Signaling Foreign Policy Interests

TLDR

The paper distinguishes two costly signaling strategies that leaders use to credibly convey foreign‑policy interests: audience‑cost signals that bind them ex post and sunk‑cost signals that bind them ex ante. The authors analyze a simple game‑theoretic model of each signaling type to derive its equilibrium properties. The model predicts that leaders never bluff with either signal, that tying hands yields higher average payoffs despite increasing war risk, and that these results help explain why many crises involve audience‑cost competition while also generating puzzles about why inference logic sometimes fails.

Abstract

The author distinguishes between two types of costly signals that state leaders might employ in trying to credibly communicate their foreign policy interests to other states, whether in the realm of grand strategy or crisis diplomacy. Leaders might either (a) tie hands by creating audience costs that they will suffer ex post if they do not follow through on their threat or commitment (i.e., costs arising from the actions of domestic political audiences) or (b) sink costs by taking actions such as mobilizing troops that are financially costly ex ante. Analysis of a game model depicting the essentials of each case yields two principal results. First, in the games' equilibria, leaders never bluff with either type of signal; they do not incur or create costs and then fail to respond if challenged. Second, leaders do better on average by tying hands, despite the fact that the ability to do so creates a greater ex ante risk of war than does the use of sunk-cost signals. These results and the logic behind them may help explain some empirical features of international signaling, such as many crises' appearance as competitions in creating domestic political audience costs. They also generate empirical puzzles, such as why the seemingly plausible logic of inference that undermines bluffing in the model does not operate in all empirical cases.