The Uncertain Business of Building New Nuclear
     A new study on Climate Progress has some surprising conclusions for nuclear power. Final costs for generating power at new plants, the study says, will be 25 to 30 cents per kilowatt-hour, which is some three times the cost of today's energy.
     Nuclear is a major part of some plans to move away from fossil fuels like coal and gas, so numbers like this should be extremely worrying to advocates of atomic power.
     One important note, though, is that the study's author, one Craig Severance, has the same problem the industry itself does — he can't be certain he's right. In fact, Severance's cost figures for nuclear are largely predicated on the uncertainties of the business.
     There's a lively debate over the validity of the report on both the Climate Progress and Center for American Progress blogs. But as the main problem seems to be that supporters and opponents of nuclear can twist the uncertainties any way they like, I thought it might be more useful to simply condense Severance's observations:

Historical costs have exceeded initial estimates, and current costs are rising
     Severance points to a Cambridge Energy Research Associates report that shows a 131 percent increase in the cost of power plants from 2000 - 2008. That's is a particular problem for nuclear, because plants can take 10 years or more to build, meaning initial estimates have to factor in future cost increases (or be wrong).
     This is a vicious circle, because the long time frame involved increases the business risks of the project, risk that in turn increases expenses. Furthermore, the length of time it takes to build plants means that utilities must pay interest on any debt they took out for construction (usually running in the the billions of dollars) for years before seeing any income.
     These might appear to be problems that can be overcome with good planning. Unfortunately, most current designs for nuclear plants haven't been proven in the field, meaning there's an element of learn-as-you-go for the people who are building. On a $10 billion project, that's not a good thing.
     Finally, the construction costs for new plants isn't progressing in a nice, easily understood curve. With China and India bulling their way into the energy markets, costs for commodities and skilled labor have shot up. The recession has depressed commodity prices, but we'll likely be back in the clear economically before any nuclear project is more than half-done.
     For any cost overruns that do occur, it will likely be the utility or ratepayers that foot the bill, as nuclear plant construction firms are wising up and adding contract provisions that mandate profits for them.

Already-lengthy construction schedules are often delayed
    Supply-chain difficulties and unforeseen problems, often of a regulatory nature, have served to drag out past projects. Severance points to the Vogtle plant near Augusta, Georgia, the construction of which dragged out for 14 years.
     Civic protests, legal or otherwise, are always a potential problem, although observation suggests, at least to me, that the fear of new nuclear is subsiding in the face of fear over global warming.
     One potential side-affect of cost-overruns and delays is outright abandonment of projects. That happened with the Washington Public Power Supply System, which defaulted on $2.25 billion in loans when it walked away from a nuclear project.

Utilities constructing nuclear plants may find their credit ratings dropping
     We all like having good credit. Utilities, especially, are often proud of their high credit ratings and rock-solid stability. But the risks above, and the prospects of a decade-long capital-intensive project, could cause rating agencies to lower rankings for utilities in the middle of a massive nuclear project, in turn causing problems for the rest of the business.
     This quote from Nuclear Engineering International, pulled by Severance, says it all: “Taking a punt on a new nuclear plant may not the first priority of a CEO in his late 50s with a distinguished career behind him.”

Even once a nuclear plant is fully built, variable costs could trip up utilities
     There are a lot of little uncertainties here, but most are fairly minor. There are two big ones: the future cost of nuclear fuel, and how to get rid of it when it's used up.
     At the moment, fuel costs are subsidized, and supplies are already tight. For a nuclear Renaissance, new supplies would need to be secured. Many of these are in the Russian sphere of influence, or in unstable republics. In short, we might have the same fights for nuclear as those we've grown accustomed to for oil.
     As to how nuclear waste should be disposed of, it's not a scientific or engineering issue; plenty of good ideas exist. But it is politically charged issue. The biggest potential containment site, Nevada's Yucca Mountain, may never be commissioned.

Higher electricity rates from nuclear may destroy demand for the energy
     If the above issues come to a head and ratepayers end up eating the costs, they may simply choose to use less power, or tap into sources like home solar and wind, which may well become cheaper than energy from the utility.