August 24th, 2015

The Rise of Nonmarital Births


An increasing proportion of American women and men have their first child, and sometimes subsequent children, outside of marriage. The upward trend began more than a half century ago, and is more pronounced among those with who are less privileged. For example, there is almost no such trend among white college graduates. Nonmarital births are typically to couples in relationships or cohabiting unions that break up within a few years, leading to substantial complexity and instability in family patterns. Panelists will consider the trends, social forces encouraging nonmarital births, and their consequences for adults and children.


PAULA ENGLAND: Good afternoon. Thank you for coming to today's plenary, one of two plenaries today, on the rise of nonmarital births. I'm Paula England, your president for about one more day, and the
organizer of this session. We have three really distinguished scholars who have done work on nonmarital births that will talk to you today. The first presentation will be by Andy Cherlin, who is at
Johns Hopkins. The second one is Larry Wu, who is at NYU and the third by Marcia Carlson, who is at the University of Wisconsin. I won't stop and introduce them again in between speakers, so please
join me in welcoming Andy Cherlin.
ANDY CHERLIN: Thank you, Paula. I was actually hoping I would get some of that great walkup music we had yesterday for the presidential address session, but I will go on without it. I'm pleased to be
here at this session. Thanks to all for inviting me. I'm sure you found Paula's presidential address yesterday as thought-provoking as I did, especially her ideas related to our presentation today
about efficacy and its role in nonmarital childbearing. I'm going to ask a slightly different question and focus on something else for this talk which is, how can we explain the change that's
occurred in nonmarital childbearing, especially, how can we explain the extent to which nonmarital childbearing has really been transformed, as I'll show you, over the last few decades and what might
account for this? Now, I want to start by showing a chart that Larry Wu, our next speaker, put together, which shows you the history of births to unmarried women over the last half century. Starts in
1950. And you can see in 1950 that for all women, and for white women separately, only a few percent of births were outside of marriage. For black women, Larry's estimates suggests a bit over 20
percent, which shows that there's a longer history in the tale we don't see of unmarried childbearing for African Americans.
But let me note that in 1950 80% of all African American were born to married parents. So there has certainly been a transformation in the African American population, as well as in ours. Now let me
show you what happens if we look at the period from 1980 to today, which is what I want to focus on. This is a chart that will show you the percentage of births to single and cohabiting mothers under
age 40. I'm showing you single and cohabiting mothers separately, for this reason. 6Most of us, when we think of nonmarital childbearing, think about a single, that is unpartnered, lone mother. Often
in our minds we have a teenager living with grandma, who has a kid outside of marriage. The government statistics that you see also include a second group of unmarried women, those who are
cohabitated. And as you'll see, that share has gone up tremendously, and most people don't realize the shift that we've seen in exactly what the family arrangements are for people who are having
children outside of marriage.
So for example, in 1980 to '84, according to these estimates from the National Center for Family and Marriage Research, 15 percent of all births were to single mothers; that is, those without
partners. And additional six percent were to cohabiting mothers. So a total of 21 percent of all births were outside of marriage, most of them to single mothers, just the way we used to think about
this topic. By 1990, 10 years afterwards, there's not much increase in the proportions of single mothers, as you can see. But the proportion of births to cohabiting mothers has almost doubled. So
it's accounting for most of the rise that we see in that decade.
The next estimates are for the late 1990s. Again, not much of an increase for single mothers. But again, a substantial increase for cohabiting mothers, and for the first time a bare majority of
unmarried births are to cohabiting rather than to single mothers, and that trend has continued. Here's what we see in the early 2000s. We see no increase in births to single mothers, but a
substantial increase in births to cohabiting mothers. So that by the early 2000s, a clear majority of all nonmarital childbearing is occurring to mothers who are cohabiting, and that's continued.
Here is what we see for today in the most recent estimates, and it's very clear that a lot - currently about 58 percent - of all nonmarital childbearing is to cohabiting mothers. And these cohabiting
mothers are not teenagers, by and large. They are people in their 20s and early 30s. That's what has been driving the increase in nonmarital childbearing since about 1980, in a new development not
everybody is aware of.
Now, let me drill down a little bit deeper on these cohabiting mothers. Who is doing these cohabiting births? What kinds of couples are we talking about? What are their characteristics? Well, it turns
out if you look at mothers without a high school degree, for a long time, cohabiting - for a long time, it's been common to see women without a high school degree cohabiting at the time of having a
first birth. There's been an increase for sure, but it was common a few decades ago.
Let me show you a chart apropos of the question of which mothers have the increase in cohabiting births been greatest among. Of the percentage of cohabiting births for high school graduates, this is
one of the places where we've seen the most change. In the early 1980s, only about five percent of births to high school graduate women were to women who were cohabiting, okay, only one out of 20
high school graduate women, without any further education, giving birth for the first time were cohabiting. But watch what happens as we get closer in time and how much this increases. So the most
recent estimate, about 30 percent of all of the births to high school graduates, are to cohabiting women. So a behavior that really wasn't very common at all 35 years ago, is pretty common now.
The same thing happens for women with some college, but not a bachelor's degree. These are people who graduated from high school; they might have gone to a junior or community college, might have an
AA degree, certainly has taken some courses, but did not get the bachelor's degree. In the early 1980s, very few of these mothers were cohabiting. It was really uncommon, less than five percent. But
there's been a very big increase over time in the proportion who are cohabiting such that by late 2009 and at the interval from 2009 to 2013, about a quarter of them are. So a behavior that was
really not seen much at all is now common. It's in that middle that we've seen the biggest percentage changes in births to cohabiting women and their partners, okay, it's in that middle that we've
seen a rise of behavior that's now common, but wasn't common a few decades ago.
And this is a very large [INAUDIBLE 00:08:00]. If you take young adults who have a high school degree but not a bachelor's degree, that's more than half the population of young adults. This is a lot
of people driving the changes that we have seen. So let me, in contrast, show you the percentage of cohabiting birth mothers having births for mothers who have a bachelor's degree or higher. So this
is the same chart, but for people with bachelor's degrees. The lines are so low in this - the bars are so low in this chart, it's almost as if it has wandered in by mistake from another session,
okay? In the early 1980s, almost none of these women had a cohabiting birth, and more recently, not even 10 percent do. In fact, the vast majority of births to women with bachelor's degrees are to
people who are married, and with a very modest share to people who are cohabiting or who are single, showing once again that if there is a class difference with respect to family life in the U.S.,
the line, the bar, is at people with a bachelor's degree or more versus everybody else; it's almost as if we have separate demographic regimes, as demographers would say, among that group.
So take-aways from what I've said so far - first, most of the increase in nonmarital births since 1980 has been in births in cohabiting couples. And secondly, the greatest percentage of increase has
occurred among mothers with a moderate level of education. Both of those conclusions help me think about what might be causing all of this.
But let me pause to ask a question. Why do we care whether couples are cohabiting versus married? What difference does it make? Well, my answer is we care because we're not yet European, okay, or if
you're here from Canada, we're not yet Quebec, okay? And there's certainly great diversity in cohabitation among European countries, but in the northern European countries such as France and
Scandinavia, and to some extent the U.K., there are long-duration cohabiting unions that last decades, with children within them, providing children with stable care arrangements that differ little
from marriage. We don't tend to do that. Not yet, at least. In the U.S., our cohabiting relationships are the shortest duration of most any wealthy country. They typically only last a few years and
then people either get married or they break up, and an increasing share has been breaking up. What this means is, when couples now in cohabiting unions go ahead and have kids, there's a very high
likelihood that that relationship will break up and that child will see another relationship start, and perhaps see that one break up, causing a great amount of instability and complexity in
children's lives, which is a concern to all of us who care about children's wellbeing. And Marcie Carlson will be telling you much more about this in her presentation later on today.
So that's why we care. Now, how can we explain this pattern of change in nonmarital childbearing? A question occurs to me when I think about that. Starting around 1980, what new social forces might
have affected moderately-educated young adults more than the bottom or the top? What might have done that? Well, what might have done that is possibly the transformation of the American economy. That
started in the late 1970s when the post-war economic boom ended. And that has affected the middle of the labor market more than the bottom or the top, as we move to an hourglass economy with the
middle shrunk by industrial jobs of the kind that high school educated people used to be able to get, either moving overseas or being automated. And so the transformation of the American economy,
also known as the polarization of the labor market, also known as rising economic inequality, is one suspect for explaining at least part of this change.
Now, I have to say inequality isn't blamed for a lot of things lately, okay? There is a well-received book by Wilkinson and Pickett called The Spirit Level which gives a laundry list, a laundry list
of outcomes that are not good that are caused by inequality. And here is the laundry list, okay? Income inequality produces all this stuff, they say. But if you look at their book, every single piece
of evidence is on a macro level. It's comparing state A to state B or country B to country A, as they move over time. It's all in macro level analyses. And many of you, if you are like me, were
warned in graduate school to be cautious about making inferences on an individual level from macro correlations.
So as I thought about this, I began to look for micro individual-level studies that show how income inequality affects various life chances, including family structure, and I couldn't really find any.
Actually there's very little micro-level evidence out there showing that inequality does very much to influence life outcomes. So I've tried to come up with some, and I want to talk with you about it
a bit. And in doing so, however, I want to say I feel an obligation to talk about what inequality means. Those of us who use this term, which is almost everybody these days, have an obligation to say
what it means because if inequality is just a repackaging for the word "poverty," if it just means what we used to mean by "disadvantage," we've gained nothing at all by using this new term. So can
we find some micro-level evidence, and what it might mean, is what I'm talking about.
This is a research project I did with David Ribar at the University of Melbourne, and Suzumi Yasutake at Johns Hopkins, in an article we have under journal review. We used the longitudinal survey of
youth, 1997 cohort, familiar to most people, I think. This is so-called the millennial generation; people born between 1981 and 1985, came of age just after the turn of the century. They were
followed from 1997 when they were 12 to 16 through 2011 in our data, it continues when they were 25 to 30. And by that time they had had most of their nonmarital first childbearing, because as Paula
told us yesterday, people who have nonmarital births, especially first births, do it young. A lot of them by age 25, nearly all to age 30. So it's a good data set for setting this.
Let me tell you a bit of the research design. In a broad session like this with lots of people with different backgrounds, I'm hesitant to be too technical but I feel some responsibility to tell the
quantitative types in the audience, or who are watching this, what we did. So if this is not your thing, I suggest you just put your hands over your ears for a minute, okay, and I promise this will
be over very, very soon. We did a competing risk discrete-time hazard model, where we took everybody in the panel who was unmarried and childless at age 16, which is most people, and we observed them
annually until one of three events happened. A first birth as a single parent, a first birth as a cohabiting parent or marriage prior to a first birth. When that happened, we removed them from the
so-called risk set; we kicked them out of our data. And we did this every year and built up a set of person-year observations, which we estimated with a multinomial logistic model. This is pretty
standard now, for ways to analyze this type of data. We had two key time-varying variables; one of them is household inequality in the local area. Okay, we had a measure of how dispersed or together
household incomes are in an area. A measured inequality in your local area. And a measure of what we call of middle-skilled jobs, that is, what share of jobs are the kind to tend to hire high school
grads or people with just a year of college, and which pay above the poverty line? We had that measure too. And so in other words, we're saying in this article, is income inequality in the local area
associated with the circumstances of first birth, and does the availability of decent jobs in the middle of the labor market partly explain disassociation?
I want to tell you the results. I'm not going to put up the numbers. As I always tell my graduate students, some people have said, I've heard them in this meeting, don't put up a forest of regression
coefficients and then say to the audience, "I know you can't see this but" - okay? If you know we can't see it, why are you putting it up, okay? Just a little editorial. I'll tell you verbally, then,
what our strongest results were. Across a whole range of models, trying to deal with issues such as statistical identification, we found that men in areas with greater inequality were less likely to
marry prior to a first birth, controlling for a whole host of individual-level variables. So we do find some micro-level outcomes. And women in areas with greater inequality were less likely to marry
prior to a first birth, and also less likely to have a first birth by cohabiting. When we put jobs into this picture, we found that men in areas with more middle-skilled jobs were more likely to
marry prior to a first birth.
So if a man lives in an area that's got lots of inequality, he's less likely to marry. If he lives in an area with more middle-skill jobs, he's more likely to marry. Actually, those go together,
because what we found is that for men, areas with high inequality have a lower level of middle-skilled jobs that people can take. We think that's what inequality means, in large part, for these young
adults, especially for men; meaning that what inequality is, is really a marker; a marker for labor market difficulties in the middle of a labor market where those moderately-educated young adults,
especially young men, used to have jobs. In addition, women in areas with more skilled jobs were more likely to marry prior to a first birth and more likely to have a first birth while cohabiting.
That is to say they, they showed two different kinds of effects; one which is kind of called an income in effect in literature. Those with greater earning power seemed to be able to, some of them, to
translate that into marrying more. But others used it to have a child in a more independent setting of living together with somebody, where the ease of leaving is much greater. So they're kind of
income independence effects for women.
The take-aways from our study, then, income inequality is associated with less marriage prior to a first birth. We find micro evidence for it. For men, income inequality may be a marker for access to
less frequent middle-skilled jobs. For women, middle-skilled jobs matter too, and may encourage not just marriage, but also cohabiting births.
So I perhaps, in your mind, built a plausible case that labor market trends have contributed to the recent post-1980 increases in nonmarital births. We have looked at some graphs, I've given you some
micro evidence, but I have not nailed it. This is just one cohort, our micro study, our individual-level study. It doesn't go back to 1980. It doesn't show the whole sweep. And looking at graphs is
very nice and comparing them is something everybody likes to do, especially in a session like this, doesn't prove it either. So we have more work to do to nail down how inequality makes a difference.
But it does seem to me we have a plausible case now for saying that, in part, it's due to labor market issues.
Let me also say quickly, however, that this isn't the whole story. Certainly cultural change has played a role, too. For the last few decades in the family policy world, there's been a war between
conservatives who said, family change is due to cultural issues, and liberals have said it's due to economic issues. It's been an unproductive war which has stymied family policy. I think there is,
in the last just year or two, coming to be more of a realization on both sides that both culture and economics are involved. Let me tell you how.
During the great depression, the economy was terrible. But almost nobody had births outside of marriage. Why? Because it was unacceptable. You couldn't do that. It was something that was shameful. And
therefore, even people suffering a great deal of economic hardships were very unlikely to do that. What has changed culturally? The great change that I think has made the difference culturally is the
greater acceptance now of family formation outside of marriage. It is much more acceptable to be a single parent. It is much more acceptable to be a cohabiting parent, which gives people options when
they have economic difficulties. And so. I think what we see today is a conjuncture, a conjuncture of cultural change, which has given more options and made marriage less central to family formation
and economic change, which has really hurt people in the middle of the labor market, the same group with moderate educations, who seem to be the ones who've made the trend over the last few decades
in nonmarital births see what it is today. That's my message. It's economics, it's culture. Together they have created a unique moment for American family structure, a moment we've not seen before,
where we've seen an enormous rise in nonmarital births, the majority of it now not to that single mom, but to cohabiting couples with a high school degree in their 20s or early 30s. Thanks very much.
LARRY WU: Okay. So Andy's presentation dovetails nicely with what I will be emphasizing today. And so I'm going to take the slightly orthogonal take on nonmarital births, and actually ask some - give
you a bunch of fairly speculative remarks. So let me start. By way of introduction, I'm a guy who studies women having births outside of formal marriage and in full disclosure, I'm a gay man who does
these things, so go figure. This is not participant observation, I can assure you. And my standing joke is, I've had the honor and privilege of working a bit with Paula on some papers, and so my
standing joke is, Paula, the straight woman and Larry, the gay man, all we do is talk about sex. Not about sex with each other, and because I'm a demographer, I manage to make my side of the
conversation about sex as boring as possible. So I'll try and live up to that moniker.
Okay, so here's the plan. Here's the plan. Yes. The plan is, a bunch of quite deliberately speculative remarks about how might we level the playing field. Therefore, almost no data, almost no
empirical results, a stark departure from my usual self. What I will do, following closely on what Andy was emphasizing, is to put a lot of focus on what I'm going to call processes underlying
nonmarital fertility. And the point here is that nonmarital fertility is trickier in that it involves at least two things that we need to think about. Whoops, back. And what I will be arguing - and
this will be the end punch line, highly speculative - is that long-acting reversible contraceptives, so-called LARCs, the argument is going to be, they will have the potential to at least level the
playing field.
Okay. Processes. So I'm going to simplify matters, the discussion by just thinking about first births because it turns out first births are consequential. Why we care, following on Andy's point about
why we care, this is also population that is at high risk of poverty, these mothers and especially the children. So we would care. So the idea is really simple. Women begin life, go figure, in a
never-married and childless state. And so the - and so we then need to worry about at least the following sorts of things. Processes. One is the marriage process, and this follows exactly on the
issues that Andy was emphasizing. So at any given age of following birth, we're going to see some women start to enter a first marriage, and at those ages, other women will remain childless and never
marry. Likewise, there will be a fertility process. At some ages, we're going to see some women proceeding to a first birth, while some other women will remain childless and never married. Those are
the two processes, at least the two processes that we need to think about. Not surprisingly, this has generated two large themes in literature. One is barriers to marriage. So there is William Julius
Wilson, dearth of marriageable black males due to structural changes in the economy. Edin and Kefalas - the idea here is that the bar for marriage - others have argued this as well, Stephanie Coontz
- bar for marriage has increased. It's increased not only for the advantaged, but also for the disadvantaged, and this leads to growing disparities in marriage. Andy's most recent book on second
gilded age, presaging his remarks today.
But then fertility, and what I'm going to call its proximate determinants. And so if you think about it, sex while people are never married can sometimes result in a pregnancy. If a pregnancy occurs,
then individuals have to think about what to do about that pregnancy. Some of those pregnancies result in these premarital first births. As Paula was emphasizing yesterday, contraceptive behavior,
therefore, is an important part of that process. And this follows on an older demographic literature. And indeed many of the remarks I'm making are not novel. Isabel Sawhill has a recent book in
which she talks about - in fact, emphasizes LARCs and has a nice phrase about individuals drifting into sex and parenthood without marriage.
Some stylized facts. Again, following on what Andy has emphasized, I'm going to be emphasizing the sex part, as you might have guessed. So premarital sex. In my own work, among the estimates are more
than nine out of 10 among the not - never married are sexually active among young adults in the United States, and that's been true for several decades now. There are actually fairly small
differences in when people become - when women become sexually active for the first time after background controls. And so small differences between advantaged and disadvantaged groups. I'm not good
at this - relationships. So as Andy emphasized, one of the things that we've certainly realized about nonmarital fertility is the very striking rise in births within cohabiting unions. I'm going to
even back up and think about individuals who may be in cohabiting unions, but who may just simply be involved. And so I'm going to refer these as relationships, just as a theoretical idea.
So what do we know about relationships? Actually less, but I think it's plausible that many of you in the classroom with college students are probably aware that they're not avoiding sex and they're
not avoiding relationships, nor are those who are not going on to college. So relationships are, at least theoretically, one thing we might want to think about. And contraception - I'm bad at this
thing - as Andy emphasized, first births, these first births are now occurring, some of them in the late teens and many in the early 20s. And by about 25 the majority of these first births have taken
place. The vast majority of these first births are reported to us as unplanned, and that's an important clue, at least in my version of the story. So from these, I'm going to think about some working
hypotheses. As Paula emphasized yesterday, it's a good guess that advantaged groups put a lot of effort into contracepting, until they decide to try - to start to have a child. It's at least a
working hypothesis that less advantaged groups contracept less effectively, and this may be pointing to Marcia's comments and talk later - this may be one of the ways the U.S. differs from Europe.
And I think we have some reasonable evidence, both qualitative and quantitative, that's at least consistent with these hypotheses. So then, I'm going to list three questions and think about two. As
Paula emphasized yesterday, one question is, why is it that we see some groups contracepting effectively and others less effectively? I am not going to engage that question in my 20 minutes. I'm
going to ask, can we level the playing field, and if so, what might be some possible consequences?
So let's think about contraception and contraceptive failure rates by contraceptive method. And so here are some numbers from CDC. These are unintended pregnancies per 1000 women. And so these
long-acting reversible methods, implant and IUD, which I've highlighted in red, has very low contraceptive failure rates. It turns out a lot of young adults are using the pill, the condom, even
things like pullout and, not surprisingly, these are - what's the word - not good methods of contraception. And the magnitudes are striking. Orders of magnitude difference. I'm good.
How about trends in LARCs? The good news is that usage of LARCs has been going up. The bad news is it's still low. So, take-aways. LARCs are highly effective. By my likes, not enough people use them,
especially in the most relevant age groups. Okay.
So here's a thought experiment, and I'm going to call this a thought experiment. What if we were to randomly assign women, especially women from disadvantaged populations, to a treatment group, let's
say give them information about LARCs, and flip a coin, put people into that group or into a controlled group where they don't get this information about LARCs? And in the treatment group, we're
going to not only give them information, but if they want a LARC, we'll give it to them. If they want the LARC out, we'll take it out, all of this at no cost to the individual. So the point here is
to minimize contraceptive efforts, the hope being to convert some unplanned births into planned births and in my nether, nether world of Larry's dreams, I'll follow these people forever, which if I
got funding for that, I would be a very happy guy. But that's a different matter.
So I'm going to claim this has the potential to level the playing field. So there's actually some nonexperimental evidence from St. Louis and Colorado relevant to this. So there was temporary funding
for initiatives, indeed providing information and free access to LARCs. Neither had random assignment, therefore, I'm labelling them as non-experimental. Trends in Colorado are confound in trying to
think about what the results mean. Nevertheless, there is some suggestive evidence of fairly market declines in the things we might care about. And also, in St. Louis, and I think these are
reasonable and real results, we see considerable satisfaction and continuation of IUD's, and especially IUD's among those who took up the treatment.
I'm going to do a slight curve here for reasons that will hopefully become apparent shortly, and talk about something seemingly totally unrelated, which is what I'm going to term as, "churning in the
schooling to work transition." So one of the things that my labor economist friends tell us is that there's considerable churning when people leave school and enter the labor market, and here is the
story that they spin out. So the idea in their heads is that young adults enter the labor market with, let's say, a vector of attributes - skills, schooling, do you work hard, do you steal money from
the till, do you get along with people, do you yell at your coworkers? Employers are seeking a vector of attributes, and it's hard sometimes to know all the things about the person you're considering
to hire. There are mismatches that then occur. These mismatches lead to labor market churning. So early in this process, people change jobs. As workers, young adults learn more about the job market
and their attributes and what is valuable in the job market, churning declines over time. And so this evens out. Okay, so that's churning in the labor market.
So I'm going to ask a "what if" question. What if relationships were at least this complicated as finding a job? Finding a person, finding your honey? So the idea is straightforward. So remember, many
of these premarital first births are occurring at relatively early ages, late teens, early 20s. Majority of these births are unplanned. Many of these nonmarital births are occurring within cohabiting
unions or relationships, and we have good evidence that these relationships and cohabiting unions, many of them are fragile, do not last. If we thought there was a higher bar for marriage, it might
also imply a higher bar for relationships. It's not just enough that he earns a good living. We might need to have similar tastes in music. It may be important that he or she is not going to vote for
Donald Trump, or whatever. Or it may be important that he or she does, you know, depending. McLanahan and Beck, in a recent paper, even have some evidence suggesting that over time - what we see
about the guy that she's with improves over time, as relationships churn.
So what if LARC usage was to increase among these disadvantaged populations? So the point here is there is the potential for young women to be able to avoid a birth in an early relationship, and
therefore churn to a later relationship, therefore to a potentially better relationship as in much the same way as better job matches occur over time. LARC usage should reduce unplanned pregnancies,
abortions, fertility to several different men, so-called multi-partner fertility, delays in age at first birth. It could potentially improve her work, and even possibly her educational outcomes.
Might even lead to more marriage as she finds a better partner. So the idea - and here highly speculative - potential to yield better outcomes for her, for him and for the kid. And hopefully, a
planned kid.
So concluding remarks: Potential for LARCs to level the playing field, churning as a central idea. I think I've said this. I've said that. I think it's important to keep in mind that we are talking
about disadvantaged populations, and so to keep in mind that LARCs are unlikely to solve all of the problems, nevertheless, they are a potential thing that could help.
Final remarks - processes. What I've tried to emphasize here is that sex is one of the processes that we need to think about and that inevitably leads to, again, full disclosure - I'm the gay guy, so
what do I know about people giving births? And it is possible to have a birth without sex, but this is not the way that most of these births are occurring. So sex in the main and as part of the way
that the current state of the world is working, sex is one of the processes that we do need to think about. Barriers to marriage are almost certainly part of the problem as well. Andy made a
compelling case, in my opinion, for that. I would point out that some parts of the literature sometimes take the birth for granted, and it's worth problematizing birth and thinking about ways in
which individuals may exert more or less control over parts of that process. And therefore, sensible to ponder both the sex and marriage parts of nonmarital fertility. Thank you. [APPLAUSE]
MARCIA CARLSON: Hello. I want to thank Paula for inviting me to be part of this panel. It's really an honor to be up here with Paula and Andy and Larry. And I want to take off from where they talked
about some of the processes about how nonmarital births have come about, and how we might change the patterns of nonmarital births, and think about what are some of the implications for families and
especially for children?
Okay. So I want to raise four key points here. The first and - it has been raised already by Andy and a bit by Larry, is that in the U.S., nonmarital births are linked to disadvantage much more so
than other Western countries. Secondly, there are very high levels of instability and complexity in the U.S. And third, many biological fathers are not highly involved, and there are many new "social
fathers" or unrelated men who will be part of children's lives. And fourth, I think this raises issues about child wellbeing, because of fewer parental resources that's not due to marriage, per se,
but what's correlated with marriage, at least in the U.S. So I want to walk through each of these four points.
So in the first point, I spent the last year out of the country. So I've become a bit more aware of how kind of unique or strange the U.S. is in many regards. One place where it's not so different is
in the dramatic rise in nonmarital births that have occurred over the past 50 years. So this figure shows you from 1960 to 2010; the red bar is 1960, the blue bar is 2010. And you can see here that
there are a lot of European countries where there has been a dramatic increase. I put a big blue arrow next to the U.S., so you can U.S. went from about five percent births occurring outside the
marriage to 41 percent over this period. And many European countries are even higher than we are in the proportion of births that occur outside of marriage. What's different, though, is that in the
U.S., nonmarital births are linked to economic, socioeconomic disadvantage much more so than other places. So this figure is somewhat similar to what Andy showed you already. The green line at the
bottom is women with a college degree; the blue line at the top is those with less than a high school degree. The red line is high school or some college, and the purple line is for all births. And
this is basically showing you the percent of births that occur outside of marriage by these three educational levels. And what you can see is that the green line holds relatively flat. Yes, it's gone
up a little bit, but still it's only about nine percent of births to these highly educated women occur outside of marriage. By contrast, for those with less than a high school degree, it's about two
thirds of all births occur outside of marriage. So somewhat ironically, exactly the women who can most afford to have a birth without necessarily a stable partner are the ones that are least likely
to do it.
So turning to some European countries, and the details here may be hard to see in the back but my point is just to show you the lines and how tightly clustered they are. And you can see here I have
six countries; France on the top left, Italy the top center, The Netherlands, Norway is the bottom left, Russia and the U.K. And you see here it's the same lines for highly educated is the green,
least educated is the blue, the red is for those in the middle and the purple is all. And you can see for all of these countries that education groups, except with the exception, perhaps, of the
United Kingdom, seem to be more in tandem. So you see the births seem to be going up, regardless of educational attainment. There is a gap, without a doubt, even in France, a place where there is a
lot of nonmarital childbearing. The most educated are least likely to have the marriage outside of marriage, but you don't see the dramatic difference, except perhaps in the very recent years in the
So the second point here is that there are high levels of instability and complexity in the United States. This is a bit of an older paper, but it's the only data I've seen that looks specifically at
parents breaking up by the child's 15th year of life, as a function of whether parents are married or cohabiting at the time of the birth. So in Europe, there are very few births that occur to
non-cohabiting women that are outside of marriage. So here I'm showing you these two categories, cohabiting and married. And you can see two basic points. The blue bars are cohabiters, the orange
bars are married. And you can see that across all of these countries, the blue bars are notably higher than the orange bars are. So, for example, in Austria, 43% of births to cohabiting mothers break
up by the child's 15th birthday. In France, 58 percent, and so on. The U.S., however, is an outlier still; 78 percent of births to cohabiters among, granted, this older group of parents, broke up by
the time the child was age 15. And the second point is that in all countries, there's a notable difference between marriage cohabitation. So in all of these countries, cohabiting is much less stable
as a union for raising children than marriage, even in places where it's quite common to have a cohabiting birth.
So once parents have separated, it's very likely that they will enter a new union. Then you can see here, I'm showing again a bunch of European countries. The U.S. is on the far right. And certainly
in the majority of countries, people will quite quickly enter a new union; at least some people, with the exception of Italy, which is the lowest bar here. But, again, the U.S. stands out. Within
only three years of the end of a relationship, almost half of U.S. parents have entered a new - a new co-residential relationship. And the figure would even be higher if I went a bit later, six years
or nine years. And the gap increases with the U.S. being unique in the fraction that are going to re-partner.
So, again, looking at stability kind of more generally, this is some work that I'm doing with Jim Raymo at Wisconsin and several other people, in progress. And we're looking at categories of
education. Here, low, medium and high, the same levels I showed you before, looking sort of overall across a bunch of different countries how stable are unions with children, so how likely is it that
children will live with both parents through age 15. And what you see here, it's a lot of bars, I realize. But the big point is that in all of the countries, compared to the U.S., which is on the far
right, the levels of stability are relatively higher, regardless of educational attainment. So you can see that, indeed, there is some gradient across all of countries. The purple bar, the highly
educated, those folks are more stable than the light green bar. So it's slanting downward. But the levels in the U.S. are much lower, so even our highly educated unions with children are stable than,
I think all of, or close to all of, the other categories by country and education. And the gradient in the U.S. is higher. So you can see that, when I highlight this section, the gap between those
with a highly educated mother and a less educated mother is about 4.2 years of spending time with both parents before age 15 in the U.S., compared to about two years in other countries.
Looking at more -- a more recent cohort of data in terms of how these unions break up, this is showing you data from the Fragile Families and Child Wellbeing study. And here we find at the time of
birth, about half of couples are living together. And this goes along with some of the other data that Andy put up. By nine years, the faction has gone up to 70 percent, so fully - more than two
thirds of kids who are born outside of marriage in the U.S. and urban areas will see their parents break up by the time they reach age nine. And again, a more recent cohort shows there are a lot of
new partners. So by five years after nonmarital birth, 22 percent of these urban mothers will find a new partner, and there's a lot of new research that's come out that shows that biological fathers
intend to stay involved with their children, but fathers' new relationships and especially mothers' new relationships often make that more difficult, and they're shown to diminish fathers'
involvement with children. So work by Edin and Nelson in their recent book about fathers and papers by Laura [INAUDIBLE 00:49:51] talk about this dynamic of how when the package deal breaks up, it's
harder for the dad to stay involved. Many of these new social fathers that move in with mothers are quite involved at the time they're there, and I'll show you some information about that in a few
minutes. But I think the question is, will those relationships also break up? And once that father is gone, he probably has very little incentive to stay involved with the child of his ex-partner, to
whom he has no biological tie.
So this can become quite complicated in terms of how families play out. In this diagram, just giving you very basic heuristic information of what sort of families and households could begin to look
like in these different situations. So on the left, I'm showing you a more traditional, kind of nuclear family where you have the mother, that's the blue oval, the father is a yellow rectangle. And
together they have green biological children, right? And so each of them, the arrows show a biological tie that both parents are biologically related to their kids, the mother and father are
together, and they live in the same household. So here, the resources of the family and household, they go into exactly the same investment, sort of arena. When the dad has a child by a new partner,
because still today, at least in the U.S., the vast majority of kids will live with mothers once unions dissolve, you can see that a second household is formed. And typically, the mother, the first
mother that had the child with the father will be living with - I'm sorry - the mother that had the first child will be living with the child she had and the father may have a new partner and be
living with the child in that household, or potentially the father will not be living with any children to whom - that he's had to the extent there's breaking up and kids live with mothers. And so
the dads are likely to be living with none of their biological children and mothers might be more likely to have biological children by different men living in the same household.
And in some - the recent Fragile Families data, we find that this situation is not at all uncommon. So if you look at urban births, these are births that were 1998 to 2000, about one year after a
nonmarital birth, about 63 percent of kids have a half sibling by either the mother or the father. So either the mom had a child by another partner, or the dad had another child by another partner,
or both. Okay, thanks. This goes up by nine years, so that 79 percent - in other words, the vast majority of kids born outside of marriage in the United States, are living in a complicated family,
such they have at least one half sibling. It's not zero among the married parents. I'm showing the married comparison group on the left. It's not to say that this doesn't happen among married
parents, but it's at a much lower level. So it's about a quarter of kids born to married parents will have a half sibling by the time the child is age nine.
In a recent cohort of data from the State of Wisconsin, where I happen to be, this is work by Maria Cancian, Dan Meyer and Steven Cook. And they have looked at nonmarital births in Wisconsin. They
have nearly the full cohort of women who had their first birth outside of marriage, and they find that as you follow kids over time, some kids at the time that their mom is having her first birth,
the dad already has a child by another partner. So the blue shows that they have a half sibling by their father. And as you follow them over time, there are some kids that then end up having a half
sibling by their mother, who has a child by a new partner. That dark purple is only a mother, the purple is if you have a child by both your mom and your dad, a half sibling. The green is only full
siblings and the yellow is no sibling. So what you see, the net year by age 10 is that fully 60 percent of kids born sort of statewide in at least one relatively large Midwestern state have a
complicated family, and that they have at least one half sibling.
Putting this in, again, cross-national perspective, we see that the U.S. is higher overall than at least several other countries that were considered here. So on the left is Australia, and then
Norway, then Sweden and the United States. The blue bar shows you the proportion of all mothers who have children with two or more fathers. So you see that in Australia, for example, it's 12 percent.
And in the U.S., it's 23 percent, almost double. The orange line shows you the percent of those who have two or more children, because obviously you can't have children by more than one partner if
you have only one child. So the fraction of those mothers who have two or more children, it's 33 percent, according to data from the national survey of family growth, have a child with - or, 33
percent of mothers have children with two or more fathers.
So amidst this family complexity, many biological fathers are not highly involved. And as I have indicated, there are many new social fathers that might be living with children. This figure here shows
you fathers' involvement after either a nonmarital birth or a marital birth within the fragile family state and what's happening in terms of the father-child contact at age nine. So earlier I told
you that about - that 30 percent of the nonmarital birth kids were living with their dad because the 70 percent broke up. So the blue bar on the left shows you the 30 percent. And then you see the
red is different levels of father-child contact with a nonresident father. So 29 percent saw their child in the past month, 14 percent saw in the past year but not the past month, 24 percent of kids
had not seen their father in the past year. For married parents, some of them do break up, so 19 percent of the married dads by age nine are not living with their kids anymore, but of these dads, the
vast majority have at least seen their children in the past month, and there's just a relatively small fraction that have very little contact. So among these unmarried kids, fathers' involvement can
be quite low, and tends to decline over time, especially as new partners become involved in mothers' lives and fathers' lives themselves.
So this, to me, raises the question of whether child wellbeing is lower because of these lower parental resources, both time and money. And I think it's important to emphasize, as Andy did earlier,
that it's not necessarily marriage per se, because in many European countries, there's a lot of stability that goes with cohabitation; people might have a child outside of marriage, they might live
together, and then they may marry or may not marry after their first or second child. In the U.S., it seems to be a very different picture with this high level of instability and these low economic
resources. So this is one figure from a paper I have done with Lonnie Berger, where we look at a child aged five, what is the total sum of parental resources that kids are getting? So in other words,
kind of what do kids get from parents if you have these two very broad categories of kind of money and time? And the slightly shaded bars that are sitting kind of next to or behind the more solid
bars show the level of mean or average income in these different households. And the married biological parents have a dramatically higher level of average income, about $77,000. The income level at
all of the other different family types is notably lower, from about $20,000 up to $37,000.
When we look at parenting, we see a different story. So you can see the black here is the level of engagement that kids have with their mothers, and it's a striking similarity, I think, across all
different family types that mothers seem to be quite highly involved, or stably involved, no matter what type of family they're living in in terms of the father partner relationship they have. And I
just saw a recent paper by Liana Sayer and colleagues that I think found a very similar thing, which is heartening. So it seems that mothers are kind of consistently involved. What changes is that -
what father involvement kids have. So the gray bar is biological father engagement, and you can see that regardless of marriage, the left two, whether you're married with a biological father,
cohabiting with a biological father, the level of that dad's involvement doesn't really look different. When you live with a kid, you're similarly involved with that child. What is different is when
you're not living with that child. So the third one over is if you're a mother with a new social father. And you can see that those social fathers, that white kind of bar, they're equally if not more
involved than the biological fathers that are living with children. And that's heartening to the extent that we think fathers' involvement is good for kids. But as I said earlier, will this last over
time if and when this relationship dissolves? You can see that the level of parental investment for the other groups, the three bars to the right, where either it's a mother and social father who are
cohabiting and not married, the mother and the biological father who are dating, or if the mother is single and doesn't have a partner, the level of biological father involvement is much lower. So
the net result is a lot lower level of fathers' involvement.
Okay, so just to conclude, I think in terms of implications, at least in the United States, I would say that children born to unmarried parents are sort of triply disadvantaged. I've talked a bit
about the lower economic resources, as have previous panelists. I tried to show you some information on unstable and complex families, and it becomes even more complex in my simple diagrams to the
extent that people re-partner again and have new kids, but that was just kind of a basic example. The third one, I didn't really focus on, but in the U.S. there's much more limited policy support
compared to many other Western industrialized countries. And that would be another topic in another panel. But I think that's another way that we differ from some of our European counterparts that
have child allowances and support that goes directly to kids, or to mothers for their kids, regardless of their family type or structure.
In terms of what to change going forward, in my view, it's very hard to change family behaviors. The Bush administration spent a lot of money trying to encourage marriage among unmarried couples, and
the net result was essentially zero. There was no positive effect of relationship classes and counseling for families to try to strengthen their relationships. People tend to do what they want to do
in terms of whether they want to be with a partner or leave a partner, at least in the current culture where that's okay. I do think that what Paula talked about yesterday, Larry talked about today,
this idea of really focusing on decreasing unintended fertility has some promise, now that we have the technology of the LARCs and ideas about efficacy and how to change those, and I think that's a
really exciting direction potentially for the future. And then I think policy, we could think a lot more about what are ways to support kids that don't depend on a particular family configuration? I
think a lot of our policies today have very strict ideas about sort of who is in the house and who is not, and who has custody of kids in terms of who gets particular benefits; means-tested programs,
the earned income tax credit and so forth. And I think we might want to think about what are some ways that we can kind of take the family as given, and take children as given that doesn't depend on
particular aspects of sort of family structure or family behavior. And I'll stop for there, and I look forward to the discussion. Thank you. [APPLAUSE]
PAULA ENGLAND: Wow, that was pretty interesting to me. So we are going to take Q and A, and I will get to you in a second. I'm going to let the panelists stay at their seats since they are miked, but
one of the panelists has asked to make a comment. Andy asked to make a comment on Larry's talk, so I'm going to let him go first. And then people can come to mikes and we'll take your questions. And
you can direct them towards a particular panelist, or you can throw it out and they can decide who will take it. Andy?
ANDY CHERLIN: I want to talk a bit more about long-acting, reversible contraceptives, the so-called LARCs. Basically, we're talking about IUDs, or implants, hormonal implants. They're undeniably
effective and, therefore, quite promising. And in a behavioral economic sense, they change the default option. The idea is, if you have to do something to become pregnant rather than doing something
to protect against being pregnant, you'll be more likely to contracept effectively. This was put forth, as Larry mentioned, in a book by Isabel Sawhill, an economist at the Brookings Institution last
fall, called Generation Unbound. I want to raise two caveats, however, which Larry might agree with on. First of all, it's not going to be enough to just hang up a shingle outside of the clinic and
say, "free LARCs available." Instead, we have to give disadvantaged women a reason to want to postpone childbearing. Unless they have that reason, they're not going to come get the LARCs. So this is
not a technological solution; one needs motivation, which is a socioeconomic issue, not a technological one. The second is we have to be aware of how much resistance will be visited upon this
approach if it becomes widespread. The people who are in favor of LARCs, if they're not careful, perhaps even if they are, will be seen as restricting low-income women's rights to have children.
We've already seen some of that. Demographers have studied issues in other countries that have had sterilization, or other long-lasting programs, we're familiar with the tremendous result.
So that doesn't mean we don't go for LARCs. I think it's an important and promising part of the solution to nonmarital childbearing. But there is no purely technological solution. Instead, one needs,
in addition, a socioeconomic solution to make all of this work.
PAULA ENGLAND: Larry, do you want to comment?
LARRY WU: So I think Andy raises absolutely pertinent points. So one of the ideas behind this "intent to treat" experiment would be to see exactly what uptake would be. And I think we don't know
that. On the flipside, many of these women do report these first births as unplanned. And the ethnographic evidence, at least some of it - some of it is old - many of these women, after having the
first birth then report that after considering things, she loves the child, but it was probably a bad idea to have the kid then with that guy. And so I think we don't know enough about these issues.
LARC usage is low. It does change the default, and the other issues about reducing fertility, again, I would return to the planning stages issue. I think there is good evidence there is enormous
ambiguity that surrounds getting pregnant in many of these populations. And so one of the early findings from Fragile Families was that the mother and father were incredibly happy when they learned
she was pregnant, incredibly happy around the birth of the child. And then a year or two later, he is gone. And she has this kid as a young woman. So I think those are the issues we need to think
PAULA ENGLAND: I will restrain myself from jumping in here. Yes, we have our first question, or second question.
SUSAN HIGGINBOTHAM: Hi. I'm Susan Higginbotham with the Society of Family Planning. I'm not a sociologist or a researcher, I fund contraceptive and abortion research. So I was coming to raise some of
the points that you just did about LARCs. And I wanted to follow up with that a little bit further because, I mean, it has been shown what the uptake is if you eliminate the cost, and they, you know,
in the choice study and at the University of Colorado, they didn't just eliminate the cost. They designed a specific counseling model to make sure that women knew what the most effective methods
were. Secondly, and I'd love to have the panel address this point - Dr. Wu, in your presentation, you were saying that you thought there should be an RCT. And I'm really not sure that it would be
particularly ethical to deny a control group information about the most effective methods. And then lastly, thank you for raising the point about because of the history in the United States, about
coercion around sterilization, particularly for low-income women and women of color, you know, there is a whole now body of work the last couple of years looking at, you know, from the reproductive
justice framework, potential coercion about targeting these methods towards low-income women. And the fact that in some, in some OB-GYN offices, there are physicians who are asking women to sign a
form saying that they will keep the IUD the full five years and not have it taken out. So anyway, thank you for raising the issues, and if you could address those.
ANDREW CHERLIN: I'm certainly not anti-LARC, but the studies that have been done which have shown that they're very effective have been of young women who've come to a contraceptive clinic because
they want family planning. It's not been a random sample of people out there in the population, which could be quite different. I'm suggesting that LARC should be a part of a larger strategy for
dealing with the problem of nonmarital childbearing, and that we should resist the idea of a quick, technological fix. If we do that, it could be quite effective and it really could make a dent in
this problem.
PAULA ENGLAND: I want to make one comment. You know, in one sense, we have a natural experiment going right now, because a feature of the Affordable Care Act was that, at least for those who have
insurance under Obama care, shall we say, any of those plans may not charge copays for any contraception; correct me if I'm wrong on this. Now, our friends at the Guttmacher Institute also say this
is not being well enforced right now, and, you know, there may be some lag time here. But it seems like we are about to be in an era where this is available freely to people, so if you want to do a
random control trial, that can't be exactly the intervention because that may actually already be the default as soon as we get the enforcement beefed up on this, which may take a long time, I don't
know. There was another question here, let's go to the next question. Yes, you had a question.
>> Thanks. Thanks for that interesting, effective presentation. Has there been, what do you all think about widespread free vasectomies?
PAULA ENGLAND: Widespread, free...?
>> Vasectomies.
>> Widespread free vasectomies. A lot of the emphasis has been on the women's responsibility. We saw the low rates of condom success, withdrawal success. Maybe I missed it, but I don't think I heard
anything about vasectomies.
PAULA ENGLAND: Comments? Larry, you want to take it?
LARRY: I have no problem with that.
PAULA ENGLAND: Larry says he has no problem with it. I'm going to make one comment on it, which is, I think that is a much better proposal when we're talking about, do people have the means to end
fertility when they've had the number of kids they want to have in their whole life, right, because it's not long-lasting, it's permanent. And we've been talking a lot about fertility among young
people like 18 to 25 when probably, and certainly when they're having the first birth, almost no one would take that up voluntarily. I mean, a few people would because most men and couples would want
another child at some point. So I think, in a way, that's a better policy intervention for can people not have more children than they want, but less so for the timing. But I agree with you that we
shouldn't only look at female controlled things. No, there was someone going up to this mike over here, if they want to come back. Okay, here's a question.
VICTOR AGADJANIAN: Victor Agadjanian, University of Kansas. I do most of my research outside the United States, and when I come today, I say I really rejoice any attempt to compare the United States
with other countries. So I appreciate these comparisons with Western Europe. Again, we don't hear too much of that, unfortunately, at ASA. But I want to turn your attention to another part of the
world, the global south, and what we can learn from their experience. I work a lot, especially in Africa, especially in rural Africa. And when I go to rural clinics and talk to nurses and the
material help clinics in, say, rural parts of Africa, the first thing I hear about from nurses is about efficacy, what we would call efficacy, and how LARCs can increase that efficacy and level the
playing field. There has been a lot of research on these issues, and well, I'm more familiar with [INAUDIBLE 01:12:16] in Africa. There's a lot of policy debate on those issues, and I think it is, of
course, we need to make a lot of corrections. You know, the playing field that we're talking about, they're talking about there is a gendered playing field. It's primarily women standing up to men,
sort of, and protecting their own reproductive health and reproductive rights. But I urge you to think and broaden sort of the geography of our comparison. We can learn a lot from many developing
countries and people who are - who don't even own a smartphone, but yet can teach us a few things. Thank you.
PAULA ENGLAND: Thank you. Anyone want to comment?
ANDREW CHERLIN: You mention the word efficacy, I believe. We did hear a talk yesterday that suggested that might be important. So are there, perhaps, some lessons that we can draw from other
developing countries where increasing the level of efficacy, especially among women who don't feel in control much at all about their fertility, could be effective? It seems so me those lessons
reinforce Paula's concern that even in the U.S., there might be levels of efficacy. And one could increase efficacy among some groups of young women and men, and let them simultaneously have more
fertility and have more likely to have the goals for their lives that they want.
MARCIA CARLSON: Just one small comment, I think it's a wonderful idea to look at other countries and other regions. I think the challenge is the data, finding comparable data. So I love ideas, if you
know of kind of data sets that are similar across countries, because when you work with these, you realize how challenging it is, even when surveys that are, ostensibly, kind of the same or similar
in countries and you try to merge them together, it becomes very complicated. So I'lI look forward to more data coming online from many other places. And I think that's a great suggestion.
ANDREW: And Victor, I was also drawing a negative lesson from the experience of sterilization campaigns in many developing countries, which were fiercely resisted. Now, one can do a better job. One
can learn from those campaigns how not to do it and how to do it. So I'm not suggesting that we'd have that level of resistance. But we can learn how much people resist anything they see as a
restriction of their rights to have children.
PAULA: Do I see [Lane?] coming up to the mike? Yes?
>> I have a question for Andy. I'm curious about the explanation using income inequality as a driver in the rise in nonmarital births, simply because we saw nonmarital births increase just as rapidly
in the '60s and '70s. And so I wonder, you didn't say anything about the size of the estimated effect from your results, but so what do you say to someone who says, "Yes, but..."?
ANDREW CHERLIN: Okay, first of all, I'm really suggesting that income inequality is a marker rather than a cause itself. It's a marker of what's happening in the labor force. But what I'm saying to
answer your question is, the nature of nonmarital childbearing changed after 1980. Prior to 1980, it was mostly single mothers, many of them younger, who were having babies. That's kind of the - the
picture that many of us have in our minds. That began to change just around 1980. And what we've seen over the last few decades is a very more specific kind of change in which the modal - mother
having children who's in her 20s or early 30s and cohabiting with the father at time of birth. It's that shift around 1980 that gets me thinking about what else happened in 1980, okay? What else
might have happened in 1980 that might have just affected, or mostly affected, people in the middle, whose position in the labor market is in the middle. So it's that analogy, that difference, the
transformation of this issue after 1980 that leads me to think, well, what else happened around then, which was an increase in economic inequality? Having said that, I really believe, as I said at
the end of my talk, I've not nailed this down. Okay, we need to do more work about this and understand if inequality is part of the story, why is it more part of the story? But it is true that this
issue and these statistics change greatly right around the time that the changes in our economy were affecting most of the people who have shown the most change in nonmarital childbearing.
>> Can I ask a quick follow-up?
>> So what do we know about the timing of the rise in cohabitation? So my prior would be that a lot of the rise has to do with people who would formally not be married but they live apart, and so we
get a change in the norm so that people in the 1960s and 1970s who have nonmarital births, but they simply didn't live together. Then they started living together. Maybe they think about getting
married and maybe they don't. So I'm just wondering whether it's a really - it's a qualitative significant shift or just one that has to do with...
ANDREW CHERLIN: The rise of cohabitation?
>> Yes. Yes.
ANDREW CHERLIN: Yes, so the question is, is this just about the rise of cohabitation, which happened starting around the 1970s and so forth. Certainly the rise in cohabitation and the greater
acceptance of nonmarital births in the population is a very important cultural part of this explanation. But one has to ask oneself, why didn't college-educated people start cohabiting and having
children within marriage? Why was it just in the middle that we saw this growth of nonmarital childbearing within cohabiting unions, but not people with a four-year college degree? I would argue
because the people with the four-year college degree are in some sense the winners in our transformed economy. And they've been using their incomes to form stable families and have children after
marriage. So again, it's the differential effect, the fact that we don't observe cohabitation increasing and having the same consequences for everybody, but rather, having the most consequences in
the middle, and not at the top, that gets me thinking there must be something more than just a rise in acceptability of cohabitation.
PAULA ENGLAND: I want to comment on part of Lane's question. I think you were - one of the things you were asking is, is the upsurge in cohabitation, is that traded out for marriage versus single. I
actually think no, that the answer there is marriage, because there is a paper, I think it's by Wendy Manning, that shows that the age at which people form their first co-residential union hasn't
changed much. It differs by education level, but it hasn't actually changed much. So that suggests that the, as marriage is getting later, the time that used to be - the things that used to be
earlier marriages are now cohabitations. So - I think there was a gentleman here that wanted to ask a question, someone? Oh, yeah, we'll go here first here and then to you.
>> Well, I've been thinking about how this idea of income inequality being a driver for the rise in nonmarital fertility kind of links up with Paula's excellent talk yesterday, in terms of the role of
efficacy being a mechanism. And I'm just kind of throwing some ideas out here, but I do work on family and health, and so there is a large literature in both mental health and physical health,
medical sociology on the importance of childhood adversity and the influence that that has on a lot of psychosocial processes through the life course. So we know that childhood adversity is a strong
predictor of this sense of mastery; personal control, efficacy and even fatalism, right, the view that you can change what happens to you, that you're going to live to a particular age, and these
kinds of things. It's also a driver of social mobility, right, because it's linked in with that. And so anyway, I am just thinking, if we look even further back the chain and go all the way to
childhood adversity, that may be fruitful in explaining some of these longer term outcomes. We also have a study that we found that childhood adversity is strongly linked particularly to black men's
ability to form secure and trusting and lasting personal relationships that really takes a toll on that. And so I think that might be another important piece of the puzzle.
ANDREW CHERLIN: That's a very nice way to link what I was saying today with what Paula was saying in her talk yesterday, which is, social forces constrain things like efficacy and income equality, or
the state of the labor market could be the kinds of constraints that may be seen.
PAULA ENGLAND: Yeah. Yes, gentleman over here.
>> I've often wondered, and I didn't hear anyone say this yet at this conference, that the rate of marriage has been decreasing since the No-Fault Divorce Act was put in effect in 1960. So, for
example, when No-Fault Divorce was implemented by all 50 states in the U.S., by 1986, I believe, the marriage rate has dropped to - from 90 percent, this year I think it was 60 percent, of eligible
adults. So and also, the tax policy of the United States rewards people that are married and who itemize. So there's, for a couple making $100,000 more or less, there is a $6,000 break from a single
person filing singly. And that that rate then increases up to about $250,000, and then the marriage penalty occurs. So for people that are in the middle class, which I guess would be between $50,000
and $250,000 income, there is a benefit to being married, if you're itemizing. So it seems that there are these incentives to marry and also incentives to divorce, because if you divorce, you get,
you split the assets, more or less. And I was just wondering if any of this, if you thought that any of these kinds of calculations had any effect.
ANDREW CHERLIN: I have been talking too much. Would one of my partners care to talk, otherwise I'll continue to filibuster.
MARCIA CARLSON: Well I think there's - I'll just way one thing - I think there's some work that's looked at kind of the effects of policy on marriage and nonmarital childbearing. And I think at least
the latest summary of research I saw, yes, there might be small effects, but they're small. And they're not driving the whole trend that we've seen. And especially when you look at other countries
that support disadvantaged families much better than the U.S., and yet have much lower rates of unintended fertility at least and so forth, you can see that it's not - I don't think that that's the
main driver, is that people are responding to these policy incentive effects. But others should comment as well.
PAULA ENGLAND: Yes, question over here. Monica?
MONICA CAUDILLO: Oh, my name is Monica [Caudillo?] from NYU. I just was wondering if you could talk a little bit more about different types of social forces that are affecting the rise in nonmarital
births, specifically long-term socioeconomic sort of [INAUDIBLE 01:23:19], say, economic instability [INAUDIBLE 01:23:24] and if you had the opportunity to encounter different effects of this on
these types of outcomes, and how you could disentangle? Because they're causing very different effects on individuals, I am guessing.
ANDREW CHERLIN: It's very hard to disentangle what is going on. That's the sense I really meant it when I said in the talk that those of us who use the trendy term "inequality" have to start thinking
more about what it means. If we're just talking about long-term, concentrated disadvantage, there's no point in using the term "inequality." It must mean something different and special. So we must
try to start ferreting out what, if anything, it is about inequality or what ways in which inequality might be a marker of other process that make a difference. And it's possible that in many cases
we'll find it's not inequality that makes a difference, but rather, long-term concentrated disadvantage. So, look, income inequality, we're told, is the issue of our time. I think I may agree with
that. But at the same time, I think that as researchers, we need to do a better job now of thinking, what do we mean by this, and how and where is it important?
LARRY WU: With respect to these instability issues, some of my early work asked to what extent instability in parents' situation while growing up might contribute to a young woman having her first
birth outside of formal marriage, and coupled with that, how about economic circumstances in her family of origin? And so those papers found effects of multiple changes in family situation didn't,
indeed, increase the risk that a young woman might have her first birth outside of formal marriage. Low income was absolutely a driver of these risks, but also if income was increasing in the family
of origin, that would decrease these births. So I think all of those are potentially at play. There is some evidence it is very difficult to disentangle these issues.
PAULA: Yes, question here.
>> A question for Dr. Carlson and one for Dr. Cherlin. For Dr. Carlson, we have been problematizing nonmarital births in this discussion today. I'm curious, in Western Europe, do they use the same
social problem language to describe nonmarital birth as we seem so enamored with? To Dr. Cherlin, 50 years after the Moynihan report, was Moynihan right?
MARCIA CARLSON: So I would say it's not problematized at all in the same way in Europe, because it's a very different phenomenon. So in many cases, as I have showed, you have much less of a gap in
terms of how it's correlated with education, right? So many people have births outside of marriage, including people with high education, and they are often quite stable. So they'll have a child or
two children, and then they might get married or they might not get married. It's very much accepted to be an unmarried partner. In France they have something called PACTS, I can't, at this moment,
remember exactly what it stands for, but it's basically a civil union that's perfectly legitimate. There is no need for an actual marriage per se, and it's really seen as kind of a personal option.
So I think the fact that it's not highly correlated to disadvantaged there, it means it's not problematized. It's more of kind of an option, even to the extent you do find some socioeconomic gradient
ANDREW CHERLIN: I guess what I would say about that questions is, I think what children need are stable parental relationships. Here in the U.S., we tend to do that still via marriage. But it's not
the marriage per se, it's the stable relationship, and in many European countries, there are many more stable, long-term relationships. Was Moynihan right? Thank you for asking me that question.
Moynihan used some unfortunate impolitic language in the Moynihan report, which he may have thought only President Johnson and a few other people would see. And I'm sure he would have taken back that
language. He also didn't have a great deal of strong, quantitative research in the report. Nevertheless, his basic point of rising single parenthood was certainly true, and it's certainly been true
across the racial and ethnic diversity of this country. And while many single parents do a fine job, I think it's safe to say that children do better with stable two-parent relationships. In that
sense, yes, I think a lot of what Moynihan said was right, but I certainly would not sign on to the whole report, nor the idea of tangle or pathology and other kinds of rhetoric that's been
associated with it.
PAULA: Do we have any other questions? If not, please join me in thanking our wonderful three panelists, and thank you for coming! [APPLAUSE]