Introduction

The United States is one of the only countries in the world that does not have a national Paid Family Leave (PFL) program. The 1993 Family and Medical Leave Act (FMLA) mandated job protection for workers for leave to bond with a new child, recover from their own illness, or provide care for a seriously ill family member, but did not mandate paid leave. As there is no federal policy, an increasing number of states have passed PFL laws. For instance, since California implemented the nation’s first PFL program in July 2004, nine more programs have been implemented, and four states have passed PFL legislation to be enacted between 2024 and 2026 (State Paid Family & Medical Leave Insurance Laws, 2023). At the same time, support for a national PFL program has gained substantial momentum in the past decade, though no federal law has been passed to date.Footnote 1 That said, PFL is not the only public support available for caregiving. The safety net may also support low-wage workers’ caregiving-related work interruptions, such as means-tested cash and near-cash benefits, tax credits for children and dependents, and state-legislated paid sick days (PSD). In this paper, we refer to these programs collectively as the social safety net.

Passage of state PFL policies is concentrated in the northeast and west coast regions of the United States, which have been recognized for having more robust social safety nets compared to other regions (Cawthorne Gaines et al., 2021; State Safety Net Interactive, 2023). Given this, it follows that individuals who reside in PFL states may have access to more generous safety net provisions that support caregiving, in addition to PFL. Therefore, passage of PFL at either the state or federal level would likely bolster safety net supports for caregiving interruptions from work in non-PFL states because of comparatively weaker safety nets in these states. To these ends, in this paper we examine the intersection of variations in state sponsored PFL and social safety net provisions between PFL and it non-PFL states to better understand the distribution of such resources between them. We also examine variation in social safety net programs across PFL states to reveal differences in both PFL and safety net provisions in existing PFL states. Our results provide guidance to policymakers, researchers, practitioners, and advocates who aim to support the simultaneous demands of caregiving and work specifically for low-wage workers with child and adult caregiving responsibilities.

Paid Family Leave in the United States

Research on PFL finds that its availability is associated with increased leave-taking (Baum & Ruhm, 2016; Houser & Vartanian, 2012; Rossin-Slater et al., 2013), positive child and maternal health outcomes (Huang & Yang, 2015; Pihl & Basso, 2019; Rossin-Slater & Uniat, 2019), and improved work and earnings (Baum & Ruhm, 2016; Das & Polachek, 2015).Footnote 2 While the evidence of state-sponsored PFL is mainly positive, program rules differ substantially across PFL states. For example, the maximum number of weeks allowed for PFL ranges from six weeks in Rhode Island to up to 12 weeks the majority of other PFL states. Wage replacement rates are similarly diverse, ranging from 60% in Rhode Island to 100% in Oregon. Table 1 summarizes PFL policy rules for all PFL states (implemented and pending) and our ascribed generosity levels. Table 2 illustrates the construction of our PFL generosity index.

Table 1 Characteristics of PFL policies
Full size table

In general, newer state sponsored PFL programs are more generous than earlier programs, measured by maximum leave lengths, wage replacement rates, covered family members, and job protection. State PFL programs trending toward more generous provisions is a major win for families, practitioners, and advocates. However, while more generous PFL programs will benefit many, even generous PFL provisions may fall short in adequately supporting low-wage workers’ time off for caregiving and transitions back to work. As such, families may turn to the safety net during these periods. Given this, it is important for practitioners, policymakers, and researchers to understand the range of public work-family supports available to low-wage workers to assess whether they meet the needs of low-wage workers, and to provide guidance to governments considering future PFL programs.

Table 2 PFL Generosity index
Full size table

PFL and the Safety Net

Some research on PFL has considered its relationship with means-tested programs, though evidence in this area remain mixed. For instance, PFL availability is associated with lower rates of participation in Temporary Assistance for Needy Families (TANF) and Supplemental Nutrition Assistance Program (SNAP) (Houser & Vartanian, 2012; Kang, 2020; Ybarra et al., 2019). These findings suggest that PFL may act as a substitute for means-tested provisions. However, Ybarra and colleagues (2019) find that the likelihood of receiving TANF, even when controlling for access to PFL, is positively associated with TANF benefit levels and earnings allowances. That is, the likelihood of receiving TANF is associated with TANF generosity. Further, in some instances, low-income individuals may receive greater benefits from TANF than from PFL, which suggests that some PFL wage replacement rates may not adequately support some low-income families (Ybarra, 2013).

Other safety net programs are also meaningful to low-wage workers during caregiving interruptions from work. Childcare subsidies, funded through the Child Care and Development Fund (CCDF) provide resources to offset childcare costs for qualifying low-wage workers with young children. Since states have the authority to set childcare subsidy policies, they too vary in generosity across states, including in income eligibility, family copayment rates, reimbursement rates to providers, required minimum work hours, and whether they serve all eligible families or use a wait list. The use of childcare subsides is associated with a greater likelihood of employment and longer work tenures among low-wage workers who participate in the program compared to those who do not (Forry et al., 2013). This suggests that childcare subsidies are important for low-wage workers in maintaining or securing employment, which may be especially useful following a caregiving-related work interruption. While Medicaid provides in-kind subsidized health insurance rather than cash or near-cash resources, it provides important healthcare resources to those who might otherwise go without. Low-income pregnant people in states with more generous Medicaid policies, including higher income limits and presumptive eligibility, are comparatively more likely to have perinatal health insurance coverage and seek earlier prenatal care (Wherry, 2018).

Tax provisions designed to support families, including low and moderate-income households, and their role in supporting caregiving remains understudied despite the promise they hold. For instance, the Earned Income Tax Credit (EITC), Child Tax Credit (CTC), and Child and Dependent Care Tax Credit (CDCTC) offset tax burdens for those who qualify and, if refundable, offer cash resources that bolster family income. Similarly, other state-legislated leave provisions such as Paid Sick Days (PSD) have proven to have positive economic effects particularly for low-wage workers (Stoddard-Dare et al., 2018). Given the likely importance of this array of social safety net provisions to caregiving interruptions from work and family resources and well-being during this period, we construct measures of generosity based on relevant policy rules for each provision (e.g., TANF; CCDF; Medicaid; state EITC, CTC and CDCTC; PSD; Medicaid Home and Community Based Services (HCBS) waivers) and then assign a generosity score for each program. Table 3, which can be found in the appendix, illustrates policy levers used for each program, index score assignment by policy lever and overall, the distribution across states, and is discussed further in the data and methods section.

The Current Study

To understand potential differences in available public supports between PFL and non-PFL states and within PFL states, we follow a rich line of scholarship that considers the impact of social safety net provisions, including design and generosity, on low wage workers and low-income families. We consider policies and programs that support caregiving across the life course, including the perinatal period and caregiving for adult family members.Footnote 3 For the perinatal period, we draw on state-level public programs that have been considered in related work including TANF (Hill, 2012; Kang, 2020; Stanczyk, 2016; Ybarra, 2013), the CCDF (Ha & Ybarra, 2013; Washbrook et al., 2011; Weber et al., 2014), Medicaid (Wherry, 2018), the EITC (Evans & Garthwaite, 2014; Hoynes et al., 2015), CTC (Marr et al., 2012), the CDCTC (Whitebook, McLean, Austin, & Edwards, 2018). For caregiving of adult members, we consider provisions that target family caregivers of older adults including the CDCTC, state-mandated PSD, and Medicaid HCBS waivers (Konetzka et al., 2024). We recognize that families may draw on other formal resources such as Supplemental Security Income (SSI) and informal support from family, friends, or nonprofits. We selected public provisions, however, based on their dual targeting of lower-income families and supporting work and caregiving responsibilities. We explore the following questions:

  1. 1.

    Do PFL states have more generous social safety net provisions that support caregiving interruptions from work compared to non-PFL states?

  2. 2.

    What is the distribution of safety net program generosity in PFL states?

Data

To answer our exploratory questions, we draw on multiple public data sources on program policies and rules that likely affect caregiving across the life course. We focus on the year 2019 because it is the most recent year of available data across selected programs that was not affected by temporary pandemic-related safety net expansions. As illustrated in Table 3 in the appendix, for each selected social safety net program we focus on criteria associated with access and benefit generosity, such as income eligibility, benefit levels, and the average benefit for a family of three.

For TANF, which provides cash assistance to qualifying low-income families with children who are primarily single-mother families, we include the maximum income eligibility for a family of three relative to the state median income (SMI), the maximum average monthly benefit for a family of three, the month of pregnancy a pregnant person becomes program-eligible, the length of exemption from work requirements for women with infants, and the program’s hassle factors (e.g. submission of required documents as part of applying to TANF). For CCDF, which provides subsidies for childcare, we include the maximum income eligibility for a family of three, the state reimbursement rate to providers (a proxy for accessibility and program quality), the share of income families must contribute as a copayment during participation, whether a state has a waitlist, and the minimum number of work hours required for participation. For Medicaid, we use the income limit for pregnant people as a percentage of the federal poverty line, and whether a state allows for presumptive eligibility for a pregnant person. For tax credits, we include indicators for state-level EITC, which provides tax credits to low-income people who work, CTC, which provides tax credits for having a child under a certain age, and CDCTC, which provides tax credits for money spent on care for children and dependents, including older adults. For each tax credit, we include whether a state has their own program, and if the program is refundable. For CDCTC, we consider this separately for children and adult dependents. For PSDs, which mandates employers provide paid time off for illness, we include whether a state has a PSD law, if the law allows for care to extended family, and if small business employees are eligible. For Medicaid HCBS, we include HCBS spending as a percent of all Long-Term Services and Supports (LTSS) spending, HCBS spending per capita, and if a waitlist is in use for seniors and adults with physical disabilities.

Method

Across all selected programs, we began with a distributional analysis of each program’s selected rules associated with caregiving to create categories that reflect a given policy lever’s range of restrictiveness or generosity. Depending on the policy rule and available data, we assigned values of 0 (restrictive) and 1 (generous), or 0 (restrictive), 1 (moderate), and 2 (generous). Next, we created a generosity index for each program by summing the assigned value for each selected program rule. The range of possible values for generosity indices varies by program, based on the number of rules considered. For example, TANF has a maximum value of ten based on five policy rules, while Medicaid has a maximum value of three based on two policy rules. For each policy we calculated tertiles based on the national distribution for each policy rule, and then assigned states a label of generous, moderate, or restrictive relative to a state policy’s distribution (i.e., top one-third = generous; mid one-third = moderate; bottom one-third = restrictive) (see appendix Table 3). We conduct two-tailed t-tests to compare generosity indices between PFL and non-PFL states.

Results

First, we compare each program’s selected individual rules between PFL and non-PFL states, and overall. CDCTC for adult dependents is not included because its generosity index includes only one program rule.

Safety Net Program Rules

TANF

As shown in Fig. 1, for the majority of selected TANF policy rules, PFL states are more generous. For policy rules concerning the maximum benefit for a family of three and the month of pregnancy a pregnant person becomes program eligible, PFL states’ indices are statistically significantly higher than non-PFL states. Work exemption rules and the hassle index are more generous but not statistically significant, and the maximum income eligibility for a family of three is lower in PFL states than in non-PFL states.

Fig. 1
figure 1

TANF Generosity Index Program Rules for PFL and non-PFL States. Asterisks indicate significant differences between mean generosity indices for PFL and Non-PFL States * p < 0.1, ** p < 0.05, *** p < 0.001. Sources Shantz et al., 2020

Full size image

CCDF

For selected CCDF rules, shown in Fig. 2, some are more generous in PFL states, and some are more generous in non-PFL states, though none of the differences are statistically significant. The mean family copayment rate for a family of three and work requirements are more generous in PFL states while non-PFL states have more generous maximum income eligibility rules for a family of three, state reimbursement rates, and not using a waitlist.

Fig. 2
figure 2

CCDF Generosity Index Program Rules for PFL and non-PFL States. Asterisks indicate significant differences between mean generosity indices for PFL and Non-PFL States. * p < 0.1, ** p < 0.05, *** p < 0.001. Sources Dwyer et al., 2020; Schulman, 2019

Full size image

Medicaid

For selected Medicaid rules, shown in Fig. 3, PFL states are more generous for most of our selected rules. Compared to non-PFL states, PFL states are statistically significantly more likely to be a Medicaid expansion state and have a more generous income limit for pregnant people. Presumptive eligibility is slightly more generous in non-PFL states, but the difference is not statistically significant.

Fig. 3
figure 3

Medicaid Generosity Index Program Rules for PFL and non-PFL States. Asterisks indicate significant differences between mean generosity indices for PFL and Non-PFL States. * p < 0.1, ** p < 0.05, *** p < 0.001. Sources Brooks et al., 2019; Medicaid and CHIP Income Eligibility Limits for Pregnant Women, 2003–2020, 2020

Full size image

Tax Credits in Perinatal Period

For policy rules that concern tax credits around the time of a birth, shown in Fig. 4, PFL states are statistically significantly more generous for state EITC’s and the CDCTC’s, and marginally significantly more generous for the CTC.

Paid Sick Days

For Paid Sick Days policy rules, shown in Fig. 5, PFL states are statistically significantly more generous on every rule, which include if the state has a PSD law, if care for at least some extended family is covered, and if employees of small businesses are eligible.

Medicaid HCBS

As shown in Fig. 6, Medicaid HCBS rules concerning spending are statistically significantly more.

Fig. 4
figure 4

Tax Credits in Perinatal Period Generosity Index Program Rules for PFL and non-PFL States. Asterisks indicate significant differences between mean generosity indices for PFL and Non-PFL States. * p < 0.1, ** p < 0.05, *** p < 0.001. Sources Making Care Less Taxing: State Child and Dependent Care Tax Provisions, Tax Year 2019, 2020; State EITC as Percentage of the Federal EITC, 2020; State Tax Credits. (n.d.)

Full size image
Fig. 5
figure 5

Paid Sick Days Generosity Index Program Rules for PFL and non-PFL States. Asterisks indicate significant differences between mean generosity indices for PFL and Non-PFL States. * p < 0.1, ** p < 0.05, *** p < 0.001. Source Paid Sick Days—State and District Statutes (2023)

Full size image

generous than non-PFL states, including HCBS spending as a percent of LTSS spending and HCBS spending per capita. However, non-PFL states are less likely to use an HCBS waitlist than PFL states, though this is not significantly different.

Safety Net Indices

Next, we compare each program’s generosity index between PFL and non-PFL states and overall. The social safety net indices are aggregate indices created by summing the policy rules discussed above. Reflected in Fig. 7 below, we find that the social safety net programs that support caregiving are more generous in PFL states than in non-PFL states. Except for CCDF, the average generosity index for each program we consider is greater (more generous) in PFL states. The average generosity index for CCDF is nearly the same for non-PFL states compared to PFL states and the difference is not statistically significant. Among programs that are more generous in PFL states, the difference is statistically significant for tax credits that support the time around a birth, tax credits for adult dependent care, PSD, and Medicaid HCBS. The Medicaid index is marginally significantly more generous in PFL states. Therefore, the findings confirm our assumption that states that have passed PFL legislation are also more likely to have more generous social safety net programs that support caregiving.

Fig. 6
figure 6

Medicaid HCBS Generosity Index Program Rules for PFL and non-PFL States. Asterisks indicate significant differences between mean generosity indices for PFL and Non-PFL States. * p < 0.1, ** p < 0.05, *** p < 0.001. Source Murray et al., 2021; Medicaid HCBS Waiver Waiting List Enrollment, by Target Population and Whether States Screen for Eligibility, 2018, n.d

Full size image
Fig. 7
figure 7

Mean Generosity Indices for PFL and Non-PFL States. Asterisks indicate significant differences between mean generosity indices for PFL and Non-PFL States. * p < 0.1, ** p < 0.05, *** p < 0.001

Full size image

Distribution of Safety Net Indices Among PFL States

While PFL states offer more generous programs overall compared to non-PFL states, we find substantial variation in the generosity of social safety net programs across PFL states. This suggests that while PFL may be filling some of the gaps in the existing safety net, the degree of supports available to low-income families differs across these states. In particular, it calls attention to the need to consider the design of state sponsored PFL programs relative to existing holes in the safety net. Figures < link rid="fig8”>8-A to 8-C show the generosity indices and ascribed generosity levels for each program by PFL state.

There are some observable patterns in the generosity of safety net programs by state. For example, California, New York, Washington D.C (Fig. 8-A), and Maine (8-C) all have generous provisions for five or more of selected social safety net provisions, with zero, one, or two moderate programs, and zero to one restrictive programs. Other states have a more even distribution of generous and moderate programs with zero to one restrictive programs, including Massachusetts, Oregon (Fig. 8-B), and Maryland, and Oregon (Fig. 8-B). Still, another set of states has a more even distribution of generous, moderate, and restrictive programs, including New Jersey (Fig. 8-A), Washington state, Connecticut (Fig. 8-B), Colorado, and Minnesota (Fig. 8-C). Delaware is the only state that has a majority of restrictive programs with some moderate programs.

There are no observable trends when considering PFL generosity along with safety net generosity. PFL generosity is indicated by the solid-colored bar in Fig. 8A, B, and C. Rather than trends associated with safety net generosity, PFL programs that have been passed in more recent years tend to be more generous than programs passed earlier on, irrespective of the generosity of the safety net. Still, this suggests that states that are considering PFL legislation should examine the existing safety net and craft legislation such that PFL can fill holes in public support for caregiving.

Fig. 8
figure 8

A PFL and Safety Net Generosity for PFL States. *Medicaid HCBS data not available for California. B PFL and Safety Net Generosity for PFL States. *Medicaid HCBS data not available for Delaware

Full size image

Limitations

While our work highlights consequential differences in the safety net between PFL and non-PFL states and within PFL states, there are limitations we hope will be addressed in future work. First, our generosity indices are relatively blunt instruments. We also assigned each program rule equal weight in index construction. In practice, one program rule may be more meaningful to some low-wage workers than others. For example, it is possible that TANF benefit amounts matter more in seeking support than the administrative hassles. Additionally, our paper is focused on program rules, not program use. We find evidence of variation in program generosity across PFL states but could not consider program use across states. Future studies should examine participation in PFL and other public provisions during caregiving interruptions from work in the context of program generosity.

Discussion

The social safety net provides low-income families with resources that can support caregiving-related work interruptions. This study finds that states that have implemented or passed PFL legislation have more generous social safety net programs than states that do not have PFL policies. This suggests that low-wage workers in non-PFL states not only lack PFL support but also, on average, have comparatively less generous social safety net caregiving supports available to them. As such, passage of a federal PFL policy would most benefit low-wage workers in non- PFL states not only by providing PFL but also strengthening comparatively weaker safety nets. Additionally, if a federal PFL policy is sufficiently generous, it could raise PFL program design standards across all states, bolstering caregiving support overall and especially in current non-PFL states.

Even though public opinion polls show strong and bipartisan support for PFL (Horowitz et al., 2017), and some southern and central states have introduced PFL legislation, PFL still remains concentrated in northeast and western states. This suggests that federal legislation may be required for people in southern and central states with weak safety nets to have access to PFL benefits. An encouraging sign of the federal government moving toward incentivizing states to implement PFL was the House of Representatives 2023 establishment of a bipartisan PFL working group, including representatives from weaker safety net states without PFL such as Iowa, Louisiana, Texas, and Oklahoma (Paid Family Leave Working Group: Legislative Framework, 2024). The working group recently released a report with options for the federal government to consider in incentivizing states and businesses to take up PFL programs or enhance existing programs. Although Medicaid delivers different resources and services and is means-tested, it is instructive in considering additional ways the federal government might incentivize PFL take-up in weaker safety net states. For example, by the end of 2014, after implementation of the Affordable Care Act, 28 states had expanded Medicaid. Today, 39 states have expanded Medicaid, including “red” states with less generous safety nets, such as Kentucky (Status of State Medicaid Expansion Decisions, 2024). The federal government incentivized states’ Medicaid expansion by paying for the bulk of expansion costs. Since existing PFL programs are largely paid for by taxes on employees, and in some cases, also by employer taxes, federal assistance with some, but not the majority, of the costs of PFL implementation may be useful to promoting expansion of state PFL programs.

While residents of non-PFL states stand to benefit the most, we know little about how individuals and families package safety net benefits during caregiving interruptions from work. Given the variation in generosity across PFL and non-PFL states and within PFL states, both researchers and policymakers need to be attuned to these differences and recognize that PFL may act as a substitute or complement to the safety net in different states, depending on the generosity of a state’s safety net and PFL program. For non-PFL states, if they introduce PFL in the future or if a federal PFL policy is passed, the program may interact with the safety net quite differently than in current PFL states given non-PFL states’, on average, weaker safety nets. For instance, it may be the case that the introduction of PFL in current non-PFL states with more restrictive safety nets does not result in a decrease in safety net participation as found in prior research on PFL and the safety net (Houser & Vartanian, 2012; Kang, 2020; Ybarra et al., 2019). As discussed in the limitations section, future research on the relationship between PFL and the safety net should account for both programs’ generosity relative to selected outcomes of interest. This can help researchers and policy makers to better understand how program generosity affects the substitutions or complementarities between the safety net and PFL programs as well as identify whether variations in PFL and safety net generosity within and across states can help us identify cross-program designs that significantly improve a host of outcomes including health, work, income, and job tenure.

Implications for Practice

Our findings suggest federal and state policymakers will want to consider variations in state-level safety net benefits when designing PFL programs. Ideally, policymakers should view PFL and the safety net as complementary programs for low-wage workers who are often without employer-provided leave benefits. Social workers engaging directly with clients in PFL states need adequate PFL and safety net policy knowledge to effectively serve clients during periods of caregiving.