Why solar parity scares big utilities
Thursday January 29, 2009, was a big money day for Victoria’s brown coal generators.
After a night of uncomfortably warm temperatures, and a dawn reading of 32°C, Victoria’s residents turned to their air-con and pedestal fans in near record numbers. By 9am, demand had spiked so high that electricity prices had soared to $10,000 a megawatt hour as utilities switched on every last generator they could find to meet demand. These wholesale prices are normally between $35-$50/MWh.
During that day, which reached a peak of 44.3°C in Melbourne in mid afternoon, the wholesale electricity price never fell below $1,000/MWh. For nearly four hours, it hovered around the $10,000/MWh price. The way the National Electricity Market works means that every generator switched on at that time receives that price, even though it still only cost the brown coal generators around $4/MWh to shovel the coal into their power plants. Over an eight-hour period, the state’s generators would have pocketed an estimated $550 million in revenue, near one fifth of their total revenue for the year.
It was, needless to say, an absolute jackpot for the generators. But while this was an extreme case, it was not an atypical event in the NEM. It is estimated that, on average, around one quarter of the revenue from electricity sales each year is generated from the prices gleaned from around 24-36 hours of peak production. The business models of the energy utilities depend on it. But now those models are under threat.
What, for instance, would have happened that day to electricity prices had there been large amounts of solar deployed along the eastern seaboard available to meet demand? According to modeling conducted by the Melbourne Energy Institute at the University of Melbourne, 5 gigawatts of solar PV would have been very effective in curbing peak demand. Prices would still have spiked, but not over $300/MWh, and for shorter periods. The total revenue for the day would have been just over $340 million – half of what it would otherwise have been.
This is what is known as the merit order effect: the effect technologies with a short-run marginal cost – i.e. with fuel that costs next to nothing, such as solar, wind – have on the market when they deliver electrons en masse to the grid. The overwhelming evidence from Australia and overseas is that they bring the wholesale cost of energy down, sometimes so much that the reduction in prices is greater than the cost of the subsidies that got them built in the first place. And established utilities with higher-cost fuel, such as coal and gas, don’t like it one bit, and are suddenly realising the extent of the threat to their business.
Normally, the overnight load in Victoria stands at around 5.5GW, not enough to even meet the output of all of the state’s coal-fired generators. Wholesale prices barely meet the cost of production. At 8.5GW of demand, an average high-load weekday, peaking gas generators are required and the wholesale price jumps to around $70/MWh. The coal-fired generators make money. But would this still be the case if 1.5GW of solar was available? The Melbourne Energy Institute says not; it would mean that gas-fired utilities normally brought into the grid would not be required and the wholesale price would remain at modest levels.
This might explain why state governments, in Victoria and NSW in particular, are happy to delay the rollout of renewable energy at a large scale. As noted here before, Victoria’s decision to defer an increase in its state-based renewable energy target was motivated by the potential impact of the merit order effect on the state’s coal-fired generators. Right now, the deployment in both wind and rooftop solar is at a virtual standstill because of policy uncertainty. That suits the established generators just fine.
Mike Sandiford, the director of the Energy Research Institute at the University of Melbourne, says that in the case of solar, this is simply delaying the inevitable. Grid parity – and the deployment of solar at a scale that the modeling contemplates – is coming whether the governments and the utilities like it or not, and it’s time policy makers faced up to the issues that it presents.
“We can either hide from grid parity or we can embrace the challenges,” Professor Sandiford said. “All we ever hear is that it is expensive, can’t deliver, or is not worth investing. We rarely hear of the opportunities.” The modeling of the 5GW solar scenario was extended to cover the entire 2009 and 2010 years. It found merit order savings of $1 billion in the first year and $600 million in the second, and avoided transmission and distributed investment.
As those figures show, it’s not just the generators that are impacted by this, it is the network operators as well. NEM data shows that while peak demand is growing, mean demand has plateaued and is now falling, possibly as a result of rising electricity prices, more solar PV, the merit order effect, or even the benefit of the pink batt program.
This divergence has meant that more infrastructure is being built to meet peak demand and is being used less during the day. And despite spending billions on network upgrades, and contributing well over half of the increased retail prices, the industry is losing productivity at a rate of 1 per cent a year. In most industries, this would be untenable. A dramatic increase in distributed energy such as solar would force the network operators to revisit the means to make money.
Professor Sandiford says most of the pubic discussion around feed-in tariffs and other green incentives is to pitch them as a form of regressive tax. That, he says, is way too simplistic. “There are other values here. It can shave peak demand, and it mitigates against extreme prices,” he says. “We are using less electricity at medium prices and more at the peak. It is important to know the answers to these problems before we go off spending. Whether we like it or not, it is going to hit us. We should try and understand what these issues are.”
This was a theme taken up by Richard Cowart, a visiting US expert on energy efficiency who spoke at the Energy Efficency Council national conference in Melbourne on Thursday. Cowart, a former head energy regulator in the American state of Vermont, said his state had avoided load growth over the past 20 years by investing in energy efficiency, rather than new power plants and networks. It did require, however, the regulatory reform that allowed utilities to be rewarded for efficiency, service quality, responsiveness to customers and environmental quality, rather than just “building more, burning more, and selling more.”
The payback was $2-$3 for every $1 invested in energy efficiency measures, because it lowered the cost of the service. In the EU, it has been shown that simply building more infrastructure is five times more expensive than implementing energy efficiency measures. Cowart now serves as chair of the electricity advisory committee for the US Department of Energy and is director of European programs for the not-for-profit Regulatory Assistance Project, which advises government on energy and environmental policy.
“Australia looks like it has a massive infrastructure investment program,” Cowart told Climate Spectator. “I don’t want to overstate the point here – I would suggest taking a very hard look at the billions that are proposed to be spent on generation, transmission and distribution in order to keep lights on. Questions should constantly be asked: Could we meet our energy needs more reliably and at lower cost if we invested more aggressively in demand-side resources? I suspect some fraction of those upgrades wouldn’t be needed if energy efficiency measures were rolled out.”
Follow @gilesparkinson on Twitter
Comments on this article
Hello,I love reading through
Hello,I love reading through your blog, I wanted to leave a little comment to support you and wish you a good continuation. Wishing you the best of luck for all your blogging efforts.1z0-043 dumps// SK0-002 dumps// XK0-002 dumps// 1Y0-A05 dumps// 310-065 dumps// MB2-631 dumps// E20-001 dumps// 1z0-048 dumps//
Economics 101
In an efficient competitive market, prices always trend back to the long run marginal cost of entry or exit. The price impact of subsidised entry (or removal) is only very temporary.
network costs are the real issue
Hi Jasper,
As far as I understand it, the team modelled 5GW. Today we have about 1.25GW on the grid.
The reason they chose 5GW is because we could have easily had that much on our grid. Germany has 6GW of PV in per capita terms and they paid a lot more than it would cost us to get the same level of penetration.
Air Conditioner growth is not fixed if the peak occurs at 6 or 7PM. Or if it happens in Sydney while it's overcast and 40C outside.
The saving in the network is something advocates often site but is not backed up by the correlation of peak demand with meterological data.
Whereas Merit Order Effect works and 85% of the high power price times correlate with PV generation times in the main cities of Adelaide, Melbourne, Sydney and Brisbane.
This is what the modelling shows.
network costs are the real issue
First of all, the article mentions 5 GW a few times, I am pretty sure this should be 1.5 GW.
While spot prices do not have an immediate impact on revenue, they do have large influence on forward contracts. So if solar PV eliminates volatility this summer, forward prices for next summer suffer (potentially very badly).
Finally, there is real potential for PV to starting to eat into the demand growth from airconditioning load that is the main justification for billions of dollars in network investment. This could be happening already.
In conclusion, low spot prices could make it harder for retailers to justify increasing electricity prices to consumers, while falling peak demand will wipe out the justification of network investment and therefore network cost increases. No wonder both retailers and network companies aren't happy.
cost to the average consumer of solar
If we investigate further the cost of installing 1.5GW's of solar PV on the average mum and dad's rooftop you can calculate the following;
1.5kW systems are currently sold to the consumers at around $3000 each, accumulating enough of these systems (1million installations) would equate to a total cost of $3billion. What must also be considered in this cost is the subsidy in price afforded to them, they are subsidised through the sale of Renewable Energy Certificates that are estimated to be generated over a future period (10 yrs i think?).
would an up-front cost of $3b be worth it for a saving of approximately $200m per annum? and this ignores the additional subsidies every electricity consumer pays for each solar pv installation - i think not.
I wonder what percentage of
I wonder what percentage of national roof top solar systems over 1.5 kw were sold by the Retailers? I wonder what the percentage of 1.5 kw systems and below we're sold by the Big Retailers?
I wonder what percentage of
.
Read ipart submissions
If you read Essential Energys submission to IPART they will tell you themselves that they don,t want PV customers on nett metering because it ruins their revenue base. Says it all really.
Concentrating Solar Thermal and House Design
I am heartened and excited by the Areva initiative installing Baseload CSP on southern EU and in the desert states of USA, a simple, cheap, quick, long life, non-threatening alternative to coal and Nuke. Areva are the regions biggest nuclear reactor manufacturers and they are moving on.
At the same time I am devastated by the apparent ignorance and lack of social conscience – or is it deliberate demand creating? of the builders of housing estates crawling across Melbourne's northern periphery. Hectares of huge mansions with NO eaves, NO balconies, BLACK roofs, and each with its new 5+KVA aircon. What ridiculous and incredibly arrogant design indifference. California bungalows were inappropriate in the 60's but now?? Who is responsible for this trend?
Where is it written that consumers have the right to demand ever increasing energy and services without thought of the consequences? The fear of blackouts is used as a persuasive tool to sell increased prices, and mock the alternatives. We should consider mandatory designs that can withstand blackouts, or even better, low energy designs that don’t create the massively expensive demand spike that cause them
At the same time we are building some of the worlds finest 6* offices. Looks like time for a uniform policy.
Forward Contracts
This statement is rather misleading "Over an eight-hour period, the state’s generators would have pocketed an estimated $550 million in revenue, near one fifth of their total revenue for the year."
What Giles doesn't tell you is that most of the electricity supplied to retailers is covered by forward contracts. So most of that money would have been returned to the retailers. Our retailers would go broke very quickly if they had to pay $1,000/MWh when they sell it to us for closer to $200/MWh including transmission and retail costs.
Duty Cycle controllers and The Smart Grid
In addition to the obvious merits of PV, consider the benefits of simple duty cycle control of domestic aircon units.
Put simply, the compressor can be cycled on and off and the fan left running.
This means that, on a very hot day with all the aircon units turned up and creating the peak load, the aggregate aircon peak load could be reduced by up to 25%.
A simple electronic unit can be retrofitted to each aircon unit and the user can adjust the duty cycle, Ie. compressor on for 3 mins, off for 1 min.
The fan is still running and the user will not notice any real reduction in comfort.
The device can be made quite intelligent and would be cheap.
They should be given out free and the cost of retrofitting by a licenced electrician subsidised.
Inclusion of this intelligence in new aircon units should be mandatory.
This functionality will be delivered by the Smart Grid when it eventually arrives, but we should be doing this in the meantime.
Unnatural conclusions
Currently, David Pethick's case falls apart at point 4: winter demand peak is mostly for heating, which is currently largely gas not electicity.
Of course, if we do the greenhouse-responsible thing and replace our gas heaters with heat pumps, it will be different.
But that's not a case for less planning. In fact, the whole point of the article is that solar is reaching grid parity regardless; so without any planning, it may actually achieve David's scenario.
If we plan to avoid climate change (as we should) we have to have strategic planning. The "free" market caused the problem; it's highly naive to expect it will also resolve it.
On that point, what we are really seeing is the creation of more genuine competition in the "market" for electricity. It has not been a real market since the NEM was created, but a narrow monopoly (or oligopoly) that jealously guards its privileges.
The recent entrance of large quantities of wind and solar are forcing the gas and coal giants to compete. Of course they don't like it!
A Cat amongst the pidgeons
What if you could store power and release it according to peaks and hollows in demand. There are several options. Batteries, water storage heat sinks even fly wheels, but at what cost and what efficiency. With the substantial base load in all states, the medium term future of coal is assured, the only real alternative other than nuclear is geothermal. This technology promises the perfect solution once it can be made to work reliably. In the interim we are looking at finding solutions to power peaks, based on scraping the cream off market costs to pay for them. Ultimately the market will compensate, and no one will lose, but if the situation is allowed to continue as it is, we will be looking at a pre paid pseudo GST.
Take it to it's natural conclusion...
1) Install *lots* of solar - get peak demand reduction in summer at a reasonable cost with solar PV running at close to nameplate capacity for a few years in peak times.
2) See further falls in spot price voltatility in summer.
3) Watch the business case for relatively low capital cost gas fired peaking plant (ie OCGT) fall apart. No one builds OCGT.
4) Winter demand peak hits. Insufficient capacity. Solar PV next to useless to reduce demand. A long, cold, expensive winter with lots of $12,500/MWh prices.
5) Brown coal fired generators rejoice - they get to run at full capacity and collect large winter price premiums from the market because there is insufficient OCGT or hydro to compete with them in the dead of winter!
I come back to a common theme. The market will (pretty efficiently) sort out the right fuel mix to meet Australia's needs. Central planning of the type you seem to advocate will have fairly predictable consequences followed by a whole lot of unintended outcomes.
a bonus, a glorious bonus!
I do remember a few years ago when a large bonus was paid to all workers form an electricity utility. The manager was very proud of the “achievement” and was “proud” that a large bonus could be paid to all workers. The large bonus was achieved because of a power shortage across Victoria and South Australia. The manager forgot to mention that the blackouts did cause a lot of harm to the wider community and that some elderly people died during the heatwave and blackouts.
If the well being of the community is at stake, then more solar PV installations are desirable. If we want more bonuses for a few and more hardship for many, forget solar PV installations.
Efficient
A good reference for espousing energy efficiency. Though i wonder if the Energy Retailers could ever get into the spirit and offer interest free insulation batts and efficient appliances for the home. It surely would take political pressure off governments to oversee efficiency programs. And $10/kWh for PV energy on hot days! Wow, is TRU Energy interested ?