Americans pay about 14% of prescription drug costs out of pocket. Although there are exceptions in very high deductible plans, most of these payments are relatively modest compared to the incomes of middle-class people. In the New York Times, Aaron Carroll recently repeated the common criticism that such cost sharing is “often preventing people from getting necessary care.” So why do people end up with insurance that encourages them to ignore their doctors’ orders and fail to adhere to prescribed medications? Poor people are covered by Medicaid which has almost no cost sharing, so this is much more likely to be an issue for those who are not poor.

One obvious answer is that such individuals do not make the insurance choice. If they are working, they can only choose among plans their employers choose, and if they are on Medicare or buy from Obamacare exchanges, they must select among the plans selected by bureaucrats and the cost sharing that goes with them. Though the out of pocket payments in Medicare have been trending downward because of Obamacare, such charges do still exist.

Before blaming employers and bureaucrats, however, we should ask ourselves—what would we want? There is one sure tradeoff here: lowering cost sharing will raise premiums that will have to be paid by employers, taxpayers, and us. Must we rely on expanding the generosity of the first two sets of agents, or could we do it ourselves?

What is the target here? Logically it has to be drugs that are much more effective than patients think they are. If middle class patients who had a heart attack were all convinced that statins provide as high a level of health benefits as physicians think, they would be adherent even if there were cost sharing.

While there is a mountain of evidence that middle class people use more of all types of care when they have to pay less out of pocket, in this case it seems like the incentives are woefully mistargeted. Why is that?

The most obvious common-sense answer does not make obvious economic sense, at least for job-based coverage. The answer is that your boss is trying to save money with a double-barreled scheme of nastiness—one that not only lowers employer payment for a given volume of drugs but also shrinks the volume. I find it hard to link this way of thinking to the kindly Human Resources department here at Penn, and even worker-readers employed by capitalist firms know that employers do not donate benefits—the money comes out of what would have been cash wages. So, would you be willing to take a moderate hit in pay or increase in your share of the premium for your drug insurance to reduce incentives to avoid needed care? That is the question we should ask as we examine our consciences on this issue.

The evidence from Obamacare exchanges is instructive. Exchanges offer plans with different levels of cost sharing at different premiums. The most expensive plans, the “platinum” ones, offer the least cost sharing. Almost nobody picks them. Instead, the most popular metal tier is the silver one where middle class people pay on average 30 percent of the cost of care out of pocket, and the next most popular is bronze with a 40 percent share. It does not look like people are eager to pay higher premiums to cover the cost of incentives to avoid bad behavior.

I don’t know about you, but here is what I think I want from my job-based coverage (recognizing that I have spent more years thinking about these things than the average American).

1) I do not want to be confused. I know that I have different levels of copay on the drugs I am supposed to take and, after filling a prescription once, I know that the copay may vary from 97 cents to $100. Contrast that with doctor visits, where I know the copay will be $40 across the board. I do not want my drug cost sharing to get even more complicated and neither does my drug insurer, because complication raises administrative costs.

2) I want to be free to ignore my doctor’s orders and benefit from it. Some drugs come with the advice “take this if you think it helps but try discontinuing and see of you feel any worse,” while others come with no excuses. I still think I can figure out the big things myself—even though I know I do not carry out other orders like more exercise, less salt, and more frequent colonoscopies.

3) Often copays are there to drive me toward a drug that offers a better deal to my insurer, even if my doctor is not so enthused or it has somewhat worse side effects than the high-priced version. That would imply that there should always be a free option in a class of drugs that can substitute and I should go for that—when I get around to talking to my doctor about it.

So as a consumer I would like to see different options—ones with different versions of so-called value-based cost sharing since different consumers place different values on things like side effects and frequency of dose. And beyond the bland information my kindly Penn HR department provides on the different plans—tell me which one gives the better health outcomes. Then I could decide how much more I want to pay to make it more likely that I will do the right thing.

As an economist I am programmed to give bad news, and as one who has thought a great deal about medical care cost sharing, I am required to object to Carroll’s conclusion. He claims “cost sharing is supposed to lower spending without sacrificing quality.” I am not sure who he thinks supposes that, but I have always supposed that among rational people there has to be some hit to quality if cost sharing makes you decide a drug that is worth something isn’t worth its cost. Noncompliance with your doctor’s orders then makes sense, though you may want to keep quiet about your disobedient ways.

Mark V. Pauly is Bendheim Professor in the department of health care management, professor of health-care management, and professor of business economics and public policy at the Wharton School, and professor of economics in the School of Arts and Sciences at the University of Pennsylvania. He is also a member of the Inquirer’s Health Advisory Panel.