Inbound international tourism to Canary Islands: a dynamic panel data model

Abstract

This paper presents a dynamic model of international tourism demand to Canary Islands. The empirical analysis exploits the panel structure of the dataset (for the 15 most important countries of origin of tourists over the period 1992–2002) by a Generalized Method of Moments estimation of a dynamic model taking into account unobserved country-specific effects. The preferred model is the GMM-DIFF proposed by Arellano and Bond, however other models are also shown for comparison. A dynamic model where the lagged dependent variable is included as regressor is used to obtain short-run and long-run elasticities. Problems arising from the non-stationarity of data are dealt with by using a model in first differences. The estimated coefficient for the lagged depended variable is always significant and may reflect a high degree of consumer loyalty or an important effect of word of mouth in determining demand of international tourism. In brief, the results suggests that tourism demand to Canary Islands must be considered as a luxury good and is highly dependent on the evolution of relative prices and cost of travel between origin and destination country.