From Generation to Generation? The Role of an Intergenerational Cycle of Teen Childbearing in Declining US Teen Birth Rates Isaac Maddow-Zimet, Laura Duberstein Lindberg, and Kathryn Kost Once described as an “epidemic,” teen pregnancy is now one of the nation’s greatest public health success stories. From a high of 117 per 1,000 teens aged 15-19 in 1990, the teen pregnancy rate plummeted to an all-time low of 57 by 2010. Between 2008 and 2010 alone, the rate dropped by an astounding 15%.1 The country’s teen birthrate fell at similar rates, declining 44% from a peak in 1991 to 2010. More recent data are available for teen birthrates than for teen pregnancy rates, and those data show that the decline in teen births has continued: It dropped 10% from 2012 to 2013, to 27 per 1,000—the lowest rate ever reported for the United States.2 These changes have been so substantial in magnitude, and so pervasive across race/ethnicity (declines occurring among non-Hispanic white, non-Hispanic black and Hispanic teens), age (including both younger and older teens) and location (occurring in all 50 states), that efforts seeking explanations have offered only limited insights. Explanations have focused on responses to the state-level policies3, the economic recession45, the influence of media6, and changes in the proximate determinants of teen pregnancy—sexual activity and contraceptive use—which may have altered the risk of pregnancy.7,8 To expand the research agenda on the forces driving the decline in teen childbearing, we will look at the implications of an important but under-investigated collateral impact of the longterm decline in teen childbearing: it is likely that fewer teens are entering their reproductive years having been raised by a teen mom. Daughters of teenage mothers becoming teenage mothers themselves resulting in a self-perpetuating generational “cycle” has been of concern to researchers for decades.9 Rendall (2003) referred to this as the “inter-generational teenage

childbearing correlation.” 10 While he investigated the role of this mother-daughter repetition in explaining differential teen childbearing trends between England and France, to our knowledge this has not been examined in the recent United States context. We propose to investigate the impact of such cycles for recent trends in US teen childbearing. To assess the long-term and ongoing consequences of past declines, we will examine links between the proportion of teens born to teen mothers, and the proportion of those teens who themselves engage in early childbearing. Given that the peak in the teen childbearing rate in the US occurred in 1991, this means that daughters of these mothers would have reached age 15 in 2006, and would continue moving through the teen childbearing ages in 2007-2010 when declines in the adolescent birth rate were substantial. Our approach is as follows. Attention to date has focused on declines in the rate of teenage childbearing. But the demographic implications of this rate play out in the number and proportion of births that are to teenage mothers. First, using national vital birth records, we will describe the number and proportions of births that are to teenage mothers over the period 1980 to 2013. This will enable us to identify inflection points in the relative size of the cohort of children born to teen mothers; tracking this cohort forward, we can estimate when we would expect to see this diminished population of at-risk teens start to have effects on the overall teen birth rate. Second, we will pool data from multiple waves of the National Survey of Family Growth (NSFG), a nationally representative in-home survey conducted by the U.S. National Center for Health Statistics. Relevant to this study are the data collected from women aged 15–44 in 1995, 2002, 2006-2010, and 2011-2013; sample size ranged from 7,643 to 12,279 in these waves Rather than comparing results from each of the four data sets, we plan to combine them and use

a birth cohort approach to identify women age 15-19 in each year from 1990-2013. In the cases where data sets overlapped, respondents from multiple data sets will be included in the birth cohort. Each NSFG round asks female respondents to report both 1) their age at first birth, and 2) their mother’s age at her first birth. (Grandmothers age at birth is not reported on the birth certificate, making it necessary to use survey data). From this we can calculate for each year from 1990-2013 the proportion of teenage girls ages 15-19 with a mother who was a teen mother, and the proportion whose mother was an adult at first birth. This provides the relative size of the population of daughters of teen mothers. In order to smooth trend lines in our analyses, we plan to report five-year centered moving averages. To understand their risk of childbearing, we will estimate the likelihood of becoming a teen mother between both groups of respondents and track changes in this over time as well. This will allow us to explore the strength and stability of any “inter-generational teenage childbearing correlation.” We will test doing these analyses separately by race-ethnicity, evaluating sample size limitations; race/ethnicity is trait that is immutable over the life course, allowing us to use measures collected at the time of the interview to examine patterns in retrospective experiences. Third, using the measures from the NSFG, we will use standard demographic decomposition to assess the impact of four component parts of the change in the teen birth rate: 1. Size of the population of daughters of teen mothers 2. Risk of childbearing among daughters of teen mothers 3. Size of the population of daughters of adult mothers 4. Risk of childbearing among daughters of adult mothers

Finally, we will project future trends in teen childbearing given both the size and risk of intergenerational childbearing. This paper will be the first to address the question of whether and to what extent a “cycle of teen childbearing” in the US plays a substantial demographic role in trends, and if so, whether future trends can be predicted by observed patterns. This analysis has the potential to address how demographic forces are shaping current fertility.

References 1. Kost K and Henshaw S, U.S. Teenage Pregnancies, Births and Abortions, 2010: National and State Trends by Age, Race and Ethnicity, New York: Guttmacher Institute, 2014, , accessed Aug. 20, 2014. 2. Hamilton BE et al., Births: preliminary data for 2013, National Vital Statistics Reports, 2014, Vol. 63, No. 2, , accessed Aug. 20, 2014. 3. Kearney MS and Levine PB, Explaining recent trends in the U.S. teen birth rate, NBER Working Paper No. 17964, March 2012. 4. Kearney MS and Levine PB, Explaining recent trends in the U.S. teen birth rate, NBER Working Paper No. 17964, March 2012. 5. Boonstra, H, What is behind the declines in teen pregnancy rates? Guttmacher Policy Review, Summer 2014, Volume 17, Number 3 6

Kearney, Melissa S., and Phillip B. Levine (2014a). “Media Influences on Social Outcomes: The Impact of MTV’s 16 and Pregnant on Teen Childbearing,” NBER Working Paper No. 19795 January 2014.

7. Santelli JS, Orr M, Lindberg LD, and Diaz D (2009), Changing Behavioral Risk for Pregnancy among High School Students in the United States, 1991-2007,” Journal of Adolescent Health; 45:25-32

8. Santelli JS, Lindberg LD, Finer LB, and Singh S, Explaining recent declines in adolescent pregnancy in the United States: The contribution of abstinence and improved contraceptive use?” American Journal of Public Health, 2007, 97(1):150-156. 9. Kahn J R and Anderson K E (1992). Intergenerational patterns of teenage fertility. Demography 29(1), pp. 39–57. 10. Rendall MS, How important are intergenerational cycles of teenage motherhood in England and Wales? A comparison with France. Popul Trends, 2003 Spring, (111):27-37.