In 2008, for example, workers at the Oconee plant in South Carolina discovered that a crucial line in the cooling system at Reactor Unit 1 was blocked by a broken gasket. The workers fixed it and the reactor was restarted.
But the two N.R.C. inspectors assigned full time to Oconee quickly began asking why Duke Energy, the operator, wasn’t also inspecting corresponding valves and lines at the plant’s other two reactors. Duke said the clogging was isolated and a blocked line could be bypassed in a pinch.
In February 2010, when the company finally agreed to look at the other two reactors, it discovered that the lines there had the same problem and that the bypass option would never have worked.I think this is a good illustration of why the current ownership model of nuclear power plants is problematic. In this situation, the NRC identified a safety issue in one reactor that could have been generic to all three of the B&W L-loop PWRs at Oconee Nuclear Station. (It also could have been generic to all L-loop plants, of which there are seven operating. Were the other four inspected?) But the owner of the power plant successfully fought off inspection of the other two reactors for two years.
Why did Duke try to stop further inspections? Because shutdowns might have been required (and ultimately were). A reactor that is shut down doesn't make money, and making money is the sole purpose of private corporations in the US. The senior management of a profitable company gets rewarded on a yearly basis, which is a much shorter time scale from when corner-cutting might result in an accident large or small. But the impact of a large accident at a nuclear power plant is potentially very large, which means safety has to be pursued aggressively at all times. That sets up a conflict between the plant owner/operator and the regulatory agency, and in today's (legally) corrupt politics, the regulator ends up backing down.