“Social Status and Unethical Behavior: Two Replications of the Field Studies in Piff Et Al 2012, Minah H. Jung, Paul Smeets, Jan Stoop, Joachim Vosgerau2023-01-05 (, )⁠:

Prominent social psychologists and major media outlets have put forward the notion that people of high socioeconomic status (SES) are more selfish and behave more unethically than people of low SES.

In contrast, other research in economics and sociology has hypothesized and found a positive relationship between SES and prosocial and ethical behavior.

We review the empirical evidence for these contradictory findings and conduct two direct, well-powered, and preregistered replications of the field studies by Piff et al 2012 to test the relationship between SES and unethical/selfish behavior.

Unlike the original findings, we find no evidence of a positive relationship between SES and unethical/selfish behavior in the two field replication studies.

[Keywords: social class, socioeconomic status, prosocial behavior, ethical behavior, replication]

…However, other articles report the opposite pattern, whereby people of higher social status behave more prosocially than those of lower social status (for example, Andreoni et al 2021; Gittell & Tebaldi2006; Hoffman2011; Hughes & Luksetich2008; James & Sharpe2007; Korndörfer et al 2015; Lee & Chang2007; Rajan et al 200915ya; Ramirez-Valles2006; Reed & Selbee2001; Schmukle et al 2019; Smeets et al 2015; von Hermanni & Tutic2019). Moreover, direct replications of laboratory studies in Piff et al 2012 (Studies 5 & 7) do not find evidence for the negative relationship between social status and prosocial behavior (Balakrishnan et al 2017a, 2017b; Clerke et al Brown2018).

…It should be noted, however, that direct replications of laboratory Studies 5 & 7 in Piff et al 2012 failed to corroborate the original findings (Balakrishnan et al 2017a, 2017b; Clerke et al 2018). For instance, SES is positively related to self-reported greed, but in 3⁄4 studies, no relationship between SES and unethical behavior was observed (Balakrishnan et al 2017b). A meta-analysis of their findings shows no moderating effect of greed on the relationship between SES and unethical behavior, as hypothesized and reported by Piff et al 201212ya. Clerke et al 2018 reported a positive relationship between SES and self-reported greed in one of their two studies but found no association between SES and the propensity to lie in a hypothetical salary negotiation. Even if the original hypothesis were true, it is unlikely for all 7 studies in Piff et al 2012 to have yielded statistically-significant results, given the low power of the studies (Francis2012).

Using the World Values Survey across 27 countries, Wang & Murnighan2014 found that income is positively correlated with the approval of unethical behavior. Specifically, individuals who reported belonging to a higher income decile were more likely to approve of claiming unentitled government benefits, avoid paying for public transport, to cheat on taxes, and accept a bribe than individuals from lower income deciles. Individuals with higher incomes are more likely to misreport their income in IRS data from 2001 (Johns & Slemrod2010). Moreover, the probability of hiding assets offshore rose sharply and statistically-significantly with wealth (Alstadsaeter et al 2019). Upper SES individuals cheat more than lower SES individuals when cheating was beneficial to them (Dubois et al 2015). But the opposite—lower SES individuals cheating more than upper SES individuals—was found when cheating benefited another person, suggesting that the relationship between SES and unethical behavior is context-dependent.

In a series of laboratory and online experiments, Piff et al 2010 documented higher SES participants to be less likely to allocate money to others in hypothetical economic games and to be less willing to help a confederate than low-SES participants. Two recent preregistered and highly powered replications of Studies 1 & 2 in Piff et al 201014ya, however, failed to corroborate these findings (Stamos et al 2020).