Is nuclear really cheaper?

Print Friendly, PDF & Email

 The Flamanville fiasco shows once again that new nuclear power plants are not being built on time or on budget, diminishing the arguments in favour of them.

On Climate and Capitalism and elsewhere, much of the debate about nuclear power has focused on technical issues — in particular, on the issues of safety (is it unacceptably dangerous?) and relative effectiveness (can any other technology do the job?). But there is another, equally important issue: can nuclear power actually deliver the cheap power it promises, in time to help stop climate change?

Perhaps those of us who remember the nuclear industry’s promises of electricity that would be “too cheap to meter” can be excused for doubting.

For over half a century, nuclear power has been a device for transferring huge sums of money out of taxpayers’ pockets and into the hands of the private corporations that design, build, and manage the plants. It is very hard to argue that nuclear has been cost-effective, except as a mechanism for boosting energy industry profits.

What’s more, given the very long lead times required to build nuclear plants, it is hard to see how nuclear power can be widely deployed in time to deal with the world’s ever-worsening environmental crises. As Peer de Rijk of the World Information Service on Energy pointed out two years ago, even if all of the nuclear industry’s well-publicized problems disappeared overnight, and all the safety concerns were satisfied at the same time, just maintaining nuclear’s present share of world energy production would require an impossible construction schedule:

For this to happen, it would need eighty new nuclear power stations to be built and brought on-line within the next ten years. That’s a new plant every six weeks or so. And, this is assuming that all existing stations would have their lifetimes extended to forty years.  … If we want nuclear power to provide just as much electricity in 20 years time as it does at present, once the first 80 new plants have been built, we would have to build another 200 between 2018 and 2028. This is a completely unrealistic prospect. (“Nuclear Energy: Relapse, Revival or Renaissance?” in Kolya Abramsky, ed., Sparking A Worldwide Energy Revolution, p.349)

Many pro-nuclear writers will agree that existing plants were expensive and took too long to build, but they argue that today nuclear plants are safer, cheaper, and quicker to install.

But the real test is practice, and as the following article from the Guardian shows, in practice the construction of new nuclear plants is taking much longer than promised,  the costs are rising fast.



The new delays and bumper cost overruns of EDF’s new reactor in France make it very hard to believe that nuclear power can fulfil the promises its supporters make

by Damian Carrington
Guardian, July 22, 2011

Time is money, they say, and the new nuclear power plant being built by EDF at Flamanville in France is now at least four years behind time and €2.7bn over budget. EDF blamed the delay on two fatal construction accidents and dealing with safety analyses prompted by the Fukushima disaster.

So what does the news mean for the role for nuclear power in delivering the low-carbon electricity essential to tackling climate change? The three ‘Cs’ of energy policy are carbon, cost and continuity of supply, and Wednesday’s announcement by EDF is relevant to the latter two.

The Flamanville fiasco shows once again that new nuclear power plants are not being built on time or on budget, diminishing the arguments in favour of them.

The only other new nuclear plant being built in Europe is at Olkiluoto in Finland. Areva, like EDF a state-controlled French company, told me this will be connected to the grid no sooner than 2013 and costs are now estimated at €5.6bn. That is four years late and €2.6bn over budget.

I wanted to understand better the effect of the delay and ballooning costs at Flamanville 3 on the ultimate cost of the electricity produced, and Jim Watson, professor of energy policy at the university of Sussex, kindly agreed to help.

The details of his working are at the end, but the summary is that the cost per kilowatt hour has jumped between about 33% and 45% in the last few years. Watson notes that the cost is particularly sensitive to delays, as this widens the gap between the heavy capital outlay and the point at which money starts to flow back in. The UK’s committee on climate change estimates nuclear power as the cheapest low carbon source, but budget busting comparable to that at Flamanville would make it the same, or higher, than wind power.

So, the construction of nuclear is stalling and the costs are rising. Contrast that with renewable energy, where installations are surging, and most of the technologies are coming down in price as they mature and reach scale.

In the UK, the government has bet heavily on nuclear to keep the lights on at a cost we can bear, while cutting carbon emissions. The plan is to build eight new plants, with EDF’s reactors operational by 2018.

But EDF has already said “we will publish an adjusted timetable in the autumn [to] take account of the final [Fukushima] report from the UK’s Chief Nuclear Inspector and the lessons we are learning from experiences and challenges at our new build projects.” I’d bet my house the adjustment will not be to an earlier date.

As I have said before, assessing what role nuclear power should have can only be done by making judgement calls on a range of key factors. Numbers are important but only get you part of the way, as there are too many uncertainties to be able to calculate your way to a definitive answer.

So do you believe EDF when it says: “Nuclear power in the UK [is] the lowest cost large scale form of low carbon electricity generation, and it can be delivered in time to meet this country’s needs.”

If you do, you have more reason to back nuclear. But, based on track record, I don’t believe them. Taking this and other factors, such as safety, into account, I see no place for nuclear power in the UK’s future energy mix.

Professor Watson’s calculation:

“First, I have assumed the Flamanville capital costs, a lifetime (the number of years to pay off the capital) of 30 years, a high load factor (the percentage of the time the station will operate) of 85%, a discount rate of 10% (which is what the UK’s Committee on Climate Change uses) and a construction period of five years. That ends up with a rough generation cost of 8p per kWh. This is in the middle of the CCC range in their Renewable Energy Review (5p to 10p/kWh).

“Second, I have varied these assumptions. I have shortened the repayment lifetime to 20 years, and increased the discount rate to 12% – which are both closer to the criteria used in recent years by investors in the UK. Then the figure rises to 11p/kWh.

“If you assume the same conditions as in the first case above, but put in an earlier construction cost estimate of €4bn, you get a cost of electricity of between 5.5p and 6p/kWh instead. [Hence, the price has risen 33-45%]

“Under both these scenarios, a significant delay in construction such as that being experienced at Flamanville would push costs up further because of a longer wait before the plant starts earning any money.”

Note: The CCC estimates the cost of nuclear at 7p/kWh (p256), including decommissioning costs, unlike Watson’s calculations. But the CCC figure excludes “some additional costs [that] may be incurred for the first few nuclear plants in the UK.” It estimates onshore wind at 9p/kWh and offshore at 10-13p/kWh.

1 Comment

  • Leaving aside the EPR for a second, the Japanese and Koreans have brought their Gen III reactors online at budget and on schedule. The overages of the French and Finnish projects notwithstanding…the Chinese have started building their 2 EPRs (to go to 4 at somepoint).

    EPRs, among pro-nuclear activists and writers, are also understood to be the *most expensive* of any Gen III reactor.

    I think the “book is still out” on the EPR and we will have to see how the costs actually develop, and pay attention to what other countries are building in more ‘bulk’, the AP1000s and the APR-1400 from S.Korea.

    David Walters