Each Monday, The Investigative Journal surveys the audits, inspector general findings, and document releases that entered the public record over the previous week. The week of August 3–9 was headlined by a Government Accountability Office review of the Department of Government Efficiency’s “Wall of Receipts,” which found that a substantial share of the initiative’s claimed savings could not be verified against documented agency actions. Below, we group the week’s notable releases by category, link to every source document, and flag the records that merit deeper reporting.
Federal Audits and Watchdog Reports
GAO: DOGE “Wall of Receipts” savings claims need more transparency
The week’s most consequential release came Thursday, August 6, when the GAO published GAO-26-108615, “DOGE Wall of Receipts: More Transparency Needed on How Savings Are Derived from Contract, Grant, and Lease Terminations.” According to the report, DOGE’s public savings tracker reported $110 billion in savings across contracts, grants, and leases as of July 7, 2026 — but GAO found that “some savings estimates are incorrect or lack supporting evidence.”
The audit’s specifics are worth reading in full (the complete report is posted here). GAO’s review indicates that DOGE did not use its own stated methodology to calculate the majority of savings associated with contracts reported as terminated, and did not provide sufficient information to verify the method behind 96 percent of reported grant savings. On leases, the records show that 108 of the 264 leases listed for termination — roughly $15.3 million of $53.5 million in claimed lease savings — were already in the termination pipeline before DOGE was established. In one example cited by auditors, DOGE reported $1.7 billion in savings on a Defense Health Agency IT contract that, in the end, was never terminated, reduced in scope, or defunded; GAO concluded “no savings were achieved” on that item.
Notably, GAO’s recommendation is modest and procedural: that the Wall of Receipts prominently display its data limitations, consistent with federal transparency practices for reporting government data. The report does not assess the broader merits of the efficiency initiative, and readers should distinguish GAO’s documentation findings from any conclusion about total savings actually achieved, which the report indicates cannot be fully verified either way from the public tracker. The report page includes the full text and agency context for those seeking DOGE’s side of the ledger.
DOJ Inspector General: victim assistance grant audits in Nevada and North Dakota
On August 6, the Justice Department’s Office of the Inspector General posted two grant audits. The first, Report 26-078, examined Office of Justice Programs victim assistance funds subawarded by the Nevada Division of Child and Family Services to United Citizens Foundation, Inc., of Las Vegas. According to the OIG’s report page, the audit identified $290,553 in questioned costs and issued 13 recommendations — a notably high count for a subrecipient audit of this size.
The companion release, Report 26-077, audited victim assistance grants awarded to the North Dakota Department of Corrections and Rehabilitation in Bismarck and produced three recommendations with no questioned costs listed. Both audits are part of the OIG’s ongoing cycle reviewing Victims of Crime Act fund flows through state administering agencies — a pipeline through which billions of federal dollars reach local nonprofits with limited public scrutiny. Grantees are afforded the opportunity to respond during the audit process, and their responses are incorporated in the report appendices.
VA Inspector General: homeless program oversight in Portland, facility inspections in Kansas and Minnesota
The Department of Veterans Affairs OIG released a cluster of healthcare oversight reports late in the week, all posted to its public reports database. The most substantive, Report 25-04102-182 (August 6), reviewed facility leaders’ oversight of care coordination within homeless programs at the VA Portland Health Care System in Oregon. According to the report’s published recommendations, case managers did not consistently comply with required monthly veteran contacts, and staff did not always discharge veterans from the Health Care for Homeless Veterans program in accordance with the facility’s case management workflow guide. The OIG issued five recommendations, two of which were closed by the publication date.
The same week brought healthcare facility inspections of the VA Wichita Healthcare System in Kansas (Report 25-00250-210, August 7) — which recommended workflows for communicating test results to providers and patients — and the St. Cloud VA Health Care System in Minnesota (Report 25-00252-204, August 7), along with a desk review of the U.S. VETS single audit reporting package (Report 26-02642-227, August 7). Individually these are routine; collectively they form a running public map of where VA facility-level compliance is slipping.
State and Municipal Records
Texas OIG evaluates UnitedHealthcare’s Medicaid fraud unit
The Texas Health and Human Services Office of Inspector General issued Report AUD-26-029 on August 4, an audit evaluating the effectiveness of UnitedHealthcare’s special investigative unit in detecting and investigating fraud, waste, and abuse in Texas Medicaid managed care, and in reporting accurate information on its activities and overpayment recoveries to the OIG. The report is listed in the Texas OIG’s audit reports database.
Managed care organizations’ in-house fraud units are a chronically under-examined control point: states delegate front-line fraud detection to the same companies whose payments are at stake. Texas has been unusually systematic here — its OIG has published a series of SIU audits covering other carriers, including Blue Cross and Blue Shield of Texas — and the UnitedHealthcare installment lands as the insurer’s Medicaid operations face scrutiny in multiple states. We have not yet reviewed the full findings; the report and any UnitedHealthcare response merit close reading.
Ohio Auditor of State: semiweekly audit batch flags recovery findings in Lorain
Ohio Auditor of State Keith Faber’s office released its semiweekly batch of audit reports on Tuesday, August 4, covering more than two dozen public entities across the state, per the office’s audit advisory. Most are routine financial audits of townships, libraries, and county governments — but the advisory flags a City of Lorain financial audit for calendar year 2023 carrying an “FFR” designation, the office’s marker for findings for recovery, indicating auditors identified public money owed back to the treasury.
Ohio’s semiweekly release rhythm makes it one of the more transparent state audit shops in the country, and the FFR flag is a reliable lead generator for local accountability reporting. The Lorain audit, alongside county-level audits in Highland and Holmes counties released the same day, can be pulled from the auditor’s audit search portal.
FOIA and Transparency Litigation
FBI projects a 158-year processing timeline for “hidden room” records
In FOIA litigation news, the watchdog group Judicial Watch reported on August 1 that the FBI, in filings in the group’s Freedom of Information Act lawsuit, projected it would take approximately 158 years to process records related to a so-called “hidden room” of documents at issue in the case, with a status conference before U.S. District Judge Dabney L. Friedrich held the following week. We note the characterization comes from the plaintiff advocacy organization; the underlying court docket is the authoritative record, and the FBI’s filings speak for themselves as to the basis for the estimate.
Whatever the litigation’s merits, agency processing-time projections of this magnitude are themselves public records of significance. FOIA backlog mathematics — where agencies propose page-per-month rates that push disclosure past every requester’s lifetime — has become a recurring flashpoint in federal courts, and judges have increasingly pushed back on multi-decade schedules. The docket in this case bears watching.
New Government Data
BLS July employment data lands with downward revisions
On the data front, the Bureau of Labor Statistics released the Employment Situation for July 2026 on Friday, August 7. According to the release, nonfarm payroll employment declined by 23,000 in July while the unemployment rate held at 4.1 percent, and BLS revised the prior two months down by a combined 103,000 jobs. The full statistical release includes the sector detail: data show declines concentrated in local government education, leisure and hospitality, and retail, with health care continuing to add jobs.
For records purposes, the revisions matter as much as the headline number. Benchmark and monthly revisions are the mechanism by which the statistical record corrects itself, and the size and direction of recent revisions have made BLS methodology itself a subject of public debate. The underlying data tables are available directly from BLS for independent analysis.
What Warrants a Closer Look
Three threads from this week’s records merit deeper TIJ investigation. First, the GAO’s DOGE findings invite a contract-by-contract reconciliation: the report identifies categories of unverifiable savings, and matching Wall of Receipts line items against the Federal Procurement Data System and USAspending records is a tractable, documents-based project. Second, the Texas OIG’s UnitedHealthcare SIU audit joins a multi-state pattern of questions about Medicaid managed care fraud referrals; obtaining the SIU audits from other large states via records requests would support a comparative analysis. Third, the DOJ OIG’s Nevada victim assistance audit — 13 recommendations and nearly $300,000 questioned at a single Las Vegas nonprofit — raises the question of how the state administering agency monitors its other subrecipients; Nevada’s subaward lists are public records and can be requested.
Editor’s note: Every item above is drawn from, and linked to, the originating public record. Where findings are characterized, the language of the source document controls. Entities named in audit reports generally respond during the audit process, and their responses appear in the linked reports; where a named entity’s response is not yet public, we will update this roundup upon receipt. Nothing here should be read as an allegation of wrongdoing beyond what the cited records state.