Custody Orders Are Only a Starting Point for Minor-Child Domicile Under Michigan No-Fault (Overruling Grange)

Case: Frownfelter v Esurance Property & Casualty Insurance Company (Docket Nos. 168356 & 168357)
Court: Supreme Court of Michigan
Date: July 22, 2026
Author: Justice ZAHRA (majority); Justice THOMAS (concurring)

1. Introduction

This consolidated no-fault insurance dispute arises from severe injuries suffered by 15-year-old McKenna Frownfelter in a January 8, 2020 motor-vehicle crash while riding to school from her father’s home. The central question is insurer priority for personal protection insurance (PIP) benefits under Michigan’s no-fault act, MCL 500.3101 et seq., specifically MCL 500.3114(1), which extends PIP coverage to certain relatives “domiciled in the same household” as the named insured.

The parties’ positions turned on whether plaintiff was domiciled with her father (Corey Frownfelter) or her mother (Stephanie Lawrence) at the time of the accident. Although the 2011 divorce judgment stated a 50/50 residence arrangement and joint physical custody, deposition testimony indicated plaintiff actually spent closer to 75% of her time at her mother’s house and 25% at her father’s. The litigation also spawned a separate, related question in a tort action: whether plaintiff was a “resident” of her father’s household for purposes of a policy “step-down” clause reducing bodily-injury liability limits for household residents.

Both the trial court and the Court of Appeals treated the custody order as dispositive of domicile under Grange Ins Co of Mich v Lawrence, 494 Mich 475 (2013), and concluded plaintiff was domiciled with her father because she stayed at his home the night before the crash. The Michigan Supreme Court used this case to reconsider—and ultimately reject—that approach.

Key issues

  • No-fault priority: Does a family-court custody order conclusively determine a minor child’s domicile for MCL 500.3114(1)?
  • Stare decisis: Should the Court overrule the Grange rule that custody orders set domicile “by operation of law” for no-fault purposes?
  • Policy “resident” step-down: If domicile analysis changes, what becomes of lower-court rulings that equated “domicile” with “resident” for policy step-down application?

2. Summary of the Opinion

The Supreme Court held that a custody order does not determine a child’s domicile for purposes of the no-fault act. The Court expressly overruled the “domicile rule set forth in Grange Ins Co of Mich v Lawrence,” which had required courts to treat custody orders as conclusive on domicile in no-fault disputes.

Going forward, a custody order is only a starting point. If the child’s actual living arrangements are clearly inconsistent with the custody order, the order should not be treated as representative of domicile. Instead, courts must determine domicile by considering the traditional no-fault domicile factors from Workman v Detroit Auto Inter-Ins Exch, 404 Mich 477 (1979), and Dairyland Ins Co v Auto-Owners Ins Co, 123 Mich App 675 (1983), plus additional joint-custody realities such as where the child actually spends most time and sleeps most nights.

The Court vacated the Court of Appeals’ holdings that (1) plaintiff was domiciled with her father and (2) plaintiff was a “resident” of her father’s home for the Esurance step-down clause, because the residency ruling “flowed directly” from the domicile conclusion. The case was remanded for further proceedings under the corrected domicile framework and for reconsideration of “resident” under the policy.

3. Analysis

3.1 Precedents Cited

A. Grange Ins Co of Mich v Lawrence, 494 Mich 475 (2013)

Grange framed the prior rule. It correctly reaffirmed a foundational principle: “a child of divorced parents has only one domicile at any given point in time.” But Grange then went further, holding that when a family court has entered a custody order, the child’s domicile is “established by operation of law,” making the custody order “determinative of the child’s domicile for all purposes, including the no-fault act.” For equal-time joint physical custody, Grange suggested the child’s domicile would “alternate” to match the parent with whom the child was living at the time.

How it influenced this case: The trial court and Court of Appeals applied Grange mechanically—treating the 50/50 divorce judgment as conclusive and deciding domicile based on the night-before location. The Supreme Court rejected this as legally unsupported and practically destabilizing (producing a day-to-day “alternating domicile” inconsistent with the concept of a fixed home).

B. Workman v Detroit Auto Inter-Ins Exch, 404 Mich 477 (1979)

Workman supplied the classic multifactor approach to “domiciled in the same household” for no-fault purposes. The Court restated Workman’s nonexhaustive factors:

  • subjective or declared intent to remain (permanently or indefinitely),
  • formality/informality of the relationship to household members,
  • physical proximity (same house/curtilage/premises), and
  • existence of another place of lodging.

How it influenced this case: The majority returned domicile to a fact-sensitive inquiry grounded in common law and no-fault practice, using Workman as the anchor rather than custody-order formalism.

C. Dairyland Ins Co v Auto-Owners Ins Co, 123 Mich App 675 (1983)

Dairyland added practical indicators (originally discussed for young adults) such as mailing address, possessions kept at parents’ home, documents listing the address, whether a room is maintained, and dependency for support.

How it influenced this case: The Court treated Dairyland as useful objective evidence of familial intent and residential reality, especially when formal custody paperwork lags behind evolving living arrangements.

D. Foundational domicile concepts: In re High, 2 Doug 515 (Mich, 1847); People v Dowdy, 489 Mich 373 (2011); Henry v Henry, 362 Mich 85 (1960); Hartzler v Radeka, 265 Mich 451 (1933); Herring v Mosher, 144 Mich 152 (1906)

These cases supplied the opinion’s doctrinal bedrock: domicile is a single “true, fixed, permanent home” concept rooted in fact and intent; every person must have one domicile; and domicile persists until replaced by choice or operation of law.

How they influenced this case: The Court used this lineage to show why an “alternating domicile” tied to custody swaps contradicts the permanence historically embedded in Michigan domicile law.

E. Domicile by operation of law and minors: Mississippi Band of Choctaw Indians v Holyfield, 490 US 30 (1989)

The Court relied on Holyfield for the proposition that minors typically lack capacity to form the requisite intent for domicile of choice, making their domicile generally track parents’ domicile “by operation of law.” The problem, the Court emphasized, is that joint custody yields two “legal residences,” so custody law alone does not answer which one is the child’s single domicile for no-fault purposes.

F. Family-court concepts (to distinguish from domicile): Sabatine v Sabatine, 513 Mich 276 (2024); Baker v Baker, 411 Mich 567 (1981); Rittershaus v Rittershaus, 273 Mich App 462 (2007); Gagnon v Glowacki, 295 Mich App 557 (2012); Pierron v Pierron, 486 Mich 81 (2010)

These authorities describe the Child Custody Act’s “established custodial environment,” which can exist in more than one home and is assessed under “best interests” principles—conceptually different from no-fault domicile.

How they influenced this case: They supported the majority’s core distinction: custody orders pursue child welfare and stability, not insurance-risk allocation; thus custody determinations cannot be treated as equivalent to domicile determinations for no-fault priority.

G. Workability and lower-court struggles post-Grange: Corbin v Meemic Ins Co, 340 Mich App 140 (2022); Davis v Auto Club Ins Ass'n (unpublished, June 23, 2025)

The majority cited examples illustrating that Grange did not deliver consistent administrability; courts distinguished it when orders were vague or reserved residence decisions to parents, and its strictness could deny domicile recognition where the child actually lived with a non-parent.

H. Stare decisis framework: Robinson v Detroit, 462 Mich 439 (2000); Hohn v United States, 524 US 236 (1998); McCormick v Carrier, 487 Mich 180 (2010)

These cases provided the Court’s factors for overruling precedent: practical workability, reliance hardship, public-interest detriment, and whether the old rule departed from precedent.

I. No-fault purpose: Shavers v Attorney General, 402 Mich 554 (1978)

The Court invoked Shavers to emphasize no-fault’s goals of “assured, adequate, and prompt reparation,” which are undermined when a child’s PIP coverage hinges on which parent’s house the child slept at the night before an accident.

3.2 Legal Reasoning

A. The Court’s core doctrinal move: disentangling custody from domicile

The opinion separates (1) family-court custody and “legal residence” determinations from (2) no-fault domicile determinations. The Court reasoned that:

  • The Child Custody Act centers on best interests and “established custodial environment,” not insurer priority or domicile.
  • An “established custodial environment” may exist with both parents and in more than one home, whereas domicile is singular.
  • Nothing in the Child Custody Act authorizes a family court to set domicile for no-fault purposes.

Therefore, the Grange assumption—equating physical custody with domicile—was “legally unsupported.”

B. Why the Grange “alternating domicile” concept was untenable

The Court attacked the practical and conceptual instability of alternating domicile in 50/50 arrangements:

  • It erodes domicile’s “permanence” by allowing weekly/daily domicile shifts.
  • It functionally recreates “dual domicile,” which Grange itself rejected, by making domicile contingent on transient custody exchanges.
  • It makes coverage indeterminate until the accident’s timing is known, impairing families’ and insurers’ ability to anticipate coverage.

C. Stare decisis applied—and Grange overruled

After concluding Grange was wrongly decided, the Court applied the Robinson v Detroit / McCormick v Carrier factors:

  • Workability: superficially simple but unreliable in practice (unclear orders; discretionary parenting-time; reality diverging from old orders; frequent 50/50 arrangements).
  • Reliance: limited (13 years) and unlikely to shape behavior ex ante in accident-driven statutes; “after-the-fact” awareness is not meaningful reliance.
  • Public-interest detriment: unpredictable coverage threatens no-fault’s remedial goals and could leave catastrophically injured children uninsured/underinsured.
  • Departure from precedent: Grange departed from the traditional multifactor domicile approach, especially in no-fault.

On balance, the Court found overruling compelled.

D. The new rule (framework) announced

For minor children of divorced/separated parents with joint custody and two legal residences:

  • Start with family-court custody and residence orders.
  • If the facts are “reasonably consistent” with the orders, the orders will “generally” be determinative in practice.
  • If the living arrangements are “clearly inconsistent” with the orders—suggesting an express or implied parental agreement different from the order—then the order should not be treated as representative of domicile.
  • Courts must then weigh all facts under Workman v Detroit Auto Inter-Ins Exch and Dairyland Ins Co v Auto-Owners Ins Co, plus joint-custody realities: where the child actually spends the majority of time and sleeps most nights.
  • No single factor is dispositive; the list is nonexhaustive.

E. The “resident” step-down issue: doctrinal caution

The Court did not decide whether plaintiff was a “resident” under the Esurance policy. It vacated the lower courts’ “resident” holdings because they largely treated residence as automatically following domicile (an approach tied to Grange). On remand, the trial court must reconsider “resident” under the policy, now unmoored from the overruled domicile-by-custody-order rule.

3.3 Impact

A. No-fault insurer priority becomes fact-driven again

The immediate effect is to restore a familiar, evidence-based domicile inquiry for minors in split-household situations. Priority disputes under MCL 500.3114(1) will depend less on the formal custody text and more on the child’s lived reality—reducing “gotcha” outcomes tied to the accident date and a custody-exchange calendar.

B. Reduced pressure on family courts to draft “insurance-proof” custody orders

Justice THOMAS’s concurrence highlights an institutional concern: family courts are not tasked with domicile determinations and do not draft orders with future no-fault implications in mind. The majority’s rule mitigates the risk that a boilerplate or outdated custody judgment later controls an insurance-coverage question the parties never anticipated.

C. Greater predictability for families—though more litigation fact development

The Court viewed the tradeoff as worthwhile: although fact-finding may increase in some cases, the approach is consistent with how domicile is determined elsewhere and avoids the public-interest harm of uncertain or shifting coverage.

D. Implications for step-down and “resident” policy litigation

Because the Court refused to equate “resident” automatically with no-fault “domicile,” insurers and claimants should expect closer attention to policy language, underwriting disclosures (e.g., household-member listings), and factual indicia of household membership when step-down clauses are litigated.

4. Complex Concepts Simplified

Domicile vs. residence (in ordinary language)

Domicile is your single, principal home—the place you ultimately return to and intend as your fixed base. You can have only one domicile at a time. Residence can be more flexible and sometimes depends on context (including contract language). This case reinforces that “resident” in an insurance policy is not automatically resolved by the no-fault domicile analysis.

Why custody orders are relevant but not conclusive

Custody orders say where a child is supposed to live (often for best-interest reasons). Domicile for no-fault asks where the child’s principal home actually is for allocating PIP responsibility under MCL 500.3114(1). Orders can become outdated or informally modified; the Court’s rule accounts for that reality.

“Established custodial environment” is not domicile

Under the Child Custody Act, a child may have an established custodial environment in more than one home (e.g., with both parents). Domicile, by contrast, is singular and tied to a principal home for legal purposes like insurance priority.

What “step-down” means

A “step-down” provision reduces liability limits in certain circumstances—here, if the injured person is a “family member” who is a resident of the insured’s household, coverage drops from $250,000 to the statutory minimum (described in the opinion as $20,000 at the time). Whether the step-down applies depends on the policy’s definition and the facts of household membership.

What the Court means by overruling and “stare decisis”

Stare decisis is the policy of following precedent to promote stability. But when a precedent is badly reasoned, unworkable, harmful to public interests, and not deeply relied upon, the Court may overrule it. Here, the Court used that framework to overrule Grange’s custody-order-as-domicile rule.

5. Conclusion

Frownfelter v Esurance Property & Casualty Insurance Company fundamentally reshapes Michigan no-fault domicile analysis for minor children of divorced parents. The Supreme Court overruled Grange Ins Co of Mich v Lawrence to hold that custody orders are not dispositive of domicile under MCL 500.3114(1). Instead, custody orders are a starting point; courts must determine domicile by weighing real-world living arrangements under Workman v Detroit Auto Inter-Ins Exch and Dairyland Ins Co v Auto-Owners Ins Co, including where the child actually spends most time and sleeps most nights.

By vacating the lower courts’ domicile and step-down “resident” rulings and remanding for fact-sensitive application, the Court prioritizes substantive accuracy and no-fault’s remedial aims over formalistic reliance on family-court paperwork—while also cautioning against collapsing policy “resident” questions into no-fault domicile determinations.