Australia’s biggest states have set stunning renewable plans. Now we need a national strategy

The planned capital spend from the NSW, Victorian and Queensland electricity coal replacement plans is about $110 billion of utility-scale wind, solar and storage out to 2035.

If you add in rooftop solar and the Australian Energy Market Operator’s Integrated System Plan’s (ISP) modelled distributed storage, then that total leaps out to $180 billion.

If the plans are achieved, the three big states will see a cumulative 80 terawatt hours (TWh) of new variable renewables by 2030, doubling to 160TWh by 2035. ITK’s current estimates have the NEM at about 70% total renewable (including hydro) by 2030, but going well beyond 80% by 2035.

These grand plans will have variable renewable energy (VRE, or wind and solar) output increasing by 15TWh per year in the 2029-2032 timeframe; more than double what we are achieving at present. Broadly speaking, about 31GW of wind and 10.5GW of utility solar will be installed.

We make surfboards, why not wind turbine blades

I personally think the wind supply chain could do with a national plan. It’s great to have Rewiring the Nation and the impact of that is already visible, but a national wind turbine supply strategy would help.

If we install 31GW of wind, that equates to about 5100 turbines or, say, 400 turbines a year. If they were coming from two suppliers, each of whom could see a guaranteed 200 turbines a year then an attractive price, whole of life maintenance and the financial future of the supplier could be managed.

Equally, blade manufacture in Australia, from maybe two factories, would likely lower transport costs and provide a decade or more of work. We make surfboards, so laying carbon cloth and glue in a mold should be within our manufacturing capability. Even having a dedicated blade onshore transport fleet might be justified.

Although the timetable is short and because production will be concentrated in renewable energy zones (REZs) there is room for AEMO to double or treble its ability to connect generation per year and to reduce costs to producers, and thus consumers, in the process.

In short, although we now have policy-driven internalised targets, the detailed execution strategy could be improved.

It seems to me that what is needed is administrators and supply chain managers. And somewhere along the line the states may need to forego their reasserted provincial independence and bend the knee a bit further to NEM-wide management.

Wind and solar production growth has slowed

Since 2018 the NEM has been growing wind and solar production, including rooftop PV, by between 6-8 TWh of energy per year.

We’ve had about five year’s practice at that – enough time to learn the problems, the bottlenecks and supposedly to move down the efficiency learning curve.

In particular, I suppose a Martian looking at the forward plan compared to how it went in the beginning would say that the concept of targeted REZs is the biggest change. We, that is the industry, now identify a series of geographies for utility wind and solar and make sure that everything is in place.

In short, there has to be a development plan. The plan has to be agreed and the tools have to be in place to enable efficient execution of development. There are lots of details in the plan – this requires ongoing oversight.

I am far from convinced that the management of electricity has good structure at the moment. The ESB is not coordinating AEMO, AEMC, AER and reporting back through one executive to a notional Board which consist of the states and the federal government.

There is no executive charged with achieving the plan and the only penalty for failure will be voter displeasure. There is not even a definition of success or failure.

What we have, instead, three state structures plus South Australia and Tasmania and each calling on the federal government and AEMO as required. At best, the AEMC and AER will have done well to stay out of the way.

The second learning is that the new capacity has to be in place before the old capacity is closed and this means controlling the rate of closure of old capacity. That seems to be broadly in hand.

The third learning is that behind the meter is a really big deal and that new technology around grid controlling inverters provides new ways mange the electricity system.

The multiple state plans

The three big states for electricity production and consumption each based on coal generation have separately released plans to replace the coal generation in their state.

No state pays as much as lip service to any other state. In fact, for Queenslanders it’s a no-brainer that their plan must be bigger and better than that of NSW; and as for Victoria, well they started the whole thing.

It’s important to note in what follows that South Australia and Tasmania will also surely add utility production. The numbers and graphs I show allow for an estimate of their rooftop growth but not other stuff. It’s just that other plans are less formalised.

The following figure is derived from the various state plans.

Even in NSW, where a detailed annual tender program is available, it’s still not clear when the energy will actually be available because there is a gap between LTESA award and then the construction and energisation and hold point testing of the facilities. I allow a two-year lag from LTESA to full year production.

Including rooftop solar, the breakup by state is:

The figure is constructed on the basis of: (1) NSW tender program as presented by AEMO Services but shifted two years out to allow for  construction, (2) The announced Queensland program in MW each three-five-year time frame evenly divided and then average capacity figures for wind and solar, (3) Taking the ISP step change numbers for Victoria as outlined in the Jacobs modelling for the Victorian plan and then the Jacobs numbers for offshore wind, and (4) allowing for a steady 2GW per year of NEM wide rooftop development.

Adding up the numbers Queensland’s plan will indeed contribute the most.

Excluding rooftop, on my count there is 62TWh of production to come on line by July 1, 2030 and including rooftop about 80TWh.

The announced programs together with what is already operating should take the NEM to at least 66% renewable by July 1 2030.

The difference can be made up from other states and from what’s under construction or otherwise committed but not included in the big state targets. Unfortunately that’s a grey area.

Already underway

ITK identifies 2500MW of solar which is committed but not operating, including New England (400MW), Wellington North (350MW), Kiamal stage 2 (166 MW), Edenvale (204MW).

Those 2,500MW solar farms will produce maybe 4.5TWh per year, but some of them will either explicitly or implicitly be included in the numbers above.

There’s about 850 committed MW in each of NSW, Queensland and Victoria. The average size of a committed farm is 108MW in Victoria, compared to over 200MW in each of NSW and Queensland.

Looking at wind, there is 6000MW more or less committed. We include both stages of Golden Plains in that (1400-1500MW in total) with far superior economics to offshore wind.

Our definition of committed is generous for this note. It includes projects like Kogan West where there is announced support from the Queensland government, but no notice to proceed has been issued. Similarly, Uungula has a PPA from Snowy but no notice to proceed.

The difference can be made up from other states and from what’s under construction, or otherwise committed, but not included in the big state targets. Unfortunately that’s a grey area.

The Victorian report was done by Jacobs and relates to work published in March 2022 and forecasts prices between $60/MWh and $80/MWh

The Queensland “Energy and Jobs Plan” modelling was done by EY and released in September 2022.. The forecast is essentially for prices in the $60/MWh range and even lower.

I am sure everyone will have their own view about these forecasts. I will content myself with saying (1) in each case the forecast supported the proposed plan (2) three separate and independent and experienced consultants have produced price forecasts that are broadly similar to historic wholesale electricity prices. There is no doubt truth in the saying that “he who pays the piper calls the tune”.

I broadly support those price forecasts – and it is broad because the range is from $40/MWh to $80/MWh. The basis for the confidence is that wind and solar require around $50/MWh and the cost is expected to fall due to learning rates.

However, the capital investment will be heavy and the directly avoided fuel costs will be modest in comparison. Using the ISP and working from the top down.

My initial thoughts looking at that graph is that, even in the utility-only case, the wind and solar are only about two-thirds of the total capital investment, or as little as 55% if I include the direct consumer capital cost for DER as modelled in the ISP. The firming cost (pumped hydro and batteries) is more uncertain than the wind, solar and transmission costs.

The VRE component is calculated as:

The MW are taken from ISP step change 2035, less what was installed in 2024. I have not included offshore wind on the basis that less investment would be required in other technologies once offshore wind is included, but I’m not sure exactly how much. I am confident that if offshore wind is included total costs are higher.

Looking at firming, I get – using some very basic rules of thumb – an estimate of $58 billion.

Clearly, if you did everything with pumped hydro you’d end up with lower storage costs but higher transmission and you’d still need a lot of batteries to do your system services and you would need more investment in “last mile” distribution over time, which is not included here and not considered in the ISP or any of the state plans to any great extent.

And for transmission.

Transmission investment in the Queensland plan is extensive and would eventually have been required in the ISP, but later. As disclosed in the Queensland Supergrid document, the breakup is:

The avoided fuel costs will not of themselves justify the capex. Assuming black thermal coal costs of A$5/GJ (about double historic costs) I can get to about $7 billion of avoided costs per year. However, there are obviously annual operating costs for the wind and solar.

It’s the total costs avoided including the capex to rebuild the existing capacity which will be incurred that make the projects stack up and keep electricity prices were they are.