Effects of Marriage on Family Economic Well-Being. Empirical Papers


Impacts of Marital Status and Parental Presence on the Material Hardship of Families with Children.” The first of the three empirical studies, this paper is based on data from the 1997 and 1999 rounds of the National Survey of America’s Families (NSAF). The primary measures of family economic well-being are poverty status (above or below the poverty line) and material hardship. While poverty reflects a family’s income, hardship measures whether a family’s material needs are met. Examples of material hardship include whether phone or utilities have been cut off, whether people are doubling up in rooms, whether family members have enough to eat, and whether the family has been evicted. Lerman examines the relationship between union formation and economic well-being among families overall, and whether this relationship is different for disadvantaged families (as defined by education levels and poverty status).

The paper begins with a rich set of descriptive results showing the degree to which union formation outcomes vary by racial subgroup, and the extent to which poverty rates vary by type of family union. These results are particularly informative because they break union formation into nine different categories reflecting marital status, living arrangements and biological relationship.(4) The results generally indicate higher poverty rates among single and cohabiting-parent families as compared to married-parent families. In 1998, for example, poverty rates ranged from 6 to 9 percent for families headed by married parents, from 7 to 21 percent for cohabiting parents and from 25 to 39 percent for single parents. This relationship between union formation status and poverty also holds true when using multivariate analysis to control for other observable family characteristics, however the differential is smaller.

The paper also examines the link between union formation status and material hardship(5). Married-parent families were still better off than either cohabiting- or single-parent families, but the differential was narrower than for poverty rates, with hardship rates ranging from 15 to 22 percent for married parents, 29 to 30 percent for cohabiting, and 32 to 39 percent for single parents. This pattern generally holds true for both the simple descriptive and the multivariate results.

Part of the likely reason why married parent families are economically better off is their greater potential for generating income; however, Lerman finds that even when holding income constant, married-parent families still have a lower likelihood of material hardship than either single- or cohabiting-parent families.(6) This relationship varies somewhat depending on the degree of hardship being examined. For example, among families experiencing either difficulty with rent or with meals, married parents are better off than cohabiting parents regardless of whether there are one or two biological parents in the household. However, among families experiencing both hardships, Lerman finds that well-being ranked highest among families having two biological married parents, then two cohabiting biological parents, followed by married with one biological parent, single parents, and cohabiting with one biological parent. These findings also hold constant a variety of other family characteristics such as mother’s education.

As Lerman’s analysis points out, human capital factors such as education can have a direct impact on material hardship, and it is important to control for these factors when estimating the benefits of marriage. It is also important to see if the benefits to marriage are actually different for people with more or less human capital (i.e. parents who are more or less disadvantaged). Lerman’s findings indicate that among parents with post-secondary education, married parents are better off than either cohabiting or single parents regardless of the number of biological parents in the household. However among parents with a high school degree or less, well-being ranked highest among families having two biological married parents, then two cohabiting biological parents, followed by married with one biological parent, cohabiting with one biological parent, and finally single parents. Thus, even among parents with lower education, families headed by two married biological parents are better off than either cohabiting or single parents, but the relationship between marriage and family well-being is generally not as straight-forward as with higher educated parents, depending also on biological status.

While such detailed findings could raise a variety of interesting hypotheses, Lerman’s basic conclusion is that “being in a married, two-parent household does indeed serve as a protective device against material hardship, even among those with the same income-to-needs ratio, immigration status, race, education, age of children and adults.”

How Do Marriage, Cohabitation and Single Parenthood Affect the Material Hardships of Families with Children.” In this second empirical study, Lerman builds on his prior work by replicating and expanding the analysis of material hardship, including the role of help from family and friends. The study also looks at the role of income available in prior years and household stability. Findings are based on data from the 1996 panel of the Survey of Income and Program Participation, covering the period from April 1996 through March 2000. As in the prior paper, Lerman examines the relationship between union formation and economic well-being among families overall, and whether this relationship is different for disadvantaged families (as defined by education levels and poverty status).

This paper begins by describing the incidence of married, cohabiting and single-parent families by race, and the household characteristics for each type of family union. The findings then turn to measures of family well-being by union formation status where well-being is captured by poverty rates, welfare ratios and material hardship. With respect to both poverty and welfare ratios, married-parent families were better off than cohabiting and single parent-families, although the impacts were slightly higher with respect to poverty rates. This relationship held true for White, Black and Hispanic families, with the returns to marriage greatest among Black families, followed by White and Hispanic families. The share of families overall having at least one material hardship ranged from 14 to 20 percent for single and cohabiting parents and 6 percent for married parents.

Lerman then turns to measuring the returns to marriage among disadvantaged families. He finds the relationship between marriage and reduced material hardship generally holds true for families below 150% of the poverty line as well. Among parents below 150 percent of poverty, the overall hardship index was nearly 70 percent greater among cohabiting and single parents compared to married parents. When looking exclusively at parents with less than a high school diploma, Lerman found similar results; the overall hardship index was between 54 and 100 percent greater for cohabiting and single parent families compared to married parent families. It should be noted, however, that for both parents with lower incomes and parents with lower education, the strength of the relationship between union status and hardship depended on the particular hardship being examined.

There are a number of reasons why material hardship may be less among married parent families even across families with similar incomes. This outcome could reflect greater union stability or commitment to long-term investments in the family, or it could also reflect greater access to resources outside the immediate family.

Lerman examines the last explanation by looking at the percent of married, cohabiting and single couples who report having access to help from family, friends or the community. In fact, reports of access to each of these types of assistance were somewhat higher among married families as compared to cohabiting or single parent families. This generally held true both for families overall, as well as families below 150 percent of poverty. Among families below poverty, the gap was particularly wide between married and cohabiting couples. For example, the percent of married parents reporting help from family was 70 percent while among cohabiting couples only 55 percent reported such assistance.

Multivariate results provide further evidence that at least some of the advantage experienced by married-parent households is attributable to greater access to help from outside the immediate family, as well as greater stability. Lerman uses a series of equations to add these factors to the model one by one. He shows that variability of welfare ratios over time, changes in household composition over time and help from friends, family and community each have a significant effect on family material hardship, and each reduces the effect that had been attributable to marital status. Nonetheless, marital status remains a significant factor in reducing hardship. These relationships held true for families below 1.5 times the poverty line as well. Among families whose household heads had less than a high school diploma the same was also true, although the differential between married and cohabiting families was weaker, as were the effects of change in household status and help from the community.

In reflecting across a broad spectrum of findings in this paper, Lerman concludes that “Married parents experience far less hardship not only than single parents with no other adults but also than cohabiting couples and single parents with another adult. The evidence for economic gains to marriage is quite strong, even for less-educated and low-income individuals.”

Married and Unmarried Parenthood and Economic Well-Being: A Dynamic Analysis of a Recent Cohort” The last paper of this series adds a longitudinal perspective to the analysis of marriage and economic well-being, and uses several techniques to control for the selectivity into marriage. Findings are based on the National Longitudinal Survey of Youth, 1979 cohort. Compared to the earlier two papers, this study provides a lengthier analysis of the factors affecting family status, particularly the likelihood of spending time as a single parent, prior to exploring economic well-being. As with his other studies, Lerman then examines the relationship between union formation and economic well-being, and whether this relationship is different for disadvantaged families (defined primarily by mothers’ aptitude scores or the experience of a pregnancy outside of marriage). This paper focuses primarily on the welfare ratio (or income-to-needs ratio) rather than economic hardship.

The basic descriptive findings in the beginning of the paper show that the likelihood of ever being a single mother is highest among Black women, followed by Hispanic women and White women (77, 52 and 32 percent respectively). Conversely, the opposite is true for the likelihood of ever being married (65, 86 and 94 percent respectively). When comparing family status by aptitude level (or AFQT score), the results are not surprising, showing mothers with higher aptitudes more likely to have ever married and less likely to have ever been a single mother.

Further exploring the longitudinal path that mothers follow over time, Lerman shows the relationship between the initial family status after a child is born and subsequent family status. Those mothers who started as married spent 84 percent of their subsequent years married and only a small fraction of time as either a cohabiting or single parent. Those starting as cohabiting spent roughly a third of their subsequent years in each of the other states. Among mothers starting off single, the breakdown was roughly one quarter of subsequent years as married, one tenth as cohabiting and over 60 percent single. When looking by race we see similar patterns but with slight variations. For example, compared to mothers overall, Black mothers were somewhat more likely to move from married or cohabiting to single and less likely to move out of single status.

Lerman also presents a variety of multivariate results on the relationship of initial family status to subsequent family status, controlling for numerous related factors. Mothers who marry early (any time prior to one year following the child’s birth.) are notably less likely to spend time later as single parents compared to mothers who cohabit. Further analyses show marriages that occur after pregnancy but prior to birth (shotgun marriages) are only marginally less beneficial compared to conventional marriages (marriages occurring prior to pregnancy) in terms of reducing time spent as a single parent.

Lerman then examines the relationship between family formation and economic well-being where well-being is measured by the welfare ratio.(7) The average welfare ratio following first birth was nearly twice as high for married mothers compared to cohabiting mothers, more than twice as high compared to single mothers living with other adults, and 60 percent higher compared to single mothers living alone. These benefits extended to Black mothers and mothers with low AFQT scores, although the comparative advantage of married mothers was smaller. Among women who experienced their first pregnancy outside of marriage, those who married prior to the birth (shotgun marriage) were about 38 percent better off than women who did not marry.

These benefits attributable to marriage held true when using multivariate analysis to control for other family characteristics that might affect well-being. Many of these results indicated not only a higher standard of living for married families, but also more stable economic outcomes. A number of factors measured prior to the pregnancy were also significant, including poverty status, work experience, education and aptitude scores. Further analyses show that while the benefits of marriage were significant for White, Black and Hispanic families, the effects were smaller among Black and Hispanic families compared to White families.

Although these estimates control for a wide variety of observable family characteristics that might affect well-being, there may be other, unobserved characteristics that affect both the decision to marry and the family’s economic well-being. In an effort to compare families whose characteristics are as similar as possible, Lerman also analyzes the returns to marriage specifically among women experiencing a pregnancy outside of marriage. Those who married prior to the birth of the child (a shot-gun marriage) were significantly better off in terms of reduced chronic poverty than those who remained unmarried at the time of birth, even after controlling for a variety of other factors that might affect well-being. Moreover, the positive effect of marriage on reducing poverty was greater among Black mothers and mothers with low aptitude scores compared to mothers overall.

Lerman fine-tunes this comparison by categorizing mothers based on their marriage prospects, or likelihood of marrying.(8) He finds that for women experiencing a pregnancy outside of marriage, those who married sometime within six months following the birth were notably better off compared to those who remained unmarried, even among those with the lowest marriage prospects. Further analyses reveal, however, that a marriage occurring after a pregnancy is not as beneficial as a marriage occurring prior to pregnancy.

As a final test to see whether the economic returns attributable to marriage are truly an impact of marital status and not due to other factors that may affect selection into marriage, Lerman estimates a series of random effects models and fixed effects models. Both are designed to net out the effects of unobserved family characteristics when estimating the impact of marriage on economic well-being. Both sets of results were similar and revealed significant impacts of marriage on well-being (measured by welfare ratios). While cohabiting parents were better off than single parents, the benefit they experienced was roughly half that of married parents. Results were positive and strongly significant for mothers overall, Blacks, Whites, Hispanics, mothers with low and mid aptitude scores, and mothers with a premarital pregnancy.

Lerman concludes that “Even among the mothers with the least qualifications and highest risks of poverty, marriage effects are consistently large and statistically significant.... The robust nature of the estimates lends some credence to the view that marriage itself generates economic benefits for mothers and children.”