ADI 7873: What STF Precedent Signals About EC 136 — and Precatorio Pricing

When Constitutional Amendment 136/2025 tightened Brazil's precatorio regime, the market repriced the asset class broadly: longer expected duration, lower carry, and a larger entry discount. ADI 7873—the Direct Action of Unconstitutionality in which the Federal Council of the Brazilian Bar Association (CFOAB, hereafter the OAB) asks the Supreme Federal Court (STF) to strike down core provisions of the amendment—then began to be treated by many market participants as a single reversal trigger: if the challenge succeeds, the prior framework returns.

That is too coarse an analytical framework. The OAB's complaint challenges seven different parts of the amendment, and the STF's case law does not treat them alike. This article organizes what the precedents actually say, what the institutional submissions now request, and what each issue means for investors pricing Brazilian precatorios.

A precatorio is a final, court-ordered payment claim against a Brazilian public entity that enters a constitutionally regulated budget and payment queue. A methodological qualification is also essential: this is not a statistical probability exercise. There are too few comparable rulings for that. What is possible is a forecast by analogy—identifying patterns in closely related cases while stating the relevant distinctions.

The Seven Issues in ADI 7873

The OAB's September 2025 complaint challenges:

1. moving the annual cut-off date—the deadline for a presented precatorio to enter the following year's budget—from April 2 to February 1;
2. an annual payment cap for states, the Federal District, and municipalities, set on a sliding scale of 1% to 5% of Net Current Revenue (Receita Corrente Líquida, or RCL), the fiscal revenue measure used to calibrate each public entity's limit;
3. the elimination of a final deadline for clearing the outstanding stock of overdue precatorios;
4. direct settlements without an express discount cap, under which a creditor accepts a discount in exchange for earlier payment;
5. the cessation of accruals charged to the public entity once the amount is deposited into a special account controlled by the Judiciary (paragraph 30), after which the claim accrues at the bank-account remuneration rate;
6. a new formula for non-tax precatorios outside the constitutional grace period: IPCA inflation adjustment plus 2% simple annual default interest, capped at the Selic rate—a combination of inflation adjustment and default compensation that must be analyzed separately. Tax claims remain subject to Selic; and
7. application of the new regime to precatorios already registered in the payment queue.

The Chamber of Deputies and the Federal Senate defended the amendment in full. The Office of the Advocate General of the Union (AGU) asked the Court to reject the challenges to the cut-off date, settlements, post-deposit accrual treatment, and the inflation-and-interest formula—but did not state a concluded position on the payment cap and proposed a public hearing.

On direct settlements, the AGU argues that the 40% discount limit in Article 102, paragraph 1, of the Transitional Constitutional Provisions Act (ADCT) continues to apply to the special payment regime, but disputes that a general 40% cap is already settled law for the new paragraph 29. The National Council of Justice (CNJ) went further in its administrative implementation: CNJ Provision 207/2025 treated discount percentages as freely negotiable and stated that the former maximum limit had been superseded.

In an opinion dated June 25, 2026, the Office of the Prosecutor General of the Republic (PGR) recommended partial relief on exactly two points: limiting the payment cap to public entities demonstrably unable to pay under the previous regime, and subjecting all paragraph 29 settlements to the constitutional order of preference and the maximum 40% discount established by the STF in 2015.

Four Constitutional Cycles, One Recurring Pattern

Since 2000, four constitutional amendments or sets of amendments have structurally reorganized precatorio payments: EC 30, EC 62, ECs 113/114, and now EC 136. Each reached the STF. Between the second and third cycles, ADI 5348 reinforced the Court's case law on inflation adjustment.

• EC 30/2000 (ADIs 2356 and 2362, merits concluded in 2023–2024): the Court rejected mandatory payment of the outstanding stock in installments over as many as ten years. None of the ten justices who reached the merits accepted the moratorium without restriction; the disagreement concerned technical issues such as mootness and the scope of invalidation.
• EC 62/2009 (ADIs 4357 and 4425, merits decided in 2013): the special regime, based on small revenue percentages and lacking an adequate final deadline, was struck down 6–5. The Reference Rate (TR) was rejected as the inflation-adjustment index by an 8–3 vote because it did not preserve purchasing power. In the Court's 2015 ruling on transition arrangements, direct settlements were preserved subject to the order of preference and a maximum 40% discount.
• ADI 5348 (2019): the Court again invalidated TR, this time by a 10–1 vote, for government judgments before precatorio issuance.
• ECs 113 and 114/2021 (ADIs 7047 and 7064, November 2023): the federal payment cap was allowed only for 2022, the exceptional fiscal year associated with the pandemic, and removed thereafter. In the same judgment, the Court unanimously upheld Selic as a single accrual index and the earlier April 2 cut-off date. It also unanimously invalidated the earmarking of borrowing proceeds exclusively for discounted settlements because it pressured creditors and undermined the payment queue.
• EC 136/2025 (ADI 7873): the fourth cycle is now before the STF, with a distinct subnational design and seven challenged issues that should not be analyzed as an indivisible package.

The pattern is not hostility to fiscal adjustment. It is a dividing line: fiscal planning that preserves payment of the claim survives; structural deferral that shifts the cost of government imbalance to creditors with no defined repayment horizon does not—or survives only in narrowed form.

One frequently misstated vote requires precision. The EC 113/114 payment-cap provision was formally decided 9–1, but the dissenting justice, André Mendonça, would have invalidated the entire article and therefore gone further than the majority. In substance, the Court was 10–0 against maintaining the cap after 2022. The formal tally understates the consensus.

The Current Court—and the Limited Weight of Its Prior Votes

As of July 28, 2026, the STF has ten sitting justices and one vacancy following Justice Luís Roberto Barroso's retirement in October 2025. Nine of the ten current justices participated in ADIs 7047/7064, and all nine opposed continuation of the federal payment cap after the acute phase of the pandemic. Only Justice Flávio Dino, who had not yet joined the Court, has no comparable vote. The same nine justices participated in the conclusion of the EC 30 litigation, and none accepted that moratorium intact.

This is the strongest indicator in our study: the current Court has a recent and substantively unanimous record against an ordinary precatorio payment cap outside an exceptional context.

The same record does not, however, guarantee the result. EC 136 is not a copy of the 2021 federal cap. It applies to subnational entities, uses permanent RCL-calibrated bands, increases the percentages over time, and purports to link fiscal capacity to amortization of the outstanding stock. Those differences leave room for intermediate remedies: upholding the framework only for entities that prove payment incapacity, requiring a verifiable path to full payment, or phasing the effects of a ruling over time.

Brazilian constitutional adjudication provides a specific technique for this: interpretação conforme, or a conforming interpretation, in which the Court leaves the provision formally in force but confines it to the only construction deemed constitutional. The precedent therefore supports a strong risk that the payment cap will be invalidated or narrowed—not certainty that it will be struck down in full.

AGU and PGR Opinions Are Inputs, Not Commands

There is a temptation to read institutional opinions as forecasts of the judgment. The four-cycle record does not support that shortcut. It contains every relevant combination: the PGR aligned with the outcome and the AGU on the losing side (EC 30 and ADI 5348); the PGR on the losing side while the AGU aligned only after changing its position during the litigation (the EC 113/114 cap); both aligned with the outcome (Selic and the cut-off date); and the STF going further than either institution (borrowing proceeds earmarked for discounted settlements, where the PGR proposed an intermediate solution and the Court invalidated the provision).

The empirically honest formulation is that the direction of these opinions, standing alone, does not explain the observed outcomes. In the closest precedents, the STF sometimes followed one institution, sometimes departed from it, and sometimes went beyond both.

For an investor, the ADI 7873 opinions do not provide a forecast. They map the remedies available to the Court. The PGR's proposed remedy for the payment cap—restricting it to entities demonstrably unable to pay—also partly overlaps with the OAB's alternative request.

Issue-by-Issue Map

In narrative form, the map shows:

February 1 cut-off date. This is the weakest part of the OAB's challenge. The previous move to April 2 was unanimously upheld in 2023 as a budget-planning measure. Without empirical evidence that the new calendar effectively defeats the right to payment, the direct precedent weighs against the claim.

The 1%–5% RCL cap plus elimination of a final deadline. This is where the three previous constitutional cycles converge—and where the OAB's argument is strongest, particularly when the cap and absence of a deadline are treated as one deferral architecture. The outcomes most consistent with the record are invalidation of the combined framework or a conforming interpretation that restricts eligibility and imposes an effective repayment horizon. For investors, either outcome shortens expected duration relative to the current base case; the difference lies in magnitude and transition.

Direct settlements (paragraph 29). The case law does not prohibit settlements. It prohibits a system that coerces creditors into accepting a discount. Our base case is the narrowing construction proposed by the PGR: preservation of the constitutional priority order and a 40% maximum discount, equivalent to a minimum recovery of 60% of the claim's adjusted amount. The 2015 order in ADI 4425 provides the safeguard model; Article 102, paragraph 1, of the ADCT incorporated it into the special regime; and significant state settlement notices issued under EC 136 have preserved that threshold.

Institutional convergence is incomplete. The AGU limits mandatory application of the 40% cap to the special regime. Article 8, paragraph 2, of CNJ Provision 207/2025 treats the former maximum as superseded for administrative application of the new paragraph 29. The forecast that the STF will restore a generally applicable cap remains strong because of the fit with the 2015 ADI 4425 order and recent settlement practice, but it is a contested forecast, not continuation of a rule that is currently undisputed. For investors, the likely outcome remains a pricing input rather than invalidation of the settlement channel.

Accrual treatment after deposit (paragraph 30). This is genuinely open and should be framed precisely: the deposited amount is not frozen. Once funds reach the special court-controlled account, accruals charged to the public entity stop, the claim leaves the outstanding stock, and the balance accrues at the bank-account remuneration rate until the payment order is released, as CNJ Provision 207/2025 provides. There is no directly controlling precedent. STF Theme 1335 addresses only the constitutional grace period and does not decide this issue. The actual question is whether the bank-account return sufficiently preserves the claim until funds are released. We treat this as an unresolved operational risk, not a conclusion.

IPCA plus 2%, capped at Selic. The formula contains two different constitutional questions. On the first, the OAB faces a difficult argument: TR failed because it did not keep pace with inflation; IPCA is Brazil's official broad consumer price index and, by construction, tracks inflation. The inflation-adjustment component has strong support in the precedents.

The second component is the formula's more difficult legal question. Simple default interest of 2% per year does not adjust for inflation; it compensates for late payment. Replacing a single Selic accrual with this component materially reduces the public entity's cost of delay. Because the interest is simple and calculated on principal, its compounded annual equivalent declines as duration increases. In a high-rate environment, the gap is significant; combined with a regime that has no final deadline, it may make nonpayment comparatively inexpensive. The merits question is not whether IPCA preserves purchasing power. It is whether 2% simple annual default interest still provides sufficient compensation or instead creates an economic incentive to delay payment of a final court judgment.

The precedents do not answer that question directly. In Theme 810, the STF held that savings-account-linked default interest was constitutional for non-tax public debts, leaving the legislature some room to design the rate. ADIs 7047/7064 upheld Selic as a single accrual index. Neither judgment, however, considered a fixed 2% annual rate within a permanent architecture that also caps payments and eliminates a final deadline. Both the AGU and PGR defend the formula. That alignment matters, but it does not erase the distinction.

The forecast must therefore be split: strong analogical support for IPCA's validity; genuine uncertainty over the constitutional sufficiency of 2% default interest. Even if the STF accepts the second objection, full restoration of Selic would not automatically follow. The remedy could affect only the interest component or establish another benchmark. For underwriting, the current rule remains the base case, but the legal optionality around carry should no longer be modeled as zero.

Application to already-registered precatorios. The stand-alone retroactivity challenge is likely to fail. The Court unanimously upheld immediate application of the 2021 rules to payment orders that had already been issued. But the provision is instrumental: if the payment cap falls, its application to the existing stock falls with it as a consequential effect.

Underwriting Implications

Three practical conclusions, in order of importance:

1. The material legal optionality is concentrated in the payment cap plus final-deadline issue. This is the only cluster on which three precedent cycles and the current Court point in the same direction against the amendment. Portfolios whose base case already assumes a full and perpetual cap carry upside asymmetry. Portfolios that require invalidation of the accrual formula or cut-off date carry the opposite risk.
2. Inflation adjustment and default carry must be modeled separately. IPCA has strong precedential support. The 2% simple annual default-interest component has institutional support but no direct precedent at that level. The base case should apply current law alongside an alternative scenario in which the STF revises only the interest component—without assuming automatic restoration of Selic.
3. Direct settlements should be modeled with safeguards, not invalidation. The constitutional priority order and a minimum recovery of 60% of the adjusted claim amount are our base case. The 2015 ADI 4425 order, later incorporation of the limit into the ADCT, the PGR's express request, and recent settlement notices point in that direction. The AGU and CNJ, however, dispute its general application to paragraph 29. Underwriting should recognize that divergence until the STF rules.

As of July 28, 2026, the case is at the following stage: in September 2025, the reporting justice removed it from the virtual session that would have considered preliminary relief and placed it under the accelerated merits procedure in Article 12 of Law 9,868/1999, sending the merits directly to the full Court. Following the PGR opinion, the case file is before reporting Justice Luiz Fux. There is no published ruling on the public hearing proposed by the AGU, and the case has not been scheduled for judgment.

Investors should monitor a decision on the public hearing, scheduling of the merits, and the composition of the Court. The vacancy will be filled through Brazil's constitutional process of presidential nomination and Senate approval; the effect of a future appointment on the case cannot be predicted.

Methodology

This article draws on a proprietary Lummen and Galvez Valencio Advogados study of the complete official records in ADI 7873 and its principal precedents: 89 filings totaling 2,542 pages collected from the STF portal and other official sources, with cross-checking of vote counts and core institutional positions. The strength grades assigned to each forecast measure analogical fit, not probability. Where the record does not support a conclusion, we say so.

Official Sources

• ADI 7873 on the STF docket: the CFOAB complaint, submissions by the Chamber of Deputies and Federal Senate, the AGU submission (Petition 149017/2025), and the PGR opinion (Petition 83647/2026).
• CFOAB complaint and PGR opinion, hosted by the OAB.
• STF: ADIs 2356 and 2362 (EC 30/2000); ADIs 4357 and 4425 and the 2015 transition order (EC 62/2009); ADI 5348; and ADIs 7047 and 7064 (ECs 113/114)—official judgments; current composition of the Court.
• EC 136/2025, EC 113/2021, EC 62/2009, and EC 30/2000—Brazilian Presidency; CNJ Provision 207/2025.
• Post-EC 136 practice: official settlement notices from São Paulo, Bahia, Paraíba, Maranhão, and Goiás compiled in this analysis, all within a maximum 40% discount in the sample reviewed.