A nurse bathes a newborn baby.

Sex, contraception, or abortion? Explaining class gaps in unintended childbearing

Richard V. Reeves and Joanna Venator

There are wide class gaps in unintended childbearing among single women in the United States, resulting from different contraceptive and abortion choices across income groups. In this paper, we use data from the National Survey of Family Growth (NSFG 2011-2013) to estimate how sexual activity, contraceptive use, and abortion use vary across income lines. Though rates of sexual activity are comparable for all women, low-income women are less likely to use contraception and are less likely to have an abortion once pregnant. We use a shift-share analysis to simulate the effect of equalizing, at the rates of affluent women, the use of the contraception and abortion services across income groups. We find that equalizing contraceptive use reduces the ratio of unintended births between affluent and poor women by half, and that equalizing abortion rates reduces the ratio by one-third.


There are large and widening gaps in rates of unintended childbearing between Americans
on different rungs of the income ladder. A poor woman is more than five times as likely as an
affluent woman to have an unintended birth. Since unintended childbearing is associated with
higher rates of poverty, less family stability, and worse outcomes for children, these gaps further
entrench inequality. Closing gaps in unintended childbearing is therefore important for greater
equality and opportunity.

But to close the gaps, we must first understand them. In most cases, a couple must take three
steps to have a child. First, have sex. Second, fail to use contraception, or use contraception
unsuccessfully. Third, decide to proceed with the pregnancy rather than have an abortion.
Income gaps in unintended childbearing must therefore reflect gaps in sexual activity,
contraceptive efficacy, rates of abortion, or—more likely—some combination of the three.
Gaining a better appreciation of the relative importance of these income gaps may assist in the
formulation of policies aimed at reducing unintended childbearing, especially for those with low

Premarital sex has been the social norm for decades, and sexual activity rates among
unmarried Americans do not vary along class lines. There is no ‘sex gap’ by income. By
contrast, lower-income individuals typically use contraception less frequently, and less
successfully, than those with higher incomes. This is why so much attention has correctly
been paid to the role that contraception—or the lack of contraception—plays in unintended
childbearing, not least by our colleague Isabel Sawhill in her latest book, Generation Unbound:
Drifting Into Sex and Parenthood Without Marriage, and by the team of researchers at Child
Trends. The ‘contraception gap’ is a big part of the story. But it is not the whole story: less
affluent women are also less likely to get an abortion when faced with an unintended pregnancy. How much do these gaps matter in terms of explaining variation in unintended birth rates by

Related Books