Spatial heterogeneity estimation of regional tourism industry elasticity in Chinese provinces: a geographically weighted regression study.

The spatial effects of both interdependence and heterogeneity are common in regional tourism growth processes,and this refers to the indirect or unintentional effects that one region's tourism industry exerts on the tourism industries of other regions.As a result,a region's tourism growth can obtain benefits from their neighbors' tourism growth through positive spillover effects.Even though the estimation research on the contributions of capital and labor on outputs in the tourism industry have received a high level of interest internationally and domestically,very few studies have considered spatial effects in theoretical tourism growth analysis and empirical tourism model specifications.The lack of research is especially obvious regarding empirical evidence on the growth of regional tourism industries in China's provinces.At present,the vast majority of empirical studies that estimate and measure tourism elasticity have been based on conditional mean estimates using the ordinary least squares(OLS) method,which ignores the role of spatial effects in model design and empirical studies of the tourism production function.As a result,estimate results of tourism elasticity coefficients lack scientificity and accuracy.As such,the conclusions obtained may not actually accurately represent growth in the tourism industry.Furthermore,policy judgments based on parameter estimation results and conclusions are often questionable.While it is pleasing to see that some studies do consider spatial effects in the tourism growth process,the aggregate production function should be homogeneous to each spatial unit in tourism research,and the parameters estimated using the OLS method are often constant parameters.As a result,it is difficult to measure regional tourism production behavior including spatial correlation and to spatially estimate the heterogeneous production function. In contrast to similar studies that include constant parameters,the present study,in its consideration of the location specificity(i.e.,resource endowment,geographical position,history and culture,market location,tourism accessibility,tourism consumption level,traffic accessibility,and information technology) of China's various regions,is based on sectional average data from the Chinese provincial tourism industry from 2001 to 2009.Using a geographically weighted regression(GWR) production function model of provincial tourism in China,which includes spatial heterogeneity,this study positively estimates and studies local varying elasticity coefficients of capital and labor on tourism industrial growth.The results,based on a spatial autocorrelation analysis of global Moran's I and local indicators of spatial association(LISA) analysis,show that there is a significantly global spatial dependency and an obvious local clustering trend for provincial tourism output.Thus,in the study of the tourism industry,we cannot ignore spatial effects,including both interdependence and heterogeneity.The globally constant parameter estimation results,with and without consideration of spatial autocorrelation in the provincial tourism production function,show that the capital investment of the elasticity coefficient is 0.6939 and 0.7824,and the labor input elasticity coefficients are 0.4932 and 0.3163,respectively,which means each factor input is in a phase of increasing returns to scale.The locally varying parameter estimation results,which consider spatial heterogeneity in the GWR models,show that the elasticity coefficients of capital investment are between 0.6905 and 0.6958,and the labor input elasticity coefficients are between 0.4859 and 0.4989.Thus,each provincial factor input is in a phase of increasing returns to scale.The GWR model can effectively estimate the regional tourism elasticity coefficients under the premise of spatial heterogeneity,which increases the differential policy implications for regional tourism development. In terms of general policy suggestions,the heterogeneity policy implication of this study is very obvious.It is not possible to achieve sustainable tourism development with a simple dependence on capital input growth.Once China's tourism enters the transformation development stage,and this is true for China's tourism development and the current mode of national economic development,the economic development pattern will rapidly transform from increasing quantity to increasing quality.To realize sustainable tourism development in China's provinces,the nation together with each province needs to promote the reform of institutions and the tourism development mechanism,and especially strengthen the reformation of tourism human resources and improve employee skills,taking into account regional heterogeneity.Furthermore,in time,the uniform nature of China's tourism development planning and tourism strategy needs to be adjusted,and tourism development strategies and travel supervision laws and regulations should be developed in accord with an area's regional character,and their regional characteristics be promulgated as soon as possible.