Game Theory and International Trade

One of the most enjoyable moments while teaching Principles of Economics is the “aha!” moment that inevitably materializes when students “get” the concept of the Prisoner’s Dilemma. I rely upon a clip from “Golden Balls”, a UK game show that made use of PD games, along with the game The Evolution of Trust while teaching game theory for the first time, and also refer to Games Indians Play.

But an area in which I could get better is in showing students how game theory is used outside of these applications that are of an introductory nature. Dixit and Nalebuff’s book is of great help in this regard, and should be read by everybody regardless of whether or not you have a background in economics.

In today’s post I want to cover a chapter in a recent book that also does a good job in this regard. The book is titled “Rebuilding the Post-Pandemic Economy” and the name of the chapter is “America and International Trade Cooperation”

Consider the workhorse economic model of international trade agreements. Trade agreements are valuable because they solve what is known, in game theoretic terms, as the prisoner’s dilemma.
In such a game, each player has two choices—“cooperate” or “do not cooperate.” (The values in each box are the payoffs to each player if that is where they jointly end up.) To start, suppose there is no coordination between the players, so that each chooses its best response. The equilibrium outcome will be that each chooses “do not cooperate,” and the payoff to each is 1. But the problem with this outcome is obvious. Even though neither of them has a unilateral incentive to change its behavior, if they
both agreed to, each can be made better off and receive a payoff of 3.

Bown, Chad P., “America and International Trade Cooperation” in Rebuilding the Post-Pandemic
Economy, ed. Melissa S. Kearney and Amy Ganz (Washington D.C.: Aspen Institute Press, 2021).
Available at
Source: From the chapter referred to above

If you have learnt game theory, it should be easy for you to understand why (1,1) is the Nash Equilibrium. If you have not learnt game theory, now’s a good time to start!

What Chad is getting at is this: “Broadly speaking, these prisoner’s dilemma models can be used to characterize the WTO and its core rules.” Countries that choose to not cooperate are boxing themselves into a corner (if you’ll excuse the pun), whereas countries that choose to cooperate and enter into a trade agreement are likely to see much better outcomes.

Ah, but hang on. Two very tricky questions arise. One, better for whom, exactly? And two, what if a country signs a trade agreement but then reneges? Let’s deal with both of these in turn.

Better for whom, exactly? This isn’t a superficial, rhetorical question, and nor is it a question that is even trying to imply that international trade is “bad”. It is a question that has had an impact on US Presidential elections (and political outcomes with real repercussions elsewhere in the world), and it is also a therefore a question that has engaged the minds of some of the best economists out there in recent years. See, for example, this article, or if you’re up for it, read this paper.

Globalization hasn’t reduced inequality, it may have increased it. As with all contentious issues, so also with this one – you’ll have passionate people arguing both sides of this story, armed to the teeth with data, models and theories. My own position is that there is, alas, at least an inconvenient iota of truth to the story. (Bonus points if you got the reference)

Listen to this podcast if you want more background information, and this symposium if you want a take by four different economists.

And read this if you want to understand how policymakers, in at least the United States of America are responding to the evolution of our understanding of this issue: better for whom, exactly? Note that you don’t need to agree that the US government has done, is doing or will do enough – all I am saying is that policies such as these are an acknowledgment of sorts that the perception on the ground is that trade hasn’t helped everybody.

And now back to game theory: what if a country signs a trade agreement but then reneges?

Historically, the United States has pushed for relatively low tariffs, applied on a nondiscriminatory basis to all members of the WTO. One interpretation of the Trump administration’s tariff war is the following. Even nearly two decades after its 2001 WTO accession, China had refused to engage in additional tariff liberalization. It was deploying other policies in ways symptomatic of noncooperative play,
imposing costly externalities on trading partners. Thus, the United States imposed trade war tariffs as its best response; as a result, each country is now imposing its noncooperative policy on the other. (Both are economically worse off than if they agreed to cooperate—see again Figure 1—but the United States may now be better than off than it was when it was cooperating but China was not.)

Bown, Chad P., “America and International Trade Cooperation” in Rebuilding the Post-Pandemic
Economy, ed. Melissa S. Kearney and Amy Ganz (Washington D.C.: Aspen Institute Press, 2021).
Available at

As Chad suggests, go back to that first diagram shared above. What Chad is saying is that we need to understand that China agreed to play nice, as did America. And so we’re in (3,3), the upper left cell of the diagram. And this is obviously better than (1,1), the lower right cell.

But what if China said it would play nice, but then went ahead and played “dirty”? Then we’re in (0,5) upper right territory, and that is not a nice place to be in at all. How then should America respond? Please play The Evolution of Trust Game if you haven’t, by the way, and ask yourself which character most closely resembles China.

And then ask yourself how USA should respond, and then read the rest of this excellent chapter.

Also read the Twitter thread that Chad Bown but up about this chapter, and if you don’t listen in already, do listen to Trade Talks. Finally, here is Chad Bown’s Google Scholar page, and here is his PIIE website.