So I got a question through Aardvark [1] asking if there's a connection between women's life expectancy and having children.

My automatic respose would be that having children decreases your life expectancy in the immediate range, because it can kill you.  Abortion is safer than pregnancy, so it would stand to reason that not even getting an abortion, just never getting pregnant, would be even safer.  Short-term, not having a baby is safer.

The long-term effects of having children could be positive, however.  If you're never pregnanty, you get more estrogen pumping around so you are more likely to get breast cancer.  If you have children, you may have an adult child looking after you when you're older so they might help you catch a problem that would kill you if you ignored it.

I couldn't find any statistics on life expectancy for a woman broken down by number of children.  I was surprised — I expected a chart with columns for "number of children" and rows for age, and the life expectancy for a woman with those characteristics. 

After all, we get statistics on how smoking, eating omega-3, even skydiving affects your life expectancy.  Yet this very essential and common experience isn't included?  Perhaps because it's women doing it, and doctors and epidemiologists assume that women would decide not to have babies if they knew the risks?

The data has to be out there.

[1] you can sign up to answer questions from your friends and from others who have questions in your area of expertise.

UPDATE
A friend found this:

Reproduction and Lifespan: Trade-offs, Overall Energy Budgets,
Intergenerational Costs, and Costs Neglected by Research
GRAZYNA JASIENSKA*

which tells us

In human females allocation of resources to support reproduction may cause their insufficient supply to other metabolic functions, resulting in compromised physiology, increased risks of diseases and, consequently, reduced lifespan. While many studies on both historical and contemporary populations show that women with high fertility indeed have shorter lifespans. This relationship is far from universal: a lack of correlation between fertility and lifespan, or even an increased lifespan of women with high fertility have also been documented.  Reduced lifespan in women with high fertility may be undetectable due to methodological weaknesses of research or it may be truly absent, and its absence may be explained from biological principles. I will discuss the following reasons for a lack of the negative relationship, described in some demographic studies, between the number of children and lifespan in women: (1) Number of children is only a proxy of the total costs of reproduction and the cost of breastfeeding is often higher than the pregnancy cost but is often not taken into account. (2) Costs of reproduction can be interpreted in a meaningful way
only when they are analyzed in relation to the overall energy budget of the woman. (3) Trade-offs between risks of  different diseases due to reproduction yield different mortality predictions depending on the socio-economic status of the studied populations. (4) Costs of reproduction are related not only to having children but also to having grandchildren. Such intergenerational costs should be included in analysis of trade-offs between costs of reproduction and longevity.

In other words, "it depends."  The author goes into good detail on the physical energy cost of childbearing and lactation and child care, hormonal issues which provide some protection against later cancers, personal interactions with children and grandchildren as a positive life extender but the countervailing tendency for grandfathers to be granted more resources out of respect while grandmothers tend to be assigned work and tend to decrease child mortality for their grandchildren.  She points out that there is a lack of good historical data on lactation duration.  More children tends to correlate with higher rates of cardiovascular disease and diabetes and sometimes obesity. 

Though the last is least worrisome it could, of course, cause some women to decide in this thinness obsessed world to avoid pregnancy. 

One interesting finding she relates is that short-term partial breastfeeding, as is often done in industrialized societies, doesn't provide as much protection as "full on" breastfeeding.  More ammo for the breastfeeding in public crowd — it's not just a moral right, it's a health benefit.

Another friend found a paper entitled, "LIFE EXPECTANCY AS AN INTEGRATING CONCEPT IN SOCIAL AND DEMOGRAPHIC ANALYSIS AND PLANNING" by Dudley Seers, Institute of Development Studies, University of Sussex.  This kind of data structuring is suggested as useful for reports, but it doesn't actually contain any such reports.