The Big Switch: from energy to capacity
Cheques to flex?
Five minute scramble
Chart of the week: electricity sector employment
Australia’s approach to climate and energy policy is a masterclass in achieving second best outcomes. We couldn’t agree on constraining and pricing emissions, so it’s just easier now to turn climate change into an R&D project. This week, to save the National Electricity Market, the agencies running it decided to sacrifice its design. They look like getting support for the introduction of capacity payments to support the continued survival of dispatchable generators. This is a huge shift away from the energy-only market which has served the market since the late 1990s. The energy-only market could have been saved if governments were willing to increase the maximum price in the market, but that wouldn’t fly politically. We’ll take second best please.
It may not be easy or quick. The last big market reform – switching from paying generators in 30 minute blocks to 5 minute blocks – was first flagged back in 2016. It’s taken five years for the structural change to start on 1 October. Retailers are still rushing to get ready. The other big market reform California is hitting the pointy end of its summer peak season without much change in its generation capacity to last year when it ran out of power. Residents have been asked to turn down demand during summer peaks, but there are signs of consumer fatigue and demand response payments might be required. We take a look at this in this week’s bulletin. It’s not like paying for demand response is a new idea or anything.
The undamaged unit at the Callide C power station is back on line, but don’t expect wholesale prices to change much as they’ve been mostly driven by high gas prices. Chevron promised to honour its 80 per cent CO2 sequestration target at the Gorgon gas project, claiming it had already put 5 million tonnes underground. Transgrid CEO Paul Italiano left his role suddenly this week.
Twiggy watch: investors are getting frustrated with Fortescue’s constant media claims of proposed investments into clean energy projects without telling shareholders.
The Big Switch: from energy to capacity
In the world of electricity nerds the enduring Ford vs Holden debate of the past decade has been about how you pay for electricity generated: this has come down to two different models: energy-only market design and capacity markets.
Energy-only markets are hard core: they pay generators only for the electricity they dispatch into the system. It’s a pure, efficient, tough love market. Capacity markets are more like a modern-day primary school sports day: everyone wins a prize. Generators get paid not just for how much electricity they generate, but for hanging around so that, in the event they are needed, they are still there to help out. Energy only markets set the price, while in capacity markets, someone has to pre-set the value of capacity. That means invariably paying too little, or more likely, too much.
Possibly the biggest energy market policy news not just this year, but this century, is that the paragon of energy-only market purity, the National Electricity Market, could be switching to a capacity basis, albeit a more low-key version than its WA equivalent. The advice from the Energy Security Board is a proposed package of reforms that would switch the NEM’s allegiance by introducing capacity payments for generators.
The details in a moment. The big winners here are the A-team of governance agencies (AEMO, AEMC, AER) who are now united under the auspices of the Energy Security Board (ESB) and are brokering these reforms. This may signify the slowing of the gradual decline of national electricity market policy into state-based, go-it-alone, anarchy.
The other big winners are owners of generators that can provide capacity, but in particular anyone who is dispatchable, like coal, gas, hydro, batteries and pumped hydro.
The losers are renewable developers who wanted more of the same conditions that have been driving world record pace for wind and solar development: state-based intervention, government handouts and continued funding of renewables projects, which they build. Builders want to build. Owners want to run. This would be a win for the owners.
On early numbers the reforms proposed by the ESB seems likely to succeed. State governments like backing renewables, but they really hate blackouts. If the big states like NSW and Victoria get behind this, then it’s a done deal.
To understand this reform it’s important to understand what has changed. Energy-only markets work in a baseload generation system. That means big coal or nuclear power stations, running all the time, turning up in the morning and evening peaks and down at night.
This market uses highly variable wholesale prices to signal the stack of other generators to jump in and out. When prices rise to the hundreds and thousands of $/MWH, that’s not consumers being ripped off, that’s just the price it takes very occasional peaking generators to fire up. Consumers don’t directly pay those eye-wateringly high prices, they pay for the insurance policy – a cap or swap contract – that protects them against those prices. This is where you find the NEM’s capacity payments – hidden in the contract market.
This design doesn’t work so well when the grid has high renewable generation. Why? Because renewables are like a wild child. Wind will run crazy all afternoon and half then night and then suddenly stop at 4am. Solar will run when there is direct sunlight. These generators are highly indifferent to the price signals being sent by the market. They only want to switch off when prices get so negative they can’t earn government subsidies.
This is sending dispatchable generators to the wall because the more renewables are suppressing the price, the less often they can jump in and make enough money to remain commercially viable. If they shut down, then the same problem bedevils any new generators. They have to sit around waiting for when it’s dark and still. And then suddenly they are needed really, really badly.
The technical solution is simple enough: increase the maximum price paid to generators (called the Value of Lost Load or VoLL) so that the risk of paying it is higher and the value of the caps market is robust enough to keep firm generators alive. But, like the idea of introducing a carbon price, this requires a level of political maturity beyond modern-day Australian parliaments.
A capacity payment as proposed by the ESB accepts this reality, and pays enough to these generators still needed to keep the lights on to hang around. This will slow renewables build, and extend the life of these generators, which is bad for emissions. The alternative is crisis funding for emergency generators. Which is expensive.
What needs to go with this design is a way of ensuring emissions are factored into the capacity payments. Higher emissions capacity should exit first. Cleaner capacity should be better rewarded to reflect its lack of emissions. Of course this was how the National Electricity Guarantee (NEG) was supposed to work
It’s a huge reform in the 20-year history of the NEM. One that needs to be fully costed and the market possibly re-designed to reflect the changes. But it’s the electricity market equivalent of Dylan going electric: it changes everything.
Cheques to flex?
California has, like much of the Western US, been suffering from extreme heatwaves. And summer is only just getting going. The California electricity system is in transition and accordingly has been struggling to meet peak demand. Last year this resulted in rolling blackouts on the hottest days, and this year the grid has been pushed to the edge again on several days, most recently on Wednesday (July 28).
One of the ways California has avoided – or minimised – forced outages is by asking customers to voluntarily cut back their consumption. It uses a system called Flex Alert to put the word out in advance of those times when the grid is under most stress and request customers to conserve energy. There’s no financial reward for participating, just the knowledge that you’re helping out your fellow Californians.
Perhaps unsurprisingly, while many users don’t mind trying to use less occasionally to help avoid blackouts, it’s a different story when they’re asked to do it repeatedly. California is starting to experience “conservation fatigue” and finding that flex alert calls are delivering less load reduction than previously.
What’s the solution (apart from build a whole lot more dispatchable supply…)? California is now thinking it needs to set up more schemes to pay people for using less, or what is known as demand response. While it may seem obvious that financial incentives will do more than appeals to people’s better nature, this isn’t always the case. There are numerous case studies of unexpected responses to introducing financial incentives to a situation – in one classic case, when a childcare centre started fining parents who were late picking up their kids, the rates of lateness actually increased, as parents stopped feeling guilty about coming late and rationalised the fine as just an additional services they were prepared to pay for.
In the case of electricity systems though, there’s reasons to expect that financial incentives will pay off. What they can do is catalyse innovation and the development of services by retailers or others to simplify and automate demand response. This could include off-peak pool pumping, smart thermostats, cycling air conditioners, etc. They will then “clip the ticket” on the demand response reward to get a return on their investment in developing these demand response tools. Everybody wins, although it seems one retailer in Texas is trying to keep all the benefits for themselves.
Demand response is already common in many electricity systems for industrial and commercial consumers. This is because there’s usually more “bang for buck” developing demand response for larger users. These users also have more consistent daily usage profiles, so baselines – or how much a consumer would have used if they hadn’t carried out the demand response- are more robust than for residential users. Capacity markets such as the US’s PJM and the WEM in Perth are popular with demand response providers as capacity payments are the easiest ways to monetise demand response, which is why the NEM is introducing a separate wholesale demand response mechanism.
The trouble is that on very hot days it is residential air conditioners that are the biggest drivers of peak demand. And few people are keen on the idea of turning those off when the mercury is rising.
Five minute scramble
As the clock ticks down to 1 October and one of the biggest changes to the NEM since it was created electricity retailers are scrambling to get their systems ready. Changing the settlement period from every thirty minutes to every five minutes so it aligns with the dispatch period may not sound like a big deal, but it is a massive change to the systems that retailers, generators, networks and AEMO use to keep the NEM running and work out how much to charge customers.
Although the rule was passed in November 2017, market participants petitioned the AEMC last year for a delay in going live. COVID certainly didn’t help anyone’s preparations. In the end they got an extra three months.
Retailers don’t often attract a lot of sympathy, but five minute settlement is pretty much just a headache for them. They are incurring millions of dollars in costs to upgrade their systems. While other parties also have significant costs, networks can simply as the AER to include these in the costs they recover from customers and AEMO sets its own fees, which have been rising faster than energy prices in the last few years. But retailers can’t pass these costs on easily, especially now their prices are heavily regulated. And it’s not obvious that five minute settlement creates new business opportunities for retailers – it’s a compliance cost.
Unfortunately for retailers, this is likely to just be one example of a regular wave of system upgrades required as the energy system transitions. Having robust and flexible IT systems that are amenable to being upgraded to meet new rules and regulations is rapidly becoming a core competency of retailers. That it wasn’t already may seem curious. It’s easy to tell when the major retailers Origin, AGL and Energy Australia were updating their billing systems by looking at annual ombudsmen reports – the company that was upgrading would shoot to the top of the charts for customer complaints.
On the plus side, these challenges may help retailers better able to defend themselves at the likely incursion of new entrants into their market. Other businesses with large customer bases are eyeing up energy as an opportunity – not necessarily to compete directly as retailers, but they will certainly be looking to eat some of their lunch. Telstra and CBA are just two of the most obvious examples. As one columnist put it, energy retail may be coming down to who has the best database.
Chart of the week: Electricity sector employment
How many people does it take to power a lightbulb? According to the Australian Bureau of Statistics, a lot more than it used to. In the pre-competitive market 1990s when monolithic state-owned utilities supplied electricity, it took around 60,000 people to make around 160 terawatt hours of electricity. The introduction of a competitive market nearly halved that number of people, while making 25 per cent more power. Over the last decade while the amount of electricity has plateaued, the number employed have shot up again, back to around pre-competitive market days.
Chart 1: Electricity production and employment (ABS)
In labour productivity terms, this means it is taking more people to make the same amount of electricity, which is not ideal. This may be the result of largely maintaining the existing infrastructure of power stations, retail businesses, networks and agencies, while adding a whole new industry on top, since the increase in employment coincides with the increased investment and spending on renewables
In particular, it tracks the boom in rooftop solar. Most of the labour in renewables projects are in the building phase: they employ very few people when they run, so the ongoing management of renewables is unlikely to fully explain the big electricity workforce increase. And once we have built out the new renewables-based grid, we can expect to see a fall in electricity sector employment.
The increased wage cost of paying another 20,000 Australians hasn’t flowed fully through to power bills. Renewable schemes have a direct cost of between $76 and $170 a year on average household power bills according to the ACCC. That’s less than 10 per cent od the total bill.
Another reason that employment doesn’t correlate directly with cost is that wholesale electricity prices are largely set by the price of gas which tends to set the marginal price. Networks costs have largely stabilised over the recent years after a big run up early in the past decade.
We’ll take a deeper look into the drivers of electricity sector employment in an upcoming Background paper.