It's all about healthcare. Well, and signaling.

This afternoon I was sitting in the third row, right behind the woman who placed the winning bid -- $13,000 -- on the auction of Harry Potter's bow tie. Well, of J. Pierpont Finch's bow tie, Finch being the lead character of How to Succeed in Business Without Really Trying; Finch is played by Daniel Radcliffe, a talented young actor/singer/dancer (well, a bit weaker as a dancer, I thought, but I'm no judge) who will probably be plagued for the rest of his life by people who can't help but think of him as Harry Potter. And he and his co-star were trying to raise money for a Good Cause, namely healthcare, after the performance. (Okay, a specific healthcare cause, but I'm still fussing about healthcare in general.) So....

So I would call the winning bid rather impressive; I think most people would. Radcliffe commented that we were "well over the record", some time before the auction closed---I presume he does this with a fresh bow tie for each performance. But I couldn't help but be reminded of the cost of my own appendectomy, not quite a year ago: it was in fact a little over the bow tie's price. Consider what that audience pays for healthcare each year...the auction was a Good Thing to Do, an Exercise in Nobility, a demonstration of the Brotherhood of Man (that being the last song&dance) but as a contribution to healthcare it was a teaspoonful in a lake.

Yes, I understand that's not the point. It's not really about healthcare, even healthcare is not really about healthcare. Robin Hanson put it rather well, some time ago, in his argument that it's about Showing That You Care:

I can explain these puzzles moderately well by assuming that humans evolved deep medical habits long ago in an environment where people gained higher status by having more allies, honestly cared about those who remained allies, were unsure ... These ancient habits would induce modern humans to treat medical care as a way to show that you care. Medical care provided by our allies would reassure us of their concern, and allies would want you and other allies to see that they had pay enough to distinguish themselves from posers who didn’t care as much as they.

That makes sense to me as the beginning of a model, and it certainly isn't a criticism of the woman who paid so much for the bow tie. She evidently does care, and presumably cares that it's evident that she cares, and that's a good thing.

Nonetheless, if you want to use healthcare provision to show you care, I think it would be a good idea to spend some time looking for actual ways to provide actual healthcare; a few more teaspoons of water in the lake won't do it. So I'd like to go back over my proposal of a bit more than a year ago. I'd organize it a bit differently now, but I don't seem to have moved all that far.

If I were (heaven forfend) In Charge, I would crowd-source as much as possible of the decision-making by pushing it into a market, with participants being given as much data for decision-making as possible, and being simultaneously milked for as much data as possible. I want incentives for innovation, to reduce the death-rate for billions yet unborn; I also want incentives for good performance now, not for the sort of regulatory capture our current system maximizes. Specifically I would:

1. Allow unlicensed health care, wherever it's clearly labeled as such; it won't get public support but people can choose to spend their money on it. The argument against this is apparently that people will make bad choices. Yeah, some will, probably including me and you. So? I've never understood the way some people believe that they (or those they select) can make good choices for others; in fact I'm moderately cynical about licensure requirements as they are now structured, whether for medics or morticians or cosmetologists.
2. Require transparent pricing, uniform no-bargaining pricing, from all providers of licensed health care. (The services producing my appendectomy really don't do this.)
3. Require that "licensing" be independent of geography; if the best/cheapest supplier of a particular treatment is two states over or on another continent, that's fine. As I've said before, I believe that telepresence medicine can enable the specialization and trade that has made markets work in other contexts since before Adam Smith wrote about it, so I expect this as the usual case, not an exception.
4. Take away the employer-based tax exemption; health care shouldn't be an employment issue.
5. Add a universal tax-funded "insurance" policy (insulation, actually): if your expenditures for "proven procedures" from licensed health care providers exceed the overall 16% (of GDP) average, then the taxpayers contribute some. Maybe if your cost is 30% of your income, then the taxpayers kick in (30-16)/2=7%, half of the overage, and the maximum you can pay is 50% of your income whether that's $0/year or $10M/year. Is that too generous? Not enough? I dunno. The point is to combine protection from catastrophe (but not from serious pain) with making sure that market prices are set by people or groups who are actually bargaining in that market, i.e. the better-off people for whom procedure X will not be covered. I want to do that combination with some simple, less-than-perfect-but-better-than-nothing rule with which I can trust a government. (Democrats and Republicans trust government on different things; just figure you want a better-than-nothing rule with which you'd trust a politician of the party you despise.)
6. If you want "unproven procedures" and you can pay for them, that's fine too; the licensed health care providers should have a strong motive to come up with new stuff and document/publish that it works. The FDA should not be able to keep you from paying for these likely-to-fail treatments, but it should keep you from charging it to the rest of us. If procedure X has no accepted studies supporting it, then it's up to you to pay for it.
7. Whatever additional insurance/insulation you want to buy for proven or unproven procedures is just fine, and can be bought across state lines. It's your problem. You want to save your money in a special bank account? Feel free.
8. Any care that has been paid or partly paid by public funds goes into an anonymized public database, so that we learn more about which treatments have what effects on which conditions. Organizations promoting not-yet-approved treatments will be encouraged to contribute data.

And that's really it, for me. I'm even less confident of this than I was when I wrote the first version, but I still don't see anything else I like as well. I think that the market I'm describing would probably evolve rather quickly into a market in which people choose and buy packaged health plans from "insurance" agents, and web sites build up crowd-sourced ratings of those health plans; there would be quite a bit of overlap with the better parts of what we have now. I hope. And I care, and I suppose I'd like to signal that I care.

Or then again, maybe not.

update:I never actually mentioned that this post was prompted by thinking about Mark Thoma's Economist's View: Discussion Question: How Can We Reduce the Growth of Health Care Costs?

there is far too much discussion of cutting services, and not enough about how to control costs without affecting services (e.g., using the government's purchasing power to reduce the amount the government pays for drugs, reducing the cost of insurance companies fighting over who pays bills, etc.)

You see, I doubt the premise: if you use government power as I believe Thoma wants, you are increasing the incentives for regulatory capture, crony capitalism, rent-seeking... you are putting yourself on a path where you have signaled your concern but healthcare is not what you're rewarding. Of course government power needs to be used -- to collect the money for treatments which research results say are crucial and which markets say are expensive. And government power needs to be used to maintain a context for innovation (rather than squelch it, as I believe our recent trends in "intellectual property" law tend to do.) But if bargaining-on-prices-with-the-government is the multi-billion$ activity you focus on, then that's what companies will have to invest in. That's a bad bad bad bad thing. I commented here.

Well, it's Easter morning. Maybe we're all saved?

Or then again, maybe not.

One Hundred Trillion Dollars

Q: Was the health care bill better or worse than doing nothing? What do you think?

A: Worse. It does some good and some harm, but I think the harm predominates. If we're very lucky, maybe it won't predominate by much.

Q: What good does it do, and what harm?

A: Well, just after the bill passed in March I wrote a long rumination-with-links, including links back to a February rumination-with-links about what is and isn't wrong with our current system and a January rumination-with-links about what I'd really like, and what worries me, and all that---

Q: No. I don't want a long rumination-with-links, and certainly not three of them. Can't you summarize? What good will the bill do, and what harm?

A: I'll try.

• The bill helps more people get health care services, at other people's expense; that, of course, is what the bill is for, and I expect that several million people will be happier for a while because they don't have to worry as much about price, they can just ask and it shall be given. Yay!
• The direct bad part then splits into two:
  • the people receiving services will worry less about price, which will make prices go up even faster than they have been. (The bill says there will be people authorized to save money, but they're not authorized to say "no" which is the only way they could save money.) And then,
  • the people providing the money will no longer have the money that has been taken from them.

Q: So it saves lives and costs money. Why is that a bad trade-off?

A: Actually, no, I don't expect it to save lives in the long run. Even in the short run, the evidence that expanded health-care coverage saves lives on net, at the current margin, is astonishingly weak. (Making people happier is important, but it's not the same as saving lives.) I hope our age-adjusted mortality rates will continue declining under the new system, but they will not drop as fast as they would have even under the old system because the incentives and resources for innovation are reduced. In the longer run the new system will cost lives, partly because it will cost money and money fuels innovation which saves lives, and partly because it is an addition to a collection of promises that will not be kept -- cannot be kept.

Q: What promises, and why can't they be kept?

A: Did you notice the title of this blog post? "One Hundred Trillion Dollars".

Q: You can't mean that's the cost of the bill -- can you?

A: Nope. That's the context of the bill. That's what we already owed -- more precisely, it's what we had already promised to pay apart from what we "owed."

Q: Huh?

A: We have a GDP, sort of a sum-of-all-incomes, of about 14 trillion dollars each year, which grows at a few percent each year (partly from population growth, partly from productivity growth.) We have an overall or "gross" debt of almost 13 trillion dollars. That includes government borrowing from Social Security payments and things like that; without those it's getting close to 9 trillion or so. That's a lot, but it's easy to imagine growing out of those debts, unless we spend so much money now that growth is slowed down even more than predicted. But...well, here's the one real link of this post, the Word from the Dallas Fed two years ago:

Add together the unfunded liabilities from Medicare and Social Security, and it comes to $99.2 trillion over the infinite horizon. Traditional Medicare composes about 69 percent, the new drug benefit roughly 17 percent and Social Security the remaining 14 percent. ... all we would have to do to fully fund our nation’s entitlement programs would be to cut discretionary spending by 97 percent....defense and national security, education, the environment ... All of them [forever]... No combination of tax hikes and spending cuts, though, will change the total burden borne by current and future generations. For the existing unfunded liabilities to be covered in the end, someone must pay $99.2 trillion more or receive $99.2 trillion less than they have been currently promised. This is a cold, hard fact. The decision we must make is whether to shoulder a substantial portion of that burden today or compel future generations to bear its full weight.

These are not "debts", and they are particularly not part of the federal debt. They are merely, ah, how shall I say this? They are merely "legally binding promises to pay".

Now, think about the health care bill in that context. We came up to that decision about what to do with the unmanageable burden of entitlements, and we decided, umm, to expand the entitlements. We, mainly my own Baby Boomer generation, have promised each other that our kids will pay for the care that we wish we deserved. These promises are actually expanding promises: if improved productivity raises wages then the Social Security promise expands, while if improved medical tech provides more expensive ways to treat whatever ails us then the Medicare promise expands. So far as I can see, these promises will not be kept. If we're very lucky, improved technology will let us say "we're not keeping the promises, but you're very well off anyway." On the other hand, extra promises like this bill make it likely that the breakage will start before technology can improve that much.

Q: But I thought that the CBO said the health care bill actually reduces the deficit. Doesn't it?

A: In a way -- but not in a way that helps much, even if the assumptions turn out right. The bill is "paid for" in the sense that Congress has named funding sources for its parts. Medicare payments are to be cut -- well, Congress has made that promise before, and broken it repeatedly: maybe they mean it this time. If they don't, it's bad. If they do, we are not only using up the flexibility we needed to pay for the promises we already made, but also discouraging doctors (some of whom already turn away Medicare patients as money-losers. And discouraging doctors of course tends to raise prices.) It's bad either way. Corporations will have to pay, which reduces employment (and growth). States have to take on extra Medicaid expenses; that's not part of the federal deficit scoring, and the states can't afford their existing entitlement programs, but so it goes. My kids will have to pay premiums adjusted to subsidize my premiums (realistic age-based premium ratios are not allowed.) There will be extra Social Security taxes, which help pay for this bill but by existing law they add to the unfunded Social Security liability. It's a mess, but it's a mess which has been carefully set up so that if you accept the assumptions then the officially budgeted part of the federal deficit is not expanded. Individual costs rise, corporate costs rise, state costs rise and unfunded liabilities rise -- we're using resources which we need. That cuts growth (and innovation.)

Q: But you're always saying that exponentially improving technology is going to make us all rich, immortal, smart -- all that good stuff. Are you taking that back?

A: Not yet, but this bill makes me less confident of it. When you look at good trends and bad trends, it always seems like a race to see whether we end up in Heaven or Hell. Here we have trends towards a productivity-spike with health and wealth and longevity for all, and there we have trends towards a financial collapse which leads to a collapse of the health system, environmental everything, etc. This bill slows what I want to speed, speeds what I want to slow. But maybe in the end it won't hurt too much.

Q: I'm still having a hard time with the doesn't-save-lives bit. Are you sure of that?

A: It's very odd and surprising, but the evidence that increases in healthcare delivery lead to increased lifespans at the margin is extremely weak. Remember that emergency care is already available, that poor people are mostly entitled to Medicaid, and that the main lifespan issues are not healthcare anyway: there's a lifestyle battle between declining smoking rates and rising obesity. The bill will certainly extend some lives and shorten others -- about 100,000 people per year die from medical errors as it is. Intuition tells me that it's gotta be a net positive at first, but I'm not sure. I am reasonably confident that it will cost lives in the end, and it adds a bit to the chance that the U.S. will eventually play Greece, with nobody to bail us out.

Or then again, maybe not. I hope not, anyway. There are times when I'm happy to think how often I've been wrong, and this is one of them.

Health Care, Insurance, Insulation, Innovation

My primary concern about reform proposals is that I want innovation, profit-fueled innovation which bankrupts current stakeholders because they're making buggy whips, except when they move fast and make painful choices. I want more innovation than all the innovation we've seen so far. I want robotic nurses to take care of me. I want stem cells to fix my aging joints. I want painless dentistry. I want tiny robots inside my body -- among other things, to deliver the stem cells. Frankly, I want to over-consume health care like most of my fellow-Americans. The more government regulation handles things, the more the people with seats at the table will be guaranteed their "fair" share; that inhibits creative destruction. It also inhibits me from getting the health care insurance I want; I live in New York State, which as Reason Magazine says is Exhibit A in the case against programs like our proposed health care insurance "reform".

Okay, I also have secondary and tertiary concerns, mostly about politics and regulatory capture and crony capitalism and politics, and I'm not sure about any of this, and I keep meaning to note down all my current notions to see which if any survive the year. But mainly, it's about innovation, and if something resembling the current plans gets through then I agree with Megan McArdle that

We've just increased substantially the supply of unrepealable, unsustainable entitlements. We've also, in my opinion, put ourselves on a road that leads eventually to less healthcare innovation, less healthcare improvement, and more dead people in the long run.

(I don't think she actually means "more dead people", but rather "higher age-adjusted mortality rates".)

I like Will Wilkinson's analysis in Will Health-Care Innovation Survive Obamacare?, where I tend to think the answer is "yes, but at a seriously diminished rate." Or maybe not.

Provisionally, I'll just note down what I would now support, until the minute that I actually realize how silly these provisions are:

1. Require transparent pricing, uniform no-bargaining pricing, from all licensed health care providers.
2. Allow unlicensed health care, wherever it's clearly labeled as such; it won't get public support but people can choose it.
3. Take away the employer-based tax exemption; health care shouldn't be an employment issue.
4. Add a universal tax-funded "insurance" policy: if your expenditures for "proven procedures" from licensed health care providers exceed the overall 16% (of GDP) average, then the taxpayers contribute. In particular if your cost is 16--32%, say 30%, then the taxpayers kick in (30-16)/2=7%, half of the overage, and the maximum you can pay is 24% of your income whether that's $0/year or $10M/year. Or something of that general magnitude; I wouldn't fuss about adjustments to these figures. The point is to combine protection from catastrophe (but not necessarily from financial pain) with making sure that market prices are set by people or groups who are actually bargaining in that market, i.e. the better-off people for whom procedure X will not be covered.
5. If you want "unproven procedures" and you can pay for them, that's fine too; the licensed health care providers should have a strong motive to come up with new stuff and document/publish that it works. The FDA should not be able to keep you from paying for weird medicines, but it should keep you from charging it to the rest of us. If procedure X has no accepted studies supporting it, then it's up to you to pay for it.
6. Whatever additional insurance you want to buy for proven or unproven procedures is just fine, and can be bought across state lines. It's your problem. You want to save your money in a special bank account? Feel free.
7. Any care that has been paid or partly paid by public funds goes into an anonymized public database. (Yes, that's hard, but I am more willing to risk failures in privacy than failures to learn about what does and doesn't work.)

And that's my plan, as of today. I'm sure it would be perfect.

Or then again, maybe not.