Blood and Ink! The Common-Interest-Game
Between Terrorists and the Media

Bruno S. Frey† Dominic Rohner‡
(10 March 2006)

Abstract:

It has often been pointed out in the literature that a symbiotic relationship exists between terrorist groups and the media. As yet, however, no formal model has been built based on this issue and only very little empirical research has been done in this field. The present contribution builds a simple game theoretic model, focussing on the social interactions between terrorists and the media. The model has features of a common-interest-game and results in multiple equilibria. After a discussion of the policy implications of the model, an empirical analysis is performed. Using newspaper coverage, terror incidents and terror fatalities data, it is shown that media attention and terrorism do mutually Granger cause each other, as predicted by the model. Moreover, it is explained why terror attacks tend to be
“bloodier” in developing countries than in Europe and the United States.

1. Terrorism and the Media Are Symbiotic

Recent history has provided plenty of examples of mutually beneficial relationships
between terrorist organisations and the media. The hostage taking by Palestinian terrorists at the 1972 Munich Olympics, the hijacking of TWA flight 847 by Lebanese terrorists in 1985, or the terrorist attack on New York’s twin towers on 9/11/2001 were all mediated megaevents, where terrorists deliberately wanted the attention of the public and where the media benefited from record sales and huge audiences. The more recent terrorist attacks on public transport services in Madrid in 2004 and in London in 2005 also follow the same pattern.

It appears that political extremists employ terror as a communication strategy, and that they deliberately choose their targets and their timing in order to maximise media attention.

Most of the time, terror attacks take place in big cities with a high density of press agencies.
Similarly, terrorists tend to attack before or during big media events, such as elections, international summits like the G8-summit or the Olympic Games. As once expressed by a leader of the terrorist organisation “United Red Army”: “There is no other way for us. Violent actions … are shocking. We want to shock people, everywhere … . It is our way of communicating with the people” (see McKnight, 1974: 168).

Obviously, the media also benefit from the public’s eagerness to obtain information about terrorist attacks. At least for sensationalist TV channels and tabloid-newspapers, the fear and fascination generated by terrorism and political extremism is a substantial part of their business.

The relationship between terrorists and the media has received little attention in the literature. Among the scholars focussing on this issue, almost all agree that a symbiotic relationship exists between terrorists and the media. Several contributions have discussed this symbiotic relationship qualitatively with the help of case study evidence (Frey, 1988; Hoffman, 1998; Wilkinson, 2000; Frey, 2004).
Very few econometric studies have been performed specifically focussing on whether the media actually increase the risk of terrorism. An interesting article by Nelson and Scott (1992) assesses empirically whether media coverage causes terrorism for the time period 1968-1984, and arrives at the conclusion that this is not the case. In the empirical part of our contribution, we will test, using more recent data, whether this conclusion still holds in today’s more globalised and media-covered world. Another important empirical paper has been written by Schbley (2004), who analyses the impact of media on the propensity of Muslim zealots for terrorism. Analysing interviews with 2619 individuals, it is found that a lot of know-how on how to organise and execute a terrorist attack is transmitted by the media.

Some other articles put emphasis on related phenomena, such as the psychological impact of the media coverage of terrorism on the public (Slone, 2000) and on the inaccuracy and nonrepresentativity of the terrorism coverage of certain media (Delli Carpini and Williams, 1987; Crelinsten, 1997; Gentzkow and Shapiro, 2004).
However, it is striking that, even though anecdotal and statistical evidence seem to
suggest that there is a link between terrorism and the media, almost no theoretical research has been done on this issue. One exception is an article by Scott (2001). But he focuses on the competition amongst different terrorist groups and does not consider the explicit interaction between media and terrorists.

We intend to fill this gap in the theoretical literature by building a simple game-theoretic model, which will have the main features of a coordination-game, or, more exactly, a common-interest-game. As the symbiotic relationship between terrorists and media is a particular form of social interaction, a game-theoretic model is more appropriate than a traditional microeconomics model with independently maximising agents. In section 2 of our paper, a simple model of the relationship between terrorism and the media is built. Section 3 is devoted to comparative statistics, section 4 performs an empirical analysis and section 5 concludes.

http://www.iew.unizh.ch/wp/iewwp285.pdf