Imperfect Knowledge and Stable Governance in Democracies: An Agent-Based Model

DOI10.1177/1065912919834655
AuthorCarlos M. Fernández-Márquez,Luis Fernando Medina,Francisco J. Vázquez
Published date01 September 2020
Date01 September 2020
Subject MatterArticles
/tmp/tmp-18OQrHXsOK190u/input
834655PRQXXX10.1177/1065912919834655Political Research QuarterlyFernández-Márquez et al.
research-article2019
Article
Political Research Quarterly
2020, Vol. 73(3) 568 –582
Imperfect Knowledge and Stable
© 2019 University of Utah
Article reuse guidelines:
sagepub.com/journals-permissions
Governance in Democracies:
https://doi.org/10.1177/1065912919834655
DOI: 10.1177/1065912919834655
journals.sagepub.com/home/prq
An Agent-Based Model
Carlos M. Fernández-Márquez1, Francisco J. Vázquez1,
and Luis Fernando Medina2
Abstract
In this paper, we introduce an agent-based model of elections and government formation where voters do not have
perfect knowledge about the parties’ ideological position. Although voters are boundedly rational, they are forward-
looking in that they try to assess the likely impact of the different parties over the resulting government. Thus, their
decision rules combine sincere and strategic voting: they form preferences about the different parties but deem
some of them as inadmissible and try to block them from office. We find that the most stable and durable coalition
governments emerge at intermediate levels of informational ambiguity. When voters have very poor information about
the parties, their votes are scattered too widely, preventing the emergence of robust majorities. But also, voters with
highly precise perceptions about the parties will cluster around tiny electoral niches with a similar aggregate effect.
Keywords
spatial model, agent-based model, political competition, democracy stability, transparency
Introduction
voters. While risk-aversion seems the most reasonable
assumption, some papers (e.g., Morgenstern and
Although the normative theory of representative democ-
Zechmeister 2001) have found that on occasions it may
racy is premised on the assumption that parties and candi-
fail. Later, Glazer (1990) showed that even if voters were
dates put in front of the voters clearly delimited alternatives
risk-averse, if parties were uncertain about the location of
so that the latter can make informed choices, it has been
the voters’ ideal points they would endogenously choose
since long understood in political science that there are
ambiguous platforms as an optimal strategic response to
many empirical difficulties with such view. The early
the resulting lack of information. In a similar vein,
efforts at understanding the emergence and role of politi-
Alesina and Cukierman (1990) developed a model where
cal parties already concluded that, as a response to a series
ambiguity could be optimal for an incumbent wishing to
of historical and institutional factors, parties could on very
exploit the lack of information from the voters’ point of
important occasions become “catchall parties” with only
view so as to increase the chances of reappointment.
very loose ideological commitments (e.g., Gunther and
Another example of a formal model where policy ambi-
Diamond 2001; Kirchheimer 1966; Panebianco 1988).
guity results as an equilibrium choice of office-seeking
When facing highly heterogeneous electorates, a fact of
parties can be found in Aragonés and Postlewaite (2002).
increasing salience in modern democracies, even parties
As an empirical counterpart, a body of literature has
characterized by higher degrees of internal coherence
endeavored to establish the degree to which it is indeed
have incentives to broaden their appeal by blurring the
the case that a strategy of ambiguity benefits parties elec-
focus of their message (Somer-Topcu 2015).
torally, but in real-life elections it is hard to disentangle
A similar concern has been clear in the tradition of for-
mal analysis of political parties from its very outset. In
1Universidad Autónoma de Madrid, Spain
his foundational work Downs (1957) recognized that can-
2Universidad Carlos III de Madrid, Getafe, Spain
didates had incentives to remain ambiguous in their pro-
posals. Shepsle (1972) added more rigor to the original
Corresponding Author:
Luis Fernando Medina, Universidad Carlos III de Madrid, c/Madrid,
idea, by studying the interaction between the ambiguity
135, 18.2.E.12, 28903 Getafe, Madrid, Spain.
of candidates’ platforms and the risk-attitudes of the
Email: lmedina@clio.uc3m.es

Fernández-Márquez et al.
569
the effects of ambiguity from other effects, giving rise to
It has been known since long that a multidimensional
issues of mismeasurement and endogeneity. Recent stud-
policy space is the most appropriate framework to deal
ies, however, using experimental techniques have with such institutional settings so that, for instance, aside
advanced in this direction. For instance, Tomz and Van
from the typical left–right economic spectrum, voters
Houweling (2009) find that in the American context,
have, and parties represent, preferences on other issues
while ambiguity itself is rather neutral in nonpartisan
such as foreign policy (“hawk vs. dove”), culture and
contexts, it is beneficial for parties in clearly partisan sit-
family (“traditionalist vs. libertarian”), or regional auton-
uations as voters tend to display a “party optimism”
omy (“centralist vs. federalist”; Kitschelt 1995; Roemer
whereby they are willing to embrace the vagueness of
2001; Shepsle and Laver 1996). But this makes it even
their own candidate if it means defeating the other one.
harder to assess the aggregate effects of uncertainty from
Other approaches have expanded the inquiry into the
the point of view of the voters. Fortunato, Stevenson, and
sources of political ambiguity in yet another direction.
Vonnahme (2016), for instance, have found that the extent
Apart from strategic motives, parties are often institution-
and usefulness of political knowledge varies across coun-
ally bound to be ambiguous in their policy pronounce-
tries in Europe, to a large extent due to the fact that the
ments so that the extent to which candidates come across
standard left–right placement does not carry the same
as more or less vague depends on several properties of the
accuracy in every country.
political system.
Third, the government formation process itself is not
Some countries, as a result of the institutional features
automatic. The end of the vote tally is just the beginning
behind government formation, have highly disciplined
of a stage of negotiation between parties, a stage that
parties so that when voters have to evaluate their stances,
can lead to many different outcomes. On occasions, as
the labels, as well as their record in office and opposition,
Spain lately has exemplified, the result can be inconclu-
convey a lot of information. Instead, in other systems par-
sive. After momentous economic challenges that frag-
ties are much more internally heterogeneous and the per-
mented the political spectrum, with new political forces
sonal vote plays a much larger role; each candidate for
appearing, the elections of December 20, 2015, gener-
office is treated almost as an isolated entity with only a
ated a parliamentary stalemate that even new elections
tenuous connection to the party he or she claims to repre-
on June 26, 2016, failed to break entirely. Only after a
sent. In fact, this is one of the key issues on the theory of
drawn-out process, with the prospect of third elections
American political parties (Aldrich 1995; Cox and
hovering over the horizon, were the negotiations con-
McCubbins 1993). For example, if a party uses primaries
cluded. Spain had thus spent almost an entire year with
for candidate selection, this very process conveys addi-
a caretaker government. Thus, the government forma-
tional information that the voters can use to inform their
tion stage acts as an added layer of uncertainty and
choices (Meirowitz 2005). Even within the same system
ambiguity for voters because, even if they could know
there are differences across parties and within the same
accurately the ideological stance of their preferred party,
party over time. Parties that are undergoing a process of
that would still not necessarily translate into an accurate
redefining their coalition, often as a response to a series
forecast about which government coalitions said party
of defeats, dilute their brand so that voters need to rethink
would join. The comparative research of Duch, May,
how to assess them (Lupu 2014).
and Armstrong (2010) shows that, indeed, voters in mul-
The extent and quality of the electorate’s political
tiparty democracies take into account the process of
knowledge has been the subject of a growing literature,
coalition-building at the post-electoral stage when
including the classic work of Delli Carpini and Keeter
deciding their vote. Facing this complexity, it is to be
(1996) for the American context. Later contributions
expected that voters will reach out to heuristic proce-
quickly recognized that many of the lessons from those
dures in assessing the choices on offer.
first efforts would not necessarily travel to, say, European
Fortunato and Stevenson (2013) have shown that vot-
democracies.
ers assign significant weight to the parties’ decision to
There are several reasons for this. First, since, as per
join cabinets as a way to evaluate them. This makes sense
Duverger’s law (e.g., Cox 1997), electoral proportional-
given that cabinet-membership is a cheap signal that con-
ity is conducive to a multiparty system, the vagaries of
veys significant, albeit...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT