Analysis of tourism demand using a multi-dimensional panel gravity model

Purpose: This study aims to estimate the determinants of international tourist flows between destinations by using the panel gravity model. Design/methodology/approach: The multi-dimensional panel gravity model was used to analyse tourism originating from 30 different countries to the 14 most-visited countries in the world between 2008 and 2016. Income (i.e. per capita gross domestic product for both the origin and destination countries), distance between countries, various economic indicators and six dummy variables were added to the gravity model as control variables. Findings: The results indicated that tourist arrivals depended mainly on economic factors, i.e. income and trade variables were significant determinants of tourist arrivals. The results also suggested that estimated international tourist flows are a negative function of distance, as is postulated in economic theory. Originality/value: In recent years, gravity models have been used frequently to analyse international tourism demand and have demonstrated their ability to evaluate the effects of various determinants of international tourism for many countries. The literature includes studies that used a two-dimensional panel gravity model to analyse the determinants of tourism demand to a single country from many different countries. This study differs in terms of specificity; in that, it relied on a three-dimensional panel gravity model that allowed for modelling of multiple destination countries. As a result, more comprehensive and general results relative to the determinants of tourism demand were obtained. In addition, the application of a non-nested three-dimensional panel data model, which has limited use, contributes a new perspective to the econometric literature.