A New Test of the Tourism-Led Growth Hypothesis Using Unconditional Quantile Regression (UQR) with Panel Data

Abstract

This study has attempted to sketch out testing Tourism-led Growth Hypothesis (TLGH) from 1995 to 2020 for 47 major countries worldwide. A confounder approach has been utilized in model specification, while Unconditional Quantile Regression (UQR) and Recentered Influence Function (RIF)-Oaxaca method are used for regressing and decomposing the factors. Unlike other studies used the income level as the quantile, this study employed the Pritchett (2000)’s different economic growth patterns for each country as a new quantile, then double quantile regression has been performed.As a result, first, we could reject the null-hypothesis that tourism growth has a positive impact on economic growth using tourism receipts variable, which the most of the previous literature used. Second, it is interesting to note that central government debt plays a big negative role in the group of low-growth countries, but not in the countries with high economic growth rates. Third, it is also worth noting that excessively high exchange rate (depreciation of the currency) might have a negative impact on economic growth rate in the low-growth patterns’ countries. Finally, total exports has a positive impact on economic growth only in the groups of the high growth countries. Moreover, this paper provides a policy implication from the decomposition results that the difference of means in growth rates between groups can not be explained by the variables used, but they siginificantly rely on country-specific effects.