Sticker shock – explaining rising nuclear construction costs

14 February 2012

There are a large number of reasons used to explain why the muted nuclear renaissance in the Western world appears to have stalled. Clearly the Fukushima accident has had an impact on political support and public acceptance in several countries, but in reality, nobody was ever expecting substantial new nuclear build in Germany or Switzerland. In other countries, the United Kingdom or United States for example, there is no substantial barrier to nuclear build coming from public opposition and many politicians are onside too. The real problem is essentially an economic one: the costs of building new reactors are not at a level where the undoubted risks justify going ahead, at least not in any large volume of projects.

The discovery of large quantities of shale gas in the United States and elsewhere has put a damper on previously rising gas prices, and lowered expectations of where these may be in the future. This makes gas plants formidable competitors to nuclear in competitive power markets, where future selling prices of electricity are also uncertain. It is also taking a long time for the true cost of carbon to be incorporated in power markets, something that will be of equal benefit to nuclear and renewable generation. These are, however, factors that the nuclear industry cannot do much about; they are part of the external environment that the industry faces. Capital investment costs of new reactors are, however, something that the industry can address and must do something to control if it is to be successful.

We can point to two interesting facts regarding the costs of nuclear plants. Firstly, they appear to have escalated sharply over the past 5-8 years and secondly, there seems to be a substantial gap between what plants cost to build in Europe and North America in comparison with Asia. This no doubt explains why most of the 63 reactors now under construction around the world today are located in Asia; indeed, 27 of these alone are within China.

Going back to the early years of the nuclear renaissance in the early years of this century, the capital investment costs of new reactors in the Western world were generally quoted at about US$2000 per kW installed. One possible problem with these estimates is that they were quoted without any recent experience of building reactors, so a degree of guesswork was no doubt employed. They also reflected experience of building nuclear plants in regulated power markets, where the costs of building reactors could be passed onto the electricity customers, who essentially bore the risk of cost overruns (rather than the plant owner or vendor).

Today the same costs are generally being quoted at around US$5000 per kW installed—a huge escalation—reaching a level where nuclear new build is seriously threatened on economic grounds. Combined cycle gas turbine (CCGT) plants cost about US$1000 per kW, so nuclear faces a severe initial cost disadvantage. Low and stable nuclear fuel costs will be a significant compensation over the life of the plant, but today’s new lower gas prices greatly lessen the relative impact of these.

What has caused this apparent degree of cost escalation? One point to make is that the US$2000 per kW numbers were probably illusory anyway, and based (to some extent at least) on the marketing spin of vendor companies. It is unlikely that many would have been willing to sign fixed-price contracts at that level in 2002 or 2003. One company which did sign such a contract at this time, namely Areva NP with Olkiluoto 3 in Finland, has no doubt regretted it ever since. The oft-quoted 3 billion Euros for the 1600 MWe reactor has now escalated by an additional 2-3 billion Euros and the final cost will probably turn out to be close to US$5000 per kW.

Some of the estimated cost increases are therefore down to a greater degree of realism today. But nobody plans on taking eight-years-plus to build a reactor, as has happened in Finland. If nuclear is to be economic, everyone recognises that reactors must be completed in no more than five years, as the cost burden (particularly the accumulated interest payments) becomes very heavy with extended construction times. Yet even with five-year construction schedules, the US$5000 per kW level still seems to be the norm today.

Some people have claimed that escalating raw material prices and major component costs can explain much of the apparent escalation. But while it is certainly true that major metals have risen sharply in price and there have potentially been shortages (for a time at least) of larger forgings for the latest reactor designs, these only constitute an element of the overall cost and cannot explain much of the escalation. And other less-expensive generation technologies, such as coal- and gas-fired stations, have also been subject to similar, if less significant, cost escalations.

“The delays experienced by the EPR projects in both Finland and now France have encouraged a degree of caution amongst potential investors. The move to competitive power markets has also meant that the risks of projects now have to be shared and vendors are generally expected to take on some of this.”

The most significant explanation for the cost escalation relates to recognition of the degree of risk of nuclear projects and risk allocation amongst the stakeholders in the projects. The delays experienced by the EPR projects in both Finland and now France have encouraged a degree of caution amongst potential investors. The move to competitive power markets has also meant that the risks of projects now have to be shared and vendors are generally expected to take on some of this. They will be cautious in offering low prices if those prices will be tied to fixed-price contracts. Nobody, neither plant vendor nor owner, is going to risk their company’s future on one single project.

It may even be argued that some companies are being a little over-cautious now. If they are confident that they can build reactors in 4-5 years, they may be able to build for less than US$5000 per kW installed. Some of the estimated costs for the AP1000 reactors planned for the United States appear to have peaked and (assuming the projects go ahead) may turn out to be rather lower than has been previously expected. But nobody really knows; the lack of recent experience in building reactors in the Western world worries potential investors, and they are likely to add a risk premium to money lent on nuclear projects, adding to the estimated costs.

By comparison, costs of building reactors in China and Korea are quoted at about US$2000 per kW. For the CPR1000 reactors in China, which can be built with a high fraction of local content, the level is more like US$1600 per kW; NPCIL in India also claims that it can build indigenous PHWRs reactors for this price. For the latest Generation III reactors sourced from overseas, such as the EPR and AP1000, the initial reactors in China will be substantially more expensive, but it is expected that subsequent units with much higher amounts of local content should achieve costs of US$2000 and below.

There are many reasons offered for this big difference between East and West. Some of it undoubtedly can be explained by much lower labour rates in Asia, while exchange rates may conceivably bring in some complications. But despite no longer being a low wage country, Korea can build reactors at these low costs, as indicated by KEPCO’s winning bid for four reactors in the UAE, which was won largely on price. The most important reason is the experience gained by building standardised reactors in some volume within tight time schedules, combined with financing provided by state-owned entities that can shoulder the remaining risks. The problem in the Western world is to some extent the problem of building first-of-a-kind reactors where there has been little recent experience of such nuclear projects. Merely building in volume will overcome some of the cost problems but that process will still take time. The initial units are going to be expensive and financiers are going to be wary.

There also has to be an admission that the Asians are now better than Western countries at completing large infrastructure projects. The speed and scale of building in China is astounding, whether it is highways, airports, bridges or power plants. Korean companies have been very successful at winning major construction contracts in the Middle East; there must be something to their way of working. Although the regulatory burden may be rather less in China than elsewhere, nobody is suggesting that the reactors they have completed are any less safe than those elsewhere. Asian project management skills have to be admired and (where possible) replicated.

In fact, this is the big hope for nuclear in the remainder of the world. In learning from the experience in Asia, we need to move to building standardised reactors in great volume. There are still too many reactor vendors offering too many reactor designs to be sold in low volume. Since the nuclear industry’s rapid expansion from the 1950s to the 1980s, the industry has essentially marked time, and its structure can be said to have become mummified (disregarding successes in Asia). The situation rather resembles the aircraft manufacturing sector in the 1950s, with many companies producing various designs. It is possible to overplay the comparison with aircraft manufacturing, since although large modules of a nuclear reactor can be built in a factory, the site is also crucial. Still, we have to recognise their example. Nuclear must move in the same direction, towards no more than three or four vendors with internationally-harmonised designs. Also, it needs an international supply chain lying behind manufacturing, where components can be sourced from the best locations.

Ultimately the Chinese (and maybe eventually the Indians) are likely to join the Koreans as significant exporters of nuclear plants and key components; this must surely help solve the nuclear cost problem. They are going to be busy with their own national nuclear programme for some time, but as they achieve high amounts of local content in both reactor design and the attendant supply chain, they will undoubtedly seek to become exporters. This is likely to bring forward the threat of trade protectionism from the industry elsewhere, but any such reactions will be very unhelpful for the future of the industry. Unless the Asian exporters are favoured by subsidies and preferential financing—which are outlawed by international trade rules—they should be allowed to compete, as the rest of the world undoubtedly needs them. The need is for the volume of reactor construction and the confidence that underlies this to become transplanted. Whatever gets in the way may condemn the nuclear industry to only small reactor construction programmes in most of the world. And with expected closures of older reactors (and not all of them for political reasons) the concern is that the number of nuclear reactors in operation outside Asia will struggle to remain constant, let alone expand.

As far as the West is concerned, globalisation today largely seems to mean cheap manufactured imports from Asia and the outsourcing of service jobs to there. But there have been imports of cars and high-tech electronic goods from Japan for many years, and Korea is now becoming a major force in these markets. The nuclear sector has been far too nationalistic in the past, with its promotion of ‘national champions,’ but these companies will gradually learn to work with the Chinese, Koreans and Indians. The future of Japanese companies is already bound up with their partnerships with Western companies, especially after Fukushima, and international links within the industry will gradually strengthen. How it will eventually end up, the mixture of corporate winners and losers, is hard to say at present, but the most important thing is that the industry as a whole is successful. There is a huge need for clean, secure and economic power generation capacity worldwide and it is essential that nuclear takes up its due share. It should not be held back by protectionism; nobody will gain by having a 100% share of a tiny market. The potential for nuclear worldwide suggests that everyone will do better with a small share of a much larger total.

Author Info:

Steve Kidd is deputy director general of the World Nuclear Association, where he has worked since 1995 (when it was still the Uranium Institute). Any views expressed are not necessarily those of the World Nuclear Association and/or its members.

Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.