Especially if they’re Black: A shortage of men for poor women to marry

One thing a lot of liberals and conservatives can agree on: not talking about race.

[If you don’t have time for the text, just skip to the figure.]

Liberals are happy when conservatives talk about inequality, which they’re doing a lot more these days. And when they debate marriage as a way to “cure” poverty, neither talks about race. For example, Annie Lowrey writes in the the NYT Magazine:

With Democrats and Republicans pitted against one another in a vicious election-year battle over how to alleviate poverty, marriage is the policy solution du jour.

First, Lowrie makes the now universal mistake in interpreting the famous Chetty et al. result:

In a new study, the economist Raj Chetty and his co-authors found that, in terms of income mobility, nothing matters more for a low-income child than the family structures she sees in her community — not neighborhood segregation, school quality or a host of other factors.

Traditionally in America, when you say “a host of other factors,” that includes race. But the Chetty et al. paper is nearly unique in its avoidance of race, partly because race isn’t specified in tax records. So “nothing matters more” is at best untested, and at worst completely wrong, since race isn’t in the model. (My argument on this is here).

To those of us old enough to remember, or have read stuff from, the 1980s, not including race in this conversation is bizarre. Of course, it is not crazy to talk about poverty as an issue. In that article, Kristi Williams is right when she says:

It isn’t that having a lasting and successful marriage is a cure for living in poverty. Living in poverty is a barrier to having a lasting and successful marriage.

But the article doesn’t address the hard demographic reality that the things that make marriage less available or attractive to poor women — Lowrey lists “globalization, the decline of labor unions, technological change and other tidal economic forces” — have done it much more for Black women, even among the poor. In addition to even worse job prospects, for Black men you need to add incarceration, mortality, and intermarriage rates much higher for men than for women.

Here’s a simple way to see this. Adapting the old formula from William Julius Wilson, I counted up the number of employed, non-married men per non-married woman (employed or not) in the age range 25-34, separately for Blacks and Whites, and by education, for the 50 biggest metropolitan areas (one not shown because of data shortage, one outlier excluded). With intermarriage rates so low for Black women, and the tendency not to marry men without jobs, this is a reasonable approximation of the marriage market for Black women, though it understates the number of men available to White women.

This is the result:


Dots in the green areas show relative surpluses of men. Dots under the red line show better markets for White women than for Black women. It takes a minute to figure out. If your jaw dropped, you got it. With or without college degrees Black women face a shortage of “mariageable” men in every single market except five (Portland OR, Minneapolis, Denver, Salt Lake City, and Providence, which was the outlier not shown). For college graduates Black women are under 75 men per 100 women in all but two markets, non-graduates are under 75 in 40 out of 48.

White women’s market is better than Black women’s in all but six (those five plus Sacramento). In most cases White women graduates have a surplus of men from which to choose.

Poverty is one thing. Race is another. They overlap, but on some questions they can’t be combined. Marriage is one of those issues. So, when you talk about the shortage of men to marry, I recommend remembering race.

Note: After I made this graph, Joanna Pepin and I decided to write a paper together on this. That is still in the pipeline, and I was going to save this for when it’s ready. But there will be plenty more.

43 thoughts on “Especially if they’re Black: A shortage of men for poor women to marry

  1. Great point. I mentioned some of our findings on race differences to Annie Lowery and others, but they never made it into the final piece. For example, we find that to the extent that marriage is associated with any benefits for the health of single mothers, these only apply to white women who enter lasting marriages with the bio father, and not to black women at all. Your point about marriage markets is especially important, though.


      1. Agreed, it’s a great piece and my work on women’s health outcomes is not as directly relevant to the poverty issue. Honestly, although I’m not surprised, I didn’t realize that the race differences in marriage markets were so stark. It would be interesting to see these specifically in neighborhoods with high poverty rates. I’ll wait for the paper!

        Liked by 1 person

        1. Hi Professor Williams.

          I would like to ask a question if Dr. Cohen willl not mind. In Lowry’s article is a quote attributed to you
          “Living in poverty is a barrier to having a lasting and successful marriage.”.
          Do you have any evidence to this? Like divorce rates per marriage of people who had been in poverty for the last year/2 years. I did some analysis that shows no such evidence. In fact, the divorce rates of people who have incomes > 60,000 is nearly twice the rate of married people in poverty.

          Liked by 1 person

        2. Vijay- Here are a couple of references for divorce that I have immediately availablebut I was also referring to the ability to get married in the first place (to find a marriage that one thinks has a realistic chance of success). Bramlett, M. D., & Mosher, W. D. (2002). Cohabitation, marriage, divorce, and remarriage in the United States.
          Raley, R. Kelly. and Bumpass, L. (2003). The topography of the divorce plateau: Levels and trends in union stability in the United States after 1980. Demographic Research, 8, 245 – 259.


  2. If you can unpack this into three graphs, white, black and hispanic, it can explain additional details such as why unmarried childbirth is increasing among Hispanics and getting close to 1990s black rates. In addition, grouping Hispanics into whites understates something, which I am not sure what it is.


      1. OK, (Now we do the usual tango between an engineer who does statistics and a liberal sociologist)

        Isnt the employed men/100 women an intermediate variable, not a starting point of an analysis? Meaning that it tells me nothing because it is a variable at a middle level.

        In my mind, Cognitive ability distribution (two or more curves differentiated by race) -> High school/college graduation rates/probabilities (again, more distributions) -> Employment outcomes at 18 and above -> Distributions of marriageable men/women at 18 and above -> children born in two person married households/female led household -> children graduating high school college-> ad-infinitum.

        You are picking a graph at the middle of this non-linear problem with tremendous feedback loops to argue women are not marrying because of unavailable men. I give kudos to you for decomposing and showing differences by race, but the root variable is something else. Genetics and heritability is much more important in the whole dynamic loop.

        Liked by 1 person

  3. I am glad that this is not just picking on black women like the usual narratives does.You did miss the fact that white and other non-white women are marrying employed black men leaving even less for black women. The intermarriage rate for black men in measured by the 2008 census was 22% v 9% for black women. If poor white women have less options, I am sure they will turn to poor black men. I live in a Midwest college town surrounded by poor rural communities and the black male interracial cohabitation & marriage rate in this area is higher than the rate of black couples. The future looks much dimmer for black women in this area.


    1. Hey, I want to jump in this race! Intermarriage rate for Asians is 27% with men having 16% interracial marriage rates whereas for women, it is 36.6%. The future for Asian men is dimmer.

      Liked by 1 person

      1. I am part South Asian & I am going to call bullshite on your attempt at Oppression Olympics. The marriage rate for black women is much lower than Asian men. Asian American women may be outmarrying more but Asian American men do not have centuries of racist ideology that deems them unmarriageable because of they are hyper-sexual, immoral, ugly, fat, rude, unladylike, bad mothers and lazy. Asian American men have a much higher incomes than black women and can pursue white women, Asian immigrants, other women of color and even, gasp, black black women!

        Liked by 1 person

  4. I see that you are congratulating CEPR inclusionist for making fun of Lowrie on twitter Re: Latino poverty rates for families (21.5%):

    It would have taken a few minutes to decompose the poverty rates for married Latino couples with families :

    1. First-generation Hispanic immigrants and their families now comprise 9 percent of the U.S. population but 18 percent of all poor persons in the U.S.;
    2. Nearly 50% of the first generation Hispanic immigrants lack a HS diploma
    3. First generation Hispanic immigrants work in very large numbers in Construction and services industry
    4. 2007-present recession resulted in significant loss of jobs and increase in poverty.

    The 21% poverty number can be directly related to first generation immigrants in and out of jobs. It changes nothing for the existing legal residents and citizens. They have a sub-10% poverty rate.

    If anything that tweet argues AGAINST import of large number of ill qialified immigrants to US; instead of being in poverty in Central America, they will be in poverty in US.

    Liked by 1 person

    1. Neither, Cochran, nor Steve is the origin of the “bulge” of black women of marriageable age, which is presented here as original research, and is attributed to be a lead to the reason for lack of marriage.
      W.E.Dubois studied the effects of Low Sex Ratios on a Community In his (1899) “The Philadelphia Negro,” where he presents one of the first comprehensive studies of a minority group that explains its deviance from the majority. Du Bois first introduces the gender imbalance in the 7th Ward of Philadelphia in Chapter 5 of The Philadelphia Negro. According to the data available to Du Bois and his research assistants about the city, the African American population in 1890 was skewed in sexual distribution, with about 4 women for every 5 men between 20 and 55.

      One would think that a sex ratio like this would lead to all men being married – but the research shows otherwise. Du Bois found in his research in Philadelphia, sex ratios this low lead to a lower incidence of married males. he presence of this large number of males with tremendous sexual competitiveness and no pressure to tie their fates to individual women increases the proportion of out-of-wedlock births. Du Bois even noted the persistence of out-of-wedlock births-> to children who behaved even more as adults, and tremendous increase in teenage birth, unmarried births, and lack of male employment (Chapter 5 and 6).

      Today, the African-American community in the United States has an adult sex ratio that is comparable to that of Philadelphia in 1890. Although boys outnumber girls at birth 103 to 100, a normal ratio worldwide, high death, incarceration and substance abuse rates among African American males have worked to lower ratios post-adolescence(Mathews, 2005, Massey, 2007). Sex ratios for the community in 2006 vary from 0.85 to 0.82 between 20 and 55.

      If you do a literature search in google, you hit more than 3000 publications explaining this skewed sex ratio. Foremost causes are teenage-adult incarceration rates, drug use, and undercutting of employment opportunities due to lack of high school, blue collar jobs, and immigration.

      In summary, there is nothing new, close to 115 years of research, it has been shown to be an outcome, and not a cause. Using this to demonstrate that people cannot marry because there are no men, are devious, at best,.


      Du Bois, W.E.B.The Philadelphia Negro.
      Philadelphia, PA: Publications of the University of Pennsylvania, 1899.

      Massey, Douglas.Categorically Unequal: The American Stratification System.New York: Russell Sage Foundation, 2007.

      Mathews, TJ, and BE Hamilton.
      Trend Analysis of the Sex Ratio at Birth in the United States.
      National Vital Statistics Report, Washington, DC: National Vital Statistics, 2000.

      Med Anthropol Q. 2004 Dec;18(4):405-28.
      Marriage promotion and missing men: African American women in a demographic double bind.
      Lane SD, et al.


  5. Well, I never thought of it that way. I basically marry who I love. I did notice that when you are happy with a black male that is not wealthy and very educated it does not shelter you from the social ills that impact you before that black male that came into your life. Its post antebellum and hopefully blacks women and men will embrace one another and put family first, or fall for nothing.
    I do feel that in a way it is like slavery days and we as black women suffer more when we are not easily coerced outside of black social groups, where as the person looks like us and has same educational attainment etc.


Comments welcome (may be moderated)

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s