Shareholders Answered FHFA's En Banc Petition. The Best Line in the Brief Is FHFA's Own Words.

If you're new here: I'm Glen Bradford. I'm long Fannie Mae and Freddie Mac junior preferred shares and have written the full Fanniegate thesis for years. On July 24 the D.C. Circuit unanimously affirmed the $812 million judgment against FHFA, Fannie, and Freddie over the Net Worth Sweep">Net Worth Sweep. On September 8 the government asked the full court to rehear it. On September 16 the court, on its own motion, ordered shareholders to respond by October 1. I said I'd read the response and post what's in it. This is that post.

The full filing is here: Response to Appellants' Rehearing Petition, Doc. #2195952 (PDF). It is 27 pages, filed October 1, 2026, in No. 25-5113 (consolidated with 25-5121, 25-5154, and 25-5155). Hamish Hume of Boies Schiller signs for the class, with Kessler Topaz, Grant & Eisenhofer, and Bernstein Litowitz. David Thompson, Brian Barnes, and John Ramer of Cooper & Kirk sign for the Berkley plaintiffs. One brief, both sets of plaintiffs, 3,890 words. They used almost every word.

The framing: this wasn't a crisis decision

The brief opens by correcting three framing points from the petition before it gets to the law.

When the Sweep happened. The petition talks about the financial crisis. The response says the decision "was made in August 2012, when the housing market had recovered and the Companies were starting to generate record profits." It cites a June 2012 memo recording the FHFA Director saying there was "no urgency" to change Treasury's dividend because the companies "will be generating large revenues over the coming years, thereby enabling them to pay the 10% annual dividend well into the future." It cites a Treasury email from a month later showing the companies were earning more than the 10% dividend and rebuilding capital for the first time since 2008, "which Treasury preferred not to happen." From there it calls the argument that the judgment will chill decisions in a future crisis "a meritless scare tactic."

Who paid in. The preferred classes "invested over $33 billion into the Companies — $19 billion of which was at the behest of regulators during the actual financial crisis in 2007-08." In its first year the Sweep "transferred over $111 billion more to Treasury than the 10% dividend would have," and by trial it had sent Treasury "over $395 billion in value." The brief says the $812 million judgment "must be understood in this overall context."

What would have happened without HERA. The petition says shareholders "would have lost the complete value of their shares in 2008 if Congress had not enacted HERA." The response says that's irrelevant and was never proven. What was proven is that without the Sweep, shareholders "would own stock in Companies that would have well over $150 billion more in assets than they currently have."

None of this is a legal argument for denying rehearing. It's there to set the scene for the judges who weren't on the panel, and it does that well.

Argument I.A: discretion is a reason for the covenant, not against it

The petition's first argument was that Collins left no "gap" in the contracts for the implied covenant to fill, because HERA gave FHFA broad discretion to do things like the Sweep. The response says that gets state law backwards.

It quotes the panel quoting the Delaware Chancery Court: "When a party has sole discretion to make a decision, that setting provides more reason for the implied covenant to apply, not less." It adds a 2026 Delaware Supreme Court decision, Johnson & Johnson v. Fortis Advisors, which held the covenant applies "when a contract allocates discretionary authority to one party over a central aspect of the contract."

Then it turns FHFA's own petition against it. The petition admits HERA's best-interests clause specifies only "whose interests FHFA may serve," not how FHFA is to exercise its discretion. The petition also admits the covenant is displaced only "when a contract provision states how a grant of discretion is to be exercised." The response's point: the contracts here contain no such standard, so by FHFA's own test, the covenant applies. In the brief's words, "Defendants' position turns the significance of a grant of broad discretion on its head."

The three Delaware cases the petition relied on (Khan, Policemen's Annuity, Blaustein) each involved a contract that spelled out a specific constraint: a minority vote, a 75% partner vote with a good-faith requirement, a director or supermajority approval. The response says those contracts filled the gap themselves. These contracts don't.

The last part of this section is short and may matter most to the full court. It says this is "entirely one of state law," that en banc review is "reserved for only the most important federal law issues," and that the government "cite[s] no case from this Court granting en banc review on a state-law issue." That's an easy reason for an active judge who didn't sit on the panel to vote no.

Argument I.B: Collins and the "no daylight" flip

The petition's second argument was that Collins already decided FHFA "could have reasonably concluded" the Sweep served the public interest, so the jury couldn't find FHFA acted unreasonably. The response gives three answers.

Collins didn't find the Sweep reasonable. It found that FHFA could have concluded it was in the public interest, which put it within FHFA's authority and ended the APA claim at the door. The brief calls that "at least two steps removed from finding the Sweep to actually have been reasonable." Being in the public interest doesn't make an act reasonable, and Collins didn't even find that. It also points out that no arbitrary-and-capricious claim was before the Court in Collins.

The covenant asks a different question. Under the covenant, reasonableness is judged against the counterparty's reasonable expectations, not in the abstract. The jury was asked whether the Sweep "arbitrarily or unreasonably violated plaintiffs' reasonable expectations under the contract." Collins never asked that.

The government changed its argument. This is the best part of the brief. The petition now argues plaintiffs had to prove two kinds of unreasonableness, and Collins settled one of them. But in its brief to the panel, the government argued the opposite. The response quotes it:

"There is no daylight between shareholders' reasonable expectations and the Supreme Court's holding that FHFA as Conservator reasonably determined that its actions served the public interest."

So before the panel, the two tests were the same thing. Before the full court, they're two different things. The response cites King v. Palmer for the rule that en banc rehearing "would be inappropriate because [this argument] was not briefed or argued to the panel." In plain words: you don't get a second hearing to try out a theory you didn't argue the first time.

Argument II: §4617(f), waiver, and the Third Circuit

The petition's third argument was that HERA's anti-injunction clause, §4617(f), bars the judgment because damages "affect" the conservator's powers, and that the panel's ruling splits with the Third Circuit's 2018 decision in Jacobs v. FHFA. The response makes four points.

Waiver. "For the first decade of this case," the brief says, the government agreed §4617(f) did not bar contract damages. Not in the 2013 motion to dismiss, not in the 2017 appeal, and not in 2016 when this court specifically asked for supplemental briefing on possible bars to the contract claims. Then the government said §4617(f) barred "injunctive and declaratory claims," and nothing about damages. The panel declined to treat this as waived, and the response politely preserves its disagreement. Its fallback: an argument the government sat on for ten years "should cause it to be viewed with great skepticism."

The text has a history. §4617(f) copies the anti-injunction language Congress wrote for bank receiverships in FIRREA, 12 U.S.C. §1821(j). Courts have read that language for decades as not barring money damages. The brief quotes the Third Circuit itself: "Courts uniformly have held that the preclusion of section 1821(j) does not affect a damages claim." When Congress copies a phrase with a settled meaning, the Supreme Court presumes it meant to keep that meaning. "Had Congress wished to foreclose all state-law liability for breaches of contract committed by FHFA during conservatorship, it could have written HERA to do so."

There's no real split with Jacobs. Jacobs barred a disgorgement claim because it would have voided the Sweep and forced Treasury to give back the dividends. But Jacobs also cited Perry approvingly on contract damages, distinguishing the claims before it as "not claims for breach of contract." The panel quoted Jacobs in holding this was "an 'appropriate damages claim,'" and it applied the same effects test Jacobs used, which is why it barred the restitution claim and let the damages claim stand.

The Collins line is out of context. The petition leaned on Collins saying FHFA's "business decisions are protected from judicial review." The response says that sentence wasn't in Collins's discussion of §4617(f) at all. It was in the section about the constitutional challenge to the Director's removal protection. Reading that remark as quietly overturning decades of settled law "would make no sense."

And it closes with the scale. The petition argued that attaching "enormous liability" to one authorized choice changes the consequences of the conservator's choices. The response says the judgment amounts to a "potential 0.5% charge on its decision to send well over $150 billion in excess dividends to Treasury," and that there's no evidence in the record that a liability that size would have deterred anything.

The argument it doesn't take head-on

The petition's procedural hook was N.S. v. Dixon (D.C. Cir. 2025). The theory: when the Supreme Court later reads a statute more broadly than a circuit precedent did, that's a change in circuit law, and only the full court can overrule the old precedent. Here that means Collins changed things, so only the en banc court can overrule Perry.

The response doesn't cite Dixon. It answers the idea indirectly, by arguing there was nothing to change: "Collins' reasoning was the same as Perry's," Collins "reached the same result as Perry on the APA claims," and Collins "said nothing whatsoever about any contract claim." If Collins and Perry agree, there's no change for the full court to reconcile.

That's a reasonable choice with a 3,900-word limit, and it may be the right one. But it's the one place where a judge who was interested in the petition could say the government's best framing went unanswered on its own terms. I said in September that the Dixon argument was the most likely reason the court asked for a response at all. We'll find out whether "Collins agrees with Perry" was enough.

What happens next

• No reply. The September 16 order says that "absent an order of the en banc court, a reply to the response will not be accepted for filing." The last thing the judges read is this brief.
• The vote. The active judges decide whether to rehear. Judge Walker and Judge Childs, from the unanimous panel, vote. Judge Ginsburg, who wrote the opinion, is a senior judge and doesn't vote on whether to rehear. Rehearing en banc "is not favored" (FRAP 40(c)).
• Timing. There's no deadline for the court to rule. My estimate is unchanged: November or December. A denial then puts the 90-day cert deadline in February or March 2027, and that's when the judgment becomes final if the government stops.
• The meter. Post-judgment interest is still running at 5.01% on about $915 million, roughly $125,000 a day.

My read

This is a strong brief, and I don't think it changes the odds much. That isn't a contradiction. A grant was always going to come down to whether enough active judges think Perry needs revisiting after Collins, not to who wrote the better 3,900 words. What the brief does is give every judge who wants to vote no a clean, short reason: it's state law, the government changed its theory, and it sat on §4617(f) for a decade.

On September 9 I put an en banc grant at "well under one in five." On September 18 I nudged it up a little because the court asked for a response, and kept it under one in five. I'm leaving it there. The most likely outcome is still a denial. What I'll be watching for is whether a judge writes separately when it comes, because a written dissent from denial would be the roadmap for a cert petition, and that would decide whether finality lands in early 2027 or slides further.

I was wrong about September 8. I'm telling you how I weight it, not what will happen.

Disclosure

Long the junior preferred, same as I've been for years. I trimmed two Freddie series in early September for reasons unrelated to the case. This filing doesn't change what I own or why.

I hold long positions in Fannie Mae and Freddie Mac junior preferred shares. This post is my personal opinion and is not financial advice, and I am not a lawyer. Quotations are from the October 1 response, the September 16 order, the petition, the July 24 opinion, and the rules linked where cited; the odds and the estimated dates are mine. Do your own research. The full thesis is at glenbradford.com/fanniegate.