Edition #3 — May 2026
Welcome back to Behind the Meter. If you're new here, this is where I share what's caught my attention in energy every couple of weeks, the stuff that actually matters for how we manage utilities, procure power, and plan portfolios.
I've been digging through AEMO's Q1 2026 Quarterly Energy Dynamics since it landed at the end of April, and one number stopped me cold. Batteries set the spot price in 32% of dispatch intervals across the NEM in Q1. In New South Wales and Queensland, more than 40%. That's the first time any technology has displaced hydro from the top of that list.
I think most people in the market haven't quite registered what that means.
The Take: When the marginal unit changes, every contract has to change with it
For as long as I've been in this industry, the mental model for how the NEM clears has been fairly stable. Solar pushes prices down through the middle of the day, sometimes hard enough to go negative. Gas and hydro do the work in the evening peak. Coal sits underneath it all, slowly retiring. The forwards curve, every PPA, every retail tariff structure, every wholesale exposure analysis has been built on top of that picture.
That picture just changed. AEMO's Q1 2026 QED (summary on RenewEconomy) shows batteries setting prices 32% of the time, with combined charge and discharge intervals ahead of hydro for the first time on record. The driver isn't a mystery: another 4.4 GW of large-scale BESS came online in the year to March, and they're being dispatched constantly through both halves of the daily price cycle.
The bit that matters for anyone holding a contract is what's happening at each end.
Through the daytime trough, batteries are charging. That floors prices well above the deeply negative levels we saw in Q4 2025, when 31% of NEM intervals went negative and the mean negative price was minus $35/MWh. In Q1 it was minus $17/MWh, and the frequency dropped sharply in the northern regions. If you're a solar generator, that's a relief. If you're an offtaker on a PPA whose merchant tail was priced against a future of aggressive midday negative excursions, your assumed shape just shifted under you.
Through the evening peak, the same batteries are discharging. They're putting downward pressure on the prices gas peakers used to set, and they're doing it at a marginal cost stack that has nothing to do with fuel. Q1 NEM spot averaged $73/MWh, 12% below Q1 2025, with Queensland down 27% year-on-year. The evening ramp is still where the volatility lives, but it's no longer where the predictable upside sits.
What I keep coming back to is that the structure of price formation has flipped from fuel-driven to storage-driven, and almost none of the contracts written between 2022 and early 2026 reflect it. Forward curves can be re-marked. PPAs can't. If your client signed a fixed-price offtake against a swap-priced merchant tail two years ago, the maths underneath that hedge has moved.
The teams I've been talking to who are getting ahead of this are doing one specific thing: they're modelling the merchant curve with batteries as the marginal unit, not gas. It changes the answer in both directions. Daytime revenue holds up better than the old model said. Evening revenue softens faster. Most of the off-the-shelf valuation tools are still running yesterday's stack.
Quick Hits
-
AEMC's draft pricing review wants to move households onto fixed network charges. The Pricing Review draft report recommends shifting a much larger share of network costs into fixed daily charges, with the final report due June 2026. The Smart Energy Council ran the numbers and a household with 8 kW of solar and a 20 kWh battery could see bills rise $400 to $700 a year. The reason this matters for the C&I crowd: the same logic, applied to commercial network tariffs in the next regulatory reset, would compress the payback on every behind-the-meter solar and storage investment already in the ground. The household debate is the dress rehearsal.
-
NSW embedded networks get a hard price cap on 1 July. The Energy Legislation Amendment Act 2025 hands IPART pricing powers from 1 July 2026, capping embedded network charges at the median of the lowest market offers, mandating ombudsman membership, and bringing real enforcement penalties. If you're operating or advising on networks in apartment buildings, retirement villages, or shopping centres in NSW, the compliance window is weeks, not months. The third-party retailer workaround that has quietly held the existing pricing gap together stops working.
-
Victoria's commercial gas ban is eight months out. From 1 January 2027, all new commercial buildings in Victoria, including commercial kitchens, have to be built all-electric. This is a design and procurement deadline dressed up as a policy date. If you've got commercial property in the Victorian pipeline, the electrical capacity, equipment specs, and connection timelines have to be locked now. October is too late.
From the blog: If the merchant-curve discussion has you thinking about how tariff structures actually flow through to bills, this piece on network charges, time-of-use, and demand pricing walks through the mechanics for commercial sites in Australia. Worth a read if you're advising on cost control or stress-testing portfolios.
The shift from fuel to storage as the marginal unit is one of those changes that will look obvious in hindsight. The contracts written before it look fine on paper for a while, until renewal arrives or the merchant tail underdelivers and someone has to explain why. The Q1 QED is the first quarter where the new picture is in the data, not in the forecasts. Worth taking the time to look.
Until next fortnight,
— Cohen
Know someone wrestling with a renewal or a PPA right now? Send them this. They can subscribe at utilified.com/behind-the-meter. I'm also on LinkedIn if you want to argue with any of it.
If something in here sparked a thought, hit reply. I read every one.