Who Gets to Write ERISA’s Rules? Exhaustion, Freedom of Contract, and the Questions Bolton Leaves Open
An important new ERISA decision asks a question courts should have been asking all along: where does this requirement come from?
In Bolton v. Inland Fresh Seafood Corp. of America, Inc., No. 24-10084 (11th Cir. Oct. 13, 2026) (en banc), the Eleventh Circuit held that ERISA itself does not require exhaustion of statutory claims, including claims for breach of fiduciary duty. It overruled Mason v. Continental Group, Inc., 763 F.2d 1219 (11th Cir. 1985), and its progeny to the extent they imposed that requirement. Their policy justifications could not supply a condition missing from the statute.
The court also rejected an alternative invitation to impose exhaustion through judicial discretion. Even assuming the discretion described in McCarthy v. Madigan, 503 U.S. 140 (1992), extended to these claims, the court declined to exercise it: exhaustion delays the access to federal courts ERISA promises, and fiduciaries accused of misconduct may have a conflict when reviewing accusations against themselves.
But Bolton expressly reserved two questions. Footnote 1 leaves the circuit’s benefits-exhaustion cases untouched. Part III leaves open whether a plan can require exhaustion of statutory claims by contract, because this plan did not.
Those reservations identify two deeper issues. If statutory silence cannot support an automatic exhaustion requirement for fiduciary claims, why does it support one for benefit claims? And if private agreements cannot eliminate ERISA’s statutory protections, why should an insurance policy’s choice-of-law clause eliminate otherwise applicable state insurance protections that ERISA expressly preserves?
Neither question is answered by saying that exhaustion promotes efficiency or that ERISA favors written plan terms. Both propositions have force. Neither establishes the limits of judicial or contractual power.
Issue One: Benefits Exhaustion After Bolton
The label obscures the question
A benefits lawsuit under § 502(a)(1)(B) is itself a statutory cause of action. Calling it “contractual” describes the source of the underlying entitlement; it does not make the federal remedy less statutory. Conversely, § 502(a)(3) can enforce plan terms as well as statutory duties. The subsection pleaded therefore cannot decide the exhaustion question. Courts examine the alleged wrong, as Hitchcock v. Cumberland University 403(b) DC Plan, 851 F.3d 552 (6th Cir. 2017), did in distinguishing genuine statutory violations from relabeled benefits disputes.
There is nevertheless a plausible basis for treating the claims differently. In Zipf v. American Telephone & Telegraph Co., 799 F.2d 889 (3d Cir. 1986), the Third Circuit identified § 503, 29 U.S.C. § 1133, as a foundation for the distinction. That section specifically addresses benefit denials and internal review. Administrators also have experience evaluating medical evidence and applying benefit formulas, with no comparable institutional advantage in determining whether fiduciaries violated federal law.
But § 503 commands the plan to provide notice and an opportunity for full and fair review. It does not expressly command the participant to pursue that opportunity before suing. Converting the plan’s obligation into a condition on the participant’s remedy requires an additional interpretive step. The question is whether Congress authorized it.
The best defense of benefits exhaustion—and the answer
The defense deserves to be stated at full strength. Congress required a claims-and-review mechanism in § 503. Section 502(a)(1)(B) enforces benefits under plan terms. The Department of Labor’s claims regulation, 29 C.F.R. § 2560.503-1, contemplates appeals before civil litigation. And administrators ordinarily should have an opportunity to determine entitlement before courts adjudicate the dispute. Together, these features arguably imply exhaustion even without an express command.
The Supreme Court has reasoned against that background. Heimeshoff v. Hartford Life & Accident Insurance Co., 571 U.S. 99 (2013), described the courts of appeals’ exhaustion practice and enforced a contractual limitations period within that framework. It did not decide whether courts properly created the exhaustion requirement. But its acceptance of the framework means reconsideration seeks doctrinal change, not recognition that benefits exhaustion has already disappeared.
The answer begins with the distinction between an opportunity and an obligation. Requiring plans to offer review remains meaningful even if participants are not invariably compelled to use it. Appeals can correct errors and avoid litigation without becoming an unstated condition of the federal remedy.
Three considerations sharpen the objection. First, Congress knows how to require a pre-suit process. The 1980 multiemployer amendments expressly require arbitration of specified withdrawal-liability disputes. 29 U.S.C. § 1401(a)(1). That later enactment does not establish Congress’s 1974 intent, but it supplies a useful textual comparison.
Second, the drafting history is suggestive. James Wooten’s A Reflection on ERISA Claims Administration and the Exhaustion Requirement, 6 Drexel L. Rev. 573 (2014), examines litigation preconditions omitted from the enacted legislation and staff expectations concerning exhaustion. Rejection of other preconditions is not necessarily rejection of exhaustion. Still, expectations that never became statutory language cannot substitute for an enacted requirement.
Third, ERISA’s federal common law has limits. Mertens v. Hewitt Associates, 508 U.S. 248 (1993), rejected using common-law authority to revise the statutory text. That constraint should apply to restrictions on enforcement as well as expansions of remedies. Interpreting an undefined plan term differs from imposing a prerequisite that can defeat an otherwise valid claim because the participant omitted a step Congress did not expressly prescribe.
Statutory silence does not invalidate every judicial exhaustion doctrine. McCarthy recognizes a role for judicial discretion, and Bolton declined to impose exhaustion even assuming the asserted discretion existed. The dispute concerns the source, scope, and exercise of that authority—and whether Bolton’s reasons for declining exhaustion apply to benefits disputes too.
What the courts that created the rule have said
The doctrine’s influential early foundation is Amato v. Bernard, 618 F.2d 559 (9th Cir. 1980). Recognizing that ERISA did not expressly address exhaustion, Amato relied on § 503, the regulation, legislative history analogizing benefit actions to Labor Management Relations Act litigation, and policy. Internal review could reduce unnecessary litigation, encourage consistent decisions, lower costs, and give courts the benefit of prior trustee consideration.
Record development was therefore part of the original justification. Nor is it accurate to say courts never offered a federal-common-law rationale. The criticism is that the asserted authority may not extend as far as the rule they created.
Later opinions candidly acknowledge the doctrine’s judicial origin. Paese v. Hartford Life & Accident Insurance Co., 449 F.3d 435 (2d Cir. 2006), and Metropolitan Life Insurance Co. v. Price, 501 F.3d 271 (3d Cir. 2007), expressly describe it as judge-made. In re Household International Tax Reduction Plan, 441 F.3d 500 (7th Cir. 2006), emphasizes judicial discretion. Watts v. BellSouth Telecommunications, Inc., 316 F.3d 1203 (11th Cir. 2003), similarly acknowledges the courts’ authorship.
Springer v. Wal-Mart Associates’ Group Health Plan, 908 F.2d 897 (11th Cir. 1990)—the benefits precedent cited in Bolton’s reservation—relied on Mason, § 503, and the plan’s appeal provision. Bolton undermines one strand of that reasoning without eliminating the others or overruling Springer.
The clearest judicial challenge appears in Wallace v. Oakwood Healthcare, Inc., 954 F.3d 879 (6th Cir. 2020). The majority held that a fiduciary invoking exhaustion must at least have described the appeal procedures in the plan document. Judge Thapar’s concurrence questioned the underlying legitimacy of imposing exhaustion where neither the statute nor the plan requires it. The majority limited the doctrine’s application; the concurrence challenged its foundation. Bolton makes that challenge newly relevant without resolving it for benefit claims.
Nonjurisdictional answers a procedural question
Paese held that failure to exhaust is a nonjurisdictional affirmative defense, which Hartford forfeited. That classification is sound: exhaustion concerns a step before litigation, not the court’s authority over the parties or the statutory category of dispute.
But the label does not supply missing authority. It addresses who must raise the defense and whether it can be forfeited; it does not explain why courts may create it. Congress can establish nonjurisdictional preconditions, and courts recognize legitimate common-law defenses. The question remains the source and limits of this one.
Nor do exceptions prove that exhaustion cannot have a statutory foundation. Statutes can contain exceptions or authorize implementing regulations. Ross v. Blake, 578 U.S. 632 (2016), distinguished unauthorized judicial exceptions to the Prison Litigation Reform Act from the statute’s own limitation to available remedies. Judicially created rules allow a different inquiry, but courts applying them remain constrained by precedent.
Jones v. Bock, 549 U.S. 199 (2007), supplies the complementary point: even where Congress expressly requires exhaustion, courts cannot freely add procedural restrictions. Neither case decides ERISA exhaustion. Both reinforce the need to identify authority for a litigation barrier rather than assume that desirable administration supplies it.
The policy case, rationale by rationale
Some early reasoning treated preservation of arbitrary-and-capricious review as a purpose of exhaustion. Denton v. First National Bank of Waco, 765 F.2d 1295 (5th Cir. 1985), identified that objective. It cannot carry the same weight after Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101 (1989), established de novo review as the default unless discretion is validly conferred. Deference is conditional, not an outcome courts should construct procedural rules to protect.
State insurance regulation further limits that rationale. The Sixth, Seventh, and Ninth Circuits have upheld restrictions on discretionary clauses: American Council of Life Insurers v. Ross, 558 F.3d 600 (6th Cir. 2009); Fontaine v. Metropolitan Life Insurance Co., 800 F.3d 883 (7th Cir. 2015); Standard Insurance Co. v. Morrison, 584 F.3d 837 (9th Cir. 2009); and Orzechowski v. Boeing Co. Non-Union Long-Term Disability Plan, 856 F.3d 686 (9th Cir. 2017).
These restrictions differ in scope, but their consequence matters: where governing law invalidates the grant of discretion, preserving deferential review cannot justify exhaustion.
Other rationales survive. Internal appeals can correct mistakes, assemble medical evidence, clarify disagreements, and avoid litigation. Those benefits remain under de novo review. Yet usefulness alone does not establish authority to impose a prerequisite to Congress’s cause of action—or explain why a missed appeal should permanently defeat enforcement rather than produce a less severe procedural consequence.
The benefits/statutory distinction also fits these policies imperfectly. Fiduciary disputes can benefit from factual development and settlement. Benefits appeals can involve delay or inaccessible procedures. And an insurer’s review process differs from adjudication by a public agency.
Where an insurer both decides and pays claims, fiduciary obligations do not eliminate its financial incentive. Metropolitan Life Insurance Co. v. Glenn, 554 U.S. 105 (2008), recognizes that conflict as a factor in review; Conkright v. Frommert, 559 U.S. 506 (2010), underscores the continued importance of validly conferred discretion. Those decisions do not establish automatic futility. But neither resolves whether courts should impose an additional pre-suit requirement directing participants back to a financially interested reviewer. John Langbein’s Trust Law as Regulatory Law: The Unum/Provident Scandal and Judicial Review of Benefit Denials Under ERISA, 101 Nw. U. L. Rev. 1315 (2007), provides important context for that institutional question.
The exceptions require a usable process
Kirkendall v. Halliburton, Inc., 707 F.3d 173 (2d Cir. 2013), excused exhaustion where the participant reasonably understood ambiguous plan language not to require it. Watts and Spinedex Physical Therapy USA Inc. v. United Healthcare of Arizona, Inc., 770 F.3d 1282 (9th Cir. 2014), likewise recognize the importance of a reasonable reading of the plan’s procedures. These decisions do not hold that every use of “may” makes an appeal optional.
Sakwa v. Hartford Life & Accident Insurance Co., No. 1:25-cv-04546 (N.D. Ill. Mar. 25, 2026), illustrates another limit. After completed appeals concerning a benefit calculation and a letter confirming exhaustion, Hartford sought another appeal. The court permitted the benefits claim to proceed, including on futility grounds, while dismissing the fiduciary claim. Vaught v. Scottsdale Healthcare Corp. Health Plan, 546 F.3d 620 (9th Cir. 2008), separately distinguishes exhaustion of a claim from exhaustion of every supporting issue.
The regulation supplies additional protections. Group health and disability plans may require no more than two mandatory appeal levels. Disability and applicable group-health procedures provide at least 180 days to appeal; the general rule for other claims, including ordinary life and accidental-death claims, provides at least 60 days. See 29 C.F.R. § 2560.503-1(c), (d), (h). Deemed exhaustion follows specified regulatory noncompliance, with a strict-adherence rule and a narrow de minimis exception for disability claims. § 2560.503-1(l).
These rules prevent plans from demanding procedural compliance while disregarding their own obligations. They also give exhaustion a substantial regulatory setting. The Department invokes delegated authority under §§ 503 and 505. Nevertheless, the regulation’s assumption of an exhaustion framework does not itself settle whether Congress authorized the universal judicial prerequisite. That question requires analysis of the statute and the delegation, not inference from the regulation’s existence alone.
Where that leaves the doctrine
Benefits exhaustion has a stronger structural foundation than exhaustion of statutory violations, principally because of § 503 and the plan-based entitlement. But the step from a required opportunity for review to a mandatory barrier to suit remains contestable. Useful policy cannot, by itself, complete it.
The most defensible version of the doctrine would tie exhaustion to clearly disclosed, mandatory, compliant plan procedures, reasonably understood by participants. Watts, Spinedex, Kirkendall, and Wallace support aspects of that approach. They do not collectively establish that exhaustion everywhere depends on an express contractual requirement, and their different holdings should not be collapsed into one rule.
Decisions enforcing exhaustion as a federal requirement despite permissive or deficient plan language—including Wert v. Liberty Life Assurance Co. of Boston, 447 F.3d 1060 (8th Cir. 2006), and Perrino v. Southern Bell Telephone & Telegraph Co., 209 F.3d 1309 (11th Cir. 2000)—present the sharper challenge. Bolton invites renewed scrutiny of their justification. It does not authorize lower courts to disregard controlling benefits precedent.
Issue Two: Freedom of Contract and the Saving Clause
The contract question begins where judicial exhaustion ends
The Department of Labor told the Bolton court that contractual requirements should not allow parties to surrender ERISA rights. The court recorded that position without adopting a categorical prohibition. Its references to Clayton v. International Union, UAW, 451 U.S. 679 (1981), and Glover v. St. Louis-San Francisco Railway Co., 393 U.S. 324 (1969), likewise support scrutiny rather than automatic invalidity.
Clayton examined internal union appeals through considerations including hostility, adequacy of relief, and delay. Glover addressed futility where employer and union allegedly acted together against employees. Neither was an ERISA case. Their relevance is that contractual procedures operate within a statutory framework and cannot be evaluated apart from access to the protected remedy.
ERISA arbitration decisions illustrate a related distinction. In Williams v. Shapiro, 161 F.4th 1313 (11th Cir. 2025), the problem was a restriction on plan-wide relief available under § 502(a)(2), not arbitration itself. Decisions including Smith v. Board of Directors of Triad Manufacturing, Inc., 13 F.4th 613 (7th Cir. 2021), Cedeno v. Sasson, 100 F.4th 386 (2d Cir. 2024), and Platt v. Sodexo, S.A., 148 F.4th 709 (9th Cir. 2025), similarly distinguish selecting a procedure from prospectively eliminating statutory relief.
ERISA also directly limits contractual freedom. Section 404(a)(1)(D) requires compliance with plan documents only insofar as they are consistent with the statute. Section 410(a) invalidates provisions relieving fiduciaries of specified statutory responsibilities or liability. Neither makes every procedural condition invalid, but both contradict an unrestricted freedom-of-contract theory.
Lawful terms retain substantial force. US Airways, Inc. v. McCutchen, 569 U.S. 88 (2013), declined to use equitable principles to override an express reimbursement provision. Heimeshoff enforced a contractual limitations period subject to limits including reasonableness and contrary controlling statutes. The question is where permissible definition of the bargain ends and impermissible displacement of governing law begins.
Congress preserved an external source of insurance rules
ERISA’s saving clause preserves state laws regulating insurance from its general preemption provision. 29 U.S.C. § 1144(b)(2)(A). The insured-plan limitation matters: under FMC Corp. v. Holliday, 498 U.S. 52 (1990), states regulate insured plans indirectly through their insurers, while the deemer clause protects self-funded plans from that route of regulation. And under Kentucky Association of Health Plans, Inc. v. Miller, 538 U.S. 329 (2003), the insurance-saving inquiry requires a law directed specifically toward insurance entities that substantially affects risk pooling.
Once a mandatory insurance rule is applicable and survives preemption, contrary policy terms do not necessarily defeat it. UNUM Life Insurance Co. of America v. Ward, 526 U.S. 358 (1999), applied California’s notice-prejudice rule despite contrary policy language and rejected reliance on ERISA’s plan-documents duty to avoid it. Rush Prudential HMO, Inc. v. Moran, 536 U.S. 355 (2002), likewise sustained insurance regulation affecting claims resolution. Larson v. United Healthcare Insurance Co., 723 F.3d 905 (7th Cir. 2013), explains how insurance requirements become terms of an insured plan.
Hughes v. United of Omaha Life Insurance Co., No. 1:17-cv-00779 (E.D. Cal. Oct. 6, 2017), illustrates the point through Heimeshoff: California’s statutory limitations floor defeated a shorter contractual period.
That raises the choice-of-law question. If an insurance protection otherwise binds the insurer, why should a provision selecting another state’s law accomplish what a directly conflicting policy term could not?
The strongest contrary answer: decide applicability first
The insurer’s response is substantial: it is selecting applicable law, not avoiding applicable law. The saving clause addresses preemption; it does not expressly provide an interstate conflicts rule. A saved law may not reach a policy because of its territorial scope, effective date, or other limitations. Multiple states may have legitimate interests. “Saved,” “applicable,” and “nonwaivable” therefore describe separate conclusions.
But that distinction can become circular. The clause establishes that the protective law is inapplicable; the resulting inapplicability is then offered as proof that the clause did not evade it. A sound analysis must consider the mandate’s intended reach, the relevant contacts, and the governing doctrine’s limits on contractual selection. A reasonable connection to the chosen state does not necessarily answer whether the clause can displace another state’s mandatory regulation.
Restatement (Second) of Conflict of Laws § 187(2)(b) requires more than a fundamental policy. It also asks whether the displaced state has a materially greater interest and whether its law would apply absent the clause. Nor does § 192 supply a general participant-domicile default for group insurance. Comment h ordinarily points toward the law governing the master policy, often associated with the employer’s principal place of business and policy delivery. Gunn v. Continental Casualty Co., 968 F.3d 802 (7th Cir. 2020), discusses that principle outside ERISA.
California Insurance Code § 10110.6 expressly reaches covered policies insuring California residents whether issued inside or outside California. That language strongly demonstrates the legislature’s intended reach. It does not, standing alone, establish a materially greater interest or determine what law governs absent the clause. Those questions still require comparison of the states’ interests and relevant contacts. Conversely, a statute limited to locally issued policies may never reach the dispute.
The selected state’s silence does not settle matters either. A prohibition in one state and permission in another can create a genuine conflict; two affirmative, incompatible commands are unnecessary. The applicable conflicts framework must determine the consequence.
The cases
The authorities employ different frameworks and address different issues:
| Authority | Approach and significance |
|---|---|
| Wang Laboratories, Inc. v. Kagan, 990 F.2d 1126 (9th Cir. 1993) | Enforced a contractual selection in choosing a limitations period, subject to reasonableness and fundamental fairness. It did not decide displacement of an affirmative insurance mandate. |
| Buce v. Allianz Life Insurance Co., 247 F.3d 1133 (11th Cir. 2001) | Allowed selected state doctrine to inform the meaning of accident. Judge Barkett’s concurrence emphasized limits arising from saved mandatory insurance law. |
| DaimlerChrysler Corp. Healthcare Benefits Plan v. Durden, 448 F.3d 918 (6th Cir. 2006) | Applied federal conflicts principles informed by the Restatement in declining to let selected law determine marital status. |
| Ellis v. Liberty Life Assurance Co. of Boston, 958 F.3d 1271 (10th Cir. 2020) | Gave strong effect to contractual selection and applied Pennsylvania law over Colorado’s discretionary-clause ban. |
| Brake v. Hutchinson Technology, 774 F.3d 1193 (8th Cir. 2014); Prudential Insurance Co. of America v. Doe, 140 F.3d 785 (8th Cir. 1998); Hankins v. Crain Automotive Holdings, 129 F.4th 1088 (8th Cir. 2025) | Reveal tension between enforcement of selected state law and statements limiting the parties’ ability to choose state law as the governing law of an ERISA plan. Their different contexts warrant care before assigning a single circuit rule. |
Ellis is the strongest appellate obstacle. Its federal approach emphasizes uniformity and administrative efficiency. The criticism is that such reasoning risks treating saved insurance regulation as law courts may choose to borrow rather than regulation Congress preserved. That challenges Ellis; it does not show that Ellis already accepts the argument. A later Tenth Circuit panel cannot disregard the decision merely because counsel supplies a different argument. Distinguishing the case or seeking en banc or Supreme Court review may be necessary. Certiorari was denied in 2021.
Several appellate decisions have left the precise issue unresolved, including White v. Guardian Life Insurance Co. of America, No. 24-2681 (9th Cir. May 14, 2025), Earle v. Unum Life Insurance Co. of America, No. 20-55868 (9th Cir. Oct. 19, 2021), and Jensen v. Life Insurance Co. of North America, No. 24-4014 (10th Cir. Apr. 4, 2025), all unpublished.
Grice v. Metropolitan Life Insurance Co., No. 25-50566 (5th Cir. Aug. 26, 2026) (unpublished), offers a particularly useful illustration. A Texas resident’s plan selected California law. MetLife argued that the choice displaced Texas’s ban, while California’s ban did not cover the Texas resident. Neither prohibition would apply. The court expressed doubt about using contractual selection to produce that result, invoking the saving clause, but affirmed on the merits assuming de novo review. Its skepticism is relevant dicta, not a holding resolving the issue.
Recent district-court decisions are less receptive. Moorhead v. Unum Life Insurance Co. of America, No. 25-cv-01826-HSG (N.D. Cal. Mar. 30, 2026), enforced a Massachusetts clause over California’s ban and addressed the saving-clause objection. Farris v. Life Insurance Co. of North America, No. 25-cv-04164-RS (N.D. Cal. May 8, 2026), enforced a North Carolina clause using Wang, employer contacts, and uniformity.
Both opinions reproduce California’s broad statutory language. The objection is not that they never encountered the issue. It is that their federal conflicts analysis gives insufficient independent weight to the legislature’s express regulatory reach and Congress’s preservation of insurance regulation. Their answers deserve direct engagement: why should a connection sufficient to make selection reasonable also suffice to displace a protection the insured’s state has declared mandatory?
Nicolopoulos supplies a powerful answer—with a qualification
In BlueCross BlueShield of Tennessee, Inc. v. Nicolopoulos, 136 F.4th 681 (6th Cir. 2025), an insurer sought to block New Hampshire’s enforcement of a fertility-coverage mandate. Its plan selected Tennessee law, and the insurer invoked its ERISA duty to follow plan documents. The Sixth Circuit rejected that defense: New Hampshire was regulating the company as an insurer, and its simultaneous fiduciary role did not confer immunity from saved regulation.
The qualification matters. This was a challenge to regulatory enforcement, not a participant’s suit deciding whether a choice-of-law clause defeated a discretionary-clause ban. The insurer was not licensed in New Hampshire. The opinion does not resolve every issue of territorial reach or conflicts law.
Nevertheless, its reasoning creates a consistency problem. If plan documents cannot defeat an applicable regulatory mandate in an enforcement proceeding, why should those documents necessarily defeat a participant seeking the same protection? An answer must identify a meaningful difference in statutory reach, remedy, or conflicts law. Repeating the duty to follow plan documents is insufficient.
Uniformity matters, but Congress did not make it absolute
Insurers gain efficiency from uniform rules. Yet Congress preserved substantial state insurance regulation, accepting resulting differences between insured and self-funded arrangements. Metropolitan Life Insurance Co. v. Massachusetts, 471 U.S. 724 (1985), and Ward recognize that consequence. The burden of complying with state insurance law cannot itself justify eliminating the statutory choice that produces it.
A discretionary-clause ban affects judicial review after denial. It may influence administration and litigation incentives, but those effects do not establish authority to override otherwise applicable regulation. The Department’s amicus brief in In re Mathias, 867 F.3d 727 (7th Cir. 2017), advanced a related argument about contractual limits in a venue dispute the court resolved against the Department. The position supplies an argument, not controlling authority.
John Coyle’s The Mystery of the Missing Choice-of-Law Clause, 56 U.C. Davis L. Rev. 707 (2022), adds broader context: insurance contracts operate within distinctive regulatory and drafting practices. Their clauses should not be analyzed as though they necessarily reflect an ordinary negotiated commercial allocation of governing law.
The proposed rule is narrower than automatic application of the participant’s home-state law. In an insured ERISA arrangement, contractual selection should operate within the space left by federal law and genuinely applicable, nonwaivable state insurance law. Judicial preferences for uniform administration should not give the clause superior force. Applying that approach requires separate findings on scope, saved status, nonwaivability, relevant contacts, and federal-law constraints—and direct confrontation with contrary circuit precedent.
One further boundary is essential: preserving an insurance protection does not preserve every state-law remedy. Aetna Health Inc. v. Davila, 542 U.S. 200 (2004), limits attempts to supplement ERISA’s enforcement scheme. Enforcing a mandatory insurance term through § 502 differs from seeking additional state tort damages.
The Same Discipline, in Both Directions
These debates require courts to identify the source of a restriction before enforcing it. Internal appeals can be excellent policy. Choice-of-law clauses can reduce uncertainty. Those advantages deserve consideration where governing law leaves room for them. They do not independently authorize judges to add conditions to federal remedies or insurers to remove mandatory protections.
For benefits exhaustion, the unresolved question is whether § 503 and ERISA’s structure justify converting an offered review process into a universal prerequisite to suit, particularly where the plan does not clearly impose one. For choice of law, it is whether federal conflicts principles properly account for the mandatory insurance regulation Congress preserved.
Bolton does not resolve either question. It supplies a reason to insist that both receive an answer grounded in legal authority, rather than the repetition of policies that have become familiar through use.
What this means for a denied claimant
Nothing in Bolton excuses a disability, life, or accidental death claimant from the appeal deadline in a denial letter. Complete every mandatory appeal, on time, in writing, and build the record as if a federal judge will read it — because one will.
What Bolton changes is the argument available when an insurer invokes exhaustion against a plan that never clearly required it, or invokes a choice-of-law clause against the insurance law of the state where you live.
Frequently Asked Questions
No. It addressed statutory claims and expressly left benefits-exhaustion precedent untouched. Claimants should continue to protect appeal deadlines and complete required appeals unless a legally supported exception applies. The argument for reconsidering the doctrine is not permission to disregard current law.
The leading appellate decisions treat it as a nonjurisdictional affirmative defense. It may be forfeited, and courts recognize exceptions. Regulatory deemed exhaustion depends on the applicable requirements, exceptions, and governing case law; it does not automatically follow from every departure from a plan’s own procedures.
The answer depends on governing precedent, the policy, and the statute. Appellate choice-of-law frameworks differ, and several courts have left the precise saving-clause issue unresolved. Ellis enforced a clause over a state ban; Moorhead and Farris did so against California’s ban. An opposing argument must establish why the protective law is saved, applicable, and nonwaivable. Neither Wang nor Buce squarely resolves every such conflict.
No. The court recorded the Department of Labor’s objection to contracting away statutory rights and discussed limits on contractual exhaustion. It reserved the issue because the plan before it did not require exhaustion.
Brent Dorian Brehm is the founding shareholder of Dorian Law P.C. His practice focuses on ERISA and insurance-denial litigation, including long-term disability, life insurance, accidental death, and beneficiary disputes. He presented on ERISA choice-of-law clauses and discretionary-clause bans at an American Bar Association program in early 2026. Learn more or request a consultation at dorianlaw.com/contact.