The Sibling That Stayed Public
Article three in this series covered the 1968 restructuring that split the original Fannie Mae in two, spinning off a privatized, shareholder owned Fannie Mae while creating a separate entity that stayed fully government owned. That second entity, the Government National Mortgage Association, or Ginnie Mae, is the subject here, and it is worth stating plainly what makes it different from everything else this series has covered: Ginnie Mae’s guarantee is not implicit, not merely assumed, not a market expectation resting on decades of federal behavior. It is explicit. Ginnie Mae securities are the only mortgage backed securities in the country carrying the full faith and credit guaranty of the United States government, a distinction the Congressional Budget Office has drawn directly against Fannie Mae and Freddie Mac’s merely effective federal support. Congress did not leave this one to implication.
Ginnie Mae does not buy or own mortgages the way Fannie Mae and Freddie Mac do. It guarantees the timely pass through of principal and interest payments on securities backed by loans that other federal agencies, principally the FHA, the VA, and USDA’s Rural Housing Service, have already insured or guaranteed, according to the same CBO analysis. The actual credit risk on the underlying mortgages sits with those insuring agencies. Ginnie Mae’s job is narrower and, in principle, cleaner: make sure investors get paid on time, every time, because the United States government said they would.
Now Standing on Nonbank Ground
Here is where the clean design runs into a genuinely current problem. Since the 2008 financial crisis, traditional banks have steadily pulled back from mortgage origination and servicing as post crisis capital and safety and soundness rules made the business less attractive, and nonbank mortgage companies filled the gap. In 2014, nonbanks serviced about 33 percent of loans in Ginnie Mae backed securities by dollar value, according to a 2025 GAO report on nonbank mortgage risk. By December 2024, the concentration had become extreme: 88.97 percent of Ginnie Mae’s roughly 2.6 trillion dollar guaranteed portfolio was held by just 30 mortgage companies, 22 of which were nonbank institutions, according to an ongoing HUD Office of Inspector General audit of Ginnie Mae’s portfolio.
A separate 2022 industry analysis found nonbanks already accounted for more than 87 percent of all Ginnie Mae originations, describing Ginnie Mae as woefully under resourced and extremely vulnerable to the failure of a large issuer or servicer, according to a HousingWire opinion piece by risk management professor Clifford Rossi.
The risk profile of a nonbank mortgage company is genuinely different from a depository bank’s. Nonbanks typically run a monoline business model concentrated entirely in mortgages, rely more heavily on short term funding lines that can dry up fast in a downturn, hold riskier mortgage servicing rights on their own books, and are not subject to the federal safety and soundness regulation, capital ratios, and liquidity requirements that apply to banks, according to the same HousingWire analysis. If one of the handful of large nonbank issuers that now dominate Ginnie Mae’s portfolio were to fail, Ginnie Mae would need to step in and either advance the missed payments itself or transfer the servicing to another company, fast enough that investors relying on the full faith and credit guarantee never notice a gap.
The Drill Nobody Wants to Run Twice
Regulators know this is not a hypothetical concern. The Financial Stability Oversight Council formed a Nonbank Mortgage Servicing Task Force with FHFA and Ginnie Mae to study exactly this risk, and in October 2023 the task force ran its first ever tabletop exercise simulating the failure of a major nonbank mortgage servicer, according to GAO’s review of the task force’s work. FSOC’s own May 2024 report, cited in the same GAO review, recommended continuing that kind of exercise given the financial stability risks nonbank servicers now pose. As of that GAO report, the October 2023 exercise remained the only one the task force had actually conducted, and FHFA officials told GAO they were still working through basic coordination issues, including how a failing servicer’s portfolio would even get transferred to a new one given that Ginnie Mae and the GSEs operate under different transfer rules.
The most pointed finding comes from HUD’s inspector general. A March 2026 audit found that despite the sharp rise in nonbank concentration and FSOC’s own 2024 warning about it, Ginnie Mae does not formally assess whether that concentration risk actually threatens its own operations, internal controls, or ability to meet its mission, according to the HUD OIG’s audit findings. The agency backing 2.6 trillion dollars in mortgage securities with the explicit, unconditional guarantee of the United States government has not, in its own watchdog’s assessment, formally evaluated what happens to that guarantee if one of the handful of companies now holding most of its portfolio goes under.
The Honest Guarantee’s Hidden Floor
Every other entity in this series eventually had to confront an implicit guarantee nobody wanted to admit was real. Ginnie Mae is the opposite case, and arguably the more instructive one: it is the single most honestly designed entity in this series, the one place where Congress said out loud exactly what taxpayers were on the hook for, with no accounting fiction and no ambiguous conservatorship required to activate it. And that honesty has not translated into caution about what actually sits underneath the guarantee. The explicit backstop remains only as good as the handful of thinly capitalized, lightly regulated companies now responsible for the daily mechanics of paying it out, and the agency whose entire job is standing behind that promise has, by its own watchdog’s account, not yet formally measured how much risk it is actually carrying. The unexamined-risk problem is not unique to Ginnie Mae. The Commodity Credit Corporation operates a $30 billion line of credit that Congress has not revisited since the mid-1980s, with GAO documenting miscalibrated payouts years after the money was already spent. The Postal Service’s pension gap is now large enough that GAO says the agency could run out of cash by 2027 — a liability Congress will have to absorb one way or another. And the Army Corps of Engineers offers a direct parallel: an agency that rejected a GAO recommendation to better document its own decision-making, leaving Congress without the information it needs to evaluate the costs it is being asked to authorize.
