Are we heading in the right direction?

There have been two reports from AEMO this year that highlight the dissonance between what is actually happening on the ground in the National Electricity Market (NEM) and what is meant to happen over the rest of the decade.   Dissonance is always interesting – when two signals are pointing in different directions – the interesting question is which is right?

Focusing on the medium-term first, the AEMO 2024 Integrated System Plan (ISP) is a planning document that provides multiple scenarios for the development of the NEM over the period to 2050.  It is interesting because it is built around projections of hourly demand data and ensures that there is sufficient capacity to meet demand over the forecast horizon (including when the wind doesn’t blow and the sun doesn’t shine).  The ISP projects potential paths for a future NEM under different scenarios.

The Step Change scenario, which is considered the most likely, shows total capacity (excluding rooftop solar) growing from 65GW in 2023-24 to 109GW in 2029-30.   Within this, there is approximately 11GW of coal and gas retirements, offset by 54GW of additional wind, solar and storage.   The amount of capacity added is much larger than that removed, both due to growth in underlying electricity demand, and because each MW of retiring coal (which operates at 60-70% on a 24x7 basis) needs to be replaced by multiple MW of wind and solar and batteries. Within this 54GW of additions, 50% is forecast to be wind, 11% is forecast to be utility scale solar, and 39% is forecast to be storage (both utility scale batteries (largest portion) and also household batteries).

The key feature of this, is that wind is forecast to be the largest single component of new generation.  The ISP forecasts 4.5GW per year of new wind farm construction over the balance of the decade.

Why is wind so much larger than utility scale solar within the ISP forecast?

Wind dominates as a source of bulk energy in the ISP because:

• Storage is expensive and wind generators naturally generate a significant proportion of their output at night (usually more than half).  This makes wind the cheapest source of zero carbon emissions electricity to meet night time power needs (and we need lots of power at night).

• Utility solar generates at the same time as rooftop solar.  Rooftop solar, under current network tariff rules, has a fundamental cost advantage over utility scale solar.  Thus, the vast build out of rooftop solar – 23GW of capacity in 2024-25 and forecast to grow by a further 15GW by 2029-30, means there is limited opportunity for utility scale solar in the overall generation mix.  This dynamic is reinforced by the extremely low dispatch weighted price for solar projects (eg in Vic the average solar dispatch weighted price was $27MWh in 2023-24)

Ok that all makes sense, but what is happening in terms of the current build out?

In the Electricity Statement of Opportunities (ESOO) AEMO trumpets 5.7GW of projects reaching final investment decision/committed status over the past year.   This comprised 3.9GW of utility scale batteries, 1.2GW of utility scale solar, 0.4GW of wind and 0.2GW of hydrogren.

What does this say:

• the overall level of new commitments is too low – 5.7GW vs 9GW per year under the ISP;

• storage is overrepresented at 68% of this year’s new projects, compared to the ISP average of 39%;

• wind is running massively behind, at 7% of new projects - compared to the ISP expectation of 50%;  and

• solar is perversely high – at 21% of new additions compared the ISP forecast of 11%.

• So there is substantial dissonance between the short term action and the medium term plan – why and what does this mean

  
  

Part of the answer to why lies in costs.  The cost of building a windfarm has risen very substantially over the past few years.  Where Infradebt’s typical cost benchmarks used to be around $2/W – most projects we see these days cost more than $3/W.   This is reinforced by the latest cost prognostications from the experts (eg the Climate Change Authority sector pathways report quotes a benchmark cost for onshore wind of more than $3/W).

The challenge for $3/W windfarms is that their levelised cost of energy is basically $100/MWh.  There aren’t a lot of electricity users out there willing to enter into long-term $100/MWh plus offtakes.  This makes the viability of building wind challenging.   Layer on top of this an extremely difficult and slow environmental approval process, and projects increasingly needing to be built in more remote and/or more technically challenging sites with on average lover wind speeds, and it all just becomes a bit hard.   That is why the pace of wind construction is slow.  By contrast, solar and battery construction costs are falling after the spikes around Covid and Russia/Ukraine and there in inherently a much larger pool of potential project locations. 

In short, what’s getting built is what is easiest to build, not necessarily what the ISP says we need.

Taking this one step further, the ISP is based on a range of assumptions around what different technologies cost (e.g. cost of wind vs solar vs batteries).    These assumptions could be wrong.  In particular, if solar and batteries are sufficiently cheap, and wind farms are sufficiently expensive, at some point it becomes optimal to deliver nighttime power needs by building more solar and batteries.

Thus, one of the lessons of the last year might be that the generation mix in 2030 ends up being different to what is expected.

Watch this space.