Brazil’s First Battery Storage Auction Pulled 297 GW of Bids. 2 to 5 GW Is Expected to Win.

Energy storage batteries

Brazil’s energy planning agency registered 6,091 projects totalling 296,807 MW for the country’s first two auctions dedicated to battery storage — a record for any Brazilian energy auction, and roughly 16 times the 18 GW the industry itself had estimated would sign up before the window opened. The two rounds run on 2 and 4 December, and one of them is reserved for equipment with Brazilian content. That single design choice has done more to shape who is bidding than any price signal: over the past six months, Brazil battery storage has stopped being an export market and started being a factory market.

296,807 MWRegistered across two auctions
6,091Projects registered
2–5 GWExpected to be contracted
15 yearsLength of each contract

What Brazil Battery Storage Auctions Are Actually Buying

The framework comes from Normative Ordinance No. 136 of 1 June 2026 from the Ministry of Mines and Energy, published in the official gazette two days later. Storage became a standalone electricity asset under Law 15.269/2025, which is what allows batteries to be contracted through a capacity reserve auction at all — the mechanism Brazil has historically used for thermal plants.

Rather than one auction, the ministry wrote two.

LRCAP 2026 — National StorageLRCAP 2026 — Storage
Auction date2 December 20264 December 2026
ProductStorage Power 2028 AStorage Power 2028 B
Local contentRequired — BNDES CFI accreditation, four qualifying routes, minimum 15% domestic contentNone
Eligibility overlapWinners of the national round may not bid in the open round

Both run under the same commercial terms: 15-year contracts signed with the electricity trading chamber CCEE, supply beginning 1 August 2028, and remuneration through a fixed annual revenue paid monthly and indexed to the IPCA inflation measure. The fixed revenue has to cover investment, grid connection, transmission and distribution use, operation and maintenance, taxes, insurance, decommissioning — and reinvestment, explicitly including replacement battery modules and new compatible inverters.

The Rules That Decided Who Shows Up

The technical requirements are where the auction stops looking like a tender and starts looking like a product specification.

RequirementLevel
Minimum power30 MW per project
Continuous discharge4 hours
Full recharge timeMaximum 6 hours
Total efficiencyMinimum 85% at the metering point
CyclingMaximum 2 full cycles a day, 366 a year
Variable costZero — revenue comes from the fixed payment only
Grid-formingMandatory for the entire contract term
EquipmentNew cells and new bidirectional inverters or converters; no prior commercial use, reconditioning, remanufacturing or reuse
ConfigurationStandalone BESS with its own authorisation, connection and metering — co-located systems are excluded from competing
LicensingNo environmental licence needed to qualify technically

Two of those lines matter more than the rest. The zero variable cost rule means a winning project cannot declare a fuel-like cost to be dispatched, the way a thermal plant does: the developer gets paid for availability and carries the risk of how the system actually uses the asset. And making grid-forming mandatory for fifteen years turns a premium feature into a baseline requirement for every project in the country’s largest storage procurement.

Then there is a 10% discount on the offered fixed revenue for projects connecting at specific grid points the system operator has identified as most in need of storage — concentrated in Alagoas, Bahia, Ceará, Minas Gerais, Paraíba, Pernambuco, Piauí and Rio Grande do Norte. That is the northeast wind corridor and its neighbours, and it is a deliberate signal about where the government wants the batteries.

The local content rule is the one with industrial consequences. Qualification for the national round requires accreditation under BNDES’s CFI supplier registry, which offers four routes and a minimum 15% domestic content threshold. Proof is required before the contract is signed and again before commercial operation begins. The regulation is explicit that accreditation does not oblige a developer to take BNDES financing — but failing to meet the minimum local content can lead to the contract being terminated.

Who Moved In

The result is visible in how suppliers positioned themselves. Shipping finished containers from Asia is now the harder route into Brazil’s largest storage procurement, and the companies with global market share have been buying or partnering their way into Brazilian manufacturing.

CATL announced a strategic partnership with Moura, a Brazilian battery manufacturer, in São Paulo in late August, aimed at joint participation in the national round. The arrangement combines CATL’s systems with Moura’s domestic footprint and, according to CATL’s own release, the support of a second strategic partner holding the number one market share in PCS and inverters in Brazil. CATL says it already holds about 45% of Brazil’s energy storage market, anchored by the Registro project commissioned in December 2022 — the country’s first utility-scale transmission-side battery system, supporting a substation that serves 15 municipalities and roughly 2 million people.

Windey opened an energy storage integration factory in Camaçari, Bahia, in late June — its first overseas storage plant, with design capacity of 1.5 GWh a year serving Brazil and the rest of Latin America. Cell maker Cornex has signed a 1.5 GWh cell supply agreement with Windey, which puts the two Chinese companies in a matched pipeline for exactly the volume the Brazilian factory was built for.

Jinko ESS signed a memorandum with UCB Power, which has plants in Manaus and Extrema and has been in the storage business for over fifty years, to industrialise Jinko’s systems locally. BYD, which already assembles cars in Brazil, is reported to be planning storage battery manufacturing on top of its existing lithium iron phosphate plant in Manaus. Sungrow does not need a partnership to sell inverters into the region, and reports cumulative Latin American orders of 25 GW of PV inverters and 10 GWh of battery storage.

None of these moves is about the December auction alone. The national round is where a developer’s localisation strategy gets tested, and Brazil’s storage pipeline is already counted among the world’s ten largest. A template proven in the northeast of Brazil is a template for Chile, Colombia and Mexico, all of which are drafting auctions with similar local content language.

The Money Behind the Auction

Two separate pools of money are feeding Brazil battery storage, and they are easy to confuse.

The first is the auction itself. The government’s published estimates point to investment potential of around BRL 10 billion, or about USD 2 billion, and annual system savings of roughly BRL 3.2 billion (about USD 630 million) compared with firming the grid with gas-fired plants. Those are the ministry’s own numbers and they describe a procurement of a few gigawatts, not the 297 GW that registered.

The second is concessional financing. On 7 October, Brazil approved its fifth Eco Invest Brasil auction, in which BRL 4.55 billion (about USD 860 million) goes to batteries and storage systems — absorbing 70% of a BRL 6.5 billion envelope covering critical minerals, batteries and electric mobility. The programme allows up to 80% foreign participation and requires at least 10.25% of its innovation fund to go to projects jointly developed with universities and research institutes. One caveat worth noting: the storage allocation is not ring-fenced for grid-scale stationary storage, so the electric-vehicle battery supply chain can apply for the same money.

The obstacles the auction does not solve are fiscal. The Brazilian solar association ABSOLAR puts the tax burden on batteries at 84.8%, against 44.3% for natural gas, and links it to the BRL 7 billion-plus that renewable generators have lost to curtailment — power that was produced and then not used, which is precisely what a battery is for.

What 297 GW Does Not Mean

The headline number needs three qualifiers.

Registration is not qualification. EPE told the market explicitly that registered projects do not automatically qualify; the agency will run technical review first, and only then will the eligible volume be known. The headline figure is a measure of developer appetite, not of what will be built. Those projects are also spread across a country whose grid constraints are real: in a joint technical note published on 30 September, EPE and the system operator ONS set out how they will calculate the remaining capacity available at each busbar, and one finding matters commercially — the assessment is bidirectional. A substation that can absorb a battery discharging at a given power may still lose that capacity, or all of it, if studies show the battery cannot charge there under safe operating conditions.

The contracted volume is still officially undefined. Market expectations pointed to between 2 and 5 GW, equivalent to 8 to 20 GWh at four hours, but no figure has been published. Against the midpoint of that range, the registration book is roughly 85 times oversubscribed, which tells you how much capital is queued up behind a market that did not formally exist a year ago.

And the first Brazil battery storage auction is one auction. ABSOLAR is already arguing that the initial contract should be a starting point rather than a ceiling, and has proposed annual calls reaching 8 GW by 2030. That is an industry association’s position, not government policy.

Author’s Take: The number that will be quoted everywhere is 297 GW, and it will be quoted wrongly. Nothing close to it will be contracted; the useful figure is the gap between what developers registered and what the government is willing to pay for, because that gap is the clearest measure yet of how much storage capital is looking for a market and not finding one. The second thing worth noticing is that Brazil ran its industrial policy through an auction rule rather than a tariff. Splitting the tender into a local-content round and an open round did not exclude foreign suppliers. It converted them into joint-venture partners, because the national round is where the volume is and the only way in is to assemble locally. CATL with Moura, Jinko with UCB Power, Windey building its own plant in Bahia, Cornex tying a cell supply deal to it — none of that is charity, and none of it would have happened if Brazil had simply bought 3 GW of imported containers. For European and American developers and policymakers who keep debating local content rules without evidence, this is the evidence: six months of supplier behaviour, observable in public, in a market with no incumbent. The third thing is the contract design. Zero variable cost, fifteen years of fixed revenue, mandatory grid-forming, and reinvestment — including battery module replacement — inside the revenue allowance. That is a contract written by people who expect the asset to be used hard and to be rebuilt once, and it is a more useful model for grid batteries than an energy arbitrage contract. Watch the prices on 2 and 4 December. They will set the benchmark lenders use for every storage deal in Latin America.

The Bottom Line: Brazil battery storage auctions drew 6,091 project registrations and 296,807 MW for the country’s first two rounds, a national record, against market expectations of 2 to 5 GW actually contracted. The rounds run on 2 and 4 December, with one reserved for equipment meeting BNDES local content rules, fifteen-year contracts and supply starting August 2028. Projects need at least 30 MW, four hours of discharge, 85% efficiency and grid-forming capability, and they are paid a fixed revenue with zero variable cost. That design is why CATL partnered with Moura, Windey built a 1.5 GWh plant in Bahia, Jinko tied up with UCB Power and BYD is looking at Manaus — and why the auction is a localisation programme disguised as a procurement.

Notes: The auction rules and technical requirements are taken from MME Normative Ordinance No. 136/2026 and from legal and trade analysis of it; the 6,091 projects and 296,807 MW figure is the energy planning agency EPE’s registration result; the 2 to 5 GW contracted range is market expectation reported by trade press, not an official figure, and the government has not published the volume it intends to contract. The localisation deals and market share figures are company statements, including CATL’s own release, and have not been independently verified. The oversubscription multiple, the 16-times comparison against the pre-window industry estimate and the four-hour energy equivalents are our arithmetic from published figures. Registration and technical qualification are separate stages and the eligible volume is not yet known. EVsays did not attend any auction event, has not inspected any project and has no relationship with any company named. See our editorial policy and correction policy.

Sources & Further Reading

SHENG HE
SHENG HE

Sheng He is the founding editor of EVsays. He launched the site as an electric-vehicle news desk and has since expanded its remit to the broader electrification transition — batteries, storage, charging, robotics and clean power.
He spent eight years in automotive sales at the dealership level, working with multiple major brands — experience that gave him a front-line read on what buyers actually ask, fear and choose. That ground-level perspective now anchors the site's coverage of cars, batteries and the wider electrification shift.
He writes original, source-backed reporting for an international readership, with a reporter's instinct for separating confirmed fact from rumor.

Articles: 172

Leave a Reply

Your email address will not be published. Required fields are marked *