Regional Tourism Cooperation And Competition Based On Network Structure

The tourism industry, which is the world’s largest industry today, has become aneffective entry point and an important breakthrough for the economic boom andsustainable development in many countries and regions. The regional tourismcooperation and competition has been paid great attention to an unprecedented heightby both academic scholars and industrial practitioners. Although it has been a hottopic to adopt social network analysis method and complex network theory to studythe system of cooperation-competition in the academic field recently, few scholarsconduct researches on regional tourism cooperation-competition system from theperspective of network structure.Combining the complex network theory, social network analysis approach, neweconomic sociology theory, new economic geography theory, new institutionaleconomics theory and borrowing ideas from the latest research findings in the field ofcooperation-competition economic relation networks, the present study proposed ananalysis method of regional tourism cooperation-competition network from fiveaspects of network graph model construction, structural feature index measurementand analysis, relevance analysis, evolution model and cooperation-competitionequilibrium mathematic model. An empirical study on Chinese inbound tourismsystem was conducted to test the validity of the analysis method and models proposed.The present study built a binary graph model of regional tourism cooperation-competition system with special features, in which regions and tourists were taken asnodes, relations between regions and tourists as edges. The project degree of regionsrepresented the regions’ market share or competition competence. The geometricmethod was adopted to project the bipartite graph to regional nodes, which led to asingle-model network of regional tourism cooperation. In the project graph, thenumber of tourists from one region to another made an edge representing thecooperation relationship between two regions. This network was not only a weightednetwork, but also a directed multiple network with self-loop. The central degree ofregions calculated by the method of DEMATEL represented the regions’ cooperationstrength.Using the data of Chinese inbound tourist flows and the number of inboundtourists in every regions in2005, the present study analyzed the main statistical properties of the network in terms of clustering coefficient, the length of the shortestpath, the distribution of project degree, the distribution of edge weight, core-periphery structure, structural hole, etc. The main findings showed that:①thenetwork had a big clustering coefficient and a short characteristic path length whichindicated the characteristics of a small world；②both the edge weight distribution andthe project degree size ranking distribution emerged as a power law which suggested ascale-free network；③the network has a typical core-periphery structure and thedensity between the core members was much higher than both the density between theperiphery members and the density between the core members and the peripherymembers；④the regions’ advantages of structure hole was not consistent with theirproject degree and central degree.The present study adopted the improved DEMATEL method to analyze therelevance of the Chinese inbound tourism network from four aspects of influencedegree, influenced degree, central degree and reason degree. Compared with therelevance indexes used in the previous social network analysis study, the centraldegree and reason degree were the more effective indicators for the position and roleof nodes in the network. The central degree presented the cooperation strength of eachnode by taking the influence degree, influenced degree into consideration. The reasondegree reported the role of each node in the relevance network. Based on the data ofcentral degree and reason degree, a cause and effect diagram of the network wasdrawn. According to this diagram, the author divided the whole network into fourareas named potential, improvement, advancement and strengths. It was followed byoptimization and upgrading strategies for each area.Based on the regional tourism collaboration-competition network bipartite graphwith two kinds of nodes of region and tourist, the author constructed an evolutionmodel of regional tourism collaboration-competition networks which was based onattract preferential attachment to describe how the accumulative number of tourists ina region grew and evolved. Then the author used SPSS software to run a regressionanalysis on the simulation data and the actual data by adopting the data of Chineseinbound tourism system from the year of2005to2009to verify the correctness of themodel. The result was F=1.005×104> F1,29(0.05)=4.18, which implicated that theattract preferential attachment was the key evolution mechanism of tourismcollaboration-competition networks.In the end, through a qualitative study on private property and fiscaldecentralization system and polymerization behavior, the author built up a regional tourism cooperation-competition equilibrium mathematic model based on PrivateProperty Rights and Fiscal Decentralization institution. The model showed that withthe precondition of private property system and fiscal decentralization system, due tothe structural constraint of “the rich get richer” resulting from the evolutionmechanism of attracting preferential attachment, the government and privateinvestors would surely invest more money in the areas with original preferenceleading to the effective tourism industrial clustering. Chinese inbound tourismindustrial cluster was a good example to indicate that private property system, fiscaldecentralization system and the structural constraint of “the rich get richer” were theimportant reasons for tourism industrial clustering.