Economic Inquiry 41 (2003): 163–183, T. S. Dee finds that unilateral divorce laws, which were found to lead to increases in divorce (see L. Friedberg, “Did Unilateral Divorce Raise Divorce Rates? Evidence from Panel Data,” American Economic Review 88 : 608–627), have a negligible effect on the incidence of husbands murdering wives, but unilateral divorce coupled with laws that favored husbands in the division of marital property led to a 21 percent increase of lethal spousal violence against husbands. Contrary to Dee’s results, B. Stevenson and J. Wolfers (see “Bargaining in the Shadow of the Law: Divorce Laws and Family Distress” Quarterly Journal of Economics 121, no. 1 : 267–288) find that no-fault divorce led to a large decline in spousal homicide against wives and no change in spousal homicide against husbands. Dee replicates the results of a 2000 version of Stevenson and Wolfers’ work and finds that their results are sensitive to a variety of assumptions and specification (see Dee, pp. 177–178). It is unclear whether any changes in their analysis that were present in Stevenson and Wolfers’ published version in 2006 render Dee’s con- cerns in 2003 moot, as Stevenson and Wolfers do not directly address these concerns in 2006. Therefore, there does not seem to be definitive evidence that unilateral divorce laws decrease spousal homicide against wives. Nevertheless, Wilcox et al., Why Marriage Matters, 31, summarize evidence that married women are subject to less violence inside and outside the home relative to single and cohabitating women. In addition, Ananat and Michaels find that divorce benefits some women with children financially, as they moved in with relatives with significant incomes and/or the former husband was not contributing anything or very much to family income (see their forth- coming article “The Effects of Marital Breakup”). Nevertheless, as discussed in the text, Ananat and Michaels also find that divorce greatly increases the odds that women with children are in the low- est income quartile. In some individual cases women and children may be better off without the children’s father because they avoid violence or poverty, but empirical evidence suggests that mar- riage tends to have the opposite effect by keeping women and children away from violence and increasing material resources.
26. There are other taxpayer-funded programs that likely experience larger expenditures due to family fragmentation such the Earned Income Tax Credit, remedial school programs, and special edu- cation programs. These programs were excluded because we did not feel comfortable making reason- able cost estimates given the available empirical literature. The likelihood that these programs would experience reduced costs if more single-adult households became married families suggests that this estimate of the taxpayer cost of family fragmentation is an underestimate of the true costs. Some state funds for TANF programs that benefit children are included in the “child welfare” calculation below, but other non-cash assistance TANF funds are excluded from this analysis.
27. Some evidence suggests that more children in single-parent families are hospitalized for asthma or childhood diabetes because single parents can be less able to manage the complex stresses of chronic illness for themselves and their children, and because they have less access to adequate health care. See Linda Waite and Maggie Gallagher, The Case for Marriage: Why Married People Are Happie , Healthie , and Better-Off Financially (New York: Doubleday, 2000).
28. Assumption 3 does not imply that only households in poverty are eligible for means-tested programs. Most federal means-tested programs serve significant numbers of households with incomes above poverty thresholds. As marriage would reduce poverty by increasing household incomes, mar- riage would also increase the income of households that already had incomes above poverty thresh- olds but were receiving means-tested transfers. At least some of these households would be rendered ineligible for these means-tested transfers due to marriage.
29. Ananat and Michaels, “The Effects of Marital Breakup” and Thomas and Sawhill, “For Richer or for Poorer.”
30. In table 3, female-headed households refers to all households with a female householder and no spouse present, including households with and without children. In 2006 there were 12,827,000 children in poverty. Among those, 7,715,000 lived with a single mother. If marriage were to lift 60 per- cent of these 7,715,000 children out of poverty, then 4,629,000 children would escape poverty. Thus, marriage would lift 4,629,000/12,827,000 or 36.1 percent of these children out of poverty. In 2006 there were 36,460,000 total persons in poverty. Among those, 19,257,000 lived with a female householder with no spouse present or were themselves the female householder with no spouse. If marriage were to lift 60 percent of these 19,257,000 individuals out of poverty, then 11,554,000 people would escape poverty. Thus, marriage would lift 11,554,000/36,460,000 or 31.7 percent of these individuals out of poverty.
31. Based on their estimate of the impact of marriage on the poverty status of female-headed households, Thomas and Sawhill find that the overall 1998 poverty rate would have been 24 percent lower if the proportion of children living in female-headed households in 1998 was the same as had existed in 1970 (see Thomas and Sawhill, “For Richer or for Poorer”).