(Anti-)egalitarianism differentially predicts empathy for members of advantaged versus disadvantaged groups.

We explore the relationship between group-based egalitarianism and empathy for members of advantaged groups (e.g., corporate executives; state officials) versus disadvantaged groups (e.g., blue-collar workers; schoolteachers) subjected to harmful actions, events, or policies. Whereas previous research suggests that anti-egalitarians (vs. egalitarians) dispositionally exhibit less empathy for others, we propose that this relationship depends on the target's position in the social hierarchy. We examined this question across eight studies (N = 3,154) conducted in the U.S. and the U.K., including online and in-person experiments and examining attitudinal and behavioral outcomes. We observed that (anti-)egalitarianism negatively predicted empathy for members of disadvantaged groups subjected to harmful situations, but positively predicted empathy for members of advantaged groups. This pattern held regardless of perceivers' own membership in advantaged or disadvantaged groups (i.e., perceiver gender, race, or SES). (Anti-)egalitarianism's differential effects on empathy for advantaged versus disadvantaged targets were attributable in part to differences in perceived degree of harm incurred (beyond roles for perceived value conflict and perceived deservingness): Egalitarians perceived the same action as more harmful than anti-egalitarians when it occurred to a disadvantaged target but less harmful than anti-egalitarians when it occurred to an advantaged target. We also explored how these patterns informed individuals' downstream policy attitudes and policy-relevant behavior (e.g., willingness to sign a petition). Our findings enrich understanding of (anti-)egalitarianism by testing competing perspectives on the link between (anti-)egalitarianism and empathy, and by demonstrating when and why individuals' preferences for social equality (vs. hierarchy) lead them to extend versus withhold empathy. (PsycINFO Database Record