For lottery tickets, people get signifigant utility out of the action of scratching/playing, which is also shown by the fact that people play free online gambling games. It can be hypothesized that the inherent risk of loss in gambling gives a higher thrill, increasing the utility over that of a free gamling simulator.
People buy lottery tickets for fun. When people buy lottery tickets as an investment, we typically assume they either have unusually poor judgement or are mentally ill.
I think that this can be explained by realizing that the expected value doesn't actually carry much information for most distributions.
Suppose I have an idea that I am 10% sure will make me $100. It's tempting to just say that the expected value is $90, and that's what's important. Any other venture with a $90 expected value I should treat the same, if I'm risk-neutral.
But where did the 10% and $100 come from? Well, they're just expected values of some other distributions, aren't they? And those distributions could look like *anything*, they don't even have to be normal! So the "variance" in my $90 expected value could not only be large, but could (and generally will) have bizarre and non-intuitive dependencies on any of its inputs.
If "risk-neutral" means only caring about the expected value, that's really saying that you don't care at all about any aspect of the actual shape of any of the distributions involved! That sounds pretty dramatically irrational to me.
Really, I think that this behavior is explained by people (heuristically) taking into account parameters other than just expected value, and that being "risk-neutral" really means insensitivity to only some of those parameters.
For lottery tickets, people get signifigant utility out of the action of scratching/playing, which is also shown by the fact that people play free online gambling games. It can be hypothesized that the inherent risk of loss in gambling gives a higher thrill, increasing the utility over that of a free gamling simulator.
People buy lottery tickets for fun. When people buy lottery tickets as an investment, we typically assume they either have unusually poor judgement or are mentally ill.
I think that this can be explained by realizing that the expected value doesn't actually carry much information for most distributions.
Suppose I have an idea that I am 10% sure will make me $100. It's tempting to just say that the expected value is $90, and that's what's important. Any other venture with a $90 expected value I should treat the same, if I'm risk-neutral.
But where did the 10% and $100 come from? Well, they're just expected values of some other distributions, aren't they? And those distributions could look like *anything*, they don't even have to be normal! So the "variance" in my $90 expected value could not only be large, but could (and generally will) have bizarre and non-intuitive dependencies on any of its inputs.
If "risk-neutral" means only caring about the expected value, that's really saying that you don't care at all about any aspect of the actual shape of any of the distributions involved! That sounds pretty dramatically irrational to me.
Really, I think that this behavior is explained by people (heuristically) taking into account parameters other than just expected value, and that being "risk-neutral" really means insensitivity to only some of those parameters.