Presidential Power and Party Strength: The “Inverse Relationship” Reconsidered
| Author | Timothy J Power,David Doyle |
| DOI | 10.1177/1478929919862431 |
| Published date | 01 February 2020 |
| Date | 01 February 2020 |
| Subject Matter | Special Issue Articles |
https://doi.org/10.1177/1478929919862431
Political Studies Review
2020, Vol. 18(1) 108 –124
© The Author(s) 2019
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/1478929919862431
journals.sagepub.com/home/psrev
Presidential Power
and Party Strength: The
“Inverse Relationship”
Reconsidered
David Doyle and Timothy J Power
Abstract
This article revisits theoretical claims that separation of power systems will display an inverse
relationship between presidential power and party strength. Shugart and Carey posited this as
a general equilibrium within presidential democracies; later, Shugart proposed that the inverse
relationship resulted from founding institutional choices shaped in turn by modes of democratic
transition. We test both the equilibrium and the genetic versions of the inverse relationship on
a large-N dataset of presidential systems from 1900 to 2016. We find strong significant empirical
support for an inverse relationship between executive power and party system institutionalization.
However, the relationship is more dynamic than would be predicted by theories based on
institutional foundings alone. Through analysis of decree reform in Argentina and Brazil, we
observe changes in the expected inverse relationship despite the absence of a constitutional
assembly and the lack of any change in electoral rules or party authority.
Keywords
presidentialism, parties, executive power, decrees, democratization
Accepted: 19 June 2019
What is the relationship between presidential power and political parties? Among the
many memorable claims of Matthew Soberg Shugart’s and John M. Carey’s (1992)
Presidents and Assemblies is that modern presidential democracies will exhibit an inverse
relationship between the extent of executive power and the strength of parties. This
hypothesis first emerged in the context of the authors’ well-known “inefficient secret”
argument (chapter 9 of Presidents and Assemblies); later, it was refined and proposed as
a more formal theory—linked even more explicitly to processes of political democratiza-
tion— in a separate article by Shugart (1998). For our purposes here, we will consider
Department of Politics and International Relations, University of Oxford, Oxford, UK
Corresponding author:
Timothy J Power, Department of Politics and International Relations, University of Oxford, Oxford OX1
3UQ, UK.
Email: timothy.power@lac.ox.ac.uk
862431PSW0010.1177/1478929919862431Political Studies ReviewDoyle and Power
research-article2019
Special Issue Article
Doyle and Power 109
these two works jointly. In this article, and following the other contributions in this
Special Issue, we revisit Shugart and Carey’s arguments 25 years later. We ask two broad
questions about the purported inverse relationship between executive strength and party
strength. First, if we expand the universe of cases beyond the relatively limited spatial and
temporal coverage of Presidents and Assemblies, does this relationship still find empiri-
cal support? Second, if the inverse relationship is confirmed, does it in fact result from the
reasons proposed by Shugart and Carey?
In the first section of the article, we briefly summarize the logic behind the hypothe-
sized inverse relationship. In the second section, we subject the argument to empirical
diagnostics in two ways, first in the equilibrium version (i.e. the inverse relationship is a
general pattern of presidentialism, as per Shugart and Carey, 1992) and second in its
dynamic version (i.e. the inverse relationship emerges due to specific characteristics of
the mode of transition to democracy, as per Shugart, 1998). In doing so, we extend the
temporal domain to well over a century of experience with the separation of powers
(1900–2016). In the third section of the article, we reflect upon alternative theoretical
accounts of an inverse relationship between presidential power and party strength. We
conclude by pointing to cumulative advances in theory-building, noting that recent work
on “presidentialized parties” both extends and clarifies some original insights between
party systems and executive type.
The Inverse Relationship in Brief
Shugart and Carey were heavily influenced by a work published several years earlier,
Gary Cox’s (1987) The Efficient Secret, which offers an historical account of how nascent
British parties came to articulate national policies despite an electoral system that would
predict extreme parochialism. For reasons that we will not recount here, Westminster
parliamentarism gradually became “electorally efficient,” meaning that voters were
offered clear alternatives in national policies. In chapter 9 of Presidents and Assemblies,
Shugart and Carey summarize Cox, reconsider the reasons that might encourage local
representatives to adopt a national outlook, and test the portability of these factors to a
presidential context. They conclude that under the separation of powers, “members of
congress need not submerge their district interests beneath a national focus in order to
maintain an executive in office” (1992: 170). This is the core of the “inefficient secret.”
As Amorim Neto and Santos (2003: 449) summarize the basic claim, “presidential
democracies in which legislators have a parochial focus of representation are electorally
inefficient because voters are not offered highly identifiable choices over national
policies.”
At first glance, the inefficient secret seems to bear little relationship to the “inverse
relationship” argument, yet in fact the former is a prequel to the latter. The inefficient
secret has clear implications for each pole of the inverse relationship, that is, the execu-
tive strength dimension and the party development dimension. If elected legislators are
preoccupied with localized, pork-barrel politics, they should be content to delegate
national policy making to the directly elected president, leading over time to an expansion
of presidential agenda-setting powers. Second, if parties are not required to maintain a
government in power, they will tend toward weaker organization and lower discipline (in
such cases there is no functional need for party leaders to have strong control over the
rank and file).1 Testing these arguments on 17 cases, Shugart and Carey (1992: 177) find
that where party organizational strength is weak, the legislative powers of presidents tend
To continue reading
Request your trial