Thursday, June 30, 2011

Eternal youth

The opening words from a Robson and Kaplan 2007 paper I happened upon:
One way to illustrate the effect of aging on longevity is to calculate the life expectancy of nine year olds if they could sustain their current mortality rate.  For the U.S. population in 2003, this life expectancy would be just over 7,000 years. (See United States Centers for Disease Control, 2006.) That is, we would not be immortal, because there is still a constant positive probability of dying; but our lives would be vastly longer if mortality risks did not increase with age. In fact, 2% of the population of nine year olds would live to almost 30,000 years of age. 
Haven't read the rest of the paper but I thought it was an interesting tidbit.

Tuesday, June 28, 2011

Why not optional exile?

[Posting will continue to be light in the near future]

The point of a prison is twofold. For one, prison is no fun, so the threat of imprisonment is a deterrent against crime. For another, locking criminals up keeps them from committing additional crimes.

The thing is, imprisonment is expensive. By contrast, exile sort of accomplishes both of the above effects -- it's less fun than not exile and it keeps prisoners from committing additional crimes against people in our society -- but we don't have to pay for our prisoners. It's sort of a jerk move towards the rest of the world, but (a) that hardly implies Americans wouldn't do it, and (b) the exile need not be unilateral.

Exile could simply be used as a sentence.  Or the prisoner could even be given a choice: optional exile, or prison term -- so it's not unfair to them.  Alternatively, we could just make escape easier for people awaiting likely imprisonment. (allow bail, don't nab them at the airport, etc).  But instead we usually get bummed out when a criminal manages to skip town. (Darn!  We almost had to deal with this guy for the rest of his life!)

Exile need not be indefinite; like prison sentences, a term can be specified.  Amnesty can even be granted for those who have escaped and been gone for a long time.  Exile could be a decent substitute for shorter-term prison sentences (they can't be so severe that exile isn't a comparable deterrent). It is obviously not the answer for everything, but I'm surprised it's not more popular, especially on those occasions when the prisons get so overcrowded that the Supreme Court orders a mass release of inmates (into not-exile).

Sort of curious why I never hear talk of this.  (Or maybe I've just missed it).

Wednesday, June 15, 2011

Why meat is yummy

Liquidity means easier access for both the owner and the thief.  People have more power when they carry around more cash, but they're also bigger targets for muggers.

Similarly, it seems that being able to use one's own fuel reserves more easily, means others being able to use them more easily as well.  Animals are delicious compared to plants because (a) they need quick access to their energy reserves -- and so those reserves are always standing by at the ready, available to anyone who might call upon them -- and (b) more generally their whole body is a dynamic system that must be able to react quickly to its surroundings, and is therefore more biochemically reactive in general.  From Harold McGee's On Food and Cooking:
The definitive characteristic of animals is the power to move the body...Almost as important to animals as their propulsive machinery is an energy supply compact enough that it doesn't weigh them down and impede their movement.  It turns out that fat packs twice as many calories into a given weight as carbohydrates do.  This is why mobile animals store up energy almost exclusively in fat, and unlike stationary plants, are rich rather than starchy.
Our taste buds in particular are designed to help us recognize and pursue important nutrients...Meat triggers all these tastes, because muscle cells are relatively fragile, and because they're biochemically very active.  The cells in a plant leaf or seed, by contrast, are protected by tough cell walls that prevent much of their contents from being freed by chewing, and their protein and starch are locked up in inert storage granules.  Meat is thus mouth-filling in a way that few plant foods are.
Interesting.  (Though this is not a complete story; for instance cows probably don't enjoy the taste of meat, yet the above does not quite separate their tastes from ours).

Tuesday, June 14, 2011

Eye colors?

A few years ago I noticed that my eyes tint the world slightly different colors.  Specifically, my left eye interprets things as slightly bluer, my right eye as slightly redder.  It's pretty subtle -- I can't tell unless I'm looking at something like a white page and actively paying attention to it -- but it is a nice reminder of the relativity of color perception.  I mean, of course it's going to be different across people, but before I noticed this, I didn't consider the possibility within a single person.

Things I wonder:

  • How much variation is there between eyes, across people?  Am I an outlier or is this a normal amount to be off by?  This finding suggests that most likely, everyone's eyes differ by some amount, whether or not they can notice in whatever lighting situation.  To quantify this, you could just feed each eye images, varying the lighting and color and whatnot, and ask them if they were the same or different. 
  • How much does this variation change over the course of a life?  Eyes change; I expect there to be some amount of drift. And while it would be difficult or impossible to  compare the color perception of my eyes now versus 10 years from now, it would not be impossible to compare their relative perception now versus 10 years from now.
  • Where are the signals actually diverging from each other?  Is it happening in the eye, or the brain? (I suspect the eye).
  • Should subtle counter-tints be part of one's glasses prescription?  How much is fidelity between eyes even worth?
Whenever someone starts getting too picky about the precise optimal color to paint a wall, I think about this.  If my left eye prefers A and my right eye prefers B, then I don't know that I can offer a meaningful opinion about which "works" better.  I suspect the differences between my eyes are smaller than the differences between my eyes (taken together) and other people's eyes.

If you're curious whether you can detect the tinting of your own eyes, I suggest giving it some attention the next time you're reading a physical book.  Alternate covering one eye or another and see if you can notice a difference.  Give it more than a passing attempt; even if you are capable of noticing, you may not notice it right away.

Sunday, June 12, 2011

Classical vs. Behavioral

Exogen(e)ous Combustion comments in response to yesterday's post about the future of behavioral economics.  Here is an excerpt:
And so, my proposed cornerstone: as costs of biases go up, people converge to the rational choice framework...Its these sort of arguments that make be believe that behavioral will never constitute a new paradigm. Instead, I see it as a perhaps-useful, perhaps-not addition to the rational choice framework (just like frictions can be to the current frictionless markets paradigm). One that has real effects, but effects that can be bounded and generally only effect the mostly-indfferent individuals.

And here is my position.  First, it annoys me when behavioral economics is treated in the popular press -- or by behavioral economists who are trying to sell it --- as the Great Overthrower of classical economics.  "Economists used to assume XYZ, but that turns out to be wrong, and a new generation of economists is proving it!"  My problem with this is not that behavioral isn't valuable.  Rather, behavioral and classical economics both have regions of behavior over which they perform better, and there is no inherent conflict between the two if they're talking about different things.  Sometimes a simple rational model works pretty well, sometimes not.

From here, we can study which regions classical and behavioral do well over; that is a worthy topic, and it's very much in dispute.  However, I don't think there's much point in attempting to delineate the boundaries of economics, so as to include or exclude one of these regions. If it is the case that behavioral anomalies tend to disappear as the market or the stakes grow, then that is a statement only about the province of behavioral economics.  It is good to know (although, for the record I'm not endorsing or rejecting this statement here), but it doesn't mean that "real" economists don't worry about these behavioral effects, or that the behavioral paradigm will never amount to anything "important."  Economics is a toolbox, and I don't think we should be telling people what region of behavior they ought to be applying it to.  Large markets are important, but a lot of behavior does not take place in them, and is it just me or are we missing something by leaving the latter to the psychologists?  I think the economic perspective has something to add.

By the way, we could say a similar thing about neuroeconomics.  I view Gul and Pesendorfer's critique as a successful rebuttal of the claim that neuroeconomics will be able to support or reject classical economic theory, but I'm not ready to draw the semantic line between neuroeconomics and economics proper, as they do:

The neuroeconomics program for change in economics ignores the fact that economists,
even when dealing with questions related to those studied in psychology, have different
objectives and address different empirical evidence.

Saturday, June 11, 2011

Elementary motives?

A very nice post by Aviad Heifetz over at Leisure of the Theory Class, comparing the current state of behavioral economics with 18th century chemistry as they attempted to transition away from alchemy and phlogiston.  I recommend checking the whole thing out (it's not long), but here's a snippet:
Are there any “elementary” motives of human behavior? Whatever the answer, it could be useful to keep the perspective that today’s behavioral economics, with its incremental efforts to classify dispositions and their implications, may eventually serve as a jump-board for a future conceptual breakthrough and a better paradigm; but that the language, approach and concepts of such a new paradigm may very well turn to be decisively different than those employed by behavioral economics today.

Thursday, June 9, 2011

A wonderful fiction

Imagine you're watching your favorite movie, right now. And suppose I can push a button and alter the plot, killing off the characters you most care about.

You probably wouldn't like that one bit. In fact you probably value their life at a positive amount, and they don't even "exist."

Well, I mean, they do exist, in your mind. Just like real people exist in your mind. Real people exist in the real world, too, and while that fact about them can influence how you think about them in your mind, it's clearly not a necessary condition for caring seriously about them. It seems that real people do not get magical "lexicographic" priority over fictional people.

To some extent everyone recognizes this, but I think there's a generally unspoken tension here. I mean, how much would it bother you if I killed hundreds of your favorite characters? It starts to add up to something significant, an amount that could actually do some real world good towards saving the life of a real person whom you evidently aren't willing to save for that price (since, after all, you aren't).

Now, if it bothers you that you could care more about a bunch of fictional characters than you care about a particular real person, maybe I can help. But only if you're looking for it to stop bothering you...if you want to stop caring about them, you're in the wrong place.  (At Economonomics we don't condemn people for not donating their emergency don't-press-the-button fund to starving-kids-in-Africa).

Okay then. Instead of trying to make any sort of an airtight case, why don't I just say what makes sense to me, and you can push back if something bothers you. Here goes:
  1. First of all, it doesn't matter what state the universe is in, if there don't exist agents who have preferences over those states. There is no "meaning," no "better or worse," before conditioning on the preferences of some agent or agents. But given agents with preferences, we can start to talk about configurations of reality that make them better or worse off. Then, and only then, do we have an up and a down.
  2. Where do preferences live? In minds. Ultimately, all value, all meaning, is created in the mind. That's not to say that the real world doesn't provide inputs into our mental production functions. But utility payoffs are realized in our heads, and there is no value realized outside of a (generalized) "head".
  3. The point is that external reality is not the point. Whether something hails from fiction or reality, it must pass into a mind to be of any value, and inside the mind the playing field is leveled. Inside the mind, nothing is anything but a thought.

The value of a relationship, for example, is not to be found in some intervening feature of the real world between the people involved. The value of a relationship is realized in their minds. It is made of the thoughts that pass through their minds, not something inherent in the fact that they are currently standing close together.

And what if one of the parties is a fictional character? That is, what if there is just one mind in the relationship? The point is, that's more like dividing by 2 than by infinity. And to varying degrees, the same answer applies to the relationship between you and your dog, or perhaps someday, your non-sentient robotic counterpart that behaves as if it were actually realizing value on the other side of the relationship. Therefore it does not bother me, the idea of caring more about fictional people who are passing through one's mind than real people who aren't. And I am not disturbed by the idea of being friends with robots who will seem like humans but never really be "the same."

Now before I go any farther with this, let me just be open about the following. It's easy to start with you caring about a fictional character, and get a very different result about the "value" of that relationship. All we have to do is
  1. Take an external observer, like a next-door neighbor, say, or perhaps someone like this guy.
  2. Endow his own preferences with some degree of importance.
  3. Assert that he cares an awful lot about external reality.
The broad point, though, is that you will find no argument for the lexicographic importance of external reality in reality itself. To get it, you always have to assume it into someone's preferences. If that's your cup of tea, fine, although for what it's worth, my personal feeling is that we're entitled to our own preferences, that other preferences are just as arbitrary and no more valid, whether they belong to neighbors or kings or anyone else. (Incidentally, this is the general professional attitude of economists, the idea of taking preferences as given, without judgment. It does not universally characterize their private attitudes, but due to the exposure, privately they tend to be much less judgmental about preferences than the average person, which many people find refreshing.  It does have a tendency to frustrate the judgy-judgers though).

Besides being tired of studying for an exam, I wrote this post to kill a few birds with one stone. Some of the carnage is postponed to the future, but here are at least a couple other topics it opens the gateway to:
  • Why one might not be overly obsessed with usefulness, which is usually taken to mean something like "real-world practical significance."
  • Making sense of emotional reactions towards events that we aren't sure happened.
I cannot promise to get to these soon, but they're going in The Urn.

    Tuesday, June 7, 2011

    Never enough time

    I'm about halfway through a couple of busy weeks...posting has been and will continue to be light.  Bear with me!  In the meantime here is something from the mental archives to tide you over.

    You have an infinite number of ping pong balls, numbered 1, 2, 3, ... and so on. You also have a bottomless urn which is, for now, empty.

    Here's the game. You start a stopwatch, put the balls numbered 1-10 in the urn, and take number 1 back out. After 30 seconds, you put in #11-20, and remove #2. After 15 more seconds, you put in #21-30 and remove #3. You continue this process, each time waiting half as long between each iteration.

    When a total time of 30+15+7.5+... seconds = 1 minute has elapsed, you're done. And the question is, what's in the urn now?

    Well on the one hand, at each stage you're putting in 10 balls and taking out 1 -- which seems equivalent to repeatedly putting in 9 balls -- and you're doing it an infinite number of times. So, shouldn't there be an infinite number of balls in the urn?

    This is what basic human intuition tells us. On some level it feels obvious. But intuition can be misleading  So to convince me that "infinitely many balls end up in the urn," you'll have to tell me which balls are in there.

    Is ball #3 in there? No, it was removed after 3 iterations, clearly 3 is too small. How about ball #47? No, you took that one out after iteration 47. OK but 47 is still pretty small, what about a really "big" number, like ball #232525609e86434574385? Again though, we know exactly when you took it out, so it's definitely not in the urn.

    Indeed, any ball you can name is not in the urn. So it seems that nothing is in the urn after a minute, even though you're continually putting 10 in for every 1 you take out!  Sometimes infinity is strange.

    Here's why this came to mind. Though this blog is pretty new, I'm already having a serious problem with overflow. For each post I write, I accumulate several posts I want to write. This would be no problem if I had an infinite amount of time (and patience)! But alas, time is our most desperately scarce resource.