Measuring compliance with minimum wages

Identifying genuine underpayment of minimum wages is not straightforward. Some well-known statistical issues affect the measurement of compliance rates, but factors such as processing or behavioural influences amongst respondents can also have an impact. We study the quantitative measurement of non-compliance with the minimum wage, using UK apprentices (who have particularly high non-compliance rates) as a case study. We show that understanding the institutional and behavioural context can be invaluable, as can triangulation of different sources. While the binary nature of compliance makes such problems easier to identify and evaluate, this analysis holds wider lessons for the understanding of the characteristics of large and complex datasets.