Nuclear fantasies
How many times do economists have to show that nuclear power is too expensive?
Last week’s roundup covered the engineering and economic reasons nuclear power is not suited to Australia’s needs. To put it simply, it’s too expensive. If we had invested in nuclear power 30 or 40 years ago, it would be worthwhile keeping those reactors running, but we didn’t build any reactors then. In fact we’re fortunate in not having to live with an existing nuclear network, because nuclear power doesn’t integrate easily with our abundant renewables. Think of trying to build a new house on a site where half the block is occupied by an old structure with a heritage order.
You may have seen an argument on the ABC’s 730 where Chris Bowen and Ted O’Brien argued the costs and timelines of the government’s policy of progressing to 100 percent firmed renewable electricity for dispatchable power, and the opposition’s idea of nuclear baseload power. Because O’Brien wasn’t able to give a coherent explanation, and because the compere wasn’t able to keep the argument focussed, it is doubtful if anyone came away particularly enlightened.
The opposition’s current line of argument is that once an investment in nuclear power is made it lasts a long time, while windmills and solar panels have to be replaced. In rough figures it’s about building something with a 60-year life rather than one with a 30-year life that has to be re-built. Economists will see the fault in making such a comparison: we shouldn’t spend money on something we won’t need for many years, because there are better things we can do in the meantime: that’s the economics of opportunity costs and discounting.
Besides this issue of the time-value of investment, there are other considerations, covered in the CSIRO-AEMO GenCost Report released earlier this month, updated to take into consideration the Coalition’s queries about its earlier GenCost Report. Even when it has taken on the Coalition’s assumptions favouring nuclear, including an unrealistically short time to have plants up and running, renewables still come out more cheaply. To quote from the GenCost summary about the supposed advantages of nuclear’s longer life:
There are several reasons for the lack of an economic advantage from longer operational life. Substantial refurbishment costs are required, and without this new investment nuclear cannot achieve safe long operational life. When renewables are completely rebuilt to achieve a similar project life to nuclear, they are rebuilt at significantly lower cost due to ongoing technological improvements whereas large-scale nuclear technology costs are not improving to any significant extent owing to their maturity. Also, due to the long lead time in nuclear deployment, the limited cost reductions achieved in the second half of nuclear technology’s operational life, when the original capital investment is no longer being repaid, are not available until around 45 years from now, significantly reducing their value to consumers compared to other options which can be deployed now.
GenCost’s assertions about the falling costs of renewables are based on solid evidence of rapidly-falling costs of large-scale photovoltaics (solar panels) and batteries. It’s all well researched and documented.
For a quick summary of the comparison between our renewable pathway and the opposition’s idea, you can turn to Giles Parkinson’s post on Renew Economy: CSIRO patiently and methodically slaps down Peter Dutton’s nuclear nonsense. Parkinson presents the figures in terms of costs per MWh: even using assumptions most favourable to nuclear and least favourable to renewables, nuclear power is way more expensive than firmed renewables.[1]
To return to the time-value of investment, accountants and economists evaluating long-term projects, such as construction of a new freeway or railroad, don’t pay too much attention to benefits in the long-distant future. That’s because the discounted present value of costs or benefits in the future diminishes over time. At a discount rate of 5 percent, the present value of $100 in 30 years’ time is only $23, and at a discount rate of 7 percent, a rate often used to evaluate large projects, it is only $13.
In the debate on 730 O’Brien threw in the opposition’s assertion that the government’s plan will cost $500 billion more than the CSIRO was showing in its earlier GenCost report. That figure is based on some wild assumptions, but the main point is that it’s not discounted to take account of the time value of money. If you take cost estimates out long enough – 50 years, 100 years – you can get any figure you want if you don’t discount. It’s all junk. Unfortunately the compere refused to let Bowen answer that assertion from O’Brien.
1. Power system people quote costs in terms of $ per MWh. At home and in small businesses we think of prices in terms of cents per kWh. To convert $ per MWh to cents per kWh, drop the $, and knock off the last figure. So $245/MWh (nuclear without extension costs) comes to 24.5 cents per kWh, to which have to be added transmission, retailing and distribution costs. By comparison solar power’s cost (without firming) is about 4 cents per kWh, and is falling.. ↩
The Coalition’s plans fail on fiscal, economic, environmental and engineering grounds: otherwise they’re sound
The Coalition’s nuclear costings and other details hadn’t been posted by the time this roundup was finished, but their general direction is fairly clear. A Sydney Morning Herald article by four experienced journalists – Dutton claims taxpayers will spend billions less on nuclear than renewables – reveals some specifics. The Coalition seems to be going for seven full-scale reactors (small modular reactors are out), and they expect the first reactor to be completed by 2035 – an extraordinarily ambitious goal, and it leaves us wondering about the other six.
The same article points to opinion polls showing that while the public don’t want to rule out nuclear power, support is falling and people definitely don’t like the idea of public funding for nuclear power.
The opposition’s claim that nuclear power will cost billions less than our present renewable pathway seems to be based on work done for the Coalition by a consulting firm known as Frontier Economics. This was analysed and critiqued in the roundup of November 23. That report used figures not too different from those in the CSIRO’s estimates (see the post above), but its main flaw was that it did not bring the future costs of renewable energy to a net present value. Rather, it used non-discounted figures that did not consider the opportunity costs of funds. That produced a grossly inflated estimate of the cost of renewable energy.
Any accountant, any economist, any person familiar with the basic mathematics of financial planning and project evaluation, would rightly assert that undiscounted future costs and benefits cannot provide a basis for comparison of projects, particularly when one project involves a large upfront outlay (nuclear), and the other involves incremental outlays over many decades (renewable). Mathematically it’s the same error as failing to consider the fact that your ultimate superannuation balance depends not only on your contributions, but also on the interest and dividends of those contributions.
This and other criticisms of the Coalition’s policies – the high cost of nuclear, the need to extend the life of coal-fired stations, the obsolescence of the “base load” model, the engineering and economic incompatibility of nuclear and renewable energy – have been covered in this roundup and in the roundups of the last three weeks.
There have been other criticisms of the Coalition’s estimates.
On Radio National’s Breakfast program on Friday Tristian Edis of Green Energy Market made the basic point that private investors are not putting new money into nuclear, but they are putting plenty of money into renewables. That provides a pretty clear indication of where the markets see the economics of nuclear versus renewables. He is also critical of the CSIRO GenCost work, because it understates the cost of nuclear power plants. He warned that nuclear power plants have notoriously high cost blowouts compared with other projects. And he produced some scary figures of nuclear’s impacts on consumers’ power prices.
On ABC Drive on Thursday evening Greg Bourne, former BP Australasia executive, former energy adviser to Margaret Thatcher, and member of the Climate Council, voiced his concern about any attempt to build seven plants on seven different locations at the same time. The practical engineering problems are formidable, and could result in construction times of up to 20 years, over which time coal has to remain in the energy mix. (If they were to be built sequentially coal would have to stay in the mix until the end of this century.)
It’s hard to see how the politics of nuclear energy will play out in the pre-election period. The ABC’s Jacob Greber notes that some Coalition members of Parliament are concerned that as people learn what nuclear power could do to their energy bills, the opposition will find it harder to pursue the government on electricity prices and other cost-of-living issues: Internal Coalition fears Dutton nuclear plan will confirm no return to good old days of energy prices a decade ago. (In this regard it is amusing to hear Patricia Karvelas trying to get Barnaby Joyce make any claims that a Coalition government would bring down households’ electricity prices.)
What could work in the opposition’s favour is that few people understand the engineering complexity of electricity generation and distribution. Nor do many people understand the standard mathematics of project evaluation. Journalists will try to explain the issues, but engineering and project evaluation aren’t on the journalism curriculum. As an example of the difficulties these issues present to journalists with a generalist background, you can read Michelle Grattan’s Conversation contribution: Dutton’s nuclear policy would have coal-fired power stations operating for a lot longer. Her article is inconclusive, leaving the lay reader believing that it’s all a matter of opinion, rather than a matter of objective analysis.
Oh – and doesn’t the title of this post suggest the Coalition’s plans are sound, in spite of all the shortcomings listed above? There’s a certain cleverness in going to an election with one big project, particularly when the opposition is challenging a government that sees reform in terms of small, incremental steps. It all fits with Dutton’s political strategy, and he can count that journalists giving a “balanced” treatment of the opposition’s and government’s plans will leave the public confused. Perhaps a suitable slogan for the government could be “if you don’t know, vote no”.