Women continue to earn lower wages than men and are underrepresented in the labour market, particularly at the top. In an earlier piece on the topic I argued that big data has enabled economists to show that overt discrimination is ubiquitous rather than anecdotal. Here, I want to look at how the influence of social psychology has led to a growing recognition among economists that discriminatory behaviour may also arise from unconscious (or "implicit") stereotyping, and at how policy is being reshaped as a result.

Implicit stereotyping is the result of a “fast” unconscious process – it does not necessarily entail ill feeling – and it may remain undetected even by discriminatory actors themselves. Common gender stereotypes are that women are worse than men at mathematics, that primary school girls are worse at sports, and that women are better at raising children than men. 

Unconscious stereotypes can lead to self-fulfilling prophecies – they can cause the discriminated groups to behave in the way that stereotypes say they do. Research shows that teachers’ stereotypes affect the gender gap in maths, choice of school subjects, and self-confidence in mathematical abilities for girls in middle school. Similarly, exposure to managers with stronger implicit bias negatively affects the performance of minorities in the workplace.

When unconscious bias is at play, coercive policies targeting conscious bias, such as equal pay legislation, transparency laws and codes of conduct, may not be effective. Instead, behavioural interventions that change behaviour of either discriminatory actors or of the group being discriminated against, can help.

Measures to overcome unconscious bias on the part of the decision-maker aim to allow more conscious (non-discriminatory) processes to kick in. These measures include changing the context in which decisions are made, promoting greater contact with the discriminated group, and spreading the decision-making process across several individuals. A popular tool for addressing unconscious bias in the decision-maker consists of revealing their biases, for example through an implicit association test (IAT), a computer-based tool developed by social psychologists, designed to minimise the risk of social desirability bias. An increasing number of firms and institutions, including Harvard University, administer the IAT to their employees.¹ In the educational context, revealing stereotypes to teachers via the IAT could be a powerful intervention to decrease gender discrimination in grading.

Interventions to prevent internalising stereotypes by the group being discriminated against can start early in childhood. For example, recent attempts to teach young girls to code counteract stereotypes and may increase resilience and grit. In a workplace setting, small changes like changing the words in a job advertisement, and the use of gender-neutral language, may incentivise women tend to put an application forward.

These “nudges” or behavioural interventions that attempt to tackle unconscious discriminatory behaviour have gained popularity in the corporate sector because they are relatively cost-effective, fast, and easy to implement. However, quick fixes can often backfire. When provided with feedback about their own implicit associations, individuals may react defensively and question the validity of the IAT. Having more women in a hiring committee does not seem to lead to a higher probability of success for a female candidate, and interventions encouraging girls and women to break with stereotypes may not always pay off. For example, women may avoid negotiations more often than men not because they shy away from competition, but because negotiating more may indeed not pay off for them.

Ultimately, neither conscious discrimination nor unconscious bias can explain the formation and evolution of stereotypical beliefs in the first place. In my next and final piece, I will look at the role that social norms play in economics, and how it is contributing to our understanding of where stereotypes come from, paving the way for interventions that aim to change unequal gender roles more directly.

1. Harvard University strongly encourages "every search committee member to take at least one Implicit Association Test (IAT)" (https://faculty.harvard.edu/recruitment-best-practices)—Alesina et al. (2018).

Almudena Sevilla is a Professor in Economics and Public Policy at UCL. This piece is based on her forthcoming article, “Gender economics: an assessment”, which will be published in the Oxford Review of Economic Policy, vol. 36 no. 4a.