5
1
9780472087679
By James Adams
All right reserved.
ISBN: 0472087673
CHAPTER 1
Political Representation and Responsible Party Government
1.1. Introduction
The linkage between the mass public and political decision makers is one of the central topics in the study of contemporary democracies (Dalton 1996). The study of political representation has stimulated extensive theoretical and empirical work concerning both the process through which citizens can influence the decisions of political elites and the degree to which they exercise such influence.
While research focusing on American politics has frequently conceptualized the representation process as one based upon the connections between the opinions of individual legislators and the policy preferences of the geographically based constituencies they represent (see, e.g., Miller and Stokes 1963; Achen 1975, 1978; and Page et al. 1988), political scientists working in non-American settings have increasingly focused on a theoretical model of representation based upon political parties as collectives. According to this responsible party government model of representation, it is political parties, not individual legislators, that are the primary vehicles that articulate citizens' policy beliefs and convert them into public policies. This notion of political representation appears to be appropriate outside of the United States because in non-American settings the members of each party's parliamentary delegation typically act in unison (see, e.g., Harmel and Janda 1982; and Thomassen 1994) so that it makes sense to focus on political parties as collectives, rather than on the behavior of individual parliamentary representatives. Thus, Giovanni Sartori maintains that "citizens in Western democracies are represented through and by parties. This is inevitable" (1968, 471, italics original; also quoted in Dalton 1996).
Indeed, important elements of the responsible party model of political representation may even apply to the United States. Some scholars argue that the policy link between party elites and their supporters is relevant to policy representation in the United States (see, e.g., Backstrom 1977; Bishop and Frankovic 1981; and Herrera, Herrera, and Smith 1992) or even that in practice policy representation in the United States revolves around parties, not individual legislators (see Ansolabehere, Snyder, and Stewart 1999).
The responsible party government model has inspired extensive theorizing about the necessary conditions that would make it possible for citizens to select parties that represent their policy beliefs and for parties' parliamentary delegations in turn to represent their supporters' preferences (Dalton 1985; Powell 1982, 1989; Iversen 1994a, 1994b; Ranney 1962). While many such preconditions have been proposed, three requirements relating to the behavior of mass electorates and the behavior of party elites appear especially crucial for responsible party government.
Of course, each of the above desirata--policy divergence and stability on the part of political parties and policy voting on the part of the electorate--need not be "completely" satisfied in order to ensure responsible party government. Parties may alter their policies to some extent without unduly confusing the electorate, in addition to which the emergence of new policy issues typically requires some change in parties' policy images. With respect to policy voting, voters may be swayed to some degree by factors not directly related to policies, such as comparative evaluations of party leaders' competence and integrity, without completely severing the connection between the policies parties propose and the views of their supporters. However, it seems plausible that the more closely the behavior of parties and voters approximates the three conditions just outlined the better the prospects are for responsible party government.
1. Policy divergence among the parties contesting the election.
Elections should involve competition between parties that offer divergent policy visions so that voters have meaningful electoral choices (Dalton 1985; Powell 1989). Without such policy divergence, electors have little opportunity to influence government policies through their votes. Furthermore, policy convergence among the competing parties may well leave large portions of the electorate without an attractive party for which to vote, in which case there is little reason to expect agreement between the policies proposed by party elites and the policy preferences of their supporters.
2. Policy stability on the part of the parties contesting the election.
Parties should present reasonably stable policies over time, first, because drastic policy shifts may leave voters confused concerning what the party actually stands for (Bartels 1986), which in turn impairs the electorate's ability to influence government policy through its votes. Second, even when voters are well informed about the parties' current policies, they will have difficulty translating their own preferences into votes if they perceive a high probability that the parties will change their policy positions. When these requirements on the behavior of party elites are satisfied--and when additionally parties vote as a bloc in parliament, as is typically the case outside the United States--then voters' choices of parties provide them with a method of exercising control over the actions of individual legislators and through these over the affairs of government (Dalton 1996, chap. 11). However, there is an additional condition the electorate must satisfy in order to exercise this policy control.
3. Policy-voting on the part of the electorate.
Voters must base their decisions at least in part on comparative evaluations of the competing parties' policy programmes in order to directly influence government policy outputs. Without such policy voting, election results do not provide a meaningful referendum on the competing policy visions the parties propose. Furthermore, if voters are not motivated by policy considerations there is little reason to expect a close match between the policies parties propose and the beliefs of their supporters, and thus the citizen-elite policy linkage that underpins the responsible party model is jeopardized.
Of course, each of the above desirata--policy divergence and stability on the part of political parties and policy voting on the part of the electorate--need not be "completely" satisfied in order to ensure responsible party government. Parties may alter their policies to some extent without unduly confusing the electorate, in addition to which the emergence of new policy issues typically requires some change in parties' policy images. With respect to policy voting, voters may be swayed to some degree by factors not directly related to policies, such as comparative evaluations of party leaders' competence and integrity, without completely severing the connection between the policies parties propose and the views of their supporters. However, it seems plausible that the more closely the behavior of parties and voters approximates the three conditions just outlined the better the prospects are for responsible party government.
In summary, the responsible party government model directs scholars' attention to the links between political parties and the electorate. In particular, it suggests that, to the extent that competing parties present divergent, stable policies and citizens use these policies as the basis for their voting decisions, parties provide effective vehicles for representing the electorate's political beliefs. According to this model, an important criterion for assessing the degree to which parties provide faithful representation is to compare the "dyadic correspondence" between the policy positions a given party proposes, on the one hand, and the beliefs of the party's supporters on the other. When the conditions outlined above are met, the degree of dyadic correspondence should be high.
1.2. The Problem
In the decades following World War II, political scientists operating from two different research perspectives argued for disturbing conclusions, suggesting that the conditions for responsible party government would not be satisfied in most contemporary democracies. One approach, dating back to Anthony Downs's (1957) An Economic Theory of Democracy, involves spatial models of party competition. Using the assumption of a policy-oriented electorate, spatial modelers attempt to deduce the policy programmes that political parties will present in order to win elections. The second approach, dating back to The American Voter (Campbell et al. 1960) and before, is behavioral research. Behaviorists emphasize the empirical study of voting behavior, and they typically analyze a variety of voter motivations in addition to policy voting such as party identification; group loyalties rooted in class, religion, or ethnic identities; comparative judgments of party leaders; and retrospective evaluations of incumbent performance.
Departing from quite different perspectives on how voters decide, each research tradition has produced results that call the preconditions for responsible party government into question. In the case of spatial modeling, the most basic result is the median voter theorem--that in elections involving exactly two parties competing over a single policy dimension the parties will converge to the position of the median voter. This result implies that the parties will present stable policies in that the parties reach an equilibrium--that is, a policy configuration such that neither party has incentives to change its position. However, this two-party equilibrium also entails policy similarity in that the parties present identical policies and thus fail to offer the electorate divergent positions that provide voters with meaningful choices. And, while results are somewhat different in two-party competition over multiple policy dimensions, recent theoretical work suggests that in these situations vote-seeking parties can again be expected to converge toward extremely similar, centrist positions. In toto, the literature on two-party competition suggests that vote-seeking parties will present similar positions that deny the electorate a meaningful choice between distinct policy visions, which is a prerequisite for responsible party government.
Although early spatial modeling results applied almost exclusively to two-party elections (but see Downs 1957, 122-32), the theoretical prediction that vote-seeking parties will present similar positions extends to three-party competition. Furthermore, such multiparty elections also encourage a high degree of policy instability in the sense that the parties have electoral incentives to continually leapfrog their competitors. A simple example illustrates these points. Consider the scenario illustrated in figure 1.1A, which shows three parties A, B, and C, located along a single policy dimension, which for convenience I label the Left-Right dimension. Under the standard Downsian logic, parties A and C are motivated to converge toward the center (fig. 1.1B), thereby "squeezing" party B because this centrist positioning gains additional support from Center-Left voters (who switch their support from B to A) and Center-Right voters (who switch their support from B to C). However, when party B is squeezed between A and C, then B is motivated to leapfrog one of its competitors, thus locating either to A's immediate left or C's immediate right (fig. 1.1C illustrates the situation in which B leapfrogs A). However, if B leapfrogs party A, then A is squeezed between B and C and is motivated to leapfrog one of its competitors in turn. This process can be expected to continue indefinitely, with the parties continually leapfrogging but always locating near the center of the policy space. Thus, in three-party elections spatial theory predicts that, from the perspective of responsible party government, the parties' behavior offers voters the worst of both worlds, namely, a scenario in which the parties' positions are unstable (in the sense that they consistently leapfrog) but they nonetheless present extremely similar, centrist positions that deny the electorate the choice between distinct policy alternatives. While results differ somewhat when four or more parties compete--roughly speaking, the problem of policy instability grows more severe and the problem of similarity somewhat less severe--the central intuitions developed here, that parties' policies will be unstable, in that they will continually leapfrog, and that rival parties' policies will be similar, typically extend to systems with any number of parties.
From their individual-level perspective, behavioral researchers' empirical findings call into question the requirement for policy voting on the part of the electorate, which appears to be essential for the party-voter policy linkage that underlies responsible party government. Behaviorists' results suggest that, although policies matter at least to some voters, their decisions are also readily swayed by considerations such as comparative evaluations of the party leaders' competence and integrity, economic conditions, group loyalties based upon class, religion, geography, and ethnicity, party identification, and so on. This conclusion, which was originally reached with respect to American voters (see, e.g., Campbell et al. 1960; Fiorina 1981; and Key 1966), has been extended outward over the past 30 years to encompass the electorates of most Western democracies (Alvarez, Bowler, and Nagler 1996; Butler and Stokes 1969; Converse and Pierce 1986; Dow 1997b, 1997c; Klingemann et al. 1994; Lewis-Beck 1988; Listhaug 1989; Pierce 1995; Rose and McAllister 1990; Strom and Svasand 1997; Whitten and Palmer 1996). Although, as I outline in chapter 2, the relative electoral impact of policy motivations and nonpolicy influences is a matter of heated scholarly debate, in toto the studies cited here strongly suggest that the cumulative impact of voters' nonpolicy motivations upon the vote is at least as great as--and perhaps much greater than--the impact of policy voting.
As I shall emphasize in greater detail in chapter two, the distinction between voting based upon policy issues and voting based on non-policy-related motivations is not always clear. For instance, voters' retrospective evaluations of incumbent performance, their reactions to government scandals, and their comparative evaluations of party leaders' competence and integrity can reasonably be described as issue voting. According to Donald Stokes, these valence issues "involve the linking of the parties with some condition that is positively or negatively evaluated by the electorate" (1963, 373). This contrasts with position issues such as Left-Right ideology, welfare spending, and the death penalty, upon which voters (and parties) may disagree. My central point is that when voters deemphasize position issues election outcomes do not provide referendums on the parties' competing (positional) policy visions and, furthermore, that the link between parties' policies and their supporters' beliefs--which are assessed by comparing dyadic correspondence on positional issues--is threatened. Therefore, throughout this book I refer to valence issues as nonpolicy related, reserving the term policy voting for voting over positional issues.
This result suggests that it may be impossible to simultaneously satisfy the requirements of policy stability and policy divergence on the part of political parties, and of policy voting on the part of the electorate, that are central to responsible party government. If voters are policy oriented, as spatial modelers assume, then (with the exception of two-party competition along a single policy dimension) parties' policies are likely to be unstable in the sense that the parties continually leapfrog, and their policies will also be similar, so that voters are not afforded the option of choosing between truly distinct policy visions. If voters deemphasize policies, as behavioral researchers suggest, then parties' policies may not be representative of their supporters' beliefs since policies are not the primary basis of these supporters' party preferences.
1.3. The Puzzle
The findings of spatial modelers and behavioral researchers suggest that political representation through responsible party government is unattainable because the preconditions for responsible party government outlined in section 1.1 are not met in most Western democracies. Yet, in surveying the polities of Western Europe and North America, what is striking is the extent to which this prediction is not borne out by empirical observation. Contrary to spatial modelers' predictions of policy instability and policy similarity, empirical studies conclude that political parties present relatively stable policy programmes over time while rarely leapfrogging each other (see, e.g., Budge 1994) and furthermore that parties differentiate their policies, thereby presenting voters with a wide range of alternatives (Castles and Mair 1984; Huber and Inglehart 1995; Laver and Budge 1992). Thus, party systems are characterized by policy stability and policy divergence. And, while the behavioralists' finding that voters are motivated in large part by nonpolicy considerations seems to emperil mass-elite policy linkages, comparative research supports the proposition that in most party systems political parties provide reasonably faithful policy representation in that the policies parties propose reflect the views of their supporters (see, e.g., Dalton 1985; Holmberg 1989; Iversen 1994b; and Jennings 1992).
While I present data in subsequent chapters on empirical findings of policy divergence, policy stability, and policy representation, it is worth briefly surveying some of this evidence here. With respect to policy divergence, table 1.1 reproduces results reported by Huber and Inglehart (1995, app. 2) on the mean Left-Right positions that expert observers assigned to the major political parties in several Western European party systems and the United States as of 1993. The results indicate that experts perceive that each party system presents its voters with a wide range of ideological options. Alternative methodologies for estimating party positions, such as analyses of roll call voting, studies of party manifestos, and opinion surveys of party elites, confirm these experts' perceptions that party systems in most Western democracies are characterized by policy divergence not policy similarity (see, e.g., Budge 1994; Budge et al. 1987; Castles and Mair 1984; Dodd 1976; Inglehart and Klingeman 1987; and Sani and Sartori 1983).
With respect to policy stability, the most comprehensive empirical study of the evolution of party policies in postwar democracies has been conducted by Ian Budge and his colleagues at the Manifesto Research Group (MRG) (Klingemann et al. 1994; Laver and Budge 1992; Budge et al. 1987). Through comparative codings of the election programmes of the major parties in some 20 democracies during the postwar period the Manifesto Research Group has assigned positions to parties along a variety of policy dimensions, including an overall Left-Right summary (I review the details of the MRG's methodology in chapter 3). Table 1.2 reproduces the results of a temporal analysis of the MRG's codings reported by Budge (1994), which bears on policy stability in postwar democracies. Column 3 reports, for every postwar Western European democracy plus the United States, the proportions of party platforms that were coded as being outside the party's traditional "ideological area," defined as left of center for the Socialist, Communist, Labour, and Democratic Parties and as right of center for the Conservative, Christian Democratic, and Republican Parties. The results indicate that parties virtually never shift their Left-Right position outside their traditional ideological area. In addition, columns 1 and 2, which report the frequency with which parties leapfrog each other--that is, the extent to which parties shift their Left-Right orientation relative to rival parties-- show that leapfrogging has been rare and that when it occurs it usually involves parties that are contiguous along the Left-Right continuum. These results testify that throughout the postwar period political parties in Western democracies have presented remarkably stable policy images.
Continues...
Party Competition and Responsible Party Government: A Theory of Spatial Competition Based Upon Insights from Behavioral Voting Research available in Paperback, eBook
Party Competition and Responsible Party Government: A Theory of Spatial Competition Based Upon Insights from Behavioral Voting Research
by James Frolik Adams
James Frolik Adams
- ISBN-10:
- 0472087673
- ISBN-13:
- 9780472087679
- Pub. Date:
- 01/01/2001
- Publisher:
- University of Michigan Press
- ISBN-10:
- 0472087673
- ISBN-13:
- 9780472087679
- Pub. Date:
- 01/01/2001
- Publisher:
- University of Michigan Press
Party Competition and Responsible Party Government: A Theory of Spatial Competition Based Upon Insights from Behavioral Voting Research
by James Frolik Adams
James Frolik Adams
$30.95
Buy New
$30.95Buy Used
$22.06
$30.95
-
-
SHIP THIS ITEM
Temporarily Out of Stock Online
Please check back later for updated availability.
-
30.95
In Stock
Overview
In countries with multiparty political systems, we assume--if the system is going to work--that parties have relatively stable positions on policy, that these positions diverge, and that voters make choices based on policy preferences. Yet much of the research on voter behavior and party competition does not support these assumptions.
In Party Competition, James Adams applies the insights of behavioral research to an examination of the policy strategies that political parties (and candidates) employ in seeking election. He argues that vote-seeking parties are motivated to present policies that appeal to voters, whose bias toward these policies is based in part on reasons that have nothing to do with policy. He demonstrates that this strategic logic has profound implications for party competition and responsible party government.
Adams's innovative fusion of research methodologies presents solutions to issues of policy stability and voter partisanship. His theory's supported by an in-depth analysis of empirical applications to party competition in Britain, France, and the United States in the postwar years.
Party Competition and Responsible Party Government will appeal to readers interested in the study of political parties, voting behavior and elections, as well as to scholars specializing in French, British, and American politics.
James Adams is Assistant Professor of Political Science, University of California, Santa Barbara.
In Party Competition, James Adams applies the insights of behavioral research to an examination of the policy strategies that political parties (and candidates) employ in seeking election. He argues that vote-seeking parties are motivated to present policies that appeal to voters, whose bias toward these policies is based in part on reasons that have nothing to do with policy. He demonstrates that this strategic logic has profound implications for party competition and responsible party government.
Adams's innovative fusion of research methodologies presents solutions to issues of policy stability and voter partisanship. His theory's supported by an in-depth analysis of empirical applications to party competition in Britain, France, and the United States in the postwar years.
Party Competition and Responsible Party Government will appeal to readers interested in the study of political parties, voting behavior and elections, as well as to scholars specializing in French, British, and American politics.
James Adams is Assistant Professor of Political Science, University of California, Santa Barbara.
Product Details
| ISBN-13: | 9780472087679 |
|---|---|
| Publisher: | University of Michigan Press |
| Publication date: | 01/01/2001 |
| Edition description: | New Edition |
| Pages: | 250 |
| Product dimensions: | 6.00(w) x 9.00(h) x (d) |
About the Author
James Adams is Assistant Professor of Political Science, University of California, Santa Barbara.
Read an Excerpt
Party Competition and Responsible Party Government: a Theory of Spatial Competition Based Upon Insights from Behavioral Voting Research
By James Adams
University of Michigan Press
Copyright © 2001 James AdamsAll right reserved.
ISBN: 0472087673
CHAPTER 1
Political Representation and Responsible Party Government
1.1. Introduction
The linkage between the mass public and political decision makers is one of the central topics in the study of contemporary democracies (Dalton 1996). The study of political representation has stimulated extensive theoretical and empirical work concerning both the process through which citizens can influence the decisions of political elites and the degree to which they exercise such influence.
While research focusing on American politics has frequently conceptualized the representation process as one based upon the connections between the opinions of individual legislators and the policy preferences of the geographically based constituencies they represent (see, e.g., Miller and Stokes 1963; Achen 1975, 1978; and Page et al. 1988), political scientists working in non-American settings have increasingly focused on a theoretical model of representation based upon political parties as collectives. According to this responsible party government model of representation, it is political parties, not individual legislators, that are the primary vehicles that articulate citizens' policy beliefs and convert them into public policies. This notion of political representation appears to be appropriate outside of the United States because in non-American settings the members of each party's parliamentary delegation typically act in unison (see, e.g., Harmel and Janda 1982; and Thomassen 1994) so that it makes sense to focus on political parties as collectives, rather than on the behavior of individual parliamentary representatives. Thus, Giovanni Sartori maintains that "citizens in Western democracies are represented through and by parties. This is inevitable" (1968, 471, italics original; also quoted in Dalton 1996).
Indeed, important elements of the responsible party model of political representation may even apply to the United States. Some scholars argue that the policy link between party elites and their supporters is relevant to policy representation in the United States (see, e.g., Backstrom 1977; Bishop and Frankovic 1981; and Herrera, Herrera, and Smith 1992) or even that in practice policy representation in the United States revolves around parties, not individual legislators (see Ansolabehere, Snyder, and Stewart 1999).
The responsible party government model has inspired extensive theorizing about the necessary conditions that would make it possible for citizens to select parties that represent their policy beliefs and for parties' parliamentary delegations in turn to represent their supporters' preferences (Dalton 1985; Powell 1982, 1989; Iversen 1994a, 1994b; Ranney 1962). While many such preconditions have been proposed, three requirements relating to the behavior of mass electorates and the behavior of party elites appear especially crucial for responsible party government.
Of course, each of the above desirata--policy divergence and stability on the part of political parties and policy voting on the part of the electorate--need not be "completely" satisfied in order to ensure responsible party government. Parties may alter their policies to some extent without unduly confusing the electorate, in addition to which the emergence of new policy issues typically requires some change in parties' policy images. With respect to policy voting, voters may be swayed to some degree by factors not directly related to policies, such as comparative evaluations of party leaders' competence and integrity, without completely severing the connection between the policies parties propose and the views of their supporters. However, it seems plausible that the more closely the behavior of parties and voters approximates the three conditions just outlined the better the prospects are for responsible party government.
1. Policy divergence among the parties contesting the election.
Elections should involve competition between parties that offer divergent policy visions so that voters have meaningful electoral choices (Dalton 1985; Powell 1989). Without such policy divergence, electors have little opportunity to influence government policies through their votes. Furthermore, policy convergence among the competing parties may well leave large portions of the electorate without an attractive party for which to vote, in which case there is little reason to expect agreement between the policies proposed by party elites and the policy preferences of their supporters.
2. Policy stability on the part of the parties contesting the election.
Parties should present reasonably stable policies over time, first, because drastic policy shifts may leave voters confused concerning what the party actually stands for (Bartels 1986), which in turn impairs the electorate's ability to influence government policy through its votes. Second, even when voters are well informed about the parties' current policies, they will have difficulty translating their own preferences into votes if they perceive a high probability that the parties will change their policy positions. When these requirements on the behavior of party elites are satisfied--and when additionally parties vote as a bloc in parliament, as is typically the case outside the United States--then voters' choices of parties provide them with a method of exercising control over the actions of individual legislators and through these over the affairs of government (Dalton 1996, chap. 11). However, there is an additional condition the electorate must satisfy in order to exercise this policy control.
3. Policy-voting on the part of the electorate.
Voters must base their decisions at least in part on comparative evaluations of the competing parties' policy programmes in order to directly influence government policy outputs. Without such policy voting, election results do not provide a meaningful referendum on the competing policy visions the parties propose. Furthermore, if voters are not motivated by policy considerations there is little reason to expect a close match between the policies parties propose and the beliefs of their supporters, and thus the citizen-elite policy linkage that underpins the responsible party model is jeopardized.
Of course, each of the above desirata--policy divergence and stability on the part of political parties and policy voting on the part of the electorate--need not be "completely" satisfied in order to ensure responsible party government. Parties may alter their policies to some extent without unduly confusing the electorate, in addition to which the emergence of new policy issues typically requires some change in parties' policy images. With respect to policy voting, voters may be swayed to some degree by factors not directly related to policies, such as comparative evaluations of party leaders' competence and integrity, without completely severing the connection between the policies parties propose and the views of their supporters. However, it seems plausible that the more closely the behavior of parties and voters approximates the three conditions just outlined the better the prospects are for responsible party government.
In summary, the responsible party government model directs scholars' attention to the links between political parties and the electorate. In particular, it suggests that, to the extent that competing parties present divergent, stable policies and citizens use these policies as the basis for their voting decisions, parties provide effective vehicles for representing the electorate's political beliefs. According to this model, an important criterion for assessing the degree to which parties provide faithful representation is to compare the "dyadic correspondence" between the policy positions a given party proposes, on the one hand, and the beliefs of the party's supporters on the other. When the conditions outlined above are met, the degree of dyadic correspondence should be high.
1.2. The Problem
In the decades following World War II, political scientists operating from two different research perspectives argued for disturbing conclusions, suggesting that the conditions for responsible party government would not be satisfied in most contemporary democracies. One approach, dating back to Anthony Downs's (1957) An Economic Theory of Democracy, involves spatial models of party competition. Using the assumption of a policy-oriented electorate, spatial modelers attempt to deduce the policy programmes that political parties will present in order to win elections. The second approach, dating back to The American Voter (Campbell et al. 1960) and before, is behavioral research. Behaviorists emphasize the empirical study of voting behavior, and they typically analyze a variety of voter motivations in addition to policy voting such as party identification; group loyalties rooted in class, religion, or ethnic identities; comparative judgments of party leaders; and retrospective evaluations of incumbent performance.
Departing from quite different perspectives on how voters decide, each research tradition has produced results that call the preconditions for responsible party government into question. In the case of spatial modeling, the most basic result is the median voter theorem--that in elections involving exactly two parties competing over a single policy dimension the parties will converge to the position of the median voter. This result implies that the parties will present stable policies in that the parties reach an equilibrium--that is, a policy configuration such that neither party has incentives to change its position. However, this two-party equilibrium also entails policy similarity in that the parties present identical policies and thus fail to offer the electorate divergent positions that provide voters with meaningful choices. And, while results are somewhat different in two-party competition over multiple policy dimensions, recent theoretical work suggests that in these situations vote-seeking parties can again be expected to converge toward extremely similar, centrist positions. In toto, the literature on two-party competition suggests that vote-seeking parties will present similar positions that deny the electorate a meaningful choice between distinct policy visions, which is a prerequisite for responsible party government.
Although early spatial modeling results applied almost exclusively to two-party elections (but see Downs 1957, 122-32), the theoretical prediction that vote-seeking parties will present similar positions extends to three-party competition. Furthermore, such multiparty elections also encourage a high degree of policy instability in the sense that the parties have electoral incentives to continually leapfrog their competitors. A simple example illustrates these points. Consider the scenario illustrated in figure 1.1A, which shows three parties A, B, and C, located along a single policy dimension, which for convenience I label the Left-Right dimension. Under the standard Downsian logic, parties A and C are motivated to converge toward the center (fig. 1.1B), thereby "squeezing" party B because this centrist positioning gains additional support from Center-Left voters (who switch their support from B to A) and Center-Right voters (who switch their support from B to C). However, when party B is squeezed between A and C, then B is motivated to leapfrog one of its competitors, thus locating either to A's immediate left or C's immediate right (fig. 1.1C illustrates the situation in which B leapfrogs A). However, if B leapfrogs party A, then A is squeezed between B and C and is motivated to leapfrog one of its competitors in turn. This process can be expected to continue indefinitely, with the parties continually leapfrogging but always locating near the center of the policy space. Thus, in three-party elections spatial theory predicts that, from the perspective of responsible party government, the parties' behavior offers voters the worst of both worlds, namely, a scenario in which the parties' positions are unstable (in the sense that they consistently leapfrog) but they nonetheless present extremely similar, centrist positions that deny the electorate the choice between distinct policy alternatives. While results differ somewhat when four or more parties compete--roughly speaking, the problem of policy instability grows more severe and the problem of similarity somewhat less severe--the central intuitions developed here, that parties' policies will be unstable, in that they will continually leapfrog, and that rival parties' policies will be similar, typically extend to systems with any number of parties.
From their individual-level perspective, behavioral researchers' empirical findings call into question the requirement for policy voting on the part of the electorate, which appears to be essential for the party-voter policy linkage that underlies responsible party government. Behaviorists' results suggest that, although policies matter at least to some voters, their decisions are also readily swayed by considerations such as comparative evaluations of the party leaders' competence and integrity, economic conditions, group loyalties based upon class, religion, geography, and ethnicity, party identification, and so on. This conclusion, which was originally reached with respect to American voters (see, e.g., Campbell et al. 1960; Fiorina 1981; and Key 1966), has been extended outward over the past 30 years to encompass the electorates of most Western democracies (Alvarez, Bowler, and Nagler 1996; Butler and Stokes 1969; Converse and Pierce 1986; Dow 1997b, 1997c; Klingemann et al. 1994; Lewis-Beck 1988; Listhaug 1989; Pierce 1995; Rose and McAllister 1990; Strom and Svasand 1997; Whitten and Palmer 1996). Although, as I outline in chapter 2, the relative electoral impact of policy motivations and nonpolicy influences is a matter of heated scholarly debate, in toto the studies cited here strongly suggest that the cumulative impact of voters' nonpolicy motivations upon the vote is at least as great as--and perhaps much greater than--the impact of policy voting.
As I shall emphasize in greater detail in chapter two, the distinction between voting based upon policy issues and voting based on non-policy-related motivations is not always clear. For instance, voters' retrospective evaluations of incumbent performance, their reactions to government scandals, and their comparative evaluations of party leaders' competence and integrity can reasonably be described as issue voting. According to Donald Stokes, these valence issues "involve the linking of the parties with some condition that is positively or negatively evaluated by the electorate" (1963, 373). This contrasts with position issues such as Left-Right ideology, welfare spending, and the death penalty, upon which voters (and parties) may disagree. My central point is that when voters deemphasize position issues election outcomes do not provide referendums on the parties' competing (positional) policy visions and, furthermore, that the link between parties' policies and their supporters' beliefs--which are assessed by comparing dyadic correspondence on positional issues--is threatened. Therefore, throughout this book I refer to valence issues as nonpolicy related, reserving the term policy voting for voting over positional issues.
This result suggests that it may be impossible to simultaneously satisfy the requirements of policy stability and policy divergence on the part of political parties, and of policy voting on the part of the electorate, that are central to responsible party government. If voters are policy oriented, as spatial modelers assume, then (with the exception of two-party competition along a single policy dimension) parties' policies are likely to be unstable in the sense that the parties continually leapfrog, and their policies will also be similar, so that voters are not afforded the option of choosing between truly distinct policy visions. If voters deemphasize policies, as behavioral researchers suggest, then parties' policies may not be representative of their supporters' beliefs since policies are not the primary basis of these supporters' party preferences.
1.3. The Puzzle
The findings of spatial modelers and behavioral researchers suggest that political representation through responsible party government is unattainable because the preconditions for responsible party government outlined in section 1.1 are not met in most Western democracies. Yet, in surveying the polities of Western Europe and North America, what is striking is the extent to which this prediction is not borne out by empirical observation. Contrary to spatial modelers' predictions of policy instability and policy similarity, empirical studies conclude that political parties present relatively stable policy programmes over time while rarely leapfrogging each other (see, e.g., Budge 1994) and furthermore that parties differentiate their policies, thereby presenting voters with a wide range of alternatives (Castles and Mair 1984; Huber and Inglehart 1995; Laver and Budge 1992). Thus, party systems are characterized by policy stability and policy divergence. And, while the behavioralists' finding that voters are motivated in large part by nonpolicy considerations seems to emperil mass-elite policy linkages, comparative research supports the proposition that in most party systems political parties provide reasonably faithful policy representation in that the policies parties propose reflect the views of their supporters (see, e.g., Dalton 1985; Holmberg 1989; Iversen 1994b; and Jennings 1992).
While I present data in subsequent chapters on empirical findings of policy divergence, policy stability, and policy representation, it is worth briefly surveying some of this evidence here. With respect to policy divergence, table 1.1 reproduces results reported by Huber and Inglehart (1995, app. 2) on the mean Left-Right positions that expert observers assigned to the major political parties in several Western European party systems and the United States as of 1993. The results indicate that experts perceive that each party system presents its voters with a wide range of ideological options. Alternative methodologies for estimating party positions, such as analyses of roll call voting, studies of party manifestos, and opinion surveys of party elites, confirm these experts' perceptions that party systems in most Western democracies are characterized by policy divergence not policy similarity (see, e.g., Budge 1994; Budge et al. 1987; Castles and Mair 1984; Dodd 1976; Inglehart and Klingeman 1987; and Sani and Sartori 1983).
With respect to policy stability, the most comprehensive empirical study of the evolution of party policies in postwar democracies has been conducted by Ian Budge and his colleagues at the Manifesto Research Group (MRG) (Klingemann et al. 1994; Laver and Budge 1992; Budge et al. 1987). Through comparative codings of the election programmes of the major parties in some 20 democracies during the postwar period the Manifesto Research Group has assigned positions to parties along a variety of policy dimensions, including an overall Left-Right summary (I review the details of the MRG's methodology in chapter 3). Table 1.2 reproduces the results of a temporal analysis of the MRG's codings reported by Budge (1994), which bears on policy stability in postwar democracies. Column 3 reports, for every postwar Western European democracy plus the United States, the proportions of party platforms that were coded as being outside the party's traditional "ideological area," defined as left of center for the Socialist, Communist, Labour, and Democratic Parties and as right of center for the Conservative, Christian Democratic, and Republican Parties. The results indicate that parties virtually never shift their Left-Right position outside their traditional ideological area. In addition, columns 1 and 2, which report the frequency with which parties leapfrog each other--that is, the extent to which parties shift their Left-Right orientation relative to rival parties-- show that leapfrogging has been rare and that when it occurs it usually involves parties that are contiguous along the Left-Right continuum. These results testify that throughout the postwar period political parties in Western democracies have presented remarkably stable policy images.
Continues...
Excerpted from Party Competition and Responsible Party Government: a Theory of Spatial Competition Based Upon Insights from Behavioral Voting Research by James Adams Copyright © 2001 by James Adams. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
From the B&N Reads Blog
Page 1 of