Media Briefings

SELF-REINFORCING INEQUALITY: Experimental evidence that people’s beliefs about their relative ability limit their upward mobility

  • Published Date: June 2016

Experiencing disadvantage can make escaping disadvantage less likely because it leads people to believe that they are of lower relative ability and therefore less willing to make the effort needed to pursue a high-achieving career. That is the central finding of a series of experiments by Dr Jeffrey Butler, which randomly assign unequal pay levels to participants and then see how it affects their beliefs and their subsequent willingness to compete.

The implications of his research, which is published in the June 2016 Economic Journal, are that policies aimed at eliminating external barriers to success may not be sufficient to reduce inequality. When individuals must choose paths leading to success, inequality itself may lead the disadvantaged to discriminate against themselves in way that perpetuates inequality.

Persistent inequality is of perennial interest on both moral grounds – we should want others to be able to attain their full potential – and economic grounds – as forgone success often represents lost aggregate productivity. The most widely accepted existing explanations rely on external barriers to success (for example, discrimination or borrowing constraints) or differences in individual traits (for example, risk tolerance or cognitive ability).

This study proposes – and provides experimental evidence for – a novel explanation in which inequality may be self-reinforcing and persist even in the absence of external barriers and without systematic differences in individual traits. This novel channel for the persistence of inequality relies on two ingredients:

• First, ability-based competition, like completing a college degree, is an important path towards upward mobility.

• Second, people believe that society is just and fair – that is, that people receive the outcomes they merit.

The latter phenomenon is well established in social psychology – the idea of ‘Just World Beliefs’ – to be universal, evincing itself in such commonplace behaviour as blaming victims for their own misfortunes. An implication of Just World Beliefs is that if merit depends on one’s own relative ability, then experiencing disadvantage may cause the disadvantaged to believe they are less relatively able.

Life paths for which success ultimately depends on relative ability (for example, attending highly competitive colleges or choosing lucrative but competitive careers) are consequently worth less to the disadvantaged, according to their own self-discriminatory expectations. Because these are exactly the paths to upward mobility in modern meritocratic economies, experiencing disadvantage may therefore make escaping disadvantage less likely.

The novel conjectured channel through which inequality is self-perpetuating involves several steps, none of which is a priori obviously true:

• It is not certain that Just World Beliefs affect beliefs about one’s own relative merit – particularly given the large body of research documenting self-serving biases such as overconfidence.

• Even then, it is not obvious whether beliefs about the ability component of merit – a stable trait – will be affected.

• Finally, it is not guaranteed that the effect of inequality on beliefs will be sufficiently strong to change behaviour, like one’s willingness to compete.

This study presents evidence from multiple experiments testing each of these steps. While there are myriad ways to engender a feeling of being treated unequally in the laboratory, these experiments implement inequality in a familiar manner: unequal pay for equal performance.

After randomly assigning unequal pay levels, the experiments elicit participants’ beliefs about their relative ability and show that even though pay inequality is transparently randomly assigned, relative ability beliefs respond in the conjectured way. Workers assigned lower (higher) pay believe that they are relatively less (more) able.

The study goes on to show that these belief patterns are strong enough to affect individuals’ willingness to compete in a way consistent with the conjecture.

Finally, to shed additional light on whether it is beliefs about ability, a stable trait, that are affected by inequality, the experiments randomly assign participants either a task that depends on ability (an IQ test) or mostly on effort (counting occurrences of a specific letter). The results are consistent with the notion that inequality affects beliefs about ability particularly; they are not consistent with the alternative story that inequality affects beliefs about the relative provision of effort by the disadvantaged.

The implications of these findings are that policies aimed at eliminating external barriers to success may not be sufficient to reduce inequality. When individuals must choose paths leading to success, inequality itself may lead the disadvantaged to select sub-optimally, discriminating against themselves in way that perpetuates inequality.


Notes for editors: ‘Inequality and Relative Ability Beliefs’ by Jeffrey V. Butler is published in the June 2016 issue of the Economic Journal.

Jeffrey Butler is at the Einaudi Institute for Economics and Finance, and the University of Nevada, Las Vegas.

For further information: contact Jeffrey Butler on +1-702-690-7713 (email:; or Romesh Vaitilingam on +44-7768-661095 (email:; Twitter: @econromesh).