Abstract
The degree of economic inequality may lead to different environments where people develop motives and behaviours that lend them higher chances of survival. However, the specific features attributed to an environment with a particular level of economic inequality have received little research attention. In this research, we explored how perceived economic inequality may influence the values inferred as normative in society. Results from three studies, one correlational and two experimental, showed that perceived normative values change according to the degree of perceived economic inequality in a given context: higher levels of perceived economic inequality are related to normative self-enhancement values, whereas lower levels of perceived economic inequality are related to normative self-transcendence values. These results are discussed in terms of how information on economic inequality is used to build a general perception of the normative climate in society and, accordingly, of the values that would best guide behaviours.
Economic inequality has an important impact on individuals (Andersen & Curtis, 2012; Chiavegatto Filho et al., 2013; Layte & Whelan, 2014; Rufrancos et al., 2013; Wilkinson & Pickett, 2006). From a broad perspective, Wilkinson and Pickett (2017) argued that economic inequality creates a specific environment wherein people have to adapt their behaviour strategically. However, the particular features attributed to a context with a specific level of economic inequality have received little attention in the literature. Only a few previous researchers have shown that a higher level of economic inequality provides a context with a masculine, individualist, and competitive normative climate (Moreno-Bella et al., 2019; Sánchez-Rodríguez, Willis, Jetten, & Rodríguez-Bailón, 2019).
In the current paper, we aim to extend the literature on the consequences of economic inequality for social normative climate, by testing whether the level of economic inequality affects perceived normative values. We suggest that level of economic inequality might work as a clue that people use to infer which values others use to guide their behaviour.
Normative Climate of Economic Inequality
In recent years, a growing body of literature has explored the consequences of economic inequality (Elgar & Aitken, 2011; Elgar et al., 2009; Kang & Kang, 2017; Petkanopoulou et al., 2018; Rufrancos et al., 2013; Snowdon, 2010; Sommet et al., 2018). The general rationale has been that economic inequality creates an environment with particular features to which people have to adapt (Wilkinson & Pickett, 2010, 2017). In particular, social strategies based on dominance and competition seem to be more appropriate in a context with higher economic inequality, in which improving or not losing rank becomes very important. Conversely, strategies based on friendship, reciprocity, and sharing are more likely to be more successful in a context with lower economic inequality (Wilkinson & Pickett, 2017). Consistent with these predictions, Nishi et al. (2015) showed that participants in groups with higher levels of economic inequality tend to cooperate less. Moreover, some studies have shown that people living in more unequal societies are less willing to trust others (Elgar & Aitken, 2011; Kawachi et al., 1997; Oishi et al., 2011). Higher economic inequality is also related to a lower willingness to promote the welfare of others, as well as the welfare of the wider community (Paskov & Dewilde, 2012). Loughnan et al. (2011) also found that countries with higher levels of economic inequality tend to show higher levels of self-enhancement bias. These effects of economic inequality are even reflected in how individuals define themselves concerning others (i.e., their self-construal): low (vs. high) economic inequality triggers a more interdependent self-construal (Sánchez-Rodríguez, Willis, & Rodríguez-Bailón, 2019).
Following this approach, economic inequality should also affect the normative climate that individuals infer from a particular level of economic inequality (Rodríguez-Bailón et al., 2020). We define normative climate as the set of features that individuals perceive to be generalized in a social context. Notably, higher levels of economic inequality would provide a normative climate based on dominance and competitiveness. By contrast, lower levels of economic inequality would provide a normative climate in which reciprocity and sharing prevail.
Indeed, previous results have shown that more unequal contexts are perceived as more competitive and less cooperative in comparison with more equal settings (Sánchez-Rodríguez, Willis, Jetten, & Rodríguez-Bailón, 2019; Sommet, Elliot, Jamieson, & Butera, 2019). Davidai (2018) similarly found that when inequality is high, individuals tend to think that social mobility is low. This effect may be because when people perceive an increase in economic inequality, they make more external attributions of economic outcomes. Recently, Sprong et al. (2019) showed that perceived high economic inequality leads people to perceive that society is breaking down—anomie.
More important to our research, people tend to infer how others are as a function of the level of economic inequality. Inhabitants of a society presented as highly (vs. low) unequal were perceived as more independent (vs. interdependent), more involved in exchange relationships (vs. communal relationships), and more prone to pursue their own goals (vs. pursuing group goals; Sánchez-Rodríguez, Willis, Jetten, & Rodríguez-Bailón, 2019); they were are also described using more masculine (vs. feminine) traits (Moreno-Bella et al., 2019). These studies suggest that economic inequality leads people to infer particular features of those living in such society. However, what remains unexplored is whether this also leads them to infer the types of motivation guiding others’ behaviours. In the current research, we address this question by focusing on the effects of economic inequality on the normative values shaping the normative climate— the set of values perceived to be generalized in a particular social context.
Normative Values
Given that values guide the principles in people’s lives (i.e., “concepts or beliefs about desirable end states or behaviours that transcend specific situations, guide selection or evaluation of behaviour and events, and are ordered by relative importance”; Schwartz & Bilsky, 1990, p. 879), information about economic inequality might lead people to infer others’ underlying values.
Following Schwartz’s theory of values (1992, 2012), reciprocity and sharing behaviours are related to self-transcendence values—caring for the welfare of others (i.e., benevolence) and societal concerns (i.e., universalism). By contrast, dominance and competition are related to self-enhancement values—pursuing personal achievement and power. Hence, equal environments should highlight the importance of values that exalt self-transcendence, whereas unequal environments should underline the importance of values that promote self-enhancement.
Previous research has shown that individuals in more unequal countries tend to self-enhance more than those in more equal countries (Loughnan et al., 2011). However, less is known about the prevalence of self-transcendence values according to the level of economic inequality. Nevertheless, Schwartz’s structure of relationships between values suggests that pursuing self-enhancement values clashes with pursuing self-transcendence values (Schwartz, 1992, 2012). Both types of values reflect a conflict between emphasizing the importance of self-interest and dominance (i.e., self-enhancement) versus emphasizing the importance of the welfare and interest of others (i.e., self-transcendence). Therefore, Schwartz’s structure of values suggests that if we expect that self-enhancement values would be inferred as more normative in unequal (vs. more egalitarian) societies, we may expect that self-transcendence values would be less inferred in these contexts.
The Current Research
In this paper, we predict that high (vs. low) inequality will increase normative self-enhancement and decrease self-transcendence values. We tested these hypotheses in three preregistered studies. First, we conducted a correlational study to test whether perceived economic inequality is related to normative values in the city where participants live. Afterward, we examined the causal relationship in an experiment manipulating the level of economic inequality in a fictional society and asking participants about the normative values of such a society. Finally, we conducted a third experiment aimed to conceptually replicate the results of Study 2 with a more ecological manipulation (all materials and data sets are available in the online supplemental material). We report all measures, manipulations, and exclusions in all studies.
Study 1
Method
Preregistered design and hypotheses
We conducted a correlation study. The original aim of this study was to explore whether perceived economic inequality was related to personal values. The relationship between economic inequality and perceived normative values—the main research question of this paper—was our secondary goal (see preregistration in the supplemental material). However, given that this paper examines the relationship between economic inequality and normative—not personal—values, we will present only this part of the research for clarity purposes (for additional details on the other measures and analyses, see supplemental material).
Sample size calculation and participants
We conducted an a priori sample size analysis using G*Power (Faul et al., 2009) for a t test, with two tails, for a correlation. Given that we did not have initial evidence to estimate our effect size, we estimated a standard medium to low effect size (r = .17) to obtain an a priori 95% power and a p value of .05. The optimal sample size was 439. Given that we anticipated that some participants would be foreign, and thus might not have good Spanish language skills, we set as a requirement that participants were native Spanish speakers (see supplemental material for exclusion criteria). Therefore, we tried to get a minimum of 439 participants after exclusions.
The initial sample was composed of 490 participants. However, 31 participants were excluded because they did not report Spanish as their native language. Therefore, the final sample consisted of 459 participants (159 men, 300 women) aged from 17 to 56 (M = 21.84, SD = 4.61).
Procedure and measurement
The sample was obtained through a convenience sampling procedure at different libraries of a city in the south of Spain. The main researcher requested the cooperation of participants, assuring them of the anonymity and confidentiality of their answers. Those who agreed to participate were given a booklet including instructions and measures. According to our rationale, we showed perceived economic inequality measures first, and after the rest of the measures. Measures are presented in the same order as in the booklet.
Perceived economic inequality in the city
Three items adapted from Sommet et al. (2019) measured the perception of economic inequality in the city (i.e., “In this city . . .” “. . . there is a huge gap between rich and poor”; “. . . there is a big difference between the top 1% of income earners and the others”; “. . . the wealth disparity between the upper and lower wage earners is large”) on a 7-point scale (1 = not at all, 7 = very much; α = .82).
Perceived economic inequality in everyday life
To measure participants’ immediate experience of economic inequality, we used the Perception of Economic Inequality in Everyday Life Scale (PEIEL; García-Castro, Willis, & Rodríguez-Bailón, 2018). This scale is composed of 12 items (e.g., “Among the people I surround myself with, some people can afford to buy a lot more and better things than others”) measured on a 7-point scale scale (1 = completely disagree, 7 = completely agree; α = .88).
Normative values
We adapted the Portrait Values Questionnaire (PVQ; Schwartz et al., 2001) to ask about the normative values in the city. Therefore, instead of asking how a particular value is important to the participant, we asked how a particular value is important to most people in Granada (e.g., “To most people in Granada is important to be rich. They want to have a lot of money and expensive things”). We only included the values in which we were interested: self-transcendence scale (five items; α = .77) and self-enhancement (four items; α = .68).
Subjective socioeconomic status (S-SES)
Participants completed the MacArthur Scale of Subjective Social Status (Adler et al., 2000) in which they had to choose which rung of a 10-rung ladder better represented their position in the social hierarchy.
Political ideology
Participants rated their political ideology on a 7-point scale (1 = extremely left-wing, 7 = extremely right-wing).
Demographic information
Finally, participants provided information about their level of education, income, gender, age, native language, and place of birth.
Results
Means, standard deviations, and Pearson’s correlations between the key variables are presented in the supplemental material. We conducted several hierarchical multiple regression analyses to test the prediction power of perceived economic inequality (in the city and in participants’ everyday life) on normative values. In the first step, we included perceived economic inequality in the city and in participants’ everyday life; we also included the mean of normative values as a covariate to control for answer bias (Schwartz, 1992). In the second step, we included socioeconomic status (SES), gender (0 = men, 1 = women), age, and political ideology 1 as covariates to control for possible confounding effects.
The first regression model included normative self-enhancement values as criterion variable (see Table 1). Model 1 was significant, F(3, 307) = 95.44, p < .001. As expected, both perceived economic inequality in the city (β = .11, p = .012) and in participants’ everyday life (β = .22, p < .001) positively predicted self-enhancement normative values. Inclusion of SES, gender, age, and political ideology did not increase significantly the explained variance, ΔF(4, 303) = 1.36, p = .247. Importantly, Model 2 remained significant, F(7, 303) = 41.88, p < .001, and both perceived economic inequality in the city (β = .12, p = .008) and in participants’ everyday life (β = .22, p < .001) remained virtually equal than those in Model 1.
Hierarchical regression analysis on predicted normative self-enhancement values.
Note. PEI = perceived economic inequality; NV = normative values; S-SES = subjective socioeconomic status; B = unstandardized coefficient; β = standardized coefficient.
p < .05. **p < .01. ***p < .001.
The second regression model included normative self-transcendence values as the criterion variable (see Table 2). Model 1 was significant, F(3, 306) = 157.90, p < .001. As expected, both perceived economic inequality in the city (β = −.10, p = .012) and in everyday life (β = −.19, p < .001) negatively predicted self-transcendence normative values. Inclusion of SES, gender, age, and political ideology did not increase significantly explained variance, ΔF(4, 302) = 1.38, p = .241. Importantly, Model 2 remained significant, F(7, 302) = 68.79, p < .001, and both perceived economic inequality in the city (β = −.11, p = .008) and in everyday life (β = −.20, p < .001) remained virtually equal than in Model 1. 2
Hierarchical regression analysis on predicted normative self-transcendence values.
Note. PEI = perceived economic inequality; NV = normative values; S-SES = subjective socioeconomic status; B = unstandardized coefficient; β = standardized coefficient.
p < .05. **p < .01. ***p < .001.
Discussion
Results of this study suggest that perceived economic inequality is related to perceived normative values. Notably, individuals who recognize higher levels of economic inequality around them think that most people living in the city tend to uphold more self-enhancement values and less self-transcendence values. It is worth noting that both perceived economic inequality in the city and in everyday life predicted normative values. Importantly, these effects remained over and above gender, age, SES, and political ideology.
The current study provides initial evidence that perceived economic inequality might be used to infer the normative values in society. However, the correlational nature of the data prevents us from assuming a causal direction of these relationships. It might be that individuals tend to over- or understate the level of economic inequality as a function of the normative values they perceive around them, or it might be a third variable (e.g., a particular worldview) that explains the relationship between them. Therefore, we conducted a second study to address these limitations. Particularly, we conducted an experiment where we manipulated the level of economic inequality in a fictional society and asked participants what normative values they thought prevailed in that society.
Study 2
Method
Preregistered design and hypotheses
The design included one independent variable with two between-subject conditions: high (50 participants) versus low (48 participants) economic inequality. We measured type of normative values (self-enhancement vs. self-transcendence) as within-participant variable. In the current experiment, we tested the following preregistered hypotheses (see supplemental material).
Hypothesis 1
There will be a significant interaction effect between the level of economic inequality (high vs. low) and the type of normative values (self-enhancement vs. self-transcendence) that people infer in a given society.
Hypothesis 1a
When economic inequality is high, participants will infer more self-enhancement than self-transcendence values.
Hypothesis 1b
When economic inequality is low, participants will infer more self-transcendence than self-enhancement values.
Sample size calculation
We conducted an a priori sample size analysis using G*Power (Faul et al., 2009) for a repeated measures ANOVA with a within–between interaction. Based on a previous study using a similar design that found a medium effect size of economic inequality on normative climate (Sánchez-Rodriguez, Willis, Jetten, & Rodríguez-Bailón, 2019), we estimated a medium effect size (f2 = 0.06) to obtain an a priori 80% power and an alpha error probability of 5%. The optimal sample size was 98 participants. 3
Participants
One hundred and one students participated in this experiment in exchange for course credit. Three participants were excluded because they were not native Spanish speakers (see exclusion criteria in the preregistration at osf.io/p6rfm). The final sample included 98 participants (82 women) aged between 18 and 29 years (M = 20.20, SD = 1.42).
Procedure and Measurement
Participants were seated in individual cubicles and asked to read and sign informed consent forms. They then indicated their age, gender, and whether they were Spanish native speakers, and were randomly assigned to the experimental conditions (high vs. low economic inequality). We manipulated economic inequality using the Bimboola paradigm (Jetten et al., 2015; Sánchez-Rodríguez, Willis, & Rodríguez-Bailón, 2019). In this paradigm, participants were asked to imagine they were going to live in a fictitious society. In this society, there were three different groups according to their income: the richest, the middle class, and the poorest. In the high-inequality condition, the gap between the richest and the poorest was presented as very large: the richest earning 13,500 Bimboolean coins per month (BC/m), and the poorest earning only 500 BC/m. By contrast, in the low-inequality condition, the gap between the richest and the poorest was small: the richest earning 8,000 BC/m, and the poorest earning 6,000 BC/m. Regardless of condition, all participants were assigned to the middle-income group, which earned 7,000 BC/m. Participants were shown the houses, cars, and holidays of the different income groups to reinforce the manipulation. Participants were then invited to imagine their lives in this new society and to choose a house, a car, and a holiday that he/she could afford to start living in Bimboola (i.e., those houses, cars and holidays belonging to the middle or poorest income group). After that and, as a manipulation check, we asked participants “To what extent is Bimboola’s economic distribution unequal [equal]?” (reversed item; 1= somewhat unequal [equal], 9 = very unequal [equal]; ρ = .94; Eisinga et al., 2013). As an additional manipulation check, we asked participants the group to which they had been assigned. 4
We then presented participants with a list of values and asked them to think about how important each of those values were as a guiding principle in life to most of the people living in Bimboola (Schwartz, 1992). Following standard procedure, we first presented all of the values, which participants read and chose those that were the most and least important (or opposed) to most of the inhabitants of Bimboola. The rationale behind this was that being aware of these extreme values should aid participants to frame other values. Self-enhancement values (α = .90) were measured by asking about power (i.e., social power, wealth, social recognition, authority, and preservation of public image; α = .89) and achievement values (i.e., ambition, influence, capability, intelligence, and success; α = .76). Self-transcendence values (α = .95) were measured by asking about universalism (i.e., equality, inner harmony, spiritual life, life meaning, world peace, unity with nature, wisdom, world beauty, social justice, open-mindedness, and protection of the environment; α = .92) and benevolence values (i.e., mature love, true friendship, loyalty, honesty, helpfulness, trustworthiness, and forgiveness; α = .90). Values were measured on a 9-point scale (7 = of supreme importance, 6 = very important, 5 = unlabelled, 4 = unlabelled, 3 = important, 2 = unlabelled, l = unlabelled, 0 = not important, −1 = opposed to my values).
Results
Manipulation check
We conducted an ANOVA on the economic inequality manipulation check. Participants assigned to the high economic inequality condition perceived more economic inequality (M = 8.35, SD = 0.88) than those in the low economic inequality condition (M = 3.09, SD = 1.42), F(1, 96) = 486.72, p < .001, η2 = .84. All participants answered correctly that they were assigned to the middle-income group.
Main results
Following our preregistration plan, and to corroborate Hypothesis 1, we conducted a 2 (self-enhancement vs. self-transcendence normative values as within-participant factor) x 2 (high vs. low economic inequality as between-group factor) repeated measures ANCOVA using the mean of all normative values as covariate in all analyses (see Schwartz, 1992).
Results showed a significant interaction effect between normative values and economic inequality, F(1, 95) = 123.07, p < .001, η2 = .56 5 (see Figure 1). We conducted planned contrast analyses within the same repeated measures ANOVA to explore the meaning of this interaction in relation to our hypothesis. Supporting Hypothesis 1a, results showed that participants in the high economic inequality condition considered that most people embrace self-enhancement values more (M = 5.12, SD = 0.96) than self-transcendence values (M = 2.10, SD = 1.23), p < .001, 95% CI [−2.47, −1.71]. 6 Moreover, in line with Hypothesis 1b, results showed that participants in the low economic inequality condition considered that most people embrace self-transcendence values more (M = 4.16, SD = 1.21) than self-enhancement values (M = 3.03, SD = 1.42), p < .001, 95% CI [1.71, 2.47]. 7
Effects of economic inequality on perceived type of normative values.
Discussion
In line with our hypotheses, Study 2 showed that degree of economic inequality influenced the type of values participants inferred as normative in society (Hypothesis 1). Specifically, in a context with high economic inequality, participants inferred that others endorsed more self-enhancement than self-transcendence values (Hypothesis 1a). By contrast, in a context with low economic inequality, participants inferred that others endorsed more self-transcendence than self-enhancement values (Hypothesis 1b). These results strengthen those from Study 1. Importantly, this study provides evidence of causal direction, suggesting that level of perceived economic inequality is used by individuals to infer the normative values in society. This result is in line with previous research showing that perceived economic inequality leads individuals to infer societal features (e.g., Davidai, 2018; Moreno-Bella et al., 2019; Sánchez-Rodríguez, Willis, Jetten, & Rodríguez-Bailón, 2019). To underpin the effect of perceived economic inequality on normative values, we conducted a third preregistered study to replicate this effect with a different methodology.
Study 3
We found evidence that perceived economic inequality in a natural setting predicts normative values (Study 1). We also showed that this may be a causal relationship (Study 2). In this study, we aimed to conceptually replicate this effect and to examine whether we could manipulate the level of inequality in the social setting where individuals live rather than in a fictional society. Accordingly, we developed a new manipulation in which we manipulated participants’ perceptions of the relative economic inequality. We expected that inducing participants to temporarily experience their country as one of the most (un)equal countries around the world would activate corresponding inferences about its normative values. Similar procedures have been used to manipulate SES (Piff et al., 2010) and cultural identity constructs (Oyserman & Lee, 2008).
Method
Preregistered design and hypotheses
We used a between-subject design with one independent variable of two conditions: high (68 participants) versus low (65 participants) economic inequality. We included self-enhancement versus self-transcendence normative values as repeated measures; we also included self-enhancement versus self-transcendence personal values 8 (see supplemental material).
In this study, we explicitly preregistered the hypotheses on the effect of economic inequality on normative values (see supplemental material).
Hypothesis 2
There will be a significant interaction effect between perceived economic inequality (high vs. low) and type of values (self-transcendence vs. self-enhancement) on perceived normative values.
Hypothesis 2a
In the high economic inequality condition, self-enhancement values will be more normative than self-transcendence values.
Hypothesis 2b
In the low economic inequality condition, self-transcendence values will be more normative than self-enhancement values.
Sample size calculation and participants
We conducted an a priori sample size analysis using G*Power (Faul et al., 2009) to calculate a repeated measures ANOVA with within–between interaction. Results showed that we needed 158 participants to obtain a 95% power with an alpha error probability of .05 to get a medium effect size (f2 = 0.06).2
Participants
One hundred and thirty-eight students participated in this experiment in exchange for course credit. Five participants were excluded because they were not Spanish native speakers (see exclusion criteria in the [preregistration] supplemental material). The final sample consisted of 133 participants (111 women) between 18 and 47 years old (M = 20.12, SD = 3.99). Although we did not reached the planned sample size, with this sample size we obtained a .05 alpha and 95% power, rendering this experiment powerful enough to find a medium effect size (f2 = 0.07).
Procedure and Measures
As in Study 2, participants were seated in individual cubicles and asked to read and sign informed consent forms. They then provided sociodemographic information—age, sex, and native language—before starting the experiment. They read that, around the world, some countries have higher economic inequality than others; however, most people are not able to accurately estimate the level of inequality in their own country in comparison with others. Because people sometimes understand economic inequality as poverty (García-Sánchez et al., 2018), we explained that level of inequality is independent of national wealth, and refers to the difference between the wealthiest 10% and the poorest 10% of the population. As part of the manipulation, we asked participants to estimate the position of Spain in comparison with other countries regarding the degree of economic inequality. They answered the question, “In comparison with other countries, where do you place Spain in the next distribution?” (1 = more equal countries, 9 = more unequal countries). Regardless of their answers, we randomly manipulated participants’ perceptions of Spain’s relative economic inequality by providing them different feedback: According to information of the World Bank on economic inequality, among the 194 countries around the world, Spain ranks 64 [149]. Therefore, the best estimation to the previous question was 3 [7]. Indeed, according to the World Bank, it is estimated that the the mean income of the wealthiest 10% is 3 [12] times higher than that of the poorest 10% in Spain. Consequently, Spain is one of the most [un]equal countries in the world.
To reinforce the manipulation and make the difference between the richest and the poorest evident, we provided the following explanation: In line with the World Bank, the National Sociology Research Centre published a report on housing last year. In this report, they demonstrated the small [large] difference between the prices of the wealthiest and those of the poorest houses in our country.
As an example, we showed participants some typical prosperous and some of the poorest houses in Spain, according to the manipulation (stimuli is available in the supplemental material). Finally, to hide the purpose of the manipulation, we informed participants that this research aimed to investigate why people cannot estimate the level of economic inequality in their country accurately, so they should write their opinion in 3 minutes. After the manipulation, we presented the following measures.
Normative values
Similar to Study 3, we asked the extent to which particular values are essential to most people in Spain (adapted from Schwartz et al., 2001): self-transcendence (five items; α = .70) and self-enhancement (four items, α = .69) values.
Manipulation check
We included two items to check whether the manipulation was successful: “In comparison with other countries, how much wealth do the richest have in Spain” and “How much wealth do the poorest have in Spain” (1 = very little, 7 = very much).
Results
Manipulation check
We conducted two independent sample t tests on perceived resources of the richest and the poorest. Results suggested that our manipulation was successful given that participants perceived the richest as having more resources in the high (M = 6.54, SD = 1.19) than in the low (M = 5.69, SD = 1.50) economic inequality condition, t(131) = −3.69, p < .001, d = 0.63, whereas they perceived the poorest as having more resources in the low (M = 4.06, SD = 1.72) than in the high (M = 2.28, SD = 1.42) economic inequality condition, t(131) = 6.52, p < .001, d = 1.13.
Effect on normative values
To corroborate Hypothesis 2, and following our preregistered analysis plan, we conducted a 2 (self-enhancement vs. self-transcendence normative values as within-participant factor) x 2 (high vs. low economic inequality as between-group factor) repeated measures ANCOVA using the mean of all normative values as a covariate (see Schwartz, 1992).
Results showed a significant interaction effect between normative values and economic inequality, F(1, 130) = 6.92, p = .010, η2 = .05, 9 supporting Hypothesis 2 (see Figure 2). Moreover, we conducted contrast analyses within the same repeated measures ANOVA to explore the meaning of this interaction in relation to our hypothesis. However, results suggested that in both conditions (high and low economic inequality), self-enhancement values were perceived as more normative (Mhigh = 5.29, SD = 0.82; Mlow = 4.97, SD = 0.88) than self-transcendence values (Mhigh = 3.99, SD = 0.96; Mlow = 4.26, SD = 0.80), phigh < .001, 95% CI [0.987, 1.614], plow < .001, 95% CI [0.38, 1.02], contrary to Hypothesis 2b, which suggested that self-transcendence values will be perceived as more normative than self-enhancement values in the low economic inequality condition. Therefore, we conducted additional analyses to understand the interaction effect. Results suggested that normative self-enhancement values were higher in the high (M = 5.28, SE = 0.79) than in the low (M = 4.98, SE = 0.81) economic inequality condition, p = .010, 95% CI [−0.52, −0.07]. 10 By contrast, normative self-transcendence values were higher in the low (M = 4.28, SE = 0.81) than in the high (M = 3.98, SE = 0.79) economic inequality condition, p = .010, 95% CI [−0.52, −0.07]. 11
Effects of economic inequality on the type of normative values (bars represent 95% confidence intervals).
Discussion
Results of Study 3 provided additional evidence that perceived economic inequality affects perceived normative values. According to our hypotheses, level of perceived economic inequality influenced perceived country values (Hypothesis 2). However, the interaction between perceived economic inequality and type of normative values was not exactly as predicted. Following the results of Study 2, we hypothesized that level of economic inequality would affect which kind of value (i.e., self-enhancement vs. self-transcendence) would be the most normative. However, in both conditions of this study, participants perceived that people in Spain are more driven by self-enhancement values. It could be that reality constraints influenced how these two values are perceived (e.g., Spain is an unequal country compared to others in the EU, so this may be driving this effect), and thus the manipulation could not counteract this reality. Relevant to our hypothesis, however, perceived values did interact with the economic inequality manipulation: high inequality (vs. low) increased self-enhancement and decreased self-transcendence values. This result is consistent with our expectation that inequality can change perceived normative values.
General Discussion
The present research provides evidence that perceived economic inequality influences perceived normative values: individuals perceived that self-enhancement values were more important for others when they observed higher (vs. lower) levels of economic inequality around them. However, when they saw lower (vs. higher) levels of economic inequality, they perceived self-transcendence values were more important for most of the people in that society. Importantly, in Study 1, we showed that these effects remained over and above other sociodemographic characteristics—SES, political ideology, gender, age. It is worth noting that Study 1 results suggest that neither SES, political ideology, gender, or age seem to be related to normative values.
Additionally, Studies 2 and 3 provided evidence of causality. Indeed, the results of these studies showed that level of economic inequality led people to infer normative values. Therefore, these results suggest that level of economic inequality is used as a clue to infer what the normative values in society are. This is in line with other research showing the importance of the level of perceived economic inequality to infer other societal and normative features (Davidai, 2018; Moreno-Bella et al., 2019; Sánchez-Rodríguez, Willis, Jetten, & Rodríguez-Bailón, 2019). It is worth noting that the found causal direction from economic inequality to values does not deny the possibility that the reverse causal direction could also be true. Future research should explore this alternative.
Despite our main results showing that perceived economic inequality affects perceived normative values, we did not find evidence that it affects participants’ personal values (see supplemental material). A possible explanation why we did not find this evidence might be because, in contrast to the effect on normative values, which seems more direct, the impact of economic inequality on personal values might be more complicated. Notably, the effect of economic inequality on personal values might need to operate through social influence processes. According to the “descriptive norms effect,” individuals tend to copy behaviours that they perceive common among others (Cialdini et al., 1990; Deutsch & Gerard, 1955). However, only those who highly identify with the group might be affected by normative values as self-categorization theory would predict (Reynolds, 2019; Terry & Hogg, 1996; Turner & Reynolds, 2012). Indeed, recently, Pryor et al. (2019) showed that conformity to an arbitrary norm depends on the degree to which people identify with the group upholding the norm—high identifiers tend to conform more. Future research should explore the role of possible moderator variables, such as identification with the group, in the effect of economic inequality on personal values.
Moreover, results from Study 1 suggest that SES does not affect normative values. Never-theless, we should be cautious in relation to this result given the lack of SES variability in our sample—most participants were university students. It would be interesting for future research to explore how perceived economic inequality might affect normative values as a function of individuals’ position in the social hierarchy. Previous research has suggested that SES is positively related to individualistic values (e.g., self-enhancement values) and negatively related to interdependent values (e.g., self-transcendence values; Manstead, 2018; Stephens et al., 2014). Therefore, SES might interact with perceived economic inequality, exacerbating its effects—that is, the effect may be stronger for high-SES individuals.
It is worth noting that the effect of economic inequality on normative values holds across contexts—economic inequality in participants’ everyday life, at the city level (Study 1), and at the country level (Studies 2 and 3). However, the larger coefficients found for the impact of perceived economic inequality in participants’ everyday life in comparison with perceived economic inequality in the city in Study 1 suggest that perceived economic inequality in our daily life might be the most powerful predictor of its consequences. This result is in line with previous research showing that perceived local economic inequality might increase social comparison processes, which would increase its effects (Dawtry et al., 2019; García-Castro et al., 2018; Krupp & Cook, 2018).
Interestingly, we found evidence for the effect of perceived economic inequality on perceived normative values in both a fictional society and the actual social context. This is important, first, because it provides ecological validity to our results and, second, because the effect is not the same in both environments. Based on Wilkinson and Pickett’s (2017) arguments, level of economic inequality should create an environment with certain features. Specifically, higher levels of economic inequality should provide a context where individuals put greater emphasis on status-seeking and competition. By contrast, an environment with low economic inequality is likely one where individuals give more importance to reciprocity and sharing. Indeed, the current research provides evidence that a higher level of economic inequality leads individuals to perceive that most people in that context are motivated by status-seeking (i.e., self-enhancement values).
Conversely, a lower level of economic inequality leads individuals to perceive that most people are motivated to care about others, and that relationships are based on reciprocity and sharing (i.e., self-transcendence). However, according to our results, this is particularly true when individuals know little or nothing about the society in question. Indeed, in Study 2, in which participants had to infer the normative values in a fictional society, the type of perceived normative values changed entirely as a function of the level of economic inequality. Nevertheless, in Study 3, in which participants had to infer the normative values in their own society, self-enhancement values were always more prevalent—although inequality did change their relative importance. This result might suggest that the effect of perceived economic inequality on normative values is stronger when individuals know little about that society, as could be the case for immigrants living in a host country. However, when individuals live in an environment with a particular level of economic inequality, its effect might be attenuated by other variables (e.g., political climate). Future research could explore in which conditions perceived economic inequality has stronger or weaker effects on normative climate, as well as its interacting variables.
A limitation of this research is that we explored the consequences of perceived economic inequality on normative values only with participants from a single country. Perceptions of the level of economic inequality might differ across societies due to different social structures and cultures. It would be worth exploring the normative climate inferred from perceived economic inequality in other contexts.
Another possible limitation of our studies might be the effect of demand characteristics on participants’ answers about normative climate. They might have thought that they should reproduce the current political discourse, which claims that economic inequality is associated with social problems. However, there is no reason to believe that participants considered self-enhancement values as dysfunctional for society. Alternatively, participants might have thought that, given that equality is associated with higher social mobility (Davidai, 2018), power and achievement (i.e., self-enhancement) should be more valued in more equal societies, as there might be more incentives to climb the social ladder. Future studies should conceptually replicate this effect with other paradigms to be able to control for demand characteristics.
In summary, in the current article, we present evidence that economic inequality affects perceived normative values. These results extend the literature on the consequences of perceived economic inequality for normative climate, providing a bigger picture of the social perception associated with economic inequality.
Supplemental Material
Supplementary_Material – Supplemental material for Economic inequality affects perceived normative values
Supplemental material, Supplementary_Material for Economic inequality affects perceived normative values by Ángel Sánchez-Rodríguez, Rosa Rodríguez-Bailón and Guillermo B. Willis in Group Processes & Intergroup Relations
Footnotes
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a fellowship from the Spanish Ministry of Education, Culture, and Sport (FPU-13/01231) and by a grant from the Spanish Ministry of Economy and Competitiveness (PSI2016–78839-P).
Notes
References
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.
