SOCIO-ECONOMIC OUTCOMES OF TEEN PREGNANCY AND PARENTHOOD: A REVIEW OF THE LITERATURE.
Subject: Teenage pregnancy (Social aspects)
Teenagers (Sexual behavior)
Young women (Psychological aspects)
Subsistence economy (Causes of)
Author: Bissell, Mary
Pub Date: 09/22/2000
Publication: Name: The Canadian Journal of Human Sexuality Publisher: SIECCAN, The Sex Information and Education Council of Canada Audience: Academic Format: Magazine/Journal Subject: Psychology and mental health Copyright: COPYRIGHT 2000 SIECCAN, The Sex Information and Education Council of Canada ISSN: 1188-4517
Issue: Date: Fall, 2000 Source Volume: 9 Source Issue: 3
Topic: Canadian Subject Form: Teenage sexual behaviour
Geographic: Geographic Scope: Canada Geographic Code: 1CANA Canada
Accession Number: 77556075
Full Text: ABSTRACT: This paper reviews literature examining the socio-economic consequences of teen pregnancy and childbearing as well as the birth intentions of teenage women who have become mothers. The literature indicates that women who become teen mothers, as compared to women who delay childbearing past the teen years, are more likely to be socio-economically disadvantaged. However, research increasingly suggests that, depending on the circumstances of the women in question, socio-economic disadvantage is correlated with but not necessarily a consequence of early childbearing. In addition, the literature suggests that a range of factors, including cultural norms and individual needs, can impact on the childrearing decisions of teenage women. It is concluded that policies and programs aimed at reducing teen pregnancy rates and eliminating the negative consequences experienced by teen mothers and their children are unlikely to be fully effective unless they realistically address the socio-economic inequities faced by many young women in Canada.

Key words: Teen pregnancy Child bearing Economic disadvantage Education Poverty Birth intentions

INTRODUCTION

The literature and political/public discourse addressing teen pregnancy and childbearing has often been rooted in the assumption that having a child is the key factor in leading teen mothers into a life of socio-economic disadvantage. However, as this paper will attempt to show, many of the fundamental issues related to teen pregnancy and childbearing do not lend themselves to simplistic notions of cause and effect. The main focus here will be to review the growing body of literature that examines the socio-economic consequences of teen pregnancy and childbearing as well as literature exploring the birth intentions of teenage women who have become mothers. Throughout the literature the question of causality vs. correlation is often addressed, as researchers attempt to disentangle what conditions predispose a woman to have a teen birth from those conditions that result from teen pregnancy. Recent research has emphasized the importance of considering racial, ethnic, cultural and socio-economic variables in the analysis of data. What emerges from such an analysis is a more complex understanding of the phenomenon of teenage pregnancy. Those women who give birth as teenagers are shown to share many similarities prior to becoming pregnant, in terms of family background, racial/ethnic profile and education. For those teenage women who intentionally conceive, research shows that they also tend to share a common pre-pregnancy socio-economic profile. The purpose of this review of the literature is to contribute to a better understanding of how socio-economic factors are related to the incidence of teen pregnancy and how the negative consequences of teen pregnancy are linked to socio-economic disadvantage.

METHODOLOGY

This paper reviews findings from a number of Canadian and U.S. studies that have investigated the socio-economic outcomes associated with teen pregnancy and parenting and the birth intentions of teens who become pregnant. Major academic databases in the social sciences such as Expanded Academic Index, PAIS International, PsychInfo and Sociofile were searched to identify papers that went beyond a simplistic interpretation of the "problem" of teen pregnancy to explore the more complex socioeconomic conditions associated with the phenomena. While the majority of papers were research oriented, some theoretical discussions and literature reviews were also selected for inclusion since they provided important contextual information.

The majority of research surveyed measured the impact that teen pregnancy and parenting had on such variables as education, employment and economic status, and welfare use. A smaller number of studies also looked at marital patterns of teen mothers and the developmental progress of children born to adolescent women. Some research on the nature of birth intentions that explores the relationship between pre-pregnancy attitudes and situations to subsequent pregnancy outcomes, has also been included. Both short and long-term studies were included, however, as is the case with most research in the area of adolescent sexuality, short-term studies outnumbered long-term ones. The sources of national data on education, employment and welfare use included census information from both the United States and Canada. Research from the United States also drew on national data from such sources as the National Longitudinal Survey of Labor Market Experience of Youth (NLSY), Panel Study Income Dynamics (PSID), and Aid to Families with Dependent Children (AFDC). Some studies that included smaller, more regionalized sample sizes were included if they contributed unique perspectives to the understanding of teen pregnancy and parenting (i.e., Nova Scotia long-term study of single mothers and Alan Guttmacher study on birth intentions).

CANADIAN LONG-TERM OUTCOMES STUDIES

This section reviews two large-scale studies (Grinstaff, 1988; Nova Scotia Department of Community Services, 1991). A long-term study of Nova Scotia mothers and their children conducted by the Nova Scotia Department of Community Services (NSDCS) presented findings on the socioeconomic outcomes related to teen pregnancy and single parenthood vs. married parenthood. The study followed a sample of women of varying ages who had their first child during the last six months of 1978 and the first two months of 1979. Data was analyzed according to age at birth and marital status. There were initially 403 cases of unmarried women and 416 cases of married women included in the study. The sample was broken down into two age groups, those 19 years or younger and 20 years and older. After the birth of children, interviews were conducted at intervals of 7 weeks, 9 months, 18 months and 10 years. At the 10-year interview, 84% of the married cases from the initial study participated, as did 62% of the unmarried cases. A follow-up interview for 15 years was originally planned but has not yet been published. Interviews sought to measure the participants' self-perceived satisfaction with 12 dimensions of life including, health, finances, family relations, employment, friendships, education, self-esteem, marital situation, religion, recreational activities and transportation. The 10-year study also included results from psychological and educational testing done on the children (NSDCS, 1991).

The Nova Scotia study presents a picture of the lives of women and children according to age at first birth and marital status. Unmarried mothers of all ages reported most dissatisfaction with educational achievement. The data reveals that those women who were in school when they became pregnant were more likely to return to school after giving birth than were those who dropped out prior to their pregnancy. The older unmarried mothers were less likely to complete their education than were the younger unmarried mothers, since they were also more likely to have dropped out of school before becoming pregnant. Young unmarried mothers (87.3%) were more likely to marry over the 10-year period than were their older unmarried counterparts (69.4%). The unmarried mothers in general waited longer to have their second child than did the married mothers of all ages. Unmarried mothers also had slightly less children (2.01) than did the married mothers (2.31). Over the 10 year period in question, only 3% of all mothers received social assistance for 9 or more years; however, 32.8% of all mothers lived in poverty, a rate twice the provincial rate. Age and marital status related poverty statistics were broken down as follows: 54% older unmarried, 52% younger unmarried, 29% young married, and 16% older married. Older, married mothers who had higher levels of completed education were more likely to have professional positions with greater remuneration. Lower levels of education and work-related experience limited the employment opportunities of younger mothers (NSDCS, 1991).

For almost all the areas of life satisfaction examined in the study, older married mothers tended to fare the best. However, the diversity of responses within each group points to the conclusion that these are not homogeneous populations. While it is evident that the younger unmarried mothers began parenting with more socio-economic handicaps, they made the most gains over 10 years but still lagged behind the older married group in terms of self-reported satisfaction and measurable gains. With reference to the situation of the young unmarried mothers the authors note

The NSDCS (1991) study also assessed the children of mothers in the sample and found few differences between groups. The children of younger mothers scored lower than those of older mothers in the area of verbal comprehension, but their general scores as a group were within the normal average range of scores for all children. There was no significant difference in the ways in which mothers of varying age and marital status dealt with child behaviour. In conclusion, the authors note:

The recommendations of the NSDCS focus on improving the educational and employment options of all women. The importance of affordable daycare and housing, improved social assistance programmes to address high poverty rates, access to education and training for all mothers regardless of age and marital status, and expansion of job opportunities for women is stressed. Educational programmes geared towards children at-risk of dropping out, particularly those whose mothers have 11 years or less of schooling, is recommended, since dropping out of school is often a precursor of early pregnancy. The need for comprehensive sexuality and relationship education to be offered at all levels of schools is stressed, with particular emphasis being placed on problem solving. Teacher training to support the delivery of such programmes is also recommended (NSDCS, 1991).

Grindstaff (1988) conducted a study of the long-term economic outcome for a sample of Canadian women who married and/or had children before the age of 20. He used information from the 1981 census to examine the economic situation of women who were 30 at the time of census taking. Using a sample of approximately 200,000 women, he subdivided the group into subgroups based on age at first birth, age at marriage and single or childless status. Findings indicated that the earlier the age at marriage and age at first birth, even for those women over 20, the lower the educational attainment of women. By the age of 30, 58% of women who married and had children as teens were in the paid labour force, with 68% of these women having full-time employment. For those women who married and had children after 25, the corresponding percentages are 54% employed and 65% full-time. These statistics reflect the long-term labour market participation of teen mothers rather than the short-term picture that generally is one of increased unemployment and limited employment. Grindstaff(1988) speculates that the increased labour market participation of teenage mothers is based on a greater economic need for income on the part of these women.

Both marriage and childbearing at an early age tend to limit a woman's job opportunities to mostly nonprofessional occupations. However, Grindstaff's (1988) study suggested that it was age at marriage and childbearing at any age that was more closely associated with employment prospects, independent of age at first birth. The patterns of less successful employment outcomes for women who had children as teenagers were similar to those for women who married as teenagers but delayed childbearing until ages 20-24. Income limitation was associated more with having children, independent of age at first birth and age at marriage. Grindstaff (1988) concludes that having a child as a teen is not significantly different from having a child between the ages of 20-24, both events impact negatively on the income level and labour market participation of the mothers. However, he recommends that the negative economic consequences of teen births and marriages should be "documented by evidence and internalized" by young people to encourage the postponement of childbearing and marriage. While the author notes that early marriage and early childbearing result in negative economic consequences, the fact that childbearing at older ages also takes an economic toll on women highlights the need "from a policy perspective ... to diminish the economic impact of having a first child" (Grindstaff, 1988, p. 56).

LITERATURE REVIEWS

This section summarizes two recently published literature reviews on teen pregnancy outcomes studies in the U.S. (Hoffman, 1998; Steven-Simon & Lowry, 1995). Steven-Simon and Lowry (1995) reviewed the literature on the socio-economic consequences of teenage pregnancy in an effort to determine if teen childbearing and poverty are causally related. Some of the studies cited in these reviews point to various socio-economic factors that predispose some women to poor educational and economic outcomes regardless of whether they experienced a teen birth (e.g., Furstenberg, Brooks-Gunn, & Morgan, 1987; Miller, 1992). The findings of these studies suggest that "the negative effect of early childbearing on the educational and vocational achievements of young Americans could be a reflection of the self-selection of academic underachievers to the teenage childbearing group" (Steven-Simon & Lowry, 1995, p. 912). The authors also note the importance of acknowledging the fact that, for many economically disadvantaged teens, pregnancy is viewed as a positive life choice that is integral to the assumption of an adult role in society. Policies and programmes that support the realistic attainment of educational and vocational goals must make it possible for such teens to fare better if they postpone pregnancy, which is not the case for most low-income, minority teens today. The authors conclude that in order to reduce teen pregnancy among this population, early parenthood must be perceived as being the "least rather than the most attractive career option" (Steven-Simon & Lowry, 1995, p. 917).

Hoffman's (1998) recent review of the literature on teenage pregnancy focuses on research that addresses long-held assumptions that teenage childbearing leads to poverty, welfare dependency, and compromised child development, unemployment and poor educational outcomes. The question of causality vs. correlation is one that many researchers are currently struggling with. As Hoffman (1998) points out, teenage mothers do not represent a random sample of the population since the majority has family backgrounds characterized by poverty and low socioeconomic status. If the age at first birth was delayed for such teenagers would this necessarily improve their life situation and that of their children? Studies in the mid-1970s and early 1980s focussed on controlling factors that might contribute to the likelihood of a teen birth, for example, parental income and education levels. However, such research failed to consider the impact of unmeasured factors such as parental involvement and attitudes. Since many unmeasured factors would be difficult if not impossible to quantify, more recent research has shifted the emphasis to comparison studies between teen mothers and "natural" companion groups (such as sisters) (Geronimus & Korenman, 1992), singleton vs. twin births (Grogger & Bonars, 1993) and teen mothers vs. teens who miscarried (Hotz, McElroy, & Sanders, 1997). The rationale behind such studies was that women in the two groups share many similarities in terms of family background, racial/cultural influences, education and environment; any discernible differences between the groups could thus be attributed to the fact of having had a teen birth or not. These individual studies will be discussed in more detail below, but, suffice to say that their results lend support to the argument that the negative consequences of teen pregnancy have been overstated in past research.

While the findings of these studies sheds new light on the teen pregnancy "problem", Hoffman (1998) is critical of some methodological points. For Geronimus and Korenman's (1992) sister study, he notes that the sample size was quite small and representative of women who were teenagers in the late 1960s to mid 1980s. Hoffman (1998) notes that many researchers have only these outdated samples to work with, and since the economic and social environment of the United States has changed dramatically during the last few decades, it is questionable to generalize from the experiences of teenagers 20 to 30 years ago to those of today. The twins study by Grogger and Bronars (1993) was based on the assumption that the socio-economic costs of having two children are twice that of having one, an assumption that Hoffman (1998) questions. The results of Hotz, McElroy, and Sander's (1996) study of women who miscarried as teenagers vs. those who had a teen birth are compromised by small sample size. Hoffman (1998) also raises a concern regarding the characteristics of women in the control group. Some of the teens that miscarried went on to have a teen birth and thus the control group did not consist exclusively of women who had avoided a teen birth.

STUDIES ON THE SOCIO-ECONOMIC CONSEQUENCES OF TEEN PREGNANCY

The foregoing literature reviews indicate that there have been a number of American studies that examine the socio-economic consequences of teenage pregnancy. Many of these studies focussed on low income, urban populations consisting of primarily African American and/or Hispanic women. While the findings of such studies are not directly applicable to the Canadian situation, the general trends and observations that are discussed provide a framework within which Canadian specific research can be situated.

The sisters studies of Geronimus and Korenman (1992) sought to address the deficiencies inherent in previous research that focused on the cross-sectional associations between teenage childbearing and measures of socio-economic success. The authors point out that such studies are "biased by failure to account for heterogeneity in the population of mothers" (Geronimus & Korenman, 1992, p. 1188). In their study of sisters, they employed the method of "within family estimation" in order to control for family background heterogeneity. However, they also point out that even within families a certain amount of heterogeneity exists. Although sisters have similar socio-economic backgrounds, personal characteristics that affect behaviour and life choices cannot always be measured and accounted for (i.e., academic achievement, attitude towards education). They also cite research (Burton, 1990; Ladner, 1971) that indicates some family units are more accepting of teenage childbearing than others.

Geronimus & Korenman (1992) analyzed data from the following surveys: 1) National Longitudinal Survey of Labor Market Experience, Young Women's Cohort (NLSYW); 2) Panel Study Income Dynamics (PSID); 3) National Longitudinal Survey of Labor Market Experience of Youth (NLSY). The NLSYW is a long-term survey begun in 1968 which follows that economic progress of a sample of women who were between the ages of 14 and 24 in that year. Data for the sisters study was taken from 1982. The PSID study focuses on the economic status of the family unit and began data collection in 1968. Geronimus and Korenman (1992) analyzed data from 1985 for women who were aged 28 to 38 at that time. The NLSY, begun in 1979, is another ongoing survey, which presents statistics on men, and women who were aged 14 to 21 in the initial year of data collection. For the purposes of their study, Geronimus and Korenman (1992) used data collected in 1988. In analyzing the data, the researchers considered such socio-economic indicators as employment and marital status, education level, poverty and welfare status, and income-to-needs ratio. Findings from the 3 data sets "yielded an unsettling wide range of estimates," and thus the authors caution against drawing firm conclusions from any of the results. For the NLSYW and NLSY samples, the negative consequences of a teen birth decline, but remain significant, when the socio-economic characteristics of the family are considered. The PSID data reveals little difference in outcomes once family background is considered. When the sisters variable was considered for all three data sets, the effects declined again for the two NLS samples, but not for the PSID sample. From their findings, Geronimus and Korenman (1992) observe that "all of the effects of a teen birth on current income is explained by differences in current marital status and high school completion in the NLSYW and PSID data sets" (p. 1204). Results from both the NLSYW and PSID samples point to higher family income as being directly associated with the likelihood of high school completion. And, interestingly, the NLSYW sample fails to draw a connection between teen birth and high school completion.

In critiquing their own methodology Geronimus and Korenman (1992) note a few concerns that could have influenced the significance of their findings. They note that the "PSID sisters subsample may be more homogeneous in terms of family backgrounds compared with the NLS sisters subsamples" (p. 1206). For example, the non-teen mothers in the PSID sisters subsample were more likely to come from single parent families and to be African American than those women in the other two samples. This tendency toward homogeneity lead the authors to conclude, "the PSID sisters sample may not be well suited for determining whether family background heterogeneity biases traditional cross-sectional estimates of the long-term socio-economic effects of teen childbearing" (p. 1206). The data presented are also limited by the fact that they represents findings from one year only and thus do not reflect the long-term effects of a teen birth. Finally, the study measures a select set of socio-economic indicators. The authors emphasize the importance of considering "alternative measures of well-being," such as infant health, child behaviour and development. Previous work by Geronimus, Korenman, and Hillemeier (1991) that looked at the impact of teen motherhood on the socio-cognitive development of children is cited as an example of such research. This study focused on two sample groups, one comprised of the children of teen mothers and the other consisting of the cousins of these children who were born to non-teen mothers. In examining a comparison of the performance of the children in such areas as verbal activity and behavioural development, and in looking at the safety and stimulation of the home environment, the research revealed little difference between the two sample groups.

Grogger and Bronars' (1993) study of twins vs. singleton births was designed to measure the socio-economic impact of an unplanned teenage pregnancy, as represented by the birth of the second twin. Using the 1970 and 1980 U.S. census, the authors selected two sample groups to study, one of women who had a twin first birth before the age of 20 and one representing women who had a singleton first birth before the age of 20. The following variables were measured for the 2 groups: family income, marital status, level of education, labour force participation, poverty level, welfare status. The authors found that results from the 1970 data samples were not significantly different from zero for any of the measured variables, thus the study essentially focuses on the 1980 sample. The results indicated that the negative effects of teen childbearing were largely dependent on race. White women experienced some degree of negative impact in the areas of income, labour market participation, education and welfare receipt but the most significant negative effects of an unplanned teen birth were experienced by black women (i.e., decreased likelihood of marriage, increased long-term rates of poverty and welfare receipt, short-term labour market participation).

Hotz, McElroy, and Sanders (1997) examined the socio-economic costs of teen pregnancy for women who had their first child before 18 years, and used as a control group a sample of teenage women who miscarried. Since data indicates that miscarriages cause an average delay of two to three years for the next pregnancy, the study addressed the question of whether a delay of 2.5 years would significantly alter the socio-economic consequences of having a birth at an earlier age. The authors used NLSY data for women who were between the ages of 13 and 19 during the years 1970 to 1985. Information from interviews conducted in 1992 formed the basis of this study. The authors caution against using the findings summarized below to make long-term predictions since they do not report on the income levels of women after the age of 34. They note the possibility that the income levels of the two samples might show different trends over a longer period of time. The results must also be viewed in light of the possible under-reporting of the incidence of pregnancy, abortion and miscarriage.

The study compared educational achievement, income level, receipt of public assistance and employment patterns of women who experienced their first pregnancy at different ages. The authors presented general trends that emerge in a comparison of women who gave birth before the age of 18 and those who delayed their childbearing until after 18. The data indicated that 61% of the teen mother group completed high school as compared to 90% of women in the delayed childbearing group. By the age of 30, teen mothers earned 58% of what those who delayed childbearing earned and received more than four times the amount of public assistance benefits than did non-teen mothers. These results, however, are unadjusted for socio-economic variables such as parental income and education level, family status, or receipt of welfare as a child. Examination of these variables showed striking differences between the teen mother group and the delayed parenting group. Teen mothers were more likely to have come from low income, single parent families and to have lived in families that received welfare benefits. The assumption that those who experience a teen birth and those who delay birth until a later age are on the same life path in terms of education, employment and social stability appears to be contradicted by these findings.

Results using comparison data from the miscarriage control group revealed that those women who had teen births actually earned more money than did those women who delayed childbearing. If women delayed childbearing by 2.5 years they would spend 20% less time as a single mother during their early to late teens and up until the age of 30..However, delaying childbearing did not affect the overall level of spousal support that a woman could expect to have by age 30. Teen mothers received more total spousal support than did the women who had a later birth. Results showed that by their mid 20s to early 30s, teen mothers worked more hours and had higher earnings than the control group. The authors speculate that since teen mothers tend to work at jobs that require on-the-job training and experience rather than formal education, "for such women, concentrating their childbearing at early ages may prove to be more compatible with their likely labour market career options than would postponing motherhood and/or spacing births over their childbearing years" (Hotz, McElroy, & Sanders, 1996, pp. 78-79). The effect of delaying birth for 2.5 years on the receipt of welfare benefits was almost negligible; there was only a 4% decrease of approximately $88 a year if childbearing was delayed. The authors compared the cost of social assistance benefits for teen mothers with the net taxes paid by them over the course of their young adult working lives and found that a delay of childbearing of 2.5 years actually increased overall government expenditures since, by the age of 30, those who delayed childbearing had earned less than those who gave birth as teenagers. The findings challenge the assumption in the U.S. that "teenage childbearing is one of the nation's most serious social problems at least when one measures its severity in terms of cost to taxpayers" (Hotz, McElroy, & Sanders, 1999, p. 85).

A study by Klepinger, Lundberg, and Plotnick (1999) examines the effect that a teen birth has on human capital accumulation (i.e., education, work experience) and the rate of return (i.e., income) for this accumulation. The findings suggested that teenage childbearing, by limiting the educational attainment of women, led to a reduction in wages. This study also found that black women faced less of a drop in wages (13%) by virtue of having a teen birth than did white teen mothers (23%). The income levels of white women were decreased due to limited education and work experience. However, white women did not experience a drop in the rates of return for educational and employment related investments. Black women had a higher rate of return for schooling but not for early work experience. The authors speculate that lower earnings of teen mothers, and the need for childcare contribute to long-term dependence on government aid. Therefore, they conclude that a reduction in teen pregnancy rates will lead to improved economic prospects for young women.

The research of Corcoran and Kunz (1997) focused on the question of whether a causal relationship exists between teen childbearing and adult poverty and welfare dependency. Drawing on the work of Geronimus and Korenman (1992), and Hoffman, Foster, and Furstenberg (1993) the authors take a within-family approach to more precisely estimate the effects of teen parenthood, independent of background (Corcoran & Kunz, 1997). Using data from the Panel Study of Income Dynamics (PSID), the authors selected a sample of 573 African American women who were between the ages of 4 and 14 in 1967 and who had had a teen birth by 1986. From this sample, a subsample of 60 women with sisters or half-sisters who had not had a teen birth was selected. A second subsample was selected, consisting of 31 pairs of sisters, one who had received Aid for Families with Dependent Children (AFDC) and the other who hadn't. The data from the general sample of 573 women was first analyzed to determine the rates and patterns of welfare receipt, poverty levels, and income-to-needs ratios for unmarried teen mothers, married teen mothers and non-teen mothers. Findings indicated that unmarried teen mothers had a 43% lower income-to-needs ratio, were 2.8 times more likely to be poor and 1.4 times more likely to receive welfare benefits than were non-teen mothers or married teen mothers. But when the sisters subsamples were considered, these differences among unmarried teen mothers, married teen mothers, and non-teen mothers were reduced dramatically (i.e., 23% lower income-to-needs ratio, 1.25 times more likely to be poor, 1.02 times more likely to receive welfare). Based on these results, the authors conclude that "much of the apparent effect of teenage out-of-wedlock births on adult women's income, poverty and welfare use are due to unobserved family and neighbourhood background shared by sisters" (Corcoran & Kunz, 1997, p. 285).

Gleason, Rangarajan, and Schochet (1998) investigated the pattern of welfare receipt among inner city teen mothers. Data on 2,325 teen mothers who began receiving AFDC benefits for the first time between the years of 1987 and mid 1991 was analyzed and compared to data on more general groups of AFDC recipients presented in other studies. The authors sought to identify trends in the length of time of welfare receipt, and reasons for entry, exit or re-entry to welfare rolls. The members of the teen mother sample group had an average age of 18, with 14% of the women being 16 years or younger. Most (75%) of the group was African American and 17% were Hispanic. The data revealed that 47% of the women left welfare for a job or because of an increase in employment earnings and 12% left after getting married. Those women most likely to leave welfare had a high school education, previous work experience and advanced math skills. The authors note that those women with limited English proficiency were more likely to find work and leave welfare and less likely to re-enter the welfare system. Those women without a high school education tended to have longer periods of welfare receipt and were more likely to re-enter welfare. When compared to data on broader groups of welfare recipients, teen mothers had longer rates of AFDC receipt, lower levels of exit rates and higher re-entry rates. The authors speculate that these trends associated with teen mothers are related to their lower levels of education and work experience, as well as to their predominantly single status. The women from this sample were primarily from racial and ethnic minorities and lived in economically depressed urban areas where job opportunities for teen mothers were limited. The authors conclude that those factors that affect the patterns of welfare use for all welfare recipients (i.e., education, work experience/skills, marital status) also affect trends associated with teen mothers (Gleason, Rangarajan, & Schochet, 1998).

While there have not been detailed Canadian studies of patterns of welfare use by teenage mothers, Canadian welfare statistics can offer some insight into welfare trends based on age, fertility and marital status (National Council on Welfare, 1998). According to 1997 statistics, teenage single parents accounted for 3% of all single parents on welfare. Few (4%) heads of families of welfare cases were under the age of 20, and 12% were between the ages of 20 and 25. Approximately 50% of single parent families on welfare had only one child and 31% had two children. Most (80%) single mothers on welfare were listed as divorced, widowed or separated. When examining reasons for welfare assistance, lack of work accounted of more than half of all welfare cases. Disability was given as the second most common reason for welfare receipt. The prevalence of single parenthood as a reason was more difficult to ascertain since not all provinces accepted it as a reason in itself. However, for those provinces that recorded single parenthood as a reason for welfare receipt, the use of this reason was highest for those people in their late 20s and early 30s. This profile of Canadian welfare recipients calls into question some of the commonly held assumptions regarding teenage mothers and welfare dependency. Teenage mothers and unmarried mothers with great numbers of children did not represent a large proportion of those on welfare. The most common reasons for assistance, lack of work and disability, do not reflect circumstances that occur as a result of teen childbearing, but rather reflect current economic conditions and the vulnerability of the disabled in Canadian society.

A study of the short-term economic consequences of teen pregnancy by Byrne, Myers, and King (1991) examined data relating to education, employment and income for women who became pregnant as teenagers. The authors included in their sample those women who went on to give birth as well as those who chose to have an abortion. Their research was designed to test the assumption that it is teenage childbearing itself that leads to a reduction in education and income rather than pregnancy, and those teens who have an abortion and never pregnant teens should therefore experience similar educational and labour market outcomes. The authors analyzed data collected for the National Longitudinal Surveys of Youth (NLSY) in interviews conducted in the years 1979, 1984 and 1985. Information regarding teenage fertility patterns was taken from the 1984 survey when the interview format was altered to allow for written, confidential responses. Following this change in format the responses show a marked increase in the reporting of abortions, leading the authors to assume that fertility information collected during this year would be more accurate than previous years' data. By the age of 25, those women who had a teen birth had an average of 11.3 years of schooling, women in the never-pregnant sample had an average of 13.3 years and women who had an abortion as a teen had an average of 12.7 years of schooling. High-school dropout rates for the 3 sample groups were as follows: 36.3% teen birth sample, 5.7% never-pregnant sample and 11.2% teen abortion sample. Labour market participation rates were 58% for teen mothers, 84% for never-pregnant women and 80% for women who had had an abortion. The authors make limited reference to racial, ethnic and socio-economic variables in the presentation of their data. They do note that, when compared to white women in all the sample groups, women of Black or Hispanic origin receive lower wages. The data also shows that married women, women with young children, women in school and Black and Hispanic women are all more likely to have less paid hours of employment.

Byrne, Myers, and King (1991) comment that "young women who undergo abortions complete less schooling, on average, than a similarly aged never-pregnant group, leading to lower wages and less attachment to the labour market" (p. 1819). Women in the never pregnant sample fared better in terms of educational achievement and labour market participation than did women in the teen birth and teen abortion samples. However, the authors emphasize that their research examines short-term consequences of adolescent pregnancy and it will be necessary to follow the sample groups for longer periods of time to determine if these negative short-term outcomes alter or disappear over time.

Butler (1992) described the changing economic consequences of teenage childbearing for the years 1968 to 1986 in the context of social and economic trends that could affect both incidence and outcomes associated with teen pregnancy. For the decades in question, the author notes that teenage mothers were "more likely to be unmarried when they give birth, which increases their chances of being a single mother later in life and therefore increases their chances of poverty" (Butler, 1992, p. 12). And while the average number of children per teen mother had decreased over the years, and the likelihood of high school completion increased, the economic climate of the late 1980s was such that even these seemingly favourable trends did not lead to improved economic conditions for teen mothers and their children. The author notes that since 1973, the average earnings of 25- to 34-year-old men declined, especially for those with a high school education or less, and for African American men. While the average wages of similarly aged young women did not change considerably between the years 1979 and 1984, "women's earnings continued to lag behind those of men and were often too low to support a family above the poverty line" (Butler, 1992, p. 13). The author speculates that the decrease in wages and employment opportunities for young men with limited education made them "less effective providers" for teen mothers and their children, therefore reducing the economic incentive for women to marry. While growing numbers of teen mothers found themselves with insufficient financial resources with which to support themselves and their children, levels of welfare benefits in the United States declined and eligibility requirements became, in many instances, more restrictive.

In order to assess the economic well being of teenage mothers in the years from 1968 to 1986, Butler (1992) analyzed data from the Panel Study of Income Dynamics (PSID). From this data set, a subset of 2,232 Black and non-Hispanic white women who were 25 years old between 1968 and 1987 was drawn. Those women who gave birth at age 14 or younger were excluded. Findings indicated that, in general, "the economic well-being of women who began childbearing as teenagers declined more steeply than did the economic well-being of women who delayed childbearing until at least their 20s" (Butler, 1992, p. 19). Once racial variables were introduced, the results showed that black women fared significantly worse than white women did in terms of economic well being. Those black women who gave birth at 15-17, as compared to white teen mothers of the same age group, experienced much more of a decline in economic well being. Only those white women who gave birth at 18-19 experienced a significant worsening of their economic situation. These results support much of the current research that has shown that "although black women who delay childbearing are better off than black women who have a child as teenagers, the differences are not nearly as great for black women as they are for white women ..." (Butler, 1992, p. 24). The author notes that despite the increasing negative economic consequences of teen childbearing, many women continue to have children as teenagers regardless of the economic costs. Prevention programmes and programmes to decrease the negative impact of teenage childbearing must recognize and address this reality.

BIRTH INTENTIONS

The literature concerned with issues relating to teenage pregnancy and prevention often begins with the assumption that most teen births are unintended or untimely. Geronimus (1997) observes that "family planning advocates blur the distinction between teenagers at risk of pregnancy who avoid childbearing when given adequate information and technology and those who would bear children even provided the same" (p. 407). Statistics published by the Alan Guttmacher Institute (Henshaw, 1998) suggests that 22% of teen pregnancies and 44% of births to 15- to 19-year-old women were intended. The factors that affect the decisions of adolescent women to conceive are as complex as those that influence the decisions of adult women. Some research in the area of birth intentions focuses on socio-economic conditions that might predispose teen women to experience either intended or unintended pregnancy, while other studies attempt to understand the more abstract culture-based and psychosocial variables at work. In order to develop effective prevention programmes as well as to ensure that economic and social supports are in place for pregnant and parenting teens, the reality of intended teen pregnancies and births cannot be ignored. The following section review U.S. and Canadian literature on this topic.

Trent and Crowder (1997) used data from the U.S. National Longitudinal Surveys of Labor Market Experiences of Youth to explore the nature of pre-stated birth intentions and subsequent fertility patterns of teen women. They sought to determine if birth intentions differed according to socio-economic background and if so, could they account for differences in fertility patterns between low and middle to high-income groups. Questions were posed to survey participants in 1979, thus the study was limited in that it examined data related to a once stated intention that may or may not be representative of a continuing held belief. The findings did not support a strong link between pre-pregnancy birth intentions and actual fertility outcomes. Before taking into account stated birth intentions, the data revealed that women from low income families were two times more likely to have had a teenage or non-marital birth over a five year period. When the variable of birth intention was considered, this finding was not significantly affected. The authors conclude that

An Alan Guttmacher Institute (1999) study of unmarried, pregnant 15- to 18-year-old women in four California counties explored the factors that influenced the decision to give birth. The sample size was small (187) and focussed on low-income minority women of African American and Hispanic heritage. The results of the study must thus be viewed as an exploration of birth intentions of these particular racial and ethnic groups, rather than being representative of a larger population. While many of the pregnancies in the study group were unintended (43%), approximately 32% of all respondents had intended to conceive while 25% reported an ambivalent attitude toward their pregnancy. However, when the statistics are viewed in light of racial/ethnic and age related groups, the percentages were altered significantly. In the 15-16 age group, 22% of participants had intended to get pregnant, compared to 36% of the 17- to 18-year-old group. The percentages of intended pregnancies for women of all age groups, according to racial/ethnic background were as follows: 14% of black teens, 34% of U.S. born Hispanic teens and 46% of foreign born Hispanic women. The findings of this study indicate that a sizeable percentage of teen pregnancies are intended and hence probably not preventable simply by the provision of contraceptive information.

Studies from the 1980s (e.g., Davis, 1989) suggested a link between teen pregnancy and childbearing and the lack of close and fulfilling personal relationships. The authors of the Alan Guttmacher study (1999) cited the desire for stability and unconditional love as being motivations for an intended teen pregnancy and noted that "many of the young women who intended pregnancy viewed childbearing positively, as a way of setting a course for their lives--a way to gain maturity and stability" (p. 24). However, in much of the prevention literature such motivations are implicitly acknowledged as being rather misguided attempts to bring love and a sense of purpose into the lives of socially and economically disadvantaged young women. The Guttmacher study characterizes those women who intended to conceive as having "no life plans," and foreign born Hispanic women are said to be not yet "acculturated" to the norms of U.S. society given their overrepresentation in the intended pregnancy group of the sample. While the study recommends the need to be sensitive to cultural norms when prevention programmes are developed, the implication is that norms that are accepting of early childbearing present obstacles for prevention programmes and are therefore not to be supported or encouraged.

In contrast, some studies have questioned this tendency of western societies to negate cultural norms which are more accepting of early childbearing. Participants at the Aboriginal Roundtable on Sexual and Reproductive Health (Aboriginal Nurses Association of Canada, 1999) noted that early parenthood was common in traditional Aboriginal societies. They identified "the breakdown in traditional support structures and values, rather than teenage pregnancy per se ... [as being] responsible for the health and social problems teenage parents and their families often face" (1999; paragraph 15). Feminist philosopher Sara Ruddick (1993) has observed, "When the dominant culture opposes self-regard to responsibility for others and identifies choice with the independent self, young women may appear to be passive even as they make what are to them self-respecting and self-defining choices" (p. 131). She cautions that "the diagnosis of untimeliness with regard to mothering also risks ethnocentricity and cultural arrogance" (p. 128). Conversely, the recognition of cultural norms can also be used to further marginalize young women from specific racial or ethno-cultural groups. Cultural stereotyping can lead to the assumption that middle-class white teens will postpone childbearing until they have achieved educational and work-related goals while black or Hispanic teens will have children early and thus suffer the consequences of poor education and limited career choices. The acknowledgement of cultural norms that are more accepting of early childbearing should not serve to inadvertently perpetuate the social and economic inequities experienced by many teen mothers.

Merrick (1995) explores the concept of adolescent childbearing as career choice for some young women. In her discussion she offers the definition of career as being a "selection of a role through which one implements one's identity as an adult and which represents one's life work" (Merrick, 1995, p. 288). Viewed in light of this definition, the decision to have a child can be seen as a deliberate career choice for some teenage women, the majority of whom are from lower socioeconomic, racial and ethnic minority groups. Merrick (1995) frames her discussion in the context of child development and human development models. She raises questions regarding commonly accepted models of adolescent development (i.e., Erickson's model) which emphasize the adolescent's need to develop a sense of identity and self through the cultivation of independence and a separation from family. She suggests that such models are male oriented and "from such a perspective, women who primarily value relationships instead of individual achievement are immature personalities with dependency needs" (Merrick, 1995, p. 289). Girls who choose motherhood as a way of seeking personal fulfilment and defining self, if viewed from a White middle-class perspective are seen as deviating from societal norms. However, "deviance may be seen as an invention of a social group that uses its own standards as ideals by which others are to be judged" (Merrick, 1995, p. 289). Merrick (1995) makes an argument for the recognition of adolescent childbearing as a "career choice" for some women. Research in the area of teen pregnancy and parenting should address the socioeconomic variables that lead some women to make such choices. It is recommended that policy initiatives in this area should focus on providing services that would offer a realistic alternative to early childbearing. For many socio-economically disadvantaged women the negative economic and educational consequences of teen pregnancy are not perceived as incentives to avoid pregnancy. As Chilman (1980) aptly states, a

CONCLUSION

The literature demonstrates that there are definite negative consequences associated with teen pregnancy. When compared to women who delay childbearing past the teen years, those women who become teen mothers are less likely to complete high-school, more likely to work at low-income jobs and experience longer periods of unemployment, more likely to receive welfare benefits during the years following birth and more likely to experience single parenthood and higher levels of poverty. However, when racial and economic variables are factored into the data analysis, the negative consequences of teen pregnancy are shown to be largely dependent on race, ethnic background and income level rather than on maternal age at birth. The fact that a large percentage of teen mothers come from socio-economically disadvantaged backgrounds means that they suffer many of the negative consequences of teen pregnancy by virtue of their family status and pre-pregnancy life situation. The research also indicates that for women from disadvantaged backgrounds, the avoidance of a teen birth does not always improve future education or employment opportunities and thus there is little perceived economic incentive to delay childbearing. Policies and programs aimed at reducing teen pregnancy rates and eliminating the negative consequences experienced by teen mothers and their children are unlikely to function effectively unless they realistically address the socio-economic inequities faced by many young Canadian women. A uniquely Canadian approach to teenage pregnancy must also take into account the ethno-cultural profile of Canadian teenagers who become pregnant.

ACKNOWLEDGEMENT: The research for this report was part of a study conducted by SIECCAN with the collaborative support of the Planned Parenthood Federation of Canada and the Sexual and Reproduction Health Promotion Unit, Health Canada.

References

Aboriginal Nurses Association of Canada (1999). Aboriginal Roundtable on Sexual and Reproductive Health (On-line). Available: www.hc-sc.gc.ca/msb/ pptsp/aborig

Alan Guttmacher Institute (1999). Teenagers' Pregnancy Intentions and Decisions: A Study of Young Women in California Choosing to Give Birth (On-line). Available: www.agi-usa.org

Burton, L. (1990). Teenage childbearing as an alternative life-course strategy in multigenerational black families. Human Nature, 1, 123-143.

Butler, A. (1992). The changing economic consequences of teenage childbearing. Social Service Review, March 1992, 1-31.

Byrne, D., Myers, S., & King, R. (1991) Short term labour market consequences of teenage pregnancy. Applied Economics, 23, 1819-1828.

Chilman, C. (1980). Social and psychological research concerning adolescent childbearing: 1970-1980. Journal of Marriage and the Family, 42, 793-805.

Corcoran, J. (1998). Consequences of adolescent pregnancy/parenting: A review of the literature. Social Work in Health Care, 27, 49-67.

Corcoran, M., & Kunz, J. (1997). Do unmarried births among African-American teens lead to adult poverty? Social Service Review, 71, 274-287.

Davis, R. (1989). Teenage pregnancy: A theoretical analysis of a social problem. Adolescence, 93, 19-28.

Frost, J.J., & Oslak, S. (1999). Teenagers' Pregnancy Intentions and Decisions: A Study of Young Women in California Choosing to Give Birth. Occasional Report No. 2. New York: The Alan Guttmacher Institute.

Furstenberg, F., Brooks-Gunn, J., & Morgan, S. (1987). Adolescent mothers and their children in later life. Family Planning Perspectives, 19, 142-151.

Geronimus, A.T. (1997). Teenage childbearing and personal responsibility: An alternative view. Political Science Quarterly, 112, 405-430.

Geronimus, A.T., & Korenman, S. (1992). The socioeconomic consequences of teen childbearing reconsidered. Quarterly Journal of Economics, 107, 1187-1214.

Geronimus, A.T., Korenman, S., & Hillemeier (1991). Does young maternal age adversely affect child development? Office of Population Research Working Paper No. 91-10, October.

Gleason, P., Rangarajan, A., & Schochet, P. (1998). The dynamics of receipt of Aid to Families with Dependent Children among teenage parents in inner cities. Journal of Human Resources, 33, 988-1003.

Grindstaff, C. (1988). Adolescent marriage and childbearing: The long-term economic outcome, Canada in the 1980s. Adolescence, 23, 45-58.

Grogger, J., & Bronars, S. (1993). The socioeconomic consequences of teenage childbearing: Findings from a natural experiment. Family Planning Perspectives, 25, 156-161.

Henshaw, S.K. (1998). Unintended pregnancy in the United States. Family Planning Perspectives, 30, 24-29, 46.

Hoffman, S.D. (1998). Teenage childbearing is not so bad after all ... or is it? A review of the new literature. Family Planning Perspectives, 30, 236-243.

Hoffman, E., Foster, M., & Furstenberg, F. (1993). Reevaluating the costs of teenage childbearing. Demography, 30, 1-13.

Hotz, J., McElroy, S., & Sanders, S. (1997). The impacts of teenage childbearing on the mothers and the consequences of those impacts for government. In Rebecca Maynard (Ed)., Kids Having Kids: Economic Costs and Social Consequences of Teen Pregnancy (pp. 55-94). Washington, D.C.: The Urban Institute Press.

Klepinger, D., Lundberg, S., & Plotnick, R. (1999). How does adolescent fertility affect the human capital and wages of young women? Journal of Human Resources, 34, 421.

Ladner, J. (1971). Tomorrow's Tomorrow: The Black Woman. New York, NY: Doubleday.

Merrick, E. (1995). Adolescent childbearing as career "choice": Perspective from an ecological context. Journal of Counseling and Development, 73, 288-303.

Miller, B. (1992). Adolescent parenthood, economic issues, and social policies. Journal of Family Economic Issues, 13, 467-475.

National Council of Welfare. Profiles of Welfare: Myths and Realities (On-line). Available: www.ncwcnbes. net/htmdocument/reportprowelfare/repprowelfare.htm

Nova Scotia Department of Community Services (1991). Mothers and Children: One Decade Later. Halifax, NS: NS Department of Community Services.

Ruddick, S. (1993). Procreative choice for adolescent women. In A. Lawson & D. Rhode (Eds.), The Politics of Pregnancy: Adolescent Sexuality and Public Policy. New Haven, CT: Yale University Press.

Stevens-Simon, C. & Lowy, R. (1995). Teenage childbearing: An adaptive strategy for the socioeconomically disadvantaged or a strategy for adapting to socioeconomic disadvantage? Archives of Pediatric and Adolescent Medicine, 149, 912-915.

Trent, K. & Crowder, K. (1997). Adolescent birth intentions, social disadvantage, and behavioral outcomes, Journal of Marriage and the Family, 59, 523-535.

Correspondence concerning this paper should be addressed to Mary Bissell, Ph.D., SIECCAN, 850 Coxwell Ave., Toronto M4C 5R1. Tel: (416) 466-5304; fax: (416) 778-0785; email:sieccan@web.net.
Their disadvantaged position vis-a-vis the older married women is not,
   however, merely a function of age. The older unmarried mothers also
   experienced more difficult circumstances than their married counterparts
   and in some instances appear more disadvantaged than the young unmarried
   mothers (NSDCS, 1991, p. 18).


The generally positive outcome of the children's development puts to rest a
   great deal of concern and speculation about how well children of unmarried
   teenage mothers develop compared to children born into two parent families.
   The study shows that mothers share similar experiences raising their
   children (e.g. constraints on their time and access to childcare) while
   differences occur not because one group is more devoted but because of the
   circumstances in which the families live (NSDCS, 1991, pp. 6-7).


persistent racial, economic, and family structure differentials in early
   and non-marital childbearing are more a function of social disadvantage
   because such circumstances indirectly place youth at greater risk through
   earlier ages at first intercourse, less reliance on contraceptives, and
   greater acceptance of unintended pregnancy (Trent & Crowder, 1997, p. 531).


delay in childbearing surely would not guarantee educational, financial,
   occupational improvements in their lives. These improvements depend far
   more on a series of social and economic reforms than they do on deferring
   childbearing until the person is no longer a teenager (p. 801).


Mary Bissell
Sex Information and Education Council of Canada
Toronto, Ontario
Gale Copyright: Copyright 2000 Gale, Cengage Learning. All rights reserved.