Administrative Costs

I take a back seat to no one in complaining about the inordinate number of bureaucrats in health care. Where I part company with many of my colleagues is that I think there are too many bureaucrats in every field. Most health policy wonks have the opposite view. They are not bothered by bureaucracy in anywhere other than health care. For example, I almost never meet anyone who complains about high administrative costs in health care who makes the same complaint about education, where the problem is much more severe.

I take a back seat to no one in complaining about the inordinate number of bureaucrats in health care. Where I part company with many of my colleagues is that I think there are too many bureaucrats in every field. Most health policy wonks have the opposite view. They are not bothered by bureaucracy in anywhere other than health care. For example, I almost never meet anyone who complains about high administrative costs in health care who makes the same complaint about education, where the problem is much more severe.

In any event, there is no problem that cannot be made worse when legislators convene. And when Congress passed the Affordable Care Act (ObamaCare) it definitely made things worse than they otherwise would have been.

A medical loss ratio (MLR) is the percent of health insurance premiums that is spent on medical care. The popular term for the remainder is “administrative costs.” The new law mandates a minimum MLR of 80% for individual and small group insurance. That means these plans must spend at least 80% of their premium income on “medical care.” Conversely, they can spend no more than 20% on “administration.” If they fall below the MLR minimum, they must give back the amount of the deficit to enrollees in the form of premium rebates. The minimum MLR is 85% for large groups.

Here’s one immediate problem: no one knows how to define “administration.” Just as there is no line item in the federal budget called “waste fraud and abuse,” there is also no line item in any organization’s budget called “administrative costs.” Most of us think we vaguely know what it is. But pinning it down is sort of like trying to nail Jell-O to the wall.

I first became aware of this problem as head of a nonprofit organization. Since many donors don’t like to think their money is paying for fundraising and overhead, most nonprofits try to make those expenses as small as possible on their financial statements. That’s when I discovered there is a lot of discretion in deciding what to call “overhead.”

Where you really find a lot of abuse in this area is in direct mail. If you send someone a dollar in response to a mail solicitation, odds are that 90 cents will pay for stationary, envelopes and postage (i.e., the cost of the mailing) and 10 cents will pay the salaries of the people who sent it for you. Almost none of your dollar will go for the purpose for which you gave it. For accounting purposes, the organization will probably call 95% of the letter making their pitch “marketing,” and allocate to “fundraising” only the one or two lines where they actually ask you for money. Even though the whole exercise is making more than a fundraising operation, the organization is able to claim that 80% to 90% goes to “programs.”

Back to health care. Think about a doctor’s office and ask yourself what goes on there that you would be inclined to call “administration” or “overhead” and what you would call “medical care.” Arguably, the physical facility, the equipment, and the utilities are all overhead. The personnel who admit you and discharge you are engaged in administration, are they not? Ditto for the taking of your medical history and your vitals — and even ascertaining the nature of the complaint that brought you there. In fact, you could make a case that unless someone is actually drawing blood, giving you a shot or ordering a prescription, it’s all overhead.

Looking at this way, you could argue that about 95% of everything that goes on in a doctor’s office is administration and overhead! Conversely, a clever accountant might also be able to argue the reverse — that only 5% is really overhead.

Now if at this point anyone is wondering why we are even having this discussion, I want to second that thought. There isn’t a good reason. It really doesn’t matter how you classify these things. But when governments pass laws that create perverse incentives, these things do matter.

Okay, let’s think about insurance companies for a moment. The MLR, remember, is a ratio. Medical care spending is in the numerator and premium income is in the denominator and there are hundreds of decisions that have to be made in deciding how to measure each of these. The other thing to remember is that the minimum MLR is a constraint that potentially limits sales efforts, profit, efforts to ferret out fraud and any other non-medical expense. So insurers will view the MLR as an undesirable constraint. For that matter, sales commissions to insurance brokers who service health policies are considered overhead, so insurers can hardly afford to compensate brokers’ services. Neither can insurers afford to perform the tasks of answering enrollees’ questions since the cost of doing so is also added to administrative costs.

Of all the perverse incentives this kind of regulation creates, I want to focus on five that I think will be especially pernicious.

First, just about any payment an insurer makes to a doctor or hospital is going to count as medical care, no matter how large the administrative costs are for the provider. So both entities have an incentive to find ways of shifting administration costs from the insurer to the provider. The most obvious way of doing that is for the insurer to contract with an HMO, give the HMO a fixed fee per enrollee and let the HMO manage all the care, no questions asked. If you thought managed care in the 1990s was abusive, you haven’t seen anything yet.

Second, almost anything the insurer does that conforms to the Obama administration’s view of “improving quality” — managed care, coordinated care, integrated care, electronic medical records, etc. — will count as “medical care.” But efforts to detect and prevent fraud will not count. Doctor “credentialing” won’t count either. One way to think about this is to realize that everything insurers are doing today to hassle good doctors will be encouraged and everything they do to get rid of bad doctors will be discouraged.

Third, the days of the insurance broker are probably numbered. Since broker commissions won’t count as “medical care” and since the MLR is averaging about 70%, rather than 80%, in the individual market, insurers will cut back on commissions and brokers may leave the market completely.

Why should you care? Because in today’s bureaucratic health insurance system, employers and brokers act as advisors and protectors. They answer questions, correct mistakes, eliminate confusion, etc. If employers begin to drop insurance in large numbers and agents vanish, millions of people will be at the mercy of impersonal insurance bureaucracies.

Fourth, the minimum MLR restricts what insurers can spend to complete with each other in a way that favors larger firms and discriminates against smaller ones. We have previously noted an exodus of insurance companies from the market that is already occurring, resulting in the inevitable slide toward monopoly. The latest episode comes from Michigan, where the state’s congressmen are asking for a waiver on the grounds that only 2 of 7 companies serving the individual market can satisfy the MLR threshold.

Finally, insurers will be constrained in their ability to realize profits on their insurance business, but there will be no constraint on their profits from invested reserves. This means that in the business of insurance, the insurer will be like a regulated utility. There will be no incentive to take risks on developing new products, because the insurer will not be able to reap the rewards of successful innovation. On the other hand, the insurer can realize the full return form risky investments outside the business of insurance.

I’ll write more about this last problem in a future Alert.