Australian BESS Investment Risk: What Is
Already Happening and What Comes Next

On 18 October 2025, a high-voltage transformer supplied by Wilson failed during final commissioning tests at the Waratah Super Battery — an 850 MW / 1,680 MWh facility operated by Akaysha Energy and backed by BlackRock. The transformer was deemed beyond repair. A second transformer also sustained damage. The facility's operational capacity fell from 850 MW to 350 MW: a 59% reduction at one of the NEM's most anticipated assets.[NARDAC/Energy Storage News]

Dr Tom Harries of NARDAC — a specialist energy and infrastructure insurance broker — estimated total losses between AU$50 million and AU$80 million.[Energy Storage News] The lower figure assumes a replacement transformer can be sourced within six months under delay-in-start-up insurance cover. Main power transformers of this specification typically require 12–18 months to manufacture and deliver, which would push losses toward the higher end. The facility entered a planned balance-of-plant shutdown from 20 November to 2 December 2025, reducing available capacity during peak summer demand. Full commissioning remains delayed into 2026.[Energy Storage News]

The Waratah incident reveals a risk that most BESS financing models treat as remote: a single critical piece of equipment — not a battery cell or an inverter — can knock out the majority of a project's revenue-generating capacity for over a year. There is no quick domestic substitute. This is the most important operational precedent set in the Australian BESS market and it will directly influence how lenders price equipment failure risk in new project finance facilities going forward. The Hornsdale Power Reserve, operated by Neoen in South Australia, provides a separate precedent from 2021: the AER fined the facility AU$900,000 for failing to deliver contracted backup power services over a four-month period in 2019.[AER] That enforcement action showed that performance obligations under ancillary service contracts carry real regulatory teeth.

BESS assets in the NEM earn revenue through two primary channels: ancillary services (frequency regulation, contingency responses) and energy arbitrage (charging cheap, discharging dear). The ancillary services channel — historically the more reliable of the two — is being eroded by volume. As AEMO's 24 GW 2030 target drives rapid commissioning, more batteries compete for the same pool of frequency regulation payments.[Allens] Per-project ancillary revenues fall. The energy arbitrage channel is exposed to NEM spot price volatility and, increasingly, to negative pricing events during high-renewable periods — exactly when batteries are most needed but also when the grid has the least capacity to absorb discharge.

The lender response confirms this is not theoretical. Australian project finance lenders are now imposing covenant restructurings on BESS facilities that have missed merchant revenue forecasts, conditioning debt extensions on the addition of adjacent storage and requiring strategic revenue resets rather than simple amortisation deferrals.[Indexbox] These restructurings are happening on assets co-located with solar, where negative pricing has cut into the revenue assumptions embedded at financial close. The signal is clear: original merchant revenue models are not holding under live NEM conditions.

The market's response to this pressure has been to move toward contracted revenue. Allens' March 2026 analysis of Australian energy investment notes that 20-year, 100% offtake agreements with major energy companies have become the dominant structure for new BESS projects, driven by developer unease over standalone merchant viability.[Allens] A specific data point: AGL and Engie have agreed a five-year virtual battery swap starting 2027, backed by a NSW BESS portfolio — a structure designed precisely to reduce merchant exposure.[Core Markets] Private equity has largely replaced traditional infrastructure funds as the dominant capital source in the merchant segment, because infrastructure funds cannot tolerate the revenue volatility that merchant BESS now involves.[Allens] No granular NEM spot price series or project-level LCOS estimates were available from the sources consulted; confidence in the precise magnitude of revenue compression is medium.

Australia holds the world's largest lithium reserves and significant cobalt resources. It is not, however, a major producer of battery cells or battery management systems. The processing and manufacturing steps that convert raw lithium and cobalt into the cells that go into BESS projects are dominated by Chinese producers — a concentration that Australian government policy has identified as a strategic vulnerability but has not yet resolved.[Austrade]

Lithium carbonate prices have rebounded sharply — up approximately 89% year-on-year to US$20,000–20,600 per tonne as of February 2026, after falling roughly 85% from their 2022 peak due to oversupply from Australian, South American, and Chinese producers.[Canadian Mining Report] This rebound reflects genuine demand growth: energy storage system deployments grew 71% in 2025 and are projected to grow a further 55% in 2026.[Canadian Mining Report] For Australian BESS developers, volatile lithium prices are a double-edged exposure — they affect both the cost of procuring battery systems and the valuation of any upstream mineral assets held by integrated players.

The cobalt picture is more structurally concerning. Cobalt Blue's Broken Hill Cobalt Project (BHCP) and its associated Kwinana Cobalt Refinery (KCR) — which were intended to provide a non-Chinese source of high-grade cobalt metal for EV and storage applications — both remain paused, with development suspended through 2024–2025 due to low cobalt prices. A three-year Major Project Status extension was granted by the federal government in July 2025, signalling continued policy support but no near-term production.[Austrade] The Sunrise Project, with approximately A$250 million invested for nickel and cobalt sulphate production, is development-ready but not yet operating. Until these projects reach production, Australian BESS projects have no credible domestic substitute for Chinese-sourced battery inputs.

Two regulatory pressures are arriving in close succession. First: the Firm Energy Reliability Mechanism (FERM) introduces liquidity obligations for long-duration storage facilities in South Australia from 1 July 2026, requiring BESS operators to demonstrate they can deliver firm energy rather than simply frequency services.[Ashurst] This restructures the revenue case for new South Australian projects — assets that were financed on an ancillary-services-heavy model now face contractual obligations that may require longer discharge durations than originally designed. Second: AEMO's General Power System Risk Review (published 2025) identifies minimum system load conditions, protection system interactions, and potential thermal overload to transmission circuits as live operational risks — all of which become more acute as coal generation retires and batteries take on a larger share of system balancing.[AEMO]

The coal retirement timeline is the underlying pressure. AEMO's 2025 Integrated System Plan work and supporting Infrastructure Australia analysis confirm that approximately 1,700 MW of coal and gas generation will exit the NEM by 2032, with the replacement mix heavily weighted toward renewables plus storage.[AEMO][Infrastructure Australia] BESS assets are therefore being asked to provide services — firm energy, inertia, system strength — that have historically been provided by synchronous generation. The technical and regulatory frameworks for compensating batteries for these services are still being developed, creating a gap between what the system needs and what the current market design pays for. The ACCC's July 2025 National Electricity Market inquiry report flagged ongoing concerns about market concentration and price formation that are relevant to how ancillary services revenues evolve as the generation mix changes.[ACCC]

The Waratah transformer failure has already changed the insurance conversation. NARDAC's analysis of the incident — the first major HV transformer failure at an Australian utility-scale battery — draws attention to delay-in-start-up cover as the critical insurance product and to the 12–18 month lead time on main power transformers as the variable that determines whether a claim is manageable or catastrophic.[Energy Storage News] The question that follows for every other large BESS project under construction or in early operation is whether their delay-in-start-up coverage reflects actual transformer delivery timelines — or the optimistic assumptions embedded when the project was first underwritten. If insurers adjust their underwriting standards in response to Waratah, developers and lenders face materially higher insurance premiums or coverage gaps on projects that were financed under earlier terms.

Grid saturation in specific NEM regions is the second risk. AEMO's forecast of 24 GW of battery capacity by 2030 is a national figure, but the NEM is not a single grid. Transmission bottlenecks in high-renewable zones — particularly in South Australia and parts of Queensland and Victoria — mean that batteries in congested regions can be physically unable to discharge when prices are highest or absorb when prices are lowest. This curtailment risk is not hypothetical: Allens' March 2026 analysis explicitly identifies grid bottlenecks limiting discharge and charge capabilities as an active constraint on NEM BESS assets.[Allens] As commissioning accelerates toward the 2030 target, the probability of regional saturation in specific corridors increases.

The third emerging risk is duration mismatch. AEMO's WEM ESOO data from September 2025 identifies a growing need for batteries with six or more hours of storage duration to support coal retirement, while most operating and pipeline Australian BESS assets are sized at two to four hours.[AEMO ESOO] Assets financed at two to four hours may face a market design environment — particularly post-FERM — that increasingly values and compensates longer-duration assets, leaving shorter-duration projects in a structurally weaker revenue position than their original models assumed. This is not yet confirmed policy but the direction is clear from the AEMO technical analysis.

Revenue compression from ancillary services saturation is the highest-priority risk because it is both already materialising and structurally driven — adding more batteries does not solve it, it accelerates it. Lenders are already responding with covenant restructurings, confirming the financial consequence is real and current.[Allens][Indexbox] Equipment failure risk is rated equally high after Waratah: a single transformer failure generated AU$50–80 million in losses and the conditions that caused it — reliance on single-source, long-lead-time components — exist at every large BESS project in Australia.[Energy Storage News]

Supply chain concentration in Chinese cells and BMS is high impact but medium likelihood in the near term — not because the concentration is less real, but because an acute disruption requires a geopolitical trigger that is not currently confirmed. The consequence if it materialises is severe: there is no credible short-term domestic substitute. Regulatory transition risk sits at medium-high: FERM commencing 1 July 2026 is confirmed and imminent, but the direction is broadly known and operators have had lead time to prepare. The unknown is implementation quality — particularly whether the firm energy compensation frameworks are priced adequately from day one.[Ashurst]

Grid saturation in specific NEM regions is currently medium in both dimensions but is the risk most visibly moving in the wrong direction. Commissioning is accelerating faster than transmission investment. If the 24 GW 2030 target is approached without regional grid augmentation, curtailment in congested corridors will convert a theoretical risk into a financial one within the next 24 months.[AEMO] Investors with projects in South Australia, western Victoria, or south-east Queensland should treat this as a near-term watch item rather than a medium-term planning assumption.