Terrorists by Immigration Status and Nationality: A Risk Analysis, 1975–2017

Including those murdered in the terrorist attacks of September 11, 2001 (9/11), the chance of a person perishing in a terrorist attack on U.S. soil committed by a foreigner over the 43‐​year period studied here is 1 in 3.8 million per year. The hazard posed by foreigners who entered on different visa categories varies considerably. For instance, the chance of an American being murdered in a terrorist attack by a refugee is about 1 in 3.86 billion per year, while the annual chance of being murdered in an attack committed by an illegal immigrant is zero. By contrast, the chance of being murdered by a tourist on a B visa, the most common tourist visa, is about 1 in 4.1 million per year. Compared to foreign‐​born terrorists, the chance of being murdered by a native‐​born terrorist is about 1 in 28 million per year.

There were 192 foreign‐​born terrorists who planned, attempted, or carried out attacks on U.S. soil from 1975 through 2017. Of those, 65 percent were Islamists, 18 percent were foreign nationalists, 6 percent were right‐​wingers, 6 percent were non‐​Islamic religious terrorists, 3 percent were left‐​wingers, and the rest were separatists or adherents of other or unknown ideologies. By comparison, there were 788 native‐​born terrorists who planned, attempted, or carried out attacks on U.S. soil from 1975 through 2017. Of those, 24 percent were right‐​wingers, 22 percent were white supremacists, 16 percent were left‐​wingers, 14 percent were Islamists, 11 percent were anti‐​abortion, and 6 percent were others. This expanded terrorism risk analysis can aid in the efficient allocation of scarce government‐​security resources to best counter the small terrorist threat.

Introduction

American‐​born Syed Rizwan Farook and his foreign‐​born wife, Tashfeen Malik, who entered the United States two years earlier on a K-1 fiancé(e) visa, murdered 14 people in a terrorist attack in San Bernardino, California, on December 2, 2015.¹ Their attack was dramatic and brutal, and it prompted calls for heightened immigration restrictions, additional security checks for K-1 immigrants, and even a complete moratorium on all immigration.² Malik’s attack was unusual because she likely entered the United States with the intent to commit an attack, whereas most foreign‐​born terrorists often live here peacefully for years before concocting their schemes.³ Regardless, the San Bernardino attack increased fears of foreign‐​born terrorism and likely contributed to Donald Trump’s winning the presidential election in 2016. In response, President Trump issued a temporary ban on the admittance of certain nationals from supposedly terror‐​prone countries, along with a reform of the refugee system and a reduction in the number of vetted refugees permitted to enter the United States each year, all to prevent attacks on U.S. soil.⁴

U.S. government security resources should be allocated to the most efficient means of reducing the costs of terrorism. Although the government initially applied cost‐​benefit methods for evaluating the risk of terrorism, the cost of terrorism, and the supposed security benefits provided by the Department of Homeland Security (DHS) in the years after the 9/11 terrorist attacks, these methods were not well developed because cost‐​benefit analyses are anathema to decisionmakers in most government agencies, who are inclined to assume the benefits provided by their agencies offset high costs.⁵ As a partial remedy to that long‐​running deficiency, the Strategic National Risk Assessment (SNRA) seeks to evaluate the risk of threats and hazards to help the government more effectively allocate resources to the security “threats that pose the greatest risk.”⁶ Even so, the SNRA did not include a thorough terrorism risk analysis of different visa categories or of native‐​born terrorists compared to those born abroad. Substantial administrative hurdles and barriers are in place to block foreign‐​born terrorist infiltration from abroad through vigorous vetting procedures that have low error rates.⁷ Any change in immigration policy for terrorism prevention should be subject to a cost‐​benefit calculation. Sensible terrorism screening policy must do more good than harm to justify its existence, meaning that the cost of the damage the policy prevents should at least equal the cost it imposes.

This policy analysis identifies 192 foreign‐​born terrorists in the United States who killed 3,037 people in attacks on U.S. soil from 1975 through the end of 2017. Nine were illegal immigrants, 57 were lawful permanent residents (LPRs), 21 were students, 1 entered on a K-1 fiancé(e) visa, 25 were refugees, 11 were asylum seekers, 41 were tourists on various visas, and 11 were from Visa Waiver Program (VWP) countries. The visas for the remaining 16 terrorists could not be determined. During that period, the chance of being murdered by a foreign‐​born terrorist on U.S. soil was 1 in 3,808,374 a year. The chance of being killed on U.S. soil in a terrorist attack committed by a refugee was 1 in 3.86 billion a year. In fact, the annual chance of being murdered by someone other than a foreign‐​born terrorist in a normal homicide was 264 times greater than the chance of dying in a foreign‐​born terrorist’s attack.

In addition, this policy analysis identifies 788 native‐​born terrorists who killed 413 people in attacks on U.S. soil from 1975 through the end of 2017. During that period, the chance of being murdered by a native‐​born terrorist on U.S. soil was 1 in 28,004,914 a year. In fact, the annual chance of being murdered by someone other than a native‐​born terrorist in a normal homicide was 1,939 times greater than the chance of dying in a native‐​born terrorist’s attack.

The first part of this policy analysis quantifies the risks of foreign‐​born and native‐​born terrorists on U.S. soil by evaluating how many people they murdered and injured in attacks, their ideologies, the visas foreign‐​born terrorists entered on, and their countries of origin. The second part estimates the costs of terrorism on U.S. soil; the third part compares the costs of foreign‐​born terrorism with the costs of proposed policy solutions, particularly an immigration moratorium in the United States.

Brief Literature Survey

Few researchers have tried to identify the specific visas used by terrorists and, with the exception of the earlier version of this policy analysis, no others have used that information to produce a risk assessment for each U.S. visa category or by nativity.⁸ John Mueller and Mark G. Stewart produced superb terrorism risk analyses but did not focus specifically on the terrorism risks broken down by visa category or nativity.⁹ Robert S. Leiken and Steven Brooke wrote the most complete survey of visas used by foreign‐​born terrorists.¹⁰ However, their published work does not separate threats by country, and their analysis ended in 2006; in addition, their data set is no longer available, and they did not produce a risk analysis.¹¹ Immigrants are overrepresented among those convicted of terrorist‐​related offenses post‐​9/​11.¹² Broader links between immigration and terrorism are the subject of additional strands of research, but they are not risk analyses.¹³

Methodology

Terrorism is the threatened or actual use of illegal force and violence by a nonstate actor to attain a political, economic, religious, or social goal through fear, coercion, or intimidation.¹⁴ This analysis focuses on terrorism during the 43‐​year period from January 1, 1975, to December 31, 2017. This time period begins with large waves of Cuban and Vietnamese refugees entering the country—which posed a terrorism risk—and ends with a bombing in New York City. It identifies foreign‐​born terrorists who were convicted of planning, attempting, or committing a terrorist attack on U.S. soil and links them with the specific visa they were first issued, as well as the number of people they individually murdered, if any, in their attacks, the number of people they injured, the countries they were born in, and their ideologies.¹⁵ This report counts terrorists who were discovered trying to enter the United States on a forged passport or visa as illegal immigrants. Asylum seekers usually arrive on a different visa with the intent of applying for asylum once they arrive, so they are counted under the asylum category unless they entered months before claiming asylum. For instance, the Tsarnaev brothers, who carried out the Boston Marathon bombing on April 15, 2013, traveled here on tourist visas but are categorized as asylum seekers because their family immediately applied for asylum. This analysis carries out the same detailed analysis for native‐​born Americans, with a few notable differences. First, visas are not catalogued for native‐​born terrorists because they originate in the United States. Second, their ideologies are broken down into more specific domestic subcategories that are of greater interest to American readers.

Next, information on individual terrorists, their visa types, and the number of victims is compared with the estimated costs per victim and the total number of visas issued in each category. Where conflicting numerical estimates exist, the highest plausible figures are used intentionally to maximize the risks and costs of terrorism in terms of human life. Appendix 1 lists all the foreign‐​born terrorists identified by relevant date, the number of murders, the number of injuries, visa type, country of birth, and ideology.

Finally, other costs of terrorism, such as injuries, property damage, losses to businesses, and reduced economic growth, are also considered. It is easy to draw comparisons between attacks regarding fatalities, but injuries are inherently difficult to compare because of the wide spectrum of severity. Although the cost of deaths is about the same in all attacks, the range of injuries is so vast that the cost of injuries is the least meaningful statistic, which the judicious reader should interpret with caution. Only three terrorist attacks committed by foreigners on U.S. soil have created significant property and business damage, as well as wider economic damage: the 1993 World Trade Center bombing, the 9/11 attacks, and the Boston Marathon bombing. The costs of the U.S. government’s responses to terrorism, such as foreign wars and domestic counterterrorism spending, are specifically excluded. This analysis is concerned primarily with the cost of human lives taken in terrorist attacks and only then considers the costs of injuries and property damage in subsequent sections.

Counting Terrorists and Their Victims

The foreign‐​born terrorist portion of this analysis excludes American‐​born terrorists. For attacks planned or carried out by native‐​born Americans in concert with foreigners, the foreign‐​born terrorists are credited entirely for the murders and injuries that resulted from the plot. For consistency, the choice to credit foreign‐​born terrorists means that native‐​born terrorists who worked with them are not credited with deaths or injuries that resulted from their joint attacks. This choice increases the estimates of harm caused by foreign‐​born terrorists while diminishing that of native‐​born terrorists. For plots that included many foreign‐​born terrorists and victims, each terrorist is credited with an equal number of victims. For instance, the 1993 World Trade Center attack was committed by six foreign‐​born terrorists. Six people were murdered, and 1,042 were injured, making each terrorist responsible for 1 murder and 173.67 injuries. Airplane hijackings that started but did not end in the United States, such as the September 10, 1976, hijacking of TWA Flight 355 by Croatian nationalists that eventually terminated in Paris, France, are also included. However, this analysis excludes terrorist attacks in which the identities of the perpetrators were unknown, as well as attacks that occurred or were intended to occur (but were not successfully carried out) abroad. Those killed or injured by the police or security forces responding directly to the terrorism are counted as victims of the terrorist as well. Terrorists are convicted of making threats only if they make an actual effort to commit the attack or have bomb‐​making experience, or if they make it appear as if they committed the attack through a hoax. Moreover, those who committed violent crimes domestically to fund terrorism, even if they never committed the actual terrorist attack, are counted as terrorists. If the accused terrorist commits suicide while awaiting trial then he is counted as a terrorist, as in the cases of Alexander David Slack and William Rodgers.¹⁶ If the terrorist is killed overseas before being arrested for plotting an attack, then he is not counted. Convictions for weapons charges are not delineated as terrorism unless the weapons are bombs, dynamite, or poisons. Mere possession of machine guns or other illegal firearms by itself is not terrorism. People who teach others how to build a bomb are not counted as terrorists and neither are those who solicit others to commit terrorist attacks. Lastly, this policy analysis counts terrorists who were entrapped by the FBI.¹⁷

The ideologies of foreign‐​born terrorists are broken into the categories of foreign nationalism, Islamism, left, political assassination with unclear motive, religious (non‐​Islamist), right, separatism of various kinds, and unknown or other. Native‐​born terrorists have been divided into a greater number of ideological categories than their foreign‐​born counterparts: anti‐​abortion, anti–specific religion such as anti‐​Jewish or anti‐​Muslim, black nationalism, foreign nationalism, Islamism, religious (non‐​Islamist), separatism of various kinds like Puerto Rican independence or Texas secessionists, white supremacy, unknown or other, and left‐​wing ideology, which includes those motivated by animal rights, environmentalism, and being anti‐​police. In the case of anti‐​police terrorists, this analysis assumes that they were left‐​wing unless there is evidence to the contrary. The citizenship of terrorists at their birth is their country of origin. In the handful of cases where a terrorist could fit into multiple ideological categories, such as being right‐​wing and anti‐​abortion, he was assigned to whichever ideology seemed more of a motivation judging by the target he chose or his statements. The most difficult challenge was distinguishing terrorism from crimes motivated by ethnic, racial, national, religious, or other forms of bigotry. Individual cases that came close to being categorized as terrorism but that were ultimately rejected are available from the author upon request.

Finally, calculating the risk of being murdered in a foreign‐​born terrorist attack on U.S. soil partly depends on the number of people in the United States. The U.S. Census Bureau and the American Community Survey record only the resident population at any year, but there are at any given time a large number of temporary travelers and tourists within the United States. Ideally, these individuals should be included in any risk calculation because they could also be murdered or injured in a terror attack. However, the previous version of this policy analysis did not include them in the denominator for the risk calculations. This may seem like a small point except that there are several million tourists in the United States at any given time, which, if included in the risk calculation, would lower the estimated chance of being murdered or injured in a terrorist attack significantly for each year. Notably, Sayfullo Habibullaevic Saipov murdered eight people in a terrorist attack in New York City on Halloween in 2017, and five of those were Argentine tourists here on vacation.¹⁸ Undoubtedly, some number of those murdered or injured in other attacks have also been tourists or other nonresidents. Despite this, estimating the number of tourists in the United States at any given time over the last 43 years would require too many assumptions and estimates to sustain statistical reliability. Therefore, this policy analysis deliberately chooses to count all people murdered or injured in terrorist attacks on U.S. soil, but in estimating the annual chance of being murdered in an attack counts only U.S. residents as the entire population. This convention overestimates the risk of dying or being injured in a terrorist attack.

Sources

The terrorists’ identities come from 17 main data sets and documents. The first is Terrorism since 9/11: The American Cases, edited by John Mueller.¹⁹ This voluminous work contains biographical and other information related to terrorist attacks and cases since September 11, 2001. Mueller’s work is indispensable because he focuses on actual terrorism cases rather than questionable instances of people who were investigated for terrorism, then cleared of terrorism and convicted under nonterrorism statutes, but whose convictions were ultimately counted as “terrorism‐​related” convictions. For instance, a 2017 Department of Justice (DOJ) National Security Division’s “Chart of Public/​Unsealed International Terrorism and Terrorism‐​Related Convictions from 9/11/01–12/31/15,” included 627 “terrorism‐​related” convictions, of which only 280, or 45 percent, were convictions under an actual terrorism statute.²⁰ Seventy of those 280 convictions were for planning or executing an attack on U.S. soil, and only 40 of the 70 were foreign‐​born. Many of those terrorism‐​related convictions were for citizenship fraud, passport fraud, or false statements to an immigration officer by immigrants who never posed an actual terrorism threat to the homeland.²¹ The convictions of Nasser Abuali, Hussein Abuali, and Rabi Ahmed provide further context for the government’s use of the term “terrorism‐​related.” An informant told the FBI that the trio had tried to purchase a rocket‐​propelled grenade launcher, but the FBI found no evidence supporting the accusation. The three individuals were instead charged with and convicted of receiving two truckloads of stolen cereal.²² The government classified their convictions as “terrorism‐​related” despite the lack of an actual terrorist connection, terror threat, planned attack, conspiracy, or any actual tentative steps toward the execution of a terror attack. While especially absurd, it is not too different from many of the other 346 terrorism‐​related convictions in the DOJ report.

The second source is the Fordham University Center on National Security’s terrorism trial report cards, a compilation of all the trials for terrorism cases for Islamic State of Iraq and Syria (ISIS) members in the United States and statistical analyses and overviews.²³ Third is the 2013 Congressional Research Service report “American Jihadist Terrorism: Combating a Complex Threat.”²⁴ The fourth source of terrorist identities is a combination of the RAND Database of Worldwide Terrorism Incidents (RDWTI), which covers the years 1968–2009, and other RAND Corporation publications on terrorism.²⁵ Fifth is the Global Terrorism Database (GTD) maintained by the National Consortium for the Study of Terrorism and Responses to Terrorism at the University of Maryland, College Park, and other research produced by the GTD.²⁶ It is important to highlight that the RDWTI and GTD overlap considerably, which provides a valuable check. Sixth are the results of numerous Freedom of Information Act (FOIA) requests by various organizations and individuals asking for all terrorism‐​related convictions since 9/11.²⁷ Sources 7 through 17 are the New America Foundation,²⁸ The Intercept,²⁹ the Investigative Project on Terrorism,³⁰ the research of University of North Carolina professor Charles Kurzman,³¹ the George Washington University Program on Extremism,³² the Center for Immigration Studies,³³ the Southern Poverty Law Center,³⁴ research by the National White Collar Crime Center,³⁵ the Terrorism Research Center in Fulbright College,³⁶ a dissertation by Catlyn Kenna Keenan,³⁷ and numerous FBI reports from 1982 to 2005 on terrorist incidents in the United States.³⁸

Individual immigration information for the terrorists, their ideologies, and countries of origin comes from the sources mentioned above, as well as news stories, court documents, and other government reports. Many of the foreign‐​born terrorists analyzed here entered the United States on one visa but committed their terrorist attack after they switched to another visa or were naturalized. This report classifies those foreign‐​born terrorists under the first visa they had when they entered. The only exception to that rule is for those seeking asylum in the United States—they are counted under the asylum visa. That exception is important because those individuals usually make their claim at the U.S. border or after they have entered on another visa, often with the intention of applying for asylum. For instance, Faisal Shahzad is counted on the student visa because he entered initially on that visa and then obtained an H-1B visa before his unsuccessful attempt at setting off a car bomb in Times Square in 2010.

Because this policy analysis is an update, it includes some corrections based on the author’s additional research and suggestions from readers. In almost all cases, these errors were solely those of the author. The corrections from the 2016 version of this policy analysis are identified and explained in Appendix 2.

The Attackers

These data sets identify 192 foreign‐​born terrorists in the United States from 1975 to the end of 2017, of which 9 were illegal immigrants, 57 were lawful permanent residents (LPRs), 21 were students, 1 entered on a K-1 fiancé(e) visa, 25 were refugees, 11 were asylum seekers, 41 were tourists on various visas, and 11 were from Visa Waiver Program (VWP) countries. The visas for 16 terrorists could not be determined.

The number of murder victims per terrorist attack comes primarily from government reports, the RDWTI, the GTD, and John Mueller’s research. From 1975 through 2017, those 192 foreign‐​born terrorists murdered 3,037 people, 98.1 percent of whom were killed on September 11, 2001. The other 1.9 percent of murder victims were dispersed over the 43‐​year period, with two spikes in 1993 and 2015. These spikes represent the 1993 World Trade Center bombing that killed 6 people and the combination of two 2015 attacks—the Chattanooga shooting on July 16, 2015, that killed 5 people and the San Bernardino attack on December 2, 2015, that killed 14 people.

From 1975 through 2017, the approximate annual chance that an American resident would be murdered in a terrorist attack carried out by a foreign‐​born terrorist was 1 in 3,808,374. At one end of the spectrum, foreigners on the VWP killed 1 American resident in a terrorist attack, resulting in a risk of about 1 in 11.57 billion per year. On the other end, those on other tourist visas killed 2,829.4, resulting in a risk of about 1 in 4.1 million a year. The approximate chance that an American would be killed in a terrorist attack committed by a refugee was 1 in 3.86 billion a year. Of the roughly 801,000 total murders committed in the United States from 1975 to the end of 2017, 3,037 (or 0.38 percent) were committed by foreign‐​born terrorists in an attack.³⁹ Those risk statistics are summarized in Table 1. The annual chance of being murdered in a criminal homicide was 264 times as great as dying in an attack committed by a foreign‐​born terrorist on U.S. soil.

The U.S. murder rate declined from 9.5 per 100,000 in 1975 to 4.6 per 100,000 in 2017, whereas the 1975–2017 rate of murder committed by foreign‐​born terrorists was 0.026 per 100,000 per year, only spiking at 1.05 in 2001 (see Figure 1). Zero Americans were killed in a domestic attack committed by foreign‐​born terrorists in 27 of the 43 examined years. In the 16 years after 9/11, only 4 years were marred by successful foreign‐​born terrorist attacks. Figure 1 shows a single perceptible blip for terrorism on the 9/11 attacks in an otherwise flat line.

From 1975 through 2017, 192 foreign‐​born terrorists injured 17,049 people in attacks on U.S. soil, injuring 5.6 people for every person they murdered (Table 2). Just over 87 percent of all people injured in foreign‐​born terrorist attacks, 14,842, were injured on 9/11. From 1975 through 2017, the annual chance of being injured in a terrorist attack carried out by a foreign‐​born terrorist was 1 in 678,399. Illegal immigrants injured zero American residents in terrorist attacks, whereas those on tourist visas injured 1 in 773,202 a year. The approximate chance that an American would be injured in a terrorist attack committed by a refugee was 1 in 798 million a year.

In 27 of the 43 years, zero people were injured in terror attacks committed by foreign‐​born terrorists. The three most injurious attacks were 9/11 (14,842 people injured), the 1993 World Trade Center attacks (1,046 people injured), and the 1984 Rajneeshee bio­terror attack in The Dalles, Oregon (751 people were sickened).

From 1975 through 2017, 788 native‐​born American terrorists killed a total of 413 people in terrorist attacks on U.S. soil. The annual chance of a resident American being killed by a native‐​born terrorist was about 1 in 28 million during that time. The annual chance of being murdered in an attack by a foreign‐​born terrorist is about 7.4 times greater than being murdered in an attack committed by a native‐​born American. That high number is due entirely to the deadliness of the 9/11 attacks. To put the number of murders committed by foreign‐​born terrorists into perspective, a total of 3,518 people were murdered in terrorist attacks on U.S. soil during the 43‐​year time period. Of those, 413 were killed by native‐​born Americans, 68 were killed by unknown terrorists, and 3,037 were killed by foreign‐​born terrorists.⁴⁰

Uniqueness of 9/11

The foreign‐​born terrorist murder rate by itself has a single spike in 2001 and is virtually a flat line for every other year (see Figure 1). The foreign‐​born terrorist murder rate of 1.05 per 100,000 in 2001 is 177 times as great as the next‐​highest annual rate of 0.0059 in 2015. The statistical mode (meaning the most common number) of the annual murder rate by foreign‐​born terrorists is zero.

The 9/11 attacks killed 2,979 people (not counting the 19 hijackers). These attacks were a horrendous crime, but they were also a dramatic statistical outlier. The year 2015 was the deadliest year excluding 2001, with 19 Americans killed by foreign‐​born terrorists. Fourteen of those victims were killed in the San Bernardino attack—the second‐​deadliest attack committed by foreign‐​born terrorists on U.S. soil. The attacks on 9/11 killed about 213 times as many people as were killed in San Bernardino.

Government officials frequently remind the public that we live in a post‐​9/​11 world, where the risk of terrorism is so extraordinarily high that it justifies enormous security expenditures and curtailments of our civil rights.⁴¹ The period from 1975 to September 11, 2001 had only 23 murders committed by 28 foreign‐​born terrorists out of a total of 97 who either attempted an attack or were successful in their attacks. During the same period, 264 people were killed in terrorist attacks committed by native‐​born Americans, and 68 were killed by unknown terrorists (see Table 3). About 64 percent of the murder victims of native‐​born terrorists (168) prior to 9/11 were killed in the 1995 Oklahoma City bombing committed by Timothy McVeigh, Terry Nichols, and Michael Fortier.

From September 12, 2001, to December 31, 2017, 35 people were murdered on U.S. soil by a total of 7 foreign‐​born terrorists, while 67 other foreign‐​born terrorists attempted or committed attacks that did not result in fatalities. During the same period, 149 people were murdered in terrorist attacks committed by native‐​born Americans, and there were zero murders committed by unknown terrorists. The largest such non‐​foreign‐​born attack was when U.S.-born Omar Mateen murdered 49 people in an Orlando nightclub shooting in 2016.⁴²

Prior to 9/11, the chance of being murdered by a foreign‐​born terrorist was about 1 in 276.7 million per year compared to the approximate annual chance of being murdered by a native‐​born terrorist of 1 in 24.1 million per year (see Table 3). Prior to 9/11, native‐​born terrorists had an annual murder rate about 11.5 times as great as foreign‐​born terrorists. After 9/11, the chance of being murdered by a foreign‐​born terrorist was about 1 in 140.5 million per year compared to the approximate annual chance of being murdered by a native‐​born terrorist of 1 in 33 million per year. In post‐​9/​11 America, the chance of being murdered by a native‐​born terrorist was 4.3 times as great as being murdered by a foreign‐​born terrorist. Relative to their population sizes, foreign‐​born terrorists were deadlier than native‐​born terrorists in each period. The horrendous death toll from the terrorist attacks of 9/11 dominates deaths from other attacks.

Prior to 9/11, 1,814 people were injured on U.S. soil by a total of 26 foreign‐​born terrorists (see Table 4). During the same period, 1,040 people were injured in terrorist attacks committed by native‐​born Americans, and there were 441 injuries inflicted by unknown terrorists. Prior to 9/11, the chance of being injured by a foreign‐​born terrorist was about 1 in 3.5 million per year compared to the annual chance of being injured by a native‐​born terrorist of 1 in 6.1 million per year (Table 4). Prior to 9/11, native‐​born terrorists had an annual injury‐​infliction rate about 43 percent lower than foreign‐​born terrorists. After 9/11, the chance of being injured by a foreign‐​born terrorist was about 1 in 12.5 million per year, compared to the annual chance of being injured by a native‐​born terrorist of 1 in 16.1 million per year. It is important to repeat that the horrendous injury rate in the terrorist attacks of 9/11 dominates injuries from other attacks.

Terrorism Deaths and Injuries by Ideology

Because of the disproportionate number of murders committed on 9/11, the Islamist ideology of those attackers is a dominant motivation for the attacks. Of the 192 foreign‐​born terrorists who were active from 1975 through 2017, 65 percent were Islamists, 18 percent were foreign nationalists, such as Armenians who murdered people in vengeance for the genocide carried out by the Turkish government or Croatians who wanted independence from Yugoslavia, while the rest were spread out over the remaining ideologies (Table 5). The number of murders committed by foreign‐​born terrorists by ideology is even more lopsided. Including the 9/11 attacks, foreign‐​born Islamist terrorists murdered 99.6 percent of all people murdered in a terrorist attack on U.S. soil. Of all people murdered by foreign‐​born terrorists, 98.1 percent were murdered in the 9/11 attacks. Of all the people murdered by native‐​born and foreign‐​born Islamist terrorists, 96.1 percent were murdered on 9/11. The approximate annual chance of being murdered in an attack committed by a foreign‐​born Islamist was about 1 in 3.8 million per year. That chance drops substantially to 1 in 2.3 billion per year for foreign‐​born nationalists and right‐​wingers. The distribution of injuries is also similar, with 95.5 percent being inflicted by foreign‐​born Islamists.

Native‐​born terrorists have a very different distribution of ideology compared to that of foreign‐​born terrorists (Table 5). Of the 788 native‐​born terrorists from 1975 through 2017, 24 percent were inspired by right‐​wing ideology, 22 percent by white supremacy, 16 percent by left‐​wing ideology, 14 percent by Islamism, and 11 percent by anti‐​abortion beliefs. The number of murders by native‐​born terrorists as categorized by ideology is quite different. Terrorists inspired by right‐​wing ideology are responsible for 47 percent of the deaths, mostly due to the Oklahoma City bombing. The next‐​deadliest native‐​born terrorist ideology was white supremacy, with 19 percent of all murders, closely followed by Islamism at 18 percent. The annual chance of being murdered in an attack committed by native‐​born right‐​wing terrorists was about 1 in 59.3 million per year. That chance drops substantially to 1 in 152.2 million per year for native‐​born Islamist terrorists. Right‐​wing native‐​born terrorists are responsible for 62 percent of all injuries inflicted in attacks by natives.

Estimating the Cost per Terrorist Victim

When regulators propose a new rule or regulation to enhance safety, they are routinely required to estimate how much it will cost to save a single life under their proposal, which acknowledges that human life is very valuable but not infinitely so.⁴³ Americans are willing to take risks that increase their chance of violent death or murder, such as enlisting in the military, living in cities that have more crime than rural areas, or driving at high speeds—actions that would be unthinkable if individuals placed infinite value on their own lives. It then stands to reason that there is a value between zero and infinity that people place on their lives. In public policy, a review of 132 federal regulatory decisions concerning public exposure to carcinogens found that regulators do not undertake action when the individual fatality risk is lower than 1 in 700,000, indicating that risks are deemed acceptable when annual fatality risk is lower than that figure.⁴⁴ Using a similar type of analysis for foreign‐​born terrorism will help guarantee that scarce resources are devoted to maximizing the number of lives saved relative to costs incurred.

In 2010, the 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.⁴⁵ An alternative valuation by 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.⁴⁶ This policy analysis uses Hahn, Lutter, and Viscusi’s $15 million estimate to remove any suspicion of undervaluation.

There are other costs of terrorism, such as property damage, medical care for the wounded, and disruptions to economic activity that ideally should be included.⁴⁷ However, those costs are highly variable and confined to four major terrorist attacks. These four attacks are the 1993 World Trade Center bombing, the 9/11 attacks, the Boston Marathon bombing, and the Oklahoma City bombing; their highest plausible cost estimates are $1 billion,⁴⁸ $170 billion,⁴⁹ $25 million,⁵⁰ and $554.5 million⁵¹ (excluding the costs of prosecution), respectively. The combined estimates of just over $171 billion for foreign‐​born terrorism and $554.5 million for native‐​born terrorism exclude the costs of the U.S. government’s response to terrorism but capture virtually the entirety of the property and other economic damage. The cost of terrorism in terms of lives lost was greater than the value of property and other economic damages in every terrorist attack examined here, except for the 1993 World Trade Center bombing and the 9/11 attacks.

Foreign‐​Born Terrorism Risk for Visas Issued by Category

The DHS annual Yearbook of Immigration Statistics,⁵² the U.S. Department of State,⁵³ and author estimates based on the long‐​established ratios between admissions and the number of visas issued⁵⁴ provided the numbers of LPRs, student visas, K-1 fiancé(e) visas, asylum seekers, B‐​tourist visas, and entrants through the VWP. The VWP numbers are available only beginning in 1986, the year the program was created. The number of visas issued is not available prior to 1982 so the number of admissions are used instead in those years. The Refugee Processing Center has recorded the number of refugees going back to 1975. The annual gross inflow of illegal immigrants is estimated on the basis of data from DHS, the Pew Research Center, the Pew Hispanic Center, and other sources.⁵⁵ For the purposes of this report, only the illegal immigrants who actually entered the country illegally are included in that category.⁵⁶ Immigrants who entered on legal visas and became illegal by overstaying are counted under the legal visa category on which they entered. There are other vastly greater estimates of the number of illegal immigrants who entered the United States from 1975 to 2017; this analysis assumes the smaller estimated number of illegal entries in order to maximize the danger posed by that class of immigrants.⁵⁷ Terrorists who successfully entered on fraudulent passports, fraudulent visas, or on another person’s legitimate passport or visa are counted toward those categories. For example, Iyman Faris originally entered the United States on a student visa and passport that belonged to another person.⁵⁸ Faris then applied for asylum four months later and received a green card through marriage more than a year after that, but for the purposes of this report he entered on a student visa. This estimation methodology could exaggerate the number of terrorists who entered the United States with a green card or LPR, thus diminishing the relative danger of other categories. At the time of writing, immigration data were mostly available for 2017. In cases where the 2017 data were unavailable, this paper extends the 2016 data one more year. The specifics of the various visa programs will be described in their individual subsections below.

The terrorist risk for each visa category can be understood in different ways. The following sections will present the number of foreign‐​born terrorists in each visa category, the number of murders and injuries carried out by terrorists in each visa category, the chance of terrorists getting a specific visa, and how many deaths can be expected by each foreign‐​born terrorist on a specific visa. Multiplying the number of murders in each visa category by the $15 million cost per victim yields the estimate of the costs of terrorism.

Each subsection that follows presents two estimates: one includes all victims from all foreign‐​born terrorist attacks from 1975 to the end of 2017, and the other excludes victims from the 9/11 attacks because the death toll from them is such an extreme outlier. The number of victims from the 9/11 attacks is more than two orders of magnitude larger than the next‐​deadliest foreign‐​born terror attack on U.S. soil.⁵⁹ That scale of attack is unlikely to be repeated, whereas other attacks on a smaller and less deadly scale will certainly occur in the future. Presenting the terrorism hazard in two formats, one including 9/11 and the other excluding it, enables the reader to focus on understanding the risks from the more common smaller‐​scale attacks that terrorists commit on U.S. soil and the complete historical hazard.

Terrorism Risk for All Visa Categories. The U.S. government issued about 813 million visas from 1975 to the end of 2017.⁶⁰ The 192 foreign‐​born terrorists were granted 0.000024 percent of all those visas issued, including the terrorists who entered on unknown visas. In other words, one foreign‐​born terrorist entered the United States for every 4.2 million nonterrorist foreigners who did so. Of the terrorists who entered in known visa categories, one foreign‐​born terrorist entered the United States for every 4.3 million nonterrorist foreigners. Table 6 and Figure 2 display these numbers, broken out by visa subcategories.

The 9/11 terrorist attacks were the deadliest in world history. Table 7 gives the same statistics as Table 6, except that it excludes the 9/11 attacks. Excluding the 9/11 terrorists and Zacarias Moussaoui, who intended to participate in the attacks but could not because he was in jail at the time, 172 foreign‐​born terrorists entered the United States out of a total of 750 million foreign‐​born persons issued visas in these categories from 1975 through 2017. That means that only 0.000023 percent of all foreigners who entered on these visas were terrorists. For each terrorist, excluding the 9/11 attackers, 4.4 million visas were granted to nonterrorist foreigners.

Of the 19 9/11 hijackers, 18 were in the country on tourist visas. The 19th hijacker was Hani Hanjour, who entered the United States on a student visa. Zacarias Moussaoui was not a hijacker on 9/11, but he was involved in the plot. His French citizenship allowed him to enter the United States on the VWP. The 2,979 deaths from the 9/11 attacks account for all but 58 murders in foreign‐​born terrorist attacks from 1975 through 2017. Omitting the 9/11 terrorist attackers would make the student and tourist visa categories look substantially safer and slightly improve the safety of the VWP.

Number and Cost of Terrorism Victims for All Visa Categories. As previously noted, 3,037 people were murdered by foreign‐​born terrorists in attacks in the United States from 1975 to the end of 2017. Those terrorist attacks cost $45.56 billion in human life or $1.1 billion per year on average as displayed in Table 8.⁶¹ The terrorism cost equals $56.06 per visa issued over that time.

Excluding the 9/11 terrorist attacks lowers the human cost of terrorism to $870 million during the period or $20.2 million per year as displayed in Table 9. The murder‐​cost of terrorism committed by the foreign‐​born inside the United States, excluding 9/11, is $1.07 per visa issued.

Of the 192 foreign‐​born terrorists, 138 did not murder anyone in a terrorist attack. Many of them were arrested before they attacked, or their attacks failed to take any lives. However, they did injure 827 people in their attacks or about 6 people per terrorist. Including all terrorists and the 9/11 hijackers, even the ones who did not kill anybody, each terrorist killed about 16 people on average, for a total human cost of $237.3 million, and injured about 89 people. Excluding 9/11, each terrorist killed an average of 0.34 people, for a total human cost of $5.1 million per foreign‐​born terrorist analyzed here, and they injured a total of 2,207 people or 12.8 people per terrorist.

Only 54 of the 192 foreign‐​born terrorists actually killed anyone in an attack. Of those terrorists, each one killed an average of 56.2 people for a cost of $843.6 million in human life. Excluding 9/11, each successful terrorist killed an average of under 1.66 people for a human cost of $24.9 million inflicted by each successful terrorist. Prior to 9/11, only two terrorists, Mir Aimal Kasi and Eduardo Arocena, killed more than one person each: Mir Aimal Kasi shot and killed CIA employees Frank Darling and Lansing Bennett as they were waiting in traffic outside CIA headquarters in Langley, Virginia, on January 25, 1998; Eduardo Arocena assassinated Eulalio José Negrín on November 25, 1979, and Félix García on September 11, 1980. Over time, the number of terrorists has shrunk but their deadliness has increased.

There were seven successful attackers after 9/11 who killed 35 people, with each terrorist responsible for an average of five murders. Egyptian‐​born Hesham Mohamed Hedayet killed 2 people on July 4, 2002, at Los Angeles International Airport; the Tsarnaev brothers killed 5 people in the Boston Marathon bombing on April 15, 2013, and their subsequent run from the law; Mohammad Abdulazeez murdered 5 people on July 16, 2015; Tashfeen Malik, along with her U.S.-born husband, killed 14 on December 2, 2015, in San Bernardino, California; Emanuel Kidega Samson murdered 1 person on September 24, 2017, in Antioch, Tennessee; and Sayfullo Habibullaevic Saipov murdered 8 people on October 31, 2017, in New York City.

Foreign‐​born terrorists on tourist visas have killed more Americans in attacks than those on any other type of visa, followed distantly by those who entered on student visas. The 2,979 deaths from the 9/11 attacks account for all but 58 murders in foreign‐​born terrorist attacks. Summary data for fatalities by the different visa categories are provided in Table 10.

The number of injuries caused by foreign‐​born terrorists in attacks on U.S. soil is also important to consider, although the range of costs caused by injuries is enormous. At the low end of costs are scratches and blown‐​out eardrums, while at the high end are brain damage, amputation, and paralysis. A fatality caused by a violent action, in contrast, is just as destructive as another fatality caused by a violent action, so the cost range is roughly zero for fatalities. As a result, it is very difficult to make comparisons of injuries in different terrorist attacks. Regardless, injuries are a source of costs that terrorist attacks impose on Americans. The number of people injured by foreign‐​born terrorists in attacks in the United States from 1975 to the end of 2017 was 17,049 (Table 11); 14,842 of these injuries, or 87 percent, were inflicted during the 9/11 attacks. Of the 192 terrorists, 132 did not injure anyone in a terrorist attack, but 15 terrorists murdered 9 people in attacks without committing any additional injuries. A full 61 percent, 117 out of the 192 foreign‐​born terrorists, neither injured nor killed anyone in a terrorist attack on U.S. soil. Summary data for the number of injuries are in Table 11.

Countries of Origin for Foreign‐​Born Terrorists

The country of origin for the largest number of foreign‐​born terrorists is Saudi Arabia, which accounts for 18 out of the 192 foreign‐​born terrorists from 1975 through 2017 (Table 12). Saudis were also the deadliest; they murdered 2,351.8 people and injured 11,717.4, entirely because 15 out of the 19 9/11 hijackers were from Saudi Arabia. Each terrorist from Saudi Arabia has murdered about 131 people in attacks on U.S. soil. There were 16 foreign‐​born terrorists from Croatia who murdered 3 people in terrorist attacks and injured 4, all in the 1970s. There were 13 foreign‐​born terrorists from Pakistan who murdered 17 people and injured 194 in terrorist attacks on U.S. soil. Notably, terrorists from the United Arab Emirates, Egypt, and Lebanon were the deadliest after Saudi Arabia because of their involvement in 9/11.

President Trump issued two executive orders that temporarily stopped or restricted visa issuance to nationals from Iran, Libya, Somalia, Syria, Yemen, Chad, North Korea, and Venezuela in order to “protect the nation from foreign terrorist entry into the United States.”⁶² The government removed Chad from the list shortly after.⁶³ However, only 18 individuals from those eight countries committed or attempted to carry out a terrorist attack on U.S. soil. Foreign‐​born terrorists from the travel ban countries killed zero people in those attacks and injured 32—accounting for 0.19 percent of all injuries caused by foreign‐​born terrorists on U.S. soil. Twenty‐​three of those injuries were inflicted by Dahir Ahmed Adan and Abdul Razak Ali Artan. Adan stabbed 10 people in St. Cloud, Minnesota, on September 20, 2016, and Artan rammed his car into a crowd at Ohio State University and then stabbed several people, ultimately injuring 13 people in his attack.⁶⁴ Both of those attackers were Somali‐​born. Adan was a refugee and Artan was an asylum seeker, and they were both killed by law enforcement officers before they had a chance to murder anybody. There was no good national security justification for President Trump’s executive orders based on the number of people murdered or injured in attacks committed by terrorists from those countries.

The Syrian civil war is also causing much worry about international terrorism. Since the 2016 publication of Cato’s first policy analysis of the foreign‐​born terrorist risk, the author has learned of a Syrian‐​born terrorist of Christian Armenian descent named Karnig Sarkissian who tried to dynamite the Turkish embassy in Philadelphia in 1982.⁶⁵ Sarkissian and his coconspirators did not succeed in their attack.⁶⁶ Sarkissian was a member of an Armenian terrorist group intent on avenging the Armenian genocide perpetrated by the Turkish nationalist government in the early 20th century. He was not targeting Americans on U.S. soil, he was not a Muslim, and he was not ethnically Arab or Syrian. Sarkissian was merely born in Syria and may have had Lebanese citizenship. Although there was one Syrian‐​born terrorist who attempted an attack on U.S. soil in 1982, there have not been any since, despite the vicious civil war in that country.

The following subsections discuss the terrorism risks, costs in terms of human life, and countries of origin for each specific visa category.

Illegal Immigrants. Only nine illegal immi­grants became terrorists, a minuscule 0.000029 percent of the estimated 31.3 million who entered from 1975 through 2017, as summarized in Table 10. In other words, about 3.5 million illegal immigrants entered the United States for each one who ended up being a terrorist. They murdered or injured zero people in terrorist attacks on U.S. soil. Three of the illegal immigrant terrorists were from Macedonia, two from Algeria, one from Palestine, and one each from Lebanon, Canada, and Somalia. The Macedonians were Shain Duka, Britan Duka, and Eljvir Duka; they crossed the Mexican border illegally as children with their parents in 1984. They were three conspirators in the incompetently planned Fort Dix plot that the FBI foiled in 2007, long after they became adults.⁶⁷ Of the other terrorists, seven were Islamists, one was an adherent of a foreign nationalism, and one was a left‐​wing terrorist. None of the 9/11 terrorists entered as an illegal immigrant.

Lawful Permanent Residents. A lawful permanent resident (LPR), also known as a green‐​card holder, can reside and work permanently in the United States until he either naturalizes or commits a serious enough crime to lose his green card and be deported.⁶⁸ More terrorists have taken advantage of the LPR category than of any other visa category. From 1975 through 2017, 57 foreign‐​born terrorists were LPRs—an average of 1.3 terrorists per year. Over the 43‐​year period, about 18 million LPRs were allowed in, meaning that just 0.00032 percent of LPRs were actual terrorists. In other words, one terrorist entered for every 314,462 nonterrorist LPR. Those 57 LPR terrorists killed 17 people in terrorist attacks. The human cost of LPR terrorism was thus $255 million, equal to $14.23 per green card issued (Table 10). Of the 9/11 hijackers, none were lawful permanent residents.

Terrorists with green cards came from 30 different countries. Six terrorists on green cards came from Pakistan, 6 from Armenia, 4 from India, and 3 each from Croatia, Egypt, Jordan, and Sudan. The deadliest terrorist who entered on a green card was Sayfullo Habibullaevic Saipov from Uzbekistan, who murdered 8 and injured 11 in his Halloween attack in New York City in 2017.⁶⁹ Two Kuwaiti terrorists on green cards killed a total of 6 people and injured 176. The first was Nidal A. Ayyad, one of the 1993 World Trade Center bombers; he is responsible for a single death and 174 injuries. The second is Mohammad Youssuf Abdulazeez, who murdered 5 and injured 2 in a 2015 shooting spree in Chattanooga, Tennessee.⁷⁰

Student Visas. Student visas allow foreig­ners to enter the United States temporarily to attend an educational institution such as a college, university, seminary, private elementary school, or vocational training program.⁷¹ A total of 21 foreign‐​born terrorists entered on student visas, and they accounted for 0.00017 percent of the 12.3 million student visas issued from 1975 through 2017.⁷² In other words, 1 terrorist was issued a student visa for about every 586,938 students who were not terrorists.

Terrorists on student visas appear especially deadly because one of them, Hani Hanjour from Egypt, was a 9/11 hijacker. Altogether, students caused 158.8 fatalities, or 1 for every 77,623 student visas issued. The human cost of terrorism caused by foreigners on student visas was thus $2.38 billion, equal to 5.23 percent of all the terrorism costs to human life. The average terrorism cost per student visa issued is $193.24. Excluding 9/11, 20 terrorists entered the United States as students, or 1 entry for every 616,285 million student visas issued. Those 20 committed a total of 2 murders that cost $30 million or $2.43 per student visa issued. Iranian students were the most likely to be terrorists because 7 were convicted or committed attacks that injured or killed zero people—4 of which were the result of a scheme to kidnap Minnesota governor Al Quie in 1979.⁷³Foreign‐​born terrorists who entered on a student visa also injured 1,065.1 people, 6.2 percent of all those injured in attacks committed by foreign‐​born terrorists on U.S. soil.

K-1 Fiancé(e) Visas. The K-1 visa permits a foreign‐​citizen fiancé or fiancée to travel to the United States to marry his or her U.S.-citizen sponsor within 90 days of arrival. Once married, the foreign citizen can then apply to adjust his or her immigration status to that of an LPR.⁷⁴

Tashfeen Malik entered the United States on a K-1 visa sponsored by her U.S.-born husband, Syed Rizwan Farook. Together they murdered 14 people during the San Bernardino terrorist attack of December 2, 2015. This policy analysis attributes all 14 murders to Malik in order to maximize the death toll from foreign‐​born terrorists. Her U.S.-born husband is included in the native‐​born terrorist category. They injured 17 people in their attack, all of which are credited to Malik.

The San Bernardino attack is the only one to involve the K-1 visa. Because of the small number of these visas—708,942—issued over the 43‐​year time frame, this single attack makes the visa look very dangerous, with a single murder for every 50,639 K-1 visas issued. The single terrorist on the K-1 visa has imposed $210 million in costs or an average of $296.22 for every K-1 visa issued—by far the highest cost per visa issued. Although it is the second‐​deadliest visa, there is no trend of K-1 visa holders committing attacks.⁷⁵

Refugees. A refugee is a person who is located outside the United States and is of special humanitarian concern and who demonstrates that he or she was persecuted or fears persecution because of race, religion, nationality, political opinion, or membership in a particular social group. A refugee is not settled in another country and does not violate other immigration bars on admission, such as posing a national security or public health risk.⁷⁶ Refugees are granted status from a third country and then enter the United States after they have been granted a visa. Refugees must apply for a green card after residing for one year in the United States.

Of the 3,391,203 refugees admitted from 1975 to the end of 2017, 25 were terrorists, which amounted to 0.00074 percent of the total. In other words, one terrorist entered as a refugee for every 135,648 refugees who were not terrorists. Refugees were not very successful at killing people in terrorist attacks on U.S. soil. Of the 25, only 3 were successful in their attacks, killing a total of 3 people and imposing a total human cost of $45 million, or $13.27 per refugee visa issued. Two of the three refugee terrorists were Cubans who committed attacks in the 1970s; the other was Croatian. All three were admitted before the Refugee Act of 1980 created the current rigorous refugee‐​screening procedures. Prior to the Refugee Act of 1980, a hodgepodge of poorly managed post–World War II refugee and displaced persons statutes, presidential grants of parole, and ad hoc congressional legislation allowed Hungarian, Cuban, Vietnamese, and other refugee groups to settle in America.⁷⁷ All the murders committed by foreign‐​born refugees in terrorist attacks were committed by those admitted prior to the 1980 Refugee Act. Refugees injured 14.5 people in terrorist attacks; 10 of those injuries were committed by Dahir Ahmed Adan in his September 20, 2016, attack.

One of the Cuban terrorists assassinated a Chilean dissident and his American aide along with another Cuban who entered as an LPR. The second Cuban terrorist assassinated a Cuban exile leader who supported a closer American diplomatic relationship with Fidel Castro. The Croatian terrorist helped hijack a plane in 1976 and was convicted of murder. The GTD and RDWTI showed many more terrorist attacks and assassinations in the 1970s and 1980s that were likely perpetrated by Cuban or Vietnamese refugees, but because there were no convictions these could not be included here.

Many of the refugees arrested after 9/11 were admitted as children, and in some cases there is doubt over whether their attacks can qualify as terrorism.⁷⁸ Other refugees have been arrested for terrorism or the vague “terrorism‐​related charges,” but they were planning terrorist attacks overseas or providing material support for foreign groups operating overseas.⁷⁹ No refugees were involved in the 9/11 attacks.

Asylum Seekers. Asylum seekers are those who ask U.S. border officials for protection because they have suffered persecution or fear persecution because of their race, religion, nationality, membership in a particular social group, or political opinions.⁸⁰ Unlike refugees, asylum seekers must apply in person at the border and are often detained before being granted asylum. Eleven asylum seekers, or 0.0015 percent of the 732,168 admitted from 1975 through 2017, later turned out to be terrorists. For every terrorist who was granted asylum, 66,561 nonterrorist asylum seekers were admitted.

Terrorists who were asylum seekers killed nine people in terrorist attacks, three of them in the Boston Marathon bombing on April 15, 2013, carried out by the Tsarnaev brothers. The brothers entered the United States as young children and later became terrorists. Ramzi Yousef and Ahmed Ajaj, both of whom helped plan the 1993 World Trade Center bombing that killed six people, planned and carried out those attacks as part of a six‐​person team; therefore, this analysis considers them to be jointly responsible for only two of the six murders. Eduardo Arocena, who murdered two in an anti‐​Castro terrorist attack, entered as an asylum seeker even after he was granted asylum in a third country.

The total human cost of terrorism by asylum seekers was $135 million, which is an average of $184.38 per asylum seeker admission. Foreign‐​born terrorists who entered as asylum seekers injured 669.3 people in terrorist attacks—93.7 percent in the 1993 World Trade Center and 2013 Boston bombings. No asylum seekers were involved in 9/11.

Tourist Visas. Tourists on the B visa are allowed to tour the United States for business or pleasure, as well as enroll in short recreational courses of study.⁸¹ These are the tourist visas available to most residents of the world.

The tourist visa categories were also the second most abused by terrorists. A total of 41 terrorists entered the United States on tourist visas from 1975 through 2017. That is an average of 0.95 terrorists who entered on a tourist visa annually. About 245 million tourists entered the United States on tourist visas, so a single terrorist was issued a visa in this category for every 6 million visas issued.

The 41 terrorists on tourist visas killed 2,829.4 people in attacks, or 1 victim for every 86,738 visas issued. The total terrorism cost in terms of human life by terrorists on tourist visas was $42.44 billion, or $172.94 per visa. Foreign‐​born terrorists who entered on tourist visas injured 14,959 people in their attacks, or about 87.7 percent of all people injured in foreign‐​born attacks on U.S. soil from 1975 through 2017.

Eighteen of the terrorists who carried out the 9/11 attacks held tourist visas, so this visa category is responsible for 93.2 percent of all deaths caused by foreign‐​born terrorists. Excluding 9/11 lowers the number of fatalities to 7.2 and the total death‐​related costs to $108 million, or $0.44 per tourist visa issued. Excluding the 9/11 hijackers, one terrorist entered on a tourist visa for every 10.7 million non­terrorist tourists. There was one murder victim for every 34.1 million non­terrorist tourists who entered, excluding the 9/11 attacks.

Visa Waiver Program. The VWP enables most citizens of the participating countries to travel to the United States for business or tourism for up to 90 days without first obtaining a visa.⁸² The participating countries are nations in Europe, East Asia, and South America that have already established security procedures to exclude terrorists and share traveler information with the U.S. government, and whose citizens rarely overstay illegally in the United States.⁸³

There were 11 terrorists on the VWP out of a total of 437.4 million entries during the life of the program (since 1986), or a single terrorist for every 39.9 million entries. That makes the VWP the safest visa category. The 11 VWP terrorists killed 1 person in a terrorist attack on U.S. soil. Islamist terrorist Glen Cusford Francis, an immigrant to Canada from Trinidad and Tobago, assassinated Rashad Khalifa in Tucson, Arizona, in 1990. Francis was not convicted of the murder until 2012. The other interesting foreign‐​born terrorists who entered through the VWP are the French national Zacarias Moussaoui, who was originally part of the 9/11 conspiracy but was in jail on unrelated charges during the attacks; the British shoe bomber Richard Reid, who attempted to ignite his shoe on a transatlantic flight en route to the United States, and Qaisar Shaffi, who cased New York buildings for a future attack that was broken up by British intelligence. Ahmed Ressam was apprehended at John F. Kennedy International Airport while attempting to enter the country illegally using forged passports from nations that were part of the VWP. Because he was captured at the border and his documents were forgeries, he is classified as an illegal immigrant.⁸⁴

In addition, a few international terrorist suspects have been apprehended while trying to enter through the VWP. These include a member of the Provisional Irish Republican Army, a French‐​Bolivian dual‐​national who was implicated in a 1990 bombing of U.S. Marines in La Paz, Mexico, and a British mercenary who tried to buy a fighter jet for the infamous Colombian drug lord Pablo Escobar.⁸⁵

Unknown Visas. The visa statuses of 16 terrorists are unknown. Eleven of the unknown terrorists killed 4.8 people in terrorist attacks for a total human cost of $72 million. They also injured 2 people.⁸⁶

Cost‐​Benefit Analysis

Immigration screening for counter­terrorism purposes is important, but it will never be perfect.⁸⁷ As Steven Camarota at the Center for Immigration Studies has written, “To be sure, in a nation as large as the United States, it is impossible to prevent terrorists from entering the country 100 percent of the time.”⁸⁸ Even though terrorists rarely achieve their ultimate policy goals, the United States will always be vulnerable to terrorist attacks in the sense that the possibility of harm will be greater than zero.⁸⁹

Confronted with the threat of Islamic terrorism, many well‐​known conservatives have called for a complete moratorium on immigration since 2015.⁹⁰ They presumably want to restrict only LPRs, student visas, fiancé(e) visas, illegal immigrants, refugees, and asylum seekers, but they may also want to prevent the entry of tourists. Although support for a complete immigration moratorium is an extreme position held by few Americans, it is a useful policy counterexample to understand the costs and benefits of immigration and tourism restrictions as a means to combat terrorism.

The following sections separate tourists from immigrants and migrants to estimate how many Americans must die from terrorism to justify a moratorium on foreigners entering the United States. Finding the breakeven point at which the benefits of reduced terrorism justify the cost incurred by stopping all legal immigration and tourism helps form the outermost boundaries of a sensible policy.⁹¹ If the benefits of the different policies proposed below outweigh the costs, then the measure is cost‐​effective. If, however, the costs of the policies proposed below are greater than the benefits, then they are not cost‐​effective.

This cost‐​benefit analysis considers the cost of human deaths, property damage, injuries, and economic disruption caused by terrorism. In virtually all cases of terrorism, with the notable exception of the 9/11 attacks, property damage is small and the cost of injuries is minor compared with the cost of the deaths. Government reactions to terrorism, such as the virtual shutdown of Boston in the wake of the Marathon bombing and the grounding of all air travel after 9/11, are indirect costs of terrorism; as such they are not considered here.⁹²

Broad Immigration Moratorium. The economic cost of a moratorium on all future immigration is tremendous. This section includes two cost projections. The first conservatively estimates the economic costs of a moratorium to be only $50.2 billion annually, which is the number used by Harvard economist George Borjas.⁹³ That $50.2 billion counts only the immigration surplus, which is the increase in American wages caused by immigration. The figure ignores other enormous economic benefits, including the economic gains to the immigrants themselves even when they naturalize. The second cost projection assumes the $229 billion annual price tag of a moratorium calculated by Texas Tech University economist Benjamin Powell.⁹⁴

The greatest possible benefit of an immigration moratorium would be the elimination of all terrorism committed by immigrants. Excluding tourists and visitors on the VWP, since 1975 206.6 people were murdered on U.S. soil in terrorist attacks committed by 141 terrorists who entered as illegal immigrants, LPRs, students, fiancé(e)s, refugees, asylum seekers, and those with unknown visa statuses, accounting for 6.8 percent of all fatalities caused by foreign‐​born terrorists on American soil. The other 2,830.4 murders, or 93.2 percent, were committed by 52 tourists and VWP visitors who would have been unaffected by an immigration moratorium, who account for 26.9 percent of all foreign‐​born terrorists and 93.2 percent of murders caused by foreign‐​born terrorist attacks. Some 95 percent of the murders committed by terrorists on tourist visas occurred on 9/11. A ban on immigration will barely diminish the costs of terrorism.

The costs of an immigration moratorium vastly exceed the benefits, even with very generous assumptions buttressing the pro‐​moratorium position. According to a breakeven analysis, which seeks to find when the cost of an immigration restriction would equal the benefit of reduced terrorism, an immigration moratorium would have to prevent 3,347 deaths annually at an estimated $15 million per death, assuming Borjas’s $50.2 billion estimate of the annual economic gain from immigration. In reality, an average of 4.8 murders were committed per year by immigrant (nontourist and non‐​VWP) terrorists during the 43‐​year period. An immigration moratorium would have to prevent 697 times as many such murders annually as actually occurred from 1975 through 2017 for the costs of a moratorium to equal the benefits.

Benjamin Powell’s more realistic $229 billion annual estimate of the economic costs of an immigration moratorium means that the ban would have to prevent 15,267 murders by terrorists each year at a cost savings of $15 million per murder for the benefits of the ban to equal the costs. That number is about 3,178 times as great as the average annual number of terrorist deaths caused by immigrants (excluding tourists) and more than 5 times as great as all the murders committed by all foreign‐​born terrorists (including tourists) from 1975 through 2017. In short, an immigration moratorium would produce huge economic costs for minuscule benefits.

Tourism Moratorium. Given the role that tourism played in the 9/11 attacks, it is tempt­ing to think that limiting an immigration ban to tourism might be a preferable policy. Yet the economic costs of a tourism moratorium are even greater. The World Travel and Tourism Council estimated that international tourists added $503.7 billion directly to the U.S. economy in 2016.⁹⁵ A moratorium on tourism would deny the U.S. economy an amount of economic activity equal to almost 3 percent of U.S. GDP.

The majority of all murders committed by foreign‐​born terrorists, 93.2 percent, were committed by 52 different terrorists on tourist visas and the VWP. A total of 99.7 percent of all terrorist murders committed by those on tourist and VWP visas were committed by 18 terrorists on 9/11. Over the entire 43‐​year period covered by this policy analysis, an average of 65.8 people were murdered each year in terrorist attacks committed by those on tourist visas, producing an average annual cost of $987.4 million—the amount that would be saved if there was a moratorium.

However, the costs of a tourist moratorium vastly exceed the benefits from lives saved. Such a moratorium would have to deter at least 33,580 murders by terrorists per year to justify the loss in economic activity and break even. The annual number of murders committed by tourists in terrorist attacks would have to be 510 times as great as they currently are to justify a moratorium. To put this in perspective, the 33,580 murders that would have to be prevented each year for the costs of a tourist moratorium to justify its benefits is over 11 times as great as all the deaths caused by all foreign‐​born terrorists over the entire 43‐​year period studied here. The threat from terrorism cannot justify a moratorium on tourists.

Including Nonhuman Costs. The destruction of private property, businesses, and economic activity caused by foreign‐​born terrorism during the 1975–2017 period is estimated to have cost $171 billion. The combined human, property, business, and economic costs of terrorism from 1975 through 2017 are thus estimated at $216.58 billion. Spread over 43 years, the average annual cost of terrorism is $5.04 billion, which is about one‐​hundredth the minimum estimated yearly benefit of $553.9 billion from immigration and tourism ($50.2 billion plus $503.7 billion). The average yearly costs of terrorism, including the loss of human life, injuries, property destruction, and economic disruptions, would have to be 522.83 times as great as they have been to justify a moratorium on all foreigners entering the United States. A moratorium on foreigners entering the country is costlier than the benefits, even when including the property, business, and economic costs caused by foreign‐​born terrorism.

Conclusion

Foreign‐​born terrorism on U.S. soil is a low‐​probability event that imposes high costs on its victims, despite relatively small risks, and low costs on Americans as a whole.⁹⁶ From 1975 through 2017, the average chance of dying in an attack committed by a foreign‐​born terrorist on U.S. soil was 1 in 3,808,374 a year, and the chance of being injured was about 1 in 678,399. For 27 of those 43 years, no people were killed on U.S. soil in terrorist attacks committed by foreign‐​born terrorists. In 27 years, most of which overlap with the years in which no one was killed, no people were injured on U.S. soil in terrorist attacks committed by foreign‐​born terrorists. During the same period, native‐​born terrorists murdered 413 people and injured 1,346 in attacks on U.S. soil.

Foreign‐​born terrorism has been a more serious hazard to American life, liberty, and private property than native‐​born terrorism from 1975 through 2017. But foreign‐​born terrorism is a manageable threat given the huge economic benefits of immigration and the relatively smaller costs of terrorism. Unknown terrorists murdered 68 people during that time. The U.S. government should continue to devote resources to screening immigrants and foreigners for terrorism and other threats, but large policy changes like an immigration or tourist moratorium would impose far greater costs than benefits.

Appendix 1

All identified foreign persons who attempted or committed terrorism in the United States over the period 1975 through 2017 are listed in Table A.1. All identified native‐​born persons who attempted or committed terrorism in the United States over the 1975–2017 period are listed in Table A.2.

Appendix 2

Appendix 2 contains a list of corrections from the first version of this policy analysis published on September 13, 2016. Any extensive empirical work like this will contain errors, and those in the first version were almost entirely my own. Additional research on my part, as well as helpful and critical comments delivered in person, via email, or on Twitter from dozens of thoughtful readers and researchers helped to correct some of these errors. The first version of this policy analysis counted 3,024 people murdered by foreign‐​born terrorists in attacks on U.S. soil from 1975 through 2015. The updated version counts 3,028 murders during that time. Although all errors are inexcusable, my initial count of murders was off by only 0.13 percent relative to the corrected updated figures. Undoubtedly, there are some additional errors in this update, but my hope is that they are less serious than the previous version’s. Below, my errors and corrections are listed in the rough order of their importance.

Notes

^1. Matt Pearce, “A Look at the K-1 Visa That Gave San Bernardino Shooter Entry into U.S.,” Los Angeles Times, December 8, 2015.

^2. Alicia A. Caldwell, “U.S. Reviewing Fiancé Visa Program after San Bernardino Shooting,” Associated Press, December 8, 2015; Larry Kudlow, “I’ve Changed. This Is War. Seal the Borders. Stop the Visas,” National Review, December 11, 2015; David Bossie, “Conservatives Should Think Bigger on Immigration Ban,” Breitbart, December 11, 2015; and Ann Coulter, interview, Breitbart News Saturday, December 12, 2015.

^3. “Intelligence Assessment: Most Foreign‐​Born, U.S.-Based Violent Extremists Radicalized after Entering Homeland; Opportunities for Tailored CVE Programs Exist,” Department of Homeland Security, Office of Intelligence and Analysis, March 1, 2017.

^4. Exec. Order No. 13769, 82 Fed. Reg. 8977 (January 27, 2017); and Exec. Order No. 13780, 82 Fed. Reg. 13209 (March 9, 2017).

^5. John Mueller and Mark G. Stewart, Terror, Security, and Money (New York: Oxford University Press, 2011), p. 3.

^6. “The Strategic National Risk Assessment in Support of PPD 8: A Comprehensive Risk‐​Based Approach toward a Secure and Resilient Nation,” Department of Homeland Security, December 8, 2011.

^7. See Jared Hatch, “Requiring a Nexus to National Security: Immigration, ‘Terrorist Activities,’ and Statutory Reform,” BYU Law Review 3 (2014): 697–732; David Bier, “Extreme Vetting of Immigrants: Estimating Terrorism Vetting Failures.,” Cato Institute Policy Analysis no. 838, April 17, 2018.

^8. Alex Nowrasteh, “Terrorism and Immigration: A Risk Analysis.,” Cato Institute Policy Analysis no. 798, September 13, 2016.

^9. John Mueller, ed., Terrorism since 9/11: The American Cases (Columbus: Ohio State University, March 2016).

^10. Robert S. Leiken and Steven Brooke, “The Quantitative Analysis of Terrorism and Immigration: An Initial Exploration,” Terrorism and Political Violence 18, no. 4 (2006): 503–21.

^11. Emails exchanged with Robert Leiken on March 14, 2016, and Steven Brooke on March 17, 2016, confirmed that the data set their paper was based on no longer exists. Emails are available upon request.

^12. U.S. Government Accountability Office, “Criminal Alien Statistics: Information on Incarcerations, Arrests, and Costs,” GAO-11–187, March 2011, p. 25.

^13. Alberto Abadie, “Poverty, Political Freedom, and the Roots of Terrorism.” American Economic Review 96, no. 2 (2006): 50–56; Subhayu Bandyopadhyay and Todd Sandler, “Immigration Policy and Counterterrorism,” Journal of Public Economics 110 (2014): 112–23; Efraim Benmelech and Claude Berrebi, “Human Capital and the Productivity of Suicide Bombers,” Journal of Economic Perspectives 21, no. 3 (2007): 223–38; Claude Berrebi, “Evidence about the Link between Education, Poverty and Terrorism among Palestinians,” Peace Economics, Peace Science, and Public Policy 13, no. 1 (2007): 1–38; Vincenzo Bove and Tobias Böhmelt, “Does Immigration Induce Terrorism?” Journal of Politics 78, no. 2 (2016): 572–88; Seung‐​Whan Choi, “Fighting Terrorism through the Rule of Law?,” Journal of Conflict Resolution 54, no. 6 (2010): 940–66; Seung‐​Whan Choi and James A. Piazza, “Ethnic Groups, Political Exclusion and Domestic Terrorism,” Defence and Peace Economics 27, no. 1 (2016): 37–63; Andreas Freytag et al., “The Origins of Terrorism: Cross‐​Country Estimates of Socio‐​Economic Determinants of Terrorism,” European Journal of Political Economy 27 (2011): 5–16; Tim Krieger and Daniel Meierrieks, “What Causes Terrorism?” Public Choice 147, no. 1–2 (2011): 3–27; Alan B Krueger, “What Makes a Homegrown Terrorist? Human Capital and Participation in Domestic Islamic Terrorist Groups in the USA,” Economics Letters 101, no. 3 (2008): 293–96; Peter Kurrild‐​Klitgaard, Mogens K. Justesen, and Robert Klemmensen, “The Political Economy of Freedom, Democracy and Transnational Terrorism,” Public Choice 128, no. 1–2 (2006): 289–315; Quan Li, “Does Democracy Promote or Reduce Transnational Terrorist Incidents?” Journal of Conflict Resolution 49 no. 2 (2005): 278–97; Daniel Milton, Megan Spencer, and Michael Findley, “Radicalism of the Hopeless: Refugee Flows and Transnational Terrorism,” International Interactions (August 2013): 621–645; Matthew C. Wilson and James A. Piazza, “Autocracies and Terrorism: Conditioning Effects of Authoritarian Regime Type on Terrorist Attacks,” American Journal of Political Science 57, no. 4 (2013): 941–55; and Andrew Forrester, Benjamin Powell, Alex Nowrasteh, and Michelangelo Landgrave, “Do Immigrants Import Terrorism?.” Cato Institute Working Paper no. 56, January 15, 2019.

^14. Global Terrorism Database, “Data Collection Methodology,” http://​www​.start​-dev​.umd​.edu/​g​t​d​/​u​s​i​n​g​-gtd/.

^15. Illegal immigrants are included in a visa category called “illegal” to improve readability.

^16. Jennifer Dobner, “Animal Rights Activist Found Dead in Garage,” Deseret News, June 30, 1999; and Associated Press, “Suspect in Ecoterror Case Found Dead,” Denver Post, December 22, 2005.

^17. Trevor Aaronson, The Terror Factory: Inside the FBI’s Manufactured War on Terrorism  (New York: Ig Publishing, 2018), p. 24.

^18. Max Radwin et al., “Old Friends from Argentina Reunited in New York. Five Died Together in a Terrorist Attack,” Washington Post, November 1, 2017.

^19. Mueller, Terrorism since 9/11: The American Cases.

^20. U.S. Department of Justice, National Security Division Statistics on Unsealed International Terrorism and Terrorism‐​Related Convictions 9/11/01–12/31/15, www​.human​rights​first​.org/​s​i​t​e​s​/​d​e​f​a​u​l​t​/​f​i​l​e​s​/​N​S​D​-​T​e​r​r​o​r​i​s​m​-​R​e​l​a​t​e​d​-​C​o​n​v​i​c​t​i​o​n​s.pdf.

^21. Department of Justice, National Security Division Statistics.

^22. The Intercept (website), Trial and Terror, “Nasser Abuali.”

^23. “Terrorist Trial Report Card: September 11, 2001–September 11, 2011,” ed. Karen J. Greenberg, Center on Law and Security, New York University School of Law; “May 2017 Updates: ISIS Cases in the United States, March 1, 2014–May 8, 2017,” Center on National Security, Fordham University School of Law, May 8, 2017; and “The American Exception: Terrorism Prosecutions in the United States: The ISIS Cases, March 2014–August 2017,” ed. Karen J. Greenberg, Center on National Security, Fordham University School of Law, 2017.

^24. Jerome P. Bjelopera, “American Jihadist Terrorism: Combating a Complex Threat,” CRS Report for Congress no. R41416, Congressional Research Service, January 23, 2013.

^25. RAND National Security Division, “RAND Database of Worldwide Terrorism Incidents”; Brian Michael Jenkins, “The Origins of America’s Jihadists,” RAND Corporation, 2017; Brian Michael Jenkins, “Would‐​Be Warriors: Incidents of Jihadist Terrorist Radicalization in the United States since September 11, 2001,” RAND Corporation Occasional Paper, 2010; and Kevin J. Strom, John S. Hollywood, and Mark Pope, “Terrorist Plots against the United States: What We Have Really Faced, and How We Might Best Defend against It,” RAND Corporation Working Paper WR-1113-DHSST, September 2016.

^26. Global Terrorism Database, http://​www​.start​.umd​.edu/gtd/.; and Roberta Belli, “Effects and Effectiveness of Law Enforcement Intelligence Measures to Counter Homegrown Terrorism: A Case Study on the Fuerzas Armadas de Liberación Nacional (FALN),” Science and Technology Directorate, Department of Homeland Security, August 2012.

^27. Nora Ellingsen and Benjamin Wittes, “Anatomy of a Presidential Untruth: What Data Did the Justice Department Really Provide the White House?” Lawfare, February 12, 2018.

^28. New America Foundation (website), “Terrorism in America after 9/11.”

^29. The Intercept (website), “Trial and Terror.”

^30. Investigative Project on Terrorism, “International Terrorism and Terrorism‐​Related Convictions 9/11/01–3/18/10,” http://​www​.inves​tiga​tive​pro​ject​.org/​d​o​c​u​m​e​n​t​s​/​m​i​s​c​/​6​2​7.pdf.

^31. Charles Kurzman, “Spreadsheet of Muslim‐​American Terrorism Cases from 9/11 through the End of 2015,” University of North Carolina, Chapel Hill, https://​kurz​man​.unc​.edu/​f​i​l​e​s​/​2​0​1​9​/​0​1​/​K​u​r​z​m​a​n​_​M​u​s​l​i​m​-​A​m​e​r​i​c​a​n​_​I​n​v​o​l​v​e​m​e​n​t​_​w​i​t​h​_​V​i​o​l​e​n​t​_​E​x​t​r​e​m​i​s​m​_​2​0​1​9​_​0​1​_​2​2.xls..

^32. Sarah Gilkes, “Not Just the Caliphate: Non‐​Islamic‐​State‐​Related Jihadist Terrorism in American,” Program on Extremism, George Washington University, 2016; George Washington University Extremism Tracker https://​extrem​ism​.gwu​.edu/​i​s​i​s​-​a​m​erica.; and J. J. MacNab, “Anti‐​Government Extremism in America: Violent Acts and Plots in the United States, 2000 to 2018,” Program on Extremism, George Washington University, 2018.

^33. Steven A. Camarota, “The Open Door: How Military Islamic Terrorists Entered and Remained in the United States, 1993–2001,” Center for Immigration Studies, Center Paper no. 21, May 2002; and Janice L. Kephart, “Immigration and Terrorism: Moving beyond the 9/11 Staff Report on Terrorist Travel,” Center for Immigration Studies, Center Paper no. 24, September 2005.

^34. Southern Poverty Law Center, “Terror from the Right: Plots, Conspiracies and Racist Rampages since Oklahoma City,” November 1, 2015; and Southern Poverty Law Center, “Terror from the Right,” July 23, 2018.

^35. John Kane and April Wall, “Identifying the Links between White‐​Collar Crime and Terrorism,” National Institute of Justice, U.S. Department of Justice, September 2004.

^36. Brian L. Smith, Kelly R. Damphousse, and Paxton Roberts, “Pre‐​Incident Indicators of Terrorist Incidents: The Identification of Behavioral, Geographic, and Temporal Patterns of Preparatory Conduct,” Terrorism Research Center in Fulbright College, May 2006.

^37. Catlyn Kenna Keenan, “Behind the Doors of White Supremacy,” University of Denver, Electronic Theses and Dissertations, Digital Commons, August 1, 2014.

^38. FBI​.gov, “FBI Analysis of Terrorist Incidents in the United States 1982” Criminal Investigative Division, 1982; FBI​.gov, “FBI Analysis of Terrorist Incidents in the United States 1986,” Terrorist Research and Analytical Center, Terrorism Section, Criminal Investigative Division, 1986; FBI​.gov, “Terrorism in the United States 1988,” Terrorist Research and Analytical Center, Terrorism Section, Criminal Investigative Division, 1988; FBI​.gov, “Terrorism in the United States 1997,” Counterterrorism Threat Assessment and Warning Unit, National Security Division, 1997; FBI​.gov, “Terrorism in the United States 1998,” Counterterrorism Threat Assessment and Warning Unit, National Security Division, 1998; FBI​.gov, “Terrorism in the United States 1999: 30 Years of Terrorism, A Special Retrospective Edition,” Counterterrorism Threat Assessment and Warning Unit, National Security Division, 1999; FBI​.gov, “Terrorism: 2000–2001,” Counterterrorism Division, 2001; and FBI​.gov, “Terrorism: 2002–2005,” Counterterrorism Division, 2005.

^39. Disaster Center (website), “United States Crime Rates 1960–2015”; and FBI​.gov, “Uniform Crime Reporting (UCR) Program.”

^40. I used the Global Terrorism Database (GTD) at the University of Maryland to estimate the total number of terrorist deaths in the United States during this time period except for 1993 because the data are missing for that year. I used data from the RAND Database to fill in the missing 1993 GTD data.

^41. John Mueller and Mark G. Stewart, Chasing Ghosts: The Policing of Terrorism (New York: Oxford University Press, 2016), pp. 13–21.

^42. Ralph Ellis et al., “Orlando Shooting: 49 Killed, Shooter Pledged ISIS Allegiance,” CNN, June 13, 2016.

^43. John Mueller and Mark G. Stewart, “Responsible Counterterrorism Policy.,” Cato Institute Policy Analysis no. 755, September 10, 2014, p. 4.

^44. Mueller and Stewart, Chasing Ghosts, p. 137.

^45. Lisa A. Robinson et al., “Valuing the Risk of Death from Terrorist Attacks,” Journal of Homeland Security and Emergency Management 7, no. 1 (2010): article 14.

^46. Robert W. Hahn, Randall W. Lutter, and W. Kip Viscusi, “Do Federal Regulations Reduce Mortality?” AEI‐​Brookings Joint Center for Regulatory Studies, 2000. See also Benjamin H. Friedman, “Managing Fear: The Politics of Homeland Security,” Political Science Quarterly 126, no. 1 (2011): 85, footnote 31.

^47. See Karen C. Tumlin, “Suspect First: How Terrorism Policy Is Reshaping Immigration Policy,” California Law Review 92, no. 4 (July 2004): 1173-239; and John Mueller and Mark G. Stewart, “Evaluating Counterterrorism Spending,” Journal of Economic Perspectives 28, no. 3 (Summer 2014): 237–48.

^48. Phil Hirschkorn, “New York Remembers 1993 WTC Victims,” CNN, February 26, 2003.

^49. Mueller and Stewart, Chasing Ghosts, pp. 144, 279.

^50. “Insurers Have Paid $1.2M for Boston Bombing P/C Claims So Far; Health Claims to Top $22M,” Insur​ance​Jour​nal​.com, August 30, 2013.

^51. Damien Hoffman, “The Cost of Terror: Oklahoma City Bombing Cost $681 Million Plus Lives,” Cheat Sheet, April 19, 2010.

^52. DHS​.gov, Immigration Data and Statistics, Yearbook of Immigration Statistics.

^53. U.S. Department of State, Bureau of Consular Affairs (trav​el​.state​.gov), Visa Statistics, “Report of the Visa Office,” 2017; Department of State, Bureau of Consular Affairs, (trav​el​.state​.gov), Visa Statistics, Nonimmigrant Visa Statistics, “Classes of Nonimmigrants Issued Visas (Detailed Breakdown),” 1991 and 1996.

^54. John Teke and Waleed Navarro, “Nonimmigrant Admissions to the United States: 2016,” Department of Homeland Security, Annual Flow Report, January 2018.

^55. Thomas J. Espenshade, “Unauthorized Immigration to the United States,” Annual Review of Sociology 21 (1995): 195–216; Doug S. Massey and Audrey Singer, “New Estimates of Undocumented Mexican Migration and the Probability of Apprehension,” Demography 32, no. 2 (May 1995): 203–13; and DHS​.gov, Publications Library, “Estimates of the Unauthorized Immigrant Population Residing in the United States: 1990 to 2000.”

^56. U.S. General Accounting Office, “Overstay Tracking: A Key Component of Homeland Security and a Layered Defense,” GAO-04–82, May 2004.

^57. Combining three sources for three time periods yields an astonishing 48.9 million illegal entries: (1) the estimated gross illegal entries from Massey and Singer, “New Estimates,” for the years 1975 to 1989; (2) Robert Warren and Donald Kerwin, “Beyond DAPA and DACA: Revisiting Legislative Reform in Light of Long‐​Term Trends in Unauthorized Immigration to the United States,” Journal on Migration and Human Security 3, no. 1 (2015): 80–108, for the years 1990 to 2009; and (3) estimating from Jeffrey S. Passel and D’Vera Cohn, “Trends in Unauthorized Immigration: Undocumented Inflow Now Trails Legal Inflow” (Washington: Pew Research Center, 2008), for the years 2009 to 2017.

^58. United States of America v. Iyman Faris, Complaint to Revoke Naturalization, March 20, 2017.

^59. “San Bernardino Shooting,” CNN Special Report, December 7, 2015.

^60. Illegal immigrants are not really a visa category, but they are listed as such for simplicity’s sake.

^61. No discount‐​rate adjustment.

^62. Exec. Order No. 13769, 82 Fed. Reg. 8977 (January 27, 2017); and Exec. Order No. 13780, 82 Fed. Reg. 13209 (March 9, 2017).

^63. Cristiano Lima, “Trump Lifts Travel Restrictions on Chad,” Politico, April 10, 2018.

^64. Pete Williams et al., “Suspect Identified in Ohio State Attack as Abdul Razak Ali Artan,” NBC News, November 28, 2016; and Nadine Comerford and Alexander Smith, “Dahir Ahmed Adan Named by Police as St. Cloud, Minnesota, Stabbing Suspect,” NBC News, September 20, 2016.

^65. There is some evidence that Sarkissian was born in Lebanon, but the majority of evidence points to Syria; see United Press International, “Three Armenians convicted of plotting to bomb the Turkish … ‚” UPI Archives, January 26, 1985.

^66. David Greenwald, “The Father of a Suspected Armenian Terrorists Accused with …” UPI Archives, June 21, 1984.

^67. Alex Nowrasteh, “Have Terrorists Illegally Crossed the Border?.” Cato at Liberty (blog), September 14, 2016.

^68. U.S. Citizenship and Immigration Services (uscis​.gov), Glossary, “Lawful Permanent Resident (LPR).”

^69. Emanuella Grinberg and Sarah Jorgensen, “New York Terror Attack Suspect Pleads Not Guilty,” CNN, November 28, 2017.

^70. Naina Bajekal and Tara John, “Everything We Know about the Chattanooga Gunman,” Time, July 17, 2015.

^71. U.S. Department of State, Bureau of Consular Affairs (trav​el​.state​.gov), U.S. Visas, “Student Visa.”

^72. F and M visas are for students.

^73. Associated Press, “Four Iranians Accused with Sudanese of Plot to Kidnap Gov. Quie,” New York Times, November 10, 1979.

^74. U.S. Department of State, Bureau of Consular Affairs (trav​el​.state​.gov), U.S. Visas, “Nonimmigrant Visa for a Fiancé(e) (K-1).”

^75. Alex Nowrasteh, “Secret Policy to Ignore Social Media? Not So Fast.,” Cato at Liberty (blog), December 15, 2015. 

^76. U.S. Citizenship and Immigration Services (uscis​.gov), Humanitarian, “Refugees.”

^77. Charlotte J. Moore, “Review of U.S. Refugee Resettlement Programs and Policies,” Congressional Research Service (Washington: Government Printing Office, 1980), pp. 3–16.

^78. David Bier, “Extreme Vetting of Immigrants: Estimating Terrorism Vetting Failure.,” Cato Institute Policy Analysis no. 838, April 17, 2018; and Matthew Hendley, “Paul Gosar Thinks Abdullatif Aldosary, Alleged Bomber, Is a ‘Known Terrorist’; He Is Not,” Phoenix New Times, December 7, 2012.

^79. U.S. Government Accountability Office (website), “Combating Terrorism: Foreign Terrorist Organization Designation Process and U.S. Agency Enforcement Actions,” GAO-15–629, June 25,2015.

^80. U.S. Citizenship and Immigration Services (uscis​.gov), “The United States Refugee Admissions Program (USRAP) Consultation & Worldwide Processing Priorities.”

^81. U.S. Department of State, Bureau of Consular Affairs (trav​el​.state​.gov), “Visitor Visa.”

^82. Bureau of Consular Affairs (trav​el​.state​.gov), “Visa Waiver Program.”

^83. Bureau of Consular Affairs, “Visa Waiver Program.”

^84. Camarota, “The Open Door: How Military Islamic Terrorists Entered and Remained in the United States, 1993–2001.”

^85. “An Evaluation of the Security Implications of the Visa Waiver Program,” OIG-04–26, Department of Homeland Security, Office of Inspector General, April 2004, pp. 11–12.

^86. Two terrorists killed one person in an attack, so they each got credit for one‐​half of the murder.

^87. DHS​.gov, “The Strategic National Risk Assessment in Support of PPD 8,” August 6, 2015.

^88. Camarota, “The Open Door: How Military Islamic Terrorists Entered and Remained in the United States, 1993–2001.”

^89. Max Abrahms, “Why Terrorism Does Not Work,” International Security 31, no. 2 (Fall 2006): 42–78.

^90. Kudlow, “I’ve Changed”; Bossie, “Conservatives Should Think Bigger”; and Ann Coulter, interview by Breitbart News Saturday.

^91. See Mueller and Stewart, “Evaluating Counterterrorism Spending.”

^92. Mueller and Stewart, Chasing Ghosts, p. 188.

^93. George Borjas, We Wanted Workers (New York: W.W. Norton & Company, 2016), p. 158.

^94. Benjamin Powell, “Coyote Ugly: The Deadweight Cost of Rent Seeking for Immigration Policy,” Public Choice 150, no. 1/2 (January 2012): 195–208.

^95. World Travel and Tourism Council, “Travel & Tourism: Economic Impact 2017,” p. 1. Copy of report in author’s files.

^96. Mueller and Stewart, “Evaluating Counterterrorism Spending,” pp. 239–40.