Modelling the interaction between tourism, energy consumption, pollutant emissions and urbanization: renewed evidence from panel VAR

In less than two decades, the global tourism industry has overtaken the construction industry as one of the biggest polluters, accounting for up to 8% of global greenhouse gas emissions as reported by the United National World Trade Organization (UNWTO2018). This position resonates the consensus of the United Nations Framework Convention on Climate Change (UNFCCC). Consequently, research into the causal link between emissions and the tourism industry has increased significantly focusing extensively on top earners from the industry. However, few studies have thoroughly assessed this relationship for small island economies that are highly dependent on tourism. Hence, this study assessed the causal relationship between CO(2)emissions, real GDP per capita (RGDP) and the tourism industry. The analysis is conducted for seven tourism-dependent countries for the period 1995 to 2014 using panel VAR approach, with support from fully modified ordinary least square and pooled mean group-autoregressive distributed lag models. Unit root tests confirm that all variables are stationary at first difference. Our VAR Granger causality/block exogeneity Wald test results show a unidirectional causality flowing from tourism to CO(2)emission, RGDP and energy consumption, but a bi-directional causality exists between tourism and urbanization. This implies that in countries that depend on tourism, the behaviour of CO(2)emission, RGDP and energy consumption can be predicted by the volume of tourist arrivals, but not the other way around. The impulse response analysis also shows that the responses of tourism to shocks in CO(2)appear negative within the 1st year, positive within the 2nd and 3rd years but revert to equilibrium in the fourth year. Finally, the reaction of tourism to shocks in energy consumption is similar to its reaction to shocks in RGDP. Tourism responds positively to shocks in urbanization throughout the periods. These outcomes were resonated by the Dumitrescu and Hurlin causality analysis where the growth-induced tourism hypothesis is validated as well as feedback causality observed between tourism and pollutant emission and urbanization and pollutant emission in the blocks over the sampled period. Consequently, this study draws pertinent energy and tourism policy implications for sustainable tourism on the panel over their growth trajectory without compromise for green environment.