Editore"s Note
Tilting at Windmills

Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

May 12, 2008
By: Kevin Drum

GENETIC DISCRIMINATION....Congress voted nearly unanimously last week to outlaw genetic discrimination by health insurance companies. Michael Kinsley comments:

The very appealing notion that genetic discrimination is unfair looks especially odd in the context of insurance. The idea of insurance is to protect against the unexpected or unlikely. Forbidding insurers to take predictable risks into account when choosing whom to insure and how much to charge is asking them to behave irrationally and make bets they are sure to lose. Not insuring people who are likely to get cancer, or charging them more, isn't evil. It's rational behavior. Of course, we outlaw a lot of behavior that would be rational if it weren't against the law. But the skeptics who say this is a step on the way to universal health care actually understate the case.

This paragraph is preceded by some silliness about Yo-Yo Ma and followed by some further silliness about a possible descent into Stalinism. But that stuff aside, Kinsley is right about this. Insurance companies have to be allowed to assess risks when they set premiums. If they don't, then they aren't insurance companies.

So what's the eventual result of forbidding healthcare insurers to rationally assess risks when they write policies? They go out of business. Slowly, to be sure, but eventually they go kaput.

So think of this as a revealed preference. Conservatives all claim to believe that the private market is the best way to provide health insurance. And yet, given a close look at exactly what that means, they voted to outlaw the very thing that makes private insurance work: rational discrimination. The reality was just too ugly to support. If a conservative is a liberal who's been mugged, I guess that means a liberal is a conservative who's been denied insurance because of a congenital condition.

Kevin Drum 12:39 PM Permalink | Trackbacks | Comments (64)

Bookmark and Share
 
Comments

Seems to me as if this issue might be rather more complicated. Should insurance companies, e.g., be able to deny women coverage for pregnancy-related problems, or people of African descent for sickle-cell anemia? One might respond that they'd be out-competed by companies that didn't do so...but that's not at all clear.

Posted by: Winston Smith on May 12, 2008 at 12:48 PM | PERMALINK

Wait, insurance companies aren't requiring genetic testing now, and they aren't exactly going out of business... how will this be any different? Really, this is more like preventing potential profit.

Posted by: ack ack ack ack on May 12, 2008 at 12:49 PM | PERMALINK

I disagree. If insurance were to routinely start keriotyping every applicant, that will increase costs for everyone across the board. While those increases might be washed out by charging genetically at-risk individuals higher premiums or denying coverage all together, or by considering certain genetic markers "pre-existing conditions," those costs come back again when medicaid/medicare has to step in.

Routinely telling people genetically at risk of certain expensive medical conditions to just "stay home and die" won't work. Neither will bankrupting insurance companies by forcing them to pay.

Short of nationalized single-payer healthcare (which, don't get me wrong, is *my* preferred solution--I'm just being pragmatic and not holding my breath for that to become a reality any time soon), the only vaguely sustainable option is to tell the companies to keep their grubby little hands off our DNA and keep doing what they've been doing--asessing the overall risks for certain population segments and spreading the costs amongst everybody's premiums.

Posted by: Orange Crush on May 12, 2008 at 12:56 PM | PERMALINK

Nothing here that is news to the insurance companies. They see this coming, and they are helping to channel the debate towards an arrangement that will continue to keep them out on the golf courses for a little while longer ... if they can't stay forever.

By structuring "universal" health care as a system based on private insurance, they are bargaining to insure the well while shunting the sick, once they're destitute, into a public system--single payer for those who can no longer provide the insurance companies with a profit.

Posted by: Boolaboola on May 12, 2008 at 12:56 PM | PERMALINK

When confronting market fundamentalists on health care, I like to ask this: If you believe that health care should be left up to the market - i.e. you get it if you're willing to pay for it, or pay for premiums, etc. - then why don't you advocate a private-market approach to police and fire services? If you didn't pay your monthly premium or fee, you don't get to call 911, you don't get to file a police report, and you don't get saved when your house is burning.

They can't come up with an answer without at the same time providing a justification for publicly-funded health care.

Posted by: DNS on May 12, 2008 at 12:58 PM | PERMALINK

I'm sorry, but as someone who works for an insurance company as a programmer... this is bunk. Our actuaries have been doing perfectly well without genetic testing for over a hundred years, and will continue to be able to assess risks afterwards.

Insurance companies assess risks in broad bases, not individually. An actuary is most out on a limb when they try to apply their statistical analyses to ONE person.


No, the problem is that we no longer think of health care as a society as a luxury, but rather as an entitlement. Health insurance, unlike almost all other insurance, makes no sense as a private agency now. What we NEED is a large enough pool of the insured that catastrophic coverage for things like cancer will not break the bank. What can be a larger pool than nationwide? 350 million people paying into an insurance plan will provide more than enough reserve for those who do end up needing catastrophic coverage.

Posted by: Kenneth Cavness on May 12, 2008 at 12:59 PM | PERMALINK

The inimitable Daniel Davies was the first I saw point this out on his personal blog going way back to about the time when he developed the whole Shorter (den Beste) thing. If I have time I will hunt it down.
Maybe this, but there are related posts (search on health insurance).
http://d-squareddigest.blogspot.com/2003/02/gene-pools-and-risk-pools-or-why.html

Posted by: theCoach on May 12, 2008 at 1:01 PM | PERMALINK

I'm with anything that will lead to single payer.

BTW, Novakula today writes about wacko religious wingnuts who are talking about voting for Obama cuz he'll lead to Armageddon or something.

ooookay. Same kind of thing. Yep, just vote for him, Rev. And let's see what happens.

Posted by: Cal Gal on May 12, 2008 at 1:03 PM | PERMALINK

So what's the eventual result of forbidding healthcare insurers to rationally assess risks when they write policies? They go out of business. Slowly, to be sure, but eventually they go kaput.

Why wouldn't they just spread the risk equally, like they do now? Please show your work.

Posted by: Boronx on May 12, 2008 at 1:04 PM | PERMALINK

Kenneth - Your actuaries have indeed been doing their jobs. But they still try to assess whether people smoke. They want to be able to use information concerning pre-existing conditions.

Upshot: Highly salient information that can be used further to subclass large populations will permit the company more aggressively to pursue business from low-risk customers at low cost. They want to do this and will do this absent regulation. The whole point of this debate is that some kinds of genetic information, like the BRCA and Huntington's genes, can have very high salience. Not all, but some.

Posted by: christor on May 12, 2008 at 1:04 PM | PERMALINK

No, Kevin, you are wrong. Insurers will not go out of business because they can't engage in "rational discrimination" so long as they are all on a level playing field and some of them can while others can't engage in some form of rational discrimination. It won't even make insurance more expensive in the aggregate, it will just tend to spread the costs out through more people, because insurers won't be able to "stick" the selected risks with the lion's share of the "extra" costs that are likely to be attributable to them. Which risks have to be "spread" versus those that can be selectively applied to individuals is what insurance regulation is all about. Making the risk more communal is a step in the right direction.

Posted by: Barbara on May 12, 2008 at 1:04 PM | PERMALINK

The problem is that you are asking conservatives to acknowledge a cognitive dissonance.

They want a 'private, free market' without actually realizing where that path leads. Sort of like complaining that universal care will lead to higher taxes, without grasping that 'we' are already paying $15k/yr AND taxes into the current broken system where 30% goes to paperwork.

Posted by: kis on May 12, 2008 at 1:05 PM | PERMALINK

Not exactly true. They can still factor in for the unknown. Of course that will cause all rates to go up. However, they will not inevitably go out of business. However, this situation does point our the irrational morality of our medical insurance system. The healty get afforadable medical care and the sick do not. Now since there are some with a conscience, the healthy pay high rate to take care of the sick. I guess we are still not ready to walk over dead bodies as we go in to by coffee at Starbucks.

Posted by: Elphage on May 12, 2008 at 1:08 PM | PERMALINK

It is a lot more complicated. On policy grounds, there are times when insurance companies are barred from making distinctions that the market says are rational. Can insurance companies refuse to sell life insurance to African Americans or charge them higher premiums because they have a shorter average life expectancy than whites? Or charge men higher premiums than women on the same grounds?

In Craig v. Boren, the Supreme Court held that a state could not legislate a different beer-drinking age for girls than for boys, even though in a statistical sense beer-drinking girls caused less harm than beer-drinking boys (One of the metrics was arrest rates, and part of the court's analysis was that this might just reflect social norms -- maybe cops arrest tipsy boys and chivalrously escort tipsy girls home). If the insurance company or state is using a suspect class like race or gender -- or in this case, people born with genetic risks -- as a proxy for a more nuanced risk analysis, then the courts have struck down the market-based rule.

I know that "people born with genetic risks" does not sound as suspect as race or sex as a basis for making this distinction, but the same kind of analysis applies. If all employers can refuse to hire a person who may down the road have health problems, then the person will be unable to earn the money that might let them ward off the health problem. In each case, a policy judgment is made that a particular risk has to enter the "pool", and the insurer is not allowed to exclude the risk.

Posted by: Genevieve on May 12, 2008 at 1:08 PM | PERMALINK

It is a lot more complicated. On policy grounds, there are times when insurance companies are barred from making distinctions that the market says are rational. Can insurance companies refuse to sell life insurance to African Americans or charge them higher premiums because they have a shorter average life expectancy than whites? Or charge men higher premiums than women on the same grounds?

In Craig v. Boren, the Supreme Court held that a state could not legislate a different beer-drinking age for girls than for boys, even though in a statistical sense beer-drinking girls caused less harm than beer-drinking boys (One of the metrics was arrest rates, and part of the court's analysis was that this might just reflect social norms -- maybe cops arrest tipsy boys and chivalrously escort tipsy girls home). If the insurance company or state is using a suspect class like race or gender -- or in this case, people born with genetic risks -- as a proxy for a more nuanced risk analysis, then the courts have struck down the market-based rule.

I know that "people born with genetic risks" does not sound as suspect as race or sex as a basis for making this distinction, but the same kind of analysis applies. If all employers can refuse to hire a person who may down the road have health problems, then the person will be unable to earn the money that might let them ward off the health problem. In each case, a policy judgment is made that a particular risk has to enter the "pool", and the insurer is not allowed to exclude the risk.

Posted by: Genevieve on May 12, 2008 at 1:08 PM | PERMALINK
"Insurance companies have to be allowed to assess risks when they set premiums. If they don't, then they aren't insurance companies."
This is painfully oversimplistic.

Just because they have to be able to assess risks does not mean they have to be able to assess risks at the level of the individual.

My health insurance comes through my company's plan. I would be utterly amazed if they did not assess the risks of insuring my company. They have never assessed the risks of insuring me personally.

If you think that this means they "aren't insurance companies", uh, more power to you, I guess.

Posted by: on May 12, 2008 at 1:09 PM | PERMALINK

Barbara beat me to it. The whole point of insurance is to spread risk over a pool of people so that no one person (or small group) suffers the full cost of a terrible event. To exclude people who might get sick might help simplify the insurance business it doesn't really address the problem. What we need is to increase the size of the pool, that stabilizes the fluctuations in the risk so that rational predictions can be made and reasonable premiums can be charged. At least in an ideal world.

Now, if you have a specific marker I can see a company requiring that you get regular checkups or engage in some other preventative practice to try to reduce future expense. Just like insurance companies encourage you to put in smoke alarms and such.

Posted by: JohnK on May 12, 2008 at 1:11 PM | PERMALINK

I would dispute this idea on the basis of taking it to its logical extreme: if the insurance companies could predict the future and see when anyone is going to get hurt or grow ill, they could make a fortune by selling insurance only to people who don't need it.

The point of insurance, from a societal point of view, is not for the insurance companies to do everything they can to ensure a sucker bet against their clients, it's to spread the costs of health care thin so that more people can afford it. From a business point of view, it is all about cleaning up on the suckers. Which helps explain so much of what is wrong with insurance companies.

This is the classic liberal vs. conservative view: liberals say everyone should take care of each other, conservatives say every man for himself. I pay for social insurance here in Japan and my premiums are way higher than the benefits I get. And I'm OK with that. It's the price you pay for more people getting better care, and you know that someday, you might be that other person.

Posted by: BlogD on May 12, 2008 at 1:11 PM | PERMALINK

I think Kevin is wrong when he says "So what's the eventual result of forbidding healthcare insurers to rationally assess risks when they write policies? They go out of business. Slowly, to be sure, but eventually they go kaput."

As a general rule, homeowner's insurance for things like fire are fairly level. The insurance company figures the odds of your house burning down is .1% and so they charge .15% and make a profit on the law of large numbers.

Health insurance for large corporations also basically do the same thing. If someone joins IBM then they get insurance.

Posted by: neil wilson on May 12, 2008 at 1:14 PM | PERMALINK

Does the decision say they can't use genetic information to understand what they are insuring? Or just that they can't use genetic information to make decisions on who to insure?

Regardless, all this will do is spread the risk out, which presumably is the purpose of insurance.

As Boronx says, "show your work". Or as my Professors would say along long time ago.

"Redo"

Posted by: jerry on May 12, 2008 at 1:22 PM | PERMALINK

I second the reccommendation for the post by dsquared. Anything I write here, he has covered better, but here are my 2 cents.

OK, health insurance companies have done fine without genetic testing so far, why would that change things?

Adverse selection - Assume that individuals know the results from their genetic tests, but the insurers don't. So they offer insurance rates based on the average risk in the respective pool. Will high-risk people buy insurance? More likely. Will low-risk people buy insurance? Less likely. The risk in the pool of people who buy insurance is greater than the risk in the whole pool. The insurer reprices, driving more people out of the market... The market unravels.

So a mandate IS necessary. Nationalized Health Care IS a subsidy from those of us with low health riks to the rest. And that's how it should be.

2)

Posted by: puzzled on May 12, 2008 at 1:23 PM | PERMALINK

I second the reccommendation for the post by dsquared. Anything I write here, he has covered better, but here are my 2 cents.

OK, health insurance companies have done fine without genetic testing so far, why would that change things?

Adverse selection - Assume that individuals know the results from their genetic tests, but the insurers don't. So they offer insurance rates based on the average risk in the respective pool. Will high-risk people buy insurance? More likely. Will low-risk people buy insurance? Less likely. The risk in the pool of people who buy insurance is greater than the risk in the whole pool. The insurer reprices, driving more people out of the market... The market unravels.

So a mandate IS necessary. Nationalized Health Care IS a subsidy from those of us with low health riks to the rest. And that's how it should be.

2)

Posted by: puzzled on May 12, 2008 at 1:23 PM | PERMALINK

BlogD is right. This post is terrible, its not insurance if you know the person won't get sick. Knowing the genetic makeup of a community is legit, but the idea that excluding individuals people who will get sick (for whatever known reason) is a good model for society wide health insurance is stupid.

Posted by: crack on May 12, 2008 at 1:24 PM | PERMALINK

And just to add: My fairness concern does not extend to people who engage in risky behavior. I believe society should help you out if you drew the bad draw in the genetic lottery, but not if you choose to do something dumb.

Posted by: puzzled on May 12, 2008 at 1:28 PM | PERMALINK

If insurance companies cannot use genetic tests in pricing or coverage decisions, but individuals can use genetic tests in making their decisions, doesn't that set up a serious imbalance?

Posted by: Wapiti on May 12, 2008 at 1:29 PM | PERMALINK

Kinsley's column (all the stalinism stuff) == whack job. Jeebus, how embarrassing.

Posted by: John McCain: More of the Same on May 12, 2008 at 1:31 PM | PERMALINK

If someone joins IBM then they get insurance.
Posted by: neil wilson on May 12, 2008 at 1:14 PM

Not under John McCain's heath plan. His idea is to make all policies individual so that all of us lack the clout to complain when we are screwed by the insurance industry for being too old or having a pre-existing condition.

Posted by: Ron Byers on May 12, 2008 at 1:31 PM | PERMALINK

If insurance companies cannot use genetic tests in pricing or coverage decisions, but individuals can use genetic tests in making their decisions, doesn't that set up a serious imbalance?
Posted by: Wapiti on May 12, 2008 at 1:29 PM

I don't know about you Wapiti, but I didn't get to choose my DNA. Just what genetic decision you get to make?

Posted by: Ron Byers on May 12, 2008 at 1:34 PM | PERMALINK

Kevin wrote:

So what's the eventual result of forbidding healthcare insurers to rationally assess risks when they write policies? They go out of business. Slowly, to be sure, but eventually they go kaput.

Well, no, that's (perhaps) the eventual result of absolutely prohibiting insurers to assess risks. It's not the eventual results of prohibiting the insurers from assessing one kind of risk they never assessed in the past anyway, yet their businesses still succeeded.

Still got random local high school students helping you type your posts up, Kev?

Anyway--

I think eventually genetics will play a much more prominent role in 1st World societies. We will confront problems we won't have a solution for, except for a gentic engineering/quasi-eugenics one, and we will take that solution so the problem can be solved-- just as we have been futzing around with this science so far.

Aldous Huxley, et al., were right to warn about eugenics becoming another basis for oppression by the state. But a limited use of genetics, I think, is going to be the answer we're confronted with as the only answer to problems man has been tried by since civilization began. Once we realize this, it will probably take generations-worth of political struggle to legally authorize the work; perhaps some countries will adopt it, but others won't overtly, and will only use it for a few secret purposes, preferring a "justice / human rights" rationale as their public face. It may be a tough question civilization is never able to come up with one "right" answer to.

But it goes without saying, the potential of genetics to correct or alleviate public health and other social problems is immense. Society could probably reap tremendous rewards from abandoning the kind of attitude that inspired the legislation.

Posted by: Swan on May 12, 2008 at 1:39 PM | PERMALINK

I am glad at least two candidates understand. Mike Gravel and Hillary Clinton. Perhaps Nader too, I don't know. The other candidates and their blogular punditry could learn a thing or two.

Posted by: jerry on May 12, 2008 at 1:39 PM | PERMALINK

I guess the comment at 1:09 makes a good point, too, though.

Posted by: Swan on May 12, 2008 at 1:40 PM | PERMALINK

Christor -- I see what you are saying, but my point is that while you can say "People X with genetic defect X' are 75% more likely to develop breast cancer", you cannot say "Person X with genetic defect X' is 75% more likely to develop breast cancer"; this is the danger of genetic screening. What they COULD do is take amalgamations from blind genetic tests and say "people in this subarea have a 20% higher likelihood of getting colon cancer; we will make premiums higher in this area", but even then a lot of people are going to say that this is not the way they want health insurance to work any more.

Posted by: Kenneth Cavness on May 12, 2008 at 1:42 PM | PERMALINK

If I know, from genetic tests, that I am very likely to develop a serious medical condition, I can decide to buy more coverage for that condition, no?

Posted by: Wapiti on May 12, 2008 at 1:43 PM | PERMALINK

For-profit health insurance doesn't make sense; insurance is supposed to be about sharing risk. Making a profit requires that risk be reduced. Reducing risk leaves the people who most need health insurance without. The upshot of all this is that health-care is simply not a commodity that can be turned over to the marketplace.

We accept socialized police and fire protection, public education, and highways. What's wrong with socialized medical care?

Posted by: cmac on May 12, 2008 at 1:43 PM | PERMALINK

I don't understand all the fuss. Democrats and Republicans agree that there is insufficient money to pay for a single-payer health plan.

http://thehill.com/leading-the-news/dems-hedge-on-healthcare-2008-04-23.html

"We all know there is not enough money to do all this stuff," said Sen. Jay Rockefeller (D-W.Va.), a Finance Committee member and an Obama supporter, referring to the presidential candidates" healthcare plans. "What they are doing is ... laying out their ambitions."
Who are we kidding?

Posted by: Everett on May 12, 2008 at 1:54 PM | PERMALINK

"If I know, from genetic tests, that I am very likely to develop a serious medical condition, I can decide to buy more coverage for that condition, no?"

Uh, Wapiti, good luck with that. Why don't you try it and get back to us?

Nine years ago, my then-28-year-old daughter was denied coverage by a major health insurer because she had a fibroid tumor of the uterus - a condition that upwards of 80% of all women will have at some point in their lives. Now she can't buy private insurance because she has been denied coverage in the past.

Private health insurance as it exists in this country today ensures one thing: only the people who don't need it will be able to buy it.

Posted by: cmac on May 12, 2008 at 1:54 PM | PERMALINK

Wapiti--
Yes, but...
How much insurance can you buy to cover your condition? (Remember that this is health insurance rather than life insurance we are talking about, so nobody is going to get rich.)
How many people are relatively sure that they are going to get a costly condition? Even if the insurance companies lose money on those people, aren't they still going to earn an overall profit?

Posted by: reino on May 12, 2008 at 1:57 PM | PERMALINK

Just to be clear, I'm not saying that Congress necessarily reached the wrong result on the particular piece of legislation, either. There are a lot of factors that go into determining public policy, and I feel like I don't know enough about the issue to have a good opinion on whether it was the right or wrong answer.

But if the attitude is reflexively, "Oh my God, this about testing people genetically to determine how the law should treat them or what can be done with them, and we can never do that," I think that attitude or some of the assumptions that give rise to it are standing in the way of a lot of good we can be doing.

They have already discovered that genetics makes Rush Limbaugh and the Republicans who didn't clean up after Hurricane Katrina and who are keeping us in this stupid Iraq war the way they are (google for "Liberal brain"- it's in an LA Times article). What if one day we as a society really recognize that all these people just got us into a stupid mess? If being that kind of a conservative is an aspect of your physical makeup, it's not really political oppression to take action to respond to it as a society. It's a response to a public health crisis.

Early genetic thinking that inspired eugenics was too crude, and people painted with too broad a brush- thinking that it justified discrimination against ethnic groups. It doesn't. But do we really need our society run by public servants who see their official purpose as arranging kickbacks and corruption for big business? Those individuals could be directed into professions where they don't wield so much power, or the number of spaces in a legislature that can be occupied by them could be limited. I'm not advocating one policy or another yet, but I want people to recognize that this is a broad new horizion, and we should carefully consider all the possibilities before we start advocating one thing or another to do.

Posted by: Swan on May 12, 2008 at 1:58 PM | PERMALINK

There is enough money, actually. There isn't enough to retain profit. But there's plenty of money to cover everyone. The barrels of money being poured into healthcare now would simply be redirected (and reduced, if the statistics on health care in Europe and Asia are to be believed) into a nationalized plan. Think of the joy in boardrooms all across America as companies were relieved of the costs of medical benefits.

Posted by: cmac on May 12, 2008 at 1:58 PM | PERMALINK

Everett--
Who are you kidding?
The US government spends more money per capita than the British government on health insurance. For that money, the Brits get health insurance, while we don't. Yet one Senator claiming that we can't afford universal coverage without even considering the different ways that universal coverage can be implemented is supposed to prove that it's impossible? It is possible, and if you don't believe me you should consider any country in the world other than ours.

Posted by: reino on May 12, 2008 at 2:01 PM | PERMALINK

Mr. Drum sounds here like about as much of a crackpot as Kinsley.

There absolutely has to be some sort of law against genetic discrimination. Genes, hard though it may be to believe, ARE NOT DESTINY. You may have a 70% chance of getting something. You may have a 90% chance of getting something. BUT YOU MIGHT NOT GET THAT DISEASE. Unless you're talking about a genetic disease that a person is born with, something really hideous, MOST DISEASES are a matter of CHANCE AND PROBABILITY, NOT ABSOLUTE DESTINY.

It is completely asinine to think or suggest that insurance companies cannot manage to figure out how to balance insurance payouts, coverage, and premiums. THAT is what they do. And they need to--they HAVE TO--do it for MOST PEOPLE.

What Mr. Drum seems to want is insurance companies that work only as a rigged system, like Vegas casinos. They COULD let skilled gamblers gamble, and just adjust costs and payouts, but INSTEAD they completely exclude them.

Skilled gamblers compared to average joes with a possibility of disease: It's a perfect metaphor. You're welcome.

Posted by: Anon on May 12, 2008 at 2:02 PM | PERMALINK

Sometimes Kevin has no idea what he's talking about. This is one. Damn shame he doesn't at least acknowledge what Kenneth Cavness and others say above.

Posted by: David in NY on May 12, 2008 at 2:07 PM | PERMALINK

Kevin writes:

So what's the eventual result of forbidding healthcare insurers to rationally assess risks when they write policies? They go out of business. Slowly, to be sure, but eventually they go kaput.

That's not necessarily so. For the insurance companies to make a profit, it's only necessary that the average premium paid in is less than the average payout. Laws about nondiscrimination increase the average payout, so the insurance companies would necessarily have to increase the average premium to compensate.

The real question is whether the premiums rise so high that significant numbers of low-risk customers drop their insurance altogether. So the critical numbers are (1) X: How much it would cost, per person, to pay for medical care for everyone who wants insurange, and (2) Y: How much are healthy people willing to pay for insurance? If X exceeds Y, then you're right, that insurance companies are doomed (unless they can be discriminatory in their coverage and/or premium structure). But it isn't necessarily the case that allowing everyone equal access to insurance would push X above Y.

Posted by: Daryl McCullough on May 12, 2008 at 2:19 PM | PERMALINK

Indeed, regarding insurance evaluation: insurers can use genetic testing information in the aggregate to estimate the projected occurrence of a given disease irrespective of lifestyle, and spread its cost throughout its insurance pool. It could even do this on a population (though not individual) basis.

Regarding adverse selection: there are actually a lot of safeguards in place to avoid adverse selection within group markets. Within individual markets, there is a greater risk of adverse selection, but it is important to remember that most genetic tests are predictive -- moreover, human behavior being what it is -- many people are quite likely to discount their own risks of developing the thing that was tested against, and thus, it is not necessarily apparent how big a role it would play in the decision to buy or not buy insurance. Which is another way of saying, that imminent health concerns are a far better predictor of adverse selection than future ones.

Posted by: Barbara on May 12, 2008 at 2:33 PM | PERMALINK

Kevin,

Your comment missed an important point.

An insurance company can choose what entity it ensures for risk. You assumed that the entity being insured has to be the individual, but it doesn't. The insurance company can insure the pool of individuals instead, then calculate the risk of events to that pool.

When they sell a policy they are simply gathering in members of the pool, but the risk remains that of the pool. If the pool is the group of all possible individuals in a given geographic area, they need not discriminate between healthy and unhealthy individuals.

There is a risk that the individuals any one insurance company would insure would, by pure chance, be more unhealthy than average, but that can be covered by buying reinsurance. The cost of reinsurance would offset the risk that they would insure unusually healthy individuals. (Risks can be either positive or negative, but are still risks.)

Adverse selection is not a problem if every individual in the pool must pay for insurance - universal mandatory insurance. How they pay is a totally separate discussion. Some might do so individually, some might be paid for through taxes, but that mix is quite separate from the discussion of how the insurance works.

Obviously this requires the government to set up the risk pool that covers everyone and mandate that everyone in some manner contribute towards the expenses of the pool. The insurance companies cannot be permitted to establish their own risk pools. But as the Swiss model apparently shows, non-profit insurance companies can operate in this environment as insurance companies.

The only viable business model outside this is for insurance companies to skim the cream of healthy clients out of the pool, get them to pay the full cost of what they would pay if they were covering the entire pool, then use a variety of tricks to avoid paying claims. They then pay most small claims, but avoid many of the really large ones. by avoiding the large claims they save money, make a profit, and they do it rarely enough to avoid being seen as acting disreputably. They certainly are acting disreputably, but hide it.

That's what we have now. It leaves those most in need of health coverage out of the system and makes a few insurers quite wealthy.

Posted by: Rick B on May 12, 2008 at 2:36 PM | PERMALINK

Genevieve

You are basicly making the argument that in our society in which more and more data is obtained on every individual, there has to be certain areas of privacy in which data is not collected or used.

I know truck drivers who cannot get work because in the past they had some form of back surgery. That's another of those suspect categories, and using it to deny employment because of the danger the employer will have to pay medical expenses doesn't make overall sense. But it makes perfect sense to a small employer who is paying for health insurance for his employers.

With all that data out there on everyone, a lot of it is misused to make discriminatory decisions against each of us. That's a major reason for privacy laws.

Frankly, too many companies know way too much about each of us. Insurance data is only the tip of that iceberg.

Posted by: Rick B on May 12, 2008 at 3:01 PM | PERMALINK

Everett,

If the government doesn't have enough money to pay for single payer health care, then how is it that the governments alone in the U.S. already pay out more per capita than the single-payer health plan in Canada does?

And somehow, the Canadian health care system provides better health care overall than the American lack-of-system that is two-and-a-half times more expensive.

The U.S. government has more than enough money for single payer, and we'd get better health care out of it. Our automotive industry (among others) would almost immediately become more competitive internationally, too.

There is simply a lack of will to face down the predatory insurance companies who are sucking up the excess money and giving nothing back for it. Private for-profit health insurance companies are a cancer or infection on the American economy.

Posted by: Rick B on May 12, 2008 at 3:18 PM | PERMALINK

reino, cmac, Rick B,

Multiple Democratic Senators were quoted in the article I cited. It seems that your argument is with the Democratic leasdership -- not with me.

Posted by: Everett on May 12, 2008 at 3:48 PM | PERMALINK

Rick B,
Thank you for your comments, but I am not sure that I was posting about privacy. I have no problem with insurers discriminating against people who smoke in private, for example.
My concerns are: (1) the one several other posters have raised more succinctly about risk-pooling/concentrating. Insurance companies should not be allowed to exclude the highest genetic risks from the pool, and will not go out of business if genetic risks are spread over a large pool rather than concentrated on the at-risk individual; and (2) the fact that a distinction is in some sense "objective"(women live longer than men on average, whites live longer than African-Americans on average, those with certain genes are more likely to get cancer) does not make it legal, moral, or good policy to allow certain financial consequences (insurance costs being a prime example) to be based on them.
We make this kind of legal and policy determination all the time, without the absurd consequences posited by Kinsley. Employers must make "reasonable accommodations" for disabled employees. This might mean allowing a seeing-eye dog into the workplace, or paying for special computer equipment -- it does not mean you have to hire chauffeurs who are blind. But legally, we require employers and the broader society to shoulder some of the costs for accommodating difference, rather than sloughing them all off on the individual.

Posted by: Genevieve on May 12, 2008 at 4:06 PM | PERMALINK

I rushed post by Kevin, but most seem to be missing his point: Conservatives imposing government rule over the free-market flow of information are being inconsistent.

And BTW,
A: Insurance companies DO deny individuals coverage.
B: Kev is not suggesting genetic discrimination is a good thing.

Posted by: Bush Lover on May 12, 2008 at 4:07 PM | PERMALINK

With regards to risky behavior. In the UK, they like those who smoke. The tax money on tobacco is more than the cost of running the entire national health service ...

Posted by: royalblue_tom on May 12, 2008 at 4:13 PM | PERMALINK

As a bunch of people have pointed out, this really makes no sense, because if it did insurance companies would already have gone belly-up. (Unless, of course, you assume that the 40-plus million uninsured are all that way entirely by choice because they've made a profit-maximizing decision not to have ready access to health care...)

The crucial point is that -- especially in a sellers' market -- genetic tests supply far more information to the insurance company than they do to prospective insured. Who do you think has a better understanding of statistics and of what the costs of a particular disease might be.

It's no longer insurance when your contract can be written to exclude coverage for all the big-ticket diseases you're actually likely to get...

Posted by: paul on May 12, 2008 at 4:20 PM | PERMALINK

It simply means that insurance companies have to pass along the costs of that kind of insurance all across the board. It doesn't mean that they can't assess what the costs will probably be.

Posted by: Jeffrey Davis on May 12, 2008 at 4:32 PM | PERMALINK

BlogD and others above hit these points already, but let me repeat them in different words.

1) The more insurance companies can rationally assign cost based on risk, the less useful insurance is. Your description of 'insurance' suffers from the following simple reductio ad absurdum argument. Suppose health insurance companies could predict future health care needs for an individual, perfectly. Then they would rationally charge that individual exactly that cost (appropriately discounted for investment growth, etc.) Naturally, this means that the individual would have absolutely no reason to buy usch insurance, and the insurance company would go out of business. More accurate risk allocation makes insurance LESS valuable.

2) Insurance is thus defined not just by risk assessment. That's only half the story. The other half is risk distribution. Without risk distribution, no one would have any reason to buy insurance. Consequently: accurate risk assessment by individuals makes insurance less useful. Accurate risk assessment by populations makes insurance financially viable. The bigger the popluation, the more valuable the insurance.

Sorry, Mr. Drum, but you missed half the story on this one!

Posted by: PQuincy on May 12, 2008 at 5:19 PM | PERMALINK

Thanks for those who point out that KD's point is really that 'conservatives' in Congress were somehow being inconsistent, here. I'm not sure about that -- because the term 'conservative' has no referent, nowadays, (any more than 'liberal'), and one might think that a (Burkean) conservative would hesitate to reform a well-established institution (insurance) on the basis of shiny new technology without being very very cautious. Just saying...

Kinsley (with Drum's agreement) still misrepresents how insurance (market-based insurance) works, though, I think.

Posted by: PQuincy on May 12, 2008 at 5:28 PM | PERMALINK

The posters here have given considerably more thought to this issue than Kinsley has. I'm still trying to figure out why the guy thinks it "unfair" that he can't play the cello like Yo-Yo Ma, design buildings like Frank Geary, or pick stocks like Warren Buffet. How much does he think he's "fairly" entitled to? After reading that, everything else he had to say in this column seemed suspect.

Posted by: on May 12, 2008 at 5:50 PM | PERMALINK
So what's the eventual result of forbidding healthcare insurers to rationally assess risks when they write policies? They go out of business. Slowly, to be sure, but eventually they go kaput.

The Blues throughout the country did fine with community underwriting for decades. If all insurers have to use the same underwriting standards, there are no problems. Look at Medicare or the VA. They don't do individual underwriting, but they manage to offer service for better prices than private insurance for comparable risks.

Posted by: freelunch on May 12, 2008 at 7:04 PM | PERMALINK

PQuincy has it right - a private insurance market fails when insurers are able to use underwriting information which gives them the ability to: precisely predict the cost a policy will incur and charge a rate commensurate with that cost. Once these two conditions are met, the risk-spreading feature of insurance goes away and there is no reason for anyone to buy it. As insurers get access to more accurate cost-estimating techniques, their profits increase, while the utility of the product to its purchasers decreases.

Posted by: Proudhon on May 12, 2008 at 7:28 PM | PERMALINK

Think about it from the patient or doctor's point of view, as well as the insurance companies' viewpoint. What the logical consequence for a parent who pays a boatload of insurance due to a genetic condition such as Marfan's? They might tell their children, "Never, never, never get the test that identifies you as having Marfan's. If you've got it, you'll pay too much for insurance even though the mere presence of Marfan's does not mean you will get all the diseases associated with it." The potential for a doctor to track that patient for onset of Marfan's symptoms--blindness, dissection of the aorta, etc--disappears for lack of the test. The test is avoided because insurance companies impose high premiums on anyone with the Marfan's gene.

Posted by: tr on May 12, 2008 at 7:33 PM | PERMALINK
So what's the eventual result of forbidding healthcare insurers to rationally assess risks when they write policies?

Um, outlawing genetic discrimination (or any other kind of price discrimination) does not prevent risk assessment, it prevents efficient (from a profit perspective) market segmentation, which the natural effect that prices rise to the cost of the most expensive to insure, until only those that most need healthcare or have the strongest aversion to risk (and thus place the highest premium on being "covered") have any incentive to purchase insurance. It narrows the insurance market, but it does not destroy the capacity to profitably survive in that market.

Posted by: cmdicely on May 12, 2008 at 7:48 PM | PERMALINK
Conservatives all claim to believe that the private market is the best way to provide health insurance.

Actually, no. Conservatives are mostly anti-insurance altogether. They mostly claim to believe that the best way to provide healthcare is by individual direct purchase of service, not insurance at all, which is why they push for "Health Savings Accounts" rather than comprehensive insurance.

So, granting your premise that this plan has the natural affect of destroying the insurance industry, this is perfectly compatible with their stated preference of replacing today's insurance-based healtchare financing system with a direct-purchase-of-service system. Its all part of the "ownership society" (a society constructed of, by, and for the owners of capital, in which everyone else is exposed to maximum risk to encourage maximum compliance.)

Posted by: cmdicely on May 12, 2008 at 7:52 PM | PERMALINK

Think of universal health care as insuring people before they're even born -- against the risk of getting a bad break in the genetic lottery.

Posted by: SqueakyRat on May 13, 2008 at 2:44 AM | PERMALINK

Then what we need is affordable health CARE, not health INSURANCE.

Posted by: lilybart on May 13, 2008 at 10:06 AM | PERMALINK

Suppose (hypothetically) that insurance companies could charge every one exactly the correct cost for their own life time health insurance, properly amortized over time. They would no longer be in the insurance business. They'd be a cost-added product to pay for their service, but there would no longer be an insurance aspect to the charges, as it would be fee for service. Health care would cost the same it does now, plus the insurance company additional charge.

If this came to pass, you could eliminate the insurance company, and reduce the price of health care!

A few further comments at my own blog, if I'm allowed to advertise. http://biostatprof.blogspot.com/

Posted by: Biostatprof on May 13, 2008 at 5:29 PM | PERMALINK




 

 

Read Jonathan Rowe remembrance and articles
Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

Advertise in WM



buy from Amazon and
support the Monthly