The Yale Law Journal


Chosen Family, Care, and the Workplace

05 Nov 2021

abstract. Although federal law offers, at best, unpaid time off work to care for family members with medical needs, recently enacted state laws guarantee paid leave. This Essay argues the laws are groundbreaking in their inclusion of nonmarital partners, extended family, and other chosen family, and it proposes strategies for effective implementation.

The wedding section in the New York Times recently published a feature on “platonic marriages”—best friends who marry as a commitment to being life partners but remain open to dating outside their marriage.1 This practice simultaneously underscores and disrupts assumptions around the primacy of marriage as the principal marker of adult commitment. It also points toward a deeper truth: modern American families come in many different shapes and sizes.

Marriage rates are declining, while cohabitation rates are rising.2 About a quarter of children live with a single parent, a significant share lives with cohabiting parents, and blended family households are increasingly common.3 There has been a marked growth in multigenerational households and in extended family members serving as caregivers for children living apart from their parents.4 Many adults—including about forty percent of adults over the age of sixty-five—live alone, typically relying on a combination of family and friends for help and companionship.5 Indeed, fewer than twenty percent of households consist of what many would consider the paradigmatic “family”: a married couple living together with their children.6

Given this variety of lived experiences, it is unsurprising that employees often request time off work to care for the medical needs of nonmarital partners and other loved ones who are part of their extended or chosen families.7 Employers sometimes grant their requests. But until recently, most employees would not have had any legal right to take such leave.8 As a threshold matter, many workers do not receive sick days or medical leave at all.9 But even in jurisdictions that mandate leave policies and specify that workers can take time off to care for their family members, the scope of eligible family is sometimes narrowly drawn to include only immediate family, defined as close blood relatives or a legal spouse.10

Recently enacted laws, mostly at the state and local level, take a dramatically different approach.11 Several guarantee leave to care for intimate partners without requiring legal formalization of the relationship and cover a broad range of extended family.12 Some laws go further and apply to any individual who has a relationship with the employee that is “like” or “equivalent to” a family relationship.13 Still others employ a functional approach that simply asks whether a sick individual depends on the employee for care.14

These new provisions are transformative: They give employees the autonomy to define their own concept of families. However, such flexibility can pose administrative challenges.15 The laws will only achieve their purpose if both the public and private personnel implementing them understand the broad scope of coverage and ensure that employees whose families depart from traditional norms are protected from workplace discrimination.

This Essay provides the first detailed analysis of inclusive family definitions found in employment-leave laws. Part I uses census data and social-science research to illustrate the variety of families and support networks in contemporary America. It also employs an intersectional lens to emphasize why this broader conception of eligible family is particularly important for people of color and the LGBTQ+ community. Part II describes recently enacted laws that provide workers paid time off to help family members with their medical needs, and Part III details how the laws, and proposed federal legislation, define an eligible “family.” Finally, Part IV proposes strategies for effectively administering the family-care provisions in these new employment-leave laws.

A final introductory note: As this essay is being finalized, Congress is actively considering legislation that would guarantee paid family and medical leave for most workers.16 The draft language recommended by the relevant committee in the House adopts a flexible definition of family that would include extended and chosen family, and it emphasizes the need to ensure that policies are administered without discrimination.17 Federal legislation in this area is long overdue. However, a silver lining of Congress's failure to act sooner is that it can build on models enacted at the state level that appropriately recognize the variety of modern families.

I. the diversity of modern families

At various points in our lives, every one of us is sick or injured and needs care.18 When this happens, we typically turn to family, or friends who feel like family, for help.19 Young children and older adults are particularly likely to need support, but care needs can arise at any age. A victim in a car crash may require emergency surgery, and a cancer diagnosis can initiate a lengthy process of chemotherapy and radiation. Some people have chronic health conditions or disabilities that require care throughout their lives. While care needs are perennial, the range of “family” who might be called upon to meet those needs has changed dramatically in recent decades. Policies premised on the expectation that care will be provided within the traditional nuclear family are out of step with modern America.

First, the assumption that adults will have a spouse to support them is often incorrect. Approximately three in ten adults identify as single, and many are not interested in dating.20 This includes adults of all ages, from young adults,21 who frequently live with their parents22 or roommates,23 to a sizeable share of senior citizens.24 Additionally, a wide spectrum of adult intimate relationships exists. Many adults live with an intimate partner without marrying.25 Indeed, more adults under age forty-four have cohabited than been married.26 Adults may also “live apart together,” identifying as part of a committed relationship but living separately from their
partner.27 Consensual nonmonogamous relationships, including stable polyamorous relationships, are also becoming more prevalent.28

Nor is it realistic to expect that only a “parent” would care for a child. Fewer than half of all children live with two parents in their first marriage.29 This is the result of a seismic growth in the nonmarital birth rate30 and a relatively high divorce rate.31 It has become much more common for unmarried adults to live together with children, including children from prior relationships; often in this scenario, only one of the adults in the household is a legally recognized parent.32 Additionally, solo parents frequently live with their own parents.33 Family structures are also often quite fluid. One study found that almost one in three children under the age of six will experience a major change in their household structure, such as divorce, separation, marriage, cohabitation, or death.34

At any given time, a significant group of children—about 2.7 million, or around four percent of all children in the United States—are being cared for by extended family members or close family friends without a parent present.35 Such “kinship care” typically takes the form of informal arrangements that do not involve the state foster-care system.36 For example, a grandparent or an aunt might step in while a parent participates in a drug-rehabilitation program or serves a military deployment. That said, child-welfare agencies also routinely look to relatives to provide foster care for children who are removed from their parent’s home because of abuse or neglect.37

Caregivers for the elderly vary widely as well. With baby boomers reaching retirement age,38 Americans over age the age of sixty-five make up a rapidly growing share of the overall population.39 Over eighty percent of senior citizens live on their own or only with a spouse or partner.40 They often turn to their middle-aged children for support, many of whom are also raising their own children (and thus belong to the “sandwich” generation).41 It is also becoming increasingly common for three or more generations of a family to live together.42 Older adults also form nontraditional networks of support. This can include sharing space in one’s house with a nonrelative in exchange for caregiving or transportation, living in independent housing with shared communal spaces, relying on help from neighbors in a naturally occurring retirement community, or joining a communal “village” that relies on staff and volunteers to assist older residents.43

There is significant class-based, as well as racial and ethnic, variation in family structures. The marriage rate is correlated with income44 and education level;45 the cohabitation rate,46 nonmarital birthrate,47 divorce rate,48 and the likelihood of having children with multiple partners,49 by contrast, are inversely related to education and income. There are considerable differences among racial and ethnic groups along many of these vectors as well.50 Additionally, people of color are more likely than white people to live in households that include extended families.51 Even when living separately, cultural norms of multigenerational caregiving—both eldercare and childcare—are prevalent in Black and many immigrant communities.52 Thus, policies premised on the assumption that caregiving needs will be met exclusively within the traditional nuclear family can compound disadvantage for already marginalized groups.

The concept of "chosen family"—close friends who are not related by blood or marriage—was first developed in the LGBTQ+ community,53 and a flexible and inclusive definition of family is particularly important for its members. Many older members were rejected by their birth families when they came out; even if their blood relatives were accepting, it is common for members of the LGBTQ+ community to move to more tolerant cities and form networks of informal kin.54 These connections remain crucial. LGBTQ+ baby boomers are much more likely to be childless and living without a partner than non-LGBTQ+ baby boomers,55 and they often turn to chosen family to provide care and support.56 Although stigma is lessening in many communities, LGBTQ+ youth are still disproportionately represented in both the foster-care and homeless-youth populations.57 Various facets of identity also intersect, multiplying disadvantage. For example, LGBTQ+ youth of color are disproportionately likely to be homeless and may ultimately be more likely to rely on informal networks of chosen family.58

In short, modern American families are extraordinarily varied, and their configurations change over time. Thus, to be effective, family-care policies must be flexible enough to accommodate a wide range of caregiving relationships. Moreover, when a need for care arises, most people consider not just the markers of “closeness” that indicate family, but also the seriousness and duration of the medical need, geographic proximity, and the other demands on potential caregivers. That said, women are disproportionately more likely to provide care for both younger and older family members than men, and support for family caregivers is essential to promote sex equality more generally.59

Even individuals who seem to have a relatively traditional family structure may turn to extended or chosen family for help with medical needs. For example, a widow in her seventies who lives far from her adult children might ask a cousin or a friend to accompany her to an out-patient procedure at a hospital. But if she has more serious surgery, with a longer recuperation period, one or more of her adult children might travel to help her. The family and household patterns discussed above, however, indicate that people of color and members of the LGBTQ+ community are particularly likely to look to an unmarried partner or a member of their extended family or chosen family for help. The next two Parts discuss legislative developments that make it much easier for family—expansively defined—to provide such care.

II. leave for medical caregiving

The COVID-19 pandemic dramatically highlighted a preexisting problem: many American workers risk being fired if they take time off work to care for their own medical needs or those of their loved ones. Fortunately, recently enacted laws, mostly at the state and local level, offer workers new important protections. This Part briefly discusses the general contours of discretionary employer policies and relevant legislation.

A. Employer Policies

American employers have wide latitude in structuring workplace absence policies. Outside the legislative mandates discussed in the following Section, employers generally have the discretion to offer employees very little or no time off for health conditions.60 To the extent employer policies do address these needs, they usually adopt a two-tiered approach. Short-term absences are typically covered by “sick day” policies.61 Longer-term absences for more serious medical needs are generally covered by policies known as “family and medical leave,” named after the federal legislation that sometimes requires such leave.62 In designing such policies, employers balance the costs associated with providing leave with the competitive advantage such policies can offer in attracting, supporting, and retaining employees.63

There are significant disparities in access to paid time off under employers' discretionary policies. Most full-time employees receive a week or two of paid sick days each year.64 However, fewer than half of part-time employees receive any paid sick leave.65 Additionally, employers can choose whether to allow employees to use sick days to care for family members or only their own medical needs, and how broadly or narrowly to define eligible families.

Access to paid leave for longer-term absences to care for family members with more serious health conditions is rare, even for full-time workers.66 Just one in four full-time workers receive this benefit, with higher earners being more likely to receive it; virtually no part-time workers receive paid family or medical leave.67 Larger employers are required to offer some employees unpaid leave for such needs;68 this sets a soft norm followed by many businesses.69 Thus, most employees technically have a right to take unpaid leave for family health conditions.70 However, few workers can afford to go very long without a pay check. This is especially true for workers of color, given stark racial disparities in wealth.71

In the absence of a formal leave policy, missing even a few days of work can lead to job loss. Low-wage jobs, in particular, frequently operate under draconian absence policies: workers are penalized with a “point” for any unplanned absence, regardless of the reason, and even a few points can be grounds for discipline or termination.72 Accordingly, when family members need care for injuries or illnesses, many workers, especially low-wage and part-time workers, are left with no good options. If they stay home to provide help, they risk being fired—a potentially devastating outcome if they also face significant medical expenses.

Changes in workplace policies in the wake of COVID-19 may alter this dynamic, but in ways that benefit certain workers while exacerbating preexisting disparities. During the pandemic, large sectors of the economy shifted online, and many employers have announced that they do not expect office workers to ever return full time to in-person offices.73 If remote work remains common, it will help many employees who need to care for family members: They can provide support without necessarily taking a day off work.74 But white-collar workers, particularly highly paid skilled workers, are more likely to be able to take advantage of this option because they are more likely to retain discretion over where they work.75 By contrast, workers in jobs that require in-person presence—many of which are low-wage service jobs that are disproportionately filled by women of color76—will generally need to take time off work every time a child stays home from school or an elderly parent needs assistance getting to and from a medical appointment. Low-wage workers were already less likely than higher paid workers to receive paid leave; they will now be more likely to need to miss work, and potentially face job loss, when family members are sick.

B. Paid Sick Day and Paid Family Leave Laws

Employers' discretionary policies are clearly inadequate, and existing federal law provides, at best, unpaid leave for some workers. However, a rapidly growing number of states have enacted laws guaranteeing paid sick days and paid family leave, and there is new momentum for comparable federal legislation.

Although federal law currently does not require employers to provide paid sick days, as of October 2021, sixteen states and the District of Columbia, as well as many large cities, guarantee this basic workplace right.77 These reforms happened exceptionally quickly, with virtually all of the laws being passed in the last ten years.78 Most of them are structured so that both part-time and full-time employees accrue paid time off based on hours worked.79 Typically, covered employers must allow workers to accrue at least forty hours of paid time off annually; larger employers are sometimes required to allow workers to accrue sixty-four or seventy-two hours.80 Workers are paid directly by their employers (in contrast to the state-sponsored paid family- and medical-leave programs, discussed below), and they receive their full regular wages or salaries.

Sick-day laws generally provide time off for a wide range of health conditions and medical needs, as well as preventive care such as routine visits to a doctor. Most also cover the time off an employee might need to take in response to domestic violence, sexual assault, or stalking.81 Most importantly for this Essay, all of the state sick-day laws currently enacted cover not only the employee’s own needs, but also time off to care for a family member with medical needs.82 That said, the scope of eligible “family” differs across each law, as discussed in Part III.

Federal and state legislation sometimes also guarantees job-protected leave for longer-term absences for more serious medical conditions. Federal law currently provides only a right to unpaid leave, under the Family and Medical Leave Act (FMLA).83 The FMLA allows some workers to take up to twelve weeks off for their own or a family member’s “serious health condition.”84 The FMLA only applies to workplaces with at least fifty employees, and within those workplaces, covers only workers who have been employed for at least one year at close to a full-time schedule.85 Collectively, these requirements exclude about forty percent of the private workforce.86 However, some states have enacted laws that, like the FMLA, mandate unpaid leave but apply to smaller employers,87 and, as noted above, many employers opt to provide unpaid leave, even if not required to do so.88 However, few workers can afford to take an extended unpaid leave.

As of October 2021, nine states and the District of Columbia have passed laws that guarantee workers paid time off for more extended medical-related absences.89 Seven of these laws have been passed since 2016, and the new laws are more generous than the older ones, suggesting growing support.90 They typically provide between two-thirds and full wages or salary, up to a cap set around the median wage.91 This means that low-wage workers receive a larger proportion of their regular income than higher-paid workers.92 The length of leave for family care varies from four to twelve weeks, with most allowing twelve weeks.93

In stark contrast to the FMLA, these laws cover employers of all sizes, and many allow independent contractors—a category that includes many workers in the gig economy—to opt in.94 States use an insurance model, funded by a very small payroll tax, to provide the benefits; this structure spreads costs so that small employers are not unduly burdened.95 Eligible workers receive the benefits from the state, rather than from their individual employers.96 Many of the laws guarantee both benefits and job-protected leave.97 Some, however, simply provide benefits without ensuring that an employee’s job will be kept open.98

In terms of eligible medical conditions, most of the state laws directly track the federal FMLA in allowing workers to take up to twelve weeks of leave for their own or a family member’s “serious health condition.”99 But the state laws depart in important ways from the FMLA’s definition of eligible “family”for whom a worker can provide care. As discussed in Part III below, many adopt a much more expansive and flexible standard.

In response to COVID-19, the federal government passed temporary legislation that expanded both paid sick leave and paid family leave to meet caregiving needs related to the pandemic.100 Although the mandatory leave under these provisions expired at the end of December 2020, the pandemic has heightened calls for a more comprehensive and permanent federal solution. In April 2021, President Biden presented a sweeping bundle of policy proposals designed to better support American families, including calls for Congress to enact both paid sick leave and paid family and medical leave.101 Bills that would provide this leave are pending as this Essay is published.102

III. recognition of nonmarital partners, extended family, and other chosen family

The state and federal legislation in this area shows a gradual expansion in the scope of eligible “family” for whom a worker can take time off to provide care. The most recently enacted laws are truly transformative. They explicitly give workers autonomy to define their own conception of family. The leave laws are thus not only important for the essential benefits that they provide, but also because they serve as models for inclusive family definitions in other contexts.

Advocates trace recognition of chosen family in leave policies back to federal regulations promulgated in the Vietnam War era.103 Those regulations allowed federal workers to take time off for the funeral of an “immediate relative,” which was defined as “[a]ny individual related by blood or affinity whose close association with the deceased was such as to have been the equivalent of a family relationship.”104 This policy was later expanded to permit federal employees to take sick leave for the same expansive range of family,105 and it was endorsed by Congress in subsequent legislation governing federal workers.106

However, the first federal legislation requiring private-sector employers to provide medical leave—the FMLA—adopts a much narrower conception of eligible family: only an employee’s child, parent, or spouse.107 This excludes even close blood relatives such as grandparents, grandchildren, and siblings, as well as more distant relatives. Additionally, although the “parent” and “child” definitions include some functional relationships,108 “spouse” only covers legal marriages.109 The law’s implementing regulations also explicitly state that “parent” does not include “parents in law.”110 Finally, children are only covered when they are under the age of eighteen, unless they are “incapable of self-care because of a mental or physical disability.”111

All of the state paid family and medical-leave laws and most of the paid sick-day laws include broader definitions of “family” than the definition in the FMLA.112 That said, some of the laws expand the scope only minimally and still limit coverage to a specific list of relations. These laws generally cover all of the relationships addressed in the FMLA and add formal legal domestic partnerships, grandparents, grandchildren, and parents of a spouse or domestic partner113—or include all of those relationships and also siblings.114 Most of the laws define “parent” and “child” similarly to the FMLA, but do not incorporate the age limit that excludes care for nondisabled adult children.115

Recent laws expand the boundaries more dramatically. States have adopted four different approaches, with some using two or more. The first approach lists specific blood relations and legally recognized relationships such as marriage, but also explicitly covers domestic partners without requiring any kind of formal registration.116 These laws typically provide a multifactor list of attributes (including shared household, shared budgeting, shared children, and length of personal relationship) that are used to determine whether a requisite closeness exists; however, the laws specify that not all factors are required.117 As discussed in Part I, this approach is significant because many Americans spend a considerable portion of their adult life in a committed romantic relationship—either cohabiting or living separately—without marrying or taking other steps to formalize a legal relationship. The definition could also potentially cover long-term, close platonic friendships that include mutual dependence and support.

The second approach lists specific covered relationships but adds a catch-all phrase: Any individual related to the employee “by blood or affinity” whose “close association” with the employee is equivalent to the enumerated family relationships.118 The pending federal bills include similar phrasing.119 The language obviously covers extended family. It may be particularly useful for kinship caregivers who might otherwise be ineligible to take time off for the medical needs associated with the children in their care.120 This approach also arguably covers nonmarital partners and other chosen family, depending on the meaning ascribed to the word “affinity.”121 Notably, the wording is very similar to the language used in the earlier federal legislation governing benefits for federal employees.122 The U.S. Office of Personnel Management, the federal agency charged with implementing that statute, indicated that the “blood or affinity” clause should be interpreted to include individuals who have close family-like relationships, whether or not those relationships rest on blood or formal legal ties.123 Given that states often follow federal interpretations for borrowed language, they might apply the same interpretation.124

The third model covers close relationships that are “like” family without requiring a connection by “blood or affinity” at all. For example, Colorado’s paid family-leave law specifies that a covered individual may take leave to care for “any other individual with whom the covered individual has a significant personal bond that is or is like a family relationship, regardless of biological or legal relationship.”125 New Jersey’s paid family-leave law was amended in 2019 to extend coverage to “any individual . . . that the employee shows to have a close association with the employee which is the equivalent of a family relationship.”126 New Jersey’s127 and New Mexico’s128 paid sick-day laws include similar language. This approach removes any doubt as to whether nonmarital partners and other chosen family is covered. It also has an important signaling effect, affirming the reality of such bonds.

Finally, a few recently enacted statutes employ a functional approach, listing specific covered relationships and then adding any additional people who depend on an employee for care. For example, Colorado’s paid sick-day law allows an employee to take leave to care for any “person for whom the employee is responsible for providing or arranging health- or safety-related care,” as well as immediate family.129 Rhode Island’s paid sick-day law includes similar language and also explicitly covers any member of the employee’s household.130 The temporary paid sick leave Congress passed in response to COVID-19 took a similar approach.131

The inclusive family definitions in the leave laws bear a superficial resemblance to earlier policies regarding domestic-partnership benefits, but there are important differences. State and local governments and private businesses began formally recognizing “domestic partnerships” in the 1980s and 1990s, primarily as a mechanism to make benefits and privileges that were premised on marriage available to same-sex couples.132 Prime among those benefits was covering a partner under employer-provided health insurance. To qualify as domestic partners, couples typically needed to show that they were in a long-term, committed relationship and lived in a shared household with intermingled finances.133 Recognition thus rested on a very traditional conception of marriage, offering same-sex couples less autonomy than heterosexual married couples.134 By contrast, the new family provisions offer employees the freedom to define and determine their own conception of what it means to be “family,” without requiring a shared household or financial dependency.

IV. administrating family autonomy

Expanding family leave to support the varied structures of modern families meets a real and pressing need. But because effective administration generally relies on standards that can be applied consistently by multiple actors, implementation of the new laws may pose some challenges. In the family-leave context, traditional definitions of family—individuals related by blood or marriage—rest on well-established legal concepts with (relatively) clear criteria. By contrast, a functional assessment of whether a person has a “close association” or a “significant” personal bond that is “like” a family relationship is necessarily more subjective. As Katherine Baker points out in a related context, this may raise privacy concerns and risks reinscribing structural inequities.135 While a full explication of this issue is beyond the scope of this Essay, this Part offers some preliminary thoughts regarding potential hurdles to effective implementation and proposes some solutions.

The expansive conception of family included in these leave laws responds to the reality that workers routinely seek to provide care for their extended and chosen family—and that legislatures believe they should be able to do so. To achieve this mandate, both public and private personnel administering leave laws must understand and properly apply the provisions. This task has at least three facets: (1) developing regulations, policies, and claim forms that appropriately represent the broad scope of statutory language; (2) ensuring that applicants’ privacy is respected and that applicants do not face workplace discrimination or harassment; and (3) fairly assessing whether the relation is “like family” or meets the relevant domestic-partnership standard when the legitimacy of a claimed connection is questioned. If administrators are unduly rigid, or substitute their own judgments for the worker’s judgment of what constitutes a “real” relationship, the underlying purpose of the laws will not be realized.

A. Developing Regulations, Policies, and Claim Forms

Government agencies in the relevant jurisdictions are generally charged with developing rules, regulations, and other informational materials.136 Once in effect, the employers independently administer paid sick-day laws. The paid family-leave laws are a hybrid; benefits are administered by the relevant state agency, but employers manage the leave portion by keeping the job open or placing the employee in a comparable position when she is ready to return.137

For the laws to achieve their objective, potential beneficiaries need to know about new protections and everyone tasked with implementing them needs to understand their scope. On this front, initial signs are promising. Publicity and informational materials developed by state agencies in relevant jurisdictions prominently include the relevant language regarding informal domestic partnerships138 and extended or chosen
family139 in describing the scope of protections.140 Agencies developing formal regulations implementing the provisions should maintain the flexible understanding of family that the relevant statutory language signals. It also might be helpful to provide examples of different kinds of relationships that might be covered.141

As these laws are phased in, employers will need to appropriately update their own employment policies regarding sick days and family leave. It also will be essential for employers to train supervisors and human-resources personnel on the new provisions, making clear that the scope of family coverage is much broader under the new state laws than it is under the FMLA. Institutional leaders in the human resources field, such as the Society for Human Resources Management (SHRM), and legal counsel can assist with this by providing model policy language and conducting trainings.

Relevant forms should also signal that nonmarital partners, extended family and other chosen family may be covered. The state administers claims for family leave. Most state forms simply ask applicants to check a box indicating the connection to the person for whom they will be caring, such as a spouse or grandparent.142 The forms sometimes require the worker to affirm or assert that the information provided is accurate, but they generally do not require any other proof of the family relationship.143 The federal agency charged with administering leave benefits for federal employees takes a similar approach.144 Claim forms for the newer laws should simply add an additional check box with the relevant statutory standard for partners, extended, or chosen family.

Some administrators might believe that a worker should be required to provide more rigorous documentation of a close relationship to claim benefits to care for family who are not part of an employee's nuclear family. While it is possible that expanding the scope of eligible family could open the door to misuse, initial implementation of leave laws with flexible standards suggests this has not been a significant problem.145 The structure of the laws includes an element of built-in deterrence: if an employee exhausts her leave allotment caring for a casual acquaintance or distant family member, she risks being fired if she subsequently needs to take time off for her own medical needs or to care for a closer family member.146 Additionally, workers who lie on a form may be subject to discipline by their employers.147 And for family leave, the family member receiving care typically must provide medical certification to establish that her health needs meet the “serious health condition” threshold,148 which further reduces the risk of frivolous or fraudulent claims.

Workers should not be required to pre-identify the full universe of persons who might be considered “family” for leave purposes.149 As Part I emphasized, both the nature of the medical need and when it arises will affect who might provide care. Nor should employers or state administrators expect that any such designations would necessarily be reciprocal. That is, an employee might legitimately ask to take leave to care for a cousin or an aunt, even if that employee would turn to her spouse if she herself needed care.

B. Addressing Privacy and Discrimination Concerns

Employers should also take steps to ensure that employees do not face discrimination, harassment, or retaliation at work for taking leave. In surveys assessing the efficacy of the FMLA, workers routinely indicate that they fear negative repercussions for taking time off work, even though such retaliation is illegal.150 This risk may be heightened for workers seeking leave to care for chosen or extended family under the new laws. Supervisors or coworkers might not only resent any inconvenience that comes from covering the leave itself, but also feel that the time off was unwarranted because the care recipient was not a “real” family member. As Part I details, marginalized communities are more likely to rely on care networks outside the nuclear family, meaning such resentment could be intertwined with racism, sexism, or animus based on sexual orientation or gender identity.151 Some kinds of chosen family are further stigmatized, such as choosing to be in a polyamorous relationship or serving as a kinship caregiver because of a parent’s drug abuse or incarceration.

I have explored elsewhere how employees can face adverse actions at work for the exercise of “intimate liberties” that merit constitutional protection, such as nonmarital pregnancy or cohabitation.152 Although the state may not punish individuals for engaging in nonmarital intimacy, courts often interpret existing antidiscrimination law to offer no protection from workplace discrimination on these grounds.153 I have argued that even under more general antidiscrimination laws, these cases frequently rely on an untenable distinction between status and conduct.154 The new paid-leave laws are more explicit that some kinds of discrimination based on family relationships would be unlawful.155 Nonetheless, illegal discrimination of all kinds occurs routinely, and most employees (or former employees) never challenge unfair treatment in

Rather than relying on after-the-fact litigation, effective implementation will depend on employers protecting worker privacy in the first place. Employers generally are required to keep health information separate from the rest of an employee’s personnel records.157 The same rules should apply to information or supporting documentation regarding family status. Additionally, as noted above, employers and outside organizations such as SHRM or management-side law firms should provide training to personnel involved in administering the programs. This training should explain not only the scope of the laws, but also why flexible understandings of family are essential. That is, training should provide a snapshot of the many varied forms of the modern family and make clear that it is reasonable and appropriate for workers to provide care for intimate partners, extended family, and other chosen family.

Training should also reaffirm that it is illegal to discriminate or retaliate against an employee for requesting or taking leave. In the FMLA context, unlawful retaliation includes discouraging employees from requesting or using leave, denying future professional opportunities based on an employee’s having requested or used leave, or counting FMLA leave under “no fault” attendance polices.158 Similar standards would likely apply under the state or local laws providing more extensive paid leave.

C. Assessing Legitimacy of Claims

Finally, if questions arise as to the legitimacy of the claimed connection, administrators—and ultimately courts—might be required to assess what kind of personal relationships are “like” family.159 This is a comparative process that necessarily calls for identifying the key markers of what constitutes a “family” relationship and evaluating factual support for the claimed personal connection.160

As a threshold matter, it is important to distinguish the recognition of family for employment-leave purposes from other contexts in which functional family relationships have been considered. Many such cases arise when one person alleges that a family-like relationship hasexisted and another person in the putative “family” challenges the claim, typically because they have since separated. For example, an adult seeking to be recognized as a de facto parent, and thus eligible for legal and physical custody of a child, will seek to prove, often over the objections of an already-recognized parent, that she has played a parental role for a child.161 Or a person who was in a long-term nonmarital relationship that has since ended may allege that during the relationship she minimized her financial earnings to provide support to her partner, and thus has a claim to a share of her partner’s property.162 In both of these contexts, even if one agrees as a matter of law that these relationships can be akin to other forms of “family,” it is generally appropriate for courts to carefully scrutinize these claims’ factual basis. This is both because the stakes are high—recognition of one person’s claim diminishes the familial rights or autonomy of another person—and because discord or animosity between the parties may heighten the risk of false claims.163 Moreover, in these contexts, the alleged family-like connection is often contested by one of the parties.

Family-leave policies will not typically pit one family member “against” another. Rather, recognizing family in this context imposes a burden on a third party—the employer or the state. Thus, the concern could be that the relevant parties might collude in falsely asserting a family relationship exists. Again, however, other contexts that impose costs on third parties often concern high-cost open-ended benefits, such as health insurance for a partner164 or a path to citizenship.165 The sick- and family-leave laws, by contrast, provide a relatively short time off work. Moreover, as noted above, they include a built-in element of deterrence against abuse because an employee who exhausts her leave on a spurious claim would have no recourse if she or a close family member subsequently needed care.166 This suggests administrators should err on the side of assuming legitimacy. This will help to ensure that the laws achieve their objective: providing immediate support to workers in times of need.

Nonetheless, administrators may sometimes need to substantively analyze whether a claimed relationship is “like” family. In answering that question, it seems relatively obvious that a romantic monogamous relationship between two adults (of whatever sex) would meet a requisite threshold. Likewise, a household that is structurally similar to a nuclear family, such as an unmarried couple and children of either member of the couple, or an aunt raising a niece, should clearly qualify. But what about a polyamorous family, who may or may not share a household?167 Or a collaborative childrearing co-op of single moms?168 A group of elderly persons who live together in a cohousing community?169 Close friends who have never lived together, but count on each other for primary emotional or financial support, rather than on romantic partners?170

To achieve the underlying purpose of the laws, administrators and courts should recognize that all of these arrangements could, at least sometimes, be appropriately characterized as “like” a family. Though they depart from the traditional nuclear family, each of these relationships might give rise to close personal bonds and each might describe persons who would support each other with medical needs. By passing laws with expansive and flexible language, legislatures have signaled their understanding that family and family-like relationships vary enormously, and that they exist in the absence of blood ties, legal ties, or other tangible markers such as shared households. However, there is a risk that administrators or courts would apply their own moral standards to deem some relationships as categorically improper. Thus, an administrator might recognize a collaborative childrearing co-op but not a polyamorous family, even if both are essentially a collection of adults sharing in the rearing of children. This kind of selective recognition would be inappropriate, just as is generally inappropriate for employers to second-guess an employee’s religious
beliefs.171 Barring special considerations that might apply if recognition of a family relationship would be in tension with an employer's religious beliefs,172 the laws should be interpreted to provide autonomy to the workers to identify their own chosen family.

A deeper critique of the new laws might ask why recognition should be premised on family at all.173 Rather than requiring a worker to assert that a nonmarital partner, distant relative, or friend is “like” close family, laws could empower workers to simply take leave for anyone they would like to help. As noted, some laws gesture in this direction by relying on a functional assessment of whether the individual depends on the worker for care, rather than a qualitative assessment of the family-like nature of the relationship.174 Ultimately, however, it seems likely that lawmakers, and probably much of the public, understands “family” to denote a particularly special kind of love, commitment, and support that employers can and should accommodate. That said, by recognizing the significance of extended- and chosen-family bonds—and ensuring that workers can take time away from work to care for those loved ones—the laws rework and expand the conception of family as far more capacious than traditionally recognized.


Under American workplace policies, many workers can be fired if they stay home to care for loved ones who are sick or injured. Fortunately, a rapidly growing number of states have passed legislation that ensures workers can take paid time off as necessary for both short-term and long-term medical conditions. These laws are groundbreaking in their expansive definitions of family. Many include intimate partners, regardless of legal status, and other extended and family. This flexibility is crucial because only a sliver of modern American households consist of a traditional nuclear family.

These laws’ enactment is an important step forward. However, effective administration will be essential to achieve their promise. State agencies charged with implementing the laws should develop regulations and claim procedures that preserve the flexibility that these statutes’ language signals. Supervisors and human-resources departments will need to take steps to ensure that employees who seek leave are protected from adverse actions at work, and particularly that those whose families depart from traditional norms do not face workplace discrimination. And finally, when assessing whether or not a claimed connection is “like” family, administrators and courts should recognize that the laws give workers the autonomy to define their own conception of family, and that this may depart considerably from the paradigm of a married couple raising their shared children. Hopefully, with time and effective administration, the new laws will not only allow workers to care for loved ones without fearing losing a job, but also serve as a model more generally for recognizing the varied and often fluid nature of the modern American family.

Deborah A. Widiss is Professor of Law, Associate Dean for Research and Faculty Affairs, and Ira C. Batman Faculty Fellow, Indiana University Maurer School of Law. I am grateful to Rachel Arnow-Richman, Naomi Cahn, Doron Dorfman, Tristin Green, Orly Lobel, Kaiponanea Matsumura, Nicole Buonocore Porter, Leticia Saucedo, and Michelle Travis for insightful suggestions on earlier drafts. I am also grateful for feedback I received when presenting this project at the 2021 Law and Society Association conference, the 2021 Family Scholars and Teachers Conference, the 2021 Roundtable on Nonmarriage, and the 2021 Colloquium on Scholarship in Employment and Labor Law. Many thanks, too, to the editors of the Yale Law Journal, particularly Raleigh Cavero, for their conscientious work finalizing this Essay for publication.