Aug. 25th, 2013

pktechgirlbackup: (pktechgirl)
The health care debate has gotten very, very muddled. Most of my reading on the subject is by progressives (writing either for other progressives or for a general audience as an attempt at conversion), or by libertarians for libertarians. There is a pattern I have noticed that I think is worth addressing: progressives tend to see either One Big Healthcare Issue, or lots of issues with the same solution. Libertarians (as personified by me) tend to see hundreds of small issues that are going to need individual solutions. Which is interesting, because it's the reverse of the usual pattern, in which I think everything could be solved by introducing competition and progressives think we need an expert to tailor a solution to the specific problem.

So here are some of the individual problems I see, and short versions of my preferred solutions.

Problem: People can't afford health insurance/health care.

Give them money. To be fair, this is my solution for all problems relating to poverty. It sounds glib, but I'm being sincere. If we want to place a floor on the consumption of adults, we should just do that. If people are not using the money on the things we want them to, we should reevaluate our wishes or their ability to take care of themselves. Alas, this gets considerably more complicated if children are involved.


Problem: Health insurance is too expensive/increasingly expensive.

Insurance profits are reasonable and have not been rising in recent years. What you are seeing is a rise in the price of health care itself.


Problem: Health care is too expensive/increasingly expensive.

This is sometimes called "health care inflation", which is misleading, because much of the increase comes from the introduction of new things to buy. It's like saying we've had smart phone inflation since 2000. You could make the problem go away by not buying new things.

Some of it is that we're consuming more things. I am maybe the only person who liked Obamacare more after she heard about the death panels, because the alternative was everyone deciding for themselves how much of other people's money they wanted to spend on their own health care. Also, it's still not inflation in the technical sense of the word, any more than we experience produce inflation when I double my banana consumption.

However, some of it is genuine inflation , brought on by Baumol's cost disease (short version: if you have two sectors, and one gets more labor efficient, the other one experiences inflation). There are three solutions: pay workers less for the same job (which every politician ever has promised not to do, because health care professionals are sympathetic, have money, and vote), shift work to lower-wage workers (which we are trying, via things like nurse practitioners and physician's assistants. I am in favor of this but think it bears monitoring before we declare it a total success), or invent labor saving devices (which we have actually done some good at, but since in many cases the replacements are themselves expensive, we won't see returns until they go off patent). I am totally in favor of investment, both public and private, in labor saving devices. But we can't do that by wishing really hard, we have to work with the market we have and adjust if it improves later.

Problem: Wait a minute, you skipped over a possible solution. We could just pay less for the new things.

Okay, technically that's true, but it's a one time savings. Profits are what motivate pharmaceutical companies to make new things that let me live longer. I think our patent system actually does an excellent job here, providing some time to make money and then providing the drugs nearly at cost for all eternity (I have no opinion on the proper length of that first period).


Problem: why don't we just cut out the middle man and do it ourselves?

That is a good idea. There are definitely inefficiencies in the current model. But how do we do that? University research isn't directly translatable into usable drugs. There are many possible improvements to this model, and I am fully in favor of government funding for them as pilot projects, but I would like to see them work before we shoot the old model in the back of the head.

Problem: Even if the increase in insurance costs are due to care costs, couldn't we get a one time boost by removing their inefficiencies?

Yeah, maybe. There certainly are inefficiencies introduced by having 40 different billing systems. But considering that the government's systems are the most complicated and time consuming of all, I don't see them fixing it.

Problem: Health insurance is linked to employment.

Yes, this is deeply stupid. There are two causes: one is that the government subsidizes employer-provided insurance by not taxing it as income. This is stupid and they need to stop doing it.

The second is that employment is a way to get around adverse selection and the resulting insurance death spiral. This one is harder to solve. We probably can't fix the information asymmetry, which leaves mandating everyone purchase insurance. There are two problems with this: done by the federal government, it is either unconstitutional or stretches the commerce clause to the point that it can do anything. State governments can do it constitutionally, but still risk what happened in Massachusetts: originally advertised as mandating only catastrophic (i.e. true) insurance, the bill bloated until it mandated all kinds of care (because lobbyists bought the clauses), raising the premiums considerably. (My impression is that the conclusion progressives draw from this is lobbyists and/or money has too much power over politicians. The conclusion I draw is that we should minimize the number of things government can mandate so that there's no point in bribing politicians).

Note that Obamacare introduced mandatory issue (insurers must insure everyone that asks) and community rating (insurers may not charge more for riskier people) years before mandating purchase (everyone *must* purchase insurance). This is the worst possible thing.

So it's a stupid system, but it will be very hard to transition to something else.

Problem: what was that snarking about "true" insurance?

Insurance means insulating you from rare, extremely costly, unpredictable events, like car accidents and cancer. What gets called insurance today is often just 3rd parties paying for routine care, like vaccines and annual check ups.

Kathleen Sbelius, United States Secretary of Health and Human Services, got this exactly backwards .

Problem: but preventative care lowers the costs for the insurance company, so it still makes sense for them to pay for it.

This is just not true. Very little preventative care lowers the total cost of care over someone's life, because we just keep treating you until you die. Everyone dies of something, and the causes get more interesting and expensive the older you get. Even if the cost was exactly the same, you've been alive for longer and sucked up more money for annual care. The only things proven to actually lower costs are vaccines, dental care, and pre-natal care. You can argue about preventative care being a more cost effective way to raise life span or Quality Adjusted Life Years, but that is a different issue. To the extent that it is true, the gains don't accrue to the insurance company. This doesn't mean preventative care isn't an excellent thing, but excellence and money saving are not synonymous

Problem: The same care costs a lot more when you pay out of pocket than when insurance covers it, and with a lot more uncertainty and anxiety too.

This is a big problem. I don't know how to solve it and am interested in hearing ideas.

This is all very hard to argue at cocktail parties. The other side has some very good visuals and easy to explain solutions. I have a lot of graphs and arguments about highly distributed long term consequences.

Profile

pktechgirlbackup: (Default)
pktechgirlbackup

May 2014

S M T W T F S
    123
45 678910
11121314151617
18192021222324
25262728293031

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Sep. 21st, 2017 01:52 pm
Powered by Dreamwidth Studios