$500 each. The partner is more likely to accept an offer he considers fair, and that’s the fairest possible split.
So I either have to assume that the guy is rational and will take anything I give to him, or cut the 1000 in half.
Giving him more money than I would receive would seem like a safe bet but I can’t imagine anyone accepting 70% but not 50%.
I think I’d offer to give him $200. It seems like a large enough sum to keep him in and the extra $300 for me would be nice. I feel unusually risky today. I’m not a gambling man, but in this case, I would give it a try.
Not getting any money isn’t going to kill me anyway.
At this point I’m just interested in polling what the person who was given the money would do. I am also interested in what people would accept, if they want to volunteer that. I think that might be a second poll.
Oops, you’re right - I should have just said “The second person,” since you most likely would not know the second person.
Yes, it probably should have - that was what I was going for, rather than breaking it down into many different splits. Writing poll questions is hard! 
So, according to the book, a “rational agent” would offer one cent to the second person, since it doesn’t matter how much the first person offers, since the options are take it or leave it for the second person, so any amount is better than no money. It goes on to say that in this experimental game, the first person should always offer only one cent, regardless of how much money is involved. I found this fascinating, as I would never have thought of that. (Yeah, I should have probably had less than a dollar as an option, too.)
I should probably note what book I’m reading - “Filthy Lucre: Economics for People Who Hate Capitalism” by Joseph Heath.
Yes, if I were going to act based on reason alone, a ha’penny is better than nothing, but 1) they don’t make those anymore, and 2) sometimes, people just need to be told to go fuck themselves.
Yeah, I’m having trouble imagining a real-world person in this situation who would look at $.01 on his side of the table and $999.99 on the other side and cheerfully suck it up and take the penny. I mean, seriously?
I’d do that for certain friends of mine. For them, a free grand would be a huge deal. Oddly enough, those particular friends are more likely to offer a 50/50 split, even though they really need the money.
Offering half is the only reasonably sure way to end up with any money at all, since the receiver can reject the offer and we both end up with zero.
If I were the receiver I would reject any offer other than 50/50 even if the difference were only a dollar. The giver risks losing everything if he makes any lop-sided offer.
This is really a game with only one true option, 50/50.
Cat Whisperer
Great, an economics book from a 43 year old philosophy professor.
That’s how people react and think in the real world, not in the World Of Economic Policies.
The book is more of an examination of the most common economic fallacies of both the Right Wing and the Left Wing, rather than some kind of exhaustive economic textbook. I’m finding it fascinating so far - the chapter I’m reading now examines why The Invisible Hand of the Marketplace never actually works.
Economists use the word “rational” to describe behavior that ordinary people know to be irrational and vice versa. The basic premise that they use in games like this is that a “rational” person always desires to have the maximum possible amount of money while caring about nothing else. In reality, only irrational people want that.
As for me, my answer would depend largely on who the other person is. If they’re somebody in need, I would give them the whole $1000. Otherwise I’d probably offer $500.
That’s why they need the money. 
I missed the ‘buddy’ part, which would mean 50/50. A random stranger? I’d offer them $450, with the extra $50 to me based on careful estimation of the value of the time I spent thinking about it.
There is an economics branch of my subfield, because you have to pay up front for a quality benefit later. How to calculate the benefit is tricky. When I was at Bell Labs my group got funded by making a benefits argument, which our internal customers bought but which was mostly garbage. (The real benefits were often not captured by our proposal.) So, the application was to explain why things really got funded, how people represented the low probability that things would go wrong, and the different discount rates people use in making funding decisions (which are usually different from the official ones.) When my daughter went over the areas from Thaler’s book (she took his course at Chicago) I could see the application.
PM me with your email and I’ll send you our paper. We also have a Psychology paper, but you’ll need access to the IEEE library for that, since I don’t have the final copy.
Probably does, but my area is computer design.
When I was Intel I was in a central group in a processor design project, handling things which had to be traded off against one another (like speed and power.) I notices that different designers had very different ways of resolving the tradeoffs. This is more psychology than economics, but if we understood how to do tradeoffs better someone could save a lot of money. I haven’t been able to convince my daughter to start a project based on this; for some reason she wants to do things her adviser like which will get her her PhD faster. Where did I go wrong?
If that’s all the book said, it’s wrong.
One cent is the correct “rational” offer only if you know that the second person is also “perfectly rational.” (And those of us with real-world experience realize that’s a dubious assumption. )
Reasoning based on “I know he knows I know … he’s rational” leads to paradoxes, most famously the village of 100 adulterous couples where a newcomer tells them something they all already knew and after 100 nights all the wives kill their husbands.
I still don’t understand the people claiming 50/50 is the only “fair offer.” Why not split it three ways, including the guy under the tree? When two people come to work on my orchard and one does all the work while the other just helps drink the beverages, is it only “fair” that I pay them equally? I doubt other guy would turn down $450, but if you think he would then you’re at least consistent. But many of the 50/50 answerers didn’t say that, instead just speaking of “fairness.”
I think you have to turn in your lawyer card for that don’t you…
That’s it for me too. And I’m surprised to see some posters seriously concerned with how much more than half they could safely take.
As for what I’d take in the second position:
If I was offered less than $200 I’d probably tell the person (asshole) making the offer that neither of us would be getting any free money today.
However, if the total amount was raised from $1000 to $1 million, I’d settle for $1000. The first guy would still be an asshole, but hey no one is perfect.
Well, you may have raised a PhD student, but at least she’s not a stripper.
I was contemplating what offers I would accept if I were the second guy, and I think I’d turn down anything less than 450. Sure, I can handle a little bit of gamesmanship from the other guy, but if he's going to screw me, I'm rather us both get nothing. I'd laugh at anyone that thinks that they could get away with an offer of .01, unless we assume that both parties are perfectly rational constructs of game theory.
Although I think my sense of pride would probably go out the window as the amount of money went up. Change the total amount to a million dollars and I wouldn’t hesitate accepting a 95/5 split.