Electoral Systems and the Representation of Ethnic Minorities: Evidence from Russia
Author(s): Robert G. Moser
Reviewed work(s):
Source: Comparative Politics, Vol. 40, No. 3 (Apr., 2008), pp. 273-292
Published by: Ph.D. Program in Political Science of the City University of New York
Stable URL: http://www.jstor.org/stable/20434082 .
Accessed: 08/05/2012 11:57
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Ph.D. Program in Political Science of the City University of New York is collaborating with JSTOR to digitize,
preserve and extend access to Comparative Politics.
http://www.jstor.org
Electoral Systemsand theRepresentationofEthnic
Minorities
Evidence from
Russia
Robert G.Moser
Scholars have long noted thatstateswith significant ethnic cleavages face difficult obsta
cles to themaintenance of democracy and that these problems are most severe in new
democracies.1 As a result, scholars have looked primarily to institutions as a means
to
helpethnically
dividedstatesovercometheseobstaclesandsuccessfully
implant
democ
racy in the face of deep ethnic cleavages.2
Achieving
broadrepresentation
ofdifferent
ethnic
groups
has important
forthe
implications
stability and quality of democracy, particularly inpolities just emerging out of long peri
odsofauthoritarian
rule.3
Legislativerepresentation
carriespowerful
symbolic
powerfor
ethnicminorities and often becomes an end in itself even when minorities have littleor
no chance of participating in the governing coalition.4
It is often taken as a matter of faith thatproportional representation (PR) increases
minorityrepresentation
while electoralsystems
with singlememberdistricts(SMD)
exclude minorities from legislative representation.Lijphart argues thatproportional rep
resentation, forexample, is not only better than singlemember districts but is superior to
all other electoral institutions(for example, the alternative vote, reserved seats forminor
itygroups,andmixed electoralsystems)inpromoting
minorityrepresentation.5
While
advocates of proportional representation acknowledge thepossibility thatminorities can
gain representation under certain conditions in other electoral systems, it is still viewed
as providing crucial advantages. As Lijphart explains, "the beauty of PR is that in addi
tiontoproducing
proportionality
andminorityrepresentation,
ittreats
all groups-eth
nic, racial, religious, or even noncommunal groups-in
a completely equal and
evenhanded fashion.Why deviate from fullPR at all?"6
Despite the scholarly consensus in favorof proportional representation, there is sur
prisingly
little
PR systems
empirical
evidencethat
providebetterrepresentation
ofethnic
minorities than SMD
systems.7This article offers a model of electoral system effects on
minority representation and a set of hypotheses concerning how PR and SMD electoral
ruleswill affectdifferent
categoriesof ethnicminoritiesundercertainconditions.
Patterns of minority representation under PR and SMD electoral arrangements are then
273
Comparative Politics
April 2008
examined inRussia using a unique dataset that identifies the ethnicity of individual leg
islators.
Russia employeda mixed electoralsystemduringtheperiodof investigation,
which allowed voters to cast two ballots, one for a party list in a PR contest and one for
an individual candidate in a singlemember district. This system affords the opportunity
representation
whilehold
of different
electoralrulesonminority
toexaminetheimpact
ing social conditions constant through the comparison of thePR and SMD
tiersofmixed
systems in a single country.8This study also has the advantage of examining minority
representation at the level of individual legislators rather than studying the electoral suc
cess of ethnic parties. Data on individual legislators are essential to captureminority rep
resentation
through
bothethnicand nonethnicparties.Finally,theRussian case is
importantbecause of thevirtual absence of ethnic parties in this country.The lack of eth
nic parties is a product of demography (most notably small ethnic groups) and actions by
theRussian state,which has passed legislation thatseverely restrictsthe formationof eth
nic or religious parties and campaigning on explicitly ethnic issues.9 Thus, this study
examines the impact of electoral rules in a contextwhere ethnic parties do not play a sig
nificantroleinelectoral
politics.
The findings presented here offer a challenge to the conventional wisdom
thatPR
systems generally are more conducive tominority representation than SMD systems.
Instead, it is shown that, first,the effectsof electoral systems are highly conditional on a
moreassimilated
factors,
ethnicgroups
and institutional
number
ofdemographic
second,
witness no significant differences in representation across electoral systems, and third,
each typeof system provides advantages over theother forcertain typesof ethnicminori
ties,with neither PR nor SMD elections being advantageous more often than the other.
Consequently, various demographic and political factors, including group size, geo
mitigatetheinflu
ethnicfederalism,
and culturalassimilation,
graphicconcentration,
ence of electoral systems on minority representation and thusmust be taken into account
thisrelationship.
whenexamining
Electoral
Systems and Minority Representation
inNational Legislatures
The ability of ethnicminorities towin legislative seats is usually seen as a consequence
of electoral systems and the size and geographic concentration ofminority populations.'0
The argument thatPR systems foster the representation of ethnicminorities has typical
ly rested on the idea that they encourage the emergence of ethnic parties. Proportional
representation encourages theiremergence by lowering the electoral thresholdnecessary
to gain representation, thereby increasing thenumber of parties and making smaller eth
nic parties more viable.11 This result can be seen particularly in ethnically divided soci
eties with PR systems that have very high district magnitudes and very low legal
274
RobertG.Moser
thresholds,
whichprovidefewobstaclestoelectionforeventhesmallest
parties.Perhaps
thebest example of how proportional representation can promote the election ofminori
tyethnicgroupsthrough
ethnicpartiesis Israel,whichhad sevenpartiesrepresenting
specific ethnic or religious constituencies out of a totalof twelve parties winning seats in
the2006parliamentary
electionstotheKnesset.12
Additionally,
thegreater
proportional
ityof PR systems provides incentives for all parties to field a more diverse set of candi
dates in order to capture ethnic voting blocs, since even small increases in a party's vote
share could mean more legislative seats.13
While generally viewed as inferiorto PR systems, singlemember districts also can
be conducive tominority representation under certain conditions. The election of ethnic
minorities in SMD systems tends to be based on geographic concentration and the abili
tyof a minority group to constitute a criticalmass within a given electoral district. In so
existfortheemergence
calledmajority-minority
ofethnic
districts,
opportunities
parties,
which can displace one of themajor parties within theirhome regions (for example, the
Bloc Quebe'cois inQuebec), while major parties feel pressure to nominate more minori
ty candidates, as seen in theUnited States."4 Thus, geographic concentration can over
come the bias against small (ethnic) parties and minority candidates under plurality
systems. 15
Thus, under the rightcircumstances, both PR and SMD systems allow forminority
betweenthemlies in theconditionsthatpromote
representation.
The distinction
mobi
lization around ethnic parties and/ornomination ofminority candidates bymajor parties.
is expectedtoprovideany ethnicminority,
Essentially,
proportionalrepresentation
whether it is geographically concentrated or not, the ability to gain representation
throughan ethnic party and/or ethnic balancing onmajor parties' lists,whereas plurality
systems are expected only to promote minority representation for ethnic groups that
reach a criticalmass within majority-minority electoral districts. From thisperspective,
giventhatproportional
representation
is expectedtopromote
minorityrepresentation
under a greater number of circumstances than singlemember districts, itspreferred sta
tus remains intact even when acknowledging the prospect of minority representation
underothersystems.
A Model
of Ethnic Representation
Clearly, studies of electoral system effectson ethnic representation tend to view the elec
tion of ethnic minorities through the prism of ethnic voting -the
tendency of ethnic
groups to vote for coethnic candidates or parties when given the opportunity.However,
to view minority representation as solely emanating fromethnicmobilization is too sim
plistic, formembers of ethnicminorities gain election through a variety of ways, often
with considerable support from themajority population. In contrast, a model of ethnic
275
Comparative Politics
April 2008
Figure 1 A Model ofMinority Representation
LOW
Level of Ethnic Mobilization
HIGH
~~~~~~~~~~~~~~~~~~~~~~~r
Assimilation:
EthnicMobilization:
Reliant upon minimal co-ethnic
Reliant upon maximal
PR: Manifest through election by
parties across the political spectrum.
PR: Manifest through election by
ethnic parties.
electoralsupport.
SMD:
SMDs
Manifest through election in
dominated by majority
electoralsupport.
SMD:
Manifest
co-ethnic
through election in
minority-majority districts.
population.
representation can be based on the notion of a continuum from ethnic mobilization
to
assimilation. The central causal element of this continuum is the relative level of support
from theircoethnic voting constituency thatminority candidates must rely upon to gain
election. Figure 1 illustrates themajor assumptions of themodel.
Ethnic mobilization represents themost common conception of how minority can
didates and ethnic parties gain election. In thismodel, the election of ethnicminorities is
contingent upon the electoral mobilization of coethnic voters."6 It is assumed that the
winning minority candidate or ethnic party gains all or a vast majority of its electoral
support from its targetethnic constituency.Under PR rules, this type ofmobilization is
manifest through ethnic parties. In SMD systems, ethnicmobilization ismarked by the
mobilization of coethnic support for a minority candidate. Such mobilization may be
channeled through an ethnic party (for example, the Bloc Quebecois
in Canada), a
nonethnic party (for example, theDemocratic Party in theUnited States), or no party at
all. Regardless of the nature of partisanship, the central dynamic of ethnicmobilization
remains the same: the candidate relies primarily upon coethnic voters forelectoral victo
ry.
Minority representationthroughethnicmobilization requires thatthewinming candidate
or party assemble an ethnic-based constituency capable of overcoming the electoral
threshold for representation.The conventional wisdom presumes that the relative thresh
old of PR systems tends to be much lower (depending on the districtmagnitude of the
system) than thatof plurality systems, inwhich thewinner must capture a plurality of the
276
Robert G. Moser
vote, typically approaching 50 percent. This lower relative threshold lies at the heart of
betweenseatsandvotesunderPR rules.However,plurality
proportionality
thegreater
systems in countries with noninstitutionalized party systems such as Russia tend to have
partysystems,
withmore institutionalized
thancountries
greaterpartyfragmentation
Moreover,
representation.17
withproportional
conditions
usuallyassociated
thuscreating
itmust be remembered that the absolute vote threshold forgaining a single seat in the
legislature is actually lower in SMD systems, since itusually takes fewervotes towin in
one district than to gain even a small percentage (for example, 5 percent) of thenational
vote. This caveat is important to keep inmind for very small ethnic groups with geo
graphically concentrated populations. Under a PR system, such groups may be unable to
field an ethnic party that could be viable and may be less attractive as candidates in a
within
awinningelectoralconstituency
strategy
butmay stillconstitute
ticket-balancing
a single or handful of singlemember electoral districts.
The assimilation model assumes thatmembers of an ethnicminority requiremini
mal or no support from coethnics in order to gain representation.This model represents
thediametrically opposite set of outcomes compared to those seen throughethnicmobi
ethnic
mobilization,theethnic
achievedthrough
lization
because,unlikerepresentation
background of the candidate from assimilated minorities becomes a virtual nonfactor,
allowing theminority candidate towin election with support from themajority popula
tion.Under PR rules thisoutcome is characterized by the election ofminority candidates
among parties across the political spectrum rather thanminority nominations concen
trated in ethnic,multiethnic, or nonethnic parties thatmake some sortof appeal to attract
assimilationis
through
minorityvoters.In SMD electionsminorityrepresentation
withhighconcentrations
candidatesinelectoraldistricts
markedbyelectionofminority
of voters from themajority ethnic group.
The twoperspectives of ethnicmobilization and assimilation are presented as a con
tinuum rather than a dichotomy because thereare scenarios under which minority candi
dates are elected that combine elements of minority and majority support. Parties may
have incentives to nominate less assimilated minority candidates but not rely solely or
even predominantly upon the support of thoseminorities forelection. Particularly inPR
systems, one can imagine that certain parties would nominate members of minority
groups in order to attract support fromparticular ethnic groups as part of a larger elec
toral constituency thatalso includes interestswithin themajority population.
Hypotheses
Based on thismodel, several hypotheses can be offered regarding the impact of electoral
systems on minority representation conditioned by the size, geographic concentration,
and relative assimilation ofminority groups. In general, differences in the level ofminor
277
Comparative Politics
April 2008
ityrepresentation
betweenPR andSMD tiersaredrivenbydemographic
characteristics
thatprovide advantages for the group in one type of electoral systemwhile at the same
timeproducing
disadvantages
undertheothersystem.
Conversely,
similarlevelsof rep
resentation
betweenthetwotiersareproducedeither
whendemographic
characteristics
produce the same dynamic in both tiers (for example, assimilation) or when a combina
tion of characteristics produce advantages
in both tiers (for example, large and geo
graphically
concentrated
groups).
AssimilationHypothesis Highly assimilatedethnicminoritieswill experienceno differ
between SMD and PR tiers,regardlessof
ence in the levelof legislativerepresentation
theirsize or geographicconcentration.
onminorityrepresenta
Assimilationtendstodiminishtheimpactof electoralsystems
tionby lesseningtheperceiveddifferences
betweentheassimilated
minorityand the
majority population. Highly assimilated ethnicminorities will be able to gamer support
from themajority population to gain election, given that theirethnic identities are a less
salient issue (and often a less perceptible voting cue) than the ethnicity of members of
likemembersof the
mem
lessassimilated
groups.Consequently,
majorityethnicgroup,
bers ofmore assimilated minorities should be attractive candidates forPR party lists and
SMD contests. There are other crucial distinctions in the patterns of representation of
highly assimilated minority groups. In the SMD tier,members of these ethnic categories
are expected to gain election in single member districts dominated by themajority pop
ulation and to be nominated by major political parties (as opposed to running as inde
pendents or members of minor parties). In the PR tier, highly assimilated minority
candidates also are expected to have distinct partisan backgrounds, being nominated by
more conservative and sometimes even nationalistic parties that tend not to nominate
other
minority
groups.
Small, ConcentratedGroup Hypothesis Groups thatare small,geographicallyconcen
trated,and less assimilatedwill achieve higher levelsof legislativerepresentationin the
SMD tierthan in thePR tierofmixed-memberelectoral systems.
they tend to be more displaced from themajority culture, less assimilated
groups are likely to rely on ethnic mobilization to elect representatives to the national
legislature. Small groups face disadvantages in PR elections because they are not large
Because
enough to sustain ethnic parties or to be nominated in large numbers on theparty lists of
major, nonethnic parties. However, if these small minorities are geographically concen
278
Robert G.Moser
togainelectionbecause they
memberdistricts
providethebestopportunity
trated,
single
aremost likely to constitute a majority in at least one singlemember electoral district.
Large, Dispersed Group Hypothesis Groups thatare large,geographicallydispersed,
and less assimilatedwill achieve higher levelsof legislativerepresentationin thePR tier
thanin theSMD tierofmixed-memberelectoral systems.
Large, geographically dispersed ethnic groups thatare less assimilated will gain greater
representation in PR as opposed to SMD systems due to two countervailing dynamics.
One provides an advantage inproportional representation,while theother presents a dis
advantage within singlemember districts.First, largerethnic groups can marshal the sup
port of a significant voting bloc on a national scale. This ability makes such groups
potent electoral forces in PR systems because they can sustain a viable ethnic party
and/or command a significantminority voting bloc thatmake them attractive candidates
parties.
for
multi-andnonethnic
Large, ConcentratedGroupHypothesis Groups thatare relativelylarge,geographically
and less assimilatedwill experienceno differencein the levelof legislative
concentrated,
representation
betweenSMD and PR tiers.
Unlike assimilated minorities, which downplay the impact of electoral systems, certain
ethnic groups have similar rates of PR and SMD representation because they combine
demographic characteristics that provide advantages in each tier.Large, unassimilated
ethnicminorities that are geographically concentrated enjoy two advantages over other
minorities that enable them to achieve relatively high rates of representation under both
PR and SMD electoral arrangements. In thePR tier, larger ethnic groups have thepoten
tial tomobilize larger voting blocs and thus should be better able to field viable ethnic
parties or be more attractivecandidates formainstream political parties hoping to capture
significant voting constituencies. In the SMD tier, these same large ethnic minorities
should also possess an electoral advantage as long as theyare geographically concentrat
ed because such groups should have at theirdisposal a cohesive voting bloc capable of
overcoming the electoral threshold in a (relatively) large number of single member dis
tricts.By constituting a critical mass of the electorate nationally and within numerous
singlemember districts, large, geographically concentrated ethnic groups should be able
to find electoral success at roughly equivalent levels inboth PR and SMD tiers.
Of course, these hypotheses are not particularly surprising.They match the general
propositions of studies of electoral systems and minority representation, but with two
thepotentialforethnic
caveats.First,current
does nothighlight
scholarship
important
in levelsof representation
underdifferent
typesof
minoritiestoexperience
no difference
electoral systems. Second, the conventional wisdom allows forSMD systems toproduce
279
Comparative Politics
April 2008
levels of representation thatwould equal those expected under a PR systemwhen ethnic
groups are concentrated but does not argue that there are situations when SMD systems
would actuallyproducelevelsofminorityrepresentation
that
would exceed thosethat
would be produced in a PR system.
Data
and Variables
Ethnicity is a socially constructed source of identitybased on racial, religious, linguistic,
or regional
While oftenbasedon seemingly
cultural,
properties
backgrounds.
primordial
such as skin color or language, most recent scholarship has emphasized themalleability
of ethnic identityand itsmanipulability as a political force, particularly by the political
elite.18These aspects of ethnic identityhave made ita difficult concept to examine sys
tematically. Clearly, the phenomenon of ethnicmobilization is an important element in
much lessmeasuring,itsinfluence
democratic
politics,butcapturing,
remains
problem
atic.
Scholarship on legislative representation of ethnicminorities faces even more fun
damental data problems.While there are extensive and reliable databases on the number
of women in legislatures around theworld, no analogous source provides information
regarding the ethnic identityof legislators. Thus, much isknown about how electoral sys
tems,culturalattitudes,
and socioeconomicfactorsinfluencetheelectionofwomen
around theworld, but very little is known about how these factors affect the election of
ethnicminorities.19 This situation is hardly surprising given the sensitivityof such infor
mation inmany contexts.
As a resultof this lack of information, theoretical propositions regarding the factors
that influenceminority representation tend not to be directly tested using empirical data
or are examined through the use of questionable proxymeasures such as the proportion
of women elected to the legislature or the electoral success of ethnic parties.20 Such indi
rectmeasures of the ethnic identityof legislators may fail to capture the full extent of
minority representation,particularly in countries thatdo not have strong ethnic parties.
For the case examined here, informationon the ethnic identityof individual legisla
torswas available. The election ofminority candidates, rather than the success of partic
ular ethnic parties, was used as the dependent variable. The data come from official
handbooks published by theRussian State Duma, presumably from information supplied
by the deputies themselves or their staffs.
EthnicMinorities inRussia
Russia is unique among postcommunist states in termsof its ethnic composition. Ithas a
20 percent-similar in size tomany countries
significantminority population-around
280
RobertG.Moser
Concentration
of
andGeographic
Table 1EthnicFederalism,
RelativeAssimilation
MinoritiesinRussia
SelectEthnic
% withRussian Largest% of the
Populationin Ethnic
Thousands Homeland as their
First
Population
of a
Language
SingleRegion
57
17
2943
No
Ukrainians
815
No
63
8
Belorussians
58
6
597
No
Germans
2
29
Georgians
199
No
148
No
63
5
Koreans
1130
No
32
4
Armenians
5593
Yes
14
48
Tatars
22
1674
Yes
10
Bashkirs
22
Yes
68
1637
Chuvash
6
516
Yes
53
Ossetians
445
Yes
13
36
Buryats
6
33
444
Yes
Yakuts
293
Yes
26
23
Komi
233
Yes
91
Jews
4
Ethnic
Minority
of the region, such as Bulgaria, Romania, and Slovakia. However, unlike these states
Russia lacks a single major minority group. Instead, thenon-Russian population is com
posed of literallydozens of very small groups. The largestminority group of theRussian
Federation, Tatars, makes up less than four percent of the population, and only three
groups (Tatars, Ukrainians, and Chuvash) comprise over one percent of the country's
population.
The composition
ofRussia's ethnicpopulationcan notbe properlyunderstood
without acknowledgment of its institutionalizationwithin a system of ethnic federalism.
The Russian Federation has continued the Soviet practice of defining some of its federal
units along ethnic lines. Russia is composed of eighty-nine subjects or regions that are
divided into twenty-one republics, fortynine oblasts, six krais, two cities (Moscow and
St. Petersburg), ten autonomous okrugs, and one autonomous oblast (the Jewish
AutonomousOblast). Republics,autonomousokrugs,and theJewish
Autonomous
Oblast make up the thirtytwo regions named after a non-Russian ethnic group.
As Harris has demonstrated, Russia's ethnic federalism has had a strong influence
on the ethnic identityof non-Russians. Groups provided with "ethnic homelands" man
aged to retain theirnative languages and resist assimilation efforts to a much greater
extent than (often larger) minority groups that lacked their own designated federal
in ethnichomelandshave cultivated
ethnic
Moreover,regionalgovernments
region.22
in
The
of ethnicfederal
units
mobilization center-periphery
relations. verydesignation
has produced incentives and resources for thepromotion of an ethnic cadre of elites and
mass mobilization tied to ethnicity.23Thus, ethnic federalism has indirectly affected
281
Comparative Politics
April 2008
minority
representation
through
itsinfluence
overethnicidentity
andbehavioritself,
cre
ating, inmost cases,
less assimilated
and more geographically
concentrated ethnic
groupsthathavecertainpoliticalopportunities
and resourcesthatothergroupswithout
their
ownethnichomelandslack.
The confluence
of certainlevelsof cultural
assimilation
andgeographic
concentra
tion
withfederal
ornonfederal
status
withinRussia's ethnicfederalsystem
hasproduced
twodiscernible
categories
of ethnic
minorities
withinRussia.The first
category,
nonfed
eral minorities, includes groups thatdo not have their own ethnic homeland within the
Russian federalsystem.
Consequently,
thesegroupstendtobemoregeographically
dis
persed and more culturally assimilated than those groups with an ethnic homeland. This
category is quite ethnically diverse with groups with a Slavic background (Ukrainians
and Belorussians),
as well as groups with origins fromEurope (Germans, Greeks, and
Finns),Eurasia (Latvians,
Georgians,andArmenians),andAsia (Koreans).
The second category, federalminorities, ismade up of groups thathave theirown fed
eral subunit. These groups tend to be more geographically
concentrated than other
minorities -on average, 66 percent of titularnationalities live in theirdesignated feder
al units-and
are relatively unassimilated
into themajority Russian
culture.24Table
1
shows the federal designation, relative assimilation (measured as percentage claiming
Russian as their first language), and level of geographic concentration of selectminority
groupsinRussia.25
By and large, the figures inTable 1 conform to the characterization thatethnic feder
alism has produced twobroad categories ofminorities-one
relativelyunassimilated and
geographically concentrated and the other relatively assimilated and geographically dis
persed. On average, 26 percent of federalminorities speak Russian as their first lan
guage, and they comprise 36 percent of the population in one of Russia's federal units,
while 50 percent of nonfederal minorities are native Russian speakers and they tend to
make up only 7 percent of thepopulation in any particular Russian region. Jews present
the one exception to the pattern that federally designated ethnic groups are geographi
cally concentrated and less assimilated thanminorities without an ethnic homeland.
While
Jews inRussia have theirown ethnic homeland, they are not geographically con
centrated in thathomeland. Less
than 2 percent of Jews living inRussia
resided in the
JewishAutonomous oblast in 1989, and the Jewish population comprised only 4 percent
of the region's population. Jews are also arguably themost highly assimilated ethnic
minority group inRussia, with over 90 percent speaking Russian as their first language.
MinorityRepresentationinRussia
Given the ethnicmake-up of Russia, one might expect both thePR and plurality tiersof
itsmixed electoral system to underrepresent non-Russian ethnic groups, based on the
assumption thatminority representation is primarily driven by themobilization of coeth
282
Robert G.Moser
Table 2Modes ofRepresentation of Specific Ethnic Groups, 1993-1999
Ethnic Group
|
Elected
HElected
in the PR tier
in SMDs
Russian Regions
HElected in SMDs
Ethnic Homelands
in
in
1993 Election
Non-Federal Minorities
Ukrainian
13
8
1
Belorussian
5
2
0
Korean
0
1
0
Arnenian
2
1
O
Other
2
4
Total
22
16
0
1
Federal Minorities
Tatar
Bashkir
2
_I2
1
0
3
0
Chuvash
0
0
Jewish
5
2
0
Othera
5
1
18
Total
13
3
23
_________________
UTkrairnian
Belorussian
Korean
Armenian
Other
Total
137
Election
Non-Federal
Minorities
~~~~1995
3
0
2
2
____________
1
0
1
1
1
1
18
1
13
0
7
Federal Minorities
1
3
Bashkir
Chuvash
0
1
0
0
2
0
Jewish
2
1
0
Othera
9
1
14
Tatar
Total
5
17
3
__
Ukrainian
Belorussian
Korean
Armenian
Other
Total
12
7
0
1999 Election
Non-F& ra)Minzaities
12
2
3
0
1
0
3
0
2
4
24
19
19
1
0
4
Federal Minorities
Tatar
Bashkir
3
3
1
0
5
3
Chuvash
0
0
3
Jewish
4
Othersa
Total
11
21
1
2
4
1
10
22
283
Comparative Politics
April 2008
nic voters. No single ethnicminority constitutes a large enough share of the population
plausibly to field an ethnic party capable of overcoming the five percent legal threshold
in thePR tier,and only the largestminority groups such as Tatars or Ukrainians offer a
potentially cohesive bloc of voters that is large enough to compel a nonethnic party to
includemembers of these groups on theirPR lists to capture the "ethnic" vote. One could
imagine
minoritiesfaring
betterinplurality
elections,giventhegeographicconcentra
minorities
withintheir
tionof certainfederal
designated
homelands.
However,there
were
only twenty-fourmajority-minority districts out of a total of 225 in 1995, and of these
only fifteen had a Russian population under 40 percent.26Despite
these demographic
obstacles, both the PR and plurality tiers of Russia's mixed system elected a relatively
large percentage of non-Russians to the State Duma. Neither the PR nor the SMD
tier
themajorityethnicgroup,despitetheadvantages
Russianspossessed
overrepresented
under both systems. Russians were elected to the legislature in almost identical numbers
in both tiers and in nearly equal proportion to their population. However,
therewere
in theoverallrepresentation
differences
of federaland nonfederalethnic
important
minorities and the impact of electoral rules on these two categories. Table 2 provides a
of individual
detailedbreakdown
threeavenues:
minority
grouprepresentation
through
election in thePR tier,election in single member districts inRussian regions, and elec
tioninsingle
inethnichomelands.27
memberdistricts
EthnicMobilization,andModes ofMinorityRepresentation
CulturalAssimilation,
inRussia
in thepatterns
Severaldifferences
of representation
fornonfederal
and
federalminorities deserve special attention because theyhighlight the impact that cul
tural assimilation and ethnicmobilization have had onminority representation inRussia.
First,despiteseveralgeographic
disadvantages,
nonfederal
minorities
werebetterrepre
sented in the State Duma than their federal counterparts.While nonfederal minorities
made up only 32 percent of theminority population inRussia, they constituted 56 per
cent of the total number of minorities elected to the State Duma from 1993 to
1999.28Moreover, thisoverrepresentationwas fueled primarily by the success of themost
assimilated groups within this category,Ukrainians and Belorussians, who share a com
mon ethnic, religious, and historical background with Russians. These two groups made
up 39 percent of the total number of minorities elected to the State Duma from 1993 to
1999, even though they comprised only 13 percent of the non-Russian population. The
non-Slavicgroupswithoutethnichomelands (forexample,Armenians,Koreans,
ethnicand religiousdifferences
from
Germans),whichpossessedsignificant
Russians,
did not gain the same degree of representation.This subset of thenonfederal category of
minority groups gained roughly the same percentage ofminority representatives (17 per
cent) as their share of theminority population (19 percent). This difference is likely due
to the fact that candidates with Slavic backgrounds were less distinguishable by name
and appearance fromRussian candidates than non-Slavic candidates. Essentially, some
groups were more easily assimilated and thusprobably more attractive toRussian voters.
284
RobertG.Moser
minorities,
which enjoyedmajority status
within crtain electoraldistricts
By conbtast,federal
Even though
theycomprised
underrepresented.
homelands,
were relatively
intheir
ethnic
67 percent of theminority population, federalminorities made up only 44 percent of the
minorities elected to the State Duma. The implication is clear: the impact of cultural
assimilation on minority representation inRussia not only affected the avenue by which
minority candidates were elected but also had an impact on the overall success of candi
dates fromcertainethnicgroups.Members ofmore assimilated,nonfederalethnic
groups appear tohave done better at thepolls than fellowminority candidates frommore
culturally distinct groups, even though the latterpossessed greater resources forelectoral
voters.
mobilization
of coethnic
Second, the data inTable 2 strongly suggest that the key to nonfederal minorities'
Russianpopulation.
Unlike lessassimi
majority
electoralsuccesswas supportfromthe
dom
minorities
gainedelectioninpoliticalcontexts
minorities,
nonfederal
latedfederal
inated by Russians. As Table 2 vividly shows, therewere dramatic differences in the types
of single member districts that elected nonfederal and federal minorities. The more
assimilated nonfederal minorities won the vast majority of their seats in regions where
Russians were in themajority, whereas the less assimilated federalminorities won virtu
ally all of theirSMD
seats in non-Russian ethnic homelands.
Jews pose the one exception to the two patterns discussed above because, although
they have their own ethnic homeland, Jewish candidates gained representation in con
textsassociated with the election of nonfederal groups. As shown inTable 2, unlike other
federalminorities Jews tended tobe overrepresented in theState Duma (given theirsmall
numbers), and Jewish candidates gained almost all of their seats in thePR tierand SMD
contests inRussian regions. However, this exception actually bolsters the argument.As
discussed above, the demographic and cultural profile of Russian Jews ismuch more
similar to nonfederal minorities than to federal ones. Jews tend to be geographically dis
persed and highly urbanized and are perhaps themost linguistically assimilated minority
in Russia
(see Table 1). Thus, in gaining election through venues dominated by the
of
generalpatterns
theexperience
of thisgrouphas simplyfollowed
majority
population,
minority representation inRussia. This observation also suggests that the impact of eth
nic federalism on minority representation is primarily indirect, through its impact on cul
tural assimilation and geographic concentration. If the existence of an ethnic homeland
does not create insulation from cultural assimilation and provide pressure toward geo
graphic concentration, then theminority group with an ethnic homeland will not have
electoral experiences typically found among other federal ethnic groups.
Partisanship,Social Background,andMinorityRepresentation
Finally,one
must consider other potential factors thatmay interactwith or supersede ethnicity as a
mobilizational
resource forcandidates. The electoral success ofmembers of certain eth
nic groups may be rooted in the relative status such groups occupy
in a country's
social, political, and economic hierarchy. Perhaps certain minority groups achieved
285
Comparative
Politics April2008
Table
3 Social and Occupational Background ofNon-Russian Duma Deputies
Ethnic
Election Residence Residence Residence National Regional Econ. Other#
year/tier inMoscow inRussian inEthnic Elite#
Elite#
Elite
(%/0)
Group
Homeland (%)
Category
(%)
(%/)
#(%/0)
Region
_
Non-
1993
_
_ # (o/o)
___
2 (13)
13 (81)
# (_
1 (6)
0 (0)
3 (19)
6 (38)
7 (44)
Federal
SMD
l____
NonFederal
1993
PR
16 (73)
5 (23)
1 (5)
3 (14)
2 (9)
7 (32)
10 (46)
NonFederal
NonFederal
1995
SMD
1995
PR
2 (10)
13 (65)
5 (25)
10 (50)
6 (30)
3 (15)
1 (5)
3 (17)
14 (78)
1 (6)
5 (28)
5 (28)
2 (11)
6 (33)
1999
10 (44)
13 (57)
0 (0)
12 (52)
6 (26)
3 (13)
2 (9)
1999
9 (38)
14 (58)
1 (4)
9 (38)
5 (21)
4 (17)
6 (25)
42 (34)
72 (58)
9 (7)
39 (31)
27 (22)
25 (20)
32 (26)
1993
5 (19)
3 (12)
18 (69)
1 (4)
10 (39)
7 (27)
8 (31)
Federal
1993
_
_ _ PR
6 (46)
3 (23)
4 (31)
1 (8)
2 (15)
1 (8)
9 (69)
1995
8 (36)
3 (14)
11 (50)
13 (59)
4 (18)
5 (23)
0 (0)
Federal
1995
_
_ _ PR
4 (24)
2 (12)
11 (65)
2 (12)
2 (12)
5 (29)
8 (47)
1999
6 (23)
2 (8)
18 (69)
6 (23)
14 (54)
2 (8)
4 (15)
1999
Federal
_ _
_ _ PR
6 (29)
4 (19)
11 (52)
3 (14)
7 (33)
6 (29)
5 (24)
35 (28)
17 (14)
73 (58)
26 (21)
39 (31)
26 (21)
34 (27)
Non-
Federal
SMD
NonFederal
PR
NonFederal
total
Federal
____SMD
_
Federal
_
Federal
____SMD
Federal
total
SMD
IIIIII
II
greater or lesser success overall or under particular electoral rules because candidates
from these groups possessed other influential social characteristics, such as partisan
affiliation, higher education, and positions of power in prestigious national or regional
membersof nonfederal
minoritieswho gained election
organizations.In particular,
throughgreater support fromRussian voters may have had electoral advantages as can
didates thatwere unrelated (or at least indirectly related) to their ethnic identities.As
hierarchically
within
Horowitzandothershave shown,ethnicgroupsareoftensituated
the social structureof a society, and thus ethnic cleavages are cumulative rather than
competing sources of identityalong with other social divisions based on economic pros
perity, education, and occupational specialization. In countries with ranked ethnic
groups, certainminorities may have advantages over other groups (including themajori
ty),while other groups suffer from chronic economic and social disadvantages.29 Such
disparities in economic and social resources and status can be expected to have a direct
impact on the ability of members of ethnicminorities to gain legislative representation.
Do members of different ethnic groups who have gained election to theRussian
286
RobertG.Moser
nationallegislature
displaydistinct
patterns
ofpartisanship
and socialbackgrounds
that
would suggest that some groups have social advantages over others?Analyses of thepar
tisanship and social background ofminority deputies elected in thePR and SMD
tiersof
Russia's mixed electoral system suggest that these other factors can not account for the
differences
observedinthe
modesof representation
forfederal
andnonfederal
minorities
inRussia.
Deputies
fromnonfederal and federalminorities did have distinct patterns of parti
sanaffiliation.30
However,thesedistinctions
arebetterconceivedof as effects
rather
than
intervening
causesof theway thatethnicity
has intersected
with electoralpoliticsin
Russia. First, in the SMD
tier federalminority deputies were much more likely to be
independents
thannonfederal
minority
deputies.
Over thethree
electionsexamined
here,
there
were forty-one
independents
amongdeputiesbelongingtofederal
minority
groups
butonlynineteenindependents
Itcould
amongthedeputiesfrom
minorities.
nonfederal
be argued that this increased partisanship may have played a role in the ability of non
federal ethnic minorities such as Ukrainians
to gain election in Russian-majority dis
tricts.
However,it ismore likelythatincreased
partisanship
amongmore assimilated,
nonfederal ethnic groups was an effectof the inherentdifferences in the electoral context
inwhich SMD candidates from federal and nonfederal ethnic groups found themselves.
Since federal
memberdistricts
minority
deputieselectedfromsingle
tendedtogainelec
tion inmajority-minority districts, ethnicitymay have replaced party as a central voting
cue.Conversely,
who gainedelectionalmostexclusively
nonfederal
candidates
minority
inRussian-majority
tomobilize support
districts
couldnotuse theirethnicity
and thus
neededotherresourcesthatpoliticalpartiescouldprovide.
Moreover,intheabsenceof
significant ethnic parties, all major parties tended to have a constituency dominated by
Russians. It stands to reason that,when nominating minority candidates, such parties
would be more likely to nominate members of ethnicminorities deemed closer cultural
ly to themajority population. Indeed, thepower of ethnicityas amobilizing force ismade
manifest here because federalminorities managed to gain election in significant numbers
despite the fact thatmany ranwithout party backing.
Other differences in partisanship between nonfederal and federal minority
groups can also be derived from the fact that the formerare more assimilated than the
latter.
Nationalistpartieselectedovertwiceasmanyassimilated,
nonfederal
minorities
than
membersofunassimilated,
federal
groupsforobviousreasons.Conversely,
parties
with a regionalist platform elected more than twice as many federalminorities as non
federalminorities because theirplatform brought greater electoral success in areas (eth
nichomelands)thatproducedthesedeputies.
Similar conclusions arise from an analysis of the social background of minority
deputies offered inTable 3. The most striking feature of the social background of non
federal and federalminority deputies is thedegree of similaritybetween the two groups.
Both categories had similar levels ofMoscow-based politicians, as well as members of
national, regional, and economic elites. Both categories had a similar proportion of
287
Comparative Politics
April 2008
deputies coming from the "other" group, which included members of professions and
occupations such as artists, academics, and low-level bureaucrats that tended to be less
prevalent in thenational legislature. The key distinction lay in residence inRussian ver
sus non-Russian regions.Members of nonfederal and federal ethnic groups were virtual
mirror images of one another in termsof patterns of regional (outside ofMoscow) resi
dency patterns.Thus, minority deputies tended to be local rather thannational politicians
constituents.
who residedamongtheir
Nonfederal
minority
deputies'constituents
simply
livedinRussian-majority
while federal
deputies'constituents
livedin
regions,
minority
ethnichomelands.
In short, analyses of levels of partisanship and social background suggest thatother
factors related to partisanship and social status provide much less explanatory power for
thedistinct
of electionofdifferent
groupsthantheinfluence
of ethnic
patterns
minority
ity itself and, in particular, the impact of cultural assimilation for one subset of ethnic
minority groups and geographic concentration for the other.
SecondaryElectoralRules, ProportionalRepresentation,and FederalMinorities
federalminorities tended to gain more representation in SMD elections, the dif
ference in the level of representation between the two tiersdropped significantly after the
initialpostcommunist election in 1993. This change over timewas driven by the fact that
While
the number of PR deputies among federalminorities doubled after the initial election in
1993 (see Table 2).
forgeographically
minoritiesinthePR
This increase
of representation
concentrated
tier ismost likely due to an institutional factor: a change in the electoral law instituted
after the initial postcommunist election. Beginning with the 1995 election, parties were
required to draw up a national list of twelve (later eighteen) candidates, followed by
regional sublists established by parties themselves.31 If a partywon enough seats to go
beyond the national list, the remaining seats were distributed according to the relative
proportion of votes won in each region outlined by the party.Regions with the greatest
number of votes for a given party were the first to receive parliamentary seats. This
change may have provided more opportunities for candidates from geographically con
centrated ethnic groups with their own federal units to obtain a winning position on
major party lists if support for such parties was more concentrated in non-Russian
regions than other parts of the country.
Conclusions
The contours of minority representation inRussia have several important implications
for the study of the effects of electoral systems on ethnic representation. First, contrary
to conventional wisdom, proportional representation did not display any greater propen
sity to elect minorities to the national legislature than plurality elections. In every elec
288
RobertG.Moser
tiontheSMD tierproduced
moreminority
representatives
thanthePR tier.
Second, theRussian experience suggests that there is a complex interactionamong
assimilation,
geographicconcentration,
and institutions.
Minoritygroupsappearedto
gain representation at varying levels under PR and SMD
rules and in differentdemo
graphic contexts depending largely on how well assimilated theywere with themajority
Russianpopulation.
Thus,more assimilated,
minorities
geographically
dispersed
gained
substantial representation (well beyond what could be expected based on their share of
the population) in both the PR and SMD tiers of Russia's mixed system. Given the
absence of a critical mass of coethnics, one must assume this electoral success was
attained through support from theRussian majority. The reason thatproportional repre
sentation did not favor the election of these groups was theirability to gain election in a
ofRussian-majority
substantial
number
memberdistricts.
Whilemanygeograph
single
icallydispersedgroups(Jews,
Armenians,
Germans),followedthispattern,
thestriking
success of themost assimilated Slavic minorities suggests thatassimilation is a key fac
tor thatneeds to be taken into consideration when ascertaining the causes of minority
representation.
lessassimilated,
concentrated
minorities
geographically
appearedto
gainelection
Conversely,
much more through themobilization of coethnics. Thus, these groups gained most of
intheSMD tier,
representation
indistricts
their
predominantly
withinfederally
designat
ed ethnic homelands. The reason that the electoral advantages within the SMD
tier
declined after 1993 was a change in the electoral law.When Russia changed itsPR sys
tem to include regional sublists in each party list, geographically concentrated ethnic
minorities benefited because
they represented regional voting blocs and had their own
federal
entities.
Finally, electoral laws are not the only institutional instrumentsaffecting the elec
toralmobilization of minorities. In theRussian case, ethnic federalism was themost
important
institutional
factorinfluencing
ethnicrepresentation.
The establishment
of
federal subunits designated as ethnic homelands for certain minorities substantially
influenced all other factors thatplayed a role inminority representation. Those groups
with theirown federal region tended tobe less assimilated and more geographically con
centrated
thantheir
nonfederalized
counterparts.
These findings suggest the need for caution when considering the impact of
electoral systems in ethnically diverse societies. Proportional representationmay not
provide ethnic minorities with the avenue to legislative representation its proponents
claim, especially in countries thatdo not have strong ethnically based parties.While the
introduction
of proportional
may increasethelikelihoodthatethnicpar
representation
tieswill emerge, itprovides no guarantee. Other electoral rules governing party forma
tion and registrationwill affect this outcome, as will the existence or nonexistence of
ethnic groups thatare large and cohesive enough to sustain such parties. 32 InRussia, the
absence of ethnic parties can be traced to both demographic factors (small minority
groups) and actions by the state that legally curtailed party elites from forming ethnic
289
Comparative
Politics April2008
parties and campaigning on ethnic themes. In a more permissible legal environment, eth
nic parties (for example, aMuslim
party) may attain greater influence and thus change
betweenelectoralrulesandparticular
ethnic
minorities.33
theinterrelationship
Finally, electoral systems do not affect all minorities in the same way. More assimi
fromthe
majoritypopulation
lated
minoritiesthatcanattract
substantial
electoralsupport
mechanismssuchasmajority-minority
single
member
maynotneedspecialinstitutional
districts to attain legislative representation,while such institutionaldevices may be the
minoritiestoachieverepresentation.
onlyway forlessassimilated
NOTES
1. See, for example, Alvin Rabushka and Kenneth A. Shepsle, Politics in Plural Societies: A Theory of
Democratic
1972); Juan Linz and Alfred Stepan, Problems of
Instability (Columbus: Charles E. Merrill,
Southern Europe, South America and Eastern Europe (Baltimore:
Democratic Transition and Consolidation:
and
The Johns Hopkins University Press, 1996); Jack Snyder, From Voting to Violence: Democratization
Conflict (New York: Norton, 2000).
in Plural Societies (New Haven: Yale University Press,
2. See, for example, Arend Lijphart, Democracy
1977); Donald Horowitz, Ethnic Groups in Conflict (Berkeley: University of California Press, 1985); Donald
in a Divided Society (Berkeley: University
Horowitz, A Democratic South Africa? Constitutional Engineering
of California Press, 1991); Ben Reilly and Andrew Reynolds, Electoral Systems and Conflict inDivided
Nationalist
National Academy Press, 1999).
(Washington, DC:
3. Arend Lijphart, "Constitutional Design forDivided Societies," Journal ofDemocracy,
Societies
15 (April 2004),
97.
4. Andrew Reynolds, "The Case for Proportionality,' 'Journal ofDemocracy,
5. Lijphart, "Constitutional Design forDivided Societies," p. 100.
6. Ibid., pp. 99-100.
6 (October
1995), 86-99.
7. On the dearth of empirical studies of electoral system effects on ethnic representation, see Pippa Norris,
Not Bullets:
Theories
of Ethnic Conflict, Electoral
Testing Consociational
Systems, and
inAndrew Reynolds, ed., The Architecture ofDemocracy:
Constitutional Design, Conflict
Democratization,"
"Ballots
and Democracy
(Oxford: Oxford University Press, 2002), pp. 211-13.
Management,
in
8. Robert G. Moser, Unexpected Outcomes: Electoral Systems, Political Parties, and Representation
Russia (Pittsburgh: University of Pittsburgh Press, 2001). For a critique of this approach, see Erik S. Herron
and Misa Nishikawa,
"Contamination Effects and the Number of Parties inMixed Superposition Electoral
Systems," Electoral Studies, 20 (January 2001), 63-86.
9. The 2001 federal law On Political Parties explicitly prohibits the formation of political parties on a racial,
national, or religious basis (Art. 9, 3) and prohibits parties from using names and symbols thatmay offend
racial, national, or religious feelings (Art. 6, 5 and Art. 7, 3). Moreover, the law requires thatpolitical parties
have regional branches in over half of Russia's federal regions (Art. 3, 2), which undermines parties with con
centrated geographic support. The 2002 federal law On theElection ofDeputies of the State Duma prohibits
electoral campaigning and the production of any type of electoral propaganda by charity and religious organi
zations and organizations founded by them (Art. 57, 5). This same law prohibits electoral campaigning that
violate the integrity of the Russian Federation or excite racial, national, or religious hatred or enmity
texts of these laws are available at www.democracy.ru.
(Art. 64, 1). Full English-language
would
10. Reilly and Reynolds, pp. 48-51.
11. Pippa Norris, Electoral Engineering:
University Press, 2004), 208-13.
290
Voting Rules
and Political
Behavior
(Cambridge:
Cambridge
RobertG.Moser
12.Determination of the Israeli parties considered to be ethnic or religious parties was based on descriptions
of parties found inEurope World online, www.europaworld.com.
13.Andrew Reynolds, "The Case for Proportionality," p. 91.
14. See, for example, David Lublin, The Paradox of Representation: Racial Gerrymandering and Minority
Interests inCongress (Princeton: Princeton University Press, 1997); David T. Canon, Race, Redistricting, and
The Unintended Consequences
of Black Majority Districts (Chicago: University of Chicago
Representation:
Press, 1999).
15. Joel D. Barkan, "Elections
inAgrarian Societies," Journal ofDemocracy, 6 (October 1995), 106-16.
Chandra, Why Ethnic Parties Succeed
(Cambridge: Cambridge University Press, 2004);
Norris, Electoral Engineering, ch. 9.
17.Moser, Unexpected Outcomes, ch. 2.
16. See Kanchan
18. See, for example, Ernest Gellner, Nations and Nationalism
(Ithaca: Cornell University Press, 1983);
rev. ed.
Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism,
(London: Verso, 1991); Snyder, From Voting to Violence.
19. Democracies
in Developed
1998), 109-25; Andrew Reynolds,
(July 1999), 547-72.
and Developing
Countries," Legislative Studies Quarterly, 28 (February
"Women in theLegislatures and Executives of theWorld," World Politics, 51
20. Norris, "Ballots Not Bullets," pp. 211-13. For examples of studies using political parties as the unit of
analysis, see Barkan, "Elections inAgrarian Societies," pp. 108-9; Reynolds, "Constitutional Engineering in
"Ethnic and Women's Representation under Mixed Election Systems,"
South Africa"; Tatiana Kastadinova,
Studies, 26 (2007).
21. The sources for this informationwere Pyataya Rossiiskaya
Electoral
Gosudarstvennaya Duma [FifthRussian State
Izdanniya Gosudarstvennaya Duma, 1994), for the 1993 election; Gosudarstvennaya Duma
Federal'naya
Sobraniya Rossiiskoi Federatsii Vtorogo Sozyva [State Duma Federal Legislature of theRussian
Federation Second Session] (Moscow:
Duma,
Izdanniya Gosudarstvennaya
1996), for the 1995 election;
Sobraniya Rossiiskoi Federatsii Tretego Sozyva [State Duma Federal
Gosudarstvennaya Duma Federal'naya
Duma]
(Moscow:
Legislature of the Russian Federation Third Session] (Moscow:
Izdanniya Gosudarstvennaya Duma, 2000);
Galina Kozhevnikova, Lyudmila Mikheyesku, and Vladimir Pribylovskii, Deputaty Gosudarstvennoi Dumy RF
Tretego Sozyva (1999-2003)
[Deputies of the State Duma of the RF Third Session] (Moscow: Panorama,
2000). The database I assembled from these sources includes twenty-eight different ethnic categories. Iwas not
able to collect data on the ethnic identity of legislators for the 2003 election.
22. Chauncy Harris, "The New Russian Minorities: A Statistical Overview," Post-Soviet Geography, 34
(1993), 571.
World Politics, 43 (1991), 196-232;
23. See Phillip Roeder, "Soviet Federalism and Ethnic Mobilization,"
Daniel Treisman, "Russia's Ethnic 'Revival': The Separatist Activism of Regional Leaders in a Postcommunist
Order," World Politics, 49 (January 1997), 212^19.
24. Harris, pp. 553, 573.
25. Geographic concentration ismeasured
by the group in question.
as the largest share of a single region's population
that is comprised
26. These observations are based on data for the 1995 election fromMichael McFaul
and Nikolai Petrov, eds.,
ofRussia (Moscow: Carnegie Endowment for International Peace, 1998), 668-71.
27. Russian regions are all oblasts and krai, and all have majority-Russian populations. Ethnic homelands are
all republics, autonomous oblasts, and the Jewish Autonomous Okrug. Some nonfederal minorities such as
Ukrainians gained election in single member districts located in the ethnic homelands of other (federal) minori
Political Almanac
ties.
28. The proportion of the totalminority population comprised by federal and nonfederal minorities was cal
culated by totaling the numbers of all groups within each category found in the dataset and then dividing by the
total population of non-Russians. All figures came from the 2002 census.
29. Horowitz, Ethnic Groups inConflict, pp. 21-36.
30. Due to space considerations, a fullbreakdown of the partisanship ofminority deputies is not provided here
291
Politics April2008
Comparative
but is available upon request from the author.
31. It is important to note thatparties used dramatically different boundaries
for their regional sublists, often
combining several federal units into one regional list.Thus, this reformwas not equivalent to the establishment
of regional multimember
districts.
32. On the role of registration rules on the emergence of ethnic parties, see Johanna Birnir, "Stabilizing
Segments of Society? The Effects of Formation Costs
Systems and Excluding
Latin America,"
in Comparative
Studies
International Development,
39 (Fall 2004),
33. It is unclear whether ethnic parties would have flourished inRussia
Muslim
social movement
in the absence of legal limitations. A
to register in 2001. The
latter changed
party retained its leadership, organization, and profile, which originated inDagestan
a pan-Islamic
less than 1 percent of the PR vote. As Ware
electoral association would
See Robert Bruce Ware,
Reform," Europe-Asia
292
in
3-27.
its name to True
due to fears of losing its registration before contesting the 2003 election. Nevertheless,
roots. Each party won
Party
Party Foundation
called Nur (light inArabic) was allowed to contest the 1995 election, and a party orig
inally named the Islamic Party of Russia was allowed
Patriots of Russia
on New
"Recent Russian
this
and had definite Islamic
notes in the case of Dagestan,
any attempt at
face serious obstacles of internal ethnic differences and competition.
Federal Elections
Studies, 57 (June 2005),
583-600.
inDagestan:
Implications for Proposed Electoral