Is it safe?

Is it safe?

We don’t know, in these coronavirus days, what it means to be “safe.” We don’t know what the question means. But we know we better come up with some sort of answer or we’re all going under the drill.

Is. It. Safe?

We are like Hoffman, bewildered by the question and offering up whatever answer might stop the pain: “Yes, it’s safe. It’s very safe. It’s so safe, you wouldn’t believe it.” Or, when that doesn’t do the trick, “No, it’s not safe. It’s very dangerous. Be careful.” We all want the pain to stop, but we don’t know how to answer the question.

We’ve been shown charts and graphs on a daily basis, representations that are being used to make policy decisions. But maybe those charts and graphs, instead of helping us make informed decisions, are actually just giving us a false sense that we know how to balance costs and benefits in the crisis. A serious analysis would be based on the kind of information we don’t have.

What kind of information do I mean? The kind insurers traditionally rely on. To get some sense of risk and the reasonable cost to insure it, one needs actuarial tables. Edward G. Robinson, in what may be the noirest of all film noirs, Double Indemnity, plays the claims manager of an insurance company. His boss has proposed stiffing Barbara Stanwyck the payout on a life insurance policy, accusing her husband of committing suicide by jumping off the back of a slow-moving train. Robinson chides the head man for not understanding his business: “Come, now. You’ve never read an actuarial table in your life, have you?” Robinson says, poking a stubby finger at the CEO. “Why, they’ve got 10 volumes on suicide alone. Suicide by race, by color, by occupation, by sex, by seasons of the year, by time of day. Suicide, how committed: by poisons, by firearms, by drowning, by leaps,” he races on. “Suicide by poison, subdivided by types of poison, such as corrosive, irritant, systemic, gaseous, narcotic, alkaloid, protein, and so forth. Suicide by leaps subdivided by leaps from high places, under the wheels of trains, under the wheels of trucks, under the feet of horses, from steamboats!” Robinson is an enthusiast, an evangelist preaching the gospel of statistics. “Of all the cases on record, there’s not one single case of suicide by leap from the rear end of a moving train. … We’re sunk, and we’ll have to pay through the nose, and you know it.”

Come to think of it, that’s the perfect epitaph for our experience with the Wuhan virus: We’re sunk, and we’ll have to pay through the nose, and you know it.

We are sunk in large part because we lack the actuarial information that Robinson’s character could reel off so effortlessly. I’ve been asking statisticians — sorry, “data scientist” is the more up-to-date title — how one might calculate the benefits and costs of social distancing, keeping the economy largely closed, and other mitigation measures. The answers have not been encouraging. This is not because the calculations lead to dire conclusions but because no one seems to have the information needed to do a credible calculation. We don’t know how the virus behaves; the models change every day. Nor do we have data detailing economic consequences compiled the last time the government sent everyone home because it’s never been tried before.

As one professor of actuarial statistics told me, “Answering the question of how safe we need to be is extremely difficult because we simply do not have the data needed to make that call.”

Another put it this way: “There are too many competing risks that are on different dimensions and impossible to assess for there to be an ‘analytical’ solution that one would take seriously,” by which I think he meant that, not only do we lack sufficient information, what data we do have are of the apples and oranges sorts.

And yet, in the face of such insurmountable ignorance, decisions must still be made. So, we are all facing the question again and again: Is it safe?