Attitudinal Ambivalence on Redistribution: Causes and Electoral Implications Across Europe

Abstract

While support for redistribution remains high across Europe, voting for left-wing parties, traditionally identified with this agenda, has been under par. Past research explains this puzzle by class-based disagreements about redistributive priorities and by second-dimension attitudes. These explanations, however, assume coherent voter preferences reacting to structural changes. By contrast, I argue that part of the puzzle also lies in attitudinal ambivalence—simultaneous negative and positive evaluations—regarding redistributive policy. Using cross-sectional public opinion and party position data, I find that such ambivalence increases with lower political sophistication, greater value conflict, and weaker economic need. Electorally, it deepens detachment between support for redistribution and left-wing self-identification and increases voting for more economically and culturally right-wing parties. These patterns hold independently of class differences and second-dimension attitudes and replicate stably in earlier data. The findings contribute to ongoing debates about attitude structures and voting patterns and illuminate an additional challenge for economically progressive parties.

Keywords

public opinion attitudes ambivalence voting behavior european politics redistribution political economy political psychology

The electoral decline of the once-dominant economic left, especially social democratic parties, constitutes one of the notable changes in contemporary Western politics (Benedetto et al., 2020). The partisan left in Europe has long been identified with the expansion of income redistribution and social policy (Huber & Stephens, 2001; Korpi, 1983). Greater support for redistribution, moreover, remains a key explanation for left-wing voting (Abou-Chadi & Hix, 2021; Lewis-Beck & Nadeau, 2011; Quinlan & Okolikj, 2020). Yet, curiously, despite the left’s underperformance in recent years, public support for government redistribution has remained stable and high. This pattern is illustrated in Figure 1, which compares the average seat share of economically left-wing parties and the level of general support for redistribution (GSR) across Europe.¹ What explains the gap between attitudes about redistribution and support for economically left-wing parties that promote it?

Figure 1.

Average Seat Share for Economically Left Parties (Panel A) and General Support for Redistribution (Panel B) Across Europe, 1999–2020. The black line plots the sample mean and gray lines mark country-specific trends.

The literature proposes two dominant answers to this puzzle. First, although GSR remains high, class-based disagreements about redistributive priorities hurt the economic left electorally (e.g., Gingrich & Häusermann, 2015; Karreth et al., 2013; Lindvall & Rueda, 2014). Second, the increased salience of so-called second-dimension issues—namely, cultural and social concerns—weakens the traditional link between redistributive preferences and voting (e.g., Bornschier, 2010; Hooghe & Marks, 2018; Kriesi et al., 2008). These explanations, however, assume coherent voter preferences that shift following large-scale sociostructural processes. As such, little attention is given to additional micro-level explanations rooted in incohesive attitudes. Yet, as other works show, many voters display inconsistent and conflicting positions on divisive issues, impacting their policy preferences and voting behavior (Alvarez & Brehm, 2002; Converse, 1964; Feldman & Zaller, 1992; Lavine, 2001). Redistributive attitudes, in particular, comprise multiple dimensions that do not always covary (Attewell, 2021; Cavaillé & Trump, 2015; Roosma et al., 2013). Nevertheless, the nature and electoral consequences of internal attitude conflicts regarding redistribution have not been thoroughly explored.

In this paper, I argue that ambivalent attitudes on redistribution—defined as the conflicted endorsement of both positive and negative evaluations of such policies—play an important part in the disconnect between general redistributive preferences and support for the parties that promote them. Specifically, I examine both the causes and the electoral implications of such ambivalence. Using cross-sectional European Social Survey (ESS) data from 2016, I find that ambivalence regarding redistribution increases among individuals with lower political sophistication, greater internal conflict in core values, and weaker economic need. Contrary to some expectations, ambivalence levels are not influenced by occupational class or by party polarization on these issues.

Then, matching the surveys with Chapel Hill Expert Survey (CHES) data on party positions, I show that greater ambivalence correlates with a deeper incongruence between GSR and left-wing self-identification and predicts voting for more economically and culturally right-wing parties. Importantly, these relationships hold independently of class and second-dimension influences, the key macro-level explanations proposed in the literature. These patterns, moreover, are replicated in data from 2008, before the Eurozone and refugee crises, implying a sustained structural challenge for the partisan left.

The paper makes several contributions to the current debates about redistributive preferences, attitude structures, and electoral politics. First, my findings outline an underdiscussed factor that influences traditional voting patterns across Europe. Whereas the literature emphasizes structural processes, my analysis shows that micro-level attitudinal inconsistencies, rooted in cognitive characteristics, play a meaningful complementary role. Furthermore, it implies different electoral lessons than commonly discussed (e.g., Abou-Chadi & Wagner, 2019): in addition to adjusting their mix of issue positions and social coalitions, left-wing parties may also benefit from better messages, frames, and information that could help less sophisticated and conflicted voters reconcile their redistributive attitudes. Similarly, right-wing parties can attempt to deepen this ambivalence and the value conflict at its basis.

Second, my analysis broadens the discussion about attitudinal ambivalence, which mostly focuses on American politics and preference instability (e.g., Alvarez & Brehm, 2002; Craig and Martinez, 2005; Feldman & Zaller, 1992). My findings show that ambivalent attitude structures, especially regarding a central issue such as redistribution, are also prevalent across Europe and have strong relationships with partisan identification and voting patterns. As such, they call for more cautious assumptions regarding the internal consistency and reasoning of voter attitudes. The findings further underscore the need for more comparative research on the links between such structures and electoral behavior in other contexts and issue domains.

Third, the paper also contributes to the growing debate about the multidimensionality of redistributive attitudes (e.g., Attewell, 2021; Cavaillé & Trump, 2015; Roosma et al., 2013). This body of work shows that general preferences for redistribution can be disaggregated into multiple considerations that can operate separately. However, these dimensions are typically analyzed independently and additively. My analysis demonstrates that they can also form complex attitude structures with internal tensions and important behavioral and electoral implications.

The paper proceeds as follows. Next, I review the existing discussions about the left’s electoral decline, ambivalent attitude structures, and the multidimensional nature of redistributive preferences. Building on these works, I define ambivalence on redistribution and outline several hypotheses about its causes and electoral implications. I then discuss my data and variables and present my findings. I conclude with several takeaways and avenues for further research.

Theoretical Background

The Declining Left-Wing Vote

The electoral underperformance of left-wing parties is commonly explained by two types of arguments. The first focuses on internal disagreements on redistributive priorities within the left’s electorate. Since the 1980s, structural changes in the labor market split the left’s voter base into two primary groups: on the one hand, strong workers and middle-class professionals with secure employment, and, on the other, weaker workers with insecure low-wage jobs (Beramendi et al., 2015; Rueda, 2005). While both groups support government redistribution, they differ on its priorities. Stronger and middle-class workers prefer social-investment policies that improve future employment prospects. Lower-class workers, meanwhile, prioritize social-consumption policies that provide immediate economic support (Garritzmann et al., 2018; Gingrich & Häusermann, 2015; Häusermann et al., 2022).

Facing this split, most left-wing parties favored the former group over the latter (Rueda, 2005). Many lower-class voters, in turn, shifted to other parties on the extreme right or left (Gingrich & Häusermann, 2015; Marx, 2014). As a result, the mainstream left experienced a net voter loss without gaining a similarly loyal new base (Karreth et al., 2013; Lindvall & Rueda, 2014). Hence, according to this explanation, class politics remain a central reason for the left’s electoral underperformance (Oesch & Rennwald, 2018).

By contrast, the second type of explanations focuses on the role of non-economic issues. The last few decades marked the emergence of a new political cleavage structured around social and cultural preferences. This division contrasts libertarian, post-materialist, and cosmopolitan attitudes with traditionalist, authoritarian, and nationalist views (Inglehart, 1977; Kitschelt & McGann, 1995). Fueled by the growth of globalization, immigration, and value changes, this axis reflects new concerns about the cultural threat of minorities and immigrants, supranational institutions such as the EU, and the declining sociopolitical status of majority groups (Hooghe & Marks, 2018; Kriesi et al., 2008).

The new sociocultural cleavage has created a two-dimensional ideological space, where most mainstream left parties are located on the economically and culturally progressive sides (Bornschier, 2010; Hix, 1999). This combination has pushed away voters that are conservative on either axis (Gidron, 2022), especially working-class voters with traditionalist-authoritarian tendencies (Lubbers et al., 2002; Norris & Inglehart, 2019). While the latter are unhappy with current inequality levels, they are also suspicious of existing welfare institutions and prefer selective redistribution that excludes undeserving recipients, immigrants, and minorities (Achterberg et al., 2011; Busemeyer et al., 2022; Derks, 2004; Reeskens & Van Oorschot, 2012). These accounts, therefore, underscore the role of non-economic preferences in the left’s decline.

Both arguments present comprehensive findings and are increasingly analyzed in tandem (e.g., Abou-Chadi & Wagner, 2019; Gidron & Hall, 2017). Nevertheless, both also emphasize large-scale sociostructural shifts, assuming that voters intently adjust their behavior by coherent interests and preferences. As such, they overlook a large literature that finds voter attitudes often inconsistent and messy (Bartels, 2003; Converse, 1964). One of the common forms of such complexity, to which I turn next, is attitudinal ambivalence.

Attitudinal Ambivalence: Definition, Implications, and Causes

Attitudinal ambivalence can be defined as “an individual’s endorsement of competing considerations relevant to evaluating an attitude object” (Lavine, 2001, 915). Ambivalence, by this definition, lends similar weight to negative and positive evaluations of a certain issue or political actor. Unlike confusion or indifference, it reflects a genuine internal conflict between sufficiently salient considerations (Thompson et al., 1995; Thornton, 2011; Yoo, 2010). Such ambivalence can appear toward candidates and parties (Basinger & Lavine, 2005; Lavine, 2001), partisan and group identities (Keele & Wolak, 2006; Shufeldt, 2018), and, most relevant for our discussion, issues and public policies (Alvarez & Brehm, 2002; Craig and Martinez, 2005; Feldman & Zaller, 1992).

The literature on attitudinal ambivalence, which predominantly focuses on American politics, outlines several behavioral implications. First, greater ambivalence destabilizes policy preferences and makes them noisier. This is particularly prevalent with politically charged issues in the US, such as race (Alvarez & Brehm, 1997), affirmative action (Steenbergen & Brewer, 2004), abortion (Alvarez & Brehm, 1995; Huckfeldt & Sprague, 2000), gay rights (Craig et al., 2005), environmental protection (Kellstedt et al., 2019), and campaign finance (Grant & Rudolph, 2003). Outside the US, similar patterns are found regarding preferences for European integration (De Vries, 2013) and immigration (Lindstam et al., 2021; Lutz, 2021). Second, ambivalent voters express lower confidence in their views (Meffert et al., 2004), take longer to access them cognitively (Huckfeldt & Sprague, 2000), and are more reactive to outside influences (Lindstam et al., 2021). Third, attitudinal ambivalence may also affect political participation, such as turnout and campaign contributions, albeit with mixed findings across studies and countries (Çakır, 2022; Johnson, 2014; Lavine, 2001; Kentmen-Cin, 2017; Shufeldt, 2018; Yoo, 2010). Less scholarly attention, however, is given to voting choices themselves.

Compared to their outcomes, the root causes of ambivalent attitudes have been studied less extensively. Nevertheless, despite some mixed results, a few explanatory factors do stand out. At the individual level, ambivalence is often linked to political sophistication. Some find that greater knowledge increases ambivalence due to higher exposure to competing facets of an issue or candidate (Rudolph & Popp, 2007; Stoeckel, 2013). Other studies, however, argue that political sophistication lowers ambivalence by better reconciling rivaling considerations (Kellstedt et al., 2019). Others still find little to no consistent influence (Feldman & Zaller, 1992).

Several studies also link attitudinal ambivalence to conflicting core values and identities. Dominant cultural values, such as individualism and egalitarianism, can cause internal conflict with personal or partisan tendencies (Feldman & Zaller, 1992; Gainous, 2008; Keele & Wolak, 2006). Similarly, affect toward social groups that gain or lose from a policy can raise ambivalence involving other core beliefs (Grant & Rudolph, 2003; Lutz, 2021; Stoeckel, 2013).

Finally, attitudinal ambivalence also reacts to the national political environment. For instance, greater elite polarization regarding controversial issues can raise the number and salience of competing arguments to which voters are exposed (Çakır, 2022; Johnson, 2014; Keele & Wolak, 2008; Stoeckel, 2013). Another contributing factor is political and economic crises, which can raise uncertainty, partisan polarization, and new or otherwise dormant concerns about certain issues (De Vries, 2013; Lutz, 2021).

Ambivalence and Attitudes on Redistribution

Despite the salience of redistributive policy in Western politics, ambivalent attitudes on this issue have not been researched extensively. A small handful of studies did analyze this issue in the US. According to some, welfare policies can increase ambivalence when they cause internal conflict between compassion for beneficiaries and core values of individualism and small government (Feldman & Zaller, 1992; Gainous, 2008; Gainous et al., 2010). Others, however, are more skeptical that such ambivalence is prevalent among Americans (Jacoby, 2005; Steenbergen & Brewer, 2004). The US, moreover, stands out in its strong individualist ethos (Hochschild, 1995; Wolak & Peterson, 2020) and fragmented welfare state (Alesina & Glaeser, 2004). Should we expect a similar complexity in Europe?

Recent research implies that we should. Whereas comparative studies of redistributive attitudes typically assume unidimensionality—that is, support or opposition to higher taxes and transfers—recent cross-sectional work reveals multiple underlying dimensions (Roosma et al., 2013). Of these various considerations, two receive particular attention. The first includes attitudes about the size of the welfare state, that is, the extent to which the government takes from the rich and benefits the rest. The second comprises attitudes about the moral and economic implications of redistribution, especially its value or damage for recipients and the economy (Achterberg et al., 2011; Attewell, 2021, 2022; Cavaillé & Trump, 2015).

This body of work, however, has two notable limitations. First, it assumes that the two redistributive dimensions operate independently of one another. Accordingly, it neglects their potential interaction and internal tensions, which can establish meaningful ambivalence. Indeed, there is an inherent discord between beliefs that the government should redistribute income more generously and that such policies damage the economy and reach undeserving recipients (or vice versa: that these policies improve society but are unjustified).

Second, this debate rarely considers the electoral implications of ambivalent views on redistribution. Nevertheless, the long-standing centrality of redistribution in European politics deems such implications quite likely. For example, Attewell (2021) finds that different redistributive dimensions correlate with voting on separate partisan axes: whereas voting for mainstream parties is linked to views about the size of redistribution and recipient deservingness, voting for extreme right or Green parties is correlated only with the latter dimension. However, this analysis, too, models these considerations independently and additively rather than interconnectedly.

Hypotheses

To address these limitations, I build upon past works to hypothesize about the causes and electoral implications of ambivalence regarding redistribution. First, however, what constitutes ambivalence on this issue? General support for government reduction of inequality, as plotted in Figure 1, obscures internal dimensions and conflicting intuitions regarding redistribution. In particular, it does not consider the type, extent, and boundaries of government involvement, redistributive priorities and criteria, or current and expected outcomes (Roosma et al., 2013).

Nevertheless, once broken down, not all multidimensional attitudes on redistribution entail ambivalence. For example, some attitude structures prefer lower income inequality but reject government intervention as the appropriate solution (Achterberg et al., 2011; Derks, 2004). Others support greater government redistribution in principle but criticize the performance of current welfare-state institutions (Roosma et al., 2014). Support for redistributive policies can also combine with preferences for more selective criteria by recipient deservingness or effort (Laenen & Meuleman, 2018; Rossetti et al., 2020). In all these cases, the interaction of different dimensions can remain consistent: one can wish to lower income inequality but prefer non-governmental solutions, disapprove of existing policies, or support different distributional criteria.

Instead, I define ambivalence regarding government redistribution as conflicting attitudes on the two key dimensions underscored in the literature: (1) the preferred scope of redistribution and (2) its expected implications. This combination is inherently tense as it contrasts support for an action and judgment of its merit. By this definition, greater ambivalence reflects simultaneous support (opposition) for greater redistribution by the government but negative (positive) evaluations of its moral and economic outcomes. As such, it taps into the deep tension between attitudes on inequality and the government’s responsibility, on the one hand, and judgments of specific beneficiaries and their deservingness, on the other (Attewell, 2021; Cavaillé & Trump, 2015; Laenen & Meuleman, 2018; Rossetti et al., 2020; Van Oorschot, 2006). This definition, nevertheless, is not the only possible form of ambivalence on redistribution. However, given its centrality and intrinsic conflict, it will be the focus of my analysis from here on.

The Causes of Ambivalence on Redistribution

My first set of hypotheses considers the factors that can increase or decrease attitudinal ambivalence regarding redistribution. As discussed earlier, previous research identifies three types of explanations for attitudinal ambivalence. The first points to political sophistication and knowledge. The literature considers political sophistication a cognitive trait that involves exposure to, retainment, and organization of political information (Delli Carpini & Keeter, 1996; Luskin, 1990; Zaller, 1992). Sophistication levels stem from personal interest, intelligence, and motivation (Luskin, 1990), and, as such, precede and shape attitude structures (Converse, 1964; Luskin, 1987). As noted, prior works disagree about the direction of its influence on attitudinal ambivalence: greater sophistication increases awareness of more competing considerations but may also help to reconcile them better. Hence, I propose two competing hypotheses:

H1a

Greater political sophistication increases ambivalence regarding redistribution.

H1b

Greater political sophistication decreases ambivalence regarding redistribution.

The second common explanation links ambivalence to conflicting core values (Feldman & Zaller, 1992; Gainous, 2008; Keele & Wolak, 2006). Core values provide general principles and guiding dispositions that inform evaluations of concrete policies and outcomes (Feldman, 1988). Two types of opposing values are particularly relevant for redistributive preferences. The first is individualism, which emphasizes self-reliance and personal achievement without outside assistance. The other involves other-regarding values such as egalitarianism and solidarity, which underscore equality and concern for other people’s well-being. An internal conflict between these core values can spill over into policy attitudes and increase their ambivalence. Thus, I suggest the following hypothesis:

Greater internal conflict between individualist and other-regarding values increases ambivalence regarding redistribution.

The third explanation focuses on the national political environment, particularly the influence of polarized elite cues on relevant issues (Çakır, 2022; Johnson, 2014; Keele & Wolak, 2008; Stoeckel, 2013). When party disagreements on a certain issue are deeper, voters are exposed to multiple competing arguments and extremer rhetoric that can weaken their internal convictions. This may be especially relevant for a central political division such as redistribution, establishing the following hypothesis:

Greater party disagreement on redistributive policy increases ambivalence regarding redistribution.

Finally, in the case of redistribution, I propose economic self-interest as a fourth possible explanation missing from the broader debate on ambivalence. Individuals in greater economic need are typically net recipients of welfare policies and support them more strongly and unconditionally (Laenen & Meuleman, 2018; Rueda & Stegmueller, 2019). As such, they should have more coherently positive views of both the need for greater redistribution and its positive social and economic value. By contrast, better-off citizens, who benefit less from redistributive policies, have more room to feel conflicted about different facets and implications. Accordingly, I propose the following hypothesis:

Greater economic need decreases ambivalence regarding redistribution.

The Electoral Implications of Ambivalence on Redistribution

My second set of hypotheses considers the influence of this ambivalence on voting patterns. One of the most consistent findings is that ambivalence on an issue leads to greater instability in related policy preferences. However, voting differs from policy attitudes in at least two aspects. First, unlike varying levels of agreement with an issue or policy, voting requires a singular and unambiguous choice. Second, contrary to closed survey questions on specific issues, redistribution is not the only topic by which people vote. Different parties offer multiple positions on other, and potentially less internally conflicting, issue domains. Indeed, past research suggests that voting is more weakly connected to redistributive attitudes where welfare policies are less visible or mature (Gingrich, 2014; Quinlan & Okolikj, 2020). Hence, rather than voting more erratically, ambivalent voters may elect parties by alternative ideological axes.

This is all the truer given the growing salience of second-dimension issues in Europe. In the current environment, parties that emphasize sociocultural grievances, particularly on the populist right, provide a salient and cognitively stabler alternative for voters who feel ambivalent regarding redistributive policies. As a caveat, ambivalence on redistribution does not inherently strengthen the cultural right, but could also benefit other alternatives. This suggestion, therefore, considers the current European context but might require adjustments when applied to other regions or periods.

By the logic above, greater ambivalence on redistribution should lead to two outcomes. First, it should weaken the link between GSR, when stated in its underspecified form, and self-identification with the left. That is, ideological identification would be either looser or attached to other issues aligned with the right. Hence, I propose the following hypothesis:

Greater ambivalence regarding redistribution increases incongruence between GSR and left-wing self-identification.

Second, ambivalence on redistribution should also lower the odds of voting for parties with economically progressive positions. Furthermore, ambivalent voters could favor parties that express traditional, authoritarian, and nationalist (TAN) positions on cultural and social issues, the most dominant alternative to the economic dimension in recent decades. I therefore suggest the following two hypotheses:

Greater ambivalence regarding redistribution decreases the odds of voting for economically left-wing parties that support greater redistribution.

Greater ambivalence regarding redistribution increases the odds of voting for parties that support traditional, authoritarian, and nationalist positions.

In sum, I hypothesize that ambivalence regarding redistribution should increase with lower or higher political sophistication, greater value conflict, deeper elite polarization, and lower economic need. Such greater ambivalence, in turn, should weaken the congruence between GSR and left-wing self-identification and increase the probability of voting for economically and culturally right-wing parties.

Data and Variables

I test these hypotheses using cross-sectional survey data from ESS Round 8, fielded across Europe in 2016, which includes an extended question battery on redistributive attitudes. These surveys are combined with CHES party position data that cover all present and prospective EU member states (Jolly et al., 2022). The final sample comprises 18 countries: Austria, Belgium, the Czech Republic, Estonia, Finland, France, Germany, Great Britain, Hungary, Ireland, Italy, Lithuania, the Netherlands, Poland, Slovenia, Spain, and Sweden.²

Key Variable: Ambivalence Regarding Redistribution

My primary variable of interest is ambivalence regarding redistribution. As discussed earlier, I define such ambivalence as an internal tension between two dimensions: (1) the preferred range of government redistribution and (2) its perceived moral and economic implications.

The first dimension is measured using a battery of questions gauging support for active government assistance to various needy groups. Specifically, I calculate the principle-component factor of three questions probing whether it is the government’s responsibility to ensure a reasonable standard of living for (1) the elderly, (2) the unemployed, and (3) working parents. Unlike GSR, this construct explicitly notes the breadth and reach of government-led provision. The principle-component factor has an eigenvalue of 1.82.

The second and opposite dimension is measured using several questions about the negative implications of redistributive policies. Specifically, I calculate the principle-component factor of four questions asking whether social benefits and services (1) place too great a strain on the economy, (2) cost businesses too much in taxes and charges, (3) make people lazy, and (4) make people less willing to care for one another. The principle-component factor has an eigenvalue of 2.23.

As anticipated, these dimensions correlate negatively with one another (r = −.22, p < .001). Also expectedly, they exhibit opposite correlations with GSR as measured in Figure 1: preference for greater government responsibility correlates positively with the latter (r = .24, p < .001), whereas negative perceptions of its merits correlate negatively (r = −.12, p < .001). Nevertheless, the correlation coefficients are not very large, indicating that the three measures are not entirely identical or exclusive and leave room for ambivalence.

I measure the level of ambivalence between the two dimensions using an adjusted version of the Griffin formula suggested by Thompson et al. (1995) for this purpose. Consider two attitudinal dimensions, A_i and B_i, that represent person i’s competing evaluations of a policy on a 0–1 scale. The Griffin formula measures their ambivalence level by intensity and closeness. Closeness examines the distance between the two opposite evaluations. Close scores indicate an ambivalent endorsement of both positive and negative dimensions. Closeness, therefore, is calculated as their negative distance in absolute values:

{Closeness}_{i} = - | A_{i} - B_{i} |

Intensity, meanwhile, measures the strength of the two dimensions’ endorsement. This component is necessary to differentiate ambivalent attitudes, which include strong conflicting positions, from indifferent views, which can assign similar noncommittal scores to both. The ESS surveys use bidirectional scales, where both high and low scores signify strong support or dismissal (e.g., “agree strongly” and “disagree strongly”) and the middle reflects neutrality (e.g., “neither agree nor disagree”).³ Accordingly, I measure intensity as the two dimensions’ average absolute distance from the mid-scale point (Ryan, 2017), such that a higher score indicates more extreme values in either direction:

{Intensity}_{i} = \frac{| 0.5 - A_{i} | + | 0.5 - B_{i} |}{2}

Adding intensity and closeness, ambivalence is calculated as follows:

{Ambivalence}_{i} = \frac{| 0.5 - A_{i} | + | 0.5 - B_{i} |}{2} - | A_{i} - B_{i} |

To ensure comparability, I rescaled the two dimensions’ principle-component scores from 0 to 1 prior to the calculation. The final ambivalence scale ranges from −.5 (strong coherence) to .5 (strong ambivalence). Respondents with close but low-intensity scores fall in the middle.

Explanatory Variables: Causes of Ambivalence

My hypotheses propose several possible causes of greater ambivalence: political sophistication, conflicting values, party polarization on redistribution, and economic need. I measure political sophistication using two separate variables: self-attested Interest in Politics and respondents’ highest level of Education.

I measure Value Conflict as similar endorsement levels of both individualist and other-regarding values. Individualist (I) values are calculated as the principal-component factor of placing more importance on (1) being rich and having money and expensive things, (2) showing abilities and being admired, (3) being successful and recognized for these achievements, and (4) getting respect from others. These statements correspond with Achievement and Power in Schwartz’s scheme of human values (Davidov, 2010; Schwartz, 1994) and produce an eigenvalue of 2.24. I gauge other-regarding (OR) values as the principal-component factor of placing more importance on (1) treating people equally and providing equal opportunities, (2) understanding different people, (3) helping people and caring for their well-being, and (4) being loyal and devoted to friends and close people. These dispositions correspond with Universalism and Benevolence in Schwartz’s framework and have an eigenvalue of 2.18. I calculate their closeness as 1 − |I_i − OR_i| after rescaling both from 0 to 1.

I measure Party Polarization on redistribution using party position data from CHES 2014, the closest to ESS 2016. The CHES dataset scores each party’s position on redistribution on a scale from 0 (strongly favors redistribution) to 10 (strongly opposes redistribution). I calculate country-level party polarization as the SD of each country’s mean party scores in this domain. The calculation weighs parties by their seat share to reflect their relative public influence.

Finally, I measure economic need using two separate variables. The first gauges respondents’ Economic Security as the mean score of their objective income (net household income) and subjective perception (personal sense of income sufficiency). The second variable signifies respondents’ occupational Class, which is often linked to unique redistributive needs and priorities (Häusermann et al., 2022; Iversen & Soskice, 2001; Rueda, 2005), using the 5-class scheme suggested by Oesch (2006).⁴

Electoral Variables: The Implications of Ambivalence

Turning to electoral implications, my hypotheses suggest that ambivalence should undermine congruence between GSR and partisan self-identification. As in Figure 1, GSR is gauged as the level of agreement that the government should take active measures to reduce differences in income levels. Partisan self-identification (PID) is measured as respondents’ self-placement on an 11-point scale from “Left” to “Right”. I calculate the level of Incongruence between the two aspects as 1 − |GSR_i − PID_i| after rescaling both from 0 to 1.

I also hypothesize that ambivalence should influence voting choices. To measure the positions of the parties to which respondents voted, I matched the latter’s self-attested vote in the most recent election with CHES data on party positions. For each country, I use the CHES round (2014 or 2019) that includes the relevant election asked about in the 2016 ESS. When the appropriate elections were unavailable in CHES, I used party positions from the closest ones in which they participated. Supplemental Information Section A4 offers additional details on these procedure and exceptions. I use CHES scores on each party’s position in three domains: (1) Redistribution, (2) Aggregate Economic Left-Right Positions, and (3) Aggregate Cultural Positions on CHES’s GAL-TAN (Green-Alternative-Libertarian vs. Traditional-Authoritarian-Nationalist) scale. All three are scored from 0 to 10, where higher values indicate more right-wing/TAN positions.

Control Variables

I include several control variables to account for additional influences on my dependent variables. In all models, I control for basic sociodemographic attributes: gender, age group, religiosity, an indicator for native-born respondents, and level of residential rurality. In the incongruence and voting model, I also control for union membership, which may increase ties to economically left-wing parties.

When estimating electoral implications, I also control for respondent positions on three second-dimension issues. The first is Anti-Immigration attitudes, calculated as the principal-component factor of six questions: (1) how many immigrants of the country’s ethnic majority should be allowed in, (2) how many immigrants different from the country’s ethnic majority should be allowed in, (3) how many immigrants from poorer non-European countries should be allowed in, (4) immigration is bad or good for the country’s economy, (5) the country’s culture is undermined or enriched by immigrants, and (6) immigrants make the country better or worse. The principle-component factor has an eigenvalue of 3.99. Higher values reflect greater animosity toward immigrants.

The second issue domain is Social Conservatism, measured by attitudes about gay rights. I calculate these positions as the principal-component factor of support for three statements: (1) gays and lesbians should be free to live as they wish, (2) expected shame if a close family member was gay or lesbian, and (3) same-sex couples should have equal rights to adopt children. The principle-component factor has an eigenvalue of 2.12. Higher values reflect a greater objection to gay rights.

Finally, the third issue domain is European integration. I measure Euroskepticism using a single question asking whether European unification should go further or has already gone too far. Higher values reflect greater criticism of European integration.

For ease of comparison and interpretation, particularly of compound variables, I recode all dependent and control variables (except party polarization, which is expressed in terms of SD, and ambivalence, which ranges from −.5 to .5) on a scale from 0 to 1. Supplemental Information Section A summarizes the operationalization of all variables and Supplemental Information Section B presents their descriptive statistics.

Findings

Causes of Ambivalence Regarding Redistribution

I begin by testing my hypotheses about the causes of greater ambivalence on redistribution (H1.a through H4). Since my analysis includes a country-level variable, party polarization, I estimate a multilevel linear regression with random intercepts by country.⁵ All analyses include post-stratification and population weights. Table 1 presents the estimation.

Table 1.

The Determinants of Ambivalence Regarding Redistribution.

	β	s.e.
Interest in politics	−.060^***	(.008)
Education	−.057^***	(.012)
Value conflict	.110^***	(.010)
Party polarization	−.018	(.032)
Economic security	.045^***	(.008)
Class: Higher grade service	−.007	(.010)
Class: Lower grade service	.007	(.005)
Class: Small business owner	.026^**	(.009)
Class: Skilled worker	.007	(.006)
Female	.007	(.006)
Age group	.009	(.010)
Religiosity	.053^***	(.010)
Native born	−.022^***	(.004)
Rurality	.034^**	(.011)
Observations	24,251
Log likelihood	9088.179

Standard errors in parentheses, ^*p < .05, ^**p < .01, ^***p < .001. The baseline category for Class is Unskilled Workers. The estimation includes post-stratification and population survey weights.

Starting with the first two hypotheses, I find that greater political sophistication (interest in politics and education) correlates with lower ambivalence. Hence, in the open debate about the role of sophistication, my findings indicate that political interest and education lessen rather than amplify ambivalence, corroborating H1.a and rejecting H1.b.

Consistent with H2, ambivalence on redistribution also increases with a greater internal conflict between individualist and other-regarding core values. As noted earlier, similar findings rely primarily on US data. Despite differences in cultural ethe and welfare institutions, I find that this pattern replicates across Europe, too.

The findings also provide partial support for the influence of economic need (H4). As expected, respondents with greater economic security express more ambivalence regarding redistributive policies. However, occupational classes do not have a meaningful correlation with ambivalence levels, except for small business owners. Hence, ambivalence regarding redistribution is reflective of economic insecurity but not interclass disagreements on redistributive priorities.

In contrast to the individual-level variables, party polarization on redistribution (H3) does not predict ambivalence on this issue. While it has the expected positive coefficient, this measure fails to reach statistical significance at the 95% level.

Some of the sociodemographic variables, too, play a part. Ambivalence regarding redistribution increases with religiosity and rural domicile, both characterizing voters outside the left’s reach in recent years. These findings suggest that greater ambivalence regarding the left’s flagship agenda is another channel sustaining this alienation. Additionally, immigrants express greater ambivalence, perhaps due to weaker familiarity and norms concerning the welfare state.

Three additional tests, detailed in Supplemental Information Sections C1–C3, add robustness to these findings. First, as verification that ambivalence does not reflect lower attention when answering the survey, I find that interview length (standard, logged, or squared) has a null effect. Second, given the scholarly disagreement about sophistication and ambivalence, I tested whether their relationship is nonlinear (e.g., similar values at low and high sophistication levels). To do so, I re-estimated my models while breaking down political interest and education into categorical dummies. This test corroborates a strictly negative relationship similar to the standard model. Third, to consider whether sociotropic concerns about the economy play a larger role than personal economic security, I added measures for individual-level satisfaction with the national economy and for country-level real GDP growth. Both measures exhibit a null effect and do not change my substantive findings.

General Support for Redistribution and Partisan Orientation

Next, I examine my hypotheses on the electoral implications of ambivalence regarding redistribution (H5 through H7). First, to test H5, I estimate whether greater ambivalence leads to a looser congruence between GSR and left-wing self-identification. Since I do not include country-level variables, I estimate this relationship using an OLS regression with country fixed effects and sample weights.⁶ Table 2 presents the results.

Table 2.

The Influence of Ambivalence on Incongruence Between Support for Redistribution and Left-Right Self-Identification.

	(1)		(2)
	Bivariate		With Controls
	β	s.e.	β	s.e.
Ambivalence	.281^***	(.013)	.227^***	(.015)
Anti-immigration			.127^***	(.015)
Euroskepticism			.001	(.010)
Social-conservatism			.106^***	(.014)
Class: Higher grade service			.002	(.010)
Class: Lower grade service			.005	(.009)
Class: Small business owner			.013	(.009)
Class: Skilled worker			−.005	(.008)
Economic security			.077^***	(.012)
Union membership			−.043^***	(.007)
Education			.006	(.010)
Female			.004	(.005)
Age group			−.042^***	(.008)
Religiosity			.091^***	(.011)
Native born			.018	(.010)
Rurality			.019^*	(.009)
Observations	28,063		20,872
R ²	.064		.107

Standard errors in parentheses, *p < .05, **p < .01, ***p < .001. The baseline category for Class is Unskilled Workers. The estimation includes post-stratification and population survey weights.

Model 1 corroborates H5 using a simple bivariate model: greater ambivalence on redistribution increases the incongruence between GSR and left-wing self-identification. Model 2 finds that this relationship holds when controlling for second-dimension attitudes and class, the two primary factors noted in the literature, as well as sociodemographic attributes. The results are also robust to the exclusion of post-communist countries, where the relationship between left-wing support and redistributive attitudes is more complex (Lipsmeyer & Nordstrom, 2003; Tavits & Letki, 2009), as elaborated in Supplemental Information Section C4.

Other covariates influence this outcome as well. Stronger right-wing attitudes on second-dimension issues, particularly anti-immigration and social conservatism, expectedly weaken the connection between redistributive orientation and left-wing self-identification. This incongruence also grows with higher economic security. However, it remains unaffected by occupational class, a null result that holds even if economic security is omitted from the model. This insignificant relationship is particularly noteworthy given past arguments on class and the decline of left-wing voting, inviting additional research into its sources. Finally, the incongruence between redistributive support and partisan orientation increases with younger age, higher religiosity, and rurality.

Party Vote

Hypotheses 6 and 7 posit that ambivalence regarding redistribution should also predict voting choices. To test this argument, Table 3 presents the estimated relationships between ambivalence on redistribution and voting for parties with more right-wing positions on redistribution (Models 1–2), general economic policies (Models 3–4), and cultural and social issues (Models 5–6). All estimations, with and without control variables, employ an OLS regression with country fixed effects and sample weights.

Table 3.

The Influence of Ambivalence on Voting for Parties With More Right-Wing Positions on Redistributive, Economic, and Cultural-Social Issues.

	Redistribution		Economics		GAL-TAN
	(1)	(2)	(3)	(4)	(5)	(6)
Ambivalence	2.557^***	1.892^***	2.539^***	1.908^***	2.295^***	1.086^***
Ambivalence	(.122)	(.139)	(.121)	(.138)	(.132)	(.150)
GSR		−1.448^***		−1.481^***		−.578^***
GSR		(.091)		(.092)		(.100)
Anti-immigration		1.530^***		1.533^***		2.586^***
Anti-immigration		(.139)		(.141)		(.155)
Euroskepticism		.087		.100		.335^***
Euroskepticism		(.087)		(.087)		(.097)
Social-conservatism		−.097		−.100		.801^***
Social-conservatism		(.132)		(.131)		(.150)
Class: Higher grade service		.329^***		.348^***		.123
Class: Higher grade service		(.090)		(.090)		(.098)
Class: Lower grade service		.284^***		.320^***		.132
Class: Lower grade service		(.085)		(.085)		(.094)
Class: Small business owner		.466^***		.458^***		.253^**
Class: Small business owner		(.092)		(.091)		(.098)
Class: Skilled worker		.204^**		.236^**		.143
Class: Skilled worker		(.075)		(.075)		(.081)
Economic security		.570^***		.651^***		.296^*
Economic security		(.121)		(.122)		(.134)
Union membership		−.443^***		−.448^***		−.359^***
Union membership		(.063)		(.064)		(.066)
Education		.088		.093		−.237^*
Education		(.099)		(.100)		(.108)
Female		−.031		−.055		−.157^**
Female		(.047)		(.047)		(.049)
Age group		.191^*		.216^*		.044
Age group		(.085)		(.087)		(.088)
Religiosity		.765^***		.677^***		.852^***
Religiosity		(.106)		(.107)		(.117)
Native born		.299^**		.301^*		.296^*
Native born		(.114)		(.119)		(.137)
Rurality		.298^***		.320^***		.398^***
Rurality		(.081)		(.082)		(.087)
Observations	18,618	14,555	18,618	14,555	18,618	14,555
R ²	.097	.184	.096	.187	.164	.253

Standard errors in parentheses, ^*p < .05, ^**p < .01, ^***p < .001. The baseline category for Class is Unskilled Workers. The estimation includes post-stratification and population survey weights.

Corroborating H6, models 1 through 4 find that ambivalence regarding redistribution has a positive correlation with voting for parties with more rightward redistributive and economic positions. Importantly, these relationships hold when controlling for GSR, demonstrating that ambivalence operates distinctly of such positions. Moreover, all other variables, including anti-immigration attitudes, have smaller coefficient sizes. Other second-dimension preferences, namely, social conservatism and Euroskepticism, do not have a meaningful influence on voting in these domains.

Models 5 and 6 estimate voting for parties with more TAN positions on sociocultural issues. Expectedly, second-dimension attitudes—mainly anti-immigration preferences, and, more weakly, Euroskepticism and social conservatism—play a more prominent role in this domain than in economic issues. Nevertheless, even when holding their influence constant, ambivalence regarding redistribution has a direct and positive influence, as expected by H7.

One potential concern is that ambivalence may primarily influence voters with already strong second-dimension positions, weakening its substantive contribution. To rule this out, I re-estimated the full models in Table 3 while interacting ambivalence with each second-dimension attitude in turn. The results, detailed in Supplemental Information Section C5, reject this concern. The interactions are either null or negative, in which case they support the opposite relationship: ambivalence is most influential among voters with weak, not strong, second-dimension attitudes.

Most control variables in Table 3 act as expected. Voting for redistributively and economically right-wing parties increases with higher-class and self-employed occupations, economic security, age, religiosity, nativity, and rurality. Union members, conversely, tend to vote leftward on these issues. Similarly, voting for socioculturally right-wing parties is more likely among more economically secure, uneducated, religious, native, rural, male, and non-unionized workers. Contrary to the economic domain, GAL-TAN voting does not correlate with occupational class and age.

Stability or Change? A Suggestive Comparison With 2008

Given the partisan left’s ongoing electoral underperformance, an interesting follow-up question is whether ambivalence on redistribution has been actively increasing. Unfortunately, a systematic answer requires unavailable long-term panel data. Nevertheless, as a suggestive answer, I replicate my findings using data from ESS 4 (2008), which fielded a near-identical module on redistributive attitudes. While only 8 years earlier, these data precede several large-scale political and electoral shocks, including the Eurozone crisis, refugee waves, and the growing legitimacy of populist parties.

To replicate my analysis, I recreated all my variables in the 2008 dataset. Two small differences are noteworthy. First, the sample is slightly different than in 2016: Austria, Italy, and Lithuania are not part of ESS 4, whereas Denmark, Greece, Croatia, Latvia, and Romania are now included. Accordingly, the 2008 sample has 20 countries compared to 18 in 2016. Second, the battery of questions on gay rights, used to gauge social conservatism, is missing in 2008. In its place, I use the single question that reappears, gauging whether gays and lesbians should be free to live their lives as they wish. As before, I match respondents’ votes with CHES data for the relevant election, with exceptions listed in Supplemental Informtion Section A4. Country-level party polarization on redistribution is calculated using data from CHES 2006 or, if missing, 2010 or 2014, in this order.

Figure 2 compares the smoothed distributions of ambivalence on redistribution in 2008 and 2016. Quite tellingly, the distributions are nearly identical. A first glance, therefore, indicates that ambivalence on redistribution is strongly stable over the relevant decade.

Figure 2.

Gaussian Kernel Density Estimations of Ambivalence Regarding Redistribution in 2008 and 2016. Bandwidth selection is based on the minimization of mean integrated squared error (MISE).

The stability is apparent in the causes of ambivalence. Figure 3 compares the coefficients from Model 2 in Table 1 with an identical estimation using the 2008 data.⁷ The results are replicated quite closely, implying that ambivalence has stable roots: lower sophistication, greater value conflict, a greater sense of economic security, and certain sociodemographic attributes.

Figure 3.

The determinants of ambivalence: 2008 versus 2016.

Figure 4 compares the estimated electoral implications in 2016 and 2008. For brevity, I only plot the coefficients for ambivalence and second-dimension attitudes. The findings show that ambivalence has a near-identical relationship with rightward voting, both in size and in statistical power, prior to the turbulent events of the 2010s. Nevertheless, the various crises of that decade did leave a mark: second-dimension attitudes, particularly anti-immigrant sentiments, increased their correlation with incongruence between GSR and partisanship and with voting for traditional-authoritarian parties.

Figure 4.

The electoral implications of ambivalence: 2008 versus 2016.

Hence, ambivalence on redistribution exhibits stable patterns and adverse electoral implications for economically left parties for at least a decade. Accordingly, the populist right’s increasing popularity throughout the 2010s may be more appropriately attributed to the decade’s economic and immigration crises and the growing salience of second-dimension attitudes. Furthermore, the stability of ambivalence indicates that it is not endogenous to these crises or to changes in second-dimension issues. Whether these factors will deepen ambivalence on redistribution over longer periods remains to be seen.

Conclusion

Voter attitudes can be complex, multifaceted, and often messy. Nevertheless, this perspective has remained at the margins of current discussions on European voting patterns, particularly the noticeable gap between support for redistribution and voting for parties identified with this agenda. Past research points to structural changes that reshaped redistributive and sociocultural priorities while assuming they remain consistent and well-reasoned. This framework, however, overlooks additional micro-level factors rooted in internal conflicts and inconsistencies within voter attitudes.

This paper presents one such explanation. Specifically, I explore the nature and role of attitudinal ambivalence regarding redistribution, a central policy domain in European politics. Such ambivalence reflects conflicting views on the desired levels of government redistribution and its perceived socioeconomic implications. Using cross-sectional public opinion and party position data, I show that greater ambivalence increases the disconnect between GSR and left-wing self-identification, lowers the inclination to vote for economically progressive parties, and increases support for socioculturally right-wing parties. These relationships hold regardless of class and second-dimension preferences, indicating a comparable and independent influence. They are also relatively stable, indicated by their replication before the political, economic, and social upheavals of the 2010s.

My analysis also examines the causes of ambivalence on redistribution, which have not been thoroughly studied from a comparative perspective and in this issue domain. I find that such ambivalence primarily grows from individual-level cognitive attributes, particularly lower political sophistication and greater internal value conflict, as well as the privilege of higher economic security. Occupational class, by contrast, is not associated with ambivalence levels. These patterns help clarify some of the previous findings and ambiguities on the causes of ambivalent attitudes, mostly studied in the US, from a broader comparative perspective.

My analysis is not without limitations, opening new avenues for further research on this phenomenon. First, my findings are observational, an appropriate strategy when outlining a broad phenomenon across multiple countries. Nevertheless, additional experimental research can shed more light on the micro-behavioral mechanisms underlying attitudinal ambivalence on redistribution, the factors that can increase or decrease it, and their interaction with structural and sociodemographic processes.

Such exploration is particularly important to understand the attitudinal malleability of ambivalent voters when faced with different party messages, frames, and information. Previous research indicates that elite cues can lower ambivalence, pull voters closer to certain positions, and, in the case of redistribution, associate welfare programs with existing biases (Gilens, 1999; Keele & Wolak, 2008; Lindstam et al., 2021). Researching these questions can be particularly important for parties seeking to win over these voters or alienate them from rival camps.

Second, whereas my analysis uses existing question batteries, additional research can expand and diversify the comparative measurement of ambivalence on redistribution. Some studies on American politics, for example, measure ambivalent attitudes using counts of half-open positive and negative statements (Alvarez & Brehm, 1995; Craig et al., 2002, 2005; Feldman & Zaller, 1992) or rank orders of different values (Jacoby, 2005). It would be useful, therefore, to examine whether these or other instruments operate similarly in comparative settings, different issue domains, and the electoral arena.

Third, my findings emphasize the need for more research on complex attitude structures and voting patterns in other policy domains. The literature on ambivalence examines its electoral implications primarily when directed at candidates and parties, not issues. There is more room, therefore, for theories and empirical analyses on the relationships between ambivalence on salient issues and voting behavior, the similarities and differences between these domains, and their interactions with contextual factors.

Finally, while I find stability in ambivalence between 2008 and 2016, more research is needed on its longer-term trend. Is such ambivalence a permanent feature of political reality? Or does it change over time, and if so, why? Moreover, such research can determine more rigorously whether ambivalence levels react to other structural political, economic, and social changes and combine with existing explanations in this vein. These questions require the systematic collection of longitudinal data on complex attitude structures. This paper, I hope, can advance such endeavors.

Supplemental Material

Supplemental Material - Attitudinal Ambivalence on Redistribution: Causes and Electoral Implications Across Europe

Supplemental Material for Attitudinal Ambivalence on Redistribution: Causes and Electoral Implications Across Europe by Alon Yakter in Comparative Political Studies

Footnotes

Acknowledgments

I wish to thank David Attewell, Juliana Chueri, Mallory Compton, Yotam Margalit, Venkat Nadella, and the CPS editors and three anonymous reviewers for their helpful comments and suggestions. I am also grateful to feedback from participants in the 2022 annual meetings of the Midwest Political Science Association (MPSA), European Political Science Association (EPSA), and Council for European Studies (CES).

Declaration of Conflicting Interests

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

Funding

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

ORCID iD

Alon Yakter

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biography

Alon Yakter is a Lecturer (Assistant Professor) in the School of Political Science, Government and International Relations at Tel Aviv University. His research focuses on intergroup relations, identities, attitudes, and conflict from a comparative perspective.

References

Abou-Chadi

Hix

(2021). Brahmin left versus merchant right? Education, class, multiparty competition, and redistribution in Western Europe. British Journal of Sociology, 72(1), 79–92. https://doi.org/10.1111/1468-4446.12834

Abou-Chadi

Wagner

(2019). The electoral appeal of party strategies in postindustrial societies: When can the mainstream left succeed? Journal of Politics, 81(4), 1405–1419. https://doi.org/10.1086/704436

Achterberg

Houtman

Derks

(2011). Two of a kind? An empirical investigation of anti-welfarism and economic egalitarianism. Public Opinion Quarterly, 75(4), 748–760. https://doi.org/10.1093/poq/nfr034

Alesina

Glaeser

E. L.

(2004). Fighting poverty in the US and Europe: A world of difference. Oxford University Press.

Alvarez

R. M.

Brehm

(1995). American ambivalence towards abortion policy: Development of a heteroskedastic probit model of competing values. American Journal of Political Science, 39(4), 1055–1082. https://doi.org/10.2307/2111669

Alvarez

R. M.

Brehm

(1997). Are Americans ambivalent towards racial policies? American Journal of Political Science, 41(2), 345–374. https://doi.org/10.2307/2111768

Alvarez

R. M.

Brehm

(2002). Hard choices, easy answers: Values, information, and American public opinion. Princeton University Press.

Attewell

(2021). Deservingness perceptions, welfare state support and vote choice in Western Europe. West European Politics, 44(3), 611–634. https://doi.org/10.1080/01402382.2020.1715704

Attewell

(2022). Redistribution attitudes and vote choice across the educational divide. European Journal of Political Research, 61(4), 1080–1101. https://doi.org/10.1111/1475-6765.12486

10.

Bartels

L. M.

(2003). Democracy with attitudes. In MacKuen

M. B.

Rabinowitz

(Eds.), Electoral dmocracy. University of Michigan Press.

11.

Basinger

S. J.

Lavine

(2005). Ambivalence, information, and electoral choice. American Political Science Review, 99(2), 169–184. https://doi.org/10.1017/s0003055405051580

12.

Benedetto

Hix

Mastrorocco

(2020). The rise and fall of social democracy, 1918-2017. American Political Science Review, 114(3), 928–939. https://doi.org/10.1017/s0003055420000234

13.

Beramendi

Häusermann

Kitschelt

Kriesi

(Eds.), (2015). The politics of advanced capitalism. Cambridge University Press.

14.

Bornschier

(2010). The new cultural divide and the two-dimensional political space in Western Europe. West European Politics, 33(3), 419–444. https://doi.org/10.1080/01402381003654387

15.

Bryan

M. L.

Jenkins

S. P.

(2016). Multilevel modelling of country effects: A cautionary tale. European Sociological Review, 32(1), 3–22. https://doi.org/10.1093/esr/jcv059

16.

Busemeyer

M. R.

Rathgeb

Sahm

A. H.

(2022). Authoritarian values and the welfare state: The social policy preferences of radical right voters. West European Politics, 45(1), 77–101. https://doi.org/10.1080/01402382.2021.1886497

17.

Çakır

(2022). Does party ambivalence decrease voter turnout? A global analysis. Party Politics, 28(4), 713–726. https://doi.org/10.1177/13540688211002486

18.

Cavaillé

Trump

K.-S.

(2015). The two facets of social policy preferences. Journal of Politics, 77(1), 146–160. https://doi.org/10.1086/678312

19.

Converse

P. E.

(1964). The nature of belief systems in mass publics. In Apter

D. E.

(Ed.), Ideology and discontent. Free Press of Glencoe.

20.

Craig

S. C.

Kane

J. G.

Martinez

M. D.

(2002). Sometimes you feel like a nut, sometimes you don’t: Citizens’ ambivalence about abortion. Political Psychology, 23(2), 285–301. https://doi.org/10.1111/0162-895x.00282

21.

Craig

S. C.

Martinez

M. D.

(Eds.), (2005). Ambivalence, politics, and public policy. Palgrave Macmillan.

22.

Craig

S. C.

Martinez

M. D.

Kane

J. G.

Gainous

(2005). Core values, value conflict, and citizens’ ambivalence about gay rights. Political Research Quarterly, 58(1), 5–17. https://doi.org/10.2307/3595591

23.

Davidov

(2010). Testing for comparability of human values across countries and time with the third round of the European social survey. International Journal of Comparative Sociology, 51(3), 171–191. https://doi.org/10.1177/0020715210363534

24.

Delli Carpini

M. X.

Keeter

(1996). What Americans know about politics and why it matters. Yale University Press.

25.

Derks

(2004). Are the underprivileged really that economically ‘leftist’? Attitudes towards economic redistribution and the welfare state in Flanders. European Journal of Political Research, 43(4), 509–521. https://doi.org/10.1111/j.1475-6765.2004.00163.x

26.

De Vries

C. E.

(2013). Ambivalent Europeans? Public support for European integration in East and West. Government and Opposition, 48(3), 434–461. https://doi.org/10.1017/gov.2013.15

27.

Elff

Heisig

J. P.

Schaeffer

Shikano

(2021). Multilevel analysis with few clusters: Improving likelihood-based methods to provide unbiased estimates and accurate inference. British Journal of Political Science, 51(1), 412–426. https://doi.org/10.1017/s0007123419000097

28.

Feldman

(1988). Structure and consistency in public opinion: The role of core beliefs and values. American Journal of Political Science, 32(2), 416–440. https://doi.org/10.2307/2111130

29.

Feldman

Zaller

(1992). The political culture of ambivalence: Ideological responses to the welfare state. American Journal of Political Science, 36(1), 268–307. https://doi.org/10.2307/2111433

30.

Gainous

(2008). Who’s ambivalent and who’s not? Social welfare ambivalence across ideology. American Politics Research, 36(2), 210–235. https://doi.org/10.1177/1532673x07312584

31.

Gainous

Martinez

M. D.

Craig

S. C.

(2010). The multiple causes of citizen ambivalence: Attitudes about social welfare policy. Journal of Elections, Public Opinion and Parties, 20(3), 335–356. https://doi.org/10.1080/17457289.2010.490717

32.

Garritzmann

J. L.

Busemeyer

M. R.

Neimanns

(2018). Public demand for social investment: New supporting coalitions for welfare state reform in Western Europe? Journal of European Public Policy, 25(6), 844–861. https://doi.org/10.1080/13501763.2017.1401107

33.

Gidron

(2022). Many ways to be right: Cross-pressured voters in Western Europe. British Journal of Political Science, 52(1), 146–161. https://doi.org/10.1017/s0007123420000228

34.

Gidron

Hall

P. A.

(2017). The politics of social status: Economic and cultural roots of the populist right. The British Journal of Sociology, 68(Suppl 1), S58–S84. https://doi.org/10.1111/1468-4446.12319

35.

Gilens

(1999). Why Americans hate welfare: Race, media, and the politics of antipoverty policy. University of Chicago Press.

36.

Gingrich

(2014). Visibility, values, and voters: The informational role of the welfare state. Journal of Politics, 76(2), 565–580. https://doi.org/10.1017/s0022381613001540

37.

Gingrich

Häusermann

(2015). The decline of the working-class vote, the reconfiguration of the welfare support coalition and consequences for the welfare state journal. Journal of European Social Policy, 25(1), 50–75. https://doi.org/10.1177/0958928714556970

38.

Grant

J. T.

Rudolph

T. J.

(2003). Value conflict, group affect, and the issue of campaign finance. American Journal of Political Science, 47(3), 453–469. https://doi.org/10.1111/1540-5907.00033

39.

Häusermann

Pinggera

Ares

Enggist

(2022). Class and social policy in the knowledge economy. European Journal of Political Research, 61(2), 462–484. https://doi.org/10.1111/1475-6765.12463

40.

Hix

(1999). Dimensions and alignments in European union politics: Cognitive constraints and partisan responses. European Journal of Political Research, 35(1), 69–106. https://doi.org/10.1111/1475-6765.00442

41.

Hochschild

J. L.

(1995). Facing up to the American dream: Race, class, and the soul of the nation. Princeton University Press.

42.

Hooghe

Marks

(2018). Cleavage theory meets Europe’s crises: Lipset, Rokkan, and the transnational cleavage. Journal of European Public Policy, 25(1), 109–135. https://doi.org/10.1080/13501763.2017.1310279

43.

Huber

Stephens

J. D.

(2001). Development and crisis of the welfare state: Parties and policies in global markets. University of Chicago Press.

44.

Huckfeldt

Sprague

(2000). Political consequences of inconsistency: The accessibility and stability of abortion attitudes. Political Psychology, 21(1), 57–79. https://doi.org/10.1111/0162-895x.00177

45.

Inglehart

(1977). The silent revolution: Changing values and political styles among Western publics. Princeton University Press.

46.

Iversen

Soskice

(2001). An asset theory of social policy preferences. American Political Science Review, 95(4), 875–893. https://doi.org/10.1017/s0003055400400079

47.

Jacoby

W. G.

(2005). Is it really ambivalence? Public opinion toward government spending. In Craig

S. C.

Martinez

M. D.

(Eds.), Ambivalence and the structure of political opinion. Palgrave Macmillan.

48.

Johnson

A. A.

(2014). Ambivalence, political engagement and context. Political Studies, 62(3), 502–521. https://doi.org/10.1111/1467-9248.12063

49.

Jolly

Bakker

Hooghe

Marks

Polk

Rovny

Steenbergen

Vachudova

M. A.

(2022). Chapel Hill expert survey trend file, 1999–2019. Electoral Studies, 75, 102420. https://doi.org/10.1016/j.electstud.2021.102420

50.

Karreth

Polk

J. T.

Allen

C. S.

(2013). Catchall or catch and release? The electoral consequences of social democratic parties’ march to the middle in Western Europe. Comparative Political Studies, 46(7), 791–822. https://doi.org/10.1177/0010414012463885

51.

Keele

Wolak

(2006). Value conflict and volatility in party identification. British Journal of Political Science, 36(4), 671–690. https://doi.org/10.1017/s0007123406000354

52.

Keele

Wolak

(2008). Contextual sources of ambivalence. Political Psychology, 29(5), 653–673. https://doi.org/10.1111/j.1467-9221.2008.00659.x

53.

Kellstedt

P. M.

Ramirez

M. D.

Vedlitz

Zahran

(2019). Does political sophistication minimize value conflict? Evidence from a heteroskedastic graded IRT model of opinions toward climate change. British Journal of Political Science, 49(4), 1309–1332. https://doi.org/10.1017/s0007123417000369

54.

Kentmen-Cin

(2017). What about ambivalence and indifference? Rethinking the effects of European attitudes on voter turnout in European parliament elections. Journal of Common Market Studies, 55(6), 1343–1359. https://doi.org/10.1111/jcms.12573

55.

Kitschelt

McGann

A. J.

(1995). The radical right in Western Europe: A comparative analysis. University of Michigan Press.

56.

Korpi

(1983). The democratic class struggle. Routledge & Kegan Paul.

57.

Kriesi

Grande

Lachat

Dolezal

Bornschier

Frey

(2008). West European politics in the age of globalization. Cambridge University Press.

58.

Laenen

Meuleman

(2018). Public support for the social rights and social obligations of the unemployed: Two sides of the same coin? International Journal of Social Welfare, 28(4), 454–467. https://doi.org/10.1111/ijsw.12369

59.

Lavine

(2001). The electoral consequences of ambivalence toward presidential candidates. American Journal of Political Science, 45(4), 915–929. https://doi.org/10.2307/2669332

60.

Lewis-Beck

M. S.

Nadeau

(2011). Economic voting theory: Testing new dimensions. Electoral Studies, 30(2), 288–294. https://doi.org/10.1016/j.electstud.2010.09.001

61.

Lindstam

Mader

Schoen

(2021). Conceptions of national identity and ambivalence towards immigration. British Journal of Political Science, 51(1), 93–114. https://doi.org/10.1017/s0007123418000522

62.

Lindvall

Rueda

(2014). The insider-outsider dilemma. British Journal of Political Science, 44(2), 460–475. https://doi.org/10.1017/s0007123412000804

63.

Lipsmeyer

C. S.

Nordstrom

(2003). East versus West: Comparing political attitudes and welfare preferences across European Societies. Journal of European Public Policy, 10(3), 339–364. https://doi.org/10.1080/1350176032000085342

64.

Lubbers

Gijsberts

Scheepers

(2002). Extreme right-wing voting in Western Europe. European Journal of Political Research, 41(3), 345–378. https://doi.org/10.1111/1475-6765.00015

65.

Luskin

R. C.

(1987). Measuring political sophistication. American Journal of Political Science, 31(4), 856–899. https://doi.org/10.2307/2111227

66.

Luskin

R. C.

(1990). Explaining political sophistication. Political Behavior, 12(4), 331–361. https://doi.org/10.1007/bf00992793

67.

Lutz

(2021). Loved and feared: Citizens’ ambivalence towards free movement in the European union. Journal of European Public Policy, 28(2), 268–288. https://doi.org/10.1080/13501763.2020.1720782

68.

Marx

(2014). Labour market risks and political preferences: The case of temporary employment. European Journal of Political Research, 53(1), 136–159. https://doi.org/10.1111/1475-6765.12027

69.

Meffert

M. F.

Guge

Lodge

(2004). Good, bad, and ambivalent: The consequences of multidimensional political attitudes. In Saris

W. E.

Sniderman

P. M.

(Eds.), Studies in public opinion: Attitudes, nonattitudes, measurement error, and change. Princeton University Press.

70.

Norris

Inglehart

(2019). Cultural backlash: Trump, Brexit, and authoritarian populism. Cambridge University Press.

71.

Oesch

(2006). Coming to grips with a changing class structure: An analysis of employment stratification in Britain, Germany, Sweden and Switzerland. International Sociology, 21(2), 263–288. https://doi.org/10.1177/0268580906061379

72.

Oesch

Rennwald

(2018). Electoral competition in Europe’s new tripolar political space: Class voting for the left, centre-right and radical right. European Journal of Political Research, 57(4), 783–807. https://doi.org/10.1111/1475-6765.12259

73.

Quinlan

Okolikj

(2020). Exploring the neglected dimension of the economic vote: A global analysis of the positional economics thesis. European Political Science Review, 12(2), 219–237. https://doi.org/10.1017/s1755773920000028

74.

Reeskens

Van Oorschot

(2012). Disentangling the ‘new liberal dilemma’: On the relation between general welfare redistribution preferences and welfare chauvinism. International Journal of Comparative Sociology, 53(2), 120–139. https://doi.org/10.1177/0020715212451987

75.

Roosma

Gelissen

van Oorschot

(2013). The multidimensionality of welfare state attitudes: A European cross-national study. Social Indicators Research, 113(1), 235–255. https://doi.org/10.1007/s11205-012-0099-4

76.

Roosma

van Oorschot

Gelissen

(2014). The preferred role and perceived performance of the welfare state: European welfare attitudes from a multidimensional perspective. Social Science Research, 44, 200–210. https://doi.org/10.1016/j.ssresearch.2013.12.005

77.

Rossetti

Roosma

Laenen

Abts

(2020). An unconditional basic income? How Dutch citizens justify their opinions about a basic income and work conditionality. Journal of International and Comparative Social Policy, 36(3), 284–300. https://doi.org/10.1017/ics.2020.15

78.

Rudolph

T. J.

Popp

(2007). An information processing theory of ambivalence. Political Psychology, 28(5), 563–585. https://doi.org/10.1111/j.1467-9221.2007.00590.x

79.

Rueda

(2005). Insider-outsider politics in industrialized democracies: The challenge to social democratic parties. American Political Science Review, 99(1), 61–74. https://doi.org/10.1017/s000305540505149x

80.

Rueda

Stegmueller

(2019). Who Wants What? Redistribution preferences in comparative perspective. Cambridge University Press.

81.

Ryan

T. J.

(2017). How do indifferent voters decide? The political importance of implicit attitudes. American Journal of Political Science, 61(4), 892–907. https://doi.org/10.1111/ajps.12307

82.

Schwartz

S. H.

(1994). Are there universal aspects in the structure and contents of human values? Journal of Social Issues, 50(4), 19–45. https://doi.org/10.1111/j.1540-4560.1994.tb01196.x

83.

Shufeldt

(2018). Party–group ambivalence and voter loyalty: Results from three experiments. American Politics Research, 46(1), 132–168. https://doi.org/10.1177/1532673x17705854

84.

Steenbergen

M. R.

Brewer

P. R.

(2004). The not-so-ambivalent public: Policy attitudes in the political culture of ambivalence. In Saris

W. E.

Sniderman

P. M.

(Eds.), Studies in public opinion: Attitudes, nonattitudes, measurement error, and change. Princeton University Press.

85.

Stegmueller

(2013). How many countries for multilevel modeling? A comparison of frequentist and Bayesian approaches. American Journal of Political Science, 57(3), 748–761. https://doi.org/10.1111/ajps.12001

86.

Stoeckel

(2013). Ambivalent or indifferent? Reconsidering the structure of EU public opinion. European Union Politics, 14(1), 23–45. https://doi.org/10.1177/1465116512460736

87.

Tavits

Letki

(2009). When left is right: Party ideology and policy in post-communist Europe. American Political Science Review, 103(4), 555–569. https://doi.org/10.1017/s0003055409990220

88.

Thompson

Zanna

M. P.

Griffin

D. W.

(1995). Let’s not be indifferent about (attitudinal) ambivalence. In Petty

R. E.

Krosnick

J. A.

(Eds.), Attitude strength: Antecedents and consequences. Lawrence Erlbaum Associates, Inc.

89.

Thornton

J. R.

(2011). Ambivalent or indifferent? Examining the validity of an objective measure of partisan ambivalence. Political Psychology, 32(5), 863–884. https://doi.org/10.1111/j.1467-9221.2011.00841.x

90.

Van Oorschot

(2006). Making the difference in social Europe: Deservingness perceptions among citizens of European welfare states. Journal of European Social Policy, 16(1), 23–42. https://doi.org/10.1177/0958928706059829

91.

Wolak

Peterson

D. A.

(2020). The dynamic American dream. American Journal of Political Science, 64(4), 968–981. https://doi.org/10.1111/ajps.12522

92.

Yakter

(2022). Replication data for: Attitudinal ambivalence on redistribution: Causes and electoral implications across europe. Harvard Dataverse. https://doi.org/10.7910/DVN/TXKLXF

93.

Yoo

S.-j.

(2010). Two types of neutrality: Ambivalence versus indifference and political participation. Journal of Politics, 72(1), 163–177. https://doi.org/10.1017/s0022381609990545

94.

Zaller

J. R.

(1992). The nature and origins of mass opinion. Cambridge University Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.29 MB