Was it really necessary?

Photographer: Alex Trautwig

Planning for the Blizzard That Wasn't

Justin Fox is a Bloomberg View columnist. He was the editorial director of Harvard Business Review and wrote for Time, Fortune and American Banker. He is the author of “The Myth of the Rational Market.”
Read More.
a | A

Elected officials have long gotten in trouble for not being well-prepared for snowstorms. The most famous case was probably Chicago in 1979. Two blizzards dropped more than 35 inches of snow on the city in two weeks in January. It took months to get rid of it, garbage wasn’t picked up, commuter trains couldn’t run. And at the end of February Mayor Michael Bilandic lost the Democratic primary to upstart candidate Jane Byrne.

Today in New York, most of the questions were about whether officials had overreacted to what turned out to be a run-of-the-mill snowstorm. “I would rather, if there’s a lean one way or the other, lean toward safety,” Governor Andrew Cuomo said at a news briefing this morning. He also kept bringing up Long Island, where tons of snow did fall less than 50 miles east of the city. Metropolitan Transit Authority Chairman and Chief Executive Officer Thomas Prendergast chimed in with the aphorism, “You’re only as good as your last snowstorm.”

But what exactly counts as “good”? Because only 6 inches of snow fell on New York City, the precautions taken by Cuomo and New York City Mayor Bill de Blasio -- commuters urged to stay home or go home home early yesterday, roads closed and public transit shut down after 11 p.m. last night, schools closed today -- feel excessive. That doesn’t mean they were excessive, though. Just ask (as I did a couple months ago) one of the world’s leading authorities on decision-making, Stanford University’s Ronald Howard:

You can’t tell by the outcome whether you made a good decision. It’s just a logical mistake to say, “I got the good outcome, I must have made a good decision.” And yet that’s what everybody thinks.

The way to tell whether you’ve made a good decision, according to Howard, is to follow the right process: frame the question, characterize the alternatives, assign probabilities to the different potential outcomes, and assess those outcomes in light of your preferences.

That’s all a little vague, so let’s get specific about the decision facing Cuomo and de Blasio yesterday. In oversimplified form (that of a decision tree), there were two choices:

  1. Take drastic measures -- shutting down roads etc. -- before the storm. Let’s say the cost of that (in both government outlays and lost economic activity) is $200 million.
  2. Wait and see. Cost: $0.

Now suppose there was an 80 percent chance of a big blizzard, and a 20 percent chance of a fizzle.

If you choose option 1, and you do get a big blizzard, the additional cost (in government spending and lost economic activity) is $200 million. If the storm fizzles, the additional cost is zero. Multiply these two potential outcomes by their likelihood of occurring (80 percent and 20 percent), and you get an expected cost of $160 million, to which you have to add the $200 million you’ve already spent. So the total expected cost is $360 million -- or, to put it in terms an economist or a gambler would understand, the expected value is negative $360 million.

What if you choose option 2? You didn’t spend anything up-front, but because you’re unprepared the cost of a big blizzard (in government spending, lost economic activity and probably lost lives) goes up to $1 billion. The cost if the storm fizzles is zero. Multiply those two potential outcomes by their likelihood of occurring (80 percent and 20 percent), and you get an expected value of negative $800 million.

If your preference is spending less money rather than more, then negative $360 million is a lot better than negative $800 million. By this calculation, Cuomo and de Blasio did the right thing. And while I have of course made up these probabilities and dollar amounts, they do provide a helpful framework for thinking about such questions.

There are harder-to-quantify factors such as political costs and personal regret. In February 2013 Massachusetts Governor Deval Patrick raised a lot of eyebrows by banning road travel before a big storm. Cuomo didn’t do that, and hundreds of drivers were stranded overnight on the Long Island Expressway -- something that was clearly on his mind at his news conference this morning.

This desire to minimize potential regret can lead to wasteful, overcautious decisions. The Boston “lockdown” of April 2013, when officials ordered area residents (I was one of them) to stay indoors while police chased after the missing Marathon bomber would be much harder to justify with the sort of decision process outlined above. The costs of the lockdown were high, and the risk to any individual Bostonian vanishingly small.

With big storms, though, taking precautions usually makes sense, especially as the science of meteorology advances. Nor’easters, winter cyclones that develop along the East Coast of the U.S., have long been notoriously hard to predict. Forecasters seem to be getting better at it, but they’re not perfect -- this time they were right about the severity and timing of the storm, but off in the location. Early this morning Gary Szatkowski of the National Weather Service apologized on Twitter for leading area officials astray. I don’t think he really did lead them astray, though.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

To contact the author on this story:
Justin Fox at justinfox@bloomberg.net

To contact the editor on this story:
James Greiff at jgreiff@bloomberg.net