Black holes are dense. Their centres are gravitational 'singularities'. They have zero volume and infinite density. There's a region around each within which gravity is so strong even light can't escape: hence the label, 'black hole'. The boundary of this region is called the 'event horizon'. Any object falling within this horizon can't go back, only further in.

Renewable energy targets (RETs) impose anthropogenic cost 'event horizons' on base-load power. These 'horizons' are power cost points of no return. Once we cross them, there's no going back on power costs, unless revolutionary energy technology improvements (eg, 'cold' fusion?) allow us to overcome the arithmetic of power cost 'gravity'. Intermittency is the problem. Some solutions to the problem (eg, nuclear power) seem to be an antipodean taboo.

The inherent *intermittency* of renewables, like wind and solar power, combined with policy-imposed RETs, *force* base-load power *itself* to operate *intermittently*. Base-load generation is *the* foundation for reliable, cheap power *because* it operates continuously. It's not designed to be turned on or off all the time (and unpredictably). 'Peaker' generators, plus wind and solar, fill that role. 'Peakers' *can* generate all the time, but cost more. Wind and solar can't. They depend on the sun and weather. Priority to renewables power necessitates 'crowding out' cheaper continuous power generators.

Advertisement

Take wind as an example. Official data for SA shows, on average, wind turbines generated only 29% of their 'plate rated' capacity in 2016-17. To illustrate, suppose governments set a RET, to be delivered only by wind power, rated capacity of 100, of exactly 29% each year. That requires 100% of wind's 'plate rated' capacity to be used at least 29% of the year. To hit this wind-based RET of 29% for a year, policy in fact imposes a 100% RET when the wind is blowing. Therefore, base-load generators cannot supply power in competition with wind turbines for 29% of the time. Instead of operating continuously, base-load generators in this case can only compete to supply power, at best, 71% of the time.

It's clear policies favouring intermittent renewables, like RETs, force *all* power sources to become intermittent. Why? Because government policies protecting renewables dictate they must – or exit the energy market. For base-load power, curtailing supply hours destroys its economics. Generator costs must be recovered if base-load power is to be viable. Spreading the costs on a 24/7 basis keeps $/MWh costs as low as possible. Forcing such base-load suppliers to turn on and off means shorter operating times across which to spread their costs (net of savings on fuel inputs). There are extra costs from stop-start operation. The cost implications of this are illustrated in the chart below.

This chart is applicable to *any* wind, solar, or other renewables mix. It applies to any given $/MWh cost of base-load power in continuous operation. Importantly, it assumes extra costs from stop-start activity roughly offset fuel input cost savings from such intermittency. The net balance of these is extra detail. It does not invalidate the basic picture.

The chart shows the % decline in generation cost-competitiveness for base-load generators as increasing RETs make such generators operate more and more intermittently. Approaching a 100% RET, the required % increase in $/MWh costs for base-load viability (the % loss of cost-competitiveness) approaches infinity. The shaded area covers the range of RETs advocated by our governments. Black hole-like, this is the base-load power cost 'gravity well' curve we face.

Relative to continuous base-load supply, the chart shows:

Advertisement

- % cost increases (competitiveness reductions) for base-load power
*themselves* increase as RETs increase.
- With a 25% RET, base-load power costs increase (competitiveness falls) by about 33%.
- With a 30% RET, base-load power costs increase (competitiveness falls) by nearly 43%.
- At a 50% RET, base-load power costs increase (competitiveness falls) by 100%.
- At a 75% RET, base-load costs increase (competitiveness falls) by 300%.
- At a 95% RET, base-load costs increase (competitiveness falls) by 1,900%.

Many renewables enthusiasts will only see 'upside' in all of this. As current base-load power is forced to become more and more intermittent, subsidies and protection costs needed to make renewables cost-competitive will decline, more base-load will close, and rising RET targets will become a self-fulfilling prophecy. Greens will applaud.

Raising RETs ensures Australian policy on renewables pushes us within our power cost 'event horizons'. Reductions in subsidies needed to ensure renewables are cost-competitive will be outweighed as viable (but increasingly intermittent) base-load power costs *themselves* increase. All generators still operating, on average, will charge more if they want to stay in business. They'll *all *be more expensive, base-load and renewables alike. Will they, together, be as reliable?

There are many ways to quantify effects of renewables on overall power costs. Assume *all* costs required for *reliable* renewables, *plu*s cost increases as base-load power becomes intermittent, are reflected in average overall power costs. We can compile all-up power cost increases for various base-load, plus renewables, power mixes. The chart below assumes the average starting cost for power capacity is $50/MWh for 'base-load', and only 75% of that for renewables.

The chart above shows results for different RETs:

- 25% RET: power costs up 25% (wind) or 128% (solar), including a 33% increase in base-load costs.
- 50% RET: power costs up 132% (wind) or 412% (solar), including a 100% increase in base-load costs.
- 75% RET: power costs up 254% (wind) or 695% (solar), including a 300% increase in base-load costs.
- 95% RET: power costs up 352% (wind) or 922% (solar), including a 1,900% increase in base-load costs.

Using the black hole analogy, where are our base-load power cost 'event horizons'? These depend on assumed starting point power costs. Define any 'event horizon' as the RET share where base-load costs exceed total *reliable* renewables costs. On the starting assumptions in the chart above ($50/MWh for base-load, and $37.50/MWh for renewables):

- For wind, a RET of 47-48% is base-load power's wind cost 'event horizon'. Power costs increase for
*all* RETs.
- For solar, a RET of 85-86% is base-load power's solar cost 'event horizon'. Power costs increase for
*all* RETs.

I'm told black holes are nasty. Their gravity stretches us into 'spaghetti' as we near their 'singularities'. Best avoid RET power black holes too. They could end up 'stretching' Australian power affordability a lot. The 'good' news (?) is that, in the example used here, if we get to 100% renewables, power cost increases are 'capped' at 'only' about 277% (wind), or 878% (solar), or 577% (unweighted average of both). While very high, these increases are a lot less than infinity. At the 100% end-point of the RET policy push, the black hole/'singularity' analogy – like all analogies – breaks down.

At any RET value, including 100%, how much will our stretched energy budgets reduce *global* emissions, as others avoid *their* power black holes? As much as the volume occupied by the 'singularity' at the centre of a real black hole.

For all the cost, effort, and energy market disruption RETs entail for Australia, that's very small beer indeed.