Understanding pay gaps

In this paper, we use micro-data from the UK Labour Force Survey to estimate unconditional and conditional pay gaps for gender and ethnicity groups since the mid-90s. Both types of gender pay gap have decreased over the sample, but remain in double digits. For ethnicity, the unconditional pay gap has been materially lower, compared to the gender pay gap, while the conditional pay gap is of similar magnitude to the gender one. These trends are apparent not only at the mean but also at the lower and upper ends of the distributions. Interaction effects between gender and ethnicity reveal that female ethnic minority workers experience a larger pay gap than both ethnic minority male, and white female workers. Half of the gender pay gap can be accounted for by compositional effects, such as the individual’s age, education, and the nature of their job, such as occupation and sector, while half remains unaccounted for. We find that the minimum wage leads to a decrease in the gender pay gap. Compositional effects for ethnicity suggest that ethnic minorities should be earning more than their white counterparts, with the unaccounted for factors driving the positive ethnicity pay gap. We find that compositional effects are heterogeneous across gender and ethnic minority groups. We assert that there is a strong case to extend compulsory pay gap reporting to ethnicity. There is also a case to extend compulsory reporting from firms with more than 250 employees to those with around 30 or more, to increase the coverage of the UK employed population.