Tourism development and economic growth - a nonlinear approach

We use cross sectional data (1995-2005 yearly averages) for 88 countries to investigate the nonlinear relationship between tourism development and economic growth when a threshold variable is used. The degree of tourism specialization (q(i), defined as receipts from international tourism as a percentage of GDP) is used as the threshold variable. The results of the tests for nonlinearity indicate that the 88 countries' data should be separated into three different groups or regimes to analyze the tourism-growth nexus. The results of the threshold regression show that when the qi is below 4.0488% (regime 1, 57 countries) or above 4.7337% (regime 3, 23 countries), there exists a significantly positive relationship between tourism growth and economic growth. However, when the q(i) is above 4.0488% and below 4.7337% (regime 2, 8 countries), we are unable to find evidence of such a significant relationship. Further in-depth analysis reveals that relatively low ratios of the value added of the service industry to GDP, and the forested area per country area are able to explain why we are unable to find a significant relationship between these two variables in regime 2's countries. (C) 2008 Elsevier B.V. All rights reserved.