I am assuming that at least some of you have taken an introductory course in microeconomics. If so, maybe you can relate to the scenario that I face each year. My class goes as follows: supply and demand, social welfare and the effects of regulation, externalities and public goods, firm behavior, consumer choice…and then game theory. I introduce the concept of strategic interaction, where the best choice that you can make is dependent on what others choose to do, and we talk about the Prisoner’s Dilemma. My class pretends to be reasonably interested, even though (as I’m told later) they are actually wondering what brand of jeans I am wearing, and I’m going through the lecture thinking “Well, this is fun and all, but how on earth do I get this to tie back to the nuts and bolts of economics? Why is this important to understand when what we are concerned about is studying markets?”
Part of the problem with the literal prisoner’s dilemma example is that, well, it’s talking about prisoners. And I don’t know your proclivity for incarcerable behavior, but personally I don’t find the prisoner context to be particularly relevant to my daily life. What the economics textbook fails to make clear is that a. the prisoner’s dilemma setup is all over the place in the real world, and b. the suboptimal outcome of the game is what actually happens in a lot of markets due to coordination failure. (Sidenote: that link is less than stellar in giving insight into what a coordination failure is, so click at your own risk.)
The example I give in my class is of Coke and Pepsi trying to decide on their advertising budgets. Consider the following setup:
- Coke and Pepsi can choose either a high ($2 million) or low ($0) budget.
- If both companies choose a low budget, they each make $10 million.
- If one company chooses a high budget and the other a low budget, the company that chose high gets $15 million while the company that chose low gets $3 million. (It makes sense for the advertising to steal away some customers, right?)
- If both companies choose a high budget, they each make $8 million. (The advertising by the two firms gets cancelled out, so there is no shift of customers, but each company has spent $2 million.)
Clearly, the best outcome collectively for the two companies is to agree to not advertise. (I will ignore for the moment that such agreements are not legal in the US.) But what will happen in practice?
Let’s pretend we’re Coke for a second. (We could pretend we’re Pepsi, since the payoffs are symmetric, but then we’d, well, be Pepsi.) What should we do? To answer this question, we need to think about the competition. If we think that Pepsi is going to choose a low budget, we want to choose a high budget, since $15 million is better than $10 million. If we think that Pepsi is going to choose a high budget, we still want to choose a high budget, since $8 million is better than $3 million.
Well, that’s convenient, no? Our best option (high budget) doesn’t actually depend on what the competition does. Even more convenient, Pepsi’s decision-making process is exactly the same, so it’s going to want to choose high also. In the end, Coke and Pepsi each end up with $8 million.
That is somewhat of a disappointing outcome, since both companies could do better by choosing the low budget. This is the classic example of a coordination failure, since in order to get to the $10 million outcome the companies would have to either have some sort of enforceable agreement or way of punishing for not cooperating.
This concept exists in plenty of settings where there are more than two participants, which brings me to the title of this post. There are plenty of people who don’t “want” Wal-Mart to exist. Well, let me clarify- they want the low prices that Wal-Mart offers without giving up the charm of the small businesses in the downtown areas of their towns. The problem comes in when people only think about their own incentives when making a decision. If I were to think about only my own incentives, I would believe that I could get away with saving money at Wal-Mart and letting everyone else support the smaller local businesses. In reality, everyone faces the same (or at least similar) incentives, most people let everyone else support the local stores, and the local stores go out of businesses, even if the loss of the businesses makes the community worse off overall. (I’m pretty sure that zoning restrictions and the like make it so the nearest Wal-Mart for me is 20 miles away. If you’ve been paying attention to the site, you can consider than an effective commitment device. Or a justification for regulation, but that sounds less palatable.)
There are lots of examples of this sort of behavior, which are generally lumped under the heading of market failures. The main thesis is as follows: free markets can sometimes result in suboptimal outcomes when the happiness of individuals is dependent not only on their own choices but also on the choices of others that they have no control over. (There are, of course, other reasons why free markets can result in suboptimal outcomes.) If this sort of thing interests you, there is in fact a book titled No One Makes You Shop at Wal-Mart: The Surprising Deceptions of Individual Choice that gives a fairly non-technical overview of this sort of problem. For now, here’s another example:
I have a friend who used to work for the Red Sox. The other day he made a comment about not wanting to go to Fenway Park and sit among the “white collar d-bags” (his words, not mine) who knew nothing about baseball and were just there because it was the cool thing to do. In this way, his decision regarding whether to watch a Red Sox game is not independent of the choices of others around him. (Note that we usually assume in Econ 101 that there is some objective utility that a good or service has, and we don’t allow for that utility to change based on other people’s choices. Clearly the basic model needs to be modified in order to accurately represent this type of setup.) Because this is Boston we’re talking about, I would venture to guess that there are a bunch of other people that feel the same way. Unfortunately, since for the most part we cannot control the behavior of others (demanding girlfriends and mothers-in-law perhaps being the exception here), we have to take the actions of others as given. And given that the real Red Sox fans are sitting in their homes in Quincy drinking their Dunkin Donuts, Sam Adams and Guinness and perhaps snacking on some lobster rolls, the best option for my friend is to stay home also. (There, did I come up with the most egregious stereotype ever? It’s just because all I can think about right now is DD iced coffee.) However, IF there was a creative way to organize these fans, they could take over the park and get to an arguably better outcome.