Understanding Delegation through Machine Learning: A Method and Application to the European Union

Forthcoming, American Political Science Review

59 Pages Posted: 25 Jul 2018 Last revised: 29 Jul 2019

See all articles by Jason Anastasopoulos

Jason Anastasopoulos

Princeton University; University of Georgia - School of Public and International Affairs

Anthony M. Bertelli

Bocconi University - DONDENA - Carlo F. Dondena Centre for Research on Social Dynamics; Pennsylvania State University

Date Written: July 3, 2018

Abstract

Delegation of powers represents a grant of authority by a legislature to one or more agents whose powers are determined by the conditions in enabling statutes. Extant empirical studies of this problem have relied on labor-intensive content analysis that ultimately restricts our knowledge of how delegation has responded to politics and institutional change in recent years. We present a machine learning approach to the empirical estimation of authority and constraint in European Union (EU) legislation and demonstrate its ability to accurately generate the same discretion measures used in an original study directly from all EU directives and regulations enacted between 1958-2017. We assess validity by training our classifier on a random sample of only 10% of hand-coded provisions and replicating an important substantive finding. While our principal interest lies in delegation, our method is extensible to any context in which human coding has been profitably produced.

Keywords: Delegation, Text Analysis, Machine Learning, European Union, Gridlock

JEL Classification: D72, D70, C40

Suggested Citation

Anastasopoulos, Jason and Bertelli, Anthony M., Understanding Delegation through Machine Learning: A Method and Application to the European Union (July 3, 2018). Forthcoming, American Political Science Review. Available at SSRN: https://ssrn.com/abstract=3207821 or http://dx.doi.org/10.2139/ssrn.3207821

Jason Anastasopoulos

Princeton University

22 Chambers Street
Princeton, NJ 08544-0708
United States

University of Georgia - School of Public and International Affairs ( email )

Athens, GA 30602-6254
United States

Anthony M. Bertelli (Contact Author)

Bocconi University - DONDENA - Carlo F. Dondena Centre for Research on Social Dynamics ( email )

Via Roentgen 1
Milan, 20136
Italy

Pennsylvania State University ( email )

University Park, PA 16802-3306
United States

HOME PAGE: http://tonybertelli.com

Register to save articles to
your library

Register

Paper statistics

Downloads
73
Abstract Views
409
rank
324,141
PlumX Metrics