— Trump, in a tweet, Jan. 13, 2019
“23% of Federal inmates are illegal immigrants. Border arrests are up 240%. In the Great State of Texas, between 2011 & 2018, there were a total of 292,000 crimes by illegal aliens, 539 murders, 32,000 assaults, 3,426 sexual assaults and 3000 weapons charges. Democrats come back!”
— Trump, in a tweet, Jan. 12, 2019
Everything is bigger in Texas. But not this big.
Trump portrays the Lone Star State as a crime hub for undocumented immigrants, with rampant cases of voter fraud and thousands of child predators behind bars. Yet his tweets are laden with misleading and inaccurate statistics. The real Texas numbers are far less scary. Let’s dig in.
Trump often claims that immigration leads to more crime, even though most studies show that immigrants commit crime at lower rates than the native-born, regardless of whether they entered the United States legally or illegally. The Trump administration has been undeterred by such research and routinely comes up with new, misleading statistics to make the case for a border wall between the United States and Mexico.
That’s the case with these three Texas tweets.
“58,000 non-citizens voted in Texas, with 95,000 non-citizens registered to vote. These numbers are just the tip of the iceberg.”
Actually, this iceberg has been melting rapidly since Trump’s tweet on Jan. 27.
Only citizens have the right to vote. The Texas secretary of state’s office issued an advisory on Jan. 25, asking county election supervisors to investigate whether 95,000 registered voters were U.S. citizens.
These 95,000 people — 58,000 of whom voted in a previous election — were flagged because they had shown documents to the Texas Department of Public Safety indicating that they were in the country legally but were not U.S. citizens when they obtained a driver’s license or ID card.
The problem with this list from the Texas secretary of state is that many of the 95,000 registered voters became naturalized citizens after they got driver’s licenses or ID cards and before they voted. The initial advisory said the 95,000 people should be considered “WEAK” matches, and on Jan. 29, two days after Trump’s tweet, Texas state officials began removing thousands of names from the list.
The actual number of ineligible voters, if any, remains unknown. The Texas Tribune called the list’s rollout by Republican state officials “a ham-handed exercise that threatened to jeopardize the votes of thousands of legitimate voters across the state,” leading to widespread confusion.
“The secretary of state’s office would eventually walk back its initial findings after embarrassing errors in the data revealed that tens of thousands of the voters the state flagged were actually citizens,” the Texas Tribune reported Feb. 1. “At least one lawsuit would be filed to halt the review, and others were likely in the pipeline. And a week into the review, no evidence of large-scale voter fraud would emerge.”
Here are some of the revisions that have been disclosed for now:
- In the Houston area, nearly 18,000 people have been removed from the initial list. Harris County officials said “more than 60 percent of nearly 30,000 names on a list the state supplied last week are being removed after new guidance from state officials,” the Houston Chronicle reported.
- In Dallas County, an election official said “the county’s voter-roll database vendor has identified 1,715 people as being incorrectly placed on an initial list of 9,938 registered voters sent to the county over the weekend. The share that was flawed, 17 percent, hasn’t been confirmed by state officials yet but could climb if the officials flag more people as being on the list erroneously,” according to the Dallas Morning News.
- “In Travis County, officials dropped 634 voters off their original list of 4,558. … In Tarrant County, it was about 1,100 voters cleared from the original 5,800,” the Texas Tribune reported.
- Chris Davis, president of the Texas Association of Elections Administrators and the election supervisor in Williamson County, told the Morning News that “more than half of the 2,033 voters on his county’s list were being removed after the state’s revision.”
- In Cameron County, officials got an initial list of more than 1,600 people and were later told to remove nearly 1,500 names, leaving around 30. The Texas Tribune reported that county officials were then told by the secretary of state’s office “that the 1,500 number had also been incorrect,” leaving roughly 85 percent of the original list.
- In McLennan County, all 366 voters on the initial list were later removed, according to the Waco Tribune-Herald.
We reached out to the office of Texas Secretary of State David Whitley (R) but received no response. We asked the White House whether Trump stands by his Jan. 27 tweet, considering all the revisions announced in the days after he posted the initial numbers. We received no answer.
The League of United Latin American Citizens, a civil rights group, has filed a lawsuit in federal court in San Antonio arguing that Texas officials are trying to intimidate naturalized citizens by forcing them to prove their eligibility to vote.
“Especially in the context of voting, we know that distortion of the facts has been long weaponized to suppress voting in communities of color,” said Vanita Gupta, president of the Leadership Conference on Civil and Human Rights and former deputy assistant attorney general for civil rights under President Barack Obama. “This must stop. Our democracy is stronger when every eligible voter is able to participate. It is weaker when fear, lies and intimidation deter voters from casting a ballot.”
Documented cases of voter fraud are extremely rare in the United States.
“Thousands of illegal aliens who have committed sexual crimes against children are right now in Texas prisons. Most came through our Southern Border.”
Trump cited the Texas Department of Public Safety for this claim. But Texas DPS data actually say there were 2,837 undocumented immigrants convicted by that state for all sexual offenses — both against adults and minors — from June 2011 through 2018. Moreover, the data do not break down how many of these individuals crossed through the southern border, so it’s unclear what basis Trump has for claiming “most” of them did.
Using data as of January 2018 from the Texas Department of Criminal Justice, Keri Blakinger, a reporter for the Houston Chronicle, found that 1,920 people in Texas prisons who had detainer requests from U.S. Immigration and Customs Enforcement were convicted of sexual assault against a child, while 539 such individuals were convicted of indecency with a child.
That adds up to 2,459. But the data don’t show where these individuals came from, whether they crossed through the southern border or, if they did, whether they crossed legally or illegally.
“Trump seems to be conflating arrests with state prison sentences, and multi-year data with a single moment in time snapshot, and broadening sex assault to sex crimes,” Blakinger wrote.
“23% of Federal inmates are illegal immigrants. . . . In the Great State of Texas, between 2011 & 2018, there were a total of 292,000 crimes by illegal aliens, 539 murders, 32,000 assaults, 3,426 sexual assaults and 3000 weapons charges.”
Actually, 21 percent of federal prison inmates are foreign-born, but not all of them are “illegal immigrants.” At the end of 2017, nearly 23,641 federal prison inmates “did not have lawful immigration status in the United States,” according to the Justice Department. That was 13 percent of the total federal prison population.
The other numbers Trump tweeted are from the Texas Department of Public Safety, covering the period from June 2011 through 2018. But he’s using a strange mix of statistics, focusing on charges instead of convictions, and at one point combining both totals. Not all charges result in convictions, so Trump’s fast-and-loose use of numbers here is highly misleading.
For example, when Trump mentions “292,000 crimes by illegal aliens,” that’s a combination of charges and convictions, so it’s possible he’s double counting many charges. He is also including charges that did not result in convictions; those would not be considered “crimes.” The total number of convictions for this period was much lower: 61,665.
For homicides, Trump uses the number of charges (539), instead of the much lower number of convictions (238). For assaults, Trump also uses charges (32,443), rather than convictions (13,559). It’s the same issue for sexual assaults (3,428 charges and 1,689 convictions) and weapons offenses (2,949 charges and 1,280 convictions).
The Pinocchio Test
Trump tweeted about the 95,000-voter list from the Texas secretary of state before officials began to remove tens of thousands of names. When we reached out to the White House to ask whether Trump stood by his tweet even after these huge revisions, we received no answer.
There’s no way to unring the bell, of course, because more than 40,000 users have retweeted Trump’s tweet. But some acknowledgment that the numbers are wrong would be much better than total indifference to spreading misinformation.
There’s even less of an excuse for Trump’s two other erroneous tweets about Texas. He’s slicing and dicing the numbers in the most misleading and extreme way to hype the rate of criminality among immigrants.
Word to the wise: Don’t mess with Texas. You’ll get Four Pinocchios.
Send us facts to check by filling out this form
Sign up for The Fact Checker weekly newsletter
The Fact Checker is a verified signatory to the International Fact-Checking Network code of principles