Nuclear power ‘will be tough to sell into the grid’

The headline cost of building nuclear reactors will dominate political debate. But requiring a return on that investment will have a far greater impact on power bills.

A Vogtle nuclear power plant under construction. Picture: Reuters

Forget the headline cost of building nuclear reactors in Australia. While not entirely irrelevant, the total cost of Peter Dutton’s ­nuclear build is far less important than the question of how his government will sell nuclear power into the grid.

Delta Energy boss Richard Wrightson made one of the most salient points so far in the nuclear debate at The Australian’s Energy Nation Forum on Wed­nesday.

Coal-fired generation is not being pushed out of the grid solely because the east coast generation fleet is ageing – though that is part of the equation.

Coal generators are being retired because they can’t compete with cheaper power being sold into the National Energy Market by solar and wind.

“I’m not that cheap as an ­energy provider. But I am a ­reliability provider. I can’t really compete with renewables, they’ll push me out of the market if I don’t get paid for reliability,” Wrightson said.

The real question hanging over the deployment of nuclear energy in Australia – and the ­Opposition Leader’s claims that it will drive down energy prices – is how the costs of building those seven reactors will translate into wholesale power pricing.

Opposition energy spokesman Ted O’Brien and treasury spokesman Angus Taylor have both made it clear that the federal government won’t be in the business of subsidising those costs in the long term, and their government would expect a return on investment from building and owning nuclear reactors.

Construction cost estimates remain under wraps, but the ­Coalition’s funding model will mirror that of the National Broadband Network, Inland Rail and Snowy Hydro 2.0.

The government will pump in public funds to take equity in the state-owned company that builds and runs them, and which will (presumably) then sell the energy to the NEM, as other generation assets do.

That means the construction costs won’t blow up the national accounts – rather than adding to spending deficits, they will, in fact, appear on the other side of the balance sheet as an asset.

But it does imply that a commercial return will be expected – both in terms of an operating profit, and the eventual repayment of the capital invested.

An image posted by Peter Dutton of a concept design of a small modular reactor.

How quickly that return is expected, and at what rate, are key details that eventually will mean much more for household bills than just the total construction costs of nuclear reactors.

The latest levelised cost of energy report from advisory giant Lazard illustrates nuclear’s cost problem. Lazard’s June report puts the cost of nuclear power in the US from fully depreciated reactors (where the construction costs have been fully repaid, effectively), at about $US32 ($48) a megawatt hour.

The same report tips the cost of power from new reactors at $US142-$US222 ($213-$333) a megawatt hour. Lazard’s costs are based on publicly available figures from the new Vogtle nuclear units in the US state of Georgia – which use the Westinghouse AP1000 technology, one of the options nominated by the opposition in its plans.

Similarly, recent WoodMac analysis estimates that the current cost of conventional large-scale nuclear power in the Asia-Pacific ranges from $US48 a megawatt hour in China to $US110 in Japan.

By way of comparison, average wholesale energy costs in the NEM in the March quarter ranged from $69 a megawatt hour in Tasmania to $137 in Queensland.

Lazard’s costs were drawn from a large and mature nuclear market, with considerably more experience with nuclear power than Australia will develop in the 10 to 15-year timeline set by Dutton’s policy.

Vogtle’s latest two units were the first major nuclear generation assets built in the US in 30 years, and suffered massive cost and timeline blowouts as a result (seven years behind schedule and $US16bn over the initial budget, according to the US Energy Information Administration).

As new nuclear construction accelerates globally, it’s fair to ­assume those costs will fall and ­become far more predictable.

But, given the massive rollout of wind turbines and solar panels globally, it is also fair to assume that the cost of renewable generation will also keep falling, along with battery firming solutions.

The key to nuclear’s role in setting electricity costs will be how the state-owned business – to be named Affordable Energy Australia – sells its power into the grid.

The NEM is a dynamic market. Some big energy users, such as Rio Tinto, Alcoa and others, can cut direct power offtake deals with generators. Their willingness to wear the electricity tariffs required to repay nuclear’s capital costs, if cheaper and reliable renewable alternatives are available, will help define the business case for ­nuclear generation.

Household and small business bills come from energy retailers, that also buy power on the market (from multiple generators) then sell it on – averaging the costs and clipping the ticket along the way.

But if depreciated coal assets can’t compete with renewables on price over time, how would new nuclear compete in an open market? Particularly if a return on investment is required by the government? Absent a requirement for energy retailers to buy baseload power from government-owned reactors as part of their mix, they could well become white elephants.

And recent history in other sectors suggests that any such requirement will drive average power prices up, at least in the short term, if a commercial return on investment is required.

The NBN is perhaps the most apt comparison. When established, the funding model was the same as that proposed for nuclear, and the government wanted a full return on its investment from ISPs using the network to deliver broadband services to households and businesses.

Also billed as a nation-building project, public money was used deliver high-speed internet faster than waiting for commercial investment – particularly to regional areas. But it came at a total cost of about $75bn, including construction costs, accumulated losses and sundry other matters.

And it hardly did much good for household bills. So great was the outcry about internet bills, the government was eventually forced to concede it probably wouldn’t recover about $31.5bn in accumulated losses and “regulatory costs” as the only way to keep precipitous rises in wholesale internet charges under control.

Even now, consumers linked to privately owned fibre networks pay far lower prices for much faster internet access than those forced to use the NBN equivalent.

The NEM is a far more complex beast than the telecommunications market. Even when the plan’s headline costs are announced, it won’t give consumers any real idea of the likely impact on their power bills.