Download BibTex file | 

Abstract This article examines the thresholds at which provision of unpaid care affects employment in England. Previous research has shown that providing care for 20 or more hours a week has a negative effect on employment. The present article explores the impact of a lower threshold and asks whether provision of care for 10 or more hours a week has a negative effect on employment. The article focuses on women and men aged between 50 and State Pension Age (60 for women, 65 for men). The study uses data from the first four waves of the English Longitudinal Study of Ageing (ELSA), collected in 2002/2003, 2004/2005, 2006/2007 and 2008/2009. Across these waves, there are 17 123 people aged 50-59/64 years, of whom 9% provide unpaid care to an adult. Using logistic regression analysis of the longitudinal data, the study finds that employed women in their fifties who start providing care for <10 hours a week are significantly more likely to remain in employment one wave later than similar women who have not started to provide care. In contrast, employed women in their fifties who start providing care for 10 or more hours a week are significantly less likely to remain in employment one wave later than similar women who have not started to provide care. Employed men aged between 50 and State Pension Age, who provide care for 10 or more hours a week at the beginning of the period have a significantly reduced employment rate one wave later than those who do not provide care. The study therefore suggests that carers' employment may be negatively affected when care is provided at a lower intensity than is generally estimated in England. This has important implications for local authorities, who have a duty to provide services to carers whose employment is at risk.