I said I would take the $600. The reason is that I am currently unemployed and in dire financial straits. $600 would help keep me housed for another month.
In better financial times, I would answer for the coin toss, because I never had the $600 to start with so it’s not really a loss. I have more to gain than to lose at that point.
My gambling logic always depends on my real life situation.
This notion long predates Thaler and Behavioral Economics.* It is simply the notion of decreasing marginal utility.
*OK all of Economics is Behavioral, but it predates the subfield called Behavioral Economics.
I don’t know about the history but Thaler did studies on this to establish it as a principle. It’s true that all economics is behavioral and it seems obvious to us today but it was not how they did economics prior to Thaler.
Not exactly. Loss aversion works in the same mathematical direction as decreasing marginal utility but is much larger in magnitude. Folks can feel very different about loss or or gains whose marginal utility to them is substantially zero.
That was the crux of Thaler’s insight: losses and gains are qualitatively different. Not just quantitatively of differing sign. And that quality difference has a hefty quantity all its own.
I’m well aware of the differences between loss aversion and risk aversion. The distinction is due, however, originally to Kahneman and Tversky not Thaler. It’s in their 1979 paper . “Prospect Theory: An Analysis of Decision under Risk”. Econometrica . 47 (4): 263–291.
Nevertheless, the preferences stated here can be explained by risk aversion equally well as loss aversion. In fact, the mentioned question about whether you’d take multiple such bets is usally referred to as the Samuelson Paradox (“Risk and uncertainty: A fallacy of large numbers.” Scientia (1963) 98: 10–13.) which predates the concept of Loss Aversion by more than a decade.
If you just keep multiplying by ten I’d eventually get to the point where I’d take the guarantee over the bet amount, but it would be at the $6,000,000 guarantee vs. a 50/50 chance of $20,000,000. It’s the first amount where I could say that the guarantee amount would be enough.
My math skills are good enough that I realize the expected value of the bet is $10,000,000 and I’m taking a big loss, but six million would be enough. Six hundred thousand wouldn’t be. I don’t need all the money, I need enough money, and six million would be enough.
Thanks for the refs. Time to get reading.
I like that.
Similar to I’m at a point in my life where I can afford to buy pretty much anything I want - but I can’t afford to buy everything I want. So that flight on a Russian rocket to the ISS is probably not going to be moved to the “been there, done that” column.