Steve Reads

Is there any non-tautological definition of “self-interested behavior”? If one asserts something like “people are fundamentally selfish,” in other words, does that mean anything?

This requires a brief detour into what economists — so far as I know — mean when they say that people are “rational utility maximizers.” Depending upon whom you ask, a “utility” either has some real significance (cardinal utility), or is really only important for what it says about a person’s choices, which are the fundamental unit of study (ordinal utility). The latter concept makes more sense to me, so I’ll stick with that.

The idea is simple: when presented with a range of options, people are assumed to be able to rank-order those options: right now I prefer ice cream to steak, and prefer steak to tofu. So in some sense ice cream is greater than steak, which in turn is greater than tofu. This “is greater than” relation is confined in certain ways which rational people are expected to follow. If a is greater than b, and b is greater than c, then a is assumed greater than c, for instance: if I prefer ice cream to steak and steak to tofu, then I by all rights should prefer ice cream to tofu if you give me a choice between just those two things.

That’s all it means, so far as I can tell, to say that “the utility of a is greater than the utility of b”: it means that I would prefer a to b in a head-to-head matchup, and that my other preferences must all cohere in certain intuitive ways. (NB: “Intuitive” and “consistent with the evidence” are two different things.)

To say, now, that I’m a utility maximizer is to say … that I always choose the things I desire. This doesn’t seem very controversial. Conversely, it could mean that if you observe someone making a sequence of choices, those choices must be the things that maximize their utility. Again, not very interesting.

Where the proposition that “people are self-interested” gets controversial is when people assert that it implies “no one is altruistic.” Economics, anyway, has nothing to say about that assertion: altruism and “pure selfishness” (whatever that means) are equally consistent with utility theory. Again, utility theory would only posit that I have well-defined preferences, and that I behave in such a way as to satisfy as many of those preferences as I can. If one of my preferences is to see people less fortunate than myself do well, economics gives me no reason to think that this is irrational. I can prefer whatever I want; my preferences just have to fit together in certain ways.

Outside of mathematical economics, people can try to assert that no one is altruistic by other means, none of which is remotely convincing. There’s the “argument [such as it is] from evolution,” namely that people have some inherent biological urge to behave selfishly. This can be swatted down in any number of ways. First might be the approach that Robert H. Frank took in Passions Within Reason: The Strategic Role Of The Emotions. In brief, his argument there was that short-term selfishness — in the sense of maximizing utility over a narrow time interval — needn’t be long-term rational. If I become known as the sort of man who will dump his girlfriend and move on to another as soon as I find someone hotter, I will soon find myself with a reputation that keeps me out of any woman’s bed. There are many repeated-interaction games where it’s in my best interests to sacrifice some short-term gain for my own long-term good.

Even without Frank’s argument, the “evolution says no to altruism” argument needs to get a little nuance before it even approaches reality. Surely I care about my family, yes? Is it altruistic to sacrifice my life for that of my children? Surely this happens, yes?

If anyone clings to the argument that altruism is biologically impossible, then, he’s usually forced to define altruism as non-self-interested behavior directed at strangers. But then he has to explain why many people, in the history of mankind, have jumped into icy rivers or rushed into burning homes to save someone, only to lose their own lives in the attempt. Surely your own death is non-self-interested, if anything is.

Or maybe not! Maybe we can take a slight modification of the hypothesis that humans maximize their utility. Instead, we now posit that they maximize their subjective expected utility. That is, they plot out all the possible courses of action that they could take, plot out all the outcomes that could come from those courses of action, estimate the probability of each outcome, and choose the course of action that maximizes the product of probability and utility. Maybe our firefighter runs into that burning building because he estimates a 50% chance of dying, but a 50% chance of living another day and fathering 500 children to women who tear off their clothes at the mere sight of his alpha maledom.

At some point we’re obliged to label that entire thought process irremediably silly, realize that it’s much easier to just assume altruism is possible, and go eat a by-now-mostly-melted bowl of ice cream.

(Not to mention that our models create our world. If you deeply internalize a belief that people could never be altruists, you encrust your soul against the beauty of the people around you. There’s something to be said for believing in a convenient fiction. Fortunately I don’t think it’s fiction.)

P.S. (2 March 2008): A friend reminds me that “self-interest,” in this context, means “you only care about what happens to you directly; your utility function is indifferent to the conditions of others. Experimentally, this is of course false.” Quite so. We also have a word for those whose utility functions don’t change with the conditions of others. We call them sociopaths.

If you have an iPhone, my blog will now look something like this:

This sexiness is courtesy of the WPTouch plugin, which I discovered through Lifehacker.

P.S.: I had to apply the first tweak to their style: I turned off justified text. Everywhere you see text-align: justify within wordpress/wp-content/plugins/wptouch/themes/default/style.css, put a /* at the start of that line and a */ at the end. I don’t know why anyone thinks it’s a good idea to use justified lines on the web, given that no web browser supports the hyphenation that would make that appetizing.

The more I think about this mortgage thing, the less I understand it. Here’s what I think I understand.

A collateralized debt obligation gets constructed from mortgages in some way or another. The simplest way to think of this, maybe, is that your bank has a big pool of mortgages — $100 billion dollars’ worth — and splits it up into “shares.” Maybe into a billion shares. I buy one share, which entitles me to the returns on $100 worth of house. If homes in general are rising at 10% per year, and your bank holds a representative sample of mortgages, your CDOs will also rise by 10% per year.

Up to here, this sounds no more complicated than the “IOs” and “POs” sketched out in When Genius Failed. So far as I know, no one ever called into question the fundamental wisdom of IOs and POs in there, and certainly not to the degree that they’re badmouthing CDOs today.

A credit default swap is kin to an insurance policy on a CDO: it’s insurance against the CDO’s defaulting. I may not be able to estimate the risk of default on this one CDO, but the arsenal of actuarial and statistical tools allows me to estimate the risk of default on the whole pool. An insurance company, that is, that specialized in CDSes would go bankrupt more often than an insurance company specializing in fire and life insurance, but the risks aren’t qualitatively different. Investing in CDOs, or insuring them with CDSes, isn’t like insuring against a meteor colliding with the earth, and it’s not like insuring against the Russian government’s being taken over by terrorists tomorrow. CDSes insure, ultimately, against macroeconomic risk, which is tricky but not outlandish.

So in principle, I don’t understand why a CDS should be any more opaque than any other complicated security (again, like IOs and POs). As for its connection to macroeconomic conditions: that’s no more tricky than a share of General Motors stock. Rational investors should take into account the risk of a recession, which classically hits automobile companies first. Likewise, they should be able to judge the risk of a recession, and the “premium” on a CDS should rise with that risk.

The step that’s throwing everyone off, including me, is the step whereby a pool of risky CDOs gets subdivided and mixed around such that it comes out AAA-rated on the other end. This was a failure at Moody’s and S&P, for reasons that have been well-reheared (e.g., that they get paid by the companies whose securities they rate).

How does that AAA-rated security get constructed out of that high-risk asset? That seems to be the crux of the issue. Whatever magic happens in there is what caused the bubble to burst. Is it that people were using the Black-Scholes options-pricing formula in unfortunate ways? Black-Scholes, recall, involves taking a share of some security, pairing it with a set of carefully chosen options, and thereby building a “riskless portfolio.” To construct this riskless portfolio, you have to assume you understand the risk of the share you started with; otherwise, how can you choose the options to match against it?

So is it possible that Black-Scholes, or something like it, is lurking inside these AAA-rated-securities-that-aren’t? We start with a risky mortgage (or CDO), and use financial wizardry to build a new security that — we postulate — has the same risk as a U.S. government bond. We can raise or lower that level of risk by adding and subtracting other securities, whose risks we also claim to understand. I don’t know if this is actually what happened inside the AAA-rated securities, but it’s a plausible story. “The Case of the Mistaken Black-Scholesization,” we could call it.

Much depends, obviously, on whether we actually know the level of risk associated with that CDO. If real-estate prices had been rising forever, we might be led to underestimate future rates of default. If I’m a company constructing CDOs out of mortgages, and I’ve bought those mortgages from the originator, the originator has every incentive to push them out the door as fast as possible. Depending upon how the incentives are constructed, the originator may have every reason to lie to me about the reliability of the mortgage.

Why did the originator have every incentive to lie, if that’s in fact what happened? If I’m building securities out of mortgages, I want those securities to come properly advertised with their risk: when I sell them to someone else, or I sell a CDS that claims to properly protect against the CDO’s risk of default, there are certain things that will go dreadfully wrong if I mis-label that risk. If people suddenly start defaulting at high rates, I’m left holding the bag. So it should be in the security-inventor’s interest to ensure that his security’s risk is properly labeled.

Laying at the bottom of this may be boundless, unjustified, and ultimately disastrous optimism about the risk of default, but I don’t see that there’s anything abnormal about this — anything that lies beyond the basic competence of markets to handle. When people discuss the recent collapse, though, they’re discussing it as though the securities themselves were beyond anyone’s comprehension. See, for instance, a recent piece in Wired by Felix Salmon (whose blog I’ve recently fallen in love with). Salmon’s basic assertion, schematically, seems to be as follows:

1. If you want to understand the risk of default, you can’t pretend that individual mortgages are uncorrelated. Macroeconomic collapse hits everyone at once.
2. Modeling correlation is a fine idea, but everything rests on your model.
3. No one was looking at the assumptions behind the model.
4. When those assumptions were violated, the model failed.
5. By this point, the model had become so embedded in financial practice (as readers of An Engine, Not A Camera will recognize) that the model shaped the reality.

Even sharper still: investors used the Gaussian copula described in that piece specifically because defaults are so rare. Since they’re so rare, no one had enough information to understand what price CDSes (and friends) should actually command. As Robert Waldmann put it,

Financial market participants managed to convince themselves that they could beat efficient markets (already a logical contradition) because the market price of a CDS was an accurate enough measure of the probability of defaults that correlations in the changes in the prices of 2 CDSs could be used to estimate the (assumed to be constant) correlation between two (assumed to be normal) latent variables making it possible to estimate the probability that two bonds would both default.

That is they assumed that market prices contained information which no one even claimed to be able to obtain any other way. So how did the information get into the prices. One hypothesis is that the Zeitgeist exists and has rational expectations. I can’t think of another explanation.

In short: a million investors, each possessing no information, cannot magically produce a market that has enough information to determine the correct price level.

All told, it’s a very strange picture:

• Markets are decent at determining some kinds of correlation; they know, for instance, that General Motors stock will get nailed as soon as a recession starts. Therefore, if you’re trying to construct a portfolio whose constituents are as uncorrelated as possible, we normally have the tools to do that.
• We can’t compute meaningful correlations, of course, when there’s not enough data.
• But we try to fake our way out of this lack of data using financial tools based on dubious assumptions.
• J. Random Quant uses these tools without understanding them.
• A million J. Random Quants use these tools without understanding them.
• Hence: utter ruin.

It actually exhausts me to think about this. I have the sort of temperament that wants desperately for someone to have a working model that can get us out of this. I want someone to figure out optimal levels of leverage such that we can arbitrage away most any incorrect pricing, while still leaving the economy stable enough that it doesn’t fall over when brushed with a feather. I want economists to understand the macroeconomy well enough to apply the brakes when things get overheated. And I want there to be such consensus in our understanding of the market that the Federal Reserve chief, heretofore viewed as a god, can’t just say “oops” and expect it all to be okay.

But the rational side of my mind says that it doesn’t matter. It says that if an economist developed the One True Model, that model would soon be incorporated into the functioning of the market, and would probably cease to be true. If everyone knew, for instance, that everyone else was expecting a housing bubble to burst in six months, they’d start short-selling houses or CDOs, thereby causing the bubble to happen sooner. We could envision any number of self-defeating or self-affirming scenarios like this. Or we could envision any number of market structures that lead to alternative possibilities. If the market were truly rational, investors would already be short-selling CDOs, and their price would already reflect the best available knowledge about future housing prices. In that case there would be no bubble.

All my economics reading over the last couple years has left me with much less faith that economic knowledge is actually possible.

The Onion was really prescient, wasn’t it? (Cached.)

It’s bugged me for a while that section 132(f)(2) of the Internal Revenue Code allows companies to deduct up to $175 per month for parking, but only up to $100 per month for mass transit. This meant that companies typically would only give their employees the fringe benefit up to the amount they were allowed to deduct. Mass-transit users got the short end of the stick.

In detail, the law reads as follows:

(f)  Qualified transportation fringe
    (1) In general
        For purposes of this section, the term “qualified transportation fringe” means any of the following provided by an employer to an employee:
            (A) Transportation in a commuter highway vehicle if such transportation is in connection with travel between the employee’s residence and place of employment.
            (B) Any transit pass.
            (C) Qualified parking.
    (2) Limitation on exclusion
        The amount of the fringe benefits which are provided by an employer to any employee and which may be excluded from gross income under subsection (a)(5) shall not exceed—
            (A) $100 per month in the case of the aggregate of the benefits described in subparagraphs (A) and (B) of paragraph (1), and
            (B) $175 per month in the case of qualified parking.

In an opaque way, the American Recovery and Reinvestment Act does away with that discrepancy:

SEC. 1151. INCREASED EXCLUSION AMOUNT FOR COMMUTER TRANSIT BENEFITS AND TRANSIT PASSES.

      (a) In General- Paragraph (2) of section 132(f) is amended by adding at the end the following flush sentence:

            `In the case of any month beginning on or after the date of the enactment of this sentence and before January 1, 2011, subparagraph (A) shall be applied as if the dollar amount therein were the same as the dollar amount in effect for such month under subparagraph (B).'.

So: rock on, Congress! Thanks for fixing a tiny injustice.

Does anyone know how to search legislative history to figure out

1. Which bill originally amended the Internal Revenue Code to introduce the injustice, and
2. Who introduced the fix into ARRA?

Have you heard The Bad Plus? If not, please do yourself and your offspring a favor and give a listen to two of their tracks:

• Their cover of Pink Floyd’s “Comfortably Numb”.
• Their own “The Empire Strikes Backwards”.

If you don’t get into either of them, that’s perfectly okay; I suspect a lot of people won’t, especially if they’re wedded to the Floyd original. For my money, their “Comfortably Numb” is sublime. I have to close my eyes and just let it wash over me. As for “The Empire Strikes Backwards,” either it will put a groove in your pants or you have no pants.

Adam Kessel introduced me to TB+ a few years back. They’re quirky and hilarious and brilliant, and their new album is a must-own.

I’m only just putting together, by the way, that The Bad Plus love covering rock and pop songs, and building new adventurous jazz tunes that are orthogonal to the originals. See, for instance, their versions of Heart’s “Barracuda” and Blondie’s “Heart of Glass.” Honestly, the latter in particular is heart-rending, which is amazing for a jazz cover with no lyrics.

Just go listen to anything of theirs you can lay your hands on. And catch them while they’re on tour to promote their latest album, For All I Care.

The story floating around in the zeitgeist about Eichmann in Jerusalem is that Hannah Arendt established how boring the Nazis were: that the men responsible for the destruction of millions were just paper pushers. This captures part of Arendt’s book, but not nearly all of it. Overall, I think it’s best to describe Eichmann in Jerusalem as a clear-eyed look at the trial of Adolf Eichmann, a study of guilt, and a dispassionate analysis of war-crimes trials. It’s a tremendous book.

Many Jews may stop reading when Arendt seems to accuse them of collaborating with the Nazis. I know virtually nothing about how this book was received, but it wouldn’t surprise me at all if it offended a lot of people. If the historical record is as clear as Arendt claims, however, then there’s nothing to get upset about. Councils of Jewish Elders, says Arendt, were formed in every country that the Nazis took over; those Councils documented the assets of the Jews within their communities, dutifully went around collecting them, turned them over to the Nazis, and only later found themselves herded into cattle cars to Auschwitz. In the death camps, she says, Jews did much of the gruesome work, like removing gold teeth from gas-chamber victims.

This is obviously sensitive stuff. Arendt’s style is to deliver it as honestly and forthrightly as possible. Her style, indeed, is tightly bound to her subject. She writes of Eichmann,

[H]e apologized, saying, “Officialese . . . is my only language.” But the point here is that officialese became his language because he was genuinely incapable of uttering a single sentence that was not a cliché … To be sure, the judges were right when they finally told the accused that all he had said was “empty talk” — except that they thought the emptiness was feigned …

She says elsewhere in Eichmann that the man’s inability to speak was a symptom of his inability to think. It is her duty, then, to view Eichmann’s trial with the clearest eye and sharpest mind possible.

Readers may recall the backstory here: the Israelis kidnapped Eichmann from Argentina in 1960, tried him in Israel, and hanged him in 1962. The trial sought to paint Eichmann as one of the masterminds behind the Holocaust and the vilest sort of monster. Arendt retorts that he was, at best, a high-level functionary, and indeed a paper pusher, and that everyone at the trial could see immediately that this was the case. He never killed anyone, and indeed it seems pretty clear that the merest sight of blood would make the man queasy.

None of this lets Eichmann off the hook, though, which is exactly the point: there’s a world of difference between passers-by, who allowed the European Jews to be destroyed, and the Eichmanns who filed away the forms to send them to their destruction.

Granted, then, that Eichmann deserved to pay in a way that the silent millions did not, why was Israel the proper forum for his punishment? Arendt is skeptical that it was. Eichmann’s crime was a crime against humanity, and he should have been punished the same way that other Nazis were punished at Nuremberg. At the same time, no other nation had stepped up to try Eichmann, and Argentina was refusing to turn over the Nazis within its borders, so Israel may have had no choice. Arendt, and the court’s decision itself, approvingly quote Grotius’s line: “punishment is necessary ‘to defend the honor or authority of him who was hurt by the offence so that the failure to punish may not cause his degradation.’”

Israel seemed to believe that it had the right to try Eichmann because of his crimes against the Jews, which makes me wonder: does Israel automatically grant itself the right to try crimes against Jews even today? Suppose some other country tried to kill all the Jews within its borders now; would Israel grant itself the right to try the leaders of that country?

Arendt says that Israel’s trial of Eichmann was much less about Eichmann and much more about the history of anti-Semitism from the time of Pharaoh all the way up to the Germans. There is an element of farce in all of this, and I think it’s fair to say that Arendt took offense: a crime as serious as Eichmann’s deserves a serious trial, rather than a circus. Fortunately the legal decision that came down at the trial’s conclusion was a model of seriousness.

Even with Eichmann swinging from the gallows, there’s still the matter of Europe’s guilt. Arendt, as ever, is only as brutal as she needs to be here: the nations of Europe stand guilty of allowing the slaughter to happen. The French, for instance, allowed foreign Jews within French borders to be shipped off to Germany, but put their foot down when the Nazis demanded French Jews. Their refusal to export their own Jews points out another matter: when nations said no to the Nazis, the Nazis often backed down. They were by no means an immovable wall of violence. The evil that European nations allowed to happen is all the more inexcusable when we know that there were exceptions.

At the heart of all of this is the basic principle, which Arendt summarizes so well:

There remains, however, one fundamental problem, which was implicitly present in all these postwar trials and which must be mentioned here because it touches upon one of the central moral questions of all time, namely upon the nature and function of human judgment. What we have demanded in these trials, where the defendants had committed “legal” crimes, is that human beings be capable of telling right from wrong even when all they have to guide them is their own judgment, which, moreover, happens to be completely at odds with what they must regard as the unanimous opinion of all those around them.

This moral question, and the denunciation that necessarily follows it, doesn’t go away even if the Nazis mercilessly destroyed those who refused to follow their orders. We wish for a clear voice calling out from the maelstrom. Arendt’s is that voice.

Headline: MBTA to lease part of the Green Line for $200 million.

Wow! $200 million! That will surely help the MBTA address its indebtedness. Continuing on in the article:

The move, authorized last week by the MBTA’s board of directors, will net $217 million for the T over the length of the 85-year lease.

Uh. Over 85 years.

So wait, who are the bigger jerks? The MBTA for feeding a bogus headline to the press, or The Metro for gobbling it up?

I know it’s awfully easy, when you’re starting up an ad campaign in 20 different cities, to pick some quick reference point about each city and move on to the next one:

“Johnson, what do we have for Philadelphia?”

“Cheesesteaks, sir!”

“Tremendous. Rukeyser: Miami!”

“Cigars, sir! And the word pendejo!“

And so forth. So I don’t exactly blame you for picking the American revolution when you come to Boston. You were going to pick one of

• baked beans
• “pahkin’ the cah in Hahvud Yahd” (though only the Harvard University grounds crew really pahk the cah in Hahvud Yahd; you pahk in Hahvud Squayuh, but only if you’re lucky)
• the Red Sox (actually, you were going to make an even narrower joke about the “curse of the Bambino” — which is why recent billboard ads for one cell-phone provider or another told us that their great reception crosses “another curse off the list”)

You may not be aware of why you’re doing this. It’s for two reasons: 1) you’re lazy, and 2) so far as this city’s branding goes, it may as well not have existed from the end of the Federalist era until the Sox won the Series. I’ve been griping about this for a while. Naturally, Virgin America is not going to advertise its flights as “like Boston’s busing crisis — only without the flagpole-stabbings!” I understand this.

Still, though: I’m asking for some originality. Here are some things about Boston that are great:

• Brilliant people.
• The fact that the T is filled with people reading interesting things all the time.
• A shabbiness that is charming at times, infuriating at others. If you’re advertising on the T, you can make an oblique joke about how the train they’re on is probably running late, and how — if it’s rush hour — they’re probably more intimate with the person next to them than they’ve been with their own husbands in the last week. If you’re clever enough about it, you can pass this joke by the T’s censors and get a wry smirk out of your viewers.
• An alarmingly high proportion of professors. Make some joke about … I dunno, tweed. Everyone loves a good tweed joke.
• Endless cute college students, especially during the summer.
• Joyous, glorious summers that make you cry, they’re so pretty, and which make up for the dismal winters.
• Cold. You could make a joke about the cold, if it ever came to that. Please don’t ever let it come to that.
• A part of town that is almost certainly dedicated to the construction of a sentient computer attached to a horrifying simulacrum of a human head. Or so I imagine.
• Traffic.
• Batshit-insane cab drivers, who nonetheless are — in my substantial experience — really pleasant to pedestrians.
• The ability to walk everywhere.
• An Irish pub within, I’d wager, a 10-minute walk of wherever you’re standing.
• Probably the bluest city in the bluest state in the nation.

This took me no time at all. I’m sure that if you look around on the web, you will find lots of other people giving you lots of other ideas about what makes Boston great.

More generally: writing generic ads based on a plainly lazy reading of 20 different cities isn’t going to get you anywhere. Bostonians don’t look up at your signs and think “Oh hell yes! An American Revolution joke! It’s funny because it’s true!” They don’t pull a three-cornered hat out of their closets, dust it off, and drive their Priuses to Concord for the annual re-enactment. Most of us don’t care about the Freedom Trail. I doubt Chicagoans think you’re awesome for writing something about thick-crust pizza. So whom are you impressing with these lazy nods to your various cities? Certainly not the cities’ own residents. Don’t you have ad writers who live in each major city? Ask them about each place. Don’t be a lazy putz.

Love and kisses,
Steve

(Attention conservation notice: Started out as a review of Daschle’s health-care book, metastasized into a compare-and-contrast exercise with Ezekiel Emanuel’s book, and ended up a 1700-word outline of health-reform plans.)

Policymaker and policy-wonk opinion seems to be converging around something like Daschle’s solution. It has a few pegs:

• Everyone must be covered. There are several ways to do this that you’ll typically read:

  • Single-payer, wherein the government is the only provider of insurance. Not many people really seem to be behind this, if only because it would be an impossible sell in the United States. And it would be a tough sell for a very good reason: we don’t trust the government to handle insurance for 300 million people well, even if Medicare seems to be a winner. “Medicare for all” is the slogan here, but it doesn’t necessarily follow that a program aimed at the elderly will scale well to the entire society. That said, Medicare does cover the fraction of the population that requires the most health care (namely the elderly). The government also insures Americans through Medicaid (for the poor) and the VA-hospital system (for veterans), so it may already have proven its competence here. The biggest argument in favor of single payer is that it would minimize overhead: since there would only be one health-insurance administrator, we wouldn’t have to pay for advertising or marketing or lobbying. Plus, we’d get the same benefit that comes whenever companies merge: economies of scale mean that employees can be laid off. (I’m aware that many people will not consider this a virtue.)

    One of the biggest arguments against single payer, at least in the short term, is that the health-insurance companies would fight extremely hard against it; it would kill them, after all. Whence we move on to other approaches.

  • Mandates: require everyone to have health insurance, and fine them if they don’t. Massachusetts does this, and has achieved near-universal coverage. I need to read more about how the fine is set.

  • Give everyone a certificate entitling them to at least basic health care, which they can redeem at any of their friendly neighborhood risk-pooling agents. This would get universal coverage, but insurers could continue to charge high rates to those with pre-existing conditions. To make coverage universal in practice, rather than just in principle, you’d need to find some way to limit costs for the unhealthy. To make this happen, you could force insurers to charge the same amount for everyone. You’d therefore need to set rates for a long list of procedures. Medicare already does this, I believe, so this isn’t especially outlandish. The health-care-certificate plan is what Ezekiel Emanuel proposes. With rates set by a governing board (more on this anon), insurance companies would compete on the basis of who could provide coverage most efficiently. That is: with revenues per customer essentially fixed, the only way to get higher profits is to reduce costs.

    Note that overhead would still be relatively high under this plan, at least compared to single payer. Firms would still market and advertise, though one can also envision a role for the government in telling customers which insurers are best along various axes.

  • Something modeled on the existing Federal Employees Health Benefits Program. Under the FEHBP, federal employees can choose from a menu of insurers. Senator Daschle’s plan, among others, would extend the FEHBP to all Americans, and would introduce “Medicare For All” as one of the plans in the menu.

• Costs must be contained. Health-care costs have been rising faster than the broader rate of inflation for years now. If we don’t save costs here, no solution — private, public, or otherwise — is sustainable. People typically propose a few components to keeping costs down:

  • Electronic medical records. If doctors could share my medical history, we could cut down on lots of administrative overhead. Naturally the question is: why hasn’t the market already provided this? The story we most often hear is that there’s no incentive to coordinate when health care is delivered in such a piecemeal way. Your insurer doesn’t care about building a long-term relationship with you if you’re likely to switch employers within a few years, the specialists you go to normally have no connection to your primary-care physician, and so forth. Solving this disjointed-care problem involves thinking beyond just health insurance to the entire health system.

  • Evidence-based medicine: only paying for care if it’s likely to do you any good. Here’s where any plan is going to get into hot water: how do we decide what “good” means? A few baselines present themselves:

    1. Compare to placebo. I.e., does the treatment do any better than a sugar pill? This is clearly not much of a standard at all. It’s the standard that I’m told the FDA uses when approving any new drug.
    2. Compare to the best existing medicine. Does this new blood thinner, for instance, help prevent heart attacks any better than an aspirin? This is a step in the right direction, but it’s not as far as Daschle and Emanuel go with this sort of thing.

    3. Make an explicit cost/benefit estimate, and refuse to pay for anything that doesn’t give a specified benefit for a specified cost. That is, even if we’re better than the best existing treatment (item 2), we may not be better by a sufficient degree to justify the cost. This is where Daschle and Emanuel go.

       The political and ethical problem here arises when people are near death. When you’re desperate, you will pay for anything — no matter how experimental or how unlikely it is to yield benefits. A government panel deciding what to pay for would immediately come under fire: it refuses to pay for Grandma Johnson’s experimental cancer treatment, she pays for it herself, and she makes a miracle recovery. Look at the big bad government, almost letting Grandma Johnson die.

       It’s clear, though, that someone will have to set limits, or medical spending will quickly swallow up all available GDP. The question is: if private insurers today don’t already set this kind of limit, then why don’t they?

       Also: the devil here is in the details, as I’ve suggested before. Who decides which treatments to approve and which to reject? Senator Daschle and Mr. Emanuel both recommend something explicitly patterned on the Federal Reserve — apolitical and insulated from the electoral process. It hurts me to see a former senator write things like

       The next election is always just around the corner, so it’s hard to get lawmakers to think in the long term. The Fed, in contrast, can make decisions based on data and a thorough analysis of what’s best for the country.

       Insulating against the variable will of the people is exactly why we have a Senate with 6-year terms in the first place! Daschle elsewhere writes that “We’ll be able to wrest power from Congress and the White House only when political leaders realize that they are incapable of making the technical decisions on benefits that are so crucial to any health-care system.” This is frankly stunning. Are we really so fed up with democratic government that we need to “wrest power” from the elected branches? Even though the Executive Branch already has a vast apparatus (FDA, Department of Defense, CIA, National Science Foundation, ad nauseum) for studying technical issues?

       The apolitical health board would guide doctors toward effective technologies; but why are doctors prescribing ineffective medicine to begin with? Clearly Daschle and Emanuel believe that doctors don’t have the right information on hand. In various places they lament how difficult it is for doctors to keep up with the extraordinary pace of medical research. If that’s so, and if government intervention is the best way around this under a public plan, then why doesn’t the government already sort through research for us? Neither Emanuel nor Daschle mentions the FDA here, but isn’t this exactly their role? Both Daschle and Emanuel seem eager to create a new bureaucracy to correct for the existing one’s failings.

       The speed of medical research isn’t the only problem. Daschle writes,

       Studies have proven that aspirin alone can save lives [of heart-attack patients], and that even more people can be saved if doctors immediately use a thrombolytic, a clot-dissolving drug. And yet doctors only give aspirin to three-quarters of the patients who should get it, and they administer a thrombolytic to only half of the people who would benefit from it.

       The bit about the thrombolytic is maybe understandable, but: do 1/4 of doctors really not know that aspirin could help their patients? This doesn’t seem like something that centralized medical information could help; we need to study the problem from the bottom up. Daschle and Emanuel, here at least, don’t work in that direction.

  • Encourage people not to be fat, encourage them to quit smoking, etc. Here’s where we see that the problem extends beyond providing insurance, and beyond caring for the sick. Those who make cigarettes, fast food, and candy will fight tooth and nail against anything under this bullet — e.g., smoking-cessation programs, or telling people not to eat trash.

    Daschle sort of misses the point here when he says (p. 198) that “My three decades on Capitol Hill taught me that nearly every piece of legislation will harm the interests of at least some people. The challenge is to craft a bill that creates more winners than losers, that mitigates the potentially negative effects on the losers, and that benefits society as a whole.” This gives equal weight to all the winners and the losers, but that’s decidedly not the case, as Mancur Olson could tell us. Compact interest groups — private insurers, junk-food companies — know exactly what they’ll lose from a public-health revolution; the general public does not, and is too diffuse to band together.

Daschle’s and Emanuel’s books are valuable for different reasons. Emanuel’s is a taxonomy of health-care proposals, with a clear bias in favor of his own. Daschle’s is yet another reminder — in the vein of Jon Cohn’s Sick — of exactly what’s wrong with our broken health-care system. Both have their problems, of course, and I don’t think that either contains exactly the plan that we want. But if you want to be part of the health-care conversation that we’re going to have within the next year, it’s vital that you read at least one of them.

….namely this:

• Lightly roasted espresso is good if you’re drinking it straight. In this case, go for something from George Howell. While you’re at it, stare with a smirk at the New York Times article noting Starbucks’ acquisition of Howell’s Coffee Connection back in 1994, in which the Times felt it necessary to explain that Starbucks was “a 300-store, specialty coffee chain based in Seattle.” (Cached)

  Here was my realization, though: lightly roasted espresso is too mild to work in a cappuccino. The milk absolutely drowns out the coffee. This is why, despite my love for Simon’s, near Porter Square in Cambridge, I don’t love their cappuccinos. I did, on the other hand, love the Classic Cappuccino at Murky Coffee (whose disappearance from Capitol Hill after I left saddened me deeply), and Murky espresso is fairly light. I wonder what their trick is.

• Darkly roasted espresso is not my cup of … um … tea. It may be yours, and that’s cool. I take my cue here from Murky’s founder, Nick Cho: the oil you see on the outside of a darkly roasted bean “should be inside.”

• Full-city roast: that’s the stuff. It’s juuuuust right.

This is why I think I’ve discovered my new favorite Boston-area café: the Equal Exchange Café, a one-minute walk away from North Station. (You pass over the Zakim Bridge on the way there. Given that you’re walking over a massive Interstate highway, the view could be worse. The Google Street View people have apparently not passed in front of 226 Causeway Street in a while.) Their cappuccinos are spectacular: the espresso is malty — just rich enough to poke out through the foam, which is thick and accompanies the espresso perfectly.

My Short Version For Boston-Area Coffee Drinkers, then, is like so:

• Drink straight espresso at Simon’s and Toscanini’s.
• Drink cappuccinos at the EECafé.

Currently checked out of the library (in no particular order):

• Robert C. Feenstra, Advanced International Trade: Theory and Evidence (recommended by Dani Rodrik via email)
• Lucas Conley, Obsessive Branding Disorder (noted by Chris Rugen)
• Jill Lepore, The Name of War : King Philip's War and the Origins of American Identity (recommended years ago by Scott Sandage)
• Jill Lepore, A Is for American : Letters and Other Characters in the Newly United States (ditto)
• The Terror Presidency : Law and Judgment Inside the Bush Administration (not only do a few of the other books I’ve read cite it, but the Balkinization guys have been all over it for years)
• David Romer, Advanced Macroeconomics (picked up because Brad DeLong suggested that it’s the canonical macro text, and I’m loath to read Mankiw’s)
• Paul Samuelson, Foundations of Economic Analysis (Paul Samuelson’s pathbreaking [so they tell me] text)
• Samuelson, Economics: An Introduction (original 1948 edition, as recommended by Krugman)
• Samantha Power, "A Problem From Hell": America and the Age of Genocide
• Albert N. Shiryaev, Essentials of Stochastic Finance (Cosma Shalizi recommendation)
• Oskar Lange and Fred M. Taylor, On the Economic Theory of Socialism (ditto, followup to my post about economies as computers)
• Joe Stiglitz, Whither Socialism? (ditto)
• Jean-Jacques Rousseau, Discourse on the Origin of Inequality
• Andrei Shleifer, Inefficient Markets : An Introduction to Behavioral Finance. I’ve seen this cited in a few places and seen it in the store a few times. It seems to enlarge upon some earlier papers of Shleifer’s, like the one on noise-trader risk.
• Johann Wolfgang von Goethe and Johann Peter Eckermann, Conversations of Goethe with Johann Peter Eckermann (highly recommended in From Shakespeare to Existentialism)
• Torture and Democracy (cited in Jane Mayer’s The Dark Side)

I wonder if I’ll be able to finish these before they’re all due.

I’d like to take a moment here, by the way, to thank my librarians. They rock. Hats off in particular to Esmé Green, head of the Central Square branch of the Cambridge Public Library.

At some point in my education, I will need to find the collection of papers that encapsulated the socialist calculation debate between Oscar Lange and Friedrich von Hayek. The story, as I understand it, is that Lange believed a sufficiently large computer could rationally allocate resources just as well as the market could.

Somewhere along the line, someone noted that this computer would have to be massive indeed. Assume there are markets for n products (wheat, bread, rubber, tires, etc., each in manifold variations). The problem of assigning prices to all n commodities seems at least as difficult as solving a linear-programming problem with an n-by-n matrix: in a perfectly competitive market, the price of each commodity would be driven down to its marginal cost, which means that the price of the nth commodity could be written like so:

p_n = p₁q₁ + p₂q₂ + … + p_nq_n-1

where p_i is the per-unit price of the ith commodity and q_i is the amount of the ith commodity that goes into the nth commodity. To solve for the prices of all n quantities simultaneously, you would need to solve n such equations at once (one for commodity 1, one for commodity 2, etc.).

Stopping here for a moment: I’m curious how far people took this challenge in practice. For one thing, this matrix would be extremely sparse: most commodities don’t take most other commodities as ingredients. Wheat is not a constituent part of an automobile, for instance. Even assuming that the average commodity has 100 parts — which seems implausibly large — you still have a matrix with millions of rows, most of whose columns are zeroes.

In short, then, the mythical socialist computer would need to solve linear-programming problems that were well beyond the capacities of computers at the time, and may well be beyond today’s computers. (What size linear program can be solved very quickly on a modern computer? If we want the computer to set prices as quickly as a financial market does, the time scale would have to be as small as the smallest scale on which prices typically change.)

In any case, I need to read more.

The question I came here with, though, is about market failures. If we envision a market as a distributed computer, whose many distributed parallel processors are all solving subproblems of the overall price-setting problem, then we have to ask how that distributed computer can screw up. In particular: under what conditions does the computer optimize all its subproblems, yet fail to optimize the overall problem? This is what happens in any number of economic paradoxes — the paradox of thrift, for instance. If we all do what we think is in our best interest, sock away money in the bank and eat only beans and rice, then we all suffer: no money is circulating in the economy, our companies have no money to pay us, and we all end up on the bread line. Individually rational behavior leads to globally suboptimal behavior (specifically, a Pareto-inferior allocation).

I guess what I’m asking is some form of a question about dynamic programming, and how far the analogy between computers and economies can reach.

And then very specifically: does anyone know of papers that tackle this sort of analogy? Does it contribute anything to either field that that field didn’t already know?

“Evangelist” comes from the Greek for “good news,” “ev” being the prefix for “good” and “angel” meaning “messenger”; the original evangelists were the authors of the Christian gospels, spreading the good news about Christ. Somehow last night, the idea of an “anti-evangelist” — someone who spreads the word that we should avoid something — came up. I wasn’t quick enough on my feet or my Greek to invent a nice Greek derivative for “anti-evangelist”. Would “kakangelist” do the trick? The Greek prefix “κακó” — whence “cacophony” — means “bad” or “ill.” So someone who’s “spreading the bad news” (the kind of messenger you’re not supposed to shoot) could be a kakangelist.

I think that’s an ugly word, though. I wonder if anyone can think of a better one. Bonus points if it has some nice connection with the etymology of “evangelist.”

Healthcare, Guaranteed is a set of talking points, and a sales pitch. Ezekiel Emanuel wants to sell us on his Guaranteed Healthcare Access Plan. It promises a radical simplification of the American healthcare-delivery system. First, everyone would be entitled to basic health care, and could pay for additional care beyond the basics. Every citizen would be given a card entitling him to coverage through a qualified insurance provider. In order to keep costs under control, Emanuel’s plan would create an Institute for Technology and Outcomes Assessment that would decide whether new technologies and medicines are beneficial, and would publish statistics on the various health plans’ success at creating healthy patients. It would be funded through a 10% value-added tax, the proceeds of which couldn’t be used to fund any other program.

Finally, central to the whole thing is a National Health Board and twelve regional Health Boards patterned on the Federal Reserve; the health boards would approve insurance plans or kick them out. Like the Fed, the Health Board’s advertised virtue is its insulation from politics. Having just read a biography of Dick Cheney, who was old friends with Alan Greenspan and met regularly with the man in private, I have my doubts about this purported insulation.

Much more importantly, it strikes me as wrongheaded, depressing, and insufficiently principled to want to take power away from the democratically elected branches of government because of the failings of politics; why not just give up on the whole democratic process, then? I can certainly understand the justification: the Fed does control, and the Health Board would control, vast swaths of the economy; one dollar in every six goes to health care under our current broken system. While not denying the logic here, I’d like to see it applied to other important sectors. The Department of Defense, for instance. Surely we wouldn’t want something as sacred as the security of our homeland to be subject to bribery, wherein the states that manufacture new weapons systems or house military bases are chosen on the basis of political power? Surely not. Defense should be walled off from those who would weaken our country in the name of politics.

I hope others find it as disheartening as I do that the first thing politicians often want to do, when confronting a new problem, is to wall it off from their own work. Emanuel is not a politician, we must note — he’s the Chair of the Department of Bioethics at The Clinical Center of the National Institutes of Health with a political-philosophy Ph.D. from Harvard and an M.D. from the same university’s medical school — but this apolitical, technocratic dream is a common one. Tom Daschle (who is a politician, and will probably be Obama’s Secretary of HHS) lays out the same dream near the start of his book Critical.

Healthcare, Guaranteed works better as a series of talking points than as a detailed policy manual. First, it’s good to be reminded of exactly how broken our system is, and the various overlapping pieces that still fail to cover one in six of our countrymen: Medicare if you’re old, Medicaid if you’re poor, private insurance while you’re employed, the emergency room when you’re not, COBRA for a short window in between. And it’s good to be reminded of the consequences of this failure. People stay in jobs that they hate because they lose their safety net when they leave. People stay married because their spouse works and they don’t. Companies employ full-time staff just to sort through the details of competing health plans, thereby taking money away from the jobs that the companies actually specialize in; they’re software companies or automobile companies, after all, not health-insurance-picking companies. A rational system would take this burden out of the hands of companies and shift it onto some entity that specializes in doing it. (Though with that in mind, it’s worth looking to see whether companies already outsource HR. If not, why not? If so, then we have less reason to believe that the government needs to do this.)

Here, though, Emanuel again makes me uneasy: his plan would give the Institute for Technology and Outcomes Assessment the job of measuring medicines, technologies, and health. But doesn’t the FDA do that? Healthcare, Guaranteed doesn’t ask why the FDA is failing at its job. Read Marion Nestle’s What To Eat and Food Politics to understand some of the details there. The food industries have lobbied hard to hem in the FDA and make it less and less able to do its job. Insulating the FDA in some way from politics could certainly help here … but why would I — or any American who sees how his government works — expect Congress to construct a better system than the one we already have, when it can’t handle the system it’s already got? Here of course we get back to the politically insulated structure that Congress will apparently construct, and again I shake my head at Congress’s perception of its own fecklessness in the face of lobbying.

At some point I need details, specifically: how will the National Health Board be kept insulated from politics? If the Fed only has a lever or two to handle (mostly interest rates), I can see how it wouldn’t be the victim of much lobbying: just about as many people will probably want low interest rates as high ones. The new health administration, by contrast, will decide which medicines can be used and which insurance plans are available. That’s a lot of machinery and a lot of little gears. How exactly do we keep it under apolitical control?

These are the sorts of detail questions that Emanuel is loath to answer in Healthcare, Guaranteed. He considers his book ground-clearing, and suggests that thinking too much about the details is a strategy that opponents typically use so that nothing gets done. Perhaps so. Perhaps it’s best to start with what we want, aim for it, and let the details fill themselves in later. I have my doubts; the structure of government-as-she-is contains a lot of hard-won information that we ignore at our peril.