Does marriage improve well-being for parents and children? It can certainly appear that way. In the contemporary United States, children who grow up with married parents perform better in school, enjoy better physical and emotional health, more often begin and finish college, and enter stable employment at higher rates compared to peers who grow up in other family arrangements (Brown, 2010). Married parents appear to be better off, too: they report being happier, healthier, and more financially secure than parents who are single or in cohabiting unions (Waite, 1995).

The relationship between marriage and well-being is of policy interest for several reasons. First, a lot of childrearing in the U.S. happens outside of marriage. Forty percent of U.S. children are born to unpartnered or cohabiting parents (Guzzo, 2021; Osterman et al., 2024, Table 9), and roughly one quarter of children under age 18 live with a single parent, usually their mother (Census Bureau, 2022, Table C3). By age 12, more than half of U.S. children have spent some time outside of a married-parent family household (Brown et al., 2016).

Further, in the U.S., children in single-parent households, and particularly those headed by single mothers, are exceptionally likely to be poor, and child poverty is strongly associated with compromised development and achievement (Duncan et al., 1998). Among families with children in 2022, 37.2% of female-headed households and 18.3% of male-headed households were in poverty under the official poverty measure, compared to just 6.9% of married-couple families (Shrider & Creamer, 2023, Table A-2). Among 30 peer countries, the U.S. ranks first for single motherhood's average marginal effect on the probability of being in relative poverty (Brady et al., 2024).

And although most Americans say that they would like to marry (Gallup, 2020), married parenthood is largely stratified by race and social class. Sixty percent of Asian adults, 54% of White adults, and 63% of college-educated adults are in married couples today, compared to fewer than half of Black or Hispanic adults and adults with a high school education (31%, 45%, and 45%, respectively; Census Bureau, 2022, Table F2; Julian, 2023). Married adults also have higher earnings at marriage compared to their same-aged unpartnered or cohabiting counterparts (Ludwig & Brüderl, 2018; Oppenheimer, 2003).

Socially patterned disparities in marriage formation and stability in the United States are not new: indeed, Dianne M. Stewart (2020) has described the profound systemic barriers to stable marriage that Black adults have encountered over four centuries in America as “this country's most camouflaged civil rights issue” (p. 217). But at the end of the Baby Boom, marriage was nearly universal across sociodemographic groups (Allred, 2018) and nonmarital fertility was infrequent. Since then, sociodemographic disparities in the likelihood of childbearing outside of marriage grew in step with rising economic inequality (McLanahan, 2004) before slightly declining in the last decade. Today, 70% of births to Black women and half of births to women with a high school diploma occur outside of marriage, compared to one third of births to White women and 14% to women with a Bachelor's degree. Most of the growth in this disparity is attributable to the share of births occurring in cohabiting unions among non-college-educated parents (Guzzo, 2021).

Thus, marriage appears to be both a product and a driver of social inequality. Would more marriage improve population well-being? Based on the evidence, my answer is no. To improve well-being for parents and children, we should look beyond marital status.

To explain, let's look first at why marriage appears to be uniquely advantageous. Much of our insight comes from looking at what happens within and between families. In most married-parent families, both parents are working (Ruggles, 2015) and have more income and assets to share with household members compared to single-parent or even cohabiting-parent families. Married parents also have more time and emotional energy to invest in children (Kalil et al., 2014; Umberson et al., 2013), greater kin support (Harknett & Knab, 2007), and more opportunities for social connections (Song, 2012). Marriages last longer than cohabiting relationships or spells of single parenthood, and the resulting stability is good for children (Cavanagh & Fomby, 2019). These factors are each associated with children's better outcomes in married-parent families.

But focusing on what is happening inside of married-parent families after they have formed distorts our view of how and why marriage matters. First, composition matters. As marriage has become more selective on educational attainment, it has also become more homogamous. The advantages to pooled resources are maximized in households where two college-educated partners each have the capacity for relatively high earnings, job security, time flexibility, and social and cultural capital to achieve their desired ends. Put another way, the positive outcomes married parents and children experience are not primarily a product of marriage itself, but rather, of the multiple forms of capital that these families accumulate and effectively deploy (Brown, 2010; McLanahan, 1985). To the extent that highly-resourced parents are overrepresented among married-parent families, the relative gains to that family arrangement may be overstated.

Second, and relatedly, not everyone benefits equally from marriage. From a sociological perspective, unequal returns to marriage largely reflect systemic inequalities. For example, both Black and White children in married-parent families have higher family income, lower poverty rates, and better outcomes compared to their same-race peers in single-parent families. And Black married couples are even more positively selected on college completion than are White couples (Bureau of Labor Statistics, 2023). Still, Black married-parent families have lower income and wealth and greater job insecurity compared to White families, and the gains to parents’ marriage are smaller for Black children compared to White children with regard to educational attainment, financial support for college, and timing of sexual initiation and childbearing (Amorim & Deming, 2022; Cross, 2020; Fomby et al., 2010). These weaker gains reflect racial differences experienced by parents and children in school quality, employment opportunity and discrimination, intergenerational transfers, home ownership, and contact with the criminal justice system (Addo et al., 2024; Cross et al., 2022; Fomby, 2021; Pager & Shepherd, 2008). Increasing marriage rates in population subgroups that benefit less from marriage without redressing other causes of inequality is unlikely to close racial and social class gaps in family resources or child outcomes.

Third, not everyone can find a marriageable partner. Evidence from marriage markets demonstrates that people seek partners who will be reliable providers. What that means varies across time and place, but when the pool of potential partners falls short on characteristics like employability, job stability, earnings growth, and good health, people are less likely to marry (Autor et al., 2019; Wilson, 1987) and more likely to favor self-reliance (Pedulla & Thébaud, 2015). Perhaps counterintuitively, the decision to eschew marriage speaks to its value. In their study of poor young single mothers in Philadelphia and Camden, New Jersey, Edin and Kefalas (2005) documented that women were holding out for all the best that marriage had to offer: security, stability, trust, and permanence. Silva (2013) similarly described working-class young adults seeking partners who would support their personal ambitions. For both groups, there was little value in entering a marriage that might fail.

Fourth, getting married is different from staying married. The divorce rate has recently declined for college-educated adults but increased for adults with less than a college education. Among adults married in 2005 to 2009, 14.2% of college-educated adults had ended their marriage through separation or divorce 10 years later. Among adults with less education, the share of marriages ending in divorce was 2 to 3 times higher (McErlean, 2021).

As this point illustrates, children experience a lot of family change. Brown and colleagues (2016) described the family structure pathways experienced from birth to age 12 among adolescents in 2006 to 2010. Only 3% of children always lived with a single parent. Even among those born to a single parent, only 17% stayed in that arrangement. Nearly all other such children saw their parent eventually marry or cohabit, either with the child's second parent or a new partner. And only a minority of adolescents—about 44%—lived continuously with their married parents to age 12.

Consequently, 84% of youth spent part of childhood in a married parent household and more than half spent time outside of marriage. That is, most children already experience marriage and something other than marriage. From this perspective, there is little to gain by encouraging more marriage. Nor is there strong evidence that we should encourage faltering marriages to endure for children's well-being; in fact, children whose parents remain in high-conflict marriages have significantly poorer outcomes compared to peers whose parents divorce (Amato et al., 1995).

Finally, studies focused on household characteristics overlook how the status of marriage benefits parents and children. Families are embedded in a network of institutions that favor married parents in employment, housing, education, and health and health care (Cherlin, 2009; Cott, 2000). More broadly, the federal government recognizes over 1100 statutory provisions under which benefits, rights, and privileges depend at least partially on marital status (Government Accountability Office, 2004). These include taxation, qualification as a caregiver under the Family and Medical Leave Act, and access to entitlement programs. There are no provisions for cohabiting partners or unpartnered adults under these federal statutes, many of which are premised on an outdated male breadwinner model of family organization (Letiecq, 2024; Smeeding, 1999). Yet the gains to the cultural, economic, and legal status of the married-parent family are largely invisible in research models that emphasize household-level characteristics.

To summarize, marriage is associated with multiple domains of parents’ and children's well-being. But marriage is a selectively accessed, variable, unstable, and exclusionary institution. To support parent and child well-being, we should look beyond marriage. A combination of the following strategies could be productive.

婚姻能改善父母和孩子的幸福吗？它肯定会出现这样的情况。在当代美国，与在其他家庭安排下长大的同龄人相比，与已婚父母一起长大的孩子在学校表现更好，身心健康，更容易开始和完成大学学业，并以更高的比例进入稳定就业。布朗， 2010 ）。已婚父母似乎也过得更好：他们报告说比单身或同居的父母更快乐、更健康、经济上更有保障（Waite， 1995 ）。

出于多种原因，婚姻与幸福之间的关系受到政策关注。首先，在美国，很多育儿都是在婚外进行的。 40% 的美国儿童是由没有伴侣或同居的父母所生（Guzzo， 2021 年；Osterman 等人， 2024 年，表 9），大约四分之一的 18 岁以下儿童与单亲父母（通常是他们的母亲）生活在一起（人口普查局， 2022 年，表 C3)。到 12 岁时，超过一半的美国儿童已经在已婚父母家庭之外度过了一段时间（Brown 等人， 2016 ）。

此外，在美国，单亲家庭中的儿童，特别是单亲家庭中的儿童，特别有可能陷入贫困，而儿童贫困与发展和成就受损密切相关（Duncan 等， 1998 ）。根据官方贫困标准，2022 年有子女的家庭中，37.2% 的女户主家庭和 18.3% 的男户主家庭处于贫困状态，而已婚夫妇家庭的这一比例仅为 6.9%（Shrider & Creamer， 2023 年，表 A） -2）。在 30 个同等国家中，美国单亲母亲对陷入相对贫困概率的平均边际效应排名第一（Brady 等， 2024 ）。

尽管大多数美国人表示他们想结婚（盖洛普， 2020 ），但已婚父母在很大程度上是按种族和社会阶层划分的。如今，60% 的亚裔成年人、54% 的白人成年人和 63% 受过大学教育的成年人已婚，而只有不到一半的黑人或西班牙裔成年人和受过高中教育的成年人（31%、45%、和 45%，分别是人口普查局， 2022 年，表 F2；朱利安， 2023 年）。与同龄未伴侣或同居的成年人相比，已婚成年人的婚姻收入也更高（Ludwig & Brüderl， 2018 ；Oppenheimer， 2003 ）。

美国婚姻形成和稳定性方面的社会模式差异并不新鲜：事实上，黛安·M·斯图尔特（Dianne M. Stewart， 2020 ）将黑人成年人在美国四个世纪以来遇到的稳定婚姻的深刻系统性障碍描述为“这个国家最隐蔽的障碍”。民权问题”（第 217 页）。但在婴儿潮末期，婚姻几乎在各个社会人口群体中普遍存在（Allred， 2018 ），非婚生育很少见。从那时起，婚外生育可能性的社会人口差异随着经济不平等的加剧而增长（McLanahan， 2004 ），然后在过去十年略有下降。如今，黑人女性 70% 的生育和拥有高中文凭的女性一半的生育是在婚外生育，而白人女性的生育比例是三分之一，拥有学士学位的女性的生育比例是 14%。这种差异的扩大主要归因于未受过大学教育的父母同居生育的比例（Guzzo， 2021 ）。

因此，婚姻似乎既是社会不平等的产物，也是社会不平等的驱动因素。更多的婚姻会改善人们的福祉吗？根据证据，我的答案是否定的。为了改善父母和孩子的福祉，我们的目光不应局限于婚姻状况。

为了解释这一点，我们首先来看看为什么婚姻似乎具有独特的优势。我们的大部分见解来自于对家庭内部和家庭之间发生的事情的观察。在大多数已婚家庭中，父母双方都有工作（Ruggles， 2015 ），与单亲甚至同居父母家庭相比，他们有更多的收入和资产与家庭成员分享。已婚父母也有更多的时间和情感精力投资于孩子（Kalil et al., 2014 ；Umberson et al., 2013 ）、更多的亲属支持（Harknett & Knab, 2007 ）以及更多的社会联系机会（Song, 2012） ）。婚姻比同居关系或单亲关系持续得更久，由此带来的稳定性对孩子有好处（Cavanagh & Fomby， 2019 ）。这些因素都与孩子在已婚家庭中的更好的结果有关。

但是，关注婚姻家庭形成后内部发生的事情会扭曲我们对婚姻如何以及为何如此重要的看法。首先，成分很重要。随着婚姻对教育程度越来越挑剔，婚姻也变得更加同质化。在两个受过大学教育的伴侣各自都有能力获得相对较高的收入、工作保障、时间灵活性以及社会和文化资本以实现其期望目标的家庭中，汇集资源的优势得到最大化。换句话说，已婚父母和孩子所经历的积极成果主要不是婚姻本身的产物，而是这些家庭积累和有效部署的多种形式资本的产物（Brown， 2010 ；McLanahan， 1985 ）。鉴于资源丰富的父母在已婚父母家庭中所占比例过高，这种家庭安排的相对收益可能被夸大了。

其次，与此相关的是，并不是每个人都能从婚姻中平等受益。从社会学的角度来看，婚姻回报的不平等很大程度上反映了系统性的不平等。例如，与单亲家庭中的同族同龄人相比，已婚家庭中的黑人和白人孩子的家庭收入更高，贫困率更低，并且成绩更好。黑人已婚夫妇在完成大学学业时比白人夫妇更受青睐（美国劳工统计局， 2023 年）。尽管如此，与白人家庭相比，黑人已婚家庭的收入和财富较低，工作不安全感也更大，而且在教育程度、大学经济支持和时间安排方面，与白人孩子相比，黑人孩子父母婚姻的收益也较小。性开始和生育的过程（Amorim & Deming, 2022 ；Cross, 2020 ；Fomby et al., 2010 ）。这些较弱的进步反映了父母和孩子在学校质量、就业机会和歧视、代际转移、住房所有权以及与刑事司法系统的联系等方面经历的种族差异（Addo 等人， 2024 年；Cross 等人， 2022 年；Fomby， 2021 ；寻呼机和牧羊人， 2008 ）。在不纠正其他不平等原因的情况下，提高从婚姻中受益较少的人口亚群体的结婚率不太可能缩小家庭资源或儿童结局方面的种族和社会阶层差距。

第三，并不是每个人都能找到适合结婚的伴侣。来自婚姻市场的证据表明，人们寻求的是可靠的伴侣。这意味着随时间和地点的不同而不同，但当潜在伴侣不具备就业能力、工作稳定性、收入增长和健康状况等特征时，人们结婚的可能性就会降低（Autor et al., 2019 ；Wilson, 1987 ） ）并且更有可能支持自力更生（Pedulla & Thébaud， 2015 ）。也许与直觉相反，回避婚姻的决定说明了婚姻的价值。 Edin 和 Kefalas（ 2005 ）在对费城和新泽西州卡姆登的贫困年轻单身母亲进行的研究中记录到，女性坚持婚姻所能提供的一切最好的东西：安全、稳定、信任和持久。 Silva ( 2013 ) 类似地描述了工薪阶层的年轻人寻找能够支持他们个人抱负的伴侣。对于这两个群体来说，缔结一段可能失败的婚姻没有什么价值。

第四，结婚与维持婚姻不同。最近，受过大学教育的成年人的离婚率有所下降，但受过大学教育以下的成年人的离婚率有所上升。在2005年至2009年结婚的成年人中，14.2%的受过大学教育的成年人在10年后通过分居或离婚结束了婚姻。在受教育程度较低的成年人中，以离婚告终的婚姻比例要高出 2 至 3 倍（McErlean， 2021 ）。

正如这一点所表明的，孩子们经历了很多家庭变化。 Brown 及其同事（ 2016 ）描述了 2006 年至 2010 年青少年从出生到 12 岁所经历的家庭结构路径。只有 3% 的儿童始终与单亲生活在一起。即使在单亲家庭出生的人中，也只有 17% 的人保持这种安排。几乎所有其他此类孩子都看到他们的父母最终结婚或同居，要么与孩子的第二个父母，要么与新伴侣。只有少数青少年（约 44%）一直与已婚父母一起生活到 12 岁。

因此，84% 的青少年在已婚父母家庭中度过了部分童年时光，超过一半的青少年在婚姻之外度过了一段时光。也就是说，大多数孩子已经经历过婚姻以及婚姻以外的事情。从这个角度来看，鼓励更多的婚姻并没有什么好处。也没有强有力的证据表明我们应该鼓励摇摇欲坠的婚姻为了孩子的福祉而持续下去。事实上，与父母离婚的同龄人相比，父母仍然处于高度冲突的婚姻中的孩子的结果要差得多（Amato et al., 1995 ）。

最后，关注家庭特征的研究忽视了婚姻状况如何使父母和孩子受益。家庭嵌入在就业、住房、教育、健康和保健方面有利于已婚父母的机构网络中（Cherlin， 2009 ；Cott， 2000 ）。更广泛地说，联邦政府承认超过 1100 条法律规定，根据这些规定，福利、权利和特权至少部分取决于婚姻状况（政府问责办公室， 2004 年）。其中包括税收、《家庭和医疗休假法》规定的护理人员资格以及享受权利计划的机会。这些联邦法规没有针对同居伴侣或无伴侣成年人的规定，其中许多法规都是以过时的男性养家糊口的家庭组织模式为前提的（Letiecq， 2024 ；Smeeding， 1999 ）。然而，在强调家庭层面特征的研究模型中，已婚父母家庭的文化、经济和法律地位的进步在很大程度上是看不见的。

总而言之，婚姻与父母和孩子的福祉的多个领域相关。但婚姻是一种有选择的、可变的、不稳定的、排他性的制度。为了支持父母和孩子的福祉，我们应该超越婚姻。以下策略的组合可能会富有成效。