This project aims to investigate how the substance of legislative reform is shaped during parliamentary con-sideration by the salience attributed to individual proposals. The project speaks to studies that have analyzed factors determining the content of legislative proposals across their legislative life cycles (Martin and Vanberg 2005, 2014; Pedrazzani and Zucchini 2013). We suggest that while assessing the content of specif-ic legislative proposals is an important step forward in understanding the dynamics of parliamentary reform, previous work has paid insufficient attention to the importance that political actors ascribe to specific reform projects. This is to say that bargaining is not only determined by actors’ preference profiles, but crucially, by how much they believe to be at stake in individual reform proposals (Thomson et al. 2006). We argue that bargaining on individual reform proposals is heavily influenced by salience attributions, such that actors make greater efforts to shape the content of legislative reform when the stakes are high. This effect should be evident in interactions between government and opposition, but also within governing coalitions. Coalition partners try to influence the substance of reform proposals, even after they are introduced into the legislative process, particularly when they have far-reaching policy consequences. Thus, the project also contributes to attempts to study interaction patterns within coalition governments (Carroll and Cox 2012; Kim and Loewenberg 2005; Martin 2004; Martin and Vanberg 2004; Thies 2001).
To investigate the effect of political attention on the content of legislative reform, we introduce a nov-el measure of political attention that captures salience perceptions for specific legislative proposals. We study plenary attendance from publicly available video footage, ordinarily collected in legislative assemblies. Furthermore, we employ recent advances in automatic face detection and recognition to gather evidence on plenary attendance, providing a systematic measure of the importance that the assembly attributes to specif-ic pieces of legislation (cf. Poole and Rosenthal 1997; Scully 1997). This measure allows a large-n analysis of the proposal-specific attention effects on the substance of legislation, as it can be automatically generat-ed. Moreover, due to its fixed metric, it is comparable across space and time and can even be retrospectively generated. The proposed salience measure is unique in that it resides at the level of individual proposals, going beyond most previous measures of salience that have ordinarily tapped into salience perceptions at the level of policy fields. For the empirical application, we study three legislative terms of the Landtag Baden-Württemberg (2006-ongoing).