Concept
The Mathews Balancing Test
Procedural due process
The Mathews Balancing Test
Mathews v. Eldridge asked whether the Fifth Amendment's Due Process Clause requires that a recipient of Social Security disability benefits be offered an evidentiary hearing before the payments end.1 The Clause says that no person shall "be deprived of life, liberty, or property, without due process of law."2 Relying on Goldberg v. Kelly, the District Court held that a hearing was required. It found the agency's existing procedures, a pretermination review of the medical information plus a chance to submit written evidence, constitutionally deficient.3 The Supreme Court reversed. It held that an evidentiary hearing is not required before disability benefits are terminated and that the present administrative procedures fully comport with due process.4 Along the way the Court articulated a balancing test for deciding what procedures are required, and it has applied that test repeatedly ever since.5
The procedures the agency already provided set the baseline the Court weighed. Before cutting off benefits, the agency told the recipient its tentative assessment, gave him a summary of the supporting evidence, let him review the medical reports in his file, and invited a written response with additional evidence.6 After the cutoff he could seek reconsideration, then an evidentiary hearing before an administrative law judge in a nonadversary proceeding, then discretionary review by the SSA Appeals Council, and finally judicial review, with counsel permitted at every stage.7 A claimant who ultimately prevailed could obtain retroactive benefits.8 The dispute therefore concerned timing: whether a hearing was due before the first cutoff.
The Court said that identifying the specific dictates of due process "generally requires consideration of three distinct factors: First, the private interest that will be affected by the official action; second, the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute procedural safeguards; and finally, the Government's interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail."9
The factors do two different jobs. The first two measure what an added procedure would be worth: how much the claimant stands to lose, and how likely the existing procedures are to err in a way the added safeguard would catch. The third measures what the added procedure would cost the government. Together they direct a court to ask whether the value of the procedures a claimant demands, beyond those already provided, justifies the cost of adding them.10 Goldberg is the benchmark for the most demanding result. Mathews noted that only in Goldberg had the Court held that a hearing closely approximating a judicial trial is necessary, and that in other cases requiring some pretermination hearing the Court had spoken sparingly about the requisite procedures.11 A trial-type hearing is thus one result the balance can produce, and it is not the model every benefit must follow. The subsections below take the factors in order and then set Goldberg beside Mathews.
The Private Interest
The first Mathews factor asks how much the claimant stands to lose: "the private interest that will be affected by the official action."12 Two things set its weight, the severity of the potential deprivation and the length of time a wrongful deprivation may last.13 The logic follows from what a procedure is for. A safeguard exists to catch mistakes, and the greater the loss an uncorrected mistake inflicts, the greater the need for safeguards.14 Setting the welfare recipient in Goldberg beside the disability recipient in Mathews shows the scale at work.
The welfare recipient sits at the heavy end. Goldberg concerned benefits for "persons on the very margin of subsistence,"15 and an AFDC beneficiary whose benefits are wrongfully terminated is in an "immediately desperate situation" because she and her children have been deprived of "the very means by which to live."16
The disability recipient sits lower. Eligibility for disability benefits is not based on financial need, and it is unrelated to the worker's income or to support from other sources such as a family member's earnings, savings, private insurance, pensions, and veterans' benefits.17 If a termination pushes the worker or the family below the subsistence level, other forms of government assistance become available,18 and the worker may even be eligible for welfare benefits.19 A worker who later prevails can also seek retroactive benefits.20 Each fact softens the interim loss. Together they led the Court to find less reason than in Goldberg "to depart from the ordinary principle, established by our decisions, that something less than an evidentiary hearing is sufficient prior to adverse administrative action."21
The Court did not push the contrast far. It called the potential deprivation "generally likely to be less than in Goldberg," while warning that "the degree of difference can be overstated."22 To stay eligible a recipient must be "unable to engage in substantial gainful activity," so, unlike the discharged federal employee in Arnett v. Kennedy, a terminated recipient has little chance of finding even temporary work to ameliorate the interim loss.23 Duration adds to the problem. The Court said that "the possible length of wrongful deprivation of . . . benefits [also] is an important factor in assessing the impact of official action on the private interests." The Secretary conceded that 10 to 11 months passed between a request for a hearing before an administrative law judge and a decision, and because a recipient must first obtain a reconsideration decision, the delay between the cutoff and a final decision after a hearing exceeds one year.24 Even so, the Court concluded that the disabled worker's need is likely to be less than a welfare recipient's, and that conclusion kept the first factor on the lighter side of the scale.25
This factor has to be kept apart from the threshold question taken up in the Protected Interests section. Whether a claimant holds a protected property or liberty interest does not turn on the strength of the interest. The first Mathews factor assumes the interest qualifies and asks how much its wrongful loss would hurt.26 A welfare recipient and a disability recipient may both clear the threshold, and the factor then distinguishes them only by weight. The factor also looks to the interest in the underlying benefit, because due process safeguards are triggered by protected property or liberty interests and not by an interest in the procedures themselves.27 A claimant who argues only that a hearing is valuable in itself has run together the two questions that Roth separated: whether a protected interest exists, and what procedures are due.28
The more important a court considers the interest, the greater the loss from wrongful deprivation and the greater the need for safeguards.29 Weight alone never settles the question, though. Even a very important interest does not call for added procedures that would be unlikely to reduce the risk of an inaccurate decision.30 The first factor tells a court how much is at stake, and the second tells it whether more procedure would protect that stake.
Risk of Error and Safeguard Value
The second Mathews factor asks how likely the existing procedures are to produce a wrongful deprivation and how much an additional or substitute safeguard would improve on them.32 The Court framed it as "the fairness and reliability of the existing pretermination procedures, and the probable value, if any, of additional procedural safeguards."33 A procedure earns its place only by catching mistakes. Where the procedures already in use produce accurate decisions, an added safeguard brings delay and expense and little gain in accuracy. Where they leave real room for error, the same safeguard is worth more. The first factor measured what an erroneous cutoff costs the claimant; this factor measures how likely such an error is and how much a proposed addition would reduce it.
The value of a safeguard depends on the kind of mistake it is meant to prevent, so the Court looked first to "the nature of the relevant inquiry."34 A typical disability termination turns on "routine, standard, and unbiased medical reports by physician specialists" about a subject they have personally examined. In Richardson v. Perales the Court recognized the reliability and probative worth of such written reports and observed that, although professionals may disagree about the medical conclusions, the "specter of questionable credibility and veracity is not present."35 A welfare dispute looks different. Where eligibility turns on credibility, Goldberg found that "written submissions are a wholly unsatisfactory basis for decision."36 A typical AFDC question is whether the recipient lives with a man able to support her children, which concerns patterns of behavior and depends largely on whom the decisionmaker believes. In that setting the chance to watch a witness's demeanor and to cross-examine is worth far more.37
The capacities of the people who must use a procedure point the same way. AFDC beneficiaries usually have limited educations that make it hard to participate effectively in written proceedings, while the primary participants in disability proceedings are doctors, who typically write well.38 A written reply is a real opportunity for a person who can use it and an empty one for a person who cannot.
The other half of the factor asks how well the current procedures already guard against error, since an added safeguard has little to contribute where the existing ones work. The disability program built in several protections before payments stopped:
- Information about the recipient's condition came from his own sources of medical treatment, and an independent consulting examination could be arranged when those sources conflicted.39
- The agency informed the recipient of its tentative assessment, the reasons for it, and a summary of the evidence it considered most relevant.40
- The recipient's representative had full access to all information the agency relied on, and the recipient could review the medical reports and other evidence in his case file.41
- The recipient could submit additional evidence or arguments before the cutoff.42
Each protection answers a particular source of mistake. The independent examination addresses conflicting medical sources, and the notice and summary show the recipient what the agency is relying on. File access and the chance to respond let him challenge both the accuracy of the file and the correctness of the agency's tentative conclusions, and they let him "mold" his argument to the precise issues the decisionmaker regards as crucial, which the procedures in Goldberg did not allow.43 Because the recipient learns the agency's reasons before the decision is final, a written reply can do much of the work an oral hearing would do. The Court concluded that the risk of erroneous deprivation was not high under these procedures, given the nature of the eligibility determination and the use of medical reports.44
The challengers also pointed to the rate at which terminations were reversed on appeal. The Court said that in this context the value of reversal-rate statistics for evaluating the adequacy of the pretermination process is diminished, and that although it viewed such information as relevant, it was "certainly not controlling."45 A reversal alone does not show that the earlier decision was careless, since the recipient may respond to the agency's tentative assessment and submit more evidence along the way.
One more point fixes how far the factor reaches. The Court conceded that credibility and veracity may matter in some disability cases, and it still held that procedural due process rules "are shaped by the risk of error inherent in the truthfinding process as applied to the generality of cases, not the rare exceptions."46 A constitutional rule attaches to a whole class of proceedings, and demanding a trial-type hearing for every disability claimant because a few disputes turn on credibility would impose the full cost on all claims to correct errors that arise in a few. A trial-type hearing would improve truthfinding here, but probably not by a great deal.47 On that reasoning the Court found the potential value of an evidentiary hearing, or even an oral presentation to the decisionmaker, substantially less for disability terminations than in Goldberg.48
The same agency can owe different amounts of procedure depending on what a termination rests on. The added benefit of an evidentiary hearing is greater when issues of witness credibility and veracity are directly involved, as when benefits are terminated because a recipient allegedly concealed outside income.49 Students who recite that disability claimants never receive a pretermination hearing therefore go wrong, because the program label settles nothing. The result follows from the inquiry and the safeguards in place.
The Government's Interest
The third Mathews factor puts a price on the added procedure: "the Government's interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail."51 The Court called this the public interest, and it includes the administrative burden and other societal costs of requiring, as a matter of constitutional right, an evidentiary hearing on demand in all cases before benefits end.52 The costs include out-of-pocket expense and also the impact of procedural formality on the character of the underlying program.53
In Mathews the most visible burden was the incremental cost of the additional hearings, together with the expense of paying benefits to ineligible recipients while they awaited a decision.54 Both costs flow from one feature of a prior-hearing rule. If full benefits continue until the hearing, a recipient loses nothing by requesting one, so most recipients would exhaust that option. The agency's theoretical right to recover benefits paid in error would not, as a practical matter, offset the added outlay in any substantial way. The parties' dollar estimates varied widely, and the Court did not choose among them. It said only that experience with constitutionalizing government procedures suggested the added cost in money and administrative burden would not be insubstantial.55 The aftermath of Goldberg bore that out. Once agencies had to provide pre-deprivation hearings, AFDC beneficiaries demanded oral evidentiary hearings far more often. A beneficiary had nothing to lose by asking, since someone claiming eligibility for AFDC rarely has assets from which a successful recoupment action could recover payments if she lost, and even a loser kept about a year's benefits.56
Cost counts, but it does not decide the question alone. The Court said that financial cost alone is not a controlling weight in deciding whether due process requires a particular safeguard, yet the government's interest, and the public's, in conserving scarce fiscal and administrative resources must be weighed, because at some point the benefit of an additional safeguard may be outweighed by its cost.57 It then said whose money is at stake: resources for any social welfare program are not unlimited, so the cost of protecting those whom the preliminary process has identified as likely to be undeserving may in the end come out of the pockets of the deserving.58 The mechanism is a fixed budget. A dollar spent on added hearings or on payments to ineligible recipients during the wait is a dollar the program cannot pay to eligible ones. A student who treats cost as irrelevant because the claimant's loss is serious has skipped a factor the Court requires, and a student who treats it as decisive has misread the same passage.
The Court also described what due process demands in terms that shape the cost inquiry. Its essence is that "a person in jeopardy of serious loss [be given] notice of the case against him and opportunity to meet it," and the procedures need only be tailored, in light of the decision to be made, to "the capacities and circumstances of those who are to be heard" so that claimants have a meaningful opportunity to present their case.59 The question is whether a trial-type format is needed to produce a fair decision on this kind of dispute. For a dispute that turns on documented medical reports, a written exchange may serve as well at far lower expense.
Courts also give substantial weight to the good-faith judgments of the officials whom Congress charged with administering the program that the procedures they provide assure fair consideration of claims.60 The deference is strongest where the scheme gives the claimant an effective way to assert the claim before any administrative action and also assures an evidentiary hearing and judicial review before a denial becomes final.61 The disability program in Mathews had exactly that structure. The deference is one weight in the balance, and it does not displace the first two factors.
Goldberg weighed the same kinds of interest and came out the other way. The Court observed that the State's own provision for a post-termination hearing was evidence that it recognized the primacy of the public interest in correct eligibility determinations, and it held that the eligible recipient's interest in uninterrupted aid, joined with the State's interest that payments not be erroneously terminated, clearly outweighed the State's concern to avoid added fiscal and administrative burdens.62 Mathews then gave cost more weight. It emphasized what the Goldberg Court had resisted, that the interest in administrative justice "[a]t some point . . . may be outweighed by the cost," and it heeded Justice Black's warning in his Goldberg dissent that adding procedure threatens to harm beneficiaries by diverting scarce public resources from benefits to administrative costs.63 Part of the shift reflects context, since the disability program carried a lighter private interest and a more documented inquiry. The comparison below shows how the factors combine.
Cost also has a limit, and Stanley v. Illinois marks it. The State there insisted on presuming rather than proving Stanley's unfitness as a parent "solely because it is more convenient to presume than to prove." The Court held that this advantage is insufficient to justify refusing a father a hearing "when the issue at stake is the dismemberment of his family," and it added that the State's interest in caring for Stanley's children is de minimis if he is shown to be a fit father.64
Mathews and Stanley fit together once you ask what each government saved. In Mathews the agency still decided the dispositive question, whether the claimant remained disabled, after giving notice of its reasons and a chance to respond, and a full hearing followed.65 What it economized was the timing and format of the hearing. The State in Stanley economized by presuming the answer to the central question about the individual. Cost can shape how and when an individualized decision is made, and it cannot substitute a presumption for the decision.
The third factor therefore never decides a case alone. Its output is a weight, large where the added procedure would multiply hearings and unrecoverable payments for little gain in accuracy, small where hearings would be few and cheap, and powerless to excuse the absence of any individualized decision on the dispositive issue.
Goldberg Compared With Mathews
The Social Security Administration in Mathews used procedures identical to the ones the Court had held inadequate for welfare in Goldberg. On a preliminary determination of ineligibility based on an ex parte investigation, the agency notified the individual of that determination and its basis and allowed a specified period to submit written materials. It then decided on the evidence it had gathered and the written submission, and it notified the individual that benefits had been terminated.67 Goldberg held that an agency cannot terminate AFDC benefits without a pretermination oral hearing that includes the right to cross-examine witnesses.68 Mathews held that the government can deprive a person of disability benefits after a pre-deprivation written exchange of data and views, followed by a post-deprivation opportunity for an oral evidentiary hearing.69 The Court did not overrule Goldberg. It explained the difference with three distinctions, and each corresponds to one of the factors.
The first distinction concerns what the claimant stands to lose while the dispute is pending. Welfare assistance goes to persons "on the very margin of subsistence," whereas eligibility for disability benefits "is not based upon financial need."70 A disability recipient may therefore have a spouse's earnings, savings, or other support, and the interim hardship is generally likely to be smaller. The Court conceded that the difference "can be overstated," because a worker who cannot work has little chance of finding temporary employment to soften the loss.71 This distinction feeds the first factor. It lowers the weight of the claimant's interest without erasing it.
The second distinction concerns the question the agency decides. A typical disability dispute asks about the claimant's health, a relatively objective matter. A typical welfare dispute asks about patterns of behavior, such as whether the recipient lives with a man able to support her children.72 Witness veracity matters more in the second kind of dispute, and so do the chance to watch a witness's demeanor and the chance to cross-examine.73
The third distinction concerns who must use the procedure. Welfare recipients usually have limited educations that make written participation difficult, while the main participants in disability proceedings are doctors, who typically write well.74 The point applies a principle the Court stated directly: procedures must be tailored, in light of the decision to be made, to "the capacities and circumstances of those who are to be heard."75 The second and third distinctions both belong to the second factor, because each bears on how much an oral hearing would add to the accuracy of the decision.
With those distinctions in place, the Court assessed the disability procedures themselves. Before benefits ended, the agency gave the claimant notice of its tentative assessment and a summary of the evidence, allowed review of the file, and invited a written response and further evidence.76 Afterward he had reconsideration, a nonadversarial hearing before an administrative law judge, discretionary review, and judicial review, with the right to counsel at every stage.77 The Court gave substantial weight to the administrators' good-faith judgment that these procedures assure fair consideration of claims, and the weight was greatest because an evidentiary hearing and judicial review were assured before a denial became final.78 It concluded that "an evidentiary hearing is not required prior to the termination of disability benefits and that the present administrative procedures fully comport with due process."79
Placed side by side, the two cases hold the paper procedure constant and change the program. A state that ends disability benefits after giving a statement of reasons, a chance to review the evidence and answer in writing, and a later full hearing with retroactive relief satisfies due process on the Mathews balance.80 A city that ends need-based welfare aid on the same written-submission procedures fails, because the stakes, the credibility issues, and the recipients' capacities all differ.81 The whole difference in outcome traces to the program facts the three factors measure.
A stem that matches one case on every feature is easy to resolve. The harder stem splits the features, and the analysis then has to weigh them.
A court reaches that answer by weighing. The more important it considers the interest at stake, the more procedural safeguards it is likely to demand,82 and the more important it considers witness veracity, the more value it places on demeanor and cross-examination.83 Answering by program label ("disability, so paper is enough") skips that weighing. So does treating a heavy private interest as decisive, because that shortcut ignores the second factor, which asks what the added procedure would contribute.
The aftermath of Goldberg shows why cost gets a formal place in the balance. Once pretermination hearings were required, AFDC claimants demanded oral hearings far more often, and studies appeared to support the Goldberg dissenters' claim that the majority had underestimated the cost of the procedures.84 The distinctions the Court drew between the programs are real but easy to overstate. Most disability recipients wrongly cut off would also face serious financial trouble. Contested disability cases often turn on differing opinions of equally able doctors about conditions such as depression, anxiety, or chronic back pain. Most welfare recipients who gained an oral hearing lacked the skill to cross-examine effectively or the money to hire someone who had it.85
The dissent in Mathews pressed the point directly. It called the Court's reliance on a limited deprivation speculative, said that the legislative choice to pay disability benefits without a need test presumes a need the Court should not denigrate, and noted that in Eldridge's own case the benefit termination was followed by foreclosure on the family home and repossession of the furniture. It added that it is no argument that a worker may seek other forms of public assistance.86 The majority's reasoning is the governing law, and the dissent shows where the balance is soft.
That softness is a defining trait of the test. Critics say Mathews gives too little weight to the dignity and fairness that procedures offer apart from accuracy and too much to quantifiable variables such as cost.87 They add that it gives courts enormous discretion in evaluating each factor and balancing them. Justice Rehnquist once called the balance "simply an ad hoc weighing which depends to a great extent upon how the Court subjectively views the underlying interests at stake."88 The test is so malleable that it often produces five-to-four decisions, and sometimes a three-way division among the Justices, and it also allows a public school student to be suspended after a brief informal meeting with a school official.89 It remains the test courts use to decide what procedures are required. A strong answer works with that flexibility by naming the extra procedure the claimant wants and identifying the fact that shows what it would add or cost.