On May 5, the Tomari nuclear plant in Hokkaido shut down for routine maintenance, leaving Japan with no operating nuclear-power plants. There is a confused debate over what to do next.

Prime Minister Yoshihiko Noda has called for restarting the reactors at Ohi plant in Fukui, insisting that bringing them online is imperative for the economy. Although the government may prevail, many Japanese remain skeptical about nuclear safety.

The public, and the local politicians who control the decisions to restart, are demanding that the government guarantee the safety of each plant. The government says new safety standards and a revamped regulatory regime will address these concerns. For example, reactors are now expected to withstand more severe earthquakes than originally assumed and backup power generators are supposed to be located on higher ground. In addition, a new commission that centralizes all the regulations for nuclear-power plants will be created.

Yet this basic approach is the same as the one in place before the Fukushima disaster last year. The idea is to avoid accidents through various mechanisms that can be effective even after an incident. Strengthening those mechanisms is important, yet the lesson of Fukushima is that this safety strategy is fundamentally misguided.

Failed Prevention

Any safety standard can only be judged relative to the maximum severity of the disaster that is assumed. When there is an incident that is substantially bigger than assumed, an accident will follow. Last year, the ineffective responses of Tokyo Electric Power Co. and the government exacerbated the unexpectedly large shock of the accident and made the aftermath worse. Still, even with a perfect response, the Fukushima quake was so huge that it was going to have disastrous consequences. Because we can never know the maximum size of disasters that can befall a nuclear plant, this prevention approach is destined to fail eventually.

The only fail-safe strategy is prohibitively expensive, so it is time for a new approach. An analogy to banking regulation is helpful here. The current conception of safety for nuclear-power plants is similar to a financial-stability policy that seeks only to prevent bank failures. Such a banking policy was pursued by the Japanese government for most of the postwar period until it failed in the late 1990s. The lack of a mechanism to close down failing banks intensified the banking crisis of 1997-1998. Similarly, during the global financial meltdown of 2007-2009, the lack of a mechanism to close down large failing financial institutions such as Lehman Brothers Holdings Inc. greatly exacerbated that crisis.

Financial regulators in the post-Lehman world have been attempting to improve resolution mechanisms for large financial institutions. In many countries, this effort has included requiring systemically important banks to prepare living wills that can help the process of winding down their operations in ways that would minimize the damage to the rest of the system. Regulators report that the very action of beginning to prepare the wills has already increased safety by leading some companies to address longstanding weaknesses.

Post-Fukushima safety standards should include a similar mechanism to deal with a large nuclear accident. For example, the owner of a plant could be required to work with the regulator to come up with a credible evacuation plan for the residents within an 80-kilometer (50-mile) radius of the installation. This would cut across prefectural boundaries, overcoming some of the poor coordination that characterized the response to the Fukushima disaster.

Evacuation Plan

The plan should aim for a complete evacuation within 24 hours; if this cannot be achieved, the barriers to its implementation should be identified and a strategy for addressing them announced. The Fukushima experience suggests many other ingredients that belong in these plans. For instance, evacuation plans should pay careful attention to assistance for patients in hospitals and the elderly in nursing homes. It should include provisions for evacuating pets and farm animals. And it should recognize that evacuees may need to stay in temporary shelters for an extended period.

U.S. regulations require each nuclear-plant owner to design an evacuation plan. But these blueprints haven’t been updated and most are out of date. Periodic reviews are a necessity. Even better, evacuation plans should be tested every few years through practice drills.

It is also important to specify who pays for an evacuation because that knowledge can influence the behavior of people who can affect the probability of accidents. Since power-plant operators have the largest influence on the probability of an accident, those companies should be required to bear this cost. This would give them additional incentive to avoid such events. The company would probably try to insure the costs, and the insurers would bring further pressure to improve safety.

In the meantime, Japan shouldn’t wait for the new safety standard to be in place before it restarts its nuclear-power plants. With just two plants operating this summer, the economy is going to suffer. The relevant tradeoff is the cost imposed on the economy by forgoing nuclear power -- which supplied 30 percent of the country’s energy before the accident -- compared with the cost of operating some plants with a less-than-perfect safety standard. The combination of the new safety standard along with some better planning for accident mitigation would make it clear that operating more than just two plants is the right decision.

(Takeo Hoshi is a professor at the School of International Relations and Pacific Studies at University of California, San Diego. Anil K Kashyap, a professor of economics and finance at the University of Chicago Booth School of Business and director of the Booth Initiative on Global Markets, is a contributor to Business Class. The opinions expressed are their own.)

Read more opinion online from Bloomberg View. Subscribe to receive a daily e-mail highlighting new View columns, editorials and op-ed articles.

Today’s highlights: The editors on how Bahrain can model Mideast reform and why Ben Bernanke should ease more; Jonathan Alter on a new Obama campaign mantra; Stephen L. Carter on the failures of capitalism; A. Gary Shilling on why the strong yen won’t last; Robert and Edward Skidelsky on markets versus the good life; Dmitri Trenin on the risk that Russia and China ally against the U.S.

To contact the writers of this article: Takeo Hoshi at thoshi@ucsd.edu; Anil K Kashyap at Anil.Kashyap@chicagobooth.edu

To contact the editor responsible for this article: Max Berley at mberley@bloomberg.net