Tag: terrorism

The Chance of Being Murdered or Injured in a Terrorist Attack in the United Kingdom

On Tuesday, a Sudanese immigrant to the United Kingdom named Salih Khater crashed his car into cyclists and pedestrians in a terrorist attack in London. Fortunately, Khater did not murder anybody in his attack but he did injure three pedestrians, one of whom was so lightly wounded that he was treated at the scene and released. The other two wounded people have since been released from the hospital. 

Terrorism has been relatively common in the United Kingdom for decades, from the Irish Republican Army to al Qaeda to ISIS. However, there is little research on the actual risk of a British person being killed or injured in a terrorist attack. This post is an attempt to quantify that risk.

According to data from the Global Terrorism Database at the University of Maryland, the RAND Corporation, and online sources for 2018, terrorists murdered 2,402 people in the United Kingdom from 1975 through August 14, 2018 (Figure 1). Despite increases in the number of murders committed by terrorists in recent years, especially a series of horrible attacks in 2017 that murdered 42 people, the long run trend is a decline in the number of people murdered by terrorists in the United Kingdom. Figure 2 shows that 5,267 people were wounded in terrorist attacks in the United Kingdom during the same period. Figures 1 and 2 only include victims and exclude the terrorists themselves from the death and injury statistics. Using existing data sources, somebody with knowledge and a lot of time could use the GTD and RAND databases to identify the nativity, ideology, and other characteristics of each terrorist like I did for the United States

Figure 1: Annual murders committed by terrorists in the United Kingdom
Figure 2: Annual injuries committed by terrorists in the United Kingdom

From 1975 through August 15, 2018, a British person’s chance of being murdered in a terrorist attack on British soil was about 1 in 1.1 million per year.  But that annual chance of being murdered in a terrorist attack obscures big shifts over time. Over the last decade, the annual chance of being murdered in a terrorist attack on British soil was about 1 in 11.4 million per year, far lower than the entire 1975-2018 period. Especially relevant is the number of injuries given that Khater only injured people in his attack. The annual chance of being injured over the entire time was 1 in 496,464 per year, but only 1 in 1.4 million per year over the last decade. 

Figure 3 tries to show how the risk has changed over time by using a moving three-year average of the annual chance of being murdered in a terrorist attack. I used a three-year moving average because zero people were killed by terrorists in many years and one cannot divide by zero. A note about reading Figures 3 and 4: the Y-axis is the annual chance of being murdered or injured in a terrorist attack, so the 2011 number of 63,280,444 in Figure 3 means that the chance of a British person being murdered in a terrorist attack was 1 in 63,280,444 that year. Thus, the higher the number, the lower the chance of being murdered in a terrorist attack. 

Figure 3 shows that the annual chance of being murdered in a terrorist attack fell rapidly from 1975 through 2004, rose over the next several years, fell again, and has been increasing since about 2012. Dropping the moving gives sharper divides: In 2016, 2017, and 2018 (so far), the annual chance of being murdered in a terrorist attack was about 1 in 7.3 million per year, 1 in 1.8 million per year, and zero (so far), respectively. 

Figure 3: Annual chance of being murdered in a terrorist attack in the United Kingdom

Figure 4 shows a similar decline in injuries inflicted by terrorists from 1975 through 2003 that abruptly reverses in 2004, falls again in the following years, and then starts to increase over the last few years. 

Figure 4: Annual chance of being injured in a terrorist attack in the United Kingdom

These above figures show that the chance of dying or being injured in a terrorist attack in the United Kingdom is small. Yet terrorism succeeds in terrifying people. None of the numbers above would give comfort to the actual victims of terrorism or their families because what happened to them is the equivalent of “winning” an evil anti-lottery. But the above numbers should show British citizens, their government, and the world at large that terrorism is a relatively small problem in the United Kingdom.    

Counterterrorism Spending

The Stimson Center’s new study group report found that the federal government spent about $2.8 trillion on counterterrorism (CT) activities since 9/11.  The report seeks to account for all federal government spending on CT efforts divided into the four broad categories of defense emergency and overseas contingency operations, war-related state/USAID, other foreign aid, and government-wide Homeland Security.  The defense emergency and overseas contingency operations spending category accounts for about $1.7 trillion or over 60 percent of the $2.8 trillion spent.  War-related state/USAID and other foreign aid account for a relatively small $138 billion and $12 billion, respectively.  Government-wide Homeland Security spending makes up the rest at $978.5 billion since 9/11.

The big question the report does not attempt to answer is: Was all that spending worth it?  Did that spending result in fewer people killed by terrorists on U.S. soil?  One of the distinguished study group members is my Cato Institute colleague John Mueller who has spilled much ink trying to estimate the effectiveness of CT spending.  Mueller provides some back of the envelope estimates to answer the question of whether this CT spending was worth it in his recent panel discussion on the Stimson Center’s report.  After talking with Mueller, I decided to add some more analysis to show that an unreasonably large number of American lives would have to have been saved for the costs of CT spending to be justified.

For the costs of CT spending to equal the benefits in terms of the value of lives saved, it would have to have saved 188,740 lives, or 11,796 lives per year, since 9/11.  Narrowing down to just domestic CT spending on government-wide Homeland Security projects shows that spending on just that set of subprograms would have to have prevented the murder of 65,233 people, or 4,077 per year, to break even.  From 2002 through 2017, my latest estimate is that 172 total people were murdered on U.S. soil by all terrorists (Islamic, non-Islamic, domestic, U.S.-born, foreign-born, white supremacists, etc.).  Thus, all CT spending would have to have saved 1,097 times as many lives as were actually taken by terrorists in attacks on U.S. soil for the costs of CT spending to equal the benefits in terms of lives saved.  Focusing on just government-wide Homeland Security CT spending shows that it would have to have saved 379 times as many lives as were actually killed in terrorist attacks on U.S. soil to break even.  It is difficult to estimate a counterfactual but it would take a very creative imagination to honestly believe that post-9/11 CT spending actually saved that many lives by preventing terrorist attacks.

Methodology

The first step is estimating the value of a statistical human life to compare with the cost of CT spending.  This is an emotional and fraught way to measure human life.  As a father and a husband, I understand this emotional reaction very well but the fact remains that if the government spends more than the statistical value of life to save a life through enhanced CT, then that means that other people died because of neglected safety in other areas.  As a hypothetical example, suppose the value of a statistical life is $15 million.  If the government spends $30 million to save one life by spending on X then that means that one person, at least, died who did not have to if that money was spent where it would save more lives.  Thus, spending that amount of money on reducing the risk of X results in more deaths than otherwise would have occurred.  Although emotional and hard to calculate, estimating the statistical value of life can help policymakers save more lives.  The death of human beings is the largest and most significant cost of terrorism but not the only one as other forms of destruction are also costly but relatively minor compared to death.  For the purposes of simplicity, I will focus on the cost in terms of human life.   

As I wrote in 2016, the Department of Homeland Security (DHS) produced an initial estimate that valued each life saved from an act of terrorism at $6.5 million, then doubled that value (for unclear reasons) to $13 million per life saved.  Adjusting for inflation raises that estimate to about $7.5 million.  Hahn, Lutter, and Viscusi use data from everyday risk-reduction choices made by the American public to estimate that the value of a statistical life is $15 million.  I use $15 million in this blog post as it is the largest number.

ICE Scraps Plans For “Extreme Vetting” Prediction Tech

During his presidential campaign Donald Trump proposed the “extreme vetting” of immigrants. Civil libertarians criticized the proposal, not least because the Extreme Vetting Initiative mandated by one of President Trump’s first executive orders sought technology that would use machine learning to determine whether visa applicants would be likely to contribute to society and the national interest. Fortunately, Immigration and Customs Enforcement (ICE) – is no longer pursuing this vetting technology.

In January 2017 President Trump issued Executive Order 13769, which stated in part (emphasis mine):

Sec. 4. Implementing Uniform Screening Standards for All Immigration Programs. (a) The Secretary of State, the Secretary of Homeland Security, the Director of National Intelligence, and the Director of the Federal Bureau of Investigation shall implement a program, as part of the adjudication process for immigration benefits, to identify individuals seeking to enter the United States on a fraudulent basis with the intent to cause harm, or who are at risk of causing harm subsequent to their admission. This program will include the development of a uniform screening standard and procedure, such as […] a process to evaluate the applicant’s likelihood of becoming a positively contributing member of society and the applicant’s ability to make contributions to the national interest; and a mechanism to assess whether or not the applicant has the intent to commit criminal or terrorist acts after entering the United States.

The Extreme Vetting Initiative tasked with implementing (among things) this feature of Trump’s executive order, included the following in its statement of objectives:

ICE must develop processes that determine and evaluate an applicant’s probability of becoming a positively contributing member of society as well as their ability to contribute to national interests in order to meet the EOs outlined by the President.

A background document on the initiative outlined requirements, including the exploitation of publicly available information found on blogs, social media, academic websites, and other online sources. The same backgrounder went on to state that the goal was for the initiative to generate 10,000 investigatory leads each year.

Earlier this year dozens of computer scientists, mathematicians, and engineers wrote a letter to then-Acting Secretary of Homeland Security Elaine Duke, outlining the numerous issued associated with the Extreme Vetting Initiative. As I noted in November last year, the letter highlighted that ICE’s proposal would likely be discriminatory as well as unreliable. From the letter:

According to its Statement of Objectives, the Extreme Vetting Initiative seeks to make “determinations via automation” about whether an individual will become a “positively contributing member of society” and will “contribute to the national interests.” As far as we are aware, neither the federal government nor anyone else has defined, much less attempted to quantify, these characteristics. Algorithms designed to predict these undefined qualities could be used to arbitrarily flag groups of immigrants under a veneer of objectivity.

Inevitably, because these characteristics are difficult (if not impossible) to define and measure, any algorithm will depend on “proxies” that are more easily observed and may bear little or no relationship to the characteristics of interest. For example, developers could stipulate that a Facebook post criticizing U.S. foreign policy would identify a visa applicant as a threat to national interests. They could also treat income as a proxy for a person’s contributions to society, despite the fact that financial compensation fails to adequately capture people’s roles in their communities or the economy.

For more information on the Extreme Vetting Initiative, including original ICE documents, visit the Brennan Center for Justice’s resource page.

The Toronto “Van Incident” and Terrorism in Canada

Toronto Police Chief Mark Saunders said that there is no evidence that yesterday’s “van incident,” where Alek Minassian murdered 10 people and injured 15 others on a busy sidewalk with a van, was a terrorist attack.  To count as a terrorist attack, Minassian’s motivations must have been political, religious, or social in nature beyond simply a desire to terrorize or murder others.  Minassian’s motives are so far unclear with much speculation regarding his social awkwardness and possible anti-women opinions but, so far, little surrounding his political or religious opinions.  This could change as police and investigators uncover new facts.

Many in media and government, prompted by Minassian’s mass murder, are commenting on terrorism in Canada but with little context.  By using the methods employed in my recent terrorism risk analysis for the United States, I’ve found that terrorism is rare in Canada.  Assuming that investigators will eventually find that Minassian’s mass-murder is not terrorism, as they currently claim, then the annual chance of being murdered in a terrorist attack on Canadian soil over the last 25 years was about one in 60.4 million per year.  The annual chance of being injured in a terrorist attack on Canadian soil during that time was about one in 7.4 million per year.

Data and Methodology

This post examines 25 years of terrorism on Canadian soil from 1993 through April 23, 2018.  Fatalities and injuries in terrorist attacks are the most important measures of the cost of terrorism. The information sources are the Global Terrorism Database (GTD) at the University of Maryland, the RAND Corporation, and others.  I excluded three fatalities counted by the GTD as they were the terrorists themselves.  I further grouped the ideology of the deadly attackers into four broad ideologies: Islamists, Anti-Muslims, anti-government, and Unknown/Other. GTD descriptions of the attackers, news stories, and wikipedia were my guide in grouping the attacks by ideology. The grouping by ideology was easy as there were so few terrorist attacks in Canada from 1993 to the present.  The number of Canadian residents and non-terrorist murders in each year comes from Statistics Canada.

Terrorism Risk in Canada

Terrorists have murdered 14 people on Canadian soil from 1993 through April 23, 2018.  Islamists murdered 3 of the victims, an anti-government terrorist murdered 3, suspected terrorists of an unknown ideology murdered 2, and 6 were murdered by an anti-Muslim terrorist named Alexandre Bissonnette in a shooting at a Quebec mosque last year (Figure 1).  Of the 63 terrorist attacks in Canada during that time, according to a wide definition of the term “terrorist” in the GTD, only 7 resulted in a fatality.  In other words, 89 percent of terrorist attacks in Canada during the last 25 years killed nobody.

Figure 1

Murders in Canadian Terrorist Attacks by the Ideology of the Attacker, 1993-2018

 

Sources: Global Terrorism Database at the University of Maryland, RAND Corporation, ESRI, and author’s calculations.

Although most of the recorded terrorist attacks targetted small groups in Canada, like Muslims or the police, it is useful to get a sense of the relative danger by looking at the annual chance of being murdered by a terrorist inspired by each ideology.  The annual chance of being murdered by an Islamist in a terrorist attack was the same as that of being murdered by an anti-government terrorist: about one in 281.7 million per year.  The annual chance of being murdered by a terrorist with an unknown ideology was about one in 422.5 million per year.  The greatest risk, but also still tiny, was being murdered by Alexandre Bissonnette in his Mosque attack last year at one in 140.8 million per year over the 25 years. 

There were 114 injuries in terrorist attacks on Canadian soil from 1993 through April 23, 2018 (Table 1).  Terrorists with unknown or other ideologies caused almost 68 percent of those injuries.  Alexandre Bissonnette, the anti-Muslim terrorist, was personally responsible for 17 percent of all injuries in terrorist attacks during this time in Canada.  Islamist terrorists were responsible for about 11 percent of injuries while anti-abortion and anti-government terrorists were responsible for 4 and 2 percent of all injuries, respectively. 

Table 1

Injuries in Canadian Terrorist Attacks by the Ideology of the Attacker, 1993-2018

  Injuries Annual Chance of Being Injured Percent of All Injuries
Unknown/Other

77

1 in 10,973,614

67.5%

Anti-Muslim

19

1 in 44,472,016

16.7%

Islamist

12

1 in 70,414,026

10.5%

Anti-abortion

4

1 in 211,242,077

3.5%

Anti-government

2

1 in 422,484,154

1.8%

Total

114

1 in 7,412,003

100%

Sources: Global Terrorism Database at the University of Maryland, RAND Corporation, ESRI, and author’s calculations.

Comparison to Murder

Fatalities and injuries in terrorist attacks are rare so a relevant comparison to non-terrorist murder puts the terrorism danger into perspective.  There were about 14,807 murders in Canada from 1993 through April 23, 2018.  Because the number of murders is not reported for 2016-2018, I assumed that the number of murders for each of those years was the same as the number in 2015.  The annual chance of being murdered outside of a terrorist attack was about one in 57,000 per year from 1993 through 2018 – about 1,058 times greater than the chance of being killed in a terrorist attack.      

Conclusion

There is a small chance of being murdered in a terrorist attack in Canada over the last 25 years.  By comparison, the annual chance of being murdered in a terrorist attack in the United States over that time was about 25 times greater than in Canada.  Similarly, the annual chance of being murdered in a terrorist attack in Canada also appears to be lower than in Europe.  The chance of being murdered in a non-terrorist murder in Canada was over 1000 times greater.  Alek Minassian’s horrific mass murder does not appear to be a terrorist attack based on the information available at this time, but if it does turn out to be terrorism then it would be the deadliest attack on Canadian soil since December 6, 1989, when Marc Lepine murdered 14 and injured 14 others in an attack inspired by his anti-feminism.  The murder or death of innocent people is tragic no matter the circumstances and the perpetrator should be punished to the fullest extent of the law.  Regardless, Canadians can at least take some comfort in the fact that the chance of being murdered in a terrorist attack in Canada is small in absolute terms, relative to the residents of other developed nations, and compared to the chance of being murdered in a non-terrorist homicide.     

 

 

 

 

More Americans Die in Animal Attacks than in Terrorist Attacks

Comparing the risk of dying in a terrorist attack to a common household accident like slipping in the bathtub is inappropriate.  After all, inanimate objects like bathtubs do not intend to kill, so people rightly distinguish them from murderers and terrorists.  My research on the hazard posed by foreign-born terrorists on U.S. soil focuses on comparing that threat to homicide, since both are intentional actions meant to kill or otherwise harm people.  Homicide is common in the United States, so it is not necessarily the best comparison to deaths in infrequent terror attacks.  Yesterday, economist Tyler Cowen wrote about another comparable hazard that people are aware of, that is infrequent, where there is a debatable element of intentionality, but that does not elicit nearly the same degree of fear: deadly animal attacks.

Cowen’s blog post linked to an academic paper by medical doctors Jared A. Forrester, Thomas G. Weiser, and Joseph H. Forrester who parsed Centers for Disease Control and Prevention (CDC) mortality data to identify those whose deaths were caused by animals in the United States. According to their paper, animals killed 1,610 people in the United States from 2008 through 2015. Hornets, wasps, and bees were the deadliest and were responsible for 29.7 percent of all deaths, while dogs were the second deadliest and responsible for 16.9 percent of all deaths. 

The annual chance of being killed by an animal was 1 in 1.6 million per year from 2008 through 2015.  The chance of being murdered in a terrorist attack on U.S. soil was 1 in 30.1 million per year during that time.  The chance of being murdered by a native-born terrorist was 1 in 43.8 million per year, more than twice as deadly as foreign-born terrorists at 1 in 104.2 million per year.  The small chance of being murdered in an attack committed by foreign-born terrorists has prompted expensive overreactions that do more harm than good, such as the so-called Trump travel ban, but address smaller risks than those posed by animals.

Recapping Immigration Week on the Cato Daily Podcast

All this week, the Cato Daily Podcast (subscribe!) has tackled the myths, errors, and underappreciated elements of immigration policy. President Donald Trump has made a massive reduction in legal immigration a centerpiece of his second year in office, and the sales pitch he’s made on behalf of that plan hinges on a number of false or misleading claims about the costs and benefits of immigration. In case you missed them, here are my discussions with Alex Nowrasteh, David Bier, and Matthew Feeney:

And, not to be left out, Jim Harper discussed his recent paper on new national ID systems including E-Verify, the deeply flawed employment verification system aimed at keeping undocumented immigrants from working in the United States.

Here’s more of Cato’s work on immigration.

Congressman Perry, Terrorists Did Not Cross the Border to Attack Las Vegas

Republican Congressman Scott Perry (PA) was a guest on Tucker Carlson Tonight last night in a segment about the continuing investigation into the Las Vegas shooting earlier this year. Congressman Perry said:

I have been made aware of what I believe to be credible evidence, credible information regarding potential terrorist infiltration through the southern border regarding this incident.

When pressed by another guest, Congressman Perry offered zero evidence but did say that “I have received what I feel to be and believe to be credible evidence of a possible terrorist nexus.”  With all due respect to Congressman Perry’s feelings, they are not evidence and a serious charge like his requires evidence.  If he does have such evidence, he should release it immediately.  

In researching my 2016 policy analysis on terrorism and immigration, I discovered nine terrorists who planned or carried out terrorist attacks on U.S. soil after entering illegally.*  Those nine terrorists killed zero people in terrorist attacks on U.S. soil from 1975 through 2017.  Of those nine terrorists who entered illegally, only three did so along the border with Mexico: Shain Duka, Britan Duka, and Eljvir Duka crossed as children with their parents in 1984.  They are ethnic Albanians from Macedonia.  They were three conspirators in the incompetently planned Fort Dix plot that the FBI foiled in 2007, long after they became adults and more than two decades after they entered illegally.  There is no evidence that the Fort Dix plot was more than 23 years in the making.  The Dukas murdered nobody in a terrorist attack.

The border with Mexico is a terrible way for terrorists to attempt to enter the United States, which is why so few have tried.  The arguments for additional border security are bad but at least they are grounded in some facts.  It is a sign of the desperation of immigration restrictionists that so many have to resort to conspiracy theories about terrorists to justify spending taxpayer dollars on more border security. 

More convicted or actual terrorists who planned an attack on U.S. soil actually entered or attempted to do so through the Canadian border.  The deadly foreign-born terrorists since 1975 have overwhelmingly used non-immigrant visas to enter.  Conspiracy theories about a wide-open southern border that terrorists cross to commit attacks are common but they should not be peddled by a member of Congress.  If Congressman Perry does have evidence that a terrorist or terrorists entered through the southern border and committed the Las Vegas attack, I implore him to release it as soon as possible to prove me wrong.

 

 

*In my original analysis, I reported that ten foreign-born terrorists initially entered illegally but I incorrectly counted 1993 World Trade Center bomber Ahmed Ajaj as an illegal immigrant when he actually initially entered on a student visa.  

Pages