In their brilliant book Smart Choices, Hammond Keeney & Raiffa warn against falling into the “anchoring trap”—a psychological quirk which makes human decision-makers glom onto the first number they see and base subsequent decisions on that number. If the number is wrong, or out of context, those subsequent decisions could be disastrous. On Tuesday of this week, the Toronto Star reported that the only “compliant” bid in the Ontario nuclear reactor competition comes with a cost of $26 billion. That’s a lot of money regardless of how you slice it. But some context might be useful. Otherwise the province is about to tie its electric future to natural gas.
The problem is, the government seems to have provided no context when it released the $26 billion figure to the Star. The Star rightly pointed out that the originally touted figure for the cost of the Darlington expansion was $7 billion. Did the government not provide any reason the figure went up to $26 billion? The Star reported only that the energy minister’s spokeswoman said the process was “complex and that no single number tells the full story.”
Nobody expects the usual anti-nuclear suspects to point this out, and true to form none of them have. But the provincial government’s failure to provide context has now anchored that $26 billion into the public mind. First the reactors were going to cost $7 billion. Now it’s 26 billion. In the facile rhetoric of the anti-nuclear crowd: where does it end?
Never mind that if you apply the cost over the 50–60 years the reactors will run, this project is actually a serious money-maker. Assuming a 95 percent capacity factor, the two ACR 1000s will each generate nearly 10 billion kilowatt-hours of carbon-free electricity every year. At 5.1 cents per kWh (the average Hourly Ontario Electricity Price since the IESO started reporting it on May 1, 2002), the two units together would generate over a billion dollars worth of electricity every year. Over a 50-year lifespan, that’s over $50 billion—more than enough to pay back their cost plus interest.
Did the only bid judged compliant by the ministry, from federally owned AECL, show more convincingly than the competing bids that the lifetime cost was less than the expected lifetime revenue? It may have: the only other reported figure, related to Areva’s bid, was lower—$23.6 billion versus AECL’s reported $26 billion. Areva’s EPR has 450 more MW of capacity. So either AECL’s lifetime cost of power was much lower than Areva’s or the figures are just out of whack. In a conference call Friday, an Areva executive said it was the latter.
Still, the $26 billion anchor puts a tricky and perhaps show-stopping twist into the federal-provincial negotiations over the extent of federal support for the Ontario project. How can the feds support something this expensive?
Well, since the alternative is gas-fired generation, one argument that could get legs is the development in the U.S., where the Waxman-Markey climate bill represents the most serious attempt yet to introduce carbon costs to U.S. power generation. Ontario trades electricity with U.S. states. With nearly 20 billion kWh of energy coming from the putative Darlington CANDUs, Ontario power exports to the U.S. would be free of any carbon penalty. The same cannot be said of gas-fired power.
If the feds and Ontario can’t come to a deal on nuclear financing at Darlington, the province could well just throw up its hands and say it has enough power for now. And, if you go by the demands on the grid so far in 2009, that claim would be correct. But it’s been a cool summer. Gas prices are extremely low, both because of the relaxed demand for gas-fired peak power for air conditioning and because of the recession. We have really had only one taste of hot weather back in late June; power generation in Ontario went to just over 22,000 megawatts on June 25.
But nobody remembers that day. It has been a cool summer. Which leads to another instance of the Anchoring Trap: everybody thinks our summers will be cool and easy from now on. Somehow, climate change is no longer real.