Expectations of trustworthiness in cross-status interactions
Introduction
In a world of increasing income inequality, the normative expectations of people from different status groups may collide, weakening the social fabric and making social cooperation more difficult (Elgar and Aitken, 2011; Putnam, 2001; Uslaner and Brown, 2005). Status hierarchies create material and normative boundaries between groups (Lamont, 2002), hence individual status characteristics could elicit different expectations of normative behavior. In this paper we address the relationship between social status and expectations of trustworthiness. In trusting others, one party, the trustor, makes itself vulnerable by taking a course of action that creates incentives for the other party, the trustee, to exploit it. If there is no vulnerability –some risk of being disappointed in relationships with others– then trust is not necessary (Rousseau et al., 1998). Trust is thus ‘calculative’, based on the expectation of trustworthiness (Hardin, 2002). Following Robbins (2016a, 2016b), we understand trust as a belief about another person's trustworthiness with respect to a particular matter at hand that emerges under conditions of uncertainty. Because this expectation is a risky belief about the future behavior of trustees (will they reciprocate?) in interactions with strangers, trustors may use social cues to infer (attribute) trustworthiness, such as phenotypical traits and markers of economic, cultural or symbolic capital. Indeed, some authors have claimed that ingroup vs. outgroup preferences and stereotyping are two social categorization mechanisms based on visible cues that lead to initial trust (how trust forms in the initial phase of a relationship, i.e. when parties are unfamiliar with each other, see McKnight and Chervany, 2006). Evidently, social cues relate to cognitive schemata (e.g., social identities, heuristics, stereotypes) that are not central to the definition of trust, but rather causal variables that impact the willingness to trust people we do not know very well. Here, we address the role of individual differences in estimating trustworthiness, and how the trustee's characteristics are assessed by respondents of different social backgrounds when deciding to trust others. We focus on whether the social status of trustees moderates the relationship between the social status of trustors and their trusting behavior.
Material conditions of life lead to behavioral dispositions that are often not idiosyncratic but shared by people in similar conditions. Status differences also shape the normative behavior of individuals and their expectations regarding the normative behavior of others. For instance, lower-status individuals exhibit more generosity and tend to react more sympathetically to the suffering of others than higher-status individuals (Piffet al., 2010; Stellar et al., 2012). Due to their harsher living conditions, lower-status individuals are also more wary of social threats. Research has documented that lower-ranking participants are more generous and trusting, but they also track the hostile emotions of others more accurately and show more antagonistic reactivity compared to higher-ranking participants (Kraus et al., 2011). We analyze whether these different types of normative reasoning across the status hierarchy also result in different expectations of trustworthiness in cross-status interactions.
We extend previous research on trust by studying how trustees' characteristics – such as phenotypic traits, gender, parental and own cultural capital, income, electoral participation, and religious denomination – are assessed by participants when asked to trust unfamiliar others. We address this question using a vignette experiment (Atzmüllerand Steiner, 2010). There is a long tradition of research on trust using vignette experiments (e.g., Buskens and Weesie, 2000; Robbins, 2017). These studies have addressed the sources of trust and trustworthiness in hypothetical social exchanges (e.g., car repair, group projects). In this paper, we implemented a vignette experiment that is a modified version of the trust game (Berg et al., 1995). The trust game involves two players, a trustor and a trustee, and trusting behavior is a measure of the willingness to take a chance that an investment will be rewarded by a stranger (Ben-Ner and Halldorsson, 2010). In our modified game, participants act as trustors, and trustees are not real players, but individuals represented in vignettes whose characteristics are manipulated to elicit participants’ expectations of trustworthiness, which we observe in their trusting behavior in the game. This design assumes that in interactions with strangers, individuals use status cues –markers that provide information regarding the status of others (Fişeket al., 2005) to form their expectations and consequently decide whether or not to trust them. We address two related questions. Firstly, whether the status characteristics of trustees outweigh other attributes, such as religious denomination and electoral participation, that might signal trustworthiness. Secondly, and more importantly, we explore the extent to which the status of participants interacts with the status characteristics of the trustee in shaping their trusting decisions.
To broaden the research agenda on trust, we implement our modified vignette-based trust game in Chile, which provides an adjudicative setting for addressing these questions. Inequality has been a structural feature of the social order in Chile (PNUD, 2017), which still has high income inequality, with a Gini index of 44.4 (The World Bank, 2017). Chile also exhibits low intergenerational economic mobility (Núñez and Miranda, 2010), characterized by high elite closure (Torche, 2005), clear occupational barriers between social classes (Espinoza and Núñez, 2014), and status attributions based on skin tone (Salgado and Castillo, 2018) especially among the closed-off elite (Torres et al., 2019). Based on previous research suggesting an inverse relationship between inequality and trust (Zmerliand Castillo, 2015; also see Fairbrother and Martin, 2013), it is not surprising that Chile exhibits comparatively low levels of generalized trust (Alganand Cahuc, 2014). Our research thus broadens the understanding of the determinants of trust by focusing on a less-studied non-WEIRD (western, educated, industrialized, rich, and democratic), unequal, and segregated context.
Section snippets
Social status and trust
Differences in normative behavior across status groups stem from diverging forms of social cognition anchored in the material conditions faced by individuals (Kraus et al., 2012). While higher-status individuals experience ongoing socioeconomic freedom that enhances a mind-set involving increased attention to their own mental states and a solipsistic attitude, the heightened interdependence and attunement of lower-status individuals to the needs and emotions of others make them more concerned
Vignettes and data
A multiple-stage data collection design was implemented on a wide range of post-secondary education programs and institutions attended by students of different SES. Post-secondary education offers a setting of a transitional nature between a deeply income-segregated school system (Valenzuela et al., 2014) and entry into the labor market. Hence, the higher educational tier – albeit still significantly segregated by social class – offers exceptional socioeconomic diversity, which most individuals
Empirical strategy
Following the standard practice for analyzing factorial survey experiments (Auspurg and Hinz, 2014), the nested nature of our repeated measurements (i.e., vignette evaluations nested within participants) was considered for testing the hypotheses, thus estimating a set of multilevel linear regression models (Gelman and Hill, 2006). A first random intercept7
Results
Our sample data reproduce the previously established positive correlation between the SES of individuals and the standard trust measure. Indeed, as shown in Fig. 1, higher-status participants showed, on average, higher levels of generalized trust, β = 0.528, p = 0.002. As previous research has claimed (Simpson et al., 2007), this result makes the experimental analysis presented here more convincing. We now turn to the analysis of the amount sent in the modified trust game.
The average amount
Discussion and conclusions
Using an original experimental survey and a lab-in-the-field approach, this study addressed the expectations of trustworthiness in cross-status interactions with strangers in a non-WEIRD and high inequality context. Findings show that, in trusting others, higher-status individuals are considered less trustworthy than lower-status individuals, even though they show higher levels of generalized trust. Thus, higher-status individuals declared more generalized trust, although on average they were
Funding
This work was funded by The National Research and Development Agency (ANID), Chilean Ministry of Science, Technology, Knowledge and Innovation, through its programs The National Fund for Scientific and Technological Development (FONDECYT 1161624), and through the Centre for Research in Inclusive Education (SCIA ANID CIE160009).
Acknowledgments
We thank Dr. Juan Carlos Oyanedel, who participated in the early stages of this research and the two anonymous reviewers of Social Science Research for their valuable comments. An earlier version was presented at the 11th Annual Conference of the International Network of Analytical Sociologists at Stanford University in 2018. Any errors are the authors' own responsibility.
References (87)
- et al.
Trust, growth, and well-being: new evidence and policy implications
- et al.
Trusting and trustworthiness: what are they, how to measure them, and what affects them
J. Econ. Psychol.
(2010) - et al.
Trust, reciprocity, and social history
Game. Econ. Behav.
(1995) - et al.
Trust and gender: an examination of behavior and beliefs in the Investment Game
J. Econ. Behav. Organ.
(2008) - et al.
oTree—an open-source platform for laboratory, online, and field experiments
Journal of Behavioral and Experimental Finance
(2016) - et al.
Does inequality erode social trust? Results from multilevel models of US states and counties
Soc. Sci. Res.
(2013) - et al.
Status cues and the formation of expectations
Soc. Sci. Res.
(2005) - et al.
How much should we trust the World Values Survey trust question?
Econ. Lett.
(2012) - et al.
Relative deprivation and revolt: current and future directions
Current Opinion in Psychology
(2020) From the general to the specific: how social trust motivates relational trust
Soc. Sci. Res.
(2016)