Asymmetric effects of exchange rate on bilateral tourism trade balance: evidence from Turkey

Abstract

As an emerging economy, Turkey has experienced a significant structural transformation in the last twenty years. Within this period, Turkey has increased to the eleventh rank for international tourist arrivals in the world (UNWTO 2023). However, in recent years, there has been a rapid depreciation of the national currency, raising the question of how this affects the tourism trade. Using bilateral data and considering non-linearities, we investigate the exchange rate-tourism balance nexus for Turkey and its most tourist sender partners by applying ARDL and NARDL modelling. We find that (i) when the Turkish Lira (TL) appreciates, the tourism balance is positively affected in the German and UK models in the short-run and in the Russian and Netherlands models in the long-run; (ii) the depreciation of the TL, on the other hand, has a positive effect on the tourism balances for France and the UK in the short-run. Therefore, either the increase or decrease in the exchange rate (except for Russia in the short-run) affects the tourism balance insignificantly or positively. These findings imply that policymakers should focus more on country-specific tourism policies to attract more tourists to the country rather than on exchange rate policies to manage the tourism balance.

KEYWORDS:

• Tourism
• trade balance
• exchange rate
• NARDL

JEL Classifications:

• C22
• F13
• F14
• F31

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1 In the empirical literature, there are also some studies which investigate the impact of income and real exchange rate on tourism demand (e.g., Ongan, Işik, and Özdemir 2017; Irandoust 2019; Işık, Sirakaya-Turk, and Ongan 2020).

2 There are also some studies that examine the determinants of tourism outflows or tourism inflows of Turkey (e.g., Uysal and Crompton 1984; Halicioglu 2010; Ulucak, Yücel, and İlkay 2020). However, since the focus of this study is the relationship between the tourism trade balance and the exchange rate for Turkey, these studies are excluded from the literature review part of this study. In addition, Kiliç and Bayar (2014) analyze the impact of real effective exchange rate volatility on tourism receipts and expenditures in Turkey. However, this study is again out of scope for our research question.

3 The data set is available from the authors upon request.

4 As noted above, the bilateral real exchange rate of the Turkish Lira (TL) against the currency of trading partner $i$’s is computed as $\mathrm{CP}{I}_{\mathrm{TR}}\ast \mathrm{NEX}/\mathrm{CP}{I}_i$, where $\mathrm{CP}{I}_{\mathrm{TR}}$ is the Turkish consumer price index ($\mathrm{CP}{I}_{\mathrm{TR}}$), $\mathrm{CP}{I}_i$ is the CPI of country $i$, and NEX is the nominal bilateral exchange rate expressed as the number of TL per unit of trade partner’s currency. Therefore, an increase in REX implies a depreciation in the home currency. In order to facilitate the interpretation of the findings, the tourism trade balance is defined as the rate of inbound tourist arrivals / outbound tourist departures.

5 We report the results in Appendix Table A1 in order to save space. No variable has an I(2) process in any of our models.

6 The depreciation of the TL, which started in 2012 and reached 30% levels by the middle of 2013, made the international tourism more expensive. Hence, international tourism demand of Turkey's citizens decreased in general (TURSAB, 2014). Therefore, there has been a serious decrease in the number of Turkish citizens who want to visit some foreign countries from Turkey. For this reason, a dummy variable is added for the UK, the Netherlands, and France models in these periods. On the other hand, dummy variables are added to the 1st and 2nd quartiles in the model due to the decrease in the number of Russian tourists coming to Turkey because of the negative impact of the terrorist attacks in Istanbul and Ankara.

7 Note that the ARDL estimates of the models are provided in the Appendix.