The nature of the relationship between international tourism and international trade: the case of German imports of Spanish wine

Abstract

This article deals with the relationship between international trade and tourism. We focus on the effect that German tourism to Spain has on German imports of Spanish wine. Due to the different stochastic properties of the series under analysis, which display different orders of integration, a methodology is used based on long memory regression models, where tourism is supposed to be exogenous. The period covered is January 1998 to November 2004. The results show that tourism has an effect on wine imports that lasts between 2 and 9 months. Disaggregating the imports across the different types of wine, it is observed that only for red wines from Navarra, Penedús and Valdepeñas, and to a certain extent for sparkling wine, tourism produces an effect on future import demand. From a policy-making perspective our results imply that the impact of tourism on the host economy is not only direct and short-term but also oblique and delayed, thus reinforcing the case for tourism as a means for economic development.

Acknowledgement

The second named author acknowledges financial support from a PIUNA Project at the University of Navarra, Pamplona, Spain.

Notes

¹For the purpose of the present article we define an I(0) process {u_t , t = 0, ±1, …} as a covariance stationary process with a spectral density function that is positive and finite at the zero frequency. An I(1) process is then defined as a process that requires first differences to get I(0) stationarity.

²As an alternative approach to model the seasonal structure, and in line with the methodology used in the article, we could have removed the seasonal component using (seasonal) fractional integration techniques (Gil-Alana and Robinson, 2001; Gil-Alana, 2002). However, since the wine import series does not present evidence of seasonality that approach would be invalid in a regression model relating the two variables (Section IV).

³See Baillie (1996) for an interesting review of I(d) models.

⁴At the other end, if d < 0, x_t is said to be ‘anti-persistent’, because the spectral density function is dominated by high frequency components. See Mandelbrot (1977).

⁵This means that the test is the most efficient one when directed against local alternatives. In other words, if we direct the tests against the alternative: H_a : d = d_o  + δT ^−1/2, the limit distribution is normal, with variance 1 and mean that cannot be exceeded in absolute value by any rival regular statistic.

⁶The periodogram is an asymptotic unbiased estimate of the spectral density function f(λ). If a series is I(0), 0 < f(0) < ∞, and if it is overdifferenced, f(0) = 0. Thus, the periodogram should mimic that behaviour.

⁷As an alternative definition of German tourism in Spain, we also employed the number of nights spent in Spanish hotels by German travellers. However, the series presents a similar pattern as the one used in the present article.

⁸The first two methods, Dickey and Fuller (1979) and Phillips and Perron (1988) tests the null hypothesis of a unit root (I(1)) against the alternative of stationarity, while Kwiatkowski et al . (1992) tests the null of stationarity against the alternative of a unit root.

⁹This is a nonparametric approach of modelling, the I(0) disturbances that produce autocorrelations decaying exponentially as in the AR(MA) case. An empirical application of this procedure can be found in Gil-Alana and Robinson (1997).

¹⁰Note that the LM procedure of Robinson (1994) is based on the Whittle function, which is an approximation to the likelihood function.

¹¹ p refers to the number of parameters required to describe the short-run dynamics. Other values of p were also employed and the results were very similar to those reported in the article with p = 1.

¹²We conducted some tests for exogeneity of tourism in the wine imports equation. To establish evidence for noncausality, an unrestricted VAR was used. Weak exogeneity appeared to be satisfied in the dynamic equation because when entering the current value of DT in the equation it is proved to be insignificantly different from zero. This finding supports the view that DT is weakly exogenous for the model.

¹³Alternatively we could have employed a multivariate system for all wine categories. However, in the context of fractional processes, multivariate models are not yet fully theoretically justified.

¹⁴One possible reason for the 5-months delayed reaction of imports to tourism flows in the case of sparkling wine and Penedús red could be that they may be goods in particular demand for New Year's Eve and Christmas, respectively. Contrary to the other analysed wine types, these two import series peak at around November while the tourism series’ main peak is in May, thus pointing to a possible link for some wine types between the choice of festivity drinks and a recent holiday destination.

¹⁵The numbers are the average (mean) of the estimates given in Tables 5 and 7. The interpretation of the estimates for the monthly growth rates series is not directly comparable to the ones obtained from the DST series, therefore, they have not been included in the summary calculation of Table 8.

¹⁶Most accurately one would have used the series of German tourists travelling to specific Spanish areas rather than the overall German-tourists-travelling-to-Spain series for the estimation of the β-parameters. However, such regional-travellers series were not available and it can be assumed that their inter-temporal variance should be highly correlated to the one of the national-travellers series, thus resulting in similar parameter estimates.