Room for Manoeuvre: Euroscepticism in the Portuguese Parties and Electorate, 1976-2005

Marina Costa Lobo and Pedro C. Magalhães, Instituto de Ciências Sociais, Universidade de Lisboa.

([email protected]; [email protected])

ABSTRACT In this paper, euroscepticism in Portugal is approached from the perspective of party positions and mass-level attitudes. We document the nature and evolution of party euroscepticism since 1976. Although the poles of party-based euroscepticism can be found, predictably, at the extremes of the party system, there is also a dynamic and contingent element to party positions that can only be accounted for on the basis of a strategic explanation. We also show that, whenever parties attempted to mobilize voters on the basis of a Eurosceptic discourse, voters responded to such attempts. Under those conditions, citizen attitudes vis-à-vis the EC/EU, both in terms of the economic and the political consequences of integration, emerged as a relevant electoral cleavage in Portuguese politics. Keywords: euroscepticism, Portugal, parties, mass attitudes.

Introduction Portugal, together with the remaining Southern European countries that became members of the European Union in the 1980s, has typically been described as a reliable redoubt of ‘euroenthusiasm’ (Gramacho and Llamazares, 2005). This is certainly true if we look at the high aggregate levels of public opinion support for EU membership, or

when we consider that the two major parties— the centre-left Socialist Party (PS) and the centre-right Social Democratic Party (PSD) — that have regularly alternated in government since the inception of democracy in Portugal have remained, after all, resolutely pro-European. Thus, Portugal would seem to constitute a good illustration of Mair’s general diagnosis about the limited impact of Europe in European party systems (Mair, 2000: 28). However, even Mair acknowledged that although party system format might not seriously change due to European integration, the competitive dynamics of the party system could be affected by the adoption of eurosceptical positions by political parties. The analysis of such dynamics in the Portuguese party system and their reflection at the level of mass political behaviour is the main concern of this paper. The analysis will focus on the four main political parties in Portugal, i.e. those which have consistently managed to elect MPs since 1976 until 20051. These are the PS and PSD as well as the Communists (PCP) and the Conservative CDS-PP. The BE, a relatively new party which has managed to elect an increasing number of MPs in the last three elections since 1999, is also included in the analysis, due to its increasing electoral importance. In the following sections, using data on parties’ electoral manifestos, we’ll begin by mapping the positions taken by each of the main Portuguese political parties on European integration since 1976. As we shall see, those positions roughly correspond to what could be expected in terms of the social and political cleavages on the basis of which the Portuguese party system has been built. However, in spite of some amount of continuity in parties’ relative placements, the extent to which their substantive positions on European integration issues have changed is such that structural/ideological explanations seem to be ultimately insufficient. Thus, we will examine whether parties’ strategic considerations and the political constraints they faced can also explain the surprising variability in party positions on the issue of European integration which exists in Portugal. Finally, we will determine whether those strategic considerations were at all reflected on mass political behaviour. Using Eurobarometer data on voting intentions 1

The PRD, (Partido Renovador Democrático), a centre-left party which was created in 1985 and obtained 19% of the national vote in elections that year, is excluded because it was an epiphenomenon, having failed to elect MP’s in 1991 and ever since. For the same reasons, we exclude micro-parties which have not polled above 3% of the national vote from our analysis.

2

around three legislative elections in the 1990s, we will suggest that, at least under conditions of high polarization of the European issue at the party and elite level, citizens’ orientations vis-à-vis European integration have been more consequential at the level of electoral choices than previously thought.

The Evolution of Portuguese-EC/EU relations from Authoritarianism to Democracy

Recent historiography on Portugal’s authoritarian regime, which lasted from 1933 until 1974, has shown that it did not remain completely on the margins of post-war European integration (Castilho, 2000, Rosas and Oliveira, 2004). Portugal was accepted as a founding member of both the OECD (1948), NATO (1949), and most importantly, EFTA (1959). This is not the place to explain the relative paradox of the participation of a colonial and authoritarian regime in major European post-war initiatives. The Cold War context, Portugal’s limited economic integration with other European countries, and the preferences of a Portuguese Europhile technical elite seem to have all been relevant factors in the successful conclusion of the various Treaties signed. In the 1960s, following the entry into EFTA, the Portuguese economy grew at a considerable speed and economic integration with Europe increased, even more with EEC countries (namely France and Germany) than with EFTA members (Leitão, 2007). With the imminent entry of the UK, Ireland and Denmark into the EC, the Portuguese government negotiated an agreement, signed in 1972, which allowed the continuance of the decrease in industrial tariffs that had existed previously with EFTA members between Portugal and the EC. The 1972 agreement with the EC coincided with the hardening of the authoritarian regime in its final years. Marcello Caetano, Salazar’s supposedly reformist successor, chose to engage in negotiations of a commercial agreement with Europe instead of seeking an associated-member status, thus showing the limits of his willingness to integrate with Europe, and to democratise Portugal (Castilho, 2000; Rosas and Oliveira, 2005; Reis, forthcoming). It was well understood that the EC was a democracies’ club and closer association with it would have sent a strong internal signal

3

of the government’s preferences regarding the future of the colonies and the authoritarian nature of the regime. Once the coup occurred in 1974, entry into the EC was not taken up immediately by the political parties which established themselves as the main players in Portuguese politics, namely the Socialist Party (PS), the Communist Party (PCP), the centre-right Social Democratic Party (PSD), and the conservative Social Democratic Center/Popular Party (CDS/PP), the latter two founded shortly after the April 1974 Revolution. In the aftermath of the Revolution, and until the first constitutional government was formed in 1976, pressing issues such as ending the wars in Africa, decolonisation, democratisation and economic development took the forefront of political discussions. The issue of Europe was nonetheless latent, subsumed in the more fundamental cleavage concerning the nature of the new regime. A clear preference for EC membership was first voiced by the PS, and in particular by its leader, Mário Soares. This symbolised a choice for a liberal democracy on Western lines, which was then followed by parties on the Right of the party spectrum. In order to contextualise the data presented below on both party and mass euroscepticism it is important to know the relative electoral strength of parties over the whole period under analysis, namely since 1976 until 2005. The table below presents the legislative and the European Parliament (EP) election results. Two parties, namely the PS and PSD, have become the dominant parties of the system at both the legislative and the EP level. Whereas between 1976 and 1987 they gathered on average 65% of the legislative elections vote, between 1991 and 2005 the two parties controlled an average of 77% of the vote. Indeed, the Portuguese party system has evolved from a relatively fragmented, to a rather bipartisan party system, where the PS and the PSD have alternated in single-party governments in power between 1987 and 2002, and again since 2005 (Lobo, 2001; Magalhães, 2003. The overall trend is still one of a continued predominance of the two main parties and of the pro-European parties. At the EP level the two main parties have also gained considerable ground: whereas in 1987 they polled 59% of the vote, their share has been increasing systematically, reaching 78% of the votes in 2004. [Table 1 about here]

4

The corollary of this overall trend of voter concentration in the two main parties, the PS and the PSD at both levels, is that the parties on the flanks have experienced substantial electoral decline in the same period. Moreover, the two major parties have also alternated in single-party governments since 1987, with one exception.2 Having established the context both regarding the issue of Europe within Portugal as well as the evolution of the party system, we can now turn to an analysis of party and mass euroscepticism.

Party euroscepticism in Portugal The literature on euroscepticism emphasizes several factors which may explain why parties take up such a stance. One strand of theoretical argument tends to emphasize the role of ideology: the further from the centre of the left-right spectrum a party positions itself, the likelier it is that it will oppose the European integration project. This is because EC/EU integration has always been a centrist project, led by the elites from the governing — centrist — parties (Mair, 2000:38; Hix, 1999). It is well-known that, when mapping the parties left-right position with an orthogonal dimension indicating level of support for European integration, the result for West European democracies tends to resemble an inverted u-curve (Hooghe, Marks, Wilson, 2004: 125). Opposition to integration seems to stem from two fundamental sources. On the one hand, from parties on the TAN side of a GAL (green, alternative, libertarian)/TAN (traditional, authoritarian, nationalist) dimension (i.e., radical right parties), that are likely to see European integration as a threat to national sovereignty. And on the other hand, from parties in the traditional extreme left, that tend to depict and criticize integration as a market-oriented capitalist project (Hooghe, Marks, Wilson, 2004; Marks, et. al., 2006). From this perspective, we should expect to find two poles of opposition to Europe in the Portuguese party system. The first should be located in the extreme-left represented by the Communist Party (PCP), opposing Europe as a ‘Trojan horse for international capitalism’ (Marks et al., 2006: 170). More recently, the PCP has been joined by the Left2

Between 2002 and 2004 the CDS-PP participated in a coalition government with the PSD.

5

Block (BE- Bloco de Esquerda) on the extreme-left. The BE is a party that derived from a fusion of extreme-left parties and movements. It contested its first elections in 1999 and defends left-libertarian ideological principles which are also largely congruent with euroscepticism, especially from an economic perspective. The other pole of opposition should be found in the party furthest to the right of the party system: the Social and Democratic Centre-Popular Party (CDS-PP). Created in a revolutionary context where the entire ideological right was delegitimised — through its association with the authoritarian regime — the CDS even presented itself initially as a centrist party, was supported by the Christian Democratic European parties, and even formed a short lived coalition government with the Socialists in 1978. Since then, although the party’s ideological discourse has shifted to the right, its original ideological and factional syncretism — as a Liberal/Conservative/Christian Democratic party (Robinson 1996) — has nevertheless prevented that shift from taking the party to the extreme right. There is, however, a second strand of arguments about why parties take eurosceptical positions, and that is related to their strategic goals. Successive Treaties have lead to the growing europeanisation of public policies. Moreover, the participation of national institutions in these supranational bodies is asymmetrical, with the bulk of power going to the government (Hix, 2005; Goetz, and Hix, 2000). Therefore, parties who are systematically out of government tend to be eurosceptical (Taggart, 1998; Sitter, 2001). Indeed, euroscepticism is, to a great extent, ‘the politics of opposition’ (Sitter 2001; 2002a), although it can be politicized or downplayed, depending on survival ambitions, coalition strategies and the pursuit of office. In particular, parties find it difficult to combine hard euroscepticism with participation in government, whenever the EU issue is salient in the polity, while eurosceptical parties that aspire to become credible coalition partners are likely to be forced to downplay their hostility to Europe. Some of the predictions about the Portuguese case that derive from this strategic account of party positions are very similar to those resulting from an ideological approach to party-based euroscepticism. More specifically, we should expect eurosceptical positions to be adopted more systematically by all parties besides the two centre-left and centre-right mainstream parties that have alternated in government since

6

1976, the PS and the PSD. We should also expect such positions to be more radical and systematic on the part of the Communist Party, whose exclusion from government has remained the unwritten rule of Portuguese politics since the first legislative elections in 1976. However, there are also some different expectations that derive from this strategic approach. More specifically, we should expect all parties in the opposition — regardless of whether they are in the centre or the flanks of the party system — to adopt more eurosceptical positions than incumbent parties (Sitter 2002b). And we should also expect marginal parties, whenever they are engaged in an office-seeking strategy, to decrease the extent to which they are willing to express opposition to Europe. Finally, given that the process of EU integration has been primarily about economic integration, we should also expect all parties to adopt more eurosceptical positions in times of economic crises.

An Overview of Parties’ Positioning towards Europe In order to examine these different arguments in the Portuguese case, we draw on parties’ electoral manifestos that have been collected and coded by the Manifesto Research Group (MRG) (Budge, et.al, 2001; Klingemann,et al., 2006)3. Figure 1 below maps the evolution of the net position of Portuguese parties on the European issue between 1976 and 2005, using MRG data. For each party, the percentage of sentences contained in each manifesto (in relation to the total number of sentences) opposing or criticizing the European Union was calculated, and then subtracted to the percentage of sentences contained in the manifesto (again in relation to the total number of sentences) that offered a positive view of European integration. A negative value means that the weight of anti-European statements in the manifesto was larger than the weight of proEuropean statements, thus producing a simple measure of the level of support for Europe expressed by parties before elections. The data are broken down over three time-periods, which were defined on the basis of three criteria: first, in order to highlight important turning points in EC/EU integration; second, to emphasise the continuities and changes in party net positioning on Europe; and third, to reflect the economic cycle and in particular the economic crises 3

The Manifest Research Group has to date published two volumes as cited here. The first volume includes manifesto data from 1945 to 1998 and the second includes data up to 2005. Thus, it was possible to position Portuguese parties since democratization until the latest 2005 legislative elections using both volumes.

7

which have occurred in the post-accession period. Thus, the first time-period starts in 1976 and ends with the signature of the Maastricht Treaty in 1991. Across this fifteen year period there is some stability in Portuguese parties’ overall stance on Europe (Appendix A)4. The signature of the Maastricht Treaty was considered an important step for European integration, in particular for the politicisation of that process and also marks the beginning of a Portuguese economic downturn that lasted until the mid-1990s. As we shall see, the first legislative election that took place after the Maastricht treaty saw important changes in the positions of parties on Europe, which were not maintained in subsequent elections. Thus, we decided to present the positions of parties for the 1995 election alone in order to single out this important election. The third period comprises three novelties: the process of EU integration takes another important step with the creation of EMU (European Monetary Union), another economic crisis looms from 19992000 onwards, and a new party obtains parliamentary representation, the left-libertarian BE. Thus, our third period depicts Portuguese parties’ positions in the advent of EMU, following the onset of an economic crisis and following a change in party system format. [Figure 1 about here]

Previous research has described the Portuguese party system as one characterized both by ‘limited contestation’ (Taggart, 2005) — with only small ‘soft’ eurosceptical parties and pro-EU positions being expressed by the main parties— and by an inverted ucurve whenever parties’ position on Europe, measured by electors’ perceptions or expert opinion, is plotted against a Left-Right scale (Hooghe, Marks, and Wilson, 2004; Eijk, and Franklin, 2004; Freire, Lobo, Magalhães, 2007). Our results, this time using manifesto data (Budge et. al., 2001; Klingemann et.al., 2006), confirm these assessments. First, as we can see, both mainstream parties — the PS and the PSD — display the most

4 An exception to this “stability trend” occurs in 1987 when around one quarter of the CDS manifesto consisted of supportive statements on Europe. That legislative election coincided with the first direct elections to the European Parliament in Portugal. At the time, the centre-right PSD seemed poised to win, and did so with an absolute majority. The CDS may have wanted to emphasise its support for the EU for fear of adding a failure to elect one deputy to the European Parliament to an almost certain poor result in the legislative election. Although this shows how this particular party was willing to use Europe for strategic purposes, it is clearly an outlier position for this party, when evaluating the whole period 19761991. Thus, it was deemed preferable to exclude that data-point from Figure 1, as well as the rest of the analysis in this paper.

8

pro-European positions, with lower levels of support for Europe concentrated in the flanks of the party system. Second, only the PCP emerges as an anti-European party between 1976 and 1987, as measured by its electoral platform, with a short-lived proEuropean message quickly replaced by renewed Euroscepticism after 1995. And while the BE predictably emerged with a anti-European position from the start, the CDS-PP’s position is perhaps the most intriguing, mirroring − albeit with larger variations −the PCP’s position: strongly Eurosceptic in the first elections after Maastricht, but resolutely pro-European in the remaining periods. Table 2 below complements the findings on Figure 1: it summarises each party’s stance on key European votes in Parliament, namely the approval of successive European Treaties. As in the manifesto data, the two mainstream parties, the PS and the PSD, emerge as unwaveringly supportive of EC/EU integration, having voted systematically in favour of all Treaties, whereas the Communists, and more recently the BE, have systematically voted against the ratification of all EC/EU treaties. The CDS is the only party which has varied its stance: having ratified the accession Treaty and the Single European Act, it decided to vote against the ratification of the Maastricht Treaty. In the following Treaty of Amsterdam, although the party decided to change its previous eurosceptic stance and vote in favour, a considerable number of MPs defected to vote against the ratification.

[Table 2 about here] Thus, between 1976 and the early 1990s, opposition to Europe was voiced exclusively by the Communist Party, and its share of the vote declined substantially from the mid-1980s onwards (Bosco, 2001). However, the adoption of a eurosceptic stance by the CDS-PP in 1995 may have contributed to the 9% share of the vote in the legislative elections, a large increase for that party which had polled 4% in the previous election.5 And although the CDS-PP quickly abandoned that Eurosceptical stance, the appearance of a new radical left party, the BE, has served to diversify and somewhat stem the decline of the eurosceptic pole, which in 2005 totaled 13% of the vote. What we need to turn to

5

In the preceding year, the Communists and the CDS-PP together scored 24% of the vote in the European Parliament elections. This was the peak of votes for the eurosceptic pole in European Parliament elections.

9

now, however, is to the more detailed description and explanation of why and how parties’ positions changed across the three periods identified above.

Party Euroscepticism since Democratization to Maastricht, 1976-1991 As has been explained elsewhere (Alvarez-Miranda, 1995; Lobo, 2003) in Portugal, until 1985, the pro/anti- Europe cleavage tends to overlap with the cleavage that separated the parties in terms of their positions on liberal democracy and the acceptance of at least some amount of market pluralism, which pitted the pro-European PS, but also the PSD and the CDS, against the Communists. This cleavage had fundamental political consequences, since it cross-cut the left-right cleavage and thus severely conditioned party relationships on the left side of the party system. The compounded cleavage which combined positions on democracy and on membership of the EC effectively prevented the Socialists and the Communists from ever forming a coalition government. Thus, until 1986, the Communist Party’s stance can be considered one of ‘hard’ euroscepticism, in that the party was simply opposed to entry into the EC. Opposition to Europe was construed mostly on the basis of economic arguments. The Communists viewed membership of the EC as a threat to key sectors of the economy and to workers’ rights that had been legislated shortly after democratization. However, political arguments were not disregarded either: the Communists considered the EC would weaken Portuguese democracy and national sovereignty in the face of international capital (Lobo, 2001:107). A few years after membership, the Communists did moderate their euroscepticism, changing from a ‘hard’ to a ‘soft’ position, as the possibility of exit from the EU seemed very unlikely and participation in the European Parliament elections warranted some kind of acceptance of the EC as a legitimate forum of political deliberation (Lobo, 2003:108). However, this did not change their relative position in the system as the most eurosceptical party.

Therefore, accession did not lead to a re-

configuration of the parties’ positions on Europe mainly because it was a congruent issue with the long-standing regime cleavage which split the left of the party system.

The Aftermath of the Maastricht Treaty: A turning point in Portuguese Party euroscepticism?

10

The analysis of party platforms for the first post-Maastricht elections, in 1995, provides a somewhat different pattern from that observed in the previous elections. Although the PCP remained the least euroenthusiatic party on the left of the party system — depicting the signature of the Maastricht Treaty as a negative development and economic and political integration as a service to ‘big capital’ (Lobo, 2003:110) — the party’s net position on Europe, as measured by MRG (Budge et. al, 2001; Klingemann et.al,, 2006), was marginally positive. The trend towards decreasing euroscepticism since the late 1980s on the part of the PCP can be best understood as a result of adaptation to changing domestic conditions. Although, until the mid-1980s, the Communists still expected to be able to "force themselves" upon a weakened Socialist Party as a future coalition partner, their own electoral stagnation and internal crisis following the fall of the Berlin Wall forced a shift on the part of the PCP in favour of both pluralist democracy and European integration, in the hope of rendering themselves legitimate and acceptable partners of a leftist coalition (Bosco, 2001: 337-338; 349-350). In an even more striking shift, the first post-Maastricht elections led to a change on the right of the party system, with the CDS-PP presenting itself as the most resolutely anti-European force in the Portuguese party system (Figure 1). Starting in 1985, the party suffered ever worsening electoral scores at successive elections, in a continuous loss of electorate to the PSD, from which it had become ideologically indistinguishable and in which the most rightist voters eager for a previously elusive governmental stability became willing to vote strategically (Bacalhau 1989; Goldey, 1992; Robinson, 1996; Magone 1999). Following the 1991 elections, in which the PSD obtained a second consecutive absolute majority and CDS-PP was left with an all-time low score of 4.4 percent, the party elected a new young leader, Manuel Monteiro, who brought in a radically new agenda to stop the party’s electoral decline. Euroscepticism, particularly anti-Maastricht positions, became at this point a central tenet of this party’s agenda. This right-wing euroscepticism was different from that which had always been emphasized by the PCP: instead of an emphasis on the decline in economic sovereignty and its consequences for the social ‘losers’ of the prointegration policies of the mainstream parties, the CDS-PP focused primarily on the loss of formal political sovereignty as a consequence of integration (CDS-PP, 1995:24). In

11

any case, 1995 brought an important difference in relation to previous elections. Although the inverted u-curve pattern persisted and a strong pole of opposition to Europe remained in the party system, that pole had suddenly shifted to the right of the party system.

Party Euroscepticism within a European Monetary Union, 1999-2005 The third period portrayed in Figure 1 covers the introduction of the single currency up to and including the last legislative elections, and a period of protracted economic crisis (1999-2005). Again, new changes are observable. First, the 1995 position of the CDS-PP seems to have been abandoned in favour of a renewed pro-Europe stance, almost indistinguishable from that adopted by the PSD. Second, the Communist party has returned to a clear eurosceptical stance, after some moderation in the 1990s. Third, the newly-created party, the Bloco de Esquerda (BE) has adopted a eurosceptical stance which is even more critical of the EU than that of the Communist Party itself. In fact, although the CDS-PP’s 1995 strategy seemed to pay off in the domestic electoral front (Table 1) those elections were soon followed by local elections in 1997, in which negotiations for many city-specific coalitions with the PSD ultimately failed, and the party’s results were disappointing (Magalhães & Doncel Luengo, 1999). Those results put into question the party's ability to independently undermine support for the PSD and, thus, the viability of any strategy that discarded a centre-right coalition. Thus, in 1997, Monteiro was replaced by his former mentor — Paulo Portas — who since then systematically played down the party's anti-European discourse. The CDS-PP’s electoral platforms since 1995 not only deemphasised the European issue but, when dealing with it, adopted a generally pro-European stance. In the 2005 election programme, the one which followed the CDS-PP participation in government with the PSD after twenty years in opposition, the party even described itself as a ‘Euro-calm’ party, by which it explicitly meant to convey a general — albeit careful and qualified — support for integration (CDS, 2005:135). In that manifesto, the Project of a European Constitutional Treaty was even discussed as a positive development, even though the party advocated the need for a popular referendum before adopting such a Treaty (CDS: 2005: 36).

12

In what concerns the PCP, following a 1999 election where the party still presented itself as relatively neutral vis-à-vis integration (see appendix A), 2002 saw their return to the eurosceptical camp (PCP, 2002:113). Despite voicing political concerns, the PCP continued to place greater emphasis on the economic consequences of EU integration. Not surprisingly, in its electoral program of 2005, the Communists opposed the Constitutional Treaty of the EU and demanded that a referendum take place before it is adopted in Portugal (PCP, 2005: 120). Finally, as we can see in figure 3, the Leftist Bloc emerged as firmly and increasingly eurosceptical. In 2005, the BE’s electoral program dwelled on the negative consequences of integration for the living conditions of Portuguese workers and citizens (Bloco de Esquerda, 2005). Although most of the criticism of Europe occurs on economic grounds, the BE also devotes a few paragraphs to the European Constitutional Treaty which it opposes (ibid., 108). Thus, the EU has relative salience in this program, and overall the emphasis is clearly negative, especially from an economic standpoint. On the basis of this broad description of the parties’ positions on the European issue, what can be said about the causes of party-based euroscepticism in Portugal? First, throughout the entire period, clear opposition to Europe stemmed from parties in both flanks of the party system. The nature of euroscepticism shows, however, some important differences on the Left and on the Right. The Communists gave more detailed attention to economic issues than the CDS-PP, which focused rather more on the political consequences of integration. Second, parties’ placements have been subjected to sometimes abrupt shifts that an ideological aaccount is hard pressed to explain. On the one hand, there is the position of the CDS-PP, which moved from a moderately proEuropean party to a clearly eurosceptical one, only to return again to pro-Europeanism before and during its participation in a coalition government with the PSD. On the other hand, the trajectory of the Communists is also intriguing: starting out as the most clearly eurosceptical party, they increasingly moderated that stance until 1999. Only after the 1999 elections, following the appearance of the BE, did the Communists return to a more resolutely eurosceptical stance. Thus, independently of fundamental structural and ideological constraints, the changes in the positions of the different forces in the flanks of

13

the Portuguese party system can only be explained in terms of strategic considerations about the best way to conciliate vote- and office-seeking strategies.

A Model of Party Positions towards Europe In this section we propose to use the party manifestos data in order to formally test some general explanations advanced above. Our hypotheses are the following: firstly, parties who position themselves on the extremes of the party system tend to be eurosceptical; secondly, parties that are systematically excluded from government tend to be eurosceptical; thirdly, variations in parties’ net position — in their electoral manifestos — in relation to Europe are related to opposition status, regardless of whether parties are located in the centre or at the extreme of the party system. Fourthly, variations on parties’ position on Europe should be related to economic conditions. In order to test this hypothesis, two economic indicators were selected: unemployment level and the yearly GDP per capita growth, both measured at election year6. Studies investigating mass attitudes to the European Union have shown the importance of economic conditions as determinants of those attitudes (Gabel, 1998). Given that the process of European integration was perceived until relatively recently as a mechanism dealing above all with economic integration, this is not surprising (Richardson, 2001; Milward, 2000). Finally, we add a dummy variable coded 1 for all post-Maastricht elections, testing the hypothesis that the general demise of the ‘permissive consensus’ on European integration that followed Maastricht and the Danish referendum produced a general effect of greater euroscepticism in party discourses.

[Table 3 about here] Table 3 shows the results of a linear regression analysis where each party’s net position on European integration — as measured in electoral manifestos — is the dependent variable (Appendix A). As we can see, in table 3, lower levels of support for Europe in party platforms tend to result not only from parties location away from the 6

Data retrieved from "The Conference Board and Groningen Growth and Development Centre, Total Economy Database, May 2006, http://www.ggdc.net". Series up to 1990 are mostly derived from Maddison (2003). From 1990 onwards, series for OECD countries are mostly derived from the most recent editions of OECD National Accounts.

14

centre but also from the mere condition of running an election from the opposition. Furthermore, the degree of euroenthusiasm expressed by parties is also affected by economic growth. In other words, in Portugal, euroscepticism (or, at least, a less euroenthusiastic discourse) seem to result not only from the marginality of parties in the system, but also from opposition status and economic crisis.

Popular Euroscepticism: Europe as an Electoral Cleavage As we have just seen, euroscepticism has provided the basis of a relevant political cleavage between Portuguese parties, distinguishing government and centrist parties from those in the opposition and positioned in the flanks of the party system. But has euroscepticism provided the basis of a relevant electoral cleavage in Portugal? The notion that ‘Europe’ has become an autonomous and behaviourally consequential dimension of domestic party competition in European countries has been met with considerable scepticism by several scholars. European integration seems to have failed to become a particularly salient issue in everyday politics in comparison with other more clearly domestic issues (Mair 2000; Taggart and Szczerbiak, 2002; Taggart and Szczerbiak, 2005). Either because mainstream parties wish to avoid internal dissent or because marginal parties are forced to solve the potential contradictions between electoral and office goals, the importance of the European issue can be explicitly downplayed in electoral competition and, thus, inter-party differences rendered imperceptible for mass publics. As a result, in spite of real levels of mass-level polarization in terms of European orientations, those orientations inconsequential in terms of political behaviour, as voters’ have remained apparently unable to perceive meaningful differences between parties in this respect (Eijk and Franklin 2004). Others, however, have suggested that the ‘sleeping giant’ of European politics may have already awakened, at least in some cases and under particular circumstances. In a seminal article, Gabel shows that, to a varying degree, ‘EU citizens’ support for EU membership appears (…) to have a statistically significant influence on vote choice in most EU member states’ (2000: 62). The existence and magnitude of that influence in concrete countries and circumstances seems to be contingent upon several factors. First,

15

the existence of some amount of polarization at the mass-level in what concerns attitudes towards integration. Second, that such mass-level attitudes towards integration are not fully subsumed under more traditional sources of vote choice among the electorate, i.e., left-right orientations (Gabel, 2000). And finally, that the European integration issue has become sufficiently salient to be used by voters in detriment of others as a basis for vote choice. The fulfilment these conditions seem to be behind the surprising electoral relevance of European integration in cases such as Britain since the 1990s, (Evans, 1998, 2002), Austria, Finland, and Sweden immediately before and after the accession (Tillman, 2004), and Denmark in the late 1990s (Netjes, 2004). What do we know about Portugal in this respect? First, a broad mass-level consensus about the benefits of integration seems to have prevailed ever since the country became a member of the EU. On the basis of Eurobarometer data from 1986 to 2006, on average, at least two out of three Portuguese have considered that the country has benefited from membership in the EU or that membership is “a good thing”, clearly above the EU average. However, as figure 2 shows, there are periods where such consensus seems to have been undermined, at least temporarily. From 1991 – year of the signature of the Maastricht Treaty — until 1997, aggregate negative perceptions regarding the benefits from integration soared from a mere 8 percent to about 20 percent. And after a new decline in such negative perceptions, they again increased since 1999, reaching about 30 percent in late 2006. Thus, there have been sizeable minorities of citizens that have been either ambivalent or downright negative about the consequences of integration, minorities that have tended to swell at crucial periods.

[Figure 2 About here] Second, it has been argued that the attitudes of this sizeable minority have been systematically correlated with certain party preferences. Bacalhau, for example, shows that, before accession, voters supporting the PCP were already less supportive of entry in the EEC in comparison with the rest of the electorate (Bacalhau 1994: 184). To a great extent, this pattern was prolonged following accession: as Lobo notes, ‘the Communist electorate is systematically less enthusiastic about membership of the EU than all other parties’ electors for the entire period’ (Lobo 2003: 112). This correlation, however, does

16

not necessarily mean that EU orientations are in and of themselves explanatory of voting decisions, especially considering that one of the most important factors behind vote choices in Portugal — left-right ideology — also seems to be correlated with EU attitudes, with the more leftist voters displaying greater scepticism vis-à-vis the advantages of membership (Bacalhau 1994: 270; Gramacho and Llamazares 2005).Thus, in order to ascertain whether EU orientations have ever, in and of themselves, formed the basis of a relevant electoral cleavage, particularly during the periods of greater polarization around the European integration issue, we first need to determine the extent to which such EU orientations have been independent from left-right orientations. This entails, first of all, dealing with the issue of how citizens’ EU attitudes can be captured empirically. As Gabel notes, an analysis of this sort focusing exclusively on voters’ support for EU membership is likely to miss out on the possibility that ‘specific issues of European integration may have distinct effects on vote choice for particular parties in certain national contexts’ (2000: 64). Furthermore, as we saw in the previous section, parties’ discourse on Europe seems to be itself multidimensional: while some stress the dimension related to the economic consequences of integration — a common pattern in the eurosceptical parties in the left — others place greater emphasis on its political consequences — the main emphasis of the CDS-PP in the 1995 elections. As it happens, the same multidimensionality seems to be present at the level of popular attitudes. Recent scholarship has suggested that, across Europe, at least two different dimensions can be found in this respect. The first is a dimension of instrumental euroscepticism, referring to the extent to which individuals ‘are sceptical about the benefits of the EC/EU for their country’, captured by indicators such as those we have described above: the perception of benefit from being a member of the EC/EU and the evaluation of the consequences of membership for one’s country (Lubbers and Scheepers 2005). This dimension could also be said to tap into the (lack of) ‘specific support’ of ‘output-oriented legitimacy’ bestowed on the European Union (Chierici, 2005). The second is a political euroscepticism dimension, referring to citizens’ rejection of the idea to provide more power to EC/EU to deal with policy issues, and captured by indicators such as individuals’ support for joint decision-making within the European Union in different policy areas (Lubbers and Scheepers 2005). This different dimension of

17

citizens’ orientations is arguably linked to the (lack of) ‘diffuse support’ to (or ‘inputlegitimacy’ of) the EC/EU (Chierici 2005), and has been linked, at the mass level, both to feelings of distrust vis-à-vis European institutions and to perceptions of threat to the preservation of national identity (McLaren 2005). Thus, we examine this multidimensionality in the Portuguese case by calculating two indices, one for instrumental and another for political euroscepticism. The former was measured by averaging, for each respondent, the values of the variables ‘(lack of) support for membership’7 and ‘(lack of) perception of benefits from membership’8. Political euroscepticism was measured in a similar way, but this time by averaging, for each respondent and survey, the responses to eight different questions concerning the (lack of) support for joint-decision making within the EC/EU across the policy areas on which longer series are available, i.e, health and welfare, education, protection of the environment, defence, scientific and technological research, rules for broadcasting and press, currency and foreign policy in relation to non-EU countries.9 The values of the two indexes range from 1 (least eurosceptical) to 3 (most eurosceptical).10 Finally, we calculated the correlations between those indexes and individuals’ left-right orientations, measured in the conventional 10-point scale, in the three Eurobarometer surveys available that were carried closest to the 1991, 1995 and 1999 elections in Portugal and contained all questions measuring our variables of interest: EB 36.0 (October 1991); EB 43.1 (May 1995); and EB 51.0 (June 1999). These elections are particularly relevant because they provide interesting contrasts in terms of the party positions and mass-level attitudes. On the one hand, in 1991 and 1995, as we saw in the previous section, a high polarization of party positions around the European issue − with great distances between the most and the least pro-European party — contrasts with a lower polarization in 1999 (see appendix A). On the other hand, there is also a contrast between 1995 and the 7

Recoded as membership ‘good thing’: 1; ‘bad thing’: 3; and neither good nor bad’ ‘ and DK/NA: 2. In this and other variables, we recoded DK/NA as the middle category, following Rohrschneider (2002). 8 Recoded as ‘benefited’: 1; DK/NA:2; ‘did not benefit’: 3. 9 Recoded as ‘jointly within Europe’:1; DK/NA:2; ‘national government’:3. 10 We conducted a factor analysis (principal axis factoring, Varimax rotation) of all items used to measure instrumental and political euroscepticism in the Eurobarometer Trend File, and the results confirmed the existence of two different dimensions, with ‘membership’ and ‘benefit’ forming one dimension and questions about joint-decision making in the EU forming another. For the items measuring instrumental euroscepticism (‘benefit’ and ‘membership’), Crombach’s alpha reaches .71, and .84 for the items used to build the political euroscepticism scale.

18

remaining elections, with euroscepticism (particularly expressed in political terms) being voiced by the right − the CDS-PP − in 1995. As the results in table 4 show, in two of the three surveys, instrumental euroscepticism does tend to be higher the more to the left voters place themselves. However, that correlation was already weak in 1991 and has, by 1999, disappeared altogether. In other words, how citizens evaluate the costs and benefits of European integration seems to be largely and increasingly independent of their ideological positions. The same could be said for citizens’ views vis-à-vis the ceding of sovereignty on crucial policy areas to the European Union, with the difference that such views are not significantly correlated with left-right self-placement in any of the three surveys. This suggests that EC/EU attitudes have had the potential to have been used, in and of themselves, as consequential guides for voting choices by citizens in Portugal.

[Table 4 about here] However, whether that has indeed been the case requires us to take an additional step: to estimate models for vote choice in the three elections under scrutiny. In the following analysis, our dependent variable is ‘vote intention’, coded as ‘vote for the incumbent’ (the PSD in 1991 and 1995, the PS in 1999), 1; ‘vote for the main opposition party’, 2; ‘vote for the Communist Party’, 3; and ‘vote for the CDS-PP’, 4.11 Since this dependent variable is nominal and has more than two categories, multinomial logit regression is an appropriate technique. We test the impact of EU attitudes on vote choice while controlling for the effects of other potentially confounding variables. Besides introducing left-right self-placement, we employ gender, age, education and income as generic socio-demographic controls.12 Furthermore, we also added ‘(dis)satisfaction with national democracy’ as a control variable.13 Such dissatisfaction has been shown to directly affect attitudes towards European integration, as it tends to used by voters as an ‘informational shortcut’ through which they evaluate the advantages of membership (Anderson 1998). On the other hand, dissatisfaction with democracy also seems to form 11

Vote intentions for other parties, blank and null votes and non-voters are excluded from the analysis. In EB 36 and 51.0, income terciles were used. In EB 43.1, since no data on respondent’s income was collected, we added education to the model, measured as age at which respondent finished education (or current age if still studying). 13 ‘Very satisfied’:1; ‘Fairly satisfied’: 2; DK/NA:3; ‘Not very satisfied’:4; ‘Not at all satisfied’:5. 12

19

part — together with evaluations of economic and government performance — of an overall syndrome of political discontent, one of whose direct consequences is to vote against the incumbent (Montero, Torcal and Gunther 1997; Gunther and Montero 2006). Thus, by adding ‘dissatisfaction’ as a control variable, we can determine the extent to which euroscepticism exerts an effect on vote choices that is independent what it already entails in terms of mere discontent with governmental performance. The results of this analysis are presented in table 5.

[Table 5 about here] The results lend some credence to the notion that EC/EU attitudes have been important in and of themselves, at least in some elections, in shaping voters’ choices. First, in both 1991 and 1995, voters who perceived greater costs resulting from European integration were less likely to vote in the incumbent and more likely to vote instead in the PS, the PCP, or even (in 1995) the CDS-PP. The magnitude of such effects can be illustrated by calculating the effect of changes in the instrumental euroscepticism variable in the predicted probabilities of voting for the different parties in the different elections, while keeping the remaining variables at their mean values (or at the mode for the dichotomous variable gender): as the value of the instrumental euroscepticism varies from its minimum to its maximum value, the probability of voting for the main opposition party increased by more than .20 both in the 1991 elections (which the Socialists lost) and the 1995 elections (which they won).14 Note, however, how in 1999 instrumental euroscepticism produced no relevant effects on voting choices. A different story emerges from the analysis of how political euroscepticism has affected voting choices. In all three surveys, voters in the PS and the PSD (regardless of which of them was the incumbent party) were no different from each other in terms of their levels of political euroscepticism. This is hardly surprising, considering that, as we have seen, at the level of party positions and discourse, the cleavage around the implications of European integration for national sovereignty has never divided the two mainstream parties. However, this does not mean that such dimension of citizens’

14

Calculated with X-Post: Post-Estimation Interpretation Using Excel, by Simon Cheng and Scott Long, downloadable at http://www.indiana.edu/~jslsoc/xpost.htm.

20

attitudes vis-à-vis the EU has lacked consequences for political behaviour. In 1995, voters in both the PCP and the CDS-PP were different from those who voted for the incumbent (and, for that matter, those who voted for the main opposition party), in that they tended to exhibit significantly higher levels of political euroscepticism. More specifically, those voters positioned at the maximum value of the political euroscepticism variable were almost seven times more likely to have voted for the CDS-PP instead of the incumbent (and six times more likely to have voted for the PCP) than those positioned at the minimum value, holding all other variables constant. Thus, there are reasons to believe that the variations existing at the level of mass attitudes in what concerns the consequences and paths of European integration have been, at least under particular circumstances, consequential in terms of political behaviour in Portugal. On the one hand, in the 1991 and 1995 elections, incumbents were punished by voters who perceived fewer benefits from integration. Although this finding certainly needs to be confirmed by further research, it suggests that the tendency of opposition parties in general to adopt more Eurosceptic positions is a cue to which voters are not oblivious, and that government parties in Portugal have been held accountable for the perceived benefits and costs of Europe. Conversely, such accountability may be neutralized under specific contexts, such as the one surrounding the 1999 elections in Portugal, when the presentation of the entry in the Euro-zone as a ‘national objective’ caused an acute depolarization of the European integration issue in party discourses and appeals. On the other hand, citizens’ views regarding the appropriateness of sharing sovereignty within European institutions have also produced an impact on voting choices. However, this was limited to only one of the three elections we analyzed — the 1995 elections — where citizens’ political euroscepticism was channelled towards a vote in the more extreme parties, the PCP and the CDS-PP. In fact, several aspects seem to make these elections special in the context of Portuguese democratic history. They were the first post-Maastricht elections where, for the first time, a clear eurosceptical tenor dominated the discourse of a party on the right of the system, the CDS-PP. This euroscepticism was especially articulated in terms of an explicit opposition to ‘European federalism’ as pre-figured by Maastricht. However, by 1999, the CDS-PP returned to a

21

moderate position on European integration in view of the need to become acceptable as a potential partner in a centre-right coalition, and political euroscepticism became again (as in 1991) an irrelevant cue for electoral behaviour.

Conclusion Our concern in this paper was to locate the positions of the Portuguese parties visà-vis European integration, signal and account for the cases and levels of euroscepticism in those positions, and evaluate whether those positions have been at all consequential at the level of mass political behaviour. We found that parties in both the left and right of the party system have, at one point or another in Portuguese democratic history, adopted clear eurosceptical positions. This has especially been the case on the left, where the Communist Party has formed the most consistent pole of party-based criticism of European integration, in which it has been accompanied by the Leftist Bloc since 1999. But also the rightist CDS-PP has been generally more eurosceptical than either of the two mainstream centre-left and centre-right parties (the PS and the PSD), particularly in the first half of the 1990s. In other words, euroscepticism has been a discourse mostly reserved to the ‘non-mainstream’ parties. This is a general expectation that results from both ideological and strategic approaches to the causes of party-based euroscepticism. However, we also found that this structural regularity coexists with important short-term shifts. On the one hand, we saw how parties, regardless of their distance vis-àvis the centre, have expressed lower levels of support for Europe whenever they were in the opposition and during periods of economic downturn, as measured by GDP per capita change. On the other hand, we also saw how the PCP and the CDS-PP have downplayed or reinvigorated their opposition to Europe depending on the specific vote- or officeseeking strategy they were pursuing at the time. Since the late 1980s, the Communists have gradually shifted to less anti-European positions, even after the decline in public support for membership that followed the Maastricht treaty and the Danish referendum. Only after the 1999 elections, which brought the emergence of a new competitor on their side of the ideological spectrum and revealed the Socialists’ intransigence in what concerned a possible coalition in the left, did the PCP return to a strong anti-European

22

position. As for the CDS-PP, their support for Europe was abruptly replaced by a clear eurosceptical message in the 1995 elections, to be completely discarded as soon as getting into office became the main goal. Portugal is thus an interesting case where euroscepticism clearly emerges both as a structural and a contingent phenomenon. Finally, we appraised whether the cleavages around the European issue that have characterized at the level of political competition have been reflected at the mass level. On the one hand, we found that voters’ attitudes vis-à-vis the benefits of integration for Portugal have indeed been related to their voting behaviour, particularly in what concerns voting for or against incumbents and in circumstances of relevant polarization of party positions on the European issue. On the other hand, we also found that, when the CDS-PP attempted to activate a cleavage around the political consequences of integration — particularly in what concerns the loss of sovereignty to Europe — it was successful in this regard, turning the political euroscepticism of some voters into a consequential guide for voting choices. Further research is needed on this issue. This is particularly the case in what concerns the need for more refined measures of party placements; better specified models to explain those placements, particularly comprising the possibility of mutual influences between party cues and mass attitudes; more fully specified models of voting behaviour; and tests of hypotheses about voting choices in other elections other than 1991, 1995 and 1999, the only ones on which we had data available on the relevant variables. However, our results clearly point to two fundamental possibilities. On the one hand, that the positions adopted by parties on the European issue seem to be much more flexible than what an exclusively ideological approach to the causes of euroscepticism is able to explain. Portuguese parties have enjoyed enough autonomy to modulate their discourse in Europe whenever their strategic goals and the context of party competition in which they were framed so dictated. On the other hand, the mobilization of Portuguese voters on the basis of specifically anti-European attitudes is not only possible — as several authors have pointed out when talking about Europe in general (Eijk and Franklin, 2004) and Southern Europe in particular (Gramacho and Llamazares, 2005) — but has, in fact, already occurred, both on the basis of the perceived economic costs of integration and of its political costs. Whether parties will be willing to explore these avenues for

23

mobilization is likely to depend not only of the overall levels of instrumental and political support for European integration but also of the constraints that their office-seeking goals pose (or not) on their vote-seeking strategies.

24

[Appendix here]

25

References Alvarez-Miranda, B., (1995) ‘A las puertas de la comunidad: consenso y disenso en el sur de Europa’, Working Paper, Juan March Institute. Anderson. C. J. (1998) ‘When in Doubt, Use Proxies: Attitudes Towards Domestic Politics and Support for European Integration’, Comparative Political Studies, vol. 31, no. 5, pp. 569601. Bacalhau, M. (1989) ‘Mobilidade e transferência de voto através das sondagens’, in Portugal: O Sistema Político e Constitucional, ed. M. B. Coelho, Lisbon, ICS, pp.237-256. Bacalhau, M. (1994) Atitudes, opiniões e comportamentos politicos dos portugueses: 1973-1993, Lisbon, author’s edition. Bosco, A. (2001) ‘Four Actors in Search of a Role: The Southern European Communist Parties’, in Parties, Politics, and Democracy in the New Southern Europe, eds. P. N. Diamandouros & R. Gunther, Baltimore, Johns Hopkins University Press, pp. 329-387. Bloco de Esquerda, (2005) ‘Tempo de Viragem- As prioridades para a Governação que imponham um novo Ciclo de Políticas’, Programa Eleitoral do Bloco de Esquerda para as eleições legislativas de 2005, Lisboa. Budge, I., Klingemann, H.D., Volkens, A., Bara, J. & Tennenbaum, E. (eds.), (2001) Mapping Policy Preferences, Estimates for Parties, Electors and Governments, 1945-1988, Oxford, Oxford University Press. Castilho, J. M., (2000) A Ideia de Europa no Marcelismo, Porto, Afrontamento. CDS-PP, (1995) Programa Eleitoral e de Governo do Partido Popular. CDS-PP, (2005) Programa de Governo. Chierici, C. (2005) ‘Is there a European Public Opinion? Public support for the European Union, theoretical concepts and empirical measurements’, paper presented at POLIS 2005, Paris. Eijk, C. van der & Franklin, M. N. (2004) ‘Potential for Contestation on European Matters at National Elections in Europe’, in European Integration and Political Conflict, eds. G. Marks & M. R. Steenbergen, Cambridge, Cambridge University Press, pp. 32-50. Evans, G. (1998) ‘Euroscepticism and Conservative Electoral Support: How an Asset Became a Liability’, British Journal of Political Science, vol. 28, no.4, pp. 573-590. Evans, G. (2002) ‘European Integration, Party Politics and Voting in the 2001 Election’, British Elections and Parties Review, vol. 12, no. 1, pp. 95-110. Freire, A, Lobo, M. C. & Magalhães, P. C. (2007) ‘Left-right and the European Parliament vote in 2004’, in European Elections after Eastern Enlargement: Preliminary Results from the

26

European Election Study 2004, eds. M. Marsh, S. Mikhaylov & H. Schmitt, Mannheim, CONNEX, pp. 97-140. Gabel, M., (1998) Interests and Integration: Market Liberalization, Public Opinion, and European Union, Michigan, University of Michigan Press. Gabel. M. (2000) ‘European Integration, Voters, and National Politics’, West European Politics, vol. 23, no. 4, pp. 52-72. Goetz, K. & Hix, S. (eds.) (2000) Europeanised Politics? European Integration and National Political Systems, London, Frank Cass. Goldey, D. B. (1992) ‘The Portuguese Elections of 1987 and 1991 and the Presidential Election of 1991’, Electoral Studies, vol. 11, no.2, pp. 171-176. Gramacho, W & Llamazares, I. (2005) ‘Euroskeptics among Euroenthusiasts: An Empirical Exploration of Lukewarm Europeanism in Greece, Portugal, and Spain’, Paper presented at the conference Euroskepticism: Causes and Consequences, July 1 - 2, 2005 Pulitzer hotel, Amsterdam. Gunther, R. & Montero, J. R. (2006) ‘The Multidimensionality of Political Support for New Democracies: Conceptual Redefinition and Empirical Refinement’, in Political Disaffection in Contemporary Democracies: Social Capital, Institutions and Policies, eds. M. Torcal & J. R. Montero, London, Routledge, pp.46-78. Hix, S. (1999) ‘Dimensions and Alignments in European Union Politics: Cognitive Constraints and Partisan Responses’, European Journal of Political Research, vol. 35, no 1, pp. 69106. Hix, S. (2005) The Political System of the European Union, Second Edition, Basingstoke, Palgrave Macmillan. Hooghe, L., Marks, G. & Wilson, C. (2004) ‘Does Left/Right Structure Party Positions on European Integration?’ in European Integration and Political Conflict, eds. G. Marks & M. Steenbergen, Cambridge, Cambridge University Press, pp.120-40. Klingemann, H.-D., Volkens, A., Bara, J. & Budge, I. (eds.) (2006) Mapping Policy Preferences II: Estimates for Parties, Electors and Governments in Central and Eastern Europe, European Union and OECD 1990-2003, Oxford, OUP. Leitão, N. Andresen (2007) O Estado Novo, a Democracia e a Europa, Lisbon, ICS. Lobo, M.C. (2001) ‘The role of parties in Portuguese Democratic Consolidation’, Party Politics, vol. 7, no.5, pp. 643-653. Lobo, M. C. (2003) ‘Portuguese Attitudes Towards EU Membership: Social and Political Perspectives’, in Spain and Portugal in the European Union: The First Fifteen Years, eds. S. Royo & P. C. Manuel, London, Frank Cass, pp. 97-118.

27

Lubbers, M. & Scheepers, P. (2005) ‘Political versus Instrumental Euro-scepticism: Mapping Scepticism in European Countries and Regions’, European Union Politics, vol. 6, no. 2, pp. 223-242. Maddison, A. (2003) ‘The World Economy: Historical Statistics’, OECD Development Centre. Magalhães, P. C. & Doncel Luengo, J. A. (1999) ‘Informe sobre la actividad políticoconstitucional portuguesa durante 1997’. In Jornadas Luso-Españolas de Derecho Constitucional (Constitución y Unión Europea), ed. P. Pérez Tremps, Mérida, Editora Regional de Extremadura. Magalhães, P. C. (2003) ‘Elections, Parties, and Policy-making Institutions in Democratic Portugal”, in Contemporary Portugal: Politics, Society and Culture, ed. A. C. Pinto, Boulder, Social Science Monographs. Magone, J. (1999) ‘Portugal: Party System Installation and Consolidation’, in Changing Party Systems in Western Europe, eds.D. Broughton & M. Donovan, London, Pinter. Mair, P. (2000) ‘The Limited Impact of Europe on National Party Systems’, in Europeanised Politics? European Integration and National Political Systems, eds. K. H. Goetz & S. Hix, London, Frank Cass, pp. 27-51. Marks, G. et al. (2006) ‘Party Competition and European Integration in the East and West: Different Structure, Same Causality’, Comparative Political Studies, vol. 39, no.2, pp. 155-175. McLaren, L. (2005) ‘Explaining Mass-Level Euroskepticism: Identity, Interests and Institutional Distrust’, paper prepared for the Annual Meeting of the American Political Science Association, 1-4 September 2005, Washington. Milward, A. (2000) The European Rescue of the nation-state. London, Routledge. Montero, J. R., Gunther R. & Torcal, M. (1997) ‘Democracy in Spain: Legitimacy, Discontent and Disaffection’, Studies in Comparative International Development, vol. 32, no. 3, pp. 124–160. Netjes, C. E. (2004) ‘Sleeping Giant Or Much Ado About Nothing? Examining the role of attitudes towards European integration on National vote choice in Denmark’, Paper submitted for the conference ‘The European construction through the prism of the June 2004 European Parliamentary Elections’ Strasbourg, 18th-19th of November 2004. PCP (2002, Programa eleitoral do PCP – Legislativas 2002. PCP (2005) Programa eleitoral do PCP – Legislativas 2005. Reis, B. Cardoso (forthcoming) ‘The Portuguese answer to the European question during the New State’, in South European Society and Politics. Richardson, J. (ed. ) (2001) European Union: Power & Policy-Making, London, Routledge.

28

Robinson, R. (1996) ‘Do CDS ao CDS-PP: o Partido do Centro Democrático Social e o seu Papel na Política Portuguesa’, Análise Social, vol. 31, no.138, pp.951-973. Rohrscheider, R. (2002) ‘The Democratic Deficit and Mass Support for an EU-Wide Government’, American Journal of Political Science, vol. 46, no. 2, pp. 463-475. Rosas, F. & Oliveira, P. (eds.) (2004) A Transição Falhada: O Marcelismo e o Fim do Estado Novo, Lisbon: D. Quixote. Sitter, N. (2001) ‘The Politics of Opposition and European Integration in Scandinavia: Is Euroscepticism a Government-Opposition Dynamic?’, West European Politics, vol. 24, no. 4, pp.22-39. Sitter, N. (2002a) ‘Opposing Europe: Euro-Scepticism, Opposition and Party Competition’, Opposing Europe Research Network Working Paper No 9. Sitter, N., (2002b) ‘Cleavages, Party Strategy and Party System Change in Europe East and West’, Perspectives on European Politics and Society, vol. 3, no. 3, pp.425-451. Taggart, P. (1998): ‘A touchstone of dissent: euroscepticism in contemporary West European Party Systems’, European Journal of Political Research, 33, pp.363-388. Taggart, P. & Szczerbiak, A. (2005) ‘Three Patterns of Party Competition Over Europe’, Paper presented at the conference Euroskepticism: Causes and Consequences, July 1 - 2, 2005 Pulitzer hotel, Amsterdam. Taggart, P. & Szczerbiak, A. (2002) ‘The Party Politics of euroscepticism in EU Member and Candidate States' OERN Working Paper No 6, April 2002. Tillman, E.R. (2004) ‘The European Union at the Ballot Box? European Integration and Voting Behavior in the New Member States’, Comparative Political Studies, vol. 37, no. 5, pp. 590-610.

29

TABLES

Marina Costa Lobo and Pedro C. Magalhães, Instituto de Ciências Sociais, Universidade de Lisboa.

Table 1: Main Parties’ Electoral Results in Legislative and European Parliament Elections, 15 1976-2005 Legislative Elections: Year: 1976 1979 BE PCP 19 14 PS 27 35 PSD 24 43* CDS-PP 16 EP elections: Year 1987 1989 BE PCP 12 14 PS 22 29 PSD 37 33 CDS-PP 15 14

1980

1983

1985

1987

1991

1995

17 27 45*

18 36 27 13

15 21 30 10

12 22 50 4

9 29 51 4

9 44 34 9

1994 11 35 34 12

1999 10 31 43 8

1999 2 9 44 32 8

2002 3 7 38 40 9

2005 6 7 45 29 7

2004 5 9 45 33* -

* In this year, the PSD and the CDS-PP ran in coalition together. Source: www.cne.pt.

15

Percentages for the parties being analysed were rounded to the unit. In 1979 and 1980, the PSD and the CDS formed a pre-election coalition called AD (Aliança Democrática). It ended in 1982 and in all the subsequent legislative elections the parties have run independently. The PCP has since 1991 run in a coalition with the Green Party – Partido Ecologista os Verdes. This is a micro party with no autonomy from the Communist Party and thus does not have to be acknowledged above.

30

5 4 3 2 1 0 -1 -2 BE

PCP

PS

PSD

CDS-PP

Party Average Positioning 1976-1991 (CDS average excludes 1987) Party Positioning in 1995 Party Average Positioning 1999-2005

Fig. 1. The parties’ net position on Europe across three periods 1976-1991; 1995; 1999-2005 Source: Budge et. al. 2001 and Klingemann, et.al., 2006.

Table 2: Parties’ Parliamentary Vote on Ratification of Successive EU Treaties16: BE PCP PS PSD CDS-PP Single European Act Against In favour In favour In favour Treaty of Accession Against In favour In favour In favour Treaty of Maastricht Against In favour In favour Against* Treaty of Amsterdam Against Against In favour In favour In favour** In favour In favour In favour Treaty of Nice Against Against * Freitas do Amaral, MP and historic CDS leader, defected the party to vote in favour of the Maastricht Treaty. ** A group of 9 CDS-PP MPs voted against the Treaty of Amsterdam. Source: Assembleia da República, Diário da Assembleia da República online.

16

Portugal is one of the member-states which had not ratified the European Constitutional Treaty before the French and Dutch failed referendums in May/June 2004, and suspended the process thereafter. The idea of holding a referendum on the Treaty which had been promised by the Socialists also seems increasingly less likely.

31

Table 3: Explaining Parties Positioning towards the EU. Dependent Variable: Net Position of Parties on Europe, 1976-2005 Independent variables Extreme party (PCP, CDS, BE=1; PS, PSD=0) Opposition dummy Unemployment Level GDP per capita growth Post-Maastricht dummy Intercept Adjusted r2 Durbin-Watson N

-1.72** (.55) -1.62* (.62) .06 (.17) .26* (.13) -.25 (.97) 2.12 (1.10) .37 2.24 46

*p<.05; **p<.01; ***p<.001; OLS coefficients; standard error in parenthesis.

90 80 70 60 50 40 30 20 10

M ar -8 6 M ar -8 7 M ar -8 8 M ar -8 9 M ar -9 0 M ar -9 1 M ar -9 2 M ar -9 3 M ar -9 4 M ar -9 5 M ar -9 6 M ar -9 7 M ar -9 8 M ar -9 9 M ar -0 0 M ar -0 1 M ar -0 2 M ar -0 3 M ar -0 4 M ar -0 5 M ar -0 6

0

Membership good thing

Membership bad thing

Country benefited

Country did not benefit

Source: Eurobarometer. Interrupted lines correspond to missing data for the period.

Fig. 2. Portuguese Attitudes vis-à-vis the Consequences of EU Membership, 1986-2006

32

Table 4. Correlations between citizens’ EU orientations and their left-right self placements 1991 (EB 36.0) 1995 (EB 43.1) 1999 (EB 51.0)

Instrumental euroscepticism -.182*** N=792 -.113** N=730 .019 N=812

Political euroscepticism .033 N=792 -.012 N=730 .037 N=812

33

Table 5. Parameter estimates for multinomial logit regression of vote choice in Portugal, 1991, 1995 and 1999 (reference category: vote for the incumbent; estimates for intercept, gender, age, education and income are not presented in the table) Instrumental euroscepticism Political euroscepticism Left-right ideology Dissatisfaction with democracy N Nagelkerke r2

1991 .603† (.344) -.068 (.232) -1.559*** (.145) .400** (.149) 574 .663

Main opposition 1995 1999 .674** -.192 (.234) (.277) .314 -.308 (.278) (.192) -1.467*** 1.022*** (.129) (.113) .131 -.002 (.123) (.127) 543 442 .683 .563

1991 .923† (.542) .104 (.498) -2.903*** (.272) .946*** (.263) 574 .663

PCP 1995 .620† (.369) 1.077* (.456) -2.487*** (.215) .564* (.229) 543 .683

1999 .250 (.390) .081 (.314) -1.069*** (.157) .072 (.175) 442 .563

1991 .514 (.778) -.130 (.555) .182 (.221) .310 (.301) 574 .663

CDS-PP 1995 .557† (.294) 1.204** (.384) -.110 (.143) .512** (.191) 543 .683

1999 -.115 (.498) .237 (.400) 1.214*** (.178) .129 (.243) 442 .563

†

<.10; *p<.05; **p<.01; ***p<.001 Standard errors in parenthesis

34

Appendix A: Parties’ Net Positioning on Europe, 1976-2005 Election Year 1976 1979 1980 1983 1985 1987 1991 1995 1999 2002 2005

BE

PCP

PS

PSD

CDS

0.00 -2.11 -2.79

0 0 -0.36 -0.94 -0.81 -1.92 1.58 0.64 0.40 -2.50 -1.21

0 2.94 1.78 1.60 9.59 0 4.47 1.85 1.46 2.73 2.07

0 2.75 1.66 0.22 3.26 2.94 3.99 4.02 2.65 0.29 0.86

0.52 2.75 1.66 0 0 24.24 1.99 -1.50 1.05 1.05 1.97

Source: (Budge, et. al, 2001; Klingemann, et. al., 2006.

35

Appendix B: Party Names and Acronyms BE CDS-PP PCP PS PSD

Bloco de Esquerda (Left Block) Centro Democrático Social-Partido Popular, (Popular Party) Partido Comunista Português, (Portuguese Communist Party) Partido Socialista (Socialist Party) Partido Social-Democrata (Social Democratic Party)

36