Hazelwood highlights our tangled web of energy policy

Tuesday, Jun 28, 2016, 01:11 PM | Source: The Conversation

Mike Sandiford

The strangeness of last Sunday …

Has there ever been better times for our electricity utilities?

Sunday is typically pretty subdued in terms of electricity demand. Consequently Sunday wholesale electricity prices are at the low-end of the weekly range, even with the extra demands of a chilly winter day.

Since the beginning of 2008, the Sunday market price in Victoria averages about $23 per megawatt hour, factoring out the carbon tax component. That is about half the average weekday price of $43.

Victorian wholesale electricity pricing for Sundays from 2008 on. Note the extraordinary nature of Sunday 26th, 2016. Data from AEMO’s half hour price and demand datasets. Coloured solid lines show Sunday averages. Coloured dashed lines show the volume weighted price for working week-days. Black line shows Sunday 26th 2016. Colour distinguishes carbon pricing (blue) and non carbon pricing (red) intervals. Data sourced from AEMO.

Sunday prices do vary across the year rising slightly in the winter months. Between 2008 and 2015, June Sunday prices have averaged about $31 with a high of $50 on Sunday June 1, 2008.

So what can we make of last Sunday (26th June 2016), when the volume-averaged price in Victoria was a staggering $120, almost four times above the long-term?

Last Sunday was not just a once off, as the price for every June Sunday this year topped $75 - more than 50 per cent higher than any previous June Sunday in the last 8 years.

And Victoria isn’t alone. Right across the National Electricity Market (or NEM), prices have been stratospheric. Last Sunday, South Australia topped the class at $158, closely followed by New South Wales at $133.

As pointed out in my post a week ago, in historical terms these prices are simply staggering, and it’s not very obvious why.

If these prices remain, they will have widespread consequences, as they flow through to the contract market. So it is important to get to grips with what is going on.

Last Sunday was Melbourne’s coldest morning of the year, with several coldest-day records set across Victoria. So we might expect some extra heating was stretching the power system, helping drive up prices.

However, while the Victorian demand was higher than average for a Sunday, it was about 10 per cent lower than the Victorian average weekday demand. So we really can’t pin Sunday’s extreme prices on unusual demand. Moreover, last Sunday net Victorian exports amounted to about 15 per cent of its electricity production (see note 1). So there was no shortage of supply to meet local demand.

Victorian average Sunday electricity demand in megawatts, from 2008 on. Note while last Sunday (June 26 2016) demand was above the Sunday average, it was substantially lower than the weekday average. Coloured solid lines show Sunday averages. Coloured dashed lines show weekday averages. Black line shows Sunday 26th 2016. Colour fills distinguish carbon pricing (blue) and non carbon pricing (red) intervals. Data from AEMO’s half hour price and demand datasets.

Why are prices so high?

Rather than originating in Victoria, the reasons for the strange pricing events reside elsewhere on the NEM. They point to the emerging issue of the inter-regional coupling, the tangle of energy policies across the NEM and other energy sectors such as gas, and the power of industry incumbents.

In comparison to Victoria, Queensland demand last Sunday morning was significantly above expectation, slightly exceeding even the weekday demand average. Demand has been growing strongly in Queensland, partly because of the commissioning of its gas export processing plants. Still, at not much more than the weekday average, it is hard to understand why last Sunday’s demand in Queensland would be sufficient to drive prices to $115. As outlined below, with gas prices rising we would expect Queensland prices to follow if more coal-fired power could not be bought online. But at the time coal generation was running at around 75 per cent capacity. Why?

Queensland average Sunday electricity demand (grid-based) in megawatts, since the beginning of 2008. Data from AEMO’s half hour price and demand datasets. Coloured solid lines show medians for this period. Colours dashed lines show the average for all times of the week. Black line shows Sunday 26th 2016. Colour distinguishes carbon pricing and non-carbon pricing days.
Queensland wholesale Sunday pricing since the beginning of 2008. Data from AEMO’s half hour price and demand datasets. Coloured solid lines show medians for this period. Colours dashed lines show the volume weighted price for all times of the week. Black line shows the morning of Sunday 26th 2016. Colour distinguishes carbon pricing and non carbon pricing days.

As with Queensland, New South Wales demand was close to the highest recorded for a Sunday. But while New South Wales was importing furiously, at what would seem to be over the odds prices, its coal generators were running at only 68 per cent capacity. Why?

In South Australia demand was not exceptional in terms of recent history, but the recent closure of Northern coal-fired power plant has significantly tightened supply settings there and made it much more sensitive to the cost of marginal gas generators. Meanwhile its cleanest, most efficient, and one of its largest gas generators, Pelican Point, sits idle. Why?

Sunday was not only cold in South Australia, but it was also not very windy. As a consequence output from wind generation averaged only about 25 per cent of installed capacity. With only modest wind output, and with no coal, South Australia generation was dominated by gas, the price of which is surging on the back of increased LNG export pipeline flows (see also). In addition, South Australian consumption was heavily augmented with Victorian imports.

South Australian dispatch in megawatts for Sunday 26th 2016, showing the dominant role played by gas generators. In addition to local generation, imports from Victoria averaged around 300 megawatts over the day (see note 1).

The important point is that, for whatever reason, prices are being signalled right across the NEM, from Queensland and South Australia through to Tasmania.

This price signalling is having some paradoxical effects. Just last week, the Office of the Tasmanian Economic Regulator approved an increase in retail prices of 3.43 per cent, effective this Saturday, on the back of the high market prices in Victoria. The surprise is that Tasmania has just emerged from its own energy crisis with recent rains together with the re-connection of Basslink allowing exports for the first time in almost 10 months. Just as supply constraints relaxed for the time being and security restored though Basslink, long-suffering Tasmanian consumers are the first to be hit with price-hikes.

And in Victoria, where prices are responding to distant drivers, brown coal generators must be laughing at their good fortune.

Hazelwood’s “super-profits”

Across the last 4 weeks, market prices in Victoria have averaged $98 per megawatt hour when weighted by volume. In Victorian wholesale market terms the total value of dispatch was about $370 million. That is almost three times higher than the equivalent period last year.

Victorian price by 5-minute intervals, for the last 4 weeks. Red line shown the average prices weighted by volume. Note how regularly 5-minute prices have risen to $300 per megawatt hour as indicated by the dashed line.
As with the above figure, for the equivalent period in 2015.

While Hazelwood’s output will be contracted at much lower prices, with futures contracts already rising it looks like it is boom times for the owners of Victoria’s brown coal stations, like the French energy utility Engie that operates the Hazelwood power station.

Value of Hazelwood dispatch in terms of Victorian wholesale market prices, for the 4 week period from May 29 - June 25. Carbon tax contribution calculated using emissions intensity of 1.56 tonnes CO2 per megawatt hour. Costs estimated at $5. See note 3.

In Victorian market terms, the value of Hazelwood’s dispatch over the last 4 weeks was worth around $70 million (see note 3). That compares with the market value of around $20 million in June last year, and less than $10 million in June 2014 when the carbon tax liability is factored out.

On an annualised basis, the market value of Hazelwood’s dispatch amounts to around $900 million, a truly phenomenal turn-around from 2013 when the Commonwealth Bank sold its 8.3% stake that it had previously valued at just $1 million (see notes 2 & 3).

Hazelwood is Australia’s most contentious power station - a proverbial lightning rod for the debate about our energy system. As Victoria’s oldest, most emission intensive coal fired power station it has been plagued by community concern. At around 1.56 tonnes per megawatt hour it produces about three times the CO2 of a modern gas plant such as the super efficient combined cycle Pelican Point gas plant in South Australia,Engie’s other main asset in Australia. Already it is some 10 years over its use-by-date.

But, in terms of its balance sheet, Hazelwood has probably never seen such good times. Meanwhile Engie reportedly disputes liability for the $18 million costs incurred by the Country Fire Authority for fighting a month long fire in Hazelwood’s mine in early 2014.

While Engie is set to enjoy remarkable “super-profits” from Hazelwood, should we ask at what cost. With Victorian electricity exports to South Australia and New South Wales averaging around 1000 megawatts, Hazelwood arguably exists to underwrite those exports (see note 1). Meanwhile Engie’s Pelican Point Power Plant in South Australia sits idle, its gas contracts reputedly on sold presumably to the export market. If Engie were to fire up Pelican Point, it would reduce our national emissions by several million tonnes each year, and arguably put downward pressures on South Australian prices that would then be signalled across the country.

But perhaps that is exactly why. Oh what a tangled web our national energy policy is.

The headache for next Sunday

I doubt next Sunday, on the morning after the federal election, many will be watching the electricity prices. But some will, hoping the electricity market magically returns some normality. After all, it will not just be the first day of the new government, it is also the start of a new financial year.

Could it be that the current market prices are being manipulated by our electricity oligarchs to improve their bottom lines before the end of the financial year? Or will the high prices persist, reflective a fundamental shift in market dynamics, as the various threads of national energy policy tangle more tightly.

Surely we can understand that it is just not right to let the market dynamics encourage the likes of Engie to operate in Australia as if there is no tomorrow, rewarding it handsomely to dump some 15 million tonnes of CO2 into our atmosphere each year leaving Pelican Point idle. Would Engie be given license to do that in France. I doubt so, so why here?

But it is not so much Engie that is the problem, or any other player, as it is the whole tangled web of our national energy policy that is tightening at each turn. Either way, next Sunday will present the new government with the massive headache of untangling this web. We are in desperate need of a bipartisan approach with a sharp focus on how we rebuild our energy system fit for the challenges ahead, adding environmental concerns as an equal to the existing priorities of security and affordability. After all, last Sunday highlights that our existing energy system is not looking particularly affordable. And, like Hazelwood, much of it is beyond its use-by-date, so probably it is not that secure. And for sure it is an environmental disaster.


[1] The exceptional electricity pricing on Sunday 26th June 2016 is illustrated by the morning period between 8am and 9 am. Electricity consumption (TOT.DEMAND), production (DISPATCH.GEN), interchanges (NETINTERCHANGE) and prices (VWP) by the NEM regions. NETINTERCHANGE are +ve for export. Flows between the regions are coloured by the exporting region, and labelled by INTERCONNECTOR. Note that Victoria was exporting over 1160 megawatts, 380 megawatts to South Australian and 880 megawatts to New South Wales. In turn Victoria was importing 200 megawatts from Tasmania. The difference is essentially equivalent to the output of Hazelwood.

gv tree dot.

[2] Hazelwood Power Station was purchased by a consortium led by International Power in 1996 for $2.35 billion, before merging with GDF Suez, now Engie. In 2010, CBA revealed it had written down its 8 per cent share to $1 million “The Hazelwood power station probably wasn’t the best investment the bank made,” CBA chairman David Turner said at the time. CBA sold that share to GDF Suez for an undisclosed amount in 2013.

[3] Costs for Hazelwood production are reputedly very low, by AEMO accounts at least $2.5 per megawatt hour. Here I assume a value of $5. Note that revenues will be determined by contracts, and thus differ from the estimates of market values shown here. It’s noteworthy that Victorian baseload forward contracts have risen about 25% over the last quarter, and 35% over the year, to around $50 per megawatt hour.

The Conversation


Mike Sandiford receives funding from the Australian Research Council for geological research.