Helping a Friend or Supporting a Cause? Disentangling Active and Passive Cosponsorship in the US Congress

In the U.S. Congress, legislators can use active and passive cosponsorship to support bills. We show that these two types of cosponsorship are driven by distinct motivations: backing political allies versus supporting bill content. To analyze this, we develop an Encoder+RGCN (Relational Graph Convolutional Network) model that learns contextualized legislator representations using bill texts and speech transcripts. Our model predicts active and passive cosponsorship with an F1-score of 0.88. We further show that these representations generalize well to the task of predicting voting behavior and are interpretable, revealing key drivers behind legislative behavior.

In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics