Effects of Ballot Type and District Magnitude on Local Public Goods Bill-Initiation Behavior: Evidence from Honduras

• Author: Juan Muñoz-Portillo
• Published date: 01 June 2021
• DOI: 10.1177/1065912920905885
• Date: 01 June 2021
• Subject Matter: Articles

https://doi.org/10.1177/1065912920905885

Political Research Quarterly

2021, Vol. 74(2) 388 –402

© 2020 University of Utah

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/1065912920905885

journals.sagepub.com/home/prq

Article

Introduction

The study of electoral systems is at the heart of political

science, and this is for good reason. It is believed that

through different mechanisms, they affect how democra-

cies work and contribute to shape public policies (Chang

et al. 2011; Lijphart 1999; Lizzeri and Persico 2001;

Milesi-Ferretti, Perotti, and Rostagno 2002; Persson and

Tabellini 2003). In spite of being one of the most studied

fields in political science, we still lack an understanding of

the consequences of these formal institutions. Particularly,

a theorized relationship between proportional representa-

tion (PR) electoral systems and legislators’ personal vote-

seeking behavior,1 through activities such as the provision

of local public goods (LPG), remains understudied

(André, Depauw, and Shugart 2014; Colomer 2011).

In the American politics literature, the single-member

district plurality rule system is a frequent culprit of pork-

barrel behavior by members of the U.S. Congress.2 This

literature is both quite old and extensive (e.g., Ansolabehere,

Gerber, and Snyder 2002; Cain, Ferejohn, and Fiorina

1987; Fenno 1978). However, in the past three decades,

scholars have become increasingly interested in explaining

whether PR has an effect on how legislators represent the

interests of their constituencies (Colomer 2011; Edwards

and Thames 2007; Hallerberg and Marier 2004; Shugart,

Valdini, and Suominen 2005; Wessel Tromborg and

Schwindt-Bayer 2019).

To a great extent, this interest has been motivated by

Carey and Shugart’s (1995) seminal contribution. PR sys-

tems consist of multimember constituencies. This means

that the number of seats per constituency—that is, district

magnitude (M)—is greater than one. PR systems also use

either candidate-centered or party-centered ballots. Carey

and Shugart (1995) suggested that variations in the ballot

type and M interaction can produce diverse incentives on

legislators’ personal vote-seeking behavior. Scholars

studying different forms of relationship between constitu-

ents and legislators have found that Ballot Type × M

interaction affects the behavior of legislators and voters

significantly, thus following the personal vote-seeking

incentive distribution expected by Carey and Shugart

(André, Depauw, and Deschouwer 2014; André, Depauw,

905885PRQXXX10.1177/1065912920905885Political Research QuarterlyMuñoz-Portillo

research-article2020

1University of Costa Rica, San Jose, Costa Rica

Corresponding Author:

Juan Muñoz-Portillo, School of Political Sciences, Faculty of Social

Sciences, University of Costa Rica, San Jose, Costa Rica.

Email: juanmanuel.munoz@ucr.ac.cr

Effects of Ballot Type and District

Magnitude on Local Public Goods Bill-

Initiation Behavior: Evidence from

Honduras

Juan Muñoz-Portillo1

Abstract

An influential literature predicts that incentives to provide local public goods are conditioned by how electoral

systems expose a legislator to the need to seek a personal vote. Carey and Shugart theorize that district magnitude

and ballot type interact affecting the legislators’ personal vote-seeking behavior. Another literature challenges the idea

that electoral systems affect the behavior of legislators, particularly in highly clientelist settings, usually associated with

high poverty. I empirically evaluate these arguments on an original data set of local goods bills presented by legislators

of the National Congress of Honduras between 1990 and 2009. Honduras changed its electoral system from closed-

list to open-list in 2004 while keeping its district magnitude constant. The results suggest that the Ballot Type ×

District Magnitude interaction does not affect the behavior of legislators in small magnitude constituencies, where

poverty is more significant. However, support for the hypotheses is found in the largest, more developed constituency

where M is equal to twenty-three seats.

Keywords

clientelism, electoral systems, Honduras, legislative studies, personal vote, political economy