Later Baby

Movement towards equal pay for women caused births to fall 8% during 1975-1978, claimed Research Fellow John Ermisch in a lunchtime talk on 8 February. Women's hourly earnings rose from 63% of men's in 1970 to 74% in 1977, and this increase in relative earnings caused women to postpone having their first child. Higher men's earnings, however, accelerated the early stages of childbearing through their effect on family income.

Ermisch observed that successive generations of women have shown a stronger attachment to paid employment. This strong attachment has led women to minimize their absence from the labour market, by compressing childbearing into their late twenties and early thirties. Ermisch's research also indicates that higher rates of male unemployment were associated with lower birth rates among childless women under 25 and among women in their late thirties.

Ermisch drew attention to the large swings in the birth rate in Britain over the past 50 years. He noted that these fluctuations in births will affect the age distribution of the population well into the next century. Changes in the age distribution have important implications for student numbers as well as youth earnings and unemployment. The age distribution also affects pressures in the housing market, health services and the state pension scheme.

Ermisch discussed the results of research done jointly with Heather Joshi and Eric De Cooman and reported in detail in CEPR Discussion Paper No. 37 , "The Next Birth and the Labour Market: A Dynamic Model of Births in England and Wales'. They investigated the extent to which economic developments explained fluctuations in births. Current wages and unemployment rates affect the decision to have a child by influencing current income and the costs of childbearing and also the couple's expectations about their future income and costs.

Labour market developments played a particularly important role in explaining birth rates, according to Ermisch. Higher wages mean higher income; thus a couple can afford more children. Higher wages also mean more income is lost by those caring for additional children. Since most child rearing is done by the mother, higher men's wages mainly affect childbearing through the couple's income, while higher women's wages raise the cost of children, by increasing the earnings foregone by the mother. Higher income also means that the couple can spend more in raising each child, and thus each child can cost more.

Short-term fluctuations in birth rates arise largely because of changes in the timing of first and second births. Ultimate family size has fluctuated much less than annual birth rates. Changes in family size are the result of longer-term trends in the birth rates of third and fourth children. Ermisch noted, however, that results for third and fourth births should be treated cautiously because of the relatively short period of the analysis.

The data suggested that increases in women's wages postponed births. This was primarily an effect on the timing of first and second births, which were delayed several years. Declines in women's wages would bring these births forward.

Ermisch also found that higher men's wages accelerate the early stages of childbearing by raising family income. These increases in men's wages were associated with more first and second births to women under the age of 30. Increases in men's wages reduced births among mothers of one aged 30 and over, and among mothers of two and three aged over 35. This effect is consistent with higher income having raised the desired expenditure on each child, so an additional child costs more.

Increases in men's unemployment were found to reduce births among childless women aged under 25 and among women in their late thirties with three children. The wives of men in lower skill occupations constitute a large proportion of the latter group of women, and that may account for the sensitivity of their childbearing to unemployment.

Births fell by 35 per cent between 1964 and 1977, but the decline was steepest after 1971. There was a recovery in births in the late 1970s. Ermisch had examined this period closely. Birth rates during the 1970s could be simulated accurately using the estimated economic model. The simulations were then used to examine how fluctuations in economic growth and other economic variables affected births during the 1970s.

One important development was the rise in women's wages relative to men's. Women's hourly earnings rose from 63% of men's in 1970 to 74% of men's in 1977. In large part this was due to the introduction of the Equal Pay Act. The simulations suggest that births would have been 8 per cent higher during 1975-78 if this rise in women's relative earnings had not occurred. The main effect was on the timing of first births and not on family size. There was a postponement of first births during the mid-1970s, and a recovery in first births during 1978-80, in reaction to this earlier postponement. Without this rise in women's relative earnings first births would have fluctuated much less, and total births would not have fallen to such a deep trough in 1977.

Ermisch also argued that the rate of second and higher order births have considerable momentum. They only begin to be affected by current economic developments four years later. This lagged reaction facilitates short-term forecasting of births. There had been changes in the definitions of some data series in 1981. Work was currently in progress using the new data in order to forecast current birth rates.

The research reported by Ermisch was based on aggregate time series data for England and Wales. This made it difficult, Ermisch acknowledged, to capture effects such as the increasing availability of child care or changes in the provision of housing, which might have important effects on couples' child-bearing decisions. The effects of social class were also difficult to capture since data on birth registrations by social class were only available for the 1970s. He outlined plans to use the Women and Employment Survey, which contained complete demographic and employment histories at an individual level. Analysis of such microdata could provide a fuller understanding of the influences on the UK birth rate.