Senator Edward Kennedy (D-MA), is proposing that Medicare, the federal government’s health-care program for people over 65 or disabled, be expanded to cover every American. In a speech at the National Press Club, he said:
An essential part of our progressive vision is an America where no citizen of any age fears the cost of health care, and no employer refuses to create new jobs or cuts back on current jobs because of the high cost of providing health insurance.
The answer is Medicare, whose 40th birthday we will celebrate in July. I propose that as a 40th birthday gift to the American people, we expand Medicare over the next decade to cover every citizen – from birth to the end of life.
It’s no secret that America is still dearly in love with Medicare. Administrative costs are low. Patients’ satisfaction is high. Unlike with many private insurers, they can still choose their doctor and their hospital.
Putting on my “health economist” hat, I have to say, sorry, but that is just not going to work.
The dirty little secret behind Medicare is that it works only because it does not cover every American. Part of the reason for this is that Medicare’s payment structure is designed to pay doctors and hospitals in such a way as to limit total spending, rather than to ensure they can break even. Clearly, they have to do better than break even to stay in business, and the people running Medicare know that. Medicare depends on the fact that there are lots of non-Medicare patients out there who (through their private insurance) can pay enough to keep the doctors and hospitals in business. This is called “cost shifting.”
Some background: Medicare determines payments for services by assigning a “relative value” to each service based on the estimated amount of “resources” the service requires. “Resources” includes the doctor’s time, the medical supplies, rent for the office, support staff, and even malpractice insurance. (This is the system for doctor’s services; hospitals are paid under a different system, but the basic idea is the same.) Every service has a set number of “relative value units,” and that number gets multiplied by a “conversion factor” to determine the payment for that service.
Every year, the Center for Medicare and Medicaid Services determines the conversion factor for that year. They do this by estimating the total number of each service that will be required that year — based on population, practice patterns, and so forth — and adding up the total number of relative value units for those services. Then — get this — they take the total budget for each type of service, and divide it by the total number of relative value units. They do this with separate budgets for physician services and for hospitals, and for certain sub-categories within those budgets. In other words, they decide in advance how much to pay doctors that year, and divide that payment by the total amount of physician services (relative value units) to get the price per unit. How much the unit is actually worth to the patient is completely irrelevant. (If this sounds like the Soviet-style approach to economic planning, that’s because it is like that!)
Furthermore, if they foul up one year (in predicting the number of units of service that will actually be billed), they make it up the next year by reducing the next year’s payments. This is something like telling your gas station that they have to reduce their price per gallon for 2005 because you drove more miles than you expected in 2004. Those whole process, in all its gory detail, is explained in this document — which also explains how in 1997 and 1999 they reduced the conversion factor. In government-speak, this is called a “negative update.” To doctors and hospitals, it is called a “pay cut.”
There is another subtle reason why this system cannot work in the absense of private medical insurance. The relative value units for each service have to be determined before Medicare can pay for the service, and they are determined on the basis of “actual” costs based on survey data. This means that somebody has to actually be providing the service before it’s covered by Medicare, and that means there have to be patients outside the Medicare system to provide it to. Currently, this is not a problem — almost everybody in the U.S. under the age of 65 is not covered by Medicare. (The only exceptions are for certain types of disabilities, and end-stage renal disease.) Medicare covers everything a year (or more) after everyone else, but at least when they finally get around to covering something, their payments have at least some basis in reality.
If everyone were in Medicare, there would be no data on “pre-Medicare” costs for new (or changed) services, so they’d have to make blind guesses as to how much to pay for things, and the fee schedule would be even more detached from reality than it is now due to the budget process. This would result in patients getting the wrong services, since doctors would be loathe to perform services on which they consistently lost money, and would have added incentive to perform services for which they made more money. If (say) they set the payment for mammograms too low, it would be tough finding anyone willing to do mammograms. That’s true now too, of course, but without the non-Medicare world providing independent cost information, they’d be much more likely to set the payment too low.
One more thing: As far as Kennedy’s claim that “administrative costs are low”: Is he still under that bridge he drove off of? I’ve talked to certain medicare providers, who make literally one penny per unit on certain items provided under Medicare. That one penny is supposed to be enough to cover all their other costs — personnel, rent, etc. That would be bad enough, except that the “administrative costs” of sending Medicare a bill and dealing with the Medicare bureaucracy to get paid is very costly, so costly they lose money on every patient, and would even if the “profit” (hah!) were substantially more than one penny per unit. When they report their costs to Medicare, it’s tough to convince the bureacrats to include the cost of submitting the bills and chasing down erroneous denials of claims in their “allowable costs.”
Why, you ask, would they continue to provide services to Medicare patients even though they lose money every time? The answer is, they depend for their business on referrals, and to keep the referrals coming, as a practical matter they have to accept every patient referred to them. They people doing the referring don’t want to have to keep track of who’s got what insurance when they make their referrals, so they will not refer anybody to providers that don’t accept a substantial proportion of patients, as the Medicare patients are. So in effect, they have to accept the Medicare patients on whom they lose money in order to get the private-insurance patients on whom they make money. (Remember what I said about cost shifting?) Now, if Kennedy’s proposal were passed into law, there would be no more non-Medicare patients, so these providers would lose money on every patient, and they would go out of business completely. And there would, therefore, be no such services provided at all.
Like I was saying: the dirty little secret behind Medicare is that it works only because it does not cover every patient.