Opportunity Perceptions and Electoral Choice

Empirical research on party system change in advanced democracies increasingly agrees that focusing on voters' objective and immediate socioeconomic status (for example, current income or employment status) is insufficient for explaining radical versus mainstream voting. Against many of the early hypotheses in this literature, according to which the unemployed, the poor and nonstandard ‘cheap labour’ was the recruiting ground for radical challenger parties (for example, Betz Reference Betz1993; Esping-Andersen Reference Esping-Andersen1999; King and Rueda Reference King and Rueda2008; Lubbers, Gijsberts, and Scheepers Reference Lubbers, Gijsberts and Scheepers2002; Mughan, Bean, and McAllister Reference Mughan, Bean and McAllister2003), we now know that the economic situation of voters needs to be theorized in their temporal and cross-sectional context. Accordingly, a burgeoning literature has adopted a more contextualized perspective on how socioeconomic circumstances might shape electoral choices.

To situate our perspective, we outline three ways in which researchers have so far adapted conceptualizations of socioeconomic status as a driving factor of political behaviour by (1) highlighting the importance of status relative to others, (2) emphasizing dynamic changes in status over time, or (3) focusing on the subjective nature of status perceptions.

First, research on relative deprivation emphasizes that voters compare their economic situation to others in society (Runciman Reference Runciman1966). If such comparisons lead voters to conclude that they are – unjustly in their view – dis-advantaged, this may fuel resentment and receptiveness for political appeals that play on that sentiment (Burgoon et al. Reference Burgoon2019; Kurer Reference Kurer2020; Pettigrew Reference Pettigrew2017). The importance of relative judgments of economic well-being is also evident in ethnographic studies, for instance, of members of rural or declining industrial communities who emphasize the struggles of their own working life while drawing comparisons to the (‘unfairly’) comfortable lives of professionals or public employees in thriving city centres (Cramer Reference Cramer2016; Gest Reference Gest2016; Hochschild Reference Hochschild2016).

Second, some researchers have looked at how dynamic changes in voters' (objective or subjective) socioeconomic status might explain political preferences. Burgoon et al. (Reference Burgoon2019) study ‘positional deprivation’, defined as a situation where the increase in disposable income of an individual is smaller relative to the growth in income of other groups in the same country's income distribution. Kurer (Reference Kurer2020) explores the employment trajectories of threatened routine workers, distinguishing relative and absolute changes in status. Gest, Reny, and Mayer (Reference Gest, Reny and Mayer2017) explicitly ask survey respondents to compare the status of ‘people like them’ today to what they think it was 30 years ago, while a dynamic perspective is also implicit in other studies on subjective social status and nostalgia (Gidron and Hall Reference Gidron and Hall2017; Steenvoorden and Harteveld Reference Steenvoorden and Harteveld2018). Again, ethnographic work has captured perceptions of relative status loss using qualitative methods. Hochschild's (Reference Hochschild2016) metaphor of a social escalator on which members of the ‘old’ middle classes feel that they are overtaken by members of ethnic minorities, immigrants, women, or public sector workers vividly illustrates how changes in relative social status might matter to individuals (for Germany, see also Nachtwey Reference Nachtwey2016).

A third related strand of work has more generally emphasized the importance of subjective measures of immediate, relative, or dynamic deprivation (Teney, Lacewell, and Wilde Reference Teney, Lacewell and Wilde2014); or of subjective social status (Engler and Weisstanner Reference Engler and Weisstanner2021; Mutz Reference Mutz2018). Notably, Gidron and Hall (Reference Gidron and Hall2017) show that lower levels of subjective social status are associated with support for the radical right. They trace how the self-perceived status of men without a college education (relative to women) has declined in the past decades due to structural social transformations. Even if the relevant temporal reference points are not always discussed upfront, these concepts inherently include a relational element. Subjective judgments of status are also ever-present in the aforementioned ethnographic studies, focusing on individuals’ interpretations of their economic circumstances.

We learn from these contributions that to understand better how voters' economic situation affects their political behaviour, we need to capture how voters perceive their situation in context. And while this insight represents important progress, many status threats voters invoke when assessing their economic situation refer to the future – for example, work conditions deteriorating, risks of automation, prospects of ‘increasing’ immigration etc. Similarly, radical parties tend to operate with prospective narratives and scenarios of threat.

Consequently, we explore how voters' expectations for their future relate to their electoral preferences. This prospective focus shares similarity with the economic voting literature, which emphasizes the relevance of subjective evaluations of economic circumstances for incumbent party support (Clarke et al. Reference Clarke2004; Duch and Stevenson Reference Duch and Stevenson2008; Lewis-Beck and Stegmaier Reference Lewis-Beck, Stegmaier, Dalton and Klingemann2007). Our approach is complementary in that it adds a more long-term structural perspective on evaluations of the economy. We distinguish between mainstream incumbent, mainstream opposition and radical parties because the contestation of established mainstream parties by radical challengers is at the core of a fundamental realignment of European party systems that cannot be conceptualized in terms of incumbent versus opposition (Hernández and Kriesi Reference Hernández and Kriesi2016).

In theorizing the electoral implications of perceived prospects, we directly build on the argument that ‘aspirational voters’ play a vital role in stabilizing advanced capitalist democracies (Iversen and Soskice Reference Iversen and Soskice2019). According to Iversen and Soskice, aspirational voters are voters who are not (yet) direct beneficiaries of democratic capitalism but who believe that they or their children have good prospects of someday joining the ranks of skilled workers employed in knowledge-intensive sectors. Aspirational voters believe they will benefit from structural change in the longer run. In the authors' account, this segment of voters stands for political continuity and ‘reformism’, thereby enabling the growth that eventually fuels demand for skilled and educated labour.

Conversely, Iversen and Soskice argue that anti-system appeals fall on fertile ground among those who feel excluded from the metaphorical escalator of progress. In this vein, the authors also point to the rapid spread of populism in the aftermath of the Great Recession. This observation resonates well with the findings of Hernández and Kriesi (Reference Hernández and Kriesi2016): With the economic crisis undermining confidence in incumbent parties, support for opposition forces is particularly likely to take the form of radical party support when voters cease to see a place for themselves and their children in an ever more knowledge-intensive, globalized society. Shorter-term, economic voting dynamics can hence accelerate the rise of radical parties.

We build on the conceptual contribution of Iversen and Soskice but differentiate more explicitly between subjective prospects and current socioeconomic status. While a positive evaluation of future opportunities may function as a buffer against supporting radical parties even among lower-status voters, a negative view of future opportunities may also drive well-off voters towards the fringes of the political spectrum. There is no reason to think that economically secure individuals should be immune from losing confidence in the capacity of the political-economic order to provide prosperity for themselves and their children. Differentiating prospects and status develops our theoretical expectations beyond a median voter argument towards conceptualizing different groups, their preferences and – importantly – their relative weight in society. It also extends a classical economic voting perspective by identifying two groups (mainstream opposition voters and radical opposition voters) who might be equally prone to punish incumbent parties for short-term economic downturns but who differ in whether their rejection of the incumbent extends to a more fundamental contestation of the political-economic system as a whole.

Theoretical expectations

Lacking confidence in the system's capacity to provide opportunities likely results in opposition against parties representing this system. However, beyond this hypothesized general anti-system effect, we are interested in whether and how the impact of voters' opportunity perceptions on party preferences varies with their immediate economic circumstances. Can confidence in future prospects compensate for a less-than-ideal present? Even lower-status voters in today's knowledge-based economy might support the current economic framework based on a positive evaluation of social mobility prospects – for themselves or the next generation. At the same time, even someone with a relatively high economic status today need not necessarily believe in a bright future. This reasoning (which extends to Iversen and Soskice Reference Iversen and Soskice2019) results in a two-by-two table in which we distinguish voters with positive and negative expectations regarding their opportunities on the one hand and high versus low current income on the other (we will consider finer breakdowns later on in the empirics). Table 1 summarizes our main expectations regarding distinctive electoral preference hierarchies for each quadrant.Footnote ¹ Our expectations are fairly straightforward regarding the two unconflicted quadrants where economic circumstances and expectations about opportunities reinforce each other. First, high-income voters with a positive evaluation of future opportunities are beneficiaries of today's political-economic set-up, and they do not expect this to change. On average, these voters (whom we label ‘comfortable voters’) have every reason to support mainstream parties, particularly incumbent parties, because these have built and defended a system that delivers for them.

Figure 1. Distributions by opportunity type (vertical dotted line indicates mean value).

Diametrically opposed are low-income voters with negative evaluations of their future opportunities, whom we expect to be prone to opposition support (or to abstention). Even more so than mainstream opposition voices, we expect the appeals of radical parties to resonate particularly with this group. Voters in this quadrant (whom we call ‘burdened voters’) are not faring well in the modern knowledge economy and they perceive their long-term perspectives to be as bleak. Hence, we expect them to support radical parties over-proportionally, followed by mainstream opposition parties.

The remaining two quadrants are interesting because of how perceptions of opportunities might counterbalance immediate economic conditions. This makes them potentially pivotal for coalitional electoral dynamics sustaining advanced capitalist democracies. For these two quadrants, we have no strong expectations about incumbent versus mainstream opposition support per se. In our view, cross-pressures stemming from current status as opposed to future opportunity perceptions, ideological considerations, or short-term economic voting might turn these groups in favour of either incumbent or mainstream opposition parties. However, these two conflicted groups' diverging confidence in opportunities delivered by the knowledge society results in clear expectations about their propensity to support radical parties over-proportionally.

Lower-income voters who are nevertheless optimistic about opportunities correspond to Iversen and Soskice's aspirational voters: they have not (yet) fared particularly well, but they are confident that the existing political framework generates opportunities for them and their children. Such voters' bet on the knowledge economy should be associated with general support for parties that defend this order, that is, both mainstream incumbent and opposition parties while working as a buffer against radical parties. Radical disruption to the ‘rules of the game’ may obstruct already internalized visions for upward mobility.

Lastly, we are left with higher-income voters who evaluate future prospects negatively (our ‘apprehensive voters’): these are ‘winners’ of the prevailing political-economic order, but they nevertheless worry about what the future might hold for them or their children. These worries may be rationally founded (think of skilled but relatively lower-educated routine workers in declining industries) or more culturally based on perceived threats to lifestyles and norms (for example, with regard to a male breadwinner role model, belonging to the ethnic majority group, etc.) We expect the perceived threat of economic and status decline to predispose these voters towards radical parties rather than established mainstream parties (whether in government or opposition).

In terms of the mechanisms translating ‘apprehensive voters' negative evaluations of the future into support for radical parties, we can build on the rich and growing literature on how subjective and relational concepts of well-being contribute to explaining voting behaviour. Precisely because we define apprehensive voters as being well-off in terms of immediate material conditions (notably having relatively higher income), we would expect their concerns to take a more diffuse form, spreading to (and easily directed at) the more encompassing ways in which the economy and society are changing. Rather than worrying about immediate risks that might be mitigated within the existing system of advanced democratic capitalism – for example, with social insurance and transfers (cf. Gingrich Reference Gingrich2019) – these voters experience a more fundamental sense of dislocation as occupational status hierarchies, family models, gender roles, and other aspects of social life transform. This should translate into a more profound and encompassing sense of discontent with the direction in which established political forces are taking society. We expect such disaffected but well-off voters to see mainstream parties (incumbents and opposition, left and right) as broadly complicit in the transformation towards a knowledge-based modern society, pushing them towards the radical (and often anti-eliteFootnote ²) fringes of the party spectrum in general.

Data and operationalization

We rely on original survey data collected in eight West European countries (1,500 respondents each): Denmark, Sweden, Germany, the Netherlands, Ireland, the UK, Italy, and Spain (http://welfarepriorities.eu/). This country selection not only maximizes variation in terms of welfare regimes and electoral institutions but, importantly, also encompasses countries with strongly differing structural and cyclical economic performance. In particular, Spain, Italy, and Ireland were on a strong upward macroeconomic trajectory in the years leading up to the survey while facing deep-rooted structural economic challenges, especially in terms of labour market performance. At the same time, Sweden, Denmark and Germany exhibit stronger structural performance but overall stability in the immediate period leading up to the survey year. This variance delivers suggestive evidence to demarcate our perspective from short-term retrospective economic performance as a determinant. The target population was a country's adult population (>18 years), with quotas on age and sex (crossed) and educational attainment. The total sample counts 12,506 completed interviews that were conducted between October and December 2018.

Independent Variables: Perceptions of Opportunities

We asked respondents about perceptions of economic opportunity concerning their own future and their children's (if any). The survey questions asked about both labour market opportunities (‘If you think of your future, how do you rate your personal chances of being in good, stable employment until you will retire?’) and general social opportunities (‘Now think beyond the labor market to your overall quality of life. How do you rate your personal chances of having a safe, fulfilled life over your life course?’) The questions were repeated to query respondents about their assessment of their children's chances on labor markets and life more generally. The precise wording and a correlation matrix of the various opportunity items is provided in Appendix B. Figure 1 shows the distributions of the four opportunity measures. The general skew to the left with mean values above the mid-point suggests that most respondents have rather positive evaluations of economic opportunity for themselves and their children.

Our procedure to test the theoretical expectations summarized in Table 1 is twofold. First, we create a categorical variable that emulates the four groups of voters at the centre of interest (aspirational, apprehensive, burdened and comfortable voters). To do so, we combine a dichotomized measure of perceived opportunities (positive versus negative) and a dichotomous measure of income level (above versus below the country-specific median income). In our main analysis, we rely on social opportunities for respondents themselves as the main indicator of positive/negative prospective evaluations. In additional analyses, we also define these groups on the basis of respondents' perceived labour market opportunities and on the basis of perceived opportunities for respondents' children to probe the robustness of our results and to discuss implications of the varying strength of the relationship depending on the precise operationalization of opportunity perceptions.

Table 1. Theoretical expectations: distinctive electoral preference hierarchies by quadrant

Second, we use the full, continuous information on respondents' income and opportunity perceptions and run regression models that include the interaction of these two variables. Note that, regardless of the level of current socioeconomic status, our framework suggests that positive evaluations of future opportunities should relate to higher mainstream party support (incumbents or opposition). We consider a possible interactive effect in the sense that high socioeconomic status and positive evaluations of future prospects may reinforce each other in how they relate to support for incumbents and mainstream parties more generally. However, our most important expectation is that negative opportunity perceptions relate to radical opposition support at high as well as low levels of socioeconomic status (that is, ‘apprehensive voters’ constitute an important group alongside ‘aspirational voters’).

Socio-demographic control variables

In order to assess how evaluations of opportunities affect electoral preferences at different levels of economic conditions, we must control for the socio-demographic profile of aspirational, apprehensive, burdened and comfortable voters. While we cannot establish the importance of opportunity perceptions without controlling for the composition of these groups, these socio-demographic profiles themselves are of interest. In advancing their original ‘aspirational voter’ argument, Iversen and Soskice (Reference Iversen and Soskice2019) hardly characterize aspirational voters other than saying they want to join the ranks of skilled workers. However, for the stability of advanced democratic capitalism, the question of who occupies our four quadrants is critical.

Figure 2 provides a general sense of each group's defining socio-demographic characteristics. The presented attributes give a qualitative summary of the characteristics of a ‘typical’ representative of the respective quadrant. The presented typical attributes are the result of a large number of descriptive quantitative analyses in which we ‘predict’ belonging to one of the four groups based on a range of important socio-demographics (see Appendix F for details).

Figure 2. Socio-demographic and socioeconomic characteristics of voter groups.

Although our four voter groups are, of course, heterogeneous in all of these respects, specific socio-demographic groups are clearly over-represented in each quadrant. For example, the over-represented among comfortable voters include managers, technicians, and sociocultural specialists. More generally, these are people with non-routine cognitive work. Meanwhile, service and production workers – in other words, people with non-routine manual or routine task profiles – are typical burdened voters. Comfortable voters are also disproportionately male and older (>55), while burdened voters tend to be younger and female. These two quadrants encompass a contrast between direct beneficiaries of the knowledge economy and those groups most clearly excluded from knowledge-intensive sectors of the economy, providing a first validation of how we operationalize our two-by-two table.

Age emerges as a key difference between aspirational and apprehensive voters. Aspirational voters are typically relatively young (18–35), while apprehensive voters are typically middle-aged (36–55). Service workers (and small business owners, but not production workers) or people in non-routine manual work are also well-represented in the aspirational group, as are people with medium levels of education. On the other hand, routine work is typical for apprehensive voters. The contrast between aspirational young service employees and apprehensive middle-aged routine employees bolsters our confidence in the operationalization of these quadrants.

Electoral Coalitions in the Knowledge Economy

We have suggested a conceptual framework in which beneficiaries of post-industrial knowledge economies are understood not only on their objective socioeconomic standing but also on their subjective perception of future economic opportunities in a changing economy and society. As a direct implication of our analysis, the prospects of structural change and political stability depend on each of the four groups' relative size and political relevance. For example, suppose more confident pro-system forces, consisting of comfortable but also aspirational voters, dominate the democratic competition. In that case, modern knowledge economies have a fairly stable backing among the population and will be able to avoid potential disruption from the minority of voters who are and/or feel left behind by economic modernization. Figure 9 shows the relative importance of each of the four constituencies in each country under study.

Figure 9. Cross-national variation in relative importance of groups.

On average, the more confident coalition indeed clearly outweighs the pessimistic coalition. At the same time, however, the plot reveals striking variations in relative group sizes. Continental and Northern European countries are characterized by a large share of voters (60–70 per cent) who believe in economic opportunity for themselves and their children. However, voters in the European South show a much stronger prevalence of negative perceptions of economic opportunity. In both Italy and Spain, pro-system aspirational and comfortable voters represent a minority coalition. They are outnumbered by burdened and apprehensive voters who are relatively more likely to support radical parties. This pattern across countries is particularly telling given the fact that the Southern European countries, as well as Ireland, were on a striking path of economic upswing and recovery in the years leading up to 2018, the year of our survey. Unemployment fell sharply, and growth was robust after 2015 in these countries. Nevertheless, the share of voters with negative structural, long-term perceptions of opportunities is massive compared to the countries of Northern and Continental Europe. This pattern gives us further confidence that our concept and measure of opportunities is theoretically and empirically distinct from cyclical macro-economic performance but instead reflects a perception of structural performance, which then feeds into an assessment of whether the status quo should be radically challenged in political terms or not.

The relative strength of the four groups in Southern Europe represents a political configuration with much more limited support for creating and sustaining modern knowledge economies. However, this also points towards a potentially self-reinforcing dynamic, whereby less knowledge-based economies are less likely to create and sustain the electoral coalitions that might support pro-system agendas set to further deepen the knowledge economy within the existing framework. What is more, the share of burdened and apprehensive voters in the Southern European countries is so high that they even represent a relevant share of mainstream party electorates (around 50 per cent), whereas they are mostly confined to the more radical party constituencies in the Continental and Northern European countries (Appendix E)Footnote ⁵. Their impact on government formation and policies is even more likely, reinforcing the suggested self-reinforcing dynamics.