How important is tourism for growth?

We revisit the tourism-led growth hypothesis by utilising a panel set of 108 countries over the period 1996–2017. We quantify the effects of tourism on the entire conditional distribution of economic growth for both relatively poor and relatively rich countries within a panel quantile regression framework. We address the unobserved heterogeneity and potential endogeneity concerns. We reveal that the lower the conditional growth rate a country experiences the more important is tourism development for the conditional growth distribution for both developing and developed countries. The size of the effect in developed countries is twice as high as in developing ones. On the other hand, tourism specialisation is beneficial only at higher quantiles of the conditional growth distribution and only for developed countries. On the contrary, it brings about an undesirable effect in developing countries. Finally, we examine the impact of a reduction in tourism activity on economic growth due to an exogenous shock (i.e., COVID-19). Simulation analysis based on the quantile regression estimates shows that countries facing relatively low growth rates conditionally to the growth distribution are affected the most. Policymakers may consider the importance of tourism activity in the growth process and formulate strategies that align with the growth experience of each country.