Underpowered Power Law Research and the Decline-of-War Question (2024) - SUPERCEDED BY Beyond Significance Testing: Can the Correlates of War Data Establish Stability (2026)

Clauset (2018) and Braumoeller (2019) find that the statistics of interstate war after the Second World War cannot be distinguished from what preceded them, widely received as showing that nothing changed. We ask the complementary question: can stability be established? Applying equivalence tests to the same data, we find it cannot — neither for the power-law exponent of war sizes nor for the onset rate. The interval on the change in the exponent is four times the width of the loosest bound we would defend, spanning a fourteenfold rise and a thirty-threefold fall in the probability of a catastrophic war. Waiting cannot resolve the size question at all, and the frequency question only over several centuries.