Killing Queensland coal?

Australia's Renewable Energy Target (RET) cacophony

The Palaszczuk caretaker Government supports a Queensland RET of 50% by 2030. This is about double the Commonwealth's Australia-wide RET to be achieved by 2020 (but with subsidies continuing until 2030). South Australia, Victoria, and, especially, the ACT, also have very ambitious RETs (for the ACT, 100% by 2020).

How these different targets are reconciled within the so-called 'National Electricity Market' (NEM) is unclear. The Commonwealth Government's National Energy 'Guarantee' (NE'G') provides no detail. The NE'G' has not been agreed by COAG, and prospects for agreement look doubtful.

One member of the Energy Security Board suggested State RET targets greater than the national target would reduce their energy reliability or further increase costs compared with the national averages. How this will work is unclear.

In general, RETs require states within the NEM to supply a specified proportion of total electricity from renewable energy. The overall target generally is set by legislation, and financial incentives to supply renewable energy are supported by certificates received and surrendered to the Clean Energy Regulator.

In effect, governments quantify RET targets and offer subsidies to renewable suppliers to meet them.

Fossil fuel generators can't receive RET subsidies. In fact, they incur penalties as the RET target increases. How so?

Coal base-load basics

Existing coal-fired generators minimise costs by spinning 24/7. Their turbines take time to spin up and spin down. They last longest and operate most cheaply when allowed to keep spinning. So they are ideal as lowest-cost, base-load, generators. Grid stability itself is supported by this, and compromised when it's gone.

Until recently, Australia has been a global leader in low-cost energy generation. This was a huge competitive advantage. Queensland was a major beneficiary, using black coal power generation to fuel Queensland industry, and to export power to other States via the Queensland-NSW interconnector links.

But this cost advantage is particularly vulnerable to the time 'crowding-out' inherent in meeting RET targets.

The 24/7 ideal operating time for coal-fired generators can be directly reduced by imposing RETs. Coal plant cost-recovery for investment viability can be compressed into shorter generation dispatch times. Costs rise – a lot.

The two blades of the RET scissors

RET policies are dressed up only as a cost subsidy for renewable power. They are also a time, and therefore cost, penalty for base-load fossil fuel generation. RETs' emissions-cutting uses policy scissors. Scissors have two blades.

While governments stress the subsidy blade of the scissors, the fossil fuel penalty blade is powerful at cutting fossil fuel energy supply too. Policy ignoring this can lead to overshooting the RET. (Some think that's sneaky, but good.)

Households and businesses pay the extra costs from the penalty blade, as well as the renewables subsidy blade.

RET overshooting means higher costs, less reliability, and less grid stability than (presumably?) is intended.

Consider this example.

Suppose the all-up long run marginal cost to supply coal-fired base-load electricity is $1,680 per mWh per day, or $70 per mWh on a 24/7 basis. Suppose, in a 2030 Palaszczuk world, renewable energy (including battery storage, etc., as necessary) on average supplies all power 12 hours every day. If, as now in Queensland, the generation is mainly solar power, the cost-minimising optimum 12 hours (including battery or other storage as required) will probably be all the daylight hours. In that case, a 50% renewables target for 24 hours becomes a 100% target during daylight.

(This is like wind in SA. With the wind blowing, SA is wind-powered and has excess power it exports to Victoria.)

Let's be very clear about the nature of RETs. It's critical for an understanding of their implications if achieved.

AllRETs, the Palaszczuk caretaker government's 50% RET included, have been set regardless of weather (sun, wind or water) and demand peaks. Nights, cloudy days, the 'wet' season, summer and winter solstices, etc, have been ignored. The 2030 target must be achieved regardless of the weather; rain or shine, winter or summer. Costly multiplied generation (especially solar) and multiplied storage effectively have been promised by supporting governments.

For a 50% RET in Queensland, with battery or other storage as required, when the sun is shining (or even when it's cloudy), solar power plus batteries will get preference for all electricity dispatch. Multiplied solar generation plus battery storage costs a lot.

In this case, the daily cost recovery to ensure any coal-fired plant remains viable could be compressed into as little as 12 hours' dispatch from 24 hours' dispatch. What then?

Suppose, for efficiency, coal generators keep spinning 24/7. At best, they can recover costs from dispatch for half that time. If so, the hourly coal generator cost could increase up to $140 per mWh. Suppose instead coal generators are switched off every day when not dispatching power (a big operation, spinning up and spinning down every day). Any (small?) coal input cost savings must be set against the increased costs of much greater wear and tear and shorter operating life for the plant. Greater costs for grid stability maintenance must also be taken into account.

Allowing for the uncertain intermittency of renewable power, a prudent system-wide strategy, supporting coal plant efficiency, might be to keep coal generators spinning 24/7, even though dispatch is only for 12 hours on average.

This suggests two possibilities. First, dispatched base-load coal electricity could as much as double in price. Second, competition from other sources means coal plants can't recover costs and close. Either way costs.

Well d'oh. RETs are intended to encourage new renewables and force closure of fossil fuel plant. AGL's reaction to subsidised renewables and penalties for fossil fuel plant (eg, Liddell) is rational. The closure of Hazelwood is too.

The Prime Minister shouldn't complain. He supports RETs, if less ambitious than the Palaszczuk caretaker government, SA, Victoria, and the ACT. All must accept such policies are scissors with two blades, not one.

They own these policies. Businesses respond rationally to them. If governments don't like the responses, change the policies.

Queensland's coal-fired generator fleet: what will remain?

Queensland has eight black coal-fired generators (in no particular order):

• Stanwell (1,446.9 MW)
• Callide B (700 MW)
• Callide C (840 MW)
• Gladstone (1,680 MW)
• Kogan Creek (744 MW)
• Tarong (1,415 MW)
• Tarong North (450 MW)
• Milmerran (852 MW)

These have a total 'plate rated' output of 8,127.9 MW. Actual total output will be less, especially for older plant.

At present, very roughly, black coal averages around 80% or so of total electricity generation in Queensland.

The remaining 20% or so is split between gas and other fuels, and renewables (mainly small solar). Renewables might average about 5-6%.

The Palaszczuk 2030 RET requires a roughly ten-fold increase in the share of renewables in the next twelve years.

If this is 100% at the expense of coal, that more than halves coal's power generation share by 2030 (down to about 35%). At around today's generation levels, that implies an average cut in coal generation supply across all eight existing coal generators of about 56%. Alternatively, if the cut comes via closures, assuming the oldest plants go first, that could halve Queensland black coal generators, with Gladstone, Tarong, Tarong North and Callide B closing by 2030.

If coal power stays in business at all, Queensland coal prices still could as much as double. Total electricity prices would rise. How would the energy market in Queensland (and the NEM) respond to that outcome? Not well, I think.

If Queensland coal prices as much as double in twelve years, any remaining coal power could shut down as well. The main question might be the order of closure. Which would come first? Presumably the oldest (Gladstone)?

Will a 50% RET become an unreliable, expensive, fossil fuel killer? If so, whither Queensland?

Will a 50% RET in Queensland kill off coal power in that State?

Isn't it really, on average, a 100%(+?) RET during the day?

The time 'crowding-out' blade of the RET scissors – not advertised by proponents – could force coal costs uncompetitively high, while renewables are given a strongly subsidised ride.

Prospects for Queensland coal power look dire.

Worse, all existing renewables targets (the 100% 2020 ACT target aside) are sold by sponsoring governments as transition goals on the way to even lower emissions. 100% reliance on renewables 24/7 is the obvious end-point.

If this outcome is imposed through policy, regardless of the economics, the consequences will be even more adverse:

• Energy costs will increase further, and substantially.
• Grid stability will be compromised or require costly extra investment to correct.
• Investment in multiplied generation and storage will be needed to offset intermittency and low energy density.

Reliability will be threatened. If that threat is avoided, affordability will be pushed much further out of reach.

Queensland's competitive position within Australia, and overseas, will be further eroded.

Palaszczuk's RET ambitions may be achieved mainly by shutting down, or even reversing, Queensland's growth.

Is that really what we want?