Environmental Planning and Management, Journal of the Knowledge Economy, Industrial Management & Data System, Journal of Organizational Change Management, Innovative Marketing, Journal of Travel Research, Tourism Review International, Tourism Review, Tourism Analysis, International Journal of Economics and Financial Issues, Tourism Planning and Development, Journal of Tourism, Hospitality & Culinary Arts, Journal of Quality Assurance in Hospitality & Tourism, Journal of Convention and Event Tourism, Asia Pacific Journal of Marketing and Logistics, Event Management, Journal of Foodservice Business Research, Journal of Hospitality Marketing & Management, Asia Pacific Journal of Tourism Research, International Journal of Tourism Research, International Journal of Tourism Sciences, British Food Journal, Multimedia Tools and Applications, International Journal of Contemporary Hospitality Management, International Journal of Tourism & Hospitality Administration, Journal of Hospitality and Tourism Management, International Journal of Economics and Financial Issues, Journal of China Tourism Research, Tourism and Hospitality Research, and Journal of Hospitality & Tourism Research, among others. He is currently working as an editor-in-chief for Probe—Business Management, a reviewing editor for Journal of Physical Education and Sport Management, Heliyon, Journal of Hospitality Management and Tourism, and Cogent Business & Management, and the editorial board members for Journal of Business Administration Research, SCIREA Journal of Sociology, International Journal of Tourism & Hotel Business Management, New Construction Technology, JResLit Journal of Science and Technology, International Journal of Tourism & Hotel Business Management, International Journal of Business and Systems Research, Probe—Environmental and Waste Management, Probe—Business Management (OA Journal), Environments (OA Journal), Sensors (OA Journal), Foods (OA Journal), Composite Materials Research, Environment and Social Psychology, Sustainability (OA Journal), Cognitive Neuroscience Journal, Tourism Analysis, International Business Research, Journal of Tourism & Hospitality, International Journal of Business and Management, International Journal of Marketing Studies, Journal of Hospitality and Tourism Research, Science Journal of Business and Management, and Business and Management Studies. His research interests include tourism economics, green marketing, hospitality management, service quality, customer satisfaction, experiential marketing, and consumer behavior.

References

Aguiar-Conraria

Azevedo

Soares

M. J.

2008. “Using Wavelets to Decompose the Time–Frequency Effects of Monetary Policy.” Physica A: Statistical Mechanics and Its Applications 387 (12): 2863–78.

Aguiar-Conraria

Soares

M. J.

2011. “Oil and the Macroeconomy: Using Wavelets to Analyze Old Issues.” Empirical Economics 40 (3): 645–55.

Aguiar-Conraria

Soares

M. J.

2014. “The Continuous Wavelet Transform: Moving beyond Uni- and Bivariate Analysis.” Journal of Economic Surveys 28 (2): 344–75.

Aguiló

Alegre

Sard

2005. “The Persistence of the Sun and Sand Tourism Model.” Tourism Management 26 (2): 219–31.

Alegre

Mateo

Pou

2013. “Tourism Participation and Expenditure by Spanish Households: The Effects of the Economic Crisis and Unemployment.” Tourism Management 39:37–49.

Andraz

J. M.

Rodrigues

P. M.

2016. “Monitoring Tourism Flows and Destination Management: Empirical Evidence for Portugal.” Tourism Management 56:1–7.

Arouri

Estay

Rault

Roubaud

2016. “Economic Policy Uncertainty and Stock Markets: Long-Run Evidence from the US.” Finance Research Letters 18:136–41.

Baker

S. R.

Bloom

Davis

S. J.

2016. “Measuring Economic Policy Uncertainty.” Quarterly Journal of Economics 131 (4): 1593–636.

Birău

F. R.

2014. “The Impact of Global Financial Crisis on Tourism Development in European Union Member Countries—An Integrated Approach.” Journal of Management and Social Science 1 (1): 70–82.

10.

Bloomfield

D. S.

McAteer

R. J.

Lites

B. W.

Judge

P. G.

Mathioudakis

Keenan

F. P.

2004. “Wavelet Phase Coherence Analysis: Application to a Quiet-SUN Magnetic Element.” Astrophysical Journal 617 (1): 623–32.

11.

Bodosca

S. L.

2015. “Sustainable Decision Making under the Crisis for Tourism Businesses: A Survey of the Literature for EU Member States.” Ecoforum Journal 4 (1): 127–31.

12.

Brito

S. P.

2014. “Portuguese Tourism–Facing the Crisis.” Worldwide Hospitality and Tourism Themes 6 (5): 442–51.

13.

Chen

M. H.

2007. “Macro and Non-macro Explanatory Factors of Chinese Hotel Stock Returns.” International Journal of Hospitality Management 26 (4): 991–1004.

14.

Costa

Gomes

Montenegro

2014. “Did the Context of Economic Crisis Affect the Image of Portugal as a Tourist Destination? Strategic Question Overview.” Worldwide Hospitality and Tourism Themes 6 (5): 392–96.

15.

del Mar Alonso-Almeida

Bremser

2013. “Strategic Responses of the Spanish Hospitality Sector to the Financial Crisis.” International Journal of Hospitality Management 32:141–48.

16.

Frechtling

D. C.

1982. “Tourism Trends and the Business Cycle: Tourism in Recession.” Tourism Management 3 (4): 285–90.

17.

Goffe

1994. “Wavelets in Macroeconomics: An Introduction.” In Computational Techniques for Econometrics and Economic Analysis, edited by Belsley

, 137–49. Boston, MA: Kluwer Academic.

18.

Graham

Nikkinen

2011. “Co-movement of Finnish and International Stock Markets: A Wavelet Analysis.” European Journal of Finance 17 (5/6): 409–25.

19.

Grinsted

Moore

J. C.

Jevrejeva

2004. “Application of the Cross Wavelet Transform and Wavelet Coherence to Geophysical Time Series.” Nonlinear Process Geophysics 11 (5/6): 561–66.

20.

Liow

H. K.

2012. “Co-movements and Correlations across Asian Securitized Real Estate and Stock Markets.” Real Estate Economics 40 (1): 97–129.

21.

Loh

2013. “Co-movement of Asia-Pacific with European and US Stock Market Returns: A Cross-Time-Frequency Analysis.” Research in International Business and Finance 29:1–13.

22.

Maraun

Kurths

2004. “Cross Wavelet Analysis: Significance Testing and Pitfalls.” Nonlinear Processes in Geophysics 11 (4): 505–14.

23.

McCarthy

Orlov

A. G.

2012. “Time-Frequency Analysis of Crude Oil and S&P500 Futures Contracts.” Quantitative Finance 12 (12): 1893–908.

24.

Perles-Ribes

J. F.

Ramón-Rodríguez

A. B.

Rubia-Serrano

Moreno-Izquierdo

2016. “Economic Crisis and Tourism Competitiveness in Spain: Permanent Effects or Transitory Shocks?” Current Issues in Tourism 19 (12): 1210–234.

25.

Ramsey

Lampart

1998a. “Decomposition of Economic Relationships by Time Scale Using Wavelets: Money and Income.” Macroeconomic Dynamics 2 (1): 49–71.

26.

Ramsey

Lampart

1998b. “The Decomposition of Economic Relationships by Time Scale Using Wavelets: Expenditure and Income.” Studies in Nonlinear Dynamics and Econometrics 3 (1): 23–42.

27.

Roueff

Von Sachs

2011. “Locally Stationary Long Memory Estimation.” Stochastic Processes and Their Appli-cations 121 (4): 813–44.

28.

Rua

Nunes

L. C.

2009. “International Co-movement of Stock Market Returns: A Wavelet Analysis.” Journal of Empirical Finance 16 (4): 632–39.

29.

Sanuy

1983. “El Turismo Y La Crisis Economica.” Estudios Turisticos 80:67–76.

30.

Schulmeister

1979. Tourism and the Business Cycle. Vienna, Austria: Austrian Institute for Economic Research.

31.

Sharif

Saha

Loganathan

2017. “Does Tourism Sustain the Economic Growth? A Wavelet Based Evidence from United States.” Tourism Analysis 22 (4): 467–82.

32.

Song

Dwyer

Zheng

2012. “Tourism Economics Research: A Review and Assessment.” Annals of Tourism Research 39 (3): 1653–682.

33.

Song

Lin

Witt

S. F.

Zhang

2011. “Impact of Financial/Economic Crisis on Demand for Hotel Rooms in Hong Kong.” Tourism Management 32 (1): 172–86.

34.

Tiwari

A. K.

Mutascu

Andries

A. M.

2013. “Decomposing Time-Frequency Relationship between Producer Price and Consumer Price Indices in Romania through Wavelet Analysis.” Economic Modelling 31:151–59.

35.

Torrence

Compo

1998. “A Practical Guide to Wavelet Analysis.” Bulletin of the American Meteorological Society 79 (1): 61–78.

36.

Vera

1994. El Cobre En El Centro De La Política. Santiago de Chile, Chile: CESOC.

37.

T. P.

H. C.

2019a. “A Multiple and Partial Wavelet Analysis of the Economic Policy Uncertainty and Tourism Nexus in BRIC.” Current Issues in Tourism 23 (7): 906–16.

38.

T. P.

H. C.

2019b. “Causality between European Economic Policy Uncertainty and Tourism Using Wavelet-Based Approaches.” Journal of Travel Research 58 (8): 1347–356.

39.

T. P.

H. C.

Chu

C. C.

Deng

2019. “Tourism and House Prices: A Wavelet Analysis.” Journal of China Tourism Research. Published online November 1. doi:10.1080/19388160.2019.1683111

40.

Zhou

2010. “Comovement of International Real Estate Securities Returns: A Wavelet Analysis.” Journal of Property Research 27 (4): 357–73.