Even one year ago the target of 50 per cent renewable energy in the National Electricity Market (NEM) by 2030 was a wish by some, and perhaps a worry for others.
Now, in early 2019 it’s starting to look like a fairly conservative forecast. The pace of change is so fast that even experienced industry executives are struggling to keep up.
Australia is rushing into the future. In my opinion these eggs can’t be unscrambled, and nor should they be. All that regulators and policy setters can do is try to make sure the toast is cooked (that is, make sure enough transmission is built) and that the tomato isn’t too hot (enough dispatchable energy is around to keep the system reliable).
(Ed: And they are showing signs of understanding that and putting it into action).
Every customer, big or small, every industry professional, every fund manager, every banker, every government executive who wants to look like they know what they are doing should have a minimum 50 per cent renewable energy in their electricity forecasts by 2030.
It’s absolutely inevitable because it’s such an easy target. It only requires another 13GW of variable renewable energy (wind and solar), even ignoring rooftop solar, of unannounced projects, or just 1.2GW per year to meet it.
We have not forgotten rooftop, but we have kept demand flat and made an extremely crude assumption that the annual increase in demand equals rooftop supply. We also ignore electric vehicles, as their impact will be further down the track.
The energy coming from committed new supply has been checked several times by the ITK team and they could be a bit out without changing the overall message at all.
It’s a strong assumption that demand remains flat, but if does it implies that thermal generation is going to decline 65TWh by 2030 from FY18, that compares to the 170TWh it produced in FY18. That’s a much bigger decline than just Liddell, of course. Much bigger.
For the thermal industry to survive to 2030, pretty much only an immediate uptake in electric vehicles or the strong electrification of industrial power can provide the increase in demand to let them keep share.
All that said, the physical problems of internationally competitive prices, building the transmission and system strength foundations, and planning out the dispatchable power mean that it may still be “ better” to go a bit slower at the beginning, and then accelerate.
At the very least, the rate of change is going to require a lot of top-down management and a lot of industry co-operation. It’s also going to require some fast thinking on the form of financing and the role of government.
Academics and industry are starting to converge
This week I was lucky enough both to hear many of the leading academic voices in the climate change and renewable energy industry in the one place (ANU’s 100% renewable energy 3 day workshop) and also to interview Brett Redman, CEO of AGL (producer of 25 per cent of the electricity consumed in the NEM in FY18). (Listen to our Energy Insiders podcast here).
Most analysts become skeptical after years in the industry. If a cynic is, as George Carlin said, “a failed romantic”, a skeptic, to me, is a “would-be romantic”, or perhaps more prosaically, someone that can’t help looking for faults. And, of course, you can find fault in anything – just ask “Eve”.
So this week the ANU team published a piece which essentially forecast that the recent rate of installations of wind and solar, which they say is 6GW per year, will continue until 2032 and the NEM will be 100 per cent renewable. Actually they say “if,” but the paper says in part:
“The Australian renewable energy pipeline is sustainable in the long term “ They also say “The price of electricity from large-scale PV and windfarms in Australia is currently about $50 per Megawatt-hour (MWh), and steadily falling.”
I’d comment:
1. It’s wonderful to see such a positive attitude. It’s this sort of view that gets the world changed. Letting people see the possibilities is how progress is achieved. Also, the world and Australia needs rapid decarbonisation – far more rapid than out mealy-mouthed Paris Commitment.
However …
2. The cost of wind and solar in 2018 and I think 2019 is higher than $50 MWh. Headline PPA prices might be at that level, but bottom-up cost estimates don’t support those numbers and when you look into the detail of PPAs you find that: (a) an element of subsidy say $10/MWh is either assumed or built in; and (b) the PPA rarely covers all of the output over all of the life.
I am confident that developers typically assume, probably wrongly in my opinion, that they will earn more than the PPA price on the merchant output they retain. In many cases significantly more. There are many more subtleties to PPAs, but I will spare readers the detail
3. Cost is not the only driver of investment. Investment is driven by return expectations and assets that have pricing power, such as dispatchable power, can be far more valuable than price-taking assets.
Ask yourself who is making the most money in South Australia, wind producers with their low costs or gas producers with their high costs? Ask yourself who is making the best returns on capital in 2019 in NSW – solar farms or black coal generators?
4. Lots of new supply and a lower subsidy will reduce returns and increase risk. No analysis was done of this, but booms and busts are a feature of many markets. That’s not to say it can’t happen, just that investors need more than a theory to put $10 billion on the table year after year. Money is cheap, but it ain’t that cheap.
5. The assumption that you can install 6GW per year of wind and solar seems to completely ignore both the physical restrictions – the long build-time of transmission, the system strength issues, and the lower marginal value of each marginal unit of investment.
There is a developing literature on “spilled generation” and it’s blindingly obvious that at some point, not too far off, a unit of solar is of very limited value unless it has a unit of storage attached to it.
6. The disruptive impact of such a fast rate of install has significant possibilities to cause system weaknesses and to see dramatic pricing volatility and cause industry to really rebel. For better or worse, Australia has a bunch of industries that depend on a globally competitive electricity price. And certainly a stable system.
It may be technically possible to decarbonise Australian electricity at the pace suggested, but I think it would be both risky and require a fantastic amount of management of the sort that is perhaps only achievable if there was, say, a war on.
There is a war against climate change but there a lot of conscientious objectors and the sense of imminent invasion if we don’t pull together is missing.
7. Notwithstanding that electricity can shoulder more than its share of the burden, other sectors of the economy have to contribute. At some point the marginal cost of reducing the next unit of emissions from, say, cars will be less than that of another solar plant. Broader policy is essential.
Whatever the capital cost of 100% renewables, 50% is < $35 billion
That brings me to AGL. AGL believes it has plenty of time to invest and that the market will need AGL because the quantity of investment – $400 billion, according to management – will leave plenty of room for everyone. My view is different. 50 per cent renewable energy by 2030 is, from a financing point of view, relatively trivial.
The requirement is about 1.1GW new wind and solar per year, say 13GW in total, and that’s no more than $2.5 billion per year. Banks will supply at least half, and if efficiently financed, banks will supply two-thirds. Even at half, that is an equity requirement of just $1.2 billion per year. That’s really not a lot of money.
Transmission spending might require another, say, $4 billion, 70 per cent debt financed, and let’s throw in Snowy Hydro 2.0 capex of $4 billion (excluding transmission). Even if it’s not Snowy Hydro 2.0, it will be something similar in dispatchables and probably 50 per cent debt financed.
So the total equity commitment is less than $17 billion over 10 years. AGL could aim for 25 per cent of that, if it wanted to. So could Origin. The reality is that the federal and state governments, via reverse auctions, will end up as guarantors of the revenue and equity providers will basically fall over themselves to provide the equity.
It’s not just the loss of market share, it’s the negative impact that the SA/NSW interconnecter will have on gas generation profitability in South Australia, where AGL is the biggest player, partly offset by some small extra supply from NSW to SA.
This interconnecter is an absolute certainty, although it won’t be operating for three to five years. A potential opening in 2022 is another eye opener. 2022 is very ambitious, even in today’s world, but there is a willingness to build it.
AGL, Origin and EnergyAustralia are all struggling to keep market share. None of them has invested at the same rate as the broader industry in recent years.
Certainly AGL’s share of investment on balance sheet has been minimal: $200 million into PARF is more of a good-will gesture than a statement of intent. Investment in maintenance of coal stations is comparable to investment in new production.
AGL also states that it looks for a 14 per cent return on equity. That’s frankly an absurd rate of return when bond rates are 3 per cent. The cost of equity, even for a risky investment, is probably not much over 10 per cent at the most. If AGL sticks to 14 per cent it won’t be doing much investment and market share will fall steadily.
If there is going to be, as I think, at least 50 per cent renewables by 2030, then the share of coal-fired generation is going to drop sharply. AGL has the low-cost assets, so it will be some of the last coal generation left, but may still face increasingly sharp price competition as higher cost competitors refuse to give up without a fight.
That’s the flip-side of buying carbon-intensive coal generation assets for a low price. You get the return in early years, but in the out years it can get pretty tough. AGL’s price to earnings ratio of 14.3 X 2020 consensus EPS already reflects that. And that’s as early as 2020.
In short, in my view time is pressing in on AGL and, to a lesser extent, EnergyAustralia and Origin, much faster than they would like. Big decisions will have to be made, and made fairly promptly.
