Q2 2026: wind earned less, but the market fell further

What happened

NEM wind average output rose from 3,492 MW in Q2 2025 to 4,198 MW in Q2 2026, up 20.2%. Over the same period AEMO's published wind volume-weighted spot price fell from A$86/MWh to A$50/MWh, down 41.9%. Grid-scale solar output rose 11.8%, and its captured price fell from A$48/MWh to A$39/MWh.

Read on their own, those numbers tell a familiar story: more wind, lower prices for wind. That story is incomplete.

The number that changes the story

The whole market fell further. The simple average of monthly AVGRRP across the five NEM regions dropped from A$140.7/MWh in Q2 2025 to A$74.1/MWh in Q2 2026, down 47%. Apart from Q2 2020, it was the cheapest second quarter since 2019 in this dataset.

So wind's captured price fell slightly less than the market, and solar's fell much less. As a rough ratio against that average, wind went from about 0.61 to 0.67 of the market price, and grid-scale solar from about 0.34 to 0.53.

That benchmark is crude. It is an unweighted average of five regions, not the demand-weighted NEM figure AEMO uses, so treat the ratios as direction rather than precision. These ratios compare a NEM-wide generation-weighted numerator with an equally weighted regional benchmark. They are indicative comparisons, not like-for-like capture rates. For wind, "roughly in line with this benchmark" is the cautious reading.

A broad market decline is important context for the fall in captured prices. This comparison alone cannot establish its main cause or rule out wind and solar cannibalisation.

Why this matters for PPA valuation (my view)

When I model a PPA, I separate two things: the level of the market, and the discount for when the project generates. The absolute captured price mixes them together. Someone looking only at wind falling from A$86 to A$50 would conclude wind value is collapsing. Someone looking at this indicative ratio would see a smaller decline relative to the benchmark, while recognising that the weighting differs.

Those two readings lead to very different conversations about strike prices. The first says renegotiate everything. The second says the forward price level has moved, and the shape discount needs separate tracking. For a long-dated contract, the shape discount is usually the harder risk to price, because it can worsen as more of the same technology connects in the same region, depending on demand, storage and other supply.

Explore the historical PPA strike price comparison. It compares published fleet-level captured spot prices with a selected strike; it is not a project valuation or forward-price model.

What it means for retailers

Many retailers are offtakers. For a fixed-strike contract-for-difference PPA, if the settlement reference is below the strike, the offtaker pays the difference on the contracted volume. The published fleet average is not the settlement price of an individual project; contract terms, output and location matter. That is the hedge doing its job. The retailer is also buying its load at lower spot prices, so the net position depends on how well the wind volume lines up with customer demand.

That is the part a lower market does not fix. Wind output does not follow residential load. A retailer with a wind PPA still needs separate cover for the evening peak and for still, cold days. Lower spot prices reduce the cost of that gap, but they do not close it. And because regulated retail offers are reset once a year from forward market prices, a change in spot prices does not translate one-for-one into customer bills, and may take time to flow through.

What to watch

My expectation is that solar's improved ratio will not last into spring. Negative midday prices are already returning. Queensland's negative-price time rose to 16.6% in September 2026, and last October it was above 30%. If that pattern repeats, solar's captured price should fall harder than the market in Q4. That is a testable call, and the Q4 2026 QED report will allow the prediction to be checked using a consistent published benchmark.

Data and version

Output and captured prices reproduce AEMO Quarterly Energy Dynamics figures, pinned to this article's data version. The market average is calculated by Energy Data Hub from AEMO's monthly AVGRRP, unweighted across five regions. Ratios are approximate. Sections marked as my view are interpretation, not findings from the data.

Explore the six-quarter generation history · Captured price history · Method and limits

Sources and fixed data version

Underlying published figures © AEMO. Figures refer to QED Q2 2026, pages 23 and 25. Derived percentages use rounded source values. The original PDF checksum is unavailable.

The article text, charts and CSVs all use the same frozen snapshot. Updating Green Energy does not change this article. A source revision requires an explicit reviewed correction.

Download the frozen source dataset · AEMO copyright permissions