The asymmetric impact of real exchange rate on tourism demand

Yalcin, Yeliz; Arikan, Cengiz; Köse, Nezir Article The asymmetric impact of real exchange rate on tourism demand The European Journal of Comparative Economics (EJCE) Provided in Cooperation with: University Carlo Cattaneo (LIUC), Castellanza Suggested Citation: Yalcin, Yeliz; Arikan, Cengiz; Köse, Nezir (2021) : The asymmetric impact of real exchange rate on tourism demand, The European Journal of Comparative Economics (EJCE), ISSN 1824-2979, University Carlo Cattaneo (LIUC), Castellanza, Vol. 18, Iss. 2, pp. 251-266, https://doi.org/10.25428/1824-2979/005 This Version is available at: https://hdl.handle.net/10419/320172 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nd/4.0/ The European Journal of Comparative Economics Vol. 18, no. 2, pp. 251-266 ISSN 1824-2979 http://dx.doi.org/10.25428/1824-2979/005 The asymmetric impact of real exchange rate on tourism demand Yeliz Yalcin*, Cengiz Arikan**, Nezir Kose*** Abstract Europe is a notable tourism region and so international tourist arrivals are getting more crucial day by day for attraction center countries. Besides many economic factors, exchange rate is also main economic determining factor of tourism demand. This paper investigates the asymmetric effects of the real exchange rate on tourism demand by utilizing asymmetric VAR methodology for 10 most popular destinations in Europe. According to empirical results, there is a negative relationship between the real exchange rate and tourism demand, with mixed effects for a few countries. The effect of the currency appreciation on the total number of tourist arrivals is more greatly than the currency depreciation for France, Netherlands, Poland and Turkey. Austria, Greece and Italy are also affected asymmetrically from the currency rate in the long term but not short term. The tourist arrivals in Spain, Germany and the UK are not asymmetrically sensitive to exchange rate. The results show that decrease in exchange rate have greater impact on the tourism demand compared to increase in the exchange rate in asymmetrically affected countries. JEL Classification: Z32, C32, F3 Keywords: Real exchange rate, Tourism demand, Asymmetric effect, VAR 1. Introduction Tourism is the fifth sector in the World’s exports sector after petroleum, chemical, food, and automotive industries. International tourist arrivals grew by 5 percent in 2016 and more than half of it was welcomed in European countries (World Tourism Organization Annual Report, 2016). Tourism is one of the main income source of global economy. Therefore, the countries invest in this sector to get big portion of the global tourism income. Tourism has an important role in providing the capital required for the development of countries. The reason why tourism sector is considered important is its foreign currency earning feature. On the other hand, tourism income has an essential function in the balance of payments due to being foreign exchange source in tourism countries (Uguz ve Topbaş, 2011). * Department of Econometrics Ankara Hacı Bayram Veli University Ankara, Turkey ** Corresponding author: C.Ari[email protected]; Republic of Turkey Ministry of Trade Ankara, Turkey *** Department of Economics and Finance İstanbul Gelisim University İstanbul, Turkey The views expressed are those of the authors and do not reflect the stance of Republic of Turkey Ministry of Trade. EJCE, vol. 18, no. 2 (2021) Available online at http://eaces.liuc.it 252 Investigating the tourism economics is substantial for tourism countries. According to the researches results countries take precautions to prevent tourism demand. Undoubtedly, there are many economic or social factors that determine the demand for tourism. One of the main economic factors affecting the international tourism is exchange rate and its volatility. Although there are many studies on the impacts of exchange rate and its volatility on tourism demand, there are varying results. A large part of studies has found that the exchange rate is statistically significant for tourism demand. However, according to some empirical studies, the exchange rate has no effect on tourism demand (Vanegas and Croes, 2000; Croes and Vanegas, 2005; Quadri and Zheng, 2011). The reason for these varying results is that different countries and different time periods have been used in these studies. The exchange rate may affect both the choice of destination for international tourists and the length of stay and expenses (Webber, 2001; Wang, et al., 2008; Crouch, 1993). According to several findings, an increase in destination country’s exchange rate causes a decrease in total number of tourists. Therefore, the exchange rate is an important indicator for international tourists as the cost of living (Martin and Witt, 1988). Previous studies focus on the symmetric impact of the exchange rate on the tourism demand but there is no consensus on the effect of real exchange rate on tourism demand. Asymmetric effect means that the increases and decreases in real exchange rate may not have same magnitude on the number of tourist arrivals. One of the reasons for conflict results may be that the effect of exchange rate on tourism demand is not symmetric. There is no example that have examined the asymmetry in terms of dynamic structure in the literature. To fill this gap, in this study the asymmetric VAR and nonlinear impulse-responses are used. This paper aims to investigate the asymmetric effects of the changes in real exchange rate on tourism demand by using asymmetric VAR proposed by Kilian and Vigfusson’s (2011) model for 10 most popular destinations in Europe using monthly data. Our study has three main contributions. First, this is the first article to investigate the asymmetric effect of the currency depreciation and appreciation on tourism demand by using nonlinear impulse responses unlike the literature. Second, it captures how the negative shock and the positive shock in currency affect the tourism demand in both Y. Yalcin, C. Arikan, N. Kose, The asymmetric impact of real exchange rate on tourism demand Available online at http://eaces.liuc.it 253 short and long run. Third, it is also the first study to examine how the tourism demands in top ten destinations in the Europe are affected by their currency rate. This paper proceeds as follows: Section 2 presents methodology. In Section 3, the data set and empirical results are discussed, and conclusions are given in Section 4. 2. Literature There are many studies on the impacts of the exchange rate and its volatility on tourism demand. Our literature review focuses on the studies investigate the effect of exchange rate on tourism demand. Vita and Kyaw (2013) found that the exchange rate and its volatility are important determinants on Turkey’s tourist arrivals from Germany by using GARCH model for the period 1996-2009. Falk (2015) showed that the impact of the exchange rate on tourism demand with Swiss application. According to his results, after the 2008 global crisis, the appreciation of the Swiss franc led Swiss tourists to go to neighboring Austria. Demir (2004) found that the appreciation of foreign currency via the high inflation rate decreased in the number of tourist. Moreover, Akar (2012) showed that the exchange rate has significantly affected Turkish tourism demand from the Eurozone and the US. Pavlic, et.al (2015) concluded that there is a long-run relationship between tourist arrivals and real effective exchange rate for Croatia. Martins et.al (2017) investigated the world tourism demand by using 218 countries’ income per capita, the nominal exchange rate and relative price for the period 1995-2012. According to their findings, arrivals in Europe are sensitive to the exchange rate and tourism demand in Africa is the most affected by macroeconomic variables. Lim and Zhu (2017) analyzed the determinants of the tourism demand growth for Singapore by using heterogeneous dynamic panel. They concluded that the Singapore tourism is not sensitive to the real exchange rate. Chi (2015) examined the impact of income and exchange rate on the US tourism for the period 1960-2011. According to findings that the real exchange rate has a significant role on tourism balance for both long and short run. Croes and Vanegas (2005) examined that the effect of price and exchange rate on the tourist arrivals to Aruba from the US, Netherlands and Venezuela by using Box-Cox transformation. They found that the effects of price and exchange rate depend on the country from which the tourists come from. Chang and Mcaleer (2012) investigated the same relationship for the tourist arrivals from the world, the US and Japan to Taiwan. EJCE, vol. 18, no. 2 (2021) Available online at http://eaces.liuc.it 254 According to their results, the exchange rate has a negative impact on tourist arrivals to Taiwan. According to Quadri and Zheng (2011)’s results, the exchange rate has no impact on Italian tourism demand. Reason for conflict results on the effect of exchange rate on tourism demand may be that the effect is not symmetric. Asymmetric effect means that the increases and decreases in real exchange rate may not have same magnitude on the number of tourist arrivals. Previous studies assume that the magnitude of these effects are same. In the literature, generally, the asymmetric impact of exchange rate volatility has been examined instead of the exchange rate. Also, these studies have used EGARCH or GJRGARCH models for asymmetry (Chang and McAleer, 2009; Daniel and Rodrigues, 2010; Demirel et al, 2013). There are limited articles, Wang, et al. (2008) and Tang, et al. (2016), investigating the asymmetric relationship between the currency rate and tourism demand and they used copula-GARCH model. However, neither these nor other studies have examined the asymmetry in terms of dynamic structure. 3. Methodology Hypothesis of this study is the increases and decreases in real exchange rate have not same magnitude on the number of tourist arrivals. In order to investigate to presence of an asymmetric effect of exchange rate on tourism demand, the asymmetric VAR model, developed by Kilian and Vigfusson (2009 and 2011) is employed. In the asymmetric VAR model, the first equation is identical to the first equation of the standard linear VAR model. REERt= α10+∑α1𝑖𝑇𝑁𝑇t-i 𝑝 𝑖=1 +∑α2𝑖𝑅𝐸𝐸𝑅t-i 𝑝 𝑖=1 +ε1𝑡 (1) where REER is real effective exchange rate and TNT is total number of tourist arrivals. However, the asymmetric VAR model differs from standard linear VAR model in that the second equation includes both REERt and REERt +, in other words, both exchange rate increases, and decreases affect tourism demand. Therefore, the second equation is given as follows Y. Yalcin, C. Arikan, N. Kose, The asymmetric impact of real exchange rate on tourism demand Available online at http://eaces.liuc.it 255 TNTt= β10+∑β1𝑖𝑇𝑁𝑇t-i 𝑝 𝑖=1 +∑β2𝑖𝑅𝐸𝐸𝑅t-i 𝑝 𝑖=1 +∑g2iREERt-i + 𝑝 𝑖=1 +ε2𝑡 (2) where εit (i=1,2) are error terms with uncorrelated white noise (0,Σ). Here REERt + is a censored variable under given threshold value. The threshold value can be estimated by using Chan (1993) or can be taken as zero. 𝛼𝑗𝑖 𝑎𝑛𝑑 𝛽𝑗𝑖 (j=1,2; i=1,…,p) are the coefficients of REER and TNT and g2i (i=1,…,p) are the coefficients of censored variable. Since the OLS residuals of equations (1) and (2) are uncorrelated, these coefficients are estimated by standard regression method. Although the parameter estimates are not efficient asymptotically, the advantage of this model is that the dynamics responses are estimated consistently without knowing the nature of the DGP (Kilian and Vigfusson, 2011). If there is an asymmetric effect the coefficients of REERt + should be zero or the impulse responses should be equal for two regimes. Therefore, there are two ways which are slope-based test and impulse response-based test to test whether there is an asymmetric effect or not. If increases and decreases in real exchange rate have same magnitude effects on tourism demand, the slope coefficients or slope line must be symmetry. For testing symmetry in the changes of real exchange rate, the null hypothesis can be defined as follows H0: g21,0=…=g21,p=0 (3) For testing this hypothesis, without require any properties Wald test is used and it has an asymptotic χp+1 2distribution. If the null hypothesis is rejected, it means that the impulse responses are asymmetry. However, rejecting this hypothesis does not give any idea about the direction of deviation from symmetry and level of the asymmetry. Since the impulse responses of this model are non-linear, the degree of asymmetry and presence of deviation from symmetry is statistically significant can be tested by impulse response-based test. Because of nonlinear VAR model, the impulse responses are obtained by history dependent method which is called generalized impulse responses (Gallant, et al., 1993 and Koop, et al., 1996). For testing of symmetric responses to positive and negative exchange rate shocks to H period, the null hypothesis is given as follows EJCE, vol. 18, no. 2 (2021) Available online at http://eaces.liuc.it 256 H0: Iy(h, δ) =- Iy(h, -δ) (4) where Iy(h, δ) and Iy(h, -δ) are responses of TNTt at horizon h=0,1,2,. . .,H to a shock of positive or negative real exchange rate shocks. Wald test of this hypothesis has an asymptotic 𝜒𝐻+1 2 distribution. 4. Data and Empirical Results In this study to examine asymmetric effects of exchange rate on tourism demand, total number of tourist and real effective exchange rate are collected from Eurostat and World Bank for the 10 most popular destinations in Europe. These countries and their availability periods are shown in Table 1. Table 1: The 10 Most Popular EU Countries* and Data Periods Variables From To Austria 1990.1 2017.6 France 2011.1 2017.5 Germany 1990.1 2017.5 Greece 1995.1 2017.4 Italy 1990.1 2017.4 Netherlands 1990.1 2017.6 Poland 2003.1 2017.6 Spain 1990.1 2017.6 Turkey 2003.1 2017.3 UK 1994.1 2016.12 *According to UNWTO Tourism Highlights, 2016 Edition Since the total number of tourism series have seasonal pattern, they are adjusted with Tramo Seats method. All variables are used in logarithm form. In order to check that variables are stationary, Augmented Dickey-Fuller (ADF, 1981) and Phillips Perron (PP, 1988) unit root tests are used. The ADF and PP tests results for the model with constant and constant and trend are given in Table 2. Y. Yalcin, C. Arikan, N. Kose, The asymmetric impact of real exchange rate on tourism demand Available online at http://eaces.liuc.it 257 Table 2: Unit Root Tests Results Data Country ADF PP Level First Difference Level First Difference Constant Constant and Trend Constant Constant Constant and Trend Constant Number of Tourist Austria 2.577 -1.533 -5.376*** -1.411 -10.870*** 78.599*** France -1.045 -4.057** -8.779*** -6.041*** -8.610*** -40.308*** Germany 1.135 -2.195 -15.462*** 1.536 -2.222 -26.486*** Greece -0.081 -2.426 -15.632*** -0.065 -3.754** -32.901*** Italy 0.118 -3.680** -19.908*** -0.240 -12.721*** -59.034*** Netherlands 1.271 -0.728 -19.650*** 0.707 -4.064*** -41.328*** Poland -0.270 -1.524 -15.855*** -0.918 -3.991** -32.173*** Spain 0.650 -1.921 -25.024*** 0.604 -2.319 -25.323*** Turkey -2.012 -2.403 -7.680*** -1.927 -2.149 -14.519*** UK -0.453 -1.980 -12.284*** -2.305 -4.871*** -30.310*** REER Austria -2.330 -2.354 -14.926*** -1.919 -1.915 -14.844*** France -1.073 -1.669 -7.268*** -1.135 -1.914 -7.237*** Germany -1.442 -2.494 -14.078*** -1.219 -2.204 -13.901*** Greece -1.375 -0.956 -14.057*** -1.475 -1.096 -13.968*** Italy -3.010** -2.890 -8.461*** -2.423 -2.319 -13.416*** Netherlands -2.528 -2.555 -13.610*** -2.141 -1.866 -13.380*** Poland -2.664* -2.606 -9.285*** -2.411 -2.350 -9.264*** Spain -1.484 -1.702 -13.653*** -1.299 -1.527 -13.837*** Turkey -2.431 -3.209* -9.918*** -2.611 -3.125 -9.750*** UK -2.070 -1.905 -13.124*** -2.019 -1.837 -13.134*** *,**,*** statistically significant at the 10% , 5%, 1% level, respectively The results show that, two variables for all countries are nonstationary at 5% significance level. Therefore, the asymmetric VAR model, equations (1) and (2) has been built up for each country’s real exchange rate and tourism demand with first differences, separately. Since tourism demand has no impact on REER determination, lags of REER have excluded from equation (1). So the estimation model is given as follows: ∆REERt= α10+∑α2𝑖∆REER p i=1 t-i+ε1𝑡 ∆TNTt= β10+∑β1𝑖∆TNT p i=1 t-i+∑β2𝑖∆REER p i=1 t-i+∑g2i∆𝑅𝐸𝐸𝑅t-i ++ p i=0 ε2𝑡 (5) where the lag length of the model, p, is determined by Akaike Information criteria (AIC) as 6. In here, the censored variable is: ∆REERt +={ ∆REERt , ∆REERt > 0 0 , ∆REERt ≤ 0 (6) EJCE, vol. 18, no. 2 (2021) Available online at http://eaces.liuc.it 258 Since the slope based test does not give any idea about the direction of deviation from symmetry and level of the asymmetry, the impulse-response-based test is used for testing whether there is an asymmetric effect of exchange rate on tourism demand. After estimating model (3.1) for each country, the generalized impulse responses are gathered by using Kilian and Vigfussion (2011) algorithm for structural analysis. The impulse responses of each country tourism demand when one standard deviation positive and negative shock is given for REER are plotted out. Figure 1 reports the history dependent impulse responses for eight periods for ten most popular destinations in Europe. In order to make some statistical inference for the impulse response analysis, confidence intervals at 95% level based on the bootstrap simulation with 500 trials are calculated. In each graph, the red straight line is for the response of TNT to a negative exchange rate shock (depreciation in exchange rate), the black straight line is for the response of TNT to a positive exchange rate shock (appreciation in exchange rate). To compare the impulse responses to both positive and negative shocks, the negative shock’s impulse responses are drawn in absolute (mirror images). The dotted lines represent the confidence bands. Figure1: Impulse Response Graphs Effects of REER on Austria's Tourisim Demand Y. Yalcin, C. Arikan, N. Kose, The asymmetric impact of real exchange rate on tourism demand Available online at http://eaces.liuc.it 265 References Akar C. (2012), ‘Modeling Turkish Tourism Demand and the Exchange Rate: the Bivariate GARCH Approach’, European Journal of Economics, Finance and Administrative Science, 50, 133-141 Chan K. S. (1993). ‘Consistency and limiting distribution of the least squares estimator of a threshold autoregressive model’, The Annals of Statistics, 21, 520–533 Chang C. L., Mcaleer M. (2009), ‘Daily Tourist Arrivals, Exchange Rates and Volatility for Korea and Taiwan’, CARF Working Paper http://www.carf.e.u-tokyo.ac.jp/pdf/workingpaper/fseries/199.pdf Access Date: 31.01.2018 Chang C. L., Mcaleer M. (2012), ‘Aggregation, Heterogeneous Autoregression and Volatility of Daily International Tourist Arrivals and Exchange Rates’, The Japanese Economic Review, 63(3), 397-419 Chi J. (2015), ‘Dynamic Impacts of Income and The Exchange Rate on US Tourism, 1960– 2011’, Tourism Economics, 2015, 21(5), 1047–1060 Croes R., Vanegas M. (2005), ‘An Econometric Study of Tourist Arrivals in Aruba and its Implications’, Tourism Management, 26, 879-890. Crouch G. I. (1993), ‘Currency Exchange Rates and the Demand for International Tourism’, Journal of Tourism Studies, 4(2), 45-53 Daniel A. C., Rodrigues P. M. M. (2010), ‘Volatility and Seasonality of Tourism Demand in Portugal’, Banco de Portugal Articles, https://www.bportugal.pt/sites/default/files/anexos/papers/ab201003_e.pdf Access Date: 31.01.2018 Demir C. (2004), ‘How Do Monetary Operations Impact Tourism Demand? The Case of Turkey’, International Journal of Tourism Research, 6, 113–117 Demirel B. et al. (2013), ‘The Impact of Exchange Rate Volatility on Tourism Sector: A Case Study, Turkey’, Niğde Üniversitesi İİBF Journal, 2013, 6(2),117-126 Dickey D. A, Fuller W. A. (1981), ‘Likelihood Ratio Statistics for Autoregressive Time Series with a Unit Root’, Econometrica, 49(4), 1057-1072. Falk M. (2015), ‘The Sensitivity of Tourism Demand to Exchange Rate Changes: An Application to Swiss Overnight Stays in Austrian Mountain Villages During the Winter Season’, Current Issues in Tourism, 18(5), 465-476 Gallant A., Rossi P., Tauchen G. (1993), ‘Nonlinear Dynamic Structures’, Econometrica, 61, 871- 907 Kilian L., Vigfusson R. (2009), ‘Pitfalls in Estimating Asymmetric Effects of Energy Price Shocks’. CEPR Working paper, 7284 Kilian L., Vigfusson R. (2011), ‘Are the Responses of the U.S. Economy Asymmetric in Energy Price Increases and Decreases?’, Quantitative Economics, 2, 419-453 Koop G., Pesaran M. H., Potter S. M. (1996), ‘Impulse Response Analysis in Nonlinear Multivariate Models’, Journal of Econometrics 74(1), 119–147 Lim C., Zhu L. (2017), ‘Dynamic Heterogeneous Panel Data Analysis of Tourism Demand for Singapore’, Journal of Travel and Tourism Marketing, 34(9), 1224 -1234 Martin C. A., Witt S. F. (1988), ‘Substitute Prices in Models of Tourism Demand’, Tourism Annals of Tourism Research, 15(2), 255-268. Martins L. F., Gani Y., Lopes A.F. (2017), ‘An Empirical Analysis of The Influence of Macroeconomic Determinants on World Tourism Demand’, Tourism Management, 61, 248-260 Pavlic I., Svilokos T., Tolic M. S. (2015), ‘Tourism, Real Effective Exchange Rate and Economic Growth: Empirical Evidence for Croatia’, International Journal of Tourism Research, 17, 282–291 EJCE, vol. 18, no. 2 (2021) Available online at http://eaces.liuc.it 266 Phillips P. C. B., Perron P. (1988), ‘Testing for a Unit Root in Time Series Regression’, Biometrika, 75: 335–346 Quadri D.L., Zheng T. (2011), ‘A Revisit to the Impact of Exchange Rates on Tourism Demand: The Case of Italy’, Journal of Hospitality Financial Management, 18 (2-4) Tang J. et al. (2016), ‘Modelling Dependence Between Tourism Demand and Exchange Rate using The Copula-Based GARCH Model’, Current Issues in Tourism, 19(9), 876-894 Uguz S. C., Topbas F. (2011), ‘Döviz Kuru Oynaklığı Turizm Talebi İlişkisi’, Paper Presented at EconAnadolu 2011: Anadolu International Conference in Economics II June 15-17, 2011, s. 1-11, Eskişehir, Turkey Vanegas M., Croes R. (2000), ‘Evaluation of Demand, US Tourist to Aruba’, Annals of Tourism Research, 27(4), 946-963. Vita G.D., Kyaw K.S. (2013), ‘Role of the Exchange Rate in Tourism Demand’, Annals of Tourism Research, 43, 624-627 Wang H. C. et al. (2008), ‘Tourism demand and exchange rates in Asian countries: New evidence from Copulas approach’, In ICCIT'08. Third International Conference on Convergence and Hybrid Information Technology, 1188-1193. Webber A. G. (2001), ‘Exchange Rate Volatility and Cointegration in Tourism Demand’, Journal of Travel Research, 39(4), 398-405. Eurostat Database http://ec.europa.eu/eurostat/web/tourism/data/database Access Date: 22.08.2017 World Bank Database https://data.worldbank.org/indicator/PX.REX.REER Access Date: 22.08.2017 World Economic Forum, The Travel and Tourism Competitiveness Report 2017 http://ev.am/sites/default/files/WEF_TTCR_2017.pdf Access Date: 31.01.2018 World Tourism Organization, Annual Report, 2016 http://media.unwto.org/publication/unwto-annual-report-2016 Access Date: 31.01.2018 World Tourism Organization, UNWTO Tourism Highlights, 2017 https://www.eunwto.org/doi/pdf/10.18111/9789284419029 Access Date: 31.01.20180 World Tourism Organization, UNWTO Tourism Highlights, 2015 http://mkt.unwto.org/publication/unwto-tourism-highlights-2016-edition Access Date: 31.01.2018