Together with The Texas Tribune and The Marshall Project, ProPublica investigates border initiatives under Gov. Greg Abbott. State funding for border security has swelled from $110 million in 2008-09 to nearly $3 billion for the 2022-23 budget cycle.
This article is co-published with The Texas Tribune, a nonprofit, nonpartisan local newsroom that informs and engages with Texans, and with The Marshall Project, a nonprofit news organization covering the U.S. criminal justice system. Sign up for newsletters from The Texas Tribune and The Marshall Project.
Thomas King-Randall had been waiting for two hours to drop his daughters off at his ex-girlfriend’s apartment in Midland, Texas. It was 10:30 on a school night in August and it was her turn to care for the two girls.
The ex-girlfriend showed up drunk and was arguing with her new boyfriend in his truck, police later wrote in a report. King-Randall, who is Black, said in an interview that the woman’s Latino boyfriend called him a racial slur, which led to a fight.
By the end of the encounter, the woman’s boyfriend had a bloody nose and swollen eyes. King-Randall was gone, and local police issued an arrest warrant for the 26-year-old California native. A month later, Texas Department of Public Safety officers arrested King-Randall when he tried to renew his driver’s license.
King-Randall’s arrest was one of thousands used to bolster claims of success for Operation Lone Star. Texas Gov. Greg Abbott launched the initiative last March, citing an urgent need to stop the flow of drugs and undocumented immigrants into the state through Mexico.
But the alleged assault had nothing to do with the border. King-Randall, a U.S. citizen, was arrested more than 250 miles from the border with Mexico. Neither DPS nor the Texas Military Department, the state agencies carrying out Operation Lone Star, played a role in the investigation. And the family violence assault charge King-Randall faced wasn’t linked to border-related crime or illegal immigration.
Operation Lone Star has helped increase the state’s budget for border security to more than $3 billion through 2023 by deploying thousands of DPS troopers and National Guard members and allocating funding to build border barriers. As part of the operation, troopers are also arresting some immigrant men crossing into the U.S. on state criminal trespassing charges.
Abbott and DPS have repeatedly boasted in news conferences, on social media and during interviews on Fox News that the border operation has disrupted drug and human smuggling networks. A year into the operation, officials touted more than 11,000 criminal arrests, drug seizures that amount to millions of “lethal doses” and the referrals of tens of thousands of unauthorized immigrants to the federal government for deportation as signs that the program is effective.
But the state’s claim of success has been based on shifting metrics that included crimes with no connection to the border, work conducted by troopers stationed in targeted counties prior to the operation, and arrest and drug seizure efforts that do not clearly distinguish DPS’s role from that of other agencies, an investigation by ProPublica, The Texas Tribune and The Marshall Project found.
King-Randall’s charges were among more than 2,000, including some for cockfighting, sexual assault and stalking, that the agency stopped counting toward Operation Lone Star more than nine months into the exercise, after the news organizations began raising questions about the ties between the arrests and border security. Of those, about 270 charges were for violent crimes, which are defined by the FBI as murder, manslaughter, rape, robbery and aggravated assault.
King-Randall said in an interview that he was fighting the allegations. The case is pending, according to the Midland County district attorney’s office.
Claiming such arrests is “inherently flawed” and misrepresents the accomplishments of the operation, said Patrick O’Burke, a law enforcement consultant and a former DPS commander who retired in 2008.
“The problem could be simply related to crimes in those communities,” O’Burke said. “It’s not battling cross-border crime.”
Asked by the news organizations why such charges were not excluded from the operation’s metrics at the start, DPS officials said they are continuously improving how they collect and report the data “to better reflect the mission” of securing the border. The governor’s office maintained that “dangerous individuals, deadly drugs, and other illegal contraband have been taken off our streets or prevented from entering the State of Texas altogether thanks to the men and women of Operation Lone Star.”
But DPS and Abbott have provided little proof to substantiate such statements. A year into the initiative, Abbott, DPS and the Texas Military Department have fought two dozen public records requests from the news organizations that would provide a clearer picture of the operation’s accomplishments.
DPS, the only agency to release some records related to Operation Lone Star’s results, has made several significant revisions to the arrest data, including removing charges. The agency did not provide details that would help determine how the cases that remained are connected to the initiative’s goal of deterring border-related crime. The agency also failed to identify arrests and drug seizures that could have occurred without the additional personnel made available through the operation.
The absence of clear metrics for measuring its accomplishments points to a larger problem with the border operation and more than a dozen others launched by the state’s two governors during the past 17 years. Lawmakers have repeatedly increased state funding for border security while providing minimal oversight of the operations launched by Abbott and his predecessor, Gov. Rick Perry.
Over the years, some legislators have balked at state agencies’ calls for more accountability from border security efforts.
“It’s almost offensive to say, ‘What are the results?’” former state Rep. Dan Flynn, a Republican from East Texas, said during a hearing in 2018. At that hearing, the Texas Sunset Advisory Commission, which determines whether there’s a continuing need for state agencies and programs, raised concerns that DPS was not providing “sufficient information to the public and policymakers about the return on investment for border security.”
Texas, which shares a 1,200-mile border with Mexico, spends more money on border security than any other state. And at a cost to taxpayers of more than $2.5 million a week, Operation Lone Star is by far the most expensive of the state’s border operations, and the one with the broadest mandate and scope.
In South Texas’ Rio Grande Valley, which was at the center of last year’s immigrant influx, Hidalgo County Judge Richard Cortez said he doesn’t know what Operation Lone Star has accomplished beyond “arresting people and making them criminals.”
Cortez said the problem is not criminal activity, but the sheer number of immigrants seeking better opportunities who sometimes attempt to cross into his community at once, straining resources and overwhelming Border Patrol. The solution, he said, is a comprehensive approach to address the reasons people are trying to come to the U.S. and provide more legal avenues to do so.
“We’re spending millions and billions of dollars in trying to manage something,” Cortez said about Operation Lone Star. “But instead of getting me the plumber to stop the leak, they’re sending me people to mop up the floor.”
Politics of Border Security
With DPS SUVs lined up behind him as if forming a wall, Abbott promoted his new initiative during a March 2021 news conference in Mission, a city in the Rio Grande Valley where more immigrants were crossing the border.
While federal officials started apprehending a greater number of immigrants during Donald Trump’s presidency, Abbott blamed newly inaugurated President Joe Biden for not doing enough to stem record levels of arrivals at the border.
During his first two months at the helm, Biden temporarily halted a policy that required people seeking asylum to wait in Mexico until their cases could be heard by U.S. immigration judges. A federal judge in Texas later ordered the administration to reinstate part of the policy. Under a Trump administration pandemic health order that Biden kept in place, more than three-fourths of immigrants apprehended at the border during that period were immediately turned away.
“If you were president in 2024, which some of us hope that you are, what’s the first thing that you would do to enact something down here?” asked a man in the crowd whom Abbott’s staff singled out for the final question.
“Secure the border. Period,” Abbott said.
With the presidential election in the distance, Abbott has made border security a cornerstone of his gubernatorial reelection campaign, playing offense against his primary opponents, attacking Biden and using the issue as a way to distinguish himself from his general election challenger, former U.S. Rep. Beto O’Rourke, a Democrat from the border city of El Paso.
The governor handily won the Republican primary early this month with Trump’s support. The former president’s success rallying the Republican base by pushing hard-line policies and promoting the construction of a border wall has become a model for Texas GOP candidates, who saw Trump make inroads with Latino voters in border counties in 2020.
The results emboldened Republicans, who doubled down on Trump’s rhetoric, pushing some of his more restrictive border measures, said James Henson, director of the Texas Politics Project at the University of Texas at Austin.
“It’s almost as if he gave permission for more straightforwardly nativist rhetoric, but he didn’t do that in a vacuum, certainly at least not here,” Henson said, pointing to anti-immigrant sentiment among Republican voters prior to Trump’s election.
In launching Operation Lone Star, Abbott went further than any other governor in recent history, attempting to curtail immigration by using state trespassing charges to directly target those who cross the border on private property.
The federal government has sole authority to enforce immigration laws, but Abbott increased trespassing penalties under a declaration that gave him more power akin to what he would have after a natural disaster.
In June, the governor shifted the operation’s emphasis from the Rio Grande Valley, where political leaders opposed some of his efforts, to a vast rural region of mostly private ranches around Val Verde County, about 170 miles west of San Antonio. Trump won the county by a 10-point margin in 2020. Until this year, Val Verde and Kinney were the only two counties prosecuting people crossing into the country through private property for trespassing.
The misdemeanor charge, punishable by up to a year in jail, makes up about 40% of the operation’s arrests from mid-July to Jan. 27, an analysis by ProPublica, the Tribune and the Marshall Project found.
The governor’s office said the operation is based on facts, not politics, and is geared to provide “maximum assistance to the counties greatest affected.” But federal statistics show some of the counties in the Rio Grande Valley that DPS shifted additional resources away from were among those experiencing the greatest influx of immigrants and drugs.
Command Sgt. Maj. Jason Featherston, a Texas Army National Guard veteran who helped oversee the guard’s deployment under the operation until his retirement in November, said he and his colleagues believed politics was the main driver for the mushrooming initiative. He said he recalls commanders saying things like, “We’re going back to the border, the governor is trying to get reelected.”
Federal and state Democratic lawmakers have urged investigations into the constitutionality of the trespassing arrests and the poor working conditions, pay delays and suicides among National Guard members assigned to Operation Lone Star, problems reported by the Tribune and the Army Times. And some state Democrats, led by the Mexican American Legislative Caucus, announced a task force early this month to investigate “many layers of grave concerns” about the operation, including alleged human rights violations and a lack of accountability. Abbott’s office has said the arrests and prosecutions under the operation “are fully constitutional.”
But the broader operation’s goals and results have received little scrutiny.
In July, DPS began counting toward Operation Lone Star a number of arrests and drug seizures from a 63-county region almost the size of Oregon that officials dubbed the area of interest. The area included counties that did not receive additional resources from the operation, and some of the newly credited actions included work already conducted by troopers stationed there before the governor’s initiative began.
Before then, DPS had been counting arrests and drug seizures from what the agency called the “more focused” area of operation, a smaller group of counties closer to the border.
The governor and DPS declined to answer questions about who ordered the change and whether all the counties in the larger area of interest received extra resources from the operation. DPS officials said the area of operation is fluid as the department is continuously monitoring the border and adjusting its use of resources as needed.
Abbott pointed to some of those arrests last year as he sought additional funding for border security efforts, bringing lawmakers back for a special legislative session. Abbott’s office received $1.3 billion of the $3 billion total, marking the first time that the governor’s allocation for border security was larger than that given to DPS.
The growing share of border security funding managed by the governor’s office raises questions about transparency, said Eva DeLuna Castro, a budget analyst for the progressive think tank Every Texan. She said such spending is harder to track because the governor’s office doesn’t report its expenditures with the same level of detail as DPS.
While the governor’s office argues that the agencies it funds have to report spending, DeLuna Castro said some are not subject to such rules.
In January, after increasing the number of National Guard members at the border to 10,000, the governor and a handful of the state’s Republican leaders moved nearly half a billion dollars from DPS, the Texas Department of Criminal Justice and the Texas Alcoholic Beverage Commission to help cover the increased costs.
“He’s just running up a tab that the Legislature, and taxpayers, will have to cover,” DeLuna Castro said.
Trouble With the Numbers
Fentanyl seizures have become shorthand for Operation Lone Star’s success.
Abbott repeatedly highlights them in press conferences and on social media, boasting that the state is helping to stop Biden’s “open border policies.” He has used seizures of the synthetic opioid, which is 100 times stronger than morphine, as a way to attack O’Rourke, who is challenging him in the November gubernatorial election.
At a February event in Austin before the primary election, Abbott’s campaign handed out pill bottles with a fake label that read “Beto Biden open border” and pointed to 1,334 Texas fentanyl deaths in 2021.
Inside was a mock warning label that credited the seizure of 887 pounds of fentanyl, or what he called more than 201 million deadly doses, to Operation Lone Star. Days later, Abbott repeated similar claims in a press release from the governor’s office.
The figure reflects seizures across the state and contradicts the number DPS has given for what is attributable to Operation Lone Star. About 160 pounds of fentanyl were seized from March 2021 to January 2022 in the regions that DPS uses when reporting metrics from the operation.
Abbott’s office defended using statewide seizure numbers, saying they are directly tied to Operation Lone Star because the drug generally enters from Mexico.
“DPS can’t always seize fentanyl right at the border; but they will not stop until they find it, even if it is in North Texas,” Nan Tolson, Abbott’s spokesperson, wrote in an email.
Including statewide seizures is “just disingenuous,” said O’Burke, the former DPS commander.
“Chicago has a border nexus. Are we going to count drugs that were seized in Chicago? That’s just not transparent,” he said. “It’s just not a measure of success. It’s just conflating these statistics because it makes the general public feel safer.”
Instead, O’Burke said, Operation Lone Star’s results should only count actions in which its added resources were used.
That number comes with its own caveats. All but 12 of the 160 pounds of fentanyl were captured in El Paso County, which was not one of the ones listed by DPS officials in November as receiving additional troopers and National Guard members from the operation. The county was one of several that declined to sign on to the governor’s border disaster declaration.
Fentanyl seizure claims are not the only example of the difficulty of measuring the return on investment for taxpayers.
DPS has a history of taking credit for work, such as drug seizures, carried out by other agencies. As part of the operation, DPS and Texas Military Department officials reported apprehending more than 200,000 migrants in the past year and referring them to the federal government for deportation. That included eight migrants who were caught rafting across the Rio Grande by DPS troopers, National Guard members and Border Patrol agents in November. But while DPS counted the immigrants it referred to Border Patrol as part of its reporting for Operation Lone Star, that same group may also have been included in the National Guard’s tally, meaning both agencies could be getting credit for the same arrests. The Texas Military Department did not answer questions about the case.
U.S. Customs and Border Protection declined multiple interview requests. Officials did say that the federal agency “does not have a role or partner in any way” with DPS on the operation and that they don’t track the state’s referrals.
DPS officials acknowledged in an interview that more than one agency could be taking credit for some of the same detentions because the Texas Military Department does not share with DPS the details of immigrants it refers to the federal government, and such data is not publicly available.
Featherston, the retired Texas Army National Guard senior enlisted adviser, said he believes immigrant apprehensions are “double counted.”
In another case, DPS posted on its Facebook page in September that it encountered more than 700 gang members as part of the operation. But officials declined requests to provide records detailing such arrests, saying gang affiliation “was not a metric the Department is tracking.”
And despite removing more than 2,000 charges from the arrest data credited to Operation Lone Star, DPS still includes other charges without explaining how they align with the operation’s goal of capturing dangerous criminals. (DPS disputed this characterization of the removed charges; a full explanation of our rebuttal is described in the methods section at the end of this piece.) In May, for example, troopers arrested a 20-year-old woman in Coke County, about 200 miles from the border in West Texas.
The woman was driving 9 mph over the speed limit in a no-passing zone on a rural highway. After troopers stopped her for speeding, they discovered a Ziploc bag with “loose leaf marijuana in the glovebox,” according to the arrest report.
The woman, who could not be reached for comment, does not appear to have a prior criminal record. The arrest report doesn’t note her immigration status. She was charged with possession of less than 2 ounces of marijuana.
“The whole reason for all this, you know, playing with statistics, is for optics so that the governor could get reelected. And so from that perspective, has it worked? Yes. It's worked for him,” said Gary Hale, a former chief of intelligence in Houston for the Drug Enforcement Administration who is now at Rice University’s Baker Institute for Public Policy. “But what’s the net gain? I don’t think there’s any. Zero. We really haven’t had any significant impact on migrant smuggling or drug trafficking.”
A Year Later
A year after Operation Lone Star launched, a panel of three Texas senators sought to better understand how to gauge the costly initiative’s accomplishments.
“What metrics are you using to measure success in terms of defining the arrests for which you’re responsible for, to make sure we’re using our DPS officers in an effective way?” state Sen. Juan “Chuy” Hinojosa, a Democrat from the border city of McAllen, asked DPS’ chief, Col. Steve McCraw, at a hearing on March 8.
Success could not be measured through arrest and seizure numbers alone, McCraw responded.
For the first time since the operation began, he offered a different metric: securing the border by stopping the flow of drugs and unauthorized immigrants in Texas’ 103 Border Patrol zones, one at a time. That is accomplished when each area has enough barriers, technology and law enforcement resources to “prevent transnational criminal activity,” according to DPS, which said it has met that goal in four zones that make up some parts of Hidalgo and Starr counties.
During the hearing, McCraw didn’t say how the agency knows it has secured a region. He also did not explain how DPS would be able to continue committing the resources needed to sustain that level of security. The senators didn’t ask.
“The challenge we have is when trying to decide what success looks like, is that if the numbers go up, do we claim success because we’re more efficient?” McCraw asked, adding that arrest and drug seizure statistics fluctuate. “You can’t have it both ways, you can’t be successful when the numbers go up and when the numbers go down.”
Since the start of the operation, DPS and Abbott have repeatedly touted success using arrests and drug seizure numbers. While continuing to cite the statistics, McCraw sought to minimize their significance, saying that what matters most is “not how much crime you’re enforcing. It’s the absence of it.”
By the end of that Senate hearing, lawmakers remained uncertain about the return on their multibillion-dollar investment.
“How do we know whether the amount of money was appropriate for what was needed?” state Sen. Bob Hall, a Republican from Rockwall, northeast of Dallas, asked the state’s financial analysts. “And how do we know when we’ve accomplished what we set out to do, so that we can figure out what to do next, other than just appropriate more money and then wonder what to do next?”
The question has plagued lawmakers since the first border security operation launched nearly two decades ago.
About the Data: How We Analyzed Criminal Charges Linked to Operation Lone Star
Beginning in June 2021, reporters from The Marshall Project, ProPublica and The Texas Tribune began making records requests to the Texas Department of Public Safety for data on arrests and charges associated with Operation Lone Star. The department was responsive to those requests and provided information over the course of several months, though the format and contents of the files they sent changed over time in notable ways.
DPS sent us two data releases, one in July and another in August 2021, with records of arrests and charges associated with Operation Lone Star. Those releases came as separate files from three branches of DPS. However, in November, agency officials said that these records were incomplete, only capturing one of two broad border regions. In December, they then said they had retroactively started removing charges that did not “reflect the mission” of Operation Lone Star.
From November 2021 to January 2022, DPS sent three data snapshots, each of which the department said represented the totality of its records of Operation Lone Star charges and arrests at the time the files were created. This data was organized with each charge on its own row. An arrest can include multiple charges.
What We Found
DPS emphasized that it is continuously improving how it collects and reports data for Operation Lone Star. As such, we used the latest data snapshot, from January 2022, when describing the criminal charges that the agency attributes to the operation, including how many charges were related to trespassing and how the charges were distributed geographically.
We also examined the evolving nature of the department’s record-keeping by looking at changes between the data snapshots provided to us. In comparing the first and second complete data snapshots (one provided in November 2021, the other in December 2021), we found more than 2,000 charges that had been removed from the data.
Vetting Our Findings
DPS said that our approach did not account for the fact that “each spreadsheet represents an extract from a live database, and information is subject to change.” The agency stated that our analysis “assumes that any row that does not appear exactly the same in each spreadsheet can be described as either ‘added’ or ‘removed.’”
We did not require rows to match exactly when identifying charges preserved or removed. Rows were matched using arrest IDs and charge descriptions, and we looked only at charges from dates covered in both files. For about half of the more than 2,000 charges we identified as being removed from the data, the arrest IDs for these charges were not included in the later data snapshots — for example, Thomas King-Randall’s arrest only appears in the first snapshot. For the other half, the arrest ID did appear in later data snapshots, but with fewer charges associated with it. Additionally, looking only at the number of charges in each dataset, we observed that for arrests that occured in the same time period, there were fewer charges in later data snapshots than there had been in the earlier snapshot. DPS declined to answer questions about why particular cases were removed and declined to answer many of our specific questions about the dataset.
The constantly changing nature of the database is not unique to Operation Lone Star. Methods for comparing datasets are commonly used and actively studied. It is valid to analyze changes in such databases (with the appropriate caveats) and to describe them as additions or removals. DPS itself told reporters the department “identified offenses that should be removed” in a December 2021 email about changes to Operation Lone Star data collection.