Brazil: A Bankruptcy/Restructuring Overview

Two Options, Different Balance Sheets: The Extrajudicial Turn – and What Judicial Reorganisation Still Does

Brazilian insolvency law still offers two main rescue procedures under Law 11,101/2005 (the LRF), as amended by Law 14,112/2020. Extrajudicial reorganisation (recuperação extrajudicial or EJ) lets the debtor select a group of types of claims, obtain majority support within that group, and ask a court to bind the dissenters. Judicial reorganisation (recuperação judicial or RJ) brings most pre-petition claims into a court-supervised process, with a broad stay and a creditors’ meeting (assembleia geral de credores). What has changed is not the statute so much as the use of each procedure.

The 2026 market is running both at once. EJ filings had, by late August, covered a record BRL174.5 billion of debt. RJ filings remain high in number, though far smaller in individual size. Listed groups with a capital-markets book have tended to take the first option. Companies that need to stop a wider set of creditors, or that are returning to court after an earlier reorganisation, have tended to take the second. The question is which problems each procedure can still solve.

What extrajudicial reorganisation now offers

The 2020 amendments made the EJ path more usable in general, but mostly from a financial point of view. The debtor chooses the claims to include, within the statutory exclusions. A filing may proceed with the support of at least one-third of the claims of each type covered; support from more than half of each type must follow within 90 days of filing (Article 163, paragraph 7). The stay applies only to the included claims and runs from the filing itself (Article 161, paragraph 4). Confirmation binds the dissenting remainder of that species or group. Tax claims and proprietary securities remain outside by statute; labour claims enter only through collective bargaining with the union; creditors left out keep their actions.

For a listed issuer whose immediate problem is a refinance wall – debentures, receivables certificates, bank lines, offshore notes – EJ is attractive. The operational creditor might not be invited. There is, as a rule, no judicial administrator (administrador judicial). The court is asked to approve the plan, not to run the case. The confirmation of the plan by Raízen, a company in the sugar, ethanol, and energy sector, in July 2026, covering some BRL65 billion of subject claims, showed that a financial restructuring of that scale can go through the extra-judicial door; the request filed by the petrochemical company Braskem in August 2026, involving approximately BRL54 billion in bond and bank debt with only one-third of the required quorum secured, is testing the same approach.

Practice has added a second use, and a cost. Debtors now often file a thin first instrument – a standstill or placeholder plan – to turn on the stay while the economics are still being written. Oncoclínicas, an oncology medical company, moved from a contractual pause to interim relief under Article 20-B, paragraph 1 and then to an extra-judicial filing. Raízen filed with support above the one-third threshold to a standstill plan and completed the terms after the group was already in the proceeding. The procedure is no longer the ratification of a deal already done.

That is also why EJ cases have begun to look more contentious. The case of the Pão de Açúcar group’s plan, for example, still before the São Paulo court at the time of writing, has drawn creditor objections and an opinion from the Public Prosecutor’s Office against confirmation, raising questions of access to information, dealings outside the plan, a less favourable default option, and whether a mixed unsecured group is homogeneous enough to bind a minority. Those are submissions, not findings. They are enough to show that a limited perimeter does not, by itself, keep litigation out of the file.

What judicial reorganisation still does that the other option cannot

The RJ procedure remains the likely process for a different balance sheet.

Its stay is not confined to the claims the debtor chose to invite. Once processing is granted (deferimento do processamento), enforcement of every claim subject to the proceeding is stayed for 180 days, extendable once, across a wider field, including trade creditors who would not have been placed by choice in an extrajudicial position. For a company whose problem is operational as well as financial, that difference is decisive.

The RJ procedure also supplies machinery the extrajudicial option does not: a judicial administrator, verification of claims, a meeting by class, and, if the debtor’s plan is rejected or the stay lapses without a vote, an alternative creditor plan (Article 6, paragraph 4-A, and Article 56, paragraphs 4 to 8). Debtor-in-possession financing has a statutory footing (Articles 69-A to 69-F), even if Brazilian law still does not permit US-style priming; new money ranks as non-subject (extraconcursal) and the court may grant only a subordinated lien over already-encumbered assets (Article 69-C). None of that framework expressly reaches the EJ file. The statute places debtor-in-possession financing in the RJ chapter alone, so money advanced while a thin first plan is being negotiated rests on contract and on whatever priority the plan itself can create; any statutory rank would have to be argued by analogy, without a textual footing. Asset sales can be structured as isolated productive units (unidades produtivas isoladas, UPIs) free of successor liability (Article 60). Consolidation of a group can be tested in court (Articles 69-G to 69-L) rather than assumed in a private context.

This procedure has a price. RJ cases are theoretically slower and more visible. They pull in parties the debtor would rather leave untouched. They are also, for some groups, unavoidable. A second filing – Oi and Coesa (ex OAS) are the familiar example – is not a problem an EJ plan can solve, insofar as a financial majority cannot be assembled, or in which excluded fiduciary and trade creditors can undo a limited pause.

Where the two procedures now meet

The textbook contrast – EJ as a contract with a stamp, RJ as a full case – no longer describes the docket.

Both procedures now turn on how a group is drawn and therefore treated. In EJ, the question is homogeneity: may banks, debenture holders and other capital market investors vote and be treated as one group (Article 163, paragraph 1)? What if the group also includes other unsecured creditors? In RJ, given the statutory classes, how can the incentives be designed in a manner that preserves pari passu whilst obtaining a majority vote in an unsecured class formed by unlike creditors. The vocabulary differs. The dispute does not. Nor does the arithmetic favour the same debtors: an EJ group is counted by value alone, whereas an RJ class of secured or unsecured creditors must approved by head-count as well as by value (Articles 45 and 163).

Both leave fiduciary owners and tax authorities at the edge of the stay. The tax edge is not symmetrical, however. The dedicated instalment and settlement regimes for federal tax debt created in 2020 (Articles 10-A to 10-C of Law 10,522/2002) are written for the debtor in RJ; whether an EJ debtor may claim them is unsettled, and in practice it negotiates under ordinary settlement rules. The EJ debtor is spared, in turn, the tax-clearance requirement that Article 57 imposes before an RJ is confirmed. For a group whose tax book is large, that trade-off can decide the procedure.

Essentiality arguments, once associated with judicial cases, now appear whenever an EJ debtor – unable by statute to include a fiduciary creditor – still needs the asset, even though the essential-asset protection of Article 49, paragraph 3, is written for the RJ stay.

Both leave guarantees in place unless the creditor consents. An EJ does not clean a holding-company guarantee by implication.

The two procedures are also used with an eye on each other. An EJ negotiation is often conducted with a RJ in view. Capital-markets holders now appear in both rooms: in a meeting of debenture holders in one, on a list of claims in the other.

Both procedures also travel. Each qualifies as a foreign proceeding for the purposes of Chapter 15 in the United States, and EJ confirmation has already been recognised there as a foreign main proceeding. Recognition turns on the debtor’s centre of main interests, not on the breadth of the perimeter; what it protects, however, is only what the perimeter contains, so a species built around New York-law notes should be drawn with that enforcement in mind.

The Superior Court of Justice has been clear on one boundary. EJ does not restrain creditors who were not included (Article 161, paragraph 4). That ruling, restated in 2026 (special appeal 2,234,939, Third Panel), makes perimeter design the central EJ decision – and sends some debtors back to the RJ procedure when those who can disrupt the business sit outside the chosen group.

Choosing an option in 2027

Four questions are more useful than a preference for one procedure.

First, does the distress that must be stopped sit inside a financial group that can honestly be treated as one group? If so, the EJ path remains the more proportionate tool. If trade, tax or excluded security-holders can still bring the company down, the limited stay may not do the work.

Second, is the filing the ratification of terms already agreed, or a pause in which those terms are still to be written? The statute designed the first use for EJ and the second for RJ, but practice emulated the latter for EJ service; boundaries yet to be set by courts. Finished plan or not, if the majority is not reached within the 90 days, the EJ proceeding ends, the stay falls away and the RJ door becomes the fallback.

Third, will holders bound without having negotiated have enough of a complaint – about information, side arrangements, or the default option – to erase the advantage of speed? The EJ file is no longer insulated from that inquiry. An RJ case at least provides a statutory process in which information is supposed to be produced (Article 51 and the administrator’s reports, against little more than the documents of Article 163, paragraph 6, on the extra-judicial side).

The two doors still lead to different rooms. One is more akin to a financial perimeter and a confirmation. The other has a broader scope by statute. The 2026 evidence shows companies choosing between those designs with some accuracy. It also shows the EJ room taking on disputes once associated only with RJ. How courts treat group design, disclosure to non-signatories, and the thin first plan may determine whether the two procedures retain textbook distinctions.