Co-integration Analysis of the Central and Eastern European Tourism Demand in Tunisia using Non-stationary Panel Data

This paper aims to model the destination choice process of tourists from Central and Eastern Europe using non stationary panel data. If the unit root hypothesis is accepted, co-integration analysis is applied to estimate common long run demand relationships for these origin countries. The behavior of tourists from this part of the world is analyzed through estimated long run demand elasticities. The main finding of this study is that the existence of these relationships is expected given identical geopolitical and historical factors characterizing these countries. In addition to prices and income factors, the supply factor is introduced as an explanatory variable in the econometric model. The cointegrating relationships show that the large elasticity magnitude may be a reflection of tourism in Tunisia being regarded as a luxury by tourists. These results are corroborated by the estimated values of the short run demand elasticities obtained through the dynamic panel data model. They show also that only 1% of the increase in the tourist arrivals is due to tourist loyalty, which reflect the failure of the tourism policy in these markets to create and retain tourist loyalty to Tunisian services.