Obamacare 5 years later. Crow, anyone?

[Francesca R]

Ok, well I don't think those systems work as well as a free market based system would, in which poor people are subsidized by the government with dollars rather than free health care services.

I suppose one could make health insurance mandatory and also have people billed for their actual risk and then reimburse everyone down to a flat level, or just reimburse poor folks. But that would be functionally equivalent.

If health insurance is voluntary, then it makes no difference if the government provides cash or not from the perspective of the self-selection problem: low risk folks would opt out, raising the average cost for everyone else, which would prompt some moderate risk folks to opt out, raising the average cost again, prompting medium risk folks to opt out, and so on.

And I think one of the reasons that universal health insurance seems to be working in other countries relative to the US is that those systems are benefiting from the health care largess of the US.

If, say, that was the only reason why other countries spend way less, then the US would be spilling out a positive externality to the tune of about 7% of non-US OECD GDP which is an order of magnitude more than it spends on foreign aid.

I would tend to agree with you that there probably is an externality, but I suspect it is a tiny part of the difference in cost. Not that cost is the only metric of "working".

 

[Architect]

http://www.nih.gov/about/impact/impact_global.pdf

This suggests that whilst the US might be paying more than it's share by size, it's not the only one and that the playing field is levelling.

 

[sunmaster14]

I suppose one could make health insurance mandatory and also have people billed for their actual risk and then reimburse everyone down to a flat level, or just reimburse poor folks. But that would be functionally equivalent.

If health insurance is voluntary, then it makes no difference if the government provides cash or not from the perspective of the self-selection problem: low risk folks would opt out, raising the average cost for everyone else, which would prompt some moderate risk folks to opt out, raising the average cost again, prompting medium risk folks to opt out, and so on.

If everybody is charged their expected cost, then it makes no difference if healthy people opt out of the system. You only need to force healthy people into the system if you are charging them an artificially high price so as to subsidize unhealthy people. If you are using taxes to subsidize unhealthy people (or poor people), then you don't need to force anybody into the system.

If, say, that was the only reason why other countries spend way less, then the US would be spilling out a positive externality to the tune of about 7% of non-US OECD GDP which is an order of magnitude more than it spends on foreign aid.

There are other reasons for the high US health care costs. We have an obesity problem, which is in large part (no pun intended) cultural. We have other cultural issues, which are wrapped up with some of our racial issues, that leads to higher health costs (teenage pregnancy being one; poor adherence to doctor prescribed regimens being another). Perhaps our biggest problem, however, is the exorbitant amounts we spend on end-of-life care, which once again is a product of cultural attitudes. So I don't think the implicit subsidy is anywhere near 7% of our GDP, but it is pretty high nonetheless. The Rand paper I cited tried to quantify it. As best as I could understand it, the subsidy for Europeans amounted to $8,000 per capita for 55-59 year olds, net present value (presumably accumulated over one's remaining lifetime). Assuming an expected remaining life of 25 years, that's about $400 per year, per capita, for people in that age range. In my opinion, that's a pretty big subsidy.

ETA: I just realized that the paper indeed came up with that exact same number - $400 per year. I computed it independently, so I must understand the results alright. I'll add, however, that this is just for prescription drugs, and it is only the short-term effect. The effect grows larger over time because it will affect the development of new drugs, not just the prices of drugs that are already on the market.

 

[sunmaster14]

http://www.nih.gov/about/impact/impact_global.pdf

This suggests that whilst the US might be paying more than it's share by size, it's not the only one and that the playing field is levelling.

This is not relevant to the discussion that I'm having at least. My argument is that the lucrative US health care market drives R&D. It doesn't matter where the R&D is happening, only why it is happening. If government price controls were instituted in the US, then there would be less private R&D spending, even in China.

 

[Francesca R]

If everybody is charged their expected cost, then it makes no difference if healthy people opt out of the system. You only need to force healthy people into the system if you are charging them an artificially high price so as to subsidize unhealthy people. If you are using taxes to subsidize unhealthy people (or poor people), then you don't need to force anybody into the system.

The adverse selection problem exists even then because--while it is not particularly difficult to price health spending risks--it is for any single individual/household because the insuree has more info than the insurer, even with quite detailed underwriting investigation as I said before. Insurance ends up being a bad deal for some, and bad quality/denied for others. From there the problem gets steadily bigger.

If what you propose worked well/better, then the US would already have had just-fine health insurance before Obamacare.

 

[Architect]

This is not relevant to the discussion that I'm having at least. My argument is that the lucrative US health care market drives R&D. It doesn't matter where the R&D is happening, only why it is happening. If government price controls were instituted in the US, then there would be less private R&D spending, even in China.

Evidence?

 

[DavidJames]

Evidence?

it says so in the Libertarian/Free Market bible.

 

[sunmaster14]

Evidence?

Oh, good *********** grief. Read the last few pages of this thread. I've provided the evidence, and the people who demanded it before have been too cowardly to return to the thread and comment on it. 3point14 for instance. And Randfan. And anduin.

 

[Architect]

No, you've made assertions. Evidence requires firm data.

 

[sunmaster14]

No, you've made assertions. Evidence requires firm data.

I have provided far more evidence and argument than anybody else has in this thread, or even in this forum that I've seen. Here is the full research paper. A relevant excerpt follows (my emphasis added):

The final parameter of quantitative relevance is the responsiveness of innovation to revenues. Estimates in the economics literature suggest that a 1 percent increase in pharmaceutical revenues leads to more than a 3 percent increase in the number of new molecules. We considered values ranging from zero to 5 percent. Because several papers in the literature suggest that innovators respond to revenues, zero seems a strict lower bound.

Here are the references that are cited in the paper to support the claim above (in bold):

A. Finkelstein, “Static and Dynamic Effects of Health Policy: Evidence from the Vaccine Industry,” Quarterly Journal of Economics 119, no. 2 (2004): 527–564;W. Yin, “Three Essays on the Economics of Health and Development” (Ph.D. thesis, PrincetonUniversity, 2005); and P.M.Danzon, Y.R.Wang, and L.Wang, “The Impact of Price Regulation on the Launch Delay of New Drugs—Evidence from Twenty-five Major Markets in the 1990s,” Health Economics 14, no. 3 (2005): 269–292.

D. Acemoglu and J. Linn, “Market Size in Innovation: Theory and Evidence from the Pharmaceutical Industry,” Quarterly Journal of Economics 119, no. 3 (2004): 1049–1090.

 

[Emily's Cat]

I don't agree. You can get insurance for all kinds of risks these days. Underwriting health insurance can't be any more complicated than underwriting life insurance, or product liability insurance, or airplane crash insurance, or flood insurance, or earthquake insurance, or insurance for valuables like art. Pianists have insured their hands, and singers have insured their vocal cords. Health care costs are actually pretty well understood, given all of the data that is available.

What are you basing this opinion on?

 

[sunmaster14]

What are you basing this opinion on?

Well, there's tons of data on the incidence rates of various diseases and ailments, based on certain risk factors, the protocols for treating them are well-known, and the costs for implementing those protocols are known in advance. The probability distribution for outcomes is also known. The only big unknown is the cost of treatments and procedures which have yet to be developed. But that doesn't change much year to year.

 

[DavidJames]

I have provided far more evidence and argument than anybody else has in this thread, or even in this forum that I've seen. Here is the full research paper. A relevant excerpt follows (my emphasis added):



Here are the references that are cited in the paper to support the claim above (in bold):

That's theory, eg, writings from a bible. Yes you provide arguments. Don't confuse arguments and theory with real world evidence.

 

[sunmaster14]

That's theory, eg, writings from a bible. Yes you provide arguments. Don't confuse arguments and theory with real world evidence.

Fine. But if you want to challenge the claim, you should come up with a plausible argument. The claim that drug innovation scales with revenue is actually based on empirical research, although I have not read those papers. One of the authors - Amy Finkelstein - is pretty well known.

 

[Emily's Cat]

I had always thought risk pooling was the insurance equivalent of portfolio diversification.

It's not quite the same, in that there aren't offsetting risks in terms of insurance.

When you're talking about portfolio diversification, you're relying on the fact that each individual investment has risk, but that the variance in their risks respond to different economic levers. When the cost of sugar cane is high, people buy more corn to compensate. The two crops grow in different climates, and are very rarely affected by the same weather phenomena. So the risk of having a bad cane year is offset by the corn investment. They move in different directions.

This doesn't really happen in health insurance. Risks don't really offset. There are risks of global things, like pandemics, but they affect everyone - there's nobody who gets healthier when there's a pandemic. There's some shifting of where a service occurs - for example, between ER and Urgent Care, or between Inpatient Hospital and Long Term Nursing.

But the inherent risks of the people involved don't offset each other. There's no group of people that get healthier when a different group gets sicker. So from that perspective, the diversification perspective really doesn't work at all.

Generally speaking, we pool risks because it makes them more predictable. Health insurance is a compound distribution. Both the frequency and the severity distributions are extremely long-tailed. The smaller the number of covered lives, the larger an impact the variance has to actual experience. When we pool a larger number of risks together, it might result in an overall higher mean (the expected cost that drives premium), but it reduces the variance.

It's also a bit of a value at risk problem, in that IF something on the tail occurs, it has a much larger impact on the experience of a small pool than a large pool.

Ok, well I don't think those systems work as well as a free market based system would, in which poor people are subsidized by the government with dollars rather than free health care services. And I think one of the reasons that universal health insurance seems to be working in other countries relative to the US is that those systems are benefiting from the health care largess of the US.

It really just depends on what you think constitutes "universal coverage". Do you consider Switzerland's schema to be universal coverage or free market?

 

[Emily's Cat]

I apologize for the snippety-snip here, but your post was very dense, and there's a lot to address.

Well, there's tons of data on the incidence rates of various diseases and ailments

There are lots of data about common diseases. There's not all that much data on orphan diseases, or on injuries - when you fall off a ladder and break your spine, there's really not much useful incidence info on that.

based on certain risk factors

The R-sqaured for individual-level predictive risk scoring is about 0.3. It might be slightly better for some specific illnesses, like diabetes... which sits at about 0.6. The risk factors increase likelihood a teensy bit... but the fit is really very poor.

Aggregate models do much better, with R-sqaureds ranging from about 0.6 to about 0.8. There's a lot of variance from model to model. Those models also ONLY predict common chronic conditions. They don't even try to estimate the incidence of hemophilia or enzyme deficiencies in the population.

the protocols for treating them are well-known

No, they really aren't. There is no "best practices" for medical treatment, and almost every condition has multiple possible treatments available. The cost variance for each treatment approach can be significant.

For example... several years ago I had surgery for a fibroid. I had three different treatment options that were all deemed reasonable by my doctor. They had drastically different costs, different risks, different potential complications, different effects on my fertility, and different recovery times. There was no "known" protocol for treating fibroids, even though they're a pretty common ailment for women, with incidence around 30% for women over age 35.

and the costs for implementing those protocols are known in advance.

No, the costs aren't known. Some of them might be estimable. But the actual cost depends on a whole host of things.

My original estimate for my surgery came in tens of thousands lower than the actual cost was. I ended up not responding to a drug therapy, and needed a blood tranfusion prior to surgery. During surgery I still needed blood, and there were delays due to the location of the fibroid.

And that was for a relatively simple surgery, with relatively few complications.

The probability distribution for outcomes is also known. The only big unknown is the cost of treatments and procedures which have yet to be developed. But that doesn't change much year to year.

You know, I don't think you really know as much about medicine as you think you do... What leads you to believe that any of this is true? Have you just assumed that it is the case?

 

[Francesca R]

It's not quite the same, in that there aren't offsetting risks in terms of insurance.

It is the same. There are not many negatively correlated equities either (although bonds are sometimes negatively correlated with them, and you can short them which makes it negative). But you don't need negatively correlated risks for diversification to work, in financial assets or in contingent liabilites; you just need correlations of less than +1.

Generally speaking, we pool risks because it makes them more predictable.

It is because pooling them makes them lower. Honestly! And lower volatility of liabilities is more predictable, yes, and that predictability is passed on to the buyer. There is a limit to how much the pooled risk can be lowered (because as you say there are few negative correlations), and that limit is the exact analogy of the systematic risk or "beta risk" of the global capital market. Only idiosyncratic risks (pertaining to individual people or individual financial assets) can be diversified away.

 

[Francesca R]

Do you consider Switzerland's schema to be universal coverage or free market?

It's compulsory, covers everybody, predominantly state funded, and excludes medical underwriting.

I don't think anybody can call it anything but universal. And certainly not a free market.

 

[Emily's Cat]

It is the same. There are not many negatively correlated equities either (although bonds are sometimes negatively correlated with them, and you can short them which makes it negative). But you don't need negatively correlated risks for diversification to work, in financial assets or in contingent liabilites; you just need correlations of less than +1.

It is because pooling them makes them lower. Honestly! And lower volatility of liabilities is more predictable, yes, and that predictability is passed on to the buyer. There is a limit to how much the pooled risk can be lowered (because as you say there are few negative correlations), and that limit is the exact analogy of the systematic risk or "beta risk" of the global capital market. Only idiosyncratic risks (pertaining to individual people or individual financial assets) can be diversified away.

I see what you're saying, but I don't think they're that much alike.

Except that in diversification, the whole point is that some of the variances are negatively correlated. Not perfectly, just negatively.

Diversification

The result from diversification is that the overall mean return is higher and the overall risk is lower than for the individual investments in the portfolio. It relies in part on the fact that the investments are not completely independent. Each investment is correlated with others in the portfolio. In fact, part of what gives it so much power is that at least some of them are negatively correlated.

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'm expressing this poorly. Adding more lives into the pool doesn't materially reduce the variance of the compound distribution... it just smooths out the tail.

 

[Emily's Cat]

It's compulsory, covers everybody, predominantly state funded, and excludes medical underwriting.

I don't think anybody can call it anything but universal. And certainly not a free market.

It defines a minimum benefit, is provided by private insurance carriers, those carriers are reimbursed for the experience related to that minimum coverage, and charge premiums for coverage above that level.

Do you consider Medicare Advantage products to be a free market product?
If Medicare Advantage were compulsory instead of Original Fee-for-Service Medicare, would you consider to be something other than free-market?
Or is it mixed?