Tourism is configured as a sector of great importance for the economy of many countries. This activity started to show signs of recovery from the economic crisis since 2009 worldwide. As Europe is the main tourist area, this article focuses on its regions (NUTS 2) with two objectives: firstly, we would like to identify the factors that have contributed to tourism recovery in the different European regions, depending on the factors listed in literature about this topic; and secondly, we would like to know the distribution of these factors in the territory in order to check the existence of different groups of regions in terms of the determinants of tourism demand.
Most of the authors who have tried to identify the determinants of tourism demand conclude that there are big groups or categories composed by a big amount of variables. Despite the absence of a global consensus on them, it has been observed a great similarity and uniformity among the different contributions. Thus, the determinants of tourism demand generally accepted in the literature can be summarized in factors of different types: economic, political, marketing and production-related, infrastructural, environmental, tourist facilities-related, customer-related and random. This paper focuses on the analysis of economic, infrastructural and tourist facilities-related variables.
In total, we have obtained nine explanatory variables of the tourism phenomenon (Tourist arrivals, tourist intensity, cultural goods shops, territorial accessibility, number of hotels and similar establishments, HIPC, exchange rate, disposable income per capita and GVA growth rate).
In order to achieve these objectives, firstly we have made several estimations by means of panel data. This will allow us, in some way, to establish the influence that the above-mentioned variables have exerted on tourist’s arrivals. Secondly, a cluster analysis has been applied using Howard-Harris algorithm, so as to find out in which way the determinants of tourism demand are grouped in the territory. Probably not all of them are present with the same relevance in the different regions. Finally, we have also applied the methodology developed by Ivanov and Webster (2007), in order to measure the contribution of tourism to economic growth.
We have obtained the following conclusions among others:
· Not all the factors identified as determinants of tourism demand are present in the European regions with the same importance. Thus, it was possible to establish different groups of classification of European regions in terms of the factors that influenced the recovery of tourist demand since 2009, distinguishing these areas:
o 23 regions of which most are traditional tourist areas, located primarily in the Mediterranean area. They have the greatest number of tourist arrivals and the highest average values in the variables directly linked to tourism.
o Hungary may be seen as another group with its own characteristics, apart from other European regions. This is a territory very affected by the crisis, and it does not stand out among other European regions in matter of tourism.
o 109 regions concentrated in Cluster 3 and a similar amount of it in Cluster 4. The greater part of European territories is concentrated in both clusters, including those which are the most developed and have the greatest economic and political relevance in the European Union (UK and the most of Germany and France). Economic and infrastructural variables stand out mainly in these regions.
o 17 regions belonging to five countries in Eastern Europe: Estonia, Latvia, Lithuania, Bulgaria and Romania. These countries show the lowest number of arrivals during the crisis on average. This may be due to a lower level of accessibility in comparison to other regions as well as their high levels of consumer prices.