How Predictable is Your State? Leveraging Lexical and Contextual Information for Predicting Legislative Floor Action at the State Level

Proceedings of the 27th International Conference on Computational Linguistics, pages 145–160 Santa Fe, New Mexico, USA, August 20-26, 2018. 145 How Predictable is Your State? Leveraging Lexical and Contextual Information for Predicting Legislative Floor Action at the State Level Vlad Eidelman Anastassia Kornilova Daniel Argyle FiscalNote Inc. Washington DC {vlad,anastassia,daniel}@fiscalnote.com Abstract Modeling U.S. Congressional legislation and roll-call votes has received significant attention in previous literature. However, while legislators across 50 state governments and D.C. propose over 100,000 bills each year, and on average enact over 30% of them, state level analysis has received relatively less attention due in part to the difficulty in obtaining the necessary data. Since each state legislature is guided by their own procedures, politics and issues, however, it is difficult to qualitatively asses the factors that affect the likelihood of a legislative initiative succeeding. Herein, we present several methods for modeling the likelihood of a bill receiving floor action across all 50 states and D.C. We utilize the lexical content of over 1 million bills, along with contextual legislature and legislator derived features to build our predictive models, allowing a comparison of the factors that are important to the lawmaking process. Furthermore, we show that these signals hold complementary predictive power, together achieving an average improvement in accuracy of 18% over state specific baselines. 1 Introduction Federal institutions in the U.S., like Congress and the Supreme Court, play a significant role in lawmak- ing, and in many observable ways define our legal system. Thus, as data and computational resources have become more readily available, political scientists have increasingly been adopting quantitative methods focused on understanding these entities and the role they play in our society (Katz et al., 2017; Poole and Rosenthal, 2007; Slapin and Proksch, 2008; Lauderdale and Clark, 2014). Although many issues are legislated and regulated primarily at the federal level, state governments have significant power over certain areas. An increasing number of important issues are being decided at the state or local levels, especially in emerging industries and technologies, such as the gig economy and autonomous vehicles (Hedge, 1998). Moreover, there are 535 members of Congress who introduce over 10,000 pieces of legislation a session,1 of which less than 5% is enacted. Similar dynamics exist at the state level, except on a much broader scale. There are over 7,000 state legislators, in aggregate introducing over 100,000 pieces of legislation, with over 30% being enacted. In order to be enacted, every bill must pass through one or more legislative committees and be considered on the chamber floor, a process we refer to as receiving floor action. This process is one of the most pivotal steps during lawmaking (Rosenthal, 1974; Hamm, 1980; Francis, 1989; Rakoff and Sarner, 1975), as on average, only 41% of bills receive floor action, with most legislation languishing in the committees. Legislative policymaking decisions are extremely complex, and are influenced by a myriad of factors, ranging from the content of the legislation, to legislators’ personal characteristics, such as profession, religion, and …

Content truncated to 500 words