Democracy and the Political Party Systems
of Slovakia and the Czech Republic
Kevin Krause
Department of Government
University of Notre Dame
Notre Dame, IN 46556-5639
Krause.4@nd.edu
Research for this article was supported a grant from the International
Research and Exchanges Board (IREX), with funds provided by the National
Endowment for the Humanities, the United States Information Agency, and
the US Department of State, which administers the Russia, Eurasian, and
East European Research Program (Title VIII), a grant from the J. William
Fulbright Foreign Scholarship Board, and a seed-money grant from the Kellogg
Institute for International Studies.
Please do not cite without permission of the author
On 6 March 1998 the United States Senate's Committee on Foreign Relations
voted to invite the Czech Republic to become member of NATO, and in late
1997 the European Union included the Czech Republic among a group of six
top candidates for EU membership. Neither NATO nor the EU extended similar
offers to Slovakia. In fact, representatives of both organizations issued
sharply worded criticisms of Slovakia to explain why it had not been invited.
The very different messages sent to Slovakia and the Czech Republic accurately
reflect a significant difference in their respective political development
of those two countries. The difference, I argue, lies in the degree to
which the governing coalitions in the two countries accept accountability
to other political institutions and to the voting public. This paper is
part of a larger effort to explain the difference in accountability through
close attention to institutional factors. I argue that the difference can
be traced directly or indirectly to differences in the two parliaments
and that the parliaments' behavior can only be understood in terms of political
parties. Understanding the role of political parties requires attention
to their relationships with voters, their internal organization, and their
relationships with other political parties in a political party system.
This paper looks only at the political party system aspects and focuses
on five key aspects that are often cited as important for democratic consolidation:
dimensions of competition, system size, polarization, stability, and coalition
choice. For each of these aspects , I show the ways in which the Slovak
and Czech party systems differ and the extent to which this difference
may help to explain the gap in accountability.
In many ways the two party systems are far more alike than the differences
in political outcomes would imply. They share a similar number of dimensions
of competition and resemble each other closely in terms of system size,
stability, coalition choice. The differences between the two, while not
insignificant, are more subtle. Although they share the same number of
competitive dimensions, the nature of those dimensions is quite different.
Although the party systems are of a similar size, the relative weights
of the parties vary. And although they share certain aspects of polarization,
they differ ultimately in the degree to which parties compete for the center.
A Difference in Accountability
In Slovakia, a coalition with a majority in parliament and government has
systematically undercut the mechanisms of horizontal accountability which
should in principle restrain those bodies. In the Czech Republic the full
complement of accountability mechanisms has remained in tact. This is not
to say that certain Czech institutions have never tried to undercut accountability
but rather that the attempts have been rather considerably milder and have
not ultimately succeeded.
The current bitterness of conflicts between political institutions
in Slovakia reflects the emergence of striking imbalances in Slovakia's
political architecture. In some areas, the mechanisms of accountability
grew stronger. Parliament and government acquired additional powers of
investigation and exercised budgetary mechanisms of control over the president.
Parliament tacitly claimed supremacy over Constitutional Court decisions.
The government acquired additional control over the intelligence service
and used that resource to exert control over other institutions. The president
and the courts experienced a corresponding decline in their ability to
affect other institutions. The president lost not only specific powers
but financial resources, personnel, and media access. The court, with few
resources of its own, saw decisions ignored by parliament. As significant
as the shift in the balance between political institutions was a shift
within several of the institutions. Within parliament, a coalition
of parties occupied all major leadership posts and severely limited the
participation of all other parties. The majority coalition which emerged
from the 1994 election excluded other parties not only from government
but also from participation in other executive agencies, from the oversight
of potentially sensitive activities, and from proportional representation
on parliamentary committees. Within the government, the same majority coalition
extended its control into areas which previously had remained outside of
partisan political control including police investigation and election
supervision. The same group of parties thereby gained effective control
over virtually all institutional political mechanisms in Slovakia outside
the increasingly limited bastions the courts and a few miscellaneous bodies
such as referendum commissions and local governments. Even these holdouts
remained fully or partially dependent on the governing coalition for their
resources.
In the process of severing these links of horizontal accountability
between political institutions the actions by governing coalition have
also potentially limited the effectiveness of vertical accountability between
political institutions and the voting public. In many rural areas, state-run
media are the primary source of information, and appointments by Parliament
have helped to ensure that the coverage of this media distinctly in favors
the government and the governing coalition. Free and fair elections require
the cooperation of the Ministry of the Interior in printing and distributing
ballots, but actions by the Interior Minister during a 1997 referendum
suggest that this cooperation may not be assumed. Free and fair elections
also require that those elected are duly appointed and allowed to remain
in office, but parliamentary exclusion of two duly elected representatives
shows that the governing coalition in Parliament may refuse to cooperate
and yet no adverse consequences. The concentration of horizontal accountability
in a few hands not only allows these violations to occur, but also limits
any means of recourse. By constraining the investigative activities of
law enforcement officials and the implementation of the Constitutional
Court's verdicts, the coalition which controls both government and Parliament
effectively shields its own violations.
The comparable story for the Czech Republic is considerably shorter.
Unlike Slovakia, where the imbalance of horizontal accountability in favor
of parliament and government allowed in turn for an increasingly unaccountable
actions by those institutions against their rivals, no similar cycle ever
developed momentum in the Czech Republic. Within parliament the ruling
coalition made early attempts at limiting the role of opposition members
in committees but the efforts did not result in severe imbalances in favor
of the ruling party or the exclusion of the opposition from supervisory
bodies. As a result, when accusations arose that the Czech intelligence
service had become involved on behalf of the coalition, the accusations
received a non-partisan investigation. Although it attempted to delay the
introduction of new institutions which might limit its power, the coalition
did not succeed in eliminating those bodies or take overt steps against
already existing institutions. Recent evidence suggests that parties did
attempt to secure electoral victory through illegal funding techniques,
but the coalition directly interfere in the electoral process, and does
not appear to have made attempts to use its influence over the criminal
prosecution process to forestall investigation. In short, although the
Czech Republic has seen no fewer political scandals than Slovakia, but
nearly of these scandals differ from their Slovak counterparts in two respects.
First, the violations do not affect the core relations of mutual control
between political institutions. Second, violations do not remain uncorrected
for indefinite periods. The web of accountability--both horizontal and
vertical--while perhaps thinner than is desirable in a democracy nevertheless
remains intact in the Czech Republic.
The Centrality of Parties
The difference in Slovak and Czech levels of accountability results directly
from decisions made by parliament and parliamentary appointees. Every one
of the above- mentioned encroachments on institutional restraint mentioned
above occurred on the basis of a parliamentary decision or the decision
of an official appointed by parliament and depending on parliamentary confidence
to remain in the position. This link extends even further, directly into
the realm of political parties. After an early post-revolution period of
loose political affiliations, political parties quickly came to dominate
parliamentary politics in both countries, not only as a means of gaining
election but also as the primary basis of cohesion.
Although complete records are unavailable, it is possible to see the
overwhelming importance of party ties in Slovak and Czech parliamentary
politics by looking at the record of voting in parliament. Calculations
made by Kopecky (Kopecky 1996) show that the similarity of voting within
party groups consistently exceeded ninety percent for virtually every major
party in the Czech Republic between 1993 and 1995. Calculations made by
Parlamentni Zpravodaj show that levels of similarity remained consistent
over time and continued at high levels even after the period studied by
Kopecky. The importance of parties in shaping political outcomes is also
suggested by the fact that despite significant splintering of party groups
the number of deputies not included in one of these party groups at any
given time between 1994 and 1996 averaged less than four percent. Kopecky
notes wide variations in the amount and locus of control over parliamentary
parties but he acknowledges that "the Czech political system is devised
for the function of (parliamentary) parties which eventually came to dominate
it"(Kopecky 1996, 21). A virtually identical set of statements may be applied
to Slovakia's political system. Although no complete set of voting statistics
is available for Slovakia's parliament,
1
a look at those votes for which records are available shows an extremely
high similarity of voting patterns among members of the same party which
appears to correspond approximately to the ninety percent level found in
the Czech Republic. Records are available for many of the votes which concern
questions of accountability and appointment discussed above many records
are available, and these show nearly unanimous agreement within parties.2
And as in the Czech Republic, party voting is particularly significant
in Slovakia because very few members of parliament remained outside of
party structures for any significant length of time, and the total has
never exceeded three percent.
Further strengthening the case for parties as the key to Slovak and
Czech differences is evidence that party representatives in parliament
and government--the figures responsible for making the accountability-related
decisions discussed above--are also linked by strong ties to other organizational
units of the party. In both Slovakia and the Czech Republic party rosters
of party leaders in parliament and government closely mirror rosters of
the party executives. Kopecky refers to the situation in the Czech Republic
as "a complete elite mix, which makes the study of 'who influences who[m]'
very difficult"(Kopecky 1996). Slovakia's parties are likewise characterized
by an extremely strong overlap between parliament, government and the party
executive. In Slovakia a small group of leaders in each party dominate
both the party organization and the operation of the party in parliament
and government (Malova and Krause 1998).
It is for these reasons that the study of parties and party system
dynamics in Slovakia and the Czech Republic is essential for understanding
the factors which have lead to such a significant difference in political
outcomes in the two countries. Parliamentary decisions in both countries
are made in near unanimity and preferences expressed in parliament and
government cannot easily be distinguished from preferences existing in
the party as a whole.
Accountability and Political Party Systems
In the last two decades scholars have begun to discuss the development
of democracy in terms of the development of political party systems. When
these systems function well, they argue, the population is well represented
and power remains in check. When these systems fail, power falls into the
hands of a few, or into hands so many and so changing that it is impossible
to speak of democratic government. But just what it means for party systems
to succeed or fail differs from author to author and each focuses on particular
aspects of a party system which must be present in a certain way for the
system to work. The consequences of failed party systems described by these
authors share much in common with the problems present in Slovakia and
it is therefore plausible that an explanation for Slovak and Czech differences
might be found in differences at the party system level.
Dimensions of Competition
An 1992 article by Herbert Kitschelt shaped much subsequent debate about
the new democracies of Central and Eastern Europe by arguing that parties
in these countries might compete simultaneously along several dimensions
which were not necessarily parallel (Kitschelt 1992). Kitschelt also argues
that "the structuring of party systems is a critical element of democratic
'consolidation'" and within the framework of structuring he includes "the
dimensionality and configuration" of the party system (Kitschelt 1994).
The number and nature of issue dimensions, he contends, directly affect
the quality and ultimately the outcome of political party competition.
The question of issue dimensions also shapes every other aspect of party
system development discussed in this paper.
Before beginning an analysis of party system competition, it is
necessary to digress first to look at a different level of analysis. When
addressing question of competition within a political party system, many
analysts, most prominently Lipset and Rokkan, begin by looking at the society
in which the parties emerge and compete for votes. A variety of analyses
of Slovak and Czech political party systems begin with an analysis of the
number and type of issue dimensions which are apparent at the mass level
and which are potentially measurable through the use of survey data. Using
discriminant analysis of results from a 1993 multi- country survey, Whitefield
and Evans find that in every country in the survey, "political competition
involves cross-cutting cleavages." While the Slovaks and Czechs share this
multiplicity of cleavages, differ in an important respect. Whitefield and
Evans argue that in Slovakia "there are three significant dimensions to
voter electoral choice" as opposed to two in the Czech Republic (and indeed
in every other country included in the survey) (Whitefield and Evans 1995,
6). According to the results, the two countries also differ in the nature
of their primary and secondary dimensions. Questions concerning ethnic
rights formed the main component of Slovakia's first dimension, while questions
on economic and social liberalism formed the respective main components
of the second and third dimensions. In the Czech Republic, economic liberalism
formed the main component of the first dimension while a combination of
social liberalism, ethnic rights, and ethnic rights comprised the bulk
of the second dimension.
A study by Markowski using factor analysis on Central European University
data also collected in 1993 finds certain limited similarities to the results
of Whitefield and Evans. Markowski finds that both countries "have a clear
economic factor (economic liberalism versus populism), a religious factor,
a libertarian-cosmopolitan factor, and a factor indicating alienation from
democratic politics" but that Slovakia also has an "idiosyncratic weak
fifth" element that relates "to the division of Czechoslovakia"(Markowski
1997, 229). In addition to showing a difference in the number of dimensions,
Markowski's methods also indicate a difference in the order and composition
of the dimensions. Like Whitefield and Evans, he identifies an "Economic
populism versus market liberalism" as the first dimension in the Czech
Republic. In Markowski's work this factor is followed, in order of importance,
by factors he labels as "participation," "libertarian-cosmopolitan," and
"religious factor." In Slovakia, by contrast, Markowski finds the "participation"
factor to be most significant followed in turn by "religion," "economic
liberalism," "libertarianism-cosmopolitanism," and, finally, the idiosyncratic
"Czechoslovakia" factor (Markowski 1997, 228). Markowski's results are
weakened by two significant factors, however. First, as is often the case
with the results of factor analysis, many of his factors do not easily
bear the names he applies to them. His economic liberalism factor for the
Czech Republic, for example, consists of a high proportion of questions
which on the surface have little to do with the economy. Likewise the libertarianism-cosmopolitanism
axis in both countries contains several components with have more to do
with the economy than they do with libertarianism. A second significant
difficulty with Markowski's for purposes of this paper involves his use
of a series of questions concerning the personal well-being of the respondent
and questions about political apathy and disaffection toward politics.
While interesting, this set of questions differs from all of the others
in asking for factual rather than normative statements. Furthermore, while
apathy may play an important role in these political systems, and may become
a key element in the party appeals at election time, it is difficult to
understand how disagreement on questions such as "parties are interested
only in votes" could become the basis of sustained and consistent issue
dimension at the party level itself.
Fortunately, resolving the latter issue about which questions to include
also helps to resolve the former question of ill-fitting factor labels.
Using the CEU data set it is possible to follow Markowski's method while
excluding the questions discussed above.3
It is also possible look at results over time by using three other CEU
data sets collected over the period from 1992 to 1996. The result of factor
analysis with varimax rotation for each of the four data sets in both Slovakia
and the Czech Republic can be found in tables
A1 through A9 in the appendix. The results are striking in both their
clarity and their consistency over time. On each of the four samples Slovakia
exhibits four relatively strong factors and these factors are virtually
identical. In fact the main three components of each of the four factors
are identical across the 1993, 1994 and 1996 data sets. Only the 1992 results
differ slightly from those that follow and even then the differences are
slight. These four dimensions are relatively compact both in the number
and the thematic consistency of their components and therefore it is possible
to provide factor labels with comparatively little risk of overstatement.
In order of clarity and ease of labeling these are:
A religious dimension which involves questions on church influence, abortion,
atheists in politics. In the 1993 and 1994 samples this also includes the
question of restitution.
A national factor which includes a question about nationalism and the split
of Czechoslovakia and in all samples except 1992 includes a question on
the relative importance of expertise and patriotism among politicians
An economic policy dimension which includes two questions on privatization,
factory- closing. In every sample except 1992 these two questions factor
together with a question on the reduction of income differentials. In 1992
and 1993 these two questions also factor together with the above-mentioned
question of restitution. In 1992 the factor also includes a question about
government responsibility for reducing unemployment and for reducing crime.
A slightly more heterogeneous dimension which in 1993 through 1996 centers
around the above-mentioned questions of the role of government in unemployment
and crime and also around the question of restitution. This dimension also
consistently includes the question of political expertise and patriotism,
and at least one question each from the economic and religious dimensions.4
While the questions differ considerably, they do share a common theme government
involvement in society an the economy, and it is not inappropriate to use
relatively neutral label of "liberalism."
In the Czech Republic, the results are less consistent in terms of
number, theme or coherence over time but certain patterns do emerge. The
data for 1992 and 1994 yield three relatively strong factors for the Czech
Republic. The data for 1993 and 1996 yield four factors but in each case
two of these appear to be extremely weak. The factors themselves also fluctuate
considerably over time. Two groupings of questions factor together consistently
in all four surveys while other questions shift groupings and produce factors
which do not remain as consistent or durable. The following list, organized
in order of a dimension's consistency over time, summarizes the factors
that appear:
-
A religious dimension that includes questions on atheists in public service
and abortion. It also includes the question of expertise versus patriotism
in politics. In 1993 and 1996 this dimension also includes the question
of church influence, but it does not do so in 1992 or 1994.
-
A broad heterogeneous dimension including questions on restitution and
government responsibility in fighting unemployment and crime and declining
morals as well as the question of the dissolution of Czechoslovakia and
questions about factory closures and privatization. Except in 1993 this
dimension also includes the question of church influence. It is difficult
to characterize this many different issues under a common heading but it
bears close resemblance to Slovakia's "liberalism" dimension in that it
links a preference for the state involvement in the economy with a preference
for active government involvement in fighting "rising crime" and "declining
morals."
-
A dimension that appears as a separate factor in 1992, 1994 and 1996 that
is almost identical in its composition to Slovakia's economic dimension
and includes questions on factory closures, privatization and income differential.
In 1992 this factor also includes the question of Czechoslovakia's dissolution.
In 1996 it includes the question of government responsibility for unemployment.
-
A dimension that appears only in 1993 and that is quite similar to the
above mentioned liberalism dimension but that also includes questions of
church influence, atheism and patriotism in a seemingly inconsistent pattern.
Low eigenvalue statistics for this dimension suggest that it need not be
given serious consideration.
-
A dimension that appears only in 1993 and 1996 which includes only a question
about the harmfulness of nationalism. This question appears to have no
stable home among the other factors. In 1992 it factors weakly with the
government involvement dimension and in 1994 it factors with the economic
dimension. As an independent factor in other years its eigenvalue is extremely
small and it cannot be considered as a significant independent dimension.
In light of these explanations it is possible to confirm and expand upon
certain of the findings of Evans and Whitefield and of Markowski regarding
the comparison of Slovakia and the Czech Republic. First, it appears that
Slovakia's population does divide across at least one more dimension than
does the Czech Republic and that this difference is stable over time. Second,
since both countries share similar religious, economic and liberalism dimensions,
the main qualitative difference between mass-level issue dimensions lies
in Slovakia's sustained and consistent national dimension. The nationality-related
and patriotism-related questions which consistently factor together in
Slovakia and which consistently factor separately in the Czech Republic,
subsumed there under religious and government-involvement dimensions.
With this background in place, it is possible to return to the question
of political competition at the party system level. Substantial of evidence
from several sources suggests that the differences between Slovak and Czech
mass-level issue dimensions have equivalents at the party system level
and that political conflict in Slovakia is characterized by competition
in multiple dimensions whereas the competition in the Czech Republic is
more easily captured by a single dimension. While in many ways persuasive,
these arguments require further examination.
For the Czech Republic, a wide variety of evidence from a large number
of sources shows a consistent pattern of competition within a single dimension.
On the basis of judgments by a panel of experts in 1993, Kitschelt argues
that the Czech Republic's party system is "clearly of a uni-dimensional
nature" and that the competition of parties on this main dimension is "driven
by their economic policy stances." He acknowledges that "libertarian/authoritarian"
and "narrow/broad citizenship" questions constitute a distinct second dimension
of competition but contends that this factor is "much weaker" and "helps
us only to distinguish two minor parties that take intermediate positions
on economics, the authoritarian and xenophobic Republican party and the
libertarian and cosmopolitan Civic Movement (OH)"(Kitschelt 1994). Brokl,
analyzing the data which resulted from a subsequent interviews of party
officials, places parties in a two dimensional space of competition defined
by an economic axis and a libertarian/authoritarian axis. Like Kitschelt's
earlier work, Brokl finds that nearly all parties compete exclusively on
the economic axis, but his placement of parties on the second axis differs
significantly. According to his methods for calculating placement on this
second axis, Brokl finds the above mentioned Republican party and Civic
Movement to fall almost perfectly in the center of the axis while the Christian
Democratic Union-Czech People's Party lies (KDU-CSL) lies near the authoritarian
extreme (Brokl 1994).
Unfortunately it is not possible to use Kitschelt's data to make direct
comparisons between Slovakia or the Czech Republic because of the absence
of comparable elite surveys.5
In fact, it is difficult make any direct elite level comparisons since
what little work does exist on competition in Slovakia's political party
system tends to be largely impressionistic. It is possible, however, to
find data which can be used in lieu of elite interviews to analyze party
dimensions in both countries and over time. One method involves measures
of sympathy and antipathy for various parties which have been taken at
the mass level over a four year period in the above-mentioned Central European
University (CEU) surveys. The 1994 Kitschelt-sponsored survey asked similar
questions at the level of the party elite and comparable data exists from
a survey of the Czech parliamentary elite taken in late 1996. Using these
two data sets in concert with the CEU data, it is possible to test the
degree to which the opinions of party supporters correspond to those of
party leaders. As table
1 and table
2 show, the evaluations of Czech party elites and party voters regarding
other parties correlate extremely closely in both 1994 and 1996. Determining
whether the views of party voters about other parties provide a reliable
indicator of elite views is more difficult for the Slovak case because
of the lack of corresponding surveys taken at mass and elite levels. The
single potential test for mass-elite correspondence regarding party position
involves self-location on a left-right scale, and responses on this question
in the 1993 CEU survey of Slovakia correlates at a level of 0.98 with answers
given by party parliamentary elites in a 1993 survey by Brokl and Mansfeldova.
This high degree of correlation--higher than the corresponding level in
the Czech Republic6--offering
at least minimal confirmation that party leaders and party voters assess
party location in similar ways.
The four CEU surveys taken in both countries between 1992 and 1996
the surveys ask respondents to rate the extent to which each major parliamentary
parties represents their interests. Because these questions ask each voter
about each party, they can be used to construct a rank ordering of party
preferences for individual voters and for party voters as a group. A spatial
theory of voting suggests that it is possible to use these rank orderings
to test the degree to which Slovak and Czech parties compete in multiple
issue dimensions. If parties compete on only one dimension, it is possible
to predict the relative ranking of every party with regard to the ranking
of other parties and thus to assess the degree to which the rankings suggest
a single dimension.7
For this assessment to have meaning, however, it is necessary to begin
with the best possible estimate of what the one-dimensional ranking actually
is. Table
3 shows the one-dimensional rankings which best fit the data for each
of the available surveys in both countries.
8
The table also includes two measures which indicate the degree to which
the hypothesized single dimension corresponds to mass and elite rankings
of the parties in question. Contrary to what might be expected from the
number of issue dimensions found above at the mass dimension and from conclusions
made previously by Kitschelt, Brokl and my own work, the table shows that
with the responses made by Slovakia's voters actually correspond more closely
to a one- dimensional ranking of parties than those in the Czech Republic.9
Just how close these systems are to a single dimension is not easily measured
using the data and methods available. It is, however, possible to compare
the results obtained here with the above-mentioned results of Kitschelt's
jury and Brokl's survey data where the nature of the data allows for factor
analysis and therefore an assessment of dimensionality. Both Kitschelt
and Brokl found that the Czech political party system could be best described
by a single major and a single minor axis. Translating the factor scores
they use in drawing their maps of party location into Euclidean distance
scores and then applying both the pairwise and rank-correlation methods
described above yields the results which can be found in the final two
rows of <
/A>table
3. As is apparent, these transformations produce extremely similar
results and suggest that other obtained results reflect a system which
operates in two dimensions but in which the placement of most parties on
the second dimension depends to some extent on their placement in the first
dimension. The results indicate that this is equally both countries. The
findings here do not necessarily contradict evidence that Slovakia's political
life is shaped by a larger number of dimensions than are present in the
Czech Republic at either the elite or the mass level. The findings do suggest,
however, that if there are more dimensions operating in Slovakia, they
either fail to shape political conflict or do so do so in ways which reinforce
one another.
Having determined the similar dimensionality of the Slovak and
Czech party systems, it is necessary to determine what the dimensions represent.
Efforts by Kitschelt and Brokl and Mansfeldova have helped to clarify the
dimensions of elite-level competition in the Czech Republic, but making
an effective comparison with Slovakia requires a return to the mass surveys
used above. While this method cannot determine the exact nature of issue
dimensions which shape competition, it can help to show some dimensions
to be more plausible than others. Here it is useful to return to the results
of the factor analysis of Slovak and Czech voters. As Lijphart points out,
issue dimensions which are apparent among voters do not necessarily translate
into dimensions of competition among political parties (Lijphart 1984).
Nevertheless since parties remain dependent on the popular vote, issues
which are salient among the voters represent an important source of information
for determining the basis on which parties compete.
If an issue dimension plays a role in party competition, then responses
on questions concerning that issue dimension should be different among
voters of different parties. Furthermore, the levels of sympathy for other
parties should reflect the degree to which they share the same opinions.
There is no easily quantifiable standard for determining party stances
on particular issue dimensions, but to the extent that the opinions of
voters do not differ dramatically from those of their chosen party, it
is possible to use the mean position of party voters as a substitute.<
A HREF="#10" name="s10">10 Table
4 and table
5 show the mean scores on each of the factors found above for each
party considered. The tables also show the range between the highest and
lowest factor scores in each country for every given survey. From these
initial results it is possible to make certain conclusions about the way
in which each of these dimensions shape party politics:
The religion (REL) dimensions in both countries show striking similarities.
Both show a consistently wide range of factor scores over time, but in
both countries the extent of this range is determined largely by the sharply
outlying position of the respective christian democratic parties. When
considered without these parties, the range of scores on religious questions
becomes considerably narrower, especially in the Czech Republic where the
space among parties almost disappears. Connected with the smaller space
of competition in the Czech Republic is a considerable variation in the
relative positions of these other parties. In Slovakia, by contrast, parties
maintain a relatively consistent rank order.
The economic (ECON) dimensions in both countries also show considerable
similarity both in the range and spread of party positions and in the consistency
of relative party positions over time. With the exception of the Republican
Party (SPR-RSC) in the Czech Republic, no party makes significant changes
in raw score or relative position from year to year.
The liberalism (LIB) dimension does differ across the two cases. In the
Czech Republic this dimension exhibits a reasonably wide range and even
distribution of party positions and extremely high consistency over time.
In Slovakia, this dimension has a considerably narrower range and shows
considerable change in relative party positions over time. Part of the
variation may stem from a significant change between 1992 and 1993 in the
factors which comprise this variable, but significant change continued--particularly
with regard to the factor score of the Hungarian Coalition (MK)--even after
the 1992-1993 period. A look at answers to the questions underlying this
factor help to explain the problem to some extent. The significant drop
in the MK factor score between 1993 and 1994 can be attributed not to a
change in the responses of MK voters away from the "liberalism" end of
the scale but rather from a change in the responses of both Christian Democratic
Movement (KDH) and Slovak National Party (SNS) voters toward the "liberalism"
pole and the entrance of the Democratic Union (DU), a party with even higher
scores on the LIB factor. While this explanation merely raises other questions
about shifts in KDH and SNS, it also emphasizes that parties in Slovakia
do make considerably more dramatic shifts in both relative and absolute
position on this dimension than they do in the Czech Republic and points
to the relative instability of this dimension in Slovakia.
The national identity dimension (NAT) differs significantly across the
two countries. In the Czech Republic the dimension appear only twice in
the four samples and shows an extremely narrow range and a high degree
of asymmetry. In both cases, the Republican Party (SPR-RSC) stands at one
extreme while all other parties stand together in a very narrow
range at the other extreme. In Slovakia the national dimension produces
quite different results, showing a wide and even distribution of party
positions and an almost perfect consistency of party positions over time.
Having looked at the individual contours of these dimensions, it is
possible to see how they relate one another. <
A NAME="tab6" HREF="#table6">Table
6 and table
7 show the correlation between mean factor scores of party voters.
Only two correlations between dimensions show sustained high levels over
time: the LIB and ECON dimensions in the Czech Republic, and the REL and
ECON dimensions in Slovakia. Other dimensions correlate only sporadically
and do not appear to be closely related. Based on this evidence it can
be argued that in the Czech Republic the LIB and ECON dimensions are closely
related and concern all parties while the REL and NAT dimensions are distinct
and concern primarily the Christian Democratic Movement (KDU-CSL) and the
Republican Party (SPR-RSC), respectively. In Slovakia, competition appears
to be more complicated, with the space of political competition shaped
by the distinct and significant ECON and NAT dimensions in concert with
the equally distinct but more limited REL dimension. From the evidence
available, the role of the more limited and inconsistent LIB variable is
unclear.
It is possible to test the ways these dimensions shape competition
among political parties by using the data for party sympathy. If particular
issues dimensions are more important than others, these should shape how
members of particular parties feel about other parties. Parties which take
similar stands on an important issue dimension should show higher mutual
regard than parties whose similar stands come only on unimportant issue
dimensions. If an issue is important enough to determine the shape of political
competitions, the sympathy of party voters towards another party should
correspond to some degree to the distance between the mean position of
the two parties on the issue dimension it question. The more defining is
the issue dimension, the higher should be the correlation. Table
8 and table
9 show the results which emerge when this test is applied to the CEU
data sets. In the Czech Republic it is the LIB factor which correlates
most closely with sympathy scores in each of the data sets. It is followed
closely by ECON during each of the three years that it appears as a separate
factor. The REL factor correlates weakly with party sympathy over all four
years. The NAT factor correlates weakly in 1993 and much more strongly
when it again appears as a factor in 1996. The factor labeled LIB2 appears
only once and its strong correlation with party sympathy likely reflects
the high degree to which it overlaps the LIB variable. In Slovakia, there
is less consistency over time but certain trends emerge quite clearly.
In 1992 all four factors correlated with party sympathy to almost the same
degree. In 1993, the correlation with LIB and NAT increased significantly.
In 1994, correlation declined for all factors, but the especially sharp
decline of LIB, REL and ECON left NAT with a level of correlation almost
twice as high as any of the other three, a situation which continued through
1996.11
Looked at in broadest terms, these calculations provide solid support
for most conclusions made regarding party system competition in the Czech
Republic and provide an preliminary evidential basis for making conclusions
about Slovakia. In line with findings by Kitschelt, Brokl, Evans and Whitefield
and Markowski, this data suggests that parties in the Czech Republic compete
primarily on an axis defined by preference for or opposition to certain
forms of liberalism. The LIB factor, which correlates most closely to party
voters feelings about other party, taps feelings about the need for a government
of experts which would keep its hands out of individual lives and restore
primacy to the private sector (and distance itself from allegedly statist
Slovakia). The ECON factor taps feelings directly related to the role of
the private sector in the economy. While not identical, party responses
on these two dimensions correspond to so closely that they form what can
be considered essentially a single dimension. It is the same dimension
noted by Kitschelt and Brokl. The outliers noted by Kitschelt and Brokl
correspond quite closely to the placement of parties on the two other,
partial dimensions noted here: the Republican Party stands apart on a national
dimension which pits the Republicans against everyone else, while the Christian
Democratic Union stands apart from the others on a religious dimension.
These dimensions exist but clearly play a less significant role in shaping
political competition and only for certain parties. 12
The complications along with the presence of two parallel but not perfectly
aligned major factors may help to explain why a one- dimensional rank ordering
of parties remains limited in its ability to predict levels of party sympathy.
In Slovakia, the similarly limited predictive ability of the one-dimensional
ranking must be attributed to a slightly different ordering of dimensions
of competition. In Slovakia, especially after 1993, the NAT dimension explains
political competition far better than any rival dimension. Those other
dimensions--unlike the tertiary dimensions in the Czech Republic--the ECON
and LIB dimensions in Slovakia are broader and less dependent upon the
outlying position of a single party. The ECON dimension in particular shows
robust and consistent party competition over time and little correlation
with the NAT dimension. These other factors do not shape the overall dimension
of competition but they do appear to affect competition among particular
groups of parties that are arrayed within that dimension. As figure
1 and figure
2 show, the range of difference on factors other than NAT in 1994 and
1996 differs considerably even among parties with similar values for NAT
and this internal difference is particularly high for low values of NAT,
which is to say for those parties whose voters tend to agree that "nationalism
is harmful" and disagree with the breakup of Czechoslovakia. These national
questions--and others which are related--may shape party competition and
even the composition of party coalitions, but parties within these coalitions
must face the possibility of internal disagreement along other dimensions.
As figures 1 and 2 show, and as Slovakia's opposition politicians have
already discovered, the potential for disagreement appears to be much more
significant on the anti-"nationalism", pro-Czechoslovak side of the NAT
axis. Unlike the main issue dimensions in the Czech Republic, the main
dimension in Slovakia yields an asymmetrical distribution of parties on
other dimensions. Put another way, there are other issue dimensions
in Slovakia's politics but their effects are felt mainly in one half of
the political spectrum.
One final consideration must be addressed with regard to issue dimensions
in Slovakia. Along with other authors, I have suggested that issues regarding
the use and abuse of political power take precedence even over national
issues in shaping the political competition among political parties (Butorova'
1997; Krause 1996a; Meseznikov 1996). Unfortunately, the otherwise inclusive
set of questions in standard CEU surveys does not include a suitably sensitive
instrument for exploring public attitudes toward the use of power. Fortunately,
however, the CEU survey taken in Slovakia just before the 1994 elections
included an extended battery of questions which do touch on precisely these
points. Table
10 compares the correlations between party sympathy scores and factor
scores including a fifth factor, AUTH, which is calculated as the sum of
responses on three questions which directly concern the use and abuse of
power by political leaders<
/A>13
The correlation between AUTH and overall party sympathy proves to be extremely
high, even higher than that for NAT. Furthermore, as table
11 indicates, party positions on AUTH correlate extremely closely with
NAT. Although the data is not yet available which could show that this
relationship holds outside of this one survey, the strength of the correlations
make it reasonable to speculate that this authority dimension does play
a strong role in shaping competition. Yet the high degree of overlap between
AUTH and NAT--even higher than between LIB and ECON in the Czech Republic--prevents
AUTH from appearing as a separate dimension or detracting from a one-dimensional
rank ordering.
Kitschelt, though he pioneered the exploration of multiple issue
dimensions in Central and Eastern European party systems, he does not offer
a developed explanation of how the number or type of dimension shapes consolidation.
An argument consistent with Kitschelt's multi-dimensional approach can
be found in the work of Offe, however. Without explicit reference to political
parties, Offe argues that the political change in Central and Eastern Europe
would occur simultaneously on three levels which he defined as "the territorial
issue," "the issue of democracy," and "the issue of economic and prosperity
order" corresponding to the vital questions concerning "who 'we' are,"
what are the "rules, procedures and rights," and "who gets what, when,
and how"(Offe 1991). Offe furthermore defines these levels of decision-making
as hierarchical and places the territorial and citizenship questions at
the "fundamental" level, built on by the constitutional and institutional
framework and finally the distribution of economic resources. By Offe's
analysis, the type of issue dimension which dominates a country's politics
must matter a great deal. The focus of the Czech parties on economic issues
along with certain limited questions of rights and government activity
suggests a willingness to refrain from making changes in more fundamental
institutional and national arrangements. In Slovakia, by contrast, the
fundamental levels remain the constant source of political conflict. While
these fundamental issues must not be resolved hurriedly, they also cannot
go unresolved for long periods without threatening the stability of the
system as a while. It is much easier for enterprising political leaders
to alter institutional relationships in settings where questions about
citizenship and institutions are still "in play" than in settings where
the matter is considered resolved. Unlike their Czech counterparts, Slovak
political elites who wished to change institutional relationships did not
first have to overcome presumptions against raising such issues at all.
It is also worthy of note here, though there is not space or time to
develop it fully here, that the dimensions of competition also shape political
development in other ways. Although no formal data is available to prove
it, a close observation of the Slovak political scene suggests the sustained
competition of parties along national and authority dimensions has resulted
in a self-sorting of party elites and party activists along these lines.
The result has been an inversion of circumstances found in democratic polities
with a strong socio- economic issue dimension. In these polities, the socio-economic
conflicts may become bitter, even extreme, but questions about institutions
and citizenship remain muted because parties themselves encompass a wide
range of opinions on these issues and it is only infrequently that any
particular stance gains an upper hand within parties. In Slovakia, by contrast,
it is the socio-economic issues which have become dispersed within coalitions
and even within parties such as Movement for a Democratic Slovakia (HZDS)--while
the institutional and national questions have become the subject of debate.
The result has been a moderate and, by most standards, successful economic
policy coupled with exacerbated national tension and fragile political
institutions. Alignment along the NAT and AUTH dimensions of competition,
has left Slovakia's current coalition with ample resources for reining
in economic extremists and few internal counterbalances against national
and institutional extremes.
Party System Size and Balance
In his work on party systems and democracy, Sartori differentiates systems
according to the size of the party system. The number of parties, he argues,
not only reflects the concentration of power within the system but also
affects the interplay of parties, affecting their methods of competition
and cooperation and affecting the success of governing coalitions. The
difference between the behavior of the Slovak and Czech governing coalitions
might be explained in Sartori's terms as a the result of differently sized
party systems. In this framework, the emergence of accountability imbalances
in Slovakia would reflect a shift away from Sartori's central categories
of "twoparty" and "limited pluralism" either in the direction of a "hegemonic
party system" or in the direction of an "extreme pluralism" which has begun
to undergo collapse of the sort experienced in Weimar Germany or the French
Fourth Republic. Yet a look at the Slovak and Czech party systems shows
that Sartori's numerical standard does not offer much help in explaining
the different political outcomes. Two independent methods for counting
party system size show the Slovak and Czech systems to be extremely similar.
Counting parties consistently and reliably is not an easy task. Difficulties
arise in how to count coalitions, splinter parties, parliamentary clubs
and fringe parties. Sartori suggests counting only those parties with show
a long term potential to affect the strategies of other parties either
through coalition or obstruction. With the passage time and an increased
basis for comparison it has become easier to apply Sartori's method in
a way which makes reasonably good sense while remaining within the bounds
of certain specific counting rules. Thresholds for entry into parliament
and for the establishment of formal parliamentary party groups are similar
in the two countries and provide a first set of guidelines for excluding
parties as irrelevant, since it is unlikely that parties which cannot gain
entry into parliament or form their own organization once there have had
scarce impact on other parties. This restriction simplifies matters considerably
in Slovakia where electoral coalitions led to the entry of numerous small
parties in 1994. Although these parties are politically active, none of
them can be considered an important independent actor in Slovakia's politics.
These rules leave certain questionable cases, however. The considerable
party splintering which has occurred in both the Slovak and Czech parliaments
has produced numerous small groups of deputies which are capable of mustering
sufficient numbers to establish a formal party group but which have little
popular base and cannot anticipate returning to parliament in the following
election. None of these parties has had significant long term impact on
either country and thus by Sartori's standards, all should be excluded.
This can be accomplished systematically simply by excluding parties with
delegations in parliament which consistently fall more than two percentage
points below the electoral threshold. The results of applying both rules
appear in table
12
When applied to Slovakia's party system these rules yield a number
which is consistent over time and corresponds to actual developments within
the system. The drop which occurs between 1991 and 1992 reflects the decline
of Slovakia's Green Party (ZS) and its Democratic Party (DS). The rise
between 1993 and 1994 reflects the emergence and electoral success of the
Party of Workers of Slovakia (ZRS) and the Democratic Union (DU). Outside
of these developments, there are few significant changes in over the entire
period,14
and the same six parties are represented in each of the nine years under
study.15
The Czech case shows a bit more variation over time. The rise in the number
of parties from 1990 to 1992 reflects the splintering of the Civic Forum
(OF) into two viable electoral parties and the emergence of the Czech Social
Democratic Party (CSSD), the Republican Party (SPR-RSC) and the Liberal
Social Union (LSU). The decline after 1992 reflects the disintegration
of the LSU and of the Movement for Self-Administered Democracy-Society
for Moravia and Silesia (HSD-SMS). The rise in 1998 reflects the split
of the Civic Democratic Party (ODS) into two distinct and almost equally
sized parties.16
While there are differences between the Slovak and Czech Republic according
to Sartori's method, the similarities more striking. In both case both
parties have remained consistently above Sartori's threshold for "extreme
pluralism," and neither system has shown a tendency to shrink over time.
These similarities are further emphasized by the alternative method for
counting party system size proposed by Taagepera and Laakso. This method
avoids the ambiguity of counting rules by assigning different relative
weights to parties rather than by excluding them. As a result it may, in
fact, yield an over-count by including parliamentary party groups with
no meaningful mass base. Yet to the extent that this paper focuses on the
parliamentary roots of accountability and since even groups of deputies
without a mass base may shape policy while within parliament, such a method
can yield relevant results.
As Table
13 shows, results for the Taagepera and Laakso method correspond roughly
to those of the Sartori method, especially in the case of Slovakia. With
this method, as with Sartori's, Slovakia begins at a fairly high level,
drops significantly in 1992 and 1993 and then returns to a higher level
where it remains until the present. Here too the Czech republic begins
at a relatively low level, increases dramatically in 1991 and 1992 and
then falls to a lower level in 1996 only to rise again in 1998. For the
Czech Republic the two methods yield potentially significant differences,
however. Whereas Sartori's method shows a drop in party size beginning
shortly after the 1992 elections as certain parties lost their cohesion,
Taagepera and Laakso's method records this as in increase since splinters
from these parties created their own parliamentary groups. Taagepera and
Laakso's method also records a rather more significant jump between 1997
and 1998 as a result of the split of the Civic Democratic Party.17
It is difficult to argue that either of these two methods represents
a clearly better understanding of party system size in these two countries
during the period in question. Sartori's method excludes parties that are
not relevant in the long run, but some of those parties certainly may remain
relevant in the short run. The many small parties which filled the Czech
parliament between 1993 and 1996 certainly increased the complexity and
difficulty of Czech politics and on occasion these parties did shape political
outcomes. In the broad framework for the effect of party system size on
political outcomes that is proposed by Sartori, it does not appear to matter
which counting method is used. In their work on Latin American party systems,
Mainwaring and Scully translate raw numbers produced by Taagepera and Laakso's
formula into Sartori's categories as follows: "Systems with an Ns between
3.0 and 3.9 usually correspond to Sartori's category of limited pluralism,
while those with an Ns of 4.0 or higher usually correspond to the category
of extreme pluralism"(Mainwaring and Scully 1995, p. 32). As with Sartori's
own counting rules, the Taagepera and Laakso rules place both countries
consistently within the extreme pluralism range for every year after 1992.
Another aspect of Sartori's study of party system size hints at greater
differences. Within his classification of party systems he includes the
category of "predominant party systems." These are systems in which multiple
parties exist freely and legally but in which the "major party is consistently
supported by a winning majority"(Sartori 1976, 196). It is therefore possible
that differences in the relative strengths of parties rather than overall
party size might explain the differences in Slovak and Czech political
outcomes. With the exception of a brief period in the Czech Republic after
the 1990 elections, neither the Slovak nor the Czech parliament meets Sartori's
formal condition for a "predominant party system" since in neither case
could the largest party command a simple majority of parliamentary seats
or rely on a "standing and efficient practice" of allowing "minority single
party governments"(Sartori 1976, 196). Nevertheless, between 1990 and 1998
both the Slovak and Czech parliaments have experienced long periods of
apparent dominance by a single party which is considerably larger than
its next largest competitor.
Table
14. shows the relative sizes of largest and second largest parties
in the two countries between 1990 and 1998. The results indicate that with
the exception of a brief period after the first elections, Slovakia largest
party has consistently outnumbered its next largest rival by a consistent
margin and that this margin has remained consistently higher than the margin
in the Czech Republic except for a brief period in 1994 after the Slovakia's
largest party suffered significant splintering.
It is possible to gain a better understanding of the significance of
these imbalances by looking at how the distribution of seats shapes the
relative influence of parties in parliamentary voting. In multi-party systems
where no single party has gained a majority, parties must cooperate to
achieve majorities. In such circumstances, a party with two percent of
parliamentary seats may be as influential as one with forty-nine percent
of parliamentary seats if both parties in producing a majority. To measure
such questions of relative influence, scholars have developed "power indices"
which determine how often a party holds the decisive vote. Using the principle
of the Banzhaf-Coleman power index, which posits all possible coalitions
and counts the number of cases in which a party can prevent a majority
by refusing to cooperate, it is possible to assess the relative influence
of parties in Slovakia and the Czech Republic (Turnovec 1995). Table
15 presents the ratio of "decisive position" of the first and second
largest parties in both countries over time. These results indicate a considerably
greater difference between the Slovak and Czech systems since even at the
point of greatest influence by this measure, no dominant party in the Czech
republic after 1990 managed to achieve a position of dominance- -according
to this measure--which even approached levels achieved by Slovakia's dominant
party in 1992 and 1993 and from the 1994 elections until the present. Of
course the Banzhaf- Coleman method is limited in what it measures since
not all possible coalitions are even remotely practical. Nevertheless,
the indicator does give a good sense of party dominance by indicating simply
how difficult it would be to assemble a winning coalition without
the dominant party. In Slovakia the distribution of seats makes that prospect
overwhelmingly more difficult than in the Czech Republic.
Slovakia's party system thus comes closer to resembling Sartori's predominant
party system and has maintained that resemblance over a longer period of
time. As Sartori himself cautions, however, predominance of one party does
not necessarily lead that party to use its position to undermine other
democratic structures, and he points to numerous historical examples of
parties with longstanding outright majorities and divided oppositions which
nevertheless remained competitive democracies. Slovakia's own development
reinforces this point since the Movement for a Democratic Slovakia (HZDS)
has held the position of pre- dominant party since 1992 but despite significant
predominance in its early years did not actively begin to undercut horizontal
authority until after the 1994 election. Clearly while predominance may
represent a contributing factor it does not alone explain the differences
between Slovak and Czech political outcomes.
Polarization
One bete noir of Sartori's Parties and Party Systems is the
outcome he calls "polarized pluralism" in which large numbers of political
parties combine with centrifugal competition that favors the extremes over
the center. Sartori refuses to draw a direct connection between polarized
pluralism and the long-term survival of democracy, but he calls attention
to the "'external' fragility and exposure to exogenous crises" of polarized
polities (Sartori 1976). It is plausible that the different political outcomes
of the Slovak and Czech cases could be accounted for by Sartori's category.
Since both of the countries have consistently been characterized by large
numbers of parties, any difference in polarization might offer a key to
Slovakia's fragility.
One of the main difficulties in applying Sartori's concepts to the
Slovak and Czech cases is his consistent reference to pluralism in terms
one-dimensional competition. Elsewhere in his work, Sartori acknowledges
the possibility of multiple dimensions but suggests that party systems
with multiple parties ultimately settle into a single predominant dimension
(Sartori 1976, 341-342).18
His reasoning appears to apply rather well to the Slovak and Czech cases,
but it is impossible to go further without knowing the nature of the predominant
dimension. Looking for polarization in Slovakia's economic dimension or
the Czech Republic's national dimension would mean little since the main
thrust of competition is elsewhere.
Although certain authors suspect the Czech Republic to be in danger
of polarization (Novak 1997), a closer look at Sartori's guidelines and
the parties in the Czech system suggest that this possibility is unlikely.
Sartori offers eight characteristics of polarization, but each of these
focuses in some way on the relationship between the political center and
the political extremes. Polarization occurs when parties abandon competition
for the center in favor of competition for the extremes. The Czech Republic
possesses certain of the elements necessary for polarized competition including
opposition parties at the extremes which are free from the limitations
of governing parties and which occasionally use anti-system rhetoric. Yet
the existence of these parties on the edges of the party system has not
redirected competition from the center. The Communist Party of Bohemia
and Moravia (KSCM) has consistently gained a 10% or greater share of votes
in elections, but its limited potential for attracting new adherents is
apparent in the gradual aging of the average KSCM voter. Furthermore, it
is not entirely clear that the party can be necessarily deserves the anti-system
label which is frequently attributed to it by other parties. Neither KSCM's
electoral strategy nor its efforts in parliament have pointed toward a
rejection of the system as a whole (Krause 1996b; To'ka 1997). The Republican
Party (SPR-RSC), with its rhetorical focus on race and its obstructionist
parliamentary tactics holds a stronger claim on the anti-system label,
but it stands alone at one end of the national dimension against almost
unified opposition. Furthermore, despite a period of strong growth in the
early 1990's, the party could not attract even ten percent of the popular
vote in any of the three parliamentary elections held after 1989. Four-fifths
of Czech voters consistently voted for parties which both in elections
and in parliament continued to aim their competition toward the center
rather than the extremes, thus forestalling the danger of polarization.
Applying Sartori's definition of polarization in Slovakia is difficult
even after the main dimension of party competition has been determined.
The national axis does, however, fit the Sartori's pattern in several ways.
At the extremes of this axis stand the two parties which appear to form
the "bilateral opposition" described by Sartori: the Slovak National Party
(SNS) and the Hungarian Coalition (MK). Whether these parties are indeed
anti-system parties is difficult to discern, but it is clear that each
perceives the other as such, and both frequently repeat the rhetoric of
the other side as evidence of the threat posed to the whole system. Furthermore,
while the political potential of both MK and SNS are limited by their small
limited electoral support and small delegations in parliament, both parties
may be perceived by the other side as able to call upon powerful outside
forces to implement their agenda. MK leaders call attention to the political
alliance between SNS and the much larger Movement for a Democratic Slovakia
(HZDS), while SNS leaders consistently call attention to ties between MK
and political leaders across the border in Hungary. There does not appear
to be "ideological patterning," of the sort described by Sartori, but substituting
the word "national" for "ideological" in his description produces a statement
which accurately describes much of the political competition in Slovakia:
"The common characteristic is ... that all the parties fight one another
in terms of [national] arguments and vie with one another in terms of [national]
mentality"(Sartori 1976, 137).
Nevertheless, many elements of Slovakia's political party system simply
do not fit Sartori's pattern. The system does not show signs of a "centrifugal
drives" by which extreme parties gain support at the expense of a prominent
center party. While this may have occurred to a certain extent in 1992
and 1994 in the form of more extreme splinter parties which abandoned the
Christian Democratic Movement (KDH) and the Party of the Democratic Left,
a countervailing process occurred at the same time in the form of center
oriented splinters from HZDS and SNS. Studies of electoral ebb and flow
show that by 1995 the movement of public support between the less "national"
parties of the opposition and the more "national" parties of the coalition
had fallen to almost nothing (MVK SRo 1996).
Slovakia during this period also confounds Sartori's definition in
terms of the nature and role of its anti-system parties. While MK and SNS
each perceived the other as anti- system, the main violations of principles
of democratic accountability in Slovakia must be attributed mainly to HZDS,
a party which fits poorly in Sartori's categories. HZDS actively subverted
the institutional mechanisms of control designed to restrict executive
power but claimed at the same time to be the only real defender of the
status quo. Furthermore its efforts did not bear many hallmarks of ideological
struggle. They frequently occurred in response to actions by others and
were frequently accompanied by ad hoc justifications which did not become
coherent until they were refined later by party spokespeople and the party
press.19
HZDS may thus represent a deviation from Sartori's scheme since it began
its anti-system efforts only after gaining power. Sartori regarded electoral
victory by an anti-system party as one of the unfortunate consequences
of polarized pluralism rather than as one of its characteristics.
With all of these considerations, it is difficult to know just how
to evaluate questions of polarization in Slovakia. Underlying Sartori's
list of characteristics is the argument that democracy faces serous dangers
when parties are not willing or able to compete for voters at the center
of the political system. This condition certainly applies more closely
to political party competition in Slovakia than in the Czech Republic,
and it helps to explain certain of the difference in political outcomes.
While there appears to have been little centrifugal tendency in Slovakia
after 1994, there was equally little centripetal pressure. The parties
in power found themselves able to count on the loyalty of a particular
segment of the population but unable to win converts from among the rest.
The solidity of their own base and the futility of appeals to opposition
voters may have thereby freed coalition leaders from electoral restraints
and allowed them to act against other centers of power. At the same time,
those encroachments promised additional powers and resources which could
be used to lure voters who might not accept the coalition's political stance.
And absent a shift in voters, the same powers offered the possibility of
shifting the whole electoral system in a way that would be favorable for
the coalition. In the Czech Republic, also seemingly contrary to Sartori's
expectations, the existence of sufficiently strong parties at the extremes
has actually forced parties to seek voters at the center, helping to keep
those parties sensitive to swings in public opinion and perhaps in the
process restraining their encroachments on other institutions.
Stability of competition
A number of authors, most recently Mainwaring and Scully (Mainwaring and
Scully 1995) and Mair cited in xx(To'ka 1997) have emphasized the importance
of stability of party competition in establishing the conditions for successful
democratic consolidation. Mainwaring and Scully in particular argue that
consolidation is enhanced by regular patterns of party competition (Mainwaring
and Scully 1995). Since party competition takes a number of forms, a variety
of measures may be used to determine the extent to which this condition
holds in Slovakia and the Czech Republic and the extent to which differences
may have contributed to the different political outcomes.
The most commonly used indicator of stability in party competition
is the volatility of party voting over time. Pedersen's index (Pedersen
1990) offers a simple and easily comparable test for volatility, though
in cases of party splits and mergers this method requires certain prudential
decisions on the part of the researcher.
Table
16 provides calculations of Slovak and Czech volatility based on varying
sets of assumptions. Toka (To'ka 1997) calculates volatility using both
a strict application of the rules, which tends to maximize the reported
volatility by regarding splinter parties and electoral coalitions as new
parties, and a more relaxed application which tends to minimize the reported
volatility by regarding splinters and coalitions as completely continuous
with the previous electoral period. Mainwaring takes a middle course by
tracing continuity between parties and their largest splinter and between
coalitions and their largest component party. As the table shows, volatility
in the Czech Republic corresponds to or exceeds that of Slovakia in each
period and according to each method of calculation.<
A NAME="s20" HREF="#20">20
A similar conclusion applies to party volatility within parliament during
parliamentary terms. Both countries have experienced periods of widespread
splintering among parties within parliament. Table
17 shows similarly high rates of intra-parliamentary volatility in
both countries between 1990 and 1994. After the 1994 election in Slovakia,
however, volatility in parliament declined to near-zero levels while remaining
elevated in the Czech Republic. The behavior of individual party voters
follows an identical pattern. According to Toka's analysis of voter party
loyalty displayed in figure
3, Czechs and Slovaks showed almost identical declines in party loyalty
after the 1992 election. In the Czech Republic, this decline continued
steadily until the 1996 election. In Slovakia pre-term elections in 1994
created a new baseline for comparison and although voters showed the same
expected pattern of decline, the degree of decline over time proved considerably
smaller than it had been after the 1992 election.
Mainwaring and Scully call attention to one final important element
of stability when they point out that in institutionalized political party
systems "a party that is markedly to the left of another party does not
suddenly move to the right of it to gain short-term electoral advantage"(Mainwaring
and Scully 1995, 5). As table
3 indicates, the best fitting one-dimensional rankings of parties show
very little change over time with the exception of a small fluctuation
in the position of the Christian Democratic Union (KDU-CSL) in the Czech
Republic. Similarly, table
4 shows that on neither the NAT, ECON or REL dimensions in Slovakia
does any party change position by more than a single rank order over time.
Only on the LIB dimension do parties shift relative positions and only
one party shifts by a significant amount. As table
5 shows, the Czech Republic is also free of significant shifts in relative
position on any of its dimensions with the exception of the Republican
Party (SPR-RSC) which shows little stability in any dimension except NAT.
Since none of these measures of overall party system stability shows
Slovakia to lag behind the Czech Republic, it is difficult to use this
variable to explain Slovakia's relatively larger difficulties with democratic
consolidation. Furthermore, the period of greatest encroachment against
institutional restraints in Slovakia corresponds precisely with the period
of least volatility in parliament and greatest voter loyalty. While by
no means reversing the relationship between stability and democracy cited
by Mainwaring and Scully, this circumstance does raise important questions
about when and how increased stability benefits democracy. The overwhelming
decrease in intra-parliamentary volatility in Slovakia after 1994--a decrease
which can be explained largely by the disciplinary mechanisms introduced
by the Movement for a Democratic Slovakia (HZDS) to control its own deputies
(Malova and Krause 1998)--brought with it an increase in the coalition
government's ability to achieve near unanimous support from its deputies
for the encroachments described above. Similarly, as is argued above, the
increased loyalty of certain groups of coalition voters appears to have
allowed the coalition to disregard public opinion and continue its efforts
to dismantle institutional restraints.
Coalition Choice
From the 1990 election until early 1991 the Civic Forum (OF) commanded
a majority of votes in the Czech parliament. Outside of this short period,
no single party has controlled a majority of votes in either Slovakia or
the Czech Republic. As a result, parties in these two countries have found
it necessary to form coalitions for the purposes of installing a government
and approving legislation. The composition of coalitions clearly has a
major impact on what those coalitions do and therefore might help to explain
why the behavior of the Slovak coalition has differed so significantly
from that of its Czech counterpart. It is necessary, therefore, to look
at the characteristics of the political party system which affect coalition
formation.
The theories of coalition formation noted by Lijphart rely upon several
factors for determining the likelihood of combinations. These include the
size of the coalition, and the ideological positions and contiguity of
the parties. While they differ in the importance they place on each of
these elements, rival theories conjecture that all else being equal coalitions
will be characterized by a fewest possible parties, the smallest possible
majority, the smallest difference in ideological position and the highest
degree of continuity (Lijphart 1984).
Table
18 and table
19 provide this information for the coalitions which did form in Slovakia
and the Czech Republic as well as for a sample of the next most likely
coalition according to these guidelines.
One striking result which is immediately apparent is that every coalition
in both of the two republics follows is consistent with every hypothesized
coalition choice rule except the rule of minimum majority which studies
cited by Lijphart show to be the least effective of the rules at predicting
coalition choice (Lijphart 1984, 52). Both countries at each time period
contained a significant number of combinations which minimize the number
of parties, but not all of these combinations fare as well according to
the standards of ideological range or continuity and only the best two
in each period are listed in the tables. The "next-best- choice" in both
countries is relatively stable over time. In Slovakia this choice involves
substituting the Party of the Democratic Left (SDL) for the Slovak National
Party (SNS) or the Association of Workers of Slovakia (ZRS) in a coalition
with the Movement for a Democratic Slovakia (HZDS). In the Czech Republic
the choice involves substituting the Czech Social Democratic Party (CSSD)
for the Civic Democratic Alliance (ODA) or the Christian Democratic Union
(KDU-CSL) in a coalition with the Civic Democratic Party (ODS). None of
these substitutions performs well in terms of the other considerations,
however.21
While a partnership between SDL and HZDS would not necessarily introduce
non-contiguity among coalition partners along the main axis of competition,
it would increase the ideological range significantly. Antipathy between
SDL and HZDS increased significantly between 1992 and 1994 and remained
at high levels in 1996, while antipathy between HZDS and SNS (and later
ZRS) remained low by comparison. In the Czech Republic the antipathy of
ODS for CSSD did not greatly exceed its antipathy for KDU-CSL, but the
difference is still of notable size, and the introduction of CSSD at the
expense of one of the coalition partners would make the coalition non-
contiguous in a quite noticeable fashion. Neither of these circumstances
necessarily excludes a HZDS-SDL or ODS-CSSD coalition, but they do show
clarify the disadvantages which both would pose. For ODS, cooperation with
CSSD would not only pose difficulties of cooperation and lead to a dramatic
increase in the breadth of the coalition's ideological position. For HZDS,
an alliance with SDL would not necessarily make the coalition more heterogenous
but would shift it away from the pro-national pole and would significantly
raise the level of antipathy within the coalition.
There also exist other potential combinations of coalitions in these
two countries which exclude the largest parties, but these possess few
of the conditions which encourage coalition choice. With the exception
of the 1996 period in the Czech Republic, all of these potential coalitions
require significantly larger numbers of parties, and while some actually
form a contiguous bloc on the main dimension of competition, they all occupy
a significantly larger range. Furthermore, few of these coalitions could
obtain a majority without including parties which lacked "coalition potential"(Lijphart
1984, 58): the Republican Party (SPR- RSC) and Communist Party of Bohemia
and Moravia (KSCM) in the Czech Republic and the Hungarian Coalition (MK)
in Slovakia.22
In general terms, the differences in constraints on coalition formation
are similar in both countries for almost all parties and in practical terms
seem limited to a slightly greater freedom on the part of Slovakia's HZDS
to chose its partners from a slightly broader list. It has been argued
that this slight difference has had a potentially significant impact on
at least one occasion but the actual effect is unclear. In June 1996 during
a crisis within the HZDS- SNS-ZRS coalition, the Party of the Democratic
Left (SDL) issued a statement in which it offered to consider support for
a HZDS minority government if HZDS would agree to meet certain conditions
including the opening of a number of parliamentary committees and supervisory
boards to opposition representation (Sme 1996). Shortly after this offer,
the coalition parties reached an accommodation and agreed to leave "the
past in the past and move on"(Narodna obroda 1996). The possibility that
HZDS could remain in government without its partners certainly did not
strengthen their bargaining position but it is unclear how the SDL offer
should be weighed against other possible inducements for a reconciliation
of coalition parties. While occasional "flirting" with SDL may serve as
a reminder that ZRS and SNS needs HZDS more than it needs them, such gestures
likely have little impact for any of the parties involved. In another work
I have argued that HZDS furthered its dominant position by choosing as
allies the parties with the weakest mass base and then by threatening them
with replacement by SDL (Krause 1997), but in light of the above analysis,
this does not appear to be correct. HZDS certainly does benefit from the
docility of its partners, but this appears more a matter of fortune than
choice. The important constraints of ideological range and contiguity,
particularly in the authority dimension limit the possibility of sustained
cooperation between HZDS and SDL and strongly favor coalition with SNS
and ZRS. In this sense HZDS appears no better able to advance its power
through coalition choice than does ODS in the Czech Republic.
Conclusion
My decision to focus strictly on party system aspects in this paper can
be characterized as a something of fishing expedition to find significant
differences in one crucial area of Slovak and Czech politics. The catch,
however, does not correspond to what experts in such expeditions might
expect. Most of the aspects of party systems do seem to matter in shaping
the different political outcomes, but they do not matter in the way or
to the degree that others predict. Sartori's argument does not explain
the differences between Slovak and Czech outcomes particularly well because
the two countries do not vary greatly in terms of party system size or
polarization. Polarization offers certain insights into difficulties in
Slovakia which are not present in the Czech Republic, but Slovakia's party
system matches Sartori's list of characteristics so incompletely that it
does not seem useful to apply his broader framework any further. Likewise,
the party system stability arguments of Mainwaring and Scully also cannot
explain the different Slovak and Czech outcomes since, again, the Slovak
and Czech cases differ very little in volatility or stability of competition.
Nor, despite my own expectation to the contrary, does coalition choice
appear to induce differences in the political behavior of the Slovak and
Czech ruling coalition.
Yet these tools of analysis cannot be quite so easily dismissed. The
analysis here focuses on those variables which cannot be measured for individual
parties but only for the system as a whole--size, relative position and
direction, and overall volatility. Although the Slovaks and Czechs may
not differ significantly on these broad variables, the results of this
paper suggests that they do differ at the level of individual parties.
Party system size is the same, but Slovakia's predominant Movement for
a Democratic Slovakia (HZDS) is more predominant than its Czech counterpart.
Polarization is unclear, but it is clear that HZDS has become an unusual
amalgam of pro- and anti-system party. Overall volatility differs little,
but low volatility for HZDS has provided a means of reinforcing control.
An examination of the party system level, while not explaining the differences,
at least points research toward the more promising party level.
Finally, there is the important background question of dimensions of
competition. Here it is the differences that are striking and the tools
that are underdeveloped. Arguments similar to Offe's about the level of
political competition are frequently used to explain the appearance of
anti-democratic activity in countries with national tensions, but these
rarely attempt to demonstrate the comparative importance of issue dimensions
or show the institutional path by which national tension causes democracy
to break down. As is apparent from the evidence above, the Slovak-Czech
comparison holds great promise for clarifying the roles that issue dimensions
play in demoratic consolidation and the role that political parties play
in developing and expressing issue dimensions.
Notes
1.A database of results is maintained by Department of
Information and Analysis of the Slovak parliament, but the information
appears to be regarded as politically sensitive by certain parliamentary
leaders, and my requests for access to the data therefore have been politely
deferred.
Back
2.Further details regarding the specifics of Slovakia's
accountability violations can be found at http://www.nd.edu/kkrause/papers/chapter2.htm
Back
3.The six excluded questions are:
| Q16A |
Voters in (country) have a real choice. |
| Q16B |
Generally speaking, those we elect to Parliament lose touch with the
people pretty quickly. |
| Q16E |
The present economic situation is very unfavorable to me and my family. |
| Q16J |
People like me have no say in what the government does. |
| Q16K |
The way things are in (country) people like me and my family have a
good chance of getting ahead in life. |
| Q16L |
Parties are only interested in peoples votes but not in their opinion. |
Back
4.In 1992 there is considerable overlap between such
questions and the economic dimension above, but even here these two dimensions
are clearly distinguishable from the remaining two.
Back
5.In 1996 I began a research effort which uses a modified
version of Kitschelt's questionnaire in interviews with deputies in the
Czech and Slovak parliament. Due to certain difficulties in data collection,
I am still compiling certain parts of the Czech questionnaire. The last-minute
refusal of coalition parties in Slovakia to cooperate in the project forced
a delay and a reorientation of the effort toward opposition parties. Both
the full Czech sample and the Slovak opposition sample should be available
by July of this year.
Back
6.The relatively low 0.89 correlation in the Czech Republic
can be attributed almost entirely to the disjunction between voter and
elite placements of the Republican Party (SPR-RSC). Factoring this party
out raises the correlation to 0.99.
Back
7.
For the following single dimensional party system consisting of parties
A, B, and C, and arranged as follows,
Party A <----> Party B <----> Party C
It is possible to predict the following rank scores:
| For party A |
A > B |
A > C |
B > C |
| For party B |
A < B |
n/a |
B > C |
| For party C |
A < B |
A < C |
B < C |
The only relationship it is not possible to predict in this situation is
the relative ranking of A and C according to voters of party B. The same
principles used in this three party example apply for any number of parties
as long the parties compete in a single dimension. For a system of four
parties there are twenty predictable relationships, for five parties there
are forty, for six parties there are seventy, and so on.
Back
8.Since there are 60 possible rank orderings in a 5
party system, 360 in a 6 party system and 2,520 in a seven party system,
it was necessary to begin by finding certain relationships common to all
parties and then to exclude rank orderings which did not contain these
relationships. This process narrowed the number of cases down to a significantly
smaller number for which it was possible to run individual correlations.
Those samples with the highest correlation scores were then tested individually
on the pairwise relationships. The pairwise relationship test and the correlation
test yielded extremely similar results for rank orderings within a given
sample.
Back
9.Because these results conflict with expectations,
it is necessary to look more closely at the indicators. The rank-correlation
indicator assumes equal space intervals between each party and thus the
close clustering of some but not all parties tends to yield a lower correlations
score than might otherwise be expected. The pairwise relationship indicator
avoids this problem by not assuming a particular distance between parties,
but as a result this measure tends to be oversensitive to small reversals
of the expected relationship.
Back
10.Available evidence suggests that this proposition
accurately describes the relationship between parties and voters in the
Czech Republic. To test its accuracy for the Slovak Republic, it is necessary
to turn again to the Brokl and Mansfeldova survey, one of the few available
surveys of both Slovak and Czech elites and the only such survey to ask
questions regarding issue positions. Comparing the mean positions of elite
party members with the mean positions of party voters on similar questions
in a CEU survey taken the same year yields similar results for both countries.
On questions concerning the regulation of abortion, the correlation between
the mean scores party elites and party voters in the two countries was
similar: 0.99 in Slovakia and 0.92 in the Czech Republic. On questions
concerning the distribution of income, the overall level of correlation
was lower overall but still similar across the two cases: 0.77 in Slovakia
and 0.81 in the Czech Republic.
Back
11.Although an analysis of correlations between party
sympathy and factor scores for individual parties does not fall into the
scope of this paper, I have included those results in the appendix in
table
A9 and
table
A10. Data from the Czech Republic shows a consistent and strong relationship
between LIB and ECON variables and sympathy for other parties among supporters
of ODS, ODA and KSCM. KDU-CSL shows the same strong pattern with LIB followed,
as might be expected, by an important role for REL. CSSD shows a strong--and
increasing--correlation between party sympathy and both LIB and ECON, though
the correlation is not as strong as among the above-mentioned parties.
The high scores of SPR-RSC supporters on the NAT factor carry through into
a significant correlation between their sympathy for other parties and
the positions of those parties' voters on national issues.
In Slovakia, the data shows a consistent strong relationship between
the NAT variable and sympathy for other parties in HZDS, SNS and MK. These
parties also exhibit reasonably strong if erratic relationships between
sympathy an ECON. For the Christian Democratic Movement (KDH), the relationship
between sympathy and REL declines over time as correlations with all three
other variables show a gradual increase. In addition to a consistently
high correlation with REL likely reflecting a continuing antipathy toward
Christian parties, the Party of the Democratic Left (SDL) showed a strong
correlation with ECON after 1993. The two parties which emerged in 1994
seem to have been strongly influenced by the dimensions which were coming
to prominence when the parties were founded. The Association of Workers
of Slovakia (ZRS) despite the class focus evident in its name actually
held its strongest correlation between party sympathy and factor scores
in 1996 on the NAT dimension. The Democratic Union, while correlating strongly
with the LIB dimension also quickly developed a high correlations with
NAT.
Back
12.Estimating the extent to which these dimensions
play a role is difficult, but the techniques used above allow certain distinctions.
The factor scores produced by Kitschelt's jury yield two dimensional coordinates.
When translated into euclidean distances and compared with party sympathy
measures from the CEU survey of the same year, they show an extremely high
correlation of 0.85, which is higher than the result obtained by using
only the factor scores from the primary dimension (0.77) or from the one
dimensional rank order suggested in the section above (0.80). The same
does not apply for the euclidean distances derived from Brokl's factor
scores. These produce a correlation of only 0.65 with the CEU survey. Part
of the problem, however, may lie in the use of equal dimensions for both
his economic and cultural dimensions, a decision which results in the placement
of the Christian Democratic Union (KDU-CSL) at a considerable distance
away from all other parties. While accepting that KDU-CSL does in fact
differ from other parties on this dimension, it may be that this dimension
does not strongly shape the views of Czech voters. It is possible to simulate
a reduced the importance for this issue dimension by dividing all factor
scores by numbers greater than one. As the divisor is increased to 3.0,
resulting factors decrease, the correlation of Brokl's model increases
to a respectable maximum point of 0.74, after which the level of correlation
begins to decline again. On the basis of this experiment it is possible
to argue that while KDU-CSL may in fact differ from other parties on the
cultural dimension, that dimension offers a much less powerful a tool for
understanding why voters in the Czech Republic like or dislike particular
parties than does the economic axis.
Back
13.Specifically, this factor sums responses on these
three questions:
| SP36b |
What is important in politics is patience at negotiations. v. What
is important is decisiveness and the firm hand of a strong personality. |
| SP37a |
The Slovak Republic is young and therefore people should refrain from
criticizing its representatives. |
| SP37b |
In the interest of the people a politician can sometimes act contrary
to the law. |
It is important to note that while I group AUTH with other factor scores,
it is generated simply from the sum of responses on these three questions
and lacks the complex weighing of scores that is involved in the calculation
of actual factors scores.
Back
14.Events in early 1998 suggest the decline of ZRS
and the rise of the Party of Civic Understanding (SOP), but a confirmation
of these developments will depend on the results of the September 1998
elections.
Back
15.This statement assumes that the Movement for a Democratic
Slovakia (HZDS) can be considered the successor to Public Against Violence
(VPN). While not true at the leadership level, this statement does apply
at the electoral level. The statement also assumes that the two major parties
representing the Hungarian ethnic group in Slovakia should be considered
as separate parties. Although these to parties have consistently campaigned
as a single coalition to avoid splitting the Hungarian vote, they have
also consistently formed separate parliamentary clubs in parliament and
disagree on important political issues).
Back
16.It is likely but not yet certain that the number
of parties may again decline to six during 1998 due to the decline of the
Civic Democratic Alliance (ODA)
Back
17.Because Taagepera and Laakso's method counts large
parties far more heavily than small ones, the split of ODS into two almost
equal halves resulted in a far larger increase in party size than would
have been the case if a small splinter group had left ODS. In fact the
mechanism of their method causes the establishment of one new party to
increase party size by nearly two full points.
Back
18.Sartori in fact suggests that the competition will
tend to spread along an dimension which the participants describe as "left-right."
That this does not appear to be the case in Slovakia reflects the dominance
of the national and authoritarian dimension. For the pro-national parties,
this dimension is more easily described in terms of "pro-Slovak" and "anti-Slovak"
than left or right, terms which carry with them economic policy connotations
that the economically centrist pro-national parties do not wish to take
on. For the anti-national parties, the dimension is best described as "pro-democracy"
and "anti- democracy" (or as certain opposition political scientists have
labeled the two parties, "standard" and "non-standard").
Back
19.In another work I detail the initial inconsistencies
in justifications offered for HZDS actions and the process by which justifications
are modified and strengthened in repeated statements by party leaders and
statements in the Slovenska republika, the party's daily newspaper.
A draft of this paper can be found at: http://www.nd.edu/kkrause/papers/chapter2.htm
Back
20.The difference between Slovak and Czech scores for
the second election period may slightly overstate the case, since the longer
period between Czech elections might be expected to yield greater volatility,
but a volatility calculation based on opinion survey taken in October 1994,
the same month as the Slovak election, shows a volatility of approximately
0.23 which corresponds closely to measurements of Slovakia's volatility
over the same period of time.
Back
21.For lack of comparative data at the elite level,
I again use CEU measures of party sympathy taken at the mass level. To
calculate the range variable, I calculated the distance between the mean
sympathy of party voters for their own party and their mean sympathy toward
every other party in the hypothesized coalition. The variable listed as
"range" is the largest such distance recorded among hypothetical coalition
partners.
Back
22.How these parties have come to be considered ineligible
for coalition is an important underlying dynamic in the interaction of
these two party systems but a full development of this topic must wait
until a later date. The most significant factor to be considered, however,
is that Czech parties with racist rhetoric (SPR-RSC) and state socialist
agendas (KSCM) become ineligible for coalition while analogous parties
in Slovakia (SNS and ZRS, respectively) became coalition members. In part
this is not as inconsistent as it seems because of the nature of the Czech
parties in question. SPR-RSC, it seems, has little agenda but racism unlike
SNS could claim the socially more acceptable themes of Slovak independence
and renewal. KSCM, while sharing important similarities in program with
ZRS, retains the air of an "unreconstructed" communist party while ZRS
is more easily able to give its demands a democratic and pluralist flavor.
Further exploration of the inconsistency would require an closer examination
of the internal politics of other parties and their reasoning for accepting
or rejecting the coalition potential of these parties.
Back
Literature
-
Brokl, Lubomir. 1994. Politicka kultura a politicky prostor Ceske republiky.
Lidove noviny, November 11, 1994.
-
Brokl, Lubomir, and Zdenka Mansfeldova. 1993. Survey of Czech and Slovak
Members of Parliament: Academy of Sciences of the Czech Republic.
-
Bu'torova', Zora. 1997. Public Opinion in Slovakia: Continuity and Change.
In Slovakia: Problems of Democratic Consolidation and the Struggle for
the Rules of the Game, edited by S. Szomolanyi and J. Gould. Bratislava:
Slovak Political Science Association.
-
Kitschelt, Herbert. 1992. The Formation of Party Systems in East Central
Europe. Politics and Society 20 (1): 7-50.
-
Kitschelt, Herbert. 1994. Party Systems in East Central Europe. Consolidation
or Fluidity. Paper read at Annual Meeting of the American Political Science
Associaiton, at New York.
-
Kitschelt, Herbert, Lubomir Brokl, and Zdenka Mansfeldova. 1994. Survey
of Czech Political Party Elites: Duke Univesity and the Academy of Science
of the Czech Republic.
-
Kopecky, Petr. 1996. Limits of Whips and Watchdogs: Cohesionof Parliamentary
Parties in the First Czech Parliament (1992-1996). Paper read at Workshop
on Parliamentary Party Groups, ECPR Joint Sessions, at Oslo.
-
Krause, Kevin. 1996a. Dimensions of Party Competition in Slovakia. Slovak
Sociological Review 1 (Fal 1996): 169-186.
-
Krause, Kevin. 1996b. The Political Party System of the Czech Republic,
Democracy and the 1996 Elections. Sociologicky casopis 34 (4): 423-438.
-
Krause, Kevin. 1996c. Survey of Czech Members of Parliament: University
of Notre Dame.
-
Krause, Kevin. 1997. Issue Dimensions and Accountability in the Slovak
and Czech Republics. Paper read at Indiana Political Science Association,
at Butler University, Indianapolis.
-
Lijphart, Arend. 1984. Democracies. New Haven, Connecticut: Yale
University Press.
-
Mainwaring, Scott, and Timothy R. Scully. 1995. Introduction: Party Systems
in Latin America. In Building Democratic Institutions: Party Systems
in Latin America, edited by S. Mainwaring and T. R. Scully. Stanford,
California: Stanford University Press.
-
Malova, Darina, and Kevin Krause. 1998. Parliamentary Party Groups in Slovakia.
In Behind Closed Doors: Parliamentary Party Groups in European Democracies,
edited by K. Heidar and R. Koole. London: Routledge.
-
Markowski, Radoslaw. 1997. Political Parties and Ideological Spaces in
East Central Europe. Communist and Post Communist Studies 221-254.
-
Meseznikov, Grigorij. 1996. Vnutropoliticky vyvoj a politicka scena SR
v roku 1995. In Slovensko 1995, edited by P. Huncik and M. Butora.
Bratislava: Nadacia Sandor Maraiho.
-
MVK SRo. 1996. Porovnanie preferencii politickych stran: MVK SRo.
-
Narodna obroda. 1996. Co bolo, bolo. Ide sa dalej. 27 June, 1.
-
Novak, Miroslav. 1997. Systemy Politickych stran. Uvod do komparativniho
studia. Prague: Sociologicke' nakladetelstvi.
-
Offe, Claus. 1991. Capitalism and Democracy by Design? Democratic Theory
Facing the Triple Transition in East Central Europe. Social Research
59 (4): 865-892.
-
Pedersen, Mogens N. 1990. Electoral Volatility in Western Europe, 1948-1977.
In The West European Party System, edited by P. Mair. Oxford: Oxford
University Press.
-
Sartori, Giovanni. 1976. Parties and Party Systems. New York: Cambridge
University Press.
-
Sme. 1996. SDL pripusta podporu mensinovej vlade HZDS. 21 June 1996,
1.
-
Toka, Gabor. 1992. Party Systems and Electoral Alignments in East Central
Europe: Central European University.
-
Toka, Gabor. 1993. Party Systems and Electoral Alignments in East Central
Europe: Central European University.
-
Toka, Gabor. 1994. Party Systems and Electoral Alignments in East Central
Europe: Central European University.
-
Toka, Gabor. 1996. Party Systems and Electoral Alignments in East Central
Europe: Central European University.
-
Toka, Gabor. 1997. Political Parties and Democratic Consolidaiton in East
Central Europe: Centre for the Study of Public Policy, University of Strathclyde.
-
Turnovec, Frantisek. 1995. The Political System and Economic Transition.
In The Czech Republic and Economic Transition in Eastern Europe,
edited by J. Svejnar. London: Academic Press.
-
Whitefield, Stephen, and Geoffrey Evans. 1995. Mass Ideology and Political
Competition in Post-Communist Societies. Paper read at Workshop on Public
Opinion and Party Formation in Post-Communist and Post Authoritarian Democracies,
at Duke University.
Tables
Table 1.
Correlation between party elite and party voter assessments of parties
in the Czech Republic, 1994
| Assessing Party |
Source of Assessment |
Sample size |
Party Assessed |
Correlation of elite/voter assessments (r) |
| ODS |
ODA |
KDU- CSL |
CSSD |
KSCM |
SPR-RSC |
|
| ODS |
Party Elites |
N=11 |
0.92 |
0.71 |
0.68 |
0.05 |
0.03 |
0.00 |
0.91** |
| Party Voters |
N=391 |
0.92 |
0.74 |
0.50 |
0.42 |
0.12 |
0.12 |
|
|
|
|
|
|
|
|
| ODA |
Party Elites |
N=19 |
0.68 |
0.93 |
0.55 |
0.20 |
0.07 |
0.07 |
0.93** |
| Party Voters |
N=123 |
0.72 |
0.89 |
0.45 |
0.49 |
0.16 |
0.11 |
|
|
|
|
|
|
|
|
| KDU-CSL |
Party Elites |
N=19 |
0.57 |
0.71 |
0.95 |
0.12 |
0.07 |
0.07 |
0.84** |
| Party Voters |
N=55 |
0.56 |
0.59 |
0.93 |
0.59 |
0.21 |
0.13 |
|
|
|
|
|
|
|
|
| CSSD |
Party Elites |
N=20 |
0.34 |
0.48 |
0.32 |
0.94 |
0.26 |
0.11 |
1.00** |
| Party Voters |
N=187 |
0.41 |
0.52 |
0.39 |
0.91 |
0.38 |
0.26 |
|
|
|
|
|
|
|
|
| KSCM |
Party Elites |
N=21 |
0.18 |
0.25 |
0.12 |
0.66 |
0.98 |
0.38 |
0.98** |
| Party Voters |
N=90 |
0.18 |
0.34 |
0.25 |
0.67 |
0.93 |
0.33 |
|
|
|
|
|
|
|
|
| SPR-RSC |
Party Elite |
N=0 |
n/a |
n/a |
n/a |
n/a |
n/a |
n/a |
n/a |
| Party Voters |
N=42 |
0.29 |
0.39 |
0.28 |
0.52 |
0.32 |
0.90 |
Source: Toka 1994 and Kitschelt, Brokl and Mansfeldova 1994
*p<0.1. **p<0.01
Back
Table 2.
Correlation between party elite and party voter assessments of parties
in the Czech Republic, 1996
| Assessing Party |
Source of Assessment |
Sample size |
Party Assessed |
Correlation of elite/voter assessments (r) |
| ODS |
ODA |
KDU- CSL |
CSSD |
KSCM |
SPR-RSC |
| ODS |
Party Elites |
N=30 |
0.95 |
0.75 |
0.59 |
0.22 |
0.02 |
0.01 |
0.99** |
| Party Voters |
N=444 |
0.91 |
0.71 |
0.54 |
0.40 |
0.09 |
0.16 |
|
|
|
|
|
| ODA |
Party Elites |
N=0 |
n/a |
n/a |
n/a |
n/a |
n/a |
n/a |
n/a |
| Party Voters |
N=94 |
0.67 |
0.88 |
0.55 |
0.46 |
0.19 |
0.178 |
|
|
|
|
|
| KDU-CSL |
Party Elites |
N=9 |
0.62 |
0.69 |
0.94 |
0.31 |
0.00 |
0.01 |
0.96** |
| Party Voters |
N=117 |
0.56 |
0.62 |
0.93 |
0.53 |
0.15 |
0.20 |
|
|
|
| CSSD |
Party Elites |
N=20 |
0.25 |
0.43 |
0.59 |
0.89 |
0.26 |
0.22 |
0.93** |
| Party Voters |
N=326 |
0.35 |
0.50 |
0.46 |
0.89 |
0.39 |
0.35 |
|
|
|
|
|
| KSCM |
Party Elites |
N=9 |
0.22 |
0.33 |
0.41 |
0.56 |
0.89 |
0.34 |
0.94** |
| Party Voters |
N=184 |
0.18 |
0.50 |
0.37 |
0.67 |
0.92 |
0.37 |
|
|
|
|
| SPR-RSC |
Party Elite |
N=5 |
0.09 |
0.04 |
0.20 |
0.32 |
0.46 |
0.98 |
0.79** |
| Party Voters |
N=86 |
0.29 |
0.41 |
0.38 |
0.55 |
0.24 |
0.92 |
|
Source: Toka 1996 and Krause 1996c
*p<0.1. **p<0.01
Back
Table 3.
Best fitting one-dimensional party arrays for Slovakia and the Czech
Republic, 1992-1996
| Counry |
Year |
Source |
Placement on one dimensional ranking |
Percentage of pairwise comparisons explained by rank order |
Correlation between voter/elite positions and rank order |
| Slovakia |
1992 |
Voters |
MK |
KDH |
SDL |
HZDS |
SNS |
|
|
95% |
0.78** |
| 1993 |
Voters |
MK |
KDH |
SDL |
SNS |
HZDS |
|
|
98% |
0.86** |
| 1994 |
Voters |
MK |
KDH |
DU |
SDL |
ZRS |
SNS |
HZDS |
95% |
0.84** |
| 1996 |
Voters |
MK |
KDH |
DU |
SDL |
ZRS |
SNS |
HZDS |
94% |
0.86** |
| Czech |
1992 |
Voters |
ODS |
ODA |
KD U-CSL |
CSSD |
KSCM |
SPR- RSC |
|
86% |
0.79** |
| 1993 |
Voters |
KDU- CSL |
ODS |
ODA |
CSSD |
KSCM |
SPR- RSC |
|
94% |
0.82** |
| 1994 |
Elites |
KDU- CSL |
ODS |
ODA |
CSSD |
KSCM |
SPR- RSC |
|
96% |
0.82** |
| 1994 |
Voters |
KDU- CSL |
ODS |
ODA |
CSSD |
KSCM |
SPR- RSC |
|
93% |
0.79** |
| 1996 |
Voters |
ODS |
KDU- CSL |
ODA |
CSSD |
KSCM |
SPR-RSC |
|
87% |
0.80** |
| ODS |
ODA |
KDU- CSL |
CSSD |
KSCM |
SPR- RSC |
89% |
0.80** |
| 1996 |
Elites |
ODS |
ODA |
KDU- CSL |
CSSD |
KSCM |
SPR- RSC |
|
95% |
0.84** |
| Calculated from factor data |
1993 |
Kitschelt jury |
ODS |
ODA |
KDU-CSL |
CSSD |
SPR- RSC |
KSCM |
|
92% |
0.80** |
| 1994 |
Brokl |
KDU- CSL |
ODS |
ODA |
SPR- RSC |
CSSD |
KSCM |
|
93% |
0.84** |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
*p<0.1. **p<0.01
Back
Table 4.
Factor scores of voters by party for Slovakia, 1992 to 1996
| Year |
Factor |
Party mean factor score |
Range |
|
|
|
|
|
|
MK |
KDH |
DU |
SDL |
ZRS |
HZDS |
SNS |
| 1992 |
LIB |
-0.4 |
-0.1 |
|
0.0 |
|
0.0 |
-0.1 |
0.3 |
| REL |
0.3 |
1.2 |
|
-0.4 |
|
-0.1 |
0.0 |
1.6 |
| NAT |
-0.6 |
-0.2 |
|
-0.4 |
|
0.2 |
0.8 |
1.4 |
| ECON |
0.4 |
0.3 |
|
-0.5 |
|
0.0 |
0.1 |
0.9 |
| 1993 |
LIB |
-0.4 |
-0.1 |
|
-0.2 |
|
0.2 |
0.2 |
0.6 |
| REL |
0.5 |
1.0 |
|
-0.5 |
|
-0.1 |
0.0 |
1.5 |
| NAT |
-0.8 |
-0.2 |
|
-0.2 |
|
0.4 |
0.8 |
1.7 |
| ECON |
-0.4 |
-0.3 |
|
0.3 |
|
-0.1 |
-0.2 |
0.7 |
| 1994 |
LIB |
-0.4 |
0.1 |
0.4 |
-0.1 |
-0.4 |
-0.2 |
0.3 |
0.8 |
| REL |
0.3 |
1.1 |
-0.2 |
-0.5 |
-0.1 |
-0.1 |
-0.1 |
1.6 |
| NAT |
-0.6 |
-0.3 |
-0.4 |
-0.3 |
0.0 |
0. 5 |
0.9 |
1.5 |
| ECON |
-0.6 |
-0.3 |
-0.3 |
0.2 |
0.4 |
0. 2 |
-0.4 |
1.0 |
| 1996 |
LIB |
-0.2 |
0.4 |
0.5 |
-0.2 |
-0.5 |
-0.3 |
0.0 |
1.0 |
| REL |
0.6 |
1.1 |
-0.1 |
-0.5 |
-0.1 |
-0.1 |
-0.1 |
1.6 |
| NAT |
-0.9 |
-0.3 |
-0.5 |
-0.2 |
0.2 |
0. 4 |
0.8 |
1.7 |
| ECON |
-0.4 |
-0.2 |
0.0 |
0.2 |
0.2 |
0. 0 |
-0.1 |
0.6 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Table 5.
Factor scores of voters by party for the Czech Republic, 1992 to 1996
| Year |
Factor |
Party mean factor score |
Range |
| ODS |
ODA |
KDU-CSL |
CSS |
KSC |
SPR- RSC |
| 1992 |
LIB |
0.4 |
0.7 |
0.4 |
-0.3 |
- 0.9 |
-0.4 |
1.6 |
| REL |
0.1 |
-0.5 |
1.2 |
-0.2 |
-0.4 |
0.2 |
1.7 |
| ECON |
-0.4 |
-0.3 |
0.2 |
0.3 |
0.6 |
0. 1 |
1.0 |
| 1993 |
LIB |
0.6 |
0.3 |
0.1 |
-0.5 |
- 1.2 |
-0.4 |
1.8 |
| REL |
0.0 |
-0.1 |
1.6 |
-0.3 |
-0.2 |
-0.3 |
1.8 |
| NAT |
0.0 |
-0.1 |
0.0 |
0.0 |
0.0 |
0.5 |
0.6 |
| LIB2 |
0.2 |
0.2 |
-0.2 |
-0.2 |
-0.3 |
-0. 5 |
0.7 |
| 1994 |
LIB |
0.6 |
0.4 |
0.3 |
-0.4 |
- 0.7 |
-0.3 |
1.3 |
| REL |
0.1 |
0.3 |
-1.5 |
0.0 |
0.3 |
0.0 |
1.8 |
| ECON |
-0.3 |
-0.2 |
0.1 |
0.1 |
0.2 |
0. 5 |
0.8 |
| 1996 |
LIB |
0.6 |
0.3 |
0.2 |
-0.2 |
- 0.6 |
-0.5 |
1.2 |
| REL |
-0.1 |
-0.2 |
1.3 |
-0.2 |
-0.1 |
0.0 |
1.5 |
| ECON |
-0.5 |
-0.2 |
0.2 |
0.2 |
0.6 |
-0. 1 |
1.1 |
| NAT |
-0.1 |
-0.2 |
-0.2 |
0.0 |
0.1 |
0.5 |
0.6 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Table 6.
Pairwise correlations between party factor scores for Slovakia, 1992-1996
| Factor Pairing |
Correlations (without signs) between party scores on paired
factors |
|
1992 |
1993 |
1994 |
1996 |
| LIB-REL |
0.36 |
0.33 |
0.07 |
0.38 |
| LIB-NAT |
0.47 |
0.95* |
0.21 |
0.31 |
| LIB-ECON |
0.78 |
0.16 |
0.4 |
0.36 |
| REL-NAT |
0.2 |
0.42 |
0.27 |
0.46 |
| REL-ECON |
0.7 |
0.74 |
0.45 |
0.74 |
| NAT-ECON |
0.07 |
0.17 |
0.21 |
0.42 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
*p<0.1. **p<0.01
Back
Table 7.
Pairwise correlations between party factor scores for the Czech Republic,
1992-1996
| Factor Pairing |
Correlations (without signs) between party scores on paired
factors |
|
|
1992 |
1993 |
1994 |
1996 |
| LIB-REL |
0.25 |
0.3 |
0.29 |
0.23 |
| LIB-NAT |
|
0.24 |
|
0.76 |
| LIB-ECON |
0.84* |
|
0.73 |
0.71 |
| REL-NAT |
|
0.24 |
|
0.33 |
| REL-ECON |
0.06 |
|
0.18 |
0.17 |
| NAT-ECON |
|
|
|
0.11 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
*p<0.1. **p<0.01
Back
Table 8.
Correlation between party factor scores and party sympathy measures
for Slovakia, 1992- 1996
| Factor |
Correlations (without signs) of distances between factor
scores and sympathy scores for all parties |
|
|
|
1992 |
1993 |
1994 |
1996 |
| LIB |
0.65** |
0.80** |
0.33* |
0.43** |
| REL |
0.60** |
0.63** |
0.35* |
0.46** |
| NAT |
0.56** |
0.76** |
0.67** |
0.80** |
| ECON |
0.56** |
0.43* |
0.37** |
0.49** |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
*p<0.1. **p<0.01
Back
Table 9.
Correlation between party factor scores and party sympathy measures
for the Czech Republic, 1992-1996
| Factor |
Correlations (without signs) of distances between factor
scores and sympathy scores for all parties |
|
1992 |
1993 |
1994 |
1996 |
| LIB |
0.79** |
0.73** |
0.76** |
0.79** |
| REL |
0.18 |
0.31* |
0.26 |
0.17 |
| ECON |
0.67** |
|
0.67** |
0.62** |
| NAT |
|
0.47** |
|
0.68** |
| LIB2 |
|
0.67** |
|
|
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
*p<0.1. **p<0.01
Back
Table 10.
Correlation between party factor scores, including AUTH, and party
sympathy measures for Slovakia, 1992-1996
| Party |
Correlation between party sympathy score and difference
in factor score for all other parties |
|
AUTH |
LIB |
REL |
NAT |
ECON |
| Overall |
0.72** |
0.33* |
0.35* |
0.67** |
0.37 ** |
| MKDM (MK) |
0.84** |
0.35* |
0.15 |
0.96** |
0.28* |
| KDH |
0.74** |
0.61** |
0.63** |
0.65** |
0.6 6** |
| DU/APR/DEUS |
0.82** |
0.61** |
-0.03 |
0.62** |
0.38** |
| SDL |
0.27* |
0.18 |
0.57** |
0.31* |
0.56** |
| ZRS |
0.73** |
0.09 |
0.49** |
0.3* |
0.74** |
| HZDS |
0.9** |
0.35* |
0.54** |
0.98** |
0.62 ** |
| SNS |
0.81** |
0.32* |
0.54** |
0.94** |
-0.12 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
*p<0.1. **p<0.01
Back
Table 11.
Correlations between party factor scores, including AUTH for Slovakia,
1994
|
Correlation between party factor scores |
|
LIB |
REL |
NAT |
ECON |
AUTH |
| LIB |
|
0.07 |
0.21 |
0.40* |
0.30* |
| REL |
0.07 |
|
0.27 |
0.45** |
0.18* |
| NAT |
0.21 |
0.27 |
|
0.21 |
0.82** |
| ECON |
0.40* |
0.45** |
0.21 |
|
0.38* |
| AUTH |
0.30 |
0.18 |
0.82** |
0.38* |
|
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
*p<0.1. **p<0.01
Back
Table 12.
Party system size according to the Sartori method for Slovakia and
the Czech Republic, 1990 to 1998
|
1990 |
1991 |
1992 |
1993 |
1994 |
1995 |
1996 |
1997 |
1998 |
| Slovakia |
8 |
8 |
6 |
6 |
8 |
8 |
8 |
8 |
8 |
| Czech Republic |
4 |
7 |
8 |
7 |
6 |
6 |
6 |
6 |
7 |
Back
Table 13.
Party system size according to the Taagepera and Laakso method for
Slovakia and the Czech Republic, 1990 to 1998
|
1990 |
1991 |
1992 |
1993 |
1994 |
1995 |
1996 |
1997 |
1998 |
| Slovakia |
5 |
7 |
3 |
4 |
5 |
5 |
5 |
5 |
5 |
| Czech Republic |
2 |
7 |
6 |
6 |
7 |
6 |
4 |
4 |
6 |
Back
Table 14.
Relative percentages of parliamentary seats of largest and second largest
parties in Slovakia and the Czech Republic, 1990 to 1998
|
|
1990 |
1991 |
1992 |
1993 |
1994 |
1995 |
1996 |
1997 |
1998 |
| Slovakia |
1st party |
32 |
21 |
49 |
43 |
36 |
41 |
41 |
41 |
41 |
| 2nd party |
21 |
15 |
19 |
19 |
19 |
12 |
12 |
12 |
11 |
| Difference |
11 |
6 |
30 |
24 |
17 |
29 |
29 |
29 |
30 |
| Ratio |
1.5 |
1.4 |
2.6 |
2.3 |
1.9 |
3.4 |
3.4 |
3.4 |
3.7 |
| Czech Republic |
1st party |
64 |
22 |
33 |
33 |
33 |
33 |
34 |
35 |
29 |
| 2nd party |
16 |
21 |
18 |
18 |
13 |
12 |
31 |
29 |
19 |
| Difference |
48 |
1 |
15 |
15 |
20 |
21 |
3 |
6 |
10 |
| Ratio |
4.0 |
1.0 |
1.8 |
1.8 |
2.5 |
2.8 |
1.1 |
1.2 |
1.5 |
Back
Table 15.
Ratio of "decisive vote opportunities" of largest party to second largest
party in Slovakia and the Czech Republic, 1990 to 1998
|
1990 |
1991 |
1992 |
1993 |
1994 |
1995 |
1996 |
1997 |
1998 |
| Slovakia |
2.7 |
1.3 |
15.0 |
9.6 |
4.1 |
13.2 |
13.2 |
13.2 |
15.0 |
| Czech Republic |
- |
1.1 |
3.4 |
3.9 |
5.4 |
5.0 |
1.3 |
1.7 |
2.0 |
Back
Table 16.
Measures of electoral volatility in Slovakia and the Czech Republic,
1990 to 1998
| Country |
Method |
Electoral Period 1 |
Electoral Period 2 |
| Slovakia |
|
1990-1992 |
1992-1996 |
| Toka |
Strict |
67.3 |
31.4 |
| Relaxed |
22.0 |
29.3 |
| Mainwaring |
35.2 |
30.1 |
|
| Slovakia |
|
1990-1992 |
1992-1996 |
| Toka |
Strict |
55.7 |
23.8 |
| Relaxed |
23.1 |
14.1 |
| Mainwaring |
25.1 |
27.9 |
Source: Mainwaring and Scully 1996 and Toka 1997
Back
Table 17.
Measures of electoral volatility and intra-parliamentary volatility
of parliamentary seats in Slovakia and the Czech Republic, 1990 to 1998
| Country |
Volatility type |
Term/Year
|
|
| Slovakia |
Electoral volatility |
1990-1992 |
1992-1994 |
1994-1998 |
| 0.24 |
0.21 |
n/a |
| Intra-Parliamentary volatility |
1990-1992 |
1992-1993 |
1993-1994 |
1994-1995 |
1995-1996 |
1996-1997 |
1997-1998 |
| 0.21 |
0.07 |
0.17 |
0.01 |
0.00 |
0.00 |
0.05 |
|
|
|
|
|
|
|
|
|
|
|
| Czech Republic |
Electoral volatility |
1990-1992 |
1992-1996 |
1996-1998 |
| 0.365 |
0.33 |
n/a |
| Intra-Parliamentary volatility |
1990-1992 |
1992-1993 |
1993-1994 |
1994-1995 |
1995-1996 |
1996-1997 |
1997-1998 |
| 0.50 |
0.07 |
0.13 |
0.10 |
0.10 |
0.01 |
0.16 |
Back
Table 18.
Characteristics of actual and hypothetical coalitions for Slovakia,
1992-1996
|
|
Coalition |
Coalition with next best coalition partners |
Outsider Coalitions |
Coalitionable Outsider Coalitions |
| 1992 |
Composition |
HZDS |
HZDS-SNS |
HZDS- SDL |
SNS-KDH-SDL-MK |
- |
- |
- |
| Plurality (Seats-75) |
-1 |
14 |
28 |
1 |
- |
- |
- |
| Range |
- |
1.6 |
1.9 |
5.3 |
- |
- |
- |
| Size |
1 |
2 |
2 |
4 |
- |
- |
- |
| Connected |
- |
Yes |
Yes |
No |
- |
- |
- |
| Coalitionable |
- |
Yes |
Yes |
No |
- |
- |
- |
| 1993 |
Composition |
HZDS-SNS |
HZDS-SDL |
H ZDS-KDH |
SNS-KDH-SDL-MK |
AD-KDH-SDL-MK |
AD-KDH-SD L |
| Plurality |
6 |
20 |
9 |
1 |
-6 |
-20 |
| Range |
1.4 |
2.6 |
3.4 |
4.8 |
4.6 (min.) |
3.6 (min.) |
| Size |
2 |
2 |
2 |
4 |
4 |
3 |
| Connected |
Yes |
Yes |
No |
No |
Yes |
Yes |
| Coalitionable |
Yes |
Yes |
Yes |
No |
No |
Yes |
| 1994 |
Composition |
HZDS-SNS-ZRS |
HZDS-SNS-SDL |
HZDS-ZRS-SDL |
ZRS-SDL-DU-KDH-MK |
SNS-SDL-DU-KDH-M K |
SNS-ZRS-SDL-DU-KDH |
| Plurality |
83 |
87 |
92 |
79 |
76 |
71 |
| Range |
1.8 |
3.8 |
3.6 |
4.8 |
5.3 |
4.4< /td> |
| Size |
3 |
3 |
3 |
5 |
5 |
5 |
| Connected |
Yes |
No |
Yes |
Yes |
No |
No |
| Coalitionable |
Yes |
Yes |
Yes |
No |
No |
Yes |
| 1996 |
Composition |
HZDS-SNS-ZRS |
HZDS-SNS-SDL |
HZDS-ZRS-SDL |
ZRS-SDL-DU-KDH-MK |
SNS-SDL-DU-KDH-M K |
SNS-ZRS-SDL-DU-KDH |
| Plurality |
82 |
86 |
91 |
77 |
75 |
69 |
| Range |
1.9 |
2.9 |
2.9 |
4.8 |
5.0 |
4.5 |
| Size |
3 |
3 |
3 |
5 |
5 |
5 |
| Connected |
Yes |
No |
Yes |
Yes |
No |
No |
| Coalitionable |
Yes |
Yes |
Yes |
No |
No |
Yes |
Back
Table 19.
Characteristics of actual and hypothetical coalitions for the Czech
Republic, 1992-1996
|
|
Coalition |
Coalition with next best coalition partners |
Outsider Coalitions |
Coalitionable Outsider |
| 1992 |
Composition |
ODS-ODA-KDU- CSL |
ODS-ODA-CSSD |
ODS-KDU-CSL-CSSD |
KDU- CSL-CSSD-KSCM
-SPR-RSC-Others |
ODA-CSSD-KSCM-SPR- RSC-Others |
ODA-KDU-CSL-CSSD-Others |
| Plurality (Seats-100) |
5 |
6 |
5 |
10 (max.) |
9 (max) |
-25 (max.) |
| Range |
1.4 |
3.2 |
3.2 |
6.0 (min.) |
6.0 (min.) |
3.3 (min.) |
| Size |
3 |
3 |
3 |
6 |
6 |
5 |
| Connected |
Yes |
No |
No |
Yes |
No |
Yes |
| Coalitionable |
Yes |
Yes |
Yes |
No |
No |
Yes |
| 1993 |
Composition |
ODS-ODA-KDU- CSL |
ODS-ODA-CSSD |
ODS-KDU-CSL-CSSD |
KDU- CSL-CSSD-KSCM-SPR-RSC-OthersODA-CSSD-KSCM-SPR- RSC-Others |
ODA-KDU-CSL-CSSD-Others |
| Plurality |
|
|
|
|
|
|
| Range |
2.8 |
3.2 |
3.2 |
4.7 (min.) |
4.7 (min.) |
3.3 (min.) |
| Size |
3 |
3 |
3 |
9 |
9 |
6 |
| Connected |
Yes |
No |
No |
Yes |
No |
Yes |
| Coalitionable |
Yes |
Yes |
Yes |
No |
No |
No |
| 1994 |
Composition |
ODS-ODA-KDU- CSL |
ODS-ODA-CSSD |
ODS-KDU-CSL-CSSD |
KDU- CSL-CSSD-KSCM
-SPR-RSC-Others |
ODA-CSSD-KSCM-SPR- RSC-Others |
ODA-KDU-CSL-CSSD-Others |
| Plurality |
8 |
13 |
2 |
8 (max.) |
9 (max.) |
-15 (max.) |
| Range |
2.5 |
3.0 |
3.1 |
4.8 (min.) |
4.6 (min.) |
3.1 (min.) |
| Size |
3 |
3 |
3 |
8 |
8 |
5 |
| Connected |
Yes |
No |
No |
Yes |
No |
Yes |
| Coalitionable |
Yes |
Yes |
Yes |
No |
No |
No |
| 1996 |
Composition |
ODS-ODA-KDU- CSL |
ODS-ODA-CSSD |
ODS-KDU-CSL-CSSD |
KDU- CSL-CSSD-
KSCM-SPR-RSC |
ODA-CSSD-KSCM-SPR-RSC |
ODA-KDU- CSL-CSSD |
| Plurality |
-1 |
42 |
47 |
19 |
14 |
-8 |
| Range |
2.3 |
3.2 |
3.2 |
4.6 |
4.2 |
4.2 |
| Size |
3 |
3 |
3 |
4 |
4 |
3 |
| Connected |
Yes |
No |
No |
Yes |
No |
Yes |
| Coalitionable |
Yes |
Yes |
Yes |
Nos |
No |
No |
Back
Figures
Figure 1.
Placement of parties in Slovakia on NAT and other dimensions, 1994 
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Figure 2.
Placement of parties in Slovakia on NAT and other dimensions, 1996 
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Figure 3.
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Appendix
Table A1.
Factor analysis for Slovakia, 1992
| Code |
Question |
Factor 1 |
Factor 2 |
Factor 3 |
Factor 4 |
|
|
|
|
|
|
| Q16F |
Giving the former state-owned companies into private hands is going
to help very much in solving the economic problems of our country. |
0.715 |
|
|
|
| Q16G |
Unprofitable factories and mines should be closed down immediately
even if this leads to unemployment. |
0.604 |
|
|
|
| Q16C |
It should be the government's responsibility to provide jobs for everyone
who wants one. |
-0.529 |
|
0.321 |
|
| Q16R |
It would be better if former owners do not receive compensation |
-0.455 |
|
|
|
|
|
|
|
|
|
| Q16N |
A woman should be allowed to have an abortion in the early weeks of
pregnancy if she decides so. |
|
0.801 |
|
|
| Q16H |
Politicians who do not believe in God should not perform public functions. |
|
-0.69 |
0.394 |
|
| Q16P |
The church has too much influence in our country |
|
0.469 |
0.391 |
|
|
|
|
|
|
|
| Q16O |
In the case of a politician I prefer a strong patriot to an expert. |
|
|
0.676 |
-0.372 |
| Q16D |
It is harmful for the economy if the government tries to reduce income
differences between rich and poor. |
|
|
0.558 |
|
| Q16M |
Politicians should care more about rising crime and deteriorating morality
than about individual freedom and human rights |
-0.332 |
|
0.407 |
|
|
|
|
|
|
|
| Q16I |
Nationalism is harmful for the development of our country |
|
|
|
0.766 |
| Q16Q |
It would be better if the Czech and Slovak republics were not separated. |
|
|
|
0.703 |
| Statistics |
|
|
Factor |
Eigenvalue |
% of Variance |
Cumulative % |
|
|
1.0 |
1.735 |
14.5 |
14.5 |
|
|
2.0 |
1.564 |
13.0 |
27.5 |
|
|
3.0 |
1.348 |
11.2 |
38.7 |
|
|
4.0 |
1.191 |
9.9 |
48.7 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Table A2.
Factor analysis for Slovakia, 1993
| Code |
Question |
Factor 1 |
Factor 2 |
Factor 3 |
Factor 4 |
| Q16N |
Allow abortions |
0.785 |
|
|
|
| Q16H |
Atheists unfit for service |
-0.711 |
0.309 |
|
|
| Q16P |
Church too influential |
0.707 |
|
|
|
| Q16C |
Jobs for all |
|
0.688 |
|
|
| Q16M |
Crime and morals key |
|
0.581 |
|
|
| Q16R |
Restitution bad |
0.401 |
0.416 |
-0.37 |
|
| Q16F |
Privatization helpful |
|
-0.343 |
0.671 |
|
| Q16G |
Close unprofitable firms |
|
|
0.642 |
|
| Q16D |
Less inequailty harmful |
|
0.306 |
0.597 |
|
| Q16Q |
Czechoslovak split bad |
|
|
|
0.734 |
| Q16I |
Nationalism harmful |
|
|
|
0.72 |
| Q16O |
Patriots not experts |
|
0.425 |
|
-0.541 |
| Statistics |
|
|
Factor |
Eigenvalue |
% of Variance |
Cumulative % |
|
|
1.0 |
2.177 |
18.1 |
18.1 |
|
|
2.0 |
1.61 |
13.4 |
31.6 |
|
|
3.0 |
1.383 |
11.5 |
43.1 |
|
|
4.0 |
1.184 |
9.9 |
52.9 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Table A3.
Factor analysis for Slovakia, 1994
| Code |
Question |
Factor 1 |
Factor 2 |
Factor 3 |
Factor 4 |
| Q16N |
Allow abortions |
0.795 |
|
|
|
| Q16H |
Atheists unfit for service |
-0.695 |
|
|
|
| Q16P |
Church too influential |
0.638 |
|
0.346 |
|
| Q16I |
Nationalism harmful |
|
0.757 |
|
|
| Q16Q |
Czechoslovak split bad |
|
0.73 |
|
|
| Q16O |
Patriots not experts |
|
-0.65 |
0.349 |
|
| Q16M |
Crime and morals key |
|
|
0.678 |
|
| Q16C |
Jobs for all |
|
|
0.613 |
|
| Q16R |
Restitution bad |
0.359 |
|
0.435 |
|
| Q16G |
Close unprofitable firms |
|
|
|
0.702 |
| Q16D |
Less inequailty harmful |
|
|
|
0.651 |
| Q16F |
Privatization helpful |
|
|
-0.375 |
0.599 |
| Statistics |
|
|
Factor |
Eigenvalue |
% of Variance |
Cumulative % |
|
|
1.0 |
2.065 |
17.2 |
17.2 |
|
|
2.0 |
1.719 |
14.3 |
31.5 |
|
|
3.0 |
1.382 |
11.5 |
43.0 |
|
|
4.0 |
1.31 |
10.9 |
54.0 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Table A4.
Factor analysis for Slovakia, 1996
| Variable |
Question |
Factor 1 |
Factor 2 |
Factor 3 |
Factor 4 |
| Q16M |
Crime and morals key |
0.592 |
|
|
|
| Q16C |
Jobs for all |
0.574 |
|
|
|
| Q16R |
Restitution bad |
0.504 |
|
|
|
| Q16N |
Allow abortions |
|
0.789 |
|
|
| Q16H |
Atheists unfit for service |
|
-0.689 |
|
|
| Q16P |
Church too influential |
0.452 |
0.583 |
|
|
| Q16Q |
Czechoslovak split bad |
|
|
0.733 |
|
| Q16I |
Nationalism harmful |
|
|
0.712 |
|
| Q16O |
Patriots not experts |
0.384 |
|
-0.59 |
|
| Q16F |
Privatization helpful |
|
|
|
0.72 |
| Q16G |
Close unprofitable firms |
-0.387 |
|
|
0.638 |
| Q16D |
Less inequailty harmful |
|
|
|
0.538 |
| Statistics |
|
|
Factor |
Eigenvalue |
% of Variance |
Cumulative % |
|
|
1.0 |
1.847 |
15.4 |
15.4 |
|
|
2.0 |
1.6 |
13.3 |
28.7 |
|
|
3.0 |
1.491 |
12.4 |
41.2 |
|
|
4.0 |
1.162 |
9.7 |
50.8 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Table A5.
Factor analysis for the Czech Republic, 1992
| Code |
Question |
Factor 1 |
Factor 2 |
Factor 3 |
|
| Q16P |
The church has too much influence in our country |
0.629 |
|
|
|
| Q16M |
Politicians should care more about rising crime and deteriorating morality
than about individual freedom and human rights |
0.578 |
|
|
|
| Q16R |
It would be better if former owners do not receive compensation |
0.576 |
|
|
|
| Q16C |
It should be the government's responsibility to provide jobs for everyone
who wants one. |
0.533 |
|
|
|
| Q16Q |
It would be better if the Czech and Slovak republics were not separated. |
0.476 |
-0.416 |
|
|
| Q16I |
Nationalism is harmful for the development of our country |
|
|
|
|
| Q16D |
It is harmful for the economy if the government tries to reduce income
differences between rich and poor. |
|
0.666 |
|
|
| Q16F |
Giving the former state-owned companies into private hands is going
to help very much in solving the economic problems of our country. |
-0.393 |
0.619 |
|
|
| Q16G |
Unprofitable factories and mines should be closed down immediately
even if this leads to unemployment. |
-0.358 |
0.492 |
|
|
| Q16H |
Politicians who do not believe in God should not perform public functions. |
|
|
0.757 |
|
| Q16N |
A woman should be allowed to have an abortion in the early weeks of
pregnancy if she decides so. |
|
|
-0.732 |
|
| Q16O |
In the case of a politician I prefer a strong patriot to an expert. |
0.349 |
|
0.499 |
|
| Statistics |
|
|
Factor |
Eigenvalue |
% Variance |
Cumulative % |
|
|
1.0 |
2.426 |
20.2 |
20.2 |
|
|
2.0 |
1.483 |
12.4 |
32.6 |
|
|
3.0 |
1.133 |
9.4 |
42.0 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Table A6.
Factor analysis for the Czech Republic, 1993
| Code |
Question |
Factor 1 |
Factor 2 |
Factor 3 |
Factor 4 |
| Q16F |
Privatization helpful |
-0.707 |
|
|
|
| Q16G |
Close unprofitable firms |
-0.672 |
|
|
|
| Q16D |
Less inequailty harmful |
-0.632 |
|
0.355 |
|
| Q16R |
Restitution bad |
0.581 |
|
0.331 |
|
| Q16Q |
Czechoslovak split bad |
0.542 |
|
|
|
| Q16C |
Jobs for all |
0.537 |
|
0.322 |
|
| Q16N |
Allow abortions |
|
0.779 |
|
|
| Q16H |
Atheists unfit for service |
|
-0.691 |
0.317 |
|
| Q16P |
Church too influential |
|
0.573 |
0.392 |
|
| Q16O |
Patriots not experts |
|
-0.301 |
0.63 |
|
| Q16M |
Crime and morals key |
0.35 |
|
0.568 |
|
| Q16I |
Nationalism harmful |
|
|
|
0.928 |
|
|
|
|
Factor |
Eigenvalue |
% Variance |
Cumulative % |
|
|
1.0 |
2.789 |
23.2 |
23.2 |
|
|
2.0 |
1.577 |
13.1 |
36.4 |
|
|
3.0 |
1.036 |
8.6 |
45.0 |
|
|
4.0 |
1.016 |
8.5 |
53.5 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Table A7.
Factor analysis for the Czech Republic, 1994
| Code |
Question |
Factor 1 |
Factor 2 |
Factor 3 |
|
| Q16R |
Restitution bad |
0.729 |
|
|
|
| Q16Q |
Czechoslovak split bad |
0.613 |
|
|
|
| Q16C |
Jobs for all |
0.607 |
|
|
|
| Q16M |
Crime and morals key |
0.607 |
|
|
|
| Q16F |
Privatization helpful |
-0.573 |
|
0.42 |
|
| Q16P |
Church too influential |
0.555 |
|
|
|
| Q16H |
Atheists unfit for service |
|
0.785 |
|
|
| Q16N |
Allow abortions |
|
-0.613 |
|
|
| Q16O |
Patriots not experts |
|
0.597 |
|
|
| Q16I |
Nationalism harmful |
|
|
0.689 |
|
| Q16D |
Less inequailty harmful |
|
|
0.603 |
|
| Q16G |
Close unprofitable firms |
-0.408 |
|
0.491 |
|
| Statistics |
|
|
Factor |
Eigenvalue |
% of Variance |
Cumulative % |
|
|
1.0 |
2.863 |
23.9 |
23.9 |
|
|
2.0 |
1.536 |
12.8 |
36.7 |
|
|
3.0 |
1.114 |
9.3 |
45.9 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Table A8.
Factor analysis for the Czech Republic, 1996
| Code |
Question |
Factor 1 |
Factor 2 |
Factor 3 |
Factor 4 |
| Q16R |
Restitution bad |
0.646 |
|
|
|
| Q16Q |
Czechoslovak split bad |
0.62 |
|
|
|
| Q16M |
Crime and morals key |
0.614 |
|
|
|
| Q16C |
Jobs for all |
0.583 |
|
-0.341 |
|
| Q16P |
Church too influential |
0.51 |
0.432 |
|
-0.366 |
| Q16O |
Patriots not experts |
0.485 |
-0.436 |
|
|
| Q16H |
Atheists unfit for service |
|
-0.758 |
|
|
| Q16N |
Allow abortions |
|
0.713 |
|
|
| Q16D |
Less inequailty harmful |
|
|
0.726 |
|
| Q16F |
Privatization helpful |
-0.516 |
|
0.55 |
|
| Q16G |
Close unprofitable firms |
-0.378 |
|
0.538 |
|
| Q16I |
Nationalism harmful |
|
|
|
0.91 |
| Statistics |
|
|
Factor |
Eigenvalue |
% of Variance |
Cumulative % |
|
|
1.0 |
2.931 |
24.4 |
24.4 |
|
|
2.0 |
1.513 |
12.6 |
37.0 |
|
|
3.0 |
1.046 |
8.7 |
45.8 |
|
|
4.0 |
1.036 |
8.6 |
54.4 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
Back
Table A9.
Correlation between party sympathy and factor positions by party in
Slovakia, 1992 to 1996
| Party |
Factor |
Correlation between party sympathy score and difference
in factor score for all other parties |
MKDH or
MKM-ES |
LIB |
0.52 |
0.97** |
0.35 |
-0.16 |
| NAT |
0.43 |
0.46 |
0.15 |
0.51 |
| NAT |
0.88* |
0.97** |
0.96** |
0.98** |
|
| ECON |
0.82* |
0.25 |
0.28 |
0.49 |
| KDH |
LIB |
0.48 |
0.84* |
0.61 |
0.87* |
| REL |
0.84* |
0.65 |
0.63 |
0.6 |
| NAT |
0.74 |
0.61 |
0.65 |
0.76 |
| ECON |
0.47 |
0.34 |
0.66 |
0.54 |
| DU/APR/DEUS |
LIB |
|
|
0.61 |
0.87* |
| REL |
|
|
-0.03 |
-0.18 |
| NAT |
|
|
0.62 |
0.79 |
| ECON |
|
|
0.38 |
0.45 |
| SDL |
LIB |
0.96* |
0.34 |
0.18 |
0.14 |
| REL |
0.8 |
0.81* |
0.57 |
0.76 |
| NAT |
0.81* |
0.29 |
0.31 |
0.6 |
| ECON |
-0.04 |
0.93* |
0.56 |
0.8 |
| ZRS |
LIB |
|
|
0.09 |
0.46 |
| REL |
|
|
0.49 |
0.58 |
| NAT |
|
|
0.3 |
0.9* |
| ECON |
|
|
0.74 |
0.74 |
| HZDS |
LIB |
0.65 |
0.95* |
0.35 |
0.47 |
| REL |
0.65 |
0.74 |
0.54 |
0.66 |
| NAT |
0.83* |
0.96** |
0.98** |
0.97** |
| ECON |
0.77 |
0.65 |
0.62 |
0.7 |
| SNS |
LIB |
0.57 |
0.91* |
0.32 |
0.5 |
| REL |
0.65 |
0.72 |
0.54 |
0.69 |
| NAT |
0.68 |
0.96** |
0.94* |
0.98** |
| ECON |
0.92* |
0.12 |
-0.12 |
0.09 |
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
*p<0.1. **p<0.01
Back
Table A10.
Correlation between party sympathy and factor positions by party in
Slovakia over time
| Party |
Factor |
Correlation between party sympathy score and difference
in factor score for all other parties |
|
|
1992 |
1993 |
1994 |
1996 |
| ODS/KDS |
LIB |
0.88* |
0.92* |
0.89* |
0.98** |
| REL |
0.02 |
0.09 |
0.02 |
-0.09 |
| ECON |
0.87* |
|
0.95* |
0.76 |
| NAT |
|
0.47 |
|
0.71 |
| LIB2 |
|
0.96** |
|
|
| ODA |
LIB |
0.94* |
0.81* |
0.78 |
0.92* |
| REL |
-0.03 |
0.14 |
0.1 |
-0.04 |
| ECON |
0.83* |
|
0.96** |
0.5 |
| NAT |
|
0.62 |
|
0.82* |
| LIB2 |
|
0.97** |
|
|
| KDU-CSL |
LIB |
0.93* |
0.79 |
0.75 |
0.94* |
| REL |
0.49 |
0.76 |
0.69 |
0.69 |
| ECON |
-0.06 |
|
0.32 |
0.45 |
| NAT |
|
0.57 |
|
0.8 |
| LIB2 |
|
0.16 |
|
|
| CSSD |
LIB |
0.3 |
0.38 |
0.35 |
0.69 |
| REL |
0.33 |
0.38 |
0.25 |
0.11 |
| ECON |
0.35 |
|
0.54 |
0.66 |
| NAT |
|
0.42 |
|
0.56 |
| LIB2 |
|
0.32 |
|
|
| KSCM |
LIB |
0.89* |
0.88* |
0.89* |
0.71 |
| REL |
0.44 |
0.42 |
0.41 |
0.27 |
| ECON |
0.91* |
|
0.7 |
0.88* |
| NAT |
|
0.25 |
|
0.74 |
| LIB2 |
|
0.69 |
|
|
| SPR-RSC |
LIB |
0.64 |
0.8 |
0.84* |
0.55 |
| REL |
0.59 |
0.5 |
0.5 |
0.23 |
| ECON |
0.79 |
|
0.66 |
0.71 |
| NAT |
|
0.87* |
|
0.78 |
| LIB2 |
|
0.63 |
|
|
Source: Toka 1992, Toka 1993, Toka 1994, Toka 1996
*p<0.1. **p<0.01
Back