Power-Sharing, Presidential Style: Issue Salience and Portfolio Allocation in Multiparty Presidential Systems

• Published date: 01 June 2023
• DOI: http://doi.org/10.1177/10659129221119205
• Author: Mariana Batista
• Date: 01 June 2023
• Subject Matter: Articles

Article

Political Research Quarterly

2023, Vol. 76(2) 915–930

© The Author(s) 2022

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/10659129221119205

journals.sagepub.com/home/prq

Power-Sharing, Presidential Style: Issue

Salience and Portfolio Allocation in

Multiparty Presidential Systems

Mariana Batista

Abstract

Minority presidents form coalitions to secure legislative support. However, such support comes at the policy cost of

delegating power to coalition partners. Given this dilemma, how do presidents decide on portfolio allocation? The

argument is that presidents consider ministries’policy relevance and the characteristics of coalition parties when

distributing positions. The president will allocate the most important ministries to aligned ministers (seeking policy

control) and to the larger coalition parties (seeking legislative support), distributing the least policy-relevant ministries to

clientelist parties. The article presents an original measure for policy salience based on topic modeling of presidential

addresses and conditional logit models to test the hypotheses. The analysis consists of a comparative study of Brazil,

Chile, and Colombia over 24 years. Results indicate that presidents use salience to sort ministries and decide portfolio

allocation. This evidence for the president-specif‌ic salience argument shows the relevance of considering the qualitative

aspects of portfolio allocation in multiparty presidential systems and how presidents share power with coalition partners

without losing control over governmental decisions.

Keywords

presidents, coalitions, portfolio salience, portfolio allocation

Introduction

The combination of presidentialism and multiparty sys-

tems creates considerable challenges for governability in

young democracies, especially in Latin America (Linz,

1990). The solution to the gridlock of multiparty presi-

dentialism lies in coalition governments. Like their peers

in parliamentary systems, minority presidents would seek

to cooperate with the legislature through the formation of

shared governments. This means exchanging cabinet

positions for legislative support (Cheibub, 2007;

Figueiredo & Limongi, 2001).

However, much of the literature on coalition govern-

ments in presidential systems focuses exclusively on the

legislative dimension of their functioning. A second di-

mension is often dismissed: the governance dimension

(Mart´

ınez-Gallardo & Schleiter, 2015). When presidents

decide on coalition formation, they have in mind not only

the potential for legislative support but also the policy

costs of delegating power to coalition partners that, by

def‌inition, do not share the same preferences as the

president. Thus, the main question is how presidents

decide on portfolio allocation.

This article argues that presidents will distribute part of

the government to coalition partners while retaining an-

other part to be handed over to ministers from their party

and non-partisans (aligned ministers). However, this

distribution is not carried out carelessly. The qualitative

dimensions of ministries are considered—their policy

salience or importance to the government agenda—so that

the logic of portfolio allocation in coalitional presi-

dentialism is the art of putting the right people in the right

places. Thus, the president is expected f‌irst to allocate the

ministries most important to their policy agenda to aligned

ministers to keep control over governmental decisions.

Second, considering the need to build legislative support,

Department of Political Science, Federal University of Pernambuco

(Universidade Federal de Pernambuco), Brazil

Corresponding Author:

Mariana Batista, Department of Political Science, Federal University of

Pernambuco (Universidade Federal de Pernambuco), Rua Acadˆ

emico

H´

elio Ramos –s/n –14

o

andar, Cidade Universit´

aria –Recife, PE –Brasil

CEP 50670-901, Brazil.

Email: mariana.batistas@ufpe.br

the president will share valuable posts with strong coa-

lition parties that contribute the most to the coalition. Last,

legislative parties are not all alike, and the president will

allocate peripheral cabinet positions to clientelist parties

that are rewarded with the perks of off‌ice.

To test this argument empirically, I propose a com-

parative analysis of Brazil, Chile, and Colombia, between

1990 and 2014. The choice is based on the 100% prev-

alence of coalition governments in these countries. This

analysis’sf‌irst contribution is constructing an original

measure of portfolio salience based on the topic modeling

of the presidential state of the union addresses. To code

these texts, I use a supervised learning technique to

identify the thematic emphases and their prevalence in

texts. This measure is dynamic and country-specif‌ic,

representing the f‌irst measure of portfolio salience in

the comparative analysis of presidential systems. Using

this indicator, I use conditional logit models to test the

president-specif‌ic salience argument proposed in the

article.

To further specify the analysis, robustness checks

consider an alternative measure of portfolio salience,

consisting of another original indicator based on the

classif‌ication of the ministries responsible for policy

coordination or the governmental core. Second, I further

specify the meaning of “aligned ministers,”analyzing the

appointment of ministers from the president’s party and

non-partisans separately. I also present the results per

country, identifying country-specif‌ic patterns that show

presidents may emphasize different characteristics of

parties in their calculus in different countries. Considering

all these specif‌ications, the results show evidence that the

salience presidents attach to specif‌ic policy areas signif-

icantly impacts the allocation of ministries inside a coa-

lition government. This evidence for the president-specif‌ic

salience argumenthighlights the importance of considering

the qualitative aspects of portfolio allocation in multiparty

presidential systems, the central role that presidents play,

and how they manage to share power with coalition

partners without losing control over governmental

decisions.

Taken together, the analyses offer three main contri-

butions to the debate about how coalitions work in

presidential systems. First, the article proposes a sys-

tematic comparative analysis of portfolio salience based

on their importance to the president’s policy agenda. The

measure is original and innovative both in the study of

presidential and parliamentary systems because it is dy-

namic and based on the agenda’s content. Second, I show

how presidents decide on portfolio allocation in multi-

party presidential systems. They consider the qualitative

attributes of the posts and the characteristics of the parties,

prioritizing high-salience positions to aligned ministers

and strong coalition parties. In contrast, clientelist parties

areofferedtheoff‌ice perks of cabinet participation in a

secondary role. Third, I also show country-specif‌ic

patterns and how the different characteristics of the

choices (parties) matter in different countries. In sum, the

article shows important differences in how presidents

and parties coalesce, considering the institutional design

of presidentialism, the central role presidents play in

portfolio allocation, and country-specif‌ic political

contexts.

The remainder of the article is organized as follows: the

next section presents the literature on portfolio allocation,

and then the following section develops the argument

about who gets what in coalitional presidentialism. The

fourth section offers an original measure of portfolio

salience based on topic classif‌ication of the state of the

union addresses. The f‌ifth section presents the effect of

portfolio salience and the characteristics of coalition

parties on the president’s allocation decision. The f‌ifth

section presents the robustness checks, focusing on the

classif‌ication of core ministries as an alternative measure

of portfolio salience and distinguishes the appointment of

ministers from the president’s party and no-partisans. The

sixth section explores country-specif‌ic patterns. Last,

there is a conclusion.

Previous Research on Portfolio Allocation

Coalition politics is a central aspect of how democratic

governments work. It def‌ines who will be responsible for

making policy decisions. For this reason, there is so much

attention to the topic with a focus on who gets what in

portfolio allocation. There are two main ways to analyze

power distribution in coalition governments. One explores

quantitative aspects of the distribution of posts, em-

phasizing the fair sharing of a prize (Gamson, 1961;

Browne & Franklin, 1973)ortheformateuradvantage

(Baron & Ferejohn, 1989). The second explores the

qualitative aspects and how portfolio allocation connects

with policymaking. This article contributes to the second

approach.

The emphasis on what happens after portfolio allo-

cation or how governments actually govern (Laver &

Shepsle, 1990) makes the allocation of ministries stra-

tegic from a policy inf‌luence perspective. In a coalition

government, where multiple parties and therefore multiple

preferences are represented, who controls which specif‌ic

ministry becomes critical. Not only how many posts a

party gets is relevant, but which ministries and their policy

attributes (Laver & Shesple, 1990). For this argument,

portfolio salience or their qualitative aspects becomes

fundamental.

The issue of portfolio salience was initially explored by

assigning weights to ministries based on the emphasis that

the policy issue they control receive from political parties

916 Political Research Quarterly 76(2)