Proposal preview

Women’s early life conditions and later-life outcomes

Social, economic and environmental experiences in early life can have large and lasting effects on human capital. Negative shocks to a child’s environment can permanently alter the trajectory of psychological and physical development. More broadly, consistent exposure over several years to different environments than peers can induce lasting long-term differences in domains as different as occupational or educational attainment, earnings, stature, and reproductive history. That is, adult economic behavior and outcomes are not simply the result of rational choices made at the time, but influenced by past choices made by parents. Much of the research on early life conditions and later life outcomes, particularly on cohorts born before World War II has focused on men, because men’s lives are easier to reconstruct from historical data. This panel shifts attention to women’s early life conditions, with particular attention to differences from male patterns.

Organizer(s)

  • Evan Roberts University of Minnesota eroberts@umn.edu United States

Session members

  • Sakari Saaritsa, University of Helsinki
  • Maarit Olkkola, Barcelona Graduate School of Economics
  • Evan Roberts, University of Minnesot
  • Wendy Rahn, University of Minnesota
  • DeAnn Lazovich, University of Minnesota
  • Matthias Rosenbaum-Feldbrügge, Radboud University
  • Claudia Goldin, Harvard University
  • Adriana Lleras-Muney, UCLA

Discussant(s)

  • Joseph Ferrie Northwestern University ferrie@northwestern.edu

Papers

Panel abstract

Social, economic and environmental experiences in early life can have large and lasting effects on human capital. Negative shocks to a child's environment can permanently alter the trajectory of psychological and physical development. More broadly, consistent exposure over several years to different environments than peers can induce lasting long-term differences in domains as different as occupational or educational attainment, earnings, stature, and reproductive history. That is, adult economic behavior and outcomes are not simply the result of rational choices made at the time, but influenced by past choices made by parents. Much of the research on early life conditions and later life outcomes, particularly on cohorts born before World War II has focused on men, because men's lives are easier to reconstruct from historical data. This panel shifts attention to women's early life conditions, with particular attention to differences from male patterns.

1st half

The Role of Public Health Nurses, Midwives and Universal Preventive Health Services

Maarit Olkkola

For the last fifty years Finland has had one of the world’s lowest infant mortality rates. In the 1930s and 1940s, however, Finland was lagging behind most of Western Europe and the United States. A dramatic decline in infant mortality took place in late 1940s and early 1950s, and by the 1970s Finland had taken the lead together with the Scandinavian countries. This was also the period when Finland introduced and expanded many universal preventive health services targeting children and pregnant women. I use different kinds of variation in the implementation of universal preventive health services to quantify their contribution to the rapid infant mortality decline. First, I document the roll-out of universal pre-natal maternity services and well-child visits for children up to age 7. Second, I use population threshold discontinuities in the laws on municipal midwives and public health nurses. Third, I interact the roll-outs with the population discontinuities....

For the last fifty years Finland has had one of the world’s lowest infant mortality rates. In the 1930s and 1940s, however, Finland was lagging behind most of Western Europe and the United States. A dramatic decline in infant mortality took place in late 1940s and early 1950s, and by the 1970s Finland had taken the lead together with the Scandinavian countries. This was also the period when Finland introduced and expanded many universal preventive health services targeting children and pregnant women. I use different kinds of variation in the implementation of universal preventive health services to quantify their contribution to the rapid infant mortality decline. First, I document the roll-out of universal pre-natal maternity services and well-child visits for children up to age 7. Second, I use population threshold discontinuities in the laws on municipal midwives and public health nurses. Third, I interact the roll-outs with the population discontinuities. Fourth, I use the invention of new preventive health products, such as vaccines, to describe how the role of preventive health services changed over time.

Early life excess female mortality, tuberculosis and development Evidence from Finnish population statistics, 19th - 20th c.

Sakari Saaritsa

Excess female mortality in age groups from late childhood till the end of the reproductive years has been observed across many populations in the 19th and 20th centuries. While the proximate cause in most cases seems to have been pulmonary tuberculosis (TB), the root causes for this important and constant observation remain contested. In an evolving literature, intrahousehold gender discrimination, unspecified physiological reasons and changing patterns of contact and care have recently been discussed. Data from Finnish population statistics indicates a stable EFM from TB already in the age group 5 to 9 from the 1880s to the 1930s. While this ratio seems largely unresponsive and unrelated to gendered structural changes occurring over the period in terms of employment, growth, industrialization and urbanization, it is equally incommensurable with epidemiological explanations based on puberty or sexual contact. This paper maps conditioning factors such as regional differences in morbidity, livelihoods and demographic...

Excess female mortality in age groups from late childhood till the end of the reproductive years has been observed across many populations in the 19th and 20th centuries. While the proximate cause in most cases seems to have been pulmonary tuberculosis (TB), the root causes for this important and constant observation remain contested. In an evolving literature, intrahousehold gender discrimination, unspecified physiological reasons and changing patterns of contact and care have recently been discussed. Data from Finnish population statistics indicates a stable EFM from TB already in the age group 5 to 9 from the 1880s to the 1930s. While this ratio seems largely unresponsive and unrelated to gendered structural changes occurring over the period in terms of employment, growth, industrialization and urbanization, it is equally incommensurable with epidemiological explanations based on puberty or sexual contact. This paper maps conditioning factors such as regional differences in morbidity, livelihoods and demographic patterns in order to explore the consistency of the finding.

The impact of parental and sibling death during childhood on women’s ages at marriage in the Netherlands, 1850-1940

Matthias Rosenbaum-Feldbrügge

Life history theory predicts that the exposure to mortality during childhood reduces female age at pubertal maturation, age at first sex, age at marriage and age at first reproduction. Studies on contemporary populations in Western European countries indeed confirm this prediction empirically. Hardly any studies, however, have tested if the proposed association between mortality exposure in the family and younger ages at marriage also holds in a historical population. For this purpose, data from the Historical Sample of the Netherlands (HSN) is exploited. The HSN contains roughly 18,000 life courses of female individuals born all over the Netherlands between 1850 and 1922. Reliable marriage information is available until 1940. Event-history analysis is employed in order to test for associations between parental and sibling death during childhood and ages at marriage. Preliminary results suggest that sibling death is indeed related to reduced female age at marriage, while parental death is not...

Life history theory predicts that the exposure to mortality during childhood reduces female age at pubertal maturation, age at first sex, age at marriage and age at first reproduction. Studies on contemporary populations in Western European countries indeed confirm this prediction empirically. Hardly any studies, however, have tested if the proposed association between mortality exposure in the family and younger ages at marriage also holds in a historical population. For this purpose, data from the Historical Sample of the Netherlands (HSN) is exploited. The HSN contains roughly 18,000 life courses of female individuals born all over the Netherlands between 1850 and 1922. Reliable marriage information is available until 1940. Event-history analysis is employed in order to test for associations between parental and sibling death during childhood and ages at marriage. Preliminary results suggest that sibling death is indeed related to reduced female age at marriage, while parental death is not associated with age at marriage. This indicates that other factors than parental death such as inheritance practices, religious denomination, place of residence and social class are more useful to explain female marriage behavior in the Netherlands in the period under consideration.

2nd half

The effects of early-life conditions on later-life health and mortality in a cohort of American women

Evan Roberts, Wendy Rahn, DeAnn Lazovich

Adverse early-life social conditions are associated with higher mortality in old-age in American men, but these effects have not been studied to the same extent in American women. We add new evidence on the long-run effects of early-life social conditions and childhood structure in the United States in the United States by studying a large cohort of women. We linked 10,375 subjects from the Iowa Women’s Health Study (IWHS) born 1916-1930 to individual early-life census records from the 1930 census. Census records provided objective measures of family structure and parental occupations in early life. The subjects matched to early-life censuses were representative of both girls their age living in Iowa at the 1930 census, and of the IWHS cohort at enlistment. Conditional on survival to IWHS enrolment in 1986 we found the following early-life factors associated with a worse survival experience in later life: growing up in an urban area,...

Adverse early-life social conditions are associated with higher mortality in old-age in American men, but these effects have not been studied to the same extent in American women. We add new evidence on the long-run effects of early-life social conditions and childhood structure in the United States in the United States by studying a large cohort of women. We linked 10,375 subjects from the Iowa Women’s Health Study (IWHS) born 1916-1930 to individual early-life census records from the 1930 census. Census records provided objective measures of family structure and parental occupations in early life. The subjects matched to early-life censuses were representative of both girls their age living in Iowa at the 1930 census, and of the IWHS cohort at enlistment. Conditional on survival to IWHS enrolment in 1986 we found the following early-life factors associated with a worse survival experience in later life: growing up in an urban area, having a father employed in manufacturing, lower levels of education, and maternal absence before age 6.

XX>XY? The Changing Female Mortality Advantage

Claudia Goldin, Adriana Lleras-Muney

Women live longer than men in most parts of the world today. Among OECD nations in recent years, the difference in life expectancy at birth has been around four to six years (seven in Japan). But women have not always lived that much longer than men. This paper examines when women began to live much longer than men and what caused the advantage to have widened then. We use mortality data from Massachusetts and find that when infectious disease was rampant, female children and youth were affected more than male children and youth. We hypothesize that the scarring effect from greater infectious disease morbidity led to more deaths at older age for females than for males. With the large decrease in infectious disease among youth in the early twentieth century, the female advantage began to appear for these cohorts as they aged in the mid-twentieth century and beyond.

Women live longer than men in most parts of the world today. Among OECD nations in recent years, the difference in life expectancy at birth has been around four to six years (seven in Japan). But women have not always lived that much longer than men. This paper examines when women began to live much longer than men and what caused the advantage to have widened then. We use mortality data from Massachusetts and find that when infectious disease was rampant, female children and youth were affected more than male children and youth. We hypothesize that the scarring effect from greater infectious disease morbidity led to more deaths at older age for females than for males. With the large decrease in infectious disease among youth in the early twentieth century, the female advantage began to appear for these cohorts as they aged in the mid-twentieth century and beyond.