January 4, 2019 12:17PM

The Government Doesn’t Understand Its Own Immigration and Crime Data

On March 6, 2017, President Trump issued Executive Order 13780.  The order was mostly concerned with reducing the number of immigrants and travelers from certain countries that his administration thought could pose a terror risk.  One portion of that Executive Order called for the Department of Justice (DOJ) and the Department of Homeland Security (DHS) to investigate the number of terrorist threats and, little noticed at the time, “information regarding the number and types of acts of gender-based violence against women, including so-called ‘honor killings,’ in the United States by foreign nationals.” 

The DOJ-DHS released their report in January 2018 and almost everybody focused on the terrorism portion – including myself and my colleagues here at Cato.  However, thanks to a brilliant lawsuit that uncovered how shoddy the report was, it is now clear that it made an absolutely false statement about the number of foreign-born people arrested for sex offenses.  The DOJ-DHS report says:

Regarding sex offenses, the Government Accountability Office (GAO) in 2011 produced an estimate regarding the population of criminal aliens incarcerated in state prisons and local jails from fiscal years 2003 through 2009. In that report, GAO estimated that over that period, aliens were convicted for 69,929 sex offenses—which, although not explicitly stated in the report, in most instances constitutes gender-based violence against women.

The DOJ-DHS authors of the report made two errors that others have made in interpreting that exact GAO report, many of whom I’ve criticized

First, 69,929 is the number of arrests for sex offenses where the arrestees were criminal aliens, not the number of sex offenses for which criminal aliens were convicted as the DOJ-DHSclaimed.

Second, those arrests occurred from 1955 through 2010, not from 2003 through 2009.

At least the DOJ-DHS have admitted they misinterpreted the GAO report – further vindication that Peter Kirsanow made numerous errors when he was given three full minutes to monologue on it last August on the Tucker Carlson Show.  Kirsanow wouldn’t appear with me on the show after that segment to debate me – I’ll let you guess the reason why.

The biggest problem here isn’t that the DOJ-DHS authors of that report didn’t read the fine print, although that is worrying, or that they likely let their political bias cloud their research findings.  The biggest problem here is that the GAO report misleads more than it illuminates and provides a legitimate looking citation for erroneous claims that are difficult to check.  The GAO is a more professional and less political department than the DOJ or DHS, at least when it comes to investigating and publishing the results of empirical research.  The GAO should retract the report and the later 2018 version that have both been so misinterpreted, rewrite them so that they are crystal clear, re-release them with a list of corrections from the previous editions, and include an FAQ section with answers.   If current government bureaucrats at the DOJ and DHS as well as former bureaucrats like Peter Kirsanow have trouble understanding the GAO report, then clearly the GAO needs to fix the problem and try to prevent it from occurring in the future.  Otherwise, what is the point of the GAO?

 

June 12, 2018 11:54AM

Another Confusing Federal Report on Immigrant Incarceration

The Departments of Justice and Homeland Security (DOJ/DHS) will be publishing a quarterly report on immigrant incarceration in federal prisons because of an Executive Order issued by President Trump last year.  The most recent report found that 20 percent of all inmates in federal prison are foreign-born and about 93 percent of them are likely illegal immigrants.  Since immigrants are only about 13.5 percent of the population and illegal immigrants are only about a quarter of all immigrants, many are misreading it and coming away with the impression that foreign-born people are more crime-prone than natives. 

That is simply not true.

This new DOJ/DHS report only includes those incarcerated in federal prisons, which is not a representative sample of all incarcerated persons in the United States.  Federal prisons include a higher percentage of foreign-born prisoners than state and local correctional facilities because violations of immigration and smuggling laws are federal offenses and violators of those laws are incarcerated in federal prisons.        

The report itself almost admits as much with this important disclaimer: 

This report does not include data on the alien populations in state prisons and local jails because state and local facilities do not routinely provide DHS or DOJ with comprehensive information about their inmates and detainees—which account for approximately 90 percent of the total U.S. incarcerated population.

There is much better information on immigrant incarcerations in other formats.  My co-author Michelangelo Landgrave and I recently released a Cato brief that estimates incarceration rates by immigration status in federal, state, and local correctional facilities.  We used a standard statistical technique known as the residual method to identify illegal immigrants in the incarcerated population in the American Community Survey in 2016.  We found that illegal immigrants are 47 percent less likely to be incarcerated than native-born Americans and legal immigrants are 78 percent less likely to be incarcerated than natives.  The results reported in our recent brief are similar to our other research into immigrant incarceration and conviction rates as well as the peer-reviewed academic research.

The evidence that legal and illegal immigrants are less likely to be incarcerated, convicted, or even arrested for crimes is so overwhelming that even immigration restrictionists like Mark Krikorian at the Center for Immigration Studies admit that, “A lot of data does suggest immigrants are less likely to be involved in crime.”

Crime data in the United States is poor, especially as it relates to details about the immigration status of the offenders.  There is no good reason for a dearth of data on this topic.  Thus, improving the quality of crime and immigration data is important so that we can better understand the relationship between immigration and criminality.  Unfortunately, the new DOJ/DHS report does not help because the information it presents is of a non-representative subsample of the entire incarcerated population, the title of the report strongly suggests that it reports all incarcerations rather than just those in federal correctional facilities, and this information is already available.    

January 16, 2018 12:55PM

New Government Terrorism Report Provides Little Useful Information

The Departments of Homeland Security and Justice (DHS/DOJ) released a report this morning on the threat of international terrorism.  This report was required by President Donald Trump’s executive order that, among other things, originally established the infamous travel ban.  The new DHS/DOJ report produces little new information on immigration and terrorism and portrays some misleading and meaningless statistics as important findings.  Interestingly, the draft version of the report had more interesting and useful information that was mysteriously edited out of the final public version.  It’s remarkable that, given almost a year to produce such a report and with the vast resources of the federal government combined with reams of government information unavailable to the public, that they were able to produce a report of so little of value.     

The DHS/DOJ report found that about 73 percent of those convicted of international terrorism-related offenses from 9/11 through the end of 2016 were foreign-born.  That means that 27 percent of them were native-born Americans.  By focusing exclusively on international terrorism-related charges, this report intentionally ignores domestic terrorists unaffiliated with international terrorists.  Thus, the results of the DHS/DOJ report are, at best, a snapshot of the international subset of terrorism that ignores the purely domestic variety. 

The DHS/DOJ report ignores the most important statistic: how many people were actually killed by these terrorists on U.S. soil.  In our updated terrorism information that runs through the end of 2017, we found that a total of 155 people were killed on U.S. soil in terrorist attacks since January 1, 2002, 34 of them by foreign-born terrorists and 121 of them by domestic terrorists (going back to September 12, 2001 does not add any deaths by identifiable terrorists on U.S. soil but would diminish the chance of dying, so I excluded it from this blog post to bias the results against me).  Since the beginning of 2002, native-born Americans were responsible for 78 percent of all murders in terrorist attacks committed on U.S. soil while foreign-born terrorists only committed 22 percent.  Including the actual number of deaths caused by terrorists flips the DHS/DOJ statistics on its head.     

From the beginning of 2002 through 2017, about the period of time covered by the DHS/DOJ report, the chance of being murdered in a terrorist attack committed by a native-born American on U.S. soil was about one in 40.6 million per year.  During the same period, the chance of being murdered by a foreign-born terrorist was about one in 145 million per year.  The total chance was about one in 32 million a year.  To put that one in 32 million a year chance in perspective, the annual chance of being murdered in a non-terrorist homicide was about one in 19,325 per year or about 1,641 times as great as being killed in any terrorist attack since 9/11.  These numbers are based on updated and expanded data that we plan on publishing in the near future (available upon request). 

The DHS/DOJ report found that at least 549 people were convicted of international terrorism-related charges in federal court from 9/11 to the end of 2016.  These are fewer than the 627 convictions that the DOJ reported through the end of 2015.  What accounts for the 78 fewer convictions over a longer period?  The DHS/DOJ report does not attempt to reconcile their report here with what they have reported previously.  Furthermore, the DHS/DOJ report does not supply the relevant information about the numbers of convictions for terrorism-related offenses, their names, or the actual offenses they committed.  The DHS/DOJ report should have published this information just as the government has done in the past in request to FOIAs.

The DHS/DOJ relies on “terrorism-related convictions” as their important metric, a definition that encompasses numerous convictions that have nothing to do with terrorism.  There is no definition of “terrorism-related” as a crime in U.S. statutes.  The phrase “terrorism-related” appears mostly in reference to actions of government officials in response to terrorism such as a “terrorism-related travel advisory.”  The anti-terrorism Information Sharing Environment, which integrates information which the GAO, defines “terrorism-related” as relating to “terrorism, homeland security, and law enforcement, as well as other information.” That is a definition that so broad “terrorism-related” is not synonymous with “terrorism.” 

The DHS/DOJ report reveals that the DHS had 2,554 encounters with individuals on the terrorist watch list via the FBI’s Terrorists Screening Database (TSDB) in FY 2017.  That means that DHS could have had multiple encounters with the same individuals who were all counted as separate “encounters.”  The TSDB includes the identities of hundreds of thousands of known and suspected terrorists who are both native-born Americans, foreign-born travelers and immigrants to the United States, and foreigners who have not traveled here.  According to a DOJ audit of the TSDB, frontline officers conducted about 270 million checks against the TSDB every month in 2007 with a total of about 3.24 billion checks per year.  Assuming those numbers were unchanged for FY 2017, even though that number has likely increased, and that only 10 percent of them were conducted by DHS, means that about 0.0008 percent of all TSDB checks conducted by DHS resulted in a TSDB hit, or about one for about every 127,000 checks.  That does sound dangerous until you realize that people flagged by the TSDB are not necessarily terrorists.  Even U.S. Senators and Congressmen have been included on the TSDB list.  Getting one’s name on the TSDB list is easy but getting off is very difficult.  As the DOJ audit of the TSDB noted:

[O]ur file review found that the State Department and the DHS’s Customs and Border Protection did not revise encounter records in a screening database in a timely fashion to reflect modified or removed terrorist identities.

Thus, the DHS/DOJ reported TSDB encounters statistic is virtually meaningless.  It’s a count of people the government is concerned about without evidence or a clear way of being removed.  The DHS/DOJ report could have told us how many of these folks actually committed a terrorist attack, eventually did so over time, or were arrested for a terrorism offense but they missed that opportunity. 

The DHS/DOJ report on international terrorism reveals little new information on the international terrorist threat to Americans on U.S. soil.  Unusual for a government report on terrorism, it isn't even capable of providing many scary-sounding statistics that could frighten people.  While that last point is an improvement, future reports on this topic should seek to provide information on this important topic that isn’t publicly known.  This report fails to do that.

December 21, 2017 5:52PM

New Report on Illegal Immigrant Criminality Reveals Little & Admits Its Own Shortcomings

The Department of Homeland Security (DHS) and the Department of Justice (DOJ) today released a report that found that about 94 percent of foreign-born inmates in Federal prisons are illegal immigrants.  That is not surprising, as illegal immigrants convicted of an immigration offense are incarcerated in federal prison and account 7.3 percent of all inmates.  Likewise, drug traffickers who cross international borders are also in federal prison and account 46.3 percent of all prisoners.  Thus, illegal immigrants are overrepresented in federal prison because the federal government enforces immigration laws and many drug trafficking laws but only a small fraction of all those incarcerated for all crimes committed in the U.S. are in federal prisons. 

The authors of this DHS/DOJ report do deserve credit for highlighting its shortcomings.  On the first page, it states:

This report does not include data on the foreign-born or alien populations in state prisons and local jails because state and local facilities do not routinely provide DHS or DOJ with comprehensive information about their inmates and detainees.  This limitation is noteworthy because state and local facilities account for approximately 90 percent of the total U.S. incarcerated population.

The federal prison population is not representative of incarcerated populations on the state and local level, so excluding them from the report means that it sheds little light on nationwide incarcerations by nativity, legal status, or type of crime.  On the last point, it is shocking how unrepresentative federal prison is regarding the types of crimes its inmates are convicted of. In 2016, 67,742 people were sentenced to federal prison.  Almost 30 percent of them were for immigration offenses.  Those immigration convictions comprised 100 percent of the convictions for immigration crimes in the United States in 2016.  By contrast, there were only 85 federal convictions for murder out of a nationwide total of 17,785 murder convictions that year, comprising less than 0.5 percent of all murders.

    

If Garcia Zarate had actually been convicted of murdering Kate Steinle, then he would have been incarcerated in California state prison and he would not show up as an illegal immigrant murderer in this DHS/DOJ report.  What good is a federal report on illegal immigrant incarceration rates if it would have excluded Kate Steinle’s killer had he been convicted? 

The DHS/DOJ report also explained why they did not include an estimate of illegal immigrants incarcerated on the state and local level:

DHS and DOJ are working to develop a reliable methodology for estimating the status of state and local incarcerated populations in future reports.

A March 2017 Cato Institute Immigration Research and Policy Brief employed a commonly used residual statistical methodology to analyze the incarcerated population in the U.S. Census for 2014.  We found that illegal immigrants were about 44 percent less likely to be incarcerated than native-born Americans.  I look forward to reviewing any methodology that the federal government comes up with but illegal immigrant criminals would have to be severely undercounted in prisons to give them an incarceration rate that even approaches native-born Americans.     

The broad finding among criminologists and economists who study this topic is that immigrants are less crime-prone than natives whether measured by the areas where they live or their incarceration rates.  Although there is less research on illegal immigrant criminals, the general finding is that they are less crime-prone or about as criminally inclined as native-born Americans.  The DHS/DOJ report reveals no new information about incarcerations on the federal level, does not provide evidence for a higher nation-wide illegal immigrant incarceration rate, nor does it support the administration’s plea for more border security.    

July 29, 2016 4:40PM

CADC Rules DOJ Discovery Blue Book Off-Limits…For Now

Earlier this month, the U.S. Court of Appeals for the D.C. Circuit (CADC) ruled that the U.S. Department of Justice Federal Criminal Discovery Blue Book for prosecutions were exempt from Freedom of Information Act (FOIA) requests. The National Association of Criminal Defense Lawyers (NACDL) filed the suit to make the book public, and for good reason.

 

For background, criminal discovery is the process by which a prosecutor's office turns over evidence to the defense team that is relevant to the criminal case before trial. Particularly, evidence that might be helpful or exculpatory to a criminal defendant must be turned over under Brady v. Maryland (1963) and subsequent cases. For example, if investigators independently found an eyewitness that supports a defendant's alibi, or discovers that a witness or police officer has a history of dishonesty, that information must be turned over to the defense counsel in the furtherance of justice. Such evidence is known as "Brady material." 

 

The origin of NACDL's case dates back to the bungled prosecution of the late Sen. Ted Stevens (R-AK). A federal judge threw out Stevens' 2008 conviction for corruption because the DOJ hid evidence from the defense team, including contradictory statements by a star witness that were crucial to proving Stevens' alleged criminal intent. Furthermore, the judge ordered an independent inquiry into the handling of the case that resulted in a damning 514-page report that faulted the DOJ for its mismanagement and "egregious misconduct" in the case.

 The prosecution and conviction ended the career of a long-serving United States Senator. If the DOJ could do that to him, they may do (and probably have done) that to people much less powerful. Consequently, Congress held hearings to consider whether or not to pass new legislation to ensure discovery would be properly handled within the Department of Justice. But as NACDL wrote in their 2014 complaint:

DOJ asserted that federal legislation was unnecessary to prevent future discovery abuses because it had instituted various internal reforms.  During the hearings, DOJ asserted it had implemented "rigorous enhanced training" to ensure that "prosecutors and agents [have] a full appreciation of their responsibilities" under federal law. As part of this effort, DOJ stated that it had created a "Federal Criminal Discovery Bluebook" that "comprehensively covers the law, policy, and practice of prosecutors' disclosure obligations" under Brady v. Maryland, 373 U.S. 83 (1963), Giglio v. United States, 405 U.S. 150 (1972), and their progeny. According to DOJ, the Blue Book was "distributed to prosecutors nationwide in 2011" and "is now electronically available on the desktop of every federal prosecutor and paralegal." [internal citations omitted]

In short, the Blue Book should assure everyone that DOJ prosecutors will play by the rules...but everyone will just have to take the DOJ at their word on that because the Blue Book is off-limits to the public.

 

Although the CADC agreed with the lower court ruling and the DOJ's interpretation of current precedent, Senior Judge David Sentelle wrote a concurrence to the opinion, joined by Senior Judge Harry Edwards, that read: 

It is often said that justice must not only be done, it must be seen to be done. Likewise, the conduct with the U.S. Attorney must not only be above board, it must be seen to be above board. If the people cannot see it at all, then they cannot see it to be appropriate, or more is the pity, to be inappropriate. I hope that we shall, in spite of Schiller, someday see the day when the people can see the operations of their Department of Justice. 

 

In short, I join the judgment of the majority, not because I want to, but because I have to.

Such a concurrence signals that the guiding precedent in Schiller should be re-examined and such information vital to the public interest should be made public. "Just trust us" is not a reasonable guarantee of governmental and prosecutorial accountability.

The NACDL released a statement that they will file for an en banc hearing at CADC. You can read the opinion and concurrence here. Judge Sentelle delivered the B. Kenneth Simon Lecture on Constitutional Thought at Cato on Constitution Day 2013 that you can read here.

April 29, 2016 11:49AM

When Should Courts Defer to White‐​Collar Prosecution Settlements?

Deferred prosecution agreements and their close relatives non-prosecution agreements (DPAs/NPAs) have become a major tool of white-collar prosecution in recent years. Typically, a business defendant in exchange for escape from the costs and perils of trial agrees to some combination of cash payment, non-monetary steps such as a shakeup of its board or manager training, and submission to future oversight by DoJ or other monitors. Not unlike plea bargains in more conventional criminal prosecution, these deals dispense with the high cost of a trial; they also dispense with the need for the government to prove its allegations in the first place. DPAs may also pledge a defendant to future behavior that a court would never have ordered, or conversely fail to include remedies that a court would probably have ordered. And they may be drawn up with the aim of shielding from harm — or, in some other cases, undermining — the interests of third parties, such as customers, employees, or business associates of the targeted defendant, or foreign governments.

So there was a flurry of interest last year when federal district judge Richard Leon in Washington, D.C., declined to approve a waiver, necessary under the Speedy Trial Act, for a DPA settling charges that Fokker Services, a Dutch aerospace company, sold U.S.-origin aircraft systems to foreign governments on the U.S. sanctions list, including Iran, Sudan, and Burma. While acknowledging that under principles of prosecutorial discretion the Department of Justice did not have to charge Fokker at all, Judge Leon said given that it had, the judiciary could appropriately scrutinize whether the penalties were too low.

Now a three-judge panel of the D.C. Circuit has unanimously overruled Judge Leon. It pointed out that under well settled law, charging decisions are entrusted to the DoJ or other executive branch prosecutors, not the judiciary, and that judges may not intervene to insist that additional or more stringent charges be filed -- and that is what the pattern in this case amounted to, in the appeals panel’s view.

So far so good, you might think. But the language of the appellate ruling in places might be read to suggest that courts should simply defer to the Justice Department’s judgment and green-light the DPAs it may negotiate, period. And that would be disturbing, since over-lenience is only one of the possible problems with these devices. Noting the rule-of-law concerns that scholars have voiced about DPAs, Michael Greve writes that the new Fokker Services decision “in sharp contrast, oozes with ‘trust your friendly prosecutor’ language” and speaks of dispensing with “seeking a conviction that the prosecution may believe would be difficult to obtain or would have undesirable collateral consequences.” Greve adds: “Inquiring minds might want to know whether the conviction would be ‘difficult to obtain’ for practical reasons — or because the charges are preposterous and brought for reasons bordering on extortion. …No judicial scrutiny means more than boundless prosecutorial discretion. It means mobilizing the courts to create a due process façade for highly suspect bargains.” Let's hope the ruling isn't read that way.

December 9, 2015 2:35PM

Arab and Muslim Americans: The New “Others”

Over the last month, GOP presidential hopeful Donald Trump's counterterrorism policy prescriptions have included creating a database of Arab and Muslim Americans, and more recently, a call for a ban on all Arab/Muslim immigration to the United States. While he has yet to call for the creation of WW II-style ethnic/religious concentration camps for our Arab/Muslim American neighbors, at this point nothing seems beyond the pale for Trump. Unfortunately, as I have noted before, when it comes to stigmatizing--if not de facto demonizing--Arab/Muslim Americans, he's getting some help from DHS, DoJ, and the legislative branch.

Indeed, in the ongoing legislative battle to pass dubious cybersecurity legislation, House Homeland Security Chairman Mike McCaul (R-TX) is being wooed to support the revised cyber information sharing bill with a new carrot: the inclusion of his "countering violent extremism" (CVE) bill in the FY16 omnibus spending bill--a measure condemned earlier this year by civil society groups from across the political spectrum.

To date, McCaul has been opposed to the Senate's approach to cybersecurity issues in the form of the Cybersecurity Information Sharing Act (CISA), and, keeping that in mind, House and Senate supporters have largely excluded him from their negotiations over a final cyber bill. By dangling the inclusion of his CVE legislation in the omnibus is a clear effort to get McCaul to drop his opposition to CISA by giving him one of his priorities: Passage of CVE legislation would create yet another bureaucracy in DHS to essentially monitor the Arab/Muslim American population for signs of extremism. 

The fact that a similar CVE effort in the U.K. failed miserably has not deterred Congressional boosters like McCaul from pursuing that same discredited approach at the expense of the civil and constitutional liberties of a vulnerable minority population. Additionally, the expense of American taxpayers is likely to be at least an additional $10 million per year for the proposed DHS CVE office. 

As former NBC Nightly News anchor Tom Brokaw reminded us this week, Arab and Muslim Americans have died for the United States in Iraq and Afghanistan. They have paid for our freedom with their blood and their lives. Proposals that would strip them of their rights and attempt to turn them into political and societal lepers should be repudiated--vocally and forcefully. Those who propose such un-American and unconstitutional discrimination are the ones who should be shunned and permanently confined to the unhinged fringes of American political and social life.