(concurring).
I concur in the result reached by the majority, but I disagree with its analysis. Based on its reading of Minn.Stat. § 548.251, subd. 1(2), the majority in essence concludes that subdivision 1(2) is not ambiguous as applied to the dispute in this ease and therefore concludes it is not necessary to go beyond the text of the statute to determine its meaning. I conclude that subdivision 1(2) is ambiguous as to whether payments by a tortfeasor’s automobile liability insurer are collateral and that a more comprehensive analysis of the statute’s meaning is necessary. After conducting such an analysis, I, like the majority, conclude that Minn.Stat. § 548.251, subd. 1(2), does not apply to a settlement payment made by a tortfeasor’s automobile insurer to an injured person who was later awarded no-fault benefits from his automobile insurer.
The crux of the issue raised by Do is how to interpret the language in subdivision 1(2). The parties dispute how we should read this language. American Family argues that we should read the *861statute as applying to payments pursuant to “automobile accident insurance,” an interpretation broad enough to include a payment from the tortfeasor’s automobile liability insurer. Do argues that we should read the statute as applying to payments pursuant to automobile accident insurance that “provides health benefits or income disability coverage,” a narrower interpretation which would exclude a tortfea-sor’s liability insurance settlement payment.
As the majority states, our goal in statutory interpretation is to give effect to the intent of the legislature. Heine v. Simon, 702 N.W.2d 752, 764 (Minn.2005). When interpreting a statute, we first examine whether the statute’s language is clear or ambiguous on its face. Id. A statute is ambiguous when its language is subject to more than one reasonable interpretation. Id. Ambiguity can arise from the statute’s application to the facts of a particular case. Id. (citing Minn.Stat. § 645.16 (2004)). We presume that statutes are consistent with the common law, and if a statute abrogates the common law, “the abrogation must be by express wording or necessary implication.” Brekke v. THM Biomedical, Inc., 683 N.W.2d 771, 776 (Minn.2004) (citation omitted) (internal quotation marks omitted).
We have previously said that subdivision 1(2) of the collateral source statute is ambiguous. Imlay v. City of Lake Crystal, 453 N.W.2d 326, 333-34 (Minn.1990). The parties in Imlay made arguments similar to the parties here. Id. at 334. But we did not settle the question of how to interpret the statute; instead, we said the statute’s language was “poorly written [and] ambiguous” and that the language “could conceivably be read as providing for one, two, three or four different types of collateral source benefits.” Id. Because we concluded there were “grammatical and analytical problems with each of the possibilities,” we recommended that the legislature clarify its intent. Id. But the legislature has not responded to this suggestion by revising the statute.1 As a result, we are still faced with language in section 548.251 that is ambiguous.
The use of the word “or” twice in section 548.251, subdivision 1(2), is key to the uncertainty regarding the meaning of the collateral source statute. Every use of “or” as a conjunction “involves some risk of ambiguity.” Bryan A. Garner, A Dictionary of Modern Legal Usage 624 (2d ed. 1995). This ambiguity can arise because the word “or has an inclusive sense as well as an exclusive sense.” Id. The use of the word “or” in subdivision 1(2) raises the question as to what the phrase “that provides health benefits or income disability coverage” modifies. How one reads the word “or” — inclusively or exclusively — can lead either to the broad reading of the text advocated by American Family or to the narrow reading urged by Do.
Here, as we did in Imlay, I conclude that our court cannot determine the meaning of the statute based on reference to the text alone. I conclude the statute is ambiguous because it is subject to more than one reasonable interpretation — one being that subdivision 1(2) refers to any payment related to automobile accident insur-*862anee, including a payment under the tort-feasor’s automobile insurance policy, and another being that the language applies only to benefits paid by the plaintiffs own no-fault insurer, which provides health benefits and income disability coverage. Thus I conclude that it is necessary to look beyond the text, to our court’s previous decisions regarding the collateral source statute and no-fault automobile insurance.
Several of our previous cases suggest that a payment by a tortfeasor’s insurer falls outside the definition of a collateral source. In Dean v. American Family Mutual Insurance Co., we stated that, “a tortfeasor’s liability insurance cannot, by definition, constitute a collateral source.” 535 N.W.2d 342, 345 (Minn.1995). We stated in Dean that although the legislature had not been “precisely clear” in defining collateral sources, “it is patently clear that a tortfeasor’s liability insurance can never be within the definition of collateral source.” Id. We reached this conclusion based in part on the definition of the common-law collateral source rule in Black’s Law Dictionary, which reads, “Under this rule, if an injured person receives compensation for his injuries from a source wholly independent of the tort-fea-sor, the payment should not be deducted from the damages which he would otherwise collect from the tort-feasor.” Id. (quoting Black’s Law Dictionary 262 (6th ed. 1990) (emphasis added)). We also noted our previous statement that “one distinguishing element of a collateral source is that the money or services in reparation of plaintiffs injury is from a source other than the tortfeasor.” Id. (emphasis added) (citing Hueper v. Goodrich, 314 N.W.2d 828, 830 (Minn.1982)). Though the collateral source statute partially abrogated the common-law collateral source rule and does not allow plaintiffs to receive double recoveries in certain situations, the fact that a payment from a tortfeasor was not and could not be a collateral source under the common-law rule is instructive to our analysis of Minn.Stat. § 548.251, subd. 1(2). In other words, the language and analysis in Dean discussing common-law definitions is helpful because though the rule on whether collateral sources can be used to reduce a plaintiffs damage award has changed, the fact that a tortfeasor’s payments have never been collateral sources informs our analysis of whether a settlement payment made by a tortfeasor’s insurer is a collateral source.
I, like the majority, disagree with the court of appeals’ suggestion that we have retreated from our statements in Dean that a tortfeasor’s liability insurance can never be within the definition of a collateral source. Do v. Am. Family Mut. Ins. Co., 752 N.W.2d 109, 116 (Minn.App.2008). The court of appeals cited our decision in Western National Mutual Insurance Co. v. Casper, 549 N.W.2d 914, 916 (Minn.1996), and stated that the facts of that case were “more closely aligned” to Do’s case than were the facts of Dean. Do, 752 N.W.2d at 116.
In Casper, the order of events was the same as in Dean: the plaintiff settled with the tortfeasor’s automobile liability insurer for $50,000, then sought UIM benefits from his employer’s insurer. Casper, 549 N.W.2d at 915. The plaintiff and the insurer could not agree on the amount of UIM benefits due and submitted the matter to an arbitration panel. Id. T he arbitration panel awarded the plaintiff $120,000 from the insurer. Id. We held that the tortfeasor’s $50,000 settlement payment should be deducted from the arbitration award. Id. at 915-16. We did not discuss whether there should be a deduction, only whether the deduction amount should be the full $50,000 settlement payment made by the tortfeasor’s liability insurer, or the $13,589 net amount that Casper received from the settlement. *863Id. It is not clear what type of offset we were applying in Casper because our analysis of the offset issue did not specifically mention the collateral source statute, nor did it discuss Dean or any other case law.2 Id. The procedural and factual history of Casper is complex, but given how it was resolved, I conclude that nothing in Casper negates the analysis that we set forth in Dean regarding the collateral source statute.
As suggested in Dean, the collateral source statute was written to classify payments by an injured person’s no-fault insurer as a collateral source because it is the no-fault insurer that provides the injured person with “health benefits or income disability coverage.” MinmStat. § 548.251, subd. 1(2). Here, it was Do’s no-fault insurance policy, not the tortfea-sor’s liability insurance, that provided Do with health benefits and income disability coverage.3 Because the tortfeasor’s liability insurance settlement payment did not provide Do with health benefits and income disability coverage, it is not a collateral source as defined by Minn.Stat. § 548.251, subd. 1(2).
The order of events in Nelson v. American Family Insurance Group is similar to Do’s case and I find our holding in Nelson to be instructive, given that it was decided after both Dean and Casper. 651 N.W.2d 499 (Minn.2002). In Nelson, after the plaintiff, Sharon Nelson, was involved in an accident in South Dakota, her insurer denied her no-fault claims, and she commenced a personal injury action against the tortfeasor. Id. at 502. The jury returned a verdict against the tortfeasor, awarding Nelson damages of $37,000 for past income loss. Id. One-third of Nelson’s recovery for past income loss was paid to her attorney under a contingency fee arrangement. Id. Nelson then sued her insurer for no-fault benefits. Id. The issue before us was whether Nelson was entitled to no-fault benefits to the full extent of attorney fees paid to achieve her tort recovery. Id. at 503.
In concluding that Nelson’s insurer was obligated to pay its proportional share of her attorney fees, we rejected proposed solutions that would have resulted in a windfall for Nelson or her insurer. Id. at 509-10. We concluded Nelson was entitled to “a recovery that mirrors what she would have received if [her insurer] had initially paid the no-fault benefits.” Id. at 509. We said that adopting the position of Nelson’s insurer “would frustrate the Act’s purpose to ensure prompt payment of no-fault benefits because it would encourage insurers to delay payment of no-fault benefits in the hope that the insured would recover up to the policy limits from a third-party tortfeasor.” Id. at 510. Based on the foregoing case law, I conclude that a core principle of the No-Fault Act is that automobile insurers bear the responsibility *864for paying basic economic loss benefits— no-fanlt benefits — to their policyholders, and any incentive to shift that responsibility to tortfeasors is not looked upon with favor.
With the foregoing analytical framework in mind, the next step in my analysis is to determine whether the collateral source statute applies to Do’s recovery from American Family. I acknowledge that if the district court had not applied the collateral source statute, and instead followed the procedure that Do advocates, then Do would have received a recovery greater than the amount the jury set as his total damages. Arguably, such a recovery amount results in a windfall to Do. But I conclude that subtracting the amount paid by the tortfeasor’s insurer from the jury award essentially rewards American Family for its resistance to paying no-fault benefits for which Do had contracted and to which the jury determined he was entitled. As a result of the court of appeals’ decision, American Family emerges at the end of these proceedings obligated to pay Do only $21,416.13 out of Do’s no-fault policy limit of $30,000. Put another way, the court of appeals’ decision means American Family does not have to pay $8,583.87 in benefits that it contracted to pay and was obligated to pay under its policy pursuant to the No-Fault Act. In essence, the court of appeals’ decision allows American Family to receive a windfall as a consequence of its refusal to pay Do’s no-fault medical expense benefits.
If the choice is to give a windfall to one party or another, our precedent informs us that we should read the collateral source statute in favor of a plaintiff in Do’s position. One of the purposes of the No-Fault Act is to “provide offsets to avoid duplicate recovery,” Minn.Stat. § 65B.42(5) (2008), but we have said that “if there is to be a windfall either to an insurer or to an insured, the windfall should go to the insured,” Stout v. AMCO Ins. Co., 645 N.W.2d 108, 114 (Minn.2002). We have also said that “if the question must be resolved on the basis of who gets a windfall, it seems more just that the insured who has paid a premium should get all he paid for rather than that the insurer should escape liability for that for which it collected a premium.” Van Tassel v. Horace Mann Mut. Ins. Co., 296 Minn. 181, 187, 207 N.W.2d 348, 352 (1973).
I am also concerned that the court of appeals’ reading of the collateral source statute will encourage insurers to delay paying no-fault benefits because of the possibility that the tortfeasors may pay some or all of what the no-fault insurer was obligated to pay its injured insureds. This likelihood risks undermining Minnesota’s no-fault system, which is predicated in part on “prompt payment of specified basic economic loss benefits to victims of automobile accidents” regardless of who was at fault for the accident, and “appropriate medical ... treatment of ... automobile accident victim[s]” through “prompt payment for such treatment.” Minn.Stat. § 65B.42(1), (3) (2008). The prospect of delayed payments by insurers was a concern we had in Nelson.4 See 651 N.W.2d at 505 (citing Minn.Stat. § 65B.42); see also Stout, 645 N.W.2d at 114. I do not believe the legislature intended the collateral source statute to operate in this fashion — allowing insurers to *865escape their obligations under the No-Fault Act and shift responsibility for no-fault benefits to a tortfeasor’s insurer.5
For the foregoing reasons, I conclude that a payment by a tortfeasor’s liability insurer is not a collateral source for purposes of the collateral source statute, Minn.Stat. § 548.251. Therefore, I agree with the majority that the district court and the court of appeals erred when they held that the collateral source statute required the district court to reduce Do’s injury award by $28,000, the amount of the settlement payment. Accordingly, I agree that we should reverse the court of appeals and remand the case to the district court for recalculation of Do’s net judgment against American Family in a manner consistent with this opinion.
Finally, I note that Do makes a second argument involving the district court’s method of calculating the net judgment. Do argues that the court should have first calculated his no-fault recovery and then calculated the UIM recovery. Do argues that in Richards v. Milwaukee Insurance Co., 518 N.W.2d 26, 28 (Minn.1994), we made it clear that the court is to first calculate a plaintiffs final no-fault recovery and then determine whether there has been a successful UIM claim. Because Do concedes that he has no UIM claim against American Family, I conclude that we need not and should not reach this issue.
. Professor Michael K. Steenson, author of a guide to Minnesota's no-fault system, suggests the legislature may have deliberately chosen not to apply the collateral source statute to a tortfeasor's payment. Steenson said:
While the collateral source statute can be construed to cover a third party's automobile liability insurance ... an alternative explanation is that the statute, by its terms, simply does not reach automobile liability insurance.
Michael K. Steenson, Minnesota No-Fault Automobile Insurance § 9.03, at 9-13 (3d ed. 2009) (emphasis added).
. Our only discussion in Casper of the collateral source statute concerned the insurer's separate argument that it was entitled to an offset based on the amount of workers’ compensation benefits paid to the plaintiff or the value of future workers' compensation benefits paid. 549 N.W.2d at 916-18. We held that the insurer was not entitled to such an offset. Id. at 918.
. Robert J. Hauer, Jr., an attorney who writes and lectures on no-fault insurance matters, has stated that a tortfeasor is not within the definition of a collateral source:
[A] "collateral source” is a third party (i.e. not the party injured and not the party causing the injury) who makes payments to or on behalf of the injured person. The collateral source is typically an insurance company, an employer, or a welfare agency that has paid medical or disability benefits.
Robert J. Hauer, Jr., Collateral Sources Issues, in Uninsured, Underinsured & No-Fault Insurance Update 1, 1 (Minn. State Bar Ass’n Continuing Legal Educ. 2008) (emphasis added).
. In Nelson, in hopes of serving both the no-fault policies promoting prompt payment and discouraging double recovery, we attributed to the no-fault insurer a pro-rated share of Nelson's attorney fees, thereby allowing Nelson "a recovery that mirrors what she would have received if [her insurer] had initially paid the no-fault benefits.” 651 N.W.2d at 509. Unlike in Nelson, the parties here have not presented any arguments or facts involving attorney fees.
. Professor Michael Steenson acknowledges that reading a tortfeasor’s liability insurance payment as outside the collateral source statute would result in a windfall to the plaintiff, but he adds:
[P]laintiffs received windfalls before the adoption of the collateral source statute, and it is arguable that the legislature simply failed to address the unique problem ... in Do. But, as in other cases, a policy, in this case the policy against double recovery, would not override the legislative omission. If the problem is to be corrected, it would be the legislature's to correct.
Steenson, supra, § 9.03, at 9-13 (emphasis added).