Nuclear game theory and its limitations
What economic theory can and can't tell us about how to avoid Armageddon
Most people who grew up in the 1980s have seen the movie War Games. In the climactic scene, an artificially intelligent computer that has been given control of the U.S. nuclear arsenal runs through a series of simulations in which it realizes that launching a global thermonuclear war against the Soviet Union would ultimately result in the destruction of both countries. When the computer realizes that the only way to win the game is not to play, the world is saved.
This is, in fact, how the Cold War ended. Realizing that nuclear war was unthinkable due to Mutually Assured Destruction (MAD), both superpowers stuck to fighting proxy wars, until the USSR’s economy collapsed and it disintegrated. In the end, the countries behaved more or less like rational actors playing a game. For this reason, game theory has always been seen as important to understanding nuclear deterrence.
Right now, as Russia’s invasion of Ukraine is stymied, there are renewed fears of nuclear war. Putin put his nuclear weapons on a somewhat heightened alert status near the beginning of the war, has repeatedly made dark warnings about the use of nukes, and is currently hiding in a nuclear bunker in the Ural mountains. A lot of effort is going into reading the tea leaves on what might prompt Putin to push the proverbial big red button, and there’s no end of people screaming on Twitter that this or that piece of U.S. assistance to Ukraine might be the thing that tips the world into Armageddon.
So it’s natural in these troubled times to wonder if game theory will help predict which U.S. actions will and won’t lead to nuclear holocaust. And indeed, game theory can help illustrate some general principles at work. But if asked to give precise answers as to what will and won’t prompt Putin to go nuclear, game theory is going to come up short.