Arnold Kling  

Administrative Costs of Medicare

PRINT
Ken Rogoff on the Current Situ... Hayek in "Unbroken", Part Two...

John Goodman and Thomas Saving write,


raw numbers show that, using Medicare's own accounting, its administrative expenses per enrollee are higher than private insurance. They are lower only when expressed as a percentage - but that may be because the average medical expense for a senior is so much higher than the expense for non-seniors.

Read the whole article.


Comments and Sharing





COMMENTS (10 to date)
Daniel Kuehn writes:

So which matters?

Presumably we'd want per person for comparable people which neither measure offers. I imagine any person that is comparable in age and medical condition to a Medicare beneficiary that is paying with private insurance is going to be a very different sort of person than the one that relies only on Medicare.

Daniel Kuehn writes:

One solution might be to compare administrative costs (either per person or as a percentage - either should work fine) for people immediately above and immediately below the age cutoff - a sort of RDD approach.

Dave writes:

I have no doubt that administrative costs of private insurance are higher than they should be. But it seems likely that this is a result of all the policies that encourage third party payments (tax subsidies to employer-provided health insurance, for instance), and does not point to public insurance as the answer. Separate insurance from employment and see what happens to admin costs when private insurers really compete in the market place.

Les Cargill writes:

Percentage sounds (naively) like a better measure
than absolute cost.

John Goodman writes:

Our post was mainly a rebuttal of the ridicuolous claims by Paul Krugman and Robert Reich that Medicare is actually efficient and that our whole health care system would be moe effcient if everyone were in Medicare.

John Goodman writes:

Tom and I used a CBO table that compared spending growth rates relative to GDP. However, Linda Gorman has produced another CBO table that reports spending rates by themselves. Linda’s table makes our point more forcefully than we did.

Kyle writes:

"Percentage sounds (naively) like a better measure than absolute cost."

Why? What possible reason is there for this?

Insurance administration is almost entirely composed of fixed costs per person (though with a slope that isn't 1). It really doesn't cost much more administratively to insure sick people than healthy ones, so it goes without saying that as people age the % admin will go down, without anything actually being more efficient.

If you’re Medicare, and you aren’t looking for fraud and are bent on paying the doctors as fast as you possibly can, insuring a sick person is not more expensive than a healthy one at all. Does that make it more efficient?? If your only criteria is blowing money out the door as fast as you possibly can, then maybe, but most rational people wouldn’t consider that a worthy goal.

I've done per-member-per-month (PMPM) comparisons of admin between private companies and medicare. My finding was actually that they were about the same, give or take a bit. Why does spending $30 a month on a 35 year old super inefficient but spending the same amount on a 65 year old super efficient? It doesn't make any sense.

You know what people who think medicare is efficient should do? Add the amount of fraud medicare incurs to it's admin, since fraud is straight up waste. Guess who wins the efficiency game then?

Kyle writes:

Hey, I have an idea. Why doesn't the government helicopter drop a billion dollars on my lawn? I'll keep 1 million, burn the rest and then we can all marvel at my .1% admin. Krugman and Obama and the Economist can all marvel at my awesomeness.

Because who needs to ask "what did we get for that million", or "how much of the rest was put to good use?" Nominal dollars are for dummies, it's the percent of total that matters.

Kyle writes:

Here’s another explanation for why using % of income (or claims) can be a spectacularly bad measure of admin efficiency.

Let’s say Medicare actually decided to tackle all that fraud that goes on. To do so, it increases admin 20%. Holy cow, 20%, that’s around two billion dollars more in admin! That must mean it’s much less efficient.

But it get’s worse. As a result of the increased admin expenditures, it reduces fraud by HALF. That’s at least 12 billion dollars in savings.

So what happened is that Medicare recipients saw no (or insignificant) change in quality of care, doctors got their payments, and taxpayers saved 10 billion dollars. But when you go to calculate Medicare admin %, the numerator will be larger, and depending on your measure, the denominator might be smaller, making for significantly higher admin as a %. Then people will all wail and nash their teeth at how much less efficient Medicare is.

(As a side note, when the Affordable Care Act initially added the minimum MLR, the joke among actuaries was “well, time to fire the fraud team!” because that directly results in a smaller admin % of total).

The entire health care reform debate - from death panels to insurance across state lines to “efficient” Medicare – was face-smashingly stupid.

Kyle writes:

By the way, discussions of trend suffer from the same problem. The line for HHS rate review is a 10% increase in premium. So if that high deductible plan full of kids in their 20’s goes up by $150 a year, SOUND THE ALARM, because that’s more than 10%. But if that ancient closed block full of 60 year olds goes up by $750 a year, meh, ignore it, because that’s less than 10%.

Why do we assume all age groups should see the same % increase in medical care? The first 2-3%, that makes sense. After that, it doesn’t.

But it does make it easier for doctors and hospitals to treat the elderly like a blank check.

Comments for this entry have been closed
Return to top