Religiosity and Regional Teen Birth Rate Differences

Note on the Kearney-Levine Hopelessness Hypothesis

Robert Cherry Chun Wang Broeklundian Professor Assistant Professor Brooklyn College Brooklyn College

Using data from 1981-2008, Melissa Kearney and Phillip Levine argue that an important determine of the persistence of high teen birth rates is the lasting impact of persistent income inequality. They (Kearney and Levine, 142) claim, “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 …” Their results have been reported in a number of influential newspapers and blogs.1 While this claim may very well had been true for the years prior to the 1990s welfare legislation and the subsequent Clinton-era economic boom (Luker

1996), we find it much less credible for the first decade of the twenty-first century.

Kearney and Levine focus on the high birth rate (and high inequality) that typifies the Deep

South states like Mississippi. They neglect to point out that there may be alternative explanations for regional differences. In our own work, we focus on the 33 states that included

92.5 percent of the national population and 92.7 percent of the teen birth rates in 2001. We separated out the seven Deep South states with the highest levels of religiosity: Louisiana,

Arkansas, Tennessee, Mississippi, Alabama, South Carolina and Georgia. Figure 1 takes the

1 Paul Krugman. 2012. “The Economics of Marginalization and Hopelessness,” NY Times (Mar 12) http://krugman.blogs.nytimes.com/2012/05/12/the-economics-of-marginalization-and-hopelessness/; Mitako Rich. 2012. “Income Inequality and Teen Pregnancy,” NY Times (Mar 3) http://economix.blogs.nytimes.com/2012/04/03/income-inequality-and-teenage-pregnancy/; David Rosen. 2012. “Inequality and Teenage Pregnancy,” Counterpunch (May 11-13). http://www.counterpunch.org/2012/05/11/the- sexual-politics-of-poverty/; and Ezra Klein. 2012. “America’s Teen Birth Rate at a New Low …,” Wonkblog (Apr 10) http://www.washingtonpost.com/blogs/ezra-klein/post/americas-teen-birth-rate-is-at-a-new-low----and-still-10- times-higher-than-switzerlands/2012/04/10/gIQAbnNY8S_blog.html

1 simple average of the birth rates of these Deep South states and that of the other 26 states in our study. It shows a striking regional disparity.

Table 1 will help us understand some of the reasons for this disparity. When teen birth rates were near their peaks in 1990, there was little regional difference in race-specific non-Hispanic teen birth rates. As a result, virtually the entire reason that the Deep South states had higher teen birth rates was the larger black non-Hispanic share of the teen population there compared to other states. Since that time, black and white non-Hispanic teen birth rates in both groups of states have declined substantially. And in all years, race-specific non-Hispanic teen birth rates for the 33 states in our study match closely the national averages.

There is one striking change: teen birth rates for both races, but especially for white non-

Hispanics, declined by a smaller percentage in the Deep South states than in the other states in our sample. For 2005-2007, the white non-Hispanic teen birth rate in these Deep South states was 75 percent higher than in the other states in our sample. By contrast, the black non-Hispanic

2 teen rate in these Deep South states was only 20 percent higher than in the other states in our

sample.

Table 1: Non-Hispanic Teen Birth Rates by Year, Race, and Group*

Average of 2005 and 2007 1990 Group White Black White Black Deep South (7 states) 41.9 71.1 59.1 112.0 Other States (26 states) 23.9 59.1 48.4 113.8 Total Sample (33 states) 26.2 62.2 51.6 113.3 National Average 26.6 63.1 50.8 112.8 (DeepSouth)/(OtherStates) 1.75 1.20 1.22 0.98

* - Birth rates here are rates for all teens living in each group of states not the simple average of the birth rates of states within each group. To obtain, the overall averages, the birth rates for states within each group are weighted by their teen female populations. Birth rates are births per 1,000 teen women. Sources: For 1990 birth rates, see Spitz et al., 1993. For 2007 birth rates, see T.J. Matthews et al., 2010. For 2005 birth rates, see Guttmacher Institute. 2010.

In 2006, the black non-Hispanic share of the teen population, 15 to 19 years old, in the seven

Deep South states was 34.0 percent as against 13.6 percent in the other 26 states in our study.

Since the black non-Hispanic rate is more than double the white rate, this disparity should

continue to cause the Deep South states to have higher teen birth rates than the other states.

Figure 2: Black-White Teen Birth Rate Ratio, by Age, 2000 and 2009

3.5 3 2.5 2 2000 1.5 2009 1 0.5 0 15-19 15-17 18-19

3 This larger non-Hispanic black share in the Deep South states, however, was offset by its much smaller Hispanic share: 3.8 percent compared to 18.3 percent in the other states. As a result, the

2005-2007 birth rate in the Deep South states would have been virtually unchanged if the

Hispanic and black non-Hispanic shares there were the same as in the other states. By contrast, virtually the entire regional difference in teen birth rates would have been eliminated if the white and black teen birth rates in the Deep South states were the same as in the other 26 states.

The dramatic rise in the regional difference in white non Hispanic teen birth rates brings into question the hopelessness hypothesis. Following Regnerus (2007), we suggest that this growing gap might be best explained by religiosity. These seven states scored highest on a number of measures of religiosity (Pew Forum and Religion and Public Life, 2008). In addition, using state-level data from the 2007 Pew Foundation survey on religious life, Strayhorn and

Strayhorn (2009) found a high correlation between religiosity and teen birth rates after controlling for income and abortion rates.

A second reason why we question the hopelessness hypothesis is our preliminary findings on the impact of economic conditions on teen birth rates. If the Kearney-Levine hypothesis was valid, we might expect improvements in the employment prospects of young women to reduce their birth rates. Using a variety of specification, however, we were unable to find any statistically significant impact of changes in the state’s female teen employment rate on its teen birth rates, for 15-19 year olds or 15-17 and 18-19 year olds separately. Instead, we found a strongly significant positive impact of the employment rate of young men, 20 to 24 years old, on all three measures of the teen employment rate (Cherry and Wang, 2012).2

2 One of the reasons our study is limited to the 33 largest states is that these gender specific employment rates are generally not available for smaller states; and gender-specific teen employment rates were not tabulated prior to 2000.

4 Rather than hopelessness, the female teen employment rate has no systematic impact on teen birth rates reflects a weak substitution effect. Twenty years ago, researchers claimed that this weak substitution effect reflected the lack of upward economic mobility of disadvantaged women even if they delayed motherhood (Geronimus and Korenman, 1992). They found that educational attainment differed little between those teens that chose motherhood and their sisters that did not.3

Disadvantaged women have much more upward mobility than twenty years ago. We believe that the weak substitution effect now reflects the substantial work and educational supports available to these women. If a nineteen or twenty year old working single woman becomes pregnant, she now gains a much more generous EITC and much more accessible subsidized childcare. Just as important, increasingly universities are making accommodations that enable these women to combine work, school, and motherhood. Thus, these women do not see a substantial cost to their long-term future by having children when they are young and unmarried.

Indeed, these women face a substantial marriage penalty since most of the supports are lost if they marry a working partner (Carasso and Steuerle, 2005). In particular, the EITC is based on family income with only modest adjustments for marriage. As a result, the vast majority of working single mothers with incomes between $10,000 and $25,000 and one child, face federal marriage penalties of more than $2,000 if they marry partners which raise their combined income into the $35,000 to $50,000 range (Cherry, 2012). There have been a number of proposals to reduce this marriage penalty, include the “New Mother Tax Relief” proposal made by one of the authors.4 Rather than focusing on eliminating hopelessness, policies to eliminate the onerous marriage penalty faced by low-income working single mothers should have priority.

3 For a critical comment on this work, see Hoffman (1998). 4 For marriage penalty reduction proposals, see Carasso et al.(2008) and Cherry with Lerman (2011).

5 Bibliography

Carasso, A. H. Holzer, E. Maag, and C.E. Steuerle. 2008. “The Next Stage for Social Policy: Encouraging Work and Family Formation among Low-Income Men,” Discussion Paper 28 (Urban Institute, Oct ).

Carasso, A. and C.E. Steuerle. 2005. “The Hefty Penalty on Marriage Facing Many Households with Children.” The Future of Children 15(2): 151-75.

Cherry, R. 2012. “New Mothers Tax Relief: Reducing Marriage and Middle-Class Parent Penalties.” Working Paper, Brooklyn College.

Cherry, R. and C. Wang, “Teen Birth Rates and Labor Market Conditions, 2001-2009.” Working paper, Brooklyn College.

Cherry, R. with R. Lerman. 2011. Moving Working Families Forward: Third Way Policies that Work (New York University Press).

Geronimus, A., & Korenman, S. 1992. “The Socioeconomic Consequences of Teen Childbearing Reconsidered.” Quarterly Journal of Economics 107: 1187-1214.

Guttmacher Institute. 2010. U.S. Teen Pregnancies, Births and Abortions. Guttmacher Institute (Jan) http://www.guttmacher.org/pubs/USTPtrends.pdf

Hoffman, S. 1998. “Teenage Childbearing Is Not So Bad After All … Or Is It?” Family Planning Perspectives 30: 236-239.

Kearney, M.S. and P.B. Levine. 2012. “Why is the teen Birth Rate in the United States So High and Why Does It Matter.” Journal of Economic Perspectives 26(Spring): 141-166.

Matthews, T. J. et al. (2010). “State Disparities in Teen Birth Rates.” NCHS Data Brief #46 (Oct) http://www.cdc.gov/nchs/data/databriefs/db46.pdf

Pew Forum on Religious and Public Life. 2008. The U.S. Religious Landscape Survey, 2007 (Pew Foundation).

Regnerus, M.D. 2007. Forbidden Fruit: Sex and Religion in the Lives of American Teenagers (Oxford).

Spitz, A. M. et al. (1993). “Surveillance for Pregnancy and Birth Rates among Teenagers.” Morbidity and Mortality Weekly Report 42 (SS-6) (Dec 17): 1-27.

Strayhorn, J.M. and J.C. Strayhorn. 2009. “Religiosity and Teen Birth Rates in the United States.” Reproductive Health 6 (Sept): 1-7.