Obamacare 5 years later. Crow, anyone?

[Emily's Cat]

a positive externality that benefits other countries and is fully voluntary on the part of the US as a side effect of transactions made by the US in the US's own interests.

Not sure... but I rather suspect that the actual consumers of drugs aren't all that gung-ho about paying exorbitant prices just because the pharmaceutical companies can get away with it in the US.

 

[Emily's Cat]

The rest of the world is subsidising the US's capture of innovation and the US is freeloading off other countries' willingness to adopt tech later.

I honestly don't know what this is supposed to mean.

 

[Emily's Cat]

Sorry, but that is not how "freedom" works. Here in the USA, you have the right to "life, liberty, and the pursuit of happiness."

You have the right to life!

You have the right to not have your life forcibly taken from you by the actions of another; You do not have the right to obligate other people to keep you alive.

You have the right to liberty. You have the right to be free and not indentured to any other person; You do not have the right to deprive others of their liberty in attaining your own.

You have the right to the pursuit of happiness. You have the right to do that which makes you happy; You do not have the right to deprive other people of their happiness in attaining your own.

You have the right to these things only so long as you are not impinging on the rights of other people in order to attain them.

 

[Francesca R]

I don't think tying health insurance to employment makes sense though.

It doesn't to the extent that there is some compulsion involved (no cash alternative to what is really part of your compensation) and it also ties you to your job more than if it was not there (here at least if you switch firms you may suffer a reset of what pre-existing things are)

But it does make sense to gear up bargaining power because otherwise insurers would have all of it.

As for the people I know who like the private market, they're the types for whom cost is really not a consideration.

Dumb people you mean? Or those with money to burn. Well reviews from those folks are not terribly representative of whether the insurance is a good thing!

 

[Francesca R]

Underlying cost structure and age are one thing. But GIVEN all of that, if Bob experiences a high claim year (his claims end up with either higher than normal frequency or higher than normal severity or both), there is NOT CHANGE in the likelihood of Joe having higher than normal claims.

That isn't a necessary condition for health costs to be positively correlated.

 

[Emily's Cat]

Dumb people you mean? Or those with money to burn.

Or people with conditions who reasonably expect they will need a higher than average level of care. Or those who are more risk averse than normal, and are willing to exchange a premium for peace of mind.

 

[Emily's Cat]

That isn't a necessary condition for health costs to be positively correlated.

Okay. It's just that the rest of those correlations are boring, well understood, and do nothing at all to deal with the predictability of an individual person's claims. They're pointless correlations. They affect the mean of the distribution, but they're meaningless beyond that.

And adding more people into the mix doesn't materially reduce the variance of the claim probability distribution because the variance of claims experience for each individual is independent of the claims experience of any other individual in any way that actually matter with respect to the aggregate variance of the pool.

 

[Francesca R]

I contend it is rather dumb for cost not to be a consideration.

 

[Francesca R]

Okay. It's just that the rest of those correlations are boring, well understood, and do nothing at all to deal with the predictability of an individual person's claims.

You can't diversify them away is what matters about those. All very well to say they are boring but you previously said they were zero.

And adding more people into the mix doesn't materially reduce the variance of the claim probability distribution because the variance of claims experience for each individual is independent of the claims experience of any other individual in any way that actually matter with respect to the aggregate variance of the pool.

Adding more people eliminates precisely the non systematic risk that you are, erm, not bored by. That part can be diversified away. Beyond a relatively small pool I suspect it shrinks to negligible for the insurer. But for the insuree, it is eliminated for every person that joins so there is no fall off in utility from the perspective of more and more people joining.

 

[Nihilianth]

You have the right to not have your life forcibly taken from you by the actions of another; You do not have the right to obligate other people to keep you alive.

Incorrect. Example: Those who are licensed to do CPR, are obligated to attempt to keep you alive.

You have the right to liberty. You have the right to be free and not indentured to any other person; You do not have the right to deprive others of their liberty in attaining your own.

No argument.

You have the right to the pursuit of happiness. You have the right to do that which makes you happy; You do not have the right to deprive other people of their happiness in attaining your own.

You have the right to these things only so long as you are not impinging on the rights of other people in order to attain them.

No argument.

 

[DavidJames]

You have the right to not have your life forcibly taken from you by the actions of another; You do not have the right to obligate other people to keep you alive.

Are you saying hospitals can turn people away, refuse to treat them?

 

[Architect]

Yes, I know. And all of the people I know who can afford it think it's great. Definitely much closer to a free market than the US system. I haven't heard anything about the relative prices for drugs in that system though. I've only seen data for what the NHS pays.

But it does rather undermine your suggestion that the NHS in its various incanations operate an effective buyers' monopoly within the UK which in turn result in a "take it or leave it" position for drug suppliers as well. If you were unaware of UK PHI operation I'm a bit unclear on how you can be so certain in your position.

Indeed as I mentioned, US based large PHI firms are in a similar position inasmuch as they negotiate coverage and treatment costs; they can apply the cost/benefit calculation behind closed doors (in contrast to the NHS, which publishes the information) and moreover have a profit incentive.

 

[Architect]

I'm certainly not against voluntary collective bargaining. I don't think tying health insurance to employment makes sense though. I thought it was a quirk of the US market, but apparently our insanity has spread across the pond.

You misunderstand the position with PHI in the UK.

The NHS covers all healthcare needs, free at the point of demand, with the exception of purely cosmetic work (which is not considered essential other than in certain restricted circumstances) and dentistry (where there is a patient contribution to total cost).

There are waiting lists for procedures which vary according to local funding and prioritisation. So, for example, a simple athoroscopic knee examination might take 3 months. Details of these waiting lists are published and there are, in England at least, maximum stipulated periods. These waiting lists do not tend to apply to urgent or serious treatment: by way of example if you are diagnosed with cancer, NHS Scotland will have you up at the Beatson in Glasgow double-quick.

PHI will generally do no more than allow you to jump the waiting list for the less urgent work. Most of the private hospitals - Carrick Glen, Murrayfield, et al - are not set up for what might be termed broad spectrum healthcare coverage. Think hips, cataracts, 'roids and knees. In fact I can only think of 3 broad spectrum hospitals in Scotland, such as Ross Hall, all of which are small.

The reason for the latter is that the NHS is frankly good and fast when it needs to be.

On that basis PHI in the UK is seen very much as a nice perk for people. Nobody I am aware of would choose a job because it came with PHI, indeed in Mrs. Architect's case I suspect we'd have taken cash in lieu.

Therefore tying it to employment really doesn't matter a monkeys. The NHS will always cover me for pre-existing conditions anyway.

Let me give you an example. I have pigmented villonodular synovitis (PVNS), a rare and potentially crippling knee condition which will recurr notwithstanding drug treatment and synovectomies every 10 years or so. Mrs. Architect's PHI covers pre-existing conditions at present. If she was to change employer and they did not do so, then that's okay because the NHS cover it anyway. And there are only a handful of surgeons in the UK (and US, for that matter) who treat PVNS so it doesn't actually mean I get any different treatment. Since my condition is treated as a priority, the only difference with PHI was that I got a private room and week (whoopee) knocked off the pre-op period.

 

[Emily's Cat]

I contend it is rather dumb for cost not to be a consideration.

I don't know what you mean by this. Do you think that cost is NOT being considered?

 

[Emily's Cat]

You can't diversify them away is what matters about those. All very well to say they are boring but you previously said they were zero.

Not on purpose I didn't.

Adding more people eliminates precisely the non systematic risk that you are, erm, not bored by. That part can be diversified away. Beyond a relatively small pool I suspect it shrinks to negligible for the insurer. But for the insuree, it is eliminated for every person that joins so there is no fall off in utility from the perspective of more and more people joining.

No, it cannot. This is what I'm trying to explain. And I'm doing it very poorly, apparently .

Look - if it were true that the non-systemic risk in health insurance could be diversified away, then insurers wouldn't care which policyholders they got, they'd only care that they had more of them. They would only care that they had a large number of policyholders.

But we don't - we need enough to make our fixed operating costs reasonable in the marketplace, but beyond that, we'd be happiest if we had a smaller number of the right kind of people than a large number that included some of the wrong kind.

The systemic variance in health insurance - the variance in negotiated rates, utilization, new technologies and treatments entering the market, regional variations in price and behavior, and the variance in cost associated with age and gender and smoking status - these are all very predictable. These are not problems for insurers. This is not where our worry is.

Our worry is around the variance of claims at the individual person level. It's the combination of the expected cost level for person A and the length of the claim cost tail on person A's distribution. And that variance doesn't get diversified away.

If we have a pool that consists of 100 healthy people, and we add one hemophiliac... the likelihood of having a claim of $1M or higher does NOT reduce. It increases. It increases because the hemophiliac's variance is NOT offset by the variances of the other 100 healthy people.

My understanding is that with stocks, every time you add a new stock to your portfolio, the variance of the portfolio is reduced, correct? So the more you add, the lower the variance. The more you add, the lower the likelihood of experience a return that is more than X standard deviations from the mean.

This is not the case for health insurance. When you add more people, it does NOT necessarily reduce the variance of the pool. It does not necessarily reduce the likelihood of experiencing a claim more than X standard deviations fro the mean. Even more than that, it can increase the likelihood that the aggregate claim level of the pool is further away from the predicted mean.

In stocks, it is often the case that the non-systemic portion of the risk is negatively correlated with a different stock, right? When domestic stocks perform well, foreign stocks perform poorly, and vice-versa. They are both reacting to the same underlying stimuli, but they move in opposite directions. My understanding is that this is why diversification works.

This does not happen with health insurance. No two peoples' non-systemic risks move in opposite directions. When Bob gets unexpectedly sicker, Joe does NOT get unexpectedly healthier. That's NOT how it works. They do NOT offset. The non-systemic portion of the risks are NOT correlated with each other - they are nearly perfectly uncorrelated.

 

[Emily's Cat]

Are you saying hospitals can turn people away, refuse to treat them?

Yes. And they do.

The only part of the hospital with any obligation is the ER - and that is a very limited obligation. The ER is only obligated to stabilize your condition. They absolutely will turn you away if you walk in asking for chemotherapy.

 

[Francesca R]

Not on purpose I didn't.

I thought you did:

In health insurance risk pooling, it's reasonable to assume independence of the "investments" - the investments are all independent people. With the exception of epidemics of communicable disease, the experience of one person does not affect the experience of another. One person's health doesn't affect any other person's health.

I do mean independence in the sense of zero correlation. Or at least very, very near zero. As I said - pandemics, hurricanes, those sorts of catastrophic events affect the whole population, so not perfectly zero. But very, very, very near zero.

. . . . Which is what I responded to. What you seem to mean in fact is that the residual, non-systematic, idiosyncratic health costs are not positively correlated. Which I have not disagreed with and indeed agree with. But that is not a sufficient condition for folks' total health costs to be independent.

And again--if the non-systematic health costs are not correlated then they can indeed be diversified away by pooling them, which is what insurance is.

Our worry is around the variance of claims at the individual person level. It's the combination of the expected cost level for person A and the length of the claim cost tail on person A's distribution. And that variance doesn't get diversified away.

Verily I say unto you, it does.

If we have a pool that consists of 100 healthy people, and we add one hemophiliac... the likelihood of having a claim of $1M or higher does NOT reduce. It increases. It increases because the hemophiliac's variance is NOT offset by the variances of the other 100 healthy people

Well yes it is. The 100 healthy people would all have a lower expected cost than "the market" and in portfolio theory jargon they would be low beta. That pool would therefore have a risk relative to the expected systematic health cost of the entire population. There would be a relatively high risk that the claims from this pool of healthy folks would be lower than the expected population average. The observation that this is "good risk" (insurers like it) does not change that. Add an unhealthy member (high beta) to that pool and reduces the likelihood that the claims coming from the pool are lower than expected population average.

Your "likelihood of having a claim of $1M or higher" is a statement about total risk not non-systematic risk. Adding an unhealthy member to a pool of healthier (than average) ones would raise the total risk but reduce the risk relative to the market.

My understanding is that with stocks, every time you add a new stock to your portfolio, the variance of the portfolio is reduced, correct? So the more you add, the lower the variance.

No it could increase. The non-systematic risk is generally lowered unless the stocks are very similar. Just like happens in an insurance pool unless the two insured members are very similar.

This is not the case for health insurance. When you add more people, it does NOT necessarily reduce the variance of the pool.

Agree and neither does it necessarily do this with financial assets.

It does not necessarily reduce the likelihood of experiencing a claim more than X standard deviations fro the mean.

I think it does. I think it does with your hypothetical of a hundred healthy (healthier than average) folks plus one unhealthy one (less healthy than average).

Even more than that, it can increase the likelihood that the aggregate claim level of the pool is further away from the predicted mean.

You have not shown this, as above I think your hypothetical argues the other way.

 

[Francesca R]

I don't know what you mean by this. Do you think that cost is NOT being considered?

sunmaster14 said that people who like (UK) private health insurance are the types for whom cost is really not a consideration. OK well if someone likes something but the cost of the something isn't really a consideration then I think there is no information in their positive view of it.

 

[RandFan]

OK well if someone likes something but the cost of the something isn't really a consideration then I think there is no information in their positive view of it.

Isn't "cost" only a consideration if it fits your definition of what govt is responsible for? How often do you hear about Democrats gutting the most expensive military in the world (never mind it is and always has been a myth that helped doom Romney's nomination)? Does cost matter when it comes the defense of America? I think it should. I think it should matter when it comes to health care. That's one of the reasons I wanted to get the cost of health care in America commensurate with other nations.

Supporting social spending for health care is not akin to not caring about costs anymore than supporting social spending on the military means one does not care about costs.

 

[Emily's Cat]

I thought you did:
. . . . Which is what I responded to. What you seem to mean in fact is that the residual, non-systematic, idiosyncratic health costs are not positively correlated. Which I have not disagreed with and indeed agree with. But that is not a sufficient condition for folks' total health costs to be independent.

Which is why I said it wasn't on purpose . What I said, and what I was trying to say weren't the same thing, and I said it wrong. Thus... Not on purpose

And again--if the non-systematic health costs are not correlated then they can indeed be diversified away by pooling them, which is what insurance is.

Okay. Take me back to day 1, would you? Clearly I'm not getting this.

In particular, where am I gone wrong with this bit:

In stocks, it is often the case that the non-systemic portion of the risk is negatively correlated with a different stock, right? When domestic stocks perform well, foreign stocks perform poorly, and vice-versa. They are both reacting to the same underlying stimuli, but they move in opposite directions. My understanding is that this is why diversification works.

This does not happen with health insurance. No two peoples' non-systemic risks move in opposite directions. When Bob gets unexpectedly sicker, Joe does NOT get unexpectedly healthier. That's NOT how it works. They do NOT offset. The non-systemic portion of the risks are NOT correlated with each other - they are nearly perfectly uncorrelated.

What am I saying wrong? What am I thinking about incorrectly? Where am I juggling terms? Please set me straight so I can speak coherently!