February 9, 2005
From the desk of Jane Galt:Anecdotes cut both ways
Dr Tony on TennCare, the generous state Medicaid programme that's being cut because it was bankrupting Tennessee:
Whenever there is an article, like this, about TennCare, they always describe some poor soul that nobody could justify removing from the roles. However, I never see any articles listing all the healthy 20-somethings I see in the ER who are on TennCare. These are potentially insurable people who simply elect not to buy insurance and get on TennCare during an acute illness, usually requiring hospitalization, and then just stay on the roles. Then they ask for a prescription for Tylenol because they have no money. Then they spill their beeper and cigarettes out of their purse as they reach for that cell phone that's ringing.
This is actually a serious problem with journalism. Whenever we have a programme cut, we go out looking for the heartrending story of someone who will lose their benefit.
But living, as we do, in an imperfect world, we have to accept that there are tradeoffs to social policy. Until Jesus comes back to sort out the deserving from the needy, we are stuck with legislators and civil servants to do the job. Since they, unlike Jesus, are not omniscient, they will get it wrong some of the time.
This means that we have to choose whether to bias towards false negatives (setting up a system that will deny benefits to some who deserve them) or false positives (setting up a system that will give benefits to some who don't). This is not a binary choice; at any given level of benefit, I guarantee you that someone, somewhere, should be getting it and isn't; while someone else shouldn't, but is. The question is how many of each you've got; society has to decide how much of each it wants to risk, even while knowing that, due to the very nature of the problem (imperfect information) it will not have a very exact idea of whether they are denying benefit to too many, or too few.
Put like that, of course, the answer seems obvious to all but the hard-core libertarian: deny benefits to as few people as possible! But of course, widening the benefit basket has costs. The most obvious are the monetary costs, which in the case of Medicaid, are considerable, but there are also all sorts of second-order economic and sociological costs. It's pretty obvious, for example, that the dramatic increase in transfer payments in the 1960's had the primary effect of lessening material want, but the secondary effect of creating an inner-city underclass. These are difficult questions.
But when journalists attack these stories, they are prone to elide the difficulties. Right-wingers will say "liberal bias" and I don't deny that that's a problem, but there are other issues. Heart-rending stories make for good journalism, the kind that gets you noticed. Plus, it's easy to find someone who's going to get their benefits cut; just call your favourite advocacy group, and they'll supply several from their stable. On the other hand, how do you go about finding someone who's basically scamming the system? They don't want to be found.
"Hard cases make bad law", and sad stories make bad policy. What journalists should be telling us is the complex web of problems that surrounds a big policy debate like this one, but too often they settle for the easy way out, reporting numbers and a few "he-said, she-said" matches between the policymakers and their interest-group opponents, which they tie together with a story about some benefit recipient guaranteed to rend the heartstrings. But anecdotes are not data.
Not that journalists shouldn't tell us these stories: I firmly believe that we should see the human face of the policy decisions we make. But we should see all the faces (or at least a representative sample), not just the ones that are easy to find, and reaffirm the average journalist's committment to spending other people's money.
Posted by Jane Galt at February 9, 2005 8:06 AM | TrackBack | Technorati inbound linksIs there a way to syndicate this post so that everyone in the country reads it? Bravo.
For some, one of the emotionally satisfying aspects of central planning is the leveling effect it has. Everybody gets access to the same amount of goods and services! What could be more fair!
For a subset of those holding those attitudes, the fact that such a regime will result in less material well-being for everyone as a whole, and in the most strict central planning regimes, will result in great material deprivation overall, is less imporant than making sure that everyone is equal in their deprivation. Others recognize this dilemma, but choose to ignore it, which is how it is posited that greatly reducing the profitability of pharmaceutical innovation will not affect the amount of private capital allocated to that purpose, or that private capital allocation is unimportant to pharmaceutical innovation. The dream of central planning will never fade completely, whatever it's track record, because it is so emotionally appealing.
Jane:
I agree; journalists should present information on both the beneficiaries and the costs of govt. largess. But they often don't. For example, for just over a week I've had to hear story after story on the majesty of watching Iraqis vote. But few point out that the massive government program called the Iraq War is likely to end up costing at least a half-trillion dollars and, along with making permanent the Administration's tax cuts and new defense spending, puts pressure on the Administration to cut programs that benefit actual citizens of the US. But there seems to be a lot more journalistic interest stories about "a real election" in Iraq and the magic of watching Iraqis vote.
Dems and Pubs both have favored massive govt. spending programs. It's just that ours (a) tend to have a modicum of thought behind them (Iraq? Star Wars? WTF?), and (b) are less risky, in that they risk treasure rather than blood.
Whenever money, goods or services is being given away, moochers will appear. The problem is that politicians are not giving away their money. Private organizations, churchs, etc. are. This gives the latter incentive to separate the deserving from the non-deserving. Sure, some cons will always slip through, but charity and compassion is best handled by those who are mostly likely to know the "victim" and are putting their own resources on the line.
You're expanding the argument rather too broadly. Is it reasonable to ask journalists, in discussing any government action, to talk about all the other things the money might have been spent on?
More to the point, this
But few point out that the massive government program called the Iraq War is likely to end up costing at least a half-trillion dollars and, along with making permanent the Administration's tax cuts and new defense spending, puts pressure on the Administration to cut programs that benefit actual citizens of the US
is just not true; journalists have been exceptionally interested in pointing out the budget pressure from tax cuts, the spending on the Iraq war, the human cost of the Iraq war, the diplomatic cost of the Iraq war, and so forth. The fact that there isn't room to deliver a laundry list of your pet peeves about the administration in every story about Iraq does not mean that the story is being underreported; it just means that the story is about something other than "Things Tim hates about Bush". Though the particular stories you cite may not have carried as much fulmination about the administration as you might have wished, others, equally prominently placed, covered all of your issues. On the other hand, I am aware of no stories on TennCare which highlight people who could be privately insured, but spend the money on cell phones and cigarettes instead, while ignoring the folks who deserve coverage but are losing it. The stories about benefit cuts, unlike the stories about the Iraq war, are all biased one way.
I don't care whether one particular story is balanced or not; what I care about is whether the stories on certain subjects are systematically unbalanced, which I believe they are in the case of spending programmes.
Your last paragraph is, as I'm sure you recognise, a matter of opinion, and not one that's really worth arguing, since it seems unlikely to change on either side.
Tim, you really may wish to look at the record of Democratic Presidents as it pertains to using military action, before proclaiming that Democrats refrain from risking blood, and I'm not necessarily being critical. Also, if Iraq can become ruled by consent of it's population, with the population in control of it's natural resources, 500 billion will have been very well spent.
What's good for the goose is good for the gander, neh? If some liberals insist that a soldier's art can't be exhibited without including pictures of the victims, because that's unbalanced, then how unreasonable is it to expect journalists to show more than one side of a story in the same article?
See http://www.bangornews.com/news/templates/?a=108388&z=179 for the story about the protest against the soldier's art exhibit.
Jane,
I'm sure you've been asked before, but why on Earth do you spell like a Canadian?
Very thoughtful post and good answer to Tim.
Jane,
You are ignoring another part of the equation. The Byzantine rule making that takes place trying to separate those who should get help from those who don't. A good example are the IRS rules concerning income.
Everyone knows the answer. Pay taxes on your what you get for your wages and money from other transactions. But how do you define "wages". Is vacation included? Is the portion of medical insurance that the employer pays included? The devil is in the details and the costs to figure it out.
Another example is the medical deduction available on the tax return. Hardly anyone gets to take it because of the high minimum, but everyone has to calculate it to make sure that it does not apply. You also must include the cost of enforcement of these rules.
Try to find these costs in any newspaper or magazine article.
This is part of the problem with reasoning by ancedote. Take a peek at www.DiscountHealthInsurance.com, I don't know how representative it is but for myself private insurance could easily hit $500 a month, adding a spouse or child can take you near the $1000/mo range or beyond! Most people can keep their cell phone bills under $100 a month.
Boonton,
If more people bought their own health insurance, would these charges go down? I don't know the answer, not being snarky, just wondered....
That's really hard to say. Insurance is risk pooling....or like going out with a bunch of friends to dinner and dividing the bill evenly. Your bill will go down if you invite more friends who are light eaters/drinkers but will go up if you invite heavy eaters/drinkers. If you require young and relatively healthy people to pay into an insurance pool it will lower the cost in general.....
Boonton:
My understanding is that my husband's company pays in the range of what you mentioned (somewhere between $500 and $1000 a month), minus his contribution that totals less than $200 a month (his company is very generous in its health plan!) for our family's coverage - sorry, the hubby's on a plane and I can't reach him to confirm amount better than that. He wishes his employer would get the heck out of the equation, bump up his pay by what they're currently paying toward our upkeep, and let us figure out our own medical insurance - possibly major medical plus a flexible-spending account that would help defray the cost of well-child visits and such.
However, he's a "professional" (a professional WHAT? is always the question I want to ask) and if his job came without health benefits it'd be unthinkable in the current environment. Therefore I opine that if his company did get out of the insurance business, they'd be likely to raise his pay accordingly - otherwise they'd never get anybody to take the job. I recall when it first came out that Starbucks offered benefits even to lowly baristas - that plus a decent wage sounded (and still sounds) like a mighty fine deal to me. That is, if Starbucks DID get out of the insurance picture, would they then bump the wage of all their benefit-eligible wage employees to enable them to buy their own insurance more readily? I don't know.
No, the problem is that "the media" doesn’t exist to inform, they exist to entertain, or more specifically to attract eyeballs for advertisers. FOX news sells Cadillac's and financial services and CNN sells deodorant and hir color, and if you look close you will be astonished to find out that the "news" they deliver almost always ends up being of interest to the demographic group that is most likely to buy whatever it is they sell.
The "media" also thrives on controversy (unless the specific controversy ends up driving away the coveted eyeballs). They will always report stuff embarrassing to the current administration because they are guaranteed that the faithful will gobble it up and the opposition will froth.
Finally, no one is really interested in a reasonable discussion of the issues. The heist rated "issue" shows on radio and TV are an insult to intelligence in the cartoonish way they "debate" the issues of the day. But those who thrive on listening to "personalities" calling the other side Nazis and Fascists will always lead the ratings. Sober, fair, and rigorous discussion on the issues is relegated to very small groups of people who have already established a respect for each other.
Jane, what you have pointed out is a symptom, not the problem.
Jane,
I'm sure you've been asked before, but why on Earth do you spell like a Canadian?
She's a "deputy Countries editor" (IIRC) for the Economist. A fine position, no doubt, but it has had the side effect of completely ruining her spelling skills ;)
This is part of the problem with reasoning by ancedote. Take a peek at www.DiscountHealthInsurance.com, I don't know how representative it is but for myself private insurance could easily hit $500 a month, adding a spouse or child can take you near the $1000/mo range or beyond! Most people can keep their cell phone bills under $100 a month.
Here is another anecdote: It is possible to get catastrophic coverage plus a limited level of preventive care coverage for under $200 a month (net of premium, per-visit copays, and any un-covered pharmaceuticals), assuming you don't have to visit your PCP too often and don't have a really nasty block of pre-existing conditions. Check with Kaiser, Blue-Cross-Blue-Shield, etc. Not an ideal setup for a family or an elderly person, but I fail to see why $500/month is necessary for a typical individual.
Jane:
1. it just means that the story is about something other than "Things Tim hates about Bush". Though the particular stories you cite may not have carried as much fulmination about the administration as you might have wished, others, equally prominently placed, covered all of your issues.
That's fair. I'd argue that the coverage tends to be fragmented in a not very useful way, but that's a fair criticism.
2. I wonder to what extent this is simply a problem of presenting aggregate stories against individual ones. As you note, the converse of "benefit cut" individual stories are "benefit scam" stories. There were a spate of them in the 80s when there was much discussion of "welfare queens." There were similar discussions in the 90s during the welfare reform debates. If Bush's immigration policy wasn't on the back-burner, you'd see discussions in both directions about illegal immigrants and the cost of services they use. There are constant discussions of lawyers who "trick" people into seemingly over-large awards.
Situations like TennCare are hard because the "scam" is over the cost of a premium, not over cost of an enormously expensive treatment. It's harder to get worked up about a cost of $200 a month vs. $100,000 of debt. One seems trivial and the other seems disabling for most.
3. The point of your last paragraph is actually the issue that interests me most. What worries me most about the last five odd years is the seeming disintegration of the culture in evaluating claims like those I made. For a long time, certainly in the 90s, there was rough agreement about how to evaluate those sorts of claims. It would be disingenous to pretend that GHWB's economic policy wasn't closer to one Dems could understand than Reagan's (and even his were better than now); similarly, Clinton's economic/benefits policies were closer to Republican understandings than prior Dem policy priorities (recall, for example, Clinton's "Eisenhower" comment).
When centrist Dems get upset (and that really is the larger share of the Dem blogosphere), it is often because a hard won deal struck in the 90s seems to have been ripped to shreds for no particular purpose.
anonymouse:
couple your catastrophic coverage with a and there is absolutely no reason why an individual can't afford insurance (at least for the type of circumstances we really worry about).
Hmm, I hear it's a good thing to close your html switches. I might have to try that some time (that and use the preview function).
I was referring to health savings accounts.
Except that you can't buy catastrophic coverage in New York State. The state, in its infinite wisdom, has decreed that health insurance has to be of the all or nothing variety.
Until Jesus comes back to sort out the deserving from the needy...
Just as soon as the last tree is cut down!-)
> If you require young and relatively healthy people to pay into an insurance pool it will lower the cost in general.....
Because everyone knows that they're just looking for an excuse to subsidize others....
Whenever a program costs some of its participants more than they think that they're getting from said program, said participants try really hard to get out of said program. Or, they change their usage so they get more. (If I'm paying anyway, I might as well waste.)
As an Australian and therefore a speller in the English fashion ("colour", "defence", etc) I'm intrigued by the notion of "Canadian" spelling as I'd always assumed they were sufficiently influenced by the giant down south to spell exactly the same way. Rather off post I admit, but generally how do they spell (in the English way?) and what words did Jane correctly (or mis from an American perspective I suppose) spell in her post?
Gee, Jane, maybe reporters have searched high & low for such undeserving patients and THERE AREN'T ANY!
As I understand your complaint is that news story are unbalanced because they only report who is hurt by a cut in a social program and do not report on those unfairly taking advantage of the program. The old welfare queen question.
Question, if a program helps 99,000 deserving people and there are 1,000 free loaders do you think you would really think a story that was 99% about those who were deserving and 1% on the problem of free loaders was a properly balanced story?
Your real problem is not freeloading, it is social welfare. Right? So you would really think a newspaper story that was all about freeloading would be the only "fair" story.
Am I correct?
Why are you ashamed to give your true beliefs, but have to conceal it behind some other misstatement.
If the policy you favor is so good, why do you have to lie about it?
cac,
Canadian spelling still tracks English spelling — I imagine it's a deliberate "point of difference." In Jane's post there are "favourite" and "programme," which would be "favorite" and "program" in ordinary American usage.
We Americans seem always to drop letters where we can — expediency? Speed? I don't know. Even "catalogue" is getting rather rare; I generally see "catalog," unless the publisher of the catalog/ue in question is going out of its way to appear unusually sophisticated (which still means "European," however much we protest otherwise).
I'm not sure about "committment," which is probably just a typo. But English spellings often do have double letters where the more common American ones don't. Think of "cancelled" vs. "canceled," "travelling" vs. "traveling," &c. The double-'l' version is universal in England, but fairly scarce here.
I've often said that legislating by anecdote (or more bluntly, by sob story) is a lousy idea.
I have normal health insurance through my job, but would have preferred catastrophic coverage with the remainder in a health savings account. With no serious problems anticipated when I started working, and none happening, I could have had a sizable emergency fund.
Whenever a program costs some of its participants more than they think that they're getting from said program, said participants try really hard to get out of said program. Or, they change their usage so they get more. (If I'm paying anyway, I might as well waste.)
Which is sort why health insurance has problems but home insurance doesn't. There's really no way for a homeowner to be sure their home won't burn down so they are getting something for the policy. There's no way for the homeowners insurance to know whose house will burn down so they accept all at more or less the same premium (with some limits...)
It's not asymetrical information that makes it work but hidden information. No one knows whose house is going to burn down until it happens, all that anyone knows is that there's a 1/1000 chance that any particular house will burn down in a year. (or whatever the real odds are).
Insurance breaks down if the information stops being hidden. If a magic fortune telling machine is able to say whose house will burn down next year then insurance collapses. Why would I even spend a $1 for insurance when I know my house won't burn down? Why would the insurance company write a policy for the poor schmuck whose house will burn?
If the hidden information is revealed only slightly the result can actually be something of a benefit. For example, an insurance company may realize a certain type of building material is more prone to burn and use its power to get that material replaced by something safer. Then everyone wins. In theory, a health insurance company may discover that doctors are performing lots of c-sections but there's no benefit...or they may find getting people to visit their doctor once a year will prevent more serious illnesses in terms of costs than the cost of seeing the doc (hence the cheap $10 co-pay which people seem to think is the most scandalous thing in health care today).
But what do you do if genetics and data profiling make it so that an insurance company knows (with some chance of error) who will get sick and who won't? You've crossed into the breakdown point where those who need insurance can't get it because the insurance companies will only want to sell to those who don't.
This is why people like employer provided healthcare so much. The employer basically tells the insurance company "I have 5,000 workers, take them or I go to the next company down the street". The insurance company, even though it may know which of those 5,000 is most likely to be expensive, takes them all at a roughly equal price because its profitable. This is why I built that point into my voucher proposal. Call it 'virtual ignorance' to keep the insurance system working.
> Why would the insurance company write a policy for the poor schmuck whose house will burn?
Yet, real insurance companies write policies covering incidents that have already happened.
I'm curious why Boonton thinks that homeowners insurance doesn't vary with risk, because it does vary significantlyh despite Boonton's assertion.
Boonton's theory also implies that high-risk-of-default people/institutions can't get loans, yet they can, at rates calculated to provide a comparable return on capital to loans made to "sure things". (Of course, when the relevant rates are banned, the loans don't happen, but that's how such regulations hurt high-risk entities.)
Why should we prefer Boonton's theorizing over actual experience? Since he's wrong on the predictions that we can easily check, why should we believe him wrt others?
> This is why I built that point into my voucher proposal. Call it 'virtual ignorance' to keep the insurance system working.
Except for it doesn't, because it isn't too hard to figure out which folks are subsidizing others.
It's nice to see that Boonton is consistent about one thing - he's always willing to resort to force when people won't behave as he'd like.
Actually, I think Boonton is basically right. Loans and such are made to people with "High risk of default," but at a huge premium, and high risk of default genenerally isn't incredibly high. This actually gets to what I think is the major problem of today's health insurance industry: insurance only works when you're pooling risk. If I have a 1/1000 chance of getting cancer, and cancer would be a negative value to me of $1 million, then (ignoring time discounting) cancer insurance should be worth $1000 to me (under a simple theory). But the insurance company shouldn't sell it for $1000, because once you add in administrative overhead they're losing money. But if I have a low risk tolerance, I may pay $1100; the extra $100 is what I'm paying to not have to deal with the overall risk. In other words, I can afford to pay $1100 a year or whatever the amount is; but if I get cancer I can't afford $1 million. I'm paying the extra $100 to guarantee I won't be wiped out.
In today's medical market, we don't just have catastrophic coverage like that, but also pay for highly regular and predictable costs (regular checkups, shots, etc). Once again simplifying greatly and pulling numbers out of a hat, let's say the insurance company and i both know I'll spend $250 on incidental expenses this year. I'll only want to pay up to $250, because otherwise I can get it cheaper on my own. The insurance company wants at least $250 plus expenses. Since they can't sell me the lack-of-risk (there's no risk because we both know the amount), no transaction shoudl take place. But because the market is screwed up in any number of ways (tax incentives for health insurance, HMO negotiations, etc), we actually pay for insurance for this stuff...as in, the insurance company can buy the same coverage for, effectively, $200. This makes the market fundamentally different. There are two things being sold: first is true insurance for low risk of high costs; second is a discount service plan. Trying to lump the two together causes lots of trouble.
An example, from personal experience: all four of my grandparents have had cancer in some form, three of them skin cancer. I myself have had a couple benign skin tumors, including one that could have developed into melanoma. So I fundamentally shouldn't be able to get cancer insurance at a reasonable rate. The question at this point isn't really whether i'll have skin cancer, but when, how badly, and of what sort. So it's basically impossible for me to buy comprehensive insurance--they don't want to put me on a plan at regular rates, and the rates they would take for me are probably prohibitive. Now, this would all be fine (well, not fine, but manageable), except that not being on insurance drives all your other medical costs up significantly (I think it can be something like doubling them in some cases). So because I'm too high a risk to be sold cancer coverage, I'd also have to pay double for my flu shots and such.
There are ways around this problem--such as the abovementioned "lump all 5000 of my employees together" strategy. Other possible solutions (not necessarily good or sensible, just possible) are forbidding discrimination based on health history (really stupid), separating true insurance from the service plan sections, and government-subsidized insurance schemes for people who are stuck (which I think is what my mother wound up doing for me).
I don't know that I like Boonton's specific scheme, though from what I remember it was light-years ahead of anybody else's plan for universal health coverage. I think it solves many of the problems involved in UHC, but not all of them (if this is the same plan I'm thinking of). Still, his point about insurance markets failing when risk is transparent is totally correct.
The homeowners analogy is somewhat simplified. Insurance companies will 'profile' you and possibly insist on a home inspection if they have reason to think your house has an above average risk of putting in a claim. They may even pay for minor improvements if it reduces the chance of a claim dramatically (some auto insurance companies will fix little windshield cracks for free because its cheaper than waiting for the whole thing to go).
I'd like Andy to explain to me how you can buy insurance to cover something that already happened.
> So I fundamentally shouldn't be able to get cancer insurance at a reasonable rate.
No, you shouldn't be able to get insurance at a the same rate as someone with significantly less risk, and that's a very different thing.
Your risk is atypical, so why should the price you pay be typical?
> I'd like Andy to explain to me how you can buy insurance to cover something that already happened.
You go to an insurance company and ask them what they'll charge to cover a fairly well understood risk.
I agree I have an atypical risk, and the price I pay shouldn't be typical. That was sort of my point, with the secondary point that because the whole medical insurance market is screwed up this causes me other difficulties that there's no reason for it to cause. As for 'reasonable price,' first of all most companies won't cover me at all. I believe, and I could be wrong, that they're allowed to tell me to stuff it when I apply for insurance, but aren't allowed to offer me insurance at ten or twenty times the standard rate. So I can't get the stuff at all. Even if someone could sell it to me at an inflated price, the risk is high enough that we start slipping from the 'insurance' model to the 'service plan' model. This means that even if someoen would sell it at a realistic price, it might not be in my interests to pay.
some years back, even MORE hypercritical http://o_nline-loan.tambernat.com/online-loa_n.html online loan of my own writing and composing than I am http://r_efinancing.tambernat.com/refinancin_g.html refinancing now, and regrettably during that time I threw http://m_ortgage.tambernat.com/mortgag_e.html mortgage rates away a lot of material which I could've done http://d_ebt-consolidation.tambernat.com/debt-consolidatio_n.html debt consolidation online something with now. I had a real block around http://r_efinance.tambernat.com/refinanc_e.html refinance help lyrics for songs I was writing, in particular. http://l_oan.tambernat.com/loa_n.html loan relief And one thing that really helped to lift me http://c_redit-counseling.tambernat.com/credit-counselin_g.html credit counseling out of that self-damning phase was to go into http://l_oans.tambernat.com/loan_s.html loans direct
Comments are Closed.