From Electoral to Policy Representation: A Comparison of 34 Democracies

 

Abstract

This study investigates the extent to which contemporary democracies are representative of the policy preferences expressed by the majority of electors, parties and governments there in. The analysis concentrates on parties’ and governments’ manifestos in 34 democracies with data provided by the Manifesto Research Group (MRG) project database. It emphasizes the importance of policy representation as a useful conceptual tool for the study of democracy, as well as the role that ideas, discourses and interaction play in the process of institution-building. The investigation also clusters countries on the basis of the correspondence between the ideological distances and policy preferences existent in the political system. The resulting classification distinguishes four groups of distinct democracies: strong policy representative, weak policy representative, strong policy unrepresentative and weak policy unrepresentative democracies.

 

Keywords: comparative political systems, electoral representation, policy representation, discursive institutionalism, political culture, political community,  democracy.

 

 

Introduction

Whereas the current literature on comparative political systems has paid a special attention on existing political institutions in order to evaluate the degree of democratic representation of a nation, the analysis of policy representation[1] – that is, the convergence between the policy preferences expressed by the citizens, the party system and the government – has only sporadically come to the attention of the international community. What follows explores the extent to which the policy preferences of the majority of electors (the median voter) are effectively represented in the political arena by the government as well as by the political parties. This article, thus, aims to find a response to the following main research question: To what extent are contemporary democracies representative of the policy preferences expressed by the majority of voters, the government and the party system?

This research differs from previous studies on comparative political systems in some important respects. From an emphasis given to political institutions and the degree of electoral representation that a political system is able to ensure to the citizens, the center of attention here shifts to policy preferences, discourses and interactions and to a resulting degree of policy representation. As this article will highlight, policy representation remains one of the most crucial understated objectives for a democracy. Indeed, not only formal representation is essential, that is to say, that the majority of citizens succeeds in having a voice in the political arena, but also the fact that practical representation really takes place, that is to say, that the preferences of citizens are effectively met, is important. How could a system be addressed as fully democratic if there is little or no convergence between what the majority has expressed and what parties and governments have implemented?

This research differs also from previous studies more directly focused on policy representation (see, for instance, Miller et al., 1999; Budge et al., 2001; Klingemann et al., 2006; Paskeviciute, 2006; Budge and McDonald, forthcoming) in that, firstly, attention here is given to the democratic repercussions of policy representation, and not to the functioning mechanisms of the political systems per se. Second, despite the existence of clear linkages, the policy preferences of citizens will not only be explained on the basis of a left-right dimension, but possible path-departures of left parties from classical left-wing policies will also be considered. Six important political dimensions (left-right, state intervention in the economy, acceptance of market economy, welfare state expansion, European Integration, and international peace) will then be investigated. Third, the ideological distances existent in the political system and not the policy position of a party in the political arena will be highlighted. The focus will thus be given to ranges of differences in policy preferences instead of means and standard deviations. Fourth, this study will attempt to provide a possible classification of countries in terms of their policy representation performance over a period of thirteen years, while emphasizing the importance of different political communities (Klingemann and Fuchs, 1995; Fuchs and Klingemann, 2006) characterized by different political ideas and discourses (Schmidt, 2006, 248-266; see also Schmidt, 2002; Campbell, 2004).

The main hypothesis leading the research is, in fact, that the degree of policy correspondence between the policy preferences of the majority of voters, the government and the party system is not simply the product of existing institutions (e.g. electoral rules, the form of government, etc.), but rather it is the combined product of pre-exiting institutional structures with their power to set “the rules of the game”, the historical background of a country, with its peculiar political culture[2], political community, and the relative ideological distances as they emerged from country specific political and social conflicts over the centuries.

In order to substantiate this argument, section one provides a brief overview of the most often cited classifications of political systems, attempting also to clarify which are the main characteristics that make a political institutional structure particularly unique. Section two discusses the method, while section three analyzes the correspondence between the policy preferences of parties, government and the median voter in 34 democracies during a period of thirteen years, from 1990 to 2003. The countries involved are: Australia, Austria, Belgium, Bulgaria, Canada, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Israel, Italy, Japan, Latvia, Lithuania, Luxembourg, Netherlands, New Zealand, Norway, Poland, Portugal, Romania, Russia, Slovakia, Slovenia, Spain, Sweden, Switzerland, United Kingdom, and the United States. Here, the analysis conducted on party programs with the most recent data provided by the Manifesto Research Group (MRG) project database (Budge et al., 2001; Klingemann et al. 2006) has the primary objective of measuring the policy preferences present in a political system, but also aims to quantify the ideological distances and the degree of policy correspondence. Finally, section four summarizes and discusses the results, while proposing the use of a complementary system of classification. This part also highlights the importance that political cultures and political communities have in influencing the level of policy representation of a country.

 

  1. Institutions and Discourses: Classifying Contemporary Political Systems

Several schemes of classification for comparative political systems have been proposed in recent years. The most common classifications describe these systems in terms of: (i) the power or relative autonomy given to the President of the Republic (presidential, semi-presidential or fully parliamentary) (Duverger, 1980); (ii) the presence of one or two chambers (Chamber of Deputies and Senate) in the political arena (unicameral vs. bicameral political systems); (iii) the number of parties in the political arena (limited vs. extreme pluralism) (Sartori, 1976, 2005); (iv) the existence of few or several veto players (partisan or institutional) (Tsebelis, 1995, 2002); (v) the electoral rules of a country (proportional, mixed and majoritarian) (Lijphart, 1984; Shugart and Wattenberg, 2001); (vi) the complexity of institutional and interactive processes (see Schmidt’s distinction between simple polities and mixed or compound polities) (Schmidt 2006); and (vii) the government’s ability to implement its policy preferences, unilaterally or through the necessity of negotiations (in numerous cases also resulting in unlikely and political unstable alliances) (consensual vs. Westminster model) (Lijphart, 1984, 1999). Political systems, however, can also be classified by the ideological distance between the parties present in the political arena (moderate vs. polarized systems); the intra-inter party system of communication (low vs. high); or by the system of state/citizen communication  (low vs. high) (Sartori, 1976, 2005). In this case, the major focus is not simply on institutions and on the interactive processes and activities of leaders and party members, but also on their elective affinities and capacity to communicate with their electorate.

As these main typologies show, contemporary political systems can be classified not only on the basis of the institutional structures present in the political arena, but also on the basis of the political ideas, ideological distances, discourses and interactions present in the political environment. By setting the rules of the game, institutions, unquestionably, influence the final form of a country’s political system, but it should not be forgotten that institutions are also influenced by the ideas, policy discourses and policy preferences of the different political, economic and social communities.

In this context, it is certainly astonishing to note how, even though the concepts of political cultures and political communities are not new entries in the field of comparative political systems, little attention has been given to them in the context of policy representation studies, and this despite the fact that Gabriel Almond (1956: 396) had already noted in the 1950s that “every political system is embedded in a particular pattern of orientations to political action” (see also Almond and Verba, 1963, 1980; Putnam, 1993, 2000).

It is these key elements that this study attempts to clarify. As the analysis below will demonstrate, if the final form of a political system must correctly be described, this has to be described not only in terms of the product of pre-existing institutional structures, as previous studies and classical new-institutionalist explanations would highlight, but also as the result of cognitive and normative processes, as its “discursive” amendment suggests (see discursive institutionalism as proposed by Schmidt, 2006, 248-266; see also Schmidt, 2002; Campbell, 2004). In what follows, after a brief methodological introduction, the ideological distances and the convergence of policy preferences in 34 democracies are explored.

 

  1. Methodology

The MRG project database, which was started in the 1970s, is the most comprehensive dataset on policy preferences available. The latest version now also includes Eastern European countries (see Klingemann et al., 2006). This analysis, which covers a period of thirteen years, from 1990 to 2003, involves the political systems of Australia, Austria, Belgium, Bulgaria, Canada, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Israel, Italy, Japan, Latvia, Lithuania, Luxembourg, Netherlands, New Zealand, Norway, Poland, Portugal, Romania, Russia, Slovakia, Slovenia, Spain, Sweden, Switzerland, United Kingdom, and the United States. The data in the database are collected through content analysis of several parties’ programs. The unit of analysis consists of quasi sentences expressed in the party manifestos. Quasi sentences are statements that concern the policy preferences of parties on numerous important political issues (for more information on the MRG project methodology, see Budge et al. 2001, 93-245; Klingemann et al., 2006, 63-236). For the purpose of this study, policy preferences have been scrutinized not only on the basis of a left-right scale, but on six main crucial dimensions. These correspond to the ideological position of parties on a left-right scale, on their support or refusal for state intervention in the economy, on their acceptance of market economy, on welfare state expansion, on European Integration[3] and on international peace.

The decision of including six and not only one dimension stems from the possible relaxation of the classical left-right cleavage (Giddens, 1994). Even though, at the theoretical level, several studies have confirmed the centrality of the left-right position as determinant factor for further policy preferences (Klingemann et al., 2006), at the practical level, left-wing parties and governments around the globe now openly agree on market liberalism and on welfare retrenchment measures (traditionally the strong point of right-wing governments) more often than prior to the fall of the Berlin Wall. Few notable examples of this trend towards market liberalism and welfare retrenchment include the Labour governments in the United Kingdom, Germany and Italy, but this list could also comprise several short-lasting governments in the new democracies of Central and Eastern Europe (see Cerami, 2006). In all these cases, state defection from an active intervention in the economy and the abandon of classical Keynesian macro-economic policies are the clearest indicators of a restructuring process, which is now powerfully emerging among left-wing parties.

In order to provide a more comprehensive overview of the policy preferences, six main scales (RILE, PLANECO, MARKECO, WELFARE, EUROP, INTPEACE), developed and provided by the methodological guide of the MRG (Klingemann et al., 2006; see also Budge et al., 2001), have been recombined into a single scale, which expresses the cumulative ideological distances of each party. In more practical terms, the newly constructed Scale of Ideological Distance, calculated as the cumulative average of the six main scales during the period 1990-2003, can be expressed by the following formula:

 

SID = Mean(RILE) + Mean(PLANECO) + Mean(MARKECO) + Mean(WELFARE) + Mean(EUROP) + MEAN(INTPEACE)

 

Where:

–   SID is the Scale of Ideological Distance;

–   Mean is the average score;

– RILE, PLANECO, MARKECO, WELFARE, EUROP, INTPEACE represent, respectively, the scores concerning the policy preferences of political parties, government and the median voter on the left-right dimension, on state intervention in the economy, on acceptance of market economy, on welfare state expansion, on European Integration, and on international peace[4].

 

It is important to note that while the MRG project database seems to be well suited to capture the ideological distance of parties and governments, the policy preferences of the majority of electors expressed by the so-called “median voter preferences” suffer some methodological weaknesses. Indeed, while the examination of parties and government policy preferences is conducted directly on the statements and on the expression of intentions that these political formations have freely declared in their programs during and after the elections, the median voter estimates are calculated according to the following formula developed by Kim and Fording (2001:163):

 

M = L + {(50 – C) / F}* W

 

Where:

–  M = Median position (ideological score);

–  L = The lower end (ideological score) of the interval containing the median;

– C = The cumulative frequency (vote share) up to but not including the interval containing the median;

–  F = The frequency (vote share) in the interval containing the median;

–  W =  The width of the interval containing the median.

 

In less complicated terms, this formula means that the ideological position of the majority of electors involves indirect and non-direct estimates of the median voter policy preferences. The voter ideology is, in fact, inferred from the individual support in the elections to a specific party and its corresponding program, rather than identified from what he or she would affirm during a classical interview with a questionnaire. According to this method, elections’ results are conceived as large scale opinion polls which are supposed to reflect the individual’s own policy preferences. Needless to say, the situation is probably more complex than this with the individual’s real ideologies, in effect, fairly distant from the party own ideology. However, despite the presence of this shortcoming, there is no reason to believe that an individual would support a party which promotes a policy manifesto completely different from his or her own policy preferences, with the exception of forms of political corruption, such as the Italian compravendita (buy-and-sell) of votes, or a protest vote, which are certainly not rare events in several political systems. Even in these cases, however, the real policy orientation of the citizens would remain extremely difficult to estimate by the more classical opinion polls. In addition, since the statistical significance of the “median voter” has extensively been explored and confirmed now in numerous studies (for a more comprehensive discussion, see Miller et al., 1999; Kim and Fording, 2001; Paskeviciute, 2006; Budge and McDonald, forthcoming), the final outcome of analysis should not drastically alter, even in presence of this shortcoming.

Before proceeding with the investigation of the ideological distances, a couple of other important methodological notes have to be discussed. As mentioned, this study analyzes the convergence of policy preferences in 34 democracies from 1990 to 2003. In this period, 130 elections have taken place, while 233 different governments have come to office. A time span of thirteen years has, of course, also meant substantial changes in the policy preferences of parties, but also implied substantial changes in the ways in which a political system has responded to the changes in policy preferences among the citizens. Previous studies have attempted to reduce these distortions, either focusing on single elections (Miller et al.,1999) or introducing a control variable called “responsiveness” (Wessels, 1999; Paskeviciute, 2006; Budge and McDonald, forthcoming). This variable calculated how far the change in the position of the median voter produced a corresponding change in the government position after each election. Despite the fact that time clearly matters in politics, the responsiveness of government policy to shifts in popular opinions has, repeatedly, been confirmed especially in the long-term period (Budge and McDonald, forthcoming). Paskeviciute (2006) has also statistically demonstrated how in both old and new democracies short-term distortions occurring in separate elections tend to cancel each other out over a long period of time. Wessels (1999) and Budge and McDonald (forthcoming) also arrived at similar results, concluding that, despite the fact that time plays an important role, electoral rules are the first determinant factor for the final outcome in policy representation. Proportional systems are, in this context, more responsive to biases and distortions than majoritarian systems, which respond difficultly to changes in citizens’ policy preferences. On the basis of these repeated findings, aggregating data from different elections over a period of thirteen years is not only possible, but also useful, since the eventual distortions occurring will be compensated by a more general overview of the entire system of policy representation. The institutional setting of a country continues to play, in this case, the major role.

A final point that has to be highlighted concerns the decision to use ideological distances and not the policy position of parties as the main indicators of analysis. Although the differences are  not extreme, since, in both cases, the policy preferences are investigated, the focus on ranges, instead of means and standard deviations has an immediate graphical outcome. It provides, in fact, information on the polarization of the political system, and not simply of the position of a party in a uni-dimensional space. This also represents a useful source for understanding the distances existent in each country in terms of political cultures and political communities.

 

  1. Ideological Distances and Policy Representation

Figure 1 and Figure 2 show the range of policy preferences of the party system, of the government and of the median voter during the period 1990-2003. The more similar are the length of the bars, the higher is the correspondence between the policy preferences of the majority (the median voter), the government and the parties. In a perfectly majoritarian political system, the ideological distance of the majority of citizens should correspond to the ideological distance of the government. The ideological distance of the opposition should, eventually, be taken into account by the members of the government, but this should not represent a practical, but rather a moral obligation. As a result, the bars of the median voter and the government should, when possible, come closer to the bar of the party system in order to ensure that also the minority of citizens is, to some extent, represented by the government’s policy position.

Fig 1

Fig 2

 

As it can immediately be seen, political systems differ not only in terms of institutional structures and the associated electoral representation, but also in terms of the ideological distance and the resulting policy representation. Countries such as Israel, Austria, Sweden, Denmark and Slovenia have clearly more polarized party systems than, for example, Bulgaria, United States, Luxembourg and Portugal, where the ideological distance between the different political formations is clearly more limited (see length of bars).

In order to have a clearer picture of the countries, a more accurate analysis of policy representation needs to be conducted. Here, a variable formed by the cumulative difference between the party system’s, government’s and median voter’s ideological distances is constructed. Then, a classification of countries through cluster analysis is provided. The variable, called Range of Ideological Distance, can be expressed by the formula:

 

RID = IDpa – IDgov – IDmedvo

Where:

  • RID is the Range of Ideological Distance;
  • IDpa is the average range of the ideological distance in the party system;
  • IDgov is the average range of the ideological distance in the government;
  • IDmedvo is the average range of the ideological distance of the median voter[5].

 

This index provides a measure of the differences existent between the different ideological distances of the political systems. The key value here is represented by the zero, which would correspond to a system almost perfectly representative of the will and policy preferences of the majority. In this case, the ideological distance of the party system, the government and the median voter would be the same. The more a country moves away from the zero, the higher will be its mismatch in terms of the ideological distances and policy preferences[6].

In order to classify countries, Hierarchical Cluster Analysis is employed. Here, Squared Euclidean Distances are calculated to measure the distances between countries, while grouping is conducted on the basis of the “furthest neighbor” method. According to this method (also known as “complete linkage”), countries are clustered on the basis of the greatest distance that exist among two distinct groups.

Fig. 3 provides an easy to read graphical description of resulting clusters. The ways how the dendrogram can be read can greatly differ from researcher to researcher, depending on the weight that one wants to give to the clustering factor. Moving from right to left of the figure, two distinct clusters of countries can be identified. The first one represents policy unrepresentative democracies, which corresponds to Canada, Estonia, Finland, France, Hungary, Ireland, Israel, Italy, Norway, Romania, Slovenia and Switzerland. The second cluster, by contrast, is comprised of policy representative democracies, which in this case, include Austria, Belgium, Bulgaria, Czech Republic, Denmark, Germany, Greece, Japan, Latvia, Lithuania, Luxembourg, Netherlands, New Zealand, Poland, Portugal, Russia, Slovakia, Spain, Sweden, United Kingdom and the United States.

 

Fig 3

 

Needless to say, this classification is over-simplistic and hardly satisfying. Even though it must be remembered that the term “unrepresentative” refers to policy representation of the majority of voters and not to electoral representation, it implies that numerous democracies would fall into the policy unrepresentative cluster without any possibility to quantify their degree of policy correspondence. In order to deal with this conceptual, as well as methodological limitation, it is preferable, to use Sartori’s words (1970), to move backward in the ladder of abstraction, until the lowest possible level of generalization (see extreme left of the figure), which also represents the first and most basic clustering suggestion resulting from the statistical analysis. At the most basic level, the dendrogram shows four clear and distinct clusters of countries, with the exception of Australia, Israel and Romania, which have some particularities that have to be discussed more into details (see next section). The resulting new classification conceptualizes countries as strongly policy representative in the cases of Lithuania, Slovakia, United States, Spain and Bulgaria, while weakly policy representative democracies are those present in Sweden, Latvia, Luxembourg, United Kingdom, Belgium, Germany, Japan, Portugal, Netherlands, Denmark, New Zealand, Russia, Austria, Greece, Poland, and the Czech Republic. By contrast, weakly policy unrepresentative democracies are those of Hungary, Canada, Ireland, France, Estonia, Slovenia, while strongly policy unrepresentative democracies are those of Finland, Italy, Norway and Switzerland.

 

Table 1

 

  1. Key Characteristics

In order to distinguish between policy and electoral representation amongst the countries studied here it is fundamental to identify the main peculiarities of each cluster, elucidating why they form unique categories. The group of strongly policy representative democracies is distinct from the others and unique, since it shows a non polarized party system (-) coupled with non polarized government (-) and non polarized median voter (-) policy preferences. This can be the result of majoritarian or mixed electoral rules, which, as argued, would favour a less polarized party system (Wessels, 1999; Budge and McDonald, forthcoming), and/or also the existence of a less polarized political community due to the historical legacies of the country. In the case of the United States, for example, its particular political culture characterized by a clear refusal of communism and nazi-fascism is likely have promoted a more unified political community of voters through a self-reinforcing mechanism which took place during the political history of the country.

The cluster of weakly policy representative democracies comprises the majority of nations and, for this reason, tends also to be the most heterogeneous. Some common features, however, can be identified. The majority of these countries show, in fact, a polarized party system (+) coupled with polarized government (+) and median voter (+) policy preferences. In Belgium, Latvia, Luxembourg, Portugal and the United Kingdom, the ideological distances of the party system tend to be more moderate if compared to the countries of the same cluster, but they should not be viewed as anomalies. If compared to the cluster of strongly policy representative democracies, this cluster shows immediately how the ideological distances of the government or the median voter tend to be more disproportionate in relation to the ideological distance of the party system. Proportional electoral rules are in this cluster preponderant.

The third cluster of weakly policy unrepresentative democracies also shows its own peculiarities. In these democracies, a polarized party system (+) has developed with non polarized government (-) and non polarized median voter (-) policy preferences. Clearly, the presence of Canada and France in this group tends to exclude the electoral rules as single explanatory factor. These countries have, in fact, a political system based on majoritarian electoral representation, while all other nations in this grouping are parliamentary democracies where the presence of non strictly majoritarian electoral rules (proportional or mixed electoral systems) should diversify the composition of the party system.

Strongly policy unrepresentative democracies are characterized by a strongly polarized party system (+ +) followed by non polarized government (-) and non polarized median voter (-) policy preferences. With Italy, for example, it can be affirmed that the highly differentiated political community (Griffin, 1997) existent in this country is the main leading factor. Italy has, in fact, alternated majoritarian and proportional representation electoral rules in the last decade, but, in no case, has this resulted in a system formed by few and stable party coalitions.

As mentioned, Australia, Romania and Israel require special attention, since Australia should be part of the policy representative democracies (see dendrogram), even though it shows higher negative distances, while Romania and Israel show unusually high values of policy non-representation (see Table 1). Possible explanations for these atypical clusters can be found in the electoral rules, as well as in the political and social conflicts that characterize these societies. In the case of Australia, the extreme polarization of the political system favoured by a majoritiarian electoral rules does not seem to have negatively influenced the final policy representation abilities of the country, while in the cases of Israel and Romania, serious doubts can be raised on the real policy representation of the political systems. Arguably, the reason could be found in proportional representation rules, which have not limited the polarization of the party system, as well as also in the need to ensure the stability in an extremely differentiated and potentially dangerous political environment, where the continuous external and internal threats tend to favour  policy action instead of policy representation.

Although this research, on the one hand, confirms previous results regarding the central importance of electoral rules and, thus, the importance of existing institutional structures (Wessels, 1999; Paskeviciute, 2006; Budge and McDonald, forthcoming), in influencing the level of policy representation of a given country, on the other hand, it shows that ambiguities and inconsistencies are certainly not absent. In fact, when the clusters of ideological distances are scrutinized more closely with the electoral rules of a political system (see Fig. 3), then it becomes clear that the impact of electoral procedures on policy representation is less clear cut. It could be argued, for instance, that, despite the presence of a majoritarian electoral system and formal democratic declarations, France with its long history of divided political cultures and communities has maintained its attitude towards weak representation between citizens and government’s policy preferences as demonstrated by the continuous restoration of previous order and rules. The American exceptionalism, Israel’s unilateral responses, and Italy’s extremely divided and differentiated political community can also be taken as further examples.

In addition, whereas it is true that proportional systems tend to be more present in weakly policy representative and weakly policy unrepresentative democracies, the exceptions to this rule, for example Latvia and the United Kingdom, are countries with clear majoritarian systems. Exceptions also exist in the club of strongly representative democracies, where majoritarian electoral systems are preponderant. Bulgaria and Slovakia, where the influence that authoritarian rulers had in the final form of government, are, probably, the best examples of how historical and political cultural legacies influence the political path of a nation. The existence of an East-West divide in political socialization cannot, in fact, be denied, as Fuchs and Klingemann (2006) have demonstrated. This has resulted in a more complex and internally differentiated political communities, where effective policy representation has often been sacrificed for maintaining political stability.

 

Conclusion

On the basis of a rich set of data provided by the MRG project database, the correspondence of the ideological distances and policy preferences between the party system, the government and the median voter have been analyzed in 34 democracies over a period of thirteen years. The resulting classification points towards 4 clusters: strongly and weakly policy representative democracies, and strongly and weakly policy unrepresentative democracies. Strongly policy representative democracies are those that present a high degree of correspondence between the ideological distance of the party system, the government and median voter. This has taken the form of systems characterized by a non polarized party system coupled with non polarized government and non polarized median voter policy preferences. Weakly policy representative are those political systems in which the correspondence between policy preferences of the party system, government and median voter is still significant, but less than the one existent in strongly policy representative democracies. These democracies have taken the form, for the most part, of a polarized party system coupled with a polarized government and median voter policy preferences. Weakly policy unrepresentative democracies and strongly policy unrepresentative democracies have been identified as those in which a mismatch between these three elements has occurred. In the case of weakly policy unrepresentative democracies, this has taken the form of a polarized party system which lives closer with non polarized government and non polarized median voter policy preferences, while, in the case of strongly policy unrepresentative democracies, this has taken the form of a strongly polarized party system followed by non polarized government and non polarized median voter policy preferences.

The causes of the differences in the level of policy representation can be, as mentioned, identified in the pre-existing institutional structure of the political system (such as the electoral rules of a nation, the composition of the polity system, etc.), but also and, perhaps, more importantly, in the political culture, historical legacies as well as in the structure of the country’s political community. The conjunction of these elements remains, in this context, the major driving factor in determining the final form of a democratic political system.

To conclude, this research emphasizes the necessity of paying a greater attention to the concept of policy representation in the study of contemporary political systems and democracies. In fact, not simply voices must be heard, but also policy preferences must be implemented. The argument is, in brief, that democracy should include not only a dimension of electoral representation, but also a dimension of policy representation, understood as the “capacity of doing things” (Ober 2006), as well as the capacity that “these things” correspond to what the citizens have voted for. A fully democratic political system, however, should not be seen as the one in which the policy preferences of the majority are perfectly represented by the government and the party system, since this would imply not only an excessive “power of the majority”, but, perhaps, even a “dictatorship of the majority”. Rather, a fully electorally, democractically and policy representative political system is, in the view of this analysis, the one in which all voices coming from citizens find a place and are equally represented in the political arena, but also the one in which if polarized ideological distances exist these have to be equally distributed in the party system’, in the government’ and in the majority of citizens’ policy preferences.

 

References

Almond, G., Verba, S., 1963. The Civic Culture: Political Attitudes and Democracy in Five Nations. Princeton University Press, Princeton, NJ.

Almond, G., Verba, S. (Eds.), 1980. The Civic Culture Revisited. Little Brown, Boston.

Budge, I., McDonald, M. D., forthcoming. Election and party system effects on policy representation: bringing time into a comparative perspective. Electoral Studies. Corrected proof. available online 28 November 2006.

Budge, I., Klingemann, H-D., Volkens, A., Bara, J., Tanenbaum E. (Eds.), 2001. Mapping Policy Preferences. Estimates for Parties, Electors, and Governments 1945-1998. Oxford University Press, Oxford.

Campbell, J. L., 2004. Institutional Change and Globalization. Princeton University Press, Princeton, PJ.

Cerami, A., 2006. Social Policy in Central and Eastern Europe. The Emergence of a New European Welfare Regime. LIT Verlag, Berlin.

Duverger, M., 1980. A New Political System Model: Semi-Presidentialism. European Journal of Political Research 8 (2), 165-187.

Fuchs, D., Klingemann, H-D., 2006. Democratic communities in Europe: a comparison between East and West. In: Klingemann, H.-D., Fuchs, D., Zielonka, J. (Eds.), Democracy and Political Culture in Eastern Europe. Routledge, New York, pp. 25-66.

Giddens, A., 1994. Beyond Left and Right. The Future of Radical Politics. Polity Press, Cambridge.

Griffin, R., 1997. Italy. In: Eatwell, R. (Ed.), European Political Cultures. Conflict or Convergence?. Routledge, London, pp. 139-156.

Hall, P., Taylor, R. 1996. Political science and the three new institutionalisms. Political Studies 44 (5), 952-973.

Kim, H-M., Fording, R.C., 2001. Extending party estimates to governments and electors. In: Budge, I., Klingemann, H-D., Volkens, A., Bara, J., Tanenbaum E. (Eds.), Mapping Policy Preferences. Estimates for Parties, Electors, and Governments 1945-1998, Oxford University Press, Oxford, pp. 157-178..

Klingemann, H-D. & Fuchs, D. (Eds), 1995. Citizens and the State. Oxford University Press, Oxford.

Klingemann, H-D., Volkens, A., Bara, J. L., Budge, J., McDonald, M. D. (Eds.), 2006. Mapping Policy Preferences II. Estimates for Parties, Electors, and Governments in Eastern Europe, European Union, and OECD 1990-2003. Oxford University Press, Oxford.

Lijphart, A., 1984. Democracies: Patterns of Majoritarian and Consensus Government in Twenty-One Countries. Yale University Press, New Haven CT.

Lijphart, A., 1999. Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries. Yale University Press, New Haven CT.

Miller, W. E., Pierce, R., Thomassen, J., Herrera, R., Holmberg, S., Esaiasson, P., Wessels, B.. 1999. Policy Representation in Western Democracies. Oxford University Press, Oxford.

Müller, W. C., 2000. Political parties in parliamentary democracies: making delegation and accountability work. European Journal of Political Research 31 (3), 309–333.

Ober, J. 2006. The original meaning of “democracy”: capacity to do things, not majority rule. Paper presented at the annual meeting of the American Political Science Association, Philadelphia PA, Aug. 31-3 Sept.

Olson, M. Jr., 1965. The Logic of Collective Action. Schocken Books, New York.

Paskeviciute, A., 2006. Elections, policy representation, and the system legitimacy in contemporary democracies. Paper resented at the annual meeting of the American Political Science Association, Philadelphia PA, Aug. 31-3 Sept.

Putnam, R. D., 1993. Making Democracy Work. Civic Traditions in Modern Italy. Princeton University Press, Princeton.

Putnam, R. D., 2000. Bowling Alone. The Collapse and Revival of American Community. Simon & Schuster, New York.

Sartori, G., 1970. Concept misformation in comparitive politics. American Political Science Review 64 (4), 1033-53.

Sartori, G., 1976. Parties and Party Systems: A Framework for Analysis. Cambridge University  Press, Cambridge.

Sartori, G., 2005. Parties and Party Systems: A Framework for Analysis. ECPR Press, Oxford.

Schmidt, V. A., 2002. The Futures of European Capitalism. Oxford University Press, Oxford.

Schmidt, V. A., 2006. Democracy in Europe. The EU and National Polities. Oxford University Press, Oxford.

Shugart, M., Wattenberg, M. P., 2001. Mixed-Member Electoral Systems: the Best of Both Worlds?. Oxford University Press, New York.

Tsebelis, G., 1995. Decision making in political systems: veto players in presidentialism, parliamentarianism, multicameralism and multipartyism. British Journal of Political Science 25 (July), 289-326.

Tsebelis, G. 2002. Veto Players: How Political Institutions Work. Princeton University Press, Princeton, NJ.

Wessels, B., 1999. System characteristics matter: empirical evidence from ten representation studies. In  Miller, W. E., Pierce, R., Thomassen, J., Herrera, R., Holmberg, S., Esaiasson, P., Wessels, B. (Eds.), Policy Representation in Western Democracies. Oxford University Press, Oxford, pp. 137-161.

[1] It is important to note that policy and electoral representation are different concepts, in which policy representation refers to the convergence of the policy preferences of the majority with the policy preferences of the government and of the party system, while electoral representation refers to the possibility of citizens finding a place where their voices can be heard in the political arena.

 

[2] On the concept of political culture, see, among many others, Almond and Verba (1963, 1980).

[3] Please note that for those countries where European Integration is not a crucial issue in the political agenda (e.g. USA, Canada, Australia, Israel and Russia), the values of the index of European Integration (EUROP) are equal to 0. Since the newly constructed Index of Ideological Distance is based on average scores, the absence of this information does not alter cross-country comparisons.

[4] The acronym of the variables may change in the different files available in the database. In the full MRG project database the variables are labeled as rile, planeco, markeco, welfare, europ, intpeace; in the government dataset as rile, planeco, markeco, welfare, eu, intpeace; while in the median voter dataset as riteleft, welfmed, planmed, markmed, eumed and peacemed.

[5] The order in this mathematical formula is based on the assumption that the range of the ideological distance of a party system should be higher than the ideological distance of the government and of the median voter. The reason for this clearly depends on the fact that a party system should, in theory, have a more polarized and representative political structure than those expressed by the government or by the median voter’s main policy preferences, which, by definition, should reflect the will of a majority of electors, either of a proportional majority, such as in presence of proportional electoral rules, or of a disproportional majority, in case of a majoritarian democracy.

[6] Please note that negative values represent situations in which the sum of the government’s and median voter’s policy preferences exceed the ideological distance of the party system. This atypical situation, which should in no case be seen as an indicator of policy non-representation, can be the result either of limited differentiation of the political environment (such as in the case of the United States) or of extreme polarization of the party system and government (such as in the case of Australia).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s