Using a seemingly unrelated regression (SUR) model, this paper examines the impact of domestic and transnational terrorism on tourism demand (as measured by visitor arrivals) to Lebanon, Turkey and Israel. The distinguishing feature of this paper is that we use a dynamic framework that takes into account (i) the intensity of terrorism and (ii) the associated spillover effects. The empirical results, based on monthly data from 1995 to 2007, reveal that terrorism in one country affects visitor arrivals not only to that country but also to other countries. We find that tourists view Lebanese and Turkish tourism markets as substitutes. Terrorism in Israel, of all types (i.e. domestic or transnational) and intensities, affects visitor arrivals to Lebanon and Turkey. In overall terms, terrorism has a much larger negative impact on visitor arrivals to Lebanon.

This paper uses a seemingly unrelated regression model (SUR) to test the individual effects of domestic and transnational terrorism on tourism demand to Lebanon, Turkey and Israel over the period 1995–2007. Tourism demand is measured by the logarithm of the number of arrivals to each country. Moreover, this paper tests whether tourism depends on the magnitude of the terrorist attacks by disaggregating terrorism into three levels of intensity – low, medium and high. The results show significant own and spillover effects for domestic and transnational terrorism on tourism demand to each of the three selected countries. In addition, the results show that the effect of terrorism on tourism depends on the intensities of the terrorist attacks.

Purpose: – The purpose of this paper is to study the direction of the causality between tourism development and economic growth in Lebanon between 1995 and 2013, after taking into consideration terrorist incidents and their intensities. These are considered as exogenous shocks that affect tourism development and economic growth instantaneously and with a lag.Design/methodology/approach: – To reach the objectives, the authors estimate a vector auto regressive model with exogenous variables, applying a series of unit root tests with and without structural breaks and the Granger causality test.Findings: – The findings suggest a positive unidirectional causality running from tourism development to economic growth in the short run. Thus, the authors find evidence for the tourism-led growth hypothesis (TLGH) in Lebanon despite the exposure of the country to frequent terrorist incidents. The impulse response functions reveal that tourism development (economic growth) responds positively to a positive shock to economic growth (tourism development).Practical implications: – The findings call for Lebanese policy makers aiming at promoting growth to design policies that encourage tourism, such as implementing tourism marketing policies and building the needed tourism infrastructure. Such policies will have positive but transitory effects on economic growth. The findings may also be useful for regional representatives of intergovernmental organizations and the offices of statistics of United Nations World Tourism Organization and the World Bank to better understand the tourism industry in Lebanon and similar countries suffering from instabilities.Originality/value: – This paper contributes to the existing literature in three points: despite the importance of the tourism industry to the Lebanese economy, this topic did not receive careful attention in the literature; it takes into consideration the presence of structural breaks and possible nonlinearities in the number of tourist arrivals; and it investigates the TLGH after accounting for instability in the country.

The literature on gravity equations has recently shown that flows between two countries are not merely impacted by their structural bilateral determinants but also affected by their relative accessibility vis-à-vis other countries. The idea can be extended to tourism analysis: bilateral tourism flows are not only determined by factors affecting the attractiveness of the destination country but also depend on the attractiveness of alternative destinations. This effect can be termed ‘multilateral resistance to tourism’ (MRT) and embodied in cross-sectional correlation across the main tourism determinants. In this paper, we show that the standard estimation techniques applied in tourism analysis fail to properly account for the MRT, yielding biased estimates and nonstationary residuals. We cast the gravity model into the common factor framework and employ the common correlated effects (CCE) estimator to obtain unbiased estimates of some standard determinants of tourism flows in the context of 35 OECD members during 1995–2015. Our CCE results show that the impact of a number of regressors becomes negligible and insignificant, highlighting their role as vectors of cross-section dependence. Our findings stress the need for analyzing the data to detect the magnitude of the cross-sectional dependence and, when the latter is omnipresent, employing MRT-robust estimations.