Terrorism and Tourism: a time-series analysis for the case of Israel

This paper will provide an analysis of the influence of terrorism on tourism in Israel. Data from the Global Terrorism Database on the number of terrorist attacks and the Bank of Israel on the number of tourist arrivals have been used to construct a multivariate vector autoregression (VAR) model. The data has been employed in a monthly format from January 2002 until December 2017. After transforming the variables where needed a VAR(2) model including two lags of the number of terrorist attacks and two lags of the number of tourist arrivals is preferred. This model indicates that there is a negative effect of terrorism on tourism and that this can be approximated by a 433 decrease in tourist arrivals compared to the previous month following a month after the attack. The effect is persistent up to 5 months characterized by a bounce-back effect. Furthermore, Granger causality tests indicate a one-sided influence of terrorism on tourism. Robustness tests strengthen the tendency of the result found in the research but point towards the importance of big terroristic events. The economic costs of terrorism should therefore also include any effects on lost tourist visits.