Congress Intelligence Surveillance & Privacy

Summary: Justice Department Inspector General Memorandum on FBI Compliance With FISA Procedures

Jeremy Gordon
Tuesday, March 31, 2020, 7:07 PM

A newly released report found errors or lost information in 29 FISA applications.

Published by The Lawfare Institute
in Cooperation With

A new report released by the Justice Department’s Office of the Inspector General found errors or lost information in all the Foreign Intelligence Surveillance Act (FISA) applications concerning U.S. persons it reviewed following its report on the FBI’s “Crossfire Hurricane” investigation into members of the Trump campaign. Each of the 29 applications sampled for review contained inaccuracies, including missing files in four FISA applications and errors or inadequately supported facts in the 25 other applications.

After finding in December 2019 that the FBI had not followed standard review procedures for four FISA applications prepared in the course of Crossfire Hurricane, the inspector general’s office initiated a larger audit, of both initial and renewal applications, to examine compliance with these procedures over the past five years. The office examined compliance with procedures designed to ensure factual accuracy—or Woods Procedures—for the 29 FISA applications, met with the case agents or supervisors responsible for the applications, and reviewed application oversight mechanisms within the FBI and the National Security Division of the Justice Department (NSD). Based on this initial review, the report states that “we do not have confidence that the FBI has executed its Woods Procedures in compliance with FBI policy.”

According to the report, Woods Procedures are intended “to minimize factual inaccuracies in FISA applications and to ensure that statements contained in applications are ‘scrupulously accurate.’” The FBI implemented the procedures in 2001 after errors were discovered in numerous FISA applications submitted to the Foreign Intelligence Surveillance Court (FISC) in FBI counterterrorism investigations.

The Report’s Findings

The report’s findings can be divided into three buckets.

1. Some FISA applications lacked required supporting documents. The Woods Procedures mandate that the FBI compile supporting documentation for each fact asserted in a FISA application, and that FBI case agents maintain the documentation in a “Woods File.” Before submitting a FISA application, both the case officer and the supervisory special agent must certify that the file contains supporting documentation for every factual assertion within the application. According to the inspector general, the FBI could not locate the Woods Files for four of the FISA applications it reviewed; in three of those instances, the report said, the FBI did not know if the Woods Files had ever existed. This failure to support the factual statements in FISA applications “undermines the FBI’s ability to achieve its ‘scrupulously accurate’ standard for FISA applications.”

2. Factual assertions in the other FISA applications did not have sufficient support. The FBI was able to locate Woods Files for the remaining 25 FISA applications reviewed by the inspector general, but when auditors tested the factual assertions in the applications against the supporting documents, they found that every application contained at least some factual assertions that lacked proper support. In some instances, the factual assertions were not supported by any material in the Woods File. In others, the assertions either were not “clearly corroborated” by the file or were inconsistent with it. According to the report, auditors identified an average of about 20 issues per application reviewed, with the number of issues per application ranging from about five to about 65.

Some of the FISA applications contained information from a confidential human source (CHS). When applications include facts attributed to a CHS, Woods Procedures require the FBI agent responsible for handling the source to include documentation in the Woods File stating that the agent reviewed facts in the application relating to the source’s reliability and background. The handling agent must also attest to the accuracy of the facts in the FISA application, based on the material in the FBI file concerning the CHS. The auditors found that about half of the FISA applications they reviewed contained facts attributed to a CHS, but many of the applications’ Woods Files lacked these required certifications. The report notes that many of the case agents handling confidential human sources had also prepared the related FISA applications and thus would have been familiar with the CHS files firsthand. But FBI policy did not waive the requirement that agents add the certifications to the file, and so this does not explain the absence of the certifications.

The report also found that FBI agents were not regularly reverifying factual assertions in initial FISA applications when they sought to renew authorization of those applications, as required by the Woods Procedures. In one instance, the report notes, errors or unsupported information in the initial application appeared in each of the renewal applications. The report states that case agents or supervisors interviewed by the inspector general’s office generally did not disagree with auditors’ findings about these deficiencies.

3. FBI and NSD oversight mechanisms had already identified similar deficiencies but were not being used to assess compliance with Woods Procedures. FBI policy requires each FBI field office to conduct an accuracy review of at least one FISA application per year, and the NSD’s Office of Intelligence separately reviews at least one FISA application each from a selection of FBI field offices. Auditors for the inspector general’s office have visited eight field offices and reviewed 34 FBI and NSD accuracy reports addressing 42 separate U.S. Person FISA applications—only one of which was also included in the 29 FISA applications the office reviewed directly.

The report states that, according to FBI and NSD officials, the accuracy reviews did not focus on compliance with Woods Procedures or the adequacy of Woods Files but, instead, sought to determine whether support existed at the time of the review for factual assertions in the applications in question. FBI field offices were given advance notice about these accuracy reviews, the report notes, and were expected to provide documentary support for assertions in the applications under review—but that evidence need not have come from an application’s Woods File.

The FBI and NSD accuracy reports identified deficiencies in 39 of the 42 FISA applications. Summaries of these accuracy reports reviewed by the inspector general’s office identified 390 individual problems, which included many of the same types of issues the office found in its own direct review of the 29 applications. The FBI and NSD accuracy reports did not deem any of the 390 issues they identified as material to the applications.

While the results of FBI and NSD accuracy reports were made available to the FBI headquarters, the report states that FBI headquarters never conducted a “comprehensive, strategic analysis of the cumulative results” to assess agents’ performance or evaluate the quality assurance mechanisms meant to ensure that FISA applications were “scrupulously accurate.” The inspector general writes that a full examination of the results of accuracy reviews would have put FBI headquarters “on notice” about shortcomings in the FBI’s compliance with Woods Procedures.

The report’s findings follow from what the inspector general describes as a limited initial review: The document does not make judgments about whether the errors or issues it identifies were material, whether the errors influenced a decision to file an application or the FISC’s approval of the application, or whether the information in the application was complete. However, the inspector general chose to release these preliminary results to help inform the FBI’s efforts to address recommendations detailed in the inspector general’s report on the Crossfire Hurricane investigation, and to provide the FBI with additional recommendations.

The inspector general’s office will continue its review of FISA applications and will expand its audit to include efforts to ensure the accuracy of FISA applications within the NSD. It will also examine the FBI’s and NSD’s FISA oversight mechanisms in more detail, including the process for ensuring that corrective action is taken when issues with a FISA application are identified. Finally, the inspector general will follow up on what the FBI is implementing regarding the FISA application process in response to the report on the Crossfire Hurricane investigation.

The report concludes with two recommendations. First, the inspector general’s office “recommend[s] that the FBI institute a requirement that it, in coordination with the NSD, systematically and regularly examine[s] the results of past and future accuracy reviews to identify patterns or trends in identified errors so that the FBI can enhance training to improve agents’ performance in completing the Woods Procedures, or improve policies to help ensure the accuracy of FISA applications.” Second, the inspector general’s office “recommend[s] that the FBI perform a physical inventory to ensure that Woods Files exist for every FISA application submitted to the FISC in all pending investigations.”

The FBI’s Response

Writing to Inspector General Horowitz, Associate Deputy Director of the FBI Paul Abbate expressed his appreciation for the inspector general’s efforts to evaluate the bureau’s compliance with Woods Procedures. He stated that the FBI is committed to procedural rigor, as demonstrated by the “sweeping corrective actions” it took in response to the report on the Crossfire Hurricane investigation, which included “foundational” FISA reforms.

Abbate’s letter emphasizes that in many areas the inspector general’s report did not turn up evidence of problems. For example, Abbate notes that the inspector general’s office did not take a position on the materiality of any of the errors it identified and did not evaluate whether support for factual assertions missing from Woods Files may have existed elsewhere, such as in case files. In areas where the report did identify problems, Abbate writes that many remedial measures the FBI began to implement after the Crossfire Hurricane report will also address process errors identified in the inspector general’s report, and states that the FBI fully accepts the report’s recommendations. Specifically, Abbate states that the FBI, in conjunction with NSD, will “build on existing accuracy reviews and enhance compliance with the Woods Procedures.” The FBI’s general counsel has directed every relevant division to ensure proper maintenance of accuracy subfiles for all FISA dockets stretching back to Jan. 1, 2015, which “exceeds the [Office of the Inspector General’s] recommendation” that the FBI do so for all pending cases.

Abbate’s letter closes by expressing gratitude for the oversight provided by the inspector general’s office to ensure that the FBI’s “indispensable” surveillance tools are properly exercised.

The Department of Justice’s Response

Associate Deputy Attorney General Bradley Weinsheimer’s response on behalf of the Department of Justice strikes a similar tone to the FBI response. Weinsheimer expresses the department’s agreement with and acceptance of the report’s recommendation and explains how reforms implemented in response to the inspector general’s report on the Crossfire Hurricane investigation would address the concerns identified here. For example, Weinsheimer points out that the FBI has revised its Woods Procedures form, developed a new checklist for dealing with confidential human sources and developed new processes to train for “rigor” in the FISA, among other steps. The FBI also now requires all Woods Procedures forms to be maintained in electronic case files.

Weinsheimer also notes that in response to the inspector general’s prior report on Crossfire Hurricane, the FBI formed a team to review and propose improvements to the FBI’s FISA processes, including Woods Procedures. And he states that the NSD will continue to examine accuracy reviews to identify patterns in errors in FISA applications and will otherwise work to ensure the accuracy of applications.

Weinsheimer concludes by highlighting FISA’s importance as a national security tool and describing the Justice Department’s and the FBI’s commitment to ensuring the integrity of the FISA process.

Jeremy Gordon is a recent graduate of the University of Virginia School of Law. He received a B.A. from the University of California, Berkeley.

Subscribe to Lawfare