In November 2025, seventeen Massachusetts ACO executives gathered in a Boston conference room to discuss work requirement implementation. Their organizations served 800,000 MassHealth members through sophisticated care management platforms with two-sided risk arrangements and quality incentive payments. They had data infrastructure connecting primary care, behavioral health, social services, and community organizations. They measured clinical outcomes, tracked social determinants, coordinated complex care. One executive asked the obvious question: how do we layer employment verification onto systems designed to improve health outcomes, not police work status? The room went quiet. They had the technical capacity. What they lacked was belief that the policy served their members’ interests.
California, New York, Massachusetts, Washington, Oregon, Colorado, Rhode Island, Vermont, and Connecticut share characteristics distinguishing them from early adopter states. All expanded Medicaid early under Democratic leadership committed to coverage expansion as health policy priority. All invested heavily in managed care infrastructure, delivery system innovation, and social determinants initiatives. All possess administrative capacity peer states envy: sophisticated data systems, established MCO relationships, mature quality measurement frameworks. Yet all approach December 2026 implementing federal mandates their political leadership fundamentally opposes. Governor Gavin Newsom called work requirements “cruel and counterproductive.” Governor Kathy Hochul warned they would “devastate coverage.” Governor Maura Healey characterized them as “policy moving backwards.”
The tension between technical capacity and political resistance defines how these nine states navigate requirements they have tools to implement but lack will to enforce. They can build verification systems. They possess MCO infrastructure for delegation. They operate data platforms for automated matching. But they will design those systems to minimize coverage loss, maximize exemptions, and create member protections that technically comply with federal law while functionally undermining policy intent. The synthesis question is whether high administrative capacity enables better outcomes when political will opposes policy goals, or whether capacity merely allows competent implementation of opposed policy producing different tensions than incompetent implementation of supported policy.
The Infrastructure They Built for Different Purposes#
These states constructed Medicaid delivery systems to achieve clinical and social goals unrelated to work requirement enforcement. The mismatch between infrastructure purpose and federal mandate creates operational tension independent of political opposition.
Massachusetts operates seventeen ACOs serving MassHealth members through two-sided risk with quality withholds. Partners HealthCare, Wellforce, Steward, and other ACO entities accept financial risk for total cost of care while being measured on preventive screenings, chronic disease management, behavioral health integration, and health equity metrics. Member relationships emphasize care coordination, needs assessment, service linkage. Navigators help members access housing support, food assistance, behavioral health treatment. The infrastructure was never designed to verify employment or terminate coverage for documentation failures.
Oregon’s CCO model creates provider-led regional entities accepting global budgets for physical, behavioral, and oral health with explicit social determinants mandates. CCOs must invest net income in community benefit programs addressing upstream health factors. Health Share of Oregon, PacificSource, Trillium, and other CCOs built care coordination platforms connecting clinical care to housing, employment support, education programs. The model presumes that addressing social needs improves health outcomes. Work requirements reverse the causal arrow: employment becomes precondition for health coverage rather than health coverage enabling employment.
California manages three distinct MCO models across fifty-eight counties. County Organized Health Systems in twenty-two counties operate as single-plan arrangements governed by county boards. The Two-Plan Model in fourteen counties (including Los Angeles) offers choice between local initiative plans and commercial plans. Geographic Managed Care in Sacramento and San Diego provides multiple plan options. L.A. Care Health Plan serves 2.8 million members with sophisticated care management infrastructure. CalOptima in Orange County serves 900,000 members. These systems coordinate complex care, manage dual eligibles, integrate specialty services. They were built for care delivery, not compliance verification.
Washington integrated behavioral health into Apple Health managed care, creating single accountability for physical and mental health. The Foundational Community Supports benefit covers housing and employment services as Medicaid benefits, recognizing these as health-related. Colorado’s RAE model similarly integrates behavioral health through Regional Accountable Entities. New York operates statewide managed care with comprehensive member services. Connecticut maintains fee-for-service by choice but with sophisticated eligibility systems and care coordination for complex populations.
The question facing these states is whether infrastructure built for care coordination can be repurposed for employment verification without undermining the care relationships that infrastructure created. When navigators who previously helped members access mental health services must now verify work hours and process exemption documentation, does that change member trust? When care managers who coordinated chronic disease treatment must now track compliance deadlines and termination notices, does that affect clinical relationships? The technical answer is that systems can be modified. The practical question is whether modification destroys the purpose systems were built to serve.
California: Scale Meets Fiscal Constraints#
California’s 4.7 million expansion adults represent more than twice New York’s population and six times Massachusetts’s. Implementation scale dwarfs all peers. The state’s January 2026 implementation plan from DHCS envisions automated verification through EDD wage records, Coverage Ambassador infrastructure for member outreach, and county welfare department coordination across fifty-eight administrative contexts. The technical sophistication is real. The fiscal constraints are equally real.
California operates at floor FMAP, receiving only 50% federal match for administrative costs. Every dollar spent on verification infrastructure requires matching state dollar. The unwinding of differential provider tax structures eliminates $3.7 billion annually that previously supported program costs. DSHP phase-out reduces fiscal flexibility for state-funded health programs. The state faces structural budget deficits before work requirement costs appear.
The Urban Institute projects 1.2 to 1.4 million Californians could lose coverage. Even if the state builds recognition-based systems minimizing documentation barriers, the sheer scale means hundreds of thousands will face verification failures. County administration across fifty-eight contexts creates consistency challenges. Los Angeles County alone processes more Medi-Cal applications than most states. Rural counties like Alpine or Sierra have eligibility staff in single digits. Training thousands of county workers on new requirements while managing concurrent changes (semi-annual redeterminations, asset limits, enrollment freezes for undocumented adults, benefit eliminations) stretches capacity regardless of sophistication.
California’s three MCO models create different implementation realities within a single state. COHS counties with single-plan arrangements can establish consistent processes. Two-Plan counties must coordinate between local initiatives and commercial plans. Geographic Managed Care regions with multiple competing MCOs face coordination complexity when work verification must flow through multiple organizational structures simultaneously. The state can issue uniform policy guidance, but operational reality varies by county model.
The synthesis insight is that administrative capacity determines what states can build but fiscal capacity determines what they will build. California possesses technical sophistication to implement recognition-based systems using automated data matching, presumptive exemptions, and proactive navigation. The state may lack fiscal resources to fund comprehensive infrastructure when administrative match is only 50% and budget constraints are acute. The result will be partial automation knowing it creates coverage losses for populations systems cannot see, or manual verification creating administrative burden at scale that overwhelms county capacity.
Massachusetts: When ACOs Meet Compliance Systems#
Massachusetts built mature ACO infrastructure as Medicaid delivery model. Seventeen ACOs serve members under performance-based contracts with quality metrics, cost containment incentives, and care coordination requirements. The model presumes continuity enables quality. Stable enrollment allows preventive care investment that reduces future costs. Care managers develop relationships with high-utilizing members to coordinate services across providers. Quality measurement compares year-over-year outcomes for enrolled populations.
Work requirements disrupt all these assumptions. When verification cycles generate enrollment volatility, quality measurement denominators become unstable. An ACO shows declining diabetes screening rates not because clinical performance worsened but because members with diabetes lost coverage and the remaining population has different characteristics. Cost trend analysis becomes meaningless when population composition changes quarterly rather than annually. Care management investment in high-utilizing members produces no return when those members lose coverage before interventions can affect outcomes.
The state must either redesign quality measurement to accommodate enrollment disruption or accept that ACO performance will appear worse due to statistical artifacts rather than care quality changes. Redesigning measurement requires new methodologies, new data systems, new contractual frameworks with ACOs. Not redesigning measurement punishes ACOs financially for policy-driven enrollment changes they cannot control. Neither option resolves the fundamental tension between continuous coverage needed for population health management and episodic coverage produced by verification requirements.
Massachusetts also faces the exemption documentation challenge at scale. The state serves approximately 700,000 expansion adults. DHCS projections suggest 60-75% could qualify for exemptions under generous exemption categories. But qualifying for exemptions and successfully obtaining them are different things. Medical frailty exemptions require provider attestation. Behavioral health exemptions require clinical documentation. Caregiver exemptions require dependent verification. Even presumptive exemptions based on diagnosis codes in claims data require member confirmation.
Processing 400,000+ exemption applications within months preceding January 2027 implementation requires infrastructure Massachusetts hasn’t built. The state has sophisticated eligibility systems for standard Medicaid determinations. Exemption processing adds different workflows requiring clinical judgment about work capacity, functional limitations, caregiving demands. Training eligibility workers to make those determinations, establishing provider attestation portals, creating appeals processes for denied exemptions, all require capacity expansion that fiscal constraints limit.
The competence paradox is that Massachusetts knows exactly what comprehensive implementation requires and understands it cannot build that infrastructure in available time with available resources. The state will implement work requirements competently relative to what’s feasible, knowing that feasible falls far short of what comprehensive protection would require.
New York: Political Will Meets Administrative Reality#
New York’s 2.7 million expansion adults make it second-largest implementation after California. The state operates statewide managed care through multiple MCOs with pharmacy carved out but comprehensive member services carved in. Governor Hochul and legislative Democrats uniformly oppose work requirements as policy. But political opposition doesn’t eliminate implementation obligations.
New York will pursue the minimalist strategy aggressive Democratic states adopt: design requirements to maximize exemptions, minimize verification frequency, and establish generous good cause provisions preventing terminations for correctable documentation errors. The state’s administrative capacity enables sophisticated exemption screening using existing data. Individuals appearing in SSDI or SSI systems trigger automatic exemptions. Those in unemployment compensation databases trigger economic downturn protections. Those with Medicaid claims showing chronic conditions trigger medical frailty reviews.
But automated screening identifies candidates for exemptions, not automatic approvals. Someone with diabetes claims doesn’t automatically qualify for medical frailty exemption unless diabetes severity prevents 80 hours monthly work. That determination requires clinical judgment automated systems cannot make. Provider attestation becomes necessary. Whether New York’s provider community will engage in wholesale exemption attestation affecting hundreds of thousands of members remains uncertain. Providers already face administrative burden from prior authorizations, quality reporting, value-based payment participation. Adding work exemption attestation may face provider resistance regardless of political alignment.
New York’s MCO infrastructure provides delegation pathway for verification and navigation. Healthfirst, Fidelis Care, MetroPlus, and other plans serving over two million Medicaid members have established member communications, care coordination platforms, and community partnerships. But H.R.1’s conflict of interest provisions prevent MCOs from conducting compliance determinations when they have financial interest in terminations. This forces New York to maintain state-operated verification separate from MCO service delivery, eliminating administrative efficiency MCO delegation was supposed to provide.
The state thus faces building parallel systems: MCO-operated navigation and support helping members comply, state-operated verification determining compliance and processing terminations. The MCO navigator helps member find employer verification. The state eligibility worker determines if that verification satisfies requirements. The member experiences both as “the Medicaid program” but they operate through different organizational structures with different incentives and different accountability.
Washington and Oregon: Integrated Models Meet Federal Requirements#
Washington and Oregon both pursued delivery system innovations integrating physical and behavioral health with explicit social determinants mandates. Washington’s Foundational Community Supports covers housing and employment assistance as Medicaid benefits. Oregon’s CCO global budgets require community benefit investment. Both states built infrastructure connecting healthcare to upstream social factors.
Work requirements invert this model. Instead of healthcare enabling social stability, social stability (employment) becomes precondition for healthcare. The philosophical conflict is fundamental, not technical. But technical capacity shapes how philosophical conflicts manifest operationally.
Oregon’s CCO structure enables the most integrated minimalist approach. CCOs already operate community benefit funds, SDOH screening, and flexible services that can be recategorized as work requirement support. The global budget structure allows CCOs to invest in navigation knowing coverage retention reduces medical costs. If a CCO spends $300 per member on navigation support that prevents coverage loss, it saves $3,000 in forgone capitation and risk adjustment degradation. The financial logic aligns even when policy philosophy conflicts.
Oregon will likely implement work requirements through CCO infrastructure designed to maximize compliance recognition rather than enforcement. Health Share of Oregon, PacificSource, and other CCOs will deploy community health workers already embedded in member neighborhoods. Those workers will help members document work or navigate exemptions using relationships already established. The system will look like enforcement but function like protection.
Washington faces different dynamics. The state’s integrated managed care doesn’t use global budgets or require community benefit investment like Oregon’s CCOs. Washington MCOs operate under capitation with risk adjustment but without the explicit SDOH infrastructure CCOs maintain. The state must build work requirement navigation capacity that Oregon can adapt from existing infrastructure.
Both states will establish expansive exemption categories. Pregnancy plus 60 days postpartum. Medical frailty with provider attestation. Student status with enrollment verification. Caregiver status for children under 14 (or under 18 if Washington successfully negotiates that). SUD treatment with provider confirmation. The goal is exempting 65-75% of expansion populations so verification burden concentrates on smaller populations more likely to be working already.
Colorado, Rhode Island, Vermont, Connecticut: Small State Variations#
Colorado’s RAE behavioral health model creates different implementation context than other Group 2 states. RAEs accept capitated risk for behavioral health services, creating financial incentive for care coordination with high-need populations. Many work requirement exemption candidates will be RAE members. The question is whether RAEs can identify exemption candidates through clinical relationships and proactively assist with documentation, or whether exemption processing remains separate from RAE care delivery.
Rhode Island serves approximately 80,000 expansion adults through single statewide MCO (Neighborhood Health Plan). Small scale creates delegation simplicity but concentration risk. If the MCO’s verification systems fail, all expansion adults face identical failures simultaneously. No geographic or plan variation provides natural redundancy. The state must build rigorous MCO oversight knowing implementation success depends entirely on single contractor performance.
Vermont maintains fee-for-service for philosophical reasons: better provider payment rates, direct state control, resistance to MCO profit extraction. But FFS structure means Vermont must build work requirement verification as state function. No MCO delegation distributes administrative burden. Vermont’s small state capacity (approximately 75,000 expansion adults) makes manual verification theoretically feasible, but small scale also means limited IT capacity and limited staff available for verification processing.
Connecticut similarly maintains FFS by choice, with sophisticated eligibility systems but no MCO infrastructure for delegation. The state serves approximately 225,000 expansion adults. Provider tax expiration in 2026 creates fiscal pressure. Legislative proposals for work requirements have been introduced but not advanced, revealing divided government tension between Democratic legislative majorities and implementation obligations. Connecticut must build verification infrastructure from stronger starting position than many states (sophisticated eligibility systems, mature data platforms) but without MCO delegation pathway other Group 2 states can pursue.
The pattern across smaller Group 2 states is that administrative sophistication helps but doesn’t eliminate fundamental tensions. Colorado’s RAE model enables behavioral health integration but doesn’t automatically convert care coordination into exemption processing. Rhode Island’s single-MCO simplicity creates dependency risk. Vermont’s FFS structure provides control but eliminates delegation. Connecticut’s political resistance creates implementation ambivalence even with technical capacity.
What Minimalist Implementation Looks Like#
These nine states will implement work requirements that technically satisfy federal law while functionally minimizing coverage impact. The approach will feature five characteristics.
First, expansive exemption categories capturing 60-75% of expansion adults. The federal law establishes exemption floors states cannot fall below. These states will exceed federal minimums by establishing broader medical frailty criteria, extending caregiver exemptions to older children, creating economic downturn protections triggering automatically when regional unemployment rises, and establishing good cause exceptions that function as de facto additional exemptions.
Second, maximum verification intervals federal law permits. Semi-annual verification represents the minimum frequency likely to satisfy CMS oversight while maximizing retention. Monthly verification like Arkansas 2018 is categorically rejected. Annual verification would be preferable but may face federal pushback. These states will advocate for verification cycles as infrequent as federal regulators allow.
Third, generous good cause provisions preventing terminations for correctable errors. First-time verification failures trigger 30-45 day cure periods with proactive outreach. Navigator assistance is offered before any termination. Appeals preserve coverage during review. The goal is treating documentation failure as correctable mistake rather than conclusive non-compliance.
Fourth, presumptive eligibility and automated exemption screening using existing data. Don’t wait for members to request exemptions; identify exemption candidates through claims data, SSDI/SSI enrollment, unemployment compensation, educational enrollment systems, and proactively process exemptions using data states already possess.
Fifth, robust navigation infrastructure funded to extent fiscal capacity allows. California’s Coverage Ambassador concept, Massachusetts’s ACO care managers, Oregon’s CCO community health workers, all represent navigation investment that protective implementation requires. Whether states fund this adequately depends on fiscal constraints these states cannot fully control.
The coverage losses will be smaller than enforcement-oriented states but larger than political leadership wanted. The fiscal costs will exceed early projections but fall short of comprehensive recognition systems. The implementation will satisfy federal oversight while frustrating both progressive advocates seeking universal protection and conservative critics seeking behavioral accountability.
The Dissonance of Competent Opposition#
These nine states face the dissonance of implementing well what they believe should not be implemented at all. The challenge is not technical incompetence but political-philosophical opposition. They know how to build verification systems. They possess MCO infrastructure for delegation. They operate sophisticated data platforms. But they will design those systems to protect coverage rather than enforce requirements.
The question is whether high administrative capacity enables better outcomes when political will opposes policy goals. The evidence suggests capacity matters less than commonly assumed. Connecticut’s sophisticated FFS administration cannot eliminate verification barriers created by policy design. California’s elaborate MCO infrastructure cannot overcome fiscal constraints created by floor FMAP and provider tax restrictions. Massachusetts’s mature ACO system faces quality measurement problems clinical excellence cannot solve. Oregon’s CCO innovation doesn’t change that work requirements philosophically contradict the integrated care model CCOs were built to deliver.
These states possess infrastructure advantages their peers lack. They will implement work requirements more competently than states starting from nothing. But competent implementation of opposed policy produces different tensions than incompetent implementation of supported policy. The early adopter states learned through failure what doesn’t work. The high-capacity blue states implement knowing what they build will work technically while failing philosophically to serve member interests as they understand them.
The competence paradox is that these states can build exactly what federal law requires while designing it to accomplish the opposite of what federal law intends. Technical compliance, philosophical resistance. Administrative capacity, political opposition. Sophisticated systems, protective intent. December 2026 approaches with these states knowing they will implement work requirements well precisely so those requirements harm people less than poorly-implemented versions would. Success becomes measured not by how many people maintain coverage through work but by how many people maintain coverage despite work requirements.