In further exploratory work we have included in the outcome regressions the time
between the first birth and the birth of the cohort member to each of the final models
presented in strand 2. In all cases the additional variable does not affect the pattern of
age at motherhood effects found and therefore does not affect the story presented. Its
inclusion produces a significant positive trend for employment of the partner,
employment of the mother and net family income group reflecting more positive
outcomes for older mothers holding the number of children constant. A significant
negative effect is found for means tested benefits, household income being less than 60%
of the median and malaise whilst no significant effect is found for the presence of a
partner or life satisfaction. This suggests that for a given number of children the larger
the gap between the first birth and the birth of the cohort member the more likely that at
interview the partner is employed, the mother is employed and the higher the net family
income.
Ethnic differentials
This paper has controlled for, but not focused on, the very distinctive family formation
patterns (and economic circumstances) of the main ethnic minority groups in the UK.
Pakistani and Bangladeshi women have a high propensity to start motherhood young
within marriage, and a low propensity to be employed. Black groups tend to show the
opposite. Low employment rates of men in all the ethnic minority groups contribute to
low incomes. These disadvantages are associated with the areas in which ethnic groups
are concentrated, which formed one of the survey strata in England, but the fuller model
26