Missing Records Suggest Strained Immigration Courts

By Suzanne Monyak | November 1, 2019, 8:51 PM EDT

A Syracuse University research center’s allegations that a U.S. Department of Justice agency deleted nearly 1 million immigration case records could signal an under-resourced immigration court system overwhelmed by a growing caseload.

The Transactional Records Access Clearinghouse at Syracuse published a bombshell report on Thursday accusing the DOJ’s Executive Office for Immigration Review of “silently but systematically deleting” more than 900,000 case records and sending the research center “inconsistent” and “garbled” datasets in response to a Freedom of Information Act request.

It’s unclear whether the omissions were deliberate or a result of incompetence and database mismanagement. EOIR spokesperson Kathryn Mattingly told Law360 the records had been withheld under a privacy exemption, but TRAC said the agency did not mark where records within a file had been withheld, leaving them in the dark about what was missing.

The magnitude of the issue — whether it's limited to the EOIR's FOIA release practices or to possible inconsistencies with the EOIR’s internal database — is also unknown.

But according to Susan Long, director of the TRAC Research Center and a professor of managerial statistics, the missing records are a red flag and an “early warning that things are going south.”

“The court’s under significant stress. It takes resources. It’s not just hiring more judges; you need all the backup in order for the organization to work well,” Long said. “It may be a cry for help.”

Jeffrey S. Chase, a former immigration judge who worked under the EOIR for more than two decades, raised similar concerns. He pointed to the departure of recent officials at the agency, recent policies encouraging immigration judges to speed through cases and what he described as “the administration’s general disregard for transparency and honesty.”

Former Chief Immigration Judge MaryBeth Keller left her post earlier this year, and David Neal, former chairman of the Board of Immigration Appeals, the immigration courts’ appellate board, announced his retirement in September.

"This is far from surprising," Chase said, referencing the "oft-heard statement that it is easier to obtain funding for judges than for all of the support resources that additional judges require." 

According to TRAC, the EOIR’s FOIA office had sent over several versions of the requested dataset, a cumulative database through September 2019. TRAC sends regular FOIA requests for this information, allowing the research center to publish data tracking the immigration court backlog nationally and by state.

The backlog recently surpassed 1 million cases, according to TRAC, though the EOIR places that figure at 987,274.

In the first version, the data was “garbled” and the columns misaligned, making it difficult to accurately read, the report said. It added that in the second version, the formatting had been fixed, but 2.8 million records that had been available in the first version had no longer appeared. TRAC again flagged the issue for the EOIR, which apparently mistakenly sent the second version again.

Finally, the office sent a third version, which the EOIR spokesperson said Friday is “accurate and up-to-date.” But TRAC said that it continued to notice omissions, prompting the research center to compare it with older versions of that data.

Those comparisons revealed a stunning 896,906 applications for relief had gone missing between the September 2018 and September 2019 versions — about 1,500 of which had disappeared over the past month.

The research center said it also discovered 1,200 missing records that flag special cases — like families seeking asylum and cases of people forced to wait in Mexico while their immigration court cases continue — and 700 missing court proceedings and 900 missing scheduling records between August and September.

While acknowledging that managing such a massive amount of data is a challenging task, TRAC scolded the EOIR for failing to carry out “basic data verification procedures” that are “industry standard for agencies managing large databases.”

“The public cannot tell if records were intentionally withheld and why they were withheld, or if records were accidentally omitted during the data copying process. The failure to address these problems also means that the public has no way to test for potential problem areas in EOIR's underlying master data files,” the report says.

Mattingly told Law360 on Friday that the EOIR had withheld that information on privacy grounds "to avoid an unwarranted invasion of personal privacy, including the requested alien ZIP codes, and certain data implicating Convention Against Torture and the Violence Against Women Act."

Under the FOIA, agencies are required to indicate where there have been deletions on privacy grounds and to justify those deletions.

The omissions are “suspicious,” said Yael Schacher, a senior U.S. advocate at Refugees International, questioning why there would be so much back-and-forth between TRAC and the EOIR’s FOIA office if TRAC’s concerns could have been explained initially by a FOIA exemption.

She said she was particularly concerned about the omitted reports on "Remain in Mexico" cases, explaining that the length of time migrants spend there is relevant to the ongoing litigation over the policy.

The majority of the missing relief applications were for relief under the United Nations Convention Against Torture, which under the Trump administration’s asylum restrictions is one of the only forms of fear-based relief available to recent border crossers who passed through another country en route to the U.S.

The administration has said that the availability of CAT relief and withholding of removal, another fear-based form of relief, are enough to fulfill the U.S.’ obligations even if asylum eligibility is restricted, though the Office of the United Nations High Commissioner for Refugees has publicly said otherwise.

But the revelations ultimately raise more questions than answers.

“The problem is at this point, there are just a lot of question marks,” Long said. “We haven’t been able to get them to work cooperatively to try to clear up the question marks. So it’s a bit unclear whether the problems are in the public releases they are making versus the underlying database.”

According to Long, the omissions are part of a move away from transparency she’s noticed under the Trump administration. Historically, she said, the EOIR has had an “amazing record” of returning full and accurate data promptly in response to FOIA requests.

But lately, she said she’s seen “a change in posture.” In the spring, the EOIR started withholding county-by-county immigration case information, she said, which had been helpful in larger cities to help attorneys and pro bono legal assistance groups see where the need and demand for attorneys was.

“Having really detailed information about where people are, and therefore where the need is, is really important in terms of the public — and attorneys in particular — being able to identify where needs are, and where they’re no longer being met,” Long said.

But at the same time, the accuracy of immigration case data has taken on new importance under the Trump administration, which has leaned heavily on data to justify its policies restricting asylum and expanding immigration detention.

When announcing the rollout of the administration’s “Remain in Mexico” policy in December, former Homeland Security Secretary Kirstjen Nielsen said that foreign citizens “trying to game the system to get into our country illegally will no longer be able to disappear into the United States, where many skip their court dates.”

In a declaration filed in a California federal court in a lawsuit challenging “Remain in Mexico,” Aaron Reichlin-Melnick, a policy analyst at the American Immigration Council, crunched the EOIR’s data to find that 87.5% of asylum seekers found to have a credible fear of persecution — an early test of asylum eligibility — showed up to their immigration court hearings, which would undermine Nielsen’s justification for the policy.

Data has also taken center stage in legal challenges over immigration detention, as federal judges mull how long is too long to be detained without a bond hearing. In 2016, the U.S. solicitor general sent a letter to the U.S. Supreme Court flagging for the justices that the EOIR had provided data that had “several significant errors” regarding how long immigrants with criminal convictions spend in mandatory detention.

The issue of prolonged detention is back before the courts, with the Ninth Circuit currently considering if migrants who show they have a credible fear of persecution are entitled to bond hearings after a certain amount of time in immigration detention.

“The data has become, I think, more important now,” said Schacher. “How long somebody’s in detention, how long it takes to get somebody through the immigration court system because of the backlog, how many people show up to court has now become the linchpin of asylum policy.”

--Editing by Philip Shea and Alanna Weissman.