The Economics of Teenage Pregnancy

Teenage motherhood is well correlated with poor economic outcomes. This of course need not mean that teenage motherhood causes poor economic outcomes; in fact, Melissa Kearney and Phillip Levine (of U. Maryland and Wellesley College) argue precisely the opposite: Being on a low economic trajectory causes teenage motherhood, and conditional on that original trajectory, teenage motherhood does little economic harm:

Why is the rate of teen childbearing is so unusually high in the United States as a whole, and in some U.S. states in particular? U.S. teens are two and a half times as likely to give birth as compared to teens in Canada, around four times as likely as teens in Germany or Norway, and almost ten times as likely as teens in Switzerland. A teenage girl in Mississippi is four times more likely to give birth than a teenage girl in New Hampshire—and 15 times more likely to give birth as a teen compared to a teenage girl in Switzerland. We examine teen birth rates alongside pregnancy, abortion, and “shotgun” marriage rates as well as the antecedent behaviors of sexual activity and contraceptive use. We demonstrate that variation in income inequality across U.S. states and developed countries can explain a sizable share of the geographic variation in teen childbearing. Our reading of the totality of evidence leads us to conclude that being on a low economic trajectory in life leads many teenage girls to have children while they are young and unmarried. Teen childbearing is explained by the low economic trajectory but is not an additional cause of later difficulties in life. Surprisingly, teen birth itself does not appear to have much direct economic consequence. Our view is that teen childbearing is so high in the United States because of underlying social and economic problems. It reflects a decision among a set of girls to “drop-out” of the economic mainstream; they choose nonmarital motherhood at a young age instead of investing in their own economic progress because they feel they have little chance of advancement.

The full paper is here.

Are you convinced?

Print Friendly, PDF & Email
Share

34 Responses to “The Economics of Teenage Pregnancy”


  1. 1 1 J Storrs Hall

    Funny how they concentrate on just the age of the mother and not on whether she gets married, which I would have thought was a much more important causal factor.

    It does seem that there is a fairly strong evolutionary pressure here to an r (as opposed to K) evolutionary strategy: http://en.wikipedia.org/wiki/R/K_selection_theory

  2. 2 2 Sconzey

    I am aware of two distinct classes of teenage mom; those who get pregnant intentionally, either for pragmatic economic reasons (lucrative welfare payments) or simply because they feel that is the only way they can make a difference in the world; and those who get pregnant unintentionally through failing to use contraception.

    With the first class, it is interesting that for a girl with few prospects of her own aspiring to be a single mother is more socially acceptable than aspiring to be a housewife. (paging Charles Murray!)

    With the second class, I wonder if the low conscientiousness and high time preference implied by the failure to use contraception also had a hand in causing the girls economic woes.

  3. 3 3 Ken B

    I bet the data behind this assertion: “that variation in income inequality across U.S. states and developed countries can explain a sizable share of the geographic variation in teen childbearing” will be confounded by generational effects: children of teen mothers. Looks like an endogeneity issue to me.

    No, I am not convinced. Aside from its inherent implausibility there are bits of wording here that suggest this is motivated research.

  4. 4 4 Kirk

    It is not unwed mothers who suffer consequences, it is the children of unwed mothers who start off in a deep hole in life. They are statistically more likely to suffer a host of consequences from drug use, crime, illiteracy etc.

    Whether this is a harm to the economy as a whole is doubtful, but it is certainly a social ill, and a real issue to the children affected.

    Obviously many children will overcome these long odds, and there are certainly many examples of smart, capable single parents, but they’re not really who she’s talking about, are they.

  5. 5 5 nobody.really

    Dunno. If evidence shows that 1) teen motherhood correlates with poor economic prospects (relative to some peer group) and 2) teens with poor economic prospects achieve similar economic outcomes, regardless of whether they have kids, then the explanation would seem plausible to me.

    Analogously, I’d heard that the US leads the world in performance on standardized tests – if you control for poverty. That is, kids from US high schools with a 1% poverty rate perform better than kids from other nation’s schools with a 1% poverty rate; kids from a US high school with a 50% poverty rate perform better than kids from other nation’s high schools with a 50% poverty rate. The US performs relatively poorly in national rankings because, among industrialized nations, the US has a relatively high percentage of its kids living in poverty.

  6. 6 6 neil wilson

    Just a minor point about another teenage mother.

    She got married to the father of her child when she was only 18 years old and 3 months pregnant. Unfortunately, she didn’t know the father was already married.

    In any event, the child of the teenaged mother turned out pretty well.

  7. 7 7 Harold

    It is striking that the USA has such a significantly higher teen birth rate than other developed countries.

    The main thrust of the paper is that teen birth rate is high because young women are choosing to have babies early. They do this because they have such low expectations that they believe delaying birth will not improve their situation. The review in the paper indicates that they are right – teen motherhood does not massively affect their life chances. They do not just see themselves as trapped in poverty, they actually are trapped.

    Policies attempting to reduce teen preganancy by increasing access to, or knowledge of contraception will have little effect, they argue. The main factor is inequality – you won’t get teen pregnancy down unless you tackle inequality.

    There is a confused bit about unwanted pregnancies. The teens themselves declared that the preganancies were mostly unwanted. This contradicts the above analysis. To rationalise the two findings, it is necessary to invalidate the stated intentions, and assume that the pregnancies were actually wanted. The literature on “wanted” or otherwise appears unclear, and it is not a black and white issue, but rather a spectrum of want. I personally can see that what people say they intended was not necessarily what they actually intended, even if they believe it themselves. Our instincts – particularly the reproductive one – is very powerful, and acts at a deep, unconscious level.

    The authors struggle to explain the drop in teen pregnancies in the USA in recent decades, even though measures of inequality have not changed. I can see one answer – overall equality might not have changed, but surely womens opportunities have increased a lot with feminism.

    They briefly address the issue of general unmarried births, as J Storrs Hall mentions, but only to say that it is a different, and complicated issue.

    An interesting article.

  8. 8 8 Tim M

    This makes the case that teenage mother’s poverty isn’t affected by their motherhood. Here is how I see it considering my position that immorality is the most likely predictor of poverty.

    Take a group of teenage girls growing up and living in poverty almost all not developing a good moral character. This is a valid assumption because if the ones who raised them had good moral characters they would have worked themselves and their families out of poverty with the opportunities available here in America. Some of these girls make the immoral choice of sexual relations many of whom become unmarried teenage mothers. No morals have changed with motherhood hence the teenage mothers and their peers remain mired in poverty.

  9. 9 9 Seth

    “…they choose nonmarital motherhood at a young age instead of investing in their own economic progress because they feel they have little chance of advancement.”

    I’m not convinced of that. Did they happen to include the amount of state transfers to unwed, young mothers in the analysis?

  10. 10 10 Mike

    I find the conclusion plausible, but the slipperiness of usage between “poverty” and “inequality” is a problem.

  11. 11 11 Ken

    nobody.really,

    That is, kids from US high schools with a 1% poverty rate perform better than kids from other nation’s schools with a 1% poverty rate

    This could be an apples to oranges comparison. After all, in the US the median income in the bottom quintile is higher than the Indian median income in the top quintile (Meaning that the US has eliminated actual poverty. What is now termed “poverty” is simply relative and not absolute). The difference is smaller when compared to European and some Asian countries, but Americans still have one of the highest per capita incomes in the world.

    And lastly, I really don’t see why motherhood is a government issue to begin with, even teen age motherhood. It certainly can’t be a federal issue as I have yet to see anything in the constitution saying that the federal government has the authority to influence who should and shouldn’t be having children and when.

  12. 12 12 Doctor Memory

    I am profoundly dubious. I’m perfectly willing to accept that the relationship between teenage childbearing and poverty is not as simple as a one-way function, but any argument that the function is one way in the other direction is going to have to carry a great deal of water to get around some basic biological and logistical facts: babies take time. Babies take money. Time and money that, for anyone not already wealthy (and/or embedded in an excellent social support structure) to begin with, are likely going to be paid at the expense of pursuing either higher education or vocational job training.

  13. 13 13 Floccina

    It reflects a decision among a set of girls to “drop-out” of the economic mainstream; they choose nonmarital motherhood at a young age instead of investing in their own economic progress because they feel they have little chance of advancement.

    I find it hard to believe that they get pregnant because they feel they have little chance of advancement. I bet that there is little evidence behind that assertion.

  14. 14 14 Floccina

    And further why would they feel they have little chance of advancement? Would it be because they have done so bad in school?

  15. 15 15 Roger

    The main point of the article is that the teen birth rate is high because teen girls have economic incentives that favor teen births. Education will not help unless the incentives are changes. For the economists here who believe that people respond to incentives, what is surprising about any of this?

  16. 16 16 Kirk

    As a Super Tax Genius I will tell you that the economics of the Earned Income Tax Credit is enough incentive to encourage children. When I worked in a low income office we were inundated with mothers (and fathers) who knew every in and out of maximizing the EIC with a minimum of work. When you consider that $10000 of income can translate into $5000 of EIC money (VERY rough numbers, but the scale is right) it is not hard to figure out why teenagers who have watched their parents ride the gravy train are ready to hop on board – or – are encouraged by their parents to produce another child that can qualify THEM for more EIC.

  17. 17 17 Martin-2

    Roger – The existence of economic incentives that favor teen births is surprising to me.

    Mike – “the slipperiness of usage between ‘poverty’ and ‘inequality’ is a problem.”

    I’m with you, and I’ll add the term “social mobility”. I can believe freezing the income of those at poverty level would increase the rate of teen pregnancy. But the article presents no evidence that taxing 6 digit earners and burning the money would have a similar effect. Even so, “social mobility” and “wealth inequality” are used interchangeably and the same evidence is attributed to both factors.

  18. 18 18 Harold

    Tim M. I had thought this sort of attitude ended the century before last. I am very out of touch.

    Seth: “Did they happen to include the amount of state transfers to unwed, young mothers in the analysis?” Yes they did, and found that it had a small effect, but nowhere near enough to be the full explanation.

    Doctor Memory: the point is that these women are not going to pursue either higher education or vocational job training, with or without children, because they see no point in it.

    Floccina: They think they have little chance of advancement because they actually do not advance. According to the paper, girls with this background do equally badly whether they have children or not. Girls with these backgrounds have a very low economic trajectory. If they had a different background, maybe they would be equiped with the skills, confidence, contacts etc that would enable them to study and work hard to advance themselves.

    Roger: What are these economic incentives for teen births? I do not see them in the paper. I interpret the paper to say there are no economic incentives to delay birth. Most proposed “solutions” to the problem of teen pregnancy assumes that the teens wish to avoid pregnancy because of an economic incentive, and so providing them with the tools to do so – education, access to contraception etc., will reduce teen pregnancy. This paper says that this assumed economic incentive is not significant, and so these solutions will not work.

    As far as we can say from the data, the USA has a slightly lower rate of teenage sexual activity than other developed countries, and a significantly lower teenage use of contraception. Differences in abortion rates were found to be insignificant in this context. I think this data is as reliable as we are going to get on these matters, and thus the explanation for the indisputably higer rate of teen births in the USA must be down to the lower use of contraception. The usual “solutions” assume that the contraception use is low through inability to use contraception, by lack of access or knowledge. This paper says it is through choice, even if that choice is not recognised by the teen mothers themselves. Further, it says that this choice is only marginally influenced by benefits and welfare payments.

  19. 19 19 nobody.really

    What are these economic incentives for teen births?

    Allegedly the life of a young poor woman can be a drag. Socially, men regard you as prey. But if you achieve the status of a mother, people accord you greater respect and deference. The sexually aggressive/drug dealers/muggers/rapists pull back when they see a young woman with kids. And even relationships with parents change, as the focus of conversation shifts from “What are you doing with your life?” to “How are the kids?”

    To coin a phrase, motherhood has its privileges.

  20. 20 20 Ken B

    @Harold: You wrote “Tim M. I had thought this sort of attitude ended the century before last. I am very out of touch.”

    Just change the word ‘morals’ in Tim’s comment to ‘welfare maximizing’ and read it again.

  21. 21 21 Jace

    I’m skeptical.

    I’m almost willing to say the article makes zero sense.

    What you’re essentially asserting is that children do not cost more, which is ostensibly incorrect.

    While I wouldn’t say that teenage pregnancy immediately consigns the young mother’s fate to that of a penurious itinerant, I also wouldn’t declare it has little to no negative pecuniary corollary.

    Having children costs more than not having children. The only time the cost does not rise to have a child (as opposed to not having one) is when there are entities extrinsic to your own aiding you and your child. In that way, it’s easy to see how middle-class teenage mothers will see their parents, and nuclear family do their best to assuage the albatross of having a child at such an unprepared age.

    These palliatives may manifest in myriad form: money, diapers, food, discounted or rent free shelter, and babysitting. All of which assist in mitigating monetary and opportunity cost, thus affording the better-off teenage mother luxuries a lower income teenage mother will (probably) not have as readily available — if at all.

    Moreover, the lack of these very same catalysts should keep a less fortunate teenage mother down, or in her own economic lane. Ergo, the conclusion of “babies have no significant economic effect” would be wrong. It is merely masked by the already existing circumstances.

    But, that’s just my theory.

  22. 22 22 nobody.really

    Having children costs more than not having children. The only time the cost does not rise to have a child (as opposed to not having one) is when there are entities extrinsic to your own aiding you and your child.

    Imagine I showed an Australian bushman a picture of a Franciscan monk and asked him to advise me about whether to become one. “How expensive!” he might respond. “Just look at those robes. Surely it would be cheaper to avoid such a profession; then you could simply wear a loin cloth as I do.”

    I couldn’t fault the bushman for concluding that robes cost money. But I might fault the bushman for assuming that, in assuming a social role that would not require me to buy robes, I would not need to incur other substitute costs. The bushman’s ignorance of my society would lead him to make such errors.

    Similarly, unless we know something about the social roles that a young, prospect-less girl may chose among, we’re likely to make similar errors. Yes, we recognize that the social role of being a mom has costs. But what do we know about the social role of NOT being a mom? Speaking for myself: not much.

    But I had the pleasure of reading Tally’s Corner once upon a time, an anthropological study of single black men in 1967 Baltimore (?). Among other insights, the author discovered that these men lived based on social networks of sharing, kind of a mutual aid society. To withhold resources from your friends would be to cut yourself off from your aid group. But, predictably, this basically meant that capital accumulation was impossible. Street men who spent their capital quickly were acting rationally within the context in which they lived, but people living outside that context would likely draw different conclusions.

    I could imagine that a young, prospect-less girl may feel the need to share whatever resources she amasses in order to maintain a social network. I could imagine such a girl might also expend resources in recreation. And I could imagine that a young mother might be excused from both such expectations, and instead be expected to harbor resources for child-rearing and to refrain from certain nightlife activities. In short, the social role of NOT being a mother may not be costless.

    But I’m just spitballing here. Unless we know something about the alternatives that society offers to prospect-less young women, we’re not in much of a position to judge. Or, as Herman Melville remarked, “Of all the preposterous assumptions of humanity over humanity, nothing exceeds most of the criticisms made on the habits of the poor by the well housed, well warmed, and well fed.”

  23. 23 23 Vald

    @nobody.really

    The example you give is completely meaningless. Australian bushmen were, essentially legendary, white men who survived out on the Australian bush living only off the land. They also wore plenty of clothes (just like other Anglo-Australians). You are talking about the Australian Aboriginal peoples.
    The term bushmen was a derogatory term used to refer, generally, to the peoples of southern Africa.

  24. 24 24 Kirk

    Super Tax Genius again. What’s interesting about unwed mothers in the tax office experience is that, oftentimes, the costs of having and raising children are borne by others (parents, boyfriend, society) while the benefits, both economical (massive tax returns, AFDC, etc.) and social (status as mother, assumed maturity, bonding with others) are received primarily by the mother.

  25. 25 25 Harold

    Ken B -regarding Tim’s post, I am not sure what “almost all not developing a good welfare maximising character” really means.

    Nobody.really: There are benefits to motherhood, which is why the majority of women become mothers. The benefits you discuss I would not have thought of as economic ones, but I suppose that could be semantics. If we assume motherhood is a desirable state, the only factor in whether the woman is a teenage mum is whether she chooses to delay the birth.

    Your comment to Jace is spot on – except the “Australian bushman” reference – I agree entirely with the point. The other point to note is that the fact that teen birth had no econmomic impact was stated by the authors to be surprising. It was a conclusion based on evidence. It is no good to say “I don’t think that’s what happens” – you have to challenge the evidence or its interpretation. Jace’s point that it costs more to have children is why the result is surprising. It does not invalidate the finding – perhaps because of factors Nobody.really identifies.

    Vald – are you sure that the term Bushmen is derogatory? I thought Bushmen was a neutral, descriptive term for certain indigenous Africans. Australian bushmen are of course a different matter.

  26. 26 26 iceman

    nobody.really – interesting re-frame, however I think the changes in our social policy since 1967 are precisely what many here (like Kirk) are pointing to as primary factors in re-shaping people’s “rational” responses, possibly in a less socially productive way. On this you might find “Losing Ground” by Charles Murray a worthwhile read. Some would say we’ve destroyed the need for the ‘social networks’ you describe. You’re probably right that the women in question would find other things to spend money on, but what of the income they spend? It seems tautological that childbirth at a minimum delays their lifetime earnings and ability to invest in their human capital.

  27. 27 27 Vald

    @Harold

    As originally used, it was essentially neutral, or as neutral as any term used by Europeans to describe people they viewed as culture-less and of the land. Anthropologists continued to use it for decades as a neutral term, and that may still be the case.

    In southern Africa, opinions vary, so far as I know, on whether it is now derogatory or not. The people who were *the Bushmen* have rejected the term in favor of calling themselves the San. Not all of them find it to be a pejorative term, which I understand to be similar to some American Indian groups that now accept externally imposed derogatory names because that is what they are commonly
    known as (Navajo, for example).

    The term IS, however, rejected across the board by African people when it is use to refer to *all* south Africans as if they were all, regardless of nation or culture group, landless heathens running across the bush in loincloths. Although nobody.really was talking about the nonexistent Australian bushmen, I was placing it in this category as a derogatory term because of the presumptions and stereotyping in his entire statement.

  28. 28 28 Harold

    Vald – if the San have rejected the term, then I am happy to oblige.

  29. 29 29 iceman

    A couple of other thoughts:

    1) As with health care stats, it seems cross-sectional analysis is complicated by the heterogeneity of the US population. Here the authors acknowledge it’s largely a southern (and particularly Hispanic) ‘problem’, yet I see no good comps for that.

    2) “We conclude that women with low socioeconomic status have more teen, nonmarital births when they live in higher-inequality locations, all else equal”.

    So this seems to be another application of the idea that relative wealth disparities *per se* are a problem; in this case, people feel less “marginalized and hopeless” when everyone else is equally poor, than where there may be examples of the possibility of improving one’s station in life? The authors acknowledge their thesis doesn’t explain the declining trend in recent decades, but even if true, do we best design policy to counter or entrench such psychology? Seems like a dangerously self-fulfilling presumptive leap from “most” to “all”; at a minimum, childbirth would seem to *ensure* the women in question remain on the same low trajectory. Of course the authors don’t actually offer policy alternatives to create more motivating “perceptions of opportunity” to overcome ‘cultural’ barriers / norms. They indicate that if anything *decreasing* transfer payments reduces births at the margin, and as a minor measure suggest more college aid, but is anyone who is *prepared* for college really currently precluded from going? (Well maybe if they have kids…) Beyond that they merely say it’s hard to know what to do and whatever we do may involve large costs. Thus in the end the ‘self-selection’ thesis comes across as “they’re simply screwed”. And the idea that there are no LT economic consequences of childbirth seems at odds with the ‘disillusionment’ thesis, since being disillusioned implies that you gave up something of potential value.

  30. 30 30 Jace

    @ Nobody.Really & Harold

    Nobody.Really as usual presented a very interesting vantage. But, as I reread his comment, one word in particular resonated with me: prospect-less.

    What does “prospect-less” mean functionally? To remain within proper context, we’re speaking of the prospects (specifically, the lack thereof) available to teenage women between the ages of thirteen and nineteen. Granted, the viable options increase considerably the higher you go within age spectrum, but then again, you could also use that same logic to circumvent the syllogism propelling the notion that having a child now is more advantageous to a teenage mother.

    Typically, prospects for a young woman would consist of a career, school, a combination of both, and perhaps finding a husband. The aforementioned caveats I have stated withstanding, at any point is the option to have a child at such an unprepared age in American society truly logical?

    I don’t know. . . .

    Completing a degree at a university typically requires four years of full-time study, but that needn’t be the only avenue. There are community colleges, and vocational schools that offer shorter (less expensive opportunities), and flexible schedules to accommodate a multitude of demographics seeking to attain training or retraining. On the other hand. . . a child could cost as much (if not more) than tuition to an institute of higher learning, and also doesn’t equip you with the intangible set of skills or the tangible piece of paper that qualifies you for a number of other opportunities which could potentially recompense you for your investment in the long-term. Not to mention it takes eighteen years to raise a child.

    Does becoming a mother award you with intangible skills, or learned abilities? I suppose so, but I don’t believe there are many extant (nor lucrative) jobs that demand a requisite knowledge of teenage motherhood, and beyond that, I don’t know of any clause within affirmative action that confers preference to mothers or fathers over single men and women.

    Gaining employment offers similar training, and experience with the added benefit of not having to pay tuition. Not only that, but if you are preparing to become a mother, aren’t you icing your cake by having an existing income, and supplementing it with the societal amenities mentioned by Nobody.Really? Moreover, couldn’t this be compounded further by the acquisition of a husband, making this trek possibly the most appealing?

    All-in-all, I don’t understand the choice of teenage motherhood to be the most logical decision, which by default compels me to postulate that perhaps it is some sort of noumena that compels these young women to believe accruing another burden is tantamount to being successful?

    I’m unsure. My thinking-cap may be broken.

    But, according to a blog entry by Paul Krugman sometime ago, American teenage pregnancy rates are actually declining, which would stand as evidence to substantiate the claim that either society is learning, or teenage motherhood is in some other way less desirable than in the past.

    Link: http://www.childtrendsdatabank.org/sites/default/files/14_fig02.jpg

  31. 31 31 Seth

    Harold: Thanks for the follow-up. That’s interesting.

    I’ve done enough modeling to know that you should be careful not to discount those small-effect variables. Their effects may be much larger, but it just so happens that another variable in the model is a better proxy for it and drowns it out — making it look less important. Income inequality, for example, could be a better proxy for state transfers.

  32. 32 32 Harold

    For the section “Teen childbearing – cause or effect?” we are dependent on the reviewers of the paper. There is a review of several papers, and those of us who do not know the literature must hope that the reviewers have assesed that this represents a reasonable interpretation of the available published data. There are some phrases that may couse concern – for example, the effects of teen birth are described as “modest” for one study, but what does this actually mean? Even these effects are “likely” to be overstated. It is possible that others may interpret these data as significant, and unlikely to be overstated – we must hope the reviewers have been diligent, and the views expressed are an accurate reflection of the literature.

    If we do assume that this section accurately reflects the literature, then it does seem that a large proportion of the poor economic outcome is due to factors other than teen birth.

    I would like to see a slightly different breakdown than “teen” – since birth at 19 may give a very different outcome than birth at 17 or 15. An OECD report from 1998 shows that the USA and the UK have not only relatively high rates of teen births, but also relatively high rates of 15-17 year old births as well. In fact, the USA 15-17 birth rate was higher than rates for 15-19 in all coutries other than UK. ( http://www.unicef-irc.org/publications/pdf/repcard3e.pdf )

  33. 33 33 Harold

    I meant referees rather than reviewers above.

  34. 34 34 Claire

    @nobody.really: your aboriginal example is werd because it implies a preoccupation with monetary cost that, in my experience, is usually not very salient amongst traditional hunter-gatherers. I did a version of this test (used a bunch of pictures from trashy magazines to get language data) on a day when none of us felt like going through more paradigms. There was a lot of discussion about how unhappy the supermodels looked.

  1. 1 JimSwift.net » Blog Archive » Wednesday Links

Leave a Reply