Arnold Kling  

Administrative costs in Health Care

State Formation and Tax Collec... Entering the Market...

There is a symposium on health care spending in the latest issue of the Journal of Economic Perspectives.

David M. Cutler and Dan P. Ly write,

In the United States, there are 1.5 administrative personnel per hospital bed, compared to 1.1 in Canada. Duke University Hospital, for example, has 900 hospital beds and 1,300 billing clerks. On top of this are the administrative workers in health insurance. Health insurance administration is 12 percent of premiums in the United States and less than half that in Canada

They argue that administrative costs are an important component of the high spending rate in the U.S.

Katherine Baicker and Dana Goldman write,

In 1988, 73 percent of private insurance policies purchased through employers were "conventional" fee-for-service plans (see Figure 2); by 2010, fee-for-service plans had virtually disappeared, comprising only 1 percent of employer-sponsored plans (Kaiser Family Foundation and Health Research and Education Trust, 2010). These plans have been displaced by various managed care plans, such as health maintenance organizations, preferred provider organizations, and point of service plans (which grew from 30 percent of employer-sponsored policies in 1988 to more than 85 percent today), and high-deductible health insurance plans (which over the last four years have grown from 4 percent of employer-sponsored policies to 13 percent).

It seems to me that managed care sets up a game between insurance companies and physicians, with the former trying to hold spending down and the latter trying to get as much as they can under the insurance companies' rules. I suspect this helps to explain the high number of administrators working in doctors' offices and hospitals.

Baicker and Goldman point out that patient cost-sharing has tended to decline over time, so that element of cost control has been taken out of the picture. They argue that it is not easy to bring cost-sharing back into play in a useful way, but they suggest that there are approaches worth trying.

Mark McLellan argues for payment systems to health care providers that provide incentives to improve quality and reduced cost. My fear is that the intended savings will be mitigated or offset by increased administrative expenses as providers devote more effort to gaming the new payment systems.

When you have third-party payments, you can have either simple rules or complex rules. With simple rules, the administrative costs are lower, but the provider's incentives may be poorly aligned with cost-effectiveness. With complex rules, you can better tune the incentives, but you raise administrative costs as providers try to game the system.

My guess is that in the U.S. we have gravitated toward complex rules, particularly in private health insurance. I think that those, like McLellan, who argue that more complex rules are the answer, are probably not taking into account of the subsequent moves by providers in this particular chess game.

I think that challenge is to come up with a way that puts most of the decision-making responsibility in the hands of consumers, while still providing genuine insurance. With something as personal and hands-on as health care, the notion that you can manage it from an insurance company headquarters or government office strikes me as implausible.

Comments and Sharing

COMMENTS (12 to date)
Shangwen writes:

The production possibility frontier in the Cutler paper posits Canada as maximally efficient. That seems psychedelically inaccurate to me, unless they are just using it as a baseline.

Americans receive more intensive health care and more procedures than Canadians, and US health systems have far more planning information at their disposal for both central and last-mile decision making. There are endless comparisons: wait times (US wins), drug costs (Canada wins), age of equipment (US wins), research activity (US wins). Yet the purpose of health care is not to create jobs for doctors and nurses, but to contribute to improving health. Health status in the two countries is basically the same. But Canada is a laggard on most measures too, largely as a result of overspending. Comparing the US to Canada is like comparing a rotten apple to a badly bruised one. The better high-level outcomes are in Japan and northern Europe.

Marcus writes:

A couple of thoughts:

1) Perhaps Americans have more outpatient services. In which case, counting administrators to hospital beds may not be a good measure of efficiency.

2) I thought managed care was an attempt by insurance companies to control costs. Didn't physicians "game" the system under fee-for-service plans too?

Steve writes:

Arnold, check out this Times article for a picture of medical practice overhead. It is tellingly titled "Family Physician Can’t Give Away Solo Practice." Relevant excerpt "Altogether, Dr. Sroka employs 10 part-time employees, or the equivalent of five full-time workers. He does not provide his staff members with health insurance. His expenses amounted to $420,000 last year, or about $200 an hour."

Sonic Charmer writes:

I think that challenge is to come up with a way that puts most of the decision-making responsibility in the hands of consumers, while still providing genuine insurance.

I think I know the ideal product that would have such properties. It's called "insurance".

I don't know exactly what to call the current complicated-health-payment-plan system most of us are currently on, but I do know that it's not insurance. Insurance would not have any role nor require any filling out of forms in paying for/defraying normal doctor checkups, for example.

Wallace Forman writes:

If managed care was such a bad deal why did everybody switch to it?

Dave writes:

Most people who point to administrative costs as a reason to move toward socialized medicine are unable to untangle the effects of the rules and regulations in the current US system that decrease competition among care providers.

Sonic Charmer writes:

If managed care was such a bad deal why did everybody switch to it?

Most people are essentially forced to. The government penalizes (via the tax code) employers if they pay their employees their entire salary in money, rather than paying part of it in Health Plan.

Wallace Forman writes:


Fee for service and managed care are insurance schemes. So the fact that employers all bought insurance doesn't explain why they chose to do so from managed care companies. If FFS was cheaper, why did they sign up for the red tape in managed care?

Sonic Charmer writes:

Don't recall a claim being on the table that the sort of (not-insurance) Health Plan you're calling "fee for service" is somehow generically/uniformly 'cheaper' than the sort of (not-insurance) Health Plan you're calling "managed care". I must have misunderstood; I guess I don't know who/what you're arguing with in the first place,

Peter writes:

In real terms, cost sharing per capita has increased, but as a percentage of health spending it has decreased. I'm curious, which measure do you think is more important and why?

Wallace Forman writes:


That is exactly what I find confusing about this post. If fee for service insurance reimbursement schemes are *not* cheaper than managed care, then how can the transition from FFS to managed care have contributed to high cost in health care? If FFS was more expensive, then managed care would have not contributed to, but mitigated against, the increasing cost of health care.

And if managed care is more expensive, as I assert the original post assumes (by the logic that I've just explained), then why did insurers and employers switch to it?

I don't have answers to these questions, but can someone explain why they aren't the right ones to ask?

Various writes:

I think the solution is not complex, but politically challenging. You have true insurance for catastrophic problems, such as high cost procedures. Everything else, patients pay directly and it is up to individual patients to self ration. Use the auto insurance industry as a model. Auto insurance does not pay for routine care and maintenance of your car, but does kick in if you're in an accident.

Separately, transfer payment type plans (such as Medicare, Medicaid, etc.) can be maitained, but would need to be modified. Instead of guaranteeing service, guarantee payment instead. Give the money (or voucher) to the covered patients, who would be empowered to select the plan of their choice. There would need to be some regulatory oversight to make sure the covered population doesn't choose foolishly, but the regulations would be a fraction of what exists currently.

Comments for this entry have been closed
Return to top