The Cost-Capitulation Paradox: Compellent Signals, Sequences, and Selection Effects

Why do some militarized threats compel opponent states to change their policy or behavior while others do not? Despite a number of game theoretic models and public opinion survey experiments, few studies evaluate the historical record of compellent signaling. An original data set compiled by analyzing over 124,000 archival documents catalogs every compellent military mobilization and verbal threat by the U.S. President since strategic parity in 1970. The results call into question theories regarding “hand-tying” and “cost-sinking” signals, reputation, and assurances. Moreover, they challenge the conventional “costly” signaling framework. Instead of a simple, positive relationship between a signal’s cost and its coercive utility, the data indicate a different dynamic: at some point, greater cost is associated with a diminished likelihood of capitulation. Case studies of compellence prior to the First Gulf War and the 2003 Iraq War suggest that this apparent incongruity may be explained by the sequencing of very costly signals. By the time they are used, compellence might already be likely to fail, and these signals may therefore be employed primarily to serve other objectives, including preparing for war, legitimizing intervention, and conveying a sense of “last resort.” Policymakers should consider that effective compellence is rare and that using costly signals to demonstrate a commitment to the use of force might not make capitulation more likely.