« Do cellphones have a role in education? | Main | Demand curves slope downward »
Tyler Cowen pretends he is a Democrat
If I were a Democrat...
First, I would not cite evidence about how Western European countries spend less on health and are healthier than U.S. citizens. This data set, if you take it seriously, also implies that the marginal product of more health care, adjusting for income and a few other variables, is zero. Expanding health care would not be important. Now I believe this is an incorrect conclusion, but that is what shows up in this data. We should not invoke this data selectively.
Second, I would recognize that American policy generally works (or doesn't work) by building upon existing institutions. The most likely form of national health care -- for better or worse -- would extend a version of Medicare to more people. This would not lower health care costs, whether in gross or quality-adjusted terms. Keep in mind that negotiating price reductions does not per se lower real resource costs at all.
I would disaggregate health care systems and see where we could do the most good:
1. Step up R&D subsidies through the NIH and our university system, both high quality institutions. Their autonomy and micro-fiefdoms provide a good framework for risk-taking and innovation. The returns to medical R&D are extremely high. Furthermore the case for market failure, based on the inability to capture the full social gains from a new idea, is simple.
2. Redo the Medicare drug bill so that people can understand it (even I can't, nor does my mother), and so more people benefit. If need be, we can do this in budget-neutral fashion. The Bush plan is a mess.
3. Invest in local public health systems. Preventive care is important, especially for the poor. Price can be an obstacle but often the relevant constraints are behavioral in nature. Public health care systems should be easy and inviting, and they have to become part of life routines. Government can be part of the solution. Strong local public health care also will improve surveillance and later surge capacity if a pandemic comes along; this added benefit is significant.
4. Borrow a page from the libertarian litany about the FDA.
5. Institute prizes for successful vaccines. We have been discouraging vaccine production when we should be encouraging it; Michael Kremer has some intriguing proposals.
All those options are doable. All would save lives. None are fiscal disasters. They offer something for both rich and poor. They lay out a positive and constructive role for government, while keeping room for the private sector. None raise the prospect of excess bureaucracy or discourage innovation. None rest on the questionable belief that government as single supplier or payer would improve efficiency. And they are all areas where the Republicans are dropping the ball.
I would cut talk of national health insurance. I would cease obsessing over the number of "40 million uninsured," however good a debating point it may be. Many of these people are either linked to immigration or get decent medical coverage in any case. I would admit that we cannot take care of everyone and that we face tough trade-offs.
Hmmm...these counterfactuals are fun. What should I try next? Pretending I am a Republican? But for now, it is back to normal life...and so we return to your regularly scheduled programming. But comments are open, in case Kevin Drum's readers wish to pretend they are libertarians...
Posted by Tyler Cowen on November 11, 2005 at 06:21 AM in Medicine | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c66b253ef00e5509794758834
Listed below are links to weblogs that reference Tyler Cowen pretends he is a Democrat:
» Healthcare and Life Expectancy from Political Animal
HEALTHCARE AND LIFE EXPECTANCY....Should we have a national healthcare plan? Tyler Cowen says no, but in the course of his argument suggests that increased spending on healthcare has no net effect on actual health. Matt Yglesias made the same claim... [Read More]
Tracked on Nov 11, 2005 1:22:23 PM
» Tyler Cowen's health-care proposals from Freedom Democrats
"Marginal Revolution" is one of my favorite blogs, presenting an economist's perspective on all types of issues. I especially like Tyler Cowen's posts because he seems to share many "Free Democrat" values -- a liberal drive to help the least fortunate mem [Read More]
Tracked on Nov 13, 2005 7:46:58 PM
Comments
Fascinating. Why, do you think, have Republicans not chewed on these sorts of ideas (I am a rather reluctant Democrat myself and concede the intellectual bankruptcy of the party at this time)? Has incumbency turned them into overcautious softies? Have the people on Capitol Hill and in the think tanks, whose job it is to bat around ideas, turned over too much of the thinking to the White House? Are insurance companies, drug companies, etc, who give so much money to campaigns, too invested in the current system? I would think the political appeal of a national candidate/party that could offer some new (and relatively inexpensive) ideas on health care reform would be enormous and irresistible.
Posted by: Linda McIntyre at Nov 11, 2005 8:21:04 AM
Yes, indeed, these counterfactuals are fun. And, if taken seriously (as here), they could also be highly constructive. If this was a standard procedure in the public debate I predict than the delineation between positive and normative arguments would become more clear to a large part of the public, and perhaps also to the debaters themselves.
Posted by: Johan at Nov 11, 2005 8:22:24 AM
TC writes:
"First, I would not cite evidence about how Western European countries spend less on health and are healthier than U.S. citizens. This data set, if you take it seriously, also implies that the marginal product of more health care, adjusting for income and a few other variables, is zero."
1) The question is not the marginal product of "more health care" but the marginal product of more spending in the health care system, however it happens to be constituted. It should not be surprising if more spending on administration has zero health care value.
2) Stepping past that distinction, it is clear that the marginal product of more health care differs based on, among other things, how much a person is already spending on health care. The marginal value of $1000 in health care to someone who currently has no resources to spend on health care is quite large. The marginal value of $1000 in additional health care to someone who is already consuming lots of health care resources may be near zero.
Posted by: alkali at Nov 11, 2005 8:37:31 AM
One further thought -- TC also writes:
"I would admit that we cannot take care of everyone and that we face tough trade-offs."
Why can't we? Other industrialized countries do it. We'd have to raise taxes by a nontrivial amount, to be sure, but we certainly could do it if we wanted to. You don't get points for intellectual honesty by ruling some policy options out of bounds a priori without explaining why.
Posted by: alkali at Nov 11, 2005 8:45:56 AM
Sorry. We true liberals and compassionate progressives can't pretned to be libertarian and Evil. We believe in government control for the greater good, not people doing their own thing (which is Evil).
Posted by: Drumfan at Nov 11, 2005 9:09:43 AM
...and there are those of us that do not want the government to control every aspect of our lives, under guise of the greater good (which can also be evil).
Posted by: Mcwop at Nov 11, 2005 10:01:52 AM
Well, I'm glad you aren't a Democrat. And I'm REALLY glad you weren't a Democrat using bad economic theory to kill the New Deal.
1) is independantly a good idea, but it is not itself a cure for health care problems anymore than spending money on Pharma r&d was our prescription drug benefit plan.
2) The medicare drug bill needs to be scrapped and rewritten, but, once again, you're thinking too small and only slightly on topic. When liberals scream universal health care, hint, hint, they're talking about the UNINSURED.
3) Investing in local public health systems is a dubious idea. First of all, its not a permanent idea. If you invest in public health systems, Republicans will use their first opportunity to uninvest in them at no cost. Second of all, its once again throwing a pebble at an invading army in terms of adequate response. Third of all, it won't work for both federalist reasons (the federal government is bad at directing local institutions, and local institutions are bad at using grant money effectively when it is bestowed upon them without asking). Fourth of all, the amount of investment needed to make these things viable alternatives to private medicine is going to be significantly less politically feasible than UHC. Fifth of all, the amount of investment needed to make these things viable alterantives to private medicine is bad policy given we could just spend the money copying the very effective policies of (non-British) Western Europe (and Canada, too!).
4) is silly and not worth discussing. Kowtowing to the GOP based on an underlying theory of government that we don't believe in is not serious when it comes from Lieberman. When it comes from a REPUBLICAN (oh wait, a "liberterian"), its a joke, right?
5) "Institute prizes for successful vaccines". Vaccinations are great, and we don't need to resort to drugs to direct NHS and University money towards this goal. Except for AIDS and cancer, however, vaccinations aren't going to do much to solve the uninsured/cost problem, particularly given that the uninsured aren't going to GET these vaccinations in the first place . . . once again, a pebble at a rock problem.
"I would admit that we cannot take care of everyone and that we face tough trade-offs."
NO. WE. DO. NOT. YES. WE. CAN.
Don't you get it? Economics, as a predictive science, is a huuuuuuuuuuuuuuuuge failure. It always has been. This is hardly surprising, since the very basis of economics is based on flawed premises about human behavior and "utility". That being said, Canada and Western Europe's systems didn't spring out of nowhere, and they're not exactly mystical concepts that can't be replicated in the United States. We can insure everyone, get better results, and CUT TAXES DOING IT (sure, there will be a premium paid to the government, but this premium will be smaller than the premiums people currently pay, giving them more "income". Overall costs going to the health care delivery system should drop by a third, which, as you know, is a TON OF MONEY). I thought liberterians were AGAINST TAXES.
And in the previous post you point out that Canada, North Korea, and Cuba have the world's only "single payer" system. While this ignores the ease one can transfer to single payer and then, through cost cutting measures, to a French or Japanese system, it might also be useful to point out that CUBA HAS A SYSTEM WITH RESULTS COMPARABLE TO OURS. Which isn't really all that helpful to my argument. Until you realize THEY SPEND LESS THAN ONE FIFTH OF THE COST, AND LESS THAN HALF THE COST IN GOVERNMENT EXPENDITURES ALONE.
That this is "not politically feasible" is:
1) Because of people like you who should be ashamed of yourself and know better, who enable the lobbyists and the red-baiters in order to preserve their preconceived notions about classical economics.
2) No more true than the idea that gay marriage was infeasible in 1980, or that the elimination of the estate tax was infeasible in 1970, or that social security, medicare, and medicaid were infeasible in 1925. Political reality is not in stasis, and the Democrats (who are no longer the majority party) will not succeed anyway unless they can use persuasion to get the population to move with them, rather than "adjust" to form majority coalitions that no longer exist for them anyway.
Posted by: Justin at Nov 11, 2005 10:07:36 AM
I am confused by the first (multi-sentence) paragraph, and do not currently have the time to read the link from Robin Hanson (although I fail to see how the title is related as well). There is an extremely facile explanation that I can think of, that feels implied, but is so intellectually dishonest that I dare not think it is what you mean. That explanation is that since there is an inverse relationship between other countries spending and outcome there is an implication that less spending = better outcomes. I am sure that is not what you mena, but it creeps out of the text, as if simply dismissing the structural differences out of hand is warranted.
I very much like the idea of prizes for vaccines, but think it should be taken a bit farther as a way to reduce patent monopolies - (John Quiggin and Dean Baker have more on this) - many of the other ideas are good as well.
I would second some of alkali's comments as well.
Another point of contention is that the comment " American policy generally works (or doesn't work) by building upon existing institutions" is generally true, but the exceptions are significant, and represent the changes that we have made in the past that are comparable to universal health care, and that once these are established they are preferred (Come on, really put on your Democrat hat, you like Social Security!!!)
Posted by: theCoach at Nov 11, 2005 10:08:31 AM
Hmm...
Suppose I buy all my groceries at a convenience store, and pay $100 a week for them. My neighbor buys the same bundle of groceries every week at a supermarket, and pays only $80 for them.
Would this data set imply that the marginal product of my spending on groceries is zero?
Maybe in some hyper-technical sense, but not in any meaningful way. It implies that we're spending at different margins. I'm spending in a wasteful way, she's spending in a thrifty way.
That's the situation that advocates of universal health coverage see the U.S. in, when compared to some European nations.
BTW, I think all five of your ideas are good ones. But I still think universal health coverage is a good idea, if done right. And rather than take the pessimistic public-choice theory perspective that "it can't be done right," I'm optimistic that we can look at what other countries have done right and wrong, and figure out something that will work for us.
Posted by: Brock at Nov 11, 2005 10:11:08 AM
OT, Is this the open comment exception that proves the rule?
Posted by: theCoach at Nov 11, 2005 10:18:28 AM
Good post. Disagree with the following:
"[evidence about how Western European countries spend less on health and are healthier than U.S. citizens]...also implies that the marginal product of more health care, adjusting for income and a few other variables, is zero."
Not really. Do you suppose we're allocating our health care dollars the same way? $1M on cheap, simple preventive care may go farther on average than $1M on bleeding-edge cancer treatments. The anti-US health care crowd seems to argue that we're spending our health care money in a non-optimal way.
There are many downsides to the American practice of linking health care to jobs. One of them is that your medical insurer has little incentive to make sure you're healthy in five years. For example, there've been many claims (which I do not know enough to evaluate) that veteran's health care in the US provides far more for far less.
Posted by: gundryggia at Nov 11, 2005 10:21:59 AM
I would admit that we cannot take care of everyone and that we face tough trade-offs.
I'm really unclear about the above. Advocates of national health insurance want precisely to take care of everyone, and if we really can't do it, they've wasted a lot of time. Some of them are pretty smart, so I'd be surprised if the evidence makes obvious that this can't be done. Also, I'm not sure why Western Europe isn't evidence that it can be done.
Posted by: SomeCallMeTim at Nov 11, 2005 10:29:42 AM
"it might also be useful to point out that CUBA HAS A SYSTEM WITH RESULTS COMPARABLE TO OURS. Which isn't really all that helpful to my argument. Until you realize THEY SPEND LESS THAN ONE FIFTH OF THE COST, AND LESS THAN HALF THE COST IN GOVERNMENT EXPENDITURES ALONE."
What's actually useful to point out is that CUBA CLAIMS TO HAVE A SYSTEM WITH RESULTS COMPARABLE TO OURS. Cuba, like all Communist dictatorships before them, has long been in the habit of claiming all sorts of things without much regard for such borgeois Western concepts as "truth", "fact", or "accuracy". The Soviets, as I recall, had similar impressive claims regarding the health and well-being of their people, the state of their armed forces, and so on that turned out not to stand up to any sort of scrutiny.
"There are many downsides to the American practice of linking health care to jobs. "
I agree. We need to stop doing that. We don't need national health care in order to stop doing that. The alternative to people having their bosses buy their health insurance isn't limited to people having the government buy their health insurance. They could buy it themselves. Granted, having people buy things for themselves has only been tried in hundreds of other industries for a handful of centuries, but I think it'll work.
Posted by: Ken at Nov 11, 2005 10:33:32 AM
Tyler, why do you deny that the marginal product of health care spending is zero? It seems to be backed by many different strands of evidence.
Why dismiss the possibility of simply using some other country's system, a system which has already been tested and which works? If I was a Democrat I would think that the returns to medical R&D are high, but I'm a contrarian, so I doubt the conventional wisdom here.
I agree "that increases in life expectancy from reductions in mortality due to cardiovascular disease over 1970-1990 has been worth over $30 trillion dollars - yes, 30 trillion dollars (for this research see: book, papers, summary). A conservative estimate is that 1/3rd of these improvements in life expectancy were due to better medical technology." but I have experience in biotech, and in both academic and government research. My strong suspicion is that marginal dollars spent on medical research are as useless as marginal dollars spent on healthcare. What I would favor is the establishment of prizes for medical research Outcomes. At $10,000,000,000 the Methusaleh prize would be grossly under-funded. Smaller prizes relating to the specific "seven deadly sins of aging" and for work in other animal models would also be appropriate. The wonderful thing about prizes is that if they don't work they are free to the taxpayer. By contrast, as a Methusaleh donor, I pay whether the work succeeds or fails. Another thing I would favor, if I was feeling really flush, is "Medical tenure". Allow any M.D., Medical researcher, and possibly even any alternative medical practitioner who can pass some difficult exams in statistics to leave their job permanently and spend the time working on any medical problem they think is important. During this time they would recieve the lesser of 50% of their most recent year's after tax income and 1.5 times the US median after tax income for personal use. An equal amount would be available without for grants or expenses related to providing any medical service or performing any sort of research that they believed to be important. They would be required to record the outcomes of their work using standard statistical tools and to relinquish any IP claims developed under government funding and would not be permitted to charge for their services in medicine or any other field while thus employed by the government.
Posted by: michael vassar at Nov 11, 2005 11:05:58 AM
Just piling on, 'cause Tyler really deserves it:
First, Tyler, you aren't a Democrat; you are a Republican.
"First, I would not cite evidence about how Western European countries spend less on health and are healthier than U.S. citizens. This data set, if you take it seriously, also implies that the marginal product of more health care, adjusting for income and a few other variables, is zero. "
Sceond, this is, at best, like looking at a raw corelation between two variables, and ignoring significant confounders. Especially as (guessing here) the ENMP (extremely naive marginal product) is probably negative.
When you see a raw relationship between two variables which is the opposite of what theory and history would predict, it's time to be suspicious of that raw relationship number.
As said above, in a certain sick, twisted and useless sense, that's true. That sense can be summed up as Barry's theorem: "for almost any input-output relationship, it is possible to find a situation where the same input magnitude would yield fall lower output values", or, more simply, "no matter who bad things are, it's possible to scr*w them up worse".
Posted by: Barry at Nov 11, 2005 11:15:26 AM
"The alternative to people having their bosses buy their health insurance isn't limited to people having the government buy their health insurance. They could buy it themselves. Granted, having people buy things for themselves has only been tried in hundreds of other industries for a handful of centuries, but I think it'll work."
So why hasn't it worked? Curiously, private insurance companies are not uniformly thrilled by the idea of sick individuals buying health insurance. I can't say I blame them.
Posted by: gundryggia at Nov 11, 2005 11:20:34 AM
It's impressive how many people are willing to dismiss Tyler's comments without even reading the evidence he supplies.
Regarding the marginal value of health care:
(1) A recent comparison of 21 developed countries also found national life expectancy did not vary significantly with medical care spending, after controlling for income, education, unemployment, animal fat intake, smoking, and consumption of pharmaceuticals...
(2) The most respected relevant study is the RAND Health Insurance Experiment, which for three to five years in the mid 1970s randomly assigned two thousand non-elderly US families to either free health care or a plan with a substantial copayment. Those with free care consumed on average about 25-30% more health care, as measured by spending, obtained more eyeglasses, and had more teeth filled. They had more appropriate medical visits, and as a result suffered one more restricted activity day per year. Beyond this, there was no significant difference in mortality, a general health index, physical functioning, physiologic measures, health practices, satisfaction, or the appropriateness of therapy.
(3) An optimistic accounting of the benefits of specific treatments attributes only five years of the forty or more years of added lifespan over the last two centuries to medicine [17].
(4) The small health effects of medicine also raises the question of why exactly lifespans have increased so dramatically. Over the last century, age-specific mortality rates have fallen at a steady exponential rate across developed countries, without noticeable changes due to major medical and public health innovations [98, 61]. Improvements in sanitation are often given great credit, but no effect on mortality has been found among individual variations in water source and sanitation, even among high mortality populations [33]. Average nutrition has greatly increased, but the fact that people who had very high nutrition a century ago had much higher mortality that we see today makes it hard to attribute most lifespan gains to nutrition.
This and more are all in the link Tyler supplied. Now, you can dispute this evidence by marshalling your own to contradict it, or you can give up on the idea that Europeans are healthier because they have national health care systems and so for some reason spend their money more effectively.
Posted by: Javier at Nov 11, 2005 11:29:24 AM
You'd be against regulations and for corporations just like you are now. Instead of being a corporate cheerleader disguised as a libertarian, you'd be a corporate cheerleader disguised as a Democrat. What am I missing here?
P.S. Smart move getting rid of that Alito ad! "Libertarians for Alito" didn't quite cut it.
Posted by: Deb at Nov 11, 2005 11:52:33 AM
No comment on the specific issue here, just a tip that I and others have mentioned before. You have to do the math for your own situation, but you can get up to 50% discounts from doctors if you pay cash. The doctors get the benefit of not dealing with the claims and bureaucracy while you get the benefit of knowing a truer market price, getting the discount and not paying health care premiums. I do recommend catestrophic insurance.
Posted by: Patinator at Nov 11, 2005 11:55:11 AM
RE: "corporate cheerleader" - I assume you live in the woods and don't work for any company, own any stock or puchase goods or services. Otherwise you are a HYPOCRITE!!!!
Posted by: Deb = Simple Minded at Nov 11, 2005 12:10:49 PM
"'I would admit that we cannot take care of everyone and that we face tough trade-offs.'
Why can't we? Other industrialized countries do it. We'd have to raise taxes by a nontrivial amount, to be sure, but we certainly could do it if we wanted to. You don't get points for intellectual honesty by ruling some policy options out of bounds a priori without explaining why."
Ok, lets start with medical innovation. The US subsidizes the whole world in pharmaceutical innovation. Other industrial countries don't have to have free market systems in drugs because ours pays for ours and theirs.
Posted by: Sebastian Holsclaw at Nov 11, 2005 12:36:54 PM
"That being said, Canada and Western Europe's systems didn't spring out of nowhere, and they're not exactly mystical concepts that can't be replicated in the United States. We can insure everyone, get better results, and CUT TAXES DOING IT"
That rather presupposes that everyone in Western Europe is in fact getting coverage form the State. Which is not true. i know of at least one person here in Portugal who does not get anti-retrovirals from the State medical system. I need to have a tetanus booster in afew weeks, something I shall pay directly for. My wife, when she broke her foot, paid for the X-Ray directly.
I think you have a rather undetailed view of exactly how Western European health care systems work.
Posted by: Tim Worstall at Nov 11, 2005 12:47:32 PM
"Ok, lets start with medical innovation. The US subsidizes the whole world in pharmaceutical innovation. Other industrial countries don't have to have free market systems in drugs because ours pays for ours and theirs."
Which is in fact an explanation. But it isn't necessarily TC's explanation, which is what people are asking for. It also suggests that TC should have written "or" rather than "and".
Posted by: SomeCallMeTim at Nov 11, 2005 1:30:37 PM
data set implies that the marginal product of more health care is zero.
That's just so wrong. It's just the simplistic conclusion an econometrician could come to.
The simplest explanation is that the spending in the US is concentrated on the wrong tasks, or even on the wrong people. A lot money is spend on dentists, pain-killers, Prozac, MRIs, etc. by people who have the money and none of it is significantly extending their lifes or rising the metrics that are usually chosen to measure health.
On the other hand people without insurance get excluded from the necessary life-saving, risk-reducing health care. And when I say other people, I mean mostly Afro-Americans and immigrant minorities. I don't have the data at hand, but one of the measures I've seen shows that their health standards are on the level of third world countries.
So the big question is whether you are content with knowing that your tax dollars are going to people of a different race. Most Europeans do not have this problem (but the rising amount of immigrants will create a lot of Euro-Republicans).
Posted by: Oskar Shapley at Nov 11, 2005 2:43:24 PM
Tyler wrote, "I would admit that we cannot take care of everyone and that we face tough trade-offs."
I agree. I take Tyler's statement to mean the following: If the U.S. were to guarantee that every resident will be given all the healthcare that he/she could arguably need, no matter the cost and no matter the person's ability to pay, and were to actually try to make good on that guarantee, the compulsory wealth transfers that would be needed would be so great that U.S. productivity would stall and its standard of living would drop accordingly, with the result that we could never reach a point where everyone would get the level of care that some people get now.
Posted by: John P. at Nov 11, 2005 3:13:13 PM
