Concepts and measures of bureaucratic constraints in European Union laws - from hand-coding to machine-learning

Aug 5, 2026

Scholars analyze how executive constraints embedded in European Union legislation are operationalized through two main measurement strategies: one based on variation in the use of restriction types and another based on the frequency of constraint use. These approaches represent distinct conceptualizations of bureaucratic control, termed the “toolbox perspective” and the “design perspective.” Evidence indicates the frequency-based measure yields fewer validity problems for estimating legislators’ intent and reduces measurement errors. The text assesses a recent computational model using hand-coded provisions and proposes a syntactic machine-learning framework.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Abstract

1. Introduction

2. The (lack of) debate on concepts and measures of bureaucratic control

Regulation & Governance (2024) 18, 688–703 doi:10.1111/rego.12543 Concepts and measures of bureaucratic constraints in European Union laws from hand-coding to machine-learning Fabio Franchino Department of Social and Political Studies, Università degli Studi di Milano, Milan, Italy Marta Migliorati Institute for European Studies, University of Malta, Msida, Malta Giovanni Pagano and Valerio Vignoli Department of Social and Political Studies, Università degli Studi di Milano, Milan, Italy Abstract Scholars employ two main measures of the executive constraints embedded in European Union laws: one is based on the variation in the use of different types of restrictions, and the second is based on the frequency of such use. They reﬂect two alternative conceptualizations of bureaucratic control. We label them, respectively, as the “toolbox perspective” and the “design perspective”. We illustrate that the constraint frequency measure poses fewer validity problems in estimating legislators’ intent to constrain implementation and tends to produce less severe measurement errors. We then evaluate the performance in estimating constraint variation of a recent computational application and identify potential drawbacks of automated learning from hand-coded provisions. We lastly introduce a skeletal framework for a machine-learning approach based on the syntactic structures employed by legislators that could improve the performance of this innovative technique. Keywords: bureaucratic constraints, conceptualization, EU law, machine-learning, measurement. 1. Introduction Pick up any law and you read a procedure or a prescription of some sort. They are ubiquitous. These constraints are the nuts and bolts of legislative design, and they are core components of the formal executive discretion of implementing agents. Their use reﬂects the important trade-offs legislators make between the beneﬁts of delegation and the risks of agency drift. Yet, measuring the legislative intent to delimit the room for maneuver of implementers is arduous, and doing so for thousands of pieces of legislation is treacherous. Scholars of legislative politics in the European Union (EU) have adopted different measurement strategies, some of them inspired by the literature on congressional politics in the United States. Yet, the choices scholars make and the assumptions that underlie these decisions are hardly discussed. In this article, we carry out such an examination, especially in light of ever more sophisticated procedures to extract estimates of constraints from legal texts that can scale up to large corpora in an efﬁcient and (presumably) valid manner. We ﬁrst review the different procedures scholars have employed to measure the executive constraints embedded in EU laws. Next, we argue that these procedures reﬂect two different conceptualizations which we call the toolbox perspective and the design perspective on bureaucratic control. Scholars that adopt the former employ measures that are based on the variation in the use of different types of restrictions. Those that prefer the latter use measures that are based on the frequency of use of such constraints. Which of the two provides a more valid estimation of legislators’ intent to constrain implementation? Employing the data from Correspondence: Fabio Franchino, Department of Social and Political Studies, Università degli Studi di Milano, Via Conservatorio 7, 20122 Milan, Italy. Email: [fabio.franchino@unimi.it](fabio.franchino@unimi.it)[ ](fabio.franchino@unimi.it)Conﬂict of interest: None declared. Accepted for publication 9 June 2023 . © 2023 The Authors. Regulation & Governance published by John Wiley & Sons Australia, Ltd. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. CONCEPTS AND MEASURES OF CONSTRAINTS F. Franchino e

FAQs

What are the two main ways EU scholars measure executive constraints?

One measure uses variation in the types of restrictions employed, while the other uses the frequency of such restrictions in legal texts.

How do the “toolbox perspective” and “design perspective” differ?

The toolbox perspective links bureaucratic control to variation across restriction types, whereas the design perspective links it to how often constraints are used.

Why does the frequency-based measure perform better in validity terms?

It poses fewer validity problems when estimating legislators’ intent to constrain implementation and tends to produce less severe measurement errors.

Aug 5, 2026

Upload by Seraphina