The Rules of the Game

Yesterday, Canada announced retaliatory tariffs on more than 700 categories of US goods, matching Trump’s tariffs »dollar for dollar, rate for rate«. The measures, worth about C$27.6 billion, will take effect on 8 September. It is the familiar logic of retaliation: you hurt me, so I hurt you back.¹ The Wall Street Journal noted that Canada has targeted products from politically sensitive states, turning the tariffs into an attempt to change the incentives facing Trump and Republican politicians ahead of the midterm elections.²

»Mechanism design is constructing the rules so that breaking them stops being attractive«

That’s not to say nobody’s thought about the problem. We have. The intergovernmental World Trade Organization (WTO) and the dispute-resolution system within the United States-Mexico-Canada Agreement (USMCA) are mechanisms designed precisely to stop trade disputes becoming cycles of retaliation. They establish rules in advance for what countries can do when their interests collide. The interesting thing about the current dispute is that those rules have proved easier to walk away from than the people who designed them probably hoped.

Economists and mathematicians have thought about this problem too. Game theory asks what rational players will do when their interests collide. But another field asks a more interesting question: what if we could design the game itself?

It is called mechanism design. Its modern foundations were laid by the economist Leonid Hurwicz in the 1960s, then developed by Eric Maskin and Roger Myerson. All three received the Nobel Prize in Economics in 2007 for their work.³

The distinction is simple. Game theory takes the rules as given and asks what rational players will do as a result. Mechanism design starts with the desired outcome and asks what rules would make rational players produce it.

Canada’s retaliation contains a crude version of this idea. By targeting politically sensitive states, it is trying not simply to impose symmetrical pain but to change the incentives of the people who can change American policy. It is mechanism design in miniature – except that it’s being improvised after the game has begun. Retaliation is what you do when the other player breaks the rules. Mechanism design is constructing the rules so that breaking them stops being attractive.

We already use the principle everywhere. Laws establish punishments before crimes happen. Auctions are designed to make people reveal what they are willing to pay. Elections turn millions of private preferences into collective decisions. A mechanism is a machine for turning individual incentives into a collective outcome.

The same principle explains why the WTO exists. Countries are better off trading than fighting over every dispute, so they designed rules intended to make predictable trade the rational strategy. It works surprisingly often. But mechanisms are only as strong as their ability to constrain the players who use them. This is where the trade war becomes interesting. It’s not really an example of a world without rules. It’s an example of what happens when the players discover that the rules are optional.

And the stakes can get much higher. A tariff dispute can become a trade war. A security dilemma can become an arms race. A border dispute can become an outright war. The players and pay-offs change, but the design question remains: can we construct systems in which escalation is less profitable than restraint?

Imagine a treaty in which the economic consequences of aggression were automatic: trade privileges, investment, and development funds would flow towards the victim, while the aggressor automatically lost access to the benefits of peaceful integration. No emergency summit. No improvised sanctions. The mechanism would be designed so that everyone had something to lose before the first shot was fired.

That’s a radically different way of thinking about peace. It treats international order not simply as diplomacy or morality, but as an engineering problem. What incentives would make cooperation rational? Where should the costs of breaking the rules fall? How could the benefits of peace be distributed so widely that starting a conflict became politically and economically difficult?

We have already built mechanisms that make trade wars less likely. We have built others that make contracts enforceable, markets function, and elections possible. The great unfinished mechanism is international security: a system in which starting a war is not merely wrong, but irrational.

The point of a good political system is not to find better players. It is to make the game difficult to play badly.

References