Background: Increasing numbers of childless men as well as fathers with reduced or no contact with their children have sparked concern about an erosion of fatherhood. Although the general trend is undisputed, claims about men's decreasing family involvement lack a sound empirical basis that enables comparisons between countries and sub-groups of society.

Objective: This study derives long-term trends in father-child co-residence over the life course in England and Wales, and provides comparisons of these trends by level of education and ethnic origin.

Methods: The paper calculates shares of father-child co-residence from the National Statistics Longitudinal Study (LS) and the British Labour-Force Surveys (LFS).

Results: There has been a decline of father-child co-residence in England and Wales for men in their thirties, an even greater decline for men in their twenties, and a small increase in the shares of father-child co-residence at higher ages. The trends for different educational groups were similar, but men with a degree had particularly low rates of father-child co-residence at younger ages, and relatively high ones at older ages. Neither less-educated men, nor men from Black-Caribbean and Black-African origins showed rates of father-child co-residence as low as one might have expected.

Conclusions: The steady decline of father-child co-residence among men born between 1930 and 1979 in England and Wales lends support to claims about an erosion of fatherhood. However, it is unwarranted to generalize findings from other countries about particularly low levels of father-child co-residence among less-educated men to men in England and Wales.