After the crisis: political protest in the aftermath of the economic recession

Abstract

Economic recessions have often led to stronger citizen activism. This article assesses the relationship between the economy and protest in 2014, 6 years after the global financial crisis took place, a long-enough period for countries to have improved their economic situation and for people’s interpretations of the economy to be more optimistic. Does the economy still matter to explain protest if it is not as salient any longer? This research employs data for available European Union member states from the 2006, 2008, and 2014 European Social Survey to test the importance of national-level objective economic indicators as well as individual-level evaluations of financial well-being. Findings from the research suggest that objective economic predictors are more relevant to understand protest in 2014 than before the crisis. Economic resources remain more important for the prediction of protest than deprivation views, with the exception of unemployment. Even at times of partial economic recovery, the state of the economy helps explain increased levels of political protest across Europe well after 2008. The link between the economy and confrontational activism before and after the economic recession looks in the end very similar and a full economic recovery in the future can lead to even more protest activism in Europe.

Introduction

Over the last decade, countries in Europe have experienced a weak economy, with high unemployment rates and the adoption of painful austerity measures. In most cases, citizens reacted to this sharp economic decline by strongly opposing their own government (Della Porta 2015; Trenz et al. 2015). The distress caused by reduced salaries and pensions, cutbacks in public services, and smaller social policy budgets was responsible for the general outrage. In the most extreme cases of economic collapse, the so-called PIIGS countries (Portugal, Ireland, Italy, Greece, and Spain), levels of confrontational activism (street demonstrations, building occupations, damages to property, or general strikes) spiked, with greater numbers of citizens participating in forms of confrontational action (Verney and Bosco 2013; Calvo 2013; Kosmidis 2014; Accornero and Ramos Pinto 2015). Even countries with more favorable financial situations experienced an increase in unconventional political activity (Vassallo and Ding 2016), which suggests the relevance of the economy in predicting protest for countries whose level of deprivation was not as severe.

As the intensity of the crisis has passed, the link between economic indicators and street marches may have weakened. Previous studies¹ on post-recession Europe (Kern et al. 2015; Quaranta 2016; Vassallo and Ding 2016) used relative deprivation theory to explain protest. This research highlighted the relevance of economic scarcity as a grievance strong enough to convince more people to take to the streets. Was this assessment in the recent literature only registering a sudden and temporary association between the economic recession and protest activism in Europe? Can the economy still explain protest when the overall financial situation is not as salient? Do people adapt to the new normal, with a lower standard of living, when the deprivation is no longer felt as much?

This article assesses the role of economic variables in explaining political protest in member states in the European Union (EU) 6 years after the 2008 financial meltdown, with the use of data from the European Social Survey (ESS) from the 2006, 2008 and 2014 rounds. As the post-recession era has created new economic standards, high unemployment or low GDP growth may not be enough to convince regular citizens to occupy a building. At the same time, if the recession has permanently changed the relationship between the economic situation and unconventional activism, a lower level of satisfaction for the economy can still lead to protest.² This study is ultimately attempting to assess the possibility of long-term consequences from the 2008 economic downturn when predicting protest behavior today.

The impact of deprivation in explaining why people protest in Europe has been challenged by other scholars. In his analysis of the PIIGS countries, Ancelovici (2015) states that relative deprivation is not a sufficient element to understand protest (p. 205). The use of economic variables to predict unconventionality has also been disputed as researchers support different interpretations of how economic performance affects contentious politics. Originally, relative deprivation theory (Gurr 1970) listed poor economic conditions, disappointment with economic policies, economic injustice, and a gap in personal economic expectations as driving factors for people ‘to rebel.’This situation comes to mind when thinking of the thousands of demonstrators on the streets in Greece since 2008 (Rüdig and Karyotis 2014) or the equally large number of empowered citizens engaged in the Podemos revolution in Spain (Calvo 2013).

In addition to studying this argument at the individual level, it has also been investigated at the country level, where economic downturns seem to rally citizens to challenge, revolt, and overtake governments and political systems through violent protest (Lichbach 1989). An alternative approach linking the economy to unconventionality has supported a positive relationship between economic performance and confrontational activism. In contrast to the relative deprivation-driven explanation, this approach contends that wealthier citizens are more likely to engage in contentious activism, including street demonstrations, general strikes, or square occupations, when the economy is actually performing well (Powell 1982; Verba et al. 1995; Dalton et al. 2010): a resource-based explanation.

As some economic measures in Europe have been improving since 2008, it is useful to study once again how the economy drives people to protest, whether objective economic indicators are as relevant as subjective interpretations of well-being in explaining demonstrations, or whether social economic protection can actually limit the impact of deprivation in predicting protest. The next section of this paper introduces the literature on protest and economic performance; the third section presents the data and the hypotheses. A section on the discussion of the findings follows, and a conclusion summarizes the article’s main contributions.

Studying the link between protest and the economy

Studies on unconventional political activism have often presented a multitude of measures of protest.³ Previous volumes on protest made the distinction between hard versus soft action, confrontational versus peaceful activism, or disruptive versus non-violent engagement (Powell 1982; Dubrow et al. 2008; Dodson 2011; Welzel and Deutsch 2012; Solt 2015). In the end, the focus on the type of protest studied depended on the measures available and the accessibility of data. For instance, information on violent protest activism is hard to gather and possibly dangerous to distribute. At the same time, few individuals are generally involved with a disruptive protest action, as people do not want to suffer possible negative consequences. Yet, confrontational protest may not be an option for most individuals in society. Citizens in advanced democracies have increasingly embraced peaceful protest actions against their governments as a form of political involvement (Inglehart and Catterberg 2002).

Actual measures of protest vary across geographic locations and time. Even in Europe, participation in unconventional activism includes a myriad of possible actions at the local, national and European levels (Imig and Tarrow 1999). Typical examples of protest in the literature include signing a petition, taking part in a general strike, occupying a building, participating in a lawful demonstration, damaging property, or getting involved in a violent riot (Barnes and Kaase 1979; Dalton 2014). An action as simple as a petition can result in a confrontational approach when large numbers of citizens openly question a governmental decision or policy, forcing political leaders to make changes.⁴

In more recent investigations, scholars addressed how to incorporate newer forms of activism that could be considered outside of the realm of conventionality. A relevant example is the use of political consumerism (Stolle et al. 2005) as a new form of activism that is not institutionalized and yet challenging enough to be associated with protest. A citizen’s choice to boycott a certain product may sound easy, but in the end is a way to express a political opinion without the need of any party or election.⁵

In this article, the main focus is on protest as measured by participation in demonstrations, boycotts of products, and support for petitions. These three items provide a good variety of actions with regard to people’s ability to act politically, allowing individuals to choose a fairly accessible action (such as signing a petition) or a more demanding activity (such as participating in a lawful demonstration). These forms of contentious political actions represent a middle ground in relation to the division in the literature between hard versus soft protest.

The use of economic indicators to predict protest relies on two features: the salience of the economy and the level of economic assessment employed. The link between economic performance and protest activism appeared to be complicated from the very first studies published on the relationship. Gurr (1970) presented a theory of unconventional political behavior that emphasized the individual’s role in order to explain how relative economic deprivation is conducive to protest. The severity of the economic downturn and the length of the financial crisis contribute to influence interpretations of lower economic well-being. The gap between expectations of economic standards and the actual economic situation is at the base of the deprivation leading to action. When the economic crisis persists, and citizens are exposed to the negative consequences of the economic recession for longer periods (Singer 2011), politically driven protest is likely. Anderson and Hecht (2014) and Armingeon and Guthmann (2014) confirmed the link between the objective economic indicators and the corresponding subjective assessments of the economic situation with regard to the economic recession in Europe after 2008: In times of crisis, citizens apparently demonstrate a good understanding of the state of the economy.

At the same time, Powell (1982) connected country-level GNP per capita with data on protest from the late 1950s through the late 1970s, demonstrating that a wealthier society supports protest involvement. Later studies on unconventionality and the economy presented the similar conundrum that a good as well as a bad economy can lead to protest (for instance Auvinen 1997; Dalton et al. 2010; Vassallo and Ding 2016).

When looking at relevant works on this topic, the choice for the appropriate level of economic variables shifts from the macro-level (usually the country) to the individual level (a citizen in a country). Some examples of macro-level economic measures often used are GDP, unemployment, inflation, or government debt (Auvinen 1997; Kern et al. 2015; Beissinger and Sasse 2014). All of them are considered objective economic indicators that can easily be measured across countries. In more detail, economic affluence (as measured by GDP or GNP) is often positively associated with protest due to people’s resources to engage in unconventional activism (Verba et al. 1995; Jenkins et al. 2008; Dalton et al. 2010; Vassallo and Ding 2016). Unemployment rates have also been used often to test the relationship. Some studies supported the interpretation of a positive relationship with protest, following deprivation theory expectations (Lahusen 2013; Kern et al. 2015; Vassallo and Ding 2016). Other publications described the negative association (Gallego 2007; Schussman and Soule 2005; Jenkins et al. 2008) when emphasizing that employed citizens were more involved in contentious politics, as they had more interests to protect.

In a similar situation, investigations on economic inequality have contradicted the relative deprivation position: Societies with higher levels of inequality are associated with lower levels of protest (Dubrow et al. 2008; Solt 2015). Lastly, measures of social protection at the national level have been useful to equally verify the link between economic grievances and protest: Expectations for a negative relationship between more social protection and lower protest have not been supported (Sanders and Bellucci 2012). Individuals who benefit from a higher level of social protection are actually more likely to engage in unconventional activity as they feel less concerned about possible negative consequences, and can still count on enough resources to be able to participate.

Recent studies on increased unconventionality in Europe have emphasized mostly the role of relative deprivation theory (Kern et al. 2015; Quaranta 2016; Vassallo and Ding 2016). Other scholars have instead dismissed the positive link between deprivation and grievance as a predictor for protest action (Solt 2015; Rüdig and Karyotis 2014). Before 2008, the research agreed on a minimal impact of economic deprivation on unconventionality (Dalton et al. 2010; Welzel and Deutsch 2012). Relative deprivation was overall useful in predicting protest only when there was a severe crisis and the prolonged impact of economic austerity remained dominant. These two conditions were present for the European recession, but have since weakened. Some countries have at least in part recovered financially, with higher GDP per capita in comparison to the 2008 data⁶ (Eurostat). In a similar approach, when considering EU-level economic indicators, real growth was 1.8% in 2014, up from − 0.4% in 2012. Even unemployment data for the EU started to improve after 2012,⁷ with values of 10.2% in 2014, 9.4% in 2015 and 8.6% in 2016 (Eurostat). The bottom of the European recession seemed to have been in 2012, although not all countries have recovered in an equal manner. Yet, by 2014, a consideration of multiple economic indicators shows that the economy has been improving in Europe overall, with a few outliers still struggling economically more than most countries. However, it remains unclear how long the economic downturn needs to last for people to notice and act, or how quickly people’s perceptions can change for the better once economic growth picks up. This is exactly the focus of this article, with the research comparing the link between protest and the economy at three different points in time: before the start of the economic recession (2006), at the beginning of the financial crisis (2008), and well after the economic downturn (2014).

In this context, the link between economic performance and unconventional activism can be reassessed to investigate whether any economic improvement after 2008 can invalidate feelings of deprivation. The combination of a more positive economic outlook and a new perception with regard to personal economic well-being can undercut the possible relationship between a poor economic position and protest action in Europe.

Modeling the economic impact: data and hypotheses

Research on the European great recession has presented a link between the economic downturn and protest activism, mostly employing data from the early 2000s through 2012. As the financial situation in many European nations has slowly improved, with citizens possibly adapting to lower economic standards, it is reasonable to assess whether the impact of the economic crisis and its length still matter 6 years after the economic collapse hit in 2008. This research uses 2006, 2008, and 2014 data from the ESS Rounds 3, 4, and 7 to test whether economic deprivation can still be relevant in explaining unconventional activism.

Hypotheses

The primary goal of the hypotheses in the research is to assess whether the relationship (in particular the direction of the association) between the economy and unconventional activism changes due to the performance of certain economic indicators, including the individual evaluation of personal financial well-being. A specific comparison of the relationship before, during, and after the recession can answer questions about how economic wealth and/or economic decline can lead to increased contentious politics.

If economic deprivation is no longer felt among Europeans and the severity of the economic austerity has softened, an analysis of 2014 data will highlight a positive relationship between better economic performance and protest, supporting the resource-based theory of unconventionality. At the same time, if a change toward deprivation-led protest has taken place, economic variables will still show a negative type of relationship with protest measures. In the case of country-level variables such as GDP per capita and social protection spending, the analysis can show that a decline in economic performance and social spending is associated with a parallel increase in contentious political activity. In the same context, higher levels of unemployment and economic inequality will be tied to more protest action. In regard to subjective economic interpretations, citizens with a lower approval for the economy and a worse perception of income satisfaction will be more likely to engage in contentious action. For this reason, the first hypotheses in the research are:

H _1a

Higher unemployment and economic inequality as well as lower social protection spending and GDP per capita lead to more protest action.

H _1b

Lower satisfaction with the economy and a worse perception of the household income lead to more protest action.

However, if generally it takes a while for individuals to understand the reality of the economic situation they find themselves in, it is also likely that it will take governments sometime before they can convince their own citizens that the economy is doing better. Consumer confidence is an important part of the economy, and unless citizens can believe the economy is improving, their level of contention will not decline. People’s opinions take longer to change and the impact of the economic crisis may linger, even when the economy is showing improvements. If this is the case, citizens will still feel deprived, and their grievances will drive them to protest, despite countries recording better economic performance values. Therefore, the corresponding second hypothesis in the study is:

H ₂

Subjective economic measures are more likely to still be significant in predicting protest than objective economic indicators 6 years after the crisis.

In the case where economic grievances are still salient to most people, relative deprivation theory can still be useful in explaining protest in Europe, long after the crisis. In detail, a lower level of satisfaction for the economy and a worse perception of income adequacy for the household will lead to more protest activism as citizens still have grievances, if their own perception of the economy is worse. Improved objective economic measures may support the resource-based theory of protest activism, whereas subjective economic evaluations will still be associated with the economic deprivation explanation.

In light of a more long-term view of important predictors of protest, the focus on the economy might reveal that overall economic variables (both objective and subjective) may not be the most influential in explaining contentious activism. In this situation, previous research has underlined the enduring importance of sociodemographic characteristics to explaining protest. That is, a person’s age, gender, education, or political ideological position is the best indicator of the propensity to engage in protest—and this is true both through varying time periods as well as across countries. For this reason, the third hypothesis in this study is:

H ₃

Sociodemographic predictors (gender, age, education, and ideological position) remain statistically significant before, during, and after the economic recession.

Variables and data⁸

The dependent variable in the research is an ordinal measure of protest that has been created using responses from EU individuals⁹ in regard to signing a petition, taking part in a lawful demonstration and boycotting certain products, all within the previous 12 months. Respondents who stated that they signed a petition or boycotted a product received one point for each action, whereas individuals who participated in a lawful demonstration received two points, as it is a more demanding type of action (time, exposure, possible consequences, risks). The final index is a scale¹⁰ (protest index) for each respondent in the sample ranging from 0 (no protest) to 4 (high protest level).¹¹

The EU-level distribution of the protest index values for individuals in the sample for 2006, 2008, and 2014 reveals that overall protest has indeed increased through time and remains higher in the European Union as a whole in 2014 in comparison with 2006 (Table 1). A lower percentage of people declared no protest activity at all in 2014 than 6 years before: a decline of 7.6%. The corresponding EU percentage of individuals engaged in high protest action in 2014 is almost twice as high as in 2006. Yet, the severe disparity in levels of unconventionality at the national level within the EU sample is significant (as also pointed out more in detail in Table 2). For instance, a national breakdown of the protest index values (see ‘Appendix 2’) confirms that there is a relevant gap in level of commitment to protest activism. Before the crisis (2006), Bulgaria ranked first in level of apathy in regard to unconventional action: 92.8% of its national sample of respondents declared no protest action within the previous 12 months. At the same time, the corresponding level of protest apathy in Sweden was 43.1%, the smallest value for 2006. The particular situation of the Eastern European countries deserves separate study. As a quick assessment, the lowest levels of high protest across the three ESS waves presented in Table 2 (0.0% for Romania (2006), 0.1% (2008), and 0.2% (2014) for Lithuania) suggest that strong contentious activism in the Eastern European countries is nonexistent.

Table 1 Protest index distribution: EU-level (%).

Table 2 Unconventional political activism: national level (%) and protest index.

Besides the specific objective and subjective economic variables discussed above, the list of independent predictors in the analysis includes sociodemographic factors, measures of political exposure, and personal satisfaction with the political system.

Research on protest has often showed a link between age and unconventionality, with older individuals less likely to engage in protest activity (Schussman and Soule 2005; Caren et al. 2011; Melo and Stockemer 2014). Younger citizens have a higher propensity to choose confrontational actions, likely because they are less inclined to contemplate possible negative consequences for their involvement in contentious politics. Studies on gender have consistently demonstrated that women are less likely to choose protest (Rucht 2007), in particular, participation in a street demonstration (Gallego 2007). When the activity considered is less risky—petition or boycott—the gender gap is less severe (Marien et al. 2010; Caren et al. 2011) or even reversed (Vassallo and Ding 2016), confirming discussions on the diminished relevance of gender to predict protest (Van Aelst and Walgrave 2001; Schussman and Soule 2005). Education has remained a very influential predictor of protest activism over time: people with more education are associated with protest (Schussman and Soule 2005; Dalton et al. 2010; Dalton 2014). Personal resources (knowledge and information) have a significant impact in the choice to become unconventionally active. Citizens on the left of the political spectrum are also more likely to mobilize and embrace protest, especially during the great recession in Europe (Rüdig and Karyotis 2014; Torcal et al. 2016). Contestation and confrontation against the regime remain important elements of political action in the leftist ideology, next to voting.

In regard to the political exposure context, information about politics (TV news), level of political interest, proximity to a political party, and voting are all useful predictors of a person’s propensity to get involved unconventionally (Barnes and Kaase 1979; Marien et al. 2010; Dalton 2014). More knowledge of politics is associated with more protest, because citizens feel better prepared for forms of political participation other than voting. Lower levels of trust in politicians are equally linked to more confrontational activism as people are not confident politicians can represent their interests. The interpretation of the political system retains its relevance in the study of mobilization. People’s perceptions of their government and democracy at large impact their choice to become political active, both conventionally and unconventionally. The relationship between satisfaction with the government or democracy and the state of the economy is salient in predicting protest. Citizens with a higher level of satisfaction for their government or the functioning of democracy are less prone to choose protest, because they have no strong grievance with regard to the political system. Consequently, they do not feel the need to intervene politically (Norris et al. 2006).

In the end, besides the economy, the models include all the above possible predictors as control variables in the explanation of high protest action in Europe.

Findings and discussion

Levels of protest after the crisis

Since the global recession protest activism in Europe has increased unevenly across countries. Table 2 presents national-level percentages of individuals who stated in 2006, 2008, and 2014 that they signed a petition, participated in a lawful demonstration, or boycotted a certain product. Six years after the beginning of the crisis, the EU average reveals a good surge in citizens involved in petitions (up to 25.3%) and boycotts (up to 19.6%), but a much smaller increase in people participating in street demonstrations (up to 7.3%).¹² If the economy was linked to confrontational activism, it affected decisions to sign petitions or boycott products with stronger long-term effects. Additionally, the EU average protest index is about 30% higher 6 years after the recession (0.59 in 2014), which supports claims of an overall stronger mobilization in favor of unconventionality among European citizens, as also suggested by the findings in Table 1.

In this context of strong contentious politics, most countries in the EU sample recorded a significant increase in unconventionality. The protest index provides the amount of change at the national level from before the crisis (2006) to 6 years after the recession (2014): Three out of the possible 16 countries in the comparison registered a decline in their protest index score. At the same time, 13 out of the same 16 countries showed an increase, with nations like Sweden (+ 0.28), Germany (+ 0.28), and Spain (+ 0.26) at the top of the protest escalation. Data in Table 2 confirm an increasing trend at the EU level, across the three measures of protest employed. Similar results can be seen for most countries in the sample over the 8-year period.¹³

In a comparative analysis for the 2008–2014 period, in particular, Spain recorded the highest percentage of people participating in demonstrations (23.2%, higher than in 2008). Sweden came in first for both petitions and boycotts with, respectively, 43.6% of individuals (down from 47.2% in 2008) and 47.5% of respondents (a meaningful increase from 37.3% in 2008). Moreover, Sweden still recorded the fourth highest percentage of demonstration attendance in 2014 (11.0%, almost double its 2008 value of 6.4%). Apparently, the Swedes master unconventional involvement.

With regard to the ranking of countries in the sample, higher protest activity at the national level in 2014 confirmed the same top performers in confrontational politics for 2008, despite different levels of economic downturn and austerity for the individual countries. For instance, although Ireland was more severely affected by the global recession than Germany, its increased protest activism did not overtake the German position in the group. The ranking of the most active protesters remains virtually unchanged. Six years after the crisis, protest engagement is still noticeably higher than at the beginning of the crisis, despite the improved financial situation and the possible lack of economic grievances. If a poor economy is associated with contentious politics, its impact seems to be long term rather than short term only. Subjective interpretations of a poor economic performance can have a lasting impact on citizens’ choices to protest. Once societies embrace protest more convincingly, people include confrontational actions into their repertoire of activism more often.

Understanding the role of economic variables

To better understand how different predictors can explain protest, the next step in the analysis employs a multilevel logistic regression, with an ordinal outcome (Heck et al. 2012), for each of the selected waves (2006, 2008 and 2014). The fixed effect component is the individual level [Model 1 (M1) in Table 3), whereas the random effect component is the country level¹⁴ (Model 2 (M2) in Table 3].

Table 3 Multilevel ordinal logistic regressions: protest index 2006, 2008, 2014

(a) Pre-recession (2006)

Two years before the recession, the relationship between the economy and protest was not straightforward. Findings from the regression models in Table 3 highlight a European society where a higher satisfaction with the economy is associated with more protest, which supports the resources theory approach to the study of unconventionality. People in a well-off economy engage in contentious politics due to the available resources. However, individuals who were personally unemployed for at least 3 months did participate more through protest actions—a deprivation-led activism, especially among younger participants (Hooghe 2012). In a parallel analysis, respondents with income problems were less likely to choose protest to participate politically: a confirmation that economic deprivation was not a useful enough theory to explain confrontational activism before the economic collapse. This was the general assessment in previous scholarship over few decades.

For the country-level model, the only significant economic variable is GDP per capita, which confirms once more a resources theory-driven understanding of unconventionality: a higher GDP for a country leads to higher protest in society. Overall, the economy mattered in 2006 to understand protest activism, but the main predictor was people’s subjective interpretation of their own financial well-being, supporting a positive link between a better economy and more protest.

An analysis of sociodemographic variables presents a clearer picture. In both individual and country-level models, all predictors related to the sociodemographic group are strongly statistically significant. Women were still less likely to be involved in high protest action, as well as older individuals, who are more prone to choose conventional forms of political participation, since they are less demanding and risky. People who positioned themselves on the right of the political ideology scale are also less likely to embrace high protest as a form of political expression. In contrast, citizens with more education sided with strong protest for their political activism. In all, sociodemographic factors confirm the theoretical expectations from the literature for understanding the protest choice in EU societies.

Finally, the political exposure components of the models suggest that apathetic individuals do not select protest. People with a lower political interest level and who have not voted at the last election are less likely to engage in strong protest. The same applies to respondents who watched more political news on TV, possibly because of the type of passive exposure to politics. In a context where political apathy does not lead individuals to strong protest, citizens who instead feel close to a political party are associated with more unconventional action.

Another group of variables with interesting findings concerns the trust level in politicians as well as the satisfaction with the national government and how democracy works. Respondents showed that they can separate their positions toward the government in power from the overall functioning of the democratic system. Individuals who trust politicians more, or are satisfied with their government, do not embrace protest. Yet, if they are satisfied with the way democracy functions in their country, they are still prone to engage in high protest activism, because citizens feel safe enough to choose confrontational actions and demonstrate a broader repertory of political behavior. The lack of grievances about the functioning of the democratic system is not enough to mute protest activity.

(b) The beginning of the recession (2008)

Findings from the multilevel regression for 2008 reveal very similar roles for most of the economic variables in comparison to the 2006 relationships. The direction of the coefficients for the personal economic perception predictors is unchanged. Additionally, two of the country-level economic variables are significant in M2 from Table 3: GDP per capita and national-level adult unemployment rate. In this case, the coefficient for the former confirms a positive relationship between the economy and protest, whereas the latter introduces evidence of a link between unemployment and contentious activism, a deprivation-oriented explanation of confrontational involvement. At the start of the economic crisis, the two subjective economic variables still point to a resource-based understanding of strong unconventionality. In contrast, both unemployment predictors (the national-level rate and the personal past unemployment experience) suggest economic grievances at the foundation of protest engagement. The economy still mattered in 2008, but various variables presented a different relationship with confrontational activism.

Consistency and statistical significance are instead features of the sociodemographic group of predictors. All variables behave in the same way as in the 2006 models, and they remain very influential explanatory factors of strong protest across Europe, even as the economy is becoming a more negatively dominant presence in the life of many citizens. Equally stable is the interpretation of the role of political engagement variables. Voting is the only predictor in this group to change significance from the individual-level model to the country-level model, where people who voted in the last elections are more likely to choose protest in 2008.

The last group of variables in the political system satisfaction bloc of the models reaffirms the 2006 relationships, with the exception of the satisfaction with democracy predictor that changes sign from the individual-level to the country-level model. Overall, results from the 2008 regressions suggest more similarity than difference in comparison with the previous wave of data analysis.

(c) The aftermath (2014)

If the economic recession in 2008 had a significant impact on the relationship between the economy and protest, an assessment 6 years after the crisis can highlight possible long-term changes to that connection. A comparison of the coefficients and their significance level in M1 and M2 for 2014 (Table 3) reveals that the two personal economic evaluation factors have lost significance. Only the household income perception in the individual-level model is still statistically significant in its relationship to protest, which confirms the resource-based theory approach. For the country-level economic variables, three out of the four are significant in the 2014 model. For the first time, the social protection expenditure variable suggests that a higher level of social protection encourages strong protest, which undermines a deprivation-led interpretation of unconventionality. Together with the GDP predictor, these two variables strengthen the resources interpretation of protest activism as based in resources and wealth.

At the same time, the adult unemployment variable at the national level and the personal unemployment experience confirm the role of economic grievances. Both predictors show that a higher unemployment rate or personal past unemployment leads to more protest for European citizens. Country-level economic variables are more helpful in understanding protest after the crisis than individual level economic predictors, suggesting that national groups do have different reactions to the economic situation they experienced. This result is corroborated by the data on national political activism presented in Table 2. Both Germany and Spain had robust increases in their protest index by 2014 (+ 0.28 and + 0.26 respectively). Yet, Germany continued to experience a persistent improvement in GDP per capita¹⁵ from 2006 to 2014, whereas Spain recorded a sharp decline during the same period.¹⁶

The role and impact of the sociodemographic variables remain very consistent and significant. Age, gender, education, and ideological position are confirmed as essential predictors of protest. This type of continuity in the predicting power of these variables and direction of the relationship is remarkable and very useful to the study of contentious politics in general. Finally, the last bloc of predictors shows only one change for the direction of the vote variable. In 2014, citizens who voted at the last elections embrace strong protest (for both M1 and M2) well after the economic crisis happened. In combination with the remaining variables in the political exposure group and the political system satisfaction section, these findings present a picture of a politically active citizen (political interest, party proximity, voting and democracy satisfaction) embracing both conventional engagement and unconventional engagement. Only individuals with a higher level of trust in politicians and satisfaction in the national government are less likely to consider protest.

Final analysis

As an overview of the general findings, results from 2014 show differences mostly in regard to economic predictors. On the contrary, sociodemographic characteristics of an individual remain important factors to understand why citizens may be more likely to choose disruptive actions when getting involved in politics. This type of conclusion suggests stability in the way citizens relate to the choice of political unconventionality: Nothing has really changed since the beginning of political behavior analysis. Women, older individuals, or less-educated citizens as well as people on the right of the political spectrum are less likely to choose protest. This was the case in the 1970s (Barnes and Kaase 1979,) and it remains so in 2014, after the most severe economic recession in history.

Changes from the findings concern primarily the economic variables, at the individual and national level. Before the global economic crisis, people’s opinions of the economy mattered in understanding protest. Subjective economic interpretations were mostly significant, but not equally so after the crisis. At the same time, national-level economic indicators seem to be more helpful to understand protest activity in 2014, but they are certainly not the only variables. This interpretation also strengthens the design choice to include a multilevel analysis, as each EU country’s economic experience may have specific characteristics.¹⁷

To sum up the hypothesis testing discussion, H_1a can only be partially accepted, because the unemployment rate does lead to more protest in 2014. Yet, social protection spending and GDP per capita have a positive relationship with protest. H_1b is rejected, as in 2014 satisfaction with the economy is not a significant variable and the direction of the coefficient for the household income perception predictor is opposite than the one hypothesized. Additionally, the inequality of income distribution variable is never significant in any of the models, confirming what previously discussed in the misunderstood link between economic inequality and unconventionality (Solt 2015). Unemployment represents the only economic variable at the individual or national level that is significantly associated with contentious activism, as explained by the economic deprivation approach.

In the same context, H₂ is rejected, because personal-level assessments of the economy are less significant after the crisis than before. National-level economic indicators more effectively link protest behavior to the actual performance of the economy, not its interpretation by citizens. Finally, H₃ is accepted, reaffirming the stability of the relationship between certain sociodemographic variables and the choice of protest for individuals across Europe.

Conclusions

In the aftermath of the 2008 economic crisis, European citizens took to the streets when their economic grievances were not heard. The recession contributed to higher levels of political engagement, especially as individuals chose protest to express their anger against national austerity policies. This article has investigated the role of economic variables in predicting unconventional political action in Europe in 2006, 2008 and 2014. Recent studies on the effect of the economic recession on political behavior suggested that since 2008 Europeans have embraced confrontational action due to a feeling of economic deprivation. Lower salaries, higher unemployment, and budget cuts in social protection and services turned citizens into activists. Yet, 6 years after the crisis people’s perception of the economic situation may not be so dire, and is not relevant when explaining why citizens protest.

The analysis of the 2014 data from the European Social Survey supports the claim that the economy still matters to understand higher rates of protest across Europe, but objective and subjective economic variables are salient in different ways. In 2014 national-level economic indicators are more significant in explaining protest, and they actually present evidence in support of a resource-based theory, where a better financial situation encourages citizens to be more politically vocal toward their governments, despite the lack of economic grievances. At the same time, people who reported dissatisfaction with their income situation were less likely to engage in contentious politics. The finding from such type of subjective economic perception measure undermines a deprivation theory-driven explanation. Only unemployment remains an economic predictor for protest, supporting an economic deprivation analysis of the relationship between the economy and political unconventional behavior.

Overall, an analysis of the relationship between protest and economic indicators at three points in time confirms that the link has not changed. In 2006, 2008, and 2014, economic resources were more relevant to explain why citizens become engaged in a form of protest activity. The severity of the recession and its prolonged impact do not seem to have favored the perception of deprivation as a variable leading to confrontational activism. Despite the relevance of unemployment in the prediction of protest, a country’s GDP per capita and social protection expenditures represent useful resources connected to citizen protest.

The persistent statistical significance and direction of the relationship for all the sociodemographic variables (age, gender, education, and ideological position) in the models reaffirm the relevance of these indicators in predicting protest.

Higher levels of unconventionality in Europe after 2008 suggest that people can still be mobilized for contentious action, regardless of the actual or perceived economic situations. With Eurostat data suggesting an economic recovery in the making since 2012, economic resources can become even more significant for future protest waves. The uniform increase in unconventional political behavior across most EU members has obviously many sources. This article has focused on the study of the role of the economy as a driving factor to convince citizens to act unconventionally. After the most serious recession ever experienced in Europe since the 1930s¹⁸ (Lindvall 2014), economic indicators continue to remain reliable predictors of protest, even when deprivation may no longer be felt.

Appendices

Appendix 1

See Table 4.

Table 4 Variable list

Appendix 2

See Table 5.

Table 5 Protest index levels (%) in EU members.