The total fertility rateREF in the United States has been far below the replacement fertility rate for decades.REF Over time, sub-replacement fertility can lead to population aging and contraction if net migration is negative or insufficient to counteract sub-replacement fertility. Population aging and slower population growth due to sub-replacement fertility can have important implications for economic growth, the size of the labor force, government revenue, and the solvency of Social Security and Medicare.
Women in the U.S., on average, are on pace to have fewer children at the end of their childbearing years than what Americans generally consider to be ideal for their family size. This suggests that U.S. adults on average may not have been able to get on track to have as many children as they would have liked.
While the period total fertility rate has fallen below the replacement level, the number of children Americans say are ideal for a family to have has consistently remained above replacement. If Americans could merely achieve their personal ideal family size, the total fertility rate in the U.S. would likely rise and could potentially rise above the replacement level.
Charts 1–3 show the differences between Americans’ ideal number of children, Americans’ intended number of additional children given the children they already have, and the projected number of children women are on pace to have by the end of their reproductive lifetimes.
Fertility and Ideal Family Size
For most of the past century, a gap has existed between the total fertility rate and ideal family size. Since 1933, Gallup has polled Americans on the question “What do you think is the ideal number of children for a family to have?” The General Social Survey (GSS), administered by the National Opinion Research Center (NORC) at the University of Chicago, has asked the same question since 1972. Chart 1 shows ideal family size, intended number of children, the total fertility rate, and the replacement fertility rate plotted against time.REF It illustrates how close Americans’ ideal fertility preferences are to the number of children they are currently on track to have. The results from the Gallup and GSS surveys have generally agreed with one another. The error bars for the GSS series show the 95 percent confidence intervals for the mean responses.
Prior to 1968, Americans’ ideal family size had been above 3.25 children. Preferences shifted sometime after Gallup’s 1967 survey. By 1977, the ideal number of children had fallen sharply to 2.6. Ideal family size has remained more or less constant since then.
Since 1965, Americans’ ideal family size has been higher than the period total fertility rate. This suggests that many Americans have not been able to have the number of children that they would ideally like to have.
Fertility and Intended Family Size
Gallup and the GSS ask U.S. adults what they believe is the ideal number of children for a family to have generally. This is a very different question from asking respondents directly how many children they plan on having or that they personally would like to have. It would be more useful to have data on personal ideal family size because it more directly measures desired fertility than the general ideal family size. However, data on the personal ideal number of children are scant.
Another way to measure desired fertility is by asking women how many children they and their partner (if any) intend on having. Unlike general ideal family size, asking women about their intended number of children measures how many children they plan on having themselves rather than how many children they think families should have in general.
Chart 1 shows the mean number of children that women of reproductive age intend to have over their lifetimes.REF The error bars show the 95 percent confidence intervals for the responses.
Data from this series comes from the National Survey of Family Growth (NSFG),REF a nationally representative survey of U.S. adults of reproductive ageREF that asks respondents about marriage and cohabitation, childbearing, fertility intentions, and sexual activity.
Mean intended number of children here is defined as the number of children women intend on having given the number of children they already have, minus unwanted live births.REF It is not a perfect approximation for the personal ideal number of children. But comparing the mean intended number of children with the general ideal number of children provides more information on Americans’ personal fertility desires.
The mean intended number of children from the NSFG is significantly lower than the general ideal number of children from Gallup and the GSS. This is because the mean intended number of children captures both (1) the ideal number of children women would like to have and (2) the number of children that women believe that they can realistically have.
For women who perceive few barriers to their ability to achieve their ideal family size, mean intended number of children probably more closely resembles personal ideal family size. Most women, however, likely report an intended number of children lower than their personal ideal because they make realistic adjustments for financial constraints, current career demands, the struggles of dating and parenthood, or advancing age. Additionally, asking respondents about “ideal” family size itself implies an abstraction from practicalities, whereas asking respondents about “intended” family size implies the number of children that respondents realistically plan on having.
As a result, the mean intended number of children is closer to the total fertility rate than to the general mean ideal number of children. Since the 2011–2013 NSFG survey, the mean intended number of children has declined significantly. While the mean intended number of children was not statistically different from the replacement fertility rate in the 2011–2013 survey, by the 2017–2019 survey, the mean intended number of children was significantly lower than replacement fertility. This is an indication that U.S. adults no longer even intend to have the number of children that would constitute replacement fertility.
Personal Ideal Number of Children vs. General Ideal Number of Children
Ideal family size is not the same concept as intended family size. As explained above, there are many reasons why the mean intended number of children falls short of providing an accurate measure of personal ideal family size. But how good of an approximation is general ideal family size for personal ideal family size?
In Europe, the Eurobarometer survey has, on multiple occasions, asked respondents a question about general ideal number of children along with a question about their personal ideal number of children. Data from the 2006 and 2011 Eurobarometer surveysREF show that men’s and women’s mean general ideal number of children correlate strongly with mean personal ideal number of children (Pearson’s correlation coefficient for women: r = 0.90, 95% C.I.: (0.84, 0.94); for men: r = 0.91, 95% C.I.: (0.85, 0.95)). The differences between mean general ideal number of children and mean personal ideal number of children for each country were also rather small (–0.01 median difference for women and +0.02 median difference for men). Therefore, the general ideal number of children is likely to serve as a good proxy for the personal ideal number of children.
The Gap Between Fertility and Ideal Family Size
The strong correlation between personal and general ideal family size suggests that general ideal number of children is a better measure of personal ideal family size than the mean intended number of children. What, then, can be inferred from Chart 1?
Since 2008, the fertility rate in the U.S. has been steadily declining. This is not because Americans want fewer children. The ideal family size has remained roughly constant since 2008.
As a result, the gap between the total fertility rate and ideal family size is widening. Since the Great Recession, fewer and fewer Americans have been able to have the number of children that would put them on pace to achieve their ideal family size.
As evident from Chart 1, the mean intended number of children is much lower than the ideal number of children, an indication that Americans have largely given up on achieving their ideal family size. U.S. adults on average do not even intend to have the number of children they would have ideally wanted. The clear downward trend in the mean intended number of children shows that U.S. adults are increasingly revising their expected family size down, even as ideal family size has remained roughly unchanged.
Americans’ general ideal family size has always been well above the replacement fertility rate. If Americans could merely achieve their ideal family size, there is evidence to believe the fertility rate could rise above the replacement level.
Breakdown of Responses for Ideal Number of Children
The mean ideal number of children shows the ideal family size for U.S. adults on average. The trend can also be broken down by the ideal number of children by family-size type.
Chart 2 breaks down the responses for ideal number of children over time using the Gallup series. Comparing Chart 2 to Chart 1, it becomes evident that the sharp decline in desired fertility during the late 1960s and early 1970s (Chart 1) is largely due to the emergence of a two-child norm and a transition away from families with four or more children (Chart 2).
Prior to 1975, two-child, three-child, and four-child families were most commonly considered the ideal, all in roughly equal measure. Larger families with five or more children were also much more commonly considered ideal than they are today.
After 1975, the percentage of Americans who considered the two-child family to be ideal sharply rose to more than 50 percent, while selection of all other family sizes, except childless and one-child child families, declined. Since the late 1990s, the two-child family has lost some of its dominance as more Americans now say that the three-child family is ideal. The four-child ideal family is also trending up slightly. Today, a plurality of Americans consider the two-child family to be the ideal family size, and nearly 30 percent of Americans consider the three-child family to be ideal.
How does Americans’ ideal number of children compare to the actual number of children U.S. women end up having by the end of their reproductive lifetimes? Chart 3 shows the Gallup poll responses for the ideal number of children in 2018 plotted alongside the estimated 2018 percentages of U.S. women between the ages of 40 and 44 who had attained each family size.
Data for the share of women 40 to 44 years of age attaining each family size are taken from the 2022–2023 NSFG.REF Number of children here is defined as the total number of live births women have had over their lifetime, so this ignores infant and child mortality and children given up for adoption. Given the low levels of infant and child mortality in the U.S., mortality can be excluded here. The error bars for number of children attained are the 95 percent confidence intervals for the point estimates.
Chart 3 shows that far more women end up without children or with only one child than what women generally consider to be ideal. It can be inferred that the vast majority of women who end up without children by ages 40 to 44 would have wanted children, and that the vast majority of women who have only one child would have wanted more.
Similarly, the percentages of women with two, three, and four children are lower than the percentages of women who want two, three, or four children. This suggests that many women who want two, three, or four children are not able to have them.
However, there is no statistical evidence that the percentages of women who want five or more children are any different from the percentages of women who have five or more children. This is evident from the fact that the point estimates for the desire for at least five children lie within the 95 percent confidence intervals for the attained number of children. This means that women who want large families are generally able to have the number of children they wanted. But women who want families with two or three children have, in many cases, not been able to have them.
Considerations for Public Policy
Americans’ family size preferences on average have not changed much since 1975, but since the Great Recession, U.S. adults have had a harder time achieving their desired family size. In fact, many U.S. adults appear to have given up on attaining ideal family size and have revised their expectations.
If Americans decide that it is in the interest of society to craft public policy to address below-replacement fertility, it is not enough for policy to merely help U.S. adults achieve their intended number of children. It is clear that Americans’ mean intended number of children is below replacement fertility. Rather, such public policy must help to address the obstaclesREF that U.S. adults face in reaching their family size goals and dreams.
Jonathan Abbamonte is Senior Research Associate for the Center for Data Analysis at The Heritage Foundation.
Appendix: References for Charts 1–3
Chart 1
Ideal Number of Children (Gallup):
