Braskem S.A. - Round 1 position (Run 4)
From: Management and Board of Braskem S.A.
To: The Braskem Room
Date: 18 August 2026, evening (cutoff approximately 20:50 CT)
Status: Round 1. Process filing this week. Economics to be written inside 90 days.
Evidence labels used below: Verified = company filings / 6-K / indenture-style facts in the common record. Reported = press (Pipeline, O Globo, Valor/Bloomberg, Estadão, Fitch commentary as reported). Inference = board judgment from those facts.
This memorandum is the company’s position. We preserve the operating company, liquidity, and enterprise value. We do not exist to protect IG4/Shine or Petrobras control as an independent objective. We also will not sell a recap that the public record says will not be signed.
1. Objective and realistic alternative if talks fail
Objective. Protocol a thin, plan-to-a-plan extrajudicial recovery (EJ) next week, before the 60-day tutela expires on Monday 24 August 2026 (verified stay granted 26 June; reported O Globo 18 August 17:36 BRT: company-side sources say we are preparing that filing). The first petition is a 90-day process plan: standstill, cash controls, information rights, a negotiation calendar, and a commitment to present economics before the 90 days run. No asset sales in the first filing - there is no time for a roadshow (reported, O Globo). Asset-sale optionality may be opened inside the 90 days if the numbers require it. The economic ask that we will take into those 90 days remains tenor: five-year extension, grace, no principal haircut. We do not have a Petrobras or IG4 equity recap to sell, and we will not pretend otherwise (reported, Pipeline Valor 18 August 12:16 BRT: no willingness from Petrobras or IG4 to inject capital or nationalize).
Ratings. Fitch already revised the parent to RD on 17 August after uncured interest defaults (verified). Fitch has said a subsequent EJ or RJ filing would take ratings to D (reported, 18 August). We file anyway. Operations, feedstock, LCs, and customer continuity matter more than the letter. RD has already done the rating damage that would have been used to scare us off a filing. D is the expected mechanical next step, not a reason to miss Monday.
Realistic alternative if talks fail. If we cannot assemble more than one-third of each affected espécie - in particular some New York notes - in time to protocol before 24 August, the realistic alternative is RJ of Braskem S.A., not cure-and-continue and not a conversion DIP. A notes-out EJ does not stay Elliott after 24 August (inference from the tutela covering invited Wind Chamber financial creditors only, plus Estadão’s reported point that without some bondholders the company cannot reach one-third). RJ is debtor-in-possession, with weaker absolute priority than Chapter 11, headcount plus art. 58, and equity that often survives; it also carries a process discount we put at 15–25 percent of going-concern value, LC/trade/utilization risk, and a live risk that the remaining Idesa contribution is attacked. We will use RJ if Monday is empty. We will not use RJ as a bluff that we are unwilling to execute.
We reject, as a company proposal: (i) DIP that converts and drops controllers below 10 percent; (ii) all-asset liens over the Brazilian crackers; (iii) a Petrobras contingent put, keep-well, or parent guarantee that Petrobras will not sign.
2. Filing bar, chairs, and how a filing happens without Elliott
Verified law, not Pipeline’s slogan. Lei 11.101 art. 163 §7: more than one-third of each affected class to protocolar an EJ; more than 50 percent of each class to homologate. Pipeline’s “two-thirds to file” is wrong. If the listed parent notes are one espécie, more than about US$2.35 billion of the listed ~US$7.02 billion is needed to file that class (verified listed outstandings at 30 June: ’28 1,250; ’30 1,500; ’31 850; ’33 1,000; ’34 850; ’41 587; ’50 750; hybrid ’81 231). Homologation of that class would need >US$3.51 billion. Debentures, CRAs, bank/ECA lines, and the drawn RCF are separate stacks totaling the remainder of US$10.3 billion gross corporate debt and may be separate espécies. We design the affected-class map to what we can file. We do not assume Elliott has a published blocking percentage, and we do not assume the AHG steering committee is 51 percent of the notes (common-record instruction).
Reported chairs. Elliott (international notes; reported largest lender on the US$1.0 billion RCF). Contrarian (hard group with Elliott on the Petrobras-cash ask, Bloomberg 10 August). SVP (notes and some RCF; reported not a formal Elliott pairing). AllianceBernstein, Capital Group, PGIM (reported, Bloomberg compilation). Local banks reported supportive of an EJ: Itaú, Safra, KfW, Bladex, DZ, SMBC and others (Pipeline).
Will Elliott / Contrarian support a one-third filing without a Petrobras put? No. That is the steering position we must take as a fact for this round (reported Tuesday meeting: bondholders demanded a Petrobras commitment to invest if metrics are missed, as the condition for supporting an EJ and 90-day talks; Pipeline reported no willingness from Petrobras or IG4). We cannot deliver that put. Lei 13.303, TCU scrutiny, the 4 October 2026 election, and IFRS 10 / consolidation of the US$10.3 billion stack if voting control is lost or common is issued are Petrobras’s hard stops (verified law and share counts: Petrobras voting 47.03 percent, total capital 36.15 percent; joint control with Shine). Offering a put we do not have would be a fraudulent solicitation.
How the filing still happens. Three sleeves, not one AHG chair:
- Local-bank / ECA / debenture / CRA espécies. Reported as supportive. This is the easy one-third - and, if we do the class map correctly, likely the easy homologation majority as well.
- NY notes espécie - peel the long-onlys. Capital Group, AllianceBernstein, and PGIM are the filing math. We need them, plus whoever else will take a process EJ with guardrails and a 90-day economic calendar, without a Petrobras put. Estadão has reported that without some bondholders we cannot reach one-third. We agree. The peel is for filing, not for Elliott’s economic clearance.
- RCF espécie. Elliott is reported as largest lender, not as the whole facility. SVP and bank lenders sit here. If we cannot show >1/3 of an affected RCF class, we have a class-map problem, not a reason to omit notes. Omitting notes leaves Elliott unstayed on 25 August. That is RJ.
O Globo’s URL slug claims creditor approval; the body still recites the one-third bar (reported). We do not treat adhesion as locked. We treat next week as a filing exercise: local banks plus a notes peel, Elliott/Contrarian hostile or outside, naphtha terms unsigned or capped, no sponsor cash equity.
June AHG “conditional support” for a temporary plan-to-a-plan EJ with protections is not adhesion to our economics (verified from the June record as summarized). July’s two Elliott offers - conversion DIP that drops controllers below 10 percent, or extension at existing coupon with all assets as collateral - were rejected and remain rejected.
3. Unsigned naphtha terms after Idesa - do they satisfy burden-sharing?
No, not as Elliott/Contrarian define burden-sharing, at any cap we can actually get signed. Supplier credit is liquidity, not loss-absorbing capital. It does not replace a Petrobras put, common-equity injection, keep-well, or parent guarantee.
Valor/Bloomberg reported (17–18 August) Petrobras weighing capped naphtha payment-term relief (purchases now often cash / short-dated, conceptually toward six months). Not a loan. Not signed. Cap, price, and duration open. Parties far apart. Goal in the press is one-third support for a generic 90-day EJ, not a finished plan. Mathematical ceiling of moving about nine-day payables to 180 days is roughly US$950 million. We will not assume the ceiling.
| Cap (inference; test cases required) | Liquidity effect after the remaining Idesa cheque | Burden-sharing claim we will make | Burden-sharing claim creditors will accept |
|---|---|---|---|
| US$250 million | Roughly offsets part of the remaining ~US$350 million Idesa use. Does not rebuild a US$1.0 billion liquidity covenant. | Petrobras is moving off cash/short-dated naphtha. | Insufficient. Looks like ordinary feedstock terms, not shareholder pain. |
| US$500 million | More than offsets the remaining Idesa cash use as liquidity. Helps LCs and inventory restocking if utilization holds. Still not capital. | This is the minimum at which we will argue Petrobras has put real working-capital skin in after Idesa. Arm’s-length, capped, documented, Lei 13.303-compatible. | Long-onlys may treat it as enough sponsor color for a process filing. Elliott/Contrarian will not. |
| US$950 million (ceiling, not a forecast) | Material. Could underwrite a US$1.0 billion minimum-liquidity covenant if LCs are rolled and Q3 conversion is not another US$547 million working-capital hole. | Strongest commercial Petrobras instrument available without equity or a put. | Still not a contingent injection if metrics are missed. Do not sell it as one. |
Board rule. We do not wait for a signed naphtha paper before next week’s protocol. The O Globo clock does not allow it. Unsigned terms are a 90-day workstream, not a filing condition. We will frame any signed cap as arm’s-length financing of profitable operations, conditional on creditor tenor and IG4 accepting warrants/dilution - not as a government rescue (inference from Lei 13.303 and election-year optics; verified that nonmarket public-policy obligations require defined conditions and compensation). REIQ/PRESIQ and the provisional PE antidumping measure are competitiveness tailwinds (verified legal/policy instruments). We will not present them as permanent subsidies or as proof that the state will protect today’s equity.
4. IG4 / Shine cash
Available amount for the parent: zero. Pipeline Valor, 18 August 12:16 BRT, reported no IG4 aporte. We will not put a US$300 million Shine junior facility in any base-case package, and we will not ask this room to “solve” the rerun by inventing one.
Shine acquired 50.1108 percent of voting capital and 34.3234 percent of total capital from NSP/Novonor through a distressed exchange, not a cash recap of Braskem S.A. (verified ownership figures in the common record). IG4 disclosed AUM around US$1 billion; that is not uncalled Braskem dry powder. The Idesa US$476 million is company / Braskem Netherlands cash, not an IG4 fund cheque (verified 18 August 6-K / company announcement; common-record instruction). Control retention is the product IG4 bought. Conversion that drops controllers below 10 percent remains rejected - because it is a value-destructive DIP at a distressed print, not because this board’s duty is to freeze IG4’s option.
If IG4 wants to keep control through homologation, the burden it can bear without writing a parent cheque is: accept creditor warrants, cash controls, information rights, no dividends, and no blockage of a capped Petrobras naphtha instrument. That is not a recap. We will not advertise it as one.
5. Opening, first concession, narrowest acceptable
The first filing is not the economic plan. Next week’s petition is thin. The grid below is what we will negotiate inside the 90 days. Opening economics remain the company ask Pipeline reported on 18 August: five-year extension, three-year grace, no haircut. That ask is still negative-NPV against a 2027–28 US$1.5 billion normalized EBITDA case, and we know it. It is the right opening because (i) Q2 recurring EBITDA was US$1.043 billion (verified), (ii) we will not give away the persistent-spread upside at an RD/D print, and (iii) we have no sponsor cash to buy a better creditor NPV with equity.
| Item | Opening (economic ask into the 90 days) | First concession (authorized without a further board fight) | Narrowest acceptable (walk-up to RJ) |
|---|---|---|---|
| Vehicle | Thin EJ next week; 90 days to write economics. No asset sales in the protocol. | Same vehicle. May open a non-core sale process inside the 90 days; nothing pre-committed. | Homologated economic EJ. If homologation fails, RJ - not conversion DIP. |
| Maturity | Five years added to affected financial debt. Three-year grace on amortizing local/bank lines. | Four-to-five years on the 2026–28 wall (RCF 31 Dec 2026, ’28s, near bank/ECA). Longer bonds can take a shorter extension if coupon is left intact. | Four years on the wall; ’33/’34/’41/’50/’81 not required to pick up five extra years as of right. |
| Principal | No haircut. No conversion. | No haircut. No conversion. | No initial principal haircut. Conversion only if the alternative is a value-destructive all-asset lien and a court would give creditors more in RJ - in which case we choose RJ rather than sign conversion DIP. |
| Cash / PIK | Historical board opening was 100% PIK toggle through December 2028 and −200 bp. We do not lead with the coupon cut this round. Pipeline’s live company offer is tenor/grace/no haircut. PIK is a toggle for the 90-day paper, not a coupon cut. | 18 months of 4% cash / 4% PIK (or equivalent), then cash coupon. | Two years partial PIK (about 4/4 or 5/3), then existing coupon +150–200 bp. Coupon cut of 200 bp is withdrawn permanently at this step. |
| LCs | US$1.5 billion unsecured LC facility, mostly roll of existing claims (historical opening). | US$1.0–1.2 billion committed LC/WC support, majority roll. | US$1.0–1.2 billion WC/LC. Existing LCs rolled to the extent issuing banks will roll. New money, if any, is LC/WC capacity - not a delayed Petrobras equity cheque. |
| Petrobras | Capped naphtha working-capital terms, unsigned. No put, no common, no keep-well, no parent guarantee. | Push to sign a US$500 million cap as the working figure; US$250 million is too small to mention as burden-sharing; US$950 million is a ceiling we will not forecast. | Signed naphtha paper at whatever cap is actually available. Still not a put. If Petrobras cannot sign before homologation, we homologate without it rather than miss the 90 days. |
| IG4 / Shine | No parent cash. Control is not a veto over a superior operating-company proposal. | Accept 15% creditor warrants as the price of remaining in control. | 15% fully diluted warrants, stepping to 22.5% if EBITDA < US$2.0 billion, liquidity < US$750 million–1.0 billion, or PIK continues beyond two years. Stretch 25% only if there is substantial third-party new liquidity (not a fake Shine facility). >35% creditor ownership or blanket liens on core Brazilian fixed assets, without real principal cancellation, is worse than RJ. We will not sign it. |
| Creditor new money | Not required for the thin filing. | Welcome on eligible inventory, receivables, and controlled proceeds, first lien. | First-lien new money on AR/inventory/proceeds only. No conversion DIP. No “rescue” loan whose economics are equity in disguise. |
| Warrants / equity | None in the opening ask. | 15% fully diluted warrants. | 15% / 22.5% step-up / 25% stretch as above. No common conversion. |
| Collateral | Crackers unencumbered. No fixed-asset package. | Limited WC collateral: eligible inventory, receivables, controlled proceeds. At most a limited second lien for legacy debt on those same pools and selected non-core shares. | Same. Camaçari, Triunfo, ABC, and the other Brazilian crackers stay unencumbered. July all-asset demand remains rejected. All-asset liens without principal cancellation fail the RJ test. |
| Covenants | Information rights and a 90-day cash-control protocol in the thin plan. | Minimum liquidity US$1.0 billion as a plan covenant once WC/LC support is in place - it is not a description of 30 June cash. No dividends/buybacks until net leverage < 3.0x. 50% excess-cash sweep above the floor; 75% asset-sale sweep. Capex, incremental debt, liens, acquisitions, related-party transfers restricted; ordinary feedstock/inventory optimization permitted. | Same grid. Alagoas safety/remediation cash is a hard carve-out from sweeps and from any “excess cash” definition. |
| Mexico | Separate estate. Fund the remaining disclosed contribution only if the parent liquidity floor holds post-filing. No new parent guarantee. | See Mexico table. | See Mexico table. If funding the remainder would breach the floor, we delay or restructure the Idesa cheque rather than insolvent the parent to save a Mexican option. |
| Alagoas | Provisions, remediation reserves, and safety payments unimpaired. No distribution to creditors or shareholders ahead of budgeted community/safety cash. | Same. Reporting preserved. | Same. A deal that looks like it subordinated Maceió to NY notes is politically and legally toxic (verified federal proceeding involving the company and former executives, June 2026; continuing administrative/socio-environmental obligations). |
Collateral bright line. First lien for new money on working-capital pools. Limited second lien for legacy debt on those pools and non-core shares only. No conversion DIP. No blanket mortgage on the crackers.
6. Liquidity, Idesa cash, and required operating cases
6.1 Verified starting point (30 June / Q2)
- Corporate gross debt US$10.3 billion; adjusted net debt US$9.5 billion; adjusted corporate net leverage 6.74x. These already exclude Braskem Idesa project debt (verified).
- Q2 recurring EBITDA US$1.043 billion (Brazil/SA US$869 million, US/Europe US$147 million, Mexico US$57 million). H1 recurring EBITDA approximately US$1.235 billion. Mexico utilization 43 percent - do not annualize Mexico (verified).
- Brazil utilization 70 percent (verified, management record).
- Q2 working capital consumed US$547 million; operating cash generation US$385 million; recurring cash generation about US$210 million (verified). Price and volume components are level resets if prices and inventory stabilize. Lost supplier finance is persistent.
- Cash and cash equivalents R$3.931 billion at 30 June, of which R$353 million inside Braskem Idesa. June plan unrestricted cash about US$795 million (verified).
- After July cure periods: R$507 million / about US$98 million of defaults under certain financial instruments. No public acceleration as of this cutoff (verified). Public bond terms generally: 30-day interest cure; 25 percent of a series can accelerate after an uncured event of default.
- US$1.0 billion standby/RCF drawn October 2025, matures 31 December 2026. H2 contractual debt service in the June plan US$2.349 billion including LCs and the standby (verified).
- Q3 contractual debt service US$878 million including US$572 million of LC runoff; June-plan unrestricted cash at 30 September about US$337 million under the status quo (verified). Q3 EBITDA US$586 million is the June plan; US$750 million and US$1.0 billion are sensitivities, not results. No Q3 print is assumed.
Spread cases we will use in the 90 days (management record, not a Q3 result): persistent spreads, 2026 EBITDA US$2.7–3.3 billion, with meaningful H2 cash generation once the price/inventory reset stops consuming cash; normalization, 2027–28 EBITDA around the June plan’s US$1.5 billion. Strong spreads improve leverage in a negotiation. They do not cure default and they do not pay the December RCF.
6.2 Idesa remaining cash - the liquidity-floor rule
Verified 18 August 6-K / announcement. Idesa senior debt about US$2.5 billion down to about US$1.6 billion. Prepack Chapter 11, SDTX, expected 60–90 days. Braskem contributes US$476 million, of which about US$126 million was already made available. About US$350 million remains. RSA support 76.55 percent of senior secured notes and 100 percent of the Inbursa term loan (first-day reporting, BankruptcyData 18 August - court-document reporting, not a 6-K). Braskem-backed DIP: up to US$279 million new money plus about US$130 million roll-up, 10 percent PIK; the DIP converts into Braskem’s equity allocation, not cash repaid by Idesa to the parent at exit; plus US$71 million effective-date cash. Interim DIP target within two business days; confirmation/final DIP within 40 days; effective date within 55 days.
Inference, arithmetic not 6-K: remaining US$350 million ≈ US$279 million DIP new money + US$71 million effective-date cash. The US$130 million roll-up is not new parent cash. Treat the US$350 million as a live parent / Netherlands cash use unless a later filing proves it is only a restatement of the existing term loan (US$180 million committed / US$129 million disbursed) or the US$82 million Idesa-secured WC loans. The term-loan lender is a separate supporting party.
Board liquidity-floor rule (inference / decision). We will not send the remaining US$350 million if doing so takes unrestricted parent cash below an operating floor of US$400 million. The US$1.0 billion covenant in the economic package is a post-deal target once LCs/WC support exist; it is not cash we have today. Sequence:
- Protocol the parent EJ first (next week). Stay invited financial-creditor service. That is Q3 contractual US$878 million of which US$572 million is LC runoff.
- Only then fund remaining Idesa, and only if post-stay, post-cheque unrestricted cash is still ≥ US$400 million.
- If LC issuing banks force cash runoff and the floor would break, Idesa waits. We will not insolvent the parent to close a Mexican prepack. Majority Idesa equity is an option, not a put on the Brazilian estate.
June-plan status-quo cash at 30 September is US$337 million before the remaining Idesa cheque. US$337 − 350 = −US$13 million (inference). That is why cure-and-continue is not a plan after Idesa, and why the thin EJ is a liquidity instrument, not a rating exercise.
Implied Q3 cash from operations in the June plan (inference):
30 Sep US$337 million = 30 Jun US$795 million + CFO − US$878 million service
⇒ CFO ≈ US$420 million against Q3 EBITDA US$586 million ≈ 72 percent conversion. We use that ratio to scale the sensitivities, and we flag it: Q2 conversion was far worse because of the US$547 million working-capital draw. Linear scaling is an inference, not a forecast. If working capital consumes another US$547 million, every case below is too optimistic by that order of magnitude.
6.3 Three Q3 EBITDA cases
No case assumes a Q3 result. December standby in every case is restructured/extended into the EJ, not repaid in cash. Unrestricted cash at 30 September cannot retire a US$1.0 billion RCF on 31 December and also keep the company in naphtha and LCs. Q4 remainder of H2 contractual service is about US$1.471 billion (US$2.349 billion − US$878 million), of which ~US$1.0 billion is the standby (inference from verified H2/Q3 figures).
LC rolled = issuing banks roll the US$572 million Q3 runoff inside the EJ (the operating assumption we will work). LC cash = runoff is paid because banks pull - the downside we must look at before we wire Idesa.
| US$ million | Q3 EBITDA US$586 (June plan) | Q3 EBITDA US$750 (sensitivity) | Q3 EBITDA US$1,000 (upside sensitivity) |
|---|---|---|---|
| Starting unrestricted cash (verified, June plan) | 795 | 795 | 795 |
| Implied Q3 CFO before contractual DS (inference; scaled from 420 at 586) | 420 | 538 | 717 |
| A. Status quo (no EJ), pay Q3 DS US$878, no Idesa remainder | 337 (verified June plan) | ~455 | ~634 |
| B. Status quo + remaining Idesa US$350 | −13 | ~105 | ~284 |
| C. Thin EJ, financial DS stayed, LCs rolled, then Idesa US$350 | 865 | 983 | 1,162 |
| D. Thin EJ, other DS stayed, LC runoff paid cash, then Idesa US$350 | 293 | 411 | 590 |
| Remaining Idesa vs US$400 million floor | C yes / D no - do not wire Idesa in D | C yes / D borderline | C yes / D yes |
| December US$1.0B standby | Restructure. Cannot pay. | Restructure. Cannot pay. | Restructure. Cannot pay. Paying it would take even case C from ~US$1.2B to ~US$0.2B before Q4 operations. |
| Cure US$98 million on top | Liquidity-destructive. Does not touch the H2 US$2.349B wall. | Same. | Same. |
| Cash conversion comment | June-plan conversion. Still a defaulted issuer. | Better conversion if WC reset has stopped. Still cannot pay the RCF. | Consistent with a path toward the US$2.7–3.3B persistent-spread year if Q4 also prints. Still a refinancing case, not a cure case. |
What this table decides.
- Cure-and-continue after the remaining Idesa cheque is not viable in the June-plan case and is cosmetic in the sensitivities. Paying ~US$98 million of July defaults does not refinance US$2.349 billion of H2 service.
- Thin EJ is the only path that both (i) funds Idesa and (ii) holds a parent floor, and even then only if LCs are substantially rolled (row C). If LCs run off in cash (row D), Idesa is delayed or cut.
- December RCF is an EJ issue, not a cash issue, in all three EBITDA cases. Opening ask: five-year extension of that facility. Narrowest: four-year extension at existing coupon plus the PIK/spread grid above, with Elliott possibly outside if the rest of the RCF class can homologate.
- US$1.0 billion EBITDA does not change the filing decision. It changes 90-day bargaining leverage (warrants toward 15 percent, coupon cut off the table, naphtha cap easier to defend as working capital rather than rescue). It does not pay the RCF and it does not create a Petrobras put.
Strongest argument against this board, stated plainly. At US$1.5 billion normalized 2027–28 EBITDA, five-year extension leaves leverage on the order of US$9.5 / 1.5 ≈ 6.3x before PIK accretion. Creditors may rationally demand permanent equity. Our answer is not “spreads are guaranteed.” Our answer is: (i) Q2 already printed US$1.043 billion in a single quarter; (ii) giving that upside away on a conversion DIP at RD/D is the wrong price; (iii) warrants at 15–22.5 percent are how creditors buy the upside without taking the crackers; (iv) if they insist on conversion or all-asset liens without principal cancellation, RJ is better for the operating company than that RE.
7. Mexico table (required)
Idesa haircut does not recapitalize the US$10.3 billion corporate stack. We take no parent-deleveraging credit for the about US$920 million Idesa reduction or for the US$825 million equitized into Idesa noteholders.
| Question | Position | Label |
|---|---|---|
| 1. Separate estate or not | Separate estate. No disclosed parent guarantee of the remaining about US$1.6 billion. No parent Chapter 11 or parent RJ as part of the Idesa prepack. Idesa was already an unrestricted subsidiary on older parent indentures. This is not a J.Crew drop-down of Camaçari. | Verified (6-K / announcement); indenture status as in common record |
| 2. Further parent cash / guarantee after the disclosed US$476 million | No new parent guarantee. No further cash beyond the remaining ~US$350 million, and that remainder only if the US$400 million parent floor holds after the EJ stay is in place. DIP is Braskem-funded and converts to Idesa equity; it is not cash repaid by Idesa to the parent. Do not model a US$350 million in-flow at Idesa exit. | Verified structure; inference on the floor |
| 3. US$82 million secured WC loan and the term loan | Sit at Idesa. Term loan: US$180 million committed / US$129 million disbursed; lender is a separate supporting party, 100 percent of the Inbursa term loan supporting (first-day reporting). US$82 million WC loans are Idesa-secured. No additional parent guarantee. No parent cash repayment assumption. Roll-up of about US$130 million is inside the Idesa DIP, not a parent recovery. | Verified commitments; first-day reporting on support |
| 4. Credit for Idesa haircut as parent deleveraging | None. Parent gross debt remains US$10.3 billion. Idesa’s move from ~US$2.5 billion to ~US$1.6 billion is an Idesa capital structure event. Pipeline’s same-day “no holding impact / earmarked cash / Slim raising the 25 percent” color is company-friendly press, not a parent-guarantee analysis (reported; we will not treat it as one). TQPM equity-support agreement exists; Slim as that lender is why they say no holding cross-default (reported). | Instruction + reported press |
| 5. Majority ownership: option worth keeping vs cash leak to stop | Post-reorg split reported (first-day): one-third to Braskem for funding value, one-third to secured noteholders (US$825 million equitized), one-third to existing shareholders (about 58 percent Braskem, about 8.3 percent Etileno XXI). Inference: Braskem ≈ 33.3% + 0.58×33.3% ≈ 52.6 percent - majority, barely. That is an option on a 43 percent-utilized Mexican cracker, not parent deleveraging. Worth keeping if and only if the remaining US$350 million does not break the parent floor. Not worth a parent guarantee, not worth unlimited further cash, not worth missing the 24 August stay. If the floor would break, stop the leak. | First-day reporting + inference |
| 6. Can a parent EJ/RJ filed before Idesa’s ~55-day effective-date freeze the remaining contribution / DIP? | Yes, it can. We will file the parent EJ next week, i.e. before Idesa confirmation (~40 days) and effective date (~55 days from 18 August ⇒ early-to-mid October). The Idesa disclosure statement already flags that certain Braskem actions could be reviewed in Brazil if the parent enters EJ or RJ (first-day reporting). Inference / legal risk: remaining US$350 million can be attacked as preferment, fraudulent conveyance, or an unauthorized related-party transfer once a Brazilian process exists. Operating response: (i) file parent EJ first so financial DS is stayed; (ii) treat the remaining contribution as performance of a pre-petition, disclosed, separate-estate commitment to an unrestricted subsidiary, with RSA support already locked; (iii) do not fund if the floor breaks; (iv) get Brazilian-court comfort in the EJ protocol that the scheduled Idesa uses are ordinary-course / going-concern preservation, not leakage to shareholders; (v) parent RJ (as opposed to thin EJ) is the scenario most likely to freeze or claw the remainder - another reason the thin EJ is preferable to a Monday-empty RJ. | First-day reporting + inference |
Mexico one-liner for the 90-day paper. Ring-fenced. Majority kept as an option. US$350 million is a scheduled use, not a receivable. No parent guarantee of the leftover US$1.6 billion. No credit against US$10.3 billion.
8. Political overlay (labeled)
Verified. First round 4 October 2026; possible second round 25 October. Lula confirmed. Petrobras is state-controlled and subject to Lei 13.303. REIQ and PRESIQ (Law 15.294/2025) exist. Provisional PE antidumping duties on certain US/Canadian imports exist as a scenario variable, not a permanent wall. Alagoas is live: June 2026 federal proceeding; continuing socio-environmental obligations.
Inference, not a guarantee. Election-year pressure favors a going-concern solution (jobs, domestic resin, industrial capacity) and raises the execution cost of a disruptive RJ and of a sudden foreign-creditor takeover. It does not order Petrobras to inject common, issue a put, or nationalize. The politically cheapest path for Brasília is a capped, documented, shared commercial instrument - which is why naphtha WC is the live Petrobras tool and equity is not. Alagoas makes a “national champion” speech fragile; every proposal in section 5 keeps safety/remediation cash ahead of distributions. We will not cite politics as a reason to reject a demonstrably superior operating-company proposal.
A 90-day clock started next week runs through mid-November, i.e. across the election. That is convenient. It is not a reason to file an empty plan that cannot homologate.
9. Outcome probabilities (sum to 100 percent)
These are this board’s probabilities as of 18 August evening. They are inferences. O Globo is company-side color, not locked adhesion. Elliott is hostile without a put we cannot deliver. Local banks are reported supportive. The notes peel is the swing variable.
| Outcome | Probability | Why |
|---|---|---|
| Thin / process EJ (filed next week; 90-day plan-to-a-plan; economics not fully agreed; Elliott/Contrarian hostile or outside; no sponsor recap; no asset sales in the protocol) | 46% | This is the path we are preparing (reported, O Globo 17:36). Filing bar is >1/3 per espécie, not 2/3 of the stack. Local banks likely get us there on bank/ECA/CRA/debenture classes. Notes class requires a long-only peel that is not locked. Modal because the tutela dies Monday and cure is dead after Idesa. |
| Economic EJ with a completed term sheet (homologable economics inside or at the end of the 90 days: tenor, partial PIK, existing coupon + spread, 15–22.5% warrants, WC collateral not crackers, no haircut, no PBR/IG4 equity) | 19% | Possible if Q3 conversion is not another working-capital hole and long-onlys plus local banks can carry homologation (>50% per class) without Elliott. Hard without a recap. Easier if naphtha signs at ~US$500 million and Q3 prints toward US$750 million+. Not the first filing. |
| Cure-and-continue (pay the disclosed ~US$98 million and refinance the wall) | 4% | Only if the remaining US$350 million Idesa use is later proven not to be live cash and someone refinances the 31 December RCF and LCs roll. Public record does not show that refinance. After Idesa, June-plan September cash is approximately zero. Residual, not a plan. |
| Creditor-favorable (conversion DIP dropping controllers below 10 percent, all-asset liens, or a Petrobras put/keep-well) | 6% | Public record: Petrobras/IG4 will not inject; we rejected both July structures; crackers stay unencumbered. Residual covers a failed notes peel plus a court/new-board forced surrender we do not currently intend to sign. We will take RJ before we sign conversion DIP or a fake put. |
| RJ of Braskem S.A. | 25% | If the notes espécie cannot show >1/3 by ~21–22 August, or if a filed thin EJ cannot homologate inside 90 days, or if LC/trade collapse on D. Election-year cost is real; process discount 15–25 percent is also real. Parent RJ is the scenario most likely to freeze remaining Idesa funding. Still the honest BATNA. |
| Total | 100% |
We are driving the 46 percent. We are prepared for the 25 percent. We are not modelling the 6 percent as a concession we will offer.
10. The single fact that would most change this position
Confirmation, before ~21 August, that Capital Group, AllianceBernstein, and PGIM will not support a process EJ without a Petrobras cash commitment / put.
That fact would mean we cannot file the notes espécie, a notes-out EJ does not stay Elliott after 24 August, and the realistic path collapses from thin EJ (46 percent) into RJ (and a worse Idesa-funding problem). Almost every other moving piece - unsigned naphtha at US$250 vs US$500 million, Q3 EBITDA US$586 vs US$750 million, Fitch D, IG4 silence - we can absorb inside a filed 90-day plan. We cannot absorb a notes-class filing failure.
A close second, which would move us the other way: a later Idesa filing proving the remaining US$350 million is only a restatement of the existing term loan / WC, not incremental parent cash. That would reopen a thin slice of cure-and-continue probability and would let us hold a higher parent floor. We will not assume it.
11. What we will say in the room on Monday
- We file a thin EJ next week. Ninety days to write economics. No asset sales in the first paper. Fitch is already RD; the filing goes to D; we care about crackers, LCs, and naphtha, not the letter.
- Economic ask remains five years / grace / no haircut. We will move on PIK, coupon increase vs the old −200 bp opening, warrants, and WC collateral. We will not move to conversion DIP or cracker mortgages.
- There is no Petrobras put and no IG4 parent cheque. Anyone whose price is that put is not a filing partner. Local banks plus a notes peel are the filing. Elliott is free to stay hostile.
- Remaining Idesa ~US$350 million is a scheduled parent/Netherlands use. The DIP converts to equity; Idesa does not cash-repay us. We fund it only after the EJ stay is in and only above a US$400 million unrestricted floor. Majority Mexico is an option we like, not a cash leak we will die for.
- If the notes class is empty on Friday, we are in RJ, not in a conversion negotiation dressed up as a rescue.
End of Round 1 position.