Stocktake of where Australia is up to with wind and solar

“The Fog of War”

11 lessons from Robert M McNamara

Tracking the (renewable) supply buildout

This occasional series will  present some information, mostly charts, on the build out of new supply in Australia. Price is a function of supply and demand. At the moment there is not much point in tracking demand because, after subtracting the behind the meter (rooftop) build out, apparent demand is flat.

Demand including that part satisfied by rooftop solar may be growing, but the change is insignificant relative to the impact of changes in supply. So that’s where we focus. We note that decision-making in fast-moving markets, but where the build lags the investment decision, is prone to error. Forecasting through the rear view mirror is the term oft used.

New supply investment still not very visible in current production

ITK forecasts something like a  22TWh addition to new supply from projects that we consider to be either under construction, have  reached final investment decision [FID] or are sufficiently likely to be included (eg rooftop solar).

Figure 1: Supply adjustments. Source: ITKe

A summary of where we are and where ITK thinks we are going  prior to the Liddell closure is as follows.

Note that the behind the meter estimates appear to be too low. Using a recent run rate (30 days to 26 September annualized) really should be seasonally adjusted to be valid. However, that adjustment will just have to wait for another day.

Figure 2: Current and new supply comparison. Source NEM Review, ITKe

In short, we expect about a 7% increase from here in aggregate NEM wide supply prior to the closure of Liddell.

Of course, Queensland is still far, far short of its stated 50 per cent 2030 target but there are signs that the state is continuing steadily along its path.

The bottom line here is that the new supply will suppress prices, but it’s still quite a guessing game to work out how much.

If new supply is going to come entirely from market driven price signals then it’s likely that new investment will slow. More importantly the transmission constraints will likely cause a slow down anyway. We note that the ESB has published on implementing stage one of the ISP and we will look at that subsequently

Price outlook. Forget the big models, its economics 101

Price and price expectations are driven by the change in supply. In the NEM there are three basic points about supply and its effect on price to remember.

• Renewable supply is increasing. The increases historically have been typically small individual units, so it’s the cumulative impact that matters and drives price down. The update in this note is that so far the increase in supply is minimal, so the impact on price is similar.
• Coal fuelled supply is reducing. Individual coal supply units in the NEM are big relative to aggregate demand so every individual reduction is a significant impact for price and economics. Each coal closure  tends to drive prices up. Hence the cycle, new renewables drive down price, coal closures drive it up again.
• Wind and solar supply can only respond to downwards price signals, not upwards ones. Wind and solar supply can be easily constrained or spilled (unlike say nuclear) but cannot increase supply in response to a positive price signal.
• As such wind and solar are typically price takers and not price setters. At the same time the zero cost of fuel for wind and solar means that they are nearly always dispatched as available.

Cyclical prices

It’s not hard given the above facts to predict that prices will be cyclical. In fact nearly everything that’s happened to prices could and was predicted years ago and by more than one person. The extent of the price move was underestimated but the pattern was and is  predictable to some extent.

An increase in renewable supply tends to drive average price down but perhaps the more important effect is it reduces the average revenue available for thermal generation (this might be gas or it might be coal). Short term prices become more volatile but initially there is excess thermal capacity to compete for dispatchable revenue and average prices fall.

Eventually thermal generation responds to the lower revenue opportunity by mothballing or closing some capacity. This may be triggered by management not wanting to sign new long term coal agreements or by maintenance capital expenditure. Because the coal units are big, the capacity withdrawal drives up overall prices.

The price increase impact may be particularly large because it may award more pricing power in the dispatchable market segment to the remaining generators. That’s particularly the case  where there is some concentration of ownership of thermal generation as in the NEM. The few remaining coal plants have even fewer owners. Dispatchable pricing is also exagerated by the scarcity pricing of gas.

PT – policy trilemma, and it is a trilemma – not a dilemma

The policy trilemma [PT] is well defined and understood around the world and the Finkel Report made it explicit in Australia – prices, reliability and decarbonization. To us, the way forward has always been clear: Policy should aim to get new low carbon supply into the system ahead of the thermal closures.

That said, the new supply will accelerate the thermal closures, so at the same time there must be policy to manage reliability. We don’t think the complex NEG was that policy, but neither do we think that relying on the oligopolistic structured market provides enough comfort to big business.

Finally because there is still far too much thermal capacity in Australia, there is no immediate drama on dispatchablity. In fact the actual  problem is how to accelerate the coal reductions but leave reliability in place.  It would be easy enough if we could agree.

Enough on policy.

14% of supply is presently coming from wind and solar

All figures in the charts below unless otherwise indicated are sourced from NEM Review

We look first at the combined wind and PV share of total supply. When I look at this chart and think about how much has already been sacrificed in Australia in terms of work and careers and policy fits and starts at Local, State and Federal levels  to achieve a relatively modest outcome it puts me in mind of the ANZAC spirit. Folks we are in the trenches and bogged down.

According to ITK and various other estimates there is enough committed and under construction wind and solar generation to get the NEM to about 24% of total production, but clearly that is coming along slowly.

Annualised output is about 27 TWh – should grow to around 47 Twh

Figure 3: Wind & solar share of output. Source: NEM Review

Let’s break the above chart down into wind, rooftop  solar, and utility solar.

Figure 4: Wind and solar share by type. Source: NEM Review

It’s clear that wind is the main game right now, and that utility solar is slow to ramp up, still only 1.3% of total generation in the past 30 months.

The higher variability of wind on a monthly basis is also clearly shown  although again as more wind is built in NSW and QLD the variance may reduce.

Individual projects

Ideally we’d to look at the capacity utilization of individual projects. However, that’s not a useful thing to do when both new wind and PV plants startup in stages.

So what we have done is to take all the individual wind and solar projects in the NEM Review database, and this comes from registered generator data from AEMO, and looked at the annualized output from the last 30 days.

This, too, is unfair to many wind farms if the last 30 days happen to be low wind, as they were, but it nevertheless gives a feel of the relative importance of some projects.

Note also that some of the bigger projects, specifically the 530 MW Stockyard Hill where first concrete pour has just taken place and the 453 MW Coopers Gap wind farm where first turbine blades have been delivered are still a some way off starting operations.

Stockyard Hill is well behind where we thought it would be, it was announced months before Coopers Gap.

25 solar farms operated during the month

However, we excluded the Gullen Range project because there may be a data issue with the wind and solar output from that project getting mixed.

The interesting feature about the solar projects is that some of the new projects are already putting out more than the well established Moree, Broken Hill and Nyngan projects.

Expect the numbers in this chart to grow a lot over the next 12 months, but I would also say that several projects are at least 3 months and maybe 6 months behind where we expected them to be right now.

It’s important to the system that as much new capacity, both wind and solar, be on line in the coming Summer. Last Summer passed without any drama but to some extent there was a bit of luck involved. Luck cuts both ways.

I’d also award some brownie points to Clare and Sun Metals operations management for getting out in front and getting their farms up and running pretty quickly.

Figure 5: Solar farm current output. Source: NEM Review

50 wind farms operated

Figure 6: Wind parks current ouput. Source: NEM Review

David Leitch is principal of ITK. He was formerly a Utility Analyst for leading investment banks over the past 30 years. The views expressed are his own. Please note our new section, Energy Markets, which will include analysis from Leitch on the energy markets and broader energy issues. And also note our live generation widget, and the APVI solar contribution.