Business Report – Cost overruns and long delays cool hopes for nuclear renaissance

September 9, 2007

By Alan Katz

Paris – Martin Landtman hunches forward in his shirtsleeves as a June storm on Finland’s Baltic coast drenches the construction site of the world’s most powerful nuclear reactor. As project manager for TVO, the joint venture buying the plant, Landtman has weathered far worse annoyances than rain.

Flawed welds for the reactor’s steel liner, unusable water-coolant pipes and suspect concrete in the foundation have already pushed the delivery date of the Olkiluoto-3 unit back by at least two years.

The reactor, the first nuclear plant ordered in western Europe since the 1986 Chernobyl disaster, is also more than 25 percent over its e3 billion (R29.6 billion) budget.

If this experience is any guide, the nuclear renaissance touted by the global atomic power industry as a viable alternative to coal and natural gas may not offer much progress from a generation ago, when schedule and budgetary overruns cost investors billions of dollars.

The UK’s Sizewell B plant, which took nearly 15 years from building application to completion, opened in 1995 and cost about £2.5 billion (R36.5 billion), against a 1987 estimate of £1.7 billion.

Today at Olkiluoto-3, a behemoth whose excavation site covers the equivalent of 55 soccer fields, the pressure is on the consortium building the reactor, led by France’s Areva. At stake is much more than Areva’s bottom line or the cost of electricity from Finland’s coast.

As the ownership of utilities around the world has shifted from state to private hands, the delivery of new reactors on time and on budget has become critical.

Keeping construction costs in check is a vital ingredient in nuclear power’s drive for economic parity with coal and natural gas generation. A new US atomic plant could cost between 30 percent and 50 percent more to build than a coal-fired plant of the same size, and the margin widens for natural gas, the cheapest option.

Nuclear costs balloon partly because plants must be built to more exacting safety standards and stand up to more stringent oversight.

Nuclear power proponents argue that the higher cost to build is balanced by lower fuel costs. Still, after accounting for construction, fuel, operation, maintenance and transmission costs, electricity from a new US nuclear plant in 2015 would be 15 percent more expensive over the reactor’s life than natural gas and 13 percent costlier than coal, according to 2007 estimates by the US energy information administration.

“The nuclear industry has put forward very optimistic construction cost estimates, but no experience comes even close to backing them up,” says Paul Joskow of the Massachusetts Institute of Technology.

Some investors are already treading cautiously, even amid rising demand for electricity.

“You have to go in with your eyes wide open,” says Robin Kendall, the director of project finance at Société Générale in Paris, which is among the banks that lent e1.95 billion to TVO for the Finland reactor.

Landtman started overseeing the European pressurised water reactor in Finland in 2003. The first big jolt came in October 2005 during installation of the base slab, which required 12 000m³ of concrete to be poured.

“An hour after it started, our supervisors saw that something was wrong,” says Landtman. “It was first too lumpy then it was fine. It wasn’t consistent.”

Autumn rain had soaked the crushed stone aggregate used to make the concrete. The pour had been intended for sunny May, says Timo Kallio, who heads civil works for TVO, but the sacks sat in the open while Areva completed detailed base designs and got them approved.

The delay meant the water content in the concrete exceeded levels allowed by Finnish nuclear regulators. Areva then had to test concrete already poured to make sure it met requirements. No more was poured in the nuclear section of the plant until April 2006, says Kallio.

Areva executives say delays are to be expected for such a huge project, especially as it is the first of a kind.

Areva is in talks to sell two reactors to China Guangdong Nuclear Power Group. It already has an order from Électricité de France, Europe’s largest electricity utility, and expects to have 35 of its latest reactors operating around the world in 2020.

The French government and parastatals own 93 percent of Areva. Its investment certificates, representing about 4 percent of capital, are up 28 percent this year and have nearly quadrupled over the past four years.

One Areva official points to a nagging issue: inexperienced contractors working for an industry that has been dormant in much of Europe and the US for 20 years.

Landtman learnt the same lesson. “It has taken a lot longer for industry to adapt to this business than we had anticipated,” he says.

But construction snags have not shaken the Finn’s belief in nuclear power.

“We need this single power plant in Finland to meet Kyoto requirements,” says Landtman, referring to the international treaty that limits greenhouse gas emissions.

“One plant cuts the equivalent emissions of all the transport in Finland. If someone says this is minor, then I don’t understand what they are talking about. To cope without nuclear? We don’t think we can make it in this country.”

But building the new reactor has been a slog. Landtman endured headaches throughout last year as a new forging method for the eight stainless steel pipes for the main water coolant line failed tests at a plant in Le Creusot, France, which is owned by Areva’s Sfarsteel unit.

The Finnish reactor is not the only new project to run into delays: the commercial start-up of China’s Tianwan project in June came more than two years later than planned.

In Taiwan the Lungmen reactor project has fallen five years behind schedule. Taiwan Power said the rising cost of steel and concrete had gutted subcontractors’ profits, causing them to stop work to renegotiate.

But Landtman remains optimistic. “I wouldn’t say it’s an unsolvable global problem,” he says. “The key element is good planning.”

He now says Olkiluoto-3 might be fully completed in 2011. The initial target was mid-2009.