(concurring in part and dissenting in part). I concur in and join parts I, II, and III of the majority’s opinion affirming the award of work-loss benefits to plaintiff.

I dissent from part IV of the majority’s opinion reversing the award of attorney fees for plaintiff. I agree with the reasons stated in the Court of Appeals opinion for concluding that the trial court did not clearly err by awarding plaintiff attorney fees for defendant’s unreasonable failure to pay plaintiffs claim for work-loss benefits.

(concurring). In her partial dissent, Justice CORRIGAN asserts that my position in this case is inconsistent with my position in Kirksey v Manitoba Pub Ins Corp.1 She is incorrect. The plaintiff in Kirksey was an independent contractor. In this case, plaintiff is not an independent contractor; he is employed by a corporation. My colleague fails to recognize that this material difference distinguishes the case at bar not only from Kirksey, but from the other cases she discusses, as well.

This case arises out of a dispute over no-fault benefits. Plaintiff Randall Ross was injured in an automobile accident and submitted a claim for work-loss benefits to defendant Auto Club Group, his no-fault insurer. Defendant denied plaintiffs claim, prompting him to file this lawsuit. The trial court not only awarded plaintiff benefits, but also awarded attorney fees. The Court of Appeals affirmed.

We granted defendant’s application for leave to appeal. We hold that the trial court properly awarded plaintiff work-loss benefits. But it clearly erred when deciding that defendant’s refusal to pay benefits was not based on a legitimate question of statutory interpretation. As a consequence, we affirm the Court of Appeals judgment that plaintiff is entitled to work-loss benefits, but reverse its affirmance of the award of attorney fees.

I. PACTS

Plaintiff was injured in an automobile accident. At the time of the accident, he was the sole shareholder and sole employee of Michigan Packing Company, Inc. Plaintiff had incorporated this entity under the Busi*5ness Corporation Act (BCA)1 and, for federal tax purposes, had filed an election under subchapter S of the Internal Revenue Code.2

As a result of his injuries, plaintiff was unable to work. He made a claim to defendant for work-loss benefits. In support of his claim, plaintiff provided defendant with W-2 forms showing that Michigan Packing had paid plaintiff wages in 2001 through 2003.3

Defendant denied plaintiffs claim. It relied on the benefit-calculation methodology set forth by the Court of Appeals in Adams v Auto Club Ins Ass’n.4 Defendant concluded that plaintiff had failed to establish a claim for lost income because Michigan Packing operated at a loss during the years at issue.5

Plaintiff filed this lawsuit on May 6, 2004. The trial court granted his motion for summary disposition, ruling that he was entitled to work-loss benefits based on his wages. The court also awarded attorney fees under MCL 500.3148(1), the no-fault act’s attorney-fee provision. It found that defendant had unreasonably delayed making payment to plaintiff. Defendant moved for reconsideration. The trial court denied the motion, and defendant appealed in the Court of Appeals.

*6The Court of Appeals affirmed in a published opinion.6 It held that the trial court had properly rejected the benefit-calculation methodology proposed by defendant and had correctly granted benefits based on plaintiffs wages.7 The Court of Appeals also held that the trial court had not clearly erred by awarding attorney fees.8 It found defendant’s denial unreasonable because defendant had relied on a case having facts dissimilar to those of this case. Moreover, plaintiff had supplied W-2 forms supporting his claim.9 The Court denied defendant’s motion for reconsideration.

Defendant applied for leave to appeal in this Court. Initially, we denied the application, but later granted defendant’s motion for reconsideration. On March 7, 2007, this Court heard oral argument concerning whether “the Court of Appeals correctly affirmed the trial court’s award of attorney fees to plaintiff pursuant to MCL 500.3148(1).”10 After hearing argument on the application, we granted leave to appeal to consider both the attorney-fee issue and the benefits issue.11

II. STANDARD OF REVIEW

This case requires us to decide whether the lower courts properly interpreted the no-fault act in determining that plaintiff is entitled to work-loss benefits. Issues of statutory interpretation are reviewed de novo.12

*7We also review the award of attorney fees. The no-fault act provides for attorney fees when an insurance carrier unreasonably withholds benefits.13 The trial court’s decision about whether the insurer acted reasonably involves a mixed question of law and fact. What constitutes reasonableness is a question of law, but whether the defendant’s denial of benefits is reasonable under the particular facts of the case is a question of fact.14

Whereas questions of law are reviewed de novo, a trial court’s findings of fact are reviewed for clear error.15 A decision is clearly erroneous when “the reviewing court is left with a definite and firm conviction that a mistake has been made.”16

III. WORK-LOSS BENEFITS

The issue concerning work-loss benefits is one of first impression. It is whether someone can recover work-loss benefits under MCL 500.3107(l)(b) if he or she is the sole employee and shareholder of a subchapter S corporation that lost more money than it paid in wages. Defendant contends that plaintiff, who is such a person, is not entitled to benefits. Defendant points out that a subchapter S corporation’s profits and losses pass through to the shareholders for tax purposes. Accordingly, it argues, plaintiff should be treated like an unincorporated sole proprietor, which means that, when his income is calculated, his gross receipts must be reduced by his business expenses. The Court of Appeals rejected this argument.

*8In holding that the corporation’s losses were irrelevant to computing plaintiffs work-loss benefits, the Court of Appeals stated:

In this case, there is no dispute that (1) plaintiff received wages as an employee of the corporation and (2) plaintiff s remuneration from the corporation was not determined on the basis of the annual net income of the corporation. Plaintiff did not assert a work-loss claim based on the lost profits of the corporation. These facts distinguish this case from Adams. We reject defendant’s argument that plaintiffs self-employment status dictates a calculation of the gross receipts of the corporation less the corporate expenses to determine plaintiffs net income. We emphasize that plaintiff as an individual received wages and was not remunerated on the basis of the gross receipts of the corporation. Defendant presents no evidence to justify the disregard of the long-held rule that “ ‘[t]he corporate entity is distinct although all its stock is owned by a single individual or corporation.’ ” Moreover, “[a corporation’s] separate existence will be respected, unless doing so would subvert justice or cause a result that would be contrary to some other clearly overriding public policy.” Because plaintiff received wages from the corporation, and because defendant has presented no evidence to the contrary, the business expenses of the corporation are irrelevant in calculating plaintiffs wage loss, and plaintiff is treated as being in no different position than an employee of any other corporation operating at a loss. The trial court correctly determined that plaintiff was entitled to work-loss benefits and properly granted his motion for summary disposition.[17]

We conclude that the Court of Appeals reached the right result for the right reasons. Accordingly, we affirm its decision and hold that plaintiff is entitled to work-loss benefits based on his wages.

*9Justice CORRIGAN argues in her partial dissent that defendant has created a factual question regarding the amount of plaintiffs loss of income from work. This ignores the legitimate distinction between a shareholder and the corporate entity that is established by Michigan law. What Justice CORRIGAN says about the nature of a subchapter S corporation is true: for federal income taxation purposes, the income and losses of a subchapter S corporation pass through to the individual shareholders as if the income and losses belonged to the members of a partnership. But her “income” analysis errs by suggesting that the blurring of corporate and shareholder identities for federal taxation purposes also blurs the separate legal identities created for those entities by the BCA. Indeed, the authority she relies on made clear that, for federal taxation purposes alone, a subchapter S corporation is merely analogous to a partnership or a sole proprietorship.

There is no authority for Justice CORRIGAN’s proposition that the distinct corporate identity created by Michigan law may be ignored. The corporation’s income or losses are not the shareholder’s income or losses for purposes of the no-fault act’s work-loss-benefits provision. Neither the BCA nor the no-fault act supports her analysis. Thus, her statement that “plaintiff and his wife had no taxable income in 2001, 2002, and 2003”18 entirely misses the central point: regardless of whether he was subject to taxation under federal law, plaintiff indisputably received actual income in the form of W-2 wages in those years. Justice CORRIGAN seems to ignore the import of her own observation that “plaintiffs work as the sole employee of his corporation resulted in no income — but only overall lossés — to the *10corporation.”19 The losses ultimately belonged to the corporation, not to plaintiff.

At its core, Justice CORRIGAN’s position would accomplish a de facto piercing of the corporate veil. It would do this even though the shareholder had not engaged in fraudulent or wrongful conduct that would justify a court’s ignoring the corporate form. It would punish plaintiff for filing an election under subchapter S, a legitimate designation that permits him to report a loss for federal taxation purposes. Justice CORRIGAN indulges in speculation that defendant created a question of material fact to defeat summary disposition under MCR 2.116(0(10). And she fails to explain how the undisputed proof of plaintiffs wages is an inaccurate reflection of his loss of income from work and how the corporation’s losses could possibly diminish that figure.20

W. ATTORNEY FEES

The second issue is whether the award of attorney fees was proper.21 The no-fault act provides for an award *11of reasonable attorney fees to a claimant if the insurer unreasonably refuses to pay the claim. Specifically, MCL 500.3148(1) provides:

An attorney is entitled to a reasonable fee for advising and representing a claimant in an action for personal or property protection insurance benefits which are overdue. The attorney’s fee shall be a charge against the insurer in addition to the benefits recovered, if the court finds that the insurer unreasonably refused to pay the claim or unreasonably delayed in making proper payment.

The purpose of the no-fault act’s attorney-fee penalty provision is to ensure prompt payment to the insured.22 Accordingly, an insurer’s refusal or delay places a burden on the insurer to justify its refusal or delay.23 The insurer can meet this burden by showing that the refusal or delay is the product of a legitimate question of statutory construction, constitutional law, or factual uncertainty.24

The trial court correctly set forth this rule of law in determining that plaintiff was entitled to attorney fees. The issue is whether it clearly erred in applying this rule and finding that defendant’s refusal was not based on a legitimate question of statutory construction, constitutional law, or factual uncertainty. The determinative factor in our inquiry is not whether the insurer ultimately is held responsible for benefits, but whether its initial refusal to pay was unreasonable.

Plaintiff sought work-loss benefits under MCL 500.3107(l)(b), which states:

*12(1) Except as provided in subsection (2), personal protection insurance benefits are payable for the following:

(b) Work loss consisting of loss of income from work an injured person would have performed during the first 3 years after the date of the accident if he or she had not been injured.

In order to be entitled to benefits under this section, a plaintiff must suffer a loss of income.

Plaintiff in this case provided W-2 forms and asserted that they adequately represented his income. He made this claim despite the fact that he was the sole shareholder and sole employee of a subchapter S corporation that had lost more money then it paid him in wages. Defendant asserted that corporate losses must be considered when calculating “income” for a sole shareholder who is also the sole employee. Defendant’s argument presents an issue of first impression. In support of the argument that plaintiff was not entitled to work-loss benefits, defendant relied on the Court of Appeals decision in Adams v Auto Club Ins Ass’n.

In Adams, a motor vehicle accident permanently disabled the plaintiff. At the time of the accident, he was a self-employed cosmetologist who worked as an independent contractor. He paid 41 percent of his weekly gross revenue as chair rental and was also required to pay all of his own business expenses, including expenses for supplies and materials. When he applied for work-loss benefits after the accident, the defendant insurance company initially approved the payment of 85 percent of the plaintiffs average daily gross receipts. Approximately one year later, however, the defendant decided that the plaintiff was entitled to only 85 percent of his net receipts. The plaintiff brought suit, claiming that *13benefits should be calculated on the basis of his average daily gross receipts. The defendant argued that it should be allowed to deduct the plaintiffs business expenses in calculating his work-loss benefits.25

The Court of Appeals began its analysis in Adams by noting that the issue to be decided was the proper method for calculating work-loss benefits under MCL 500.3107(l)(b).26 The Court noted that the statute allows an injured party to collect benefits for “loss of income” but does not define that phrase.27 The Court of Appeals examined the statutory language, Michigan precedent, and decisions from sister states. It decided that, under the facts of the case, the term “loss of income” contemplated deducting the plaintiffs business expenses from his gross income in order to determine his work-loss benefits.28 The Court determined that this procedure was necessary to avoid awarding the plaintiff more in benefits than he would have taken home from his job had he been able to work.29 The Court was satisfied that this result was consistent with the no-fault act’s goal of placing individuals in the same, but no better, position than they were before their accidents.30

In this case, defendant relied on Adams. It argued that, because plaintiff was the sole shareholder and employee of Michigan Packing and the company lost more than it paid plaintiff in wages, plaintiff was not entitled to work-loss benefits. He suffered no loss of income. Defendant asserted that the benefit-calculation *14methodology set forth in Adams applies whenever a self-employed person is involved. Its theory was that, because a self-employed individual is solely responsible for profits and losses, the losses should be considered when determining whether there was a loss of income. Otherwise, defendant contended, plaintiff will end up in a better position financially than he was before the accident.

We acknowledge that this case differs from Adams in that the plaintiff in Adams was an unincorporated independent contractor, whereas Michigan Packing is incorporated. However, the inquiry is not whether defendant is responsible for the benefits, but only whether defendant’s refusal to pay them was unreasonable. As defendant points out, a subchapter S corporation does not pay income taxes; the business’s profits and losses pass through to the owners.31 Accordingly, because the profits and losses of Michigan Packing belonged to plaintiff for tax purposes, just as they belonged to the plaintiff in Adams, defendant’s reliance on Adams was reasonable. Adams is not directly on point, and, ultimately, we do not extend its reasoning to the facts of this case. Nonetheless, we conclude that the trial court clearly erred in deciding that defendant’s argument was not based on a legitimate question of statutory interpretation.

As the Court of Appeals acknowledged, how to calculate the “income” of an individual in plaintiffs situation for the purpose of determining work-loss benefits is an issue of first impression.32 Because MCL 500.3107(l)(b) does not define the term “loss of income,” we conclude that it was reasonable for defen*15dant to rely on the factually similar Adams decision. Its position that, in calculating plaintiffs loss of income, the losses suffered by Michigan Packing should be subtracted from the wages paid to plaintiff had support in law and fact.

V CONCLUSION

The issues we decide in this case are whether plaintiff was properly awarded work-loss benefits and whether defendant’s refusal to pay work-loss benefits was reasonable. The trial court held that plaintiff was entitled to benefits. It also awarded attorney fees after finding that defendant’s refusal to pay benefits was not reasonable. The Court of Appeals affirmed on both issues. We affirm the award of benefits but reverse the award of attorney fees. Although defendant was ultimately responsible for paying the benefits, its refusal to pay was not unreasonable. Defendant relied on a factually similar Court of Appeals decision to adopt a reasonable position on an issue of first impression.

(concurring in part and dissenting in part). I concur in part IV of the majority’s decision, which reverses the award of attorney fees to plaintiff. I respectfully dissent, however, from the majority’s conclusion in part III that plaintiffs W-2 wages established compensable loss of income from work under § 3107 of the no-fault act, MCL 500.3107(l)(b).

Plaintiff was the sole owner and sole employee of a subchapter S corporation that he operated at a loss that exceeded his wages. To establish compensable “loss of income from work” under MCL 500.3107(l)(b), plaintiff merely provided evidence that he paid himself W-2 wages. Defendant persuasively argues that plaintiffs W-2 wages were not a true measure of his income from work because plaintiffs work resulted in no actual income, but only created losses. Further, because of the unique tax status of S corporations, plaintiff reported the corporate losses on his personal tax returns and, as a result, paid no income tax on his wages. The no-fault act explicitly recognizes the relationship between income from work and taxable income in MCL 500.3107(l)(b), which provides that, generally, “[b]e-cause the benefits received from personal protection *17insurance for loss of income are not taxable income, the benefits payable for such loss of income shall be reduced 15%

Under these circumstances, for our purposes, the W-2 is a meaningless form that merely reflects the cash flow plaintiff allowed himself from a business that generated no income from work. Reimbursing him for this lost cash flow would, therefore, subsidize his preexisting business losses; it would not compensate him for actual loss of income from work. I acknowledge that plaintiff and his corporation have separate legal identities. See ante at 8. But this fact does not alleviate plaintiffs burden to establish, as a matter of fact, that he suffered loss of income from work. Because defendant created a genuine issue of material fact regarding whether plaintiff can establish any “loss of income from work an injured person would have performed,” MCL 500.3107(l)(b), I would reverse and remand this case to the trial court for further proceedings.

I. STANDARD OP REVIEW

This Court reviews de novo a trial court’s decision granting or denying a motion for summary disposition. City of Taylor v Detroit Edison Co, 475 Mich 109, 115; 715 NW2d 28 (2006). This case involves a question of statutory interpretation, which we also review de novo. Haynes v Neshewat, 477 Mich 29, 34; 729 NW2d 488 (2007). The goal of statutory interpretation is to effectuate the Legislature’s intent as demonstrated by the text of the statute. Casco Twp v Secretary of State, 472 Mich 566, 571; 701 NW2d 102 (2005). “If the statutory language is unambiguous, the Legislature is presumed to have intended the meaning expressed in the statute and judicial construction is not permissible.” Id.

*18II. ANALYSIS

The no-fault act describes the benefits available for “work loss” in pertinent part as those for

\w]ork loss consisting of loss of income from work an injured person would have performed during the first 3 years after the date of the accident if he or she had not been injured. Work loss does not include any loss after the date on which the injured person dies. Because the benefits received from personal protection insurance for loss of income are not taxable income, the benefits payable for such loss of income shall be reduced 15% unless the claimant presents to the insurer in support of his or her claim reasonable proof of a lower value of the income tax advantage in his or her case, in which case the lower value shall apply. [MCL 500.3107(l)(b) (emphasis added).]

“Work-loss benefits replace income that a claimant would have earned had he not been injured.” Popma v Auto Club Ins Ass’n, 446 Mich 460, 472; 521 NW2d 831 (1994). These benefits “are meant primarily to provide claimants with simple income insurance and are intended to compensate claimants approximately dollar for dollar for the amount of wages lost because of the injury or disability.” Id. Compensable work loss is not always measured by reference to a claimant’s preaccident wages, however. The statute defines “work loss” not as “lost wages,” but as “loss of income from work.” MCL 500.3107(l)(b) (emphasis added). In accord, for example, an independent contractor may seek work-loss benefits because “work loss includes not only lost wages, but also lost profit which is attributable to personal effort and self-employment.” Kirksey v Manitoba Pub Ins Corp, 191 Mich App 12, 17; 477 NW2d 442 (1991). Most significantly, “[i]n all cases, claimants are left to their proofs.” Popma, supra at 472. A plaintiff “will not be allowed to manipulate the statutory scheme to avoid this burden of proof.” Id.

*19As noted, plaintiff is the sole shareholder and sole employee of an S corporation. Shareholders of a small-business corporation may elect for the corporation to be treated as an S corporation under subchapter S of the Internal Revenue Code. 26 USC 1361(a)(1); 26 USC 1362(a)(1). An S corporation is generally not taxed directly by the federal government. 26 USC 1363(a). Instead, its tax liabilities and deductions, including income and losses, pass through to the individual shareholders on a pro rata basis. 26 USC 1366. “The effect is to treat electing corporations more like partnerships, since partnership income flows through to the partners and is taxed accordingly.” Chocola v Dep’t of Treasury, 422 Mich 229, 236; 369 NW2d 843 (1985).

Plaintiffs corporation reported overall losses that exceeded plaintiffs wages in 2001, 2002, and 2003. He listed the corporation’s losses on his personal income tax return. Accordingly, defendant offered as evidence the opinion of an accounting expert who concluded that, although plaintiff reported W-2 wages, he suffered no actual loss of income from work. Defendant argues that, under these circumstances, plaintiff should not be treated as a wage-earning employee of a distinct corporation. Rather, plaintiffs yearly income from work should be calculated by subtracting his business expenses from his gross receipts, as was done in Adams v Auto Club Ins Ass’n, 154 Mich App 186; 397 NW2d 262 (1986).

As the majority explains, ante at 13, the Adams panel acknowledged that the “goal of the no-fault act is to place individuals in the same, but no better, position than they were before their automobile accident.” Adams, supra at 193. It opined that the self-employed “plaintiff [could] not claim that his actual expendable income included even that income which he was re*20quired to pay out as business expenses.” Id. Therefore, to avoid overcompensating the self-employed plaintiff in Adams, it was appropriate to conduct a factual inquiry into whether certain business-related expenses should be deductible for purposes of determining work-loss benefits. Id. at 193-194.

I agree with defendant and find Adams applicable. Here, plaintiff operated his business at a loss that exceeded his W-2 income, and he reported his corporate losses for tax purposes. Indeed, plaintiff and his wife had no taxable income in 2001, 2002, and 2003. Most significantly, plaintiffs work as the sole employee of his corporation resulted in no income — but only overall losses — to the corporation. The expert opinion offered by defendant, which recounted these facts and conclusions, created a genuine issue of material fact regarding whether plaintiff can establish any “loss of income from work an injured person would have performed. ...” MCL 500.3107(l)(b). Put otherwise, defendant created a genuine issue of material fact with regard to whether plaintiffs W-2 wages constituted a correct measure of his loss of income from work.

In adopting the reasoning of the Court of Appeals in this case, the majority stresses that, to the contrary, plaintiff should merely be treated as any employee of a corporation that is operating at a loss. Ante at 8. But plaintiff cannot be compared directly to an employee of any corporation that is operating at a loss. Employees of subchapter C corporations receive real income, which is in no way offset by corporate losses, and must pay income taxes on their wages regardless of whether the corporation, itself, operates at a loss. Here, plaintiff used the S corporation’s losses to offset his wages and, as a result, he was relieved from paying any federal income tax.

*21Further, this and other courts explicitly recognize the unique nature of S corporations, which are often more comparable to partnerships than to corporations. Chocola, supra at 243 (“Subchapter S corporations enjoy unique characteristics that provide a compelling analogy to partnerships, which produce apportionable business income in the hands of member partners . . . .”); Tetlak v Village of Bratenahl, 92 Ohio St 3d 46, 48; 748 NE2d 51 (2001) (stating that subchapter S “treat[s] corporate income, losses, deductions, and credits as if incurred by individual shareholders in a manner akin to the tax treatment of partnerships”), citing Bufferd v Internal Revenue Comm’r, 506 US 523, 524-525; 113 S Ct 927; 122 L Ed 2d 306 (1993). Shareholders of an S corporation are taxed on the basis of their pro rata shares of all items of corporate income and loss, regardless of whether the income or loss is separately computed. 26 USC 1366(a). Similarly, when it is necessary to compute a shareholder’s annual gross income, it is calculated as his pro rata share of the corporation’s gross income. 26 USC 1366(c). As the Ohio Supreme Court observed in Tetlak, supra at 49, S corporations are not taxed as C corporations; rather, taxable income is computed essentially as if the S corporation were an individual. Therefore, shareholder income is characterized as if it originated from whatever source generated the income for the corporation. Id.; 26 USC 1366(b).

For these reasons, I conclude that a work-loss claim of a sole shareholder and sole employee of an S corporation is subject to a factual inquiry concerning the actual amount of lost income from work. In a given case, I may not disagree with the majority of my colleagues that the W-2 wages of an employee of an S corporation may be comparable to the wages of an employee of a C corporation and will be the appropriate *22measure of loss of income from work. The facts in this case, however, reveal a sole shareholder operating his S corporation essentially as a sole proprietorship. Here, plaintiff does not offer a principled argument that his gross income and operating expenses fail to reflect the most accurate measure of his actual income from work. Rather, he advocates a rule treating W-2 wages as the measure of loss of income from work under all circumstances simply because he is an employee of a corporation.

By adopting such a rule, the majority treats otherwise similarly situated sole proprietors differently on the mere basis of whether they choose to incorporate. Further, contrary to the majority’s assertion, ante at 10, I do not think that my position requires a “de facto piercing of the corporate veil” in any traditional sense. Rather, I recognize that plaintiffs profits or losses as the sole shareholder and sole proprietor of an S corporation may bear on whether he lost income from work as a matter of fact. Moreover, to any extent my view may be cast as requiring us to pierce the corporate veil, it is not at all clear that doing so would be inappropriate under these circumstances. As I have observed on more than one occasion, this Court has failed to establish clear standards for piercing the corporate veil. See L & R Homes, Inc v Jack Christenson Rochester, Inc, 475 Mich 853 (2006) (CORRIGAN, J., dissenting). As the Court of Appeals observed in Kline v Kline, 104 Mich App 700, 702-703; 305 NW2d 297 (1981), this Court has historically treated corporations and sole shareholders “as one for certain purposes,” in part because the “fiction of a corporate entity different from the stockholders themselves was introduced for convenience and to serve the ends of justice, but when it is invoked to subvert the ends of justice it should be and is disregarded by the *23courts.” “Each case involving disregard of the corporate entity rests on its own special facts.” Id. at 703.1

Most significantly, a rule treating W-2 wages as the measure of loss of income from work under all circumstances is not consistent with the no-fault act’s intent to compensate for the actual loss of work-related income caused by an accident. The no-fault act “is not designed to provide compensation for all economic losses suffered as a result of an automobile accident injury.” Belcher v Aetna Cas & Surety Co, 409 Mich 231, 245; 293 NW2d 594 (1980). Michigan courts have consistently engaged in factual inquiries to determine the true measure of an *24injured person’s compensable accident-related losses. For instance, in MacDonald v State Farm Mut Ins Co, 419 Mich 146, 150; 350 NW2d 233 (1984), the plaintiffs injuries from a car accident would have prevented him from working for a period of 28 months. During that period, however, he suffered a heart attack that disabled him from work for an indefinite amount of time. Id. This Court examined the language of former MCL 500.3107(b), a predecessor of MCL 500.3107(l)(b), both versions of which nearly are identical to § 1(a)(5)(h) of the Uniform Motor Vehicle Accident Reparations Act (UMVARA). Id. at 151. We observed that, “by adopting the language of such a model act, it is evident that the Legislature ‘was cognizant of, and in agreement with, the policies which underlie the model acts’ language’.” Id., quoting Miller v State Farm Mut Automobile Ins Co, 410 Mich 538, 559; 302 NW2d 537 (1981). Accordingly, we found the comments to § 1(a)(5) of the UMVARA relevant to Michigan’s act. MacDonald, supra at 151. The relevant comments read:

“Work loss”, as are the other components of loss, is restricted to accrued loss, and thus covers only actual loss of earnings as contrasted to loss of earning capacity. Thus, an unemployed person suffers no work loss from injury until the time he would have been employed but for his injury. On the other hand, an employed person who loses time from work he would have performed had he not been injured has suffered work loss * * *. Work loss is not restricted to the injured person’s wage level at the time of injury. For example, an unemployed college student who was permanently disabled could claim loss, at an appropriate time after the injury, for work he would then be performing had he not been injured. Conversely, an employed person’s claim for work loss would be appropriately adjusted at the time he would have retired from his employment.” [Id., quoting the comments to § 1(a)(5) of the UMVARA, found in 14 ULA 46-47.]

*25Both the clear language of former MCL 500.3107(b) and the comments to the UMVARA led the MacDonald Court to conclude “that work-loss benefits are available to compensate only for that amount that the injured person would have received had his automobile accident not occurred.” Id. at 152. The plaintiffs wages before the accident were not an automatic, true measure of his ongoing loss. Rather, the plaintiff “would have worked and earned wages for two weeks, until the date of his heart attack. After that date plaintiff would have earned no wage even had the accident not occurred and, therefore, [he was] ineligible for work-loss benefits after that date under § 3107(b).” Id. The import of the MacDonald Court’s decision was to “allow insurers to use the act as it was intended and avoid paying compensation not due the claimant.” Id. at 154 (emphasis added).

This Court employed similar reasoning, and relied on MacDonald, to hold in favor of the plaintiff in Marquis v Hartford Accident & Indemnity (On Remand), 444 Mich 638; 513 NW2d 799 (1994). There, the plaintiff was injured in an accident and alleged that her resulting temporary injury caused her to lose her job. During her period of disability, her job was offered to a permanent replacement employee. She was then unable to find a job that paid a similar amount. Id. at 640. She alleged that, although her period of disability had ended, she qualified for full work-loss benefits during the statutory three-year period because, but for the accident, she would have remained employed in her previous position. Id. at 642. This Court agreed that, although the availability of lower-paying work could be considered in terms of the plaintiffs obligation to mitigate her damages, she had created a genuine issue of material fact regarding whether the accident was the but-for cause of her loss of income at a higher wage. Id. at 649-650.

*26In sum, this Court has consistently condoned careful factual inquiry regarding the true measure of actual income lost as a direct result of an automobile accident. I also find it striking that Justice KELLY herself employed reasoning similar to that of the MacDonald and Marquis courts as the author of the Court of Appeals opinion in Kirksey. In Kirksey, she was presented with a plaintiff independent contractor who was injured while working for a trucking company. Kirksey, supra at 13-14. At the time of his injury, the plaintiff had the option of working for a second trucking company, which guaranteed him more hours and pay, if the first company was unable to fulfill its promises of more hours and certain benefits. Id. at 13. The plaintiff claimed that, because the first company offered him reduced hours at the time of his injury, the amount of his loss of income from work should not have been based on his earnings at the time of the accident. Rather, because he would have returned to work full-time at the second company but for his accident, his benefits should have been based on his earnings during his previous work with the second company. Id. at 14. Judge KELLY recognized that wages alone are not always a measure of loss of income from work; rather, an independent contractor such as the plaintiff could seek work-loss benefits because “work loss includes not only lost wages, but also lost profit which is attributable to personal effort and self-employment.” Id. at 17. I note that this comment directly reflects the comments to § 1(a)(5) of the UMVARA, which comments state, in part: “Work loss includes not only lost wages, but lost profit which is attributable to personal effort in self-employment (as distinguished from profit attributable from investment) .... [T]he issue is whether claimed work loss is justly attributable to the injury.” 14 ULA 47 (emphasis added). In Kirksey, Judge KELLY held that the plaintiffs *27earnings at the time of his accident were not necessarily the true measure of his loss of income as a result of the accident. Kirksey, supra at 16. Rather, if a jury were to find that, absent his injury, the plaintiff would have earned a higher income from the second company, his loss of income from work might be premised on that higher income. Id.

I am unable to square the majority’s holding in this case with Michigan jurisprudence, including MacDonald, Marquis, and Kirksey. The clear import of the no-fault act and these cases interpreting it is that an injured plaintiff may recover work-loss benefits on the basis of his actual loss of income, as reflected by the factual record.2 Accordingly, I cannot agree with the majority’s decision to establish a rule that, just because a sole proprietor incorporates his business, he will be treated differently than a sole proprietor with the same actual income and losses.31 would hold that plaintiff did not meet his burden to prove his actual amount of work *28loss merely by submitting his yearly W-2 wage amounts. Rather, because defendant showed that plaintiff operated his S corporation at a loss, defendant created a genuine issue of material fact concerning the amount, if any, of plaintiffs actual loss of income from work. I would reverse the Court of Appeals judgment affirming summary disposition under MCR 2.116(C)(10) in plaintiffs favor and remand this case to the trial court. I would further require plaintiff to offer proof on remand of the true measure of income from work that is necessary to put him in the same, but no better, position than the one he occupied before his accident.