I was watching the 1966 classic film, A Man For All Seasons, about the Thomas Moore’s principled stance in opposition to Henry VIII’s grab for power. Ultimately, Moore is found guilty of treason, and in the final scene, after giving his executioner the customary tip for a clean blow, tells him, ‘don’t worry, you are sending me to God’. Considering the fear of death is one of the greatest anxieties for a conscious being, what a fabulous delusion! Clearly, biases can be helpful if looked at in a broader context, in this case, the belief in God is not evaluated based on evidence of God’s existence, but rather, the effect it has on the believer.
Considering the fear of death is one of the greatest anxieties for a conscious being . . .
Is this proscriptive or descriptive? If descriptive, the anecdote which follows would seem to be a counterexample, with others easy to find; if proscriptive, why should one fear death?
The skills of financial company should produce average returns for their clients and its should be worth for both the sides. Both Finance company and clients should be benefited.
I think this can be boiled down to "more people think they are competent to do X than are competent to do X". For example, stock picking *on average* produces a negative result over picking an index fund, but many people do much better than the average, while many do worse. The same is true for enterprise in general. People have more faith in their own ability on average than they warrant, but some have the knowledge or skills necessary to make it worth while.
Comment 1 - to the people discussing the value of comfort, moments before death: In a Kahneman paper titled "Experience Utility and Objective Happiness", he compares the memory of patients' experience and reaction to colonoscopy pain. From the study, he draws the following conclusions:1) Duration neglect - in hindsight, there was only a correlation of .03 between the duration of the procedure and the patients evaluation of it.2) Peak-End rule - The experience in the last three minutes was highly correlated with the patients evaluation of the overall procedure.
Based on this, if there is any brain function in the final moments of my life where I get to reflect on all the things I've done, I'd like for the last few moments to be pleasant so that I have fond memories. It could be argued that comfort in the final moments is more important than general comfort throughout life. If that moment is frozen in time for the rest of eternity...
Comment 2 - to the people discussing the topic of this post: According to Buffett, the relationship between risk and diversification varies as a function of your knowledge and understanding of the investments under consideration. The more you understand, the more you stand to benefit from non-diversification. So when we say "investors diversify too little", we need to specify which investors we're criticizing. I think most people would agree that no matter how much Buffett diversifies, it's just the right amount.
Also, Benartzi and Thaler have an extremely interesting take on the question of trade frequency (or in this case, the frequency at which you look at your portfolio). They attempt to explain the seemingly high equity risk premium.
Roko: the question wasn't whether they're ethically equivalent; it was whether you'd have any preference between the two ("be tortured then killed" or "do whatever you like then be killed").
If given the choice, would you prefer not to be tortured? It has nothing to do with ethics.
I share Roko's attitude towards happiness. In particular, in a hypothetical situation in which I will certainly be in solitary confinement for an interval of time, then I will certainly die, I am almost completely indifferent if I spend the interval of time being tortured or being left to myself. I would choose the torture to gain what most people would regard as a trivial gain, such as saving someone I approve of from getting a parking ticket or proving a point to you guys.
Note that if the torture is severe or continues for a long time then (like any human) I will probably do anything to stop the pain. I do not claim to be any better than the average human at using my force of will to persist in an aversive course of action. What I am saying is that I would choose the torture even if I know that there is no way for me to change my mind -- no way for me to stop the torture once it has started. But again I would need the "help" of my torturers to hold me to my choice.
When I am in pain, my pain tends to "be transmitted" to interlocutors. I.e., when I am in pain, I have the choice of keeping to myself or running a real risk of bringing pain to the people I talk to. (Some people seem much more vulnerable to this "contagious" pain than others.) To remove this consideration from the discussion is why the hypothetical above stipulates that I will be in solitary confinment in either case.
It is almost impossible to impose severe pain on someone without also causing disability. This is not surprising because to a rough, first-order approximation, pain is a signal the central nervous system uses to indicate that the organism is on a course that will lead to low inclusive fitness, and something that reduces a person's capacity to achieve that goal has a strong tendency to reduce a person's ability to achieve the other goals humans tend have. This consideration is why Roko and I stipulated that we would be killed at the end of the torture or non-torture: because that makes it irrelevant whether the experience disables us.
Nick, I cannot speak for Roko, but yes, in my system of values, if there is no way to avoid heat death then nothing matters. So for example, if there is no way to avoid heat death, I will not mind if Eliezer and his allies implement a CEV which in turn devotes the entire future light cone to a Nice Place to Live.
Eliezer, again I do not speak for Roko, but I would reply that there are still cause-and-effect relationships in the 4D crystal. Define the end result of my choice as the effect of the effect of . . . my choice. Then remove from the set any effects which are not causes. So for example, my making my girlfriend happy has utility only to the extent that the happy girlfriend exerts a continued effect on reality.
My previous comment was poorly thought through; I should have said: what is the "end result"? Does only the effect at infinity matter, or are intermediate points counted?
Joseph Knecht: "Also, why should what is to happen in the future affect the current meaning of happiness? If in fact it turns out that you are not killed in 5 seconds, does the happiness that you are in at death-5-seconds *now* have value whereas if you would have been killed it would have had *no* value?"
Well, I think that contemporary professional philosophy (and the general attitude towards morality and values that pervades our society) has made something of a mistake by considering notions like the "current meaning of happiness".
In ethical considerations, one has to choose whether to maximize over histories of the universe or states of the universe. For example, a hedonistic utilitarian might define the "total happiness" at time T, and integrate this quantity over time to get the integrated total happiness over a world history.
I disagree with this approach, and think that ethical considerations should only take into account the end result of a sequence of events. Since the end result of me being tortured and killed is the same as the end result of me being left alone in a room and then killed, they qualify (for me) as being ethically equivalent.
I think that things only matter insomuch as they affect the future - I am a consequentialist.
I heartily agree that people trade too much, and so, perhaps, 75% of finance professionals should be in different occupations, because they merely encourage negative sum 'gambling'. In that sense, it is similar to health care, too much of an essential thing. But unlike Las Vegas, there is a real value here, the value of allocating capital to the most efficient use, in that, without 'craps' life would be unaffected (except for those who love losing money on craps), while without capital markets, a capitalist economy would not allocate capital to its most productive use.
But then you could say, 75% of all academic articles are useless agitprop or hairsplitting, surely a conservative estimate. Why do they do it? Why do we allow it?
the fact that millions of people work in finance, and many are very well paid, suggest this option value is nontrivial.
Really? Lots of people work in Las Vegas too, but gambling is still a net loss for the (vast majority of) gamblers.
The people who work in casinos are experts at getting you to continue playing, so that they can tax your time and thus enrich themselves.
The finance industry as a whole (esp. the stock-picking part) seems similar. Their skills are not in producing above-average returns for their clients. Their skills are in convincing their clients to invest through them, so that they can tax the capital and thus enrich themselves.
OK, I realize that you're saying that learning to be a finance professional, can have positive utility for you personally. As can learning to be a con man, or a drug dealer. A net drain on society, but a personal benefit for you.
I suppose the expected value of being a finance professional might be positive for a given individual, while much of criminal activity probably has a negative lifetime expected value.
But still. Putting this forth as a benefit to picking individual stocks is kind of lame. The world would be a worse place if there were more of them. Society needs engineers, not lawyers. Or IRS agents.
There exists a global economy containing 1000 (let's say) stock pickers. (SP's try to outsmart the market rather than invest in the index)
Each SP has a strange property: he is capable of identifying any investor's unlikelihood of beating the market UNLESS that investor is himself.
Each SP thus believes that anyone not doing what he is doing, is likely to underperform. Further, each SP recognizes that all other SPs hold this belief.
Then one day somebody that all the SPs trust (say, Warren Buffet) announces: "At least one of you will underperform the market." Then what happens?
(Note the parallels to the 100 couples + cheating problem.)
Considering the fear of death is one of the greatest anxieties for a conscious being . . .
Is this proscriptive or descriptive? If descriptive, the anecdote which follows would seem to be a counterexample, with others easy to find; if proscriptive, why should one fear death?
The skills of financial company should produce average returns for their clients and its should be worth for both the sides. Both Finance company and clients should be benefited.
I think this can be boiled down to "more people think they are competent to do X than are competent to do X". For example, stock picking *on average* produces a negative result over picking an index fund, but many people do much better than the average, while many do worse. The same is true for enterprise in general. People have more faith in their own ability on average than they warrant, but some have the knowledge or skills necessary to make it worth while.
Comment 1 - to the people discussing the value of comfort, moments before death: In a Kahneman paper titled "Experience Utility and Objective Happiness", he compares the memory of patients' experience and reaction to colonoscopy pain. From the study, he draws the following conclusions:1) Duration neglect - in hindsight, there was only a correlation of .03 between the duration of the procedure and the patients evaluation of it.2) Peak-End rule - The experience in the last three minutes was highly correlated with the patients evaluation of the overall procedure.
Based on this, if there is any brain function in the final moments of my life where I get to reflect on all the things I've done, I'd like for the last few moments to be pleasant so that I have fond memories. It could be argued that comfort in the final moments is more important than general comfort throughout life. If that moment is frozen in time for the rest of eternity...
Comment 2 - to the people discussing the topic of this post: According to Buffett, the relationship between risk and diversification varies as a function of your knowledge and understanding of the investments under consideration. The more you understand, the more you stand to benefit from non-diversification. So when we say "investors diversify too little", we need to specify which investors we're criticizing. I think most people would agree that no matter how much Buffett diversifies, it's just the right amount.
Also, Benartzi and Thaler have an extremely interesting take on the question of trade frequency (or in this case, the frequency at which you look at your portfolio). They attempt to explain the seemingly high equity risk premium.
Roko: the question wasn't whether they're ethically equivalent; it was whether you'd have any preference between the two ("be tortured then killed" or "do whatever you like then be killed").
If given the choice, would you prefer not to be tortured? It has nothing to do with ethics.
I share Roko's attitude towards happiness. In particular, in a hypothetical situation in which I will certainly be in solitary confinement for an interval of time, then I will certainly die, I am almost completely indifferent if I spend the interval of time being tortured or being left to myself. I would choose the torture to gain what most people would regard as a trivial gain, such as saving someone I approve of from getting a parking ticket or proving a point to you guys.
Note that if the torture is severe or continues for a long time then (like any human) I will probably do anything to stop the pain. I do not claim to be any better than the average human at using my force of will to persist in an aversive course of action. What I am saying is that I would choose the torture even if I know that there is no way for me to change my mind -- no way for me to stop the torture once it has started. But again I would need the "help" of my torturers to hold me to my choice.
When I am in pain, my pain tends to "be transmitted" to interlocutors. I.e., when I am in pain, I have the choice of keeping to myself or running a real risk of bringing pain to the people I talk to. (Some people seem much more vulnerable to this "contagious" pain than others.) To remove this consideration from the discussion is why the hypothetical above stipulates that I will be in solitary confinment in either case.
It is almost impossible to impose severe pain on someone without also causing disability. This is not surprising because to a rough, first-order approximation, pain is a signal the central nervous system uses to indicate that the organism is on a course that will lead to low inclusive fitness, and something that reduces a person's capacity to achieve that goal has a strong tendency to reduce a person's ability to achieve the other goals humans tend have. This consideration is why Roko and I stipulated that we would be killed at the end of the torture or non-torture: because that makes it irrelevant whether the experience disables us.
Nick, I cannot speak for Roko, but yes, in my system of values, if there is no way to avoid heat death then nothing matters. So for example, if there is no way to avoid heat death, I will not mind if Eliezer and his allies implement a CEV which in turn devotes the entire future light cone to a Nice Place to Live.
Eliezer, again I do not speak for Roko, but I would reply that there are still cause-and-effect relationships in the 4D crystal. Define the end result of my choice as the effect of the effect of . . . my choice. Then remove from the set any effects which are not causes. So for example, my making my girlfriend happy has utility only to the extent that the happy girlfriend exerts a continued effect on reality.
My previous comment was poorly thought through; I should have said: what is the "end result"? Does only the effect at infinity matter, or are intermediate points counted?
Roko, this view seems hardly compatible with timeless physics! The entire 4D crystal of time is the end result.
So you would say that if there's no way to avoid heat death (or other $END_OF_UNIVERSE), nothing matters?
Joseph Knecht: "Also, why should what is to happen in the future affect the current meaning of happiness? If in fact it turns out that you are not killed in 5 seconds, does the happiness that you are in at death-5-seconds *now* have value whereas if you would have been killed it would have had *no* value?"
Well, I think that contemporary professional philosophy (and the general attitude towards morality and values that pervades our society) has made something of a mistake by considering notions like the "current meaning of happiness".
In ethical considerations, one has to choose whether to maximize over histories of the universe or states of the universe. For example, a hedonistic utilitarian might define the "total happiness" at time T, and integrate this quantity over time to get the integrated total happiness over a world history.
I disagree with this approach, and think that ethical considerations should only take into account the end result of a sequence of events. Since the end result of me being tortured and killed is the same as the end result of me being left alone in a room and then killed, they qualify (for me) as being ethically equivalent.
I think that things only matter insomuch as they affect the future - I am a consequentialist.
I heartily agree that people trade too much, and so, perhaps, 75% of finance professionals should be in different occupations, because they merely encourage negative sum 'gambling'. In that sense, it is similar to health care, too much of an essential thing. But unlike Las Vegas, there is a real value here, the value of allocating capital to the most efficient use, in that, without 'craps' life would be unaffected (except for those who love losing money on craps), while without capital markets, a capitalist economy would not allocate capital to its most productive use.
But then you could say, 75% of all academic articles are useless agitprop or hairsplitting, surely a conservative estimate. Why do they do it? Why do we allow it?
Remarkable opening story, very similar to that of Sir Thomas More.
the fact that millions of people work in finance, and many are very well paid, suggest this option value is nontrivial.
Really? Lots of people work in Las Vegas too, but gambling is still a net loss for the (vast majority of) gamblers.
The people who work in casinos are experts at getting you to continue playing, so that they can tax your time and thus enrich themselves.
The finance industry as a whole (esp. the stock-picking part) seems similar. Their skills are not in producing above-average returns for their clients. Their skills are in convincing their clients to invest through them, so that they can tax the capital and thus enrich themselves.
OK, I realize that you're saying that learning to be a finance professional, can have positive utility for you personally. As can learning to be a con man, or a drug dealer. A net drain on society, but a personal benefit for you.
I suppose the expected value of being a finance professional might be positive for a given individual, while much of criminal activity probably has a negative lifetime expected value.
But still. Putting this forth as a benefit to picking individual stocks is kind of lame. The world would be a worse place if there were more of them. Society needs engineers, not lawyers. Or IRS agents.
I don't understand why Eric singles out "option value". What is special about "option value" as opposed to other components of expected return?
Oops, I meant "Roko" and not "Roki" in my previous post. My apologies.
Put this into a math puzzle:
There exists a global economy containing 1000 (let's say) stock pickers. (SP's try to outsmart the market rather than invest in the index)
Each SP has a strange property: he is capable of identifying any investor's unlikelihood of beating the market UNLESS that investor is himself.
Each SP thus believes that anyone not doing what he is doing, is likely to underperform. Further, each SP recognizes that all other SPs hold this belief.
Then one day somebody that all the SPs trust (say, Warren Buffet) announces: "At least one of you will underperform the market." Then what happens?
(Note the parallels to the 100 couples + cheating problem.)