Voting for the president: The Supreme Court during war

An extraordinary body of scholarship suggests that wars, especially major wars, stimulate presidential power. And central to this argument is a conviction that judges predictably uphold elements of presidents' policy agendas in war that would not withstand judicial scrutiny in peace. Few scholars, however, have actually subjected this claim to quantitative investigation. This article does so. Examining the universe of Supreme Court cases to which the US Government, a cabinet member, or a president was a named party over a 75-year period, and estimating a series of fixed effects and matching models, we find that during war Justices were 15 percentage points more likely to side with the government on the statutory cases that most directly implicated the president. We also document sizable effects associated with both the transitions from peace to war and from war to peace. On constitutional cases, however, null effects are consistently observed. These various estimates are robust to a wide variety of model specifications and do not appear to derive from the deep selection biases that pervade empirical studies of the courts. (JEL K0, K3, Z0).