Whether a federal court reviews your ERISA benefit denial fresh or defers to the insurance company usually comes down to one thing: whether your plan document contains a discretionary clause. That single piece of language is the largest factor in the standard of review courts apply to ERISA claims, and it often decides the case before the evidence is weighed. Understanding how discretionary clauses work, when they hold up, and when they don’t is the difference between a realistic shot at reversal and a fight against built-in deference.
What a Discretionary Clause Does
A discretionary clause is language in your plan giving the administrator authority to interpret the plan’s terms and decide who qualifies for benefits. “Sole and absolute discretion” and “full power to determine eligibility” are typical formulations. The wording looks like boilerplate, but its legal effect is significant: it controls how much room a judge has when reviewing a denial.
The Supreme Court set the framework in Firestone Tire & Rubber Co. v. Bruch. The Court held that an ERISA benefit denial is reviewed under a de novo standard “unless the benefit plan gives the administrator or fiduciary discretionary authority to determine eligibility for benefits or to construe the terms of the plan.”1Justia. Firestone Tire and Rubber Co. v. Bruch, 489 U.S. 101 (1989) That single sentence created two tracks for every ERISA lawsuit. Without the discretionary language, the administrator loses its protective layer of deference entirely.
De Novo Review When the Clause Is Absent
When a plan lacks discretionary language, the judge examines the evidence from scratch. The insurer’s reasoning gets no special weight, no presumption of correctness, no deference. The court looks at the plan terms and the record, then makes its own independent judgment about whether you qualify.1Justia. Firestone Tire and Rubber Co. v. Bruch, 489 U.S. 101 (1989)
You still carry the burden of showing you’re entitled to benefits under the plan’s terms. And review is generally limited to the documents and evidence the administrator had when it issued its final denial. Courts disagree on how strict that limit is. The Eleventh Circuit has explicitly rejected any requirement that a claimant show “exceptional circumstances” before introducing evidence outside the administrative record during de novo review, reasoning that the court’s job is to make the same judgment the administrator should have made. Other circuits are more restrictive. If new medical evidence is on the table after a denial, the rules of your circuit matter.
Claimants generally prefer de novo review because the question changes. The judge is no longer asking whether the insurer’s decision was reasonable. The judge is asking whether you qualify. Those are very different inquiries, and the second one is far more favorable when strong medical evidence has been discounted by the administrator.
Abuse of Discretion When the Clause Is Present
A valid discretionary clause pushes the court to the abuse of discretion standard, also called the arbitrary and capricious standard. Courts treat these terms as interchangeable.2Digital Commons at St. Mary’s University. ERISA: The Arbitrary and Capricious Rule Under Siege Under this framework, a judge will not overturn a denial simply because the judge would have reached a different conclusion. The question is whether the administrator’s decision was reasonable and supported by substantial evidence.
Substantial evidence means more than a token amount but less than a preponderance. If the administrator followed a logical process, considered the plan language, and connected the facts to its conclusion, the denial will typically stand. This is where most claimants’ cases go to die. Even a judge who finds your arguments more persuasive will uphold the denial when the insurer’s interpretation falls within a zone of reasonableness.
The standard isn’t a blank check. A decision that contradicts the plan’s plain language, ignores relevant medical evidence, or relies on a selective reading of the record can still be overturned. Courts also watch for procedural irregularities: unreasonable delays in processing an appeal, failure to provide required notices, and similar breakdowns. These don’t automatically flip the review to de novo, but they can reduce the level of deference, sometimes requiring the administrator to support its decision with evidence “bordering on a preponderance” rather than the usual lower bar.
When a Conflict of Interest Enters the Picture
Review gets more skeptical when the same entity that decides your claim also pays it. An insurance company that both evaluates disability claims and cuts the benefit checks has a direct financial incentive to deny. The Supreme Court addressed this in Metropolitan Life Insurance Co. v. Glenn, holding that such a conflict “must be weighed as a factor in determining whether there is an abuse of discretion.”3Justia. Metropolitan Life Ins. Co. v. Glenn, 554 U.S. 105 (2008)
The conflict doesn’t change the standard. It stays abuse of discretion. But the Court laid out practical markers for how much weight the conflict should carry. The conflict matters more when the insurer has a history of biased claims handling, when it emphasized medical reports favoring denial while downplaying contrary evidence, or when it encouraged the claimant to apply for Social Security disability benefits and then ignored the resulting finding of disability. The conflict matters less when the insurer has taken structural steps to reduce bias, like walling off claims staff from financial decision-makers or imposing internal accuracy checks that penalize wrong decisions in either direction.3Justia. Metropolitan Life Ins. Co. v. Glenn, 554 U.S. 105 (2008)
Proving that financial incentives actually tainted a specific decision is harder than it sounds. Courts typically limit review to the administrative record and generally don’t allow broad discovery into an insurer’s internal operations, training materials, or claims-handler compensation structures. Some circuits recognize an exception for investigating decision-maker bias, but the rule is far from universal. Targeted motions for internal claim manuals or bonus structures sometimes succeed, but results vary by jurisdiction.
State Bans on Discretionary Clauses
More than 20 states have prohibited discretionary clauses in insurance policies outright. The National Association of Insurance Commissioners adopted a model act that bars health carriers and disability insurers from including any provision “purporting to reserve discretion to the health carrier to interpret the terms of the contract.”4NAIC. Prohibition on the Use of Discretionary Clauses Model Act The number of states with some form of this prohibition continues to grow.
These bans work because of ERISA’s preemption structure. ERISA broadly preempts state laws that “relate to” employee benefit plans, but its saving clause carves out an exception for state laws that regulate the business of insurance.5Notre Dame Journal of Legislation. ERISA Preemption Under the Deemer Clause as Applied to Employer Health Care Plans with Stop-Loss Insurance Courts have consistently held that state bans on discretionary clauses qualify as insurance regulation, so they survive preemption when applied to insured plans. The practical effect: if you live in a state with such a ban and your benefits come through an insured plan, the court applies de novo review regardless of what the plan document says.
Determining which state’s law governs the insurance contract is a threshold question, and it isn’t always where you live. It can be where the policy was issued or where the employer is headquartered. Getting that analysis wrong means building the case around the wrong standard of review.
Self-Funded Plans and the Deemer Clause
State bans have a critical blind spot. They don’t reach self-funded plans. When an employer pays claims directly out of its own assets rather than purchasing an insurance policy, the plan falls outside state insurance regulation. ERISA’s deemer clause, codified at 29 U.S.C. ยง 1144(b)(2)(B), provides that an employee benefit plan “shall not be deemed to be an insurance company or other insurer… for purposes of any law of any State purporting to regulate insurance companies.”6Office of the Law Revision Counsel. 29 USC 1144 – Preemption of State Laws
The Supreme Court confirmed this reading in FMC Corp. v. Holliday, holding that the deemer clause exempts self-funded ERISA plans from state laws regulating insurance within the meaning of the saving clause. The result is a two-tier system. Participants in insured plans may benefit from state discretionary-clause bans. Participants in self-funded plans are stuck with whatever the plan document says. Large employers increasingly self-fund their benefit programs, so this distinction affects millions of workers who may not realize their plan operates differently from traditional insurance.
Finding out whether your plan is insured or self-funded is a necessary step before assessing which standard of review applies. The Summary Plan Description should identify the funding arrangement, though the language isn’t always obvious. If you can’t tell, the plan administrator is required to provide this information upon request.
Building the Record Before You Sue
Before you can challenge a benefit denial in federal court, you almost certainly need to complete your plan’s internal appeals process. The exhaustion requirement is court-imposed rather than statutory, but federal courts enforce it consistently. Skipping the internal appeal and going straight to a lawsuit will usually end in dismissal.
The Department of Labor’s claims procedure regulations set minimum timelines the plan must follow. For disability claims, the plan has 45 days to issue an initial decision, with two possible 30-day extensions for a maximum of 105 days. After a denial, you have at least 180 days to file an internal appeal.7U.S. Department of Labor. Benefit Claims Procedure Regulation FAQs Courts recognize narrow exceptions, most commonly futility when the internal process is so clearly stacked against you that completing it would accomplish nothing. A plan’s failure to follow its own review procedures can also excuse exhaustion. Neither exception is applied generously.
The appeal stage is where the standard-of-review question starts to matter practically. Under both de novo and abuse of discretion review, courts generally limit their analysis to the administrative record. The evidence you submit during your internal appeal is likely the same evidence the court will consider. Treating the appeal as a formality before the “real” fight in court is one of the costliest mistakes claimants make.
Check the plan document for a contractual limitations provision before you begin the appeals process. The Supreme Court held in Heimeshoff v. Hartford Life & Accident Insurance Co. that plans can contractually set their own limitation periods, and those periods are enforceable even if they begin running before you receive a final denial.8Oyez. Heimeshoff v. Hartford Life and Accident Insurance Co. The Court upheld a three-year limitation period that began on the date proof of loss was due, not on the date of the final denial. If the deadline in your plan runs from proof of loss rather than final denial, the clock may already be ticking during your internal appeal.
What Winning Actually Looks Like
ERISA’s remedies are narrower than most claimants expect. Under Section 502(a)(1)(B), a successful participant can recover “benefits due to him under the terms of his plan,” enforce rights under the plan, or obtain a court declaration clarifying future benefit rights.9Office of the Law Revision Counsel. 29 USC 1132 – Civil Enforcement You cannot recover punitive damages, emotional distress damages, or other extracontractual relief. Every federal circuit to address the question has reached the same conclusion, and the Supreme Court’s reasoning in Mertens v. Hewitt Associates confirmed that ERISA’s equitable framework does not support punitive awards.10Legal Information Institute. Mertens v. Hewitt Associates, 508 U.S. 248 (1993) The insurer’s worst-case outcome is paying benefits it arguably owed all along, plus potentially your attorney’s fees. The economic incentive to deny borderline claims and force litigation is baked into the statute’s design.
Winning in court also doesn’t always mean receiving a check. When a court finds the administrator’s denial flawed, it faces a choice: order the plan to pay benefits immediately, or send the case back to the administrator for a new review. Courts frequently choose remand when they identify procedural problems without resolving whether the claimant actually qualifies. That sends you back to the same administrator, often the same insurance company, for another round of evaluation. An immediate award is appropriate when the record leaves no reasonable doubt about eligibility, and some circuits have held that an administrator doesn’t get a “second bite at the denial apple” when the first decision was contrary to the facts. If the court’s concern is procedural rather than substantive, remand is the more common outcome.
Building an administrative record that eliminates ambiguity about eligibility is the best way to push a court toward a direct award rather than a remand. Comprehensive medical evidence, vocational assessments if relevant, and clear documentation tying your condition to the plan’s definition of disability all narrow the room for a different conclusion on remand. The less wiggle room you leave, the more likely the court is to end the fight rather than restart it.