Advertisement

Acta Politica

, Volume 53, Issue 2, pp 283–304 | Cite as

Conventional and unconventional political participation in times of financial crisis in the Netherlands, 2002–2012

  • Rik Linssen
  • Peer Scheepers
  • Manfred te Grotenhuis
  • Hans Schmeets
Article
  • 119 Downloads

Abstract

In this contribution, we investigate the extent to which the recent financial crisis has affected levels of political participation in general and more particularly within privileged and underprivileged societal groups in the Netherlands. We derive competing and complementary theoretical propositions about the possible effects of the economic downturn on conventional and unconventional modes of political participation. Economic decline might mobilize people to voice their concerns in the political arena, especially via unconventional modes of political participation such as demonstrating. As privileged societal groups are more likely to participate in politics, economic decline may widen the initial differences between privileged and underprivileged societal groups in their level of political participation. We use the Dutch Parliamentary Election Studies collected before (2002–2006), at the onset of the Eurocrisis (2006–2010) and after prolonged periods of recession (2008–2012) to empirically assess these competing claims. Our results show a slight decrease in conventional modes of political participation and a slight increase in unconventional modes of political participation during the recent financial and economic crisis. We do not find that the relationship between the economic crisis and political participation changes significantly differently for privileged and underprivileged groups in the Netherlands.

Keywords

Political participation Economic crisis Protesting The Netherlands 

Introduction and research question

After the financial crisis hit Europe in 2008, various protest movements took the stage in the media, and the political and societal realm. Europe (and beyond) witnessed the uprising of a wide variety of protest movements, for instance protests against austerity measures in Greece, the ‘15 M—movement’ in Spain and the global ‘occupy movement’. These protests are indications that hard economic times might induce people to participate in (protest) politics (Muñoz et al. 2013; Ponticelli and Voth 2011).

Beyond the quintessential literature on economic voting, it dealt with the question whether economic conditions influence party choice and turnout (Anderson 2007; Lewis-Beck and Stegmaier 2000; Radcliff 1992); surprisingly, little attention has been paid to the link between macro-economic conditions and modes of political involvement other than voting such as demonstrating and petitioning. In this contribution, we transpose propositions that stem from the economic voting literature and analyse modes of political participation: political actions conducted within a five-year period.

The Netherlands, a country with traditionally high levels of political and community participation (Gesthuizen et al. 2013; Linssen and Schmeets 2010) makes an interesting case in this respect. The once stable and pillarized Dutch political landscape is characterized by increased political polarization (Aarts et al. 2007) in more recent times. The Netherlands saw no less than 5 general elections between 2002 and 2012. The three most recent Dutch Parliamentary Elections were held in 2006, 2010, and 2012 and coincided with the rise of the global financial and economic crisis. In 2006, the election came before the global financial crisis. In 2010, the parliamentary elections coincided with the onset of the Eurocrisis and the global economic crisis. Finally, in 2012 after prolonged periods of recession in the Netherlands, the administration lead by Mark Rutte collapsed while negotiating on harsh austerity measures in response to the economic decline. The timing of these elections and thereby the timing of the election studies, provides a unique ‘natural experiment’ to explore the effect of economic downturn on levels of political participation, both for privileged and deprived strata in society.

We attempt to explore the link between macro-economic conditions in recent years in the Netherlands and levels of political participation. Did the recent financial and economic crises induce political participation or did citizens refrain from participating in politics during economic hardship? We set out to assess competing and complementary theoretical propositions concerning the effect of economic conditions on political participation in the Netherlands. Using the Dutch Parliamentary Election Studies, we aim to assess to what extent economic conditions affected individual-level political participation in the 2002–2012 period among the general population as well as among subpopulations with higher and lower resources. Therefore, our research question reads: To what extent are levels of political participation affected in general, and among privileged and underprivileged societal groups during times of economic downturn in the Netherlands between 2002 and 2012?

Theory and hypotheses

Political participation

Political participation is broadly defined as those activities aimed at influencing the political decision-making process. To take into account a wide range of political actions, we use conventional and unconventional modes of political participation. Conventional political participation refers to all modes of participation directly embedded in legal institutional frameworks, or directly referring to the electoral process and representational system, such as voting, contacting politicians or attending hearings (Barnes and Kaase 1979). Unconventional political participation includes all modes of political participation not formally linked to the electoral process such as petitioning, demonstrating, and boycotting products (Barnes and Kaase 1979).1

Macro economic conditions

The Netherlands has witnessed economic decline on several domains in the period 2002–2012. In 2002, 4.1% of the Dutch labour force was unemployed, this rose to 6.4% of the labour force in 2012 (CBS 2014a). Prime Minister Balkenende’s fourth cabinet that was elected in 2006 was confronted with the global credit crunch. In response to the global financial crisis in 2008, the Dutch government had to support the largest Dutch bank with capital injections and nationalize the second largest bank in the Netherlands. During the last year of Balkenende’s fourth cabinet, the global credit crunch started to affect the real economy and economic growth plummeted, especially in the first (−4.0%) and second quarter of 2009 (−5.0%) (CBS 2013). Moreover, the situation on the Dutch housing market deteriorated between 2006 and 2012. The number of households whose mortgage exceeded the appraised value of their house doubled between 2006 and 2012 (CBS 2014b). Against the backdrop of the global financial crisis the national debt rose from 47.4% of GDP in 2006 to 71.3% of GDP in 2012 (CBS 2014c). This was also reflected in the most important national problems mentioned by the electorate. In 2006, only 12% of the Dutch electorate reported that the economic situation was the most important national problem and this tripled in 2010 (to 38%) and rose even further to 47% in 2012 (CBS 2014d). Moreover, especially the lower-educated and lower social classes report about the deterioration of their own financial situation. In line with the figures in 2010 (Hackert et al. 2012), no less than 80% of the working class and of the lower-educated agreed in 2012 with the statement that their own financial situation would deteriorate; when we move up the social ladder, this share gradually declines to 40% among the higher educated and 30% among the higher social class (Schmeets and Gielen 2015). The European sovereign debt crisis and austerity measures in response to the economic downturn were the dominant themes in the election debates preceding the installation of the minority cabinet headed by Prime Minister Rutte in 2010. Finally, the minority cabinet Rutte collapsed when the coalition partners failed to reach an agreement over the new austerity measures resulting in new general elections in 2012. If there would be an effect of macro-economic conditions on political participation, this would especially be observable within the time frame studied here.

The effect of economic downturn on political participation may take several directions. First, economic adversity might fuel political participation. During economic hardship, governments are forced to resort to retrenchments that cause a gap between what electorates expect and what governments are able to offer (Thomassen 1990). Governments are blamed for economic duress and this blame spurs political action. This argument is very close to the relative deprivation argument presented above and can be traced back to Marx’ concept of ‘Verelendung’, who argued that in deteriorating economic conditions citizens will resort to protest to voice their political concerns.

In the same vein, Davies’ (1962) J-curve hypothesis argued that economic conditions mobilize political participation and might even overthrow regimes if a period of economic prosperity is followed by a (short) period of sharp economic decline. This sharp economic decline would lead to dissatisfaction that in turn induces political action. The competing perspective argues that economic adversity does not provide an incentive to participate but instead depresses political participation. Citizens would be more preoccupied with their personal situation in a sour economy and less able and willing to connect to the remote concerns of politics (Rosenstone 1982). Another line of reasoning is linked with the civic voluntarism model (Brady et al. 1995). Not only resources such as time, money and civic skills foster political participation, but political participation is also largely shaped by citizens’ involvement in non-political institutions such as work and voluntary organizations. This could particularly be true for the Netherlands, a country in which almost half of the population is volunteering in, e.g. sport clubs, care institutions and churches. Economic decline would diminish the number of volunteers and consequently reduce political participation. In addition, in line with the resource mobilization theory (Wilson 1973), applied to the study of social movements (Jenkins 1983), social classes respond to different kinds of incentives. Purposive incentives work for middle- and upper-class groups, while lower-class groups respond to selective incentives and collective solidarity (Wilson 1973). As, in particular, the lower social classes perceive economic threat (Schmeets and Gielen 2015), however, selective incentives and collective solidarity will simply not work in the shadow of their positions being under threat. Thus, in times of economic hardship, conventional and unconventional modes of political participation will either increase (hypothesis 1) or decrease (hypothesis 2). Since economic hardship is more prolonged and possibly more severe during the 2008-2012 period compared with that during the 2006–2010 period in the Netherlands, we assume that the effects of economic hardship on political participation will be more pronounced during the 2008–2012 period compared with the 2006–2010 period. At the individual level, various studies have empirically demonstrated that political participation, both conventional and unconventional modes, is more prevalent among higher-educated and higher-class individuals (Dalton 2008; Desposato and Norrander 2009; Norris et al. 2005). Higher social classes, those with higher income, and the higher-educated people possess more skills needed for participation (Brady et al. 1995). They have greater confidence that they understand politics and that their efforts to participate in the political arena will bear fruit (Klandermans et al. 2008; Lassen and Serritzlew 2011; Morrell 2003). Moreover, the higher educated and those in higher social classes are also more likely to be involved in civic associations that act as ‘schools of democracy’ (Van der Meer 2009). These ‘civic skills’ might reinforce the associations between resources and political participation. Hence, we expect that privileged groups in terms of education and social class participate more in conventional and unconventional modes (hypothesis 3).

Following the resource-based theory, we argue that lower-class and lower-educated citizens have lower resources to participate in politics and therefore persist to refrain from participating, whereas the more resourceful people are more likely to voice their economic concerns in the political domain. Hence, we hypothesize that the economic downturn would increase the gap in participation between privileged and underprivileged groups in society since economic adversity is expected to disproportionally affect the underprivileged, for which claim we provided the evidence already derived from previous studies (Schmeets and Gielen 2015). These underprivileged people do not have the resources to voice their concerns in the political arena. We expect that the underprivileged, who lack resources to participate to begin with, participate even less in times of economic downturn as their disposable income will decrease and consequently this will have a detrimental impact on their abilities in getting politically involved. We propose that such increases in perceived financial deterioration would further decrease political participation. Hence, our hypothesis reads: The initial differences between the underprivileged and the more privileged categories of people will increase over time during the economic crisis (hypothesis 4).

Control variables

We control for the following socio-demographic characteristics in our analyses: income, country of origin, gender, and age. People of non-Dutch origin participate less in modes of conventional and unconventional political participation (Schmeets and Houwen 2010). Political participation also varies with age. Younger people are more likely to participate in unconventional modes whereas older people are more likely to participate in conventional modes of political participation (CBS 2014e): therefore we include age as a control variable in our analyses. Men are more likely to participate in various political actions that range from conventional modes such as contacting politicians to unconventional actions as attending demonstrations (CBS 2014e).

Data and measurements

To test our hypotheses, we used the Dutch Parliamentary Election Studies (DPES 2006, 2010, 2012) conducted during each general election in the Netherlands. The DPES’ aim is to collect high-quality data on the backgrounds of voting behaviour of the Dutch electorate. The sampling frame of the DPES covers the Dutch electorate eligible to vote in parliamentary elections (Dutch citizens aged 18 or older). In 2006 and 2010, respondents were interviewed in a pre-election survey within six weeks before, and shortly after Election Day. We will only use the post-election waves of the DPES (2006) and (2010) as these contained the items on political participation. In 2012, only a post-election survey was carried out within a six-week time frame after Election Day. The post-election waves for the DPES were primarily collected by Computer Assisted Personal Interviewing (CAPI). Additionally, in 2006 and 2010 non-contacts and refusals were re-approached with a shortened questionnaire by telephone or mail. This resulted in response rates of 64.3 and 57.0% in the post-election wave (compared with the initial sample) in 2006 and 2010, respectively (Schmeets 2011). In 2012, no refusal conversion techniques were applied, and all respondents were interviewed using CAPI. This resulted in a response rate of 61.9%.

Dependent variables: conventional and unconventional political participation

Our measure for unconventional political participation refers to participation in political discussions on the internet, participating in action groups and participation in demonstrations or protest meetings. We used involving political parties or organizations, attending hearings, and contacting politicians or civil servants as indicators of conventional political participation. See Table 1 for the exact question wording in the DPES, expressing that it concerns political activities employed within the previous five years before the date of the interview.
Table 1

Question wording political participation Dutch Parliamentary Election Studies 2006–2012

 

Item

Answer categories

 

There are several ways to influence politicians, civil servants or the government. Please list which one you used during the previous five years.

 

Unconventional political participation

Participated in a political discussion on the internet, via sms or e-mail.

Yes/no

Participated in an action group.

Yes/no

Participated in a demonstration or protest meeting.

Yes/no

Conventional political participation

Contacted a politician or government official.

Yes/no

Participate in a hearing or consultation meeting organized by the government.

Yes/no

Tried to involve political party or organization.

Yes/no

Scale construction: conventional and unconventional political participation

We constructed separate scales for conventional and unconventional political participation that represent the average score on the relevant dichotomous indicators.2 We assessed the scalability of these items using probabilistic scale analysis techniques (Mokken 1971; van Schuur 2003). Mokken-scale analysis is the probabilistic version of the deterministic Guttmann scale. Mokken-scale analysis uses a set of dichotomous indicators, for instance, involving a political party yes or no, and evaluates whether certain items, such as political discussion on the internet, may be easier and thereby more popular activities compared with others, such as attending a demonstration. The decisive notion is that those who engage in more difficult, i.e. less popular, activities will probably (not necessarily) also engage in easier or more popular activities.3 This means that the probability of a positive response to an item increases in concordance with the value of a subject’s latent trait. Hence, we test whether individuals that engage in a more difficult or less popular activity also engage in a less difficult political activity. Applied to unconventional modes of political participation, this means we test whether those that engage in a more difficult activity such as attending a demonstration also engage in easier unconventional political activities such as engaging in a political discussion on the internet.

The cumulative nature of the response on the items for political participation also has important theoretical implications that are neglected when analysing these items separately. It is theoretically (implicitly or explicitly) assumed that people specialize within either conventional or unconventional modes of participation and that participation is cumulative (c.f. Millbrath 1965; Verba et al. 1978; Zukin 2006). Mokken scaling incorporates the respective ‘difficulty’ of certain acts of political participation vis-à-vis other, easier or more mainstream acts of political participation. By assessing the respective difficulty of acts of political participation using Mokken scaling, we acknowledge this cumulative nature of participation.

In terms of comparability of measurements, or equivalence, we analyse the extent to which the ordering in terms of difficulty of the items is similar over time. If the ordering in modes of political participation is similar over time, measurements are considered equivalent and scale-scores can be compared.

The results of the Mokken-scale analysis are presented in Table 2. For each act of political participation the proportion of people that engaged in these acts the past five years is shown for conventional and unconventional modes. In the context of Mokken-scale analyses, these proportions represent the ‘item difficulties’. We find that for unconventional modes of political participation the item ordering pattern from most popular to least popular activity is political discussion on the internet, participation in an action group, and demonstrating, respectively. For conventional modes of political participation, Table 2 shows that contacting a politician or government official is the most popular activity, followed by attending a hearing. Involving a political party is the least popular conventional political activity.
Table 2

Mokken-scale analysis: Item difficulties (proportion positive responses) and Loevinger’s H, by years and subgroups

Item

Unconventional political participation

Proportion positive responses (item popularity)

Conventional political participation

Proportion positive responses (item popularity)

Participated in political discussion on the internet, via sms or e-mail

Participated in an action group

Participated in demonstration or protest meeting

Loevinger’s H

Contacted a politician or government official

Participate in a hearing or consultation meeting organized by the government

Tried to involve political party or organization.

Loevinger’s H

Years

 2006

0.19

0.08

0.04

0.32

0.12

0.12

0.06

0.42

 2010

0.22

0.05

0.03

0.31

0.13

0.12

0.05

0.45

 2012

0.24

0.07

0.04

0.33

0.10

0.09

0.04

0.45

 Total

0.21

0.07

0.04

0.32

0.12

0.11

0.05

0.44

Table 2 also shows that the item ordering pattern is similar across time and thus longitudinally equivalent. This is represented in the Loevinger’s H coefficients. These represent the scalability of the Mokken scale based on the number of violations of the item-ordering pattern. A violation of the item ordering pattern would occur if respondents do engage in more difficult acts (for instance demonstrating), but do not engage in easier more popular acts (for instance joining a political discussion on the internet). The Loevinger’s H coefficients are above the cut-off value of 0.3 (Sijtsma and Molenaar 2002). Based on the results of these Mokken-scale analyses, we construct a scale that consists of the average score on the items pertaining to conventional and unconventional political participation that ranges between 0 and 100 so that our coefficients in the analyses represent the percentage of active citizens.

Resources

We operationalized resources in two different ways. First, we used education as a proxy for someone’s civic skills and resources. Education is measured in five categories: elementary education, lower vocational education, secondary education, middle-level vocational education or higher-level secondary education, and, finally higher-level vocational education or university. We include education as a continuous variable in our analyses.4 Second, we used social class as a proxy for resources and relative deprivation. Respondents were asked which social class they perceived themselves to be a member of: “One sometimes speaks of the existence of various social classes and groups. If you were to assign yourself to a particular social class, which one would that be?” The categories range from upper class, upper middle class, middle class, upper working class, to working class. Social class is included as having a linear relationship with the dependent variables in our analyses. .5

The DPES data were enriched with registry-based information on income drawn from the Dutch tax office. We used the standardized disposable annual income. The disposable income is composed of wages, profits (for self-employed persons) and other allowances minus social contributions and taxes, standardized for household size and composition. To arrive at a longitudinally comparative measure of income, the standardized household income was classified in vigintiles according to the Dutch population. For reference, the lowest vigintile in 2006 represents spendable incomes lower than € 9530 per year, whereas the highest category represents spendable incomes of € 41,243 and higher. To take into account possible nonlinear associations between income and political participation, we included a quadratic term for income.

Control variables

We controlled for age, gender, and country of origin. Age was defined as age at Election Day. In the Netherlands, the age threshold for participating in elections is 18 years. To control for possible nonlinear effects of age, we included a quadratic term for age as well. For origin we distinguished between Dutch origin and non-Dutch origin. Respondents who were born in the Netherlands and have parents that were born in the Netherlands were classified as Dutch origin. Those who were born outside the Netherlands themselves or their parents were classified as non-Dutch. The descriptive statistics for all relevant variables are presented in Table 3.
Table 3

Descriptive statistics

 

Min

Max

Mean

SD

Conventional political participation

0.00

100.00

9.32

21.54

Unconventional political participation

0.00

100.00

10.51

19.37

Income

1.00

20.00

10.92

5.67

Age

18.00

96.00

47.55

17.56

 

%

n

Highest education completed

6.6

249

 Elementary

16.6

727

 Lower vocational

8.3

347

 Secondary

41.8

1875

 Middle-level vocational/Higher-level secondary

26.7

1401

 Higher-level vocational/university

  

Social class

 Working class

18.7

740

 Upper working class

12.9

578

 Middle class

49.1

2302

 Upper middle class

17.2

874

 Upper class

2.1

105

Gender

 Male

50.9

2307

 Female

49.1

2292

Origin

 Dutch origin

85.4

4127

 Non-Dutch

14.6

472

Total

 

4599

Analyses

The results for conventional political participation are displayed in Table 4. In Table 5, we present the results of the analyses for unconventional modes of participation. We use the same estimation strategy for both modes of participation. In the first model, we include dummy variables for the years 2010 and 2012 to assess whether there is significant longitudinal variation compared with 2006 in political participation in the Netherlands in times of economic crises. Model 2 includes the main independent variables that refer to resources (education and social class) and the control variables. In model 3 and model 4, we assess whether the effects of social class and education on political participation systematically diverge in times of economic crises by including interaction terms. We ran OLS-regression analysis with the Mokken scales for conventional and unconventional political participation presented above as dependent variables.6,7,8
Table 4

Linear regression analysis: conventional political participation (n = 4, 599)

 

Model 1

Model 2

Model 3

Model 4

b

SE

b

SE

b

SE

b

SE

2006

Ref.

 

Ref.

 

Ref.

 

Ref.

 

2010

0.527

(0.763)

0.127

(0.739)

−1.599

(2.436)

−2.175

(2.673)

2012

−2.134**

(0.776)

−2.236**

(0.754)

−1.036

(2.451)

0.739

(2.705)

2012a

−2.662*

(0.801)

−2.363*

(0.779)

0.563

(2.580)

2.913

(2.851)

Level of education

  

2.686**

(0.308)

2.647**

(0.454)

2.545**

(0.487)

Social class

  

1.903**

(0.360)

1.890**

(0.360)

2.142**

(0.568)

Level of education

        

 2010

    

0.456

(0.620)

0.297

(0.700)

 2012

    

−0.316

(0.623)

0.145

(0.697)

 2012a

    

−0.771

(0.648)

−0.152

(0.725)

Social class

        

 2010

      

0.411

(0.828)

 2012

      

−1.268

(0.837)

 2012a

      

−1.679

(0.872)

Income

  

−0.419*

(0.236)

−0.443*

(0.237)

−0.453*

(0.237)

Income2

  

0.024*

(0.011)

0.025**

(0.011)

0.026**

(0.011)

Female

  

Ref.

 

Ref.

 

Ref.

 

Male

  

3.752**

(0.617)

3.734**

(0.617)

3.720**

(0.617)

Dutch origin

  

Ref.

 

Ref.

 

Ref.

 

Non−Dutch origin

  

−0.111

(1.016)

−0.157

(1.017)

−0.168

(1.017)

Age

  

0.695**

(0.101)

0.690**

(0.101)

0.693**

(0.101)

Age2

  

−0.006**

(0.001)

−0.006**

(0.001)

−0.006**

(0.001)

Intercept

9.809**

(0.521)

−24.668**

(2.867)

−24.285**

(3.151)

−24.642**

(3.193)

R 2

0.003

 

0.072

 

0.072

 

0.073

 

p < 0.05; ** p < 0.01 (two tailed)

aTests for significance with the reference category being 2010

Table 5

Linear regression analysis: unconventional political participation (n = 4, 559)

 

Model 1

Model 2

Model 3

Model 4

b

SE

b

SE

b

SE

b

SE

2006

Ref.

 

Ref.

 

Ref.

 

Ref.

 

2010

−0.208

(0.687)

−0.497

(0.667)

2.000

(2.201)

2.214

(2.416)

2012

1.430*

(0.698)

1.355*

(0.681)

1.090

(2.214)

2.357

(2.445)

2012a

1.638*

(0.720)

1.852**

(0.704)

−0.910

(2.331)

0.142

(2.577)

Level of education

  

2.528**

(0.278)

2.712**

(0.411)

2.579**

(0.440)

Social class

  

1.201**

(0.325)

1.212**

(0.325)

1.545**

(0.514)

Level of education

        

 2010

    

−0.663

(0.560)

−0.592

(0.633)

 2012

    

0.065

(0.563)

0.404

(0.630)

 2012a

    

0.728

(0.586)

0.997

(0.655)

Social class

        

 2010

      

−0.174

(0.748)

 2012

      

−0.918

(0.757)

 2012a

      

−0.744

(0.789)

Income

  

−0.247

(0.213)

−0.224

(0.214)

−0.231

(0.214)

Income2

  

0.008

(0.010)

0.007

(0.010)

0.007

(0.010)

Female

  

Ref.

 

Ref.

 

Ref.

 

Male

  

0.235

(0.557)

0.256

(0.557)

0.254

(0.557)

Dutch origin

  

Ref.

 

Ref.

 

Ref.

 

Non−Dutch origin

  

−0.143

(0.918)

−0.103

(0.919)

−0.105

(0.919)

Age

  

0.115

(0.091)

0.121

(0.092)

0.121

(0.092)

Age2

  

−0.002**

(0.001)

−0.002**

(0.001)

−0.003**

(0.001)

Intercept

10.141**

(0.469)

−0.255

(2.590)

−1.186

(2.847)

−1.606

(2.886)

R 2

0.001

 

0.062

 

0.063

 

0.063

 

p < 0.05; ** p < 0.01 (two tailed)

aTests for significance with the reference category being 2010

Results

Let us first look at conventional political participation in Table 4. In model 1, we find that there is no significant difference in levels of political participation during the periods of 2002–2006 and 2006–2010 whereas a significant (albeit small) decline is observed during the 2008–2012 period compared with 2002–2006 period (and also in comparison with 2006–2010 period). This suggests that levels of conventional political participation are slightly in decline in times of prolonged economic hardship. Moreover, this effect also holds after including all control variables in model 2. Model 1 in Table 5 demonstrates that unconventional political participation remains stable between 2002–2006 and 2006–2010 but is slightly and significantly on the rise when the economy turns sour during the 2008–2012 period. This also holds when controlling for relevant characteristics in model 2. These results do not firmly support the first hypothesis as only the unconventional activities increased, but not the conventional activities. Likewise, we do not find a confirmation for the second hypothesis as there is only a small decline for the conventional modes found.

Furthermore, our results demonstrate that with higher levels of education, conventional as well as unconventional political participation increases. Also, those who consider themselves belonging to higher social classes consistently participate more in both modes. This confirms the third hypothesis. The interaction terms in model 3 and model 4 demonstrate that the economic crisis does not influence the strength of the effect of education on conventional political participation. Hence, levels of conventional and unconventional political participation do not diverge between privileged and underprivileged groups in times of economic crisis. These findings clearly reject the fourth hypothesis.

Concerning our control variables we find that income affects conventional political participation in curvilinear fashion. The lowest levels of conventional political participation are found in the 8th and 9th vigintile and increases from the 10th vigintile onwards. However, income is not significantly related to unconventional modes of participation. Moreover we demonstrate that men are more likely to participate in conventional modes but there is no gender difference in their level of unconventional political participation (c.f. Tables 4 and 5). Origin is not related to both conventional and unconventional modes of participation. The calculation based on the age and age × age variables reveals that the conventional political participation increases to the top found among people aged 59 after which it decreases. For unconventional political participation the top is 30 year, and evidently lower.

Conclusion and discussion

In this contribution, we attempted to explore the effects of the economic crisis on levels of political participation in the Netherlands between 2002 and 2012, in general and more particularly among privileged and underprivileged groups in society. We used the classic distinction between conventional and unconventional modes of political participation as proposed by Barnes and Kaase (1979). Against the background of the recent economic crises, we argued that economic downturn would either incentivize citizens to voice their concerns in the political domain (Thomassen 1990; Muñoz et al. 2013; Ponticelli and Voth 2011) or that economic downturn would induce political apathy (Rosenstone 1982). We combined this with the expectations following from the civic voluntarism model and the resource mobilization theory proposed that during times of economic crisis privileged and underprivileged groups in society systematically diverge in their levels of participation.

Using probabilistic scale modelling techniques, we demonstrated that both conventional and unconventional political participation are cumulative behavioural patterns. Hence, people who engage in more difficult political acts, such as demonstrating, also engage in easier or more mainstream acts of political participation such as joining political discussion on the internet. Moreover, we demonstrated that the pattern in modes of both conventional and unconventional political activities does not change substantially over time.

Our results indicate, first and foremost that most people do not engage in any conventional and unconventional mode of political participation studied here. Only some one in three citizens employed an activity in trying to influence politicians, civil servants or the government within a five years period. However, we find that in recent times of economic downturn in the Netherlands, conventional political participation decreases while unconventional political participation is on the rise. We demonstrate that those with more skills and resources (higher educated) and people in higher social classes participate more in both conventional and unconventional modes of participation which is consistent with the resource-based explanation of political participation as proposed by Brady et al. (1995). We hypothesized that underprivileged groups would be disproportionally affected by the economic crisis compared with privileged groups, thereby increasing the gap between these strata in society. We distinguished between higher- and lower social classes and higher- and lower-educated individuals. However, we do not find that the difference between privileged and underprivileged societal groups is affected by the economic downturn.

One might argue that an increase in unconventional political participation combined with a decrease in conventional modes would be emblematic of a more structural trend of switching political repertoires (Dalton 2008; Norris 2011) from conventional to unconventional modes. Citizens would increasingly voice their concerns through unconventional modes of political participation at the expense of conventional modes (Dalton 2008). However, the decrease in conventional political participation and the increase in unconventional modes only occur in the 2008-2012 period—in a time when the Netherlands suffered from economic adversity. We speculate that this increase in unconventional modes is due to the difference in required skills for conventional and unconventional political participation. It takes more civic skills to involve a political party compared to joining in a political discussion on the internet, and during times of economic downturn, citizens might increasingly voice their concerns through more accessible means of political participation. The underlying mechanisms governing this effect need to be taken into account in more detail in further research to assess the validity of this assumption.

Footnotes

  1. 1.

    Unconventional political participation has been labelled differently throughout time. One might argue that unconventional activities are increasingly accepted and regarded as ‘normal’ modes of (Dalton 2008; Lamprianou 2013; Norris et al. 2005). Thus, some of the activities such as attending a demonstration lost its ‘unconventional’ connotation. This renders the term ‘unconventional’ political participation somewhat old-fashioned. However, other labels used such as ‘extra-institutional’ participation, ‘emerging forms of political participation’, and ‘non-electoral participation’ still refer to the same political actions empirically. Thus, although labelled differently, the acts referred to when describing ‘unconventional’, ‘non-institutionalized’ or ‘protest participation’ are identical since they still refer to non-legally embedded political actions such as petitioning and demonstrating whereas conventional action still refer to legally embedded modes such as attending hearings and writing to government officials.

  2. 2.

    We omitted voting out of our analyses since this is a different kind of political activity compared with the conventional and unconventional activities studied here. People vote once every few years while most of the conventional and unconventional political activities studied here, such as attending a demonstration or writing to government officials, requires more prolonged time commitments.

  3. 3.

    Mokken-scale analysis has numerous advantages over more mainstream scaling methods, such as factor analyses and measurement models specified in structural equation modelling. These methods are based on the decomposition of covariances and assume that frequency distributions of the items can be regarded as ‘parallel’ and the items have more or less the same mean and standard deviation. Thus, all items need to be equally ‘popular’ to be adequately used for scaling (van Schuur 2003). Distribution of the items for political participation clearly demonstrates that this is not the case, e.g. the proportion of people who engage in political discussion on the internet, is considerably larger compared with the proportion of people who attended a demonstration.

  4. 4.

    We compared a model using education as a continuous variable to a model including dummy variables for each category of education. Model 2 including dummy variables results in a slightly better model fit compared with a model including education as a continuous variable for both dependent variables (R 2 change = 0.012 with p = 0.000 for unconventional political participation and R 2 change = 0.011 with p = 0.000 for conventional political participation). However, the results including dummy variables for education, which are presented in the appendix, do not qualitatively differ from the results presented in Tables 4 and 5. For reasons of brevity and to gain statistical power, we present the parsimonious models including education as a continuous measurement.

  5. 5.

    We assessed whether social class can be modelled as a pseudo-interval variable by testing to what extent the association between social class and conventional and unconventional political participation can be modelled with linear terms only. We ran model 2 including social class as a set of dummy variables and compared these to models using social class as a linear effect. No difference was found in fit between the two models (R 2 change = 0.000 with p = 0.943 and R 2 change = 0.001 with p = 0.115 for unconventional and conventional political participation, respectively).

  6. 6.

    Given the ordinal and very skewed nature (most people do not participate politically), one might argue that ordered logit regression analyses is more appropriate. We compared the estimates presented here with the estimates from ordered logit models; the results did not substantially differ from OLS results, which are presented in this paper for ease of interpretation.

  7. 7.

    These analyses presented here were performed on the unweighted sample. Analysis with the sample weighted according to age, gender, marital status, urbanization, region, origin, turnout (i.e.. voted in most recent parliamentary elections yes/no), and voting behaviour did not differ from the results presented here for both conventional and unconventional political participation.

  8. 8.

    There is no substantial difference between the models including all missing values on each variable as separate categories and the estimates presented in Tables 4 and 5.

References

  1. Aarts, K., H. van der Kolk, and M. Rosema. 2007. Een verdeeld electoraat? Houten: Spectrum.Google Scholar
  2. Anderson, C.J. 2007. The end of economic voting? Contingency dilemmas and the limits of democratic accountability. Annual Review of Political Science 10: 271–296.CrossRefGoogle Scholar
  3. Barnes, S.H., and M. Kaase. 1979. Political action: Mass participation in five western democracies. Beverly Hills, CA: Sage.Google Scholar
  4. Brady, H.E., S. Verba, and K.L. Schlozman. 1995. Beyond ses: A resource model of political participation. American Political Science Review 89 (2): 271–294.CrossRefGoogle Scholar
  5. CBS. 2013. Sustained economic contraction. Statistics Netherlands press release, PB13-011, 14 February 2013. The Hague/Heerlen: Statistics Netherlands.Google Scholar
  6. CBS. 2014a. Labour force; main figures by sex and other personal characteristics. CBS Statline. Retrieved 2 April, 2014 from: http://statline.cbs.nl/.
  7. CBS. 2014b. Financieel risico hypotheekschuld; eigenwoningbezitters (Financial risk mortgage debt; homeowners). CBS Statline. http://statline.cbs.nl/. Accessed 2 April 2014.
  8. CBS. 2014c. Government finance statistics; key figures. CBS Statline. http://statline.cbs.nl/. Accessed 2 April 2014.
  9. CBS. 2014d. Nationale problemen volgens stemgerechtigden (National problems according to the electorate). CBS Statline. http://statline.cbs.nl/. Accessed 3 April 2014.
  10. CBS. 2014e. Sociale contacten en maatschappelijke participatie (Social contact and societal participation). http://statline.cbs.nl/. Accessed 6 May 2014.
  11. Dalton, R.J. 2008. Citizenship norms and the expansion of political participation. Political Studies 56 (1): 76–98.CrossRefGoogle Scholar
  12. Davies, J.C. 1962. Toward a theory of revolution. American Sociological Review 27 (1): 5–19.CrossRefGoogle Scholar
  13. Desposato, S., and B. Norrander. 2009. The gender gap in latin america: contextual and individual influences on gender and political participation. British Journal of Political Science 39 (1): 141–162.CrossRefGoogle Scholar
  14. DPES. 2006. Dutch parliamentary election surveys 2006. Amsterdam: DANS KNAW.Google Scholar
  15. DPES. 2010. Dutch parliamentary election studies 2010. Amsterdam: DANS KNAW.Google Scholar
  16. DPES. 2012. Dutch parliamentary election studies 2012. Amsterdam: DANS/KNAW.Google Scholar
  17. Gesthuizen, M., P. Scheepers, W. Van Der Veld, and B. Völker. 2013. Structural aspects of social capital: Tests for cross-national equivalence in europe. Quality & Quantity 47 (2): 909–922.CrossRefGoogle Scholar
  18. Hackert, M., R. Linssen, and H. Schmeets. 2012. Economische en culturele dreiging: Wie ervaart dreiging en wie niet? [Economic and cultural threat: Who feels threatened and who doesn’t?]. Sociaal-economische trends 2012 (4): 45–52.Google Scholar
  19. Jenkins, J.C. 1983. Resource mobilization theory and the study of social movements. Annual Review of Sociology 9 (1983): 527–553.CrossRefGoogle Scholar
  20. Klandermans, B., J. Van der Toorn, and J. Van Stekelenburg. 2008. Embeddedness and identity: How immigrants turn grievances into action. American Sociological Review 73 (6): 992–1012.CrossRefGoogle Scholar
  21. Lamprianou, I. 2013. Contemporary political participation research: A critical assessment. In Democracy in transition, ed. K. Demetriou, 21–42. Berlin: Springer.CrossRefGoogle Scholar
  22. Lassen, D.D., and S. Serritzlew. 2011. Jurisdiction size and local democracy: Evidence on internal political efficacy from large-scale municipal reform. American Political Science Review 105 (2): 238–258.CrossRefGoogle Scholar
  23. Lewis-Beck, M.S., and M. Stegmaier. 2000. Economic determinants of electoral outcomes. Annual Review of Political Science 3 (1): 183–219.CrossRefGoogle Scholar
  24. Linssen, R., and H. Schmeets. 2010. Participatie en vertrouwen in europa (Participation and trust in europe). In Sociale samenhang: Participatie, vertrouwen en integratie (Social cohesion: Participation, trust, and integration), ed. H. Schmeets, 135–151. The Hague/Heerlen: Statistics Netherlands.Google Scholar
  25. Millbrath, L. 1965. Political participation: How and why do people get involved in politics?. Chicago: Rand McNally.Google Scholar
  26. Mokken, R.J. 1971. A theory and procedure of scale analysis with applications in political research. The Hague: Mouton & Co.CrossRefGoogle Scholar
  27. Morrell, M.E. 2003. Survey and experimental evidence for a reliable and valid measure of internal political efficacy. Public Opinion Quarterly 67 (4): 589–602.CrossRefGoogle Scholar
  28. Muñoz, J., G. Rico and E. Anduiza. 2013. Austerity policies and political involvement. Evidence from a panel survey in spain (2010-11) Working paper Universitat Autonoma de Barcelona.Google Scholar
  29. Norris, P. 2011. Democratic deficit: Critical citizens revisited. Cambridge: Cambridge University Press.CrossRefGoogle Scholar
  30. Norris, P., S. Walgrave, and P. Van Aelst. 2005. Who demonstrates? Antistate rebels, conventional participants or everyone? Comparative Politics 37 (2): 189–205.CrossRefGoogle Scholar
  31. Ponticelli, J. and H.A. Voth. 2011. Austerity and anarchy: Budget cuts and social unrest in europe, 1919-2008. http://dx.doi.org/10.2139/ssrn.1899287. Accessed 26 July 2013.
  32. Radcliff, B. 1992. The welfare state, turnout, and the economy: A comparative analysis. American Political Science Review 86 (2): 444–454.CrossRefGoogle Scholar
  33. Rosenstone, S.J. 1982. Economic adversity and voter turnout. American Journal of Political Science 26: 26–46.CrossRefGoogle Scholar
  34. Schmeets, H. 2011. Verkiezingen: Participatie, vertrouwen en integratie (Elections: Participation, trust, and integration). The Hague/Heerlen: Statistics Netherlands.Google Scholar
  35. Schmeets, H. and W. Gielen. 2015. Econische en Culturele dreiging in Nederland (Econonomic and cultural threat in the Netherlands) In Nationaal Kiezersonderzoek 2006-2012 (Dutch Parliamentary Election Sudy 2006-2012) ed. Schmeets, H, 62-76. The Hague/Heerlen: Statistics Netherlands.Google Scholar
  36. Schmeets, H. and K. Van der Houwen. 2010. Politieke betrokkenheid van allochtonen (Political engagement of minorities), ed Schmeets, H, 186-196. The Hague/Heerlen: Statistics Netherlands.Google Scholar
  37. Sijtsma, K., and W. Molenaar. 2002. Introduction to nonparametric item response theory. Thousand Oaks: Sage.CrossRefGoogle Scholar
  38. Thomassen, J.J. 1990. Economic crisis, dissatisfaction, and protest. In Continuities in political action: A longitudinal study of political orientations in three western democracies, eds. M.K. Jennings, J.W. Van Deth, S.H. Barnes, D. Fuchs, F.J. Heunks, R.F. Inglehart, and J.J. Thomassen, 103–134. Berlin: Walter de Gruyter.Google Scholar
  39. Van der Meer, T. 2009. States of freely associating citizens: Comparative studies into the impact of state institutions on social, civic and political participation. Nijmegen: Radboud Universiteit Nijmegen ICS dissertation.Google Scholar
  40. Van Schuur, W. 2003. Mokken scale analysis: Between the guttman scale and parametric item-response theory. Political Analysis 11 (2): 139–163.CrossRefGoogle Scholar
  41. Verba, S., N. Nie, and J. Kim. 1978. Participation and political equality: A seven-nation comparison. Cambridge: Cambridge University Press.Google Scholar
  42. Wilson, J. 1973. Introduction to social movements. New York: Praeger.Google Scholar
  43. Zukin, C. 2006. A new engagement: Political participation, civic life, and the changing american citizen. Oxford: Oxford University Press.CrossRefGoogle Scholar

Copyright information

© Macmillan Publishers Ltd 2017

Authors and Affiliations

  • Rik Linssen
    • 1
  • Peer Scheepers
    • 2
  • Manfred te Grotenhuis
    • 2
  • Hans Schmeets
    • 1
    • 3
  1. 1.Department of Political ScienceMaastricht UniversityMaastrichtThe Netherlands
  2. 2.Department of SociologyRadboud UniversityNijmegenThe Netherlands
  3. 3.Division of Social and Spatial StatisticsStatistics NetherlandsHeerlenThe Netherlands

Personalised recommendations