The Caregiver Workforce: The 50-Million-Strong Industry We Built on Unpaid Labor

Section 01The Workforce Hiding Inside the Workforce

An estimated 53 million U.S. adults provided unpaid care to a family member or friend in the most recent reference year — roughly one in five adults [1]. The paid direct-care workforce — home health aides, personal-care aides, certified nursing assistants — numbers approximately 4.6 million workers and is the single fastest-growing job category in U.S. Bureau of Labor Statistics employment projections [2]. Taken together, caregiving is the largest hidden industry in the United States.

It is also the industry most absent from the workforce-policy conversation. The 2026 conversations about AI displacement, the future of work, and worker rights are almost entirely about paid wage labor inside formal firms. The 53 million unpaid caregivers and 4.6 million paid direct-care workers — overwhelmingly women, disproportionately women of color, materially underpaid — barely show up. This piece walks the numbers, the policy reforms with strongest evidence, and what employers and individual workers can do.

Section 02The Numbers: 53M Unpaid Caregivers, 4.6M Paid Direct-Care Workers, $600B+ in Foregone GDP

Four patterns the numbers actually show.

Pattern 1: Unpaid caregiving is large, growing, and disproportionately female. The AARP / National Alliance for Caregiving Caregiving in the U.S. report (2020 reference, with 2023 update) places the unpaid caregiver count at roughly 53 million, up from 43.5 million in the 2015 estimate. Approximately 61% of caregivers are women. The median caregiver provides 24+ hours of unpaid care per week. The estimated economic value of unpaid caregiving — what it would cost to replace at market rates — exceeds $600 billion annually, larger than the GDP of Sweden [1].

Pattern 2: Direct care is the largest growing occupation in the economy. BLS Employment Projections to 2032 show home health and personal-care aides adding more jobs in absolute terms than any other detailed occupation — approximately 800,000 net new positions in the projection window. The sector's growth is structural (the aging of the U.S. population), not cyclical. The supply gap is the constraint, not demand [2].

Pattern 3: Direct-care wages are categorically depressed. The 2023 median wage of $16.72/hour for U.S. direct-care workers is lower in real terms than the 1980 figure. The sector has the highest poverty rate of any major occupational category — approximately 16% of direct-care workers live below the federal poverty line, and roughly half participate in some form of public assistance program. Approximately 87% of direct-care workers are women; roughly 60% are women of color; approximately one in four were born outside the U.S. [3].

Pattern 4: The career penalty for caregivers is real and measurable. Caregivers — particularly women caregivers — report higher rates of reduced hours, missed promotions, and labor-force exit than non-caregiver peers controlling for other variables. The estimated lifetime earnings loss for a primary family caregiver is approximately $300,000 in wages and Social Security combined (MetLife / Caring Across Generations analyses). The gender skew makes the caregiving penalty also a gender pay-gap mechanism.

Section 03Why Caregiving Penalizes Women's Careers Disproportionately

The 2023 unpaid-caregiver demographic data is unambiguous: women provide more unpaid care than men (61% vs 39% of caregivers), more hours per week of unpaid care (~24 vs ~18 hours), and more intensive types of care (medical/nursing tasks, end-of-life care, dementia care). These patterns hold even controlling for employment status, marital status, and household income.

The career-economics implications, documented across multiple sources:

• Labor-force participation. The U.S. female labor-force participation rate has flatlined since the late 1990s in part because of caregiving demands that didn't exist at the same scale in the prior generation. The OECD comparative data shows the U.S. female LFP rate is now noticeably below the OECD median; the gap is plausibly explained by the absence of universal paid family leave, the limited public childcare infrastructure, and the absence of comprehensive eldercare supports [4].
• Hours and intensity reductions. Approximately 30% of caregivers report reducing work hours; approximately 16% report leaving a job entirely; approximately 13% report retiring earlier than planned. The aggregate effect is a measurable reduction in women's lifetime earnings, separate from any direct discrimination in the workplace.
• The "sandwich generation" intensification. The cohort caring simultaneously for aging parents and minor children is now an estimated 28% of U.S. adults aged 40–59 (Pew Research Center). The 2026 squeeze is structurally larger than the 1990s version, because parents live longer and women have children later.

The Women, Work and the Future pillar covers the broader gender-economics pattern; the caregiving-specific story is that without changes to the caregiving infrastructure (paid leave, childcare access, eldercare access, direct-care workforce wages), the gender-economics story does not improve, regardless of progress on workplace discrimination.

Section 04The Direct-Care Workforce: Underpaid, Understaffed, Overwhelmingly Women of Color

PHI's annual Direct Care Workforce State Data Reports are the cleanest quantitative reference for U.S. direct-care employment [3]. The 2024 report establishes the following composition:

• Total workforce: approximately 4.6 million paid direct-care workers (home health aides, personal-care aides, nursing assistants in nursing homes, residential-care aides).
• Gender: approximately 87% women.
• Race/ethnicity: approximately 60% workers of color, with Black, Hispanic, and immigrant workers especially over-represented in home-care and residential-care segments.
• Immigration: approximately 27% foreign-born — higher than the U.S. workforce average; in some states (California, New York, New Jersey, Florida) the share approaches half.
• Wages: 2023 median hourly wage of $16.72; approximately one in six lives below the federal poverty line.
• Benefits: approximately one-third of direct-care workers do not have employer-sponsored health insurance; rates of paid sick leave, paid time off, and retirement benefits are well below U.S. workforce averages.
• Turnover: annual turnover in home-care and nursing-home roles regularly exceeds 60%; in some operators it exceeds 100% — meaning the average worker holds the role for less than a year.

Why the wages are what they are

The single largest structural reason: most U.S. direct-care wages are functionally set by Medicaid reimbursement rates, since Medicaid (state and federal) is the largest single payer for long-term care services in the U.S. Medicaid Home and Community-Based Services (HCBS) reimbursement is set by state legislatures and has historically been increased at well below inflation. The wages employers can pay are bounded above by what the reimbursement allows, regardless of labor-market conditions. The result: a workforce in chronic supply shortage, but with wage-setting almost entirely insulated from labor-market signals.

Section 05What Policy Actually Moves

The reforms with the strongest evidence base, ranked roughly by expected impact on caregiver and worker outcomes:

1. Paid family and medical leave

The U.S. is the only OECD country without a federal paid-leave program. Thirteen states (plus DC) had paid family and medical leave programs operational or scheduled to launch by 2026. The empirical evidence from California's 2004 program (the longest-running) is consistent: paid leave increases caregiver labor-force attachment, reduces the "leaving a job" share of caregiver outcomes, and produces measurable child- and elder-health benefits. The expected federal-program annual cost would be modest relative to other social-insurance programs; the political feasibility shifts cycle by cycle.

2. Universal child tax credit (refundable)

The 2021 expansion of the Child Tax Credit — fully refundable, paid monthly — reduced child poverty by approximately 30% in a single year (Census Bureau, 2022). The expansion expired at the end of 2021. Restoration with the refundability feature is the most-evidenced single anti-poverty intervention available for families with caregiving responsibilities. The CBO cost estimates have been published; the policy debate is political rather than empirical.

3. Medicaid HCBS wage and access reform

Raising Medicaid Home and Community-Based Services reimbursement, with wage pass-through requirements, is the most direct lever on direct-care worker wages. The 2024 CMS Access Rule's 80%-to-compensation requirement is the operational template; expansion of HCBS funding (the "Better Care Better Jobs" framework that surfaced in 2021–2022 federal legislative debate) would be the larger-scale version.

4. Immigration policy for care work

Direct-care work is structurally dependent on immigrant labor, and the U.S. has no functional immigration visa pathway for care workers. The proposals to create a sector-specific visa (analogous to the H-2A agricultural visa, or to Canada's Caregiver Program) have circulated for two decades without passage. The supply gap is going to widen materially in the 2025–2035 demographic window; the policy gap is structural.

5. Universal pre-K and childcare access

The childcare supply gap is the largest single mechanical reason for U.S. female labor-force participation underperformance. Federal proposals (the Build Back Better childcare provisions of 2021; state-level efforts in Vermont, New Mexico, Massachusetts) are operational starting points. The OECD comparative data is unambiguous on the magnitude of the labor-supply gain.

Section 06What Employers Can Do

The list that actually correlates with caregiver retention and engagement is narrower than the conventional "family-friendly workplace" framing implies. Six interventions with the strongest evidence base.

1. Use-it-or-lose-it caregiving leave (paid, separate from PTO)

A separate paid-leave bucket explicitly for caregiving — not aggregated into general PTO that employees feel pressure not to use — produces meaningfully higher actual utilization. The pattern in firm-level data (Care.com, Bright Horizons, Vivvi reports) is that use-it-or-lose-it leave correlates with retention and engagement in a way that nominally-available but functionally-discouraged leave doesn't.

2. Backup care benefits

Employer-subsidized backup care (child, adult, sick-care) when normal care arrangements break down is one of the most-utilized caregiving benefits when offered. The Bright Horizons and Care@Work platforms operate at scale; employer adoption has roughly doubled since 2019. The retention return is large enough to clear the per-employee cost in most published case studies.

3. Schedule flexibility (predictability, not just remote)

The remote-work conversation focused on location; the caregiving conversation requires predictability. For shift-work caregivers in particular, the ability to know one's schedule 1–2 weeks in advance (rather than 2–3 days) is the single largest accommodation factor in published surveys. Several state-level Fair Workweek laws (Oregon, San Francisco, New York City, Seattle, Philadelphia) codify this; voluntary employer adoption beyond legal mandate is associated with lower turnover.

4. Eldercare-specific benefits

Childcare benefits are common; eldercare benefits are rare. Approximately 70% of large U.S. employers offer some form of child-care benefit; under 10% offer comparable eldercare benefits, despite roughly equal numbers of employees in both categories. Care concierge services, geriatric-care-manager subsidies, eldercare leave separate from sick leave, and adult-backup-care all have growing employer pilots.

5. Caregiver ERGs and peer support

Employee resource groups for working caregivers, where they exist, consistently surface workplace-policy needs the HR function would not have surfaced otherwise. The Holding Co.'s research on the caregiving economy documents this pattern across multiple employers [5]. The activation cost is modest; the engagement and policy-discovery return is meaningful.

6. Caregiving-aware manager training

Frontline managers' responses to caregiver employees are the single largest variable in the employee-experience data. Manager training that explicitly covers caregiving stigma, legal accommodation requirements (FMLA, state-level paid leave), and the practical mechanics of supporting a caregiver report produces measurable differences in retention and engagement.

The U.S. care economy is the foundational infrastructure no one is building. We have a transit infrastructure, a digital infrastructure, an energy infrastructure. The care infrastructure is the one we have most fully outsourced to unpaid family labor — and the bill is coming due. Ai-jen Poo, The Age of Dignity (The New Press, 2015), updated 2023 paperback

Section 07What's New in 2026

Four developments worth tracking in the 2026 caregiving-economy conversation.

AI and robotics in direct care

The deployment of AI scheduling and care-coordination tools, telehealth monitoring, and a small but expanding set of assistive robotics (mobility, transfer, medication reminding) is changing direct-care work. The honest framing: these tools augment direct-care workers more often than they replace them, because the human-touch and judgment requirements of personal care remain durable. The Augmented or Replaced pillar's framework applies — direct care is a clear "augmented, not replaced" category. The risk to watch is staffing-ratio degradation: employers using AI tools to justify lower workers-per-resident ratios in nursing homes is a documented pattern, and one that direct-care worker organizations are actively pushing back on.

Dependent-care FSA expansions

The federal Dependent Care Flexible Spending Account cap (currently $5,000) has not been updated since the 1980s. Several state-level analogs and IRS guidance updates in 2024 expanded the practical utility of dependent-care benefits. Federal expansion to $7,500–$10,000 has been proposed in multiple cycles; passage would meaningfully increase the value of employer caregiving benefits.

State-level paid family leave waves

The state-level paid-family-leave map has expanded from 5 states in 2018 to 13+ states with operational or scheduled programs by 2026. Maine, Minnesota, Maryland, and Delaware are among the newer additions. The state-by-state pattern means workers' caregiving rights are now meaningfully geographic — a Massachusetts worker has very different leave entitlements from an identical worker in Alabama. This produces both political pressure for federal action and a meaningful body of state-level evidence.

The Care Workers' Bill of Rights conversation

The "Domestic Workers' Bill of Rights" framework, originally passed in New York State in 2010 and now in 11 states, is the legislative template for extending basic labor protections (overtime, sick leave, anti-discrimination) to the historically excluded categories of in-home care work. The 2026 expansion is a watch-item.

There are 53 million unpaid caregivers and 4.6 million paid direct-care workers in the U.S. There is no other industry of comparable size that gets this little policy attention. The caregiving conversation isn't a niche workforce conversation. It's the workforce conversation we keep deferring.

Section 08For Workers Who Are Caregivers

Three things the published guidance and community conversations consistently support:

1. Know your rights under FMLA and state-level paid leave. The federal Family and Medical Leave Act covers employers with 50+ employees and provides up to 12 weeks unpaid, job-protected leave for caregiving. Many workers do not realize they qualify, or do not use it because the leave is unpaid; state-level paid programs (in 13+ states by 2026) have meaningfully different rules. The Department of Labor's FMLA fact sheets and your state's paid-leave program documentation are the authoritative sources.
2. Tell your employer earlier than feels comfortable, in writing. The empirical pattern in workplace-discrimination research is that caregivers who disclose their caregiving responsibilities — strategically, professionally, and in writing — get better accommodation than caregivers who try to keep caregiving invisible until it produces an absence or performance issue. The disclosure framing matters; the literature on the "flexibility stigma" suggests the most effective framing is solution-oriented ("Here's what I need, here's what I'll deliver").
3. The Caring Across Generations and AARP resource pages are the most-updated practical references. Beyond formal employer benefits, the actual mechanics of arranging care — finding home-care workers, navigating Medicare/Medicaid, accessing community resources — are the things working caregivers spend the most time on and have the least training in. AARP's Family Caregivers hub and Caring Across Generations' resource library are the most consistently maintained free references.

And the larger reframe: caregiving is not a personal problem to be managed around your career. It is one of the largest economic activities in the country, performed disproportionately by women, structurally underpaid and underrecognized when it is paid, and structurally invisible when it is not. The policy debate is finally — slowly — catching up. The next decade's workforce policy is going to be substantially defined by whether the U.S. builds the care infrastructure its labor force already needs.