Portal de Periódicos da CAPES

Party Political Representation in Portugal

2008; Taylor & Francis; Volume: 13; Issue: 4 Linguagem: Inglês

10.1080/13608740902738384

1743-9612

Ana Maria Belchior,

European Union Policy and Governance

Abstract This study analyses the degree to which Portuguese parties represent their electorates. In this instance, representation is measured by the degree of congruence between those who vote and those who are elected. The analysis has two dimensions: ideological positioning and attitudes towards democracy. On the first dimension, it appears that, compared to the voters, the parliamentary elite is ideologically more extreme. On the second dimension, one finds that elites are more favourable towards democracy than the electorates they represent. Keywords: PortugalPolitical PartiesDemocracyPolitical RepresentationPolitical Congruence Acknowledgements The author wishes to thank José M. Leite Viegas, André Freire and Helena Carreiras for their insightful suggestions and comments on previous versions of this article. Special thanks are due to José M. Leite Viegas for his invitation to join the project Participation and Democratic Deliberation. Notes [1] For detailed information about the Portuguese party system, see, e.g., Lopes (2004 Lopes, F. F. 2004. Os Partidos Políticos. Modelos e Realidades na Europa Ocidental e em Portugal, Oeiras: Celta Editora. [Google Scholar]); Gunther (2007 Gunther, R. 2007. "Portuguese elections in comparative perspective: parties and electoral behaviour in Southern Europe". In Portugal at the Polls, Edited by: Freire, A., Costa Lobo, M. and Magalhães, P. 11–47. Lanham, MD: Lexington Books. [Google Scholar]). [2] Other studies approached this subject with a higher degree of complexity—namely by allowing for influence by different intra-party echelons of leadership on levels of congruence (May 1973 May, J. D. 1973. Opinion structure of political parties: the special law of curvilinear disparity. Political Studies, 21(2): 135–151. [Crossref], [Web of Science ®] , [Google Scholar]). [3] Other works could be added to this list. However, either because they are theoretical or simple methodological replications of one of the studies mentioned above, they are not included here. [4] As in the case of other authors (e.g., Huber & Powell 1994 Huber, J. D. and Powell, G. B. Jr. 1994. Congruence between citizens and policy makers in two visions of liberal democracy. World Politics, 46(3): 291–326. [Crossref], [Web of Science ®] , [Google Scholar]), the median is used instead of the mean in order to overcome the bias that the latter may introduce in variables that do not have a normal distribution. [5] In contrast to other electoral systems to which this measure has been applied, in the Portuguese case it is only possible to establish five types of dyadic correspondence. It is not possible to establish correspondences at the level of electoral constituencies because voters do not vote directly for candidate MPs, but for parties. Therefore, correspondences are only established at the level of each of the five Portuguese parties. The regression and correlation coefficients are therefore calculated for the mean positions of the elected and electorate of these five parties. Despite their statistical drawbacks, I opted for these measurements for the following reasons. First, I am interested in replicating in the analysis for Portugal the different aspects of measuring representation through congruence. Second, I have inserted some safeguards—namely through the additional action of applying other statistical measures. Third, the additional information obtained from the regression and correlation coefficients, though it necessarily requires care in interpretation, may become important when read together with the other measures. Despite the statistical limitations, from the analytical viewpoint the measures described above enable confirmation of the direction and intensity of the relationship between the elite and the voters shown by other measures. [6] On the statistical grounds for this measure, see especially Pierce (1999 Pierce, R. 1999. "Mass-elite issue linkages and the responsible party model representation". In Policy Representation in Western Democracies, Edited by: Miller, W. 9–32. Oxford/New York: Oxford University Press. [Google Scholar], pp. 16–17). [7] The voters' survey comprised a sample of 1,001 eligible voters (18–70 years old). Proportional quota sampling and face-to-face interviews were used. The survey of parliamentarians used face-to-face interviews. The response rate of 79 out of 230 MPs corresponded to 34.3 per cent of the universe. The number of respondents from each party were as follows: 38 from PS (out of the total of 121 party MPs), 28 from PSD (out of 75 MPs), seven (out of 12) from CDS/PP, two (out of 12) from CDU, four (out of 14) from BE and two (out of two) for PEV. Some methodological problems arise from the comparison between voters and MPs. These problems have been widely reported in the literature on this subject (e.g., Eulau 1987 Eulau, H. 1987. The congruence model revisited. Legislative Studies Quarterly, 12(2): 171–214. [Crossref], [Web of Science ®] , [Google Scholar], pp. 171–172; Pierce 1999 Pierce, R. 1999. "Mass-elite issue linkages and the responsible party model representation". In Policy Representation in Western Democracies, Edited by: Miller, W. 9–32. Oxford/New York: Oxford University Press. [Google Scholar], pp. 13–15), with two of the most important being the conceptual differences between voters' and MPs' political and ideological positions, and, as a consequence, the higher dispersion of voters' positions comparative to those of MPs. [8] It should be noted that, although the PS displays the least difference in the means, it presents the greatest standard deviation for the MPs, followed by the PSD. This is understandable on account of the more centrist positioning of these two parties. The tendency towards less internal cohesion in the larger parties at the centre in comparison to the smaller and more extreme parties has already been demonstrated at a European level (Schmitt & Thomassen 1999 Schmitt, H. and Thomassen, J. 1999. "Distinctiveness and cohesion of parties". In Political Representation and Legitimacy in the European Union, Edited by: Schmitt, H. and Thomassen, J. 111–128. Oxford/New York: Oxford University Press. [Crossref] , [Google Scholar], pp. 124–125). This lower degree of internal homogeneity in the elite, however, affects the extent of the congruence encountered for the PS. [9] As there is no data available to measure the political parties' actual policies and ideological positions, and given the sense of near-convergence of position between how MPs define their own positions and how they place their parties, I used the positions of the former to measure those of the latter. The authoritative nature of party elites' positions in defining those of their parties (Budge 2000 Budge, I. 2000. Expert judgements of party policy positions: uses and limitations in political research. Research note. European Journal of Political Research, 37(1): 103–113. [Crossref], [Web of Science ®] , [Google Scholar]) and, in the case of Portugal, MPs' relative lack of independence in relation to the political parties (see, e.g., Opello 1988 Opello, W. C. Jr. 1988. O Parlamento português: análise organizacional da actividade legislativa. Análise Social, : 127–150. [Google Scholar], pp. 134–143), supports this option. However, this is not an optimal, but only an analytical, solution: MPs' political positions are not necessarily identical to the party line. [10] Greater congruence in non-structural as opposed to structural matters has been registered in other European countries (Thomassen 1999 Thomassen, J. 1999. "Political communication between political elites and mass publics". In Policy Representation in Western Democracies, Edited by: Miller, W. 33–58. Oxford/New York: Oxford University Press. [Google Scholar], pp. 46–51).