In international bargaining situations with multiple actor – such as the climate change negotiations – states do not fight alone for their preferences. Instead, they have various ways to support each other, form coalitions, reinforce each other’s view, etc. when they share common negotiations positions. It can thus be said that negotiating countries in the climate negotiations generate cooperative networks. In this paper, we capture such networks using one form of cooperation: joint official statements of two (or more) countries during the negotiations. Based on a novel dataset coded from the Earth Negotiation Bulletins, we build such cooperative networks for all negotiation periods since 1997. Thus, we are on the one hand able to analyse cooperative patterns in the various negotiation periods, but also how these patterns evolved over time when different topics gained (or lost) in importance. The aim of this paper is, in a first step, to provide an overview of these shifts in cooperation patterns descriptively. In a second step, we employ temporal exponential random graph models (TERGMs), a relatively novel approach to model networks which allow the researcher to formulate hypotheses, and to test them on the repeated networks which serve as the dependent variable. In this part of the paper, we test how various country characteristics influence the cooperative behaviour, but also whether cooperative patterns, once established, are “sticky” and persists over time (when controlling for other factors).
