The relevance of meritocratic beliefs for redistributive preferences increases with income

In light of rising global inequality, public concerns for inequality have been declining. This paradox of inequality has been explained by the rise in meritocratic beliefs in some developed countries. Untested in South Africa, the most unequal society, this study aims to report on these issues by investigating the impact of meritocratic and nonmeritocratic beliefs on redistribution preferences in South Africa. Using the International Social Survey Programme data for 2009 and 2019, we find that contrary to some developed countries, meritocratic beliefs have declined despite persistent high inequality. Nonmeritocratic factors like sex and race are perceived as less critical for personal success, while bribery and political connections gain prominence. Results reveal that stronger meritocratic beliefs correlate positively with redistribution preferences, whereas nonmeritocratic beliefs have minimal impact. This challenges the notion that high inequality intensifies meritocratic beliefs and reduces demand for redistribution, suggesting the need to consider South Africa's unique historical and socioeconomic context.

Can an inequality in rewards result in an erosion of broad-based support for meritocratic norms? We examine whether unequal rewards can affect social preferences for redistribution by driving a cognitive gap in the meritocratic beliefs of those who are successful and those who are not. Two separate experiments (conducted in the USA and the UK) show that the elite develop and maintain ‘meritocratic bias’ in the redistributive taxes they propose. This bias results in lower taxes on the rich and fewer transfers to the poor, including those who failed despite high effort. These social preferences at least partially reflect a self- serving meritocratic illusion that their own high income was deserved and reflected their ability. An incentivized Wason Card task confirms that individuals prefer to maintain their illusion of being meritocratic, by not expending cognitive effort to process any information that may undermine their self-image of being deserving.

Experimental economists study the Public Goods Game (PGG) to investigate cooperative behavior. This study embedded inequality and income mobility into the PGG to investigate its effect on cooperation. Treatment 1 embedded fixed transfers in-between rounds of the PGG, representing equally distributed income growth. Treatments 2 and 3 embedded variable transfers in-between rounds, representing heterogeneously distributed income growth based on equal proportion of luck and effort (Treatment 2) or a disproportionate amount of luck (Treatment 3). Treatments 2 and 3 also enabled relative income mobility, while treatment 1 restricted any possibility of relative mobility. Results showed that introducing relative income mobility (Treatments 2 and 3) did not lead to greater cooperation (when compared with Treatment 1) among the players. Comparing Treatments 2 and 3, participants contributed more to the PGG when relative income mobility was dependent on effort rather than on luck. Despite theories linking weakened meritocratic beliefs with greater preferences for redistribution, in this study, contribution to the PGG fell when the income mobility process was predominantly luck-based. Unfairness observed in the income mobility process discouraged contribution, even when equity of final outcomes could be achieved through cooperation. This could be because a behavioral component of dissatisfaction against inequity in the income mobility process outweighed the motive of using the PGG to redistribute unfair income. The study also investigated within-treatment behavior of different endowment types (low class, middle class, high class) when various forms of income mobility were integrated. Under these conditions, and when relative income mobility was integrated into the PGG on a predominantly luck-based process, a significant reduction in contribution from the middle class was observed. Endowment inequality did not reduce cooperation, presumably because the associated relative income mobility incorporation into the PGG threatened downward mobility, such that higher-class participants contributed to establish a safety-net.