Good Health

Near the beginning of the health care section of her diavlog with Jon Chait, Megan McArdle correctly observes that no one health care system can serve every person as well as they might be served, and then says "What you're looking for is the average or the median." Jon gives this the old "right, right" hoping to move on to more debatable concerns, but I think it is worth saying that a health care system that this is less obviously true than one might think. Simply abolishing Medicaid would, after all, have no real impact on the typical middle-class, middle-aged American and would leave room for him to pay lower taxes. It would just be, you know, wrong.

A lot of the time, I feel like the health care debate gets a little unduly technocratic. It's unlikely that there's such a thing as a unique "best" health care system in the sense of "system that works best." Value issues about allocation of resources to health care versus to other things, and about allocation of health care resources to whom both come into play.

3/23/07 11:04AM del.icio.us this incoming links

Comments

"Simply abolishing Medicaid would, after all, have no real impact on the typical middle-class, middle-aged American and would leave room for him to pay lower taxes. It would just be, you know, wrong." (Matt)

No. We need Medicaid to prevent all kinds of tangible problems, like communicable diseases spreading everywhere, or the possibility that enough could go wrong in my life that I'd need Medicaid.

Posted by: Gary Sugar on March 23, 2007 12:07 PM

Perhaps, Gary, but that's hardly why Medicaid is important. It may benefit middle class dudes like you, but we didn't implement Medicaid in order to benefit the middle class, and we wouldn't repeal it if turned out not to benefit the middle class as much as it seems to.

Posted by: DivGuy on March 23, 2007 12:10 PM

Matthew,

I couldn't agree more. In the health care field, one of the hardest quantitative measurements is 'patient satisfaction'. For example, it is widely accepted in the medical field that most bronchitis episodes are viral and the use of antibiotics is not only unhelpfu, but actually harmful - in so far as antibiotic overuse promotes drug resistant bacteria. But the patient who feels miserable from a bronchitis doesn't want to hear from a clinician, 'well, no antibiotics for you'. She wants antibiotics to help her feel better! damn, it.

From a public health perspective, it's not good to give antibiotics. But from a private practicioners point of view, not only is it quicker to just prescribe antibiotics but also profitable - as, well, patients wouldn't come back if you don't prescribe antibiotics; or pain meds, or etc.

To extrapolate this over a larger system then, it's almost impossible to measure overall 'satisfaction'. Some patients may be satisfied with clinicians who order more tests and do more procedures (which corrupts the private insurance system when clinicians 'own' the testing - for example, cardiologists owning their own echocardiograms) and other patients just like their clinicians to be simple and prescribe the fewest medications. But at no point will it be the case that we're near 'full satisfaction'. Someone will always be unhappy.

We have to figure out what level of dissatisfaction we are willing to accept - both from an individual patient perspective and from a public health perspective.

(by the way, this is also the argument against the national insurance system - as opposed to a national health system....but more on that later)

Posted by: Adi on March 23, 2007 12:17 PM

Actually, abolishing Medicaid would have a huge effect on the middle class. For starters, medical care in rural areas and for the urban poor would just disappear. Think foodhandlers and communicable disease.

Secondly, research hospitals and many independents would be stuck with huge bills, as would the relatives of disabled and elderly people, at least until society came up with some rational for simply letting people die in the streets.

Medicaid and Medicare serve as a sort of economic stabilizer, making it possible for care to be provided at rates that are meager but do keep the system going. And all of the care provided is essential care, the foundation of the health-care system, not the fripparies and vanities of the superrich.

To understand how big this is, flip the coin and start talking about how much Medicaid-Medicare costs. I can't give the figures off the top of my head, but I know they are huge. Even people who think the whole thing should be abolished will plainly tell you they think that because it's so big.

The major clue here would be that successful countries do something about healthcare for the poor. A steady revenue stream and controls on disbursement are essential parts of national economies that work.

Posted by: serial catowner on March 23, 2007 12:29 PM

Adi, I'm sure that a creative practitioner can figure out something to give the patient that isn't counterproductive and harmful to society. Prescribing antibiotics for a viral disease might have a placebo effect, but there are other placebos that don't help breed antibiotic-resistant bacteria. And if the patient is insistent, then the doctor can take the time to explain why antibiotics will do more harm than good. If the doctor has no time, because she's scheduled to see five patients per hour, then have printed handouts prepared.

The doctor could prescribe vitamins, minerals, herbs, chicken soup, and sympathize with the cold sufferer, if nothing else. If the doctor is considering prescribing a useless medication, prescribe something else that might actually improve the patient's nutrition.

Posted by: Joe Buck on March 23, 2007 12:30 PM

Unless our medical system devolves to one in which even the local public sector has no obligation to provide treatment, it is roughly a zero sum financial game.

Scuttle Medicaid and grandma has to live with the kids or is out on the street. And the grandkids go to the community clinic supported by local property taxes. And the even larger legions of uninsured end up at the city or county hospital on the local nickel.

So, kill Medicaid and local taxes go up and so do insurance rates due to uncompensated care.

Posted by: Nat on March 23, 2007 12:33 PM

Adi's comment is an illustration of how tight cost controls, as in Medicaid-Medicare, stabilize the healthcare markets.

The doctor is paid about $20 for your office visit under Medicaid. In a sense, they don't care if you don't come too often. Doctors who take Medicaid patients will always have patients to see and if you don't want to come, someone else will. The major practical problem here is that they don't get paid for spending a lot of time educating you.

But educating you isn't really their job. Most people will seek opinions from two or three other people (a friend, a relative, a pharmacist) before deciding that they should see a doctor. The burden is on society to have at least a modest ability to sustain health, and here is where the great evil of unrestrained advertising and the lack of safeguards in the workplace becomes most evident.

People generally don't want to go to the doctor or take pills, and this is illustrated by the figures on "non-compliance with medical regimes", which always show a lot of people not taking the pills their doctor ordered, or returning for a follow-up appointment.

Patient satisfaction might be hard to measure or achieve among the people who always buy 'gourmet' coffee, but most of us are a lot more satisfied with our doctors than we are with the hiring process we encounter when we look for work, and the reason for that is simple- a visibly higher level of fairness. And that's not so hard to achieve.

Posted by: serial catowner on March 23, 2007 12:49 PM

Joe,

The problem comes with the pressures of private practice. On average it would take 5-10 minutes to explain to each patient the dangers of antibiotic overuse, while it would take 15 seconds to write an rx. And as for patient handouts, these are wonderful - but patients do doctor shop often and will just find another doctor who will give them antibiotics. In the end, in the private system, a percentage of patients will only be satisfied with getting what they want - whether it be prescriptions, MRIs, etc. A few clinicians are trying other creative methods - like handing a prescription to a patient and asking them not to fill it unless their symptoms don't improve within a certain time frame, 48h, 1 week, etc.

In the end, having a private provider system (even with public spending), will lead to a percentage of overuse/abuse. Some patients will not be satisfied with their life and will be able to find some clinician who promises to do more - order more tests, write more prescriptions, do more procedures, etc. in the hope that they can marginally help (whether or not these tests would be standard care). Now these aren't necessarily bad or corrupt clinicians - so much of medicine has only empiric credibility and idiosyncratic effects. Futher, sometimes, it's ok to try to 'shoot-the-moon' if nothing else works. But overall, these inefficiencies and uncertainty lead to marked waste.

Posted by: Adi on March 23, 2007 01:09 PM

I'd like to piggy-back on "serial catowner"'s comment about the process of finding a doctor. The system as it is now is insane. It's a hassle for most patients AND for clinicians. Having to fill out various forms, turning away patients, getting prior authorization from different insurance companies, etc. adds an unnecessary transaction cost to each medical encounter. The medical encounter itself isn't usually the problem.

That said, anything to simplify the system would be well appreciated by everyone (except, I suppose, by the people who make money off the inefficiencies - insurers, pharmaceuticals, biomed firms, etc.).

Posted by: Adi on March 23, 2007 01:25 PM

Matt, you misunderstood me, probably because I was unclear. What I meant was, you want to look at the mean or the median, not the tails (except insofar as the tails drive innovation that eventually reaches everyone else.) So if we could get some sort of utilitarian sum, and then divide it by the number of people, that would be the mean. Obviously, suddenly kicking everyone off Medicare would cause that mean to drop.

Posted by: Jane Galt on March 23, 2007 01:36 PM

Jane,
Are we looking at a Bell Curve here? What are the dimensions?

Posted by: theCoach on March 23, 2007 01:52 PM

You don't need a bell curve for the distribution to have tails . . . but it might be. Not that I have any idea how to measure. It's a conceptual, not a quantitative, argument.

Posted by: Jane Galt on March 23, 2007 02:06 PM

On the communicable diseases issue, we can still get to Matt's point by simply positing a Medicaid system that treats the poor only for communicable diseases. No benefits for heart disease, diabetes, etc. That would arguably benefit the taxpayers in other economic classes. Matt says this would be wrong. Myself, given the link between so many health costs and bad health habits, I'm not sure if there's a compelling argument for taxpayers funding the treatment of these ailments.

Matt has a point about the health care debate being unduly technocratic. But that's mainly the fault of his ideological compadres, who try to bundle the supposed "efficiency" benefits of single payer with the redistributive implications of a national health care system funded by progressive income taxes.

But as I have pointed out before, we could easily separate the technocratic arguments from the "moral" ones by positing a "single-payer" or whatever system is most efficient funded in such a way so the expected benefits minus tax costs aren't progressive. In other words, if the benefits of a government-run health care system are progressive in the distribution (more benefits to the poor), then the taxes to fund it should be equally regressive (more taxes on the poor). If the system is then truly efficient from a technocratic point of view, then this should make everyone better off.

Posted by: Keith on March 23, 2007 02:35 PM

Let me also point out that, according at least to Tim Worstall, the French system is less generous than Medicaid, so the adoption of at least one major universal health care scheme would involve cutting medicaid (and medicare) benefits. Is this also just plain wrong?

Posted by: Keith on March 23, 2007 02:38 PM

On the larger point, if you want to enact national health care or any other kind of social program that costs money, you're going to need a lot more than liberal moral arguments, which might persuade most liberal moralists, but won't persuade anyone else. Do conservative moral arguments persuade you? There aren't any moral arguments that persuade me. But I think nearly all liberal policies can be adequately defended with practical arguments. For example, I want government health insurance because I'm afraid that if I get sick, Aetna will dump me, or that some of my family members might not get good enough jobs that provide decent insurance. I promise, those are better arguments than expecting me to care about people I'll never meet.

Posted by: Gary Sugar on March 23, 2007 03:22 PM

Jane,

I ask because I think the details are important to the concept.

Posted by: theCoach on March 23, 2007 03:29 PM

Financing may be the big barrier to getting anything done. Even if you stipulate that there will be significant cost savings on the administrative side, how do you fund it so that the people who already have health care coverage actually get some benefit from the cost savings, and don't end up paying more in taxes than they currently pay for their own coverage? Absent those people actually paying less, it's a hard thing to sell politically.

Posted by: henry evans on March 23, 2007 03:37 PM

henry, theoretically the health care benefit would be swapped for higher salaries. In practice that would probably take a little while. Just looking at the estimates I have seen though the cost savings for covering the uninsured is less than the cost of the additional overhead.

Posted by: theCoach on March 23, 2007 04:22 PM

Coach, if you're right that means it's possible to fund it in a manner that saves money for people who currently have coverage. Which would be great politically. But the question remains: what is the financing method -- payroll tax, income tax, excise taxes, sales tax, or something else? Voters are likely to be interested.

Posted by: henry evans on March 23, 2007 04:31 PM

Actually spelling out the tax that needs to replace the paycheck deduction/contribution would be politically difficult, I imagine.
Probably the easiest to slide buy would be something that closely mirrored what is being paid now - not unlike SS payroll taxes - probably labeled something different so it could be compared to the previous values taken out for vaious health care plans.

Alex Tabarrok expounded on a point that Tyler Cowen made, that I had not thought of -- that even if the amount spend on money is exactly the same, that amount goes from a benefit from working to a tax on working. Interesting I think. My intuition is that this would affect total work, has some sort of weird binary calculation (health benefits or not), and that incentive to work would smooth out and could be addressed in other ways, but I thought the argument was interesting, and would love to here more.

For Tyler's post that Alex comments on go here.

Posted by: theCoach on March 23, 2007 04:57 PM

I thought she was probably referring to the "utilitarian" mean -- which must take into account the fact that poor people need health-care aid a lot more desperately than nonpoor people do -- and, judging from her comment above, I was right. The question remains whether her proposal is adequate to this.

Posted by: Bruce Moomaw on March 23, 2007 05:35 PM

theCoach,

Tyler and Alex did address this issue and it seems like a point a lot of people are missing. If you shift the cost of something from the private sector to the public sector through an income tax you do incur the full deadweight loss which is non-recoverable and not "smoothed out" over time.

Anytime you increase taxes you increase this deadweight loss.

I believe, however, that there is a solution to this problem that would still get the universal health care people what they want. For example, any universal plan that forces people who can afford it to pay for their own health insurance avoids adding a distortive tax to those people's incomes. This would have to be a majority of Americans, by definition, or else that level of health care would not be affordable for the society collectively. A greatly lesser income tax on the general population would result in less distortion. There would also be a range of incomes where there would be partial subsidization of healthcare and in this range there would also be distortion to the labor supply- the more you make the less subsidization you get, hence discouraging work. But I think the overall distortion from this type of plan would be much less than the alternative.

Posted by: mpowell on March 23, 2007 05:59 PM

mpowell,

My intuition is that Alex is wrong on this (but to be fair he is an economist, and I took econ in high school, and I have not yet read his papers)-- at the very least the extent, and your explanation does not help. First, I fail to see how forcing people to pay is different than a tax. Second, the would not have to be on wealth - it could be on inheritance, say.

Posted by: theCoach on March 23, 2007 10:44 PM

theCoach, you may not read this, but I'll respond just in case you do.

The economist's explanation of this principle is correct, but does not lead to an intuition of what is going on to a non-economist. The problem with an income tax is that as you make more you pay more. This is a disincentive to an activity that is otherwise profitable to both you and your employer (in theory at least, and in the aggregate, probably true). That is where the disincentive creates a deadweight loss.

But if you are paying for your health care coverage privately, paying for health care does not discourage working more. (if anything, its the other way around) Also, if you are paying a fixed amount for health care through a government required that is not a function of your wages within a reasonable range, that payment does serve as an incentive or disincentive on how much you work.

Now an estate tax is great b/c it doesn't produce as much deadweight loss b/c it has very little impact on economic activity. I greatly favor the estate tax. And we could increase tax revenues through the estate tax and maybe the increase could cover health care (though I doubt it). But we could just as easily increase the estate tax and decrease income taxes, thus saving the deadweight cost in income taxes without involving health care, so I don't see how that would apply to this debate.

Posted by: mpowell on March 26, 2007 04:04 PM

In the 2nd paragraph, required should read requirement.

Posted by: mpowell on March 26, 2007 04:05 PM

Interesting post

Posted by: Kim on May 31, 2007 01:43 AM

Posted by: sohbet on August 12, 2007 06:01 PM

...

Posted by: porno on August 24, 2007 11:16 PM

...

Posted by: seks on August 24, 2007 11:17 PM

....

Posted by: youporno on August 24, 2007 11:17 PM