The Representativeness of the Annual Survey of Hours and Earnings and Its Implications for UK Wage Policy

ABSTRACT The Annual Survey of Hours and Earnings (ASHE) is based on an annual 1% sample of employee jobs and provides many of the UK's official earnings statistics. These statistics are produced using official weights designed to make the achieved sample in each year representative of the population of employee jobs in Britain by gender, age, occupation and region. However, we show that jobs in small, young, private‐sector organisations remain significantly under‐represented after applying these weights. To address this issue, we develop new weights and demonstrate their importance through policy‐relevant examples. Our new estimates suggest that the bite of the National Living Wage is greater than previously reported, and the gender pay gap is wider. We conclude that a new official review of the methodology for ASHE is merited to improve the accuracy and reliability of data informing earnings analysis and research in the United Kingdom.