The Sturdy Child vs. the Sword of Damocles: Nuclear Weapons and the Expected Cost of War

Abstract

Is the world better off with nuclear weapons or without? Nuclear pessimists point to the potentially devastating costs of a nuclear war. Nuclear optimists argue that nuclear weapons reduce the likelihood of war and are thus beneficial. This debate is inconclusive in part because it misses an important conceptual point. We should care both about the cost of war and the likelihood of war, as they combine to form the expected cost of war, which is the product of the two. I discuss five implications of focusing on expected costs. Three support the pessimists: (1) nuclear weapons raise the upper limit on how destructive wars can be; (2) there may be a floor on how low the likelihood of war can go; and (3) risk aversion over damage will raise the expected cost of nuclear war. The remaining two support the optimists: (4) strategic models exhibit a declining expected cost of war; and (5) casualty data show that the expected cost of war is declining over its observed range in the past two hundred years.