33

In conservative responses to racial disparities involving wealth and other things a lack of fatherhood figures within the black community is often cited as causative. For example, Ben Shapiro said about 4 years ago:

[If racial disparities are not largely a "black" cultural problem, then explain why] the single motherhood rate in the black community jumped from 20% to 70% in the same course of time that the civil rights movement has made such tremendous strides.

I was reading around on similar topics on Skeptics and found a quote from MLK that criticized the black community in 1957 for their "illegitimacy rate".

We have eight times more illegitimacy than white persons.

I readily see that these two claims are not exactly comparable nor mutually exclusive. Both could be true at the same time, but it made me immediately think that perhaps there's a significant twist on both of these quotes that is not captured in their exact words.

I'm looking for a brief historical analysis on the "illegitimacy rates" in the black community1. Shapiro is claiming an increase from a low rate (by today's standards) in the early sixties up to a high rate today, while MLK states pre-sixties that it was already high compared to whites. I'm supposing that perhaps Shapiro's claim misses a large bit of nuance in the relative differences, since his claim is absolute in context, or MLK's factoid is actually false (but was maybe a common misconception at the time).

  1. I understand "illegitimacy" to mean children born without married parents. It would be great if "committed relationship" were more quantifiable, in 1960 and now, because my impression of modern culture is that it is far more acceptable and common for committed couples to have children even if they are not legally married, but this seems unlikely to be differentiated in official statistics.
  • 25
    While the numbers can be checked, the author's conclusion falls afoul of [correlation vs causality](https://en.wikipedia.org/wiki/Correlation_does_not_imply_causation). – Schwern Jan 25 '21 at 21:31
  • 33
    @Schwern That's kind of his point, that the correlation doesn't necessitate "racism" as a causative factor, and that he thinks there's other more believable things as causes. But the bigger question is, does every question on uncomfortable stats really need this "correlation vs causation" disclaimer? –  Jan 25 '21 at 21:35
  • 39
    This is the Internet. Yes. – Schwern Jan 25 '21 at 21:59
  • 5
    Well played, Sir! –  Jan 25 '21 at 22:04
  • 8
    I think a good answer should also compare the amount of single mothers in other groups, too. At least for Brazil, being a single mother is a thing that happens quite often in all groups, specially among teenagers. – T. Sar Jan 25 '21 at 22:28
  • 4
    As a progressive, I view most conclusions like Shapiro's as self-fulfilling prophecies and the result of racist feedback effects. E.g. more single-parent households because systemic racism results in more black men being imprisoned for minor offenses, which causes more poverty among blacks, which causes more of them to turn to crime to survive. – Barmar Jan 26 '21 at 15:27
  • 3
    @Barmar That's not really this question though. You can make an answer if you like, but if I thought the causes of racial disparities could be answered tidily in an SE post I would have asked *that*. –  Jan 26 '21 at 16:12
  • 2
    @T. Sar: Should also consider whether they're counting single mothers at the child's birth, or single at some point between birth and adulthood. Certainly I know a number of the latter, including those that later remarried. – jamesqf Jan 26 '21 at 16:43
  • 2
    Single motherhood is also strongly correlated with living in a deeply conservative state, see [this map](https://www.motherjones.com/kevin-drum/2021/01/do-you-remember-what-conservatives-feared-about-hillary/) and [the rate of single motherhood by state](https://www.statista.com/statistics/242302/percentage-of-single-mother-households-in-the-us-by-state/). So if one can interpret race as a causative factor, one should also interpret "living in a place ruled by Republicans" as a causative factor. – Erwan Jan 27 '21 at 15:35
  • 1
    I (and [Werriam-Webster](https://www.merriam-webster.com/dictionary/illegitimate)) would define "illegitimate" in this sense as "born to parents who are not married *to each other*". Probably this is what you meant, too, but it is relevant to the question that "not married to each other" does not imply "single". – John Bollinger Jan 27 '21 at 15:35
  • 2
    @Erwan *Strongly*? You should punch out the data first and see what number you get because [the map you linked](https://www.motherjones.com/wp-content/uploads/2021/01/blog_single_mothers.gif) shows geographic concentration in the South, while many of the midwest, central, and western states are also "deeply conservative". –  Jan 27 '21 at 16:00
  • 3
    @fredsbend I was curious enough to collect the data and I stand corrected: the correlation is close to 0 indeed. I took the blog post I linked at face value since the author is usually honest and careful with statistics, but in this case he was wrong. – Erwan Jan 28 '21 at 01:48
  • 1
    @Erwan I didn't bother reading the article, but if this author is not careful enough to actually calculate a correlation while claiming one, then I would reconsider any reliability you believe he holds. But good on you for checking up on it and following up here. A truly skeptical and honest approach. :) –  Jan 28 '21 at 04:18
  • Comments are not for extended discussion; this conversation has been [moved to chat](https://chat.stackexchange.com/rooms/119031/discussion-on-question-by-fredsbend-did-the-single-motherhood-rate-among-america). –  Jan 28 '21 at 04:22
  • We should be careful of the false equivalency between "illegitimacy" and "single motherhood". Many couples raise their children together without being married, and many children are raised by two parents even if their biological father is not one of them. – Evargalo Jan 28 '21 at 11:44
  • 2
    This was big news in the 1960s. Look up the "Moynihan Report", 1965. This is where it first came to the forefront. – Mayo Jan 28 '21 at 18:17

4 Answers4

43

I'm just providing the numbers. No claim about the cause can be made.

Shapiro is claiming an increase from a low rate (by today's standards) in the early sixties up to a high rate today, while MLK states pre-sixties that it was already high compared to whites.

I'd say both are correct.

The trends for one vs two parent families has been basically the same for Americans since 1960, the earliest I could find complete data for all groups. It started low, climbed until about 2000, and is now slowly going down again.

The magnitude is quite different by race. In 1960 there was already a large disparity in magnitude and has continued to be.

  • 1960
    • Black - 22%
    • White - 8%
    • All - 10%
  • 2020
    • Black - 51%
    • White - 27%
    • All - 36%

enter image description here

References

Schwern
  • 17,034
  • 7
  • 63
  • 66
  • 14
    You say no claim about the cause can be made, but the WIlson Hypothesis (that the Black-White marriage gap is due to unemployment, incarceration and low earning capacity of Black men) has [some support](https://voxeu.org/article/incarceration-unemployment-and-black-white-marriage-gap-us) (admittedly [some more limited](https://www.jstor.org/stable/146195?seq=1)). – Oddthinking Jan 26 '21 at 00:21
  • 11
    @Oddthinking Not from this data alone. – Schwern Jan 26 '21 at 00:27
  • 23
    You say they are both right, but maybe you mean in their point only. MLK said "eight times that of whites", but 22% is not 8x. Shapiro says "20% to 70%". 20% is close enough, but 70 is pretty far from 51. –  Jan 26 '21 at 01:36
  • 2
    @Oddthinking That's very interesting information, but Shapiro's claim I think is in the other direction, that is to say that the high black illegitimacy rate is causative of high black poverty, etc. If I understood my brief read of the abstract, your source flips that on its head. I would take the disclaimer in this post to mean that these statistics do not support that black illegitimacy is causative of black poverty (Shapiro's claim). –  Jan 26 '21 at 01:54
  • 1
    Coincidentally, I think I heard recently from Shapiro again that of 3 things that make you very unlikely to be not in poverty in America one is marriage. In that claim he definitively claims causality, so that might make a great question in tandem with this one. The other two seem pretty straight forward: finish high school and get a job. –  Jan 26 '21 at 02:08
  • 24
    Shapiro (apparently) claims low marriage causes poverty. Wilson (apparently) claims poverty and imprisonment (especially due to drug crimes) cause low marriage. I don't want to fall into the same trap, and assume everything has one nice neat cause. They could reinforce each other. (Single-motherhood is different to being unmarried, but I imagine there is a strong bi-way relationship there too.) – Oddthinking Jan 26 '21 at 06:34
  • 20
    @fredsbend: It looks to me like two slightly different things are being measured. At least as I read things, Dr. King and Ben Shapiro are both talking about the percentage of *children* who have only one parent at home, but this seems to be measuring the number of *families* with only one parent. So, the 70% would indicate that the average family size is around 1.4-1.5 children. I can't vouch for that being accurate, but it strikes me as *quite* believable. – Jerry Coffin Jan 26 '21 at 09:56
  • 4
    These numbers seem to be about living arrangements. I suspect MLK's comment was for birth outside marriage, and those numbers are [now higher](https://www.childtrends.org/publications/dramatic-increase-in-percentage-of-births-outside-marriage-among-whites-hispanics-and-women-with-higher-education-levels) with a smaller ratio – Henry Jan 26 '21 at 10:21
  • 1
    One parent families are not the same as single mother families (but it's probably pretty close). – gerrit Jan 26 '21 at 11:25
  • 1
    @Oddthinking : if *"imprisonment due to drug crimes"* was a significant cause, then either the differences should be minimal, or it would imply that almost half the black population is in prison. – vsz Jan 26 '21 at 14:35
  • 2
    @fredsbend Take a look at the link in Henry's comment. Data in that link are claimed to be sourced from the CDC's National Vital Statistics System and show that the 70% number is indeed about right for _births to unmarried mothers_, as opposed to the "living in a single-parent home" numbers presented in this answer. The numbers in the answer would exclude cases where the mother is living with someone (the father or otherwise) or married someone (father or otherwise) after the child's birth, but "births to unmarried mothers" would include those cases. – reirab Jan 26 '21 at 16:24
  • @reirab Ah, I see. That's great that there seems to be some differentiation in the data. So perhaps Shapiro's "single motherhood rate" is confusing these two items. I suppose it wouldn't be too unusual for an orthodox Jew (whom Shapiro is) to confuse "unmarried parents" with "single parent". MLK being a reverend was probably highly concerned with marriage specifically as well, and perhaps too quickly assumed that meant single parenthood. –  Jan 26 '21 at 17:08
  • 3
    @vsz the lifetime incarceration rate of black men is something like 30%, and if you consider that likelihood of going to jail ending a relationship then it's not absurdly far off what you say is required (pdf) https://www.bjs.gov/content/pub/pdf/Llgsfp.pdf – llama Jan 26 '21 at 18:25
  • 2
    @fredsbend I'm not sure that it's so much a mistake as just ambiguous language. At least in my experience, "single" can mean either "not married" or "not in any sort of relationship." I would list my marital status as "single" regardless of whether I'm dating anyone at the time, for example. Looking further at the sources in this answer, it's actually a bit unclear to me how they're defining 'parent' for purposes of measuring 1 vs. 2-parent households. For example, is the rather common scenario of mother living with a boyfriend who is not the father of her child considered 1 or 2-parent here? – reirab Jan 26 '21 at 19:47
  • 1
    The plot title is confusing. It should just say "% of 1 parent families by race and year" not "1 vs. 2" – spacetyper Jan 27 '21 at 01:34
  • 1
    This statistic is very depressing, even for the white part. Do I understand this correctly, 1 in 3 families are single parent in the US in 2020? – kutschkem Jan 27 '21 at 07:28
  • I wonder how much of a factor the Vietnam War is in this - this just about coincides with the steepest part of the line at the beginning of the chart, and represents a period of history that resulted in a large number of specifically male deaths, and especially among lower-income people who could not afford to avoid the draft by various means. – Darrel Hoffman Jan 27 '21 at 14:35
  • 2
    @DarrelHoffman Vietnam War dead for the US were about 60,000 with about 500,000 soldiers serving in Vietnam at the peak against tens of millions of families. Deaths and deployments alone would not be enough to move the curve. – Schwern Jan 27 '21 at 20:44
  • @spacetyper The title is correct. This data specifically excludes children with no parents or who are raised by other relatives, which would need to be included with your suggested title. – Darrel Hoffman Jan 28 '21 at 14:36
  • Is "all" Black and White? Or actually all races? – Azor Ahai -him- Jan 28 '21 at 19:17
33

I'm looking for a brief historical analysis on the "illegitimacy rates" in the black community

  • Tldr; The rate for non-whites was ~21% in 1960 and 72% for blacks in 1998. The figures and references are below.

Did the single motherhood rate jump to 70% since the 1960s?

It has risen beyond 70% according to the CDC. This data was published by the CDC in 2000, in a report titled Nonmarital Childbearing in the United States, 1940–99. On page 7 is the Figure 10:

enter image description here

Was the single motherhood rate among blacks 20% in the 1960s?

This is tough to confirm, partly because at that time this data was grouped into "non-white" collectively with other minorities. That being said, the CDC listed the illegitimate rate of birth among this group at around 21% for 1960 (and less before that):

enter image description here

In a report titled The Negro Family: The Case For National Action by D.P. Moynihan (also called the Moynihan report) was written in 1965, and lists the single motherhood rate among blacks at 25%, and it appears to use the same data from the CDC as well as data from the most recent Censuses, and data from the BLS, which would include other non-white groups. That being said, after 1969 Blacks and Hispanics were recorded separately and the data is more clear.

The Moynihan report makes the same argument as Shapiro and other conservatives make today, so the position itself is nothing new.

BUT UNWED MOTHERS AREN'T NECESSARILY SINGLE!

Single motherhood is defined in this answer as a mother who birthed a child while not married. The author of the question specifically asked for "illegitimacy rates", which historically has meant illegitimate birth rates, meaning a child born to parents who were not legally married at the time. See the original question, and the first line of this answer.

That being said, this does not account for cohabiting parents, but it also does not account for homes where the couple is married and the father is away on active duty. Statistics is an imperfect science.

Also, many comments are concerned with cohabitation and claim that the true number of single mothers is (much?) lower than what the CDC is recording as such.

While that is true to some extent, it does need context:

Conclusion

  • Changing the definition of single mother from what the OP has stated, and the implications thereof, would be a new claim and should be asked in a separate question. There is more research done on family structure that would detract from the core claim being examined in this question, which is whether or not what Shapiro and King said was related, accurate, and/or taken out of context.

  • This answer asserts that if they are both referring to the same statistic - births to unwed mothers - then their claims are related, accurate, and not taken out of context, even if the specific language used was different.

  • Both King and Shapiro's claims formed the premise/foundation for a broader argument. The validity of that argument is itself a separate claim and should be posted as another question.

Disclaimer: I am not interested in having the debate here (I see a lot of causation vs. correlation comments), only providing the data for others to study. But, I will debate in the chat if necessary.

MDMoore313
  • 426
  • 3
  • 7
  • 3
    What is startling about that table is how large a proportion of mothers aged 10-14 seem to have been married (over half of white mothers and a quarter to a fifth of nonwhite mothers) – Henry Jan 26 '21 at 15:31
  • Your 'TLDR' conclusion is misleading. "Nonwhite" is not black. – Dom Jan 26 '21 at 16:23
  • 3
    @Henry likely due to laws (IRC, still not repealed in a few states) that sort of forced child into marriage with their abuser. – Dan M. Jan 26 '21 at 16:38
  • 2
    Excellent work and well organized answer. I usually wait several weeks before making a selection, but to let other potential answers know, you'll have to beat this one. –  Jan 26 '21 at 17:02
  • 3
    As soon as I read the title, the Moynahan Report jumped straight to mind. And he was far from a Republican!!! – RonJohn Jan 26 '21 at 17:47
  • Your first example is a list of NON MARRIED rates, not single mother rates. There are likely plenty of couples that live together and thus would not count as a child being raised by a single mother but where the couple is not married. Thus the non-married rate of blacks is likely higher then the single-mother rate in blacks and therefor I don't think this proves the original claim is true. – dsollen Jan 26 '21 at 19:22
  • 4
    @dsollen there is no "original claim" by the OP. I think we have to agree on language used by Shapiro and others, that single motherhood is taken to mean non-married women. There is an historical precedent there, and also a current precedent with the IRS when filing taxes - either you're married or single: There's no middleground there. – MDMoore313 Jan 26 '21 at 19:39
  • @MDMoore313 dsollen is right, unmarried does not mean single, so again it's misleading. I didn't notice that before. The best you can say here is it has very likely risen, how much is impossible to conclude from the data. – Dom Jan 27 '21 at 09:50
  • @RonJohn Just bear in mind that this was before the Southern Strategy. Back in those days it was the Democrats who were the party for racists in the South. https://en.wikipedia.org/wiki/Southern_strategy – Paul Johnson Jan 28 '21 at 11:20
  • Caveat for the first source : "births to unmarried women" is not the same as "single motherhood". – Evargalo Jan 28 '21 at 11:47
  • 2
    @Evargalo it sure was when the Moynahan Report was written. "Shacking up" was virtually unheard of. – RonJohn Jan 28 '21 at 13:20
  • @ronjohn But there were already divorce and widows (producing single motherhood out of "legitimate births"), or adoptions (producing 2-parents families out of "illegitimate births")... And even if the USA are much more conservative than Europe, I would be surprised if no unmarried parents were raising their children in the big cities in 1998... – Evargalo Jan 28 '21 at 13:24
  • 2
    @Evargalo divorce and widowhood does *not* produce single-motherhood, according to US usage. (Technically it does, because there's a *single* parent, but like I said, that term is not used in the US. "Single motherhood" is never-married.) – RonJohn Jan 28 '21 at 13:34
  • @ronjohn oh, if that's so I wasn't aware of the US-meaning of "single motherhood" and misunderstood most the question. (The concept of (US) "single motherhood" seems much less relevant to me than when I thought it was about the family environment a child is raised in, but so be it) – Evargalo Jan 28 '21 at 14:21
  • @Evergalo divorced and widowed parents are routinely called single parents in the US. It's just that they're not included in the statistics given here. – phoog Feb 05 '21 at 07:20
12

Just to complement @Schwern's excellent answer, and similarly not attempting make any claims about cause, I note that Shapiro is talking about single parents (which @Schwern's answer addresses), whereas King is talking about "illegitimacy", that is, births to parents who are unmarried (although may still live in a 2 parent household).

The following graph, produced by The Social Capital Project (an ongoing research effort by The Congress Joint Economic Committee), shows the percentage of births that were to unmarried people, broken down by race, from 1952 to 2007, for women between 15-44, in the US.

The report details its sources in the Source Notes, with the main source being this CDC report.

Breakdown of percentage of births to unmarried people by race over time in the US

James_pic
  • 221
  • 1
  • 4
  • 1
    The sources for this data are [cited here](https://web.archive.org/web/20181222074543/https://www.lee.senate.gov/public/index.cfm/socialcapitalproject?ID=24993658-359C-4039-B672-19AF5E444B0F). Look at the section for "Figure 7." – reirab Jan 26 '21 at 16:32
  • 1
    Also probably worth noting that the "[Social Capital Project](https://www.jec.senate.gov/public/index.cfm/republicans/socialcapitalproject)" is an ongoing research effort from the [U.S. Congress' Joint Economic Committee](https://www.jec.senate.gov/public/index.cfm/home/). – reirab Jan 26 '21 at 16:41
  • @reirab Thanks. I'll update the answer. – James_pic Jan 26 '21 at 17:03
  • Shapiro says "single motherhood rate" which is an odd vernacular way to say whatever it is he's exactly referring to. I don't think it's too unreasonable to assume that MLK and Shapiro were referring to roughly the same thing, that is a lack of fatherhood figures. –  Jan 26 '21 at 17:04
  • 5
    @fredsbend I can't speak to MLK or Shapiro's exact intent. It's certainly conceivable that they may have been alluding to the same concept but had access to different data points, so spoke about the data they had available. It's also conceivable that they were referring to genuinely different concepts, since childbirth out of wedlock is an issue that has religious implications, independent of the role fathers play in their children's lives. In any case the data on both statistics are largely similar, so it's likely that the same conclusions would apply whichever interpretation you chose. – James_pic Jan 26 '21 at 17:18
  • 7
    When interpreting MLK's 1957 words, you need to understand the 1957 moral mindset. The birth control pill (1960) did not yet exist. It was illegal to distribute contraceptive **information** to both married couples (1965) and single people (1971). Out-of-wedlock births were so shameful that pregnant girls were sent away to "visit an aunt in the country" i.e. to give birth in a "home for unwed mothers" somewhere far enough away that they wouldn't bring shame on their families, and the babies were frequently given up in closed adoptions. The alternative was for the pregnant girls to get married. – shoover Jan 26 '21 at 19:56
4

Something notably absent from other answers (so far) has been data on the ratio of cohabiting parents. This Pew Research article using 2017 data (not long after Shapiro's comments) shows a total single-parent rate for Black families of 51%, with 43% having two parents (36% married and 7% cohabiting).

The highest accurate representation of children living with single mothers (including cohabitation) would be 54%. Even if you were to sum all children without married parents (including children living with no parents at all), the resulting 64% would still fall a bit short of his claim.

At best, Shapiro was exaggerating to make a point. The previous statement cannot be asserted, since we don't know what data Shapiro was referencing; differences in datasets could have resulted in something closer to his claim. When considering the actual data for all children under 18 around the time of his claim, it was not true.

enter image description here

An interesting note is that the solo-father and cohabitation rates are much closer across racial boundaries than the married and solo-mother rates.

AjimOthy
  • 149
  • 2
  • 4
    "_At best, Shapiro was exaggerating to make a point_" - Or he was referring to births to unwed mothers. How could he be exaggerating on data that wasn't available until _after_ the claim was made? Don't take my comment to mean this isn't good research by Pew, I think it is, and a good find on your part. – MDMoore313 Jan 28 '21 at 12:04
  • Fair point - it's not clear what data he was discussing, so there could be differences between this (covering all children under 18) and something like births, or between the 2017 data and whatever year(s) he was citing. – AjimOthy Jan 28 '21 at 14:36
  • Would be interesting to see how many of these "single parents" are due to the death of the other parent or divorce – warren Jan 29 '21 at 18:27
  • Does the data indicate if there is a relative difference in the *total number* of children in a given family structure (eg, to unwed moms, it's an average of 4.1 children, but to married moms it's an average of 2.8 children)? Thinking about the social problem of sometimes-called "baby factories" (women who have another kid every X often to benefit from social services programs) - and how that may (or may not) affect the totals. – warren Jan 29 '21 at 18:29