Why the Medicaid Statute Is a Bargain, Not a Benefit Card

Most explanations of Medicaid begin with the wrong picture. They describe a program, as though a single national health plan called Medicaid covers a defined population with a defined benefit package, and as though the details vary only at the margins from state to state. The statute tells a different story. Title XIX of the Social Security Act does not create a program in that sense at all. It creates a federal offer, written into law, extended to every state on identical statutory terms, and accepted in practice by every state through a distinct state plan. What a person experiences as Medicaid is therefore not the federal statute alone. It is the statute as a particular state has agreed to administer it, with the particular populations that state has chosen to cover, the particular benefits that state has chosen to furnish, and the particular payment methods that state has negotiated, all of it framed by federal requirements and federal money.

This distinction matters because it controls nearly every serious argument about the program. When a legislature debates whether to expand or contract coverage, the question is not what Medicaid covers in the abstract. There is no such answer. When a provider complains that reimbursement is too low, the relevant answer lives in a state plan amendment approved by the Secretary of Health and Human Services, not in the federal statute standing alone. And when reformers propose national changes, they must reckon with the fact that the federal government does not operate the program. It pays for a share of it, sets the conditions on which it will pay, and approves the documents in which states promise to meet those conditions. The rest is administration by more than fifty sovereign partners.

The most useful key for unlocking the structure is a set of three commitments, and every serious reader of the statute should memorize them first. A state that accepts federal Medicaid money promises, in the language of the statute and its implementing rules, statewideness, comparability, and freedom of choice of provider. statewideness means the program must operate across the state, not in a handful of counties the state happens to favor. Comparability means the coverage furnished to one group of beneficiaries cannot be meaningfully inferior to the coverage furnished to another, so that the federal money buys a coherent program rather than a patchwork of favors. Freedom of choice of provider means a beneficiary may obtain covered services from any qualified provider who is willing to furnish them, a requirement that once stood as a bulwark against closed networks and restricted panels. These three requirements are the core of the bargain. Nearly everything else that is interesting about Medicaid’s modern form is an exception to one of them. A reader who can name which of the three a given policy sets aside already understands more about the program than most of the commentary written about it.

That last sentence deserves emphasis, because it is the organizing claim of this article. Managed care, which assigns beneficiaries to networks and restricts which providers they may see, is an exception to freedom of choice. Home and community based services waivers, which furnish long term care services in some counties or to some populations before others, are exceptions to statewideness and comparability. Section 1115 demonstration projects, the broadest waiver authority in the statute, can set aside provisions of the state plan requirements altogether for experimental purposes, subject to Secretarial approval and to terms and conditions the Secretary imposes. The history of the program since the 1980s is largely the history of states negotiating exceptions to these three rules and the federal government deciding which exceptions to allow, under what terms, and for how long. To read the statute without the waivers is to read a law that no longer describes the program. To read the waivers without the statute is to see flexibility without understanding what it is flexing against.

The second fact to hold in mind is the complication. Medicaid is not a single national program. It is one federal statute administered as more than fifty distinct programs, one in each state, in the District of Columbia, and in the territories. The federal share of spending varies by state under a formula that reflects relative per capita income, so that federal money matters more in some states than in others. The populations covered vary, because the statute divides eligibility into mandatory groups, which a participating state must cover, and optional groups, which it may cover at its election. The benefits furnished vary, because the statute similarly divides covered services into mandatory categories and optional ones. Two beneficiaries with identical incomes, identical health conditions, and identical family situations can receive meaningfully different coverage depending on which state they live in, and that difference is not a malfunction of the program. It is the design of the statute operating as intended. Any claim about what Medicaid covers that does not name a state is therefore incomplete, and often misleading.

This article proceeds through six structural features that together explain how the statute works. First, voluntary participation, the choice Congress made to invite rather than compel the states, and the consequences that choice carries for federal power. Second, the state plan and the Secretary’s approval of it, the mechanism by which the federal government reviews and blesses each state’s design. Third, the Federal Medical Assistance Percentage, the matching formula through which federal dollars flow and through which the federal government exercises its leverage. Fourth, the division between mandatory and optional eligibility groups, which explains why the population covered differs across states. Fifth, the division between mandatory and optional benefits, which explains why the services covered differ as well. Sixth, the three core requirements and the waiver authorities that set them aside, together with the question of where the enforceability line runs, meaning which requirements can actually be compelled against a state and by whom. Taken together, these six features explain the program’s resilience, its variation, and the constant tension between federal standards and state administration that defines it. A note on reading method: wherever this article describes what the law requires, the requirement described is a requirement on a participating state, not a right a beneficiary holds directly against the federal government. Keeping that distinction in view prevents the most common misreading of the statute, which treats a condition on federal funding as though it were a federally administered entitlement. It is not. It is the price of the money, and the money is optional.

One caution on scope is in order before the analysis begins. The statute discussed here is the statute as it stood on February 1, 2011. That includes the amendments enacted in 2010, which among other things added an expansion of eligibility framed, as enacted, as a mandatory condition of state participation in the program. Whether that condition would survive judicial review, and how courts would treat the federal government’s authority to condition continued participation on acceptance of new terms, were questions being actively litigated after the date of this article, and they fall outside its scope. This article describes the law as Congress wrote it and as the executive branch administered it through the date stated above. Later litigation is not described here, and no inference about its outcome should be drawn from this account. Describing enacted law is a different task from forecasting what courts will do with it, and this article undertakes only the first of those two tasks. What follows is an explanation of the structure, and the structure is best understood on its own terms before any court has the last word on any part of it.

Title XIX and the Voluntary Compact

The Medicaid statute begins, as federal health law so often does, with an amendment to the Social Security Act. Congress added Title XIX to the Act through Public Law 89-97, the Social Security Amendments of 1965, the same legislation that created the Medicare program. The reader who wants the full account of that legislative moment will find it in this series’ companion article on the 1965 amendments, Social Security Amendments of 1965: Medicare and the Title XIX Companion. Title XIX is codified in the United States Code at 42 U.S.C. 1396 and the sections that follow it, and its central operative provision, section 1902 of the Social Security Act, appears at 42 U.S.C. 1396a. That section contains the state plan requirements, the long list of promises a state must make to qualify for federal matching funds. Everything distinctive about Medicaid’s legal architecture flows from the decision to build the program on this foundation: a title added to a Depression era statute, drafted as a cooperative federal state arrangement, and administered ever since through state plans rather than through direct federal operation.

The voluntary character of that arrangement is not a technicality. It is the premise on which the entire statute rests, and it shapes the federal government’s power in ways that are still felt in every negotiation between Washington and a state capital. Congress could have created a federal health program for low income people administered directly by a federal agency, on the model of Medicare. It chose not to. Instead, it offered the states money on conditions. A state is free to decline. If it declines, it receives no federal Medicaid funds and owes the federal government no Medicaid obligations. If it accepts, it must do so on the statute’s terms, embodied in a state plan that the Secretary of Health and Human Services must approve. The choice to build on invitation rather than command was deliberate, and it reflected both constitutional caution and administrative practicality.

The constitutional dimension is worth stating plainly. Under the spending power, Congress may attach conditions to federal money that it offers the states, but it may not commandeer the states’ legislative processes or compel them to administer a federal regulatory program. The distinction between encouragement and coercion has generated a substantial body of case law, and the precise boundary is contested. But the design of Title XIX was meant to sit safely on the permissible side of that line. By making participation voluntary, Congress gave the states a genuine choice, at least as the choice was understood in 1965, and thereby secured the constitutional footing for attaching extensive conditions to the money. A state that finds the conditions too burdensome is free, in theory, to walk away. That freedom is the consideration that makes the bargain enforceable, and it is the reason the state plan requirements can be as detailed as they are without raising the constitutional objections that a direct federal mandate would invite.

Administrative practicality pointed in the same direction. In 1965, the states already operated public assistance programs, including medical assistance programs for particular categories of the poor under the earlier Kerr Mills framework. They had caseworkers, eligibility systems, provider relationships, and administrative know how that no federal agency could replicate quickly. Congress therefore drafted Title XIX to build on existing state machinery rather than to replace it. The state plan became the instrument of that design. Each participating state would submit a plan describing how it would administer medical assistance in accordance with the federal requirements, the Secretary would review the plan for compliance with the statute, and federal matching funds would flow once the plan was approved. The federal role was standard setting, financing, and oversight. The state role was everything else: determining eligibility, enrolling providers, paying claims, and running the program day to day.

Section 1902, codified at 42 U.S.C. 1396a, is where the conditions live, and it repays close reading. The section opens with the requirement that a state desiring to participate submit a state plan, and it then enumerates the provisions such a plan must contain. The list is long, and its length is itself a statement about the program’s character. The plan must designate a single state agency to administer or supervise the administration of the program, so that federal officials know whom to hold accountable. It must provide for financial participation by the state, so that the state has its own money at stake and a corresponding incentive to administer the program responsibly. It must describe the eligibility groups the state will cover and the services it will furnish, within the mandatory and optional categories the statute establishes. It must set out the methods and standards for determining eligibility, the safeguards for the use and disclosure of applicant information, the procedures for fair hearings for individuals whose claims are denied, and the mechanisms for assuring the quality and proper utilization of care. And it must contain the three core commitments described in the opening of this article: statewideness, comparability, and freedom of choice of provider, each stated as a requirement the plan must satisfy.

The density of these requirements reflects a legislative judgment about the risks of the cooperative design. Congress understood that federal money administered by fifty different states would produce fifty different programs unless the statute imposed meaningful uniformity, and it understood that the states, left to their own devices, would face fiscal and political pressures to narrow coverage, restrict access, or favor some populations over others. The state plan requirements are the answer to those risks. They are not suggestions. A plan that fails to satisfy them may not be approved, and a state operating under an unapproved plan receives no federal matching funds. The Secretary’s approval is therefore the gate through which every state’s program must pass, and the substantive review that approval entails is one of the principal instruments of federal control over a program the federal government does not operate.

What Secretarial review actually involves is less mechanical than the phrase plan approval might suggest. The Secretary does not merely check that the state’s submission contains the required headings. The review examines whether the state’s proposed methods of administration comply with the statute and with the regulations and guidance the Department has issued to implement it. When a state seeks to amend its plan, as states do constantly, the amendment receives the same review. Disapproval is a real prospect, and the statute provides the Secretary with the authority to withhold federal funds, in whole or in part, from a state whose plan no longer complies with the federal requirements. That sanction, the power to cut off the money, is the ultimate enforcement tool, and its severity is both its strength and its weakness. It is strong because no state can afford to lose its federal Medicaid matching funds. It is weak, or at least unwieldy, for the same reason, because the federal government is understandably reluctant to impose a penalty that would immediately harm the very beneficiaries the program exists to serve. Much of the real work of federal oversight therefore happens in negotiation: in the review of plan amendments, in the conditions attached to waiver approvals, and in the steady administrative dialogue between the Centers for Medicare and Medicaid Services and the state agencies.

The voluntary design also explains a feature of the program that surprises newcomers: the states did not all join at once, and the last state to join did so nearly two decades after the statute was enacted. Arizona, the final state to enter the program, did not begin participation until 1982, when it launched the Arizona Health Care Cost Containment System, known as AHCCCS. The Arizona program entered not through an ordinary state plan but as a demonstration project under the waiver authority of section 1115 of the Social Security Act, codified at 42 U.S.C. 1315, and it became the first statewide managed care Medicaid program in the country. Arizona’s long absence from the program is better understood as proof of the design than as a historical curiosity. A state could, and one did, decline the federal offer for seventeen years, and no federal authority could compel it to join. When Arizona finally entered, it did so through the demonstration authority that has since become one of the principal instruments of state level innovation and federal state bargaining.

That Arizona entered through section 1115 points to the other half of the statutory architecture: the waiver authorities that allow the Secretary to set aside otherwise applicable requirements. Section 1115, at 42 U.S.C. 1315, authorizes the Secretary to waive compliance with the state plan requirements of section 1902 for experimental, pilot, or demonstration projects that are likely to assist in promoting the objectives of the program. The authority is broad, and its exercise has grown enormously since the 1980s. States have used it to expand eligibility beyond the statutory categories, to restructure delivery systems, to impose cost sharing or premiums not otherwise permitted, and to test managed care models on a statewide basis. Each demonstration operates under terms and conditions negotiated with the Secretary, typically for a fixed period with the possibility of renewal, and each must satisfy the statutory standard of promoting the program’s objectives. The breadth of that standard makes section 1115 one of the most consequential provisions in American health law, and the primary channel through which the modern program has diverged from the statute’s original blueprint.

Alongside section 1115 sits section 1915 of the Act, codified at 42 U.S.C. 1396n, which provides narrower and more targeted waiver authorities. Section 1915(b) permits waivers of freedom of choice and related requirements for managed care programs, allowing states to require beneficiaries to enroll in managed care plans or to restrict provider choice in specified ways. Section 1915(c) authorizes home and community based services waivers, under which states may furnish long term care services in home and community settings to individuals who would otherwise require institutional care, and may limit those services to particular geographic areas or populations in ways the statewideness and comparability requirements would otherwise forbid. These provisions are the statutory home of the two great transformations of the program’s delivery system: the shift toward managed care and the shift toward home and community based long term care. Both transformations required setting aside the three core commitments described at the outset of this article, and Congress created specific authorities for doing so rather than leaving the matter to the general demonstration power alone.

The financing structure completes the picture of the compact. Federal matching funds flow through the Federal Medical Assistance Percentage, the FMAP, established at 42 U.S.C. 1396d(b), which is section 1905(b) of the Social Security Act. The formula is stated in the statute with unusual mathematical precision. The federal share equals 100 percent minus the state percentage, where the state percentage is 45 percent multiplied by the ratio of the square of the jurisdiction’s per capita income to the square of the national per capita income. Squaring the income ratio has the effect of magnifying differences between richer and poorer states, so that the federal share rises more than proportionally as a state’s relative income falls. The statute then imposes a floor and a ceiling: the FMAP for a state may never fall below 50 percent and may never rise above 83 percent, regardless of what the formula produces. The District of Columbia receives 70 percent. Certain populations and services carry enhanced matching rates set elsewhere in the statute, with family planning services matched at 90 percent as the most prominent example. The formula is recalculated on the basis of updated income data, so that a state’s federal share moves with its relative economic position over time.

The FMAP does more than move money. It is the mechanism through which the federal government calibrates its fiscal commitment to each state, and through which the voluntary compact acquires its financial logic. A poorer state receives a higher federal share, which lowers the effective price of each additional dollar of program spending and encourages broader coverage. A wealthier state receives the 50 percent floor, which means it pays half of every dollar itself and faces the full marginal cost of expansion. The enhanced rates for particular services operate as targeted incentives, steering state behavior toward services Congress has deemed especially worthy of federal support. And because the federal match is open ended, meaning the federal government matches state spending without a cap, the formula creates a standing incentive for states to spend, constrained mainly by their own willingness to appropriate the state share. Critics have long argued that open ended matching encourages spending growth and dulls state incentives for efficiency. Defenders respond that it is precisely what allows the program to respond to recessions and other surges in need without congressional action, since federal dollars automatically follow state spending upward when need rises. The statute does not resolve the argument. It creates the conditions under which the argument continues.

Enforceability is the final structural feature, and it is the one most often misunderstood. The state plan requirements are legally binding on participating states, but the question of who can enforce them, and how, is narrower than it first appears. The Secretary may disapprove a state plan or plan amendment, and may withhold federal funds from a noncompliant state, subject to administrative procedures and judicial review. Beneficiaries and providers have at times sought to enforce particular requirements through private lawsuits, and the availability of such suits has depended on whether the provision at issue creates an individually enforceable right, a question courts have answered differently for different provisions across different eras. What is clear as a structural matter is that the primary enforcement relationship runs between the federal government and the jurisdiction, not between the beneficiary and either sovereign. The beneficiary’s protections flow through the state’s plan and the federal oversight of it, rather than through direct federal administration.

Stepping back, the voluntary compact can now be stated as a whole. Congress offered the states an open ended federal match for medical assistance to specified populations, conditioned on compliance with an extensive list of requirements embodied in a federally approved state plan. The states were free to decline, and one did for seventeen years. The requirements include the three core commitments of statewideness, comparability, and freedom of choice, and the statute provides waiver authorities, general and specific, through which those commitments may be set aside under federal supervision. The financing formula ties the federal share to relative state income within a statutory floor and ceiling. Enforcement runs primarily through the Secretary’s control of the money. Every distinctive feature of the program, from its variation across states to its constant administrative negotiation to its periodic reinvention through demonstrations, follows from this architecture. The statute is not a program. It is the terms of a bargain, and the program is what the states have built within those terms.

Why did Congress make Medicaid participation voluntary for states?

Congress chose the voluntary design for constitutional safety and administrative practicality. A federal mandate conscripting state legislatures would have invited serious constitutional challenge, while an offer of money on conditions sat on firmer ground. The states also possessed the eligibility systems, caseworkers, and provider networks that no federal agency could replicate quickly in 1965.

The voluntary structure did more than solve a constitutional problem. It bought political durability. A program the states chose to join generated a different politics than a program imposed on them, and governors and state legislatures became stakeholders rather than conscripts. Arizona’s long delay in joining demonstrates that the option to decline was real, and the fact that every state eventually joined demonstrates the power of the financial offer.

What does Secretarial approval of a state plan actually decide?

Secretarial approval decides whether a state’s proposed administration of medical assistance satisfies the federal statutory requirements and therefore qualifies for federal matching funds. The Secretary reviews the plan and its amendments for compliance with section 1902 and the implementing regulations, and approval is the legal predicate for the flow of federal dollars.

The review is substantive rather than ministerial. The Department examines the state’s eligibility methods, benefit descriptions, payment methodologies, and administrative procedures against the statute’s requirements, and it negotiates changes where it finds deficiencies. Disapproval blocks federal funding for the noncompliant portion of the program. Because withdrawing funds would harm beneficiaries, the sanction functions mainly as a threat that concentrates the minds of state officials during negotiation, and the real decisions emerge from the iterative process of submission, review, revision, and approval, a process that never fully ends because states amend their plans continuously.

With the compact described, the next step is to examine each of its mechanisms in turn, beginning with the money that makes the bargain work.

The money: FMAP and the federal match

The federal government does not fund Title XIX with a lump sum. It matches what each participating state spends, at a rate the statute assigns to that state, and the assignment runs through a single mechanism: the Federal Medical Assistance Percentage, the FMAP. The formula lives at 42 U.S.C. 1396d(b), which is section 1905(b) of the Social Security Act in the Act’s own numbering. The two citations describe the same provision, one in the United States Code and one in the Act as Congress wrote it, and readers who want the full mapping between the two systems can consult the guide to reading a federal statute. Nearly everything fiscal in the program flows from this mechanism: the federal share of each qualifying state expenditure, the floor and ceiling that bound that share, and the enhanced rates the statute sets for particular categories.

The formula works in two steps. First it computes a state percentage equal to 45 percent multiplied by the ratio of the square of the jurisdiction’s per capita income to the square of the national per capita income. Then it subtracts that state percentage from 100 percent, and the remainder is the FMAP, the federal share of program costs. The relationship is inverse: the higher a state’s per capita income, the larger its state percentage and the smaller the federal share; the lower the jurisdiction’s per capita income, the smaller its state percentage and the larger the federal share. Per capita income serves here as the statute’s proxy for fiscal capacity, a rough measure of the revenue a state can raise from its own residents. The formula asks nothing about a state’s actual spending, its tax rates, or its health needs. It asks only how the income of its people compares with the income of the nation’s people, and it sets the federal share from that comparison alone. Two states that spend identically on identical populations can therefore draw very different federal shares, because the match prices fiscal capacity rather than effort. A hypothetical state modestly above the national average and one modestly below it face different federal prices for the same program: the poorer of the two draws more federal cents per state dollar, which narrows but does not erase the gap in what each state can afford from its own revenues.

The most consequential design choice inside the formula is the squaring. A plain ratio of state to national per capita income would already tilt the federal share toward poorer states, but the statute squares both incomes before dividing, and that single operation reshapes the match across the entire range of states. The effect is mathematical rather than mysterious, and it determines how much federal support each state actually draws per dollar it spends.

Why does the FMAP formula square per capita income?

Squaring makes the federal share respond more than proportionally to differences in state fiscal capacity. A state at half the national per capita income does not receive a modestly higher match; the square drives its state percentage down sharply, which pushes the federal share up. Poorer states thus receive progressively larger federal support per dollar spent.

The square amplifies the formula’s tilt in both directions at once. Consider the arithmetic as an illustration of the formula rather than a claim about any particular state. A state at half the national per capita income produces a ratio of one half; squared, that ratio becomes one quarter, so the state percentage is 45 percent of one quarter, or 11.25 percent, and the federal share is 88.75 percent before the ceiling applies. Without the square, the same state would face a state percentage of 22.5 percent and a federal share of 77.5 percent. The square thus moves this hypothetical poorer state from a strong match to the maximum the statute allows. The mirror image holds for wealthier states: squaring pushes their state percentages up faster than a plain ratio would, driving the federal share down toward the floor. In practice the floor and ceiling truncate the formula’s extremes, which means the squaring does its real work in the middle of the range, differentiating states of middling income more sharply than a linear ratio ever could. A linear ratio would cluster the middle states close together; the square spreads them out, so that modest differences in per capita income produce meaningful differences in the federal share.

The 45 percent constant in the formula deserves attention of its own. It scales the squared income ratio before the subtraction from 100 percent, and its size determines where the formula’s raw output lands relative to the floor and ceiling. A larger constant would push more states against the 50 percent floor; a smaller one would push more states against the 83 percent ceiling. The constant therefore centers the distribution: it positions the formula so that the squaring does the differentiating work in the middle of the income range while the bounds catch the extremes. Like the squaring and the bounds, the constant is a policy choice embedded in arithmetic, and changing any one of the three would re-price the program for many states at once.

The statute then bounds the result. No state’s FMAP falls below 50 percent, and none rises above 83 percent. The floor guarantees that even the wealthiest participating state pays no more than half of each program dollar, with the federal government paying the other half, and that guarantee keeps participation attractive across states of very different wealth: no state is asked to carry the program largely alone. The ceiling bounds the federal obligation in the poorest states. Without it, the formula could push the federal share toward the whole cost of the program in the lowest-income states, and the statute caps that exposure at 83 percent. Together the two bounds compress the formula’s output into a fixed band, so the effective federal share always runs from one half to just over four fifths of each qualifying dollar. The band also steadies state budgeting, because no state’s share can swing past the guardrails no matter how far its relative income moves.

One jurisdiction sits outside the formula entirely. The District of Columbia receives a fixed FMAP of 70 percent. The statute assigns this rate directly rather than computing it from per capita income, which means the District’s federal share does not move with relative economic performance the way a formula state’s share does. The fixed rate removes the District from the movement of the formula. That stability cuts both ways: the District cannot see its federal share rise when its relative income falls, but it is likewise insulated when the formula would otherwise move against it. The statute trades responsiveness for predictability for the District, a different bargain from the one it strikes with the states.

On top of the base FMAP, the statute assigns enhanced matching rates to particular services and populations. Family planning services match at 90 percent, a rate well above any state’s ordinary FMAP, and the breast and cervical cancer treatment group carries an enhanced FMAP under 1396d(b)(4). Administrative costs, by contrast, generally match at a flat 50 percent, untethered from per capita income. The economics are straightforward. An enhanced rate lowers the effective price to the jurisdiction for the favored category, which makes the jurisdiction more likely to furnish it; a 90 percent match means the jurisdiction pays ten cents of each family planning dollar, a price few states will refuse. The flat administrative rate separates the cost of running the program from the cost of the benefits it pays, and it does not vary with state wealth, so a poor state and a rich state face the same federal terms for administration.

The enhanced rates reveal a second use of the match: steering. The statute does not merely price each state’s spending; it prices categories differently to pull state behavior toward federal priorities. A 90 percent match for family planning does not order any state to furnish those services, but it makes refusal expensive in forgone federal dollars, which is a gentler instrument than a mandate and a stronger one than a suggestion. The enhanced rate for the breast and cervical cancer treatment group works the same way, attaching favorable financing to a population the statute favors. Differential matching is thus the statute’s way of expressing priorities without adding to the mandatory list: it buys the behavior it wants at a discount to the states.

Because the formula turns on relative per capita income, a state’s FMAP is not fixed. A state whose economy grows faster than the nation’s sees its state percentage rise and its federal share drift down; a state that falls behind the nation sees the reverse. The federal share thus tracks relative economic fortunes, and every shift in the ratio re-prices the program for the jurisdiction. This movement is gradual rather than sudden, since per capita income itself moves gradually, but it means the federal bargain is re-struck continuously: the same program costs the jurisdiction a different share from one period to the next as its economy converges with or diverges from the national average.

The formula is also notable for what it ignores. It takes no account of a state’s poverty rate, its disease burden, its health care prices, or the size of its eligible population; two states with identical per capita incomes receive identical FMAPs even if one faces far greater health needs than the other. Per capita income is the statute’s single measure of ability to pay, and the match therefore equalizes fiscal capacity rather than need. The text commits to that measure without resolving whether it is the right one, and every fiscal argument about the program starts from that commitment rather than from a freestanding theory of fairness.

The matching design itself deserves a structural note. The statute could have funded the program entirely from the federal treasury or left it entirely to the states; instead it requires both to pay for every dollar. That shared payment disciplines both sides. The jurisdiction cannot expand the program without appropriating its own share, which forces state legislators to weigh each expansion against other state priorities. The federal government cannot restrain spending by unilateral fiat, because its payment is a function of state choices rather than a number it sets. Each level thus holds a check on the other: the jurisdiction checks federal ambition by controlling the spending base, and the federal government checks state retrenchment by making every state dollar cut forfeit matching federal dollars. The FMAP calibrates how heavy each side’s share of that joint burden feels.

The match is open-ended. The statute places no aggregate cap on the federal payment: every qualifying dollar a state spends draws its FMAP share from the federal treasury, so the federal outlay rises and falls with state spending. This shapes state fiscal behavior in a direct way. A state with a 50 percent FMAP pays fifty cents of each program dollar and draws fifty federal cents alongside it; a state at the 83 percent ceiling pays seventeen cents and draws eighty-three. The lower the jurisdiction’s effective price, the stronger the incentive to extend covered services and populations, because each additional state dollar buys more than a dollar of program. The same structure exposes the federal treasury: when states spend more, the federal government automatically spends more, with no statutory brake beyond the FMAP percentages themselves. Neither side of that bargain can move unilaterally without moving the other, which is the point of a matching design: the jurisdiction cannot expand without spending its own funds, and the federal government cannot cap its exposure without changing the statute.

The open-ended match also has a limit worth stating: it sets the price of the program but does not insure the jurisdiction against its own revenue shortfalls. When a state’s economy contracts in step with the nation’s, its relative per capita income barely moves, so the FMAP may hold steady while the jurisdiction’s ability to fund its share deteriorates. The federal government keeps paying its percentage, but the jurisdiction’s percentage must still be appropriated from shrinking revenues. The match shares costs; it does not guarantee that the jurisdiction’s share will be affordable when the jurisdiction needs the program most. That gap between shared cost and fiscal capacity is built into a formula that measures relative income rather than budget stress.

That open-ended design is best understood against its structural alternative, capped or block-style funding, in which the federal government would pay each state a fixed sum regardless of what the jurisdiction spends. Under a cap, the marginal federal contribution of the next state dollar is zero; under the match, it is positive at the FMAP rate, and that single difference rearranges every incentive. The strongest case for the open-ended match is that it tracks real costs. When enrollment grows or prices rise, federal dollars grow with them, so states are not left to absorb economic downturns alone, and the federal government retains a financial stake in the program’s terms, which preserves its leverage over how the program runs. The strongest case for a cap is fiscal control and state discretion. A fixed federal payment is predictable in the federal budget, a state that holds its program within the cap keeps the savings, which rewards efficiency, and fixed funding typically arrives with fewer federal conditions, which enlarges state autonomy. The supported verdict is structural rather than partisan: the two designs allocate fiscal risk differently. The match places the risk of cost growth on the federal ledger and preserves federal leverage over program terms; a cap would place that risk on the jurisdiction and enlarge state discretion. Choosing between them is choosing who bears the risk when costs move, and the statute, through the FMAP, assigns that risk to the federal treasury while keeping the states as spending partners.

Who must be covered: mandatory and optional eligibility

Eligibility under Title XIX is not a single national standard. The statute defines categories of people, assigns some to a mandatory list every participating state must cover and others to an optional list each state may cover at its election, and attaches the federal match to the jurisdiction’s choices within that architecture. The categories are defined by the statute’s text, not by state preference, and the federal match flows only to spending on people the statute places in one of the two lists. The mandatory-optional distinction is the load-bearing wall of the program’s eligibility design: benefits, financing, and state variation all hang from it, and every eligibility question is really two questions, which category the person falls into and what the person’s state has elected.

What separates a mandatory eligibility group from an optional one?

A mandatory group is one the statute requires every participating state to cover; refusing it means forfeiting federal matching funds. An optional group is one the statute permits a state to cover at its election, with federal matching available if it does. The difference is compulsion versus permission, and it is the statute’s central eligibility architecture.

The distinction operates as a condition of participation. A state that covers all mandatory groups and complies with the statute’s terms draws federal matching funds; a state that refused a mandatory group would step outside the bargain and lose the federal money that comes with it. An optional group carries no such compulsion: the jurisdiction may cover it and draw matching funds for its members, or decline it and remain fully compliant. The mandatory list is therefore the national floor of the program, identical in every state, while the optional list is where states write their own variations above that floor. The two lists together define the universe of people for whom federal matching is available; no one outside the lists draws federal funds no matter how great the need, and everyone inside the mandatory list must be covered regardless of state preference. The asymmetry is deliberate in its structure: the jurisdiction may add groups above the floor at will, but it may not subtract groups from below it. The floor is one-directional, and that one-directionality is what makes the federal guarantee credible.

Eligibility is also the gatekeeper of the federal match. The FMAP determines the federal share of each dollar, but the eligibility lists determine which dollars qualify at all: spending on a person outside both lists draws no federal match, while spending on a person inside a covered group draws it at the jurisdiction’s rate. This makes the eligibility architecture the program’s fiscal perimeter. Debates about benefits and payment rates matter only within the perimeter the eligibility rules draw, which is why the mandatory-optional distinction carries more fiscal weight than any single benefit decision.

The mandatory groups form the program’s core. The statute requires coverage of SSI recipients who are aged, blind, or disabled, tying health coverage to the federal cash-assistance program for those populations, so that the poorest aged, blind, and disabled people carry health coverage as a complement to their cash support. It requires coverage of families linked to the former AFDC cash-assistance categories, preserving the historical connection between cash aid and health coverage for low-income families with children. It requires coverage of pregnant women and children at poverty-related thresholds the statute sets in its own text, the income-based categories that reach beyond any linked cash program to protect mothers and children on the basis of low income alone. And it requires coverage of children in foster care, a categorical group defined by circumstance rather than income, reflecting the jurisdiction’s direct responsibility for children in its custody. The common thread is a federal determination of who must be protected in every state without exception. A person in one of these groups must be covered in any participating state, and the jurisdiction has no discretion to exclude the group, narrow its definition, or condition its coverage on state budget preferences.

The family category carries a structural feature worth noting. The statute ties eligibility to the former AFDC cash-assistance categories, which embeds a historical definition inside the eligibility rules: the reference point is fixed in the past rather than tracking later changes in cash assistance. Family eligibility is therefore measured against preserved categories instead of a live program’s current terms. Whatever the reasons for freezing the reference, the effect is that one of the largest mandatory groups is defined by history rather than by current program design.

The mandatory list is what makes the program national rather than merely federal-state. Because these groups must be covered everywhere, a person who qualifies in one state as an SSI recipient or a foster child qualifies on the same categorical basis in any other participating state; the floor travels with the category, not with state generosity. This portability is the federal guarantee inside a state-administered program: the statute does not leave the core populations to state discretion, and it uses the prospect of lost federal funds to make the guarantee stick. Everything above the floor is negotiation; the floor itself is not.

The optional groups let states extend the program’s edge at their election. The most significant is the medically needy: people whose income exceeds the mandatory thresholds but who incur large medical expenses. The statute’s spend-down concept makes this group workable. The jurisdiction subtracts a person’s incurred medical expenses from countable income and tests the remainder against the eligibility threshold, so medical bills large enough to push the remainder below the line bring the person into coverage. The mechanism recognizes that income alone mismeasures need when illness strikes: a family above the line on paper can be below it once the hospital bills are counted, and the spend-down converts catastrophic medical costs into the very evidence that qualifies the person for help with those costs. In effect the statute asks not what the person earns but what the person has left after the illness has taken its share: a worker whose income sits above the threshold but who faces medical bills larger than the gap between income and threshold spends down into eligibility, while a neighbor with identical income and no medical bills remains outside. Need is measured net of medical catastrophe rather than gross of it.

The medically needy category thus creates two different measures of poverty inside the program: poverty by income, which the mandatory thresholds capture, and poverty after medical costs, which only the spend-down captures. A state that elects the category effectively agrees that the second measure counts; a state that declines it treats income alone as the test. The election is the jurisdiction’s in both directions, and the federal match follows whichever choice the jurisdiction makes. Other optional groups follow the same elective logic. States may cover persons with tuberculosis, a public health category the statute singles out for optional coverage. They may cover women found through the breast and cervical cancer early detection screening program, connecting the screening program’s findings to treatment coverage at state option. They may cover working persons with disabilities under buy-in options that let those workers contribute toward the cost of their coverage, so that employment does not cost a disabled worker health protection. And they may cover aged, blind, or disabled persons with incomes above SSI levels where the jurisdiction so elects, extending the aged, blind, and disabled categories past the cash program’s income line. Each is a doorway the statute builds and leaves to the jurisdiction to open, and each, once opened, draws federal matching funds at the jurisdiction’s FMAP.

Because each optional group is a separate election, states assemble very different programs from the same menu. One state may cover the medically needy, the tuberculosis group, and the working disabled buy-in while its neighbor covers none of them; both draw federal matching for what they cover and both remain compliant. The statute thus produces many variations on a common theme rather than a single national program above the floor, and the differences reflect state politics, state budgets, and state judgments about need rather than any federal ranking of the groups.

The health reform legislation enacted in 2010 added a new mandatory eligibility group, making its coverage a condition of state participation as enacted. That addition belongs to the mandatory architecture described here: a category the statute requires rather than permits, with the same compulsion and the same federal matching behind it. As enacted, the new group sits alongside the older mandatory categories rather than replacing any of them, so the mandatory floor grew while the optional list stayed intact. Disputes over that requirement reached the federal courts after this article’s date, and that litigation falls outside the article’s scope.

Because the addition is mandatory, it also illustrates the stakes of the mandatory-optional line. A category placed on the mandatory side binds every participating state at once; the same category placed on the optional side would have produced a patchwork, covered in some states and absent in others. Where the statute draws that line determines whether a policy is national or merely available, and the 2010 addition drew it on the national side.

The optional list also carries a fiscal logic worth stating plainly. When a state elects an optional group, the federal match follows the election: every qualifying dollar spent on the newly covered population draws federal funds at the jurisdiction’s FMAP, including any enhanced rate the statute assigns to the category. Expansion is therefore subsidized, and a poorer state with a high FMAP pays only a fraction of each expansion dollar while the federal government pays the rest. The match lowers the price of saying yes, and there is no symmetric reward for saying no, which is why the optional architecture tends to expand as states discover that each new doorway costs them less than its face value. A state weighing whether to open a new doorway therefore asks not what the coverage costs in total but what it costs the jurisdiction after the match, and the match systematically makes the answer smaller than the sticker price. The federal subsidy does not compel the jurisdiction’s choice, but it shapes it, and understanding any state’s optional coverage requires understanding the price the match set for it.

This architecture explains why two states can both comply with the same statute and cover different populations. Compliance is measured against the mandatory floor, not against uniformity above it: a state covering only mandatory groups satisfies the statute, and so does a state covering mandatory groups plus every optional group it has elected. Neither state violates the text; they exercise the discretion the text grants. The result is a program that is national in its floor and local in its ceiling, and the variation is designed in rather than tolerated as noncompliance. Observers who expect identical programs in every state misread the design: the statute does not promise identical programs, it promises an identical floor beneath different programs.

The structural consequence is that eligibility questions have no national answer. Whether a particular person qualifies depends first on which state the person lives in, because the jurisdiction’s elections determine which optional groups exist there at all. Analysts, applicants, and advocates must therefore begin with the jurisdiction: its mandatory coverage, which is the same everywhere in principle, and its optional elections, which are the same nowhere else. The statute provides the categories and the rules for choosing among them; the jurisdiction provides the final shape of the program its residents actually encounter. Any account of who the program covers that omits the jurisdiction has omitted the mechanism that decides the question.

For the applicant, this means the federal statute alone never answers the eligibility question; it supplies the categories and the rules, but the jurisdiction’s elections supply the answer. Two neighbors with identical incomes and identical medical bills can face different outcomes on opposite sides of a state line, both correctly under the same federal text. That is not a malfunction of the design. It is the design: a national floor that guarantees the core, and state elections that shape everything above it.

Money and eligibility set the perimeter of the program. What the program actually buys within that perimeter is the next question.

What must be covered: mandatory and optional benefits

Title XIX does not hand the states a single benefit package. It hands them two lists. One names the services every state plan must cover for the categorically needy; the other names the services a state may cover if it chooses. The first list is the federal floor, and no state plan is approved without it. The second is the menu from which each state builds upward, and no two states build in quite the same way. Every benefit a state offers, from the hospital ward to the dental chair, begins with the question of which list it sits on. The mandatory list is codified at 42 CFR 440.210, and it repays close reading, because the services Congress made nonnegotiable reveal the shape of the program Congress intended to buy.

The mandatory services begin with the core of acute care: inpatient hospital services, outpatient hospital services, physician services, and other laboratory and x-ray services. These four are the spine of the benefit, and their logic is architectural rather than political. A program that paid for hospital stays but not the physicians who staff them, or for physicians but not the tests on which diagnosis depends, would not be a health program in any recognizable sense. The statute does not permit a state to assemble the benefit piecemeal at this level; the acute-care core travels together, and a state plan that omitted any of the four would fail the federal floor before the analysis even reached the more specialized services. Whatever else a state does with its program, the emergency room, the ward, the doctor, and the laboratory are not negotiable. The word “other” in “other laboratory and x-ray services” does quiet work here: it reaches the tests ordered outside the hospital walls, so that the diagnostic core follows the patient from the ward to the clinic.

From the hospital the list moves to the settings where prolonged need lives. Nursing facility services for individuals twenty-one and older are mandatory, and home health services stand beside them as the community-based counterpart, so that the statute covers both the institution and the alternative to institutional care. The pairing matters because it expresses a preference without imposing one: the jurisdiction must pay for the nursing facility bed, and it must also pay for the home health visit that may keep a beneficiary out of that bed. Family planning services and supplies are mandatory as well, which places reproductive care inside the basic package rather than among the extras, and guarantees that a service whose consequences reach far beyond the clinic visit cannot be treated as discretionary.

The statute then names the providers it refuses to leave out of a state plan. Rural health clinic services and federally qualified health center services are mandatory, a guarantee that the clinics serving remote counties and low-income neighborhoods cannot be cut when a budget tightens, because the communities with the fewest alternatives are the ones the floor protects most explicitly. Nurse-midwife services and nurse practitioner services are mandatory too, an acknowledgment written into the benefit list that much of the care received by the poorest patients comes from clinicians other than physicians, and that a benefit covering only physicians’ services would miss where the care actually happens. Pregnancy-related services complete this portion of the list, tying the mandatory benefit to one of the life events around which the program was most plainly organized. Taken together, these entries show a list built from observed reality: where poor patients go, who treats them, and which needs cannot wait for a legislature to get around to them.

Read as a whole, the mandatory list has a discernible logic, and it is worth naming because it explains why these services and not others. The list covers the episodes no one can postpone, which is the hospital and physician core. It covers the life stages in which need is concentrated and deferral is dangerous, which is pregnancy, childhood, and frail old age. And it covers the providers who serve the populations the private market serves least, which is the rural clinic, the community health center, and the nurse practitioner. Nothing on the list is there by accident. Each entry answers a version of the same question: what would it mean to promise health coverage to the poorest residents of a state and then leave this out? The services that survive that question are the mandatory ones.

The list also uses age twenty-one as a hinge, and the symmetry is deliberate. Nursing facility services are mandatory for individuals twenty-one and older; early and periodic screening, diagnostic and treatment services are mandatory for individuals under twenty-one. The statute divides the life course at the threshold of adulthood and assigns each side its characteristic need: the institution for the frail elderly, the screening sequence for the growing child. A beneficiary who ages across the line moves from one mandatory regime to another, and the state plan must be built to receive her on both sides.

The broadest mandatory benefit belongs to children. Early and periodic screening, diagnostic and treatment services, known as EPSDT, must be covered for every beneficiary under twenty-one. The benefit is a sequence with no gaps: screen at intervals to discover what is wrong, diagnose what the screening finds, and treat what the diagnosis identifies. The statute pairs the screening duty with a treatment duty, so that what screening discovers, the program must address. For adults the line between mandatory and optional services holds; for children EPSDT is the high-water mark of the statute’s demands, because the treatment obligation follows the child’s condition rather than stopping at the boundary of the mandatory list. The asymmetry reflects a judgment about time: a condition discovered in childhood can be treated before it hardens into the adult disability the program would otherwise carry for decades, so the statute places its broadest mandate where early action buys the most health. No other mandatory benefit reaches as far into a beneficiary’s needs, and no state may narrow the sequence or defer its steps. The clinicians who treat children in the program work inside this sequence whether they name it or not: the screening visit creates the obligation that the follow-up visit satisfies.

The optional list is shorter than the mandatory one but just as consequential in practice. A state may elect to cover prescription drugs, dental services, personal care services, physical therapy, eyeglasses, prosthetic devices, intermediate care facility services, rehabilitative services, and tuberculosis-related services. None of these is required of any state, and all of them are available to every state. The distance between those two sentences is where state variation lives. Two states can both satisfy the federal floor while offering visibly different programs, because one has drawn heavily on the optional menu and the other has left most of it alone. The mandatory list makes the states alike; the optional list makes them different, and the difference is deliberate, because the statute assigns each state the task of judging what its residents need beyond the floor.

The texture of the optional list is worth attention, because its specificity reveals how Congress thought about state choice. Some entries name discrete, well-understood needs: eyeglasses, dental services, prosthetic devices. Others name modes of care rather than items: personal care services, physical therapy, rehabilitative services, tuberculosis-related services. Still others name settings: intermediate care facility services. A state electing from this menu is not choosing abstractions; it is deciding whether an older beneficiary who needs help bathing gets it, whether a child who needs glasses gets them, whether a beneficiary recovering from a stroke gets the therapy that restores function. These are the decisions legislatures recognize from their own districts, which is why the optional list, for all its federal origin, behaves in practice like an instrument of state judgment.

In practice the menu is heavily used, and the programs that result are far richer than the mandatory list alone would suggest. State after state covers long rosters of optional services, filling in the gaps the federal floor leaves open: the teeth, the eyes, the prescriptions, the therapies that restore function after illness or injury. Medical need does not respect the line between the two lists, and state legislatures answer to the patients and providers who live with the consequences of where the line is drawn. The statute permits generosity. It is each state’s own appropriations process, not the federal floor, that usually sets the limit on how far up the menu a state climbs. The program therefore looks spare on paper, in the mandatory list alone, and substantial in operation, in the state plans that layer optional benefits onto the floor.

The most instructive entry on the optional list is prescription drugs. Ask beneficiaries whether the program covers their prescriptions and the answer will be yes in every state, because every state covers prescription drugs in some form. The statute nevertheless classifies the benefit as optional, and the classification is not a technicality. It is the source of the most common misconception about the benefit structure: universality of practice is mistaken for a federal command, and the drug benefit is assumed to be mandatory because no state has chosen to go without it. The distinction matters for who decides. A mandatory benefit is defined by Congress; an optional benefit is defined by the state plan that elects it. The federal government can hold a state only to the promise the jurisdiction made, and the promise to cover prescription drugs was made by each state rather than imposed by the statute. That is why the drug benefit can differ in its contours from state to state even though it exists in all of them: the coverage is universal, but its shape is state-drawn, and a state revising its drug benefit answers to its own plan rather than to a federal mandate. Beneficiaries experience the benefit as an entitlement, and for practical purposes it functions as one; the statute’s classification changes nothing about the prescriptions a pharmacist dispenses in any given week. It changes everything about who could revise the benefit in a future budget year, and under what authority the revision would proceed.

The two lists interact with state choice at every point, and the interaction runs through the state plan. The benefit lists tell a state what it must cover and what it may cover, and the state plan is the document in which the jurisdiction records its answers: the mandatory services it will furnish, the optional services it elects, and the terms on which each will be offered. A state that elects dental services for one categorically needy group will discover, in the next section’s comparability requirement, that the election cannot easily be confined to that group. Three further requirements, statewideness, comparability, and freedom of choice, govern how evenly the chosen benefits must be offered and how freely beneficiaries may obtain them, and three waiver authorities allow a state to set those requirements aside for defined purposes. The benefit lists are the what of the program’s coverage. The requirements and the waivers are the how, and they are the subject of the next section. Once approved, the state plan is the program as the jurisdiction’s beneficiaries experience it: the lists are federal, but the plan is the jurisdiction’s, and the plan is what the clinic sees.

The three-requirement key

If the benefit lists are the what of Title XIX, three requirements are the how. Every state plan must satisfy them as the price of federal funds: the plan must operate statewide, it must furnish comparable assistance to the categorically needy, and it must let beneficiaries choose their providers. These are not suggestions, and they are not defaults a state may quietly revise. They are the commitments a state makes in exchange for federal money, and they shape every benefit design the jurisdiction attempts, because they answer three questions the benefit lists leave open: where must the program operate, how evenly must it treat its beneficiaries, and how freely may those beneficiaries obtain care. The waivers that set the requirements aside are best understood against them, since a waiver is simply permission to break, for a defined purpose, a rule that would otherwise bind.

How do the three core requirements limit state plan design?

They lock a state into three commitments once it takes federal funds. statewideness bars county-by-county benefit designs. Comparability bars giving one categorically needy group less in amount, duration, or scope than another. Freedom of choice bars steering beneficiaries to preferred providers. A state may depart from these only through an approved waiver.

statewideness, at 42 U.S.C. 1396a(a)(1), is the simplest of the three and the easiest to picture. It forbids a state from operating the program in some counties and not others, from confining a benefit to the counties that can most easily afford it, and from piloting a richer package in one region while the rest of the jurisdiction makes do with less. The program must be the program everywhere inside the jurisdiction’s borders; geography may explain why delivering care is harder in one county than another, but it may not explain why the benefit differs. The requirement is waivable under section 1115 and section 1915(c), and the waiver is what allows a demonstration or a home and community-based program to operate in selected counties. Without the waiver, a county line is not a permissible basis for distinguishing among beneficiaries.

The requirement also does quiet work against a subtler temptation. Without it, a state could concentrate its generosity where its voters are watching and its frugality where they are not, offering the full program in the capital county and a thinner one two counties over, and no beneficiary in the thinner county would have a federal claim against the difference. statewideness converts the state plan from a collection of local bargains into a single statewide promise. It is also the requirement that most visibly yields to the waiver authorities, because demonstrations are by nature local: a state cannot test an approach everywhere at once, and section 1115 exists to let it test the approach somewhere first.

Comparability, at 42 U.S.C. 1396a(a)(10)(B), governs the horizontal equity of the plan. It requires that the medical assistance furnished to any categorically needy group be equal in amount, duration, and scope to that furnished to any other such group. Amount is how much of a service a beneficiary may receive; duration is how long the coverage of it lasts; scope is how broad the covered services are. What the requirement forbids is tiering: a state may not furnish one categorically needy population a generous package and another a lean one, even if the lean package covers every mandatory service on the federal list. This is where the optional menu meets its discipline. Once a state elects an optional benefit, comparability spreads the election across the categorically needy; the jurisdiction that covers physical therapy for one group covers it, in equal amount, duration, and scope, for the others. Like statewideness, comparability is waivable under section 1115 and section 1915(c), which is what permits a waiver to define its own population and furnish that population, and only that population, the waiver’s services.

The triplet of amount, duration, and scope closes the loopholes a leaner standard would leave open. A state bound only to cover the same services for each group could cover them thinly for the group it favors least: the same physician benefit in name, but fewer visits in amount, shorter authorization windows in duration, narrower covered diagnoses in scope. Comparability forbids all three evasions at once by measuring equality across each dimension. Consider a state that elects to cover physical therapy, an optional benefit. Comparability does not oblige the jurisdiction to cover physical therapy at all; the election is the jurisdiction’s alone. But once elected, the benefit must reach every categorically needy group in equal amount, duration, and scope, so the jurisdiction cannot offer twenty visits to one group and five to another. The jurisdiction’s freedom is the freedom to elect, not the freedom to discriminate among the groups it has elected to serve.

Freedom of choice of provider, at 42 U.S.C. 1396a(a)(23), protects the beneficiary’s side of the transaction. It provides that a beneficiary may obtain covered services from any qualified provider willing to furnish them, and it forbids the jurisdiction from conditioning coverage on the use of a state-preferred doctor, clinic, or plan. A state may inform beneficiaries about their options, and it may encourage them toward particular providers, but it may not compel them, and it may not build a benefit design that compels them indirectly by paying only the preferred. The requirement is waivable under section 1115 and section 1915(b). The 1915(b) authority exists for exactly this purpose: it lets a state require beneficiaries to enroll in managed-care plans with restricted provider networks, trading the beneficiary’s free choice of any willing provider for enrollment in a plan the jurisdiction has chosen.

The requirement sits in productive tension with the direction state programs have taken. Managed care promises coordination and predictability, and both are easier to deliver when beneficiaries are enrolled in defined plans with defined networks; freedom of choice promises the beneficiary the last word on who provides her care, and it is hardest to honor inside exactly those arrangements. The statute resolves the tension the way it resolves most tensions in Title XIX: the rule stands, and the waiver provides the exception. A state that wants the managed-care model without the free-choice rule must seek section 1915(b) or section 1115 permission and accept the federal terms on which that permission is granted, rather than building the restriction into its state plan and daring the federal government to object.

The three requirements reinforce one another, and their combined effect is stronger than any one of them alone. statewideness without comparability would let a state offer a thin program everywhere; comparability without statewideness would let a state offer a generous program somewhere; freedom of choice without either would let beneficiaries choose freely among the providers of a program that varied by county and by group. Together they describe a single commitment: the same program, for the same groups, obtainable from the providers the beneficiary chooses, in every county of the jurisdiction. The waivers then unbundle that commitment, lifting one requirement here and another there, which is why the table at the end of this section matters: it shows exactly which strand each authority may loosen, and which strands no authority may touch.

The three waiver authorities are the statute’s pressure valves, and each is aimed at a different kind of departure. Section 1115, the research and demonstration authority at 42 U.S.C. 1315, is the broadest: it allows the federal government to set aside state-plan requirements so that a state may test an approach the regular rules would forbid, and it reaches all three of the requirements described above. Section 1915(b), the managed-care waiver at 42 U.S.C. 1396n, is the targeted instrument for delivery-system change; it sets aside freedom of choice so that a state may move beneficiaries into managed-care plans. Section 1915(c), the home and community-based services waiver at the same citation, sets aside statewideness and comparability so that a state may serve a defined population in defined places. Each authority names the requirements it may set aside, and none sets aside everything. The two requirements the waivers leave standing are as important as the three they lift, and they are taken up below.

Section 1115’s breadth is its defining feature and its defining limit. Because it can set aside all three requirements at once, it can authorize a program that looks very little like the regular state plan: different benefits in different counties for different populations, delivered through restricted networks. The authority is cabined by its purpose, which the statute names research and demonstration: the departure must be an experiment or a demonstration, undertaken to test something, not a permanent exemption from rules the jurisdiction finds inconvenient. The statute’s answer is written into the authority’s name. The jurisdiction must be trying to learn something, and the federal government must agree that the lesson is worth the departure.

What does a section 1915(c) waiver let a state do that the statute otherwise forbids?

It lets a state cover home and community-based services for a defined population in defined areas as an alternative to institutional care. The waiver sets aside statewideness and comparability, so a state may limit services to certain counties or to people with certain conditions, which the statute otherwise forbids.

The logic of the 1915(c) waiver is clearest when held against the two requirements it sets aside. Without the waiver, statewideness would oblige the jurisdiction to offer its home and community-based benefit in every county, and comparability would oblige it to offer that benefit to every categorically needy group in equal amount, duration, and scope. With the waiver, the jurisdiction may define the served population and the served area, county by county, and furnish the waiver’s services to that population in that area alone. The waiver is how the statute accommodates targeting. The regular program is universal within the jurisdiction’s borders; the waiver program is particular by design, and the particularity is the point, because home and community-based services are worth furnishing precisely where the need for them is concentrated.

The targeting the waiver permits is precise. A state may draw the served population around a particular need and the served area around the counties where that need can be met, rather than furnishing the same services to every categorically needy beneficiary in every county. The regular requirements would treat that precision as discrimination: statewideness would call the county limitation a violation, and comparability would call the population limitation one. The waiver treats the same precision as design. This is the sense in which section 1915(c) lets a state do what the statute otherwise forbids: not by repealing the requirements, but by licensing a defined departure from them.

Two requirements survive the waivers, and their survival is the point of the exercise. Reasonable promptness, at 42 U.S.C. 1396a(a)(8), obliges the jurisdiction to determine eligibility and furnish assistance without delay. No demonstration, however innovative, is permitted to let applications languish in a queue or benefits stall after approval; the duty to decide and the duty to deliver run underneath every waiver the statute authorizes. The equal-access provision, at 42 U.S.C. 1396a(a)(30)(A), requires payment rates sufficient to enlist enough providers that covered services are actually available to beneficiaries. No managed-care design and no demonstration may pay so little that the provider network exists only on paper while beneficiaries wait for care that never comes. The provision does not set a federal fee schedule; it sets a federal test. Rates are sufficient when they enlist enough providers that beneficiaries can obtain the covered services, and insufficient when providers exit faster than beneficiaries can find them. The waivers set aside the architecture of how benefits are organized. They do not set aside the duties to decide promptly and to pay adequately, because a program that rationed by delay or by starvation rates would keep the form of the federal bargain while abandoning its substance.

This is why the program looks different in every state. Every state begins with the same two benefit lists and the same three requirements, and then makes its own choices: which optional services to elect from the federal menu, whether to seek a research and demonstration waiver under section 1115, whether to move beneficiaries into managed care under section 1915(b), whether to build home and community-based programs under section 1915(c). The mandatory floor is national. Everything above the floor, and much of the machinery that delivers the floor itself, is the product of state decisions made inside a federal frame. Two beneficiaries with the same condition can receive different benefits on opposite sides of a state line without either state violating the statute, because the statute commands a floor and a set of rules, not a uniform program. The variation is not a failure of the design. It is the design, working as the waiver authorities intended.

Arizona supplied the early proof that the frame could hold a different shape. When the jurisdiction entered the program in 1982, it did so through a section 1115 research and demonstration waiver built around managed care, setting aside the standard requirements to test a delivery model the regular rules would not have permitted. The opening of this article tells that story in full. Its lesson for the statute’s structure is the one this section has been making throughout: the waiver did not repeal the requirements but suspended them for a defined purpose, and the program that resulted was still Title XIX.

The same architecture explains the change Congress enacted in 2010. That legislation made expansion of the program a mandatory condition of state participation, adding new federal commands to the structure this article describes. Litigation challenging the condition came later, and it falls outside this article’s scope. What matters for the statute’s structure is the mechanism rather than the outcome: Congress did not need a new program to change the program. It attached new conditions to the old bargain of lists, requirements, and waivers, and the states made their choices inside that bargain as they always had. The three-requirement key still turns the same lock; Congress simply added another tumbler.

The state plan requirement table

Requirement	Statutory citation	Waivable?	Under which authority	What a waiver typically changes
statewideness	42 U.S.C. 1396a(a)(1)	Yes	1115 and 1915(c)	Lets a state limit a benefit or demonstration to certain counties or populations
Comparability	42 U.S.C. 1396a(a)(10)(B)	Yes	1115 and 1915(c)	Lets a state offer different benefit packages to different groups
Freedom of choice of provider	42 U.S.C. 1396a(a)(23)	Yes	1115 and 1915(b)	Lets a state require enrollment in managed-care plans with restricted networks
Reasonable promptness	42 U.S.C. 1396a(a)(8)	No, not waived under the standard waiver authorities	None	States must still determine eligibility and furnish assistance without delay
Equal access / sufficient provider payments	42 U.S.C. 1396a(a)(30)(A)	No, not waived under the standard waiver authorities	None	Payments must remain sufficient to enlist enough providers

Benefits, requirements, and waivers describe what the statute commands. The remaining question is who can compel compliance when a state falls short.

Can you sue over Medicaid: the enforceability line

Every statute raises one blunt question before any other, which is who may sue when the law is broken. For the Medicaid statute, the answer begins with an absence. Congress created no express private right of action in Title XIX. No provision of the Medicaid Act states, in so many words, that a beneficiary denied covered services or a provider paid below the statutory standard may file suit to enforce the statute. That silence shapes everything that follows, because private enforcement must find a door that Congress did not label.

The door litigants use is 42 U.S.C. 1983, the civil rights statute that permits suits against state officials for the deprivation of federal rights under color of state law. Because the states administer the program, state officials are the natural defendants, and section 1983 supplies the cause of action that Title XIX itself omits. The route carries a demanding toll. A section 1983 plaintiff must assert the violation of a federal RIGHT, not merely a violation of federal law. The distinction controls the entire field. A statutory provision can command a state to act, and a state can fail to act, without any individual holding a personal right to demand compliance in court. Medicaid litigation therefore turns, provision by provision, on whether the statutory text confers a right on the plaintiff or simply announces a duty of the jurisdiction.

The Supreme Court first treated a Medicaid reimbursement provision as privately enforceable in Wilder v. Virginia Hospital Association (1990). Wilder read the reimbursement language as conferring an enforceable right on providers and allowed the section 1983 action to proceed. For a brief period the decision stood as the high-water mark of private enforcement of the Medicaid Act. Two years later the Court narrowed that approach in Suter v. Artist M. (1992). Suter held that a statutory requirement framed in broad and aspirational terms did not create an enforceable right even though it imposed a genuine obligation on the jurisdiction, and the decision sent litigants back to the text of each Medicaid provision with fresh uncertainty about which ones could survive a motion to dismiss. Congress answered in 1994. In 42 U.S.C. 1320a-2, Congress provided that a provision of the Social Security Act is not deemed unenforceable merely because it appears in a section that requires a state plan. The message was pointed. The jurisdiction-plan structure of the program, in which federal requirements take the form of conditions a state plan must satisfy, does not by itself defeat private enforcement. A provision located in a state-plan section can still create an enforceable right. Whether it does depends on its language.

The 1994 enactment deserves a careful reading, because both sides in later litigation claimed it. The statute says only that a provision is not deemed unenforceable merely because of where it appears. It does not say that every provision in a state-plan section creates an enforceable right, and it does not lower the substantive test for what counts as a right. It removed a categorical objection and left the hard work of textual analysis intact. That is why provision-by-provision litigation continued after 1994 rather than ending: Congress had closed one door to dismissal and left the others open.

The Court supplied the working test in Blessing v. Freestone (1997), and the test has three parts, stated here as prose because they operate as a sequence rather than a checklist. First, Congress must have intended the provision to benefit the plaintiff, which means the plaintiff must belong to the class the provision was designed to protect. Second, the asserted right must not be so vague and amorphous that its enforcement would strain judicial competence, which excludes provisions whose content a court cannot define without making the kind of policy judgments reserved to administrators. Third, the provision must impose a binding obligation on the jurisdiction, stated in mandatory rather than precatory terms, so the claimed right has something concrete to attach to. A provision satisfying all three may be enforced through section 1983. A provision failing any one of them may not.

The design of the test explains why Medicaid enforceability litigation became a war of close textual readings. Because the first factor turns on congressional intent toward the plaintiff and the second on judicial competence, the two sides read the same statutory sentence differently: the plaintiff hears a promise made to individuals, the jurisdiction hears an instruction addressed to its planners. The third factor then decides whether the promise was stated as a command or a hope. Courts of appeals applying this framework in early 2011 therefore produced no single answer for the Medicaid Act. Each provision received its own analysis, and the results accumulated into the pattern this section traces.

What did Gonzaga v. Doe change about enforcing Medicaid in federal court?

Gonzaga tightened the first Blessing factor so that only an unambiguously conferred right, expressed in rights-creating language focused on the individual, can support a section 1983 suit. After Gonzaga, a plaintiff must point to statutory text that confers the right in clear terms, not merely to a provision that benefits the plaintiff.

Before Gonzaga, the first Blessing factor asked whether Congress intended the provision to benefit the plaintiff, and some lower courts answered that question loosely, finding benefit wherever beneficiaries stood downstream of a state duty. Gonzaga v. Doe (2002) replaced benefit with conferral. The Court held that only an unambiguously conferred right, stated in rights-creating language with an unmistakable focus on the benefited class, can support a section 1983 action. Benefit was no longer enough. The statute had to confer, and confer clearly. The difference sounds technical and proved decisive. Under the looser reading, a provision directing a state to furnish services to eligible individuals could be read as benefiting those individuals and therefore as enforceable by them. Under Gonzaga, the court asks the harder question whether the provision speaks in the language of rights, addressed to the individual, with the clarity of an entitlement. Provisions framed as directives to the jurisdiction, as conditions on the state plan, or as aggregate standards administered by the federal agency fail at this first step no matter how much their performance would benefit real people. The practical effect on Medicaid litigation was to raise the bar on every claim. Provisions that had looked enforceable under the earlier understanding had to be reexamined, and many did not survive the reexamination. Litigation strategy shifted on both sides. Plaintiffs comb the statute for rights-creating verbs and individual-focused phrasing, while states point to language that runs to the plan, to the agency, or to the population in the aggregate. In early 2011, Gonzaga remained the controlling gloss on Blessing, and every Medicaid enforceability dispute in the federal courts was argued in its terms.

The practical question after Gonzaga was what counts as rights-creating language. Courts looked for phrasing addressed to the individual rather than the institution: language describing what an eligible person may obtain or choose, rather than language describing what a state plan must contain or what the federal agency must approve. The distinction is not always visible at a glance, because Medicaid’s plan requirements often blend the two, directing a state to establish procedures through which individuals receive something. In those blended provisions, courts asked which element the plaintiff’s claim actually invoked: the individual’s entitlement or the jurisdiction’s procedure. Where the claim rested on the entitlement, it had a chance; where it rested on the procedure, Gonzaga usually ended it.

Why is the enforceability line still contested?

The test turns on the language of each separate provision, and reasonable judges disagree about where statutory text confers a right and where it states an administrative standard. Because Medicaid contains both rights-like guarantees and broad payment standards, courts applying Blessing and Gonzaga provision by provision reached different results on different provisions.

The provision-by-provision character of the test is the engine of the disagreement. Two illustrations from the courts of appeals show opposite outcomes under the same framework. The free-choice-of-provider provision guarantees beneficiaries the ability to obtain covered services from any qualified provider willing to furnish them. Courts held the provision enforceable because its language runs to the individual beneficiary in terms that read as a guarantee: the beneficiary chooses, and the jurisdiction must honor the choice. The phrasing is rights-creating and individually focused in exactly the way Gonzaga demands, and the obligation it imposes on the jurisdiction is concrete rather than aspirational. The equal-access provision produced the opposite result. That provision addresses payment adequacy, requiring that state payment rates be sufficient to enlist enough providers that covered services remain available to beneficiaries at least to the extent they are available to the general population. Courts treated this language as a judgment-laden standard committed to the administering agency rather than a right conferred on any individual. Determining whether payments are sufficient to enlist enough providers calls for the kind of economic and policy judgment that strains judicial competence, and the provision speaks to aggregate adequacy across a population rather than to any one person’s entitlement. The same three-factor test thus yielded enforceability for the individual guarantee and non-enforceability for the aggregate standard, and neither outcome is obviously wrong on the test’s own terms.

The strongest case for broad private enforceability starts with Congress’s own 1994 intervention. By enacting 42 U.S.C. 1320a-2, Congress rejected the idea that placement in a state-plan section categorically defeats enforcement, which signals that the jurisdiction-plan requirements were meant to bite in real disputes rather than operating only in federal-state negotiations. Many of the statute’s mandatory plan provisions speak directly to eligible individuals and describe what the jurisdiction must furnish to them, language that reads naturally as conferring rights on the class the program exists to serve. Beneficiaries are, after all, the intended beneficiaries of the entire title, and a reading that leaves them without a judicial remedy for violations of provisions written for their protection renders the statutory guarantees precatory in practice. Wilder confirms that the Supreme Court once read a reimbursement provision exactly this way, treating payment language as conferring an enforceable right on providers. Nothing in the intervening decisions overruled Wilder outright, and the 1994 enactment narrowed the ground on which Suter-style arguments can rest. On this view, the Blessing and Gonzaga tests screen out vague or precatory language rather than licensing courts to reclassify every state-plan requirement as an unenforceable administrative aspiration.

Supporters of broad enforceability also emphasized who the statute’s requirements protect. The plan requirements at issue in these cases regulate what states must do for eligible individuals: furnish services, ensure choice, pay adequately. The intended beneficiaries are not the states, which are the regulated parties, and not the federal government, which administers the funding. When every plausible beneficiary of a provision is an individual, the argument that Congress intended no individual to enforce it grows strained. The 1994 enactment reinforced this point by insisting that the jurisdiction-plan form of these requirements not be treated as a reason to deny enforcement, which only makes sense if Congress understood at least some of them as judicially enforceable.

The strongest case for the narrow reading starts with the structure of the program. The program operates through federal funding conditioned on state plan compliance, and the statute’s design contemplates enforcement through the federal-state relationship, with federal review of state plans and federal control of the funding that makes state participation possible. Many provisions address the jurisdiction or the Secretary rather than the individual, prescribing what a plan must contain or what the agency must find, and Gonzaga teaches that language of this kind does not confer rights no matter how much individuals benefit from its performance. The ambiguity standard cuts the other way as well. Where a provision states a broad, judgment-laden standard such as payment adequacy, enforcement would require courts to make the policy determinations assigned to the administering agency, which is precisely what the second Blessing factor forbids. Suter and Gonzaga together trace a narrowing arc, and the 1994 enactment, on this reading, merely preserves the possibility of enforcement for provisions that independently satisfy the test rather than lowering the test itself. On this view, private enforcement is the exception, available only where Congress spoke to the individual in unmistakable rights-creating terms.

Supporters of the narrow reading also emphasized the third Blessing factor alongside the first. Even where a provision uses mandatory terms, the obligation often runs to the content of the state plan rather than to any individual’s treatment: the jurisdiction must have a plan that provides for certain things, and the federal agency reviews the plan. A mandatory duty owed within the plan-approval process is not the same as a right owed to the beneficiary in the clinic, and Gonzaga’s insistence on unambiguous conferral keeps the two distinct. On this view, the statute’s many uses of mandatory language describe the terms of the federal-state bargain, and the remedy for breach of that bargain is the federal government’s, exercised through the plan-approval process rather than through private suits.

The supported verdict is that the line, in early 2011, ran through the statutory text itself. Beneficiary-facing provisions stated in rights-creating, individually focused language survived the Blessing and Gonzaga framework, as the free-choice-of-provider cases show. Aggregate, judgment-laden standards directed at state planning and payment adequacy did not, as the equal-access cases show. The pattern was stable even though its application to any given provision remained contested, and the contest followed from the framework’s design: a test that turns on the precise wording of each provision generates border disputes wherever Congress wrote in broad terms. The enforceability question in Medicaid litigation was therefore never answered once for the whole statute. It was answered repeatedly, provision by provision, with the individual’s guarantee on one side of the line and the administrator’s standard on the other.

Verdict: the statute is the map

The namable claim of this article can be stated in one sentence. Statewideness, comparability, and freedom of choice are the three central state plan commitments, and waiver-driven program design is understood by identifying which commitment is altered. Nearly every distinctive feature of the modern program, from managed care enrollment to home and community based services to statewide demonstrations, is an exception to one of the three, authorized through a specific waiver power. A reader who can name which commitment a given policy sets aside already holds the map of the statute.

The key earns its name because it is usable. Take any Medicaid policy and ask which of the three it touches. Does it limit the program to certain counties or populations, implicating statewideness. Does it furnish different benefits to different groups, implicating comparability. Does it restrict the beneficiary’s choice of provider, implicating freedom of choice. Then ask which waiver authority permits the departure: the general demonstration power of section 1115, the managed care authority of section 1915(b), or the home and community based services authority of section 1915(c). The answer places the policy on the map the statute draws.

The complication is that Medicaid is not one national program. The federal statute sets uniform requirements, but each state writes its own plan, and the plan is where federal requirements become operational promises about eligibility, services, and payment. Two beneficiaries in two states live under the same title of the Social Security Act and different legal realities. The statute is therefore best understood as a map rather than a territory: it shows the required roads that every state plan must contain, while the plan itself shows the roads as built. Anyone who reads only the federal text knows the requirements but not the program as experienced; anyone who reads only a state plan knows the program but not the federal rights that constrain it. The enforceability analysis traced in this article operates at the federal level, provision by provision, and its results then travel into each state’s plan, where the same questions recur in local form.

This federal-state architecture explains why the enforceability line matters so much in practice. The federal requirements are the same everywhere, but the plans differ, which means the same federal provision can generate different disputes in different states depending on how each state implemented it. A beneficiary’s claim always runs through two documents: the federal statute, which supplies the right if the three requirements are met, and the state plan, which supplies the local facts about what the jurisdiction promised and did. Lawyers who master only one document argue half a case.

The practical payoff follows for each audience. For lawyers, the lesson is to plead the provision rather than the program. A complaint that invokes Medicaid in general invites dismissal, while a complaint that quotes the specific rights-creating language of a specific provision, shows its individual focus, and walks through the three Blessing requirements gives a court something to enforce. Anticipate the Gonzaga challenge in every case, because defendants will argue that the language runs to the jurisdiction or the agency rather than to the plaintiff, and know before filing which provisions the courts of appeals have already held enforceable. For state officials, the lesson is that mandatory plan language carries litigation risk proportional to its clarity. Where Congress spoke to the individual in unmistakable terms, implementation choices are made in the shadow of section 1983; where the statute states an aggregate standard committed to administrative judgment, the agency’s reasonable determinations retain primacy, as the equal-access cases demonstrate. For advocates, the lesson is strategic: build cases on provisions with rights-creating, individually focused text, of which the free-choice-of-provider guarantee is the model, and avoid staking claims on broad adequacy standards that courts have treated as judgment-laden. For students, the lesson is methodological. Learn the Blessing framework as Gonzaga refined it, then apply it provision by provision to the statute’s plan requirements, always asking the same three questions, and compare the free-choice and equal-access outcomes as the canonical pair that shows how an identical test produces opposite results on different texts.

Across all four audiences, one discipline matters more than any other: read the provision before reading the commentary about it. The enforceability cases turn on single sentences, and paraphrases smooth over the exact wording that decided the outcome. A student who learns to compare the free-choice language with the equal-access language, side by side, will understand the Blessing and Gonzaga framework better than one who memorizes the three factors in the abstract. The habit transfers directly to practice, where the first task in any Medicaid dispute is to put the operative sentence on the page and ask whether it speaks to the client.

The study path that follows from this article is straightforward. Read the plan requirements of Title XIX with the three requirements in hand, marking each provision as rights-creating or administrative, individual or aggregate, mandatory or precatory. Test the marking against the decided outcomes, especially the two worked illustrations in this article, and revise the marking where the courts saw the text differently. Then carry the framework into a state plan and watch the federal rights take local shape. The notebook repays a simple method: one page per provision, with the statutory language at the top, the three-factor analysis below it, and the decided outcome recorded alongside. Over the course of the title’s plan requirements, those pages accumulate into a personal map of the enforceability line, marked provision by provision exactly as the courts have marked it. A structured place to build and keep that map is the VaultBook legislation study notebook, which is designed for exactly this kind of provision-by-provision study.

Two questions that readers commonly bring to this article belong to companion pieces. Readers who arrive carrying the comparison question, asking how Medicaid differs from Medicare as a legal matter, will find it answered in the series article on the legal difference between Medicare and Medicaid. The long-term-care gap that Medicare leaves and that Medicaid fills, one of the most practically important divisions between the two programs, is traced in the series article on the Medicare Parts A and B statute. Finally, this article describes the statute as it stood on February 1, 2011, including the 2010 expansion enacted as a mandatory condition of state participation. Later litigation over that condition falls outside this article’s scope; this later series article carries the discussion of post-2011 litigation over the expansion.

The larger lesson of this article is that the Medicaid statute rewards the reader who stays inside it. Its structure, federal requirements implemented through state plans; its enforcement design, rights where Congress conferred them and agency administration elsewhere; and its vocabulary, precise where courts have enforced it and broad where they have not, all emerge from the text itself. The statute is the map. Learn to read it provision by provision, and the program’s complexity resolves into a set of answerable questions, each with its own text, its own test, and its own line.

Frequently Asked Questions

How is the Medicaid statute structured?

Medicaid is Title XIX of the Social Security Act, codified at 42 U.S.C. 1396 and following. Congress organized it as a cooperative federal-state grant program rather than a directly administered federal benefit. The statute opens with the authorization for federal appropriations and then lays down the conditions a state must meet to receive them. The center of the structure is section 1902, 42 U.S.C. 1396a, which lists the required contents of a state plan. Around that core the statute arranges the categories of people a state may or must cover, the medical services that are mandatory or optional, the federal payment formula keyed to the federal medical assistance percentage, and the waiver authorities in sections 1115, 1915(b), and 1915(c) that let the Secretary excuse compliance with specified requirements. Because federal money flows only through an approved state plan, every substantive duty in Title XIX is expressed as a condition of participation.

What is FMAP in Medicaid?

The federal medical assistance percentage, abbreviated FMAP, is the share of a state’s Medicaid service expenditures paid by the federal government. It appears at 42 U.S.C. 1396d(b), section 1905(b) of the Social Security Act. The statute ties the percentage inversely to state per capita income, so states with lower average incomes draw a higher federal match and wealthier states draw a lower one. Congress placed a floor and a ceiling on the rate: no state receives less than 50 percent or more than 83 percent federal funding for covered services. The FMAP applies to most medical assistance spending, while the statute assigns fixed higher rates to certain specified services and administrative activities. Because the percentage moves with relative state income, a state’s federal share reflects its fiscal capacity rather than a uniform national rate.

Which Medicaid benefits are mandatory and which are optional?

Section 1902 obliges every state plan to cover a defined set of mandatory services. Those include inpatient and outpatient hospital services, physician services, laboratory and x-ray services, nursing facility services for beneficiaries age 21 and older, home health services, family planning services and supplies, the early and periodic screening, diagnostic, and treatment benefit for beneficiaries under 21, rural health clinic and federally qualified health center services, nurse-midwife and nurse practitioner services, and pregnancy-related services. Beyond that floor, the statute permits states to add optional benefits through their state plans. The optional category includes prescription drugs, dental services, personal care services, physical therapy, eyeglasses, and prosthetic devices. The mandatory list establishes the national minimum every participating state must furnish, while the optional list gives states room to tailor coverage to their populations and budgets, subject to Secretarial approval of the state plan.

What is a Medicaid state plan?

A state plan is the written document in which a state describes how it will administer its Medicaid program, required by section 1902 of the Social Security Act, 42 U.S.C. 1396a. The statute lists the provisions the plan must contain, including the three core requirements of statewideness, comparability, and freedom of choice of provider, along with the eligibility groups and benefit package the jurisdiction has chosen. The plan functions as both the jurisdiction’s application for federal funds and its binding commitment to operate the program as described. The Secretary of Health and Human Services must approve a plan that satisfies the statutory requirements, and federal matching payments flow only under an approved plan. When a state changes eligibility, benefits, or payment methods, it submits a state plan amendment for Secretarial review before the change takes effect.

What is a section 1115 Medicaid waiver?

A section 1115 waiver, authorized at 42 U.S.C. 1315, lets the Secretary of Health and Human Services set aside specified Medicaid requirements for an experimental, pilot, or demonstration project that, in the Secretary’s judgment, is likely to promote the objectives of the program. The waiver suspends the ordinary rules only within the terms and conditions the Secretary approves, so the demonstration operates under a negotiated framework rather than under the standard state plan. Arizona’s program illustrates the device: when Arizona joined Medicaid in 1982, it did so through a section 1115 demonstration called the Arizona Health Care Cost Containment System, or AHCCCS, which became the first statewide managed-care Medicaid program in the country. Section 1115 thus gives states a path to test delivery and financing models the statute would otherwise forbid, with the Secretary acting as gatekeeper.

What is EPSDT in Medicaid?

EPSDT stands for early and periodic screening, diagnostic, and treatment, a mandatory Medicaid benefit for beneficiaries under age 21. Congress designed it as a comprehensive child health benefit rather than a narrow list of covered services. The screening component calls for regular examinations of physical and mental health, vision, hearing, and dental health on a schedule the jurisdiction establishes. When a screening reveals a condition, the diagnostic and treatment components require the jurisdiction to furnish the care needed to correct or ameliorate it, even if the particular service falls outside the jurisdiction’s standard benefit package for adults. In practice this makes EPSDT the broadest benefit in the Medicaid statute for children, because coverage turns on medical necessity for the child rather than on whether the service appears on the adult list.

Can you sue a state over Medicaid?

A beneficiary may sue only in narrow circumstances. Because Medicaid is Spending Clause legislation, its provisions are generally understood as conditions on federal funding rather than as rights enforceable by private plaintiffs. The Supreme Court’s pre-2011 framework begins with Blessing v. Freestone, decided in 1997, which set the test for when a Spending Clause provision creates an enforceable right, and continues through Gonzaga v. Doe, decided in 2002, which held that a private suit under 42 U.S.C. 1983 requires a right unambiguously conferred on the individual by the statute. Most Medicaid requirements, including the broad state plan duties, fail that test because they speak to what the jurisdiction must do to keep federal funds rather than to what the individual may demand in court. The Secretary’s remedy for a noncompliant state is administrative, principally the withholding of federal matching funds.

Which state joined Medicaid last?

Arizona was the last state to join Medicaid, entering the program in 1982. Rather than adopting a conventional fee-for-service program, Arizona launched the Arizona Health Care Cost Containment System, known as AHCCCS, in October 1982 under a section 1115 demonstration waiver. The demonstration made Arizona the first state to run its entire Medicaid program through managed care, with the jurisdiction contracting with health plans to deliver services instead of paying providers directly for each service. The 1115 authority allowed this departure from the ordinary requirements, and the Secretary approved the demonstration as a test of whether managed care could control costs while preserving access. Every other state had joined Medicaid before Arizona, making its 1982 entry the close of the program’s original state-by-state adoption.

What does the statewideness requirement demand?

The statewideness requirement, at 42 U.S.C. 1396a(a)(1), provides that the state plan must be in effect in all political subdivisions of the state. In ordinary terms, a state may not offer Medicaid in only some counties or regions while excluding others. Whatever eligibility rules and benefit package the jurisdiction adopts must apply uniformly across the state, so a beneficiary’s access does not depend on where within the state the beneficiary lives. The requirement guards against a patchwork program in which wealthier or more urban areas receive coverage that rural or poorer areas do not. A state that wishes to test a different arrangement in part of the state, such as a pilot program in a few counties, must seek a waiver of this requirement, most commonly under section 1115 or section 1915(b).

What does comparability require of Medicaid benefits?

The comparability requirement, at 42 U.S.C. 1396a(a)(10)(B), provides that the medical assistance available to any categorically needy individual may not be less in amount, duration, or scope than what is available to any other categorically needy individual. Once a state covers a service for one group of mandatory beneficiaries, it must furnish that service on equal terms to the others. The rule prevents a state from, for example, covering robust physician services for children while offering only a thin version to aged or disabled beneficiaries within the mandatory population. Comparability does not require the jurisdiction to cover every optional service, but it does require evenhandedness for the services the jurisdiction chooses to cover. States seeking to vary benefits among populations, as in a targeted managed-care or home-based program, typically need a waiver of this provision.

What does freedom of choice of provider guarantee beneficiaries?

The freedom-of-choice requirement, at 42 U.S.C. 1396a(a)(23), guarantees that a Medicaid beneficiary may obtain covered services from any qualified provider willing to furnish them. A state may not confine beneficiaries to a closed panel of providers or steer them away from a qualified hospital, physician, or clinic that accepts Medicaid payment. The provision protects the beneficiary’s side of the transaction: the jurisdiction sets the payment rates and the qualifications, but the choice of which qualified provider to see belongs to the patient. This requirement is one of the principal obstacles to mandatory managed care, because assigning beneficiaries to a single plan or network restricts the very choice the statute protects. States that wish to require managed-care enrollment therefore seek a waiver of this provision, most often under section 1915(b).

What is the medically needy option?

The medically needy option lets a state extend Medicaid to people who meet the nonfinancial criteria for coverage but whose income or resources exceed the limits for the mandatory groups. Under this option, the jurisdiction counts the medical expenses the person incurs against the excess income, a process often called spend-down. Once the incurred expenses reduce countable income to the jurisdiction’s medically needy threshold, the person qualifies for coverage. Congress made the option voluntary: a state may cover the medically needy, but the statute does not require it. States that adopt it must define the income standard and apply it evenhandedly, and the option gives states a way to reach people, such as those facing prolonged nursing facility or hospital stays, whose medical costs dwarf their incomes.

What is a section 1915(b) waiver used for?

A section 1915(b) waiver, authorized at 42 U.S.C. 1396n, lets a state set aside the statewideness, comparability, and freedom-of-choice requirements to implement managed care and related delivery reforms. The typical use is to require beneficiaries to enroll in managed-care plans or to receive services through a limited network of providers, arrangements the unwaived statute would forbid. States also use 1915(b) authority for selective contracting with providers and for central enrollment brokers that assign beneficiaries to plans. Unlike a section 1115 demonstration, a 1915(b) waiver does not suspend the statute’s eligibility or benefit rules; it changes how services are delivered and how providers are chosen. The Secretary approves these waivers for limited periods, and the jurisdiction must show that the arrangement serves the program’s objectives without reducing access.

Who can receive services under a section 1915(c) waiver?

A section 1915(c) waiver serves people who would otherwise require the level of care furnished in an institution, such as a hospital or nursing facility, but who can be served safely in the home or community. The waiver targets beneficiaries whose needs would ordinarily lead to institutional placement: older adults needing nursing facility care, people with physical disabilities, and individuals who would otherwise enter long-term institutions. To qualify, the person must meet the jurisdiction’s criteria for the institutional level of care the waiver substitutes for, and the home and community services must be an appropriate alternative. Congress also required the jurisdiction to show that serving these beneficiaries in the community costs no more than serving them in the institution, a cost-neutrality condition that keeps the waiver from becoming an open-ended expansion. The provision thus redirects institutional spending toward home-based care for a defined population.

Why do some services receive a higher federal match than the FMAP?

Congress did not apply the FMAP to every dollar a state spends. For certain services and administrative activities that lawmakers wanted to encourage, the statute assigns fixed federal percentages above the jurisdiction’s FMAP. The design uses the matching rate as a policy lever: a higher federal share lowers the jurisdiction’s cost of the favored activity and nudges states toward it. The FMAP itself remains the rate for most medical assistance, moving inversely with state per capita income between the 50 percent floor and the 83 percent ceiling, while the enhanced rates operate as exceptions written directly into the payment provisions. A state therefore faces a blended federal share across its program, with the enhanced categories drawing more federal money per dollar spent than the services paid at the ordinary FMAP.

What happens when the Secretary disapproves a state plan amendment?

When the Secretary disapproves a state plan amendment, the proposed change does not take effect and the jurisdiction receives no federal matching funds for it. Disapproval follows a finding that the amendment fails one or more of the requirements in section 1902, such as the statewideness, comparability, or freedom-of-choice provisions. The statute gives the jurisdiction procedural protections: the jurisdiction is entitled to notice of the reasons for disapproval and an opportunity for a hearing before the decision becomes final. In practice, disapproval often opens a negotiation in which the jurisdiction revises the amendment to satisfy the Secretary’s objections and resubmits it. A state that believes the disapproval is unlawful may seek judicial review of the final agency action. The process underscores that the Secretary, not the jurisdiction alone, determines whether a plan change comports with Title XIX.

How does Title XIX differ structurally from Title XVIII?

Title XIX and Title XVIII of the Social Security Act create mirror-image structures. Title XIX establishes Medicaid as a cooperative federal-state program: states administer the program under Secretarially approved state plans, set eligibility and benefits within statutory bounds, and receive federal matching payments keyed to the FMAP. Title XVIII establishes Medicare as a federally administered program: the federal government sets uniform national rules, contracts directly with providers and plans, and pays claims without state intermediation. Medicaid’s duties therefore run through the state plan mechanism, with the three core requirements of statewideness, comparability, and freedom of choice constraining state design, while Medicare’s duties run directly between the federal government and providers. The two titles share the Social Security Act as their vehicle but assign the states fundamentally different roles.

Must every state cover prescription drugs?

No. Prescription drugs are an optional Medicaid benefit, so the statute does not require any state to cover them. A state that wishes to cover outpatient drugs adds the benefit through its state plan, subject to Secretarial approval, and a state that declines the benefit remains in full compliance with Title XIX. This surprises many observers because drug coverage feels like a core part of health insurance, but the mandatory benefit list Congress wrote into section 1902 does not include it. The optional status gives states budgetary control: drugs are a large expenditure, and a state may define the scope of coverage, such as formularies and prior authorization, within its plan. The consequence is variation, with drug coverage differing from state to state rather than guaranteed nationally.

What does the reasonable-promptness requirement guarantee?

The reasonable-promptness requirement obliges a state to furnish Medicaid assistance to eligible individuals without undue delay. Once a person is determined eligible, the jurisdiction may not leave the application languishing or postpone covered services indefinitely. The provision addresses the gap between eligibility on paper and care in fact: a benefit that arrives months late is little better than no benefit for a person needing treatment. Courts and the Secretary have read the requirement to forbid both bureaucratic foot-dragging in eligibility determinations and unreasonable waits for covered services once eligibility is established. The standard is reasonableness rather than immediacy, so the jurisdiction may take the time genuinely needed to verify eligibility and arrange care, but it may not use administrative backlog as a reason to withhold what the statute promises.

Why is Medicaid called a federal-state partnership?

Medicaid earns the partnership label because neither level of government runs it alone. Congress set the program’s requirements and pays a share of its costs through the FMAP, which ranges from 50 to 83 percent depending on state per capita income. The states administer the program day to day, determine eligibility under their approved state plans, pay providers, and contribute the nonfederal share. Within the federal guardrails, states choose which optional eligibility groups to cover and which optional benefits, such as prescription drugs, to include. The Secretary of Health and Human Services oversees the arrangement by approving state plans and plan amendments and by granting waivers that let states depart from specified requirements. The result is fifty-plus distinct programs operating under one federal statute, which is why descriptions of Medicaid’s rules must always be qualified by state.