The San Onofre Controversy: What Should We Criticize … and What Should We Praise?
On May 16, Ken Silverstein interviewed me by telephone about the controversy over whether Southern California Edison (SCE) had withheld information about problems at its San Onofre nuclear generating station (SONGS) – problems that later led to radiation leaks that required taking the power plant offline. I followed up with an email. Ken’s May 17 Forbes story and my email are both online.
On May 29, Ken sent me a link to a follow-up story he had posted on EnergyBiz, and asked for further comment. When he read the email I sent in return, he requested my okay to post it as a two-part guest column. Part one, posted on May 30, focused on what SCE should do and on whether the issue is a “crisis” or merely a “controversy.” This is part two.
Reading your posts on the SONGS controversy, I am impressed by a problem of “hindsight bias”: how to interpret warnings after they come true.
Decades ago I did a lot of work with manufacturing companies worried about toxic tort liability. In particular, they were afraid of getting sued by cancer victims arguing that emissions from a neighborhood factory had caused their cancers. One of the most depressing things I learned from my clients was that their attorneys had consistently advised them to do as little investigation as the law permitted of the carcinogenicity of their emissions.
As I wrote in a 2002 website column on “Yellow Flags: The Acid Test of Transparency”: “A chemical company, for example, wants to have nothing in its files suggesting that dimethylmeatloaf might be a carcinogen. A chain of memos in which the company’s people debate the issue and ultimately decide, no, it’s not a carcinogen (reasoning cogently even if mistakenly) is seen as ammunition for the plaintiff; far better never to have considered the possibility.”
Now, suppose two energy companies are both planning a new power plant. Company X asks itself “What might go wrong?” It aggressively seeks answers to that important question – so it ends up with dozens of letters in its files from contractors and outsiders warning of possible problems. It does its best to think through all the worst case scenarios it has solicited – so it also ends up with dozens of internal memos in its files debating the various problems it was warned about. In some cases it initiates design changes to make a problem less likely to occur or less damaging or easier to respond to. In other cases it decides after due consideration that a problem is unlikely to materialize, or that there’s nothing feasible to be done about it.
Company Y, by contrast, is less cautious. It doesn’t do as much to investigate possible problems, so it doesn’t end up with as many warning letters and internal memos in its files.
Presumably, X’s power plant will be safer than Y’s, since X made some design improvements in response to warnings that Y never even considered. But if something goes wrong at X’s power plant, there is likely to be a relevant warning in the files, leading to charges that X “ignored” the warning. Y is in better shape to defend itself after a possible problem turns into an actual problem, because Y remained determinedly ignorant that it was a possible problem in the first place.
I’m not insisting that this analogy accurately reflects the situation at SONGS – that SCE is an “X” sort of company. I don’t know enough to know whether SCE should be praised or excoriated.
Judging from your posts, and the quotations from various letters and memo in your posts, it sounds like SCE seriously debated the possibility that the steam generator design it was considering might cause vibrations that would erode the tubes holding radioactive fuel, and that the eroded tubes might leak radiation. It decided this possibility wasn’t serious enough to justify a redesign that looked like it wasn’t going to be an easy fix. It also decided that the problem wasn’t serious enough to require an explicit heads-up to the Nuclear Regulatory Commission.
Both of these decisions – not to redesign the steam generators and not to tell the NRC – deserve to be second-guessed. But the right question isn’t whether the two decisions were correct. Obviously they were incorrect. The tubes eroded and radiation leaked and SONGS got shut down. SCE would presumably have saved a lot of money if it had insisted on a redesign. And it would presumably be in better shape today if it had told the NRC, regardless of whether the NRC responded by requiring additional public hearings or not, and regardless of whether the NRC ended up okaying the original design or forcing a redesign.
In hindsight, obviously, SCE made two wrong calls.
Still open in my mind is whether the two wrong calls were sensible and legal at the time, without benefit of hindsight. I’m not saying they were; I don’t know. But surely the mere fact that a decision turned out incorrect doesn’t prove it was scandalous or incompetent or illegal.
And even if we conclude that we are right to criticize SCE for deciding not to redesign the steam generators and not to tell the NRC, let’s try not to criticize it for finding out there was a possible vibration problem and debating that possibility seriously. The existence in the company’s files of warnings and an internal debate proves only that it did decent due diligence.
Of course none of this justifies being less than candid after the vibration problem surfaced. I haven’t looked carefully at exactly what SCE said when. But if SCE executives told Congress and the media that the company was never warned about the problem, then obviously they were either mistaken or lying. That’s the evidence of the recently released letters.
They would have been far wiser if they had shown everyone the warning letters, explained how and why they decided not to redesign and not to tell the NRC, and then wished aloud that they had decided differently. If they made some honest wrong calls back in 2004 and 2005, they should have said so.
I’m fine with criticizing (and perhaps even prosecuting) SCE for misleading everyone about those warnings, if that’s what it did. But I don’t know whether it’s fair or not to criticize SCE for deciding in the first place that the warnings didn’t justify a redesign or a call to the NRC. And I’m pretty certain it’s terribly unfair – and terribly unwise – to criticize SCE for having warnings in its files about possible problems it decided not to fix.
As a society, we want companies to think about all the things that might go wrong, but we don’t want companies to feel they must fix all the things they think about. That means sometimes something will go wrong that a company thought about and didn’t fix. If that in itself becomes punishable – legally punishable or reputationally punishable – we will have fewer companies like X and more like Y … and more things will go wrong.
Peter M. Sandman is a risk communication consultant based in New York City, with more than 40 years consulting for companies, governments, and others facing controversies and crises. For more on his approach, see his risk communication website, especially the section on what he calls “outrage management.”