A ‘necessary evil’ even during hard times? Public support for political parties in Portugal before and after the bailout (2008 and 2012)

Abstract

Notwithstanding the prolific research on the crisis of democracy since the 1980s, the attention that has been given to political parties has not been enough to fully understand the (increasing lack of) citizen support for them. Additionally, there is too little research on how the economic context can contribute to changes in support for political parties. Focusing on the Portuguese case, this article has three main goals: to assess citizen support for parties before and after the economic crisis; to explore the contribution of the crisis to changes in the explanatory models of support; and finally, to identify the consequences of support for parties, with regard to electoral turnout. Findings reveal the economic crisis has affected support for parties in Portugal, specifically regarding diffuse-institutional support and party legitimacy. Changes in public support for parties between 2008 and 2012 have had consequences on voter turnout.

Keywords

economic crisis political parties political trust Portugal public support for parties

Introduction: Democratic crisis and the declining support for parties

Since the third wave of democratization in the 1970s and 1980s, the democratic system in Europe has lived under a singular paradox. On the one hand, the expansion of democracy to various European countries and the triumph of the democratic form of government; on the other, the debate on the crisis of democracy since the 1980s, which has emphasized a growing distrust of major political institutions and actors by citizens and has widened the gap between those who govern and those who are governed (for a discussion and literature review see Brandenburg, 2011; Fuchs et al., 1995). There is a type of ‘fatigue’ or ‘malaise’ regarding the functioning of the democratic system and its main institutions and actors. However, this does not necessarily mean the legitimacy of democracy as a political system is in question because its autocratic alternatives are still being rejected (see especially Dalton, 1996; Norris, 1999; and on the Portuguese case Magalhães, 2005). The so-called fatigue or malaise has been explained by the emergence of a more ‘critical’ (Norris, 1999, 2011) and ‘sophisticated‘ (Dalton, 1996, 2004) citizen, who is younger, more educated and is guided by post-materialist values (Inglehart, 1977).

But democratic deficits also result from the increasing expectations and the cognitive awareness of ‘critical citizens’ who are dissatisfied with the supply side of democratic governance (e.g. process and policy performance, and power-sharing arrangements). There are a host of symptoms to corroborate this thesis: civic engagement is declining, conventional political participation is eroding and people are becoming unsatisfied with and sceptical about political institutions, parties and politicians (Pharr and Putnam, 2000; Schmitt and Holmberg, 1995; Teorell et al. 2007).

Explanations of this multifaceted phenomenon abound. In addition to social modernization, cognitive mobilization, individualization and the spread of post-materialist values, the most commonly cited culprits are: the erosion of class lines and centripetal party competition, the absence of meaningful party competition and tangible policy alternatives in the era of globalization and powerful supranational institutions, the impact of mass media—especially television—which has replaced more traditional agents of political socialization, the personalization of politics and power-sharing arrangements in some countries that blur the lines of accountability (Abramson and Inglehart, 1995; Dalton, 1996; Putnam, 2006). This list is indicative, and by no means exhaustive.

Until recently, economic depression has not been one of the explanations cited. Despite the double oil crisis that affected western countries in the 1970s and the economic ups and downs associated with normal business cycles, from a macroscopic point of view these trends and this debate have taken place during a period of rising standards of living throughout the western world. Since 2008, the entire world—and since 2010, countries of the southern European periphery—have entered an unprecedented period during which many of the gains associated with the post-war ‘economic miracle’ are being reversed.

In Portugal, the feelings of disaffection and mistrust towards political institutions, and particularly towards established political parties, have been accentuated since the outbreak of the economic crisis (Teixeira et al., 2014). A similar tendency can be seen in other South European countries affected by the crisis (Bosco and Verney, 2012). As a consequence, in these circumstances it seems reasonable to ask ‘What can explain the acceleration in the decline of confidence in political institutions in general and political parties in particular?’

The ‘so what’ question might seem pertinent at this point. Why should we care if positive feelings towards political parties are in decline? The most obvious answer is that parties are one of the most central institutions of contemporary representative democracy. As Schattschneider (1942: 1) famously put it, ‘modern democracy is unthinkable save in terms of political parties’. However, very little attention has been paid to the direct behavioural implications of this declining trust in political parties. Yet, we know that how people feel about the political system as a whole, or about its specific component parts, can have a significant effect in terms of how they engage (or not) with politics (Bélanger and Nadeau, 2005; Hetherington and Husser, 2012). Dalton and Weldon (2005) have shown that the impact of declining trust in political parties in particular is associated with lower levels of political engagement—specifically election turnout and campaign activity.

Using the analytical framework proposed by David Easton in respect of support for the political system as a whole, we intend to examine support for political parties in particular. Easton (1965) took the lead in proposing a qualitative distinction between types of support for the political system by focusing on the level of generality and abstraction. More specifically, he distinguishes between diffuse and specific support, which in their actual manifestations have a tendency to correspond to different objects of support. Diffuse support (or lack thereof) is directed to the fundamental aspects of the political system. Specific support is directed at the officeholders themselves (political authorities) and is closely related to matters concerning their performance and output.

Diffuse support tends, therefore, to be more durable and stable because it derives from deep-seated political values and attachments. Conversely, specific support is contingent on perceptions and evaluations of political system performance and its policy outputs; thus it tends to be in sync with the political ebb and flow and is prone to short-term fluctuations. Support for specific parties (government or opposition), as well as attitudes towards leadership elites and authorities, is expected to fluctuate over time in response to short- and medium-term contextual factors, such as the performance of particular governments, major shifts in public policies, leadership changes or merely a cyclical economic crisis.

In comparative studies, little attention has been paid to party support as such. Most studies rely on indirect indicators (Dalton and Wattenberg, 2000; Schmitt and Holmberg, 1995; Van Biezen et al., 2012; Webb et al., 2003)—such as party identification and party membership—to assess the legitimacy of parties as political institutions. Easton believed political parties to be a useful mid-range indicator of political system support (Easton, 1975; Martin, 2013), but one could argue that support for parties, like political support in general, can have generalized and specific manifestations. More specifically, support for political parties can be expressed as support for their institutional role in contemporary democracies in general, in what we label institutional support for parties. On the other hand, support for political parties can be based more on the evaluation of the performance of parties based on experience than on deep-seated political values and beliefs concerning the necessity of political parties as institutions in a contemporary democracy. This is what we call performance support for parties.

There have been a handful of studies, both comparative (Dalton and Weldon, 2005) and case studies (Martin, 2013), confirming the existence of this paradox for support for political parties in different countries. On the one hand, large majorities tend to view political parties as indispensable to democracy while, at the same time, confidence and support for existing and established political parties are shrinking progressively to ever-smaller minorities within the same societies.

As Dalton and Weldon (2005: 937) argue, citizens in contemporary democracies tend to view political parties as a ‘necessary evil’—essential for organizing interests and competing in elections—while harbouring doubts about how well they tend to represent their interests in reality. A convincing interpretation is offered by Piero Ignazi: parties have strengthened their ties to the state in order to counterbalance their eroding ties to civil society. This has created the double effect of having parties that are more powerful than ever in terms of resources and their ability to colonize the state, but at the same time less able to penetrate civil society and legitimize their role in the eyes of the citizens (Ignazi, 2014). This dynamic of parties has also meant recognizing the obsolescence of their more traditional functions combined with the priority given to their government action.

This article focuses on a specific case—Portugal—and its main objectives are to assess:

How Portuguese citizens have supported political parties both before and after the crisis (2008 and 2012).

To what extent the crisis has contributed to changes in the explicative models of citizen support for political parties.

The consequences of public support for political parties in terms of possible changes to electoral turnout.

The case of Portugal is of particular importance in this regard, since it is one of the countries suffering the brunt of the economic crisis while also being the one that initiated the third wave of democratization (Huntington, 1991). If the question of the long-term fate of democracy and political parties during times of growth and relative affluence seemed important, the same issue becomes more crucial in a context in which the forces that have long been affecting the quality of contemporary representative democracies are joined by conditions of rapidly declining standards of living.

Portuguese party system and political parties

Our analysis focuses on the five parties currently holding seats in the Portuguese parliament. The configuration of the Portuguese party system was established soon after the revolution of 25 April 1974. The centre-left Socialist Party (PS—Partido Socialista), which was founded in 1973, and the centre-right Social Democratic Party (PSD—Partido Social Democrata), which was founded in 1974, immediately became the country’s largest parties. They are commonly considered to be catch-all or cartel parties (Lisi, 2011; Lopes, 2004) and have alternated in government (either alone or in coalition) since 1987. With a few exceptions, the tendency has been for voting to be concentrated in these two centrist parties, which combined consistently receive around 70 percent of the votes.

To the right of the PSD is the Democratic and Social Centre-People’s Party (CDS-PP—Partido do Centro Democrático e Social-Partido Popular). Closest to the modern cadre type (Lopes, 2004), it represents mainly Christian-democratic values and conservative voters. Despite its small size in electoral terms (usually attracting less than 10 percent of the vote), the CDS-PP has managed to present itself as a party with the potential to form part of government coalitions and, as such, has been included in the government on a number of occasions.

The Portuguese Communist Party (PCP—Partido Comunista Português) is a Marxist-Leninist party founded in 1921. It is the only political formation that resisted the major hardships of the dictatorship period and had a concomitantly clandestine existence. The PCP is the only one that can properly be considered a mass party (Lisi, 2011; Lopes, 2004; Teixeira, 2009). Its electoral results have experienced a general decline, to slightly below eight percent in the 2011 legislative elections, although it did experience some recovery in the 2014 European elections (where it gained almost 13 percent of the vote).

Finally, the left-libertarian Left Bloc (BE—Bloco de Esquerda), which gained its electoral breakthrough in 1999. While its electoral performance was initially very low, around two percent, it has since grown, albeit never attaining 10 percent. In the 2011 and 2014 elections it consolidated its share of the vote at around five percent.

Portuguese parties in the context of the economic crisis

As in other southern European countries, the international effects of the 2008 economic crisis were strongly felt in legislative elections in Portugal (Bosco and Verney, 2012). In fact, the Portuguese 2009 election was marked by the polarization between the leaders of the two main parties (PS and PSD). The campaign was dominated by the effects of the international economic crisis and the question of appropriate policy responses (specifically the alternatives of a Keynesian demand-side policy versus fiscal discipline and economic orthodoxy). The incumbent PS won the election but lost a large number of seats in parliament and was unable to secure an absolute majority (36.5 percent). The Prime Minister, José Socrates, was obliged to form a minority government before a very hostile majority in Parliament, a quite radicalized opposition and a President hailing from the centre-right, at an extraordinarily difficult time for the country just as the global financial crisis was about to turn into a sovereign debt crisis in the European periphery. The abstention of the two main opposition centre-right parties (PSD and CDS-PP) secured the approval of the 2010 budget, which despite the commitment to a more expansive economic policy was the toughest austerity budget in nearly 30 years. Both President Cavaco Silva and the new leader of the PSD, Passos Coelho, supported reductions in public expenditure and the austerity policies adopted by the PS government. In March 2011, however, a fourth austerity package was announced by the minister of finance without the participation or knowledge of other parties or the president, a move that cost the government the support of PSD, leading to José Socrates tendering his resignation on 23 March 2011 and the calling of elections in June 2011.

A few days after his resignation, and following the continuous growth in Portugal’s borrowing costs (interest rates on 10-year sovereign bonds reached seven percent), Socrates announced Portugal had agreed to accept financial assistance from the European Union (EU) and International Monetary Fund (IMF) in the form of a 78 billion euro loan upon signing a Memorandum of Understanding (MoU)—which included immediate measures to substantially reduce the budget deficit (which stood at 9.8 percent in 2010). Formally, this agreement was only signed by the PS government, but in fact it had the agreement of both the PSD and the CDS-PP. The first austerity measures included progressive reductions in the salaries of public officials, cuts to public sector pensions and general tax increases. These measures were severely criticized by the opposition parties and trade union organizations, as well as by many within the PS itself, which was a signatory to the MoU.

In the June 2011 elections, the so-called parties of the ‘area of governability’ obtained the following results: PS—28 percent of the vote (78 deputies); opposition PSD 38.6 percent and CDS-PP 11.7 percent. As a result, the PSD and CDS-PP, with 132 of the 230 parliamentary deputies, formed a parliamentary majority, creating the conditions to ensure compliance with the commitments to the country’s creditors.

Levels of confidence in Portuguese parties in a comparative perspective

Before beginning this analysis of the Portuguese case, it is necessary to have a general picture of Europe and a clear understanding of Portugal’s position. In terms of national political institutions, parties enjoy the least confidence among European citizens. Looking at the levels of public confidence in parties throughout Europe (Table 1) in the period 2000–2012, Portugal occupies 13th position, with 18 percent of respondents stating they trust political parties, below the average for all countries which is in itself a rather low 20 percent. It is lower than Spain with 24 percent, but above Greece with 17 percent and Italy with 15 percent, and also Germany, the UK and France. Results for Portugal are closer to those observed in the new democracies of eastern Europe, rather than the old democracies of northern and western Europe which still present relatively high values for citizens’ confidence in political parties.

Table 1.

Evolution of confidence in political parties (%), 2000–2012.

	2000	2002	2004	2006	2008	2010	2012	Net change (2000–12)	Year	Sig.
Austria	21	25	19	38	30	40	33	12	1.308***	0.007
Belgium	21	22	20	29	28	20	23	2	0.341	0.269
Bulgaria				10	9	15	14	3	0.821**	0.047
Cyprus				20	33	23	19	–2	–0.512	0.589
Czech Republic				15	11	12	8	–3	–0.524	0.195
Denmark	32	35	32	49	50	49	35	3	0.863	0.192
Estonia				21	15	20	19	2	0.048	0.912
Finland	20	21	21	36	31	30	36	16	1.203***	0.005
France	15	13	13	10	10	14	12	–3	–0.291*	0.091
Germany	17	17	11	22	18	19	21	4	0.313	0.203
Greece	19	16	28	25	14	9	5	–14	–1.363***	0.007
Hungary				29	9	29	18	6	0.429	0.771
Ireland	22	24	23	32	27	17	12	–10	–0.879**	0.030
Italy	13	15	13	26	13	18	8	–5	–0.159	0.666
Latvia				6	5	4	6	–4	–0.321	0.314
Lithuania				10	7	6	13	3	0.107	0.785
Luxemburg	41	32	31	41	29	40	22	–19	–0.275	0.657
Malta				25	35	25	20	–9	–1.417	0.123
Netherlands	39	35	27	42	40	45	33	–6	–0.269	0.477
Poland				9	7	15	17	12	1.845***	0.001
Portugal	18	21	16	19	19	15	17	–1	–0.423**	0.015
Romania				14	18	10	9	–13	–1.548**	0.020
Slovakia				10	11	25	19	8	1.500**	0.040
Slovenia				19	13	11	16	–7	–1.393**	0.066
Spain	28	26	27	31	40	14	6	–22	–1.104**	0.098
Sweden	16	23	21	23	28	37	35	19	1.511***	0.000
United Kingdom	14	15	10	18	13	18	12	–2	–0.159	0.542

Source: Eurobarometer series: 2000-2012.

The relationship between parties and civil society has always been marked, for reasons connected with the genesis of the democratic system in Portugal, by a progressive distancing of the parties from the voters and, at the same time, the growing approximation of the parties to the state. For example, using data from the late-1990s to 2009, Van Biezen and colleagues (2012) showed that party membership in Greece and Portugal has fallen. However, most of the remaining European countries have experienced larger declines in this period, showing that this is not a phenomenon that is exclusive to either Portugal or Greece. This distance seems to have been accentuated by the economic crisis. In terms of Portuguese attitudes towards parties, the already very low levels of confidence fell even further, as shown in Table 1, and party identification plunged from 50 percent in 2008 to 36 percent in 2012.

It is interesting to note the evolution in the levels of confidence during the period considered, noting the statistically significant increase, between 2002 and 2012, of public confidence in political parties in countries in which it tends to be greater, as in the case of Sweden (+19 percent), Finland (+16 percent) and Austria (+12 percent). At the same time, there has been a statistically significant fall in countries where confidence levels are normally very low, and which were also those most affected by the sovereign debt crisis and subject to international intervention between 2008 and 2012. For example, in Spain, where the level of public confidence in the parties fell from 22 percent in 2002 to six percent in 2012; Greece, where it fell from 19 percent in 2002 to a mere five percent in 2012; and Ireland, where confidence in parties stood at 22 percent in 2002 and fell by more than half by 2012.

Portugal also presents a statistically significant decrease, but one that is much less pronounced, as the level of citizen confidence in parties was already extremely low in 2002, at 18 percent, falling by just one point to 17 percent by 2012. The lowest levels of confidence in parties exist in the new democracies of southern and eastern Europe—a trend that is related to factors associated with authoritarian or totalitarian regime legacy and the late advent of democracy contributing to established parties having more institutional than societal origins, and which consequently have much shallower roots in civil society (Van Biezen, 2003).

Hypotheses and explanatory models of popular support for parties in Portugal before and after the bailout

Our main goal is to discover whether the financial and economic crisis contributed to a change in the explanatory models of popular support for parties in Portugal. To discern the individual determinants of the two types of support (institutional and performance) we employ four blocs of independent variables. The first includes the usual socio-demographic variables. Age and education are expected to be particularly important. Younger people, due to generally higher levels of political alienation and mistrust, are expected to show lower levels of support for parties, especially of diffuse performance support, when compared with older people. Older people lived under two different regimes (the dictatorship for almost 50 years and a western type of representative democracy since 1974) and are therefore expected to value more the democratic institutions in general, including political parties as institutions, and their performance (Morlino and Montero, 1995). This leads to the following hypothesis:

H1– Decreasing age results in lower party support, especially performance support.

Education is expected to have as well a positive relationship with institutional support, as individuals with lower levels of education tend to feel more alienated from the political process and its institutions. Moreover, higher levels of education are expected to be associated with higher expectations for democratic performance. Consequently, it may be expected that:

H2 – Higher levels of education result in lower performance support and higher institutional support.

The second bloc includes respondents’ attitudes and values in respect of politics, namely: their interest in political affairs; their cognitive mobilization, measured here by their level of exposure to political information; their degree of identification with a political party; their ideological self-placement on the left-right scale; and, finally, their adherence to post-materialist values. The hypothesis can be formulated as follows:

H3 – Increasing interest in politics, higher levels of cognitive mobilization, post-material values and more left-wing political orientation result in higher institutional support and lower performance support.

The third bloc includes some variables related to how respondents assess the democratic performance of the Portuguese form of government: that is, in terms of institutional and procedural criteria. Citizens, who expect the political regime to meet certain democratic standards—such as a fair electoral system that gives equal opportunities to all political parties—should have little reason to be satisfied with the major parties if the regime is perceived to be failing to meet the benchmarks. Consequently:

H4 – Institutional support and performance support for parties tend to be greater among those who are party identifiers and are satisfied with the electoral system adopted in the country: that is, the proportional electoral system according to the formula of the highest D’Hondt average.

The fourth bloc gathers the variables relating to policy performance: the individual-level retrospective evaluations citizens make regarding the country’s economy and the government’s performance. Following Easton’s framework, this last bloc is expected to be more associated with performance support (closer to specific support and, therefore, more prone to political and economic fluctuations) than with institutional support for political parties.

H5 – A more negative evaluation of the state of the economy and government performance does not necessarily result in decreasing institutional support but results in lower performance support.

In terms of the impact of the crisis on our explanatory models, we expect that short-term evaluations will acquire more explanatory weight in both institutional and, especially, performance support.

H6 – Evaluations of government performance and of the state of the economy will assume more explanatory weight in our post-crisis explanatory model in relation to our pre-crisis model.

In the following section we present the data employed in our study before proceeding to the findings of our analyses.

Data

In order to understand how Portuguese attitudes towards parties evolved before and after the bailout we will use four standard questions included in two national surveys, one conducted in 2008, the other in 2012. Both surveys used multi-stage probability samples of citizens aged 18 years or older and living on the mainland (2008, N = 1350; 2012, N = 1209). The two national surveys have a corresponding questionnaire covering basically the same topics with respect to attitudes towards Portuguese parties (Freire et al., 2009). The four questions included in both are as follows:

To what extent do you strongly disagree, disagree, agree or strongly agree with the following statements:

Without parties there can be no democracy;

Political parties are indispensable for political participation;

Political parties only serve to divide people;

Political parties criticize each other a lot, but in reality they are all the same.

Items a) and b) correspond to views on the indispensability of political parties in contemporary representative democracies. In other words, such questions gauge the importance people attribute to the political party as an institution in the abstract and the extent to which citizens consider democracy and political parties as inextricably linked. For this reason we chose to employ these two items to capture plain diffuse support for political parties.

Items c) and d) tap into views and evaluations of the actual performance of the parties. It can be argued these two items implicitly include an experiential component, making them better suited to the measurement of evaluations of the contemporary role and behaviour of political parties. They are more contingent on individual’s perceptions and evaluations, and more prone to short-term fluctuations than the former two. For these reasons we believe they are closer to the concept of performance support than the first two. We know, however, that the plain specific level of party support concerns more concrete issues, such as confidence in party leaders or trust in political parties. Our goal is to measure popular support for political parties in Portugal using a continuum in party support, tapping the most abstract level, or plain diffuse support, which we call diffuse-institutional support, or, for analysis simplicity, institutional support; and other more specific levels of support based on party performance and its consequences, which while not fully matching Easton’s specific support can be considered very close to it. We decided to call this diffuse-performance or performance support.

Examining popular support for political parties in Portugal

When we look at Figure 1, we see that most Portuguese, more so in 2008 than in 2012, agree that without political parties there can be no democracy (84.4 percent in 2008 and 66.6 percent in 2012) and that political parties are indispensable for citizen participation in the country’s political life (87.2 percent in 2008 and 75.1 percent in 2012). These results show there is an attitude of normative approval of political parties as institutions. However, when looking at citizen attitudes towards the actual activities of established parties in Portugal, a different picture emerges.

Figure 1.

Support for parties in Portugal, 2008 and 2012. Notes: 2008 N = 1350; 2012 N = 1197–1127.

When asked if political parties only serve to divide people, since one of their main functions is to present differentiated alternative policies, respondents mostly tend to agree with this statement, both before and after the bailout: in 2008, about 62 percent of respondents agreed, rising to 72 percent in 2012. The explanation for this increase can be found in the fact that, notwithstanding the degree of contention between the two biggest parties with regard to austerity policies, both the PS and PSD signed the MoU with the Troika, while from a very early stage the PS sought to disassociate itself (at least rhetorically) from the implications of the financial adjustment programme. Equally high is the percentage of Portuguese who agree with the statement ‘the parties criticize each other a lot, but in reality they are all the same’ (85 percent in 2008 and 85.7 percent in 2012). Portuguese voters realize this contention is more artificial than real and that, if both the parties in government (PSD-CDS-PP) and the largest opposition party (PS) were already quite indistinguishable in terms of government practice, the bailout only accentuated this and did nothing to reverse it, leaving both with much too narrow a margin for political manoeuvring.

The percentage differences observed between 2008 and 2012 show that the mean value of agreement with the statements ‘political parties are indispensable for political participation’ and ‘without parties there can be no democracy’ falls from 2008 to 2012, and the differences are statistically significant, indicating a decline in the ‘legitimacy’ of political parties during the period of the economic crisis. The mean value of agreement with the statements ‘political parties criticize each other a lot, but in reality they are all the same’ and ‘political parties only serve to divide people’ decreases slightly from 2008 to 2012 (3.3 to 3.2 and 2.83 to 2.82, respectively); however, the differences are not statistically significant.

In short, examining all Portuguese attitudes towards parties seems to end in a sort of paradox: they have a positive image of the parties, more so in 2008 than in 2012, because they recognize they are necessary for the functioning of democracy; but they also have an extremely negative image of political parties in respect of their representative function in defending the interests of different segments of society and providing meaningful political alternatives.

How then does one interpret this apparent paradox? As in other countries, Portuguese political parties are still seen to be a kind of ‘necessary evil’ (Dalton and Weldon, 2005). Applying Easton’s framework, these preliminary findings suggest that support for political parties (like support for the democratic form of government) has indeed a multidimensional nature. It seems to emerge in two fundamental dimensions that are not necessarily related to one another: on the one hand, a dimension that recognizes the democratic legitimacy of parties as institutions essential to the functioning of democratic government (institutional support for parties); and on the other, a dimension expressed in the negative or positive evaluation of the actual performance of existing political parties, which tends to be determined less by core normative values and more by concrete experience and performance-oriented evaluative factors (performance support for parties).

The multidimensionality of the concept of public support for parties

To confirm whether the two types of party support are conceptually and empirically distinct, we first proceeded to an exploratory factor analysis. Table 2 presents this analysis, showing that citizens clearly distinguish two dimensions of support for political parties, whether in 2008 or in 2012, as suggested by the previous findings. Two clearly independent factors emerge with different factorial weights and with explained variances ranging between 34.4 and 35.5 percent, which are subsequently the dependent variables in our multivariate analyses.

Table 2.

Exploratory factor analysis: Support for political parties in Portugal, 2008 and 2012.

	2008		2012
	First factor	Second factor	First factor	Second factor
Political parties are indispensable for political participation	0.003	0.821	0.848	0.004
Without parties there can be no democracy	0.003	0.832	0.834	–0.100
Political parties criticize each other a lot, but in reality they are all the same (inverted scale)	0.838	0.077	–0.024	0.842
Political parties only serve to divide people (inverted scale)	0.839	–0.072	–0.070	0.836
Variance explained %	35.2	34.4	35.5	35.5

Notes: 2008 N = 1350; 2012 N = 1209. Rotated component matrix. Extraction Method: Principal Component Analysis. Rotation Method: Varimax. 2008: KMO = 0.484; 2012: KMO = 0.520.

This multidimensional composition of the concept of support for political parties is also supported by a confirmatory factor analysis we developed for both years, in Figures 2 and 3, as well as in the measurements related to the adjustment of the models revealing a good performance (for 2008: CFI > 0.90, NFI > 0.80, and RMSEA p < 0.08; for 2012: CFI > 0.90, NFI > 0.80 and the RMSEA is close to the reference value p < 0.08).

Figure 2.

Confirmatory factor analysis (standardized estimates): Support for political parties in Portugal, 2008.

Figure 3.

Confirmatory factor analysis (standardized estimates): Support for political parties in Portugal, 2012.

Findings from multivariate explanatory models

To determine the individual determinants of the two types of support (institutional and performance support), four regression models were developed, which include the same independent variables in both years. The dependent variables, as mentioned before, correspond to the factor scores from each of the two factors formed in each year and are used as measures of institutional and performance support respectively.

The multivariate models in Table 3 show that overall there are a lot of similarities when it comes to the factors that explain both institutional and performance support, thus partially refuting our hypotheses, at least when it comes to the expectation of the inverse impact of certain factors on each type of support. The model for institutional support in 2008 is the weakest of all the models (only explaining 3.8 percent of the dependent variable’s variance), and the model for performance support in the same year is the strongest (even so it is weak, explaining 14.3 percent of the variance). In 2012 both models explain around 9 percent of the variance. This globally means that the specification of the models does not perform better after the crisis has emerged, especially regarding performance support, even considering variables related to the economic crisis such as government and economic performance.

Table 3.

Explanatory factors of the two types of support for political parties in 2008 and 2012.

	Diffuse–Institutional 2008			Diffuse–Performance 2008			Diffuse–Institutional 2012			Diffuse– Performance 2012
		B	p		B	p		B	p		B	p
(Constant)		–1.457 (.484)	0.003		–2.846 (.449)	0.000		–1.032 (.310)	0.001		–0.907 (.330)	0.006
Bloc 1
– Sex (1 = male)					–0.150 (.088)	0.087
– Age					0.188 (0.039)	0.002		0.095 (0.036)	0.008		0.081 (0.038)	0.035
– Education					0.111 (0.029)	0.000		–0.095 (.024)	0.000		0.102 (0.026)	0.000
– Residence (1 = urban)								0.180 (0.077)	0.020
Adjusted R²	0.000			0.060			0.066			0.043
Bloc 2
– Interest in politics					0.108 (0.060)	0.074					0.219 (0.052)	0.000
– Cognitive mobilization					0.028 (0.015)	0.059
– Left–Right self-placement		–0.040 (.023)	0.084		–0.045 (.022)	0.037
– Post-materialism		0.241 (0.063)	0.000		0.232 (0.058)	0.000		0.134 (0.047)	0.005		–0.128 (0.050)	0.011
– Identification with party (1= yes)								0.169 (0.080)	0.035		0.148 (0.085)	0.081
Change in Adjusted R²	0.018			0.053			0.014			0.047
Bloc 3
– Portuguese electoral system better than first–past–the–post systems					0.154 (0.054)	0.004		0.180 (0.087)	0.039
Change in Adjusted R²	0.000			0.011			0.004			0.000
Bloc 4
– Positive evaluation of government performance		0.195 (0.079)	0.014
– Positive evaluation of state of economy in past 12 months					0.140 (0.058)	0.016		0.134 (0.061)	0.029
Change in Adjusted R²	0.028			0.019			0.005			0.000
Total Adjusted R²	0.038			0.143			0.089			0.090
N	463 463						603

Notes: Standard errors in parentheses. Only statistically significant coefficients (p < 0.10) are presented.

Regarding socio-demographic variables, coefficients are statistically significant in almost all cases for age, in the expected direction: the older are globally more supportive of political parties. The coefficient is stronger for performance support in 2008, supporting the expectation of H1 of a stronger relationship between age and performance support. In 2012 the two coefficients are similar in magnitude.

Education plays opposite to our expectations. In both years the levels of education of respondents have a significant role in the explanation of support for the performance of political parties; the relationship is opposite regarding institutional support in 2012. Contrary to what we would expect, the more educated seem to be more critical of parties as institutions after the crisis, simultaneously keeping an understanding view of their performance.

Generally, in 2008 a positive correlation of left-wing ideology and post-material values with institutional support is observed as expected, as well as with performance support against expectations. Performance support in 2008 is also associated with cognitive mobilization. Paradoxically, we had hypothesized that those factors would contribute to institutional rather than performance support. Conforming to expectations, evidence for 2012 shows that respondents who hold post-material values tend to offer more institutional support and less performance support.

Preference for proportional electoral systems—reflecting a satisfaction with the Portuguese electoral system—is positively correlated with performance support in 2008 and with institutional support in 2012, as well as party identification for both kinds of support in 2012, expectedly reflecting respondents’ identification with the political institutions and institutional arrangements.

Additionally, institutional support appears to be positively correlated with a positive evaluation of government performance in 2008, and with positive evaluation of the economy in 2012 (as for 2008 regarding performance). This suggests that evaluations on the economy and government functioning are relevant for parties’ support, but not in a systematic or standardized manner (as generally seen in the preceding relevant variables).

To sum up, the older and better educated generally tend to offer more support (especially performance) for parties than the younger and less educated, supporting H1, but not H2. Against our expectation, interest in politics is correlated with performance support and not with institutional support, as cognitive mobilization in 2008. Left-wingers and post-materialists explain both kinds of support especially in 2008, but only post-materialist values support H3. Party identification quite robustly explains support for parties in 2012, while the support for the proportional electoral system does it in different types of support across years. The support for H4 is therefore partial. Moreover, short-term evaluative factors appear to especially explain variation in institutional, and not performance, support—again contrary to what we would expect (H5). Moreover, after the crisis has hatched, only positive evaluation of the state of the economy plays a role in explaining institutional support (H6).

The general conclusions that we can withdraw are, in short, firstly that although two different dimensions are identifiable regarding party support, its explaining models do not show significantly different patterns of explanation, but rather disperse and non-standardized relationships across years and types of support. Secondly, the models after the crisis generally show that while other variables play a role in explaining parties’ support, party identification emerges has a consistent predictor, suggesting that it is the level of closeness to parties that leads citizens to better understand and accept the institutional role and performance of parties in times of crisis.

Behavioural correlates of public support for parties

But what are the consequences of support for parties in behavioural terms? Is decreased support for political parties, particularly with respect to institutional support, translated into abstention from politics, at least in its conventional form? Bélanger and Nadeau (2005) demonstrated that declining political trust had an effect on party support (for third-parties alternatives) and on electoral participation.

Support for political parties, both institutional and performance, appears to be unrelated to electoral turnout in Portugal before the economic crisis in 2008 (see Table 4). However, in 2012 (see Table 5), in the wake of the financial and economic crisis, support for parties becomes a significant predictor: high levels of either institutional or performance support increase the probability of basting a ballot at the national election. Nevertheless, socio-demographic variables appear to be much better predictors of electoral participation, especially age and education which correlate highly with voting in both years. The better educated an individual is, and the older they are, the greater the probability they will vote.

Table 4.

Individual determinants of electoral turnout in 2008.

Independent variables	Voted in the last elections = 1
Independent variables	B	Wald	p	Exp(B)
Constant	1.792	7.743	0.005	6.000
Sex (1 – Female)	0.234	2.848	0.091	1.263
Age		80.638	0.000
(1 – 18–24 years)	–2.014	51.494	0.000	0.134
(2 – 25–34)	–1.137	12.901	0.000	0.321
(3 – 35–44)	1.677	7.288	0.007	0.508
(4 – 45–54)	1.048	6.036	0.015	0.953
(5 – 55–64)	–0.124	0.243	0.622	0.883
Education		19.335	0.007
(None = 1)	–1.780	14.119	0.000	0.169
(2 – Primary school incomplete)	–1.147	7.594	0.006	0.318
(3 – 4 full years)	–0.964	6.924	0.009	0.381
(4 – 6 full years)	–0.670	3.124	0.077	0.512
(5 – 9 full years)	2.136	9.903	0.002	0.321
(6 – 12 full years)	1.913	6.534	0.015	0.401
(7 – Bachelor complete)	1.445	6.299	0.012	0.236
Diffuse-institutional support for parties	0.119	3.709	0.154	0.445
Diffuse-performance support for parties	–0.039	0.130	0.719	0.962
R² Nagelkerke global	0.153
N	1239

Source: Freire and Viegas (2008).

Table 5.

Individual determinants of electoral turnout in 2012.

Independent variables	Voted in the last elections = 1
Independent variables	B	Wald	p	Exp(B)
Constant	3.305	26.740	0.000	27.246
Sex (1 – Female)	–0.044	0.088	0.767	0.957
Age		120.024	0.000
(1 – 18–24 years)	–3.360	83.318	0.000	0.035
(2 – 25–34)	–2.604	66.607	0.000	0.074
(3 – 35–44)	1.808	34.435	0.000	0.164
(4 – 45–54)	1.684	7.210	0.007	0.505
(5 – 55–64)		49.941	0.000
Education	0.157	0.108	0.743	1.169
(None = 1)	–1.561	8.216	0.004	0.210
(2 – Primary school incomplete)	–1.468	24.388	0.000	0.230
(3 – 4 Full years)	0.826	7.566	0.006	0.438
(4 – 6 Full years)	0.815	8.880	0.003	0.443
(5 – 9 Full years)	0.194	0.479	0.489	0.824
(6 – 12 years complete)	1.107	2.830	0.093	3.027
(7 – Bachelor complete)		14.024	0.007
Diffuse-institutional support for parties	0.219	3.709	0.054	1.245
Diffuse-performance support for parties	0.239	4.130	0.033	1.962
R² Nagelkerke global	0.259
N	1174

Source: Freire, Viegas and Lisi (2012).

Concluding remarks

Inspired by Easton’s typology, this article examines the question of the support for political parties in Portugal both before and after the economic crisis in two dimensions: institutional support and performance support for political parties. Findings reveal some familiar patterns in terms of the contrast between high levels of institutional support and low levels of performance support. More surprisingly, and alarmingly, the economic and financial crisis and harsh austerity policies attached to the bailout deal appear to have affected institutional rather than performance support. Even though large majorities continue to consider political parties central and necessary to the democratic process, these majorities have declined between 2008 and 2012. Political parties continue to be considered a ‘necessary evil’, perhaps with a declining emphasis on the ‘necessary’ part. Possibly even more alarmingly, institutional support for political parties after the crisis appears to be more strongly associated with older and less educated segments of Portuguese society, a fact that does not bode well for the future of generalized support for parties as institutions in Portugal. Based on this evidence, it is not clear whether this trend can be attributed to long-term value change or to a backlash produced by the effects of the crisis among the educated young. However, at the same time, the more educated and dynamic strata of society seem more likely to exhibit higher performance support for political parties. This is probably an indication that institutional and performance support are not completely independent from each other and certainly do not run in opposite directions as hypothesized. As stated above, both types of political support can be conceptualized as types of diffuse support, even if—when compared to institutional support for parties—performance support is clearly closer to what Easton defined as specific support.

Moreover, the behavioural correlates of public support for parties can be seen in 2012, both in respect of electoral participation and voting behaviour for the incumbent parties. A plausible explanation for this is that Portugal has one of the most fixed party systems in Europe, where government is taken up alternately between the PS, PSD or the PSD and CDS-PP in coalition—what some left-wing politicians call ‘rotativism without alternatives’.¹ Simply consider the fact that, among the countries of southern Europe that most felt the effects of the sovereign debt crisis within the Eurozone, Portugal is the one that presented the lowest electoral volatility between the elections preceding the crisis and those that succeeded it: approximately 13.3 percent compared with Greece’s 60 percent and Italy’s 41.3 percent. Even though there are no immediately observable behavioural components to the support for political parties, one could posit that more favourable conditions are being created in Portugal for the emergence of personalist, ‘anti-party’ parties, such as those that have emerged in Italy, Greece and several eastern European countries.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Note

References

Abramson

Inglehart

(1995) Value Change in Global Perspective. Ann Arbor, MI: University of Michigan Press.

BE (Bloco de Esquerda) (2014) Recuperar a Confiança: Contribuição para Moção à IX Convenção do Bloco. 11 July. Available at: http://www.bloco.org/content/view/2971/46/ (accessed 5 October 2014).

Bélanger

Nadeau

(2005) Political trust and the vote in multiparty elections: The Canadian case. European Journal of Political Research 44: 121–146.

Bosco

Verney

(2012) Electoral epidemic: The political cost of economic crisis in southern Europe, 2010–2011. South European Society and Politics 17(2): 129–154.

Brandenburg

(2011) What gives politics such a bad name? Irish Political Studies 26(4): 501–511.

Dalton

(1996) Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies. 2nd ed. New York, NY: Chatham House Publishers.

Dalton

(2004) Democratic Challenges, Democratic Choices. Oxford: Oxford University Press.

Dalton

Wattenberg

(eds) (2000) Parties without Partisans: Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press.

Dalton

Weldon

(2005) Public images of political parties: A necessary evil? West European Politics 28(5): 931–951.

10.

Demo.cratica (2011) XI Legislatura 2.^a Sessão Legislativa (2010–2011): Reunião Plenária de 2 de Fevereiro de 2011. 2 February. Available at: http://demo.cratica.org/sessoes/2011/2/2/ (accessed 5 October 2014).

11.

Easton

(1965) A Systems Analysis of Political Life. New York, NY: Wiley.

12.

Easton

(1975) A reassessment of the concept of political support. British Journal of Political Science 5: 435–457.

13.

Freire

Viegas

(2008) Portuguese Citizens Survey 2008. Research project at ISCTE-IUL, University Institute of Lisbon, and CIES-IUL, Portuguese deputies in comparative perspective: Elections, leadership and political representation (PTDC/CPO/64469/2006). Available online at: http://er.cies.iscte-iul.pt/

14.

Freire

Viegas

Lisi

(2012) Portuguese Citizens Survey 2012, Research project at ISCTE-IUL, University Institute of Lisbon, and CIES-IUL, Elections, Leadership and Accountability: Political Representation in Portugal in a Longitudinal and Comparative Perspective (FCT: ‘PTDC/CPJ-CPO/119307/2010’). Available at: http://er.cies.iscte-iul.pt/

15.

Freire

Leite Viegas

Seiceira

(eds) (2009) Representação Política em Portugal: Inquéritos e Bases de Dados. Lisbon: Sextante.

16.

Fuchs

Guidorossi

Svensson

(1995) Support for the democratic system. In: Klingemann

H-D

Fuchs

(eds) Citizens and the State. New York, NY: Oxford University Press, pp. 323–353.

17.

Hetherington

Husser

(2012) How trust matters: The changing political relevance of political trust. American Journal of Political Science 56(2): 312–325.

18.

Huntington

(1991) The Third Wave: Democratization in the Late Twentieth Century. Norman, OK: University of Oklahoma Press.

19.

Ignazi

(2014) Power and the (il)legitimacy of political parties: An unavoidable paradox of contemporary democracy? Party Politics 20(2): 160–169.

20.

Inglehart

(1977) The Silent Revolution. Princeton, NJ: Princeton University Press.

21.

Lisi

(2011) Os Partidos Políticos em Portugal: Continuidade e Transformação. Coimbra: Almedina.

22.

Lopes

(2004) Os Partidos Políticos: Modelos e Realidades na Europa Ocidental e em Portugal. Lisbon: Celta.

23.

Magalhães

(2005) Disaffected democrats: Political attitudes and political action in Portugal. West European Politics 28(5): 973–991.

24.

Martin

(2013) The party is not over: Explaining attitudes towards political parties in Australia. International Journal of Public Opinion Research 25(3): 1–17.

25.

Martins

(2012) Jerónimo de Sousa exige demissão do Governo e eleições antecipadas. Jornal de Notícias, 30 November. Available at: http://www.jn.pt/PaginaInicial/Politica/Interior.aspx?content_id=2919217 (accessed 5 October 2014).

26.

Morlino

Montero

(1995) Legitimacy and democracy in Southern Europe. In: Gunther

Nikiforos Diamandouros

Puhle

H-J

(eds) The Politics of Democratic Consolidation: Southern Europe in Comparative Perspective. Baltimore, MD and London: John Hopkins University Press, pp. 231–260.

27.

Norris

(ed.) (1999) Critical Citizens: Global Support for Democratic Governance. New York, NY: Oxford University Press.

28.

Norris

(2011) Democratic Deficit: Critical Citizens Revisited. Cambridge: Cambridge University Press.

29.

Pharr

Putnam

(2000) Disaffected Democracies. Princeton, NJ: Princeton University Press.

30.

Putnam

(ed.) (2006) The Dynamics of Social Capital. Oxford: Oxford University Press.

31.

Schattschneider

(1942) Party Government. New York, NY: Rinehart.

32.

Schmitt

Holmberg

(1995) Political parties in decline? In: Klingemann

H-D

Fuchs

(eds) Citizens and the State. New York, NY: Oxford University Press, pp. 95–133.

33.

Teixeira

(2009) O Povo Semi-Soberano? Partidos Políticos e Recrutamento Parlamentar em Portugal. Coimbra: Almedina.

34.

Teixeira

Tsatsanis

Belchior

(2014) Support for Democracy in Times of Crisis: Diffuse and Specific Regime Support in Portugal and Greece. South European Society & Politics. Epub ahead of print 7 November 2014. DOI: 10.1080/13608746.2014.975770.

35.

Teorell

Torcal

Montero

(2007) Political participation: Mapping the Terrain. In: van Deth

Montero

Westholm

(eds) Citizenship and Involvement in European Democracies: A Comparative Analysis. New York, NY: Routledge, pp. 334–357.

36.

Van Biezen

(2003) Political Parties in New Democracies. London: Macmillan.

37.

Van Biezen

Mair

Poguntke

(2012) Going, going…gone? The decline of party membership in contemporary Europe. European Journal of Political Science 51(1): 24–56.

38.

Webb

Farrell

Holliday

(eds) (2003) Political Parties in Advanced Industrial Democracies. Oxford: Oxford University Press.