BookPDF Available

Towards Electoral Control in Central and Eastern Europe



In modern democracy, voters should be able to use elections to control parties and politicians. Citizens lose control when their electoral voice does not compel them to act according to the will of the people. Repeated free and fair elections are supposed to function as a mechanism of electoral control. To evaluate electoral control, citizens need the right data on candidates, parties, and parliamentarians. In this edited book, we present a step towards electoral control by presenting the methodology of the East European Parliamentarian and Candidate dataset (EAST PaC). These data are the universe of candidates and parties who stood for national parliamentary elections in Ukraine, Poland, and Hungary from the 1990s to the 2010s. Candidates are matched over time, rendering a dataset that allows researchers to track the political careers of every candidate, from the thousands who never won to the few political lifers whose parliamentary careers are decades long. With EAST PaC, we can achieve new insights into electoral politics of Central and Eastern Europe. This book contains everything that scholars need to use EAST PaC. We designed the book for regional specialists and non-specialists; for seasoned scholars of representation, accountability, and political inequality, and for those who are newly interested in these concepts. We hope to intrigue potential users of EAST PaC and to attract new scholars to study electoral politics in Ukraine, Poland, and Hungary. This book is funded by Poland’s National Science Centre (Sonata Bis decision number 2012/05/E/HS6/03556).
Joshua Kjerulf Dubrow
and Nika Palaguta (eds.)
IFiS Publishers
warsaw 2016
Projekt został s nansowany ze środków
Narodowego Centrum Nauki przyznanych na podstawie
decyzji numer DEC-2012/05/E/HS6/03556
is publication has been prepared under the project Kto wygrywa,
a kto przegrywa w parlamentarnych wyborach. Od formalnej teorii
do analiz empirycznych,  nance by  e National Science Centre,
Contract No. UMO-2012/05/E/HS6/03556
Recenzja wydawnicza
dr Michał Kotnarowski
e monograph Towards Electoral Control In Central
and Eastern Europe
by Joshua Dubrow and Nika Palaguta (eds.)
was reviewed by Michał Kotnarowski
Projekt okładki/Cover design
Andrzej Łubniewski
John Fells
Copyright C by Authors and Wydawnictwo IFiS PAN
ISBN 978-83-7683-123-7
IFiS PAN Publishers
00-330 Warsaw, Nowy Świat 72
Tel. (48) 22 6572897
e-mail: publish@i
Table of Contents
Preface and Acknowledgements ................................................... 7
Towards Electoral Control in Central and Eastern Europe
Joshua Kjerulf Dubrow and Nika Palaguta ................................ 9
C 
Representation and Accountability: Intellectual Foundations
of EAST PaC – Joshua Kjerulf Dubrow, Anna Kurowicka,
Nika Palaguta, and Kazimierz M. Słomczyński ............................. 21
C 
e Electoral Market as a Mechanism of Political Inequality
Joshua Kjerulf Dubrow ............................................................. 43
C 
e Content and Structure of EAST PaC for Ukraine, Poland,
and Hungary, 1985–2014 – Joshua Kjerulf Dubrow .................... 77
C 
Collecting and Matching Data on the Candidates for the
Ukrainian Parliament, 1990–2014 – Nataliia Pohorila ............... 85
C 
Collecting, Cleaning, and Matching Electoral Data on
Candidates and Parliamentarians in Poland, 1985–2011
Zbigniew Sawiński and Joshua Kjerulf Dubrow ......................... 99
6Table of Contents
C 
Challenges of Matching Candidate Data in Hungary,
1990–2010 – Zsó a Papp ............................................................ 115
C 
A Brief Political History of Ukraine in the 20th and 21st
Centuries – Nataliia Pohorila and Yurii Taran ............................. 131
C 
e Electoral Laws, Election Outcomes, and Political Parties
of Ukraine, 1990–2014 – Nika Palaguta and Anna Kurowicka .... 151
C 
A Brief Electoral History of Poland, 1985–2015
Marcin Ślarzyński, Anna Kurowicka, and Nika Palaguta ........... 177
C 
Electoral Gender Quotas in Poland – Joshua Kjerulf Dubrow ...... 191
C 
Electoral System and Election Outcomes in Hungary,
1990–2014 – Nika Palaguta and Anna Kurowicka ...................... 207
C 
Multi-disciplinary Scienti c Collaboration across Countries
and Time: Managing the Electoral Control Project
Joshua Kjerulf Dubrow and Olga Zelinska ................................. 219
Lessons Learned and New Questions – Joshua Kjerulf Dubrow
and Nika Palaguta ...................................................................... 249
A A: Coding Issue Position of Ukrainian Parties,
2012–2014 – Nika Palaguta ...................................................... 263
A B: Codebook for EAST PaC Hungary,
1990–2010 – Zsó a Papp ........................................................... 292
About the Authors ..................................................................... 305
Preface and
is book is about electoral control in Central and Eastern Europe
(CEE) and was produced by the Electoral Control project.  e Elec-
toral Control project was based on the research grant, “Who Wins and
Who Loses in the Parliamentary Elections? From Formal  eory to Em-
pirical Analysis,” funded by Poland’s National Science Centre (Sonata
Bis decision number 2012/05/E/HS6/03556) from 2013 to 2016.  e
Principle Investigator of this project was Joshua Kjerulf Dubrow, As-
sociate Professor at the Institute of Philosophy and Sociology (IFiS),
Polish Academy of Sciences (PAN) and administrator at Cross-national
Studies: Interdisciplinary Research and Training program (CONSIRT. of  e Ohio State University (OSU) and PAN.
e Electoral Control project built an international multi-disci-
plinary scienti c team focused on the collection and use of data on
parliamentarians and candidates in CEE to address critical issues in
representation, accountability and political inequality. To that end,
this project produced the East European Parliamentarian and Can-
didate dataset (EAST PaC), composed of the candidates who stood
for national parliamentary elections in Ukraine, Poland, and Hungary
from the  rst post-Communist election to the most recent and who are
matched over successive elections. In addition to bringing the project
team together to various events held in Warsaw and in scienti c venues
elsewhere, Polands National Science Centre funded EAST PaC data
collection and this book. We are grateful for their support.
We are also grateful for the support of IFiS PAN and the Graduate
School for Social Research (GSSR) for the events held under the
banner of this project. We also thank CONSIRT which provided
Preface and Acknowledgements
organizational support for the administration of this grant and the
project as awhole. Kazimierz M. Słomczyński, Director of CONSIRT,
Professor of Sociology at IFiS PAN and Professor Emeritus of OSU’s
Department of Sociology, was a key  gure, both intellectually and
administratively, for this project. We thank Goldie Shabad, Professor
Emeritus of OSU’s Department of Political Science, and Jakub Zie-
linski, former Professor at OSU’s Department of Political Science and
now at the Boston Consulting Group, for their intellectual contribu-
tions to this grant project. We are thankful for the intellectual support
of the Institute of Political Studies at PAN, especially Mikołaj Cześnik
and Michal Kotnarowski. We thank Irina Tomescu-Dubrow, Associ-
ate Professor at IFiS PAN and Projects Manager at CONSIRT, for
her administrative and intellectual guidance. We are thankful to Ola
Wójcicka, Ewa Dworniak, and Grażyna Drążyk for their administra-
tive assistance, and to John Fells for proofreading the text. We thank
Ireneusz Sadowski for his assistance in analyzing EAST PaC data.
e Electoral Control project brought together young and estab-
lished scholars in sociology, political science, and area studies to form
amulti-disciplinary scienti c research team.  e core of this team was
Mihail Chiru, Tomek Kołodziej, Sheri Kunovich, Justyna Nyckowiak,
Zsó a Papp, Nataliia Pohorila, Zbigniew Sawiński and Peter Tunkis.
We were fortunate to work with such atalented group of scholars.
We were also fortunate to have the enthusiastic participation of excel-
lent PhD students at GSSR, including Kateryna Gryniuk, Anna Kuro-
wicka, Sanjaya Chaudhary Mahato, Nika Palaguta, Marcin Ślarzyński,
Anastas Vangeli, and Olga Zelinska. We thank the many attendees and
presenters at our events, including Amy Alexander, Ti any Barnes,
Catherine Bolzendahl, Nazar Boyko, Henryk Domański, Gábor
Molnár, Iulian Stanescu, and Claudiu Tu s. We are also thankful to
Karim Khan and Melanie Hughes for their assistance to the project.
We dedicate this book to Bohdan Solchanyk, auniversity lecturer
in Lviv and amember of the team that collected the 2006–2012 EAST
PaC data in Ukraine, who died on February 20, 2014 during aprotest
at Kyiv Maidan. He was 28 years old. According to an article about him
in e New York Times, Bohdan w as “killed by abullet ... in Kiev in the
carnage on and around Independence Square.” Bohdan was adoctoral
student at GSSR, and an historian and sociologist. We celebrate his life
and we mourn his loss.
Towards Electoral Control in Central
and Eastern Europe
oshua jerulf ubrow and ika alaguta
Comprehensive information about parliamentarians, candidates, and
parties is afundamental resource that citizens need to cast judgment
on the electoral system. It allows citizens to critically evaluate the par-
liament, an important political body that represents citizen interests
and produces the policy that impacts daily life. How can we hold the
legislature accountable if we do not have aclear record of who the par-
liamentarians, candidates, and parties are, and of what they have done?
eories of democratic accountability tell us that voters can use
elections to control parties and politicians.  eorists assume that po-
litical representatives anticipate sanctions for poor individual and party
performance and thus have an incentive to implement policies that
correspond with citizens’ interests. Citizens lose control when their
electoral voice does not compel parties and politicians to act according
to the interests of the people who put them in power. Repeated free
and fair elections are supposed to function, then, as amechanism of
electoral control.
To empirically examine electoral control, we need the right data.
Many previous empirical studies on parliamentary elections, even in
Central and Eastern Europe, have critical shortcomings, as they typi-
cally (i) limit analyses to elected parliamentarians although all can-
didates are equally important for understanding the process of elec-
toral control; (ii) fail to adequately incorporate the history of elections
although voter decisions are usually based on the assessment of the
past performance of candidates’ political parties; and (iii) do not have
Joshua Kjerulf Dubrow and Nika Palaguta
appropriate contrasts in cross-national settings: examining di erent
electoral systems allows researchers to properly assess the extent to
which our knowledge about the determinants of winning and losing
can be generalized across the region.
In this edited book, we present astep towards electoral control
with the East European Parliamentarian and Candidate dataset (EAST
PaC). EAST PaC is composed of the candidates who stood for national
parliamentary elections in Ukraine, Poland, and Hungary and spans
the 1990s to the 2010s (Polish data goes back to 1985). Candidates
are matched over time, rendering adataset that allows researchers to
track the political careers of every candidate, from the thousands who
never won to the few political lifers whose parliamentary careers are
decades long. By covering every parliamentary election and situat-
ing them in their particular historical moments, scholars can identify
trends and dynamics of the political party systems and achieve major
insights into electoral politics of the region. EAST PaC is an oppor-
tunity for scholars to better test theories of accountability, representa-
tion, and political inequality in Central and Eastern Europe from the
fall of Communism to the present.
e goals of this book meet the main goals of the Electoral Control
project funded by Poland’s National Science Centre (see Preface and
Acknowledgments).  e Electoral Control project brought together
young and established scholars in sociology, political science, and area
studies to form amulti-disciplinary scienti c research team that (a)
updated existing data on parliamentarians and candidates used in prior
publications and made these data easily accessible and freely available;
(b) trained graduate students and set the research agenda for the use
of these data around the topics of accountability, representation, and
political inequality; and (c) produced methodological and substantive
works based on these data.
is book is designed to meet the methodological aims of the
project, which were to provide users with all of the information they
need to use these data for substantive analyses and to provide aroadmap
for collecting these data in future elections and other countries (see
“Conclusion: Lessons Learned and New Questions,” this book). As
for substantive works, an aim of this project was to specify how and
to what extent winning and losing in the parliamentary elections
depend on acombination of party and candidate characteristics within
Introduction: Towards Electoral Control in Central … 11
social, economic, and political contexts. To that end, we are preparing
aseparate edited book based entirely on substantive analyses of EAST
PaC data, in addition to aseries of publications in academic journals
due in 2016 and 2017.  is methodological book both accompanies
the substantive analyses that will appear in various academic outlets
over the next few years and serves as the means to further methodologi-
cal advancement in the  eld of electoral politics in CEE and elsewhere.
Several high-pro le studies on electoral control in Central and Eastern
Europe based on earlier versions of EAST PaC data formed the basis
for the Electoral Control project. Jakub Zielinski, Goldie Shabad, and
Kazimierz M. Słomczyński’s (2005) “Electoral Control in New De-
mocracies:  e Perverse Incentives of Fluid Party Systems,” published
in World Politics, was acrucial work.  ey asked whether repeated elec-
tions function as ameans of accountability; their case study was Poland
during the 1990s. Using adataset that was an earlier version of EAST
PaC, they found that economic performance in uences vote decisions:
incumbents are elected (or not) based on how well their constituencies
fare economically.  ough 1990s Poland is characterized by high levels
of party switching, electoral control works through political parties
in amanner substantively similar to Western Europe. Zielinski et al
(2005: 391) expressed surprise at their  ndings:
“Given the historical legacy of dictatorship and the broad sense of un-
certainty that accompanies the early stages of democratization, one
might have expected that it would take considerable time before voters
learn how to use elections to control their representatives systemati-
cally. It appears, however, that such learning is very rapid. As aresult,
there seems to be no fundamental di erence between new and old
democracies with regard to this basic nature of the electoral process.
Later, Shabad and Słomczyński (2011) returned to the idea of electoral
control in Poland, but this time from the voters’ perspective.  ey used
waves from the Polish Panel Survey (POLPAN) and again found evidence
that electoral control in Poland functions as theorized: “Voters take both
Joshua Kjerulf Dubrow and Nika Palaguta
economic performance and political performance into account,” they
wrote, “in deciding whether to reward or sanction governing political
parties” (316).  ey warned that polarization in Polish politics could
lead voters to harden their political loyalties, making them less likely to
push out parties with poor economic performance and thus diminish the
ability of elections to function as aform of democratic control.
Słomczyński, Shabad, and Zielinski (2008) turned their attention
to Ukraine. “Fluid Party Systems, Electoral Rules and Accountabil-
ity of Legislators in Emerging Democracies:  e Case of Ukraine,”
published in Party Politics, asked the same question they had posed to
Poland: do repeated elections function as amechanism of accountabil-
ity? Examining an earlier version of EAST PaC Ukraine, from 1994 –
2002, again they were surprised: Ukrainian parliamentarians are sanc-
tioned for poor economic performance.
“Our analyses also show that electoral control in Ukraine operates
through political parties or partisan blocs. Despite the informational
noise that exists in Ukraine’s party system and the deliberate confusion
perpetrated by ‘the party of power’, partisan labels communicate to the
voters which incumbents were or are associated with political forces
supportive of the presidential administration and therefore should be
held responsible for economic outcomes” (104).
Similarly, Andrew Roberts (2008), in his article in Electoral Studies,
“Hyperaccountability: Economic voting in Central and Eastern
Europe,” asked the same question as Słomczyński and colleagues,
namely, whether repeated elections engender accountability in
post-Communist nations. Roberts examined aggregate data of party
vote shares across ten countries.  e data are based in large part on
the Project on Political Transformation and the Electoral Process in
Post-Communist Europe at the University of Essex and span from
the 1990s to the mid-2000s. Roberts, also expressing surprise, found
strong evidence for electoral control based on economic performance.
“... the communist inheritance erected anumber of barriers to elector-
al accountability. In particular, voters’ lack of experience and participa-
tion in politics, the high uncertainty of the transition, and unformed
party systems might make accountability di cult to practice. Why did
reality subvert these expectations?” (542)
Introduction: Towards Electoral Control in Central … 13
Roberts suggested two reasons for this. First, he argued that Com-
munism as practiced in the CEE produced educated and urbanized
citizens who “were equipped with the skills to participate intelligently
in politics if they so chose” (542). And, according to Roberts, they did
so choose. Second, the regions sustained radical social and political
change heightened citizen interest in politics.
Ahealthy rate of entry and incumbency in electoral competition
might indicate arobust participatory political culture expected in both
new and old democracies (Almond and Verba 1963).  e extreme po-
litical sensitivity of CEE voters that allow for electoral control also
allow them to continually throw out incumbent parties (see Markows-
ki 2006 and Svolik 2013). Poland, for example, had its longest period
of political stability from 2007 to 2015. In 2015, despite a stable
economy, Poles voted out the incumbents and voted in the opposition.
e winning and losing of parliamentary elections should be consid-
ered in the broad context of demographic characteristics of candidates,
of the party that they represent, and the electoral system.  us, akey
issue in this project is the relationship between the personal character-
istics of legislative candidates and electoral outcomes.  e Electoral
Control project is concerned with questions directly related to how
individual traits prove bene cial or detrimental to the winning of
o ce in CEE democracies.  e implications of such arelationship are
wide-ranging – who political candidates are and where they are from,
not to mention who the winners and losers of repeated elections may
be, is consequential for the evolution of the quality of democracy in
post-Communist countries in terms of representation, accountability,
and political inequality of voice.
Political biographies and the demographic characteristics of those
running for public o ce matter for the quality of democracy. Rep-
resentation and accountability are central principles to a system of
governance that goes beyond procedure alone (Dahl 2005; Schmit-
ter & Karl 1991). Candidates’ backgrounds and participation in free
and fair elections speak to the representativeness and accountability of
Joshua Kjerulf Dubrow and Nika Palaguta
parliaments and parliamentarians.  e characteristics of the winners
provide us with apicture of the representativeness of parliaments and
the professionalization of politics (Shabad and Słomczyński 2002).
Depending on the type of electoral system, incumbent success matters
in discussions about accountability.
Two key features of EAST PaC data are completeness and histori-
cal relevance: (a) we include all elections that took place in Poland,
Hungary, and Ukraine in the post-Communist era, and (b) we cover
along period of time that accounts for dynamics of the political party
For each country, the data for all elections are pooled so that the
candidate is the unit of observation, while personal characteristics and
characteristics of electoral participation make up the values of variables.
ese data allow us to track the political experience of candidates, in-
cluding dynamics of their partisan a liations, across consecutive elec-
tions.  e project, then, is concerned with the exact populations of
candidates.  e main sources of data on parliamentary candidates are
o cial records from governments and state agencies responsible for
maintaining election archives. For elections conducted in the early
1990s, these records were in the form of paper documents, while in
later years they were stored as electronic  les.
is book contains everything that scholars need to use EAST PaC
data for various academic purposes – whether to use these data as they
are or to combine them with survey and non-survey data. We designed
the book for various audiences: for CEE experts and for those who
are unfamiliar with the region; for seasoned scholars of representation,
accountability, and political inequality, and for those who are newly in-
terested in the analysis of these concepts. We hope to intrigue potential
users of EAST PaC data and to attract new scholars to study electoral
politics in Ukraine, Poland, and Hungary.
Introduction: Towards Electoral Control in Central … 15
We divided the book into four parts.
Part 1  eory and Concepts presents an overview of the main theories
and concepts of the project that are fundamental for research using
EAST PaC data and suggests theoretical directions. In Chapter One,
Dubrow et al use recent and classic texts to explore representation and
accountability that are the driving concepts of the Electoral Control
project and the rationale for collecting the EAST PaC data. In Chapter
Two, Dubrow considers the idea that, for as long as the structure of
candidates and parliamentarians resemble the top of the strati ca-
tion ladder rather than its whole, issues of electoral control are  rmly
connected to the issue of political inequality. We connect these ideas
through descriptive representation, an ideal that is also an indicator
of political equality.  is chapter posits atheory of how parliaments
become politically unequal places through the metaphor of an electoral
Part 2 Methodology covers the content and structure of EAST PaC
data and the collecting, cleaning, and matching for Ukraine (Pohorila),
Poland (Sawiński and Dubrow), and Hungary (Papp) written by the
leaders of the data collection teams for those countries.  e goal of
EAST PaC was to create ahigh quality dataset of the universe of can-
didates who ran for national o ce and in which users can track can-
didates across elections. EAST PaC is matched data that covers three
countries, 29 years, 23 elections, and 97,439 unique candidates.  e
problems and errors encountered were of various kinds: technological,
bureaucratic, social and political.  e solutions for matching were gen-
erally based on acombination of automatic coding based on gender,
age, and political a liation, and of manual coding, where the data
collection team had to comb through the data to identify and resolve
duplicate cases.  e process required multiple technological solutions,
some of them dedicated to this task and all of which we developed over
time and improved our data. We hope that by describing these data
and the process, future scholars and other data users can improve and
build on these data for future elections.
Part 3 Context is composed of aseries of chapters on historical and
contemporary electoral politics in each country, including detailed
descriptions of parties and changes to electoral laws and electoral
outcomes.  e purpose of this part is to provide the broader electoral
context of Ukraine, Poland, and Hungary to pro tably analyze EAST
Joshua Kjerulf Dubrow and Nika Palaguta
PaC data. It proceeds country by country and begins with abrief po-
litical history of Ukraine that starts at the Bolshevik revolution in 1917
that created the Union of Soviet Socialist Republics to the electoral
and social turmoil of Ukraine’s post-Independent period (Pohorila
and Taran). What follows is an in-depth description, with informa-
tion found in EAST PaC, of the political parties, electoral systems
and electoral outcomes in Ukraine from the  rst post-Independence
election to 2014 (Palaguta and Kurowicka). Ślarzyński, Kurowicka,
and Palaguta then cover the Polish electoral scene since 1985 (the  rst
election covered in EAST PaC Poland), focusing on changes to elec-
toral laws and the outcomes of parliamentary elections. Next, con-
sidering that gender and representation are key features of electoral
politics, Dubrow provides abrief history of gender quotas in Poland.
Kurowicka and Palaguta end this part with afascinating description of
the Hungarian electoral scene since 1990.
We then turn to the Electoral Control project itself as an object of
study. Social scientists rarely provide in-depth reporting on the pro-
cesses that led to their research. Knowledge of process is adeep layer
of information that enables future scholars to critically evaluate past
projects and make good decisions for their own projects. We present the
“who, what, when, where, why, and how” of the project; it is arecord
of this project’s history, aims, and critical decisions that led to these
data, this book, and the seeds of future products. Part 4 Collabora-
tion is asingle chapter that features adescription of the cross-national
multi-disciplinary Electoral Control project (Dubrow and Zelinska).
We describe the progression of the project and the process of facili-
tating collaboration among the multi-disciplinary mix of young and
established social scientists that comprised the team. We learned best
practices about the process of scienti c collaborative work by develop-
ing and adapting our ideas over time.
e appendices present technical information. First is a coding
scheme developed by Palaguta on how to code the issue stances of all
major parties in Ukraine 2012 – 2014.  is chapter goes well beyond
the scope of the well-known Manifesto project and allows future
scholars to usefully code new variables based on Ukrainian parties’ po-
litical platforms. After, Zsó a Papp provides acodebook for the com-
plicated EAST PaC Hungary data.
Introduction: Towards Electoral Control in Central … 17
e Electoral Control project was designed to build an international
and interdisciplinary scienti c team focused on the use of data on par-
liamentarians and candidates in Central and Eastern Europe to address
critical issues in representation, accountability and political inequal-
ity.  e future of democratic progress rests on knowing the past and
present of electoral politics. In this regard, we hope that this book does
its part for Central and Eastern Europe.
Almond, Gabriel A. and Sidney Verba. 1963. e civic culture: Political atti-
tudes and democracy in  ve nations. Princeton: Princeton University Press.
Dahl, Robert A. 2005. “What political institutions does large-scale democ-
racy require?” Political Science Quarterly 120: 187–197.
Higley, John and Jan Pakulski. 1999. “Elite power games and democratic
consolidation in Central and Eastern Europe.Soviet and Post-Soviet
Review 26: 115–137.
Markowski, Radosl aw. 2006. “Political Accountability and Institutional Design
in New Democracies.International Journal of Sociology 36(2): 45–75.
Roberts, Andrew. 2008. “Hyperaccountability: Economic voting in Eastern
Europe.Electoral Studies 27(3): 533–546.
Schmitter, Philippe C. and Terry Lynn Karl. 1991. ”What democracy is…
and is not.” Journal of Democracy 2: 75–88.
Shabad, Goldie and Kazimierz M. Słomczyński. 2002. “ e emergence of
career politicians in post-Communist Poland and the Czech Republic.
Legislative Studies Quarterly 27: 333–359.
Słomczyński, Kazimierz M., Goldie Shabad and Zielinski, Jakub. 2008.
“Fluid party systems, electoral rules and accountability of legislators in
an emerging democracy:  e case of Ukraine.Party Politics 14: 91–112.
Svolik, Milan W. 2013. “Learning to love democracy: Electoral accountabil-
ity and the success of democracy.” American Journal of Political Science
57(3): 685–702.
Tavits, Margit. 2009.  e making of mavericks: Local loyalties and party
defection. Comparative Political Studies 42: 793–815.
Zielinski, Jakub, Kazimierz M. Słomczyński, and Goldie Shabad. 2005.
“Electoral control in new democracies:  e perverse incentives of  uid
party systems.” World Politics 57: 365–395.
P 1
C 
Representation and Accountability:
Intellectual Foundations of EAST PaC
oshua jerulf ubrow, nna urowicka, ika alaguta, and
azimierz . łomczyński
eory tells us that by conditioning their ballots on policy outcomes,
voters can use elections to control politicians. Presumably, politicians
anticipate that they will be sanctioned for poor party-performance
and thus have an incentive to implement policies – through their own
e orts, or through parties and other political units – that correspond
to the preferences of the electorate. Does the system of repeated elec-
tions function as amechanism of electoral control, and if it does, what
factors in uence its e ectiveness?
In the Electoral Control project, we consider this question in
abroad context of studies on parliamentary elections.  is chapter sets
out the intellectual foundations of this project (see also Introduction,
this book) and thus the rationale for collecting the EAST PaC data. We
consider the two main and interrelated concepts that have guided the
project from the beginning: representation and accountability.
To address whether elections are amechanism of electoral control, we
begin with adiscussion of representation.  e literature on political
representation is vast and multi-disciplinary as it includes views of
political scientists, sociologists, psychologists, and area studies spe-
cialists, among others. When it comes to pinpointing afoundational
post-Second World War work on political representation, Hanna F.
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
Pitkins e Concept of Representation, published in 1967, is aleading
In her seminal book, Pitkin presents atypology of various theo-
retical approaches to representation. She begins by introducing the
notions of authorization and descriptive views on representation. Ac-
cording to the authorization view, also referred to as the formalistic
type of representation, arepresentative is granted apreviously absent
right to perform a limited range of actions within certain bounds.
Keeping within the bounds means that the representative performs the
function of representation. Stepping outside of them indicates repre-
sentation failure.  e represented are, at the same time, responsible for
the actions of representatives that they granted the right of representa-
tion to (Pitkin 1967: 38–39).
Some subtypes of descriptive representation, or “standing for” type
of representation, include proportional representation, which is con-
sidered to be astrong re ection or even replication of the population
structure on asmaller scale and assumes representation of all of its
groups (Pitkin 1967: 61–62).  eorists who hold this view often use
the metaphor of arepresentative body as aportrait of the constitu-
ency they represent. Representation of this type calls for aparticular
methodology. Representation by sample involves randomly choosing
agroup from the whole population.  is method assumes that such
kind of non-biased selection allows for representation of the popula-
tion by asmall group of representatives (Swabey 1937: 25 as quoted in
Pitkin 1967: 74). Gri thss and Wollheim’s (1960) theory of descrip-
tive representation suggest that arepresentative has to share distinct
similarities with the represented (188).
To reduce representational inequality, many governments, political
leaders, social justice advocates, and researchers champion the concept
of descriptive representation. Proponents of descriptive representa-
tion assert that those elected o cials who resemble the demographic
and experiential characteristics of their constituencies have su cient
empathy to evaluate and construct representative policy (Mansbridge
1999; Phillips 1995; Young 1990). In this sense, political structures
encourage representation by empathetic demographic insiders. In
contrast, delegative representation makes no provisions for demo-
graphic representation, relying instead on the stewardship of sympa-
thetic demographic outsiders (Birch 2001). In practice, descriptive
Representation and Accountability: Intellectual Foundations… 23
representation attempts to ameliorate inequitable social conditions
by providing historically marginalized groups with opportunities to
become political elites. In so doing, proponents assert, descriptive rep-
resentation safeguards the interests of the disadvantaged.
Descriptive representation, while perceived by Pitkin as limited,
has proved to be an important point of reference for contemporary
theorists working on the exclusion of marginalized groups based
on gender, race, ethnicity, or other factors.  e most important of
these include Will Kymlicka who in his Multicultural Citizenship:
A Liberal  eory of Minority Rights (1995) argues for group rather
than just individual representation; Anne Phillips who in e Politics
of Presence (1995) explains the importance of representation of di-
versity in society, and Melissa S. Williams who in Voice, Trust, and
Memory: Marginalized Groups and the Failings of Liberal Representa-
tion (1998) explores patterns of group-based advantage and exclusion
that tends to be ignored by individual representation.  e criticisms
of traditional concepts of representation expressed by those and other
thinkers focus on the con ict between group fairness of representa-
tion and individual proportional representation fairness (e.g. Urbinati
and Warren 2008).
Descriptive representation is more of a concept than a theory,
designed to stimulate praxis rather than merely academic research. Ad-
dressing inequitable political representation, theoretical debates focus
on the tenability and “philosophy and ethics” of descriptive representa-
tion as agovernance solution, especially in the light of the current state
of disadvantaged group representation (Chaney and Fevre 2002: 897).
us, it refers to both an ideal and areality; the ideal being the govern-
ance solution, the reality being the degree to which the legislative body
represents the demographics and experiences of the citizenry.
Descriptive representation has been criticized on various grounds
(for areview, see (Mansbridge 1999). Most common is that descrip-
tive representation would not lead to substantive representation, such
that demographic qualities bear little to no relationship to deliberative
capabilities (Mansbridge 2000: 101). Others argue that by overem-
phasizing group di erences through claims of supra- representational
abilities, descriptive representation erodes the bonds among legislators
whose job it is to produce policies for all, rather than ademographic
subset, of their constituency (Bird 2003).
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
Many other complaints focus on the di culties of implementing
descriptive representation. Choosing which groups from amultiplicity
of genders, races, ethnicities, religions, age groups, physical handicaps,
and social classes are worthy of descriptive representation could be so
complex that random or arbitrary assignment to legislative bodies is
the only reasonable way (Andeweg 2003: 149; Kymlicka 1995). Some
fear that implementation of this form of representation would lead
to a selection of less quali ed legislators drawn from, among other
places, the bottom of the talent pool. Akin to this is the argument that
descriptive representatives vary as much within their group, including
multiple social identities such as Muslim lower class woman, or young
Silesian émigré, etc., as they do between groups. In its implementation,
descriptive representation oversimpli es acomplex set of demograph-
ics, leaving some subgroups underrepresented, thereby undermining
the very purpose for which it was intended.
Counter to these criticisms, most proponents assert that descriptive
representation is not acall for an exact microcosm of the citizenry,
“such that children represent children, lunatics represent lunatics
(Bird 2003: 2). Instead, the goal is (a) substantive representation
through making the legislative body demographically closer to the citi-
zenry and (b) situation speci c, in that selection of groups in need of
representation should be made after careful, rational deliberation and
under particular conditions (Dovi 2002; Mansbridge 1999).
Symbolic representation is based on anotion of symbol, which
does not contain ameaning by itself, but one which is attributed to it
by social convention.  us asymbolic representation is only in place,
when it is believed in (Pitkin 1967: 100).
Substantive representation, referred to by Pitkin as “acting for” rep-
resentation, assumes that the actions and the goals of representatives
should coincide and there is aprimacy of represented over representa-
tive in away that only the latter is accountable. (Pitkin 1967: 163–165
in Celis et al 2008: 100). Celis et al (2008) further explain two di erent
types of substantive representation and the major di erences between
them by highlighting the role of represented and representative. Under
the  rst type of representation, the representative has no possibilities
of actions beyond those granted them by the represented. Under the
second type, the representative has more possibilities for independent
actions than that strictly authorized by the represented. Pitkin argues
Representation and Accountability: Intellectual Foundations… 25
that there is adi erence in how the representative should act in di er-
ent circumstances: if representatives are more experienced and are able
to propose rational, well grounded solutions in the national interests,
they should act more independently.  eir actions however should be
limited in cases of explicitly con icting interests with the represented
or when signi cant di erences in interests of the represented and the
state occur, and in cases where both the represented and the repre-
sentatives have su cient experience and information to decide upon
an issue (Celis et al et al 2008: 100–101).
While extremely in uential, Pitkin’s typology could stand some im-
provements, especially for the ever-changing circumstances of demo-
cratic representation. It was expanded on by, among others, Mansbridge
(1999 and 2003), who discusses the traditional model of representation
and adds three more models, whose features suggest that they should be
judged according to adi erent set of criteria than the traditional one.
e types of representation proposed by Mansbridge (2003) are
Promissory (traditional) representation that follows the classical prin-
cipal-agent format and assumes that the agent (the representative) fails
to meet their obligations to the principal (the represented) and that
the represented will try to make representative accountable through
the promises that the representative makes (515). According to Mans-
bridge this type of representation uses a“forward-looking concept of
power” (515). In contrast, Anticipatory representation is aresult of the
idea of retrospective voting, under which representatives look forward
to the expectations of their voters on the next election but not to those
promises they made for the elections they have already won (2003:
515). According to Mansbridge, there is no moral attitude of the agent
in aiming to meet their previous promises to the principal, but rather
there is acareful anticipation with the goal of reelection (2003: 518).
is new model of representation based on the idea of retrospective
voting and rejection of the traditional principal-agent model has im-
portant consequences for the way the quality of such representation is
judged. As Mansbridge (2003) explains,
“replacing morality with prudence in the incentive structure of anticipa-
tory representation leads us to judge the process with new normative
criteria. It makes us shift our normative focus from the individual to
the system, from aggregative democracy to deliberative democracy, from
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
preferences to interests, from the way the legislator votes to the way the
legislator communicates, and from the quality of promise-keeping to the
quality of mutual education between legislator and constituents” (518).
One of Mansbridge’s great contributions to the literature on represen-
tation is her critique of previous representation theories and the devel-
opment of anew one, what she refers to as “gyroscopic representation.
Gyroscopic representation is based on an idea of representation from
di erent from classic accountability models: “ e representatives act
like gyroscopes, rotating on their own axes, maintaining acertain direc-
tion, pursuing certain built-in (although not fully immutable) goals”
(2003: 520). Representatives are not expected to be accountable before
the voters, like in traditional models, but rather to follow their goals
(2003: 520).  ey are guided by some general principles and logically
grounded necessities trying to meet voters expectations, comply with
moral standards and demonstrate required pro ciency levels (2003:
520–1). While afocus on the characteristics of an individual candidate
is more typical of the American political context, the European model
precludes politicians’ orientation towards the interests of the party
that consists both of their own beliefs and of the line of the party as
awhole.  e party member is responsible for his disobedience before
the party and the party before the citizens (2003: 521).
e Surrogate model of representation is atype of representation
which “occurs when legislators represent constituents outside their
own districts” (2003: 515). Under the circumstances where the repre-
sentative is not controlled by  nancial or power instruments, account-
ability of the representative to the represented is absent. In cases where
nancial control is possible, voters are able to control representatives
through the ordinary mechanism of representation and by electing
representatives that would act according to pre-determined patterns
pursuing their goals (“as in gyroscopic representation”) (2003: 523).
is type of representation occurs, for example, in the case of minority
groups having a“surrogate” representative elected from one constitu-
ency who “speaks for” an entire group, as in the case of the US Senator
Barney Frank who represented the interests of the LGBT. It is often
seen as apossible answer to some problems inherent in the limita-
tions of the traditional model of aterritorially-based representation (cf.
Urbinati and Warren 2008).
Representation and Accountability: Intellectual Foundations… 27
According to Mansbridge, the last three should be judged accord-
ing to di erent criteria than the traditional one; multiple criteria of
assessment have to be applied as there are many ways of representation.
“ e criteria are almost all deliberative rather than aggregative. And,
in keeping with the conclusion that there is more than one way to be
represented legitimately in ademocracy, the criteria are plural rather
than singular” (2003: 515). As aresult, those three types of representa-
tion do not  t into the “mandate”/“delegate” vs. “trustee” dichotomy.
Another crucial shift in the theories of representation was initiated by
Michael Saward (2006, 2010). He argues that representation is not ex-
clusively aprocess of choice of representatives by the represented – where
representatives remain static – but that it also involves an active process
of making “representative claims” (2006: 298) that singles out the rep-
resentatives who wish to be chosen. Future representatives highlight the
need to represent the interests of agroup, geographic unit, or some other
bounded territory by framing the existence of their special needs that
should be represented.  e process of making claims is not limited to
political life only, as “interest group or NGO  gures, local  gures, rock
stars [and] celebrities” (2006: 306) also take part in the process (Saward
2006 in Celis et al 2008: 101–102). Saward’s approach is yet another sig-
ni cant attempt (following Mansbridge 2003) to rede ne representation
so that it is more congruent with the trends in contemporary democra-
cies and politics, discussed in Urbinati and Warren (2008).
Issues of representation and accountability are closely linked. We brie y
present various approaches to electoral accountability and theories that
use this term to describe features of various democratic systems.
One of the most comprehensive and widely cited texts on account-
ability is Democracy, Accountability, and Representation edited by Adam
Przeworski, Susan C. Stokes, and Bernard Manin (1999).  is collec-
tion includes such key theorists as James D. Fearon and John Ferejohn
and provides an overview of many issues crucial to the topic.  e rst
chapter of this book written by the editors provides an informative
take on two concepts of representation that highlights the connection
between representation and accountability.
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
Manin et al (1999) discuss two di erent concepts of representation:
mandate and accountability.  e mandate model occurs under three
conditions: (i) interests of politicians coincide with those of the voters;
(ii) politicians want to be elected and reelected and expect that it will
happen if they implement policies they promised in the campaign; and
(iii) to avoid losing the elections, politicians want their promises to be
credible in the future (Manin et al 1999: 31–33).  is vision is idealis-
tic as it ignores other potential incentives for politicians.  is mandate
model also brings to the fore one of the crucial problems of representa-
tion: the potential incongruence between voters’ interests and prefer-
ences. Sometimes politicians face achoice between doing what they
consider to be in the voters’ best interests and the expressed wants of
the voters. According to Manin (1997), in current democratic systems,
“politicians are not legally compelled to abide by their platform“ or are
“subject to binding instructions,” which means that there is no legal
way for the voters to force politicians to ful ll their promises (Manin et
al 1999: 38). Historically, this is because: representatives were expected
to deliberate and consult experts; voters do not always trust their own
judgment; and conditions (the facts on the ground) may change.  e
only way of sanctioning politicians at voters’ disposal is the institution
of periodic election, in which they can be held accountable.
e accountability model assumes that voters may control elected
o cials by having them anticipate that they will be held accountable
for their past actions. Incumbent politicians are considered account-
able when the electorate is able to assess their actions and punish or
reward them during the elections according to the results of their
activity (Manin et al 1999: 40).  e following conditions are necessary
for this type of representation to occur: voters vote for the incumbent
only when the representative acts in the best interests of their constitu-
ency, and the incumbent chooses policies necessary to get reelected
e standard view on accountability assumes retrospective voting,
according to which the electorate would observe the performance of
previously elected political parties and punish them if they do not
meet their expectations (Ashworth 2012: 187). According to V.O. Key
(1966: 61), citizens assess politicians` activity retrospectively in any
case, even during elections: while it seems that the population votes
for the future bene ts, indeed voters choose on the basis of previous
Representation and Accountability: Intellectual Foundations… 29
knowledge of what politicians in question did before; citizens do not
vote for political promises made for the future.
Morris Fiorina (1981) wrote about how mechanisms of electoral ac-
countability are incorporated in voting decisions of individual voters.
He argues that citizens evaluate and make decisions based on the per-
formance of the politicians’past performance, and estimate prospective
actions the politicians will likely make in future. Manin et al (1999)
note that such theory of accountability only works under the assump-
tion that voters have complete information. Complete information is
rare, as it is di cult, if not altogether impossible, to accurately judge
politicians, and politicians do not know what they need to do to assure
their reelection (1999: 42–43). Consequently, in their view, informa-
tion asymmetry provokes asituation in which accountability does not
lead to representation (Manin et al 1999: 44). Manin et al conclude by
stating that rather than voting according to the pure mandate model
(use the vote to choose the better candidate) or to the pure account-
ability model (use to vote to sanction the incumbent) voters try to use
their vote for both purposes (1999: 45).
ere are various de nitions and approaches to accountability itself,
and one of the most in uential de nitions of accountability can be
found in Fearon’s (1999) chapter, “Electoral Accountability and the
Control of Politicians: Selecting Good Types versus Sanctioning Poor
Performance.” Fearon suggests that an agent Ais accountable to princi-
pal B, if “(i) there is an understanding that Ais obliged to act in some
way on behalf of B and (ii) B is empowered by some formal institution
or perhaps informal rules to sanction or reward Afor his/her activities
or performance in this capacity” (1999: 55).
Besley (2006) in his “Principled Agents:  e Political Economy
of Good Government,” points out the importance of distinguishing
between formal (de jure) and real (de facto) accountability. Accord-
ing to Ashworth (2012: 184), accountability leads to such relations
between politicians and voters that the position of politicians can be
in uenced through the institution of elections. Besley wrote that the
institution of elections determines the rules of representation, but it
does not guarantee reelection of the politician that meets the formal
requirements of accountability. Instead, reelection depends on the pol-
iticians’ performance that meets voters’ requirements. In weak democ-
racies, where politicians have lower incentives to meet voters` needs
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
or where voters possess very limited information about their repre-
sentatives, incumbent politicians are less accountable than in stable
democracies.  us, the level of accountability may be abenchmark for
democracy assessment (see also Markowski 2006).
Acommonly employed theory of accountability is related to the
concept of political agency. Besley (2006: 98) argues that electoral ac-
countability arises from aprincipal-agent relationship between voters
and political actors. Elections are meant to solve two major incentive
problems: monitoring, which refers to the fact that politicians may
act only in their own interests and thus control over them is necessary
to reward or punish them accordingly; and there is selection, which
focuses on the necessity of choosing the most competent politicians
and/or those whose motivations are congruent with those of voters
(Besley 2006: 99). Besley (2006) further comments on the aforemen-
tioned problem of retrospective voting. In the political agency model
“voters are learning from past actions and use Bayes’ rule to update
their beliefs”, and that is why the di erence between retrospective and
prospective voting vanishes (2006: 106).
Ashworth (2012) points out two basic elements of electoral ac-
countability. First, voters decide to keep apolitician in power. Second,
apolitician who tries to meet voters’ expectations and two policy-mak-
ing periods: in the  rst one the incumbent is in o ce, he or she chooses
an action, and the voter decides how to vote, followed by the second
period in which whoever won chooses an action again, only this
time with no election at the end (the game ends at this point) (2012:
184–5). Ashworth (2012: 185) explains that incumbent’s behavior is
conditioned by the incentives as they will adjust their behavior in order
to be reelected. Consequently, the greater the incentive, such as perks,
salary and other bene ts, the greater responsiveness to the voters, and
when there is no incentive (e.g. term limits, no possibility of reelec-
tion), the incumbent will not be responsive. According to the voters
reelection rule (Ashworth 2012: 185), the incumbent is reelected when
performance meets some standard. In most contemporary literature,
this “standard [is] determined in equilibrium by the voters’ desire to
choose better-performing politicians in the future” which stands in
contrast to more traditional approaches, in which all politicians were
identical and thus there is no reasonable scope for selection (see also
Barro 1973, Ferejohn 1986, Austen-Smith and Banks 1989).
Representation and Accountability: Intellectual Foundations… 31
Besley (2006: 103) elaborates further on some modeling issues
crucial for various conceptions of accountability: the nature of the un-
certainty, the motives for holding o ce, the nature of accountability,
and retrospective voting. In terms of uncertainty, Besley (2006) focuses
on the types of politicians and the question whether they themselves
know their own types. “Amodel of uninformed politicians is more in
tune with the celebrated Holmstrom (1999) ‘career concerns’ model in
which individuals put in e ort which reveals their type to the ‘market’.
Persson and Tabellini (2000) develop apolitical agency model along
these lines, in which voters and politicians are symmetrically informed”
(Besley 2006: 103). Crucially, voters are generally poorly informed
about the best policy, and may be uncertain about the exact policy im-
plemented, and the wisdom of agiven policy is often unclear for some
time (Besley 2006: 104). As for the motives for holding o ce, Besley
enumerates “ego rent,” material gain, public goods concerns, as well
as possibly a“legacy e ect” (Maskin and Tirole 2004). Besley (2006)
argues that the agency models work best with individual directly
elected politicians: mayors, governors, presidents, where the basis of
accountability is clearly de ned relative to responsibilities (2006: 105).
According to the view of Barro (1973) and Ferejohn (1986), moral
hazard creates asituation under which politicians acting according to
their goals may be to some extent constrained by institute of elections
and the theoretical expectations of their voters. In this situation in-
cumbent politicians would have to decide on how prudent they should
be in terms of rent seeking (Besley 2006: 106–107). Besley and Prat
(2006) address adverse selection model under which the voters need
to make achoice of an appropriate incumbent politician, who has no
in uence on their choice.  is model does not consider that, under
normal circumstances, politicians have the means to manipulate in-
formation  ow (Besley 2006: 107; theories that combine elements of
both models include those of Banks and Sundaram (1993) and Aus-
ten-Smith and Banks (1989)).
Others pay attention to the di erences in behavior of politicians
in terms of policy choice (Besley and Case (1995a); Coate and Morris
(1995); Fearon (1999) and Rogo (1990).  ey highlight that some
of them intending to perform well and to be reelected on this ground
may adopt more sophisticated policies that cannot be adopted by poli-
ticians with lower standards of governance, but the latter ones may try
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
to form alliances with the former, so that they would look better in the
eyes of electorate (Besley 2006: 107).
Referring to the political agenda model by Persson and Tabellini
(2000), Besely explains that it is related to the “career concerns model”
of Holmstrom (1999) because neither the electorate nor the repre-
sentatives do not possess full information on the level of representa-
tive performance. Decision making is oriented to the future bene ts
available after the elections take place (Besley 2006: 107).  e career
concerns model is also abasis for models created by Ashworth (2005)
and Ashworth and Bueno de Mesquita (2006, 2008). Besley highlights
that “ is kind of approach also gets away from some of the signaling
issues as the politicians do not know their own types. For some appli-
cations (for example where atype is some kind of competence issue)
this may be natural. However, when preferences are part of the type,
this is aless plausible assumption” (Besley 2006: 108).
Besley (2006: 108) summarizes his typology of accountability
models by stating that the agency models have apositive impact on
accountability because the behavior of politicians is regulated by their
own incentives to adopt better decisions.  e exception to this rule
occurs when politicians are discouraged from doing what is in the
voters’ best interests because they know they will not be rewarded for
it: this is the case when the voters’ interests and preferences are incon-
gruent, e.g. when it comes to long-term policies. Besley explains this
further in connection with the moral hazard models of accountability,
when he notes that under the condition of information asymmetry,
when politicians have more information on their performance, politi-
cal decisions may be adopted in away that would please the electorate
with little regard to the results of the policy (2006: 136).
Ashworth (2012) o ers an alternative yet complementary typology
of accountability models. He presents pure moral hazard models, in
which the prediction is made by selecting the equilibrium that maxi-
mizes voters’ payo (see also Seabright 1996, Persson et al. 1997, Shi
and Svensson 2006, Bueno de Mesquita 2007, Fearon 2011).  ose
are criticized in Fearon (1999) who “shows that the set of equilibria
of the pure moral hazard model is often not robust to allowing can-
didate heterogeneity” as even asmall di erence between candidates
can change the outcome if it is relevant to the voter (Ashworth 2012:
Representation and Accountability: Intellectual Foundations… 33
ere are some prominent examples of models with assumptions
about candidate heterogeneity as proposed by Fearon.  e spatial
policy making model (based on Fearon 1999) argues that both the
incumbent and the voter have ideal points in policy space.  e incum-
bent chooses an action in this space, but the voter is uncertain about
where the incumbent’s ideal point is, as it may be close to the voter’s
or it may be more extreme.  e program can be implemented (based
on de Janvry et al 2010) e ciently or with corruption/graft.  e voter
wants e ciency but does not know what the incumbent prefers, while
the incumbent has all the information.  e constituency service (based
on Ashworth and Bueno de Mesquita 2006) is when the incumbent
must allocate e ort to constituency service and other activities.  e
voter observes the quality of service, which depends on many factors,
such as oversight, competence, and others.  e voter knows that the
incumbent wants to spend less time on oversight, but the voter does
not know the competence level of the incumbent.
e issue of potentially misaligned incentives is explored by
Ashworth (2012) with two examples of models in which this occurs:
the multitask model and the pandering model.  e multitask model
used by Holmstrom and Milgrom (1991), Lohmann (1998), Ashworth
(2005), Ashworth and Bueno de Mesquita (2006), and Gehlbach
(2007), and Daley and Snowberg (2011) is based on the assumption
that the incumbent has multiple tasks in various arenas and the choice
of which to focus on depends largely on the incumbent’s own prefer-
ences. If those preferences align with those of the voters and if electoral
incentives were not in play, the incumbent would do exactly what is
best for the voter. However, since the incumbent does want to win ree-
lection, they will likely take actions with the highest impact on voter’s
beliefs about their type.  is leads to adistortion between high-impact
actions and the optimal actions for the voters’ interests (Ashworth
2012: 188–9).
e pandering model derives from the “reputational herding” lit-
erature from Scharfstein and Stein (1990) and is adopted for elections
by Coate and Morris (1995), Canes-Wrone et al. (2001), Maskin and
Tirole (2004), and Besley (2006) (Ashworth 2012: 189). It has its roots
in the ex-ante uncertainty about which policy is best for the voters. It
also relies on the imbalance of access to information, as the incumbent
tends to have more data on which policy is best but cannot share it. In
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
this scenario if the voter believes the incumbent to be the good type, the
incumbent will still be reelected based on this conviction. However, if
the voter is uncertain about the incumbent’s type, an unexpected policy
choice may cost them the election. As aresult, the incumbent has in-
centives to make the popular decision, but not the best decision; they
pander to the voters.  is problem might be controlled if there is enough
time to resolve this uncertainty, which suggests that the problem is exac-
erbated close to elections (Ashworth 2012: 189). Indeed, according to
Canes-Wrone and Shotts (2004) presidential budgets tend to be closer
to the voters’ preferences right before an election (quoted in Ashworth
2012: 189). Ashworth (2012: 189) argues that some possible solutions
to multitask and pandering problems should be based on weakening
the link between the tempting actions and reelection.  ose include
limiting transparency as proposed by Prat (2005) and Fox and Van
Weelden (2012) and term limits Ashworth (2005).
When discussing empirical evidence for the models, Ashworth
(2012: 190) presents Healy and Malhotra’s (2009) work on two ap-
proaches to natural disasters: disaster preparedness and relief spending,
nding that relief spending is rewarded electorally.  eir conclusion is
that voters behave in an irrational way and that situation results in re-
duction of public welfare as politicians will not invest in preparedness.
ere are two alternative explanations to the conclusion put forward
by Healy and Malhotra.  e rst is based on the multitask model, ac-
cording to which relief spending in uences voters’ beliefs about the
candidate type, while levels of damage depend on so many factors that
they blur the image of the candidate. If taken in this light, Healy and
Malhotras evidence can be proof of voter rationality in the multitask
model (Ashworth 2012: 190).  e second alternative conclusion is
proposed by Bueno de Mesquita (2007) who claims that relief e ort is
more visible than prevention, and thus there is less chance of corrup-
tion while choosing this action, which increases community welfare
(Ashworth 2012: 190–1).
Access to information is one of the most important variables in
accountability; the most basic claim is that an increase in informa-
tion should result in an increase in responsiveness. Incumbent poli-
ticians and voters are connected by the  ow of information: infor-
mation changes the perception of voters and voter opinion impacts
the behavior of politicians (Ashworth 2012: 191). Another problem is
Representation and Accountability: Intellectual Foundations… 35
incumbency advantage. Ashworth (2012: 192) argues that “the selec-
tion mechanism implicit in political agency models predicts an incum-
bency advantage. Ashworth and Bueno de Mesquita (2008) show that,
assuming incumbent elections and open seat elections get comparable
coverage, this incumbency advantage is larger when media coverage
is stronger. And, indeed, Snyder and Stromberg  nd that the incum-
bency advantage is larger in more congruent districts.
One of the roles of challengers in ademocratic system is as asource
of information on the incumbent’s performance, whether it is in the
form of a direct revelation (Ashworth and Shotts 2011) or via an
indirect channel, such as in the case when the very fact of entry into
the race provides information (Gordon et al. 2007; Ashworth 2012:
193). Incumbents with a challenger should be more responsive to
voters, but this claim turns out to be di cult to check, as the incum-
bent typically runs unopposed only in very speci c circumstances, e.g.
when he or she is in avery strong position.  is has been analyzed in
research on retention elections for judges in Kansas by Gordon and
Huber (2007), who found that there was asigni cant incentive for the
incumbents to be more punitive; this suggests that increased respon-
siveness may under certain circumstances lead to rather problematic
results (Ashworth 2012: 193).  e level of partisanship in adistrict
complicates matters. Ashworth and Bueno de Mesquita (2006) sug-
gested that, in less partisan districts, incumbents would focus on con-
stituency service rather than policy making, aproposition checked em-
pirically by Dropp and Peskowitz 2012; Ashworth 2012: 194).
Simple models of accountability become complicated when multiple
agents are involved. Simple models work best with individually chosen
politicians who have clearly de ned responsibilities. It is more complex
in the case of multiple agents which “could be legislators who jointly
pass legislation in alegislature or they could be thought of as the acts
of abicameral legislature, with agents located in each chamber” (Besley
2006: 159).  is type of situation is described by Persson and al (2000)
who discussed the unifying nature of agenda setting powers in aparlia-
mentary system versus apresidential system: “Aparliamentary system
uni es agenda setting powers while apresidential system diversi es it.
e latter tends to reduce rent extraction and tends towards smaller
government” (Besley 2006: 164). In addition, term limits can impact
responsiveness to voters (Ashworth 2012: 194).
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
e involvement of third parties – media outlets and the branches of
government – also in uences accountability.  e research on this topic
includes Ashworth and Shotts (2010) proposing amodel with amedia
outlet as athird party, Fox and Stephenson (2011) on judicial review,
Fox and Van Weelden (2010) on asituation with aveto player who has
an incentive to appear competent, and Fox and Jordan (2011) on the
possibility of delegation to another bureaucrat (Ashworth 2012: 197–8).
Most studies on parliamentary elections have two major shortcom-
ings. First, they have focused on electoral success by limiting analyses
to “winners” and not realizing that “losers” are equally important for
the entire electoral process. From aformal point of view “who wins
and “who loses” are inseparable dual problems. Only operating on the
full set of candidates for aparticular election can one properly assess the
determinants of winning and losing.  e second shortcoming stems
from ignoring the “history” of elections despite the fact that, for voters,
the decision for whom to vote is usually based on the assessment of
the past performance of candidates’ political parties. In consecutive
elections the reappearances of “old” candidates and appearances of
new ones on the political party lists are crucial for the winning/losing
outcomes as is well documented in the fragmented studies.
AGame  eoretic Approach to Electoral Accountability
We brie y summarize the game-theoretic approach to elections that
was an intellectual impetus for our project (see Fearon 1999; Banks and
Sunderam 1990, and Ferejohn 1986). Our contribution consists of con-
necting this approach with characteristics of the parties and electoral
systems.  e model represents the interaction between politicians and
voters as a“principal-agent” relation. Voters are the principals, and they
elect their agents, politicians, to implement apolicy. Voters are uncertain
about the policy chosen by the politicians and about the politicians’ pref-
erences. In short, they are simultaneously confronted with the problem of
“moral hazard” and the problem of “adverse selection.” As aresult, voters
have to perform two tasks in asingle stroke: they have to induce politi-
cians to implement agood policy, and they have to sort between good
and bad politicians. To accomplish all this, voters have only the right to
vote the representatives of the party out of o ce and to replace them
with the challenger.  is is the essential problem of electoral control.
Representation and Accountability: Intellectual Foundations… 37
e central result of the model is that, in the equilibrium, voters
condition their voting decision on policy consequences.  is is not just
amatter of simple retrospective voting: Voters use their information
about policy consequences to update their beliefs about the party repre-
sentatives. If they conclude that these people are more likely than their
challengers to be “good” then they elect them. Otherwise, they opt for
the challengers. In short, when voter evaluation of policy consequences
is high, voters elect the representatives from that party because good
policy consequences indicate that they are likely to be “good” and thus
likely to behave well in the future. Conversely, when voter evaluation
of policy consequences is low, voters elect the challengers because bad
policy consequences indicate that the party representatives are likely
to be “bad” and thus behave poorly in the future. By using their past
information prospectively, voters induce “bad” incumbents to behave
better. In the equilibrium, “bad” incumbents implement policy that
is between the “bad” party representatives ideal point and the voters’
ideal point. In short, by sorting politicians, voters also sanction them.
In the model, sanctioning is abyproduct of sorting.
In addition to this central theoretical insight, the model generates
two comparative static results.  e rst result is that the reelection
threshold set by the voters depends on the quality of information that
voters have about the policy selected by the incumbent. In the equi-
librium, the reelection of party representatives threshold is adecreas-
ing function of the variance.  e second result is that the reelection
threshold depends on voters’ beliefs about the quality of the challeng-
ers (non-party representatives). When voters believe that the challeng-
ers are likely to be good, they set ahigher standard for the party rep-
resentatives. Conversely, when they are convinced that the challengers
are likely to be “bad,” they become more lenient towards the party
representatives. In the equilibrium, the election of party representa-
tives threshold is an increasing function of beliefs.
In this chapter we summarized the main concepts and theories of rep-
resentation and accountability which are central to electoral control.
Since we are primarily concerned with modern democracies, the
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
notions of representation and accountability are theoretically and sub-
stantively related.
ere are avariety of approaches and perspectives, but all of them
deal with the basics of the ballot box: voters choose candidates and
parties, parliamentarians represent their constituencies, and everyone
pursues their interests through the system of electoral laws in an en-
vironment of imperfect information.  e hope is that democracy will
withstand the pursuit of divergent interests and that good governance
will prevail.
Two other key ideas that interact with representation and account-
ability are democracy and political inequality.  eir interrelationship is
the subject of Chapter Two.
Andeweg, Rudy B. 2003. “Beyond representativeness? Trends in political rep-
resentation.European Review 11(2): 147–161.
Ashworth, Scott. 2005. “Reputational Dynamics and Political Careers.
Journal of Law, Economics, and Organization 21(2): 441–466.
Ashworth, Scott and Ethan Bueno de Mesquita. 2006. “Delivering the
Goods: Constituency Service with Electoral and Institutional Variation.”
Journal of Politics 68(1): 168–179.
Ashworth, Scott and Ethan Bueno de Mesquita. 2008. “Electoral Selection,
Strategic Challenger Entry, and the Incumbency Advantage.Journal of
Politics 70(4): 1006–1025.
Shotts, Kenneth W. and Scott Ashworth. 2011. “Challengers, Democratic Con-
testation, and Electoral Accountability.APSA 2011 Annual Meeting Paper.
Ashworth, Scott. 2012. “Electoral accountability: recent theoretical and em-
pirical work.” Annual Review of Political Science 15: 183–201.
Austen-Smith, David and Je rey S. Banks. 1989. “Electoral Accountability
and Incumbency.” Pp. 121–150 in Models of Strategic Choice in Politics,
edited by Peter Ordeshook, Ann Arbor: University of Michigan Press.
Banks, Je rey, and R. Sundaram. 1993. “Long-Lived Principals, Short-Lived
Agents.” University of Rochester, Working Paper.
Barro R. 1973. “ e Control of Politicians: An Economic Model.” Public
Choice 14(1): 19–42.
Besley, Timothy and Anne Case. 1995. “Incumbent Behavior: Vote Seeking,
Tax Setting and Yardstick Competition.” American Economic Review 85
(1): 25–45.
Representation and Accountability: Intellectual Foundations… 39
Besley, Timothy. 2006. Principled agents:  e political economy of good govern-
ment. New York: Oxford University Press.
Birch, H. Anthony. 2001. e Concepts and  eories of Modern Democracy.
London: Routledge.
Besley, Timothy and Andrea Prat. 2006. “Handcu s for the Grabbing Hand?
Media Capture and Political Accountability.American Economic Review
96(3): 720–736.
Bird, Karen. 2003. “ e Political Representation of Women and Ethnic
Minorities in Established Democracies: AFramework for Comparative
Research.Working Paper Presented for the Academy of Migration Studies in
Denmark (AMID) Aalborg University. Denmark.
Bueno de Mesquita, Ethan. 2007. “Politics and the Suboptimal Provision of
Counterterror.International Organization 61(1): 9–36.
Canes-Wrone, Brandice, Michael C. Herron and Kenneth W. Shotts. 2001.
“Leadership and Pandering: A eory of Executive Policymaking.
American Journal of Political Science 45(3): 532–550.
Canes-Wrone, Brandice and Kenneth W. Shotts. 2004. “ e Conditional
Nature of Presidential Responsiveness to Public Opinion.” American
Journal of Political Science 48(4): 690–706.
Celis, Karen, Sarah Childs, Johanna Kantola and Mona L. Krook. 2008. “Re-
thinking women’s substantive representation.Representation 44 (2): 99–110.
Chaney, Paul and Ralph Fevre. 2002. “Is  ere aDemand for Descriptive
Representation? Evidence from the UK’s Devolution Programme.” Politi-
cal Studies 50(5): 897–915.
Coate, Stephen and Morris, Stephen, 1995. “On the Form of Transfers in
Special Interests,Journal of Political Economy 103(6): 1210–1235.
Alain de Janvry,Frederico Finan,Elisabeth Sadoulet “Local Electoral Incen-
tives and Decentralized Program Performance.” NBER Working Papers
16636, National Bureau of Economic Research, Inc.
Ferejohn, John. 1986. “Incumbent Performance and Electoral Control.”
Public Choice 50: 5–25.
Dovi, Suzanne. 2002. “Preferable Descriptive Representatives: Will Just Any
Woman, Black, or Latino Do?” American Political Science Review 96(4):
Daley, Brendan and Erik Snowberg. 2011. “Even if it’s not Bribery:  e Case
for Campaign Finance Reform.Journal of Law, Economics, and Organi-
zation 27(2): 324–349.
Dropp, Kyle and Zac Peskowitz. 2012. “Electoral Security and the Provision
of Constituent Service.” e Journal of Politics 74(1): 220–234.
Gehlbach, Scott. 2007. “Electoral Institutions and the National Provision of
Local Public Goods.Quarterly Journal of Political Science 2: 5–25.
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
Gordon, Sanford C., Gregory Huber and Dimitri Landa. 2007. “Challeng-
er Entry and Voter Learning.American Political Science Review 101(2):
Gordon, Sanford C. and Gregory A. Huber. 2007. “ e E ect of Electoral
Competitiveness on Incumbent Behavior.” Quarterly Journal of Political
Science 2(2): 107–138.
Gri ths, Phillips A, and Richard Wollheim. 1960. “How can one person
represent another?” Proceedings of the Aristotelian Society 34: 187–224.
Fearon, James D. 1999. “Electoral accountability and the control of politicians:
selecting good types versus sanctioning poor performance.” Pp. 55–97 in
Democracy, accountability, and representation, edited by A. Przeworski, S.C.
Stokes, and B. Manin. Cambridge: Cambridge University Press.
Fearon, James D. “Self-Enforcing Democracy.Quarterly Journal of Economics
126(4): 1661–1708.
Fiorina, Morris P. 1981.Retrospective voting in American national elections.
New Haven: Yale University Press.
Fox, Justin and Matthew C. Stephenson. 2011. “Judicial Review as aResponse
to Political Posturing.American Political Science Review 105(2): 397–414.
Fox, Justin and Stuart V. Jordan. 2011. “Delegation and Accountability.” e
Journal of Politics 73(3): 831–844.
Fox, Justin and Richard Van Weelden. 2012. “Costly Transparency.Journal
of Public Economics 96 (1–2): 142–150.
Healy, Andrew, and Neil Malhotra. 2009. “Myopic Voters and Natural
Disaster Policy.” American Political Science Review 103(3): 387–406.
Holmström, Bengt and Paul Milgrom. 1991. “Multitask Principal-Agent
Analyses: Incentive Contracts, Asset Ownership, and Job Design” Journal
of Law, Economics, and Organization 7 (Special Issue): 24–52.
Holmström, Bengt. 1999. “Managerial Incentive Problems: ADynamic Per-
spective.” e Review of Economic Studies 66 (1): 169–182.
Key, Valdimer O. 1966. e responsible electorate: rationality in presidential
voting, 1936–1960. Cambridge, MA: Belknap.
Kymlicka, Will. 1995. Multicultural Citizenship: ALiberal  eory of Minority
Rights. Oxford: Clarendon Press.
Lohmann, Susanne. 1998. “An information rationale for the power of special
interests.American Political Science Review 92 (4): 809–827.
Manin, Bernard, Adam Przeworski and Susan C. Stokes. 1999. “Elections
and representation.” Pp. 29–54 in Democracy, accountability, and repre-
sentation, ed. A. Przeworski, S. C. Stokes, and B. Manin. Cambridge:
Cambridge University Press.
Markowski, Radoslaw. 2006. “Political Accountability and Institutional Design
in New Democracies.International Journal of Sociology 36(2): 45–75.
Representation and Accountability: Intellectual Foundations… 41
Maskin, Eric and Jean Tirole. 2004. “ e Politician and the Judge: Account-
ability in Government.American Economic Review 94 (4): 1034–1054.
Phillips, Anne. 1995. e Politics of Presence. Oxford: Clarendon Press.
Persson, Torsten and Guido Tabellini. 2000. Political Economics: Explaining
Economic Policy, Cambridge, MIT Press.
Prat, Andrea. 2005. “ e Wrong Kind of Transparency.American Economic
Review 95(3): 862–877.
Mansbridge, Jane. 1999. “Should Blacks Represent Blacks and Women Rep-
resent Women? AContingent ‘Yes’.Journal of Politics 61(3): 628–57.
Mansbridge, Jane. 2000. “What Does aRepresentative Do? Descriptive Rep-
resentation in Communicative Settings of Distrust, Uncrystallized Inter-
ests, and Historically Denigrated Status.” Pp. 99–123 in Citizenship in
Diverse Societies, edited by Will Kymlicka and Wayne Norman, Oxford
University Press.
Mansbridge, Jane. 2003. Rethinking representation. American Political
Science Review 97(4): 515–528.
Pitkin, Hanna F. 1967. e concept of representation. Berkeley, CA: University
of California Press.
Przeworski, A., Stokes, S.C., and Manin, B. 1999. Democracy, accountability,
and representation. Cambridge: Cambridge University Press.
Rogo , Kenneth. 1990. “Equilibrium Political Budget Cycles.American
Economic Review 80(1): 21–36.
Saward, Michael. 2006. “ e representative claim.” Contemporary Political
eory 5(3): 297–318.
Saward, Michael. 2010. e Representative Claim. Oxford: Oxford University
Seabright, Paul, 1996. “Accountability and decentralisation in government:
An incomplete contracts model.” European Economic Review Elsevier
40(1): 61–89, January.
Scharfstein, David S, and Jeremy C Stein. 1990. “Herd Behavior and Invest-
ment.” American Economic Review 80(3): 465–479.
Shabad, Goldie and Kazimierz M. Słomczyński. 2002. “ e emergence of
career politicians in post-Communist Poland and the Czech Republic.
Legislative Studies Quarterly 27(3): 333–359.
Shi, Min and Jakob Svensson. 2006. “Political budget cycles: Do they di er
across countries and why?” Journal of Public Economics 90(8–9): 1367–1389.
Słomczyński, Kazimierz M., Goldie Shabad and Zielinski, Jakub. 2008.
“Fluid Party Systems, Electoral Rules and Accountability of Legislators in
Emerging Democracy.”  e Case of Ukraine. Party Politics 14(1): 91–112.
Swabey, Marie Collins. 1937. “ e eory of the Democratic State.Journal
of Philosophy34 (23): 641–641.
J.K. Dubrow, A. Kurowicka, N. Palaguta, and K.M. Słomczyński
Tavits, Margit. 2009. “ e making of mavericks: Local loyalties and party
defection.” Comparative Political Studies 42(6): 793–815.
Urbinati, Nadia., and Mark E. Warren. 2008. “ e concept of representation
in contemporary democratic theory.Annual Review of Political Science
11: 387–412.
Williams, Melissa S. 1998. Voice, trust, and memory: marginalized groups and
the failings of liberal representation, Princeton University Press, Princeton.
Young, Iris Marion. 1990. Justice and the Politics of Di erence. Princeton Uni-
versity Press.
Zielinski, Jakub, Kazimierz M. Słomczyński, and Goldie Shabad. 2005.
“Electoral Control in New Democracies:  e Perverse Incentives of Fluid
Party Systems.World Politics 57(3): 365–395.
C 
e Electoral Market as a Mechanism
of Political Inequality
oshua jerulf ubrow
For as long as the structure of candidates and parliamentarians resemble
the top of the strati cation ladder rather than its whole, issues of elec-
toral control are  rmly connected to the issue of political inequality.
Despite a trend towards demographic and experiential diversity in
the political elite, it is axiomatic that disadvantaged groups of gender
(Kenworthy and Malami 1999; Paxton and Kunovich 2003), race and
ethnicities (Swain 1995), and other social groups (Norris and Lov-
enduski 1993; Zweigenhaft and Domho 1998) are numerically and
substantively underrepresented in the national legislatures of modern
democracies. Political equality, according to social scientists and de-
mocracy theorists, is afoundation of modern democracy (e.g. Dahl
2006; Verba 2006). Yet, everywhere political power is exercised, there
is political inequality. Political inequality refers to the unequal in u-
ence over decisions made by political bodies and the unequal outcomes
of those decisions (Dubrow 2015). Political inequality is asubtype of
power inequality, visible within the political processes of all kinds of
political structures. In modern democracies, political inequality is si-
multaneously adimension of democracy and adimension of strati ca-
tion. Under-representation is akey indicator of political inequality.
Our understanding of how and why disadvantaged groups are
simultaneously unequal in economic, status and political domains
would be enhanced by specifying the mechanisms that link mi-
cro-level voter behavior to macro-level electoral representation.
Micro- and macro-level studies of women’s and ethnic minorities’
Joshua Kjerulf Dubrow
political representation dominate the empirical  eld while normative
defense of descriptive representation of various groups dominates the
theoretical  eld (see also Bird 2003 and Schwindt-Bayer and Mishler
2005: 407). In descriptive representation, the composition of the rep-
resentative body resembles the demographics and experiences of the
citizenry (Mansbridge 1999: 629; Pitkin 1972: Chapter 4).
As ameans to explain how micro-level actions of voters and parlia-
mentarians construct the composition of anational legislature, many
scholars co-opt the economic language of market dynamics, where
supply of and demand for types of candidates governs anation’s state
of descriptive representation (Cannon 1999; Norris and Franklin
1997; Paxton and Kunovich 2003).  e implication is that when the
supply of descriptive candidates increases, voters will vote for them,
and descriptive representation will be achieved (Dubrow 2007). Use
of market imagery creates astate policy continuum with government
intervention at one extreme and laissez-faire market solutions at the
other. While the concept of an electoral market is auseful theoretical
guideline, its empirical parameters are more often assumed than tested.
Research into government intervention centers mainly on consti-
tutional and electoral law quotas and reserved seats for women and
ethnicities (see also Htun 2003 and Chapter 10 of this book). Whether
government intervention, in and of itself, enhances descriptive rep-
resentation depends on the type of law implemented and how it is
enforced (Dahlerup and Freidenvall 2005). For example, reserved seats
can actually serve to reduce womens progress in gaining representation
above the level set for reservation. In addition, quota management can
be used against womens interests when party leaders are more interest-
ed in recruiting docile descriptive representatives who do not challenge
the status quo (Dahlerup 2002). Government intervention for ethnici-
ties follows asimilar logic. Active descriptive representation policy for
ethnicities is a potential means for securing political stability, espe-
cially during the formation of heterogeneous countries born of ethnic
con ict, and can be achieved through electoral support for ethnic
parties and formal guarantees of political representation (Htun 2004;
Juberias 2000). Government intervention for ethnicities, however, can
have negative, unintended consequences, depending on how the law
was written, implemented, and enforced, and how it is perceived by
other ethnic groups (Stein 2000).
The Electoral Market as a Mechanism of Political Inequality 45
Research into market solutions is largely limited to the e ectiveness
of voluntary party quotas for enhancing women’s representation and
government support for ethnic party competition without electoral
guarantees (Kittilson 2006; Juberias 2000). Voluntary party quotas for
gender have been shown to increase womens representation, though
like other types of descriptive representation policy, success depends
on the form and magnitude of the quota, along with the sanctions for
non-compliance.  ough endorsement of gender quotas is widespread
in the womens representation literature, there is disagreement as to the
necessity of quotas. Dahlerup and Freidenvall (2005) argue, “Major
historical leaps in womens parliamentary representation can occur
without quota provisions” (27). Jane Mansbridge, anoted descriptive
representation scholar, presents acontrary view; “… as Iwrite, signi -
cant representation by gender cannot be achieved in any existing polity
without some form of quota” (2005: 622).
Quotas, however, are not the entirety of market solutions. Arange
of non-government action, such as party level a rmative action, social
movement pressures, and changing ideology, all have the possibility
to increase descriptive representation for the disadvantaged.  us, in
order to understand whether market solutions can enhance descrip-
tive representation for the disadvantaged, we need to explore beyond
just party quotas and analyze the entirety of the political market as
afeasible mechanism for change.
Linking attitudes and behaviors of political leaders and voters to
the demographic composition of parliaments, Iadvocate the use of
concept of an electoral market as amicro to macro level mechanism
to explore the determinants of how new democracies create their
state of descriptive representation in the national legislature for dis-
advantaged social categories.
Elections link masses to political elites and are acrucial mecha-
nism through which political leaders understand mass interests.  us,
repeated elections are supposedly designed to enable the masses to
achieve one or more of the following goals: (a) sanction directly the po-
litical elite by either voting for or against the extension of incumbents’
term in o ce and (b) vote for the political leaders that best represent
their interests (Manin et al. 1999). Central and Eastern Europe has had
great electoral volatility (Lawson 1999: 32–33; Zielinski et al 2005),
caused in part by social cleavages translating their dissatisfaction into
Joshua Kjerulf Dubrow
vote choice during subsequent, and frequent, economic downturns
(Tavits 2005, Roberts 2008). If political stability is aresult of parties
rooted in social cleavages (Elster et al. 1998: Chapter 7; Lipset and
Rokkan 1967) , then adequate gender, class, ethnic and other social
groups’ representation in national legislatures in the course of repeated,
free, and fair elections is crucial.
Perhaps the most convincing way to counter to criticisms, demon-
strating the importance of studying the subject, is by assessing whether
the demographics of the representative make ameasurable di erence in
the representation of the disadvantaged (Mansbridge 1999; Ogmund-
son 2005: 319–323). Any impact of descriptive representation could
be felt in two main ways; (a) raising constituents’ political engagement
and/or (b) descriptive representatives’ impact on legislative processes.
While some  nd that descriptive representatives make little dif-
ference in either of these areas (Lawless 2004; Swain 1995), others
show that descriptive representatives do have ameasurable impact. In
terms of raising constituents’ political engagement, whites and blacks
are more likely to contact representatives of the same race (Gay 2002)
and women are more likely to become politically active in states with
competitive and visible women candidates (Atkeson 2003). Some
explain this phenomenon in terms of legitimacy, in that “constituents
are more likely to identify with the legislature and to defer to its deci-
sions to the extent that they perceive asigni cant percentage of ‘people
like themselves’ in the legislature” (Schwindt-Bayer and Mishler 2005:
As for impacting legislative processes, research is mixed on whether
demographic characteristics in uence roll-call behavior, but is more
de nitive on policy introduction (for areview, see Swers 2002: 7–17
and Reynolds 1999: 548). Partisanship seems to have alarger impact
on roll-call voting than voter’s demographic ties to their constituency
(e.g., Latino dominated districts exert little in uence on their Latino
representatives (Hero and Tolbert 1999). As for gender, net of partisan-
ship, women legislators used to consistently vote more liberally than
men, though this behavior has attenuated over time (Welch 1995).
Blacks consistently vote more liberally than whites and support more
Black issue bills, aphenomenon dating back to Reconstruction (Cobb
and Jenkins 2001). One cross-national study of wealthy, industrial-
ized countries found that greater percentages of women in legislative
The Electoral Market as a Mechanism of Political Inequality 47
positions positively impacts foreign policy in terms of providing devel-
opment assistance (Breuning 2001).
Dubrow (2007) used the concept of demographic cues, i.e. when
voters vote based on the individual characteristics of the candidates,
to argue that it is theoretically possible for the disadvantaged to “vote
in” their own descriptive representatives.  e main hypothesis was
that voters are more likely to vote for parties that are demographically
similar to them. For example, women are more likely to vote for parties
that have women in them. He found that, for the demographic groups
who do vote based on demographic cues, the contribution to overall
voting behavior is much smaller than that of political attitudes toward
the economy, the State Socialist past, and the role of the Church in po-
litical life. Dubrow (2007) concluded that demographic cues voting in
Poland from the 1990s to the early 2000s, with its laissez faire political
market consisting only of voluntary gender quotas, could not produce
adescriptively similar government. Stronger government intervention
could produce what alaissez faire approach did not.
Stronger impacts are felt in terms of policy introduction. Swers’
(2002) study of the policy impacts of women legislators strongly
suggests that women are more likely to introduce womens issues and
feminist bills into legislative consideration than their male colleagues
(see also  omas 1994). Others  nd that Blacks are more likely to in-
troduce Black issue bills and work harder for their passage than whites
(Cannon 1999: Chapter 4; Whitby 2002), though some  nd the
results somewhat mixed (Tate 2003: Chapter 4). In Eastern Europe,
descriptive representation of disadvantaged groups in uences the form
and magnitude of policy, but primarily through ethnic political parties
(Chiribuca and Magyari 2003; see also Xydias 2008).
All told, it appears that descriptive representation does matter in
that demographic groups both raise constituents’ political engagement
and impact the legislative process. It should be noted, however, that
because disadvantaged groups have only recently entered the political
elite (Zweigenhaft and Domho 1998), research on the direct impact
of descriptive representatives is in its early stages.
Joshua Kjerulf Dubrow
Under-representation of demographic groups in the parliament is aform
of political inequality. Political inequalities are found throughout the po-
litical process, and include (A) voice, identi ed in the political process as
participation (electoral and non-electoral) and representation (govern-
mental and non-governmental, i.e. by third sector organizations); and
(B) response, identi ed primarily in terms of policy (symbolic, formal
and informal). For afull discussion, see Dubrow (2015).
Dahl’s (2006) work suggests that democracy and political equality
are distinguishable in that democracy is asystem of rights and political
processes, whereas political equality is the extent to which di erent par-
ticipants have reasonably similar opportunities for voice and response
within that system of rights and processes.  at political inequality is
acreature of the political process does not reduce all political processes
to political inequality – it merely points out where we should suspect it
(Dubrow 2015). From the equality of political opportunities side, the
study of political inequality is ahunt for these structured di erences
in individual, group or organizational in uence over government deci-
sions. From the equality of political bene ts side, it suggests how we
should view any situation in which political processes systematically
and historically lead to apattern of unequal political outcomes.
e APSA Task Force has initiated what can be called an American
school of economic inequality in politics.  is school includes  eda
Skocpol et al in the American Political Science Task Force on Inequal-
ity and American Democracy (2004), Robert Dahl’s in “On Political
Equality” (2006), as well as Gilens and Page (2014), among others,
who argues that there are three major dimensions of political inequal-
ity: Citizen voice, government response, and patterns of public policy–
making. Citizen voice refers to how people transmit their opinions and
interests to their representatives. Governments can choose to respond,
or not, under the threat of electoral accountability. Patterns of public
policy-making can be discerned from studying who gets what and why.
Economic inequality depresses participation and opens gaps between
the political power of social groups (Solt 2015).  ey warn that the
pluralistic aspects of American democracy – the interplay of voice
and response – are imperiled by economic inequalities.  e reason
is that the disadvantaged are less represented and less involved in
The Electoral Market as a Mechanism of Political Inequality 49
political participation and government o cials and policy makers are
less inclined to be responsive to the preferences of the disadvantaged.
Adescription of political inequality should consist of an interlinked
set of components that can be applied to a wide array of contexts
across levels of decision-making, time, and place, with asimple, inter-
disciplinary vocabulary. As such, Idescribe ageneric political process
that can be used to explain structured inequalities in in uence over
the decisions of decision-making bodies.  e purpose of identifying
the process is to indicate where political inequality is and how it is
connected to political inequality found throughout.It allows us to
localize where the political inequality occurs, and where it a ects other
parts. In sum, the aim of the model is to show (a) various types of po-
litical inequalities, (b) how these political inequalities interact, and (c)
how these multiple, simultaneous and interacting political inequalities
contribute to aperson’s, groups, or organizations overall level of politi-
cal inequality. My description of the process is based on the American
school, and as such Ihave accepted their assumption that voice and
response is where democracy and inequality meet, and Icannot think
of other relevant parts of the political process that they did not cover.
Voice refers to how constituencies express their interests to deci-
sion-makers directly or through representatives. Voice has two main
components of participation and representation and each has subcom-
ponents. Participation can be electoral – such as voting or standing
for o ce in elections – or non-electoral, such as attending ademon-
stration, contacting apublic o cial, or joining a political organiza-
tion, among others. Representation means that aperson or an organi-
zation interprets the political voice and transmits their interpretation
to the decision-making body. Government representation refers to
those individuals and government agencies whose function is to listen
to the interests of their constituency. Examples include parliamen-
tarians, Ombudspersons, and special o ces directed by the govern-
ment. Non-governmental representatives include NGOs that can be
non-pro t or for-pro t and whose function is to transmit the interests
of their selected constituency to decision-making bodies. NGOs can
be non-violent organizations such as Kongres Kobiet in Poland and
the NAACP in America, or they can be organizations that advocate
violence. People and organizations who interpret voices may or may
not be expressly or directly appointed by their constituency.
Joshua Kjerulf Dubrow
Response refers to how decision-makers act and react to their
constituencies and is expressed via policy and symbols. Policy can be
de ned as adeliberate plan of action to guide decisions to achieve
speci c outcomes with the intention of providing speci c guidelines
for future, related decisions. Policy has two main forms: formal and
informal. Formal policy refers to legislation, judicial precedent or ex-
ecutive directives that are written and have the force of codi ed law.
Informal policy consists of rules that lack force of codi ed law yet
impact how decisions are made. For example, when apolitical party
leader directs the organization to include more minorities in their can-
didate lists, this can either be formal within the party rule book or an
informal rule, something that the rank-and- le know about but on
which there is no paper-trail.
Symbolism is aresponse that does not set  rm guidelines for future
decisions. Symbols are often publicly expressed in limited-time events,
such as aspeech on the parliamentary  oor or introducing – but not
passing – legislation. Symbols also manifest in commemorative events
such as “Black History Month” in the U.S. or “ e Decade of the
Roma” as proclaimed by the United Nations.
Symbols and policy are distinguished not by their intent, but by
whether the response is arecognizable directive to guide future action
and the extent to which the directive is enforced. To declare a“Decade
of the Roma,” for example, is symbolic in the absence of policy
measures. Enforcement is the bright line between policy and symbol-
ism: loosely enforced policy is tantamount to symbol. Acomplicating
factor is when policy is in the form of asymbol, such as “resolutions
and other non-legislative statements. “Symbolic policy” is expressed as
policy yet carries the force of symbol (see also Tomescu-Dubrow et al
Sensitizing Principles
Iposit the following sensitizing principles as guidelines for the interac-
tion between components of the conceptual framework.  e principles
are designed to cover awide range of political contexts.
Dominating Principle of Context Dependence: Social
context rules all principles and all components.
The Electoral Market as a Mechanism of Political Inequality 51
Principle of Interaction: At any point, all or most of the com-
ponents in uence each other. At an abstract level, voice and response
are in acontinuous direct feedback loop.
Principle of Direct and Indirect Effects: Within voice
and response, relationships between subcomponents may be indirect
or direct. At some point, avoice subcomponent has adirect in uence
on a response subcomponent. For example, representatives [voice]
make policy [response]; via campaign contributions (legal or other-
wise), non-electoral political participation [voice] can directly in u-
ence symbols [response].
Principle of Unequal Influence of Components:
Components di er in the extent to which they in uence political in-
equality. Di erence depends on context, where in Country Aat one
point in time some components emerge as more important than in
other countries and times. In general, symbolic responses are less e ec-
tive than policy in in uencing the extent of political inequality.
Principle of Disproportionate Response: Loudness of
voice is not always met with aproportionate response. At times, very
loud voices are met with zero response, or with symbolic response. At
other times, very quiet voice, or even implicit silence (e.g. elite con-
sensus), is met with high response, as in the case of economic policy
favoring the wealthiest, whether or not they expressly voice their inter-
ests to decision-makers.
Principle of Unintended Consequences: Voice may elicit
an unintended response. Loud voice may amplify political inequal-
ity, as in the case of regimes suppressing dissent.  is assumes voice
is intentional, and the intention can be correctly interpreted by deci-
sion-makers.  e term “backlash”  ts here.
Principle of Simultaneous Inequalities: Inequality
between individuals, groups or organizations can occur within any of
the subcomponents. Inequalities between subcomponents interact and
contribute to an overall level of political inequality. For example, in-
equality is possible in participation and representation. As for their
interaction, low political participation by aspeci c group – either elec-
toral or non-electoral – can contribute mightily to their under-rep-
resentation in governmental and non-governmental bodies. Voice
Joshua Kjerulf Dubrow
inequality then in uences response inequality, where policy does not
re ect their interests.
Debate on how best to achieve descriptive representation produces
awide spectrum of policy types. Half of the countries in the United
Nations have had some descriptive representation policy within their
electoral market (Dubrow 2006). Examples of government interven-
tion in the electoral market include “selective” forms of descriptive rep-
resentation policy (Mansbridge 1999: 632–33) such as race-conscious
re-districting, lower thresholds for ethnic minority parties (Juberias
2000: 35), the establishment of representative councils for consulta-
tion on certain types of legislation, quotas mandated by electoral or
constitutional law, and reserved seats. Examples of market solutions
include a rmative action or quotas voluntarily adopted by political
e electoral market imagery is attractive, most likely because the
assumptions of market dynamics are well known and well tested in
other areas of research. Electoral market is often used as analogy and
metaphor rather than theory (Udehn 1996: Chapter 3). Use of the
concept of an electoral market is decidedly situational, depending
wholly on the phenomenon under study.  e concept of an electoral
market remains atheoretically fruitful means of understanding social
phenomena, as it provides one of the only ways to link micro-level
behavior to macro-level outcomes in the social sciences (see also
Coleman 1990).
ough combining political and economic concepts to form theories
of social behavior are common, including the use of the term “elec-
toral market,” few explicitly de ne what they mean (see also Udehn
1996: Chapter 1). Since Becker ‘s e Economic Approach to Human
Behavior (1976), researchers have used market language of supply and
demand to explain social phenomena such as occupational attainment
(Kaufman 2002), marriage (Cherlin 1992), religion (Gill 2001), and
crime (Becker 1976). Economists examining political processes in
the context of public choice theory are most likely to use the term
The Electoral Market as a Mechanism of Political Inequality 53
political or electoral market, but few discuss it substantively (Wittman
1995:1). In e Myth of Democratic Failure (1995), Donald Wittman,
an American economist, asserts the primacy of the market in writing,
“Most controversies in the social sciences are ultimately arguments
over the nature of the market” (1). Recently, market language and the
analogy of apolitical market has been used to explain variations in
the representation of disadvantaged groups (Cannon 1999; Norris and
Franklin 1997; Mackay 2004; Paxton and Kunovich 2003).
Since the goal is to understand how micro-level political market
dynamics produces agiven state of descriptive representation, Ineed
to construct an electoral market as alink between micro-level behavior
and macro-level outcomes. Supply and demand dynamics need to be
well de ned, as well as their constituent levels of analyses and the actors
within them. Moreover, social structural, contextual factors need to be
clearly delineated as both conceptually distinct yet theoretically inte-
grated into the actions of the various actors. Additionally, the concept
should be durable enough to be comparable across time and space.
Conceptions of the electoral market are similar to the more
well-known labor market. In studies of the labor market, the supply
side is generally conceived as those factors that in uence the kind of
job aspirants available. Demand side consists of the employers who
seek types of job aspirants. In the electoral market, especially when
used to explain candidate emergence, the supply side factors are such
as candidate resources and motivation (Mackay 2004) that in uence
the kind of candidate aspirants available; the demand side as the parties
who seek types of candidates.
Following previous use of market imagery, de nitions of who or
what in uences supply and demand vary substantially from study to
study. In macro-level studies of women’s representation, macro-po-
litical and economic forces de ne supply and demand and govern
the dynamics of the market. In explaining cross-cultural variation in
womens representation in national legislatures, Paxton and Kunovich
(2003) de ne the supply side as structural factors, such as education
levels and workforce participation of women, and the demand side as
political parties and electoral systems.
ere are alternative visions of the concept of the electoral market.
Micro-level studies seem to use supply and demand as all factors
that contribute to candidate emergence and selection. In explaining
Joshua Kjerulf Dubrow
descriptive representation in the European Parliament, Norris and
Franklin (1997) conceive of a“political market-place;” on the supply
side are candidates and their characteristics, including motivations and
political capital, and on the demand side are attitudes of political gate-
keepers and party rules. In explaining Black minority representation
in the U.S. Congress, Cannon (1999: Chapter 3) de nes the supply
side as individual politicians, the demand side as voters and their social
Voters are rarely considered as important actors in the supply and
demand dynamic. Acontrast with how labor markets work is illustra-
tive.  e key di erences between the labor market and the electoral
market are the factors that potentially in uence supply. Labor market
employers rarely have to consider how the public at-large thinks about
hiring decisions and thus have agreat capacity to conceal discrimina-
tory hiring practices. Parties, on the other hand, must consider public,
or voter, reactions. Moreover, voter demands are vital for party survival
and hence must be signi cant in in uencing supply.
us, my conception of the electoral market is more similar to that
of Cannon (1999: Chapter 3), who includes voters into the supply
and demand dynamic. Including the voter forces achange in the entire
concept, de ning voters as those who demand descriptive representa-
tion and political leaders as those who control the supply of demo-
graphic types of candidates.  is change allows researchers to ask new
questions concerning the ability of market solutions to produce de-
scriptive representation and the extent to which voters play arole in
this process.
Amarket can be de ned as an institution that governs distribution of
resources (Carruthers and Babb 2000). Apolitical market is an institu-
tion that governs the distribution of representation and other political
goods using the concepts of supply and demand.  is leads to the
following proposition:  e state of descriptive representation at the
macro-level is afunction of the electoral market such that micro- and
meso-level actors and their interactions determine supply and demand
for types of candidates under varying contextual restraints.
The Electoral Market as a Mechanism of Political Inequality 55
Levels and  eir Actors
ree levels of analyses, each with their own actors, comprise the po-
litical market (Table 1).
Table 1. Levels and Actors in an Electoral Market
Micro Meso Macro
Actors Legislators Political Parties Legislature
Candidates Social Change Organizations
At the micro-level, individuals are the units of analysis. Actors
include legislators, candidates, and voters. For both the meso- and
macro-levels, organizations are the units of analysis. In uencing both
micro-level behaviors and the state of descriptive representation is the
meso-level, comprised of political parties and social change organiza-
tions. At the macro-level is the legislature, whose demographic compo-
sition is aproduct of both micro and meso-level actions. In uencing
all actors at all levels are the social structural, or contextual factors.
Actors of the supply side are individuals in the political institutions,
e.g. political leaders, including political party leaders and the rank and
le legislators. Political institutions in uence supply by impacting
the structure of opportunities in the political market, providing the
contexts necessary for candidates of all demographic stripes to emerge
(Kunovich and Paxton 2005; Norris and Franklin 1997; Wilcox et al.
2003: 44). Political leaders’ attitudes towards descriptive representa-
tion heavily in uence the supply of available descriptive representatives
(Cannon 1999; Norris and Franklin 1997). Political elites are capable
of holding strong attitudes toward descriptive representation as agov-
ernance solution (Chaney and Fevre 2002), lending support for the
contention that these in uential subgroups help foster attitudes, posi-
tively or negatively, towards descriptive representation.
Actors of the demand side are individual voters. Voter demand is
measured in two ways; (1) using attitudes towards descriptive repre-
sentation in theory and (2) by voting behavior through demographic
Joshua Kjerulf Dubrow
cues. Attitudes toward descriptive representation in theory are the in-
tellectual foundations of behavior outcomes, in uencing propensity to
vote for descriptive representatives.  e second demand signal, voting
behavior, assumes that individual traits on the part of the candidate
cues voting behavior which in turn signals to parties ademand for
types of candidates. While the intellectual foundations of demand are
concealed from suppliers, vote choice is visible.
Market language as used in explaining socio-political phenomena has
its roots in sociological rational choice theory (RCT), and from eco-
nomics in public choice theory and the assumption of economic ra-
tionality in actors’ approach to political decisions (Buchanan 1968;
Downs 1957; Hechter and Kanazawa 1997; Kiser and Bauldry 2005).
As amultilevel theory, RCT assumes that individuals are the starting
point, though in uenced by macro-level conditions (see also (Coleman
1990: 5). In essence, RCT assumes Beckers (1976) homo economicus,
where individuals engage in utility maximization, sel shly seeking to
increase bene ts and minimize costs in any exchange situation.  us,
rational action is that which provides the greatest personal good with
the least bit of bad.
RCT has been criticized on many fronts (for areview see Green and
Shapiro 1994).  ere are two main criticisms of RCT’s main assump-
tions; the stability of preferences and the capability of individuals to
make rational decisions based on available information. Iaddress each
in turn.
Iassume relative stability of preferences across time which, for my
purposes, is reasonable. Ipresent what Hechter and Kanazawa (1997)
call a“thick” model of rational choice. Opposite to “thin” models, thick
ones “specify the individual’s existing values and beliefs” (Hechter and
Kanazawa 1997: 194). For disadvantag ed groups, whose structural
position is heavily determined by their lack of political resources, pref-
erences are dependent on whether they are voters or political leaders. As
explained and tested in later chapters, preferences for both are for de-
scriptive representation for their particular social group.  us, women
prefer women representatives, farmers prefer farmer representatives,
The Electoral Market as a Mechanism of Political Inequality 57
young people prefer youth representatives, and so on. However, demo-
graphic intersectionality and within-group heterogeneity of social and
political attitudes cause variation in preferences.
Ialso assume that individuals, be they voters or parliamentarians,
have su cient information to realize their preferences, with some key
quali cations. Iassume abounded rationality within institutional con-
straints (Ingram and Clay 2000; see also Chapter 1 on accountability
and information availability).  is leads to two quali cations. First,
assuming an open-information environment, politicians are aware
of voter sentiment towards descriptive representation through voter
behavior and have the possibility of reacting in asupply and demand
appropriate fashion. Also, in the absence of perfect information,
Iargue that both have just enough information to realize preferences,
e.g. voters can determine demographic types of candidates.
Second, in an electoral market, hidden information can damage the
e ectiveness of the electoral process to produce adescriptively similar
legislature.  us, I argue that it is exactly the de cient information
possessed by both demanders and suppliers that can fuel electoral
market ine ectiveness. However, there is reason to support the conten-
tion that even if there was perfect information, descriptive representa-
tion without government intervention would be di cult to achieve.
us, the extent to which the electoral market is rational is an open
empirical question.
Dynamics of supply and demand should function such that if voters
signal demand, political leaders would furnish supply proportional to
the demand.  us, supply of demographic types of candidates should
be directly related to demand for those candidates.
How demand in uences supply is atheoretically thorny problem.
Some previous researchers consider voter demand to be an inadequate
way of in uencing supply. Demand for women representatives as ex-
pressed by the voting public, for example, has been referred to as un-
organized, di used and as ageneral function of attitudes toward the
role of women (Wilcox et al. 2003: 43; see also Dubrow 2007). While
demand by voters is not signaled so clearly as that of social movement
Joshua Kjerulf Dubrow
groups working speci cally for descriptive representation, the degree to
which the signal is unorganized and di used is an empirical question, as
is the basic assumption that demand is signaled within elections at all.
us, in the electoral market as Ide ne it, the supply of candidates
is partially driven by demand of voters, which is antithetical to current
literature on candidate emergence (Fox and Lawless 2004; Norris and
Lovenduski 1993, 1995). Candidate selection by parties is largely secre-
tive (Kunovich and Paxton 2005). Some qualitative evidence suggests,
however, that political leaders exercise prejudice against certain types
of candidates:
People start with prejudices about the candidates. In the old days,
they used to band them and say, ‘We’re not having anybody under
40, nobody over 50, or we’re not having alawyer, or we’re not having
somebody from the south-east,’ or whatever it might be. Whatever
prejudice they decided to start with knocked out awhole lot of people
many of whom might have been exactly what they really wanted.
(AConservative member of the British parliament, as quoted in Norris
and Lovenduski 1993: 379).
In another research project, in Poland, some members of parlia-
ment thought the idea of descriptive representation was untenable
(Pawlowski and Dubrow 2011):
“ e Parliament is alegislative body, whose aim is to create laws for
all citizens, not for the special interests of any social, ethnic, religious,
etc group.  e state stands for all equally, whether they are red-headed
or blond. If we accept the idea that the make-up of Parliament is to
mirror that of society it would mean that we are returning to the time
of socialist realism, where a32-year old teacher with three children
from asmall town could become arepresentative.  is is nonsense.”
(Member of the Sejm, quoted in Pawlowski and Dubrow 2011: 311).
Norris and Lovenduski (1993) argued that examining supply side
explanations for candidate emergence, de ned as those factors that in-
uence the propensity of people to apply, is more fruitful than examin-
ing prejudice on the part of political leaders. Some assume prejudice,
they argue, by looking only at the outcomes of elections, aquestionable
strategy considering that supply side factors may be more in uential.
The Electoral Market as a Mechanism of Political Inequality 59
Iargue, however, that candidate emergence is at least partially due
to voter demand for types of candidates, asituation that can be gleaned
by doing exactly what Norris and Lovenduski (1993) caution against;
examining outcomes (379). If parties are rational, then they seek to
maximize seat allotments. If vote outcomes are the sum of rational
decisions, then party reactions are also the sum of rational decisions.
Assuming parties know how voters vote, which is likely considering the
extent to which exit polls are taken in modern industrialized nations,
parties are aware of which demographic types are voting for them (or
not).  us, demand increases from aparticular voting demographic
should lead to supply increases for that demographic’s type of candi-
dates, asituation gleaned from examining vote outcomes.
Voting and voter demand for types of candidates are primar-
ily driven by three relatively interrelated factors; policy, accountabil-
ity, and demographics of the candidate. Policy preferences in terms
of partisanship and economic voting dominate all recent theoretical
models of vote demand (Brooks and Brady 1999; Lewis-Beck and
Stegmaier 2000). Accountability, or the sanctioning of governments,
is often measured in terms of policy preferences, though it includes
other factors such as the in uence of aparliamentarian’s roll-call voting
history (Przeworski et al. 1999; Zielinski et al. 2005). Finally, demo-
graphics of the candidate have recently been shown to be afactor in the
voting calculus (Dubrow 2007). As disadvantaged groups have only
recently penetrated the spheres of the political elite, examining the
contribution of individual qualities of candidates to the vote calculus
is ayoung – but rapidly maturing – literature, and its contributions are
just beginning to be understood.
For laissez-faire electoral markets to produce adescriptively similar
legislature, two main assumptions need to be met. Electoral market
analysis assumes that parties are responsive to voter demands. Voter
voice is met by party response. Whether aparticular electoral market
is capable of producing adescriptively similar legislature in the light
of the relative capability of suppliers to recognize the demand signals
and react in arational fashion is an open empirical question. Another
Joshua Kjerulf Dubrow
principle concern is the extent to which there is achoice of descrip-
tive representatives from which the voting public can choose. Electoral
market analysis assumes that every major social cleavage, gender, social
class, and age, being primary demographics, are available on the ballot.
Electoral markets would not be able to produce adescriptively similar
legislature if voting ballots are overly demographically homogenous.
Market success and failure is dependent on the extent to which
these assumptions are met. If agroup loses descriptive representatives
while signaling preferences for descriptive representatives of their type,
and there is no government intervention on their behalf, then purely
market solutions to representational inequality contribute to political
market failure.
is chapter connects the ideas of political inequality and electoral
control in modern democracies by focusing on descriptive representa-
tion. To summarize:  e legislatures of Central and Eastern Europe,
as is the case around the world, re ect the top of the strati cation
ladder rather than its whole.  is fact violates the principle of descrip-
tive representation, which states that the parliament should resemble
the demographics and experiences of its constituency.  us, poor de-
scriptive representation is an indicator of political inequality, de ned
as the extent to which there are structured di erences in in uence over
decisions made by political decision-making bodies, as well as in the
outcomes of those decisions. We get closer to achieving afundamen-
tal principle of democracy – that of political equality – when there is
reasonable descriptive representation.  is is not the case, and thus
modern democracies are systems in which political inequality thrives.
is chapter then posits atheory of how parliaments become politi-
cally unequal places through the metaphor of an electoral market. An
electoral market contains dynamics of supply and demand for di erent
types of demographic candidates and parliamentarians. While some
put voters on the supply side and political parties on the demand side,
Iargue for the opposite: Voters demand types of candidates and political
parties supply them. An e cient electoral market is when voters signal
demand to parties and parties respond to voter demand by supplying
The Electoral Market as a Mechanism of Political Inequality 61
the types of candidates that voters want. An ine cient market is where
voters demand but parties do not supply what the voters demand.
Akey factor is governmental intervention. Governments can intervene
in the electoral market by insisting that parties supply aparticular type
of candidate, under threat of sanctions. Governmental intervention
does not necessarily lead to amore e cient market and by its interven-
tion, it can privilege some demographic types over others. However,
without such an intervention, some demographic types of candidates
would not appear, or appear rarely, and the parliament would not get
appreciably closer to resembling the demographic and experiential di-
versity of the citizenry.
Electoral markets exist within the political process and can be un-
derstood as asubset of the relationship between political voice and po-
litical response. In an electoral market, we talk of voters and parties. In
the political process as awhole, we talk of everyone with avoice – in-
dividuals, groups, and organizations – and the response to those voices
by the government. Electoral markets are thus connected to political
equality and can be used to explain how legislatures become more or
less equal over time.
Ajzen, Icek. 2001. “Nature and Operation of Attitudes.” Annual Review of
Psychology 52:27–58.
Alvarez, Michael R. and Jonathan Nagler. 1998. “When Politics and Models
Collide: Estimating Models of Multiparty Elections.American Journal of
Political Science 42(1):55–96.
Alvarez, Michael R. and Jonathan Nagler. 1998. “When Politics and Models
Collide: Estimating Models of Multiparty Elections.American Journal of
Political Science 42(1):55–96.
American Political Science Association (APSA) Task Force on Inequality and
American Democracy. 2004. “American Democracy in an Age of Rising
Inequality.” Retrieved July 4, 2007 (
Andweg, Rudy B. 2003. “Beyond Representativeness? Trends in Political
Representation.European Review 11(2):147–61.
Atkeson, Lonna R. 2003. “Not All Cues Are Created Equal:  e Conditional
Impact of Female Candidates on Political Engagement.Journal of Politics
Joshua Kjerulf Dubrow
Becker, Gary S. 1976. e Economic Approach to Human Behavior. Chicago:
University of Chicago Press.
Bernstein, Robert A. and Jayne D. Polly. 1975. “Race, Class and Support for
Female Candidates.Western Political Quarterly 28(4):733–36.
Bielby, William T. and James N. Baron. 1986. “Men and Women at Work:
Sex Segregation and Statistical Discrimination.” American Journal of Soci-
ology 91(4):759–99.
Birch, Anthony H. 2001. Concepts and  eories of Modern Democracy, 2nd
Edition. London: Routledge.
Birch, Sarah. 2003. “Women and Political Representation in Contempo-
rary Ukraine” pp. 130–152 in Women’s Access to Power in Post-Communist
Europe edited by Richard Matland and Kathleen Montgomery, Oxford
University Press: Oxford.
Birch, Sarah. 2000. Elections and Democratization in Ukraine. Macmillan
Press: London.
Bird, Karen. 2003. “ e Political Representation of Women and Ethnic
Minorities in Established Democracies: AFramework for Comparative
Research.Working Paper Presented for the Academy of Migration Studies in
Denmark (AMID) Aalborg University. Denmark.
Box-Ste ensmeier, Janet M., David C. Kimball, Scott R. Meinke, and Kath-
erine Tate. 2003. “ e E ects of Political Representation on the Elec-
toral Advantages of House Incumbents.Political Research Quarterly
Brambor,  omas, William R. Clark, and Matt Golder. 2006. “Understand-
ing Interaction Models: Improving Empirical Analyses.” Political Analysis
Breuning, Marijke. 2001. “Womens Representation and Development As-
sistance: ACross-National Study.” Women & Politics 23(3):35–55.
Brooks, Clem and David Brady. 1999. “Income, Economic Voting, and
Long-Term Political Change in the U.S., 1952–1996.” Social Forces
Brooks, Clem. 1994. “Class Consciousness and Politics in Comparative Per-
spective.” Social Science Research 23: 167–195.
Cannon, David T. 1999. Race, Redistricting, and Representation:  e Unin-
tended Consequences of Black Majority Districts. Chicago: University of
Chicago Press.
Carruthers, Bruce G., and Sarah L. Babb. 2000. Economy/Society: Markets,
Meanings, and Social Structure. ousand Oaks, CA, and London: Pine
Forge Press.
Castle, Marjorie and Ray Taras. 2002. Democracy in Poland. Boulder, CO:
Westview Press.
The Electoral Market as a Mechanism of Political Inequality 63
Caul Kittilson, Miki. 2006. Challenging Parties, Changing Parliaments: Women
and Elected O ce in Contemporary Western Europe. Columbus, OH:  e
Ohio State University Press.
Caul, Miki. 2001. “Political Parties and the Adoption of Candidate Gender
Quotas: ACross-National Analysis.” e Journal of Politics 63(4):1214–29.
Chan, Kenneth K. L. 2001. “Idealism Versus Realism in Institutional
Choice: Explaining Electoral Reform in Poland.” West European Politics
Chaney, Paul and Ralph Fevre. 2002. “Is  ere aDemand for Descriptive
Representation? Evidence from the UK’s Devolution Programme.” Politi-
cal Studies 50:897–915.
Cherlin, Andrew J. 1992. Marriage, divorce, remarriage. Cambridge, Mass.
:Harvard University Press.
Chiribuca, Dan and Tivadar Magyari. 2003. “ e Impact of Minority Par-
ticipation in Romanian Government.ANew Balance: Democracy and
Minorities in Post-Communist Europe, edited by Monica Robotin and
Levente Salat. Budapest: LGI Books.
Chiva, Cristina. 2005. “Women in Post-Communist Politics: Explaining
Under-Representation in the Hungarian and Romanian Parliaments.
Europe-Asia Studies 57(7):969–94.
Chrypinski, Vincent C. 1990. “ e Catholic Church in 1944–1989 Poland.
Pp. 117–41 in Catholicism and Politics in Communist Societies, edited by
Pedro Ramet. Durham, NC: Duke University Press.
Cobb, Michael D. and Je ery A. Jenkins. 2001. “Race and the Representa-
tion of Blacks’ Interests during Reconstruction.” Political Research Quar-
terly 54(1):181–204.
Coleman, James. 1990. Foundations of Social  eory. Cambridge: Harvard
University Press.
Cook, Elizabeth A. and Clyde Wilcox. 1995. “Women Voters in the ‘Year of
the Woman’.” Pp. 195–219 in Democracy’s Feast: Elections in America, edited
by Herbert F. Weisberg. Chatham, NJ: Chatham House Publishers, Inc.
Cutler, Fred. 2002. “ e Simplest Shortcut of All: Sociodemographic Char-
acteristics and Electoral Choice.Journal of Politics 64(2):466–90.
Dahl, Robert A. 2006. On Political Equality. New Haven, London: Yale Uni-
versity Press.
Dahlerup, Drude and Lenita Freidenvall. 2005. “Quotas as a‘Fast Track’ to
Equal Representation for Women: Why Scandinavia Is No Longer the
Model.International Feminist Journal of Politics 7(1):26–48.
Dahlerup, Drude. 2002. “Using Quotas to Increase Womens Political Rep-
resentation.” Pp. in Women in Parliament: Beyond
Numbers, International IDEA. Stockholm, Sweden: International IDEA.
Joshua Kjerulf Dubrow
Dahlreup, Drude and Anja T. Nordlund. 2004. “Gender Quotas- AKey to
Equality: ACase Study of Iraq and Afghanistan.International IDEA .
Dolan, Kathleen. 2001. “Electoral Context, Issues, and Voting for Women in
the 1990’s.” Women & Politics 23(1/2):21–36.
Domanski, Henryk. 2007. “New Dimension of Social Strati cation in
Poland? Class Membership and Electoral Voting in 1991–2001.” European
Sociological Review.
Domho , G. William. 2000. Who Rules America Now? Power and Politics in
the Year 2000: May eld Publishing Company: May eld.
Dovi, Suzanne. 2002. “Preferable Descriptive Representatives: Will Just
Any Woman, Black, or Latino Do?” American Political Science Review
Dow, Jay K. and James W. Endersby. 2004. “Multinomial Probit and Mul-
tinomial Logit: AComparison of Choice Models for Voting Research.”
Electoral Studies 23(1):107–22.
Downs, Anthony. 1957. An Economic  eory of Democracy. New York: Harper.
Dubrow, Joshua K. 2005. “Descriptive Representation in the Balkans:
AComparative Analysis.” 7th Congress of the European Sociological As-
sociation, Torun, Poland, September 9–12.
Dubrow, Joshua Kjerulf. 2006. “Women’s Representation in the Romanian
Chamber of Deputies, 1992–2005.” International Journal of Sociology
Dubrow, Joshua Kjerulf. 2006. Enhancing Descriptive Representation in aNew
Democracy: APolitical Market Approach. PhD dissertation,  e Ohio State
Dubrow, Joshua Kjerulf. 2007. “Choosing among Discrete Choice Models
for Voting Behavior.ASK: Society, Research, Methods 16: 9–24.
Dubrow, Joshua Kjerulf. “Voting for Descriptive Representation: Demo-
graphic Cues across Multiple Elections in Poland, 1993–2001.” Forth-
coming in Continuity and Change in Social Life: Structural and Psycho-
logical Adjustment in Poland, edited by Kazimierz M. Słomczyński and
Sandra Marquart-Pyatt. IFiS: Warsaw.
Dubrow, Joshua Kjerulf. 2006. Enhancing Descriptive Representation in aNew
Democracy: A Political Market Approach. PhD Dissertation,  e Ohio
State University.
Dubrow, Joshua Kjerulf. 2007. “Voting for Descriptive Representation: De-
mographic Cues across Multiple Elections” pp. 271–286 in Continuity and
Change in Social Life: Structural and Psychological Adjustment in Poland, edited
by Kazimierz M. Słomczyński and Sandra Marquart-Pyatt. IFiS: Warsaw.
Dubrow, Joshua Kjerulf. 2015. “Political inequality is International, Interdis-
ciplinary and Intersectional.” Sociology Compass 9(6): 477–486.
The Electoral Market as a Mechanism of Political Inequality 65
Eckstrand, Laurie E. and William A. Eckert. 1981. “ e Impact of Candi-
date’s Sex on Voter Choice.” Western Political Quarterly 34(1):78–87.
Elster, John, Claus O e, and Ulrich K. Preuss. 1998. Institutional Design in
Post-Communist Societies. Cambridge: Cambridge University Press.
Evans, Geo . 1992. “Is Britain aClass-Divided Society? ARe-Analysis and
Extension of Marshall et al.’s Study of Class Consciousness.Sociology 26
(2): 233–258.
Fantasia, Rick. 2005. “From Class Consciousness to Culture, Action, and
Social Organization.Annual Review of Sociology 21: 269–287.
Fidrmuc, Jan. 2000. “Economics of Voting in Post-Communist Countries.”
Electoral Studies 19:199–217.
Fischer, Mary E. 1985. “Women in Romanian Politics: Elena Ceausescu, Pro-
natalism, and the Promotion of Women.” Pp. 121–37 in Women, State,
and Party in Eastern Europe, edited by Sharon L. Wolchik and Alfred G.
Meyer. Durham, NC: Duke University Press.
Foryś, Grzegorz. 2005. “Protesty Rolnikow wPolsce wLatach 1989–1993
I1997 – 2001: Proba Analizy Socjologicznej” pp. 281–297 in WObliczu
Zmiany: Wybrane Strategie Działania Mieszkanców Polskiej Wsi, edited by
Krzysztof Gorlach and Grzegorz Forys, Jagiellonia, Poland: University of
Jagiellonia Press.
Fox, Richard L. and Jennifer L. Lawless. 2004. “Entering the Arena? Gender
and the Decision to Run for O ce.” American Journal of Political Science.
48 (2): 264–280.
Gallego, Aina. 2008. “Unequal Political Participation in Europe.” Interna-
tional Journal of Sociology 37 (4): 10–25.
Gay, Claudine. 2002. “Spirals of Trust?  e E ect of Descriptive Represen-
tation on the Relationship Between Citizens and  eir Government.
American Journal of Political Science 46(4):717–33.
Gerteis, Joseph and Mike Savage. 1998. “ e Salience of Class in Britain and
America: AComparative Analysis.” e British Journal of Sociology, Vol.
49, No. 2. pp. 252–274.
Gidengil, Elisabeth. 1996. “Gender and Attitudes Toward Quotas for Women
Candidates in Canada .” Women & Politics 16(4):21–44.
Gilens, Martin and Benjamin I. Page. 2014. “Testing  eories of American
Politics: Elites, Interest Groups, and Average Citizens.” Perspectives on
Politics 12: 564–581.
Gill, Anthony. 2001. “Religion and Comparative Politics.” In Annual Review
of Political Science Vol. 4, ed. Nelson W. Polsby. Palo Alto: Annual
Glantz, Oscar. 1958. “Class Consciousness and Political Solidarity.” American
Sociological Review 23 (4): 375–383.
Joshua Kjerulf Dubrow
Gorlach, Krzysztof and Patrick H. Mooney. 1998. “Defending Class Inter-
ests: Polish Peasants in the First Years of Transformation.” Pp. 262–283
in eorising Transition: e Political Economy of Post-Communist Trans-
formations edited by John Pickles and Adrian Smith. London: Routledge.
Harper, Shannon and Barbara Reskin. 2005. “A rmative Action at School
and on the Job.Annual Review of Sociology 31(357–79).
Hechter, Michael and Satoshi Kanazawa. 1997. “Sociological rational choice
theory”. Annual review of sociology 23, 191–214.
Hero, Rodney E. and Caroline J. Tolbert. 1999. “Latinos and Substantive
Representation in the U.S. House of Representatives: Direct, Indirect, or
Nonexistent?” American Journal of Political Science 39(3):640–652.
Holland, Rob W., Bas Verplanken, and Ad V. Knippenberg. 2002. “On the
Nature of Attitude-Behavior Relations:  e Strong Guide, the Weak
Follow.” European Journal of Social Psychology 32(6):869–76.
Hout, Michael, Clem Brooks and Je Manza. 1995. “ e Democratic Class
Struggle in the United States, 1948–1992.” American Sociological Review,
Vol. 60, No. 6 pp. 805–828.
Howard, Judith A. 2000. “Social Psychology of Identities.” Annual Review of
Sociology 26:367–93.
Htun, Mala. 2003. Sex and the State: Abortion, Divorce, and the Family under
Latin American Dictatorships and Democracies. Cambridge: Cambridge
University Press.