On the asymmetric link between exchange rate variability and tourism inflows: recent evidence from the asean-5 countries

ABSTRACT

We shed new light on the long- and short-run asymmetric impact of exchange rate on tourism inflows over the period of 1995–2019 for selected ASEAN countries (Indonesia, Malaysia, Philippines, Singapore and Thailand; ASEAN-5). The investigation utilized Shin et al. (2014) recently developed NARDL model and the Toda and Yamamoto (1995) causality test. The asymmetric effect was introduced via decomposing the real exchange rate into positive and negative innovations. The key findings show that long-run relationship was found for all countries, when incorporating structural breaks. Additionally, findings support asymmetric long-run effect of exchange rate for Indonesia, Malaysia and Singapore. Finally, the Toda and Yamamoto causality test revealed diverse findings regarding the exchange rate- tourism inflows nexus.

KEYWORDS:

• Tourism
• exchange rate
• ASEAN
• cointegration
• NARDL

Disclosure statement

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

Notes

1 Since World Bank data does not show the origin of the tourists (origin country), in this paper we use world real GDP instead of the income of origin country.

2 Although many studies use effective exchange rate, but due to data availability we use real exchange rate as effective exchange rate was not available for two countries of the sample. Martin and Witt (1987) show that exchange rates adjusted by consumer price indices can be a suitable measure of the cost of tourism.

3 We follow Meo et al. (2018) in adding inflation rate as a separate independent variable.

4 By the same token, $GD{P}_t$, $P_t$, $CO{2}_t$ and $IN{F}_t$ were decomposed into positive and negative innovations

5 To get further information about this test, consult Toda and Yamamoto (1995).

6 See Zapata and Rambaldi (1997), Wolde-Rufael (2005) for more details

7 See Kisswani et al. (2020) for more explanation about the method and its advantages over other alternative methods.

8 Error Correction Model: $\begin{array}{rl}& \mathrm{\Delta }{N}_t={\alpha }_0+\sum _{i=1}^n{\delta }_{1i}\mathrm{\Delta }{N}_{t-i}+\sum _{i=0}^n{\delta }_{2i}\mathrm{\Delta }E{X}_{t-1}+\sum _{i=0}^n{\delta }_{3i}\mathrm{\Delta }GD{P}_{t-1}+\sum _{i=0}^n{\delta }_{4i}\mathrm{\Delta }{P}_{t-1}+\\ & \sum _{i=0}^n{\delta }_{5i}\mathrm{\Delta }CO{2}_{t-1}+\sum _{i=0}^n{\delta }_{6i}\mathrm{\Delta }IN{F}_{t-1}+{\theta }_1{N}_{t-1}+{\theta }_{2}E{X}_{t-1}+{\theta }_{3}GD{P}_{t-1}+{\theta }_4{P}_{t-1}+{\theta }_{5}CO{2}_{t-1}+{\theta }_{6}IN{F}_{t-1}+u_t\end{array}$

9 For more details, see Bai and Perron (1998)

10 Our analysis focuses on the asymmetric effect of the exchange rate. Nonetheless, Table 6 reports the long-run asymmetry for all independent variables.

11 Given space limitations, we have only reported the short-run effects of the exchange rate. For the other independent variables, the short-run results are not reported here but are available upon request.