Trends in marriage and fertility by race in the United States

Last Updated on December 10, 2023

This is a brief data post illustrating trends in marriage and fertility by race during all years for which data is available. I decided to make this post after failing to find useful charts clearly documenting this information (existing charts had various problems, such as limited time periods, not splitting data by race, marital status, etc.). So I constructed the charts that I wanted by piecing together information from different publications of CDC and Census data. All of the data is (at the moment) publicly available online and all of my charts/calculations are available in a Google spreadsheet that is cited in each section.

Out-of-wedlock births and single-mother households


I was first interested in a chart illustrating rates of out-of-wedlock births (i.e. percent of births to unmarried mothers) by race over time. Unlike the other charts in this post, I was able to find an existing chart [archived] used in several Wikipedia pages documenting this information. However, that chart ends in 2014 and there are many years of missing data for certain racial groups (e.g., Native Americans, Asians, etc.) despite such data being available in various CDC reports. Anyway, here’s an updated chart with data from 1940 to 2021:

Sources:

As you can see, from about the 1960s to early 2000s, there was a sharp increase in percent of births to unmarried mothers, particularly for blacks, American Indians, and Hispanics. There has been increases for whites and Asians as well, but to a much lower magnitude.

There are a few important things to note about this data:

  • Prior to 1969, there was no data available for specific non-white races. Instead, data was simply published for “whites” and “non-whites”. We can reasonably assume that data for “non-whites” indicates data for blacks, because the vast majority of non-whites were black during this time period. For example, according to Census Data, 96% (12,865,518 out of 13,454,405) of non-Whites were black in 1940. 89% (22,580,289 out of 25,462,951) of non-Whites were black in 1970.
  • Prior to 1990, data for “whites” included Hispanics. Data for non-Hispanic whites specifically was not published until 1990.

These two points are important to keep in mind for the remaining data below, as they apply to much of the sources used in this post.

As one could expect from the above post, there has also been a very large increase in the percent of children living single-mother households over the same time period.

There’s also been a corresponding decrease in the percent of children in two-parent households (whether married or unmarried).

Sources:

There seems to have been a steady increase in single-mother households from 1960 to the 1990s, with the increase being slightly greater for black households compared to white households. Since the 1990s, the rate has mostly stabilized. In fact, recent years suggest a possible decrease in single-mother households since 2015, particularly for blacks.

It’s not clear if Hispanics were included in the “white” category in this data. However, this shouldn’t distort the data too much if so (see later comments on how the distinction between “white” and “non-Hispanic white” has little impact on marital status).

The above charts show large racial differences in rates of single-mother household rates. The percent of black children in single-mother households has been about 2-3 times the same percent of white children for the past several decades. However, this actually understates larger racial differences. For white children in single-mother households, the mother tends to be divorced. However, for black children, the mother tends to be never married.

For example, Vespa et al. (2013) [archived] showed that, among family groups maintained by white mothers in 2012, 47% (2,506 out of 5,371) of mothers were divorced and 31% (1,649 out of 5,371) were never married (Table 6). Among family groups maintained by black mothers, 20% (692 out of 3,545) of mothers were divorced and 64% (2,262 out of 3,545) were never married (Table 6). So, not only are black families more likely to be maintained by a single mother than white families, black single mothers are vastly more likely to be never married than white single mothers. In fact, the absolute number of black single, never-married, mothers is considerably greater than the absolute number for white mothers (2,262 vs 1,649).

Thus, when looking at the percent of children in never-married single-parent households, we see even larger racial disparities than when looking at just percent of children in single-parent households simpliciter:

Note: these statistics include Hispanics in the “White” category before 2001.

Sources:

Marital status


Now, consider trends in marriage throughout the 20th century. In the first section, I’ll focus on the rates of marriage. In the next section, I’ll look at rates of marital separation specifically, since I noticed interesting drops in separation following the 1960s which is worth focusing on. All of the charts focus on women aged 15 years or older.

Rates of marriage

I’ll start by presenting marital rates for women by race from 1890 to 2022. I focus specifically on women since that is consistent with other sources I’ve seen presenting similar data. Here are the rates for white women:

Here are the rates for black women:

As you can see, from 1890 until 2022, it appears that rates of marriage peaked in the 1950s and has since steadily decreased for both black and white women. However, there are two differences between black and white women worth noting:

  • Congruent with the other trends presented so far, the decrease in marriage since the 1950s has been much greater for blacks than whites. There has been about a 14 percentage point decrease in marriage rates for white women since the peak in 1960 (from about 67% in 1960 to about 53% in 2022). By contrast, there has been about a 30 percentage point decrease in marriage rates for black women since the peak in 1950 (from about 62% in 1950 to about 32% in 2022).
  • For white women, the increase in the percent of women divorced and in the percent of women never married both equally contribute to the decrease in marriage rates. Since 1960, there has been about a 9 percentage point increase in both the percent of white women divorced and the percent of white women never married. The story is different for black women. For black women, the increase in never married women is the predominant explanation for the decrease in marriage rates. Like white women, there has been about a 8-9 percentage point increase in percent of black women divorced since 1950/1960. However, there has been about a 28 percentage point increase in the percent of black women never married (from 21% in 1950 to 49% in 2022).

The data for Hispanic and Asian women does not go back nearly as far as the data for black and white women. However, it may still be worth considering. Here is the data for Hispanic women:

And the data for Asian women:

Interestingly, while all racial groups have seen drops in marriage rates since 1990, Asians have remained remarkably stable, with about 62% of Asian women remaining married over the entire time period. Hispanics seem to have rates of marriage that mirror that of whites, although they have slightly higher never married rates and slightly lower divorce rates.

To aid in comparisons across race, I created charts showing percent of married women and percent of never married women by race. The following shows the percent of married women by race:

As you can see, black and white women had very similar rates of marriage until around 1950. Between 1890 and 1950, the gap in marriage rates for black and white women was consistently 2 to 4 percentage points. The gap has since grown and stabilized to about 20 percentage points since the mid 90s.

And here’s the chart showing the percent of women never married by race:

The change in racial patterns has been even more dramatic when looking at percent of women never married. Until 1950, black women were less likely to be never married than white women (meaning: black women were more likely to have been married at some point). In 1960, the percentage of never married black women was just 3 percentage points greater than the percent for white women. The gap has grown and stabilized at around 20 percentage points since the mid 90s.

Notes:

  • For the 1890 to 1950 time period, I inferred the rates for black based on data reported for “non-whites”. As I’ve shown earlier, the vast majority of non-whites during this period were black.
  • The data here includes Hispanics as “whites”. However, this shouldn’t distort the data too much as marital rates for “whites” should be similar regardless of the inclusion of Hispanics. For example, in 2022, “whites” and “non-Hispanic whites” both had a similar percent of individuals who were currently married (50.9% and 52.1%, respectively). Both groups also had a similar percent of individuals who were never married (30% vs 28.1%). The difference in earlier years would be much lower since there would have been far fewer Hispanics.

Sources for charts in this section:

Rates of separation

The similar rates of marriage between black and white women before 1960 may be misleading, as it hides the fact that married black women had far higher rates of separation compared to married white women during this time. This is displayed in the following chart:

Among white women, about 2 to 2.5% were currently married and separated from their spouse from 1960 until 2010. However, among black women, there has been significant decreases in the rate of separation since the 1970s, with the percent decreasing from a high of over 10% in 1976 to just under 5% in 2014. Note that these are percentages of all women aged 15 or older, not a percentage of married women specifically.

When rates of spousal absence are taken into account, the black and white marital relations do not appear so similar even back in 1960. For example, the following chart shows percent of women married with their spouse present by race since 1960.

As you can see, even back in 1960, there were large racial differences in the percent of women married with their spouse present. About 64% of white women were married with a spouse present compared to only about 49% of black women. Thus, the gap was initially 15 percentage points in 1960, and has since grown to about 25 percentage points in 2014. This change is much smaller than the chance implied when one just looks at rates of marriage by race.

Rates of separation were particularly large for black women between the ages of 25 and 44. To illustrate this, the following chart shows the percent of women aged 30-34 by race who were separated from their spouse:

Surprisingly, about 20% of black women aged 30-34 were married and separated from their spouse in 1976. Again, this is not the percent of black women who are married. This is the percent of all black women. Thus, out of all black women aged 30-34 in 1976, 1 in 5 reported that they were married and living separate from their spouse. However, since 1976, the black-white disparity in separation has significantly decreased.

Finally, consider the percent of women aged 30-34 by race who were married with their spouse present.

When looking at rates of marriage with spouse present (rather than rates of marriage simpliciter), the black-white disparities actually seem to have been fairly stable since 1960. The gap has increased from about 21 percentage points in 1960 to about 33 percentage points in 2014.

Sources for charts in this section:

Birth and fertility rates


Now, I want to consider changes in birth and fertility rates by race since 1910. I’ll start by considering general birth rates by race. Next, I’ll consider birth rates by race and age.

Before providing the data, it is important to distinguish between two kinds of birth rates that are reported:

  • Birth rates: numbers of births per 1,000 population.
  • Fertility rates: numbers of births per 1,000 women aged 15-44.

Thus, fertility rates will always be greater than birth rates. In general, it will be more appropriate to use fertility rates rather than birth rates when comparing groups in order to control for potential age differences.

General rates

The following chart shows birth rates by race from 1910 to 2018.

Here is the same data for fertility rates.

As you can see, fertility rates for all racial groups have decreased significantly since their peak in the 1950s. In fact, fertility rates in the 1950s were about 2 to 3 times the rates today. Fertility rates seem to have stabilized since the mid 1990s, except for American Indians and Hispanics which have decreased significantly in recent years. Interestingly, there does not appear to be very large racial differences in fertility rates in recent years. Blacks, whites, and Asians all seem to currently have similar fertility rates, and Hispanics only have slightly higher fertility rates. American Indians, however, have significantly lower fertility rates than other racial groups, which is surprising because this pattern was reversed in the 1980s and 1990s.

Sources for charts in this section:

Birth rates by age

It will also be interesting to look at birth rates by race broken out by age. Here is the data for white women.

And the data for black women.

Birth rates for both whites and blacks have decreased from 1960 to 1980 for all age groups. Since 1980, the birth rates seem to have only decreased for the 18-19 and 20-24 age groups. In fact, the birth rates have actually increased since 1980 for some of the age groups.

Similar patterns seem to be present for Hispanic women.

And Asian women.

For Hispanics and whites, there also seems to be a decrease in birth rates for the younger age groups but stable birth rates for the older age groups. One thing worth noting is that the relative birth rates by race differ based on the age considering. For example, when considering women aged 20-24, blacks and Hispanics have the greater birth rates.

However, among women aged 30-34, whites and Asians have the greater birth rates, in recent years at least. In fact, among this age group, Asian women have had the highest birth rates for any racial group for every year that we have data available.

Sources for charts in this section:

Fertility rates by marital status


Finally, let us consider fertility rates by marital status. The following chart shows fertility rates for married and unmarried women from 1940 to 2010 (with some data in later years for unmarried women).

As you can see, there has been a gradual increase in fertility rates for unmarried women over this time period. However, the fertility rate for married women decreased sharply from 1960 to about 1975 and has remained stable since then. Thus, much of the change in percent of births to unmarried mothers over the 20th century seems to be attributable to the sharp drop in fertility rates for married women over this 15-year period.

The drop in fertility rates for married women has been particularly large for black women in particular. See the following chart which shows the fertility rates by marital status and by race. The solid lines indicate fertility rates for married women whereas the dashed line indicates rates for unmarried women.

First, consider the racial differences in fertility rates for married women. Perhaps surprisingly, the fertility rates for married black women have been lower than the rates for married white women since the mid 1970s, with the gap slightly growing in recent years. This pattern emerged in 1970, as married black women had greater fertility rates than married white women before then. Another interesting finding is that the fertility rate for married Asian women was greater than the fertility rate for any other race-marital group of women for some years.

A different story emerges when considering unmarried births. The fertility rates for unmarried black women have been consistently greater than the rate for unmarried white women since the data has been recorded. However, the gap has decreased since 1960, as the fertility rate for unmarried black women has slightly decreased whereas the fertility rate for unmarried white women has slightly increased.

Comparing married vs unmarried fertility within groups, there are also interesting racial differences. For black and Hispanic women, the married and unmarried fertility rates are similar (at least since 1970). In fact, for black women, the unmarried fertility rate has exceeded the married fertility rate for many years. By contrast, for white and Asian women, the married fertility rate has been consistently far greater than the unmarried fertility rate, with the married rate being about 2-3 times the unmarried rate in recent years.

Sources for charts in this section:

Concluding thoughts


This data has interesting implications on potential explanations of the rising out-of-wedlock birth rate throughout the 20th century. The percent of children born to unmarried women in a given year can be calculated based on the following values:

  1. The percent of women who are married
  2. The birth rates for married women
  3. The birth rates for unmarried women

Most people tend to focus on (3) when in fact (1) and (2) are also important factors as well, possibly equally or more important. In fact, for black women, (3) doesn’t seem to be relevant at all in explaining the rising percent of births to unmarried mothers. Since 1960, the unmarried birth rate for black women has not decreased at all. This suggests that the increase in births to unmarried mothers is entirely explained by (1) and (2): decreasing share of women who marry and decreasing birth rates among women who do marry. In other words, unmarried black women are no more likely to reproduce today than they were 60 years ago. Rather, black women are less likely to marry, and, when they do marry, they are much less likely to reproduce.

Part of the explanation for (1) and (2) may be decreasing shotgun marriage rates. If couples are less likely to marry following conception of a child, that will obviously lower (1) because there would be fewer marriages. But it would also lower (2) because it would decrease the share of marriages that are shotgun marriages, which must also decrease the share of marriages with children (since 100% of shotgun marriages would have a child present). In fact, Akerlof and Yellen (1996) [archived] argue that decreasing shotgun marriage rates explain most of the increase in out-of-wedlock births between 1965 and 1990:

Since 1969, however, shotgun marriage has gradually disappeared (see table 1). For whites, in particular, the shotgun marriage rate began its decline at almost the same time as the reproductive technology shock. And the disappearance of shotgun marriages has contributed heavily to the rise in the out-of-wedlock birth rate for both white and black women. In fact, about 75 percent of the increase in the white out-of-wedlock first-birth rate, and about 60 percent of the black increase, between 1965 and 1990 is directly attributable to the decline in shotgun marriages. If the shotgun marriage rate had remained steady from 1965 to 1990, white out-of-wedlock births would have risen only 25 percent as much as they have. Black out-of-wedlock births would have increased only 40 percent as much.

The data on marital separation also suggests another possible interesting finding. Much of the widening black-white disparity in marriage can be attributed to decreasing rates of black women who are married with an absent/separated spouse. When one focuses only on the percent of women who are married with their spouse present (as opposed to marital rates simpliciter), the marital gap does not widen nearly as much. This suggests that much of the widening black-white gap in marriage may be due to a decreased willingness among black individuals to remain in unsatisfying marriages (if we assume that marital separation is a proxy for unsatisfying marriages).

 

Leave a Reply

Your email address will not be published. Required fields are marked *