This is The Marshall Project’s Closing Argument newsletter, a weekly deep dive into a key criminal justice issue. Want this delivered to your inbox? Sign up for future newsletters.
Like previous government shutdowns, the one that ended this week didn’t halt federal agencies’ work across the board. Some things were paused; others kept chugging along.
Immigration enforcement, for example, was largely unaffected. The Trump administration shuffled funding to pay federal law enforcement officers, including those working for Immigration and Customs Enforcement and Customs and Border Protection. This allowed enforcement surges to continue in places like Chicago, where federal agents shot a woman and arrested another woman at a day care, a location that had traditionally been off-limits. Agents also targeted Portland, where costumed protestors faced off with authorities outside an ICE facility. Washington County, part of the Portland metro area, declared a state of emergency over immigration enforcement.
But while the raids continued, the flow of information about them did not. Throughout the shutdown, researchers, journalists, activists and the rest of the public could not access data on these enforcement activities — including the number of arrests and the average daily population of detention facilities — that federal agencies are required by Congress to release. The shutdown also stalled the release of data through public records lawsuits.
It has been more than seven weeks since ICE last published a spreadsheet of aggregate detention statistics. Typically, these are published every two weeks and offer a range of data, including snapshot counts of people in detention, people booked into ICE custody and people removed from the U.S. It also includes the average daily population of facilities where ICE has detained immigrants.
The last release on Sept. 25 showed nearly 60,000 people in detention. It also showed that ICE and CBP had booked more than 310,000 people into custody since October 2024, and deported a similar number during that period.
The data, while imperfect, is invaluable to understand the dramatic changes in immigration enforcement under the second Trump administration. Breakdowns by criminal history, for example, allowed The Guardian to analyze the numbers and find that ICE now detains more people without a criminal record than with one. At The Marshall Project, in collaboration with Univision, we used the data to reveal how ICE is placing a growing number of immigrants in solitary confinement.
The contents of the data release have been required by law since 2020. Adam Sawyer, director of research at Relevant Research, said this has created a collection of data that’s up-to-date and spans multiple administrations.
Sawyer and his collaborators developed a website that gathers average daily population statistics for detention facilities from the ICE data, makes it easily searchable and shows the change over time.
“So now we're kind of walking blind,” Sawyer said. “We can only figure out what the population is at the detention centers in a very piecemeal fashion.” That’s a problem as ICE has expanded its network of detention facilities — including local, state, federal and privately-run facilities. A local jail in Missouri may hold someone arrested recently in the state, while a privately-run facility in Louisiana may be that person’s last stop before deportation. Without the ICE data, it’s difficult to contextualize population changes at any one facility, Sawyer said.
Detention statistics aren’t the only immigration enforcement numbers missing. Data that researchers have sought through Freedom of Information Act requests has been stalled. Graeme Blair, co-director of the Deportation Data Project, an effort by academics and lawyers that compiles government data, said the project’s FOIA litigation was suspended during the shutdown, and that full or partial closure of agency FOIA offices has slowed the release of data and records.
The latest batch of the project’s data only extends through the end of July so it doesn’t capture records about more recent enforcement operations.
Through records requests, the project has obtained granular details about people that immigration authorities have encountered, arrested, detained and deported. Because the project allows the public to trace an individual throughout their time in ICE custody, the records provide richer detail than the counts ICE releases on its own website. The project’s data has been widely used by journalists and researchers to uncover dynamics in enforcement, including how hundreds of people deported this year have only minor criminal offenses that often occurred years in the past.
Local governments have used the Deportation Data Project’s records to push for accountability. New York City Mayor Eric Adams cited the project’s data in his request for an inspection of a building that ICE claimed was not a detention facility, even though people were being held there for more than four days at a time. In Portland, city officials relied on the data when issuing a land use violation to ICE for detaining people for more than 12 hours.
Phil Neff, research coordinator at the University of Washington Center for Human Rights, said the shutdown paused settlement discussions in a FOIA lawsuit the center filed. The suit seeks the continuous release of detailed apprehensions forms, reports of deaths and other significant detention incidents and solitary confinement data.
“Transparency is especially important at times like today when we have an administration that uses data to tout their successes,” said Aaron Reichlin-Melnick, senior fellow at the American Immigration Council, a pro-immigration advocacy group. Without recent data, Reichlin-Melnick said it’s difficult to know whether the Department of Homeland Security's claims of record-high deportations and arrests are truthful.
A line-item from ICE’s spreadsheets, like the number of deportations per fiscal year, “lets us fact-check them using their own data and lets us understand to cut through the maze of propaganda and braggadocio,” Reichlin-Melnick added.
While ICE’s data has been absent during the shutdown, journalists, community members and court documents have tried to fill some of the gaps. Los Angeles news outlet LA Taco has been publishing daily accounts of immigration operations across Southern California, many from social posts by witnesses, often with tallies of arrests. Student journalists at the University of Chicago and Loyola University Chicago published maps of immigration enforcement in neighborhoods around their campuses, including counts of arrests in some incidents. Local news coverage has also surfaced snapshot detention numbers at individual facilities.
Court records have also enumerated arrests in some places. For example, a federal court order in Illinois counted more than 1,800 arrests through early October that could have violated a court-monitored agreement limiting warrantless immigration arrests. This week, the judge ordered a review of more than 600 arrests in which immigrants may be eligible for release. Chicago residents have also taken to social media to track ICE sightings and arrests, though Facebook took down one such page at the request of the Justice Department.
Some internal numbers have also filtered out of DHS. CBS News, citing internal DHS records, reported that the number of immigrants in detention reached an unprecedented 66,000 last week. David Bier, the director of immigration studies at the libertarian think-tank Cato Institute, said DHS documents he reviewed showed a similar number.
Because federal law requires the release of detention statistics, Reichlin-Melnick said he expects those publications to resume now that the shutdown has ended.
The Marshall Project asked ICE about the availability of data during the shutdown first in mid-October, and again last week, but the agency has not responded.
Though the return of data should help fill in the gaps about ICE’s operations over the past few months, some information could be permanently lost. Small facilities, such as county jails that only held a few immigrants for short periods during the shutdown, may not appear in new data. Because of the delay, it will also be harder to spot any sudden spikes in detention at particular facilities during recent weeks.
Even when data returns, there will still be less known about immigration detention and deportation than before the second Trump administration. Numbers from the Office of Homeland Security Statistics, such as data that shows the different legal processes the government uses to send someone out of the U.S., are not covered by the same laws that make detention reporting mandatory, and has not been updated since Trump took office. ICE’s arrest statistics dashboard ends with data from 2024, despite a note that says it will be updated quarterly.
Those seeking to understand immigration enforcement have long struggled to get data, and records that can be turned into data — but when they do, it can lead to policy changes. Geoff Boyce, an assistant professor of geography at University College Dublin, fought for years to get records on Border Patrol arrests. Those records were the basis for a report, published by the ACLU of Michigan in 2021, that found Border Patrol disproportionately apprehended Latino people compared to their share of Michigan’s population. The report helped push state officials to put limits on state police detaining someone for suspected immigration violations.
“If you don't have hard data to show the scope of it,” Boyce said, “I think it becomes really easy for those policymakers … to dismiss what's happening.”