Barry Nalebuff
Professor, Yale School of Management

Barry Nalebuff on Punishments, Relationships and Game Theory

To embed this video, copy this code:

Yale's Barry Nalebuff discusses game theory's myriad applications.

Barry Nalebuff

Barry Nalebuff is the Milton Steinbach Professor of Management at the Yale School of Management.  Professor Nalebuff has written on a wide variety of subjects ranging from strategy to pricing, bidding to bargaining, and innovation to incentives. He is an expert on game theory and has written extensively on its application for managers.  His most recent book, The Art of Strategy, is an update of the best-selling Thinking Strategically, which explains the fundamentals of game theory using real world examples.

Professor Nalebuff's work on strategy focuses on the fundamental duality in business—the conflict between cooperating to create a pie and competing to divide it up—which he presents in Co-opetition. His book, Why Not?, focuses on providing a framework for problem solving and ingenuity. His work on product bundling was featured in the European Union's investigation of the proposed GE-Honeywell merger.


Question: Can game theory be applied daily human interactions?

Barry Nalebuff: One challenge for people is figuring out, is this relationship going anywhere? And I think the following example will nicely illustrate the strategic thinking that goes into this.

So my friend was dating this fellow who had gotten divorced, and they are living together, they were in love, they were happy. And he kept on saying that they get married, but it had been kind of a while. Could they buy a house together? Could they live in together?

They had a daughter from the previous marriage and she wasn’t really quite ready to see dad make the step. And the question was, how could she feel comfortable that they were really going to go somewhere and yet somehow not make this public? And what she came up with was the following idea that he needs to get a tattoo. It could be a discreet tattoo with her name. And if he was willing to do that, she was willing to stay with him.

And when he wasn’t, she realized that he was, I guess, the classic Peter Pan in this regard. And this illustrates the point of signaling. Signaling is taking an action which is really convincing other people, and perhaps yourself, that what you are saying is true.

So getting a tattoo is, relatively speaking, inexpensive, if you are planning to spend the rest of your life with this person. But it isn’t so inexpensive if you have other plans.

And we are always trying to go and send credible signals to other people because talk is cheap.

Another example; my own sort of theory; it’s important. Why do many women go to business school? It’s pretty expensive. It takes two years of your life. But many employers are worried that women are going to leave the career to go and have children, start a family.

And you ask the person you interview -- well, one that might be illegal to ask them -- but even if you did, they are all the same, "No, no, I’m here for the career."

On the other hand, if the person says, "I just spent $70,000 in tuition, two years of my life getting this training. I wouldn’t have done that unless I was intending to pay it back. And so the fact that I went and did this activity should be great evidence to you that I am serious about my career."

Other people will say it, I’ve taken an action which is really of that category and then the little one and this is the Hermes tie, right? I mean, why spend/waste so much money on this tie? Well because I’m going to be the person who ultimately has the job that means I’ll be wearing this thing, and it was actually going to pay it back.

Somebody else can say, oh, they are going to have this, but I’ve already proven that it’s worth my while to do this.

Question: How can game theory be applied to relationships?

Barry Nalebuff: Well, appreciating that many contracts are subject to renegotiation, and so one when you have a kid, there’s a question of who’s going to get up in the middle of the night. And then part of the problem is that, if it turns out initially it’s only the mother, then the kid only responds to the mother. And so then the father said, look, I’m just no use here.

And so understanding this is a repeated game, it turns out that unless you get the dad involved early on, the dad could make then credible that he is not involved.

This is, again, an example of looking forward, and reason backward, and understanding how it will play out.

Kids, of course, are the most natural strategists because, one, they don’t care if they are up at two in the morning; what else have they got have to do? They are bored.

Or the high school kids who want the car and you’ve got friends over, your boss over, and just thinking about making a scene then, and you don’t want that to happen, they understand that.

But the great one is the question of punishment. How do you make punishments credible? Because, you know that you don’t really want to punish the kid, it’s not your interest, and you don’t like doing that, and they know that and they are trying to take advantage of it.

So I’ll give you two fun or amusing examples of this. When one kid misbehaves, you punish the other one for it. So the kid will say that’s totally unfair, but that kid will be sure to punish the one who misbehaved.  So, Tommy, if you misbehave, I am going to punish your older brother. You know that Tommy fears his older brother much more than he ever fears anything you do, and so that will be one way of doing it.

I’ll give you an example in terms of weight loss. Two of my colleagues want to lose weight and so they bet each other $5,000 that they will get down to some weight level and that they will stay below it. Now it worked, and then after about 6 months, one of them starts creeping up a little bit. And his friend called him on the bet and took the $5,000 from him which really, really got her friend mad. In fact, may even have broken their friendship.

And when I asked what was going on? Why do you do this? And he said, well, I wanted to be sure that if I went up in my weight, he would call me, and I was afraid that if I let him off, well, then he’d let me off. But I can tell you this now, that if I go above my proposed weight, I know that he is going to call me on it because he is so mad at me for having called him. And that way of making that enforcement credible has kept my friend actually at the weight he wants.

And turning that back to kids, well, I punished the kid not because I want to, but because if I don’t, he or she will punish me for not punishing them, and so that is the type of argument you make.


Recorded on: Oct 2, 2008