You might be an establishment libertarian ... if public choice theory applies to all government activity but climate science.
Note: futarchy and cryonics are not orthodox, but neither will they earn you the ire of the New York Times.
Monday, December 7, 2009
Best Paragraph I Read Today
Barack Obama, understanding the histrionics required in climate change debates, promises that U.S. emissions in 2050 will be 83 percent below 2005 levels. If so, 2050 emissions will equal those in 1910, when there were 92 million Americans. But there will be 420 million in 2050, so Obama's promise means that per capita emissions then will be about what they were in 1875. That. Will. Not. Happen.
link
Wednesday, December 2, 2009
Consider Biased Thinking
Tyler warns against good and evil thinking (although clumsily). That's good advice. But I think we should embrace biased thinking. I don't mean adopting biases. I mean thinking about biases. I subscribe to the Caplan/Edgerton view of biases. What Robin Hanson calls the One Ring to Rule Them All of biases. What does that mean? It means that we can gain utility out of beliefs that are not aimed at the truth. We can use beliefs to feel good about ourselves ("I'm smart enough, I'm good enough, and doggone it, people like me!"). But only in Lake Wobegon can everyone be above average. We can use our beliefs to signal group membership. Part of the reason why abortion and samesex marriage are flashpoints is because they singal group membership into Christian or secular groups. We use our beliefs to gain status. That's why Danniel Dennett calls atheists "brights." The goal is not to appeal to atheism on rational grounds, but as a way to gain status. We can do all these things with our beliefs but the cost is a loss of the truth.
That's my take on the Hansonian side of self-serving baises. I think the Caplan/Edgerton view is equally valuable. Irrational beliefs flourish when we are protected from the consequences of our actions. That happens in politics because no one voter can change the outcome of an election (at least, not usually). It also happens with beliefs that take a long time to be proven true or false. That includes beliefs about how to choose a career, how to choose a spouse, how much to save for retirement, and how much to eat, and how much to exercise. If we are isolated from consequences of having false beliefs then their most useful function is to use them to promote our self-serving biases.
I am not utopian. Deep down almost everyone still views the world in good and evil terms. But in a world characterized by biased thinking the way to prove that your side is good and the other side is bad is by proving that you are above biased thinking. That means bending over backwards to be polite and evenhanded. Good and evil thinking leads to an arms race of nastiness. Biased thinking leads to an arms race of niceness.
That's my take on the Hansonian side of self-serving baises. I think the Caplan/Edgerton view is equally valuable. Irrational beliefs flourish when we are protected from the consequences of our actions. That happens in politics because no one voter can change the outcome of an election (at least, not usually). It also happens with beliefs that take a long time to be proven true or false. That includes beliefs about how to choose a career, how to choose a spouse, how much to save for retirement, and how much to eat, and how much to exercise. If we are isolated from consequences of having false beliefs then their most useful function is to use them to promote our self-serving biases.
I am not utopian. Deep down almost everyone still views the world in good and evil terms. But in a world characterized by biased thinking the way to prove that your side is good and the other side is bad is by proving that you are above biased thinking. That means bending over backwards to be polite and evenhanded. Good and evil thinking leads to an arms race of nastiness. Biased thinking leads to an arms race of niceness.
Are Atheists More Status Oriented?
Here is a twist on Pascal's Wager. Which do you choose?
Not all beliefs can be settled in the short run. This raises an interesting source of bias. In theory we should aim our beliefs at the truth. But there is the temptation to choose beliefs that will get you a short term reward. That's doubly true for people with a high rate of time discounting. Could that be an important presupposition that explains why people choose atheism?
Here is another example.
The upshot is that there is a strong incentive to choose beliefs that are convenient. If you are wrong you will not have to suffer the consequences, at least, not anytime soon. On this view atheists self-select (or adjust their beliefs) for wealth, status, and worldly merriment.
Option A: Five thousand dollars if the National League wins the World Series in 2050
Option B: One thousand dollars if the National League wins the world series in 2010
Not all beliefs can be settled in the short run. This raises an interesting source of bias. In theory we should aim our beliefs at the truth. But there is the temptation to choose beliefs that will get you a short term reward. That's doubly true for people with a high rate of time discounting. Could that be an important presupposition that explains why people choose atheism?
Option C: An eternity with Jesus in heaven
Option D: A freewheeling life of merriment on Earth
Here is another example.
Option E: Delaying sex until marriage
Option F: Having premarital sex
The upshot is that there is a strong incentive to choose beliefs that are convenient. If you are wrong you will not have to suffer the consequences, at least, not anytime soon. On this view atheists self-select (or adjust their beliefs) for wealth, status, and worldly merriment.
Thursday, November 26, 2009
Humans, Bonobos, and Ants
I've been participating in a really interesting discussion in the comments on Overcoming Bias. It really highlights why monogamy is the only way to have a society that (1) cooperative, and (2) egalitarian.
Chimpanzee politics is a bloodsport in which male chimps fight to become alpha males. This does not happen with bonobos because the females have disguised fertility. It is relatively easy for male chimps to monopolize sexual access to the two or three females in the group who are fertile. It is harder for male bonobos to monopolize sexual access to all eight or ten females. Guarding more people is tougher than guarding fewer. The upshot is that the incentive to become an aggressive status seeking alpha male is lower for male bonobos. Females take up the slack which is why bonobos are female-dominated.
My point was that while the "disguised fertility plus promiscuity" strategy successfully tamed the aggressiveness of males it came with a price. Bonobo males do not have an incentive to invest hard for the welfare of the group. They do not have children as an incentive to work hard gathering food. An anonymous commenter made a very perceptive comment:
The commenter is talking about group selection (I have a bit to say about group selection here). D.S. Wilson has the best way to think about group selection. Imagine an island filled with half good people and half evil people. The good people would be shark food and then the island society would collapse into a Hobbesian "war of all against war." But now imagine two islands. The first is filled with good people and the second is filled with bad people. The first island would thrive and the second would collapse. That's group selection.
In a eusocial species like ants the group level dominates. The way that happens is by creating a reproductive caste. The queen(s) and male drones reproduce whereas the legions of female workers slave away for the good of the group. Since the female workers can't have babies of their own the best they can do is spread their genes indirectly. Work hard so that their little island prospers. That is basicaly what the anonymous commenter is proposing. There is basically nothing that the male bonobos can do to intentionally reproduce because female fertility is disuised. So they can work hard for the group like good little worker ants.
But as the primatologist Frans De Waal points out in Our Inner Ape, this is not what male bonobos do. Although the incentive to monopolize sexual access to females is lower, it still exists. Males do divert their energies into gaining access. De Waal explains in Bonobo: The Forgotten Ape:
This explains the difference between bonobo males and ant female workers. Eusocial species like ants are highly inegalitarian. We can imagine a queen saying the following to a worker female, "Listen. Here's the deal. You will never reproduce. I've been emitting a pheremone that makes you sterile. But if by some miracle you do manage reproduce then we will kill you and kill your larvae. What you will do is work hard for the group. Do that and your genes will be passed on indirectly." The worker has no choice. She will work hard and even sacrifice her life for the good of the group. That's the power of group selection.
Monogamy accomplishes the same thing but it is egalitarian. We can anthropomorphize God and imagine Him saying, "Listen. Here's the deal. You can reproduce if you want. But except for some fine print, you can only have sex with one person. Anything more than that is selfish and immoral."
The unifying theme between monogamy and the eusocial species is clear boundaries specified ex ante about reproductive opportunity. If you don't have these strong boundaries then you create a free rider problem. People will try to increase their reproductive opportunity. But that is a (1) a zero sum game. Each male who gains social status causes others to lose status. (2) it takes energy away from developing a cooperative surpluss for the good of the group. The crucial difference between monogamy and a eusocial species is that the latter are highly inegalitarian. Their essential feature is the creation of a separate reproductive caste.
Monogamy is better than promiscuity because it is group-minded. Monogamy is better than a caste system because it is egalitarian. My take on alternatives to monogamy: "A strange game. The only winning move is not to play". But of course, that assumes a group-oriented perspective.
Chimpanzee politics is a bloodsport in which male chimps fight to become alpha males. This does not happen with bonobos because the females have disguised fertility. It is relatively easy for male chimps to monopolize sexual access to the two or three females in the group who are fertile. It is harder for male bonobos to monopolize sexual access to all eight or ten females. Guarding more people is tougher than guarding fewer. The upshot is that the incentive to become an aggressive status seeking alpha male is lower for male bonobos. Females take up the slack which is why bonobos are female-dominated.
My point was that while the "disguised fertility plus promiscuity" strategy successfully tamed the aggressiveness of males it came with a price. Bonobo males do not have an incentive to invest hard for the welfare of the group. They do not have children as an incentive to work hard gathering food. An anonymous commenter made a very perceptive comment:
On the contrary, the most peculiar thing about (systematically) disguised paternity is all males are genetically incented to cooperate towards the same group goal. It’s a form of higher-level selection–sort of like the worker ants in a colony. You’re not going to see anything like this with chimps or humans.
The commenter is talking about group selection (I have a bit to say about group selection here). D.S. Wilson has the best way to think about group selection. Imagine an island filled with half good people and half evil people. The good people would be shark food and then the island society would collapse into a Hobbesian "war of all against war." But now imagine two islands. The first is filled with good people and the second is filled with bad people. The first island would thrive and the second would collapse. That's group selection.
In a eusocial species like ants the group level dominates. The way that happens is by creating a reproductive caste. The queen(s) and male drones reproduce whereas the legions of female workers slave away for the good of the group. Since the female workers can't have babies of their own the best they can do is spread their genes indirectly. Work hard so that their little island prospers. That is basicaly what the anonymous commenter is proposing. There is basically nothing that the male bonobos can do to intentionally reproduce because female fertility is disuised. So they can work hard for the group like good little worker ants.
But as the primatologist Frans De Waal points out in Our Inner Ape, this is not what male bonobos do. Although the incentive to monopolize sexual access to females is lower, it still exists. Males do divert their energies into gaining access. De Waal explains in Bonobo: The Forgotten Ape:
There is a sharp decline in sexual involvement during a male’s adolescence due to the tendency of dominant males to occupy the core of the traveling parties, where the females are. Only when they enter adulthood and rise in rank do males regain access to receptive females. Not that male bonobos are egalitarian with respect to sexual privileges. In contrast to its peaceable image, the species conforms to the general patterns in the animal kingdom of male competition for females. Bonobo males may compete less fiercely than chimpanzee males, but a recent report from Wamba leaves no doubt that dominant males mate more often than others. Since the two top-ranking males in any traveling party generally do most of the mating, it is assumed that they suppress the sexual activity of other males.
This explains the difference between bonobo males and ant female workers. Eusocial species like ants are highly inegalitarian. We can imagine a queen saying the following to a worker female, "Listen. Here's the deal. You will never reproduce. I've been emitting a pheremone that makes you sterile. But if by some miracle you do manage reproduce then we will kill you and kill your larvae. What you will do is work hard for the group. Do that and your genes will be passed on indirectly." The worker has no choice. She will work hard and even sacrifice her life for the good of the group. That's the power of group selection.
Monogamy accomplishes the same thing but it is egalitarian. We can anthropomorphize God and imagine Him saying, "Listen. Here's the deal. You can reproduce if you want. But except for some fine print, you can only have sex with one person. Anything more than that is selfish and immoral."
The unifying theme between monogamy and the eusocial species is clear boundaries specified ex ante about reproductive opportunity. If you don't have these strong boundaries then you create a free rider problem. People will try to increase their reproductive opportunity. But that is a (1) a zero sum game. Each male who gains social status causes others to lose status. (2) it takes energy away from developing a cooperative surpluss for the good of the group. The crucial difference between monogamy and a eusocial species is that the latter are highly inegalitarian. Their essential feature is the creation of a separate reproductive caste.
Monogamy is better than promiscuity because it is group-minded. Monogamy is better than a caste system because it is egalitarian. My take on alternatives to monogamy: "A strange game. The only winning move is not to play". But of course, that assumes a group-oriented perspective.
Wednesday, November 25, 2009
The Definition of an Alpha Male
March Hare has requested a definition of an alpha male. March can correct me if I'm wrong, but I'm assuming the point of this objection is "if you can't define alpha male then you need to abandon your claim that there is status-seeking behavior in the realm of sex and reproduction." I don't think that is correct. First, this is a blog not a philosophy journal. I already write long posts and I don't want to lose even more readers by providing necessary and sufficient definitions for concepts that are intuitively accessible to everyone.
And as Wittgenstein pointed out, no one has created a satisfactory definition of a game, but that doesn't mean that we can't meaningfully talk about games, or that there is no such thing as a game. As the Supreme Court justice Potter Stewart pointed out, "I don't know how to define pornography but I know it when I see it." Both men and women can make the same claim about alpha males. As long as sexual attractiveness is non-randomly distributed in the male population then there exists an alpha male. Defining it is a tougher job. But let me take a crack at it for the same reason that I spent time trying to create a necessary and sufficient definition of a game (not to be confused with "Game", which I define as a deceptive signal of social status). Creating definitions is fun.
One method is to give an operational definition. (1) Alpha males are those who have more sexual partners than average. Or (2) Alpha males are those who are rated as more sexually attractive than average. A less useful operational definition is that alpha males have a higher social rank than others, but this is skewed by the "stone age brain inside a modern skull" effect. The tall and well-built man in the mail room may be low status in his career but high status out with his buddies on Friday night.
I think the real definition of an alpha male is that they have better genes. That is the root cause of why women like alpha males. Or that alpha males have produced a costly signal of this high genetic quality. Some of those signals are highly developed male secondary sexual characteristics like height, big muscles, a deep voice, and a square jaw. I think you could probably quantify those traits and use that to create a reasonable definition that corresponds to our intuitions.
And as Wittgenstein pointed out, no one has created a satisfactory definition of a game, but that doesn't mean that we can't meaningfully talk about games, or that there is no such thing as a game. As the Supreme Court justice Potter Stewart pointed out, "I don't know how to define pornography but I know it when I see it." Both men and women can make the same claim about alpha males. As long as sexual attractiveness is non-randomly distributed in the male population then there exists an alpha male. Defining it is a tougher job. But let me take a crack at it for the same reason that I spent time trying to create a necessary and sufficient definition of a game (not to be confused with "Game", which I define as a deceptive signal of social status). Creating definitions is fun.
One method is to give an operational definition. (1) Alpha males are those who have more sexual partners than average. Or (2) Alpha males are those who are rated as more sexually attractive than average. A less useful operational definition is that alpha males have a higher social rank than others, but this is skewed by the "stone age brain inside a modern skull" effect. The tall and well-built man in the mail room may be low status in his career but high status out with his buddies on Friday night.
I think the real definition of an alpha male is that they have better genes. That is the root cause of why women like alpha males. Or that alpha males have produced a costly signal of this high genetic quality. Some of those signals are highly developed male secondary sexual characteristics like height, big muscles, a deep voice, and a square jaw. I think you could probably quantify those traits and use that to create a reasonable definition that corresponds to our intuitions.
Bloggingheads Review: Wilkinson and Heath
I’ve been on a bloggingheads kick lately. One of the most enjoyable that I’ve seen was between Will Wilkinson and Joseph Heath. It is about Heath’s new book Filthy Lucre. It is an economics book for the general reader from a slightly left-leaning perspective. Heath ultimately champions what Wilkinson himself has called a laissez-fair welfare state. The “Nordic Tigers” like Iceland and Denmark do have high taxes and lots of social insurance. But they have low corporate income and capital gains taxes and minimal regulation so they actually score quite high in the Heritage Foundation’s index of economic freedom. They would actually have even freer markets than Hong Kong if you exclude taxes and social insurance.
There were two interesting points about this bloggingheads. The first was that Heath is a warm, intelligent, and engaging speaker. One of the tastes that I’ve recently been acquiring as an appreciation for idiosyncratic but effective speech. Heath uses some unusual rhythms and intonations but they, at least to me, engaging and effective. I’d take it over the smooth-talking robots on TV any day.
The second interesting point was that Heath effectively crystallized how to defend a welfare state in terms of insurance and adverse selection. The gist comes down to George Akerlof’s famous point about the market for lemons. Imagine dividing usedcars into two categories: peaches and lemons. Peaches are worth $15,000 and lemons are worth $10,000. Let’s suppose that half the cars for sale are peaches and the other half are lemons. Then consumers would be willing to spend $12,500 for a used car. They come out ahead if they get a peach and behind if they get a lemon, but the gamble is fair. But that means that people who own a lemon would not want to sell it. No one wants to sell a car worth $15,000 for only $12,500. Thus people only sell lemons and you can’t buy a used peach. Peaches are driven out of the market. That is adverse selection.
The same principle applies to health care (although with some important caveats). Suppose peaches only need $5,000 of health insurance (or have a 1 in 100 chance of needing $500,000), and lemons need $10,000 in health insurance. And let’s also suppose that half the insured are peaches and the other half are lemons. Then the insurance company would charge its customers about $7,500 per year (plus profit) for insurance. The means that peaches would be paying $7,500 to get $5,000 in insurance. Some other insurance company would come along and “cream-skim” the peaches by charging them only $1,000 (plus profit). In the real world people are simply peaches or lemons so what this effectively does is destroy the concept of insurance. People are no longer pooling risk but are being charged for their exact condition. This is the crucial argument for government health care. If you force everyone to go into the same risk pool then the premiums will go down for most people because private firms can’t “cream skim” the healthy people with lower premiums.
Social Insurance
And that takes us to the welfare state. Welfare is basically insurance against ‘being female and getting pregnant before you finish your education’ (interestingly enough, no modern welfare state provides child support social insurance). It would be nice to provide insurance against that condition. Unemployment insurance is insurance against losing your job. Disability programs like Supplemental Security Income are basically insurance against getting a disability. Once upon a time these forms of insurance were provided by extended families, churches, and mutual aid societies. But they fell victim to modern statistics as profit-seeking insurance companies began to cream skim (well, at least for health and unemployment insurance. I don’t think that was ever the case with welfare). Heath argues that personal responsibility is a red herring. What we need is the government to create society-wide risk pools. The proper way to think of the government is as “insurer of last resort”.
Wilkinson makes few important points in response. The first is that government welfare programs crowd out private insurance. The second is more important. He points out that some societies are more cohesive than others. The research on trust and social capital supports this. Where social cohesion is high you can have either public or private social insurance. Conversely, where it is low you can’t have either. Heath is claiming that welfare states are necessary but his real claim should be that cohesive societies with lots of social capital are necessary. Thirdly, private insurance can be more effectively monitored. Churches have better “ground level” knowledge of those who are down on their luck than government bureaucrats. Moreover, churches align the incentives of those who receive charity with the monitors of the charity and the congregation which funds it. They all want the person back on their feet as soon as possible. By contrast, government bureaucracies become special interest groups which benefit from increasing the number of enrollees. I wish Wilkinson pushed this point a little harder. It was an agreeable bloggingheads and I think he could have had a couple of hardnosed followups without pushing the bounds of good taste.
There were two interesting points about this bloggingheads. The first was that Heath is a warm, intelligent, and engaging speaker. One of the tastes that I’ve recently been acquiring as an appreciation for idiosyncratic but effective speech. Heath uses some unusual rhythms and intonations but they, at least to me, engaging and effective. I’d take it over the smooth-talking robots on TV any day.
The second interesting point was that Heath effectively crystallized how to defend a welfare state in terms of insurance and adverse selection. The gist comes down to George Akerlof’s famous point about the market for lemons. Imagine dividing usedcars into two categories: peaches and lemons. Peaches are worth $15,000 and lemons are worth $10,000. Let’s suppose that half the cars for sale are peaches and the other half are lemons. Then consumers would be willing to spend $12,500 for a used car. They come out ahead if they get a peach and behind if they get a lemon, but the gamble is fair. But that means that people who own a lemon would not want to sell it. No one wants to sell a car worth $15,000 for only $12,500. Thus people only sell lemons and you can’t buy a used peach. Peaches are driven out of the market. That is adverse selection.
The same principle applies to health care (although with some important caveats). Suppose peaches only need $5,000 of health insurance (or have a 1 in 100 chance of needing $500,000), and lemons need $10,000 in health insurance. And let’s also suppose that half the insured are peaches and the other half are lemons. Then the insurance company would charge its customers about $7,500 per year (plus profit) for insurance. The means that peaches would be paying $7,500 to get $5,000 in insurance. Some other insurance company would come along and “cream-skim” the peaches by charging them only $1,000 (plus profit). In the real world people are simply peaches or lemons so what this effectively does is destroy the concept of insurance. People are no longer pooling risk but are being charged for their exact condition. This is the crucial argument for government health care. If you force everyone to go into the same risk pool then the premiums will go down for most people because private firms can’t “cream skim” the healthy people with lower premiums.
Social Insurance
And that takes us to the welfare state. Welfare is basically insurance against ‘being female and getting pregnant before you finish your education’ (interestingly enough, no modern welfare state provides child support social insurance). It would be nice to provide insurance against that condition. Unemployment insurance is insurance against losing your job. Disability programs like Supplemental Security Income are basically insurance against getting a disability. Once upon a time these forms of insurance were provided by extended families, churches, and mutual aid societies. But they fell victim to modern statistics as profit-seeking insurance companies began to cream skim (well, at least for health and unemployment insurance. I don’t think that was ever the case with welfare). Heath argues that personal responsibility is a red herring. What we need is the government to create society-wide risk pools. The proper way to think of the government is as “insurer of last resort”.
Wilkinson makes few important points in response. The first is that government welfare programs crowd out private insurance. The second is more important. He points out that some societies are more cohesive than others. The research on trust and social capital supports this. Where social cohesion is high you can have either public or private social insurance. Conversely, where it is low you can’t have either. Heath is claiming that welfare states are necessary but his real claim should be that cohesive societies with lots of social capital are necessary. Thirdly, private insurance can be more effectively monitored. Churches have better “ground level” knowledge of those who are down on their luck than government bureaucrats. Moreover, churches align the incentives of those who receive charity with the monitors of the charity and the congregation which funds it. They all want the person back on their feet as soon as possible. By contrast, government bureaucracies become special interest groups which benefit from increasing the number of enrollees. I wish Wilkinson pushed this point a little harder. It was an agreeable bloggingheads and I think he could have had a couple of hardnosed followups without pushing the bounds of good taste.
Subscribe to:
Posts (Atom)
