Bureaucracy Times

All the news fit to be declassified.

  • The Government That Isn’t: How America Outsourced Its Regulators

    The New York Stock Exchange, as any diligent reader of an SEC filing will notice buried in the letterhead, is an LLC. Not a bureau, not a commission, not a branch of anything resembling the federal government. It is a subsidiary of Intercontinental Exchange, a publicly traded company with shareholders who would very much like a return on their investment.

    And yet the NYSE writes rules, in this case NYSE Rule 1210, that determine how long a stockbroker who fails a licensing exam must sit in the penalty box before trying again. It publishes those rules through the Federal Register. It submits them to a Commission with the word “Securities” in its name. It behaves, in every visible respect, like a government agency, because Congress decided in 1934 that it should be allowed to.

    This is the self-regulatory organization, or SRO, and it is one of the more elegant tricks in the American regulatory playbook.

    Rather than build a federal bureaucracy large enough to police every trading floor and brokerage in the country, Congress simply deputized the industry to police itself, with the government reserving a veto and a badge. The result is a hybrid creature that cashes private paychecks while wielding public authority, and that hybrid nature is exactly why nobody quite knows what to call it when asked.

    A Regulatory Agency That Turns a Profit

    The legal architecture is straightforward enough on paper. Section 6 of the Securities Exchange Act of 1934 requires securities exchanges to register with the SEC and to adopt rules governing their own members, rules that must, among other things, prevent fraud and manipulation and promote “just and equitable principles of trade.”

    The SEC’s own historical account of this arrangement traces the impulse back to a practical concession, one first voiced by an Assistant Secretary of Commerce in 1963, that direct government regulation of markets “on a wide scale” was simply too unwieldy to attempt (SEC Historical Society). Better, the thinking went, to let the exchanges regulate themselves, with Washington holding a leash.

    That leash is real. Under Section 19 of the Act, an SRO must submit every proposed rule change to the SEC, which is precisely the mechanism on display in Release No. 34-106460, the filing that prompted this piece; NYSE proposing to shave FINRA’s own qualification exam waiting periods down to match, filed under a provision that lets “immediately effective” changes take hold while the Commission merely watches from the wings (SEC Release No. 34-106460, File No. SR-NYSE-2026-44).

    The SEC can suspend an SRO rule within 60 days if it senses trouble. It rarely bothers. FINRA, the successor to the old National Association of Securities Dealers and now the primary regulator of broker dealers nationwide, occupies the same strange perch, a nonprofit corporation that nonetheless carries the statutory power to fine, suspend, and expel the very firms that fund it through membership dues.

    Cornell’s Legal Information Institute puts the arrangement in blunter terms than the SEC ever would, noting that Congress built in “more stringent oversight” precisely because lawmakers recognized the conflict of interest lurking inside the design, the fox, as the encyclopedia entry puts it, guarding the chicken coop (Cornell LII).

    The fox has so far kept the coop reasonably tidy, though the 2008 financial crisis and a steady drizzle of enforcement settlements suggest the guarding is more aspirational than absolute.

    The Robe Wears the Badge Too

    Securities markets are the flashiest example, but they are far from the only industry that gets to write its own rulebook.

    Ask a lawyer who regulates lawyers and the honest answer, in most states, is other lawyers, operating under a license from the state’s highest court. The Washington State Bar Association, to take one plainly worded example, “operates under the delegated authority of the Washington Supreme Court” to license and discipline the state’s attorneys, administering exams, maintaining records, and running the entire professional discipline system “without public funding” (Washington State Bar Association).

    California’s arrangement is stranger still. The State Bar there was established by statute in 1927 as a public corporation, yet the state’s own Supreme Court has insisted, in litigation over just this question, that the legislature’s involvement “does not alter or affect its constitutional and inherent regulatory authority over the bar” (State Bar of California).

    Nobody in Sacramento is quite sure whether the State Bar is a government agency, a private association, or some administrative changeling that shape shifts depending on which court is asking.

    The justification offered for this arrangement, at least by the people who benefit from it, is that self-regulation preserves judicial independence from the other two branches, a genuinely serious argument dressed up, not incidentally, as a professional cartel’s best defense of its own gatekeeping power.

    The American Bar Association’s Model Rules make the same case for lawyers that the SEC once made for stockbrokers, namely that the industry, not the state, should hold the whip, and that the whip will somehow be wielded more wisely for it.

    Medicine, Accounting, and the Long Tail of Delegated Trust

    The pattern repeats across American professional life with a consistency that ought to unsettle anyone who assumes a license means a government inspector signed off on it. State medical boards, composed overwhelmingly of physicians, license and discipline doctors under authority delegated by state legislatures.

    Accounting has its own version in state boards dominated by CPAs. Even arbitration, that quiet corner of civil law where most consumer and employment disputes now go to die, runs through the American Arbitration Association, itself recognized as an SRO with statutory duties under securities law (Wikipedia, Self-regulatory organization).

    In every case the pitch is identical, that professionals understand their own profession better than a generalist bureaucrat ever could, so let them write the rules, subject to a government stamp that arrives after the fact and rarely gets withheld.

    What ties the securities exchange to the state bar to the medical board is not statutory language, which varies wildly, but a shared bet that the regulated industry, given the pen, will regulate itself more competently and more cheaply than the government could manage alone.

    Sometimes that bet pays off in exactly the dull, bureaucratic fashion intended, a waiting period trimmed from 30 days to 15, filed, published, forgotten. Other times the fox remembers what it is.

    The Ledger So Far

    None of this is secret and none of it is new; the SEC’s own historians trace SRO structures back more than two centuries and treat the arrangement as a settled compromise rather than a scandal (SEC Historical Society).

    But settled is not the same as scrutinized, and an institutional structure that quietly licenses lawyers, disciplines brokers, and adjudicates arbitration claims, all while carrying letterhead that says “LLC” or “Association” rather than “Department” or “Bureau,” deserves more attention from the public it nominally serves than a stray footnote in a Federal Register notice.

    The government did not disappear when it delegated these powers. It simply changed its clothes.

    Fediverse reactions

  • USPS Said Its Lost-Mail Center Returned 39% of Packages. The Inspector General Found Less Than 1%

    The U.S. Postal Service thought its national lost-package operation was returning roughly four out of every 10 eligible packages to customers.

    Its inspector general took another look at the math. Measured against all packages arriving at the Mail Recovery Center, the return rate was less than 1 percent.

    That is one of the more striking findings in a new USPS Office of Inspector General audit of the Mail Recovery Center, the Postal Service’s Atlanta-based destination for “dead mail” — packages that cannot be delivered to their intended recipient or returned to the sender.

    The audit found that USPS was dramatically underestimating how many packages were entering the operation, using a return-rate calculation that excluded the overwhelming majority of incoming packages and relying on a spreadsheet containing a formula error that had gone undetected since 2016.

    All told, the OIG estimates roughly 36 million packages arrived at the Mail Recovery Center between October 2023 and February 2026. USPS operational figures showed only about 9 million.

    The 27 million-package math problem

    The discrepancy begins with how USPS estimated incoming package volume.

    Because every package entering the Mail Recovery Center was not individually scanned, the Postal Service estimated package counts using the weight of incoming containers. The calculation assumed an average package weighed 3.56 pounds.

    OIG examined actual scan data and found the average was closer to 0.89 pounds.

    That difference is not exactly rounding error. A 200-pound container that would be estimated to contain roughly 56 packages using the Postal Service’s assumption could contain about 225 using the OIG’s observed average.

    Applied across the operation, OIG concluded the Mail Recovery Center received approximately 36 million packages during the audit period — about four times the 9 million reflected in the center’s reported volume.

    Then came the return rate.

    Mail Recovery Center management reported returning about 39 percent of packages to customers. But the denominator used for that calculation included only 607,233 packages the center considered eligible for return, rather than the roughly 36 million packages OIG estimated actually entered the facility.

    That excluded about 98 percent of incoming packages from the calculation.

    Measured against all incoming packages, OIG calculated the customer return rate at less than 1 percent.

    The distinction was not merely academic. According to OIG, Postal Service executives cited return rates in the 30-to-40-percent range believing the figures represented all incoming packages. They did not.

    Most packages entering the center were ultimately auctioned, recycled, donated or discarded.

    And then there was the spreadsheet

    OIG also discovered an error in the Excel spreadsheet used to calculate Mail Recovery Center performance metrics.

    The formula double-counted certain successfully returned packages, further inflating the reported performance figure. According to the audit, the error dated to 2016 and remained undetected until inspectors examined the calculations.

    USPS management continues to disagree with OIG over how the return rate should be presented. The Postal Service argues the existing metric appropriately measures packages that are actually eligible to be returned. OIG argues an overall return rate is also necessary because senior management had interpreted the narrower measure as describing the entire stream of packages entering the facility.

    That recommendation remains unresolved.

    A free disposal service for commercial shippers

    The audit identified another expensive problem hiding inside the mountain of undeliverable packages.

    At high-volume facilities reviewed by OIG, roughly 75 percent of dead-mail packages came from commercial shippers that repeatedly used the Mail Recovery Center by refusing returned packages carrying postage due or otherwise avoiding taking the merchandise back.

    In effect, OIG said, the Mail Recovery Center was providing some commercial shippers with a free disposal service.

    One fulfillment center outside Los Angeles reportedly refused approximately 250 returned packages every week for three to four years rather than pay the return postage.

    OIG estimated USPS lost $109 million in return-postage revenue between October 2023 and February 2026 and calculated that a refusal fee could have generated another $54 million during the period. USPS disputed those monetary calculations, saying it could not replicate OIG’s estimates and arguing that revenue generated from reselling abandoned merchandise should be considered against processing costs. OIG stood by its methodology.

    Meanwhile, customers were waiting

    The operational problems were visible outside the accounting system.

    Missing packages generated approximately 5 million customer complaints in fiscal 2025, more than any other USPS service or product, according to the audit.

    OIG found that only 7.6 percent of packages scanned when they arrived at the Mail Recovery Center had previously received the required scan showing they were being sent there. Without that scan, customers — and sometimes postal employees — could not readily determine where a package had gone.

    Inspectors also found packages looping repeatedly through the postal network because employees failed to mark them properly. One package reached the Mail Recovery Center five separate times over nearly four months.

    As of May, the center also had a four-month backlog of damaged packages awaiting processing. One package mailed from Cleveland on Dec. 19, 2025 — apparently intended as a Christmas shipment — finally reached its customer May 11, 143 days later.

    USPS agreed or partially agreed with eight of OIG’s nine recommendations, which include improving package-volume measurements, strengthening required scanning, addressing repeat commercial refusals and improving performance reporting.

    The audit ultimately describes something larger than a warehouse struggling with lost Christmas presents. The Postal Service had a national operation handling tens of millions of undeliverable packages, but its own measurement system understated the volume by roughly 27 million packages and produced a success rate senior executives misunderstood.

    For an operation devoted to finding things that went missing, the Mail Recovery Center appears to have temporarily misplaced its denominator.

    Fediverse reactions

  • GAO Measured the Delay; A Judge Had Already Called It Illegal

    The Numbers GAO Was Willing to Count

    The Government Accountability Office released a report this week on FEMA‘s Building Resilient Infrastructure and Communities program, the agency’s flagship pre disaster mitigation grant, and the numbers it counted are damning enough on their own terms. As of March 2025, FEMA had awarded 1,245 BRIC subgrants worth about $2.5 billion, which sounds respectable until you notice that figure is half of the $4.8 billion Congress actually made available. Of what had been awarded, FEMA had reimbursed $62 million and closed out exactly 37 subgrants. The median time for FEMA to finish reviewing a subapplication and award the money, GAO found, is 7 to 9 months. Another 700 subapplications, tied to roughly $2.2 billion, were sitting in the second of FEMA’s two review stages with no decision made at all.

    GAO quotes communities saying, in the mild register that federal audit reports reserve for people who are furious but trying to stay eligible for future funding, that these timeframes “could extend project timelines and increase costs.” The report’s seven recommendations are the standard genre of fix; identify efficiencies in the subapplication review, apply lessons learned to communication, establish performance goals. All reasonable. All aimed at a process problem.

    Here is the sentence in the report that is doing more work than it appears to be doing. From April 2025 to March 2026, FEMA awarded no subgrants and obligated no funds. Eleven months, stated as a fact about program administration, sitting inside a report about review timelines.

    What Happened Between April and March

    That eleven month gap was not FEMA finishing its paperwork slowly. It was FEMA terminating the program.

    On April 4, 2025, FEMA formally ended BRIC, cancelling pending applications going back to fiscal year 2020 and moving to redirect the money elsewhere, a move detailed in the federal court’s later ruling on the matter. BRIC is not a discretionary line item FEMA can turn off by memo. It is funded through direct congressional appropriations under the Infrastructure Investment and Jobs Act and set asides from the Disaster Relief Fund under the Stafford Act, which is precisely what a coalition of states argued when they sued that July.

    On December 11, 2025, U.S. District Judge Richard Stearns agreed with them, and did not agree gently. He ruled that FEMA’s termination violated the Further Consolidated Appropriations Act of 2024, the constitutional separation of powers, the Appropriations Clause, and the Administrative Procedure Act, and issued a permanent injunction ordering the program restored. FEMA did not restore it. Two months later, the same states had to go back to court and ask a judge to force compliance with his own order, which the Washington state attorney general’s office documented in March 2026, when Judge Stearns issued an enforcement order giving FEMA 21 days to publish a new funding notice. FEMA published it on March 25, 2026, three weeks past the deadline it had already been given twice.

    So the eleven months in GAO’s report is not a measurement of bureaucratic friction. It is the interval between an unlawful act and a federal court forcing its reversal, with an extra two months tacked on for the agency’s noncompliance with its own defeat. GAO’s highlights page does not use the words termination, lawsuit, or injunction. It uses the word timeframes.

    The plaintiffs were not abstractions. Maryland’s attorney general noted that the termination left more than $80 million in BRIC funding for the state in question, including flood mitigation work for Crisfield and South Baltimore. Arizona’s attorney general listed 25 selected projects worth $9.8 million, among them a $4.6 million infrastructure project in Buckeye. These were not requests still being drafted. They were projects that had cleared FEMA’s review and were waiting on a check, from a government that had already told the public the check was cancelled.

    FEMA’s Version of the Same Year

    FEMA’s own account of this period, issued as a public notice to states, tribes, and territories, tells a different story than the one a federal judge told. It explains that BRIC “became bogged down in bureaucracy, focused on climate change initiatives, and riddled with inefficiencies” under the previous administration, which is why Secretary Kristi Noem “directed FEMA to conduct a methodical evaluation of the program.” The notice adds that FEMA will resume normal operations “once the lapse in appropriations has ended,” folding the government shutdown into the same sentence as the agency’s own program suspension, as though the two were the same kind of interruption.

    Nowhere in FEMA’s telling does a judge find that the agency broke the law. The “methodical evaluation” and the illegal termination occupy the exact same calendar days; FEMA simply gets to narrate them as a virtue rather than a defeat. This is the closest thing this story has to a punchline. A federal court says the government violated the separation of powers to shut down a program Congress funded. The agency responsible calls it stewardship.

    Reapply, Please

    The 700 subapplications GAO found stalled in review as of March 2025 were caught in that stage precisely when the termination hit. The program that eventually reopened is not, procedurally, the one they applied to. FEMA’s own fact sheet for the restored funding opportunity states plainly that “previous subapplications submitted for the initial Fiscal Year 2024 BRIC funding opportunity will not be reviewed,” and instructs applicants to “review, update, and resubmit in accordance with this new funding opportunity.”

    Communities that had already cleared FEMA’s first review stage, in some cases years earlier, were told their applications no longer exist. Whatever work went into a benefit cost analysis, a scope of work, a local match commitment, has to be redone against new evaluation criteria FEMA wrote after Congress had already appropriated the money for the old ones.

    None of GAO’s seven recommendations addresses this, because GAO’s mandate is process efficiency, not legality, and by the time its auditors were writing, the termination was already adjudicated elsewhere. What the report actually documents, whether it means to or not, is what happens to an audit’s own numbers when a program spends nearly a year not existing. The 7 to 9 month review time GAO flags as the problem to fix was, for the 700 communities caught in it, the least of their delays. They were not waiting on FEMA to finish reading their file. They were waiting on FEMA to admit the file still counted.

    Sources: Government Accountability Office, Building Resilient Infrastructure and Communities: FEMA Should Improve Subgrant Review Timelines (GAO-26-107774, September 2026); Washington v. FEMA, U.S. District Court, Memorandum and Order (December 11, 2025); Washington State Attorney General, Court Order Enforcing BRIC Ruling (March 2026).

  • The Department Started Fixing Its Data, Then Fired the People Fixing It

    Late in 2024, the Department of Education did something almost unheard of in federal data governance. It admitted a flaw in one of its own signature statistics and started building the fix. Officials began requiring states to report the number of students enrolled in a school across the full school year, the same timeframe already used to count chronically absent students, instead of the single frozen October snapshot that had been producing chronic absenteeism rates north of one hundred percent. It was a small, boring, correct decision. Then the Department gutted the office responsible for finishing it, and decided the fix wasn’t worth finishing after all.

    A Reorganization With a Body Count

    In March 2025, the Department announced a reduction in force and reorganization. The numbers, drawn from Office of Personnel Management data cited in GAO’s report, are not subtle. The Institute of Education Sciences, the statistical and research arm responsible for the EDFacts data collection that produces chronic absenteeism figures, went from 191 staff at the end of calendar year 2024 to 28 staff at the end of 2025, a reduction of more than 85 percent. The Office of Elementary and Secondary Education, which officials told GAO serves as data steward for the chronic absenteeism collection specifically, lost 43 percent of its workforce over the same period.

    Officials were direct with GAO about the sequence of events. They began the enrollment fix in late 2024. They later decided not to proceed, “due to staffing changes at the agency.” That sentence, delivered without elaboration, is doing an enormous amount of work. It is the Department’s own explanation for why a known, documented flaw in a nationally reported education statistic will keep producing implausible numbers indefinitely. Not because the fix was hard. Because the people assigned to do it no longer worked there.

    The Burden That Was Already Being Carried

    What makes the abandonment particularly hard to defend on the merits, as opposed to the staffing chart, is how little the fix would have actually cost anyone. States are already required to identify which students were enrolled in a given school for at least 10 days, because that is the threshold used to determine who counts as chronically absent in the first place. GAO interviewed an outside expert on chronic absenteeism data who made the obvious point plainly, since states are already identifying and collecting these enrollment figures internally to comply with an existing requirement, reporting that same number to EDFacts, rather than the separate October 1 snapshot, would add minimal additional burden.

    This is not a case of the Department discovering a fix that would require new infrastructure, new surveys, or new cooperation from fifty-one state agencies that would rather not be bothered. The data was already sitting in state files. The only missing step was an internal decision, at the Department, to accept it as the new denominator and update the collection specification accordingly. That decision got made, reversed, and shelved, all inside about a year, and the reversal had nothing to do with the states.

    It is also worth noting what this same reorganization did not spare. The same Department that could not staff a straightforward denominator fix also suspended its consolidated monitoring of ESEA fiscal and programmatic requirements across all fifty one states in 2026, the subject of a separate finding in this same GAO report. These are not two unrelated cost-cutting decisions that happened to land in the same document. They are two downstream consequences of one upstream event, a reduction in force large enough to leave the agency unable to sustain either its oversight function or its data quality commitments, at the same time, in the same fiscal year. When an agency loses eighty five percent of a statistical office’s staff, it does not get to treat the resulting gaps as a series of isolated, defensible tradeoffs. It has one explanation, repeated in two different sections of the same audit.

    A Strategic Plan That Reads Differently in Hindsight

    The Department’s own Strategic Plan for fiscal years 2022 through 2026 identifies strengthening agencywide data governance as a stated goal, and names improving the quality of data collected from grantees and subrecipients as one of the strategies for getting there. GAO quotes this language back at the Department in its report, not unkindly, but not gently either. An agency does not get credit for a data governance strategy it wrote down and then failed to staff.

    The Department’s public defense of the current approach, offered in its formal response to GAO’s recommendations, does not mention the reduction in force at all. It says only that the Department has “chosen to continue working with currently reported EDFacts data elements, highlighting limitations as needed, rather than imposing additional reporting burden on states.” That framing casts the decision as a considered tradeoff between data quality and administrative burden. GAO’s own reporting elsewhere in the same document shows it was not a tradeoff at all. It was a project that started, ran into a staffing cliff, and stopped, and the public explanation retroactively describes the stopping point as a choice.

    There is a version of this story where an agency looks at its own reduced capacity, decides which existing commitments it can no longer keep, and says so plainly to Congress and the public. There is another version where the agency keeps the sentence about “highlighting limitations as needed” and quietly drops the sentence about staffing changes into an interview with an auditor, where it will only surface if someone reads a forty page report closely enough to notice the connection. The Department picked the second version. GAO just happened to be reading closely, and the version of events available to anyone who does not read forty page audit reports, meaning most parents, most school board members, and most of Congress, is still just the tidy line about choosing not to impose additional burden on states.

    Source: U.S. Government Accountability Office, K-12 Education: Actions to Improve Oversight of Key Federal Programs and Address High Chronic Absenteeism (GAO-26-107920, September 2026). Related: Eighteen Billion Dollars, and Nobody Assigned to Watch It and A Chronic Absenteeism Rate of 125 Percent Is Still an Official Federal Statistic.

  • Federal Register Watch: October 8, 2026

    The federal government proposed charging international students as much as $70,000 for a single work authorization today, the most eye-catching item in a 133-document Federal Register that also includes a Customs and Border Protection overhaul of how low-value packages enter the country, a pair of quiet deregulatory moves from the Indian Health Service, and the Army Corps of Engineers walking back a 23-year-old promise of rigorous environmental review for a Louisiana canal project.

    • Homeland Security: Homeland Security proposed charging international students on F-1 visas $70,000 for their first work authorization under Optional Practical Training and $30,000 for every one after that, including STEM extensions. The fee has to be paid by the student’s school before a school official can even enter the authorization into the federal tracking system, and the department says plainly that without new revenue it may have to shut the program down entirely. Comments are due by November 9, giving universities, employers, and students roughly a month to weigh in before the fee can be finalized. Read the proposal.
    • Customs and Border Protection: CBP proposed replacing the old postal entry process for low-value shipments with an all-electronic “Informal Mail Entry” category, adding tracking-number and sender-identity requirements along with a bonding requirement that low-value importers haven’t needed before. The rule is CBP building the plumbing for a post-de-minimis world: the $800 duty-free exemption for small packages has been suspended since August 2025 and is set for permanent repeal in 2027, so this is what replaces the old light-touch system once that exemption is gone for good. Comments are due December 7, and CBP’s own estimate puts the net cost at roughly $9.3 billion over the next decade, a number commenters now have something concrete to push back on. Read the proposal.
    • Indian Health Service: IHS finalized two separate rules on the same day to strip old language out of its section of the federal regulatory code, one about billing non-beneficiaries for care, the other covering provisions the agency says have been duplicative or formally “suspended” since a 1999 notice. Neither rule went through a public comment period, both take effect December 7, and the agency insists current practice won’t change either way, which leaves the question of why clearing obsolete language off the books took two separate rulemakings instead of one, and why it sat there for decades before anyone bothered. Read the first rule and the second.
    • Army Corps of Engineers: The Corps withdrew a notice of intent it issued back in 2003 to prepare a full Environmental Impact Statement, the government’s most searching form of environmental review, for deepening the Houma Navigation Canal in Louisiana. In its place, the Corps says a shorter Environmental Assessment is now the appropriate level of review, citing a 2020 change to the project’s legal authorization. An EIS typically means years of detailed study and extensive public input, while an EA is a faster process meant for actions not expected to cause significant impact, so swapping one for the other after sitting on the more rigorous review for 23 years is a real downgrade in scrutiny, and the notice offers no comment period on the switch itself. Read the notice.
    • Institute of Museum and Library Services: IMLS, the small federal agency the administration’s own budget request has proposed eliminating, published a request for comment on a brand-new grant program aimed at small history museums, historical societies, and historic sites. Agencies generally don’t start the months-long process of standing up a funding opportunity, which this notice is the first formal step toward, unless they expect to still be around to award the money, so a new program announcement from an agency slated for closure is worth watching to see whether it survives long enough to actually make a grant. Read the notice.

    This edition draws on the full Federal Register issue for October 8, 2026, which ran to 133 documents. Categories screened out as routine include airworthiness directives, antidumping and countervailing duty determinations, a cluster of nearly identical Securities and Exchange Commission self-regulatory-organization filings tied to exchanges adopting nearly round-the-clock trading hours, routine EPA state implementation plan approvals, Native American Graves Protection and Repatriation Act notices, advisory committee meeting announcements, and standard OMB/Paperwork Reduction Act information-collection notices.

  • The Paperwork Costs More Than the Pipe

    Imagine two identical tribal water projects, same pipe, same pump, same size crew, funded through two different federal agencies. One costs the tribal health consortium administering it about $15,000 in extra paperwork to complete. The other costs nearly $100,000. The physical infrastructure is the same. The only variable that changed is which federal agency’s money paid for it, and specifically, whether that agency had already bothered to sign a standardized agreement with the Indian Health Service. According to a new Government Accountability Office report, that gap is not hypothetical; it is a real cost estimate from a tribal health consortium that runs water programs in Alaska, and it is one of the clearest illustrations in the entire document of how interagency plumbing can cost more than the plumbing itself (Government Accountability Office, GAO-26-107178).

    One Agreement, Two Outcomes

    IHS routinely works with EPA to jointly fund tribal water projects, and over time the two agencies have built something useful, a standard interagency agreement, or IA, with agreed upon terms and conditions that both sides already understand (GAO-26-107178). When a project moves through that channel, the paperwork is largely settled in advance. USDA, by contrast, does not often contribute to projects already listed in IHS’s own project tracking database, and IHS area officials and USDA state office staff told GAO there is often little to no collaboration between the two agencies on the same tribal water projects (GAO-26-107178). When USDA money does end up funding pieces of an IHS project, it typically arrives through a different, less standardized channel, with the tribe itself acting as project owner, negotiating USDA’s application requirements directly and retaining responsibility for administering the funds on its own (GAO-26-107178).

    That difference in plumbing between agencies produces a difference in cost that has nothing to do with the water system being built. The tribal health consortium that administers Alaska’s water program estimated that using funding sources without a standard IA in place, USDA being the example GAO cites by name, can add close to $100,000 in additional administrative costs to a single project (GAO-26-107178). The comparable figure for an EPA-funded project moving through the existing standard IA with IHS was about $15,000 (GAO-26-107178). Same category of project. A roughly sevenfold difference in overhead, driven entirely by whether the funding agency and IHS had already done the paperwork once, in advance, at the program level, instead of leaving each individual project to reinvent it.

    Where the Extra Ninety Thousand Actually Goes

    It would be easy to wave this off as an abstract accounting exercise, so it is worth being specific about where that nearly six figure gap comes from. According to the tribal health consortium’s own cost breakdown, the most significant drivers are the application process itself, the development and legal review of individual, one off project agreements negotiated from scratch, spending and billing procedures that differ from IHS’s own systems, and the ongoing coordination and reporting required to keep two agencies’ bookkeeping in sync across the life of a project (GAO-26-107178). None of that is water infrastructure. None of it is engineering, materials, or labor on the actual system tribal members will eventually drink from. It is entirely the cost of two federal bureaucracies figuring out, one project at a time, how to talk to each other and hand off money.

    GAO is explicit that most of these costs are the kind that a standardized program level agreement could eliminate; the consortium’s own estimate flags application processes, individual agreement development, billing procedures, and reporting coordination as exactly the categories that standardized terms and conditions are designed to address (GAO-26-107178). In other words, nearly all of the $85,000 difference between the USDA path and the EPA path is not an inherent cost of moving federal water infrastructure money to a tribe. It is the cost of doing that negotiation over and over, at the individual project level, because no template exists.

    The Fix Already Exists, Just Not for USDA

    The frustrating part of this story is that GAO is not asking anyone to invent something new. IHS and EPA have already proven a standardized IA works; it is the reason EPA funded projects only run up about $15,000 in overhead rather than $100,000. GAO’s recommendation is essentially to extend that same template to USDA. GAO recommends that IHS, working with USDA’s Under Secretary for Rural Development, develop program level standardized interagency agreements, with agreed upon terms and conditions, for USDA’s Native American and Water and Waste Disposal programs, and use those agreements whenever tribes request that USDA funds be transferred to IHS (GAO-26-107178). Where legally permissible, GAO suggests those standard terms include advance lump sum payments and the removal of duplicative USDA administrative requirements, mirroring what already works in the EPA relationship (GAO-26-107178).

    USDA’s own response to this recommendation is worth sitting with. The agency neither agreed nor disagreed, noting instead that it was already working on state level interagency agreements and that expanded use of those might be possible (GAO-26-107178). GAO pushed back on that framing directly, arguing that state by state agreements do not solve the underlying problem, since USDA needs an effective tool it can leverage nationwide, not a patchwork negotiated separately in every state where a tribal water project happens to sit (GAO-26-107178). A program level agreement, done once, would apply everywhere USDA and IHS work together. Fifty separate state level versions of the same negotiation simply relocates the inefficiency instead of removing it.

    What a Sevenfold Markup Actually Buys

    Step back from the specific numbers and the pattern here is almost embarrassingly simple. Two federal agencies can jointly fund the same category of tribal water project for wildly different amounts of overhead, and the only variable is whether they had already agreed, once, on how to work together before the money started moving. IHS and EPA did that homework years ago and the administrative cost dropped accordingly. USDA has not, and every tribal water project that touches USDA money pays a tax for that gap, a tax that shows up nowhere in a press release about funding awarded, but shows up very precisely in a tribal health consortium’s own internal accounting.

    Ninety thousand extra dollars per project does not buy better water. It buys more lawyers reviewing individual agreements, more staff hours reconciling two accounting systems that were never designed to talk to each other, and more time before anyone actually turns on a tap. GAO has already shown, using IHS’s own relationship with EPA, exactly what the alternative costs. USDA’s job now is to stop treating that as optional. The overhead problem is one of three structural failures documented in the same GAO report. Even before the interagency funding question arises, many of these systems are built under a forty-year-old statutory prohibition that bars IHS from connecting water to homes built with HUD funds — leaving tribes to choose between homes without water and fewer homes altogether. And even when a system does get built and funded, nearly half of Alaska Native village water systems had no certified operator on staff in fiscal year 2025 — a direct consequence of IHS funding construction but not the ongoing salaries needed to run what it builds.

    Fediverse reactions

  • DOJ Rewrites the Rules on Whistleblower Suits, Right After Quietly Killing Four Times as Many of Them

    The Justice Department announced on September 18 that it was revising two sections of the Justice Manual governing the False Claims Act, the statute the government has used since 1986 to recover more than $85 billion from contractors, hospitals, and grant recipients who lied to Uncle Sam. Associate Attorney General Stanley E. Woodward Jr. described the changes in suitably poetic terms: “The Department of Justice should enforce the law, not make law through enforcement.”

    A fine sentiment, and one that would carry more weight if the policy in question were not primarily a formal blessing for something DOJ was already doing at quadruple its previous rate.

    This is not the Department’s first pass at the Manual, and it won’t be the last; the document is a living internal rulebook, revised whenever an administration wants to signal, in writing, how it intends to point the FCA’s considerable weaponry.

    What makes this revision worth a second look is not its stated content but its timing, arriving four and a half months after a senator started asking pointed, on-the-record questions about exactly the authority the Manual now blesses.

    The Manual Gets a Rewrite

    The first revision reinstates the 2017 “Brand Memo” doctrine, which holds that sub-regulatory guidance (the informal circulars, bulletins, and advisory letters agencies issue between actual rulemakings) cannot create legal obligations that don’t already exist in statute or regulation.

    This is presented as a fair notice measure, to wit, a company should be liable for breaking the law, not for failing to intuit an agency’s unpublished preferences. Reasonable enough, and DOJ is right that guidance documents have a way of acquiring the force of law without ever going through the trouble of becoming law.

    The second revision is the one that matters. It directs DOJ attorneys to affirmatively consider seeking dismissal of qui tam suits (the whistleblower-initiated cases that make up the bulk of FCA litigation) whenever the Department declines to intervene, and to keep revisiting that option as the case proceeds. This authority comes from 31 U.S.C. § 3730(c)(2)(A), and the Supreme Court blessed its broad use in the 2023 case United States ex rel. Polansky v. Executive Health Resources.

    Assistant Attorney General Brett Shumate framed the update as keeping “enforcement efforts remain aligned with those interests” of the United States, which is bureaucratic for “we intend to throw more of these cases out.”

    The Numbers DOJ Would Rather You Learned From a Senator

    Here is what the press release does not mention. Senator Chuck Grassley, the Iowa Republican who wrote the 1986 amendments strengthening the FCA in the first place, sent a letter to Shumate on April 2 asking pointed questions about dismissal practices. In it, Grassley noted that DOJ dismissed roughly 25 qui tam cases under § 3730(c)(2)(A) in 2025, compared to an average of about six per year under the prior administration.

    That’s not a policy adjustment; that’s a quadrupling, delivered months before the Manual itself caught up to reflect it. Deputy Assistant Attorney General Brenna Jenny made the posture explicit at an industry conference in January, telling a room full of FCA defense lawyers that “(c)(2)(A) is back.”

    None of this is happening in a vacuum of litigation. FY2025 was a record year for whistleblowers by every measure that counts, with 1,297 new qui tam suits filed, up from 980 the year before and 712 the year before that, nearly doubling in two years. Those suits generated $5.3 billion of the $6.8 billion the government recovered under the FCA in FY2025, or roughly 78 percent of the total, a figure DOJ’s own Civil Division touted as a historic high the same month Grassley was asking why the Department was dismissing whistleblower cases at four times the prior rate.

    The Department is, in other words, simultaneously bragging about relators bringing in most of its money and building the institutional machinery to throw more of their cases out. Both statements can be true. Neither makes the other look good.

    It’s worth noting where that money actually comes from. Health care accounted for roughly 84 percent of FY2025 recoveries, north of $5.7 billion, which means the cases most exposed to a more dismissal-friendly Department are precisely the ones involving hospital systems, pharmaceutical manufacturers, and Medicare and Medicaid billing practices, the same institutions that tend to have the resources to make a dismissal motion attractive to a Department weighing its limited litigation bandwidth against a case’s “alignment with the interests of the United States.”

    The Article II Question Nobody in the Press Release Mentions

    There’s a reason DOJ might want its dismissal muscle memory in good working order right now, and it has nothing to do with fair notice. The qui tam mechanism itself is under constitutional siege. In Polansky, three Supreme Court justices (Thomas, Kavanaugh, and Barrett) suggested in a concurrence and dissent that there are “substantial arguments” the entire relator-driven structure violates Article II, on the theory that private citizens cannot represent the interests of the United States in litigation the Executive Branch hasn’t chosen to bring.

    A Florida federal district court subsequently agreed and held the qui tam provisions unconstitutional outright; that ruling is now sitting with the Eleventh Circuit.

    A more aggressive dismissal posture is one of the cleanest ways for DOJ to reassert that it, not some private relator’s contingency-fee attorney, controls whether a False Claims Act case lives or dies, regardless of how the Eleventh Circuit eventually rules on the underlying constitutional question.

    Revising the Justice Manual to formalize that posture, quietly, four months after a sitting senator started asking uncomfortable questions about it, reads less like a course correction and more like an agency building the paper trail to justify a practice it had already adopted.

    What We’re Requesting

    Grassley’s 25-dismissals figure and DOJ’s own “six per year” baseline both trace back to the same letter and the same set of conference remarks. Neither is an independently verifiable dataset, and DOJ has not published a public accounting of which qui tam cases it moved to dismiss in 2025, on what grounds, or with what success rate in front of the actual district courts hearing them.

    Bureaucracy Times is filing a Freedom of Information Act request with the Civil Division’s Commercial Litigation Branch for the underlying dismissal log, including case names, filing dates, the grounds cited under § 3730(c)(2)(A), and the disposition of each motion. If DOJ’s new posture really is about “fair and effective enforcement,” as Associate Deputy Attorney General Paul Perkins put it, the underlying record should hold up to daylight.

    We’ll report back on what, if anything, we get.

    Fediverse reactions

  • Pentagon Spent Decades and Billions Hardening GPS. The Military Still Isn’t Ready to Lose It.

    A $6.27 billion replacement GPS control system was canceled this year, aircraft receivers remain years away, and GAO says fragmented Pentagon management is hampering efforts to protect U.S. forces from a threat already demonstrated in Ukraine.

    WASHINGTON — For more than two decades, the Pentagon has been trying to solve a problem that has since stopped being hypothetical.

    America’s military depends heavily on the Global Positioning System to navigate ships, aircraft and vehicles, coordinate communications and guide precision weapons. Adversaries have meanwhile demonstrated that GPS signals can be jammed, spoofed or otherwise disrupted.

    The Department of Defense has spent tens of billions of dollars trying to make that system harder to defeat and developing alternatives for when it is.

    According to a new Government Accountability Office report released Tuesday, considerable work remains.

    GAO describes GPS as the military’s principal source of positioning, navigation and timing information, but also warns that the Pentagon’s dependence on it creates a potential “single point of failure.” The department has spent more than 20 years developing a more secure and jam-resistant military GPS signal known as M-code, yet modernization has repeatedly encountered delays across satellites, ground-control systems and the equipment necessary for weapons and military platforms to actually use the signal.

    The stakes are no longer confined to acquisition schedules.

    Russia has used GPS jamming during the war in Ukraine to degrade some U.S.-provided precision-guided munitions, according to GAO, while both Russia and Ukraine have used spoofing to interfere with legitimate GPS signals.

    Research cited by GAO found intensified GPS jamming in eastern Ukraine reduced the effectiveness of one type of GPS-dependent precision-guided munition by as much as 80 percent. Another munition remained disrupted as much as 40 percent of the time even after countermeasures were employed.

    The Pentagon is therefore racing not merely to improve GPS, but to ensure American forces can continue operating when GPS is unavailable.

    It is doing so while still trying to finish a modernization effort that began around the turn of the century.

    Satellites Since 2005, Receivers Still Coming

    The military’s M-code capability depends upon three pieces working together: satellites capable of transmitting the signal, a ground system capable of controlling the constellation, and receivers installed in weapons and military equipment capable of using it.

    The first M-code-capable GPS satellites have been in orbit since 2005.

    Development of the Next Generation Operational Control System, known as OCX, began in 2010. Development of newer receiver equipment began in 2017.

    Previous GAO investigations have documented schedule delays reaching a decade or longer and cost increases reaching billions of dollars.

    The government watchdog was raising alarms about OCX by at least 2015, when it reported persistent cost and schedule growth, a high defect rate and concerns that unrealistic estimates were limiting Pentagon oversight. GAO’s 2015 report is available here.

    Eleven years later, the Pentagon pulled the plug.

    The Space Force formally accepted OCX from contractor Raytheon in July 2025 following years of development and testing. But acceptance came with waivers and deviations related to technical shortcomings.

    Among them was an unmet requirement involving “time-steering,” the mechanism used to maintain accurate time dissemination through the GPS constellation. Pentagon testing officials also considered the system’s cyber survivability a high-risk area requiring continued mitigation and investment.

    Integrated testing then uncovered still more problems.

    Space Force officials later said the difficulties spanned numerous capability areas and potentially put existing military and civilian GPS capabilities at risk.

    By January 2026, the government had spent approximately $6.27 billion on OCX, including Raytheon funding and government testing and program-support costs.

    Three months later, DOD canceled it.

    Space Force concluded that attempting to complete OCX posed greater risk than continuing to upgrade the GPS control system that OCX had been designed to replace. The Space Force’s account of the termination is available here.

    The Pentagon had effectively spent years and billions developing a replacement for its old GPS control system, accepted that replacement with unresolved deficiencies, and then concluded that modernizing the old system was the safer option.

    The Replacement for the Replacement

    DOD is now pursuing what it calls Architecture Evolution Plan Modernization, or AEP-M, a multiphase modernization of the existing GPS Operational Control Segment.

    But changing direction carries consequences of its own.

    The Space Force had planned to use OCX capabilities to launch, check out and operate the next generation of GPS IIIF satellites. Those satellites and their software were being prepared around OCX.

    Now the existing control system has to be modified to handle them instead.

    GAO says the decision creates schedule risk for incorporating the new satellites into the GPS constellation. The first ten GPS IIIF satellites were already averaging nine months of delay against a revised December 2024 delivery schedule as of September.

    The first satellite was projected for delivery in July 2027, while ground-control capability is needed in time to support planned launches beginning in 2028.

    It is another dependency in a modernization program full of them.

    And satellites that can broadcast M-code are only useful if American weapons can receive it.

    Aircraft Remain a Problem

    GAO found that development and fielding of M-code receiver equipment continue to face software, funding, technical and integration problems, particularly aboard aircraft.

    Integrating the new receivers into aircraft requires tying them into flight software and satisfying airworthiness requirements, making aviation among the most difficult pieces of the modernization puzzle.

    One Air Force program, the Embedded GPS Inertial Navigation System-Modernized, or EGI-M, lost one of its contractors after the company failed to provide what officials considered a realistic updated development schedule.

    The remaining contractor underestimated the complexity of software associated with security and airworthiness certification, according to GAO.

    Funding shortages created another problem: the contractor could not maintain the staff necessary to complete development on schedule.

    Production decisions for EGI-M variants used by the Navy’s E-2D and Air Force F-22 are now estimated for mid-2028. A decision on the F/A-18 variant is projected for 2030.

    Initial EGI-M fielding is expected in March 2032.

    The Navy, meanwhile, is developing an interim receiver to provide M-code capability to aircraft while it waits for the longer-term system.

    There has been progress elsewhere. The Army and Navy are fielding M-code-capable and multi-PNT receivers in some ground vehicles, ships and Army aircraft, and all the services are pursuing systems capable of combining GPS with alternative navigation information.

    Those alternatives are becoming increasingly important.

    When GPS Goes Dark

    DOD is building what amounts to a navigation toolbox for a battlefield where GPS can no longer be assumed.

    Some systems would obtain positioning signals from satellites in low-Earth orbit. Others could navigate without satellites at all.

    Inertial systems measure a vehicle’s direction, acceleration and orientation. Visual-navigation systems compare camera imagery against stored maps. Other technologies navigate by comparing measurements against maps of Earth’s magnetic or gravitational fields.

    Still others look upward.

    Modernized celestial-navigation systems can determine location using stars, planets and even artificial satellites whose orbits are known.

    The Pentagon is also deploying backup timing systems, including regional clocks and increasingly small atomic clocks, because disruption of GPS can affect far more than knowing where a vehicle is located. Loss of GPS timing can interfere with military communications and networking.

    Relatedly, a previous Bureaucracy Times investigation examined how the Pentagon audits the clocks that underpin GPS timing and what details remain classified.

    GAO’s illustration on page 22 depicts the intended result: GPS satellites operating alongside low-Earth-orbit constellations, celestial navigation, inertial sensors, visual navigation, magnetic and gravity navigation and alternative timing sources while adversaries attempt jamming, spoofing, cyberattacks and direct attacks against satellites.

    The technology exists in various stages of maturity.

    Getting all of it into the right weapons at the right time appears to be another matter.

    Everyone Is in Charge

    That may be the most consequential finding buried inside GAO’s highly technical report.

    Congress created a Positioning, Navigation and Timing Oversight Council to coordinate DOD’s sprawling PNT enterprise. But individual military services retain authority over the budgets funding their respective programs.

    The council can establish priorities and make recommendations.

    The services control the money.

    GAO found that this fragmented structure makes it difficult to synchronize programs that depend upon one another.

    A receiver card might be developed by one organization. Another program develops the receiver containing it. A military service then has to integrate that receiver into an aircraft, ship or vehicle. Munitions fired by those platforms may operate on yet another modernization schedule.

    The result can be technologically sophisticated pieces arriving at different times.

    Internal meeting records reviewed by GAO show that the problem has not gone unnoticed.

    Officials discussed inadequate Air Force funding for aviation receiver programs, and program officials attributed missed development goals partly to “chronic underfunding.”

    At a May 2025 PNT Oversight Council meeting, one official identified what may be the simplest description of the organizational problem: there was a need to “designate a single owner to the entire PNT value chain.”

    DOD has since consolidated several aviation receiver and M-code efforts into an Air Force-managed joint program office. GAO found that consolidation has produced closer reviews of costs, schedules, technical risks and performance and could serve as a model elsewhere.

    But the broader enterprise remains fragmented.

    GAO concluded that under the existing structure, DOD “lacks reasonable assurance” that its PNT investment decisions are addressing pressing warfighting needs.

    Two Decades of Warnings

    The new report does not arrive in isolation.

    GAO warned in 2021 that M-code remained years from widespread use and that receiver delays were producing ripple effects across weapons programs. Read GAO-21-145.

    In 2022, auditors found that Pentagon databases used to manage GPS modernization contained incomplete or inaccurate information. DOD was ultimately planning to integrate modernized GPS equipment into roughly 700 weapon systems, making reliable schedules and coordination particularly important. Read GAO-22-105086.

    In 2023, GAO again reported ground-system delays and questioned other portions of the modernization strategy. Read GAO-23-106018.

    And in September 2024, GAO reported that significant challenges remained across the space, ground and user-equipment segments of the modernization effort. Read GAO-24-106841.

    Two years later, the ground system examined in those reports no longer exists as a development program.

    GAO’s latest report puts the broader history in unusually stark terms: DOD has worked for more than two decades and spent tens of billions of dollars laying the foundation for fully modernized GPS and complementary navigation systems, yet “considerable work remains” before those capabilities are fielded.

    GAO issued three recommendations calling on the Army, Navy and Air Force to assess where GPS modernization and complementary PNT programs should be consolidated and to develop consolidation plans where doing so would improve strategic alignment.

    DOD agreed with all three recommendations.

    The Army maintains that its modernization efforts are already well coordinated and that it has found no duplication, but told GAO it will participate in a broader Pentagon assessment and develop a consolidation plan where appropriate.

    The Navy and Air Force likewise agreed to examine consolidation.

    The technology race, meanwhile, is unlikely to wait for the organizational chart.

    The Pentagon began developing a GPS signal resistant to jamming more than two decades ago. The first satellites capable of transmitting it reached orbit in 2005. Since then, wars have supplied real-world evidence of what happens when precision weapons encounter an enemy capable of interfering with satellite navigation.

    The United States now possesses satellites broadcasting the hardened signal, weapons waiting for modernized receivers, alternative navigation technologies under development and a new program to upgrade the old ground-control system after abandoning its $6.27 billion replacement.

    What it still does not have, according to GAO, is assurance that all those pieces will arrive where they are needed when the next battlefield starts interfering with GPS.

    Sources

    Fediverse reactions

  • The IRS Lost $304 Billion to Fraud. Its Best Defense Was a Mission Statement.

    A Number Too Big to Be Comfortable

    Somewhere between $116 billion and $304 billion vanishes from the federal treasury every year, and it does so not through some singular, cinematic heist but through the accumulated, unglamorous churn of stolen identities, falsified deductions, and return preparers who treat the tax code as a suggestion.

    That is the range the Government Accountability Office arrived at in a new report, built from a Monte Carlo simulation stitching together adjudicated fraud cases, detected fraud that never made it to court, and the vast, unlit expanse of the shadow economy.

    Put another way, somewhere between two and six percent of what Americans owe the federal government simply disappears into fraud each year, an amount that could fund a meaningful slice of veterans’ health care or national security spending, if only anyone at the IRS had been assigned to stop it.

    Here is the part that should trouble you more than the number itself. The IRS has spent years assessing its fraud risks with real diligence, in keeping with GAO’s own Fraud Risk Framework. It knows where the exposure sits. It has documented, biannually, a fraud risk profile that identifies individual income tax fraud as one of its highest exposures, and it flagged, as far back as 2023, that new tax credits arrive faster than the fraud controls meant to guard them.

    Diagnosis, in other words, is not the problem. The problem is that having correctly identified the disease, the agency has declined to write a prescription.

    The Excuses, Collected Like Specimens

    GAO went looking for the IRS’s antifraud strategy, the coordinated document that leading fraud-risk practice says should translate a risk assessment into an actual plan of controls.

    It does not exist, at the agency level or within any of the divisions responsible for chasing fraud. What GAO got instead, when it asked division officials why, reads less like institutional analysis and more like a set of alibis assembled after the fact.

    Some officials believed the agency’s Office of the Chief Risk and Control Officer already owned the antifraud strategy. That office told GAO, on the record, that no such strategy exists there either. Other officials had simply never identified a specific need for one.

    Others still offered up their division’s routine audit activity as though conducting fraud detection were the same thing as having a documented strategy for it, a distinction that GAO’s own Fraud Risk Framework treats as fairly foundational. And at least one division pointed to its mission statement, the kind of aspirational paragraph that adorns an agency homepage, as the functional equivalent of a plan to stop nine and ten figure annual losses.

    None of these things are strategies. A mission statement does not specify how existing and new controls will address a risk the agency has already flagged as high. Standard operating procedure is not the same as a coordinated response to an evolving threat. GAO says as much, plainly, and notes that nobody at IRS could tell them who was actually supposed to own this problem, what the plan was, or when it might materialize.

    Recommendation one asks IRS to write the strategy down.

    Recommendation two asks it to name someone whose job that is.

    Partial Agreement, Fully Argued

    IRS’s written response to the draft report is where this gets genuinely entertaining, in the grim way that bureaucratic self-defense often is.

    The agency “partially agreed” with both recommendations, which is administrative language for agreeing just enough to avoid the appearance of stonewalling while conceding nothing structural. It pointed to its Chief Tax Compliance Officer as the entity coordinating fraud initiatives, though GAO notes this designation arrived only in response to being asked. Even then, it does not cover the full list of responsibilities a genuine antifraud entity is supposed to hold, from managing the risk assessment process to serving as the agency’s institutional memory on fraud controls.

    IRS also pushed back on the report’s framing, arguing that GAO had blurred the line between fraud and ordinary taxpayer noncompliance, and that judging its fraud posture through a narrow lens missed the fuller picture of its compliance framework.

    GAO’s rebuttal is the sharpest passage in the document. It reiterates, patiently, that its definition of fraud tracks the IRS’s own definition, that its estimate specifically excludes nonfraudulent noncompliance like taxpayer error, and that it interviewed officials across essentially every division with fraud responsibility, not merely the criminal investigators.

    Then it lands the number IRS handed it itself. The agency’s own projected net tax gap for 2022 was $606 billion. Between that figure and GAO’s fraud estimate, the report notes dryly, additional action seems warranted regardless of how the taxonomy gets sliced.

    What a Strategy Actually Buys

    Strip away the institutional throat clearing and the case here is almost embarrassingly simple. The IRS knows, with real precision, what its highest fraud risks look like. It has a Return Review Program that legitimately prevented an estimated $88 billion in fraudulent refund payments from 2018 through 2024, proof that targeted controls work when someone builds them.

    What it lacks is the connective tissue between knowing and doing, the single document and single office that would let those controls scale to match risks the agency itself already identified years ago and simply left unaddressed.

    GAO’s closing argument is worth sitting with, because it reframes what “strategy” even means in this context.

    A pay-and-chase model, where the agency waits for fraud to happen and then tries to claw the money back through audits and collections, is fundamentally mismatched against fraudsters who are, by definition, trying to avoid exactly that kind of detection.

    Proactive controls, designed and coordinated ahead of the fraud rather than in response to it, are cheaper and more effective, but they require someone to actually own the coordinating. Absent that person and that document, the IRS is left doing what it has always done—identifying the fire, describing the fire in exhaustive and well organized detail, and then waiting for someone else to decide who holds the hose.

    Fediverse reactions