C H A P T E R 2 2
PARTY SYSTEMS
H E R B E R T K I T S C H E L T
T H E concept of party system, while ubiquitous in political science texts, hardly
receives systematic treatment, if handbooks by Greenstein and Polsby (1975) and
Goodin and Klingemann (1996) are the reference points (cf. Epstein 1975; Pappi
1996). In a similar vein, all editions of the American Political Science Associations
Political Science: The State of the Discipline (1983, 1993, 2002) discuss parties only
within the micro-political context of individual political behavior and preference
formation, but have no room for party systems. In the most recent volume, party
systems appear only in Fiorina's (2002) article centered exclusively on US parties.
The subject of political party systems may be too complex and heterogeneous to
deserve coherent treatment in key political science handbooks. Therefore entire
handbooks have been devoted to the study of parties and party systems (cf. Katz
and Crotty 2006). Or the proliferation of party system typologies in the 1950s and
1960s may have led to a "confusion and profusion" (Sartori 1976, 119) not even
resolved by Sartori's own last-ditch effort. Or comparative politics at least in America
has turned its attention so decisively toward comparative political economy, political
regime change, and ethnocultural identity politics as to ignore the study of parties
and party systems.1
Nevertheless, party system attributes continue to play a critical
role in treatments of political economy and public policy. The substantive alignments
of interests and the competitiveness of party systems representing such interests are
critical variables in studies of political economy, public policy, and democratic
regime survival.
In this article, I first conceptualize party systems separate from parties in analogy to
Waltz's (1954,1979) treatment of international systems separate from states(Section 1).
I then identify systemic properties of party systems for the comparative-static analysis
1
Not by chance, these are the three prominent themes of comparative politics singled out by Laitin
2002 in his overview of the comparative politics subfield.
P A R T Y S Y S T E M S 523
of competition (Section 2). Subsequently, I probe into the historical-evolutionary
competitive dynamic of party systems (Section 3). Here historical-comparative analysis
comes into its own beyond the study of formal properties of party systems and
competition. My contribution refrains from discussing party systems as independent
variables that may account for outputs and outcomes of democratic politics, as this
subject is covered in other handbook chapters.
1 T H E C O N C E P T OF P A R T Y S Y S T E M
Waltz (1954) distinguished three analytical levels or "images" of international
politics. The first deals with human behavior, the motivations and actions of
individual policy makers and members of societies. The second focuses on processes
of group decision making internal to state organizations, as they produce binding
collective decisions about foreign policy. The third examines state strategies as
a consequence of "systemic" features. The system is conceived as a set of interacting
units (Waltz 1979, 40). In a system, the action of each participant entity is affected by
the actions of all others. Systemic theory must hence "show how the systems level, or
structure, is distinct from the level of interacting units" (ibid.). In game-theoretic
language, systemic features map the structure of the game, as defined by actors'
resources, preference schedules, and feasible moves that translate into positive or
negative outcomes contingent upon the other players' moves. If preferences are fixed
and exogenous, equilibrium states of a system are entirely determined by systemic
features concerning the numbers of players, the rules of movement, and resources
distributed among the actors. As in economic markets, hegemonic or oligopolistic
configurations permit actors to coordinate around different equilibria (relative
prices, states of war and peace in the system) than competitive markets with many
suppliers and purchasers.
Also party system theory identifies numbers of players, distributions of resources
and capabilities among them, and permissible rules of movement to arrive at predictions
that hold true regardless of internal idiosyncrasies of the individual elements.
Equilibria concern the number of sustainable players, their profile of payoffs, and their
relations of alliance and conflict among each other. These then translate into practices
of creating and maintaining government executives, extracting and allocating scarce
resources to constituencies, and maintaining or abandoning democracy more generally.
Even if such systemic propositions are successful, however, they may require
qualifications and further specifications based on knowledge about the internal
behavior of individual parties, thus setting limits to a purely systemic analysis.
At least tacitly the "three images" of international relations theory have always
been a staple also of comparative party system theory, as Sorauf's (1964) distinction
between "party in the electorate" (individual behavior and orientations), "party as
524 H E R B E R T K I T S C H E L T
organization" (polities as organizations), and "party systems" suggests. Party systems
theory is driven by a particular parsimony of focus: Net of idiosyncrasies characterizing
individual actors (citizens, politicians) and modes of intra-party decision
making, does the structure and dynamics of party systems causally account for
identifiable outputs and outcomes of the political process?
Let me begin by outlining first and second image assumptions without which no
useful hypotheses about third image (systemic) features and processes can be derived.
Just as international systems presuppose historically distinctive first and second image
features (cf. Ruggie 1989; Spruyt 1994), also party "systemness" and "systemic
processes" take place only when certain lower order conditions are satisfied.
First image assumptions about individual actors (citizens, politicians). Systemic
strategic interactions among parties presuppose that at least some citizens compare
candidates and parties for electoral office with respect to some of the rewards they offer
citizens. If all citizens abstain from voting, vote in a random fashion, or vote based on
immutable affective collective group affiliations rather than the comparative alignment
of principals' and potential agents' preferences, then there can be no systemic
processes. In the sense of Lupia and McCubbins (1998) or Erickson, MacKuen, and
Stimson (2002), at least some voters must be "rational information misers" whose
strategic choices (voting or non-voting, supporting one candidate/party rather than
another) are contingent upon the expected behavior of other voters and of electoral
candidates who offer to serve as their agents in legislatures and executives.
In a similar vein, the candidate agents ("politicians") in the electoral polity must
strategically act so as to take the preferences and strategic options of at least some
principals (voters) and rival candidates into account in their own choice of a course of
action. Just as states in international relations theory are postulated to seek survival,
politicians seek (re)election to political office—executive office, and as a second best
legislative office—as the baseline objective, whatever other goals they may pursue beyond
that (personal rents, glory, policy, or targeted benefits for constituencies). Whether and
how they pursue these higher-order objectives is endogenous to the competitive
situation, characterized by the rules of the game, the stances of their competitors, and
the demands of the voters. It is these constraints that prevent politicians in some
circumstances from becoming just utterly cynical self-regarding rent maximizers and
predators.2
In some circumstances, the pursuit of executive office may presuppose that
politicians credibly commit to collective goods producing public policies.
Systemic processes in electoral democracies presuppose the existence of an
"electoral market" in which choices of principals and agents are contingent upon
each other. There must be some "elasticities" between supply and demand. Where
empirically this condition is not met, systemic party theory is inapplicable. Principals
may lack material and cognitive resources to participate in an electoral market, e.g. in
extremely poor countries, or they may be so committed to a particular political agent
("party identification") as to pre-empt systemic processes, e.g. in ethnically highly
divided polities.
2
On systemic conditions for the choice of parties' and politicians' preferences, see Strom 1990«.
P A R T Y S Y S T E M S 525
Second image assumptions about constituent entities of the party system (collective
agents). In mass democracies with universal franchise, principals and agents can act
effectively in electoral markets only through intermediary vehicles of coordination that
help them to overcome collective action problems, to facilitate the flow of information
in the market, and to simplify the range of service options based on which principals
and politicians may enter direct or indirect contracts with each other. Political parties,
the constituent elements of a party system, may provide some or all of such services
(Aldrich 1995). Party is here used in a generic sense as a set of politicians pooling
resources, not necessarily the label that demarcates parties in a legal-institutional sense.
The effective locus of coordination may sometimes be factions within party labels or
coalitions combining party labels (Morgenstern 2004). To simplify matters, parties are
henceforth the effective collective agents, not necessarily the legal labels.
Parties may help to overcome collective action problems by reducing voters' costs
of information gathering and candidates' costs of information distribution in the
run-up to the electoral choice. Parties may also reduce problems of "social choice"
that surface in unstable and cycling majority decisions in legislatures and
governments by bundling and binding sets of politicians with different individual
preference schedules to work together in pursuit of a single collective preference
schedule ("party program").3
There may be other vehicles of collective mobilization
that contribute to the articulation and aggregation of interests, such as social
movements and interest groups. Only in a very few limitational empirical cases,
such as Papua New Guinea, does democracy appear to exist without parties in the
generic sense of a system of collective agents intermediating in the electoral process.
At the other end of the spectrum, where most parties exhibit some durability and
capacity to coordinate citizens and politicians time and again, we speak of party
system institutionalization (Huntington 1968, ch. 7; Mainwaring and Scully 1995). It
is akin to what Sartori (1968, 288-97; 1986, 55-6) has called a "structured" party
system and Mair (1997, 213-14) refers to as "systemness" through "closure," namely
the identity of interacting corporate units (parties) over some extended time period.
2 V A R I E T I E S OF P A R T Y S Y S T E M S
Party system theory aims at predicting strategies of the competitors and preferably
identifying equilibria of such strategies. The critical elements are the number of
competitors and the "currency" of competition for voter support, namely the policy
issues and issue bundles politicians promise to enact to shore up electoral support.
Theories typically assume an indirect exchange between voters and politicians.
3
On the theory of party formation, see especially Aldrich (1995); Cox and McCubbins (1993, chs. 4
and 5); and Snyder and Ting (2002). Whether or not they solve collective action and social choice
problems, as Aldrich 1995 postulates, however, is a contingent process (see below).
526 H E R B E R T K I T S C H E L T
Citizens surrender their vote at the beginning of the electoral term in exchange for
the winning politicians implementing campaign promises during the electoral term.
Democratic accountability operates indirectly because of (1) the time elapsed
between election and policy delivery; (2) the benefits and costs of policy accruing to
all voters, regardless of whether they supported winners; and (3) voters speaking their
verdict over the record of governing politicians (and the opposition) retrospectively
at the end of the electoral term and taking that evaluation into account in their
prospective assessment of politicians' promises for the subsequent electoral term.
The policy-based "responsible partisan" model, however, is only one special case of
principal-agent relations within a broader set of mechanisms expressing democratic
accountability. Before turning to the key elements of the common models of party
competition—numbers of competitors and numbers of dimensions of competition—let
us therefore distinguish modes of democratic accountability in terms of
different principal-agent exchanges (Section 2.1). Moreover, and related to this point,
critics have argued that responsible partisan models home in on a highly constrained
view of the currency of competition, namely policy positions rather than a variety of
valence goods broadly conceived (Section 2.2). Once the special place of positional
issue competition has been characterized, we then can turn to numbers of players
and dimensions of policy issues as structural properties of party systems (Sections 2.3
and 2.4). Finally, for all party systems we can distinguish greater or lesser intensity of
competition or "competitiveness" (Section 2.5).
2.1 Modes of Democratic Accountability
Why do voters support parties and how can politicians in calculated fashion appeal
to voters for support? Party systems theories focus on mechanisms that involve
rational deliberation, as opposed to affective psychological attachments, such as
party identification, voter identification with the objective traits of candidates
(gender, ethnicity), or the personal inspirational ("charismatic") qualities of
a candidate. Inasmuch as support based on such criteria treats them as tracers
of candidates' cumulative policy records and policy commitments, such as in
Fiorina's (1977, 1997) felicitous phrase of party identification as the "running
tally" of a party's past record, of course, they are incorporated into theories of
party competition.
Among rational modes of accountability, let us distinguish between indirect and
direct exchange between voters (as principals) and politicians (as their agents). In
the indirect policy exchange, citizens surrender their vote in accordance with the
responsible partisan model. The exchange is indirect because it involves an intertemporally
long drawn out process between the principal delivering the vote and
the agent putting authoritative measures into place that allocate costs of benefits to
all members of abstract categories of voters, regardless of whether individual
members of each category actually voted for the decision maker or not. Politicians
may have only a general sense of where their supporters are located in society.
P A R T Y S Y S T E M S 527
They are unable to pinpoint, monitor, or sanction their voters. In contrast to this
indirect "policy" exchange, in direct, targeted, "clientelistic" exchange, individuals and
small groups of voters obtain immediate gratification in exchange for their vote or
suffer negative consequences in case of supporting the loser. The currency of exchange
here involves gifts or money, public sector jobs, public housing, privileged access
to social policy transfers, favorable regulatory rulings, or procurement contracts
that allow firms to hire workers who supported the winning party and candidate.4
Clientelistic politics comes with direct or indirect social mechanisms permitting
politicians to monitor and even sanction the electoral behavior exhibited by small
groups.
Numerous theories have tried to account for the relative prominence of clientelistic
exchange relations in party competition (cf. Scott 1969; Schmidt et al. 1977;
Shelter 1994; Kitschelt 2000a; Piattoni 2001; Keefer 2005; Kitschelt and Wilkinson
2006). Increasing affluence and eradication of poverty may make the relative value of
clientelistic inducements meaningless for voters and heightens their sensitivity to the
opportunity costs of such practices, e.g. in political production of collective goods.
Net of development, clientelism hinges upon the economic viability of state-owned,
state-subsidized, or state-regulated firms and entire sectors. Eonomic entities
operating under a state-provided "soft budget" umbrella are more amenable to
crony appointments and thus clientelism. The presence of mobilized and electorally
vocal ethnocultural groups in divided societies furthermore tends to fuel clientelistic
practices (cf. Horowitz 1985; Chandra 2004; Wilkinson 2004). Furthermore, all
these factors may interact with the competitiveness of a party system (see below).
Greater competitiveness may fuel more intensive efforts by politicians to engage in
either clientelistic and/or programmatic policy competition.
Whether electoral and executive institutions affect the balance of clientelistic and
programmatic competition in party systems, however, is a matter of disagreement.
Electoral rules that require candidates to carve out narrowly circumscribed electoral
constituencies with whom candidates have direct dealings may induce clientelistic
exchange (cf. Katz 1980; Ames 2001). But it is easy to find examples of closed-list
multi-member district electoral systems where most parties have practiced clientelism,
such as Venezuela (-1999) and Austria, or programmatic parties in open list
preference voting systems (cf. Samuels 2004).
2.2 Valence or Positional Competition
Critics of conventional theories of party competition have introduced another useful
distinction that can be related to modes of democratic accountability: that between
valence and positional issues or party offers (Stokes 1963). Citizens' preference
4
Clientelism always involves material incentives to turn out the vote, not just a monetary transfer by a
rich citizen to a party in exchange for economic favors. Such material provisions, of course, make it easier
for politicians to establish clientelism.
528 H E R B E R T K I T S C H E L T
distribution over some salient, prized good is highly skewed so that most citizens
want more rather than less of a good (honest politicians, competent management of
the economy...). Parties do not take "positional" stances over whether or not to
supply some good, but whether they can credibly supply that good better or to
a greater extent than their rivals. Each party claims to have "more" attractive
candidates and technical advisers, demonstrate "greater" competence in producing
collective goods (such as facilitating economic stability and growth, protecting the
environment, preventing terrorism), and/or distribute "more copious" targeted
benefits to anyone who is asking for them.
Positional competition, by contrast, assumes a broad distribution of voter preferences
over the merits of the parties' offers of goods or services. Parties may then
promise different things to different voters on the same dimension (see Section 2.3).
Positional offers mostly concern policy issues and bundles thereof. But critics of
positional theory claim that for voters valence issues trump positional issues most of
the time. Retrospective economic voting, for example, has to do with the perceived
"competence" of a party's politicians in delivering good economic performance, such
as low inflation and high growth. Moreover, non-policy modes of principal-agent
relations also operate in the realm of valence competition. In clientelistic politics,
parties compete for votes by advertising themselves as suppliers of the most copious,
reliable, and expediently delivered targeted benefits. And competition with a candidate's
personal charisma may turn on widely desired qualities such as leadership,
compassion, or youthful dynamism.
Nevertheless, there is no one-to-one relationship between modes of democratic
accountability and the prevalence of valence or positional offers in party competition.
With respect to candidate qualities, while no voter would want incompetent
politicians, some citizens may prefer compassion and careful deliberation as a quality
of political leadership over decisiveness and expedient action. In a similar vein,
descriptive representation of electoral constituencies (by means of the candidate's
gender or ethnicity) may be a "positional" strategic move in diverse constituencies
where candidates with different ethnocultural markers are competing for political
office. Also clientelistic exchange may evolve according to a positional dynamic.
There may be electoral situations with highly diversified constituencies that make
it attractive for some parties to embrace clientelism and imply that one of its
correlates, corruption, should be treated leniently, whereas other parties take the
opposite position.
Most importantly, however, one might directly contradict Stokes (1963) and
actually assert that most policy issue appeals are at least implicitly positional rather
than valence based. Whereas many ultimate objectives in political life may be of the
valence type, politics is about the choice of means to obtain those ends, and here one
may be firmly in the realm of positional competition because of cognitive and
evaluative disagreements. People may have different assessments about the causal
efficacy of a policy means to reach an end, given the complexity and uncertainty
surrounding causal relations in social life. People may also disagree on the distributive
implications that the choice of policy means involves. Politicians may use valence
P A R T Y S Y S T E M S 529
codes—such as fighting crime, reducing inflation, or creating jobs—to pursue a
distributive agenda. For politicians it is part of the art of heresthetics (Riker 1986) to
conceal the distributive implications of their own appeal to valence issues, but to
highlight those of their opponents' valence issue frames. It is important to realize the
limits of valence competition because the Party Manifestoes Project, as the most
comprehensive and systematic enterprise to register the programmatic appeals of
political parties, was at least initially based on the supremacy of a valence-based
characterization of party competition (cf. Budge, Robertson, and Hearl 1987;
Budge et al. 2001).
Figure 22.1 summarizes the relationship between accountability mechanisms and
the prevalence of valence or positional competition. Empirically, I claim the following
testable regularities. Political candidate appeals play out in most instances into
valence competition and only rarely as positional competition. Clientelistic accountability
works mostly as valence competition among parties (who can deliver the most
and most reliably? alternatively: Who is the "cleanest" in rejecting clientelistic
inducements?). Under certain conditions of economically highly stratified constituencies
with great disparities of income, clientelism may become a matter of positional
competition, with some parties defending and others attacking it.
Fig. 22.1 Modes of Political Linkage in Democracies
530 H E R B E R T K I T S C H E L T
2.3 Numerical Properties: Fractionalization, Effective
Number, and Volatility
From early on, party systems have been divided into two-party and multiparty
systems (cf. Duverger 1954; Downs 1957), ultimately giving way to a proliferation of
numerical criteria (Mair 1997, 200-6). Most prominent may have been Sartori's
(1976) further distinction between moderate and polarized multiparty systems
dependent on the presence of "anti-system" spoiler parties. But since the 1970s
typologies of party systems have fallen out of favor to the advantage of a variablebased,
finer instrument to gauge the size of party systems. It is the measure of party
system fractionalization (Rae 1967), or its mathematical inversion proposed by
Laakso and Taagepera (1979), the "effective number of parties," whether calculated
in terms of voter support for parties (ENVP) or size of parliamentary parties
(ENPP). The basic idea here and in further mathematical iterations of the measure
(Molinar 1991) is to combine the number and the size distribution of parties in
a polity in a single coefficient of fragmentation that sums up the parties in a polity
weighted by their size. Fractionalization measures employ partisan labels as their unit
of counting. Such measures are meaningful only as long as parties can be treated as
unitary collective actors (cf. Morgenstern 2004).
The same qualification applies to a widely used structural parameter of party
systems in the temporal dimension, the volatility of party systems. The volatility
index summarizes the percentage differences of electoral support obtained by the
same parties in two subsequent elections (usually divided by two to give a maximum
value of 100) (cf. Pedersen 1983). It is almost self-evident that fractionalization
and volatility are closely related. But where several parties are close to each other
and operate as one "bloc" in legislatures and elections, a party-based volatility index
may seriously overstate volatility by not focusing on the "inter-bloc" volatility of
party systems (cf. Bartolini and Mair 1990). The differential conceptualization
of volatility may have major consequences, if one employs the concept to gauge the
stability and consolidation of party systems over time (e.g. Mainwaring and Scully
1995; Roberts and Wibbels 1999).
2.4 Policy-Based Programmatic Party Systems: Social and
Political Divides, Cleavages, Competitive Dimensions
In addition to numbers of players, spatial-positional theories of programmatic party
systems consider the number of dimensions on which parties compete, something
that empirical comparative analysis often refers to as "cleavages." Because of the
variability of language that prevails in this literature, it is important to draw clear
terminological distinctions. There are lines of division running through every society
generated by social, political, economic, and cultural group interests and sentiments
of deprivation. If such divides of traits, affiliations, and opinions are durable we may
call them cleavages (Rae and Taylor 1970), particularly if they mutually reinforce each
P A R T Y S Y S T E M S 531
other (Bartolini and Mair 1990). They are separate from mere "divisions" that denote
more fleeting group divides typically associated with a single point decision (e.g. to
take an example from Europe: driving on the left or the right side of the road).
Cleavages tend to have the qualities of social entrapment and closure. Individuals face
costly barriers to enter and to exit a social or political category and the rewards and
deprivations associated with membership. Therefore they tend to organize as that
category in order to acquire or defend certain economic, political, or cultural
resources, rights, and privileges.
Only few of these divides ever translate into collective action to change the
allocation of gratifications, let alone the very specific and challenging form of party
politics. A political partisan divide appears where parties represent different sides of
a social divide. Statistically, such partisan mapping of divides can be detected with
techniques of factor and discriminant analysis as well as regression analysis, with
party choice as the dependent variable, especially multinomial logistic models. The
number of social divides that map onto the party system may be larger than
the number of partisan divides, if there are several reinforcing divides captured by
the same party alternatives. Thus, if all working-class voters are also secular and all
non-working-class voters are religious, there will be no separate religious and class
partisan political divides, even if parties map both issues onto the party system.
Conversely, where group memberships on social divides cross-cut each other and are
mapped onto parties, they tend to generate multiple partisan divides.
From the perspective of office-seeking strategic politicians, what matters for their
strategic moves to win elections may be neither social nor even partisan divides, but
only the minimal set of competitive divides or "competitive dimensions" in a party
system. These are only those divides on which voters display some elasticity of
partisan choices, responding to modifications of the competing parties' appeals and
offers. By contrast many political divides are a matter of political identification rather
than competition (cf. Sani and Sartori 1983). In this instance, group membership
predicts the propensity to favor a party, but there is no open electoral market in
which voters would change their partisan choice, were competing parties to modify
their appeals on the given political dimension. In case of a competitive dimension,
a critical subset of rational voters is responsive to parties' changing electoral appeals.
These elasticities are elusive to measure, as they would require a panel data design.
A weak tracer of the competitive status of a dimension is the salience of the
underlying issues for voters and parties.
Table 22.1 summarizes the terminological conventions introduced in the preceding
paragraphs. An example might illustrate the usefulness of the distinctions in anticipation
of the stylized historical sketch provided later (Section 2.4). In Belgium,
until the 1950s, there were two cross-cutting political partisan divides that were
both competitive, a social class-based one pitting the working-class socialists, at
one extreme, against the cross-class Christian Democrats in the center and the
business-oriented liberals at the other extreme, and a religious divide separating a
secular socialist-liberal sector from a Catholic Christian Democratic camp. Over
time, the religious divide lost its competitiveness and became a pure partisan
532 H E R B E R T K I T S C H E L T
Table 22.1 The organization of issue opinions in democratic party competition
CENTRALITY OF DIVISIONS DURABILITY OF ISSUE DIVISIONS?
FOR THE ORGANIZATION
OF THE PARTY SYSTEM?
LOW: "DIVIDES" HIGH: "CLEAVAGESLOW:
SOCIAL and SOCIAL and
IDEOLOGICAL DIVISIONS AT "IDEOLOGICAL" DIVIDES "IDEOLOGICAL"
THE SOCIETAL LEVEL CLEAVAGES
INTERMEDIATE: POLITICAL PARTISAN POLITICAL PARTISAN
PARTISAN DIVISIONS AT DIVIDES CLEAVAGES
THE POLITICAL LEVEL (transitory)
HIGH: COMPETITIVE COMPETITIVE
COMPETITIVE DIMENSIONS DIVIDES CLEAVAGES
(transitory)
identification divide. As a parallel movement since the 1950s a hitherto politically
unmapped, but long-standing ethnocultural divide over language and region began
to articulate itself on the plane of party competition, but much more so in Flanders
than in Wallonia. By the 1990s, a realigned socioeconomic distributive divide, the
ethnolinguistic divide, and a newly arising libertarian-authoritarian divide over
political governance all surface in Belgian party competition, particularly in Flanders.
At the same time, the old socioeconomic working class versus business divide as well
as the religious divide had lost their capacity not only to shape party competition,
but even to maintain a partisan identification divide.
Does the number of parties reflect the number of cleavages in a party system
(Taagepera and Grofman 1985; Lijphart 1999, 81-3)? While there may be some
tendency that a proliferation of societal divides boosts the number of political
partisan divides and the latter boosts the number of competitive dimensions, this
is far from a foregone conclusion. The relationship between numbers of parties and
positional divides in a polity is theoretically problematic and empirically untested
because existing research has taken insufficient care in conceptualizing political
divides and competitive dimensions.
In many instances, but not in the Belgian example above, political parties reduce
the number of active dimensions of electoral competition to one or two only. The
literature offers several not necessarily exclusive reasons for a reduction in the
dimensionality in party competition. In all instances, the baseline assumption is
that parties cannot simply cherry-pick issues and refrain from taking a stance on the
full scope of salient issues, except if they are very small niche parties. This is so
because party politicians are elected in territorial districts to represent constituencies
over an uncertain and unlimited range of issues in legislatures where they have only
very limited agenda control, as is evidenced by the necessity to vote on a state budget
that covers a bewildering range of issues.
P A R T Y S Y S T E M S 533
First, where institutional barriers to entry favor a two-party system, politicians in
the established parties have powerful incentives to prevent internal party divisions
through cross-cutting issues and therefore map positions on new and salient issues
on the existing divides (Stimson 2005). Second, general cognitive limits of politicians'
and citizens' information processing of political alternatives give a strong
advantage to parties that can articulate their positions in a very low-dimensional
space of ideological alternatives (cf. Downs 1957; Hinich and Munger 1994; Lupia and
McCubbins 1998). Third, the evolution of social structure and the effects of policies,
such as the growth of the welfare state, on the distribution of preferences in society
might facilitate a bundling of political preferences around a very low-dimensional
space (Kitschelt 1994). None of these hypotheses suggests that there is a logically
compelling constraint according to which particular issue positions fit together.
2.5 The Competitiveness of Party Systems
Party systems are more "competitive," when (1) there is great uncertainty of electoral
outcomes and (2) uncertainty matters, i.e. small variances in parties' electoral support
translate into large variance in their legislative representation and/or bargaining
power over executive appointments, patronage, or policy. Where competitiveness is
intense, politicians make greater efforts to mobilize support and voters pay more
attention to politics (campaign contributions, turnout, information processing).
In two-party systems, competitiveness has often been measured as the ex-ante
closeness of two candidates in the electoral race, i.e. the expected margin of victory.
But this operationalization is not sufficiently general and does not take the "stakes"
of the electoral contest into account. Do voters and candidates make a great effort, if
the alternatives on offer are essentially the same?
Competitiveness of a party system is intense, if the following five conditions
prevail (see Figure 22.2).5
(1) For strategic politicians, marginally greater support
translates into large increases in bargaining power over legislative majorities
(coalitions) and executive office appointments. "Majoritarian" democracy with
single-member districts and plurality formula that tend to manufacture singleparty
unified majority government, at least under parliamentarism, and lack of
outside institutional veto points, thus giving high institutional leverage to legislative
or executive coalitions, tends to increase the competitiveness of elections (cf. Lijphart
1999; Powell 2000; Tsebelis 2002). (2) Where more than two effective contenders
prevail, politicians shore up competitiveness if they configure around identifiable
alternative parties or party blocs vying for political power.6
On the side of voters,
preference distribution must make all actors perceive the outcome as both (3) close
(low margin of victory between party blocs) and (4) open in the sense that there is
5
For a related discussion of electoral competitiveness, also there referred to as "executive responsiveness,"
see now Franklin (2004, 112-14).
6
For a discussion of identifiability and its operationalization see Strom (1990b, 47, 73-5).
534 H E R B E R T K I T S C H E L T
Fig. 22.2 Variables Influencing the Competitiveness of Party Systems
a sizeable electoral market of floating voters situated between the electoral alternatives
and responsive to small modulations of candidates' appeals.
Even if these four conditions are met, competitiveness is intense, however, only if
also (5) the "stakes" of the competition are high, i.e. the disparity of the cost-benefit
allocation by rival camps of politicians is great. Politicians raise or lower the stakes in
part as a function of conditions (1) through (4), but as the next section will show,
these relations are far from unambiguous. On the face of it, one might expect the
median voter theorem to hold: Where two identifiable blocs compete to win majority
status that endows great institutional leverage on the winner and there is an electoral
market between the competitors in a close race, both camps of politicians actually
reduce the stakes by offering similar cost-benefit allocations in case of victory, and
these commitments are most pleasing to the median voter. As we shall see, there are
complications that contradict this logic of countervailing forces between strong
competitiveness at the level of majority formation and weak competitiveness of
majority action ("stakes").
Given the complexity of the conditions that affect the competitiveness of elections
both from the perspective of politicians' as well as voters' incentives to make an effort
in the electoral contest, simple measures such as party system fragmentation and
volatility cannot serve as empirical tracers of competitiveness. Nevertheless, they
P A R T Y S Y S T E M S 535
have often been employed for such purposes, although they only indirectly affect
some of the conditions that determine competitiveness. Moreover, the causal links
attributed to such measures are debatable at best. While party system fragmentation
has often been considered to boost electoral competitiveness by increasing uncertainty
of electoral victory, the opposite may be true because fragmentation tends to
reduce the identifiability of governing coalitions. Party system volatility may be a
tracer of the size of the electoral market, but not necessarily of its location (between
rival camps?). Moreover, following Bartolini and Mair (1990) for the task of predicting
politicians' and voters' strategic choices in the party competition, volatility would
have to be measured at the level of party blocs rather than individual labels, a practice
rarely followed in the literature.
3 C O M P A R A T I V E S T A T I C S : S T R A T E G I C
C H O I C E I N P A R T Y S Y S T E M S
Most theories of party system competition work with the assumption that (1)
principal-agent relations concern indirect programmatic exchange about (2) positional
issues and offers. Strategic choices vary according to the number of competitors
and the relevant competitive dimensions of party systems only. The key objective
is to find equilibria contingent upon numbers and dimensions of competition such
that no strategic actor could alter her choice without lowering her payoff. Because
formal research over half a century has found that the identification of equilibria
under such conditions is elusive, more recent theorizing has relaxed model assumptions,
including those about principal-agent relations and positional issues, to obtain
equilibrium predictions. Alternatively, the quest for equilibria has been abandoned
altogether and been replaced by agent-based modeling in computer simulations.
3.1 Simple Spatial Theory: The Elusiveness of Equilibria
The most simple case and the starting point of the literature is Downs's (1957)
median voter theorem according to which two parties will both choose policy appeals
proximate to the position of the median voter. To derive this equilibrium, one must
postulate among many other things (1) office-motivated politicians with (2) perfect
knowledge of the situation (including voter preferences), (3) not having to fear the
entry of additional competitors, (4) relying on the selfless support of political
activists whose objectives are perfectly aligned with that of the candidates, (5)
competing in a unidimensional space of voter distribution for the support of rational
voters who (6) have explicit preference schedules and knowledge of the situation, (7)
must not abstain, and (8) cast their vote for the party whose announced position is
536 H E R B E R T K I T S C H E L T
closest to their personal ideal point (9) at the very moment of the election. In a
similar vein, under highly restrictive conditions of unidimensional competition,
some formal theories can show that in systems with four or more candidates rivals
disperse over the competitive space and generate an equilibrium distribution.7
Relaxing any one or several of the numerous assumptions necessary to derive the
median voter theorem, however, reveals its fragility (for an overview: Grofman 2004).
This dovetails with the empirical observation that even in unidimensional two-party
competition often enough the positions of the competitors diverge rather than
converge. Also equilibrium conditions in multiparty and/or multidimensional competition
are fragile and elusive. Shepsle (1991) sees no promise to find equilibria when
both more than one competitive dimension or more than two candidates are allowed
and certain other reasonable assumptions apply. In a survey, one of the most prolific
contributors to spatial theorizing of party competition concludes "that simple
theoretical generalizations about the structure of competition are unlikely to be
forthcoming" (Ordeshook 1997, 266). Theories that try to gain empirical relevance
have therefore made additional assumptions or abandoned the search for equilibria.
In both instances, the key aspiration is to account for both conditions of party
dispersion as well as stability, even if the size of electoral districts (M) and the
electoral formula would permit larger party systems with more entry (cf. Cox 1997:
M + i as outer bounds of the size of party systems).
3.2 Complex Spatial Theory: Equilibria under
Special Conditions
Because of the proliferating literature, I confine myself to listing a few prominent
proposals to relax spatial-positional theories of competition. I sidestep valence-based
issue theories of competition (Budge and Farlie 1983), as I am convinced that issues
are always positional, when choices are properly framed. "Valence" comes into play,
however, through non-issue considerations of candidate attractiveness, party identification,
including the competence of both to deliver selective benefits or "good"
public policy, and here we get to two prominent recent proposals to account for
stability and dispersion of party positions in two- and N-party systems.
First, Adams, Merrill, and Grofman (2005) develop a spatial model in which
ingredients of (1) voters' non-policy partisan predilections (including identification),
(2) discounting of the candidates' credibility or effectiveness in delivering on their
promises, and (3) voters' ability to abstain bring about stable equilibria of programmatically
dispersed parties in unidimensional or multidimensional spaces. The
non-policy partisan preferences are key, while discounting and the option to abstain
only amplify their effect on the dispersal of the partisan vote. The logic is clear. If
voters identify with a party for non-policy reasons, they support it even if its current
issue positions are further removed from the voter's ideal point than those of a
7
See especially Enelow and Hinich (1990) and Shepsle (1991).
P A R T Y S Y S T E M S 537
competitor. While plausible, the trouble with this argument is that "non-policy"
factors involve a whole host of variables that must be unpacked and that indirectly
often may have a subtle policy base, for example when the influence of people's
occupation or party identification may amount to a long-term assessment of a
party's policy commitments.
Second, Schofield (2003, 2004) has refined a valence model of competition in
which strategic parties disperse over a programmatic issue space so long as their
advantage or disadvantage in capturing voters on an additional valence dimension,
incorporating their candidates' reputation for competence and leadership, gives them
flexibility in their programmatic appeals. While formally elegant, in empirical terms
this proposal may generate a post hoc opportunistic account of party system strategic
dispersal. Just as in Adams et al.'s (2005) investigation, given the flexibility of the key
independent variable, researchers will always be able to locate some sort of valence
factor, if dispersal of parties occurs.
Third, starting with May (1973) and Robertson (1976) through Aldrich (1983) and
Schlesinger (1984) to McGann (2002) and Miller and Schofield (2003), theorists have
introduced preference heterogeneity among the principals who select a party's
electoral candidates and office holders. If such candidates rely not only on voters,
but also on party "activists" who contribute labor and capital to mobilize voters
without being candidates themselves, then the aspirations and preferences of the
latter may matter for the strategic appeals of the former. To preserve the electoral
credibility of their party, leaders may need to give activists some voice in the strategic
decision-making process, thus demonstrating that unity around a set of objectives is
more than tactical lip service of a few leaders, but a broadly shared commitment
(Caillaud and Tirole 2002). But party activists tend to be ideologs who join a party to
express programmatic preferences rather than to win elections (cf. Panebianco 1988).
To secure indispensable activist input, candidates may be compelled to adopt issue
positions distinctly removed from their optimal voter issue appeal. Whether or not
activists hold such radicalizing positions, however, may depend on the format of the
party system and on societal preference mobilization around a class of issues (cf.
Kitschelt 1989«). In multiparty systems where dissatisfied activists can join competing
party labels it is less likely that activists express systematically different views than
instrumental for the pursuit of votes and office.
Fourth, a long line of modeling has postulated that electoral candidates are not
just office, but also policy seeking, and therefore diverge from their spatially optimal
vote-getting programmatic appeal. The most encompassing and complex elaboration
of that perspective can be found in Roemer (2001) who shows that even in the twoparty
case the presence of policy-motivated candidates, faced with uncertainty over
voters' preferences in a two-dimensional policy space and the task to build a winning
coalition among three different intra-party factions around a winning joint electoral
strategy, will yield equilibrium positions that clearly set the competitors apart from
each other.
Fifth, voters may be strategic and not vote based on the proximity between their
own policy ideal points and that of individual parties, but that of likely future
538 H E R B E R T K I T S C H E L T
partisan coalitions (Kedar 2005). In that case more extreme parties may gain larger
shares of votes and yield a more polarized spectrum of alternatives. Voters support
more radical parties than is warranted by their own policy ideal points in the
expectation that these have to bargain policy compromises with moderate collaborators
that ultimately bring the center of policy gravity of a coalition government close
to the voters' sincere ideal points. A further modification of this point may be a
model of lexical voting (Kitschelt and Rehm 2005). If parties, constrained by their
past record of action, do not substantially diverge from each other on a highly salient
policy dimension so that voters are basically indifferent between the partisan alternatives,
in a manner of "lexical" ordering in their choice among parties voters may
focus on a second, third, or n-th dimension of competition just as long as partisan
alternatives on that dimension are stark and salient for the voters.
Sixth, voters may not act on a simple spatial rationale in which they gauge the
Euclidean distance, weighted by salience, between their own ideal policy schedules
and those of the partisan competitors, but support parties in a "directional" fashion
based on whether they take a pronounced position on the "correct" side of a political
issue, thus giving parties an incentive to disperse their issue positions (cf. Rabinowitz
and McDonald 1989). A huge theoretical and empirical literature surrounds this
proposal that ultimately appears to conclude that both spatial and directional
elements enter voters' calculation, but that empirically the directional component
only adds a vanishingly small modification to the basic spatial set-up of voting
behavior (cf. Merrill and Grofman 1999).
3.3 Agent-Based Modeling of Party Competition
As a backlash against formal theory, but also voicing unease with purely historical
narratives of party competition, a new computational approach of agent-based
modeling of political behavior has tried to gain theoretical insights in the comparative
statics and dynamics of party systems (cf. Kollman, Miller, and Page 1992,1998).
Critical assumptions are here that voters and politicians have very limited knowledge-processing
capacity and therefore act on simple rules rather than on a survey of
everyone's preferences and strategic options. Because voters vote spatially, but process
little information, parties can only slowly move in the issue space without
wrecking their reputation. Following Laver (2005), parties act on simple rules of
thumb, such as that of "hunter" who repeats appeals that have increased electoral
support recently and modifies them, if elections were lost, or that of "predator" who
always moves toward the electorally strongest party. In a two-dimensional space with
randomly distributed voters, such conduct may yield a gradual gravitation of the
partisan actors to the center region of the space, but with no party moving directly
dead center and continuous oscillation of positions that prevents stable equilibria.
The advantage, but also downside, of agent-based computational models is that
an infinite number of modifications and complications can be introduced without
knowing the epistemological advantage of each move in the enterprise. Do we
P A R T Y S Y S T E M S 539
achieve a theoretical explanation of observable behavior if the simulation results of
a certain model specification coincide with empirical patterns? What if many
different specifications reproduce the same empirical patterns? What this suggests
is that agent-based modeling must be combined with empirical research that lends
robustness to the behavioral assumptions employed in the computer simulations.
In this sense, Laver's model could be enriched by a simple calculus of voters'
abstention or participation in elections, contingent upon the observed parties in
the vicinity of voters.
3.4 Entry of New Parties
Formal spatial theories have scored only very limited success in accounting for party
entry (for a critique, see Laver and Schilperoord 2005, 8-9). More promising may be
a recent non-spatial game-theoretical model with incomplete information where a
potential entrant interacts with an established party, although it makes questionable
assumptions about the distribution of incumbent and challenger (private) knowledge
in the game and generates rather mixed empirical results (Hug 2001).
The informal, empirical literature has implicitly been driven by a behaviorally
constrained quasi-spatial framework of competition in which the entry and exit of
parties is seen as a result of an interplay between demand and supply (cf. Hauss and
Rayside 1978; Harmel and Robertson 1984; Kitschelt 1988,1995a). Induced by sociological
and political-economic developments, new political demands become salient
that established political parties are not willing to service. This intransigence may
result from an interaction of (1) the reputation of an established party that can be
changed only slowly at considerable electoral cost combined with (2) the electoral
tradeoffs involved in modified programmatic appeals. While a new issue appeal may
attract new electoral constituencies only gradually, established voters may be alienated
quickly, plunging an established party into an electoral crisis. Barriers of entry to
new challengers, as erected by electoral systems, mass media access, or party finance,
may make it more or less comfortable for existing parties to ignore new political
demands. Computational models can capture both the strategic immobility of
established parties as well as the barriers to entry encountered by new parties (cf.
Laver and Schilperoord 2005).
While much of the informal and computational literature on party entry implicitly
subscribes to a spatial model of party competition, though with relaxed rationality
endowments for voters and politicians, critics have modified this perspective
through salience models. Meguid (2005) argues that new niche parties may arise if
a party antagonistic to its claims nevertheless raises the salience of the issue by
engaging in an adversarial strategy in the hope to hurt an existing competitor who
prefers to dismiss the issue because it might internally divide and make it lose some
of its current party constituencies, if the issue were to become salient. Meguid tries to
endogenize the dimensions of party competition itself. In a more radical fashion this
was anticipated by Riker's (1982, 213-32) theory that a permanent loser party on an
540 H E R B E R T K I T S C H E L T
existing dimension of party competition may try to create a new competitive
dimension in the party system that internally divides the hegemonic party and
creates an opening for a new party or an old loser to displace it electorally. Riker's
historical reference point is the rise of the Republicans with the slavery issue. The
example also shows, however, the limits of a voluntarist theory in which strategic
politicians can "manufacture" salient issue dimensions. As Weingast's (1998) alternative
account of the slavery issue in party competition suggests, politicians may
create new parties and alignments only when political-economic conditions enable
them to count on an exogenous process in which sufficiently large constituencies
develop new political claims that are not mapped onto the existing party system.
4 H I S T O R I C A L D Y N A M I C S OF
P A R T Y S Y S T E M S
Students of the historical dynamics of party systems, the trailblazer of which was
Lipset and Rokkan's (1967) article about the emergence and persistence of political
cleavages in Western Europe, implicitly build on and apply many elements explicitly
modeled in spatial theories of party competition and in models of party entry and
exit. Thus there is no contradiction between the formal or informal general analytical
literature on party competition, on one side, and the comparative-historical
analysis of party system evolution. As socioeconomic, political, and cultural conditions
create new divides of interests and values in society, different issue bundles will
be mapped onto the arena of party competition, contingent upon the institutional
constraints and strategic opportunities politicians see in the game of jockeying for
votes, political office, and control of public policy. Ideally, the general analytical and
the historical-comparative literatures on party systems complement and crossfertilize
each other. Whereas the former is mostly a comparative-static analysis of
strategic moves when the political preferences of voters and party politicians are
given, but the number of partisan players is either exogenous or endogenous,
the latter fills this ahistorical framework with flesh and blood by identifying the
sociological, political-economic, and cultural developments that shape preferences
as well as the institutional and strategic conditions that influence the set of political
strategies seen as feasible by the political actors.
4.1 Classical Analysis of Party System Formation
in Western Europe
Lipset and Rokkan (1967) analyze the development of European party systems from
the nineteenth to the mid-twentieth century against the backdrop of the twin
P A R T Y S Y S T E M S
541
challenges of the national and industrial revolutions that began to take place since the
seventeenth century. But in no way is their analysis one of sociological determinism
(Sartori 1968). First of all, the historical conditions that shaped the mobilization of
societal divisions were shot through with political action. The development of parties
and party systems takes place against the backdrop of strategic political choices and
interactions among conflicting elites in the process of building territorial states,
subduing religious associations under state authority, coping with the reticence of
agrarian elites against relinquishing political control, and including the growing
working-class movements in institutionalized politics. Second, they emphasize the
complex and varied political process of electoral enfranchisement and institution
building both as consequence and as cause of party system formation. Agrarian and
religious divides therefore do not naturally flow from sociological conditions, but
result from a complex strategic interaction among political elites.
The finest examples of post-Lipset-Rokkan comparative historical analysis capturing
the interrelations of demand and supply conditions in the formation and
realignment of European party systems are probably the works of Luebbert (1991),
Kalyvas (1996), and Bartolini (2000). Luebbert emphasizes the different strategic
conflict between socialist, liberal, and conservative parties in the mobilization of
agrarian constituencies to account for different pathways of party systems in the
inter-war period. Kalyvas (1996) highlights the strategic calculations of the Catholic
Church and of Catholic lay politicians involved in the formation of confessional
parties since the late nineteenth century. And Bartolini (2000) develops an allinclusive
landscape of demand and supply conditions that have shaped the mobilization
of the class cleavage in European politics as the last and therefore residual line
of conflict strategic politicians had to wedge into already party systems already
constituted along other divides.
These books render a more subtle and empirically plausible picture of party
formation than two analytically leaner, but historically far less insightful perspectives.
Przeworski and Sprague's (1986) intentionally voluntarist account of partisan class
politics emphasizes strategic politicians and their capacity to shape the terms of
working-class formation, although the empirical analysis is compelled to concede the
powerful role of pre-existing cross-nationally varying cultural diversity, corporatist
interest intermediation, and socioeconomic development of blue-collar electoral
constituencies. At the opposite end of the spectrum, Rogowski (1989) offers an
economically determinist account of political coalitions and partisan cleavages in
Europe and around the world based on relative scarcities of domestic land, labor, and
capital in world markets and resulting group interests over trade openness or
protectionism under conditions of an expanding or a contracting world economy.
While yielding important novel insights, the analysis overstates the importance of
external economic exposure for the formation of political divides and competitive
dimensions, probably because it lacks an analysis of the conditions under which
collective mobilization of economic interests and their translation into party competition
takes place.
542 H E R B E R T K I T S C H E L T
4.2 The Transformation of Party Politics in
Post-Industrial Democracies
Lipset and Rokkan's (1967) famous dictum about the "freezing" of European party
systems in the 1920s was hugely overrated in the literature. What started out as
a simple observational suggestion in the conclusion to a lengthy comparativehistorical
analysis of European political cleavage formation was subsequently
blown up into a fundamental theoretical and empirical claim about the nature of
mature, institutionalized party systems. The empirical observation of relative party
system stability in Europe over some period of time, however, did not compel Lipset
and Rokkan to deny that such systems may get caught up in a profound process of
systemic dealignment and realignment (cf. Mair 1997, 4). At least three different
themes in the comparative literature about the transformation of party systems in
affluent post-industrial democracies deserve highlighting.
First, inspired by Lipset and Rokkan's work, many scholars have probed into
continuity or decline of existing European political cleavage structures. Studies of
aggregate party system volatility usually found only moderate increases (cf. Maguire
1983; Shamir 1984; Bartolini and Mair 1990). But individual-level voting analysis
shows a strong, though cross-nationally variable decline in conventional class voting
(cf. Franklin, Mackie, and Valen 1992). On the one hand, this gave rise to a perspective
that postulates a "dealignment" of voters from parties (Dalton, Flanagan, and Beck
1984; Dalton 2004). Post-industrialization has made especially educated citizens
distrustful of parties and prepared to engage in a variety of forms of political interest
mobilization that sidestep the electoral process. That trend is associated with declining
voter turnout, disjointed single-issue voting, and vanishing partisan identification,
resulting in a detachment of economic and social structures of conflict from
partisan-level divides.
As a second theme contradicting the dealignment perspective, other scholars
have emphasized the emergence of new partisan divides and competitive dimensions
with post-industrial economic structure. Realignments of political-economic interests
with the implosion of the manual working class, the differentiation of educationalprofessional
skills, and the rise of a vast non-profit sector of social services, often
configured around the welfare state, create new opportunities for political parties to
realign political divides and competitive dimensions (cf. Brooks, Nieuwbeerta, and
Manza 2006; Evans 1999; Knutsen 2006; Manza and Brooks 1999). Again, partisan
divides and competitive dimensions are no direct reflection of underlying social
change, but result from the strategic positioning of parties and their ability to craft
electoral coalitions (cf. Kitschelt 1994; Kitschelt and Rehm 2005). These party system
changes may not so much signal a demise of economic-distributive politics, as diagnosed
in the postmaterialism literature (Inglehart 1990, 1997), as a novel fusion of
economic interest alignments and demands about political and cultural governance.
The combination of economic and non-economic interests by entrepreneurial
politicians faced with cross-nationally varying strategic configurations among existing
parties is also at the heart of a burgeoning literature on new party formation and
P A R T Y S Y S T E M S
543
success in post-industrial democracies. While this literature initially focused on a
libertarian left (cf. Kitschelt 1988, 1989b; Redding and Viterna 1999), much more
attention has recently been devoted to the rise of extreme rightist parties in many
European polities and Anglo-Saxon settler democracies. While there is widespread
agreement on the socioeconomic transformations that bring about electoral constituencies
available for such parties (primarily manual laborers at different skill
levels and traditional small business owners, such as farmers, craftsmen and
shopkeepers, men with low skills more generally) and pit them against other groups
impervious to rightist political appeals (primarily highly trained professionals,
particularly women and especially in the social service sector), it is more contentious
how political opportunity structures have affected the nature of the radical right's
appeals and its electoral success (cf. Kitschelt 19950; Lubbers, Gilsberts, and Scheepers
2002; Norris 2005). Central controversies concern the extent to which the radical
right incorporates liberal market economics into its menu of political appeals (cf.
Cole 2005; Ivarsflaten 2005; Kitschelt 1995a; Schain, Zolbergi and Hossau 2002), the
causal efficacy of electoral laws in promoting or preventing the rise of new radical
rightist parties (Carter 2005; Golder 2003; Jackman and Volpert 1996; Norris 2005;
Veugelers and Magnan 2005) and the role the convergence and similarity among
conventional left and right parties in their policies and governing practices has played
for the success of new rightist parties (Carter 2005; Ignazi 2003; Kitschelt 1995a;
Meguid 2005; Norris 2005; van der Brug, Fennema, and Tillie 2005; Veugelers and
Magnan 2005).
A further interesting question of realignment concerns the way divisions over
European integration have inserted themselves into national party systems (cf. Gabel
1998; Hix 1999; Marks and Wilson 2000; Marks, Wilson, and Ray 2002; Marks and
Steenbergen 2004). In many countries, it is unlikely that European integration
becomes a competitive dimension in the sense specified above (cf. Mair 2000).
Beyond that, contextual conditions related to the perceived and anticipated consequences
of EU integration for national political economies may bring about a rather
diverse insertion of the EU issue into domestic politics (cf. Bringar, Jolly, and
Kitschelt 2004; Ray 2004; Scheve 2000).
A third and final theme concerns the extent to which citizen-politician relations in
contemporary post-industrial polities can still be conceived within a principal-agent
framework. Some have argued that the transition to capital-intensive campaign
strategies with an overwhelming role for the mass media and increasingly funded
by public party finance has created unaccountable "party cartels" impervious to
voter demands (Blyth and Katz 2005; Katz and Mair 1995, reprinted in Mair 1997),
while others have invoked the power of competition and voter exit to contradict that
thesis (Kitschelt 2000b). In other words, does the undeniable tendency of voters to
express greater dissatisfaction with parties than in previous decades indicate that
there is a crisis of political representation precipitated by unaccountable elites, or are
these misgivings by-products of weaker economic performance and structural economic
change that opens opportunities for partisan realignment?
544 H E R B E R T K I T S C H E L T
4.3 Party Systems in New Democracies of
the Developing World
Whereas comparative literature on Western OECD polities worries about the erosion
of relations of democratic accountability, students of democracy in developing
countries are preoccupied with the reverse question of whether accountability relations
and "institutionalized" party systems will ever emerge in the first place.
Particularly students of Latin American and post-communist politics have been
impressed by the high volatility of many parties and party systems signaling difficulty
in establishing lasting relations between voters and political agents (cf. Mainwaring
and Scully 1995; Mair 1997, ch. 8; Roberts and Wibbels 1999; Rose and Munro 2003).
In countries where party systems have developed some staying power, it is not
programmatic politics based on indirect exchange, but clientelistic principal-agent
relations that appear to dominate the scene and adapt to new constituencies and
political challenges, whether in South and South-East Asia (cf. Kohli 1990; Chandra
2004; Chhibber 1998; Krishna 2002; Sachsenroder 1998; Wilkinson 2006), in Latin
America (Fox 1994; Gibson 1997; Levitsky 2003) or post-communist Eastern Europe
(Hale 2006; Kitschelt et al. 1999). The persistence or demise of clientelistic conditions
does not simply depend on economic poverty and unequal asset distribution in a
polity, but also on the strategic incentives generated within the arena of party
competition to switch to a different accountability relationship (cf. Kitschelt and
Wilkinson 2006). Also weak performance of public sector enterprises or of publicly
regulated companies that are often shot through with clientelistic exchange relations
may affect how democratic political accountability relations evolve.
Upon closer inspection, within each region of the developing world the current
state of party system consolidation and the practices of principal-agent relations
varies widely. Both in post-communist Europe as well as in Latin America a
number of party systems have quite clearly structured programmatic political
cleavages and rather stable competitive partisan divides, particularly if we follow
Bartolini and Mair's (1990) focus not on the volatility of individual parties, but on
party blocs with roughly similar appeals within a cleavage system. A growing
literature has examined the extent and the nature of political cleavages and
competitive party divides in the post-communist region (cf. Bielasiak 2002; de
Waele 2004; Evans and Whitefield 1993, 2000; Kitschelt 1992, 1995b; Lewis 2000;
Pridham and Lewis 1996; Tavits 2005; Whitefield 2002). Particular attention has
been devoted to the insertion of the former communist ruling parties into democratic
partisan politicis (cf. Bozoki and Ishiyama 2002; Grzymala-Busse 2002).
Controversies surround both the descriptive characterization of the political
divides and competitive dimensions as well as the explanation for more or less
programmatic structuring. Is it a consequence of political experiences of the past
("legacies") in each country, of democratic institutions (such as electoral systems
and relations between the executive and the legislature), or of the momentous
political-economic reforms that generate new divides between interests?
P A R T Y S Y S T E M S
545
Comparative scholarship on Latin America has asked closely parallel questions.
Some authors have ventured to identify the historical origins, profile, and durability
of political cleavages in at least some party systems (Dix 1989; Coller and Collier 1991;
Coppedge 1998). Others have focused on general patterns of stability and change in
Latin American party systems in order to explore the causes of democratic party
system institutionalization (cf. Dix 1992; Mainwaring and Scully 1995; Geddes 2003).
In Latin America, just as in Eastern Europe, those party systems appear more consolidated
and structured around mechanisms of programmatic accountability in which
there had been other episodes of democratic competition before the current spell of
democratic competition beginning in the 1980s. Such episodes of broad political
mobilization enabled people to gain political experience and sometimes even to
"lock in" certain political economic achievements, such as the beginnings of a welfare
state, that provided a focal point to crystallize electorates around programmatic
alternatives, particularly in an era of conomic reform and market liberalization.
There is a curious asymmetry, however, when comparing Eastern Europe and
Latin America. In Latin America party system consolidation and programmatic
structuring tend to have undergone the greatest erosion in the 1990s and since
2000 precisely in countries with historically more established party systems. This
erosion is greatest in Venezuela, followed by Argentina, but also present to a lesser
extent even in Costa Rica, Uruguay, Mexico, and Chile. At the same time, Latin
American countries with always inchoate party systems show few signs of changing
that state of affairs. In Eastern Europe, by contrast, the polities with the most
promising historical priors for party system institutionalization around programmatic
accountability are also those that have achieved the comparatively greatest
institutionalization. But even many less hospitable places have shown signs of
moving toward patterns of programmatic accountability.
In Eastern Europe and also in South and South-East Asia sustained economic
growth for at least the past decade and often longer has most certainly benefited the
gradual establishment of robust structures of representation. In Latin America, by
contrast, the demise of import-substituting industrialization strategies in the 1980s
and the inability of political elites to embrace a definite new strategy of politicaleconomic
development, as evidenced by anemic growth and repeated monetary
stabilization crises, may have contributed not only to the region's continuing economic
hardship, but also the fragility of its democratic party systems.
5 C O N C L U S I O N
My review of the party system literature has been highly selective, driven by my
personal research interests in the area and an effort to stress certain agenda points for
future research. Thus I believe more emphasis has to be placed on the comparative
546 H E R B E R T K I T S C H E L T
study of the varieties of mechanisms that may govern the relationship between
principals and agents in democratic party systems. I also believe that in the study
of the "dimensionality" of party competition, more attention needs to be paid to the
distinction between social, political, and competitive partisan divides. Third, and
intimately linked to the previous point, the competitiveness of party systems deserves
better conceptualization and more intensive study than in the past. Conversely, I
submit that too much significance has been attached to certain relatively easily
measured macro-level properties of party systems, such as party system fragmentation,
polarization, and volatility, none of which are good measures of party system
competitiveness.
My treatment of party systems has ignored, however, any discussion of the concept
as independent variable. After all, we might develop concepts and theorems of party
systems not for their own sake, but as fruitful tools to study the consequences of
party competition for a variety of political and economic processes. Among them I
would count the formation of legislative and executive majorities, the resulting
process of policy formation and implementation, and ultimately the consequences
of party system dynamics for the stability and survival of the political regime form
itself. Since these topics are treated elsewhere in this volume, I could do without a
detailed discussion in this entry on party systems. At the same time, a more
sophisticated conceptualization of party systems, particularly of mechanisms of
democratic accountability and partisan competitiveness, may perform wonders in
improving the causal efficacy of explanations that employ party system attributes to
predict political economic developments and political regime trajectories.
REFERENCES
A D A M S , J., M E R R I L L , S., Ill, and G R O F M A N , A. 2005. A Unified Theory of Party Competition.
Cambridge: Cambridge University Press.
A L D R I C H , J. 1983. A Downsian spatial model with party activism. American Politicai Science
Review, 77: 974-90.
1995. Why Parties? The Origin and Transformation of Party Politics in America. Chicago:
Chicago University Press.
A M E R I C A N P O L I T I C A L S C I E N C E A S S O C I A T I O N . 1983. Political Science: The State of the Discipline.
1983, ed. S. M. Lipset. Washington, DC: APSA.
1993. Political Science: The State of the Discipline. 1993, ed. A. W. Finifter. Washington,
DC: APSA.
2002. Political Science: The State of the Discipline. 2002, ed. I. Katznelson and H. Milner.
New York: W. W. Norton.
A M E S , B. 2001. The Deadlock of Democracy in Brazil. Ann Arbor: University of Michigan Press.
B A R T O L I N I , S. 2000. The Political Mobilization of the European Left, 1860-1980: The Class
Cleavage. Cambridge: Cambridge University Press.
and M A I R , P. 1990. Identity, Competition and Electoral Availability. The Stability of
European Electorates 1885-1985. Cambridge: Cambridge University Press.
P A R T Y S Y S T E M S 547
BIELASIAK, J. 2002. The institutionalization of electoral and party systems in postcommunist
states. Comparative Politics, 34:189-210.
BLYTH, M., and KATZ, R. 2005. From catch-all politics to cartelisation. West European Politics,
28: 33-60.
BOZOKI, A., and ISHIYAMA, J. eds. 2002. The Communist Successor Parties of Central and
Eastern Europe. Armonk, NY: M. E. Sharpe.
BRINEGAR, A., JOLLY, S., and KITSCHELT, H. 2004. Varieties of capitalism and political
divisions over European integration. Pp. 62-89 in Dimensions of Contestation in the
European Union, ed. G. Marks and M. Steenbergen. Cambridge: Cambridge University
Press.
BROOKS, C, NIEUWBEERTA, P., and MANZA, J. 2006. Cleavage-based voting behavior in crossnational
perspective: evidence from six postwar democracies. Social Science Research, 35:
88-128.
BUDGE, I., and FARLIE, D. 1983. Explaining and Predicting Elections. London: Allen and Unwin.
ROBERTSON, D. and HEARL, D., eds. 1987. Ideology, Strategy and Party Change: Spatial
Analyses ofPost-War Election Programmes in Nineteen Democracies. Cambridge: Cambridge
University Press.
KLINGEMANN, H.-D., VOLKENS, A., BARA, J., and TANENBAUM, A. (with group of further
co-authors). 2001. Mapping Policy Preferences: Estimates for Parties, Electors, and
Governments 1945-1998. Oxford: Oxford University Press.
CAILLAUD, D., and TIROLE, J. 2002. Parties as political intermediaries. Quarterly Journal of
Economics, 117:1453-89.
CARTER, E. 2005. The Extreme Right in Western Europe: Success or Failure? Manchester:
Manchester University Press.
CHANDRA, K. 2004. Why Ethnic Parties Succeed: Patronage and Ethnic Headcounts in India.
Cambridge: Cambridge University Press.
CHHIBBER, P. K. 1998. Democracy without Associations. Ann Arbor: Michigan University Press.
COLE, A. 2005. Old right or new right? The ideological positioning of parties of the far right.
European Journal of Political Research, 44: 203-30.
COLLIER, D., and COLLIER, R. 1991. Shaping the Political Arena. Princeton: Princeton University
Press.
COPPEDGE, M. 1998. The evolution of Latin American party systems. Pp. 171-96 in Politics,
Society, and Democracy, ed. S. Mainwaring and A. Valenzuela. Boulder, Colo.: Westview
Press.
Cox, G. 1997. Making Votes Count. Cambridge: Cambridge University Press.
and MCCUBBINS, M. 1993. Legislative Leviathan: Party Government in the House. Berkeley
and Los Angeles: University of California Press.
DALTON, R. J. 2004. Parties without Partisans: Political Change in Advanced Industrial Democracies.
Oxford: Oxford University Press.
FLANAGAN, S., and BECK, P. A. 1984. Electoral Change in Advanced Industrial Democracies:
Realignment or Dealignment? Princeton: Princeton University Press.
DE WAELE, J.-M. ed. 2004. Les Clivages politiques en Europe centrale et orientale. Brussels:
Édition de l'Université de Bruxelles.
DIX, R. 1989. Cleavage structures and party systems in Latin America. Comparative Politics, 22:
23-37-
1992. Democratization and the institutionalization of Latin American political parties.
Comparative Political Studies, 24: 488-511.
DOWNS, A. 1957. An Economic Theory of Democracy. New York: Harper and Row.
DUVERGER, M. 1954. Political Parties. London: Methuen.
548 H E R B E R T K I T S C H E L T
ENELOW, J. M., and HINICH, M. eds. 1990. Advances in the Spatial Theory of Voting.
Cambridge, Mass.: Cambridge University Press.
EPSTEIN, L. D. 1975. Political parties. Pp. 229-78 in Handbook of Political Science, iv: NonGovernmental
Politics, ed. F. I. Greenstein and N. W. Polsby. Reading, Mass.: Addison-
Wesley.
ERICKSON, R. S., MACKUEN, M. B., and STIMSON, J. A. 2002. The Macro-Polity. Cambridge:
Cambridge University Press.
EVANS, G. ed. 1999. The End of Class Politics? Class Voting in Comparative Context. Oxford:
Oxford University Press.
and WHITEFIELD, S. 1993. Identifying the bases of party competition in eastern Europe.
British Journal of Political Science, 23: 521-48.
2000. Explaining the formation of electoral cleavages in post-communist
democracies. Pp. 36-68 in Elections in Central and Eastern Europe: The First Wave, ed.
H. D. Klingemann, E. Mochmann, and K. Newton. Berlin: Edition Sigma.
FIORINA, M. 1977. An outline for a model of party choice. American Journal of Political Science,
21: 601-25.
1997. Voting behavior. Pp. 391-414 in Perspectives on Public Choice, ed. D. C. Mueller.
Cambridge: Cambridge University Press.
2002. Parties, participation, and representation in America: old theories face new
realities. Pp. 511-41 in Political Science: The State of the Discipline, ed. H. V. Milner and
I. Katznelson. New York: Norton.
Fox, J. 1994. The difficult transition from clientelism to citizenship. World Politics, 46:151-84.
FRANKLIN, M. 2004. Voter Turnout and the Dynamics of Electoral Competition in Established
Democracies since 1945. Cambridge: Cambridge University Press.
MACKIE, T., and VALEN, H. eds. 1992. Electoral Change: Responses to Evolving Social and
Attitudinal Structures in Western Democracies. Cambridge: Cambridge University Press.
GABEL, M. 1998. Political support for European integration: an empirical test of five theories.
Journal of Politics, 60 (2): 333-54.
GEDDES, B. 2003. Paradigms and Sandcastles. Ann Arbor: University of Michigan Press.
GIBSON, E. 1997. The populist road to market reform: policy and electoral coalitions in Mexico
and Argentina. World Politics, 49: 339-70.
GOLDER, M. 2003. Explaining variation in the success of extreme right parties in Western
Europe. Comparative Political Studies, 36 (4): 432-66.
GOODIN, R. E., and KLINGEMANN, H.-D. eds. 1996. A New Handbook of Political Science.
Oxford: Oxford University Press.
GREENSTEIN, F. I., and POLSBY, N. W. eds. 1975. Handbook of Political Science, iv: NonGovernmental
Politics. Reading, Mass.: Addison-Wesley.
GROFMAN, B. 2004. Downs and two-party convergence. Annual Review of Politics, 7: 25-46.
GRZYMALA-BUSSE, A. M. 2002. Redeeming the Communist Past: The Regeneration of
Communist Parties in East Central Europe. Cambridge: Cambridge University Press.
2003. Redeeming the past: communist successor parties after 1989. Pp. 157-81 in
Capitalism and Democracy in Central and Eastern Europe, ed. G. Ekiert and S. E. Hanson.
Cambridge: Cambridge University Press.
HALE, H. 2006. Correlates of clientelism: political economy, politicized ethnicity, and postcommunist
transition. Forthcoming in Patrons, Clients, and Policies: Patterns of Democratic
Accountability and Political Competition, ed. H. Kitschelt and S. Wilkinson. Cambridge:
Cambridge University Press.
HARMEL, R., and ROBERTSON, J. D. 1984. Formation and success of new parties: a crossanalysis.
International Political Science Review, 6: 501-23.
P A R T Y S Y S T E M S 549
HAUSS, C, and RAYSIDE, D. 1978. The development of new parties in western democracies
since 1945. Pp. 31-57 in Political Parties: Development and Decay, ed. L. Maisel and J. Cooper.
Beverly Hills, Calif.: Sage.
HINICH, M., and MUNGER, M. 1994. Ideology and the Theory of Political Choice. Ann Arbor:
Michigan University Press.
Hix, S. 1999. Dimensions and alignments in European Union politics: cognitive constraints
and partisan responses. European Journal of Political Research, 35: 69-125.
HOROWITZ, D. 1985. Ethnic Groups in Conflict. Berkeley and Los Angeles: University of
California Press.
HUG, S. 2001. Altering Party Systems: Strategic Behavior and the Emergence of New Political
Parties in Western Democracies. Ann Arbor: University of Michigan Press.
HUNTINGTON, S. P. 1968. Political Order in Changing Societies. New Haven: Yale University
Press.
IGNAZI, P. 2003. Extreme Right Parties in Western Europe. Oxford: Oxford University Press.
INGLEHART, R. 1990. Culture Shift. Princeton: Princeton University Press.
1997- Modernization and Postmodernization. Princeton: Princeton University Press.
IVARSFLATEN, E. 2005. The vulnerable populist right parties: no economic realignment
fuelling their electoral success. European Journal of Political Research, 44: 465-92.
JACKMAN, R., and VOLPERT, K. 1996. Conditions favouring parties of the extreme right in
Western Europe. British Journal of Political Science, 26: 501-22.
KALYVAS, S. 1996. The Rise of Christian Democracy in Europe. Ithaca, NY: Cornell University
Press.
KATZ, R. S. 1980. A Theory of Parties and Electoral Systems. Baltimore: Johns Hopkins
University Press.
and CROTTY, W. eds. 2006. Handbook of Party Politics. London: Sage Publications.
and MAIR, P. 1995. Changing models of party organization and party democracy: the
emergence of the cartel party. Party Politics, 1 (1): 5-28. Reprinted in P. Mair, Party System
Change: Approaches and Interpretations. Oxford: Oxford University Press, 1997.
KEDAR, O. 2005. When moderate voters prefer extreme parties: policy balancing in parliamentary
elections. American Political Science Review, 99: 185-200.
KEEFER, P. 2005. Democratization and clientelism: why are young democracies badly
governed? World Bank Policy Research Paper 3594
KITSCHELT, H. 1988. The rise of left-libertarian parties in western democracies: explaining
innovation in competitive party systems. World Politics, 40: 194-234.
1989«. The internal politics of parties: the law of curvilinear disparity revisited. Political
Studies, 37: 400-21.
1989b. The Logics of Party Formation. Ithaca, NY: Cornell University Press.
1992. The formation of party systems in east central Europe. Politics and Society, 20: 7-50.
1994- The Transformation of European Social Democracy. Cambridge: Cambridge
University Press.
(in collaboration with A. ]. McGann). 19950. The Radical Right in Western Europe. Ann
Arbor: University of Michigan Press.
1995^. The formation of party cleavages in post-communist democracies: theoretical
propositions. Party Politics, 1: 447-72.
2000a. Linkages between citizens and politicians in democratic polities. Comparative
Political Studies, 33: 845-79.
2000 b. Citizens, politicians, and party cartellization: political representation and state
failure in post-industrial democracies. European Journal of Political Research, 37: 149-79.
550 H E R B E R T K I T S C H E L T
HAWKINS, K., ROSAS, G., and ZECHMEISTER, L. FORTHCOMING. Latin American Party
Systems.
and REHM, P. 2005. Work, family, and politics: foundations of electoral partisan alignments
in postindustrial democracies. Paper prepared for delivery at the 2005 Annual
Meeting of the American Political Science Association, Washington, DC.
and WILKINSON, S. eds. 2006. Patrons, Clients and Policies: Patterns of Democratic
Accountability and Political Competition. Cambridge: Cambridge University Press.
and ZECHMEISTER, E. 2003. Patterns of party competition and political accountability in
Latin America. Paper prepared for delivery at the 2003 Annual Meeting of the American
Political Science Association, Philadelphia.
MANSFELDOVA, Z., MARKOWSKI, R., and TOKA, G. 1999. Post-Communist Party Systems:
Competition, Representation, and Inter-Party Cooperation. Cambridge: Cambridge University
Press.
KNUTSEN, O. 2006. Social Class and Party Choice in Eight Countries: A Comparative Longitudinal
Study. Boulder, Colo.: Westview Press.
KOHLI, A. 1990. Democracy and Discontent: India's Growing Crisis of Governability. Cambridge:
Cambridge University Press.
KOLLMAN, K., MILLER, J., and PAGE, S. 1992. Adaptive parties in spatial elections. American
Political Science Review, 86: 929-37.
1998. Political parties and electoral landscapes. British Journal of Political
Science, 28:139-58.
KRISHNA, A. 2002. Active Social Capital: Tracing the Roots of Democracy and Development. New
York: Columbia University Press.
LAAKSO, M., and TAAGEPERA, R. 1979. Effective number of parties: a measure with application
to western Europe. European Journal of Political Research, 12: 3-27.
LAITIN, D. 2003. Comparative politics: the state of the subdiscipline. Pp. 630-59 in Political
Science: The State of the Discipline, ed. H. V. Milner and I. Katznelson. New York: W. W.
Norton.
LAVER, M. 2005. Policy and the dynamics of party competition. American Political Science
Review, 99: 263-82.
and HUNT, B. W. 1992. Policy and Party Competition. London: Routledge.
and SCHILPEROORD, M. 2005. The birth and death of political parties. Draft prepared for
Philosophical Transactions of the Royal Society B. Edinburgh, 30 July-5 August.
LEVITSKY, S. 2003. Transforming Labor-Based Parties in Latin America. Cambridge: Cambridge
University Press.
LEWIS, P. G. 2000. Political Parties in Post-Communist Eastern Europe. London: Routledge.
LIJPHART, A. 1999. Patterns of Democracy. New Haven: Yale University Press.
LIPSET, S. M., and ROKKAN, S. 1967. Cleavage structures, party systems, and voter alignments:
an introduction. Pp. 1-64 in Party Systems and Voter Alignments: Cross-National Perspectives,
ed. S. M. Lipset and S. Rokkan. New York: Free Press.
LUBBERS, M., GIJSBERTS, M., and SCHEEPERS, P. 2002. Extreme right-wing voting in Western
Europe. European Journal of Political Research, 41: 345-78.
LUEBBERT, G. 1991. Liberalism, Fascism, or Social Democracy: Social Classes and the Political
Origins of Regimes in Interwar Europe. New York: Oxford University Press.
LUPIA, A., and MCCUBBINS, M. 1998. The Democratic Dilemma: Can Citizens Learn What They
Need to Know? Cambridge: Cambridge University Press.
MCGANN, A. J. 2002. The advantages of ideological cohesion: a model of constituency
representation and electoral competition in multi-party democracies. Journal of Theoretical
Politics, 14: 37-70.
P A R T Y S Y S T E M S 551
MAGUIRE, M. 1983. Is there still persistence? Electoral change in western Europe, 1948-1979.
Pp. 67-94 in Western European Party Systems: Continuity and Change, ed. H. Daalder and P.
Mair. Beverly Hills, Calif.: Sage.
MAINWARING, S., and SCULLY, T. 1995. Introduction: party systems in Latin America. Pp. 1-35
in Building Democratic Institutions: Party Systems in Latin America, ed. S. Mainwaring and
T. Scully. Stanford, Calif.: Stanford University Press.
MAIR, P. 1997. Party System Change: Approaches and Interpretations. Oxford: Oxford
University Press, 1997.
2000. The limited impact of Europe on national party systems. West European Politics,
23: 27-51.
MANZA, J., and BROOKS, C. 1999. Social Cleavages and Political Change: Voter Alignments and
U.S. Party Coalitions. Oxford: Oxford University Press.
MARKS, G., and STEENBERGEN, M. eds. 2004. European Integration and Political Conflict.
Cambridge: Cambridge University Press.
and WILSON, C. 2000. The past in the present: a cleavage theory of party response to
European integration. British Journal of Political Science, 30: 433-59.
and RAY, L. 2002. National political parties and European integration. American
Journal of Political Science, 46: 585-94.
MAY, J. D. 1973. Opinion structure and political parties: the special law of curvilinear disparity.
Political Studies, 21: 135-51.
MAYHEW, D. R. 2000. Electoral realignments. Annual Review of Political Science, 3: 449-74.
MEGUID, B. 2005. Competition between unequals: the role of mainstream party strategy in
niche party success. American Political Science Review, 99: 347-60.
MERRILL, S., Ill, and GROFMAN, B. 1999. A Unified Theory of Voting: Directional and Proximity
Spatial Models. Cambridge: Cambridge University Press.
MILLER, G., and SCHOFIELD, N. 2003. Activists and partisan realignment in the United States.
American Political Science Review, 97 (2): 245-60.
MOLINAR, J. 1991. Counting the number of parties: an alternative index. American Political
Science Review, 85: 1383-91.
MORGENSTERN, S. 2004. Patterns of Legislative Politics. Cambridge: Cambridge University
Press.
NORRIS, P. 2005. Radical Right: Voters and Parties in the Electoral Market. Cambridge:
Cambridge University Press.
ORDESHOOK, P. 1997. The spatial analysis of elections and committees: four decades of
research. Pp. 247-70 in Perspectives on Public Choice, ed. D. C. Mueller. Cambridge:
Cambridge University Press.
PANEBIANCO, A. 1988. Political Parties: Organization and Power. Cambridge: Cambridge
University Press.
PAPPI, F. U. 1996. Political behavior: reasoning voters in multi-party systems. Pp. 255-74 in A
New Handbook of Political Science, ed. R. E. Goodin and H.-D. Klingemann. Oxford:
Oxford University Press.
PEDERSEN, M. 1983. Changing patterns of electoral volatility in European party systems,
1948-1977: explorations in explanation. Pp. 29-66 in Western European Party Systems:
Continuity and Change, ed. H. Daalder and P. Mair. Beverly Hills, Calif.: Sage.
PIATTONI, S. ed. 2001. Clientelism, Interests, and Democratic Representation. Cambridge:
Cambridge University Press.
POWELL, G. B. 2000. Elections as Instruments of Democracy: Majoritarian and Proportional
Visions. New Haven: Yale University Press.
552 H E R B E R T K I T S C H E L T
PRIDHAM, G., and LEWIS, P. eds. 1996. Stabilizing Fragile Democracies: Comparing New Party
Systems in Southern and Eastern Europe. London: Routledge.
PRZEWORSKI, A., and SPRAGUE, J. 1986. Paper Stones. Chicago: University of Chicago Press.
RABINOWITZ, G., and MCDONALD, S. E. 1989. A directional theory of issue voting. American
Political Science Review, 83: 93-121.
RAE, D. W. 1967. The Political Consequences of Electoral Laws. New Haven: Yale University
Press.
and TAYLOR, M. 1970. The Analysis of Cleavages. New Haven: Yale University Press.
RAY, L. 2004. Don't rock the boat: expectations, fears, and opposition to EU level policymaking.
Pp. 51-61 in Dimensions of Contestation in the European Union, ed. G. Marks and
M. Steenbergen. Cambridge: Cambridge University Press.
REDDING, K., and VITERNA, J. 1999. Political demands, political opportunities: explaining the
differential success of left-libertarian parties. Social Forces, 78: 491-510.
RIKER, W. 1982. Liberalism versus Populism. San Francisco: Freeman.
1986. The Art of Political Manipulation. New Haven: Yale University Press.
ROBERTS, K., and WIBBELS, E. 1999. Party systems and electoral volatility in Latin America: a
test of economic, institutional, and structural explanations. American Political Science
Review, 93: 575-90.
ROBERTSON, D. 1976. A Theory of Party Competition. New York: Wiley.
ROEMER, J. 2001. Political Competition: Theory and Applications. Cambridge, Mass.: Harvard
University Press.
ROGOWSKI, R. 1989. Commerce and Coalitions. Princeton: Princeton University Press.
ROSE, R., and MUNRO, R. 2003. Elections and Parties in New European Democracies. Washington,
DC: Congressional Quarterly Press.
RUGGIE, J. G. 1989. International structure and international transformation: space, time, and
method. Pp. 21-35 in Global Changes and Theoretical Challenges, ed. E.-O. Czempiel and
}. Rosenau. Lexington, Mass.: D. C. Heath.
—•—1993. Territoriality and beyond: problematizing modernity in international relations.
International Organization, 47:139-74.
SACHSENRÖDER, W. 1998. Party politics and democratic development in East and Southeast
Asia: a comparative view. Pp. 1-35 in Political Party Systems and Democratic Development in
East and Southeast Asia, vol. i, ed. W. Sachsenröder and U. E. Frings. Ashgate: Aldershot.
SAMUELS, D. 2004. From socialism to social democracy: party organization and the
transformation of the Workers' Party in Brazil. Comparative Political Studies, 37:
999-1024.
SANI, G., and SARTORI, G. 1983. Polarization, fragmentation and competition in western
democracies. Pp. 307-340 in Western European Party Systems. Continuity and Change, ed.
H. Daalder and P. Mair. Beverly Hills, Calif.: Sage.
SARTORI, G. 1968. The sociology of parties: a critical review. Pp. 1-25 in Party Systems, Party
Organisation and the Politics of the New Masses, ed. O. Stammer. Berlin: Institut für
Politische Wissenschaften.
—•—1976. Parties and Party Systems: A Framework for Analysis. Cambridge: Cambridge
University Press.
1986. The influence of electoral systems: faulty laws or faulty method? Pp. 43-68 in
Electoral Laws and their Political Consequences, ed. B. Grofman and A. Lijphart. New York:
Agathon Press.
SCHAIN, M., ZOLBERG, A., and HOSSAU, P. eds. 2002. Shadows over Europe: The Development
and Impact of the Extreme Right in Western Europe. Houndmills: Palgrave Macmillan.
P A R T Y S Y S T E M S
553
SCHEVE, K. 2000. Comparative context and public preferences over regional economic
integration. Paper presented at the Annual Meeting of the American Political Science
Association. Washington, DC.
SCHLESINGER, J. 1984. On the theory of party organization. Journal of Politics, 46: 369-400.
SCHMIDT, S. W., GUASTI, L., LAND, C. H., and SCOTT, J. C. eds. 1977. Friends, Followers, and
Factions. Berkeley and Los Angeles: University of California Press.
SCHOFIELD, N. 2003. Valence competition in the spatial stochastic model. Journal of Theoretical
Politics, 15: 371-83.
2004. Equilibrium in the spatial "valence" model of politics. Journal of Theoretical
Politics, 16: 447-81.
SCOTT, J. C. 1969. Corruption, machine politics, and political change. American Political
Science Review, 62: 1142-58.
SHAMIR, M. 1984. Are western party systems "frozen"? A comparative dynamic analysis.
Comparative Political Studies, 12: 35-79.
SHEFTER, M. 1994. Political Parties and the State: The American Historical Experience. Princeton
University Press.
SHEPSLE, K. 1991. Models of Multiparty Electoral Competition. Chur: Harwood Academic
Publishers.
SNYDER, J. M., Jr., and TING, M. M. 2002. An informational rationale for political parties.
American Journal of Political Science, 46: 90-110.
SORAUF, F. J. 1964. Party Politics in America. Boston: Little, Brown.
SPRUYT, H. 1994. The Sovereign State and its Competitors. Princeton: Princeton University
Press.
STIMSON, J. 2005. Tides of Consent: How Public Opinion Shapes American Politics. Cambridge:
Cambridge University Press.
STOKES, D. 1963. Spatial models of party competition. American Political Science Review, 57:
368-77STROM,
K. 19900. A behavioral theory of competitive political parties. American Journal of
Political Science, 34: 565-98.
1990b. Minority Government and Majority Control. Cambridge: Cambridge University
Press.
TAAGEPERA, R., and GROFMAN, B. 1985. Rethinking Duverger's law: predicting the effective
number of parties in plurality and PR systems: parties minus issues equals one. European
Journal of Political Research, 13: 341-53.
TAVITS, M. 2005. The development of stable party support: electoral dynamics in postcommunist
Europe. American Journal of Political Science, 49 (2): 283-98.
2006. Party systems in the making, the emergence and success of new parties in new
democracies. British Journal of Political Science (forthcoming).
TSEBELIS, G. 2002. Veto Players. Princeton: Princeton University Press.
VAN DER BRUG, W, FENNEMA, M., and TILLIE, J. 2005. Why some anti-immigrant parties fail
and others succeed: a two-step model of aggregate electoral support. Comparative Political
Studies, 38: 537-73VEUGELERS,
J., and MAGNAN, A. 2005. Conditions of far-right strength in contemporary
Western Europe: an application of Kitschelt's theory. European Journal of Political Research,
44: 837-60.
WALTZ, K. 1954. Man, the State, and War: A Theoretical Analysis. New York: Columbia
University Press.
1979- Theory of International Politics. Reading, Mass.: Addison-Wesley.
554 H E R B E R T K I T S C H E L T
WEINGAST, B. 1998. Political stability and civil war: institutions, commitments, and American
democracy. Pp. 148-93 in Analytical Narratives, ed. R. H. Bates, A. Greif, M. Levi, J.-L.
Rosenthal, and B. R. Weingast. Princeton: Princeton University Press.
WHITEFIELD, S. 2002. Political cleavages and post-communist politics. Annual Review of
Political Science, 5:181-200.
WILKINSON, S. I. 2004. Votes and Violence: Electoral Competition and Ethnic Riots in India.
Cambridge: Cambridge University Press.
2006. Explaining changing patterns of party-voter linkages in India. Pp. 110-40 in
Patrons, Clients and Policies, ed. H. Kitschelt and S. I. Wilkinson. Cambridge: Cambridge
University Press.