Berrebi and Lakdawalla's "How Does Terrorism Risk Vary Across Space and Time?" focuses on Israel's experience, but it can be applied to the question of why al-Qaida hasn't attacked the United States since 9/11. The "space" referred to in the title is the particular site chosen for a terrorist attack. Proximity to the terrorist headquarters and easy access to an international border are, Berrebi and Lakdawalla write, hugely important in deciding whether to carry out a terrorist attack. "When distance to a terrorist home base doubles," they calculate, "the frequency of attacks falls by around 30 percent." Areas near international borders "are more than twice as likely to be hit" as areas far from international borders. Following this logic, Israel is a veritable paradise for Islamist terrorists. It's located right in the Middle East, and, east to west, it's only 85 miles wide. The United States, by contrast, is a jihadi's worst nightmare: halfway across the world and 3,000 miles wide. The two countries' comparative experience with terrorism reflects these realities.
The "time" referred to in the title is the interval between attacks, and it's a lot less comforting to people who live in the United States. To simplify things as we spoke in a RAND conference room, Berrebi drew a rudimentary graph on a whiteboard. The vertical axis represented the risk of attack in a regional capital. The horizontal axis represented the passage of time. It looked like this:
The risk of attack increases sharply before an attack occurs (duh), then falls, then levels off and eventually starts to rise again. I mentioned to Berrebi and Lakdawalla that my girlfriend had travelled to New Delhi a couple of weeks after the Mumbai terror attack and that I'd worried about her. Guess I shouldn't have worried! Berrebi shook his head and tapped on the line just to the right of the first peak. The risk immediately after an attack, he said, is still very high, because it takes the authorities some time to figure out what's happening and to beef up security—and because the terrorists' planned mission may not be completed. Gradually, security measures are put in place. Then gradually, these security measures slacken, creating new opportunities for attack. In Jerusalem, Berrebi and Lakdawalla found that after a terror attack the risk of a follow-up attack begins to increase after only two incident-free months. "This suggests," they conclude, "that long periods of quiet actually indicate elevated risk for sensitive areas." Berrebi and Lakdawalla are restating the familiar war-movie cliché in which two soldiers stand guard over a peaceful nighttime landscape. "It's quiet," says one. "Yeah," says the other. "Too quiet." Then the enemy emits a battle cry and the fighting begins.
In Jerusalem, it's two months from peak to trough. In the United States, it's X months, where X has an unknown value greater than 44. * If the United States suffers another major terror attack, Berrebi and Lakdawalla will be able to calculate the value of X. They think they'll probably get that second data point. They just don't know when.