Human rights "naming & shaming" and civil war violence

The aim of this PEPS Letter is to clarify the effects of human rights "Naming and Shaming" by international actors, such as IOs, NGOs and the international media, on the intensity of violence in domestic conflict. The note carries out, evaluates and proposes empirical strategies to study such effects. We find that a classic panel-year data structure is insufficient to disentangle and assess possible mechanisms. Though the UN-labeled attempts to "name and shame" seem to decrease the level of violence, this and other results cannot be defended as robust and coherent in light of likely endogeneity bias. Results based on just lagged explanatory variables differ in the sign and significance of relationships with results based on instrumented-variable estimation. We highlight the value and necessity of collecting and analyzing data that is more disaggregated with respect to time and target levels. Only such further steps will allow full study the role of "naming and shaming" on domestic conflict dynamics.