Defendant appeals as of right, challenging the trial court’s order overruling defendant’s claims representative’s decision to discontinue disability benefits to plaintiff and ordering the reinstatement of monthly benefits. We agree with the trial court that the disability policy at issue is not governed by the Employee Retirement Income Security Act (ERISA), 29 USC 1001 et seq. Moreover, although we believe that the “arbitrary and capricious” standard of review should have applied to the evaluation of the claims representative’s decision, we are bound to follow a decision of this Court mandating a de novo standard of review under the circumstances. Therefore, we affirm.
I. FACTS
In 1968, plaintiff and a friend, Kent Seidman, began a vending machine supply company while they were students at Eastern Michigan University; they formed a corporation, originally known as Manimark Company, in 1970. After Seidman died in 1981, plaintiff continued the business by himself as president and chief executive officer. In 1981, plaintiff formed two corporations, Mani-mark Corporation and Manimark Associates. Manimark Corporation was formed to handle the general vending machine business activities, while Manimark Associates was created as a real estate holding company.
On March 20, 1987, defendant issued a disability policy to plaintiff. Plaintiff testified:
*118I believe that I purchased the Paul Revere policy personally on my own behalf, not as any kind of company benefit, but as a personal purchase. It was Manimark Corporation [that] received the bill and paid the premium because of the huge amount of premium it was.
And in order for me to pay that check or that bill personally, I would have had to take approximately 40 percent more out of the company in order to net down to the premium that Paul Revere wanted. So it was easier for me to have the company pay the bill and make a[n] end of the year tax adjustment on my taxes.
According to plaintiff, his employer, Manimark Corporation, paid one hundred percent of the premiums for defendant’s policy using checks drawn on the corporation’s bank account.
Plaintiff also testified:
[fit’s as though I paid [the premium] myself with my own personal funds. The corporate account was almost, not quite, but almost an extension of my checking — my personal checking account. It was my company. I owned all of it. I could do anything I want[ed] with it. If I wanted to buy something out of the corporation, I’d buy something out of the corporation. I also did that with life insurance for me. I mean, I don’t know how else to explain it. I just took money. It was a salary. It was my money. The corporate money is my money. It was my money. It wasn’t anybody else’s. It was my money.
Plaintiff acknowledged that he did not reimburse Mani-mark Corporation for those payments. In addition, plaintiff stated on his application for the policy that his occupation was “President — CEO” and that Manimark Corporation was his “employer.”
In the late 1980s, a steel hauler’s union sought to unionize plaintiffs company, and plaintiff successfully resisted the unionization. However, a second attempt at unionization in 1990 proved successful. The initial *119contact by the union bargaining agent was apparently unannounced. According to plaintiff, the agent came into his office, sat down in a chair, and placed a handgun on the table. Plaintiff claimed that the unionization of his company and the trauma of the contact with the union bargaining agent caused him to become agitated and anxious. He testified that about six months before he sold the business in 1996, he began to suffer chest pains, became despondent and anxious, and thought that he was going to have a heart attack.
Plaintiff filed an application for disability benefits under defendant’s policy on May 4,1996, alleging that he was completely unable to perform his occupational duties because of depression and “post-traumatic stress disorder,” which plaintiff claimed were caused by the incident with the union official in 1990. In his claim, plaintiff indicated that the policy was not an employer-sponsored policy. After defendant conducted an initial investigation and plaintiff submitted medical documentation in support of his claim, defendant paid benefits to plaintiff monthly for three years, beginning in 1996. In its initial investigation, defendant concluded that the policy was purchased individually and thus was not subject to ERISA.
Later, defendant1 had plaintiff examined by a psychiatrist who opined that plaintiff was not “so depressed that he cannot work.” Plaintiffs original physician disagreed. Following a final review of all the medical information, defendant denied plaintiffs claim, concluding, on the basis of the medical evidence, that *120plaintiff did not have a disabling condition. Defendant subsequently discontinued the monthly disability payments to plaintiff.
Plaintiff filed a complaint in the Wayne Circuit Court on February 11, 2000, alleging breach of contract, bad faith, and egregious conduct based on defendant’s termination of his monthly benefits under the policy. Although defendant did not seek to remove the action to federal court by asserting that plaintiffs claims were preempted by ERISA, the trial court requested, as a preliminary matter, that the parties address (1) whether the policy was governed by ERISA and (2) the appropriate standard of review to be applied in evaluating the decision of defendant’s claims representative to discontinue monthly benefits. The trial court subsequently ruled that the policy did not fall within the scope of ERISA and that “de novo” was the correct standard of review to be employed in evaluating the decision of defendant’s claims representative. The court then found “in favor of the plaintiff finding that they [sic] have borne their burden of proof to show that Mr. Krochmal is entitled to benefits from Paul Revere under the terms of [the] policy, and that this Court will overrule the decision of the Paul Revere adjuster finding Mr. Krochmal should be discontinued from benefits.”
II. POLICY NOT GOVERNED BY ERISA
Defendant argues that the trial court erred in determining that defendant’s policy was not governed by ERISA. Defendant claims that ERISA governs because plaintiffs employer established or maintained the policy. We disagree.
Whether the disability policy in question is part of an employee welfare benefit plan governed by ERISA involves a certain amount of statutory interpretation. We *121review issues of statutory interpretation de novo. Cruz v State Farm Mut Automobile Ins Co, 466 Mich 588, 594; 648 NW2d 591 (2002). Moreover, it appears that the trial court made its ruling as part of a request for a declaratory judgment. This Court reviews de novo a trial court’s decision with regard to a declaratory judgment action. Taylor v Blue Cross & Blue Shield of Michigan, 205 Mich App 644, 649; 517 NW2d 864 (1994). However, we review any factual findings of the trial court for clear error. Id.
Title I of ERISA applies to “ ‘any employee benefit plan.’ ” Fugarino v Hartford Life & Accident Ins Co, 969 F2d 178, 183 (CA 6, 1992), abrogated on other grounds by Raymond B Yates, MD, PC Profit Sharing Plan v Hendon, _US_; 124 S Ct 1330; 158 L Ed 2d 40 (2004), quoting 29 USC 1003(a). An employee benefit plan includes “ ‘an employee welfare benefit plan or an employee pension benefit plan or a plan which is both....’” Fugarino, supra at 183, quoting 29 USC 1002(3). An employee welfare benefit plan is defined by ERISA, in pertinent part, as “any plan, fund, or program ... established or maintained by an employer ... for the purpose of providing for its participants or their beneficiaries, through the purchase of insurance or otherwise ... benefits in the event of... disability[.]” 29 USC 1002(1).
In Thompson v American Home Assurance Co, 95 F3d 429, 434 (CA 6, 1996), the court stated that “[t]he existence of an ERISA plan is a question of fact, to be answered in light of all the surrounding circumstances and facts from the point of view of a reasonable person.” If an insurance policy is an employee welfare benefit plan under ERISA, then plaintiffs state-law claims relating to that policy, with some exceptions, are preempted, and federal common law will apply to determine any *122recovery. See, generally, Teper v Park West Galleries, Inc, 431 Mich 202, 207-214; 427 NW2d 535 (1988).
In Thompson, supra at 434-435, the Sixth Circuit Court of Appeals set forth a three-step factual inquiry to determine whether a benefit plan is an EEISA plan:
First, the court must apply the so-called “safe harbor” regulations established by the Department of Labor to determine whether the program was exempt from ERISA. Second, the court must look to see if there was a “plan” by inquiring whether from the surrounding circumstances a reasonable person [could] ascertain the intended benefits, the class of beneficiaries, the source of financing, and procedures for receiving benefits. Finally, the court must ask whether the employer “established or maintained” the plan with the intent of providing benefits to its employees. [Citations and some internal quotations omitted.]
The court went on to state:
Department of Labor’s (“DOL”) regulations set out a “safe harbor” provision that excludes an employee insurance policy from ERISA coverage if: (1) the employer makes no contribution to the policy; (2) employee participation in the policy is completely voluntary; (3) the employer’s sole functions are, without endorsing the policy, to permit the insurer to publicize the policy to employees, collect premiums through payroll deductions and remit them to the insurer; and (4) the employer receives no consideration in connection with the policy other than reasonable compensation for administrative services actually rendered in connection with payroll deduction. 29 C.F.R. § 2510.3-1(j). A policy will be exempted under ERISA only if all four of the “safe harbor” criteria are satisfied. [Thompson, supra at 435.]
In applying the three-part test set forth in Thompson, id. at 434-435, the first consideration is whether the four “safe harbor” provisions exclude the plan in question from ERISA coverage. Regarding the first safe harbor condition, whether the employer made a contri*123bution to the policy, the record indicates that Manimark Corporation paid the annual premiums with corporate checks drawn on the corporation’s own bank account. Although plaintiff testified that he did not reimburse Manimark Corporation for those premium payments, he testified that “my accountant at the end of the year put it on my taxes as though I had received a bonus or draw from the company.” According to plaintiff,
[t]here was no settling up. My accountant merely put it on my taxes like he would my salary. There was no settling up. The bill came to Manimark Corporation, Manimark Corporation paid the premium just as it would a draw or a salary.
Thus, plaintiff acknowledged in his deposition testimony that, although Manimark Corporation paid the premiums, the payment was reflected on his personal income tax form and he paid taxes on the amount of the premiums.2
Plaintiff claims that Manimark Corporation made no contribution to the policy because he paid the taxes on the premium payments, while defendant maintains that the record indicates that Manimark Corporation established or maintained the insurance. Specifically, defendant contends that “[s]imply paying the taxes on the premium payments in no way alters the fact that Manimark Corporation made the annual payments, without reimbursement from Krochmal, to maintain the Paul Revere Policy.” In support, defendant relies on an unpublished federal district court case from Arizona. *124We decline to find this unpublished case persuasive and instead rely on B-T Dissolution, Inc v Provident Life & Accident Ins Co, 175 F Supp 2d 978 (SD Ohio, 2001).
In B-T Dissolution, the court addressed whether Steven Matthews’s state-law claims for breach of contract and bad faith were preempted by ERISA. Id. at 979. After Matthews resigned his position as a managerial employee and minority shareholder in B-T Dissolution, Inc., he sought to recover disability benefits under separate insurance policies issued by the defendants. Id. Although the defendants initially paid Matthews’s claims for benefits, they stopped making the payments when they determined that Matthews was not “disabled” under their policies. Id. Matthews then filed suit in state court, alleging breach of contract and bad faith. Id. The defendant insurance companies removed the case to federal district court, seeking summary dismissal of Matthews’s state-law claims on the basis of ERISA preemption. Id.
In determining whether the insurance policies were part of an “employee welfare benefit plan” governed by ERISA, the district court in B-T Dissolution determined that the defendant insurance companies failed to establish the applicability of ERISA to Matthews’s disability insurance policies. Id. at 983. Specifically, the district court rejected the defendants’ contention that Matthews failed to satisfy the safe harbor criteria on the ground that his employer contributed to the premium payments for his two disability insurance policies. Id. at 983-986. In pertinent part, the court noted:
The evidence presented at the August 9, 2000, oral and evidentiary hearing persuades the Court that B-T did not “contribute” to the payment of Matthews’ policy premiums, within the meaning of 29 C.F.R. § 2510.3-1(j), the safe-harbor regulation. The Defendants insist that B-T plainly “contributed” to the premiums because it “paid” *125them. In particular, the Defendants note that B-T wrote the checks for the premiums and deducted them as business expenses. Although these assertions are correct, the evidence also reflects that the full amount of the premiums, at least in 1994, was included on Matthews’ W-2 forms as gross income. As a practical matter, then, the Court concludes that Matthews, rather than B-T, actually paid the premiums. Although B-T wrote premium checks directly to Provident and Guardian, those funds necessarily first “passed through” Matthews, who was required to report the payments made by B-T as gross income and to pay taxes on those amounts. As a result, despite the fact that B-T wrote the checks, it is apparent that the company did so using Matthews’ money. Indeed, if the money used by B-T to pay the premiums were not his money, then he would not have been required to report it on his W-2 forms or to pay taxes on it. Given that B-T paid the premiums with Matthews’ own income, the Court rejects the argument that the company “contributed” to those payments.
In opposition to the foregoing conclusion, the Defendants repeatedly stress that B-T deducted the premium payments for tax purposes. The Defendants reason that B-T must have “paid” the premiums, or else it could not have taken these deductions. Upon review, the Court finds this argument to be unpersuasive. Without question, the premium payments cost B-T a substantial amount of money, and the company properly deducted them as expenses. The flaw in the Defendants’ argument, however, is that the money for the premium payments flowed from B-T to Matthews. This is evident from the fact that the cost of the premiums appeared on his W-2 forms as taxable income. As a result, although B-T actually wrote the checks, it did so with his money. The fact that B-T deducted the cost of the premiums does not establish that the company “contributed” to them. The Court harbors no doubt that Matthews’ taxable salary also cost B-T a substantial amount of money, which the company was entitled to deduct on its tax returns. See, e.g., Eberl’s Claim Service, Inc. v. Commissioner of Internal Revenue, 249 F.3d 994, 998 (10th Cir. 2001) (recognizing that salaries paid by closely held corporations are deductible expenses). If Matthews *126then used his taxable salary to pay his mortgage, the Defendants could not seriously contend that B-T “contributed” to his mortgage payment. By the same token, the premium payments at issue were deducted hy B-T and were taxable gross income to Matthews. The fact that the money used to pay the Provident and Guardian disability insurance premiums originated with B-T does not mean that the company “contributed” to those premiums any more than it could be said to have “contributed” to Matthews’ mortgage payments if he used his salary income to make those payments. An obvious difference in the two situations, of course, is that B-T wrote the checks to pay the Provident and Guardian premiums, whereas Matthews presumably would write his own check to make a mortgage payment. This distinction is immaterial. In either case, the money used to pay the expense was Matthews’ own taxable gross income. As a result, the Court cannot agree that B-T “contributed” to the premiums, within the meaning of the safe harbor regulation. [B-T Dissolution, supra at 983-984 (emphasis in original).]
As pointed out in B-T Dissolution, the salient fact is that while the employer may have written a check to make the premium payment, it used the employee’s money to do so, even if the employer deducted the premium payments for tax purposes. Applying the reasoning set forth in B-T Dissolution to the facts of this case, we conclude that plaintiff, not his employer, contributed to the payment of the insurance premiums. Specifically, the record indicates that while Manimark Corporation was billed for the premiums and paid the premiums on its own corporate account, and while plaintiff ostensibly did not reimburse the corporation for those payments, the entire amount of the premium payments was accounted for on plaintiffs annual W-2 forms as taxable income, and plaintiff paid the taxes on it. Indeed, the reasoning set forth in B-T Dissolution applies with greater force to the facts of this case because there is no indication that Manimark Corpora*127tion deducted the premium payments for tax purposes. Accordingly, plaintiff satisfied the first prong of the safe harbor criteria because his employer did not make any contribution to the policy.
Regarding the second safe harbor condition, whether the employee’s participation in the policy is voluntary, the record indicates that plaintiff satisfied this condition because the purchase of the policy was independent of any action by Manimark Corporation.
Regarding the third safe harbor condition, whether plaintiffs employer endorsed the policy in question, there is no indication in the record that Manimark Corporation was involved in endorsing the policy. In Thompson, supra at 436-437, the Sixth Circuit Court of Appeals noted that “endorsement” involves the absence of neutrality regarding an insurance plan or policy. See also B-T Dissolution, supra at 986.3 In this case, there is nothing in the record to show that Manimark Corporation was substantially involved in the creation and administration of the program to support a finding of endorsement. Id.-, Thompson, supra at 437. Thus, plaintiff satisfied the third safe harbor condition.
*128Finally, plaintiff satisfied the fourth safe harbor condition because there was no indication that his employer received any consideration in connection with the policy. Because plaintiff satisfied all four of the safe harbor criteria, the insurance policy in question is not governed by ERISA, and we need not reach the two additional steps in the Thompson three-part analysis. See, generally, Thompson, supra at 437-438. The trial court did not clearly err in determining that the disability policy in question was not part of an “employee welfare benefit plan” that is governed by ERISA.4
Given our conclusion that plaintiff satisfied all four of the safe harbor criteria, we need not address the next issue raised by defendant, namely, whether plaintiff “wholly owned” Manimark Corporation in such a manner that the insurance policy in question was not governed by ERISA. (As noted in Yates, supra at 1344 n 6, an insurance policy covering a sole shareholder of a company and no other people is not subject to ERISA.) For the sake of completeness, however, we note briefly that defendant’s attempt to portray plaintiff as anything other than a sole owner of Manimark Corporation is without merit. While plaintiff admitted that he owned less than one hundred percent of the shares of Manimark Corporation because one percent was held by Manimark Associates, plaintiff testified that he wholly owned Manimark Associates. Moreover, plaintiff received all the stock of Manimark Corporation after his business partner’s death, and there is no evidence in the *129record of stock being transferred to Manimark Associates. The trial court correctly noted that “[a]ll of the stock was in the name of Mr. Krochmal even though there may have been some reference to one percent ownership for the purposes of a bond.”5 ERISA did not govern the policy in question for the additional reason that plaintiff was the sole owner of Manimark Corporation. Id.
III. STANDARD OF REVIEW APPLICABLE TO THE CLAIMS ADJUSTOR’S DECISION
Defendant argues that the trial court should not have reviewed the decision of its claims adjustor using a “de novo” standard of review but instead should have used an “arbitrary and capricious” or abuse of discretion standard. We agree. Nevertheless, because we are bound to follow the precedent established in Guiles v Univ of Michigan Bd of Regents, 193 Mich App 39, 47 n 4; 483 NW2d 637 (1992), we must reject defendant’s argument.
This issue involves a question of law, and we review questions of law de novo. Sun Communities v Leroy Twp, 241 Mich App 665, 668; 617 NW2d 42 (2000).
In Firestone Tire & Rubber Co v Bruch, 489 US 101; 109 S Ct 948; 103 L Ed 2d 80 (1989), the Supreme Court addressed the standard of review that is appropriate when reviewing an administrator’s decision denying benefits under 29 USC 1132(a)(1)(B). In Firestone Tire, supra at 115, the Supreme Court held that review de novo is appropriate in reviewing a challenge to the *130denial of benefits “unless the benefit plan gives the administrator or fiduciary discretionary authority to determine eligibility for benefits or to construe the terms of the plan.” If the plan grants discretion to the administrator, then the court must apply the deferential “abuse of discretion” or “arbitrary and capricious” standard. See id. at 114-115.
In this case, defendant relies on Perez v Aetna Life Ins Co, 150 F3d 550 (CA 6,1998), in which the Sixth Circuit Court of Appeals held that the arbitrary and capricious standard of review was applicable to a review of the defendant’s decision to terminate benefits because the insurance policy in question vested discretion in the defendant, the plan administrator, by requiring the insured to submit satisfactory evidence of disability. The Sixth Circuit Court of Appeals noted:
[The Supreme] Court in Firestone ... did not suggest that discretionary authority hinges on incantation of the word discretion or any other magic word. Rather, the Supreme Court directed lower courts to focus on the breadth of the administrators’ power — their authority to determine eligibility for benefits or to construe the terms of the plan. While magic words are unnecessary to vest discretion in the plan administrator and trigger the arbitrary and capricious standard of review, this circuit has consistently required that a plan contain a clear grant of discretion [to the administrator] to determine benefits or interpret the plan. [Id. at 555 (citations and internal quotations and emphasis omitted).]
The defendant in Perez argued that “the Plan contains the requisite clear grant of discretion to the plan administrator” by pointing to the following language contained in the policy:
“Written proof of total disability must be furnished to [Aetna] within ninety days after the expiration of the [first twelve months of disability]. Subsequent written proof of *131the continuance of such disability must be furnished to [Aetna] at such intervals as [Aetna] may reasonably require. ...
“[Aetna] shall have the right to require as part of the proof of claim, satisfactory evidence ... that [the claimant] has furnished all required proofs for such benefits....” [Id. (emphasis added by Perez).]
The Sixth Circuit Court of Appeals held that the language requiring “satisfactory evidence” to establish “the proof of claim” vested discretion in the defendant. Id. at 557. The Perez court noted that “[n]umerous federal courts, including our own, have held that language similar to that contained in the Plan clearly grants discretion to the plan administrator.” Id. at 556. The court further observed:
Because the Plan is governed by ERISA, we apply federal common law rules of contract interpretation in making our determination. In developing federal common law rules of contract interpretation, we take direction from both state law and general contract law principles. The general principles of contract law dictate that we interpret the Plan’s provisions according to their plain meaning, in an ordinary and popular sense.... In applying this plain meaning analysis, we must give effect to the unambiguous terms of an ERISA plan.
Although many of our prior cases finding a clear grant of discretion involved ERISA plans which explicitly provided that the evidence be satisfactory “to the insurer,” “to the company” or “to us,” it does not automatically follow that in the absence of such language discretion has not been granted to the plan administrator. Both parties acknowledge that the Plan allows for Aetna to request and receive satisfactory evidence of total disability before an individual is entitled to receive continued benefits. We agree with Aetna that this “right to require as part of the proof of claim satisfactory evidence” means, semantically, that the evidence must be satisfactory to Aetna, the only named party with the right to request such evidence. It naturally follows *132that Aetna, the receiver of the evidence, would review that evidence to determine if it constitutes satisfactory proof of total disability. It is simply implausible to think that Aetna would merely hold the evidence as a safekeeper or depository for a third party unnamed in the contract to review in making benefits determinations. This is all the more true when one considers that an insurance contract, even one governed by ERISA, is after all simply a contract — a mutual agreement between the two contracting parties.
In short, reading the contractual language in an ordinary and popular sense as we must, the only reasonable interpretation of the Plan is that Aetna requests the evidence, reviews it, and then makes a benefits determination. To reach any other conclusion would violate the basic principle of contract law that courts are not permitted to rewrite contracts by adding additional terms. We therefore conclude that the plan clearly grants discretion to Aetna because, under the only reasonable interpretation of the language, Aetna retains the authority to determine whether the submitted proof of disability is satisfactory. [Id. at 556-557 (citations and some internal quotations omitted).]
Accordingly, the Sixth Circuit Court of Appeals in Perez held that “the arbitrary and capricious standard should have been applied” and remanded “the case to the original panel to review Aetna’s decision to terminate benefits under the arbitrary and capricious standard.” Id. at 558.
The Perez court noted that its decision reaffirmed the holding in Yeager v Reliance Standard Life Ins Co, 88 F3d 376, 380-381 (CA 6, 1996), a case on which defendant also relies in support of its position that its claims representative was vested with discretionary power in determining plaintiffs benefits claim. Perez, supra at 558. In Yeager, the Sixth Circuit held that discretion was vested in the insurance company to determine whether the insured submitted “ ‘satisfactory proof of Total Disability to us.’ ” Yeager, supra at 380-381. As explained in Yeager.
*133A determination that evidence is satisfactory is a subjective judgment that requires a plan administrator to exercise his discretion. . .. [T]he Plan at issue in this case requires “satisfactory proof of total disability.” It would not be rational to think that the proof would be required to be satisfactory to anyone other than [the insurance company]. Even if the phrase “to us” is interpreted as defining to whom the proof should be submitted, there is no reason to believe that someone other than the party that received the proof would make a determination regarding its adequacy. Furthermore, the district court’s emphasis on the fact that the Plan language could have been clearer is misplaced. The mere fact that language could have been clearer does not necessarily mean that it is not clear enough. Therefore, we conclude that the Plan language granted the administrator discretion to determine eligibility for benefits, and the district court should have applied an arbitrary and capricious standard of review. [Id. at 381.]
In the instant case, the policy provides, in pertinent part:
After we receive satisfactory written proof of loss:
a. We will pay any benefits then due that are not payable periodically; and
b. We will pay at the end of each 30 days any benefits due that are payable periodically — subject to continuing proof of loss.
This language grants discretion to the plan administrator, just as did the pertinent language in Perez and Yeager. We find those authorities persuasive. We acknowledge that Perez and Yeager involved ERISA plans and that the instant case does not. Nevertheless, the reasoning from these cases applies with equal force to the instant, non-ERISA policy. As noted in Perez, supra at 556, “[t]he general principles of contract law dictate that we interpret the Plan’s provisions according to their plain meaning, in an ordinary and popular sense.” We find no salient reason why the general principles of *134contract law should not also apply to the provisions of a non-ERISA plan. See Bianchi v Automobile Club of Michigan, 437 Mich 65, 71 n 1; 467 NW2d 17 (1991) (setting forth the general rule that courts should construe contractual language according to its ordinary and plain meaning). The ordinary and plain meaning of the contract at issue indicates that defendant has discretion to determine whether plaintiff has submitted adequate proof of loss. We conclude that the arbitrary and capricious standard of review should have applied to the evaluation of the claims adjustor’s decision.6
However, in Guiles, supra at 47 n 4, this Court held, in evaluating a non-ERISA benefits plan, that the requirement of “satisfactory proof” of loss was insufficient to trigger the arbitrary and capricious standard of review. Specifically, the Guiles Court held:
Defendant submits that because the plan requires that a claimant submit “satisfactory proof” of total disability, the university reserved to itself complete discretion to determine eligibility. We find this argument disingenuous and accordingly reject it. Under Firestone, discretion is the exception, not the rule. Where an employer wishes to retain discretion, it may do so but it must do so clearly. In this case, the language relied on by defendant does not clearly imply that the university shall have the last word on entitlement to benefits. [Id. (citations omitted).]
We acknowledge that Guiles was decided before Perez and Yeager. Nevertheless, because it is a Court of Appeals decision addressing an issue of state law, we are bound to follow its holding. See MCR 7.215(J)(1). Therefore, we must conclude that the trial court did not err in rejecting the arbitrary and capricious standard of *135review. Instead, according to Guiles, supra at 43, a de novo standard of review applied.
IV APPLICATION OF THE DE NOVO STANDARD OF REVIEW
The court, employing the de novo standard of review, considered the available evidence and determined that plaintiff was “unable to perform the important duties of [his] Occupation” under the terms of the disability policy. While certain medical professionals concluded that plaintiff was not disabled, two others concluded that he was disabled. Given that the evidence did not significantly favor one conclusion over the other, we simply cannot say that the trial court clearly erred in finding that plaintiff was entitled to disability benefits. See Taylor, supra at 649 (this Court reviews a trial court’s factual findings for clear error).7
Affirmed.
We note that the Paul Revere Life Insurance Company underwent some corporate merging and restructuring and was known by different names throughout the instant proceedings. This opinion uses the term “defendant” to refer to Paul Revere in its various forms.
Defendant points out that plaintiff submitted an affidavit in which he averred that he, not his employer, paid the policy premiums. Defendant, citing Griffith v Brant, 177 Mich App 583, 587-588; 442 NW2d 652 (1989), claims that “a party cannot renounce by affidavit testimony that he has unambiguously given by way of deposition.” We note, however, that whether plaintiff or his employer paid the policy premiums was not unambiguously clear during the deposition testimony.
Specifically, in a footnote, the district court in B-T Dissolution, supra at 986 n 12, stated:
[T]he Court notes that an employer does not “endorse” an insurance program, within the meaning of the safe-harbor regulation, merely by collecting premiums through payroll deductions and sending them to the insurer. 29 C.F.R. § 2510.3-1(j)(3). Although the Defendants contend that the payment process in the present case did not involve “payroll deductions,” the Court concludes, based on the reasoning set forth, supra, that it was sufficiently similar. B-T’s role in the process involved nothing more than sending a portion of Matthews’ gross income to Provident and Guardian on his behalf. Based upon its reading of the safe-harbor regulation, the Court concludes that this limited level of employer involvement does not constitute “endorsement” of an insurance program, and the Defendants cite nothing to the contrary.
We have employed a clear error standard of review in reaching our ultimate conclusion here because of the Thompson court’s conclusion that “[t]he existence of an erisa plan is a question of fact,” see Thompson, supra at 434, and because this Court generally reviews a trial’s courts factual findings for clear error. See Taylor, supra at 649. We note, however, that we would reach the same conclusion even if we used the de novo standard of review advocated by the parties on appeal.
In Ms deposition, plaintiff explained that Manimark Associates owned one percent of MaMmark Corporation because “[i]t goes back to the financmg on the front building was a[]... bond and my partner died and we couldn’t have just one shareholder or the ... bond wasn’t valid so when he died we had to make .. . sometMng... else a one percent and it was really paperwork.”
We note that, in applying the arbitrary and capricious standard, courts may consider whether the administrator of the benefit plan is operating under a conflict of interest. See Firestone, supra at 115.
We note that defendant fails to argue on appeal that the trial court was not empowered to make the factual finding concerning plaintiffs disability. In other words, defendant does not make the alternative argument that, if a de novo standard of review applied, the case should have proceeded to a full trial. We further note that the application of the arbitrary and capricious standard of review to this case would change the outcome, because it cannot be said, given the available evidence and the considerable testimony that plaintiff could work, that the denial of benefits by defendant was arbitrary and capricious, even if we were to apply the “heightened” arbitrary and capricious standard (i.e., a standard less deferential to the insurance company) advocated by cases such as Pinto v Reliance Standard Life Ins Co, 214 F3d 377 (CA 3, 2000), in situations where a conflict of interest may exist (i.e., where the insurance company is both the funder and the administrator of the plan). Because it is not necessary under the facts of this case, we do not decide whether this “heightened” standard should be adopted in Michigan. We conclude only that some form of the arbitrary and capricious standard should apply.