Peace agreement design and public support for peace: Evidence from Colombia

Abstract

Conflict negotiations are often met with backlash in the public sphere. A substantial literature has explored why civilians support or oppose peace agreements in general. Yet, the terms underlying peace agreements are often absent in this literature, even though (a) settlement negotiators must craft agreement provisions covering a host of issues that are complex, multidimensional, and vary across conflicts, and (b) civilian support is likely to vary depending on what peace agreements look like. As a result, we know much less about how settlement design molds overall public response, which settlement provisions are more or less controversial, or what citizens prioritize in conflict termination. In this article, I identify four key types of peace agreement provisions and derive expectations for how they might shape civilian attitudes toward conflict termination. Using novel conjoint experiments fielded during the Colombian peace process, I find evidence that citizens evaluate agreements based primarily on how provisions mete out justice to out-group combatants, and further that transitional justice provisions produced sharp divisions among urban voters in the 2016 referendum. Additional analysis suggests that material, distributive concerns were particularly salient for rural citizens. The results have implications for understanding the challenge of generating public buy-in for conflict termination and sheds light on the polarizing Colombian peace process.

Who Shot the Bullets? Exposure to Violence and Attitudes Toward Peace: Evidence from the 2016 Colombian Referendum

“…We contend that one way forward is to further disaggregate the key terms of the relationship; that is, exposure to violence and peace. Some scholars have begun to disaggregate peace, finding that, for example, the peace agreement overall was more popular than some of its specific provisions (Matanock and Garbiras-Díaz 2018;Tellez 2019a). However, few studies have attempted to unpack exposure to violence.…”

“…Given this intense campaign leading up to the referendum, the high levels of political polarization that characterized the peace process, and the complexity of the agreement (laid out in a three-hundred-page document), scholars have looked at how both the referendum campaign and the agreement's design shaped people's vote. In fact, some of the agreement's provisions and concessions (Matanock and Garbiras-Díaz 2018;Tellez 2019a) and the role of campaign information and arguments (Masullo and Morisi 2019a) have been found to have shaped Colombians' attitudes toward the peace agreement. This study takes a different perspective and explores whether exposure to violence shaped vote choice in the 2016 referendum.…”

Measurement That Matches Theory: Theory-Driven Identification in IRT Models

“…The attributes of a peace agreement, for instance, might consist of features that reflect state strength (such as disarmament) and those that reflect the strength of the rebels (such as rebel integration into the political process). Those dimensions can be politically salient: research suggests that peace agreements can experience backlash when the electorate feels that rebels are being disproportionately rewarded relative to their strength (Tellez, 2019).…”

Citizen preferences about border arrangements in divided societies: Evidence from a conjoint experiment in Northern Ireland

“…The methodological innovation that we propose in this paper is to apply conjoint analysis to identify citizen preferences about borders. Conjoint analysis has been applied to a range of questions of interest to political scientists, including preferences about the attributes of political candidates (Teele et al, 2018), immigration (Hainmueller and Hopkins, 2015), welfare policy regimes (Häusermann et al, 2019) and peace settlements (Morgan-Jones et al, 2019;Tellez, 2019), among others. In conjoint analysis, respondents rank or rate two or more hypothetical choices with multiple attributes; the objective is to estimate the influence of each attribute on respondents' choices or ratings (Hainmueller et al, 2014).…”