Richard Zeckhauser is Harvard through and through. He is Frank P. Ramsey Professor of Political Economy at the Kennedy School of Government at Harvard. He got a Harvard A.B. (what everybody else calls a B.A.) in 1962, summa cum laude no less. He got his PhD. in economics there in 1968. And he has been teaching at Harvard ever since. His time served in Cambridge now spans 60 years.

So when he tells you getting health care can be an ordeal, as he does in a new paper published by the National Bureau of Economic Research, it’s not just some libertarian hack saying so. It’s one of Harvard’s longtime leading lights.

Of course, he doesn’t use “ordeal” lightly or colloquially. He uses it in its strict economic sense (which I’m afraid to say I didn’t actually know it had until I read his paper). As he says: “In economic terminology, ordeals are burdens placed on individuals that yield no direct benefits to others. An ordeal causes what economists call a dead-weight loss.” (“Deadweight” because there’s no “counterweight” benefit to offset it: it’s pure loss. One of the most confounding aspects of economics for outsiders is our tendency to give very particular meanings to ordinary English words.)

Doing time in a waiting room is pure loss, unless you actually enjoy forced meditation or reading old magazines. Doing time on a health-care waiting list is also pure loss: no one else gains from your waiting. In the United States, the stigma associated with using food stamps is pure loss: no one benefits from it—unless maybe you count non-users whose “utility function” puts positive weight on feeling superior. Having to fill out long forms to apply for various types of health care may also feel like—and may also be—pure waste.

The title of Zeckhauser’s paper—“Strategic sorting: The role of ordeals in health care”—suggests that despite this assumption that such ordeals are pure waste they may actually produce systemic benefit. How so? People’s willingness to undergo ordeals in order to get care tells you something about how much they value the care. That information is useful because most health care systems, even in the U.S., don’t rely very much on the price system to generate such information.

With ordinary goods, if people value the good, they pay for it. If they don’t, they take a pass. So for goods people buy with their own funds we can reasonably infer that their benefit from them is at least what they paid for them. But health care is usually delivered at a highly-subsidized price. Yes, Americans with private insurance often have to come up with a “co-pay” but that co-pay will usually be a small percentage of the actual cost of providing the care. There are also tax subsidies for middle-class people and direct subsidies for poor people and old people. As Zeckhauser puts it, “a cascade of subsidies spills into most significant health care purchases.”

All these subsidies likely mean overuse. No, people don’t have gall bladder surgery just because it’s subsidized. But in lots of other cases, when they’re on the margin of going to the doctor’s office or the emergency ward, and when doctors are on the margin of ordering tests and procedures or not, the subsidies will tip them over it. But governments and health-care plans can’t afford to provide service up to the point where the marginal benefit to patients is equal to the subsidized price of nearly zero. Instead, to make their costs manageable, they ration care, often by imposing waiting times on non-urgent services, a concept Canadians are only too familiar with. And Zeckhauser is familiar with our familiarity: One of his sources is the Fraser Institute’s own now almost legendary, and deservedly so, “Waiting your turn: Wait times for health care in Canada” (whose website naturally pictures people sitting in a waiting room).

People’s willingness to submit to ordeals is only imperfectly related to the benefit they expect to get from the care they’re trying to get access to. Sitting in the waiting room is less costly for retirees, for instance, than for $500-an-hour lawyers. But the correlation probably isn’t zero. One example Zeckhauser cites concerns specialized back surgery that works only if patients faithfully undertake a demanding exercise regime in advance. If people do make it through all the exercise, it’s reasonable to infer their motivation is so great because their pain is great. In the same way, if resources for difficult or expensive surgery are limited, forcing doctors to go through lengthy paperwork in order to get their patients access to it is a way to ensure they only recommend it for those patients who need it most.

Make no mistake. This is all “second-best” economics. “First-best” might well be to let the price system determine need, after giving poor people the resources to participate in it. But that’s an argument for another paper, Zeckhauser says. If we don’t go first-best and do persist in rationing care, it may be that some of these non-price mechanisms—ordeals—will help make sure that what care is available goes to the people who need it most. That makes the system more efficient (in the economists’ special meaning of “efficiency,” namely, that the greatest needs are served first).

Zeckhauser does not pretend, however, that the various ordeals different countries’ national health care systems currently impose on people are optimal. “Too often,” he writes, “ordeals are treated as natural phenomena, with little thought that they might be replaced or improved. Those who foster an ordeal, say the manager of a health plan or head of a hospital, may track personnel costs, insurance revenues and patient revenues to the dollar, but never seek even a crude assessment of the costs and benefits deriving from an ordeal.” But precisely because the application of ordeals has been mainly thoughtless in the past, he believes there may be big gains from (n-effect) smarter ordeals in future.

“Ordeal specialist” sounds a little medieval as a career pursuit. But it seems we need such people.