The changing distribution of global tourism: Evidence from gini coefficients and markov matrixes

R. Geoffrey Lacher, Sanjay K. Nepal

    Research output: Contribution to journalArticlepeer-review

    9 Scopus citations


    This article examines the global distribution of tourism arrivals over 1995-2008 to determine whether there is a pattern of concentration or dispersal of tourist arrivals at a global scale, and then predicts the possible future distribution of global tourists arrival based on changes in those years. The study employs Gini coefficients and a Markov matrix to international arrival data in 153 countries for the period between 1995 and 2008. The Gini coefficient is used to measure the dispersion of total international tourist arrivals (ITA) in each country. Results show that the Gini coefficient has decreased over time (i.e., the distribution is gradually dispersed but the overall pattern remains unchanged). Using the same data, Markov matrix is used to predict the future distribution based on changes over the 14-year period. These results suggest future dispersion of international tourist arrivals would be somewhat different than it is today but the overall dominance of the leading countries (i.e., those with high arrival numbers) will continue. The implication is that the leading countries must develop strategies to continue to remain competitive, as other less visited countries make stronger efforts to promote tourism to counterbalance the current imbalance in international arrivals.

    Original languageEnglish (US)
    Pages (from-to)133-144
    Number of pages12
    JournalTourism Analysis
    Issue number2
    StatePublished - 2013


    • Economic geography
    • Gini coefficient
    • International tourist arrivals
    • Markov matrix

    ASJC Scopus subject areas

    • Tourism, Leisure and Hospitality Management


    Dive into the research topics of 'The changing distribution of global tourism: Evidence from gini coefficients and markov matrixes'. Together they form a unique fingerprint.

    Cite this