Effective Factors of Tourism Agglomeration in Selected Developing Countries

Rapid growth of tourism industry after 1950s is one of the main and most important characteristics of tourism industry.But according to the statistics of World Tourism Organization, tourism activities and its incomes are not distributed equally across the world. It can certainly be said that the share of developed countries in tourism income is higher than the developing countries (WTO,2012).Interestingly, the distribution of tourism among developing countries is non-uniform. What are the determinants of tourism agglomeration?To answer this question, it can be said that geographical variables such as economics of scale are some of the factors among these. Thus, the distribution of tourism activities in the framework of new economic geography (NEG) is considered.In a new economic geography framework (NEG), firms tend to agglomerate where the large regional marketsare (Chen et. al, 2008). In terms of distribution of tourism activities, it seems that in addition to factors such as accessing sea, appropriate climate, historical sites in countries with a high tourism agglomeration, accessing larger markets, economies of scale, tourism costs and development level are important. Regarding the lack of equal distribution of tourism activities in selected developing countries,assessing determinants of the tourism agglomeration is essential. In this research, in order to examine the hypothesis and to estimate the model, the Econometric Method for panel data is used for 67 selected developing countries (based on data availability) from 1995 to 2010. The model we use adopts the following form: Where, is tourism agglomeration (the number of tourists in the studied country to the total global tourists); is the relative economies of scale (relative market size); is the tourism cost (relative consumer price index); is the relative development level (relative human development index) and is the behavioral expectations and habits previous time tourism agglomeration), in s country. Discussion and The results of the unit root test indicated that none of the studied variables were on stationary level and all of them would be stationary through making one time difference. But, based on Kao-Cointegration Test, the hypothesis based on lack of agglomeration is rejected and non-spurious regression is approved. Then in order to evaluate the Panel model, first according to the statistics of F-Limer, a selection is done between Panel data and Pooled data methods. Regarding the F-Limer test, Panel data can be used in the evaluating process. After making sure that model evaluating is carried out as Panel data, the most important question that remains is to finding out whether sectional effects are fixed or random. According to the results of Hausman test, the fixed effects model is accepted as the evaluating model. Regarding the results obtained from F-Limer and Hausman tests, the pattern is estimated by using the fixed effects model. According to results, regression adjusted determination coefficient is equal to %99; thus, independent variables describe %99 of dependent variables changes. Also, common F test reflects that all regression is meaningful. According to the obtained results from this study, coefficients of Variables are meaningful and work in accordance with theory. According to the results of this research, a relative amount of economics of scale has a direct and meaningful relationship with tourism agglomeration. The relative cost of tourism has a meaningful and negative effect on tourism agglomeration. Also, development has positive and meaningful effect on tourism agglomeration in selected countries. So the economic and social development of countries will increase tourism agglomeration. In addition, expectations and behavioral habits have positive and meaningful effects on tourism agglomeration in selected countries.