Does Marriage Reduce Poverty?

Executive Summary

As marriage rates decline, particularly among low-income Americans, there is a growing debate over the role that marriage plays in reducing poverty. Some policy experts cite evidence that children from two-parent homes are less likely to live in poverty and enjoy significantly better outcomes than do children raised in single-parent families. Even without the presence of children, there appear to be social and economic benefits to marriage. Others worry that government programs and policies that explicitly promote marriage would pressure women to enter or stay in marriages that are harmful to the well-being of those women and their children. While recognizing the benefits of marriage as an anti-poverty tool, some remain skeptical of the ability of the government to successfully promote marriage. 

Although context is extremely important, the weight of evidence suggests that increasing the incidence of marriage among low-income Americans could play a limited but important role in reducing poverty. This is particularly true for parents with young children. 

And while serious questions remain about the role that government can or should play, certain policy reforms could have a positive effect on marriage rates. In particular, policymakers should focus on reforms that would make non-college-educated, working-class men more marriageable and better able to contribute to a household. These reforms  would involve the education system, job creation, criminal justice, and housing affordability, among other things.

At the very least, government policies should not create barriers to marriage. Yet, too often, social welfare programs can cause a drastic loss of benefits if couples marry. This form of  “welfare cliff” can actively discourage marriage. Cohabitation, then, is often a better financial decision, at least in the short term, than marriage. To the degree that welfare programs discourage marriage, they are undoubtedly counterproductive.

The role of marriage in reducing poverty—and the government’s role in influencing it—can be overstated. But if properly designed and contextualized, efforts to encourage marriage can be an important tool in helping to reduce poverty. 

Introduction

Any strategy for reducing poverty in America has to consider the “success sequence,” which posits that if someone finishes high school, gets a job, gets married, and only then has a child, they are unlikely to live in poverty. There is a strong circumstantial case to support the success sequence. Few Americans who finish school, work, and delay childbearing until they are married live in poverty. In fact, as Richard Reeves writes for the Brookings Institution, 73 percent of poor whites and 59 percent of poor African Americans who follow the sequence not only escape poverty but actually reach the middle class.

However, not all components of the success sequence carry the same weight or can be linked directly to outcomes. For instance, work seems obviously linked to avoiding poverty. Just four percent of full-time workers are poor, though even part-time work makes a measurable difference. Only 10 percent of part-time workers are poor compared to nearly 30 percent of those who do not work at all. Education’s impact is less direct, but linked with employability and other opportunities. Among people 25 and older, almost one-quarter of those without a high school degree are poor, nearly double the percentage for high school graduates. Marriage, though, may be the most tenuous and contentious aspect of the success sequence. 

Marriage’s unclear effect has led to an intense debate among scholars who study poverty. Many social and religiously inspired conservatives view marriage as a positive good on its own terms, but many secular scholars have also concluded that marriage can make a significant difference in reducing poverty. Yet, others raise concerns that such policies would pressure women to enter or stay in marriages that are harmful to the well-being of those women and their children.

In addition to debates over whether government policies should encourage marriage, there are also important questions about whether government policies can improve marriage rates and which policies might be effective in doing so. 

It is certainly true that married couples have a lower poverty rate than single adults. Roughly 13 percent of single childless adults live in poverty, compared to just 4.6 percent of married couples. On average, single adults have lower earnings than married adults and are more likely to be unemployed. They also have lower educational attainment and are more likely to live with their parents. Other research suggests that married and cohabiting adults fare better than those who are unpartnered when it comes to some health outcomes. Moreover, the economic gap between married and unmarried people has grown over the last 30 years. 

Writing for the Federal Reserve Bank of Chicago, Jonathan Davis and Bhash Mazunder suggest that much of the decline in social mobility of recent decades can be attributed to declining marriage rates among low-income Americans. 

Much of the discussion of marriage concerns whether it benefits children (see below), but there appear to be advantages to marriage for childless couples as well, especially for men who marry. The average married man lives longer, is healthier, is less lonely, and earns more money than his single counterparts.  

Therefore, would increasing the incidence of marriage among low-income Americans reduce poverty? And, if so, are there policies that the government can pursue that will help bring about such an increase in marriage? The answer is surprisingly unclear.

Marriage rates have been declining for some time. As of 2021, just 53 percent of American adults were married, while another nine percent were unmarried but living with a partner. To compare, as recently as 1990, two-thirds of Americans were married, and four percent were cohabiting. Of course, the unmarried population does include some previously married people (i.e., divorced, widowed, etc.), but all of the growth in the unpartnered has been the result of an increase in the number of never-married men and women.

While some of the marriage decline reflects a shift to cohabitation, it is an imperfect substitute for marriage. Cohabiting relationships tend to be shorter and less stable than marital ones. On average, a cohabiting couple in the United States lives together for roughly 14 months and ends without marriage. If cohabiting partners have children, the relationships last longer, but even so, nearly two-thirds of the women who were single at the time of a birth were no longer in a relationship with the father within five years. Cohabiting families are also 74 percent more likely to be poor than married couples, and the breakup of cohabiting couples tends to be harder on women economically than divorce. 

Marriage rates vary significantly by race, gender, education, and income: 59 percent of African-American adults were unmarried in 2019, compared to 38 percent of Latinos, 33 percent of white Americans, and 29 percent of Asian Americans. For most demographics, men are more likely to be unmarried than women. However, African-American women (62 percent) are more likely to be unmarried than African-American men (55 percent).

Attitudes toward marriage have changed significantly over the last half-century. More than half of men and women still believe that marriage is important, but only about one in six believe that it is essential for a fulfilling life. That is considerably fewer than those who believe that having a job or career that they enjoy is essential, and roughly the same as the percentage that say that “having lots of money” is essential to fulfillment. Notably, younger adults are more likely than their older counterparts to believe that marriage is not important to a fulfilling life: 37 percent of adults under 30, compared with 30 percent of those ages 30–49, 27 percent of those 50–64, and 20 percent of those 65 and older. Younger adults are also more likely to say that being married is not important for a woman to live a fulfilling life. And women (31 percent) are more likely than men to say being married is not important for a woman to live a fulfilling life (24 percent).

A growing body of evidence suggests women are increasingly dropping out of the dating and marriage markets. This trend is not unique to the United States, as Japan, South Korea, most countries in Europe, and even Iran have seen similar declines in marriage rates.   

It is important not to over-read the trends, of course. Just five percent of never-married Americans say they don’t want to marry. More than half of never-married adults want to marry, and even 20 percent of previously married adults want to remarry. The vast majority of Americans—roughly 93 percent—will marry at some point in their lives. Still, the trend lines are concerning, particularly in the context of poverty and the effects on non-marital births (see below). 

Marriage and Children

Policy experts across the ideological spectrum agree that marriage makes an important difference in outcomes once children become part of the picture. 

In 1970, two thirds of prime-age adults (ages 25–49) lived with a spouse and children. Today, just one-third do. Concomitantly, the share of children born outside marriage has increased from roughly 28 percent in 1990 to more than 40 percent today. Add in children born to married parents who subsequently divorced, and roughly half of all children are growing up in households headed by single women. As of 2023, there were roughly 7.3 million custodial single mothers and 1.6 million custodial single fathers. 

Policymakers often associate single parenthood with women having children without being married. However, roughly 28 percent of single-mother households result from divorce, another 15.5 percent are separated, and 4.3 percent are widows. This data suggests that policymakers must pursue a multifaceted approach to the issue.

Single mothers fare poorly economically. Using the Census Bureau’s official poverty measure, single mothers suffer a poverty rate as high as 28 percent. Even using the supplemental poverty measure, which counts income and benefits not included in the official measure, 26 percent of single mothers live in poverty, compared to just five percent for married mothers. The official poverty rates for single Latino and African-American women are higher than their white counterparts, at 33 and 31 percent, respectively. Single fathers have an official poverty rate of roughly 15 percent, which is much better than single mothers, but remains substantially worse than the five percent for married fathers. 

Nearly three-quarters of single mothers work, most of whom work full-time, but they still struggle to make ends meet. In 2022, working single mothers earned a median annual income of roughly $40,000, compared to $57,000 for single fathers, $60,000 for married mothers, and $76,000 for married fathers. Black ($38,000) and Latino ($34,000) single mothers earned less, as the poverty rates suggest. 

Other statistics are equally troubling. For example, 35 percent of female-headed households are food insecure—lacking access to sufficient nutrition for normal growth and development—compared to just 14 percent of the general population. Slightly more than 10 percent of single mothers have no health insurance, and approximately one in six have not completed high school. 

Single mothers are also far more likely to receive public assistance. According to the Center for American Progress: 77 percent of single mothers receive the child tax credit; half receive the earned income tax credit (EITC), 40 percent rely on Medicaid for health care; 38 percent participate in the Supplemental Nutrition Assistance Program (SNAP), also known as food stamps; 17 percent live in public housing and another 8 percent receive rent subsidies; 14 percent receive benefits through the Supplemental Nutrition Program for Women, Infants, and Children (WIC); and 85 percent of single mothers have children that eat free or reduced price lunch at school.

In her 2014 book Generation Unbound, Isabel Sawhill of the Brookings Institution writes that differences in family structure have increased income inequality by 25 percent. She posits that if the United States returned to the marriage rates of 1970, child poverty would be as much as 20 percent lower. 

None of this should be surprising. Raising a child is likely to be more difficult, relying on one rather than two incomes. A single parent is more likely to have problems with child care that could interrupt employment or educational opportunities. Employers may be hesitant to hire single parents who may have to take time off to care for their children. Even if having a child outside of marriage does not throw a person into poverty, it certainly can make it more difficult to climb out of poverty.

Significantly, the consequences of non-marital childbearing do not just fall on the parents; children also suffer. Studies show that children raised in single-parent households have lower incomes as adults than their peers who grew up in two-parent families. The non-economic consequences for children of single parents are equally stark. Melissa Kearney writes in her important and influential new book, The Two-Parent Privilege, “Children from single-parent homes have more behavioral problems, are more likely to get in trouble in school or with the law, achieve lower levels of education, and tend to earn lower incomes in adulthood.” This relationship appears to hold regardless of the age, race, or educational status of the mother.  

Of course, many single parents and their children do just fine. But the data make clear that single-parent households often make life more difficult for both the parent and children.  

Marriage is Different for Poorer Americans

It is important to recognize that marriage does not have the same characteristics across all economic or social groupings. Notably, the positive outcomes of marriage for higher-income and more educated couples cannot necessarily be applied to the lower-income and less educated. Thus, when studies show that married men and women are likely to be healthier, happier, and wealthier on average than unmarried men and women, it is important to contextualize  that data 

Couples with more education and higher incomes tend to postpone marriage and childbirth, but eventually marry at high rates and have a lower risk for divorce. For example, studies show that women with  college degrees are 12–17 percent more likely to marry than women without a degree, a gap that has been growing steadily over time. And, college-educated women are as much as 40 percent less likely to divorce. Low-income women, on the other hand, are both less likely to be married and less likely to benefit from marriage. Conservative scholars such as Charles Murray and David Brooks have written extensively about the differences in family structure between those at the top and bottom of the income and education scales. 

In his book, Coming Apart: The State of White America, 1960–2010, Murray contemplates  two hypothetical communities, the upper-middle-class community of Belmont and the lower-income community of Fish Town. Those living in Belmont have at least a bachelor’s degree and work in professional, well-paying jobs, while those in Fish Town have no education higher than high school, if that, and work in blue-collar, service, or other low-level occupations. Murray explains that: 

In 1960, extremely high proportions of whites ages 30–49 in both Belmont and Fishtown were married—94 percent in Belmont and 84 percent in Fishtown. The unquestioned norm in both neighborhoods was marriage. In the 1970s, those percentages declined about equally in Belmont and Fishtown. Then came the great divergence. In Belmont, marriage among prime-age adults stabilized during the mid-1980s and remained flat thereafter, standing at 83 percent in 2010. In Fishtown, marriage continued a slide that had not slackened as of 2010, when the percentage of married whites ages 30–49 had fallen to a minority of 48 percent. What had been a 10 percentage point difference between Belmont and Fishtown in the 1960s stood at 35 percentage points in 2010. The culprits—divorce and failure to marry in the first place—split responsibility for the divergence about equally.  

Similarly, Brooks writes:

In 1970, the family structures of the rich and poor did not differ that greatly. Now there is a chasm between them. As of 2005, 85 percent of children born to upper-middle-class families were living with both biological parents when the mom was 40. Among working-class families, only 30 percent were. According to a 2012 report from the National Center for Health Statistics, college-educated women ages 22 to 44 have a 78 percent chance of having their first marriage last at least 20 years. Women in the same age range with a high-school degree or less have only about a 40 percent chance. Among Americans ages 18 to 55, only 26 percent of the poor and 39 percent of the working class are currently married. […] As Andrew Cherlin, a sociologist at Johns Hopkins University, once put it, “It is the privileged Americans who are marrying, and marrying helps them stay privileged.

Studies suggest that marriages among less educated, low-income couples tend to be more stressful, less happy, and more likely to succumb to intimate partner violence. Given these facts, it may be unsurprising that low-income couples are more likely to divorce. Tellingly, similar dynamics apply to cohabitation. Higher-income cohabiting couples tend to form more stable, longer-lasting relationships. 

All of this suggests exercising caution when extrapolating outcomes for middle-income or high-income marriages and projecting them onto marriages among the poor. Policy recommendations cast too broadly are likely to sweep-up large numbers of wealthier, more educated Americans who will benefit little while bypassing those who may most need assistance. Likewise, it may lead policymakers to endorse solutions that do not meet low-income couples’ needs and therefore will be less likely to succeed.  

Chicken or Egg?

Many marriage advocates operate on the assumption that if marriage has been beneficial to the currently married, it would be equally beneficial to those who are not married. Therefore, if policymakers can simply convince, cajole, or coerce more couples into marrying, those men and women—and their children—will be better off. While this may be true in some instances, there are a number of problems with this projection. 

Most important, correlation is not causation. When discussing the relationship between marriage and poverty, there is a serious chicken-and-egg problem. That is, do the circumstances and characteristics that make a person more likely to be poor also make that person more likely to be unmarried, or vice versa? For example, a study published by the American Sociological Association found that married men earn more because high-earning men are more likely to marry in the first place. 

As Kristi Williams of Ohio State University told New York Times economics writer Annie Lowery in 2014, “It isn’t that having a lasting and successful marriage is a cure for living in poverty. Living in poverty is a barrier to having a lasting and successful marriage.”

Characteristics like happiness, health, attractiveness, income, education, and community all correlate strongly with each other and with marriage. On the other hand, as Cathy Reisenwitz writes in Sex and the State, “Pairing two sick, miserable, poor, etc. people together not only demonstrably fails to confer the same benefits as pairing two happy, healthy, wealthy, etc. people, but there’s good evidence to suggest it actually leaves them worse off.” Kearney warns that the benefits of marriage depend heavily on context. As she writes, “If a father doesn’t have any positive resources (financial or otherwise) to bring into the household, then there would be no household gain to the marriage. Conversely, if that father would bring violence, or chaos, or stress into the household, the gains to marriage could be, and probably would be, negative.” 

A study by Philip Cohen of the University of Maryland found that the economic gains from marriage are greater for women who are more likely to be married. Therefore, he concludes that if currently unmarried women were to marry, it would reduce the average observed benefits of marriage. Or, as Kearney puts it, “If the mothers who are not married are not married precisely because the men with whom they have fathered children would not meaningfully contribute positive resources to the raising of their children, then the observed marriage gap in children’s outcomes is not a good approximation of what children would gain from parental marriage.”

Even if there is widespread agreement that happy, healthy, stable marriages provide important benefits and reduce poverty, policymakers must deal with the fact many marriages do not meet this description. Pushing couples into unhappy or unhealthy marriages might, in fact, make matters worse. For instance, at least one study shows that while children born to married teenagers fared much better than children born to teenagers who were unmarried, children of teen parents who married after the child’s birth do worse on many measures. The authors speculate that this is likely the result of the stress and conflict that occurs in marriages that result from social, familial, and economic pressure.

These outcomes do not negate a role for policymakers to encourage marriage, but they should be careful not to overestimate the potential benefits. As Kearney summarizes, “increased rates of marriage among unmarried parents might be beneficial to children in some instances, but likely not all.”

The Role of Divorce

Not only are low-income Americans less likely to marry, they are also more likely to divorce. Roughly 20 percent of married couples with incomes under $25,000 per year divorce compared to less than 12.3 percent of those earning $100,000 per year or more. And while overall U.S. divorce rates have been declining since the late 1970s, that trend has not been the case for lower-income Americans. 

Moreover, divorce tends to leave low-income women worse off while improving the economic position of low-income men. Roughly 22 percent of recently divorced women live below the poverty level compared to just 11 percent of recently divorced men. In addition, children of divorced parents are more likely to grow up in poverty than are children whose parents remain married. On the other hand, making divorce easier to obtain significantly reduces domestic violence for both men and women, lowers the number of women murdered by their partners, and also decreases incidents of female suicide. Some studies suggest that there are negative consequences for children growing up in households where the parents are unhappy but remain married, such as psychological problems and emotional distress, behavioral issues, and problems with future relationships. 

Contrary to the stereotype of women taking men to the cleaners in divorce, spousal support remains relatively uncommon in divorce settlements. Child support is ordered for custodial mothers in only about 75 percent of cases. Even when ordered, the amount is likely to be quite low: less than $4,800 per year, on average. In reality, men frequently default on such payments and custodial parents receive only about $2,500 in support. This falls far short of the $29,500 per year average cost of raising a child. Child care alone costs $10,000–15,000 per year.

Beyond the financial consequences of divorce, studies have shown a wide range of emotional and behavioral problems for children whose parents divorce. A study by Jonathan Gruber of MIT, for instance, found that children of divorced parents had lower levels of educational attainment, lower incomes as adults, and a greater likelihood of divorce themselves. Studies also show that children of divorced parents are more likely to grow up poor. All of these problems become worse if the father becomes disengaged from the child’s life post-divorce. 

Despite the likelihood that divorce will leave them worse off, low-income women continue to be the primary initiators of divorce. This comes despite the extremely negative attitudes that low-income women express about divorce. Specifically, poorer women fear divorce to the extent that it can actually reduce their likelihood of getting married in the first place. As Andrew Cherlin and others write in the Journal of Marriage and Family, “low-income women fear divorce and that this fear is an important factor in their reluctance to marry. They fear divorce so much, it is said, that they will not marry unless they are sure the marriage will last a lifetime.”  

In their widely cited book on attitudes toward marriage and divorce, Kathryn Edin and Maria Keflas conclude, “Most poor women we spoke with say that it is better to have children outside of marriage than to marry foolishly and risk divorce, for divorce desecrates the institution of marriage.” This suggests that efforts by some conservatives to make divorce more difficult to obtain—several states, including Louisiana, Oklahoma, and Texas, are considering legislation to abolish “no fault” divorce—could backfire by leading to lower marriage rates.  

Overall, policymakers should be aware of the effects that divorce imposes on single-parent households beyond non-marital births. However, the ability of the government to substantially reduce divorce without creating more problems than it solves appears limited.    

Policies to Address the Marriage Divide

Recognizing marriage’s role in reducing poverty is less difficult than developing sound policies that might increase marriage rates for low-income Americans. In fact, government efforts to increase marriage rates have almost universally failed. For instance, President George W. Bush proposed the Healthy Marriage Initiative, which funded state and local efforts to promote marriage for low-income couples. However, independent evaluations found that the programs had little effect on marriage and divorce rates.

Even so, there are some policies and policy approaches that could make a difference.  

As Nobel laureate Gary Becker pointed out, marriage is a market. As with most other markets, both the quantity and quality of the possible choices matter. Unfortunately, many women find both the quantity and quality of marriageable men increasingly scarce.  

One of the first to write about the concept of a shrinking marriage pool was Harvard’s William Julius Wilson in his 1987 book, The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. Wilson developed the idea of a Marriageable Men Index that measured the ratio of single employed men with a given age and race compared to the number of single women. That ratio has been shrinking rapidly for decades. The reasons for this lie on both sides of the gender divide. 

For much of American history, marriage was a route to upward mobility for women, one of few widely available. In 1960, for instance, 75 percent of men with college degrees married women without one. That is no longer the case. Marriage is increasingly assortative. That is, men and women both tend to marry partners of the same socioeconomic status.   

In large part, this sorting is the result of increased economic and educational opportunities for women. Women are more likely to attend college and have a high earning job. In fact, women are now more likely than men to attend college. The share of young women with a bachelor’s degree has increased by 22 percentage points since 1995, from 25 to 47 percent. Over the same period, the number of men with degrees has also risen, but at a slower rate from 25 to 37 percent. As a result, prior to 1990, women with college degrees were less likely to be married than those without, but that has now completely turned around. Today, college-educated women are the most likely to be married, tending to marry college-educated men. Roughly half of college-educated women marry a man with a degree. The other half are split between those who marry someone without a degree and those who don’t marry at all. 

Importantly, when women marry down educationally, it is to men who nonetheless have higher incomes and strong economic prospects. Average earnings for non-college-educated men who marry college-educated women have increased over the last 50 years, while falling for non-college men who do not. A geographic analysis by the Census Bureau and Harvard’s Opportunity Insight shows that in areas where non-college men have higher rates of employment and lower rates of incarceration, the difference in marriage rates between college-educated and non-college women were as much as 50 percent smaller. 

Second, while women’s wages still tend to trail men’s, they are actually rising at a faster rate. This trend is likely to increase since the jobs of the future are likely to rely on analytical, managerial, and interpersonal skills rather than physical strength.  

Logically, then, an increase in women’s relative economic and educational position relative to men will result in fewer marriages and more divorces because the option of being able to support themselves outside the marriage is much improved while the economic benefits of remaining married are not getting any better. 

The relationship between the declining economic status of low-wage, low-education men and their declining marriage rates is well established. Men without a college degree experienced the largest decline in marriage rates from 1980 to 2020. Significantly, this is also the demographic that experienced higher rates of joblessness and stagnant earnings. In contrast, there has been relatively little change in marriage rates among college-educated men, whose income and employment rates have increased. 

On average, unmarried men earn $21,000 less per year than married men. More than one-third have incomes below 150 percent of the poverty line. Fully one-quarter are unemployed, compared to just 18 percent of married men. In addition, unmarried men are far less likely to have a college degree (29 percent compared to 41 percent). Unmarried men are also more likely to suffer from addiction or alcohol abuse and to be socially isolated. In fact, 28 percent of unmarried men live in their parents’ home. 

In 1960, a married couple, with neither spouse having graduated high school could expect a household income equal to roughly 77 percent of the national average. By 2005, however, that couple’s household income would be only about 41 percent of the national average. Similarly, in 1960, a married couple with two high school degrees but no college had a household income slightly above the national average (103 percent), but that had fallen to 83 percent by 2005. Over that time period, if both partners had college degrees, household income increased from 153 to 160 percent of the national average.  

Stagnating economic prospects for men at the low end of the earnings and education spectrums make them less desirable marriage partners. As McGill University economist Lyman Stone notes, “Men’s odds of being in a relationship today are still highly correlated with their income. Women do not typically invest in long-term relationships with men who have nothing to contribute economically.” 

A study by Kristan Harknett of the University of California-San Francisco and Daniel Schneider of the University of California-Berkeley estimates that as much as one-third of the decline in marriage rates since 1969 can be attributed to a decline in earnings and economic stability among men who graduated high school but did not attend college. 

It is not simply a question of declining wages for non-college educated men, but the decline in male earnings relative to women’s reduces marriage rates. Na’ama Shenhav with the Goldman School of Public Policy at the University of California-Berkeley estimates that 20 percent of the decline in marriage from 1980 to 2010 can be attributed to relative wages. That is, as men are less likely to provide an income premium, women are more likely to forgo marriage. Given the trends in male and female incomes, this suggests that marriage rates will continue to decline.  

Marriage is a high-stakes gamble, especially for women. But as women increase their earning potential without a partner, the potential return on that gamble decreases. With less potential financial upside to marriage, women are less likely to marry. Viewed this way, it makes sense that women are more selective about potential mates.   

If high-status men are increasingly marrying horizontally rather than down, and high-status women are both increasing as a share of the population and also marrying horizontally or pairing with high-income, low-education men, lower-status men and women draw from an ever-shrinking pool of potential mates.  

For African-American women, the problem is compounded by inequities in the criminal justice system. An analysis of the research by the New York Times suggests that as many as 1.5 million black men have been taken out of the marriage pool because they are involved in the criminal justice system, imprisoned, on parole, or on probation. This means that for every 100 black women, there are just 83 available black men, even before one considers income, employment, and education. This will undoubtedly become less of an issue as interracial marriage continues to increase, but it is worth noting that black women remain the racial group least likely to marry a spouse of a different race.  

Richard Reeves of the Institute for Boys and Men cites research indicating that “improving economic opportunities for men without a college degree could substantially increase marriage rates, particularly in poorer communities and among non-college-educated women.” The precise reforms necessary and how best to implement opportunities are generally beyond the scope of this paper. However, effective programs would likely involve reforms to education, employment, the criminal justice system, housing, and more. 

Education

Any program to increase marriageability for lower status men should begin with education reform. 

As noted above, women are now more likely to enroll in and graduate from college. Today, men receive just 42 percent of bachelor degrees awarded in the United States. To put that in perspective, that’s roughly equal to the percentage of women who received bachelor degrees in 1970, and this divergence is likely to grow. There are 2.5 million fewer male than female undergraduate students.

But men’s educational problems do not start with college. At age five, girls are 14 percentage points more likely to be ready for school—that is, having the social, cognitive, language, and physical development needed to be able to focus and learn. Once in school, boys are more likely to underperform. In high school, for instance, boys generally perform at a level one letter grade below girls (e.g., earning a B to a girl’s A) and are less likely to take Advanced Placement, honors, and other advanced coursework. Boys are also far more likely to be disciplined, suspended, or expelled from school. Dropout rates are higher for boys than girls. And, while boys still outperform girls on ACTs and other standardized tests, that gap is narrowing rapidly. 

All of this indicates a need for significant educational reform, particularly for boys. Numerous reformers have suggested changes such as increasing the number of male teachers, delaying the age at which boys start school, employing single-sex schools, and changes in teaching methods among other ideas. But none of these are likely to happen without significant disruption of the current education establishment. In particular, expanding parental choice and control is key. Charter schools, vouchers, and educational savings accounts all offer varying alternatives to the failing status quo.

While it is essential to provide more young men with the educational foundation necessary for them to attend and graduate from college, it is important to recognize that college is not the answer for everyone. There will always be students who are unsuited for college. Taking on a heavy burden of college debt is a mistake when the prospective student lacks desire, temperament, or aptitude.  

The medium-to-long-term financial and opportunity costs of four-year college education makes alternatives such as vocational training all the more important. Although there is little scholarship directly linking vocational education to marriage rates, studies draw connections to increased employment rates and income, which demonstrably improve marriageability. For example, a study of nine vocational and technical academies in New York by the Manpower Demonstration Resource Corporation found that vocational graduates earned an additional $30,000 per year, roughly a 17 percent premium. Similarly, a study of 11,000 vocational students in Connecticut found that male students at these schools had a graduation rate 10 percentage points higher than in traditional schools and that their wages were 33–35 percent higher by age 23.

More students, largely young men, have enrolled in vocational or technical training over the last few years. In particular, they have enrolled in vocational training at the community college level. Richard Reeves convincingly makes the argument for investing more vocational resources at the high school level. Providing vocational opportunities earlier in a student’s educational path can help them improve their earning potential earlier in life, enter the workforce, and support themselves at an age their college peers still rely on parental aid. 

Employment

Male workforce participation has declined from 96 percent in 1970 to just 89 percent today. There are more than 6.9 million able-bodied men who are not working or looking for work. This reduces the availability of appealing male marriage partners. Policymakers should thus increase employment opportunities for those lower-education men who are losing out in today’s changing economy, but there are limits to what can be done. 

The 1950s are not coming back. Neither corporate subsidies nor tariffs are going to re-create the factory jobs of the past. The United States actually produces more manufactured goods than it ever has, but it achieves that with fewer workers. The overwhelming majority of “missing” jobs were lost not to offshoring or foreign competition, but to automation. Attempts to change this are not only likely to fail, but may lead to increased hardship, especially for low-income Americans.  

That said, there is a broad consensus on policies that contribute to a growing economy and job creation: 1) growth in the labor supply, 2) improvements in the quality of labor participation, and 3) policies that increase innovation and productivity, notably a tolerable level of taxation and regulation. Federal and state governments should take steps to ensure economic growth.

Beyond this, there are some specific actions that governments could take to improve the economic opportunities for non-college-educated men. For instance, degree requirements could be eliminated for most government jobs. Several states, including Maryland, New Jersey, Ohio, Pennsylvania, and Utah, have taken steps to implement this change for up to 90 percent of state jobs. Minnesota has put similar but more modest reforms in place.

States can also revise their occupational licensing laws to make it easier for non-college-educated men to find work or start businesses. Today, more than 1,100 occupations, from barbers to funeral attendants, require a license in at least one state. As much as 25–30 percent of the workforce is currently covered by some form of occupational licensing. Although such licensing requirements amount to little more than a formality in some cases, the education, training, and financial burdens can be substantial in far too many. These burdens effectively put professions off-limits to Americans who lack the time or money necessary to complete the training requirements. 

Another reform to boost low-income, low-education men would be to expand apprenticeship programs. Apprenticeships have a strong track record of success. Roughly 93 percent of those who enroll in an apprenticeship program find a job at the end of it, with an average annual salary of $77,000. 

In his book, Apprentice Nation, Ryan Craig, co-founder of Apprenticeships for America, points out that the government spends less than $400 million annually to support apprenticeships, compared to more than $500 billion to support college education. And, as Robert Lerman of the Urban Institute notes, “The United States has lagged far behind other developed countries—countries like Germany and Switzerland, but also Australia, Canada, and England—in creating apprenticeships. In these countries, apprentices constitute about 2.5–3.0 percent of the labor force, or about 10 times the U.S. rate.”

Criminal Justice Reform

As noted above, many young men are effectively removed from the marriage pool because of their involvement with the criminal justice system. This would be a problem even if there were not substantial inequities in the system.  

Several studies suggest that youthful incarceration in particular reduces opportunities for marriage. A 2018 paper by Kristen Harknett of the UC–San Francisco and Daniel Schneider and Matthew Stimpson of the UC–Berkeley looks across 50 years of data from the Panel Study of Income Dynamics. The authors concluded that incarceration rates accounted for as much as 28 percent of the decline in marriage rates. According to a study by the Russell Sage Foundation, 84.1 percent of ex-inmates eventually marry, compared to 87.2 percent of non-inmates with otherwise similar characteristics.

Those numbers are much worse, however, for African Americans. A black man without a prison record is 54.4 percent likely to get married by his late thirties compared to a 43 percent likelihood for a black ex-inmate. A study by Nazih Guner, Christopher Ruah, and Elizabeth Caucut for the Center for Economic Policy Research estimates that incarceration accounted for 10.4 percent of the gap in black/white marriage rates. 

Incarceration also significantly increases the likelihood of divorce. Unsurprisingly, the longer one partner spends incarcerated, the higher the likelihood that the couple will divorce. One study, for instance, estimates that for every year a spouse spends incarcerated, the chances of divorce increase by one-third. Even shorter sentences make divorce more frequent. The risk of divorce remains high after the spouse’s release from prison. This should not be a surprise given the collateral consequences on employment, education, and housing that can often accompany a criminal record.    

Housing

Common sense suggests that if a couple is unable to find an affordable place to live, they are less likely to marry. According to a survey by Lending Tree, more than one-third of Gen Z and 21 percent of millennials say they plan to postpone marriage until they can purchase a home. 

Unfortunately, housing costs are prohibitively high. Many financial experts agree that families should not spend more than about 30 percent of their gross income on housing. However, the Federal Reserve Bank of Atlanta’s Home Ownership Affordability Monitor puts the median cost of a new house at $385,000, with a total monthly payment of more than $3,000, roughly 47 percent of median income. The affordability crisis is even more pronounced in major metropolitan areas, with median prices exceeding 100 percent of the median household after tax income in most such areas. 

While there has been limited study of housing costs and marriage rates, extant data suggests that high housing costs drive down marriage rates. For instance, a 2018 analysis by the Census Bureau concluded that owning a home was associated with higher marriage rates, while living in the home of a parent was associated with lower marriage rates. Similarly, a much older study published in the Eastern Economic Journal found that “the burden of housing costs appears to play a role in the decision to marry. A higher ratio of the cost of owner-occupied housing to per capita income…is associated with a lower marriage rate.” In addition, several studies from other countries show similar results. Research has also shown that increasing housing costs reduce fertility rates, a similar dynamic to what we can expect for marriage.  

Far too often, misguided government policies significantly contribute to high housing costs. Basic economics teaches that as demand for housing increases but supply remains low, prices rise. In a functioning market, suppliers respond to growing demand by increasing supply. When supply increases faster than demand, prices fall. However, government regulation artificially constrains the supply of housing in most growing cities in the United States.

With this in mind, there are several steps that federal, state, and local governments can take to help make housing more affordable. As FREOPP scholars Roger Valdez, Jon Hartley, and Avik Roy have described, these steps include simplifying the Low-Income Housing Tax Credit; shifting from subsidized housing to Section 8 voucher-style rental assistance; reducing the systemic risks to housing markets caused by Fannie Mae and Freddie Mac; examining the role that the Federal Reserve has played in creating housing bubbles; and, most importantly, eliminating exclusionary zoning regulations.

During his first term, President Trump issued an executive order establishing a White House Council on Eliminating Regulatory Barriers to Affordable Housing. The report calls out federal, state, and local governments for erecting a host of regulatory barriers to building more housing, including: 

Overly restrictive zoning and growth management controls; rent controls; cumbersome building and rehabilitation codes; excessive energy and water efficiency mandates; unreasonable maximum-density allowances; historic preservation requirements; overly burdensome wetland or environmental regulations; outdated manufactured-housing regulations and restrictions; undue parking requirements; cumbersome and time-consuming permitting and review procedures; tax policies that discourage investment or reinvestment; overly complex labor requirements; and inordinate impact or developer fees.

Kearney and others worry that changes to marriage patterns are now so deeply ingrained in culture that even improving the marriageability of men may not reverse trends. She cites a study of communities that experienced positive economic shocks, such as the fracking boom in Pennsylvania, without a significant impact on marriage rates. She warns that economic changes alone may not be enough to restore marriage rates to where they once were.

That may well turn out to be true, but no improvement in marriage rates is likely without increasing the pool of marriageable men. Given that the reforms outlined above are likely to produce positive outcomes even beyond the marriage arena, they are certainly worth undertaking. 

End Government Disincentives to Marriage

The evidence that the simple availability of welfare benefits by itself decreases marriage rates or increases non-marital births has weakened since Charles Murray first wrote about it in Losing Ground. As Kearney notes, whatever the influence of welfare on marriage rates in the past, changes in welfare since the passage of the Personal Responsibility and Work Opportunities Act in 1996 have reduced the role that welfare plays.

However, that does not mean that welfare programs have no effect on marriage rates. Too many welfare programs are designed in ways that create a substantial marriage penalty. For instance, in recent years there has been growing recognition of the problem posed by welfare or benefits cliffs, in which the rapid phase-out of benefits can discourage work. Welfare cliffs occur when a small increase in outside income results in a large decrease in welfare benefits, potentially leaving recipients worse off. These benefit reductions are functionally the equivalent of marginal tax hikes with much the same deleterious effect on work.

Less attention has been paid to the ways in which such cliffs can also discourage marriage. Marriage, even to someone with lower income, can cause a family to completely lose benefits or be caught in a phaseout. Such benefit cliffs effectively make marriage more expensive than remaining single, creating a “marriage cliff.” If a low-income woman marries, she can lose some or all of her benefits from food stamps, Medicaid, Section 8 housing, and the Earned Income Tax Credit (EITC). These penalties compound and can reach tens of thousands of dollars. For example, a study by Bradford Wilcox and others for the Department of Health and Human Services (HHS) warns that these penalties can amount to 10–30 percent of household incomes. The Urban Institute created a calculator that can be used to provide an estimate for how marriage cliffs might affect couples.  

Take the EITC, for example, and a couple with two children. One partner earns $15 per hour, while the other earns $10 per hour. If the couple remains unmarried and the lower-earning partner files for the EITC as head of household, the couple would receive $5,605 in EITC benefits annually. If, however, the couple marries, the benefit would drop to just $275. Marriage would cost the couple $5,330 in benefits.

The Supplemental Nutrition Assistance Program provides another example. In the case of the couple described above, if they remain unmarried and the lower-income partner files for benefits, they would receive $6,108 per year in benefits. On the other hand, if they married, their combined income would make them ineligible for any assistance. (Technically, even unmarried, they would be ineligible for SNAP as long as they shared the same food budget, but few states have the inclination or resources to determine this.) 

Medicaid is yet another program with a significant marriage penalty. Medicaid eligibility is complex and based in part on the category of recipient: mother, child, pregnant woman, disabled, etc., and tied directly to individual income, meaning that unmarried individuals are judged solely on their income, even if they cohabit. However, should they marry, eligibility becomes based on combined income. Medicaid generally does not provide partial benefits or have a phaseout. Eligibility is all-or-nothing based on meeting income thresholds. Families with incomes exceeding the permissible amount for each category will not qualify for benefits, making for an extreme cliff, though the family may become eligible for subsidies under the Affordable Care Act, which could offset some of the loss. A study by scholars at Syracuse University’s Maxwell School for Public Policy Research found that Medicaid’s benefit structure led to higher rates of non-marital childbearing and cohabitation, as well as a reduction in marriage rates.  

Child care subsidies are another area where marriage cliffs are a problem. Indeed, child care programs can have some of the largest marriage penalties. Other social welfare programs also have significant marriage cliffs, including Pre-K education and Supplemental Security Income (SSI). 

Moreover, few welfare recipients participate in just one program. Therefore, the penalties compound as multiple programs come into play. Writing for the Institute of Family Studies (IFS), Ed Dolan of the Niskanen Center considers a common combination of welfare benefits, Medicaid, SNAP, the Child Tax Credit (CTC), and the EITC. A couple living in deep poverty would receive a small net income boost if they marry and combine resources. However, if their combined income exceeds the federal poverty level ($21,223 for a family of two, $26,650 if they have a child), they would quickly lose benefits. Their effective marginal tax rate could be as high as 91 percent. 

HHS estimates that overall marriage penalties can run to 10–30 percent of household income. The result decreases marriage, increases cohabitation, and increases non-marital births among low-income women. In fact, roughly one-third of Americans say they personally know someone who has not married because they feared the loss of social welfare benefits. And, a survey by the Institute for Family Studies and the Wheatley Institute estimates that more than 1 in 10 unmarried Americans whose income falls below the median reported they were not married for fear of losing “access to government benefits.” 

Figure 4 below, using a tool produced by the Atlanta Fed, illustrates how marriage cliffs would play out for a low-income family in Boston, for example, receiving a typical basket of benefits. On a purely economic basis, the family would be better off if they don’t marry. 

The Heritage Foundation estimates that if marriage penalties were eliminated for single mothers with incomes less than $26,000 per year, their annual marriage rate would increase by nearly 250 percent, rising from 9.4 percent to 23.1 percent per year. Among single mothers with incomes between $26,000 and $40,000 per year, the annual marriage rate would more than double, rising from 11.8 percent to 25.6 percent. 

The HHS report concludes, “Financially, it may make more sense to individual men and women to remain single or cohabit due to either a reduction of public assistance benefits or an increase in taxes were they to marry.” This is clearly counterproductive. Whatever one thinks of the government actively trying to encourage marriage, it should not be making it harder or more costly to marry. Federal and state governments should reexamine social welfare programs with an eye toward reducing or eliminating marriage cliffs. 

To be sure, it is not just a loss of welfare benefits that creates disincentives to marriage. Despite reforms in recent years, the tax code also contains marriage penalties, meaning couples would pay a higher tax rate if they marry and file jointly than if they remained single. This is especially true for couples in which one partner earns substantially more than the other. The marriage penalty can run from 4–12 percent of a family’s income. In addition to the federal penalty, 15 states have similar provisions in their income taxes.    

The Tax Cut and Jobs Act (TCJA) of 2017 significantly reduced marriage penalties for low-income workers. That is something that Congress should keep in mind as it debates TCJA extension this year. 

Even if low-income workers are not as directly affected by marriage penalties today, the continuation of such penalties—even for higher-income workers—sends a cultural signal about how the government values marriage. 

The government role in marriage and its encouragement is debatable, and skepticism is reasonable. However, at the very least, lawmakers should review government policies and programs to ensure that the government does not penalize or disincentivize marriage.  

A Final Caveat: Marriage Has Changed

Too often, policymakers limit their vision of marriage to a 1950s model: a single-earner, male-headed household that is increasingly outdated. Even if all the reforms designed to help low-education, low-income men discussed above were enacted, the share of marriages fitting into the 1950s model would remain small. As Richard Reeves explains, “Today, the very institution of marriage, which is central to human societies, has been fundamentally transformed. It is an institution that is now entered into on the basis of egalitarian principles.”

Nearly three-quarters of married women with children are either in the labor force or looking for work. Mothers with younger children were, unsurprisingly, less likely to work, but even among this demographic, nearly 69 percent were employed. It is not just part-time work either; 80 percent of working mothers are employed full-time.  

Still, the nostalgia for the old marriage model is hard to shake. According to a Pew survey, fully 71 percent of Americans said that it is important for a man to earn enough money to be able to support a family. But only one third felt the same about women.  

Women continue to struggle with what sociologist Arlie Hochschild calls the “second shift,” where they are expected to be responsible for the majority of housework and child care responsibilities even if they are employed outside the home. On average, women perform 7.3 hours of housework, such as laundry, cleaning, and meal preparation, compared to just 1.4 hours for men. Women also spend roughly double the amount of time their spouses spend in taking care of their children. Many women are effectively working two jobs but only being paid for one. This imbalance actually becomes worse when a woman outearns her partner. In those marriages, women perform more housework, while men perform even less. 

Domestic violence and spousal abuse also increase in marriages where the wife is the higher earning partner. Men married to women who outearn them are more likely to cheat on their spouse as well. Outcomes are so bad that women often lie about their actual earnings. Divorce rates are as much as 50 percent higher for couples where the woman out-earns her husband. For all these reasons, a study from the National Bureau of Economic Research found that as much as 23 percent of the decline in marriage rates can be linked to concerns about higher-earning women. 

This suggests that any long-term strategy to encourage marriage will require men to adjust to a changing marriage landscape. Of course, any such change will be driven by cultural shifts, not government policies. This is already occurring in women’s education. At one time, marriages in which the wife had a higher level of education than her partner had the same sort of negative outcomes as those where the wife out-earned her spouse. However, as those marriages became more common, the stigma faded and outcomes began to more closely resemble those of more traditional marriages. 

Conclusion

Marriage may well play an important—if limited—role in reducing poverty. There appear to be social and economic benefits to marriage. Certainly, children brought up in two-parent families are at an advantage over those raised by single parents, as they are less likely to live in poverty and have better outcomes across a wide range of developmental measures. 

These outcomes suggest that government policies should support and encourage marriage if possible, without trapping women in marriages that are harmful to the well-being of those women and children.

For example, policies that increase the education, employment, and earnings potential of young unmarried men—such as educational choice, pro-employment policies, criminal justice reform, and housing reform—could help boost marriage rates and reduce poverty rates. Ensuring that government programs and policies do not penalize marriage could do the same, while improving the social safety net for all families. 

All this should take place while recognizing the changing nature of marriage today. 

The role of marriage in reducing poverty—and the government’s role in influencing it—is often overstated. But if properly designed and contextualized, efforts to encourage marriage can be important tools in reducing poverty.