odissenting). We granted leave in this case to address two important questions: (1) whether this Court has adopted 1 Restatement Agency, 2d, § 219(2)(d) and, if not, (2) whether we should adopt it now.
Regarding the first question, this Court has not explicitly adopted § 219(2)(d). However, ten years ago in *233Champion v Nation Wide Security Inc,1 we did implicitly adopt it. And regarding the second question, we should now explicitly adopt § 219(2) (d) and apply it to the facts of this case.
Accordingly, I would affirm the Court of Appeals use and recognition of § 219(2)(d) and reverse the trial court’s grant of summary disposition. Summary disposition was improper because a factual question exists concerning whether the person who assaulted plaintiff was aided in committing the tort by his agency relationship with defendant.
I. THIS COURT HAS IMPLICITLY ADOPTED § 219(2)(d)
As a general rule, an employer is not responsible for an employee’s intentional or reckless torts that exceed the scope of employment. Bradley v Stevens, 329 Mich 556, 562; 46 NW2d 382 (1951). But § 219(2)(d) of the Restatement of Agency provides:
(2) A master is not subject to liability for the torts of his servants acting outside the scope of their employment, unless:
(d) the servant purported to act or to speak on behalf of the principal and there was reliance upon apparent authority, or he was aided in accomplishing the tort by the existence of the agency relation. [1 Restatement Agency, 2d, § 219(2)(d).]
The exception essentially holds an employer liable for an employee’s abuse of the authority that the employer granted.2 Our concern in this case surrounds the excep*234tion’s phrase “aided in accomplishing the tort by the existence of the agency relation” and whether this Court has previously adopted § 219(2)(d). Id.
The majority holds that this Court did not adopt § 219(2)(d) in Champion. I agree that the magic words “we adopt the Restatement” do not appear. But I disagree that the Court’s reference to § 219(2) (d) was merely in passing and that its application of the section was limited to the facts of that case.3 Rather, a close reading of Champion suggests that the citation of § 219(2)(d) was part of the Court’s rationale. Also, the citation of § 219(2)(d) was not expressly or implicitly limited to the facts presented in Champion, and its inclusion was designed to give guidance to the bench and bar.
Additionally, the citation of § 219(2)(d) was not just a cursory statement. This Court’s citation of § 219(2)(d) in Champion was in response to one of the defendant’s arguments in that- case. The defendant-employer had asserted that it could not be responsible for its supervisor’s rape of the plaintiff-employee because it never authorized the supervisor to rape the employee. Champion, supra at 712. In direct response, this Court stated, “This construction of agency principles is far too nar*235row.” Id. The reader is then directed to § 219(2)(d) to determine how a court should determine the proper scope of agency principles.
This Court further stated that, when an employer gives a supervisor certain authority over other employees, the employer must take responsibility to remedy any harm caused by the supervisor’s misuse of the authority granted. Id., citing Henson v City of Dundee, 682 F2d 897, 909 (CA 11, 1982). Champion’s citation of the Henson decision is especially noteworthy because Henson includes a discussion of § 219(2) (d):
The common law rules of respondeat superior will not always be appropriate to suit the broad remedial purposes of Title VU[4].... In this case, however, the imposition of liability upon an employer for quid pro quo sexual harassment committed by supervisors appears to be in general agreement with common law principles. See Restatement (Second) of Agency § 219(2) (d) (master is liable for tort of his servant if the servant “was aided in accomplishing the tort by the existence of the agency relation”). [Henson, supra at 910 n 21.]
Given Champion’s direction to readers to refer to the Restatement and Champion’s citation of Henson, I do not believe that § 219(2) (d) was mentioned only in passing.4 5
Nonetheless, the majority seeks, by smoke and mirrors, to hide the fact that this Court appears to have implicitly adopted § 219(2)(d). Framing the issue as *236being whether this Court adopted § 219(2)(d) allows the majority to overrule Champion without the need to show that it was wrongly decided. Moreover, Champion’s validity was not questioned below. But even if this Court has not already adopted § 219(2)(d), we should explicitly adopt it now.
II. THE COURT SHOULD NOW EXPLICITLY ADOPT § 219(2)(d)
Section 219(2)(d) correctly places responsibility on the employer to ensure that any grant of authority it makes to an employee is proper. The employer has the ultimate power to decide whom it will hire. The employer is responsible for determining what authority its employees are allowed. Therefore, it is the employer who should be responsible when its employees abuse the authority the employer gave them and the authority granted enables the employees to cause harm.
But in seeking to shield employers from liability, the majority instead places the burden of preventing an abuse of authority and the corresponding harm on people powerless to prevent it. This case presents a perfect example. Plaintiff was taken to the defendant hospital against her will. She was strapped to a bed. Defendant’s employee then used his employer-given authority to enter plaintiffs room to sexually assault plaintiff. She had no power over who could enter her hospital room, and she could not prevent the assault. The entity with the power to protect plaintiff was the hospital. Yet, the majority leaves plaintiff to bear the full burden for the harm she was powerless to prevent.
However, § 219(2) (d) recognizes that the majority’s approach of placing the burden on the victim is unworkable. It also recognizes that such an action would create a situation where an employer has much less reason to *237monitor its employees’ use of authority.6 Therefore, this Court should explicitly adopt § 219(2)(d) and apply it to the facts of this case. Moreover, the majority’s reasons for not adopting § 219(2)(d) are unpersuasive.
The majority’s main reason for not embracing § 219(2)(d) is that the exception would swallow the rule. According to the majority, this would create “ ‘vicarious liability that knows no borders’ for acts committed by employees that are clearly outside the scope of employment.” Ante at 231 (citation omitted). But this generic rationale misunderstands the scope of § 219(2) (d). It also avoids acknowledging that this Court can adopt a narrow interpretation of § 219(2) (d).
Indeed, the majority seems to accept without explanation that § 219(2)(d) must be broadly construed. This is understandable in light of the majority’s calculated fear that adoption of § 219(2)(d) will open a Pandora’s box. But this rationale ignores the fact that the employer liability that § 219(2)(d) provides for is the tortious use of authority by an employee. Liability is not created by the employer-employee relationship alone. And § 219(2)(d) requires more than mere opportunity *238to commit the tort.7 Bozarth v Harper Creek Bd of Ed, 94 Mich App 351, 355; 288 NW2d 424 (1979). Moreover, the majority’s blanket assertion that adoption of § 219(2)(d) will create “virtual” strict liability ignores the fact that several other courts have interpreted the exception narrowly.
For example, courts have taken several approaches to interpreting the scope of § 219(2)(d). One is to adopt the instrumentality rule which is explained in Costos v Coconut Island Corp, 137 F3d 46 (CA 1, 1998). Another is to adopt a balancing approach as explained by the Supreme Court of Vermont in Doe v Forrest, 176 Vt 476; 853 A2d 48 (2004). Both seek to balance the scope of § 219(2)(d) so as not to impose strict liability based solely on the employer-employee relationship or on the mere opportunity to commit the tort.
With respect to the instrumentality approach, in Costos, supra, the United States Court of Appeals for the First Circuit interpreted the scope of § 219(2)(d). In that case, a hotel manager gained access to a guest’s room and raped the guest. The court found that the owner and corporate manager of the hotel could be held liable for the rape. Costos, supra at 50. The court reasoned:
By virtue of his agency relationship with the defendants, as manager of the inn, [the manager] was entrusted with the keys to the rooms, including [the victim’s] room, at the Bernard House. Because he was the manager of the inn, [he] knew exactly where to find [the victim]. The jury could find that [the manager] had responsibilities to be at the inn or to have others there late at night. In short, because he was the defendants’ agent, [the manger] knew that [the victim] was staying at the Bernard House, he was able to find [the victim’s] room late at night, he had the key to the *239room and used the key to unlock the door, slip into bed beside her as she slept, and rape her. [Id.]
Thus, the instrumentality approach requires that the tort be accomplished by an instrumentality or through conduct associated with the agency status. Accordingly, this interpretation in Costos reads § 219(2)(d) narrowly and balances interests so that employers do not become liable simply because an employer-employee relationship exists. In other words, the instrumentality approach to § 219(2) (d) does not result in strict liability for employers.
With respect to the second approach, the Supreme Court of Vermont in Doe, supra, explored the application of § 219(2)(d) to a sexual assault committed by a police officer while the officer was on duty. In response to the dissent’s contention that unfathomable strict liability would result, the court explained that it was “sensitive to the concern expressed by the trial court that plaintiffs arguments could lead to a rule that makes a principal liable for all intentional torts of an agent in all circumstances.” Doe, supra at 491.
In addressing this concern, the Vermont court, id. at 488, turned to the United States Supreme Court decisions in Burlington Industries, Inc v Ellerth,8 and Faragher v City of Boca Raton, 9 two sexual harassment cases brought under Title VII. In those cases, the United States Supreme Court concerned itself with the last phrase of § 219(2)(d) and rejected a narrow reading of its language.10 Doe, supra at 489-490. As such, while observing that it was not strictly bound by those *240decisions, the Doe court viewed Ellerth and Faragher as strong persuasive authority and helpful to the proper application of § 219(2)(d). Doe, supra at 490.
Accordingly, in following the lead of the United States Supreme Court, the Vermont court in Doe reasoned that it is “important not to adopt too narrow an interpretation of the last clause of § 219(2)(d), but it is equally important not to adopt too broad an interpretation.” Id. at 491 (emphasis added). The court also reasoned that it should give appropriate deference to the policy reasons underlying the United States Supreme Court’s decisions. It decided to apply those policy reasons in the context of an intentional sexual tort committed by a police officer while on duty.
As such, the Vermont court eventually determined that the three considerations noted in Faragher correctly balanced the scope of § 219(2)(d) and, thus, adopted them as its own. The considerations are (1) the opportunity created by the relationship, (2) the powerlessness of the victim to resist the perpetrator and prevent the unwanted contact, and (3) the opportunity to prevent and guard against the conduct. Id. at 491.11
*241According to the Vermont court, when all three factors weigh in favor of the victim, liability may be imposed on the employer under the exception set forth in § 219(2) (d). Thus, it is clear that the Vermont court’s approach does not result in strict liability and also serves to protect those who cannot protect themselves.
*242I agree with the Vermont court that § 219(2)(d) as interpreted by the United States Supreme Court in Faragher, supra, reflects the correct balance between reading § 219(2)(d) too narrowly and reading it too broadly. I would adopt its approach for Michigan law.12 Again, strict liability does not result from this application of § 219(2)(d). Only in those cases where (1) the opportunity created by the relationship, (2) the powerlessness of the victim to resist the perpetrator and prevent the unwanted contact, and (3) the opportunity to prevent and guard against the conduct are properly balanced will a defendant be held vicariously liable. Doe, supra at 491.
Still, the majority prefers to ignore or discount the fact that this Court has the power to adopt an interpretation of § 219(2)(d) that does not cause strict liability. This Court could adopt an interpretation that encompasses its previous statement that
when an employer gives its supervisors certain authority over other employees, it must also accept responsibility to remedy the harm caused by the supervisors’ unlawful exercise of that authority. [Champion, supra at 712.]
In my view, the most disturbing aspect of the majority’s refusal to adopt or apply § 219(2)(d) is that its rationale is based solely on an unproven hypothesis. The majority reasons that adoption of the Restatement could lead to “virtual” strict liability for employers. But this is simply an unproven assertion designed to cause fear. More importantly, the majority does not acknowledge that it can interpret § 219(2)(d) to fairly balance the interests that § 219(2)(d) seeks to protect, just as other courts have done. Instead, it merely states that it refuses to do so because of its fear that strict liability *243would result. The majority is like a farmer holding a can of red paint saying, “I dare not paint my barn because the barnyard will become red.”
Unlike the majority, I would carefully adopt and amend the common law to embrace the reasonable interpretation of § 219(2)(d) expressed in Doe and Faragher. The truth about § 219(2)(d) is that it functions as good public policy and as practical law when interpreted properly. In light of the discussion above, I am unpersuaded by the majority’s rationale that boundless liability will result from a careful adoption of § 219(2)(d). The majority’s rationale misunderstands the scope of § 219(2) (d) and fails to acknowledge this Court’s ability to craft a rule that would properly balance the interests protected under § 219(2)(d). Therefore, had we not adopted it in the past, we should adopt § 219(2) (d) today, thereby placing the burden on the party most capable of preventing loss or injury. Moreover, I would apply that interpretation to this case.
Here, under the approach detailed above, a factual question exists whether his agency relationship assisted the nurse’s aide in committing the tort. Powell’s position in defendant’s emergency room gave him the opportunity to sexually assault a restrained patient. It appears that plaintiff was powerless to prevent the unwanted contact because she was physically bound and was suffering from a manic episode. In general, patients are vulnerable and trust hospital staff and their care. Therefore, I would affirm the result of the decision of the Court of Appeals and remand this case to the trial court for further proceedings.
III. CONCLUSION
I would find that this Court implicitly adopted § 219(2)(d) in Champion. And even if we did not adopt *244§ 219(2)(d) before, we should adopt it now. We should interpret the exception as did the Supreme Court of Vermont in Doe, supra, and apply it to the facts of this case. Consequently, I dissent from the majority opinion.
I would affirm in part and reverse in part the decision of the Court of Appeals, and remand this case to the trial court for further proceedings.
CAVANAGH, J., concurred with KELLY, J.450 Mich 702; 545 NW2d 596 (1996).
The majority claims that § 219(2)(d) is primarily applicable in cases of deceit or misrepresentation. I agree but do not believe that § 219(2)(d) is *234limited to those cases. The comments on § 219(2) (d) state that liability may exist where the servant is able to cause harm because of his or her position as an agent. The comments also provide that the enumeration of situations where § 219(2)(d) applies is not exhaustive, and the section applies where an agent causes physical harm. Therefore, I would not read § 219(2)(d) so narrowly as the majority does.
The majority also argues that the placement of the citation in a footnote should determine its precedential effect. This is a dubious argument, considering that footnotes do sometimes set the state of the law. Perhaps this is unfortunate, but it is true. See, e.g., United States v Carotene Products Co, 304 US 144, 153 n 4; 58 S Ct 778; 82 L Ed 1234 (1938), which laid the groundwork for heightened constitutional scrutiny of laws that discriminate on the basis of race and religion.
Title VII is the section of the Civil Rights Act of 1964 that prohibits employment discrimination. 42 USC 2000e et seq.
The majority argues that Champion was carefully crafted to apply only to quid pro quo sexual harassment cases. Ante at 223-224.1 disagree. Champion was a sexual harassment case. But this Court’s broad statement there that Michigan’s agency principles are in line with § 219(2)(d) cannot fairly be read as an expression to limit § 219(2)(d) to sexual harassment cases.
The majority believes that § 219(2) (d) is unnecessary in light of the existence of other tort remedies. However a review of case law involving negligent hiring, training, and supervising shows that the majority is incorrect. Negligence in hiring requires knowledge on the part of the employer that the employee has criminal tendencies. Hersh v Kentfield Builders, Inc, 385 Mich 410, 412; 189 NW2d 286 (1971). Negligent training is inapplicable here because there is no allegation that the nurse’s aide was improperly trained. Negligent supervising, like negligent hiring and retention, requires knowledge on the part of the employer that special circumstances exist that could establish a duty to protect third persons. Millross v Plum Hollow Golf Club, 429 Mich 178, 196-197; 413 NW2d 17 (1987). This review of the torts listed by the majority shows that none adequately covers a situation where the employer does not have knowledge that an employee may misuse granted authority. Therefore, § 219(2)(d) is needed to protect victims like plaintiff.
The majority’s focus is misplaced. The focus is not on the employment relationship, but on the authority that the employer granted to the employee.
524 US 742; 118 S Ct 2257; 141 L Ed 2d 633 (1998).
524 US 775; 118 S Ct 2275; 141 L Ed 2d 662 (1998).
In Faragher, the Court specifically rejected the proffered reading that the aid-in-accomplishing theory merely refined the apparent authority theory.
The majority asserts that the Vermont Supreme Court in Doe incorrectly relied on these factors from Faragher because Faragher was limited to cases involving a supervisor-employee relationship. See ante at 229. In support of this assertion, the majority relies on the Doe dissent. But I find the Doe court’s response to that dissent persuasive and fitting in this case. The Doe court observed:
In following the United States Supreme Court decisions, we reject the dissent’s claim that the Supreme Court “never intended for its decisions ... to have any influence on the development of common-law agency principles or the application of § 219(2)(d) outside the specific context of Title VII.” ... The Supreme Court applied the Restatement of Agency because it found that “Congress wanted courts to look to agency principles for guidance” in deciding hostile environment sex discrimination cases under Title VII.... Thus, in [Ellerth, supra at 754], the Court noted that it *241was relying on “the general common law of agency.” (Citation omitted). The Court noted that state court decisions could be “instructive,” but they often relied upon federal decisions, id. at 755, and found the Restatement of Agency a useful starting point to find the general common law. Id. It went through the various sections of the Restatement (Second) of Agency and finally centered on § 219(2)(d) as the most useful. It then applied the “aided in the agency relation principle” of § 219(2)(d) to the situation before it. Id. at 760-65.. .. The analysis in [Faragher, supra at 801-802], is similar, and as noted in the text, the Court resolved a dispute over the meaning of the language of § 219(2)(d), holding that the “aided-by-agency-relation principle” was not merely a refinement of apparent authority.
It is, of course, the nature of the common law that every appellate decision represents the development of the common law, and nothing in the Supreme Court decisions suggests they are not an integral part of that process. Indeed, the resolution of the dispute over the meaning of § 219(2)(d) in Faragher is exactly the kind of decision that best defines and develops the common law. No common-law court engaged in this process, and certainly not the highest court of this country, would expect that a common-law decision on one set of facts would have no influence on future decisions applying the same legal principle to a different factual scenario. [Doe, supra at 490 n 3.]
In any event, the reasons underlying the Faragher Court’s use of these factors applies with equal force here even though there is not a supervisor-employee relationship. Therefore, extension of those factors is logical and appropriate. For example, just like a supervisor (Faragher) and a police office (Doe), a hospital aide has unique access to a patient who is depending on the aide for care. The patient is often defenseless. Imposing liability on the hospital may prevent recurrence of an assault by creating an incentive for vigilance. Accordingly, I disagree with the majority’s assertion that I am ignoring the specific context in which Faragher announced and applied the factors noted above. Rather, the context and underlying policy considerations are largely the same.
I do not reject out of hand the instrumentality approach adopted by the Costos court, but find the balancing approach in Doe more compelling.