Friday, March 25, 2005

Unarguments.

From left2right, a passage which, just by reading it, makes me think the opposite of what they're trying to argue:

One point on which conservatives and liberals ought to be able to achieve consensus is the wrongfulness of inflicting prenatal injury. Prenatal injury raises issues quite different from those raised by abortion because nothing about the morality of inflicting prenatal injury hinges on the controversial matter of the moral status of the fetus. This is because the real victim of prenatal injury is not the fetus but the person into whom the fetus, if it survives, will develop. Prenatal injury typically inflicts lasting damage that may affect an individual’s entire life for the worse. For this reason, to injure a fetus in a certain way is little different morally from painlessly inflicting the same injury on a small child.

And so: why is abortion legal, again?

I have no position on this question: I just sort of tune out when I hear it, in fact. I wish it weren't so, but there you have it.

But it's quite extraordinary when an author states an assumption ("Prenatal injury raises issues quite different from those raised by abortion") so clearly yet unappealingly that his own argument goes into reverse, un-convincing his audience. (Or at least me.) Are there many examples of this?

[Also, I just realized that the "it's a woman's body" argument for abortion has to be fairly tightly drawn to avoid bad consequences. It must be an "infestation" argument, that the fetus is a parasite. Because if the "woman's body" argument takes the form of "it's a product of my body, and I can do what I want with it," it's unclear why this argument stops when the baby is born. ("I made it. I can kill it.")]

Tuesday, March 22, 2005

Consumer surplus robbery!

Generally, microeconomists assume that it's better to leave the collective gains of trade in the form of a consumer surplus rather than monopoly profits. This isn't just because of the deadweight loss: they'd still believe this even if the monopolist could perfectly discriminate, and if the transferred inputs were put to an equally effective use.

The arguments usually rely on diminishing marginal utility of wealth.

Here's one reason that may not go through. Consider the case of a government-regulated (or subsidized) monopolist providing an industrial good, like energy or whatnot--that is, one that is used predominantly by businesses.

In that case, the "consumer surplus" provided by a competitive market or regulatory pricing decisions goes to rich businesses. (Not that there's anything wrong with that!) And the monopoly profits, that disappear in the price-regulated case, may have gone more or less to the government, cutting subsidies, etc. That is, it might have lowered taxes, or provided services.

So, ignoring all the complications, in some cases consumers can get screwed by government regulation to ensure a "consumer surplus"!

Monday, March 21, 2005

Voter ignorance and the countermajoritarian difficulty.

I'm sure it's been said before, but the realizations of public choice theory--that while the market may fail, alternative proposals can fail worse--also apply tolimiting legislative power.

It may be true that voter ignorance means that the will of the people is inchoate, and so judicial review isn't necessarily a great imposition. But it's also true that judicial law-making might have all the flaws of legislation and ten times more.

These points are very well made by Waldron.

---
Funny: a congressperson said, trying to stop congress from re-opening the Schiavo case:

The judiciary has spoken.
Didn't that used to be, like, "the people have spoken"?

Personhood and the image of God.

I continue to find interesting articles discussing tough issues (like the Schiavo case) from a theological, even evangelical, perspective.

Black market in drugs.

Gary Becker writes that a black market in drugs is unlikely if there is a legal option to produce. He cites taxation on cigarettes in NY, originally predicted to lead to tax avoidance, but which he claims to be fairly successful.

Some problems. First, drug demand is fairly inelastic with respect to price: he himself notes this above, but in the context of attacking one justification (pushing prices up) for the drug war! He notes that the elasticity is (minus) 0.5. Suppose the demand for drugs, with no taxes or legal sanctions, were to be ten times higher than at present (with legal sanctions). If the elasticity is constant, you would need a tax of twenty times the market price of drugs even to bring drug use back to its current level. If the demand for drug use increases with a decrease in social stigma and the end of the enormous legal penalties, it seems likely that the tax on drugs would be far larger, percentage-wise, than the cigarette tax. That makes the situation not strictly comparable. A 20X tax holds out greater black market profit possibilities than a $4 tax on a $2 pack of cigarettes (roughly the NYC level).

Furthermore, cigarette users are a fairly broad cross section of society. The fact that most fairly law-abiding citizens will not illegally produce nor buy cigarettes on the black market is not surprising. Drug users, however, are not a good cross section. If social stigmas do not entirely dwindle away, you will continue to see drug use concentrated among those who--because they have the least to lose and have fewer reservations about shady connections--care least about legal penalties.

Finally, the NYC program is not exactly an unqualified success. Google for "black market tax cigarettes", and read all the newspaper articles. Or, read Cato's policy paper on the situation.

I actually am probably in favor of drug legalization. I just don't think we can blithely assume away the problems.

Friday, March 18, 2005

Ideal amount of ice in a coke.

My little brother spent too much time in Europe, and now drinks his cokes without ice--as long as it's not too warm to start with.

For the rest of us, uncorrupted, how much ice should there be in a cup of coke? Too much, and you're wasting coke space. Too little, and the ice is completely gone by the time you've finished, and your coke warms up. (Horrors!)

The solution involves balancing the heat loss with the melting of the ice, and making lots of assumptions. I've assumed a perfectly-insulating right circular cylinder open at one end, so the area of heat exchange is constant, I've used a simple model of what the heat exchange area between the melting ice and the coke is, the ice is assumed to exchange heat only with the coke, equilibration of liquid temperature is taken to be instantaneous, and also I've neglected a lot of small details about the final minutes, when a lot of these approximations break down. (For example, the ice starts to exchange heat with the air, too.) I've decided that I can't even tolerate a fraction of a Kelvin temperature rise in my coke, so when the last ice disappears, so should the coke.

Plugging in some simple numbers (for example, I drink my 16 oz. coke at a constant linear rate over 20 min.), I get about 10% ice as ideal.

Experiments will be conducted tomorrow. Conjectures and refutations!

How much do we love our families?

There's been a lot of talk recently about how acting against short-run rationality functions as a signal of commitment. I proposed one experiment, but it was pretty expensive in that it involved plane tickets to Norway and Mexico, and also involved some bodily risk of sore cheeks and std's.

Fortunately for my medical bills, an experiment like this has already been carried out: it's called the after-Christmas sale. If two markets are assumed perfectly separable, an imperfect-marketeer can earn higher profits through price discrimination, screwing the folks in the less elastic market. If there is a cost to transporting goods between these two markets, the monopolist is slightly more constrained, and will maximize profits subject to the constraint that the two prices differ by this cost.

Basically, the two markets are pre-Xmas and after-Xmas purchases. The separability depends on our commitment to the familial meaningfulness of Xmas. The transport cost is the price of waiting it out, handing your children an IOU for a train set on Xmas morning, and then going to the store the next day.

Given the fact that prices for many non-luxury goods (like sweaters, etc.) are often 20-30% lower, we apparently love our families very much.

Thursday, March 17, 2005

Popper and essentialism.

This is the first half of the Popper-and-induction post promised below. Here we show that if testable theories put us in touch with reality--however proximately--then models of causality put us in touch with essences in the same way. Next, we show that the Aristotelian conception of essences has a similar defense as Popper's "genuine conjectures" theory of science, and that a book called "The Growth of Essential Knowledge" could be written as a counterpart to Popper's "The Growth of Scientific Knowledge."

Popper starts optimistically about science:

If a theory is testable, then it implies that events of a certain kind cannot happen; and so it asserts something about reality... Our falsifications thus indicate the points where we have touched reality, as it were.
But then he attacks essentialism:
Essentialism looks upon our ordinary world as mere appearance behind which it discovers the real world. This view has to be discarded once we become conscious of the fact that the world of each of our theories may be explained, in its turn, by further worlds, which are described by futher theory--theories of a higher level of abstraction, of universality, and of testability. The doctrine of an essential or ultimate reality collapses together with that of ultimate explanation.
Yet do not things happen? Something causes things.

Popper admits that we touch reality where our theories fail. Yet we also touch essences when things happen, even if only proximately: does he think the chain of explanation can go back forever? Aristotle, were he alive today, would certainly agree that quarks are real-er than the piano. But whatever the fundamental particle is, it contains in itself the ultimate explanation, the reason-by-which things happen. Things happen a certain way, and not another, and there is a reason for that, an internal principle. And therefore essentialism returns--with a vengeance.

I think we get closer to the essence whenever we find a more fundamental explanation. Of course we never are sure we're finally there. (Is there some way we could be?*) But if Popper thinks that we're in touch with reality through the failures of our theories, we are clearly in touch with the causes of things, their essences, through the parsing of cause and mere effect down to the Planck length.

---
* One of the book's epigrams is from (surprise) a pre-Socratic, Xenophanes:
The gods did not reveal, from the beginning,
All things to us; but in the course of time,
Through seeking we may learn, and know things better.

But as for certain truth, no man has known it,
Nor will he know it; neither of the gods,
Nor yet of all the things of which I speak.
And even if by chance he were to utter
The final truth, he would himself not know it;
For all is but a woven web of guesses.

Marshall and Walras.

There are two easy, classic models for how supply and demand equilibrate in the face of imperfect information. In the Walras model, the price is given by an equation of the form

p' = ED * (p-P)
where p is the present (non-eqm) price, ED is the excess demand, and P is the equilibrium price. The prime is change in time. The Marshall model says something similar about the quantity,
q' = q_D - q_S
and here q is the quantity, q_D is the price that demanders will pay at a given quantity of production, and q_S is the price suppliers want--their marginal cost.

As Nicholson's Micro book says, the Walras model is more appropriate when quantity is fixed (like at an auction, in the very short run), and Marshall's model is better where prices are fixed (say, by long-term contracts), and quantity can be shifted easily, say from long-term inventory.

Unemployment is clearly an example of Marshallian shifting. Wages stay roughly constant, but people are thrown out of work. Why is that? I can understand this for minimum wage workers, but Marshall seems to describe almost everyone. Why shouldn't people's wages be reduced if the labor supply outpaces production in a recession? You can't answer this by saying that workers are risk-averse, and so they want constant wages. By hypothesis workers' wages are varying--down to zero.

So, why?

Social and private divergence.

When social and private costs diverge, we have externalities: micro texts will tell you when costs are not fully internalized, the production point on the possibility frontier shifts away from tangency with the efficient-price ratio line, and, voila, too much of good X. (Public goods are the opposite.)

When some people (I'm not saying who, you know who you are) choose a high level of product differentiation, then they impose external costs on the rest of us in the form of the tyranny of choice. We are perfectly happy to satisfice, but the rental cost of extra shelf space devoted to eleventy-one sorts of pasta sauce drives all prices up, as well as making me feel bad that all I ever get is Prego.

Of course, it's also true that people with extreme preferences for product differentiation keep schnazzy bobo products in production, so that on first dates we can impress the girl with pasta sauce #83. In that sense, we are all free riders on Waddling Thunder's restaurant-going. So maybe this is just a pecuniary externality...

Wednesday, March 16, 2005

Popper on the reality of his piano.

Since according to [a proper view of science] the new scientific theories are, like the old ones, genuine conjectures, they are genuine attempts to describe [reality]. Thus we are led to take all these worlds, including our ordinary world, as equally real aspects or layers of the real world. (If looking through a microscope we change its magnification, then we may see various completely different aspects or layers of the same thing, all equally real.) It is thus mistaken to say that my piano, as I know it, is real, while its alleged molecules and atoms are mere 'logical constructions'.
Popper notes that a piano is real in a less fundamental sense than its atoms are (but it's still real). And the color black of the piano is real in a less fundamental sense than the piano itself is (but they're still real). So we have
piano : atoms :: black : piano.
He's reiterating the Aristotelian doctrine of the difference between the substance and its accidents: black is real, but less real than the substance it inheres in. I find it strange that Popper defends himself more ably against the instrumentalists and positivists--like Mach and Bishop Berkeley, as seen in the final sentence of the quote--than against naive realists with a belief in essences. If his piano is real, then it has effects (and affects, like being black! ha!). And if it has effects, then it actuates them by a principle. And so it has an essence.

Does Popper believe in universals?

This could be retitled, "Why does Popper hate Aristotle so much?" I think it's because he sees Aristotle as the true end of the pre-Socratics, and as diverting science away into essentialism from the atomistic later developments of Democritus.

(By the way: did you know that Democritus didn't just believe in atoms, but also thought there was a fundamentally smallest unit of length and time? Quite a quantum leap!)

This is all tied up into induction, as well. For Aristotle, we experience, induce, and thus grasp the essence. Aristotle's main contributions to science were in biology, where observation of squirmy things looms large. Since Popper downplays observation leading to conclusions, and upplays conjectures leading to corroboration, there's not much space for induction.

The problem is, induction still squeezes in. Popper accurately says that repetition is based upon similarity, and the idea of similarity already includes a point of view: we interpret things as similar with respect to the characteristics of the problem we're trying to solve. (It shouldn't matter whether we use a blue plastic ruler to measure, and then a red one.) To classify things together, you need a universal, and if you believe in universals, you believe in essences, and if you believe in essences, you believe in induction.

This is a complicated chain, and I'll try to post some clear thoughts on it tomorrow.

Nozick isn't true!

An update on the marginal return on reading another page of various types of books. I think that Nozick's Anarchy, S, & U is in fact a textbook. My evidence: the marginal return is constant! Even the last sections are highly informative. In fact, they get more and more brilliant: so it may be fiction. (Oh, no, too far...)

Satisficing and extreme preferences.

To clarify a point I made below, about a Postrel article on Schwartz. Postrel comments that I've confused satisficing and extreme preferences, and that there's no real contradiction between them.

Indeed, there isn't any contradiction between satisficing and extreme preferences on an individual level, because they're both tools in our decision kit, along with calcs of sunk costs, search costs, etc. On a group level, though, the attempt to draw an average minimum standard--which I take to be a pretty direct implication of Schwartz's 90% progressive tax--does in fact make an individual's extreme preferences a problem.

Collective satisficing is okay if the entire collective has extreme preferences (extreme relative to another collective). Individual satisficing and individual extreme preferences are fine too. They're both efficient: we stop maximizing when we want to.

Collective satisficing and individual extreme preferences, unfortunately, leads to clipping: some people consume less brand-differentiation than is efficient for them. Schwartz probably thinks of the progressive-tax/group-satisficing-level in terms of a decision-by-rule-method vs a case-by-case-method distinction: he chooses a rule. The utility loss to extreme individuals is more than offset by the social utility of getting rid of the tyranny of choice for the "rest" of us. So he says.

Normally I wouldn't interpret what someone's thinking (well, that's not true:), but Schwartz's extra-book thoughts described in Postrel's article, esp a 90% rate set by experts to increase the social welfare function by a positive-sum utility transfer, makes what I've guessed plausible.

How to know what book you're reading.

Here's how to tell if you're reading a textbook, a fiction book, or a discursive nonfiction work.

Consider the marginal benefit to you, the reader, of perusing one more page of the book.

If this figure stays constant as you read the book, it's a textbook: usually, if you read half a math book, you learn half the math, and at 90% you've learned 90%. There are occasional variances, like the "extensions to the theory" appendices that may not add that much (or, if you're creative, they may suggest unexplored horizons, and be invaluable).

If the marginal benefit of the next page increases, you're reading fiction. Getting to the denouement is the goal. If you don't read the last 10% of War and Peace, you really haven't read it at all.

If the marginal benefit drops off, you're reading discursive nonfiction. An example is H. L. A. Hart's The Concept of Law. Great book. But the main ideas--the rule of recognition, the internal point of view, etc.--are early on.

And that's how you know what type of book you're reading.

Siblings, taboos, libertarianism.

No, this is not about incest, you shameful monkey!

Remember Steven Pinker's article in the New Republic, and The Fly Bottle's commentary on them? They discussed how asking, "why does it matter if she's my sister?" can get you slapped.

Now I notice a blogpost about "my-sister libertarians": it asks, if you'd be unwilling to send your sister to jail for smoking dope, why you're so eager to do it for people you don't know. In short, everyone's your sister, and the family feeling that keeps us from harshing our bro's buzz should make us all a bit more tolerant and less prone to felonize behavior.

Steven Landsburg, a wonderfully Lewis Carroll economist if ever there was one (whose Cheshire cat puzzles leave you with a grin long after you've solved them, and who is also the author of a great micro text), says something related in Forbes, on "Why protectionism is a lot like racism":

It is just plain ugly to care more about total strangers in Detroit than about total strangers in Juarez.
Of course, he doesn't really say whether we should care lots about Detroiters (and thus lots about Juarezians), or very little about both. Just that they should be equal.

Both demands, unfortunately, are a bit rhapsodic and utopian. When libertarianism goes up against human nature, and insists on spreading the luv, then it makes the only sane option look like statism.

If my brother killed someone, I would be sure it wasn't what it looks like, and I'd try to make sure he didn't get caught--because he's such a nice boy. (How many times have you heard that from relatives and neighbors on the witness stand?)

And nothing in Landsburg's analysis prevents his equal-care principle from applying beyond buying American. While I cared deeply about the victims of the Bali bombings, and was frustrated at the lack of continuing American press coverage, it didn't affect me as deeply as 9/11. (And I'd assume Bali affected Australians and Indonesians more deeply than 9/11.)

I'm afraid that a purported ethics of caring is a poor basis to build up a theory of rights on, because it either proves too much, or it proves nothing. James Buchanan analyzed the difference between the application of rules and the making of rules to make just these points.

Tuesday, March 15, 2005

Slap happy.

The Fly Bottle, perhaps the best blog ever, discusses Pinker's response to the Summers fiasco, and further expounds Pinker's view that taboos are a way of preserving the family, and have real consequences, but need to be limited to the right contexts.

Since the economics portion of the blogosphere (except Quiggin) is always saying everything should be testable, here's a test:

Choose two countries where people agree the standards of the family are very different: say, Norway, where it has declined (perhaps?), and Mexico, where it's quite traditional.

Ask a hubby in both places if he'll take $X to sleep with his wife. If you are more likely to be slapped in Norway than Mexico for all values of X, his theory is corroborated.

Also, try it with sisters. The benefit of this experiment is that you'll get good at dodging slaps, and, at least in Norway, you'll get a lot of loving. (You don't have to pay $X, since it was just an experiment, you can explain.)

During a physics videoconference,

we were discussing the use of laser interferometry to measure plasma density. I playfully suggested that we use a graser--a gamma-ray laser, an unfulfilled dream for fifty years in physics--thinking this would get us far better resolution. A colleague said that it would in fact be worse, since in interferometry you're looking "longitudinally" at the shift in wavelength caused by the dispersion relation of the plasma, while the more typical uses of grasers involve transverse scattering, in which the small wavelength of the graser would give you atomic-level scanning resolution.

I was so wigged out by this--the fact that one application of graser physics had incredibly good resolution, while another is very bad--that I started saying dreamily, "I'll need to think about that while I--".

Suddenly I realized I was going to say, "smoke some pot." I don't smoke pot; it would have been just a joke; but the weirdness of the situation seemed to make the joke appropriate. But. Not at work.

What could I say? I latched onto the fact that a friend of mine smokes pot and watches the Simpsons regularly: they seem to go together well. So I finished up, in a loopy sort of voice:

"I'll need to think about that while I watch the Simpsons at home."

My advisor cocked an eyebrow, and said, "If you need to do that, that's what you need to do."

Puzzlement existed.

The formal and final causes.

Aristotle's teleology isn't just an insistence on respecting the final cause: it also sets up relationships between the four causes, e.g. linking the formal and final causes together very closely:

The formal cause constitutes the essence of something whereas the final cause is the purpose of something. For example, Aristotle believed the tongue to be for the purpose of talking. If the tongue was for the purpose of talking (final cause), then it had to be shaped in a certain way, wide and supple so that it might form subtle differences in sound (formal cause). In this way the purpose of the tongue for speaking dovetails with the structural way it might be brought about (P.A. 660a 27-32).
[From the Internet Encyclopedia of Philosophy.]

This view of that-for-the-sake-of-which sets up an undying enmity between teleology and nominalism. Without essences--or real kinds in modern speak--there is no purposefulness.

The tyranny of choice vs the tyranny of academics.

Virginia Postrel writing for any periodical's business pages is like Charles Ives humming while he sells insurance. Or something like that, something unfulfilled but better than nothing. She writes again on one of her favorite subjects--product differentiation and personal choice:

"As the number of choices grows further, the negatives escalate until we become overloaded," writes psychologist Barry Schwartz in The Paradox of Choice, published last year. "At this point, choice no longer liberates, but debilitates. It might even be said to tyrannize." Schwartz misses the good old days when he didn't expect his jeans to fit perfectly and it took only five minutes to buy a new pair.

The book is a lucid overview of the psychological literature, and within its pages Schwartz, a professor at Swarthmore, sticks to personal advice. He urges readers not to fixate on finding the very best alternative but rather to set standards and accept the first choice that meets those criteria.
Because she's writing in a quality paper and can't call Schwartz an idiot, I'll do it for her here.

Preferences vary in two ways: first, between concrete goods (like coke and pepsi), and second, along a spectrum of how important getting our preference is. I prefer coke to pepsi (unlike Fafnir), but it doesn't matter much. On the other hand, I want my jeans to fit in a certain way. And that way is tight. I mean tight. Really tight. And low around the hips. And the pockets have to be placed just so. Not one in a hundred jeans fits my specifications. As you can imagine, I often go about naked.

Schwartz's advice, to set a minimum-standard and choose the first product that meets it, is good if the marginal return of further searching, under your preference structure, is low. But it sucks if you have extreme preferences: and almost all of us have extreme preferences about something or other, if only about who we love.

Another way to say this is that people have different minimum-standards, so that Schwartz's advice is hardly the recipe for sweeping the shelves clean of competing brands that he claims. It's really an attempt to convert a "preference," which has an individualistic sound to it, into a "standard," which could be decided by experts.

(The consequences of variation in the intensity of people's preferences is well developed by James Buchanan in his The Calculus of Consent. The strict democratic one-man-one-vote rule is inefficient if 49% of folks really want pepsi but 51% with a faint preference for coke make it the national drink. We have some fairly reasonable moral problems with buying votes, although
in this case it would be beneficial, enabling the minority to pay off the majority. Buchanan notes that log-rolling over repeated votes over different issues can squeeze out some of the inefficiencies.)

Great summary analogy

by Mickey Kaus:

The oversimplified, possible implication: The War in Iraq set two trains running. One was the increasing-anger-against-us and more-people-who-will-try-to-kill-us Terrorist Blowback train. The second was the bellicose idealists' Democracy Domino Effect train. It seemed last year as if the first train would pose a threat to the U.S. for decades before the second, rescuing train could catch up with it. Now it looks as if there's at least a chance the second train will catch up sooner than could have been reasonably hoped.