Monday, August 24, 2009

Health Care. Insurance.

Some observations over the debate:

First, there may not be a death panel, but we are going to ration end-of-life care, and maybe even care for the simply very old. We can now extend the life of the very aged by a few years--at enormous cost. The cost of extending a life from, say eighty-five to ninety can often be achieved through the application of millions of dollars. We simply can't afford millions of dollars for every citizen in the last few years of life. Hence, most, if not all, will be denied the opportunity for such life extension.

But we can ration by government fiat or by market forces acting on the ability to pay. Government fiat is no doubt fairer. But such rationing comes at a huge price.

If we ration through the ability to pay, the cost of extending life from eighty-five to ninety may be millions of dollars, but ten years from now, the cost for extending from eighty-five to ninety may only be hundreds of thousands of dollars. Meanwhile, the cost of extending from ninety to ninety-five may now be millions of dollars--instead of being simply impossible at any price. I would call this "progress" and it's usually considered a good thing.

Such progress is not a feature of a "fair" system. In the fair system, it is simply illegal to extend life from eighty-five to ninety if the cost can't be borne by society as a whole. Therefore, there is no incentive to figure out how to lower the cost, to say nothing of figuring out how to charge millions of dollars for extending life to an even more advanced age.

Next, on the "public option": I have no idea whether a public option will make insurers more efficient, or whether it will drive them out of business. And I daresay nobody else has any idea either. The dynamics of this sort of change are so non-linear, so chaotic, that anybody who says they have a predictive model for this is lying to you, if not lying to themselves. A public option may be needed, but it is the very last thing that we should try as a cost control measure.

There are plenty of things with much more tractable dynamics that we should try before anything as drastic as a public option:
  • Guaranteed coverage. Back in the 70's, we made red-lining poor neighborhoods for property insurance illegal. Prohibiting insurers from cherry-picking customers has a huge public benefit, in addition to forcing the insurers to look for more direct ways of containing costs. Note, however, that the direct effect of requiring all insurers to accept all customers will be an increase in the average policy premium, at least for a while.
  • Tort reform. Malpractice insurance accounts directly for only about .5% of total medical expenditures. However, there is very little data on how much "defensive medicine"--the ordering of excessively cautious tests and procedures to avoid liability--drives up overall medical expenditures. I'll estimate another .5% by rectal extraction. Now, assuming that tort reform could halve the total expenditures, a half percent reduction in total health care costs ain't chicken feed.
  • Health savings accounts. I'll hold off on this one for a bit--see below.
  • Interstate competition. There are hundreds of insurers throughout the US, but many fewer insurers than that available to a resident of any given state. This is because each state regulates its own health care insurance market, and the cost of compliance is quite high. I'm usually an opponent of any form of federalization, but this seems to be a situation where the feds subsuming health care regulation makes a lot of sense. More competition is always good.
  • Elimination of tax deductibility for employer-provided premiums. This will force employees to be more conscious of what they're consuming, and it's a reasonable fairness issue. It also will raise a lot of money for health care.
  • Increased subsidies for the poor. I have no problem with the goal of universal coverage. I am also willing to pay more in taxes and premiums to achieve that goal. (Note: the Obama administration doesn't quite consider me one of the those evil rich people, but I'm pretty close.)
  • Centralized record-keeping and IT reforms. This will probably squeeze a couple of percent out of overall expenditures and has very few privacy issues--assuming that we enact portability and pre-existing condition regulations on the insurers.
  • Greater autonomy for physician assistants, nurse practitioners, and other medical technicians. Oddly enough, when you de-professionalize services that don't really require a professional, the cost drops precipitously. Hmm--I wonder why the AMA doesn't support this?
But now we come to the crux of the problem. Assuming we enacted all of the reforms listed a above, we are still left with a system of what we all call "health care insurance" but which really is performing three completely separate functions:
  1. It provides insurance for genuinely catastrophic conditions.
  2. It provides clearinghouse and payments-transfer services between the consumer and the actual health care providers.
  3. It acts as a set of bargaining collectives, representing consumers to health care providers and negotiating prices.
Of these, only the first is a function usually associated with insurance. This is an absolutely vital service and will obviously be a major component of any sort of reform. But let's take a look at these other two functions.

I've said this before, but it bears repeating: On average, all Americans consume some nominal amount of health care each year. We usually refer to this as "routine" health care: Well baby exams and vaccinations. Doctor consultation and medication for minor infections and injuries. Physical exams. Obstetrics. Stuff like that.

You'll find that the amount spent on these services varies significantly with income; poor people do without things that they shouldn't do without. But if you subsidized--and educated--the poor for this stuff, the amount of this yearly expense would be fairly constant: the standard deviation from the mean would be small.

In other words, all your insurance company is doing for you for this routine care is transferring your money from point A (you) to several different points B (your doctors and pharmacy), plus they're taking their cut. Consumption of these routine services are not insurable events, since the probability of their occurrence is so high. So the insurance company simply builds these costs directly into their premiums, takes your money, and shells it out, less a little for profit.

This is nuts. First, it's better for you to see these payments and understand what you're spending your money on. By my estimates, more than a third of all health care expenditures fall into this "routine" category. If you don't like the deal you're getting when your kid gets an ear infection, you're likely to go elsewhere the next time.

Second, there are plenty of other ways to pay for this than including it in your premium. The best way seems to be in Health Savings Accounts. Currently, HSAs are awkward, opt-in administrative nightmares. If you instituted a series of reforms, you could make the HSA the primary vehicle for all Americans' routine health care:
  • Make HSAs an opt-out system.
  • Make contributions to HSAs tax-free.
  • Give employers a big tax deduction on direct contributions to their employees' HSAs or for matching funds against their employee's contributions.
  • Centrally manage HSAs, so that everybody can have an HSA debit card to pay for medical care directly.
  • HSAs can also become a mechanism by which the poor receive sequestered subsidies for health care. It's easy for the government to dump your health care tax credit directly into your HSA or, in the event of overdraft, provide the funds as needed.
When you couple this with centralized health records initiatives and de-professionalization of routine care, you now have a powerful incentive for individuals to shop for inexpensive, high-quality care. Seems to me that this can probably shrink routine care expenditures by a solid 15%, which in turn lops 5-6% off total health care expenditures.

The second odd feature of US health insurance is its collective bargaining function. Whether you go to a "preferred provider" or are a member of an HMO, your insurance company is negotiating the prices for your health services directly with the providers, rather than putting you in the loop as well. To a large extent, this is not a bad thing. Collective bargaining is generally a good deal for the collective being represented.

But it's no job for an insurance company. First of all, there's zero consumer involvement, so there's no incentive for the consumer to look for alternative ways of receiving the service. Second, binding the negotiated deal to insurance carrier is far from optimal. Networks of patients, cooperatives, and even groups of insurers could get better economies of scale than individual insurers.

The consumer is also likely to be a more agile creator of collective bargaining arrangements than the insurer is. Consumers know what they need. If you're twenty-eight, healthy, and starting a family, you're likely to want a co-op that goes heavy on OBs and pediatricians. If you're fifty-five, you may want a co-op that emphasizes cardiologists and nutritionists. These co-ops are unlikely to appeal to an individual insurer, but both of these groups would be able to attract members from multiple different insurers and provide the economies of scale that would make them attractive sources of patients for the health care providers themselves.

Of course, there are plenty of genuinely insurable events that require catastrophic insurance. Such insurance needs to be a requirement. Such insurance would have a deductible that's high enough that the HSA payments cover the routine stuff, before the insurance kicks in to handle the nasty stuff. Whether you pay for your catastrophic insurance premiums out of your HSA, or your employer pays the premiums, or, if you're poor, the government pays the premiums to the insurer (or is the insurer) is important, but ultimately straightforward. The trick is to separate the routine care from the catastrophic care, so that you get excellent cost-containment for the easy stuff.

Finally, there's one other issue that needs to be addressed in this debate. There is an insidious, unspoken assumption that goes mostly unchallenged: It is that health care is so complex that poor, uneducated people are incapable of making decisions in their own best interest, or in the interest of their children.

Having poor, uneducated people in my immediate family, it's hard to argue that there's an issue here, at one level. But the liberal solution to this problem always turns out to be paternalistic--the government will take care of you, but in return you have to do what it says with regard to how you spend your health care dollars.

It's not that there aren't real problems to be addressed here, but most of them have simple, common-sense solutions that don't require cradle-to-grave interference by the government. Worried that somebody will spend their health care subsidy on booze and cigarettes? Make sure that HSAs can only be spent on medical care. Worried that somebody's kid winds up in the emergency room every time they get an ear infection? Require that the parents take a course on the most cost-effective treatment for a given condition before they can use their HSA funds.

Any policy that is based upon the idea that a section of the citizenry is too ignorant, helpless, and incompetent to make decisions in their own best interest is guaranteed to produce a citizenry that is too ignorant, helpless, and incompetent to make decisions in the own best interest forever. The trick to improving health care in the US is to make individuals responsible for their own well-being. Any system that doesn't start with this as its underlying premise is doomed to failure.

1 comment:

Douglas Watts said...

TRM -- thanks for the excellent comments at Cosmic Variance re: debating creationists.

Well aimed and well said.