Distributional Aspects of Human Fertility

Lutz, W. ORCID: https://orcid.org/0000-0001-7975-8145 (1989). Distributional Aspects of Human Fertility. London: Academic Press. ISBN 0-12-460470-6

Preview

Text XB-89-602.pdf Download (5MB) | Preview

Abstract

This book opens a new window on human fertility. Most of the literature, both popular and scholarly, deals with averages, as indicated by the observation that for long-term stationarity we need 2.1 children per woman at present low mortality rates. This could mean that 90% of women have two children and 10% have 3; it could alternatively mean that 30% have no children and 70% have 3. It could even mean that a few women have 5 to 10 children, and most have none. To enact policy that aims to raise or lower the birthrate it makes a great deal of difference which distribution applies, and on this the mean provides little information.

This question of distribution is conveniently discussed in this book in terms of parity progression: of women who have one child, what fraction goes on to have a second child; of those who have a second child, what fraction goes on to have a third; and so forth. We know that in all countries a high proportion of women try to give birth to at least two children; of those who try to bear two, what fraction goes on to have a third is in one sense the essence of the question of replacement fertility in the developed countries. One cannot imagine a policy succeeding without its main effect being to influence omen to want three children rather than two.

Distribution is important for the interpretation of cohort and period fertility data. In the former, one follows statistically groups of women born at the same time through their childbearing careers; in the latter, one tries to draw conclusions from the childbearing behavior of a cross section of women observed in a particular calendar year or over a period of years. The cohort information discloses more about fertility than does the period data; however, the period data is more up to date, by about 30 years. To study thoroughly human fertility we need the averages and distributions of both.

Actions (login required)

View Item