Michigan Supreme Court Lansing, Michigan Syllabus Chief Justice: Justices: Bridget M. McCormack Stephen J. Markman Brian K. Zahra Chief Justice Pro Tem: Richard H. Bernstein David F. Viviano Elizabeth T. Clement Megan K. Cavanagh This syllabus constitutes no part of the opinion of the Court but has been Reporter of Decisions: prepared by the Reporter of Decisions for the convenience of the reader. Kathryn L. Loomis DYE v ESURANCE PROPERTY & CASUALTY INSURANCE COMPANY Docket No. 155784. Argued October 9, 2018 (Calendar No. 1). Decided July 11, 2019. Matthew Dye brought an action in the Washtenaw Circuit Court against Esurance Property and Casualty Insurance Company and GEICO Indemnity Company, seeking personal protection insurance (PIP) benefits under the no-fault act, MCL 500.3101 et seq., for injuries he sustained in a motor vehicle accident while driving a vehicle he had recently purchased. At plaintiff’s request, plaintiff’s father had registered the vehicle in plaintiff’s name at the Secretary of State’s office and obtained a no-fault insurance policy from Esurance. The declarations page of the policy identified only plaintiff’s father as the named insured. At the time of the accident, plaintiff was living with his wife, who owned a vehicle that was insured by GEICO. After Esurance and GEICO refused to cover plaintiff’s claim, plaintiff filed a breach-of-contract claim against both insurers along with a declaratory action, alleging that either Esurance or GEICO was obligated to pay his no-fault PIP benefits and requesting that the trial court determine the parties’ respective rights and duties. Eventually, Esurance paid plaintiff more than $388,000 in PIP benefits, but it continued to maintain that GEICO was the responsible insurer. GEICO acknowledged that it was the primary insurer and began settlement negotiations with plaintiff and Esurance. Then, on November 13, 2014, the Court of Appeals’ opinion in Barnes v Farmers Ins Exch, 308 Mich App 1 (2014), was published. Barnes held that under MCL 500.3113(b), when none of the owners of a vehicle maintains the requisite coverage, no owner may recover PIP benefits. After Barnes was published, GEICO reevaluated its legal position and ceased settlement discussions. Esurance filed a cross-claim against GEICO, arguing that GEICO had breached a settlement agreement. GEICO moved for summary disposition of plaintiff’s claim, arguing that plaintiff was not entitled to PIP benefits in light of Barnes. Plaintiff also moved for summary disposition, arguing that Barnes was wrongly decided and, regardless, that his father was an owner and registrant for purposes of the no-fault act. The trial court, Timothy C. Connors, J., granted Esurance summary disposition on its cross-claim, ruling that GEICO and Esurance had entered into a valid settlement agreement and that GEICO had priority over plaintiff’s claim. The court denied GEICO’s motion for summary disposition and granted plaintiff’s motion for summary disposition, thus determining that GEICO was required to provide no-fault benefits to plaintiff. The court granted plaintiff’s motion against GEICO with regard to no-fault coverage and priority, stating that it did not need to address Barnes because plaintiff’s father was an owner and registrant of the vehicle and ruling that the only issue remaining between plaintiff and GEICO was the amount of damages. GEICO filed an interlocutory application for leave to appeal, which the Court of Appeals granted. In an unpublished per curiam opinion issued April 4, 2017 (Docket No. 330308), the Court of Appeals, BECKERING, P.J., and BORRELLO, J. (O’CONNELL, J., concurring in part and dissenting in part), held that the trial court had erred by granting summary disposition to Esurance because the parties had not yet reached a meeting of the minds on all the essential terms of the settlement agreement. The majority agreed with Barnes’s interpretation of MCL 500.3101(1), and it held that the trial court had erred as a matter of law by finding that plaintiff’s father was a “registrant” of the vehicle for purposes of MCL 500.3101(1). However, the majority held that there remained genuine issues of material fact as to whether plaintiff’s father was an “owner” of the vehicle, and it therefore remanded the case to the trial court for further proceedings. Esurance applied for leave to appeal in the Supreme Court, and both plaintiff and GEICO filed cross- appeals. The Supreme Court granted plaintiff’s cross-application for leave to appeal and denied Esurance’s application and GEICO’s cross-application. 501 Mich 944 (2017).1 In an opinion by Justice ZAHRA, joined by Chief Justice MCCORMACK and Justices MARKMAN, VIVIANO, and BERNSTEIN, the Supreme Court held: An owner or registrant of a motor vehicle is not required to personally purchase no-fault insurance for his or her vehicle in order to avoid the statutory bar to PIP benefits. MCL 500.3101(1) does not prescribe any particular manner by which no-fault insurance must be maintained, and it contains no requirement that the insurance be purchased or obtained by a vehicle’s owner or registrant. Barnes and other cases suggesting to the contrary were overruled to the extent that they were inconsistent with this holding. The Court of Appeals judgment was reversed in part. The part of the Court of Appeals judgment regarding the purported settlement agreement between Esurance and GEICO was left undisturbed. 1. Under the no-fault act, an insurer is liable to pay PIP benefits to any Michigan resident for accidental bodily injury arising out of the ownership, operation, maintenance, or use of a motor vehicle as a motor vehicle. These statutory benefits arise regardless of whether an injured person has obtained a no-fault insurance policy. Therefore, determining whether no-fault benefits are available to an injured person does not depend on who purchased, obtained, or otherwise procured no-fault insurance. The only relevant inquiry is whether the injured person can establish an accidental bodily injury arising out of the ownership, operation, maintenance or use of a motor vehicle as a motor vehicle. This relatively low threshold for statutory no-fault coverage was enacted to provide victims of motor vehicle accidents assured, adequate, and prompt reparation for certain economic losses. Given that these benefits would be available to all Michigan residents, the Legislature sought to achieve this purpose by enacting the system of compulsory insurance set forth in MCL 500.3101(1). To ensure compliance with MCL 500.3101(1), the Legislature excluded persons from receiving PIP benefits under various circumstances listed in MCL 500.3113, including if the person was the owner or registrant of a motor vehicle or motorcycle involved in the accident with respect to which the security required by MCL 500.3101 was not in effect. 1 The Legislature recently made substantial amendments to the no-fault act. See 2019 PAs 21 and 22. The opinions do not address those amendments, and all references in the opinions and this syllabus are to the preamendment version of the act. 2. MCL 500.3101(1) provides that “[t]he owner or registrant of a motor vehicle required to be registered in this state shall maintain security for payment of benefits under personal protection insurance, property protection insurance, and residual liability insurance.” MCL 500.3101(1) requires only that the owner or registrant “maintain” no-fault insurance, which means to keep in an existing state. MCL 500.3101(1) does not prescribe any particular manner by which a registrant or owner must keep no-fault insurance in an existing state, and MCL 500.3101(4) expressly contemplates that the security required by MCL 500.3101(1) may be provided by any other method approved by the Secretary of State as affording security equivalent to that afforded by a policy of insurance. If MCL 500.3101(1) were to be interpreted so that only a registrant or owner could obtain insurance on a vehicle, it would limit the Secretary of State’s power under MCL 500.3101(4) to allow security by any other method, and it would effectively read a requirement into MCL 500.3101(1) that the Legislature did not manifest through the words of MCL 500.3101(1) itself. When read together, MCL 500.3101(1) and MCL 500.3113(b) did not preclude plaintiff from receiving PIP benefits. Iqbal v Bristol West Ins Group, 278 Mich App 31 (2008), did not hold that at least one owner must obtain no-fault insurance; instead, it held that MCL 500.3113(b) refers to the required security or insurance under MCL 500.3101 only as it relates to the vehicle and therefore a plaintiff may be entitled to PIP benefits if the vehicle was insured, regardless of whether that plaintiff was the owner of the vehicle. Thus, the factual distinctions between Barnes and Iqbal did not place Barnes outside the ambit of Iqbal’s holding. 3. MCL 500.3113(b) provides that a person is not entitled to PIP benefits if at the time of the accident “[t]he person was the owner or registrant of a motor vehicle or motorcycle involved in the accident with respect to which the security required by section 3101 or 3103 was not in effect.” The term “which” in this context represents one of two possible antecedents: the first is a person, and the last is a vehicle. The last-antecedent rule provides that a modifying or restrictive word or clause contained in a statute is confined solely to the immediately preceding clause or last antecedent unless something in the statute requires a different interpretation, and nothing in MCL 500.3113(b) requires a different result. Moreover, the usage notes for the definition of “which” state that the term regularly refers to inanimate objects and never to individual persons. Accordingly, the phrase “with respect to which the security required by section 3101 or 3103 was not in effect” refers to the vehicle, not the person. Because the conclusion in Barnes that “when none of the owners maintains the requisite coverage, no owner may recover PIP benefits” was contrary to the plain language of the no-fault act, Barnes was overruled. Reversed in part and remanded to the Washtenaw Circuit Court for further proceedings. Justice CLEMENT, dissenting, would have held that the no-fault act disqualifies an owner from PIP benefits if the owner is injured in his or her own vehicle and no owner has maintained security. While she agreed with the majority that MCL 500.3113(b) contemplates a relationship between “security” and “vehicle,” she disagreed with the majority’s conclusion that security “with respect to” a vehicle is the same as insurance “for” or “on” that vehicle, given that nothing in the no-fault act requires a vehicle to be insured. Rather, she concluded that MCL 500.3113(b) requires a certain person—namely, the vehicle’s owner or registrant—to maintain security against liability as provided in MCL 500.3101, and the phrase “with respect to” connects the security to the vehicle by way of the person. Justice CLEMENT stated that the critical nexus among owner, security, and vehicle was of a whole with the rest of the no-fault act, including the priority schemes set out in MCL 500.3114 and MCL 500.3115, and that the purpose of the no- fault act would fall apart if an owner named in no policy was nonetheless understood to have maintained security through a nonowner third party’s policy. Accordingly, she concluded that plaintiff’s statutory duty to maintain security was not met merely by asking his father to get no- fault insurance. Because the record did not reveal whether plaintiff’s father was an owner of plaintiff’s car, she would have remanded the case to the trial court to determine whether plaintiff was disqualified from PIP benefits. Justice CAVANAGH did not participate in the disposition of this case because the Court considered it before she assumed office. ©2019 State of Michigan Michigan Supreme Court Lansing, Michigan OPINION Chief Justice: Justices: Bridget M. McCormack Stephen J. Markman Brian K. Zahra Chief Justice Pro Tem: Richard H. Bernstein David F. Viviano Elizabeth T. Clement Megan K. Cavanagh FILED July 11, 2019 STATE OF MICHIGAN SUPREME COURT MATTHEW DYE, by his Guardian, SIPORIN & ASSOCIATES, INC., Plaintiff-Appellee/Cross- Appellant, v No. 155784 ESURANCE PROPERTY & CASUALTY INSURANCE COMPANY, Defendant/Cross- Plaintiff/Appellant/Cross- Appellee, and GEICO INDEMNITY COMPANY, Defendant/Cross- Defendant/Appellee/Cross- Appellant, and PRIORITY HEALTH, Defendant/Cross- Plaintiff/Appellee, and BLUE CROSS BLUE SHIELD OF MICHIGAN, Defendant-Appellee. BEFORE THE ENTIRE BENCH (except CAVANAGH, J.) ZAHRA, J. This case presents the significant question of whether an owner or registrant of a motor vehicle involved in an accident is excluded from receiving statutory no-fault insurance benefits under the no-fault act, MCL 500.3101 et seq., when someone other than an owner or registrant purchased no-fault insurance for that vehicle.1 Relying on Barnes v Farmers Ins Exch,2 the Court of Appeals concluded that “[a]t least one owner or registrant must have the insurance required by MCL 500.3101(1), and ‘when none of the owners maintains the requisite coverage, no owner may recover [personal injury protection (PIP)] 1 The Legislature recently made substantial amendments to the no-fault act. 2019 PAs 21 and 22. This opinion does not address those amendments and quotes the preamendment version of the act. 2 Barnes v Farmers Ins Exch, 308 Mich App 1; 862 NW2d 681 (2014). 2 benefits.’ ”3 The insured sought leave to appeal in this Court, and we granted the application in part to consider this question.4 We conclude that an owner or registrant of a motor vehicle is not required to personally purchase no-fault insurance for his or her vehicle in order to avoid the statutory bar to PIP benefits. Rather, MCL 500.3101(1) only requires that the owner or registrant “maintain” no-fault insurance, and the term “maintain,” as commonly understood, means to keep in an existing state. Because MCL 500.3101(1) does not prescribe any particular manner by which no-fault insurance must be maintained, we will not read into the statute a requirement that the insurance be purchased or obtained by a vehicle’s owner or registrant. Further, the grammatical composition of the MCL 500.3113(b) benefits exclusion, including the use of the term “which” within that provision, signifies that the exclusion does not apply if the security required by MCL 500.3101(1) was “in effect” at the time of the accident.5 Though defendant maintains that the term “which” in this provision refers to the owner or operator of the motor vehicle, the usage of “which” at the time the no-fault act was enacted, as well as currently, reflects that this term is not properly used to refer to individual persons. 3 Dye v Esurance Prop & Cas Ins Co, unpublished per curiam opinion of the Court of Appeals, issued April 4, 2017 (Docket No. 330308), p 6, quoting Barnes, 308 Mich App at 8-9. 4 Dye v Esurance Prop & Cas Ins Co, 501 Mich 944 (2017). 5 MCL 500.3113(b) states that “[a] person is not entitled to be paid personal protection insurance benefits for accidental bodily injury if at the time of the accident . . . [t]he person was the owner or registrant of a motor vehicle . . . involved in the accident with respect to which the security required by section 3101 or 3103 was not in effect.” 3 We therefore hold that an owner or registrant of a motor vehicle involved in an accident is not excluded from receiving no-fault benefits when someone other than that owner or registrant purchased no-fault insurance for that vehicle because the owner or registrant of the vehicle may “maintain” the insurance coverage required under the no-fault act even if he or she did not purchase the insurance. The Court of Appeals’ decision in Barnes and other caselaw suggesting to the contrary are overruled to the extent that they are inconsistent with our holding.6 We reverse in part the judgment of the Court of Appeals and remand this case to the Washtenaw Circuit Court for further proceedings not inconsistent with this opinion. 6 This Court stated in Citizens Ins Co of America v Federated Mut Ins Co, 448 Mich 225, 228; 531 NW2d 138 (1995), that “Michigan’s no-fault act requires the owner or registrant of a motor vehicle to purchase an automobile insurance policy . . . .” The Court apparently used the word “purchase” because owners commonly are the persons who maintain insurance on their vehicles, but in so using the word “purchase” the Court strayed from the actual text of the no-fault act. The use of the word “purchase” instead of the word “maintain” was inconsequential to the analysis in Citizens, because the word “maintain” had no particular significance to the issue addressed in that case. The question decided in Citizens was whether an insurance policy could shift residual liability insurance from the owner or registrant of a motor vehicle to the driver of the vehicle. Citizens held that the policy was essentially unenforceable in this respect because “the no-fault act unambiguously requires that a policy of automobile insurance, sold to a vehicle owner pursuant to the act, must provide coverage for residual liability arising from use of the vehicle so insured.” Id. at 230. Unlike this case, there was no dispute in Citizens about whether insurance was properly maintained on the vehicles at issue. Accordingly, we clarify that Citizens should not be read to suggest that only the owner or registrant of a motor vehicle may “maintain” an automobile insurance policy. 4 I. FACTS AND PROCEDURAL HISTORY On September 26, 2013, plaintiff Matthew Dye was involved in a motor vehicle accident and suffered serious injuries that included a traumatic brain injury. At that time, plaintiff was 32 years old, fully employed, a member of the National Guard who had spent time in Afghanistan, and recently married. At some point before the accident, plaintiff had granted his father power of attorney “to do bussiness [sic] at the secretary of state on [plaintiff’s] behalf.” At the time of the accident, plaintiff was driving a 1997 BMW that he had purchased two months earlier. After the purchase, plaintiff asked his father to register the vehicle for him and to obtain no-fault insurance. His father registered the vehicle in plaintiff’s name at the secretary of state’s office and obtained a no-fault insurance policy from Esurance via the Internet. The declarations page of the policy identified only plaintiff’s father as the named insured. At the time of the accident, plaintiff was living with his wife, who owned a Dodge Caravan that was insured by GEICO. After Esurance and GEICO refused to cover plaintiff’s claim, plaintiff filed a breach of contract claim against both insurers along with a declaratory action, alleging that either Esurance or GEICO was obligated to pay his no-fault PIP benefits and requesting that the trial court determine the parties’ respective rights and duties. A priority dispute between Esurance and GEICO ensued.7 Eventually, Esurance paid plaintiff more than $388,000 in 7 A personal protection insurance policy applies to accidental bodily injury that occurs “to the person named in the policy, the person’s spouse, and a relative of either domiciled in the same household, if the injury arises from a motor vehicle accident.” MCL 500.3114(1). In this case, Esurance maintained that because plaintiff was not a named insured under the no-fault policy on his own car and because plaintiff was living with his wife—not his father—at the time of the accident, plaintiff should look to GEICO rather than Esurance for the payment of PIP benefits. 5 PIP benefits, but Esurance continued to maintain that GEICO was the responsible insurer. GEICO acknowledged that it was the primary insurer and began settlement negotiations with plaintiff and Esurance. Then, on November 13, 2014, the Court of Appeals’ opinion in Barnes v Farmers Ins Exch was redesignated “for publication.”8 In Barnes, the panel held that “under the plain language of MCL 500.3113(b), when none of the owners maintains the requisite coverage, no owner may recover PIP benefits.”9 After Barnes was published, GEICO reevaluated its legal position and ceased settlement discussions. Essentially, what had been—before Barnes—merely a priority dispute among potential insurers over who had to pay plaintiff’s claim turned into a dispute about whether plaintiff was covered under the no-fault act. Esurance filed a cross-claim against GEICO arguing that GEICO had breached a settlement agreement. GEICO moved for summary disposition of plaintiff’s claim, arguing that plaintiff was not entitled to PIP benefits in light of the now-published decision in Barnes because plaintiff owned the subject vehicle but had not insured it and the person who had insured it (plaintiff’s father) was not an “owner” as defined in MCL 500.3101. Plaintiff, in turn, moved for summary disposition, arguing that Barnes was wrongly decided and, regardless, that his father was an owner and registrant for purposes of the no- 8 Barnes was originally released as an unpublished opinion having no precedential value in the Court of Appeals or lower courts. But pursuant to MCR 7.215(D)(3), the panel granted the defendant’s request to publish the opinion, providing Barnes precedential effect under the rule of stare decisis. MCR 7.215(C)(2). 9 Barnes, 308 Mich App at 8-9. 6 fault act because he had the right to use the BMW and because he had physically registered the vehicle. The trial court granted Esurance summary disposition on its cross-claim, ruling that GEICO and Esurance had entered into a valid settlement agreement and that GEICO had priority over plaintiff’s claim. The trial court denied GEICO’s motion for summary disposition and granted plaintiff’s motion for summary disposition, thus determining that GEICO was required to provide no-fault benefits to plaintiff. The trial court stated that it did not need to address the decision in Barnes because plaintiff’s father was an owner and registrant of the BMW. According to the trial court, the only issue remaining between plaintiff and GEICO was the amount of damages. The trial court thus entered an order granting Esurance summary disposition on its cross-claim, denying GEICO’s motion for summary disposition with regard to plaintiff, and granting plaintiff’s motion against GEICO with regard to no-fault coverage and priority. GEICO filed an interlocutory application for leave to appeal, which the Court of Appeals granted.10 In an unpublished per curiam opinion, the Court of Appeals reversed the trial court’s decision that granted summary disposition to Esurance.11 The panel held that the trial court erred by enforcing the settlement agreement. The panel explained that “[a]lthough the issuance of Barnes promptly snuffed out what appears to have been a ‘nearly done’ deal, the parties had not yet reached a meeting of the minds on all of the 10 Dye v Esurance Prop & Cas Ins Co, unpublished order of the Court of Appeals, entered April 5, 2016 (Docket No. 330308). 11 Dye, unpub op at 1. 7 essential terms, and the trial court erred in granting Esurance’s motion for summary disposition on its cross-claim for enforcement of the alleged agreement.”12 In regard to the trial court’s decision denying GEICO summary disposition against plaintiff on the basis of Barnes, the panel embraced Barnes’s interpretation of MCL 500.3101(1) without reservation, stating: Although a motor vehicle may have more than one owner for purposes of the no fault act, it is not sufficient that a vehicle is insured by just anyone. At least one owner or registrant must have the insurance required by MCL 500.3101(1), and “when none of the owners maintains the requisite coverage, no owner may recover PIP benefits.”[13] The Court of Appeals then held that the trial court erred as a matter of law by finding that plaintiff’s father was a “registrant” for purposes of MCL 500.3101(1).14 A majority of the panel, however, agreed with the trial court that there remained genuine issues of material fact as to whether plaintiff’s father was an “owner” of the BMW plaintiff was driving at the time of the accident.15 Therefore, the Court of Appeals remanded this case to the trial court for further proceedings.16 12 Id. at 11. 13 Id. at 6, quoting Barnes, 308 Mich App at 8-9 (citations omitted). 14 Dye, unpub op at 6. 15 Id. at 9. Judge O’CONNELL dissented from the majority’s conclusion that there were questions of fact about whether plaintiff’s father was an “owner” of the vehicle but agreed with the panel in all other respects. Id. (O’CONNELL, J., concurring in part and dissenting in part) at 1. 16 Id. (opinion of the Court) at 12. 8 Esurance filed an application in this Court arguing that the Court of Appeals improperly reversed the trial court’s decision to enforce the purported settlement agreement by granting Esurance summary disposition on its cross-claim against GEICO. Plaintiff filed a cross-appeal arguing that Barnes was improperly decided and that Esurance could not deny liability simply because plaintiff himself did not obtain no-fault insurance. GEICO filed a cross-appeal arguing that the Court of Appeals erred by ruling that a question of fact precluded summary disposition in favor of GEICO on the issue of whether plaintiff’s father was a co-owner of the BMW that plaintiff, in GEICO’s view, failed to insure. This Court granted plaintiff’s cross-application for leave to appeal and denied Esurance’s application and GEICO’s cross-application.17 Accordingly, the sole issue before the Court is “whether an owner or registrant of a motor vehicle involved in an accident may be entitled to personal protection insurance benefits for accidental bodily injury where no owner or registrant of the motor vehicle maintains security for payment of benefits under personal protection insurance.”18 II. STANDARD OF REVIEW AND APPLICABLE RULES OF STATUTORY INTERPRETATION This Court reviews de novo a trial court’s decision on a motion for summary disposition.19 The parties brought their respective summary disposition motions under 17 Dye, 501 Mich 944. 18 Id. 19 Maiden v Rozwood, 461 Mich 109, 118; 597 NW2d 817 (1999). 9 MCR 2.116(C)(10), which tests the factual sufficiency of a claim.20 “In reviewing a motion for summary disposition brought under MCR 2.116(C)(10), a trial court considers affidavits, pleadings, depositions, admissions, and documentary evidence filed in the action or submitted by the parties, MCR 2.116(G)(5), in the light most favorable to the party opposing the motion.”21 If, “[e]xcept as to the amount of damages, there is no genuine issue as to any material fact, . . . the moving party is entitled to judgment or partial judgment as a matter of law,”22 and the trial court must grant the motion without delay.23 This Court also reviews de novo questions of statutory interpretation.24 “The role of this Court in interpreting statutory language is to ‘ascertain the legislative intent that may reasonably be inferred from the words in a statute.’ ”25 “The focus of our analysis must be the statute’s express language, which offers the most reliable evidence of the Legislature’s intent.”26 “ ‘[W]here the statutory language is clear and unambiguous, the statute must be applied as written.’ ”27 “ ‘[A] court may read nothing into an unambiguous 20 Smith v Globe Life Ins Co, 460 Mich 446, 454; 597 NW2d 28 (1999). 21 Quinto v Cross & Peters Co, 451 Mich 358, 362; 547 NW2d 314 (1996). 22 MCR 2.116(C)(10). 23 MCR 2.116(I)(1). 24 Badeen v PAR, Inc, 496 Mich 75, 81; 853 NW2d 303 (2014). 25 Hannay v Dep’t of Transp, 497 Mich 45, 57; 860 NW2d 67 (2014) (citation omitted). 26 Badeen, 496 Mich at 81. 27 McQueer v Perfect Fence Co, 502 Mich 276, 286; 917 NW2d 584 (2018) (citation omitted). 10 statute that is not within the manifest intent of the Legislature as derived from the words of the statute itself.’ ”28 Neither will this Court “rewrite the plain statutory language and substitute our own policy decisions for those already made by the Legislature.”29 III. ANALYSIS A. LEGAL BACKGROUND Under the no-fault act, “an insurer is liable to pay [PIP] benefits [to any Michigan resident] for accidental bodily injury arising out of the ownership, operation, maintenance or use of a motor vehicle as a motor vehicle . . . .”30 Although designated as “personal protection insurance” under the no-fault act, PIP benefits are in fact statutory benefits, arising regardless of whether an injured person has obtained a no-fault insurance policy. Indeed, a no-fault insurance carrier can be liable for no-fault benefits even if the motor vehicle it insures was not the actual motor vehicle involved in the accident.31 PIP benefits are paid to injured persons solely by insurers who are authorized to write no-fault insurance policies in this state or who have voluntarily filed a certificate complying with MCL 500.3163.32 28 Id., quoting Roberts v Mecosta Co Gen Hosp, 466 Mich 57, 63; 642 NW2d 663 (2002) (alteration in original). 29 DiBenedetto v West Shore Hosp, 461 Mich 394, 405; 605 NW2d 300 (2000). 30 MCL 500.3105(1). 31 See Detroit Auto Inter-Ins Exch v Home Ins Co, 428 Mich 43, 49; 405 NW2d 85 (1987). Amicus Michigan Defense Trial Counsel agrees, acknowledging that “[e]ligibility to claim no-fault benefits for motor-vehicle-related injuries that occur in the state is not contingent upon the injured person having his or her own no-fault insurance policy.” 32 The grand scope of these insurers’ obligations became evident soon after the no-fault act 11 For these reasons, determining whether no-fault benefits are available to an injured person does not depend on “who” purchased, obtained, or otherwise procured no-fault insurance. The only relevant inquiry is whether the injured person can establish an “accidental bodily injury arising out of the ownership, operation, maintenance or use of a motor vehicle as a motor vehicle . . . .”33 By establishing this relatively low threshold for statutory no-fault coverage, the no-fault act seeks to “provide victims of motor vehicle accidents assured, adequate, and prompt reparation for certain economic losses.”34 And given that these statutory benefits would be available to all Michigan residents, “[t]he Legislature believed this . . . could be most effectively achieved through a system of compulsory insurance, whereby every Michigan motorist would be required to purchase no-fault insurance or be unable to operate a motor vehicle legally in this state.”35 Accordingly, the Legislature enacted MCL 500.3101(1), which provides, in part, that “[t]he owner or registrant of a motor vehicle required to be registered in this state shall maintain security for payment of benefits under personal protection insurance . . . .” And to ensure compliance with MCL 500.3101(1), the Legislature excluded persons from receiving PIP benefits under various circumstances listed in MCL 500.3113. In this case, GEICO was enacted. In response to concerns that “Michigan’s no-fault law provision for unlimited personal injury protection benefits placed too great a burden on insurers, particularly small insurers, in the event of ‘catastrophic’ injury claims,” In re Certified Question, 433 Mich 710, 714; 449 NW2d 660 (1989), the Legislature created the Michigan Catastrophic Claims Association through Public Act 136 of 1978, codified at MCL 500.3104. 33 MCL 500.3105(1). 34 Shavers v Attorney General, 402 Mich 554, 579; 267 NW2d 72 (1978). 35 Id. at 579 (emphasis omitted). 12 maintains that MCL 500.3113(b) precludes plaintiff from obtaining PIP benefits because, in its view, plaintiff “was the owner or registrant of a motor vehicle or motorcycle involved in the accident with respect to which the security required by section 3101 or 3103 was not in effect.” In GEICO’s view, “the security required by section 3101 . . . was not in effect” because, as emphasized by GEICO, MCL 500.3101(1) mandates that an “owner or registrant of a motor vehicle required to be registered in this state shall maintain security for payment of benefits under personal protection insurance . . . .” (Emphasis added.) Against this backdrop, the legal issue before the Court, as aptly stated by amicus Property Casualty Insurers Association of America, hinges on whether the phrase “the owner or registrant of a motor vehicle required to be registered in this state shall maintain security,” means that the owner (or at least an owner) must be the one to acquire the insurance policy, or whether it suffices for any person to provide the required security such that all that matters is that the vehicle is insured.[36] GEICO maintains that the former interpretation is correct, and accordingly argues that plaintiff is excluded from receiving PIP benefits because he owned the BMW and was not the one to obtain the no-fault insurance policy. Plaintiff maintains that the latter interpretation is correct, and he argues that he is not excluded from coverage because he owned the BMW and the BMW was insured. While this Court has not yet addressed this issue, the Court of Appeals has considered very similar arguments in two published opinions, which we now review. 36 Some emphasis omitted. 13 B. COURT OF APPEALS CASELAW In Iqbal v Bristol West Ins Group,37 the Court of Appeals first addressed in a published decision whether every owner of a vehicle is “required to maintain insurance on the vehicle under the no-fault act . . . .” In that case, the plaintiff did not have title to any vehicle, but he frequently used his brother’s BMW.38 The plaintiff was injured while driving the BMW and requested no-fault benefits. In his answers to interrogatories, the plaintiff indicated that the BMW “ ‘belonged to my brother but I had primary possession.’ ”39 The insurer claimed that the plaintiff should also be considered an owner of the car and that as an owner, the plaintiff must have obtained the insurance policy to obtain PIP benefits. The panel rejected this argument, stating: Viewing the statutory language in the context of the given facts, the statute would preclude plaintiff from being entitled to PIP benefits if plaintiff “was the owner . . . of [the BMW] . . . involved in the accident with respect to which the security required by section 3101 . . . was not in effect.” As part of the process of construing MCL 500.3113(b), we shall make the assumption that plaintiff was an “owner” of the BMW, as that term is defined in MCL 500.3101(2)(g)(i). Next, the phrase “with respect to which the security required by section 3101 . . . was not in effect,” § 3113(b), when read in proper grammatical context, defines or modifies the preceding reference to the motor vehicle involved in the accident, here the BMW, and not the person standing in the shoes of an owner or registrant. The statutory language links the required security or insurance solely to the vehicle. Thus, the question becomes whether the BMW, and not plaintiff, had the coverage or security required by MCL 500.3101. As indicated above, the coverage mandated by MCL 500.3101(1) consists of “personal protection insurance, property protection insurance, and residual liability insurance.” While 37 Iqbal v Bristol West Ins Group, 278 Mich App 31, 33; 748 NW2d 574 (2008). 38 Id. at 34. 39 Id. 14 plaintiff did not obtain this coverage, there is no dispute that the BMW had the coverage, and that is the only requirement under MCL 500.3113(b), making it irrelevant whether it was plaintiff’s brother who procured the vehicle’s coverage or plaintiff. Stated differently, the security required by MCL 500.3101(1) was in effect for purposes of MCL 500.3113(b) as it related to the BMW.[40] In sum, Iqbal held that the only requirement under MCL 500.3113(b) was that there be no- fault insurance on the vehicle—who purchased the policy was irrelevant. Plaintiff’s position is in accordance with Iqbal. In Barnes v Farmers Ins Exch,41 the Court of Appeals was presented with a very similar question. In that case, the plaintiff and her mother were the titled owners of a vehicle.42 The plaintiff’s mother gave a friend from church, Richard Huling, money to obtain insurance for the vehicle.43 Huling purchased a policy.44 The plaintiff was later injured in an accident, and requested PIP benefits under Huling’s policy.45 The insurer denied the request, and the plaintiff brought suit.46 The insurer moved for summary disposition, contending that the “plaintiff could not recover PIP benefits from it under the 40 Id. at 39-40 (alteration in original). 41 Barnes, 308 Mich App 1. 42 Id. at 2-3. 43 Id. at 3. 44 Id. 45 Id. 46 Id. 15 policy because the policy only covered the named insured, Huling, and was never intended to benefit plaintiff.”47 In Barnes, after hearing arguments, “the trial court ruled that the no-fault act required at least one of the ‘owners’ to have insurance. It reasoned that because neither plaintiff nor [her mother] had insurance, plaintiff was barred from seeking benefits under the no-fault act.”48 The trial court granted summary disposition to the insurer.49 The plaintiff in Barnes appealed, arguing that Iqbal required the opposite result. The Court of Appeals stated: In the present case, plaintiff cites Iqbal and argues that the fact that neither she nor [her mother] insured the Cavalier does not matter because Huling did. Plaintiff contends that this is so regardless of whether Huling was an owner of the Cavalier. Iqbal should not be read so broadly as to apply to even nonowners. The Court made it clear that it was addressing the problem of whether the statute required “each and every owner” to maintain insurance on a vehicle. The Court opined that to so hold would preclude an owner who obtained insurance from receiving PIP benefits as long as any other co-owner did not maintain coverage as well.[50] Thus, Barnes distinguished Iqbal, stating that “while Iqbal held that each and every owner need not obtain insurance, it did not allow for owners to avoid the consequences of MCL 500.3113(b) if no owner obtained the required insurance.”51 In sum, Barnes held that only a registrant or owner may procure no-fault insurance for a vehicle. Barnes is 47 Id. at 3-4. 48 Id. at 5. 49 Id. 50 Id. at 7-8 (citations omitted). 51 Id. at 8. 16 consistent with GEICO’s position, and plaintiff argues that Barnes improperly distinguished Iqbal. C. APPLICATION While the Court of Appeals in Iqbal and Barnes focused primarily on the language of MCL 500.3113(b), our analysis primarily concerns the language of MCL 500.3101(1). After examining this provision, we conclude that the Legislature carefully chose its words when it prescribed that “[t]he owner or registrant of a motor vehicle required to be registered in this state shall maintain security for payment of benefits under personal protection insurance, property protection insurance, and residual liability insurance.”52 The first and most relevant definition of “maintain” is: “to keep in an existing state (as of repair, efficiency, or validity) : preserve from failure or decline[.]”53 MCL 500.3101(1) only 52 MCL 500.3101(1) (emphasis added). 53 Merriam-Webster’s Collegiate Dictionary (11th ed). See also Oxford University Press, English Oxford Living Dictionaries, (accessed November 30, 2018) [https://perma.cc/T3A8-8EDR], which defines “maintain,” in part, as follows: 1 Cause or enable (a condition or situation) to continue. ‘the need to maintain close links between industry and schools’ * * * 1.1 Keep (something) at the same level or rate. ‘agricultural prices will have to be maintained’ * * * 1.2 Keep (a building, machine, or road) in good condition by checking or repairing it regularly. 17 requires that the owner or registrant “maintain” no-fault insurance, which, as commonly understood, simply means to keep in an existing state. Further, MCL 500.3101(1) does not prescribe any particular manner by which a registrant or owner must keep no-fault insurance in an existing state. Indeed, MCL 500.3101(4) expressly contemplates that the “[s]ecurity required by subsection (1) may be provided by any other method approved by the secretary of state as affording security equivalent to that afforded by a policy of insurance, if proof of the security is filed and continuously maintained with the secretary of state throughout the period the motor vehicle is driven or moved on a highway.” If we were to accept GEICO’s interpretation that only a registrant or owner may obtain insurance on a vehicle, we would limit the secretary of state’s power to allow security “by any other method” and we would also have effectively read a requirement into MCL 500.3101(1) that the Legislature did not manifest through the words of MCL 500.3101(1) itself.54 GEICO argues that “the common thread in all of these definitions [of maintenance] is that some affirmative act is necessary by the person required to ‘maintain’ the insurance.” We conclude that this argument lacks merit. Even if the word “maintain” were to imply an affirmative act, plaintiff here undeniably undertook an affirmative act when he instructed his father to obtain no-fault insurance, the same way any person instructs a mechanic to “maintain” his vehicle or a father instructs his son to “maintain” the lawn. ‘the Department for Transport is responsible for maintaining the main roads in England[.]’ 54 Covenant Med Ctr, Inc v State Farm Mut Auto Ins Co, 500 Mich 191, 207-208; 895 NW2d 490 (2017), citing Roberts, 466 Mich at 63. 18 Thus, we conclude that the language of MCL 500.3101(1) does not require an owner or a registrant of a motor vehicle to personally obtain no-fault insurance. We further conclude, contrary to Barnes, that when read together, MCL 500.3101(1) and MCL 500.3113(b) do not preclude plaintiff from receiving PIP benefits. Again, MCL 500.3113 provides: A person is not entitled to be paid personal protection insurance benefits for accidental bodily injury if at the time of the accident any of the following circumstances existed: * * * (b) The person was the owner or registrant of a motor vehicle or motorcycle involved in the accident with respect to which the security required by section 3101 or 3103 was not in effect. As previously discussed, the Court of Appeals in Iqbal opined that “the phrase ‘with respect to which the security required by section 3101 . . . was not in effect,’ § 3113(b), when read in proper grammatical context, defines or modifies the preceding reference to the motor vehicle involved in the accident, here the BMW, and not the person standing in the shoes of an owner or registrant. The statutory language links the required security or insurance solely to the vehicle.”55 Despite acknowledging that Iqbal also stated that “ ‘there is no dispute that the BMW had the coverage, and that is the only requirement under MCL 500.3113(b),’ ”56 the Barnes panel nonetheless concluded that “Iqbal does not protect owners of vehicles if no owner provides the insurance . . . .”57 In our view, a fair 55 Iqbal, 278 Mich App at 39-40. 56 Barnes, 308 Mich App at 7, quoting Iqbal, 278 Mich App at 40. 57 Barnes, 308 Mich App at 8. 19 reading of Iqbal does not indicate that at least one owner must obtain no-fault insurance. Indeed, Iqbal concludes that [b]ecause the language in MCL 500.3113(b) precluding recovery of PIP benefits links the security or insurance requirement to the vehicle only and not the person, the trial court correctly ruled that plaintiff was entitled to PIP benefits because the vehicle was in fact insured, regardless of whether plaintiff was the “owner” of the vehicle.[58] Thus, while Barnes may be distinguishable from Iqbal on its facts, we conclude that those factual distinctions did not place Barnes outside the ambit of Iqbal’s holding. In sum, we agree with Iqbal that MCL 500.3113(b) refers to the required security or insurance under MCL 500.3101 only as it relates to the vehicle. GEICO acknowledges that “[p]laintiff may be partially correct that [MCL 500.3113(b)] ties ‘security’ to the motor vehicle itself . . . .” Yet, GEICO contends that the reference in MCL 500.3113(b) to the “ ‘security required by section 3101—and [that section’s] use of the mandatory term ‘shall’ in reference to ‘[t]he owner or registrant’—means that the vehicle is not properly insured unless that security is maintained by an ‘owner or registrant.’ ”59 We disagree. Initially, we iterate that GEICO’s interpretation would ignore that MCL 500.3101(1) does not expressly prescribe any particular manner by which a registrant or an owner must keep no-fault insurance in an existing state.60 In regard to MCL 58 Iqbal, 278 Mich App at 46 (emphasis added). 59 Second alteration in original. 60 GEICO also argues that “the only approach that carries out the legislative intent embodied in [MCL 500.3101(1)] is to equate ‘shall maintain security’ with being a named insured on a policy of automobile no-fault insurance.” In something of a “plain folks” appeal (at least for a legal brief in an insurance case), GEICO explains that “anyone who owns a vehicle in Michigan knows that it is the named insured who applies for auto 20 500.3113(b), there is no dispute that the phrase “the security required by section 3101 or 3103 was not in effect” refers to either the “owner or registrant” or the “vehicle.” Obviously, plaintiff believes that the phrase refers to the vehicle, and the dissent and defendant believe that the phrase refers to the owner or registrant.61 In our view, the dispute insurance. That’s the person who fills out the form developed by the insurer to obtain the information needed to accurately underwrite the coverage.” The shortcoming of this argument is that statutory coverage under the no-fault act does not depend on whether a person is a “named insured” in a no-fault policy. The phrase “named insured” is not even contained in MCL 500.3101 or, for that matter, in MCL 500.3113. Rather, whether a person is a “named insured” under a no-fault insurance policy is generally only relevant in deciding which potential insurer is liable for the claim. The only notable exception is MCL 500.3111, which pertains to Michigan residents involved in out-of-state accidents. Under those circumstances, coverage is predicated on whether “the person whose injury is the basis of the claim was at the time of the accident a named insured under a personal protection insurance policy . . . .” MCL 500.3111. But this provision is not applicable in this case. 61 The dissent states that MCL 500.3113(b) contemplates a relationship between the security and the vehicle but questions the basis for our “conclusion that security ‘with respect to’ a vehicle is the same as insurance ‘for’ or ‘on’ that vehicle . . . .” Post at 5. While we agree that many provisions of the no-fault act are inartfully drafted and require interpretation, there is no statutory basis to conclude that the owner of an insured vehicle who is not a “named insured” in the policy is ineligible for PIP benefits. If this were the case, the Legislature could have readily predicated coverage on whether “the person whose injury is the basis of the claim was at the time of the accident a named insured under a personal protection insurance policy,” as in MCL 500.3111. The dissenting justice argues that her interpretation “is consistent with, and therefore supported by, the act’s other sections, like the priority schemes set out in MCL 500.3114 and MCL 500.3115.” Post at 6. She claims that “the priority scheme reflects the prenominate nexus among owner, security, and vehicle.” Post at 7. In our view, the dissent has improperly conflated coverage under the no-fault act with priority under the no-fault act. Priority provisions do not expand coverage under the no-fault act. The priority provisions do not provide additional coverage; they merely dictate which insurer will pay claims that have already been established. For this reason, the dissent’s reliance on the priority provisions is misplaced. 21 is best resolved by examining the use of the term “which” in MCL 500.3113(b). The term “which,” as applied in this context, is “used relatively in restrictive and nonrestrictive clauses to represent a specified antecedent[]: This book, which I read last night, was exciting. The socialism which Owen preached was unpalatable to many. The lawyer represented five families, of which ours was the largest.”62 Here, two possible antecedents precede the phrase “the security required by section 3101 or 3103 was not in effect”: the first possible antecedent mentioned is a person, and the last possible antecedent mentioned is a vehicle. Plaintiff relies on the last-antecedent rule, “a rule of statutory construction that provides that ‘a modifying or restrictive word or clause contained in a statute is confined solely to the immediately preceding clause or last antecedent, unless something in the statute requires a different interpretation.’ ”63 “As we have warned before, the last antecedent rule should not be applied if ‘something in the statute requires a different interpretation’ than the one that would result from applying the rule.”64 But here, nothing in the statute requires a different result. Moreover, the use of the term “which” plainly favors plaintiff’s interpretation. The usage notes for the definition of “which” state that the term “is used regularly in referring to inanimate objects and, usually, animals and never, in modern usage, to individual 62 Random House Webster’s College Dictionary (2000). 63 Hardaway v Wayne Co, 494 Mich 423, 427; 835 NW2d 336 (2013), quoting Stanton v Battle Creek, 466 Mich 611, 616; 647 NW2d 508 (2002). See also Scalia & Garner, Reading Law: The Interpretation of Legal Texts (St. Paul: Thomson/West, 2012), pp 144- 146. 64 Hardaway, 494 Mich at 428, quoting Stanton, 466 Mich at 416. 22 persons . . . .”65 Given that the term “which” by the time the no-fault system was enacted no longer referred to individual persons, we conclude that the phrase “with respect to which the security required by section 3101 or 3103 was not in effect” refers to the vehicle, not the person. Because the conclusion in Barnes that “when none of the owners maintains the requisite coverage, no owner may recover PIP benefits” is contrary to the plain language of the no-fault act, we overrule Barnes.66 65 The Random House College Dictionary (rev ed, 1975). 66 Barnes, 308 Mich App at 8-9. GEICO also raises the specter of fraud to favor its interpretation by claiming that [f]or the system to work for all members of the pool, risk must be allocated and managed as accurately as possible. Through MCL 500.3101(1), the Michigan Legislature recognized that what matters most for no-fault insurance is the identity of the vehicle owner or registrant. Otherwise, vehicle owners with high risk factors would be able to avoid premiums applicable to the risk they present by adding their vehicles to the policies of others, including friends and even roommates. And the problem is not resolved by requiring owners of other vehicles to be listed as drivers because listed drivers do not fill out applications; they do not receive the same scrutiny as an applicant. First, as plaintiff rightly points out, there is no indication of fraud in this case. Second, “[t]his Court has been clear that the policy behind a statute cannot prevail over what the text actually says. The text must prevail.” Elezovic v Ford Motor Co, 472 Mich 408, 421- 422; 697 NW2d 851 (2005). In other words, the specter of fraud does not distract us from our goal of interpreting the applicable statutory language to determine the rule of law. Third, the Legislature clearly understands how to enact laws to mitigate fraud within the no-fault act. In fact, the Legislature recently did so when it enacted 2016 PA 346, which is now codified at MCL 500.3009(2): If authorized by the insured, automobile liability or motor vehicle liability coverage may be excluded when a vehicle is operated by a named person. An exclusion under this subsection is not valid unless the following notice is on the face of the policy or the declaration page or certificate of the policy and on the certificate of insurance: 23 IV. CONCLUSION We hold that an owner or a registrant of a motor vehicle involved in an accident is not excluded from receiving no-fault benefits when someone other than that owner or registrant purchased no-fault insurance for that vehicle because the owner or registrant of the motor vehicle has nonetheless “maintained” no-fault insurance. The Court of Appeals’ decision in Barnes and caselaw suggesting to the contrary are overruled to the extent that they are inconsistent with this holding. We reverse in part67 the judgment of the Court of Appeals in this case and remand the case to the Washtenaw Circuit Court for further proceedings not inconsistent with this opinion. Brian K. Zahra Bridget M. McCormack Stephen J. Markman David F. Viviano Richard H. Bernstein Warning—when a named excluded person operates a vehicle all liability coverage is void—no one is insured. Owners of the vehicle and others legally responsible for the acts of the named excluded person remain fully personally liable. 67 Although Esurance filed an application in this Court seeking enforcement of its purported settlement agreement with GEICO, this Court was not persuaded that the questions presented should be reviewed by this Court. See Dye, 501 Mich 944. This opinion does not disturb the Court of Appeals’ decision regarding the purported settlement agreement. 24 STATE OF MICHIGAN SUPREME COURT MATTHEW DYE, by his Guardian, SIPORIN & ASSOCIATES, INC., Plaintiff-Appellee/Cross- Appellant, v No. 155784 ESURANCE PROPERTY & CASUALTY INSURANCE COMPANY, Defendant/Cross- Plaintiff/Appellant/Cross- Appellee, and GEICO INDEMNITY COMPANY, Defendant/Cross- Defendant/Appellee/Cross- Appellant, and PRIORITY HEALTH, Defendant/Cross- Plaintiff/Appellee, and BLUE CROSS BLUE SHIELD OF MICHIGAN, Defendant-Appellee. CLEMENT, J. (dissenting). Plaintiff bought a car and asked his father to buy no-fault insurance. His father bought a policy naming the father, not plaintiff, as the insured. Under the no-fault act, as the owner of the car, plaintiff had to “maintain security for payment of benefits under” insurance for bodily injury and property damage. MCL 500.3101. Although the act provides for benefits for all persons injured in motor-vehicle accidents, it disqualifies from bodily-injury benefits a person injured in his or her own vehicle “with respect to which the security required by section 3101 . . . was not in effect.” MCL 500.3113. Plaintiff was injured in his own car and sought benefits under his wife’s policy, issued by defendant GEICO Indemnity Company.1 See MCL 500.3114. Defendant denied benefits on the basis that plaintiff’s car was one “with respect to which the security required by section 3101 . . . was not in effect.” To decide the issue on appeal, we must determine whether that security was or was not in effect. Plaintiff argues that the security was in effect because he instructed his father to take out a policy listing the car and his father did so. The majority agrees with plaintiff, reading the disqualification provision as operating only when an owner’s vehicle is uninsured. But the no-fault act has very little to do with insuring vehicles as such; rather, it aims to create comprehensive insurer liability for bodily injury and property damage resulting from motor-vehicle accidents. The act’s text, context, and purpose support that understanding. The majority gives short shrift to each of these, analyzing fractions of statutory text in 1 See generally ante at 4-5. 2 isolation, and as a result announces a rule that undermines the act. For these reasons, and as explained below, I dissent.2 Defendant argues that it is not liable for plaintiff’s PIP3 claim because plaintiff’s circumstances meet the statutory disqualification provision in MCL 500.3113(b): The person [seeking benefits] was the owner or registrant of a motor vehicle . . . involved in the accident with respect to which the security required by section 3101 . . . was not in effect. All agree that plaintiff owned a vehicle “involved in the accident,” so the question here is whether that vehicle was one “with respect to which the security required by section 3101 . . . was not in effect.” To answer that question, we must understand what is meant by “the security required by section 3101.” “Section 3101” refers to MCL 500.3101, which says, in relevant part: (1) The owner or registrant of a motor vehicle required to be registered in this state shall maintain security for payment of benefits under personal protection insurance, property protection insurance, and residual liability insurance. . . . * * * (3) Security required by subsection (1) may be provided under a policy issued by an authorized insurer that affords insurance for the payment of benefits described in subsection (1). . . . 2 As the majority notes, some of the statutes at issue in this case recently have been amended. See ante at 2 n 1. Like the majority, I address the preamendment version of the act. 3 PIP is common shorthand for “personal protection insurance” (a.k.a. “personal injury protection insurance”). 3 Subsection (1) requires a vehicle’s owner to “maintain security for payment of benefits under personal protection insurance, property protection insurance, and residual liability insurance.” There are two pieces here: first, the owner must “maintain security”; second, that security must ensure payment of benefits under PIP insurance (which pays for bodily injury) and property-protection insurance (which pays for property damage).4 As Subsection (3) explains, security “may be provided under a policy issued by an authorized insurer that affords insurance for the payment of benefits described in subsection (1)”— put more simply, to maintain security under § 3101 is to have a no-fault policy with PIP and property-protection insurance. According to the majority, as long as the vehicle is insured, MCL 500.3113(b)’s disqualification provision has not been triggered, no matter who is named in the policy. But in my view, the majority misreads that provision. Section 3113 requires security (i.e., a no-fault policy) “with respect to” the vehicle owned by the person claiming PIP benefits.5 The majority assumes that a policy is “with respect to” a vehicle if the policy insures the vehicle. But § 3113(b) governs PIP coverage; it isn’t aimed at vehicle coverage. Indeed, the no-fault act excludes damaged vehicles from coverage under property-protection 4 See, e.g., MCL 500.3105; MCL 500.3121; MCL 500.3131. 5 The majority explains in detail why the “which” in § 3113(b)’s “with respect to which” refers to “motor vehicle” and not to “owner.” See ante at 20-23. I generally agree. (I also observe that a similar “with respect to which” phrase is in MCL 500.3135(3), which affords limited immunity from “tort liability arising from the ownership, maintenance, or use within this state of a motor vehicle with respect to which the security required by section 3101 was in effect.”) 4 insurance. See MCL 500.3123(1)(a). For that reason, I cannot credit the majority’s assumption.6 The majority doesn’t rely on the statute’s phrase “with respect to”; indeed, it appears to toss out that language in favor of its own formulation, stating that insurance is “on the vehicle” or “for the vehicle.”7 That formulation lets the majority assume, contrary to the Legislature’s text, that the statute connects only the security and the vehicle. I agree with the majority that § 3113(b) contemplates a relationship between “security” and “vehicle,” but the majority’s conclusion that security “with respect to” a vehicle is the same as insurance “for” or “on” that vehicle is not, in my view, correct since nothing in the no-fault act requires a vehicle to be insured. Rather, MCL 500.3101(1) requires a certain person (the vehicle’s owner or registrant) to maintain security against liability, and “with respect to” connects the security to the vehicle by way of the person.8 Throughout the no-fault act, and in § 3113(b) in particular, the owner or registrant is front and center. In sum, there 6 The lapse is understandable—our own precedent has made the same assumption. See, e.g., Cruz v State Farm Mut Auto Ins Co, 466 Mich 588, 595; 648 NW2d 591 (2002) (the act was “designed to regulate the insurance of motor vehicles”). But to protect against accident damage to one’s vehicle, one must have collision coverage, see MCL 500.3037, which is neither required nor regulated by the no-fault act. What’s strange about the majority’s focus on insuring the vehicle is that, as the record shows, collision coverage was absent from plaintiff’s father’s policy, and so that policy didn’t insure plaintiff’s car. And recall that plaintiff’s claim was not for damage to his car but rather for PIP benefits. 7 See ante at 4 n 6 (“insurance . . . on the vehicles”), 15 (“insurance on the vehicle,” “insurance for the vehicle”), 16 (“insurance for a vehicle”), and 18 (“insurance on a vehicle”). 8 Cf. MCL 500.3101(1) (referring to “the insured owner or registrant”) (emphasis added). 5 exists security “with respect to” a vehicle not when that vehicle is insured but rather when that vehicle’s owner “maintains security.”9 This critical nexus among owner, security, and vehicle is of a whole with the rest of the no-fault act. While the majority hasn’t grappled with this contextual consideration, my reading is consistent with, and therefore supported by, the act’s other sections, like the priority schemes set out in MCL 500.3114 and MCL 500.3115. As those provisions explain, accident victims not otherwise covered by a personal or household10 policy must submit claims to other insurers in accordance with a priority scheme. Under § 3114(4), an injured occupant of a vehicle must claim against insurers “in the following order of priority”: (a) The insurer of the owner or registrant of the vehicle occupied. (b) The insurer of the operator of the vehicle occupied. 9 The majority determines that “maintain” means “to keep in an existing state (as of repair, efficiency, or validity) : preserve from failure or decline.” That’s an odd choice of definition for a few reasons. To start with, if “maintain” carries that meaning, it’s not clear what “existing state” is being “kept.” Before plaintiff bought the car, he lacked insurance (for himself and for the car), so under the majority’s reasoning, plaintiff’s father’s buying insurance disrupted (rather than “kept”) the existing state. And I’m not sure what it means for an insurance policy to be in a state “of repair, efficiency, or validity.” Nor do I know what it means to “preserve” an insurance policy “from failure or decline.” Still, the majority refers to its chosen meaning as “commonly understood” in this context. Ante at 17-18. By my lights, a better fit is the definition “to support or provide for.” Merriam- Webster’s Collegiate Dictionary (11th ed). In any event, the larger flaw in the majority’s reasoning is its misunderstanding not of the meaning of “maintain” but rather of what’s maintained, as explained throughout this opinion. 10 Under MCL 500.3114(1), a “personal protection insurance policy” reaches “accidental bodily injury to the person named in the policy, the person’s spouse, and a relative of either domiciled in the same household.” 6 Similarly, under § 3115(1) an injured nonoccupant (e.g., a pedestrian) must claim in this order: (a) Insurers of owners or registrants of motor vehicles involved in the accident. (b) Insurers of operators of motor vehicles involved in the accident. At the outset, note that these provisions refer as the object of no-fault coverage to persons (owners, registrants, operators), not to vehicles.11 More crucially, the priority scheme reflects the prenominate nexus among owner, security, and vehicle. That nexus is reflected too by the act’s design and purpose, which falls apart if an owner named in no policy is nonetheless understood to have “maintained security” through a nonowner third party’s policy. As we’ve said, the no-fault act created a “system of compulsory insurance, whereby every Michigan motorist would be required to purchase no-fault insurance . . . [and] victims of motor vehicle accidents would receive insurance benefits for their injuries as a substitute for their common-law remedy in tort.” Shavers v Attorney General, 402 Mich 554, 579; 267 NW2d 72 (1978) (emphasis omitted). The Legislature thus struck a balance between “assured, adequate, and prompt reparation” for accident victims and “compulsory insurance” to cover owners’ potential liability for claims by those victims. Id. We have described the PIP scheme as “comprehensive,” id., which reflects the tight fit between the owners’ compulsory insurance and the victims’ assured reparation. 11 See also MCL 500.3125 (priority for property-protection claims: “insurers of owners or registrants of vehicles involved in the accident; and insurers of operators of vehicles involved in the accident”) (emphasis added). 7 In other words, the no-fault scheme aims to assure (insofar as possible) a liable insurer for every victim. To meet that goal, the Legislature required two things: (1) that every owner (or registrant) “maintain security,” MCL 500.3101(1), and (2) that certain injured persons submit their claims according to a priority scheme.12 It’s plain that the act’s comprehensiveness—the tight fit between its means and its ends—falls apart under the majority’s reading, which relieves an owner of the burden to “maintain security” for liability for bodily injury and property damage. To illustrate, let’s say plaintiff had hit a pedestrian not covered by a personal or household policy. The priority scheme, MCL 500.3115(1), directs the hypothetical pedestrian to submit a claim to the “insurers of owners . . . of motor vehicles involved in the accident,” but since plaintiff has no insurer, the pedestrian’s claim would be outside the priority scheme,13 and he or she would be limited to recovery through the assigned-claims 12 See MCL 500.3114(4); MCL 500.3115(1); see also Royal Globe Ins Cos v Frankenmuth Mut Ins Co, 419 Mich 565, 575; 357 NW2d 652 (1984) (“The priority provisions of the act are designed to help implement [the act’s] goals.”). 13 Unless, of course, plaintiff’s father were determined to be an owner, in which case his personal policy’s issuer would be within the priority scheme. See MCL 500.3115(1). And as the Court of Appeals explained in Iqbal v Bristol West Ins Group, 278 Mich App 31; 748 NW2d 574 (2008), MCL 500.3113(b) doesn’t disqualify an owner lacking insurance as long as another owner has a no-fault policy. In Iqbal, the Court of Appeals determined that the plaintiff, a vehicle owner lacking insurance, was not disqualified from PIP benefits under MCL 500.3113(b) because his co-owner had a no-fault policy. The Court observed that “the BMW had the coverage” and so appears at first blush to have made the same mistake as the majority here, that the no-fault act is concerned with insuring vehicles. Id. at 40. But the Court of Appeals’ reasoning was more nuanced than that, recognizing that the issue was “whether the BMW, and not plaintiff, had the coverage or security required by MCL 500.3101.” Id. at 39 (emphasis added). Although inartful, the Court of Appeals’ framing of the issue reveals its understanding that security is related to a vehicle not 8 plan.14 The pedestrian’s PIP benefits then would be funded through increased rates for all policyholders, as though the pedestrian were a hit-and-run victim.15 This is how the majority understands the no-fault act. Yet under my reading, the act would require security for payment of PIP benefits to be maintained by an owner (or registrant), and so the pedestrian would be able to recover within the no-fault act’s priority scheme, rather than through the assigned-claims plan. Much puzzles me about the majority’s interpretive approach. It makes some obvious lapses in its understanding of the no-fault act’s text—I again point out the majority’s insistence that the no-fault act concerns insurance “on” (or “for”) a vehicle, despite the act’s exclusion of damaged vehicles. And its interpretation of “maintain” in MCL 500.3101(1) lacks nuance because it analyzes the term in isolation from the rest of MCL 500.3101(1). According to the majority, plaintiff “maintained security” (albeit because the security insures the vehicle but because of the nexus among owner, security, and vehicle. 14 See MCL 500.3171 et seq.; see also Rohlman v Hawkeye-Security Ins Co, 442 Mich 520, 529; 502 NW2d 310 (1993) (“[E]ssentially all accidents are now covered by personal injury protection benefits or the assigned claims plan.”). 15 Note that under the majority’s understanding of the security requirement, plaintiff in this hypothetical would enjoy limited immunity from tort liability because, according to the majority, plaintiff’s car was one “with respect to which the security required by section 3101 was in effect.” MCL 500.3135(3); see also note 5 of this opinion. In other words, unlike most Michigan vehicle owners, who merit this limited immunity by getting no-fault policies, plaintiff gets immunity while the cost of the injuries he’s responsible for is spread, through the assigned-claims plan, across other policyholders’ premiums. The majority now has charted the course for others to do the same. 9 indirectly) when “he instructed his father to obtain no-fault insurance.”16 Ante at 18. The majority suggests that plaintiff’s instruction to his father met the duty to maintain security in the same way that a son’s duty to maintain his lawn could be met by the son’s instructing his father to mow the lawn.17 But the son’s duty in those circumstances would not be met if the father mowed only the father’s own lawn. Likewise here, the son’s statutory duty to “maintain security” is not met by asking his father to get no-fault insurance if his father insures only himself. For this reason, it is to me neither here nor there that plaintiff’s father took out a no-fault policy in response to plaintiff’s instruction because that policy named the father, not plaintiff.18 The majority’s approach to statutory interpretation also gives short shrift to context and purpose. As I read the no-fault act, §§ 3101 and 3113 dovetail with other parts of the act, like the priority scheme; and my reading advances the act’s purposes, which we recognized in Shavers. The majority’s reading, on the other hand, barely acknowledges that we’re interpreting a small part of a larger system, let alone contemplates how its reading affects that larger system and undermines its purposes. The majority’s approach suggests (albeit implicitly) that neither context nor purpose plays a role in statutory interpretation. Yet the light thrown by context and purpose can cast the text in sharper 16 What, I wonder, stops a risky driver (like one with an OWI conviction) from saving money on insurance premiums by “instructing” a third party “to obtain no-fault insurance”? 17 The majority actually uses the example of a father instructing his son to maintain the lawn. I’ve swapped the father and son, to better track the facts in this case. 18 In other words, I don’t get caught up, as the majority does, in whether MCL 500.3101(1) requires plaintiff to have engaged in an “affirmative act,” ante at 18, because even if plaintiff’s instructing his father was an affirmative act, that act didn’t result in plaintiff’s compliance with § 3101. 10 relief. Indeed, had the majority paused for a moment to consider the act’s goals, it might have recognized that some of the premises underlying its decision are mistaken. As explained above, I read the no-fault act as disqualifying an owner from PIP benefits if the owner is injured in his or her own vehicle and no owner (or co-owner) has “maintained security.” But the record doesn’t reveal whether plaintiff’s father, who does have a no-fault policy, is an owner of plaintiff’s car, and so I cannot determine whether plaintiff is disqualified from PIP benefits. For this reason, I favor the Court of Appeals’ resolution—remand to the trial court for further proceedings. Elizabeth T. Clement CAVANAGH, J., did not participate in the disposition of this case because the Court considered it before she assumed office. 11