An Interesting Interview and Some Economic Nonsense

by on November 28, 2008 · 8 comments

Via Jeff Jonas, who oh-so-carefully assessed the treatment he received in Stephen Baker’s book The Numerati, I came across this NPR interview with Baker.

In the latter part of the interview, Baker discusses pretty accurately Jonas’ dissent from the passion for predictive data mining in the national security world. That dissent was given expression in the paper Jeff and I wrote, “Effective Counterterrorism and the Limited Role of Predictive Data Mining.”

The data intelligentsia are an interesting subject for a book, of course – it looks The Numerati may have a lot of similarities to Robert O’Harrow’s No Place to Hide – and the NPR interview is interesting. But what makes it notable is Baker’s economic literacy. Or, more accurately, his lack of economic literacy.

Now, I’m not an economist either, so I’ll stand for correction in the comments (actual economists preferred, not just people with strong opinions, please).

At about minute 5:00, for example, Baker says that retailers “detest” a group of shoppers called “barnacles.” These are the hardcore bargain hunters who don’t buy any luxury items or indulge in impulse buying. Baker says that retailers might discourage barnacles from coming into their stores “by giving them coupons or advertisements for things that the barnacles would never want, things that are incredibly expensive.” The conversation turns to how this discouragement is wrongfully discriminatory against poor people and how it denies them savings.

I think Baker’s premise about false advertising to non-customers is pure economic fantasy. It would be commercial suicide to advertise inaccurately toward non-customers, trying to keep those customers away. I see no case where it would make sense for a retailer to do that. More likely, retailers will treat “barnacles” as future potential customers or, at worst, as subjects of indifference. They may not pursue them very hard because barnacles are not their customer demographic, but they will always (correctly) regard a barnacle as a future potential customer. Retailers want everyone to want what they sell. They will never discourage anyone from being interested in their wares. Baker’s ideas here are just silly, as are the conclusions he draws from them.

Baker’s ideas about health insurance are a closer call, but I still think he’s wrong. A little after 7:30, he describes health insurance as a subsidy from the healthy to the less healthy. As everybody learns more details (about health or the subsidy – it’s not clear), he says, insurers will be able to discriminate against us and we will be able to demand special treatment based on what we know about ourselves. This “breaks the social contract we’ve had through the ages.”

Now, the social contract is our agreement to give up the right to use force and give it to the government because of the better security we get from that arrangement. It doesn’t have anything to do with health insurance.

But on the economics, Baker is confused about what health insurance is. Health insurance is a financial service in which people pay a small amount of money regularly over time in exchange for the right to access a large amount of money should certain contingencies (disease or injury) come to pass. Health insurance works as a business because these contingencies occur rarely. Few enough people exercise their rights under insurance contracts that insurers can profit. But every insured person gets the benefit of the right to access needed money in the event of health problems.

Now, as better health information moves many health events from the “risk” category to “certainty” or “impossibility,” those events will become uninsurable. For example, if it’s a certainty that I’m going to get skin cancer, I can’t insure against it because no insurer would accept a small amount of money knowing it will have to pay me a large amount. I just have to save up for it (while medical science comes up with a cure). Likewise, if it’s a certainty that I’m not going to get skin cancer, I’d be a fool to buy insurance for it. Uninsurable.

It’s too bad when we learn that some illness is more likely or a certainty, and many people would like insurance to continue paying for medical conditions like this, even though they are uninsurable. But that’s not insurance; that’s just health care subsidy. Baker appears to think that health insurance is only the subsidy part, and that when conditions become uninsurable that’s insurers discriminating against people. He’s mistaken.

On the issues with data collection and use, I’m sure it’s a better book. Baker’s treatment of Jeff Jonas gets Jeff’s approval, so that’s one in its favor.

Previous post:

Next post: