3 ms·
This theory makes very little sense to me. First of all, I don't believe that it is rational to ignore time or impact on your life. Would you take 1 dollar tod
by fauigerzigerk 15y ago
This theory makes very little sense to me. First of all, I don't believe that it is rational to ignore time or impact on your life.
Would you take 1 dollar today or 10 in a year? I think what matters more than the actual answer is that the choice has zero impact on your life either way. Would you take 1 billion today or 10 billion in a year? Is that even the same question? In terms of supposedly "rational" financial accounting it is, but I think it would be totally irrational for an average person not to think about these choices in completely different terms.
At some point a quantitative difference becomes a qualitative difference and it would not be rational to act as if this leap didn't exist just because the original quantitative model is unsuitable to account for it. I believe nonlinearity in things like these is the core of what we call intelligence and calling it irrational is nonsense. (Not sure if that theory does that, but it sounds like it might)
The other example they give, studying for an exam earlier or keep playing a game is a different matter altogether. It's a multi factor optimization and there is no way to transform it to a single factor problem (unlike the financial example). Is the pleasure gained from playing the game worth more or less than reducing stress and risk ahead of the exam? On what scale should that be determined?
- andylei 15y ago> I don't believe that it is rational to ignore time It's not. The question is how you factor time into your discounting. Take the $1 vs $10 example. There are several risk factors at play here, and we can discount for each of them. First is the time value of money. Let's be generous and give you a 5% risk free rate of return. That $10 is now worth about $9.50. Next is just the default risk. What if the guy asking you the question doesn't pay in a year? This risk, however, is artificial. We can modify the experiment to compensate for it: the guy can put it in an escrow account guaranteed to pay you in a year. No discount. Catastrophe risk: what if the bank underwriting the escrow account goes under / what if the entire banking system is destroyed / what if dollars are no longer valuable / what if I'm too dead to enjoy the dollars? Let's be generous and say the combined possibility that some catastrophe will render the $9.50 worthless to you is 20%. Now the money is worth about $7.6. Now the question is, what is this money actual worth to you in terms of utility. Suppose your total wealth was $100,000. $1 is 0.0001% of your wealth. The $7.6 is 7.6 times that value. That means that in order for the $7.6 to have a smaller impact on your life than the $1, your wealth would have to increase by over 7.6 times in the next year. If you're in a situation where the probability that your overall wealth will increase by over 7.6 times in the next year, then your decision to take the dollar now is likely rational. If not, your decision is not utility maximizing (and thus irrational).
- fauigerzigerk 15y agoThe problem with your concept of utility maximization is that utility is subjective and non linear. Non linear is not the same as irrational. For me, 10 billion isn't nearly 10 times as useful as 1 billion. And if I had 10 billion and you offered me another 100 billion for debugging your 1990s style Word Basic macro I wouldn't do it. But all of that would change in an instant if I was in love with a woman who adored men with a net worth of more than 100 billion. Is that irrational? I believe not.
- andylei 15y agomy comment doesn't assume too much about the shape of your utility curve except that it is continuous and monotonic. multiplying everything by a billion does change things, but it doesn't say anything about the rationality of the people who choose 1 dollar now over 10 in the future. we're also not comparing 100 billion dollars to the love a woman. we're comparing 1 dollar now to 10 dollars in a year. my argument is that if you have any kind of sane (rational) discounting system, comparing 1 dollar now to 10 in the future is at worst like comparing 1 dollar now to 7.6 dollars now, with the only uncertainty being your changing utility preferences for dollars. if you honestly believe that the probability that your utility preference for marginal dollars will change by more 7.6 times in one year is greater than 0.5, then choosing 1 dollar may not be irrational. but for most people, that is highly unlikely. thus, your discounting system and thus preferences may be irrational.
- im3w1l 15y agoHow much effort do you have to put in to get those dollars in a year? An email, a call, actually going somewhere? This could easily render the value of those 10$ below 0