(April 21, 2006) Nuclear power plants play varying yet important roles in the power supply structures of the countries that use them, and thus in these countries' respective economic systems. In the absence of overriding strategic or military interests, the energy economy itself is what largely determines their future and it normally does so on the basis of sober economic considerations. The question of whether a nuclear power plant equals a licence to print money or rather a bottomless pit of expenditure is decided on the basis of its individual circumstances. If the reactor has been generating electricity reliably for twenty years and there is reason to believe that it will continue to do so for the same period of time again, then the former metaphor is more appropriate. At least as long as the latent potential for disaster at this plant, like that at all others, does not become a reality. On the other hand, if the nuclear power plant still has to be built, and if it will also be the prototype of a series, then it is better to steer clear of the project. Unless, of course, the financial risk can be shifted to a third party.
For investors trying to decide whether to replace or build new power stations under market conditions, nuclear plants are clearly not their first choice. This is amply demonstrated by empirical evidence. In the USA, reactor builders have not been awarded a single new contract since 1973 that was not subsequently cancelled. In Western Europe - with the exception of France - reactor builders waited a quarter of a century before receiving a contract for a new plant in 2004. Now they have one at Olkiluoto in Finland. According to the International Atomic Energy Agency (IAEA), 28 nuclear power plants with a total capacity of around 27,000 megawatts were under construction worldwide in 2005. Almost half of these projects have been plodding along for 18 to 30 years now. As far as a number of them are concerned, no one believes they will ever generate electricity - in fact, the normal term for such projects is "abandoned". The remaining plants that are expected to be completed in the near future are almost all in East Asia, and are being built under conditions that have little or nothing to do with a market economy. In short, the order situation for nuclear power plants is calamitous. All the more so when one considers the competition.
Worldwide electricity capacity has increased by around 150,000 megawatts per year since the turn of the millennium, but nuclear plants have accounted for barely two percent of this. In the USA alone, an additional capacity of 144,000 megawatts was added to the grid from 1999 to 2002 from conventional power plants using fossil fuels. From 2002 to 2005 in China, a new coal-driven power plant park with a capacity of 160,000 megawatts was constructed. Even wind energy, which is still in its infancy, managed to contribute an overall new capacity of more than 10,000 megawatts.
As marginal as the role of nuclear energy is compared to the gigantic expansion in power capacities worldwide, operators of nuclear plants are making determined efforts to extend the licences of existing reactors far longer than originally planned. The average age of all the reactors in operation in 2005 was just around 22 years but this did not prevent former Siemens CEO Heinrich von Pierer from urging chancellor candidate Angela Merkel to consider extending operating lives to 60 years during the German election campaign that same year, despite the formal agreement in Germany to phase out nuclear power plants. After all, most nuclear power advocates in Europe and North America are now calling for operating lives this long. Extensions to the licences of most of the 103 nuclear power plants in the USA have already been approved, applied for, or are expected to be applied for. Von Pierer cited "business sense" as the basis for his position; and it does in fact make sense. As long as there are no serious failures or expensive repairs, and as long as wear or corrosion do not require replacing central components such as the steam generator, electricity can be generated at virtually unparalleled low cost by old reactors of the 1000-megawatt category, which have long since depreciated. Extending plant licences also postpones the so-called "fat problem" of ending nuclear power. This means closing and dismantling the big reactors, which poses a real challenge not only to safety but also to financing. In addition, because fuel costs for nuclear plants make up a relatively low share of total costs, operators can expect substantial extra yields. If German reactors could remain in operation for 45 years instead of the 32 years stipulated by the phase-out agreement - 45 being the average operating life for large-scale fossil-fuel plants - the industry could expect handsome additional profits of around 30 billion euros. The magnitude of these figures explains why plant operators are urging discussion of licence extensions in many countries but this haggling has nothing to do with a potential renaissance of nuclear energy - rather the reverse. The fact that nuclear plant operators are calling for an "overtime" period demonstrates their unwillingness to invest in new plants for business reasons. Instead of investing in new nuclear or non-nuclear technologies, these companies are sapping the substance of their reactors without regard for their growing susceptibility to failure.
The decades of decline in the nuclear power industry have by no means come to a halt. There is a single new construction site in the USA and Western Europe combined, namely on the Baltic Sea coast of Finland. This site is treated in more detail below. At the same time, an increasing number of extensive studies in recent years have suggested that new nuclear power plants are more competitive than their fossil-fuel counterparts. The major drawback of these studies is that they convince no one except their authors and publishers - and certainly not potential funders of new plant projects. This is the main reason for the unprecedented degree of uncertainty about what exactly a new generation of nuclear power plants would cost. Hardly any reliable data is available on the large cost blocks, especially construction, waste disposal and decommissioning, or for that matter on operations and maintenance. One reason for this is because analysts greet nearly all published estimates with a high degree of scepticism. After all, these figures generally come from vendors seeking to build power plants, who therefore tend to set their estimates on the low, rather than high, side or from governments, associations and lobbyists trying to sway reluctant public opinion by holding out the incentive of supposedly low electricity costs.
Beyond the special interests, there are also objective problems. Costly "teething problems" have plagued every new reactor series and long shutdown periods, therefore potential financiers view vendors' consistently cheerful and optimistic forecasts with considerable suspicion. It is impossible to predict the "performance" of a new power plant. Even less so for new reactor types that are based on largely new and thus unproven technology. In nearly all technical fields - including those outside the power plant sector - builders can follow a "learning curve" at a relatively consistent and predictable rate to ever lower prices. Yet reactor builders are still starting from scratch half a century after the launch of commercial nuclear fission. In the 1970s and 1980s, reactor vendors offered larger and larger reactors based on the partially justified assumption that bigger plants could generate electricity more cheaply than smaller ones but this shift to an "economy of scale" has not solved the problem. A clear trend toward less expensive reactors has yet to materialise. In the meantime, the situation is exacerbated by prolonged stagnation on the market, which means that further developed nuclear power plants exist only as blueprints - or more recently as computer animated displays. This in turn increases the imponderables for potential funders. Nuclear energy has become a high-risk technology not only in terms of safety but also with respect to financing. Building a new reactor certainly means attracting risk capital, with its high attendant costs; besides construction, capital costs represent the largest block of funding for these projects. This problem, too, has worsened in major industrial countries with the deregulation of energy markets. Back during the time of large-scale, state-sponsored monopolies, investors could assume that consumers would eventually refinance their capital even if the reactor performed poorly. In today's deregulated electricity markets, however, this is no longer the case. With exorbitant initial investments and payback periods extending over decades, nuclear power is not compatible with deregulated markets. The capital costs explode - assuming financiers do not prefer other technologies that do not have these problems in the first place. Indeed in many countries that have witnessed a boom in highly efficient gas power plants over the past two decades, the construction costs per installed kilowatt hour are significantly lower, periods between contract allocation and start-up are short, and many plant components are made in factories under "controlled conditions". Moreover, due to the relatively low cost of natural gas, which accounts for a higher share of total operating expenses than uranium fuel, nuclear power plants have hardly had a chance.
A series of additional imponderables make nuclear power plants a gamble for any investor. The period from the investment decision to the start of operations is far longer than for all other power plant types. There can be enormous planning problems as well as delays in authorisation either because government agencies work especially carefully under public scrutiny, because new safety-related developments have lead to changes in authorisation criteria, or because anti-nuclear interests block progress in the courts. The decision to construct the latest British reactor, Sizewell B, was made in 1979, for example, but it only started commercial operations 16 years later. When a prototype starts up, no one can be sure that it will attain the anticipated performance levels, which of course ultimately determine revenue levels. An even more important factor is the reactor's reliability over the full course of its operating life. Unlike the capital costs, this so-called load factor can be calculated. It is generally known how long a nuclear power plant has been in operation and how long it has been shut down for repairs, fuel rod replacement, or failures. The load factor is the output (kilowatt hours) as a percentage of total possible output for uninterrupted operation. Vendors' load factor forecasts have regularly proven to be high, especially for the first reactors in a series. If a reactor achieves a load factor of only 60 as opposed to 90 percent, costs increase by one third. Extra maintenance and repair costs also accrue. Only around two percent of all reactors achieve load factors of 90 percent or more; only around one hundred of the world's reactors exceed 80 percent.
Back in the euphoric early days, operators eagerly promised that nuclear power plants would run essentially automatically and thus incur lower costs than other plants with comparable outputs but this forecast too has proven overly optimistic. It is true that fuel accounts for a relatively small share of total operating costs but this share increases if so-called mixed oxide (MOX) with an element of reprocessed plutonium is used instead of "fresh" uranium oxide. Operation and maintenance costs are higher, because personnel costs considerably exceed those for gas power plants, for example. Some nuclear power plants were even closed down in the USA in the late 1980s and early 1990s because it was more economical to build and run new gas power plants.
In contrast to other systems, nuclear power plants incur enormous costs even after decades of operation. These include disposing of radioactive waste, guarding closed reactors, and ultimately decommissioning the reactors following a more or less lengthy "cool-down" period. All these investments have to be earned over the course of plant operation as well as put aside for use at a much later period of time. These costs, including accident insurance, differ from country to country. They are all the more difficult to estimate given that normal discount trajectories do not apply to the anticipated time periods. At a discount rate of 15 percent, for example, costs incurred after 15 or more years are negligible. However, because they will burden our children in the real world, these costs represent another source of uncertainty in reactor financing and in determining the price of generating electricity by nuclear power.
The discussion that has started in some countries about resuscitating the nuclear boom of the 1970s has thus far not been reflected in reality. Little has happened aside from the debate over extending plant licences. Concrete new projects represent an absolute exception. By far the majority of plants currently under construction are based on Indian, Russian, or Chinese technology. Leading Western vendors continue to show completely empty order books. The US based company Westinghouse has received one reactor order in a quarter of a century. For Framatome ANP (66 percent owned by the French nuclear group Areva and 34 percent by German Siemens) and its predecessor companies, the Okiluoto reactor in Finland is the first contract in about 15 years. It is politicians and journalists more so than vendors who are promoting the idea of a renaissance in nuclear energy. They believe that adding nuclear power to existing energy policies will make it easier to meet short-term climate control obligations, and to avoid power shortages. This has consequences. For the more forcefully politicians and the public call for a renaissance in nuclear technology, the more baldly potential investors call for state support.
In the USA, the Bush administration is strongly in favour of extending the licences of the country's ageing reactor fleet. Following electricity shortages in major states such as California as well as spectacular power outages, it is also advocating the construction of new nuclear power stations. Discussion is being fuelled by increased concern about global warming, which in turn was triggered by the disastrous hurricanes of 2005. Thus far, it has not yet led to the construction of a new reactor, or even a construction permit. Several consortiums are trying to obtain a combined licence for building and operating new reactors but as they never tire of saying, it will not work without government support. The authorisation process alone for a new reactor series is expected to cost around US$500 million and thus far, no one knows how expensive the reactors themselves will be. To remain on the safe side, the companies are calling for subsidies of billions of dollars, which President Bush is now planning to provide. The new energy bill passed by Congress in the summer of 2005 allocates US$3.1 billion in subsidies for nuclear energy over a period of ten years. Among other risks, the government is also supposed to insure against delays. Potential investors have already called for an all-round, carefree package: they have demanded tax-free financing and subsequent sales of electricity at prices to be guaranteed by the state as conditions for investing. The state is also supposed to assume liability for serious accidents, and not least of all, solve the question of final waste disposal. Following a long delay, the now partially privatised French group EDF named the site for a pilot European Pressurized Water Reactor (EPR) in 2004; namely Flamanville in the département of Manche but the usual willingness of the French government to finance such projects has flagged. Former EDF director Francois Roussely has also stated that the reasons for building this type of reactor over the foreseeable future have less to do with generating electricity than with "maintaining European industrial expertise in this field".13 In other words, the motives for building an EPR pilot plant in France are not based on energy policy but on industrial/political objectives.
13 Francois Roussely, op.cit.
Political motives also played a substantial role in the - very controversial - decision by the Finnish Parliament to build a new reactor. The basic thrust came from the country's ever-greater appetite for electrical power over the past two decades, which has placed Finnish per capita consumption at more than twice the EU average. At the same time, politicians are worried about excessive dependence on Russian gas, and about not being able to meet the country's obligations under the Kyoto Protocol without greater reliance on nuclear energy. The contract awarded to the French/German reactor manufacturer Framatome ANP to build a pilot European Pressurized Water Reactor (EPR) on the Finnish Baltic Sea coast ultimately came from the TVO power utility - 43 percent of which is owned by the state . Since construction officially began in August of 2005, the international nuclear community has viewed the Olkiluoto 3 project as proof that nuclear energy is a good investment again, even in a deregulated electricity market. This position should be viewed with scepticism though since it is unlikely that this type of reactor would have had a chance under normal competitive conditions.
Funding was made possible by an agreement that compensated the (approximately) 60 shareholders, mainly electrical utilities, by guaranteeing that the electricity generated by the reactor would be sold at comparatively high prices. TVO and Framatome ANP also agreed on a fixed price for the finished reactor - "ready for use" - of 3.2 billion euros. This type of contract, as attractive as it is unusual for the purchaser, was made possible because Framatome ANP needed a construction permit at literally any price after more than a decade of development work on the EPR. Even before the first ground was broken, it was clear that the Areva/Siemens manufacturing consortium had made extremely tight calculations in order to boost their prototype reactor ahead of nuclear as well as fossil fuel competitors.
Reactor capacity steadily increased during the EPR development period in the 1990s. The sheer dimensions were intended to ensure profitability. With a projected capacity of 1,750 megawatts (gross) and an output of 1,600 megawatts, the EPR is by far the most powerful nuclear power plant in the world - which would considerably complicate its integration into most electricity grids. A series of additional projections that gave the reactor a competitive edge over other options, including non-nuclear ones, on paper could prove to be a hard pledge to redeem in the future. Promises included a construction period of only 57 months, a load factor of 90 percent, a degree of efficiency of 36 percent, a technical operating life of 60 years, a 15 percent lower consumption rate of uranium than for earlier reactors, and considerably lower operating and maintenance costs than at existing reactors.
Experts consider every one of these projections to be extremely optimistic. No pilot plant has ever achieved its projected construction period or promised load factor. Nor can this joint German/French venture expect to be spared construction delays, glitches in early operations, or unplanned shutdowns. Despite that, operating and maintenance costs are supposed to be lower than those of existing standard reactors, and that over a service life of 60 years. At the same time, supplementary safety facilities such as the core catcher are supposed to make the EPR safer but not more expensive than its predecessors.
It does not seem possible that all of these promises can be fulfilled at Olkiluoto. Even if all targets are met - such as the construction period - the calculated price of 3.2 billion euros is viewed as massaged. It was originally cited in the context of producing a series of about ten reactors but this is not even remotely on the cards. In other sectors there is a clear term for this type of pricing behaviour - "dumping".
If construction costs should in fact multiply, the project will quickly turn into a financial nightmare for Framatome ANP due to the fixed price agreed with the Finnish customers. A cry for help to the state will not be long in coming; this was already the case when it came to securing the financing where the Bayerische Landesbank played a significant role. The State of Bavaria owns 50 percent of this bank, and is headquartered in Munich, as is the reactor builder Siemens. The bank is a partner in an international consortium that is backing a low-interest loan for the Finnish EPR (at a reported rate of 2.6 percent) of 1.95 billion euros. The French government is supporting the Framatome ANP parent company Areva with an export loan guarantee - actually reserved for investments in politically and economically unstable countries - of 610 million euros via the export loan agency Coface. Given these concerted efforts by several countries with special interests in the project, the European Renewable Energies Federation (EREF) has filed a complaint with the EU Commission alleging violation of European rules of competition.
One thing is clear: without state support, a different decision would also have been made about the Finnish reactor. In this case, support came from both the builders' and purchaser's countries. Nuclear energy is evidently only competitive when it receives considerable subsidies or in countries where nuclear technology is more or less anchored in state doctrine, and consequently where costs play a subordinate role. Thus wherever plans are afoot to build new reactors in functioning market economies, we must expect that investors will rely on state support to insure against increased construction costs, unanticipated down times, fluctuations in fuel costs, and the difficulty of estimating shutdown, dismantling, and waste disposal costs. Ultimately, governments will have to deal with the consequences of every serious accident involving a massive release of radioactivity. No country in the world can do that alone. While insurance companies issue policies that differ from country to country based on respective anticipated total costs, the share of damages they will assume in every case is ridiculously small.
Nuclear technology thus occupies an absolutely unique position. Half a century after entering commercial markets, fuelled by subsidies in the billions, it still requires and receives state support for every new project - precisely as if it needed assistance to enter the market for the first time. Astonishingly, this extraordinary practice is also advocated and demanded by those politicians who otherwise loudly insist on "more market conditions" in the energy sector. In many industrial countries, these very same politicians produce market theory arguments to campaign against subsidising the launch of renewable energy from solar, wind, hydro, biomass and geothermal sources. But there is yet another essential difference: the future of nuclear energy is past, whereas the future of renewable energies is just beginning.