By The Numbers: Europe’s Terror Problem by Daniel Bier, Foundation For Economic Education
Is Terrorism Getting Worse? Or Does It Just Feel That Way?
Last week’s suicide attacks in Brussels claimed 35 innocent lives, injured more than 300, and reignited the debate that has smoldered since the massacre at Charlie Hebdo last January. How big a problem is terrorism? How has it changed over time? Is it getting worse?
I dug into the Global Terrorism Database — an exhaustive database of over 140,000 terrorist attacks worldwide from 1970 to 2014 — to get a picture of terrorism in Europe.
Europe – Is Terrorism on the Rise?
The data show that the threat of terrorism in Western Europe has hugely declined since the 1970s, even accounting for Brussels, Charlie Hebdo, and the Paris massacre last year, which killed 137.
The average deaths per year from terrorism has fallen precipitously since the 1970s and 1980s.
The 2010s have a slight uptick from the Paris attack in November 2015, but attacks like Paris are extremely rare. In 45 years, there have only been two other attacks in Western Europe that killed more than 100 people: the 2004 Madrid train bombings (191 killed) and the 1988 bombing of Pan Am 103, over Lockerbie, Scotland (270 killed).
So it is very likely that the recent uptick will erode as the decade wears on. (And the average for the last decade, 2006-2015, is in fact lower than the average for the 2000s.)
Since 1970, 6,435 people have been killed by terrorist attacks in Western Europe. More than twice as many people are murdered every year in the United States.
Out of nearly 16,000 attacks, 76 percent had zero fatalities; 19 percent had only one death (often the attacker himself). Major attacks, killing ten people or more, comprise just 0.3 percent (about 40 attacks).
Of course, most terrorist attacks are symbolic and not intended to kill anyone — burning a building at night, blowing up a statue, etc. But both lethal and nonlethal attacks have declined.
Rarer, But Deadlier
Terrorism has never been very common in Western Europe, and it has become even less so in recent decades. Both attacks and fatalities have been in steady (if uneven) decline since the 1970s.
But, paradoxically, while terrorism is rarer than ever, it is also deadlier.
Measuring this is tricky; simply dividing deaths by attacks won’t do it, because most attacks aren’t intended to kill. But if we assume that lethal terrorist attacks, which did actually kill people, were meant to kill people, we have a good proxy for terrorist attacks that are intended to kill.
The data shows that fatal terrorist attacks have become increasingly deadly in recent decades. Attacks that kill people are killing more people. So while the total number of attacks and total number of deaths are in decline, the efficiency of terrorists in killing people is increasing.
New Terrorists, New Targets
Why might this be? I hazard two possible explanations: the rise of suicide terrorism and the decline of separatist terrorism.
Suicide terrorism has a higher body count compared to other kinds of terrorism because the attackers aren’t concerned about surviving or getting away with it. This allows brazen, public attacks, like Paris and Charlie Hebdo, where the attackers simply got rifles and walked into a crowd. Attackers who expect to survive and escape can’t execute attacks like this very easily.
Why is suicide terrorism on the rise? It’s become the favorite tactic of Islamist terrorist groups, starting in the Lebanese Civil War, perfected over the Israel-Palestine conflict, taken to deadly new heights by al-Qaeda, and then exploding across Iraq after 2003.
As European countries and the United States have come into conflict with Islamist groups in the Middle East, such as al Qaeda and ISIS, they have used suicide attacks to strike civilian and political targets. The reason why suicide terrorism is particular popular with Islamist terrorist groups is probably related to its strategic logic (they are weak, and suicide attacks are more effective), as well as specific beliefs about jihad and martyrdom.
The second factor is the decline of separatist terrorism in Europe.
Many Western European countries, including Spain, France, Italy, and the UK, have regions with militant separatist groups. For much of the 20th century, these conflicts produced a large number of small but deadly attacks. While most attacks in Western Europe are still related to separatism, these conflicts have long been in decline. Data on arrests for separatist terrorism in recent years show a sharp decline.
This is important because the 1970s and 1980s had much higher total death tolls, but on average fewer fatalities per lethal attack. I speculate that this is because separatist groups are able to carry out many more attacks than Islamist or transnational terrorists, and (for strategic and PR reasons) they are not trying to maximize body counts the way that modern terrorist groups seem to be.
By contrast, attacks sponsored or inspired by the Islamic State and al-Qaeda are calculated for higher body counts. Their supporters view European civilians as an out-group and legitimate targets, and their strategies (for different reasons) call for deadly, high-profile attacks.
Combine these trends and you get fewer but deadlier attacks.
The upshot of this is that Europe is actually getting safer, but — because deadlier attacks are more widely reported, and therefore easier to remember — it will feel like it’s getting more dangerous.
Methodological Notes: The Global Terrorism Database lost all data from 1993, and it has never been restored. Therefore, where annual rates are given for decades, totals for the 1990s have been corrected to divide by 9, instead of 10.
GTB’s last complete year is 2014. Death totals for 2015 have been drawn from news reports, based on Wikipedia’s list of attacks since 2015. It is important to include 2015, where possible, because it had two major attacks in Paris that substantially affect the recent picture. However, this list of attacks is not comprehensive and likely misses some small or nonlethal attacks. Where I look at deaths, I use GTB through 2014 and news reports for 2015. Where I look at attacks and rates per attack, I only use GTB’s complete data, which ends in 2014.