TERRORISM AND TOURISM: THE CASE OF TURKEY

Abstract

Using a time series method called ‘transfer function’, this paper examines the effect of terrorism on tourism in Turkey. The results indicate that there exists a negative but small impact of terrorism, which is observed within approximately one year. However, terrorist attacks in Turkey have accounted for a reduction of six million foreign tourists over the last nine years. Moreover, the economic cost of terrorism in the tourism industry was more than $700 million in 2006. However, terrorist attacks in continental Europe and America, and an active war involving a neighboring country, had no effects on tourism in Turkey.

Keywords:

• Transfer function
• Turkey
• Terrorism
• Tourism
• Economic impact
• 9/11
• Second Persian Gulf War

JEL Classifications:

• C32
• D74
• F51

ACKNOWLEDGEMENTS

The author wishes to thank Dr Walter Enders, Dr Henry Thompson, Dr Christian Thomann, session participants of the 78th Southern Economic Association Meetings, and anonymous referees for their helpful comments. Any errors are the author's.

Notes

¹Literature has mixed results on the issue. Abadie (2006) found no significant effect of the ethnic fractionalization on the overall level of terrorism risk while Santos Bravo and Mendes Dias (2006) found negative relationship between the ethnic fractionalization and the number of terrorist attacks in their cross‐sectional regression analysis.

²Due to the violence and bloodshed, Turkey has announced an ‘Emergency State Governance’ in the southeastern region, which is still in effect in the region.

³Becker (1968) and Rubbelke (2005) examined the motivations of crime and terrorism with a marginal benefit/marginal cost analysis that can be applied to terrorist acts. Moreover, Sandler et al. (1983) modeled a negotiation model between the terrorists and the government. However, such modeling and the causes of the terrorism in Turkey is beyond the scope of this paper.

⁴The details of these incidents can found at the archives of Turkish daily newspapers such as Hurriyet.

⁵Becker and Murphy (2001) have also acknowledged the negative impact of terrorism on tourism.

⁶Tourism Strategy of Turkey 2023, by Ministry of Culture & Tourism. Available online at http://www.kulturturizm.gov.tr/genel/text/eng/TST2023.pdf.

⁷Indeed, Enders and Sandler (1991) and Ozsoy and Sahin (2006) found that the causality between terrorism and tourism is unidirectional.

⁸Karagol and Palaz (2004), Gunluk‐Senesen (2004), and Yildirim and Sezgin (2003) have studied the effect of defense spending, which is a direct consequence of terrorism, on various macroeconomic variables such as economic growth, external indebtedness, and employment in Turkey respectively.

⁹Blomberg et al. (2004), Abadie and Gardeazabal (2003).

¹⁰Eckstein and Tsiddon (2004) define the terror index (TER) as the natural log of an index that is equal to one plus the equal weighted sum of the number of incidents, fatalities, and injuries.

¹¹See Abadie and Gardeazabal (2003).

¹²Eckstein and Tsiddon (2004) have also studied the effect of terrorism on tourism in Israel.

¹³Sandler and Enders (2004) highlighted the selection bias of ITERATE due to the inclusion of only ‘newsworthy’ terrorist events.

¹⁴MIPT has only reported the transnational terrorism until 1997 and has started to include domestic terrorism after that date.

¹⁵ AIC = T ln(SSR) + 2n, BIC = T ln(SSR) + n ln(T) where n is the number of parameters estimated, T is number of usable observations (Enders, 1995, 1996).

¹⁶However, DF results for the fatalities, injuries, TER Index are available upon request from the author.

¹⁷Takay‐Araz et al. (2009), on the other hand, found a bidirectional relationship between terror and economic activity using GDP and the Terror Index, which is constructed with the same methodology but from a different data source.

¹⁸These series is provided by MIPT/Rand Terrorism Database that includes both domestic and transnational terrorism at the beginning of 1997.

¹⁹For brevity, the results are not reported here but they are available upon request.

²⁰Information on the tourism revenue per tourist in Turkey is calculated based on the information available at the Turkish Statistical Institute (www.turkstat.gov.tr). The institute collects data on the total number of tourists visiting Turkey in a given year and the amount of money spent by these foreign tourists. Tourist revenue per tourist is simply the ratio of total revenue divided by the total number of tourists.