Our study provides the first national analysis of the labour market implications of workers who are licensed by any agency of the government in the USA. Using a specially designed Gallup survey of a nationally representative sample of Americans, we provide an analysis of the influence of this form of occupational regulation. We find that 29 per cent of the workforce is required to hold a licence, which is a higher percentage than that found in other studies that rely on state-level occupational licensing data or single states. Workers who have higher levels of education are more likely to work in jobs that require a licence. Union workers and government employees are more likely to have a licence requirement than are non-union or private sector employees. Our multivariate estimates suggest that licensing has about the same quantitative impact on wages as do unions - that is about 15 per cent - and that being both licensed and in a union can increase wages by more than 24 per cent. However, unlike unions which reduce variance in wages, licensing does not significantly reduce wage dispersion for individuals in licensed jobs.