Deescalating Trade Dispute: Conjoint Experiments in Japan and Korea

• In Event: Experimental Approaches to IPE: Evidence from Survey and Field Experiments

Abstract

Political gains for leaders from the rally-around-the-flag effect tend to be fleeting (Tomz 2007, Weiss 2013). Yet, hawkish public attitudes against an external enemy might last and make it difficult for leaders to de-escalate and exit costly conflicts. This should be particularly the case for a conflict between traditional enemies/rivals (Haynes 2017) over an issue loaded with a sense of nationalism (Incerti et al. 2020, Tanaka 2016). Leaders on both sides would find it politically risky to make a first conciliatory move lest their hawkish voters punish such a move. Leaders who are (intentionally or unintentionally) embroiled in the rally effect would subsequently need a well-timed and face-saving exit plan. How might the government tame its hawkish public and exit the dispute without political repercussions?

To answer the question, we work with the assumption that the public reacts to international disputes strategically (Colaresi 2007). If so, there exist some conditions under which even the most hawkish citizens become more accepting of the de-escalation, rather than categorically rejecting conciliatory policies. Building on the reputation mechanism posited by the audience cost theory (Fearon 1994), we argue that citizens rallying around the flag care about the country’s reputation and show a strong aversion to a de-escalation that may tarnish their country’s honor. The hawkish public is more likely to accept de-escalation when they perceive that their government is strategically at an advantageous position, which can deliver a proud exit.

We posit that two kinds of information about the international dimension of a dispute can be used by the public to infer relative bargaining positions. First, the information about the adversary’s public opinion may contribute to citizens’ strategic evaluation of bargaining positions. Second, information about international actors’ authoritative endorsement or accusation should also affect the public’s assessment of bargaining positions and associated reputation costs. More specifically, they should perceive more reputation costs to de-escalate when authoritative international actors endorse the opponent and condemn their government, as this would look like they admit their fault.

We test the empirical implications of our argument by employing original paired surveys in Japan and South Korea. The surveys were administered by Dynata in August 2019; we collected about 1,200 randomly chosen participants from their panel in each country (i.e., a total of about 2,400 respondents). The surveys reveal that the dispute features hawkish citizens on both sides. We investigate the conditions under which hawkish publics become more accepting of a conciliatory move by their own government.

Our empirical analysis adopts several innovative approaches. First, we examine how public opinion reacts to information about the international dimensions of the ongoing dispute. We do so by employing conjoint analyses that randomly scramble multiple attributes including the rival state’s public opinion and international organization’s stance on the dispute along with other conventional factors such as material costs of the dispute. Second, we conduct paired experiments in both countries at dispute. Policy implications derived from studying only one country can be misleading if the same international information shapes public opinion differently in both countries. Our approach allows us to explore whether and how public reactions to the same type of information vary in Japan and Korea.

To preview, we find surprisingly symmetric results in both countries that are consistent with our argument. Citizens in both countries are more willing to de-escalate when told that the opponent’s public is calling for reconciliation, and when authoritative international actors (the WTO and the US in our experiment) endorse their own government’s action or condemn the opponent’s. In addition, we find that, in both countries, the dispute’s costs on their own economy decrease support for reconciliation while the costs on the adversary's economy increase acceptance of reconciliation. This seemingly counter-intuitive finding is actually consistent with our argument that hawkish citizens care about their government’s relative bargaining position when deciding about reconciliation. Citizens show reluctance to accept a de-escalation when they learn that their country is at an economically disadvantageous position.

Our findings shed light on why some disputes among liberal democracies prolong despite high material costs and little tangible political rewards on both governments. Political risks of de-escalation loom large as free media in both countries report the hawkish rally in the adversary country and growing economic costs from disputes, which grab viewer/reader attention. In light of our findings, we suggest how governments under such circumstances might be able to coordinate an exit plan.

Authors

• Sijeong Lim, Korea University

• Seiki Tanaka, University of Groningen