Estimating Policy Trajectories During the Financial Crisis
NLP Unshared Task in PoliInformatics 2014
5 Pages Posted: 9 Jun 2014
Date Written: June 26, 2014
Abstract
We apply text-matching techniques to trace the trajectory of policy ideas contained in four bills related to the Financial Crisis during the 110th (2007-08) and 111th (2009-10) Congresses. By identifying the first appearance of bill text, visualizing the results, and constructing metrics to quantify the congressional “consideration time” of a bill’s ideas, our analysis reveals that two of the four bills were dominated by ideas that were first introduced many months before their eventual passage, while the other two bills contained mostly new text and were truly novel responses to the Crisis. In addition, we also apply the method to find policy ideas related to the Financial Crisis that were not included in successful bills. We suggest possible applications by both researchers and open-government advocates.
Keywords: financial crisis, public policy, natural language processing, analytics, machine learning, data mining
JEL Classification: G18
Suggested Citation: Suggested Citation