Game theory and value-based strategy

Harborne "Gus" Stuart explains why cooperative game theory is less a tool than a mind-set and how business leaders can use it to create and capture value for their companies.
June 15, 2005
Print this page

Can you briefly explain the concept of value-based business strategy?

It sounds simple, but basic economics doesn’t talk much about what the potential economic value is, what the potential pie is. And it really doesn’t account for the fact that business is a creative venture. The key insight of value-based business strategy was trying to figure out who might be the relevant people — identifying the players in the game. Once you do that, you can think about the possibilities for creating value. That’s what cooperative game theory does; rather than trying to figure out moves and countermoves, it just says think broadly about who’s in the game, and then think about the possibilities.

The negotiation people have known about this for years. If you’re trying to fight City Hall, as it were, the trick is to think broadly about who the players are. Negotiation people talk about a party diagram. It turns out that this is the core idea for any sort of business venture: think broadly about everybody involved — not only customers, but different kinds of suppliers and other people you can work with such as complementors. That was a big insight. The next thing that people worry about is how much are they going to get? They typically answer with, “What should our pricing be?” — and again, that is a limiting worldview.

A better way to answer the question is to do the experiment of pulling yourself out of the game and seeing how much the pie shrinks by. That’s your added value. It’s completely intuitive. The foundation of this idea’s been around for a hundred years — although somewhat ignored — in the notion of unrestricted bargaining. All we did was formalize it in a business strategy context. Added value is the key concept. What do you bring to the party? To determine your own added value, you have to think about how other people value you.

How does your research on game theory relate to value-based strategy?

My colleague Adam Brandenburger and I set out to figure out what game theory was relevant to business. The book that started the game theory field off in 1944 was Theory of Games and Economic Behavior by von Neumann and Morgenstern, and if you look at that book, it’s mostly about cooperative game theory. The other clue we had was that frameworks were wildly successful with MBA students, and yet most economists had a hard time teaching these frameworks. A framework can be anything from a checklist or a two-by-two matrix to Porter’s five forces. Students love these frameworks. Even if the theory behind them isn’t great, they’re an attempt to make the ideas accessible.

What these frameworks were about was thinking structurally: a large, big-picture view. The right structural approach turned out to be cooperative game theory, and so we started doing that. Ultimately what grew out of that was, first, a way to think about strategy. Adam Brandenburger ended up popularizing those ideas in a book with Barry Nalebuff called Co-opetition. One of their contributions was to think broadly about the kinds of players, in particular complementors, because you can’t even begin to think about business strategy without figuring out who’s in the game.

The second big thing was to think about what the pie could be. The strategy wasn’t about beating the competition. It was about first understanding what the opportunities were for value creation and then worrying about whether you’re getting your share of the pie. So the competition is, almost logically, second. If you pick up a traditional microeconomic textbook, it’s about the allocation of scarce resources, that people face trade-offs — very much a narrow worldview of a limited pie — whereas the reality is that business is about making deals happen.

What is the difference between cooperative, noncooperative and biform game theory?

The first question is, what’s the right way to describe a competitive situation, to look at a situation and answer the question, Should you be making money? It turns out that looking at cooperative games is the right way. Even though most of von Neumann and Morgenstern’s book was about cooperative game theory, the idea that originally appealed to people was so-called noncooperative game theory. In noncooperative game theory, you’re actually specifying the moves and countermoves. The problem is that in business the things you can do are almost beyond comprehension. The strategy set is unlimited.

And yet there are times when you can talk about a strategy set. If you’re negotiating something, you don’t want to specify the strategy set. Negotiation is too free-form. But there are times when you can talk about a specific decision: to build a plant or not, for example, to enter a market or not. So those had to be integrated. The general insight that you needed cooperative game theory for the complexity of business but that you also needed some set of specified moves precipitated the hybrid game, called the biform game.

Could you give a specific example of how a company might use biform game theory in its strategic planning?

In a recent version of the “Biform Games” paper, we look at a simplified version of the “Intel Inside” campaign. If you want a simple model explaining why the campaign would work, then the standard approach would be a game theory model that would say, OK, we spend a certain amount of money on branding our microprocessor and then start talking about how each player will change its prices. Compaq will change its prices or IBM will change its prices based on whether they have it in their computers or not. That game gets really confusing really quickly.

If you take the biform model, all you really have to worry about is, should you do the “Intel Inside” campaign or not? You don’t have to worry about the pricing strategy at all, because you’re going to look at the pie and then you’re going to look at added value and see what changes. When you do that analysis, you see that the “Intel Inside” campaign probably doesn’t increase the pie at all because Intel’s probably going to still sell to the same people.

Let’s say only IBM has the Intel chip. If Intel starts offering it to IBM’s competitor, then the competitor PC becomes more favorable in consumers’ eyes. Saying “Intel Inside” sort of elevates the lesser products. Because of that, the PC makers are competing more for the Intel chip. So with a pretty simple diagram, you can show that the reason this campaign is going to work is that it’s going to create competition among the PC makers, and therefore Intel will get a higher price for its chips. It won’t be a case of the pie increasing, but there will be greater competition for Intel’s chips, so it will get a bigger piece of the pie.

What about the results of this research surprised you?

Number one, economic value is a relative concept. Number two, ultimately it resides in people’s minds. So the idea that there’s a true economic value is nonsense. It’s always relative, it’s always contextual. This can be disturbing to a lot of people. My colleague Scott Borg came up with a wonderful comparison. Before the Copernican revolution, your position was the center of the world — it was an absolute fixed thing. Then all of a sudden you find out that you’re not the center of the world and your physical position is always relative. But that’s not disturbing to anyone. We can quite happily live with that. The fact that our physical position is relative to a frame of reference doesn’t mean that we don’t know where we are or that we get lost. The same thing can be said about economic value.

Another huge surprise that came out of this is that you should think of price as a consequence of the strategic structure. When people talk about setting prices, if they’re really thinking correctly, the right question is, what prices are consistent with the competitive situation? Thinking of price as a consequence is a huge “Aha” because the analysis allows you to understand whether you should be making money without getting into all sorts of fancy pricing models. So the whole pricing question became not a strategic question, but almost an execution question: are your prices consistent with your competitive situation?

What are the implications of your research?

It’s a lot more constructive to think in terms of value creation first because that’s an optimistic sense of possibility. As always, the business world is ahead of the academic world. Even though the terms may be sloppy in the business world, the notions are correct. You immediately think in terms of creating a market niche or coming up with a better product or creating differentiation by being a company that people would prefer to work for rather than working for someone else. Those are all ultimately ways to survive by creating value. There are still industries that are effectively zero-sum. But still, at the first cut you should always have at the back of your mind, What are the possibilities for value creation? This is consistent with what happens. Businesses are constantly trying new things, trying to create value. So you need a theory to go with it.

What’s good about our theory is that it puts the central activity of what business is about, value creation, at the center. There are a lot of buzzwords about value creation, but the fact is that every time someone does a deal, value is created. Business is about that, so the economics should be about that. The fundamental building block of what we’re doing starts out with a bilateral trade, the simplest of deals. So there is a sense that the theory is very much in sync with what people do. Another idea that pervades the research is that there’s no inappropriate certainty. A lot of times there’s a range of things that could happen, and we don’t try to say that a particular thing should happen. A lot of things are gray or indeterminate, and the theory embodies that.

What is your goal in terms of getting people to apply this research?

In a basic negotiations course, you teach people a different way of thinking and a different way of looking at the problem. That’s what we’re doing here. If game theory is done right, it isn’t about moves and countermoves. It’s a way of thinking. Students come into a negotiations course and they want tactical advice on how to get a better price on their car. They learn that negotiations are really about deals and the people involved and understanding the other players’ perspectives. In the same way, cooperative game theory just helps people to think better about their business strategy. It’s a mind-set.


Harborne Stuart is associate professor of decision, risk and operations at Columbia Business School.

Harborne Stuart

Professor Stuart teaches Managerial Negotiations and Operations Management. His research focuses on the development of business theory using game-theoretic approaches. It includes the further development of “interactive decision theory,” which takes strategic uncertainty as the primary focus, and “added value theory,” which studies businesses as the central players in economic value creation. Application of his research is principally in the...

View full profilePersonal Website

Read the Research

Adam Brandenburger, Harborne Stuart

"Biform Games"


Download PDF View abstract/citation

Adam Brandenburger, Harborne Stuart

"Value-Based Business Strategy"


Download PDF View abstract/citation