ROSENBAUM, Circuit Judge:

Upon receiving their master's degrees, certifications, and licenses, Plaintiff-Appellant student registered nurse anesthetists are legally able to put people to sleep. We have heard, though never ourselves experienced, that some legal opinions can do the same thing. We are hopeful that this one will not.1

Plaintiffs in this case include twenty-five former student registered nurse anesthetists ("SRNAs" or "Students")2 who attended a master's degree program at Wolford College, LLC, with the goal of becoming certified registered nurse anesthetists ("CRNAs").3 During the course of their study, the Students participated in a clinical curriculum, which, under Florida law, was a prerequisite to obtaining their master's degrees.

Through this legal action, the Students sought to recover unpaid wages and overtime under the Fair Labor Standards Act, 29 U.S.C. § 201, et seq. ("FLSA"), for their clinical hours. After considering the six factors that the Department of Labor identified in guidance that, in turn, does no more than reduce the specific facts of Walling v. Portland Terminal Co., 330 U.S. 148, 67 S.Ct. 639, 91 L.Ed. 809 (1947), to a test, the district court determined that the SRNAs were not "employees" of Defendants and entered summary judgment for Defendants.

But, with all due respect to the Department of Labor, it has no more expertise in construing a Supreme Court case than does the Judiciary. Portland Terminal is nearly seven decades old and, in our view, addresses a very different factual situation involving a seven-or-eight-day, railroad-yard-brakeman training program offered by a specific company for the purpose of creating a labor pool for its own future use. This case, however, concerns a universal clinical-placement requirement necessary to obtain a generally applicable advanced academic degree and professional certification and licensure in the field.

So, while we follow Portland Terminal's "primary beneficiary" test here, we do not believe that measuring the facts in this case by a strict comparison to those in Portland Terminal allows us to identify the primary beneficiary of a modern-day internship for academic credit and professional certification. As a result, we now adopt an application of Portland Terminal's "primary beneficiary" test specifically tailored to account for the unique qualities of the type of internship at issue in this case. To allow the district court to apply this test in the first instance and, if the district court desires, to give the parties an opportunity to further develop the record to address the components of the test, we remand this case for further proceedings consistent with this opinion.

I.

A.

Florida tightly regulates the practice of nurse anesthesia to protect patients, since anesthesia delivery can carry a high risk. Performing the duties of a CRNA in Florida without a proper license or knowingly employing an unlicensed person to engage in CRNA duties constitutes a felony. See Fla. Stat. § 464.016. To obtain a CRNA license under Florida law, among other requirements, a person must graduate from an accredited program and be certified by the National Commission on Certification of Anesthesiologist Assistants. See Fla. Stat. §§ 458.3475, 459.023.

B.

Defendant Wolford College is a for-profit college that is wholly owned by Defendant Lynda Waterhouse (Wolford College's chief financial officer and secretary-treasurer) and several anesthesiologists who also have an ownership interest in Defendant Collier Anesthesia, P.A., a Florida corporation that provides anesthesia services. In addition to her duties at Wolford College, Waterhouse serves as the executive director of Collier Anesthesia. Defendant Dr. Thomas L. Cook is the president and a part-owner of Collier.

Wolford College offers one of 113 accredited CRNA programs in the country, providing a 28-month curriculum that culminates in a Master of Science degree in Nurse Anesthesia. While classroom learning dominates the first three semesters of the master's program, the last four semesters consist mainly of clinical experience — a requirement that Florida law, the Council on Accreditation for Nurse Anesthesia Educational Programs,4 and the National Board of Certification and Recertification of Nurse Anesthetists all mandate.

Under the Council on Accreditation's standards, accredited schools must require students to participate in a minimum of 550 clinical cases in a variety of surgical procedures. This requirement is designed to ensure that when a student graduates and becomes licensed, she will be able to safely and competently monitor the status of her patients without another licensed professional in the room. Among other tasks that must be mastered during the clinical phase of training, SRNAs must learn to complete preoperative forms for patients; set up anesthesia equipment; draw proper medications; monitor patients through the induction, maintenance, and emergence phases of anesthesia; stock and re-stock anesthesia carts; prepare rooms for use; clean equipment; and serve while "on call."

In Wolford College's clinical phase of education, each course has an instructor and a syllabus, and the school requires daily evaluations that must be completed by both the student and the CRNA or anesthesiologist who supervises the student. Every day, the supervising CRNA or anesthesiologist must grade the student in several areas, depending on the particular course. For example, in some of the courses, the supervisor must evaluate the SRNA every day in nine different categories, including anesthesia cart, anesthesia machine, airway set up, patient assessment, record keeping, induction, maintenance, emergence, and interpersonal behavior. In addition, on the same form, the supervising CRNA or physician prepares brief comments regarding the day's events. The clinical courses also require end-of-semester self-evaluations prepared by the student and summative semester evaluations completed by the clinical instructor or coordinator.

In order to sit for the Board examination, students must graduate from an accredited nurse anesthesia program. For each class graduating in the years 2009 through 2013, between 96% and 100% of all Wolford graduates passed their Board certifications.

C.

In this case, the Students obtained some, if not all, of their clinical education at facilities where Collier Anesthesia practices anesthesiology. But the Students viewed their clinical efforts as more than just education; they filed suit alleging that they served as "employees" of Defendants for purposes of the FLSA and that Defendants unlawfully failed to compensate them with wages and overtime pay. During the proceedings, Defendants moved for summary judgment, and the Students filed a competing motion for partial summary judgment.

In support of their motion for summary judgment, Defendants submitted evidence that all Students were notified at the start of their education that they were not guaranteed employment with Collier upon graduation, and, in fact, none of the Students involved in this case ever worked for Collier after they obtained their master's degrees. The Students also agreed at the beginning of their educations (by signing Wolford's Handbook) that although they would be undertaking a clinical program, they would not become employed as nurse anesthetists through their participation.

In addition, Defendants' evidence showed that when the Students were at a clinical location, they were identified as SRNAs, and they were required to wear scrubs with the Wolford College logo. And, although while participating in the clinical program, in some instances, the Students, without direct supervision from an anesthesiologist or a CRNA, readied rooms, stocked carts, prepared preoperative forms, and performed other functions, a licensed anesthesiologist or CRNA was required to review the SRNAs' work as part of the SRNAs' daily evaluations.

For their part, the Students claimed that Collier benefited financially by using their services as SRNAs in place of licensed CRNAs. Although Wolford's curriculum contemplated that SRNAs would work in the clinical program for 40 hours per week, the Students submitted evidence that Collier routinely scheduled SRNAs in excess of 40 hours per week. They further presented testimony that they were scheduled to perform work at Collier-staffed facilities 365 days per year, including weekends, holidays, and the days in between semesters. And, although Wolford and Collier represented to the Students that their shifts would be eight hours long, the Students stated that they were required to arrive at facilities in advance of their scheduled shifts. Consequently, the Students indicated, an eight-hour shift actually required SRNAs to work for a minimum of 8.75 to 10 hours per day.

In addition to their own statements, the Students relied upon the testimony of Barbara Rose, a former Collier employee who the Students contend was responsible for CRNA and SRNA scheduling. From April 2010 through April 2012, Rose prepared the SRNA monthly and daily schedules for assignments at Collier's clinical sites.

Collier usually scheduled SRNAs for five shifts per week. Rose indicated that in preparing the daily schedule, she strived to use SRNAs to reduce the number of Collier CRNAs needed for the schedule. According to Rose, Collier removed CRNAs from the daily schedule after the fact in favor of SRNAs and, in Rose's opinion, if the SRNAs had not been scheduled, Collier would have needed CRNAs to cover shifts. The Students pointed to Rose's testimony to support their theory that SRNAs at Collier displaced CRNAs on the schedule and that Collier affirmatively tried to use as few licensed nurses as possible. Based on this evidence, the Students argued that the displacement of CRNAs allowed Collier to save money in running its practice.

Rose, however, admitted that she lacked important first-hand knowledge about Collier's scheduling practices. In particular, Rose did not know about Collier's payrolls, the number of patient cases on which Collier worked, the number of CRNAs on Collier's payroll, the number of anesthesiologists on its payroll, other personal knowledge necessary to support the Students' claims of displacement of regular workers, or the economic benefit to Collier from the Students' services.

To counteract Rose's testimony, Defendants presented the testimony of Keri Ortega, who served as the Assistant Program Director and Associate Director of Graduate Education at Wolford College during the period in question. Ortega was one of two Wolford employees who were primarily responsible for scheduling the students in the clinical program. She attested that, typically, Rose sent her a proposed monthly schedule. Ortega, Wolford Program Director Dr. Lauren Corder, or Wolford Dean Dr. John Nolan then reviewed the schedule. According to Ortega, the schedule is a "living document" — it constantly changes for a number of reasons, including, among others, cases are removed from and added to the schedule, patients refuse to allow student participation, and cases are changed from one procedure to another due to illness, equipment problems, and other last-minute circumstances. So Ortega viewed the schedules that Rose prepared as merely initial schedules which were subject to review and significant change.

Defense expert Dr. Daniel Janyja, an anesthesiologist at Collier, also provided evidence that contradicted Rose's testimony. He echoed Ortega, stating that the scheduling of CRNAs and SRNAs in operating rooms is a highly complex and fluid process that changes up to the last minute on a daily basis. According to Dr. Janyja, Collier was capable of meeting its patient safety and legal obligations with existing licensed personnel, without using the Students and without incurring additional personnel costs. Dr. Janyja opined that Wolford students did not displace CRNAs. To the contrary, Dr. Janyja viewed the Students as more of a burden than a benefit to Collier because, among other reasons, the learning process impedes the actual delivery of anesthesia.

Defendants also presented evidence supporting their contention that it is sometimes difficult to place students in a clinical environment. Certain surgeons and hospital locations refuse to allow students in the operating room. When a surgeon indicates a preference not to use students, Collier honors that preference, and Wolford does not place any students in those situations. Patients also sometimes decline student participation in their case.

And, from the perspective of a CRNA, allowing a student to participate in the administration of anesthesia under that individual's license creates an added stress that would not otherwise be present. Moreover, student participation can slow down the administration of anesthesia because the CRNA may need to take time to respond to questions posed by a student. Or a student may attempt a procedure, fail, and require the CRNA (or anesthesiologist) to complete the procedure. And, as previously described, Wolford also requires the anesthesiologist or CRNA supervising students to complete paperwork pertaining to each student's presence, including daily evaluations. This paperwork detracts time from the CRNA or anesthesiologist's day.

The Students responded to this evidence by pointing out that, under what is known as the "Revised Teaching Rule,"5 Collier could receive reimbursement for student activities. According to the Students, Collier was able to use one CRNA to obtain 100% of the CRNA fee for two cases at the same time with one student assigned to each case. And, following the rule change, Wolford and Collier decided to institute a two-to-one SRNA-to-CRNA supervision ratio. The Students argued that supervising SRNAs at a ratio of two to one made teaching "more advantageous" because Collier was reimbursed for two students' concurrent participation where only one CRNA provided supervision.

Collier acknowledged that it used the CRNA Teaching Rule and billed Medicare for some of its patients' procedures using the Revised Teaching Rule. According to Collier, however, during this time, the number of CRNAs on its payroll and the amount of Collier's payroll costs remained "substantially unchanged" despite the fluctuations in the number of Wolford students that it used in the clinical program from semester to semester. In addition, it asserted that at all times that billing occurred under this new rule, Collier anesthesiologists continued to supervise CRNAs and SRNAs without charge to Medicare. Collier's Executive Director, Waterhouse, also refuted the Students' claim that Collier saved money by using the Students in place of the CRNAs. According to Waterhouse, Collier did not save on labor costs, but rather, it lost money as a result of using the SRNAs because of the time spent training the Students. Besides these costs, Defendants presented evidence to show that Collier paid a "clinical fee" of $1,500 per student to Wolford to underwrite Wolford's costs and to assist Wolford in remaining eligible for federal student loan funding.

After considering the facts presented by each party, the district court granted summary judgment in favor of Defendants, finding that the Students were not employees under the FLSA, so they were not entitled to a minimum wage or overtime pay.

On appeal, the Students contend that, in rendering its decision, the district court improperly declined to follow the six-factor test promulgated by the Department of Labor's Wage and Hour Division. The Students also assert that genuine issues of material fact exist in the case, which preclude the entry of summary judgment in favor of Defendants.

II.

Because a determination of an individual's employment status under the FLSA is a question of law, we review de novo the district court's finding that no employment relationship existed between the Students and Defendants. Scantland v. Jeffry Knight, Inc., 721 F.3d 1308, 1310 (11th Cir.2013).

We also review de novo an order granting summary judgment and apply the same legal standards that control the district court. Id. at 1310. Under Rule 56(a), Fed.R.Civ.P., summary judgment is proper where "the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law." We view all evidence and draw all justifiable inferences in the nonmoving party's favor. Scantland, 721 F.3d at 1310.

III.

A.

Congress enacted the FLSA "to aid the unprotected, unorganized and lowest paid of the nation's working population; that is, those employees who lacked sufficient bargaining power to secure for themselves a minimum subsistence wage." Brooklyn Sav. Bank v. O'Neil, 324 U.S. 697, 707 n. 18, 65 S.Ct. 895, 902 n. 18, 89 L.Ed. 1296 (1945). In addition, Congress sought "to lessen, so far as seemed then practicable, the distribution in commerce of goods produced under subnormal labor conditions." Rutherford Food Corp. v. McComb, 331 U.S. 722, 727, 67 S.Ct. 1473, 1475, 91 L.Ed. 1772 (1947).

The protections the FLSA affords, however, extend to "employees" only. As a result, only individuals falling within the Act's definition of "employee" are entitled to minimum wages and overtime.

The tricky part arises in determining who falls within the FLSA's definition. As other courts have observed,6 the FLSA's definitions as they relate to who qualifies as an "employee" are not precise. The statute defines an "employee" as "any individual employed by an employer," and an "employer," in turn, includes "any person acting directly or indirectly in the interest of an employer in relation to an employee." 29 U.S.C. §§ 203(d) and (e)(1). The Act also provides that the term "employ" means "to suffer or permit to work." Id. at § 203(g). Congress intended for these broad definitions to be "`comprehensive enough' to include `working relationships, which prior to this Act, were not deemed to fall within an employer-employee category.'" Scantland, 721 F.3d at 1311 (citing Rutherford Food, 331 U.S. at 729, 67 S.Ct. at 1476 (quoting Portland Terminal, 330 U.S. at 150-51, 67 S.Ct. at 640)).

Nevertheless, the terms "employee" and "employer" "cannot be interpreted so as to make a person whose work serves only his own interest an employee of another person who gives him aid and instruction." Portland Terminal, 330 U.S. at 152, 67 S.Ct. at 641. As the Supreme Court has cautioned, the FLSA was "not intended to stamp all persons as employees who, without any express or implied compensation agreement, might work for their own advantage on the promises of another. Otherwise, all students would be employees of the school or college they attended, and as such entitled to receive minimum wages." Id.

B.

In Portland Terminal, the seminal case involving whether trainees are "employees" for purposes of the FLSA, the defendant railroad company offered a practical-training course for prospective yard brakemen. Id. at 149, 67 S.Ct. at 640. While participants were not guaranteed a job upon completion of the course, they were required to successfully finish the course to be eligible to serve as brakemen for the railroad. Id. at 149-50, 67 S.Ct. at 640. On average, the course lasted seven or eight days. Id. at 149, 67 S.Ct. at 640. During training, a yard crew instructed and supervised the trainees, gradually allowing them to perform actual work under close scrutiny. Id. The trainees' work did not displace any regular employees, who continued to do most of the work themselves. Id. at 150, 67 S.Ct. at 640. Nor did the trainees' work expedite company business. Id. In fact, at times, it impeded it. Id. In holding that the trainees were not "employees" for purposes of the FLSA, the Supreme Court reasoned,

Id. at 152-53, 67 S.Ct. at 641. Ultimately, the Supreme Court explained, "The Fair Labor Standards Act was not intended to penalize railroads for providing, free of charge, the same kind of instruction at a place and in a manner which would most greatly benefit the trainees." Id. at 153, 67 S.Ct. at 641 (emphasis added).

C.

The Department of Labor ("DOL") refers without attribution to Portland Terminal in its Field Operations Handbook's guidance on identifying whether a trainee or a student is an "employee" under the FLSA. Specifically, the Handbook states, "The Supreme Court has held that the words `to suffer or permit to work, as used in the FLSA to define `employ', do not make all persons employees who, without any express or implied compensation agreement, may work for their own advantages on the premises of another," See Wage & Hour Div., U.S. Dep't of Labor, Field Operations Handbook ch.10b11 (Oct. 20, 1993), http://www.dol.gov/whd/FOH/FOH_Ch10.pdf (last visited Aug. 28, 2015). This statement paraphrases Portland Terminal. See Portland Terminal, 330 U.S. at 152, 67 S.Ct. at 641 ("The definition `suffer or permit to work' was obviously not intended to stamp all persons as employees who, without any express or implied compensation agreement, might work for their own advantage on the premises of another.").

After referring to the content of Portland Terminal, the Handbook then opines that "[i]f all of the following criteria are met, the trainees or students are not employees within the meaning of the FLSA:"

Id. (emphasis in original). The Students assert that we should defer to this guidance, but we respectfully disagree.

Just as it is clear that the Handbook refers to Portland Terminal in its introduction to the six factors it sets forth, it is equally plain from reviewing the six factors that the Handbook derived them by simply reducing the facts of Portland Terminal to a test. This test is not a regulation, and it did not arise as a result of rule-making or an adversarial process. At most, it is entitled to Skidmore deference, meaning that the deference it is due is "proportional to its `power to persuade.'" See United States v. Mead Corp., 533 U.S. 218, 235, 121 S.Ct. 2164, 2175-76, 150 L.Ed.2d 292 (2001) (citing Skidmore v. Swift & Co., 323 U.S. 134, 140, 65 S.Ct. 161, 164, 89 L.Ed. 124 (1944)).

We do not defer to this test because, with all due respect to the DOL and the important work that it does, we do not find it persuasive. First, "an agency has no special competence or role in interpreting a judicial decision." Glatt v. Fox Searchlight Pictures, Inc., 791 F.3d 376, 383 (2d Cir.2015) (citation omitted). Second, as the Second Circuit has observed, the test "attempts to fit Portland Terminal's particular facts to all workplaces, and ... is too rigid...." Id. Third, while some circuits have given some deference to the test, no circuit has adopted it wholesale and has deferred to the test's requirement that "all" factors be met for a trainee not to qualify as an "employee" under the FLSA. In short, we prefer to take our guidance on this issue directly from Portland Terminal and not from the DOL's interpretation of it.

D.

We therefore return to Portland Terminal. Since Portland Terminal, courts reviewing cases involving students and trainees have focused on the Supreme Court's language describing the program at issue in that case as having "most greatly benefit[ed]" the trainees. As a result, these courts have, for the most part, concentrated on evaluating the "primary beneficiary" of the training or school program to determine whether participants constituted "employees" under the FLSA,7 generally concluding that such an approach reveals the "economic reality" of the situation.8 In doing so, they have considered the entirety of the circumstances, balancing a variety of factors9 that often entail comparing the facts of the case to the facts in Portland Terminal or to the six factors that the DOL sets forth in its Handbook. As we have explained, both forms of comparison are effectively the same.

E.

But most recently, in Glatt, 791 F.3d at 384, the Second Circuit has reflected on the limitations of comparing the characteristics of the modern internship to the specific facts at issue in Portland Terminal. As the Second Circuit observed, Portland Terminal is now 68 years old. Id. The facts of that case do not necessarily "reflect[] the role of internships in today's economy...." See id. (referring to the DOL's Handbook guidance).

We add to these points the significant fact that the training involved in Portland Terminal was not a universal requirement for a particular type of educational degree or for professional certification or professional licensure in the field. Instead, the Portland Terminal training was offered by a company for its own, specific purposes, to create a ready labor pool for itself. So trying to evaluate the program at issue here by comparing it to all of the facts from Portland Terminal that were relevant and helpful to assessing the training program at issue in that case, is like trying to use a fork to eat soup. Like the fork and the spoon, the training at issue in Portland Terminal and in the case under review have similarities and may be in the same general category (eating utensils and training programs). But comparison to the facts from Portland Terminal alone can cover the gamut of relevant considerations in a case like the one before us no better than a fork can do a spoon's job in ladling soup.10

Longer-term, intensive modern internships that are required to obtain academic degrees and professional certification and licensure in a field are just too different from the short training class offered by the railroad in Portland Terminal for the purpose of creating its own labor pool. As exemplified by the facts of the pending case, modern internships can play an important — indeed critical — role in preparing students for their chosen careers. Imagine if a CRNA could report to work on her first day and be allowed unsupervised to conduct the induction, maintenance, and emergence phases of anesthesia administration, having only ever read about or watched someone else perform them. The potential danger and discomfort to the patient under such circumstances is self-evident and startling. So we need anesthesiologists and CRNAs who are willing to teach SRNAs their trade through internships.

Yet taking on the responsibility of supervising and teaching SRNAs is a heavy one with serious potential costs. We cannot realistically expect anesthesiology practices to expose themselves to these costs by providing students with the opportunity to participate in 550 cases each, without receiving some type of benefit from the arrangement. See Am. Airlines, 686 F.2d at 272 ("if attendance were solely for the trainee's benefit, the company would not conduct the [training] except as a matter of altruism or public pro bono").

But the mere fact that an anesthesiology practice obtains benefits from offering SRNAs internships cannot, standing alone, render the student interns "employees" for purposes of the FLSA. See, e.g., Solis, 642 F.3d at 530-31 (though the school derived benefits from students' work at its facilities, the value of the benefits to the students from the work arrangement out-weighed the benefits to the school, so the students were not "employees"). Indeed, there is nothing inherently wrong with an employer's benefiting from an internship that also plainly benefits the interns.

Nevertheless, we recognize the potential for some employers to maximize their benefits at the unfair expense and abuse of student interns. And that is a problem.

So our dilemma arises in determining how to discern the primary beneficiary in a relationship where both the intern and the employer may obtain significant benefits. We think that the best way to do this is to focus on the benefits to the student while still considering whether the manner in which the employer implements the internship program takes unfair advantage of or is otherwise abusive towards the student. This orientation allows for student internships to accomplish their important goals but still accounts for congressional concerns in enacting the FLSA.

We also believe that the Second Circuit's articulation of "a non-exhaustive set of considerations" for evaluation in determining the "primary beneficiary" in cases involving modern internships goes far towards fulfilling this function. In particular, the Second Circuit has identified the following factors:

Glatt, 791 F.3d at 384. Under the Second Circuit's approach, "[n]o one factor is dispositive and every factor need not point in the same direction for the court to conclude that the intern is not an employee...." Id. Rather, courts must engage in a "weighing and balancing [of] all of the circumstances," including, where appropriate, other considerations not expressed in the seven factors. Id. The Second Circuit has described this approach as "flexible" and "faithful to Portland Terminal," reasoning that "[n]othing in the Supreme Court's decision suggests that any particular fact was essential to its conclusion or that the facts on which it relied would have the same relevance in every workplace." Id. at 384-85.

We agree with the Second Circuit's reasoning and its interpretation of Portland Terminal. The factors that the Second Circuit has identified effectively tweak the Supreme Court's considerations in evaluating the training program in Portland Terminal to make them applicable to modern-day internships like the type at issue here.

In many ways, the Glatt factors involve consideration of the same or similar facts to those that the Supreme Court found important in Portland Terminal and that the DOL Handbook guidance deemed relevant factors for consideration. Indeed, factors 2, 3, and 5 are more detailed expressions of Portland Terminal's concern that the training be similar to that available in a vocational or other educational environment. Likewise, factors 2 through 6 reflect Portland Terminal's attention to the benefit to the intern. In addition, factors 2 and 6 relate directly to Portland Terminal's consideration of whether the intern displaces regular employees and whether the intern works under the close supervision of existing employees. Finally, factors 1 and 7 are essentially the same as Portland Terminal's considerations that the intern and employer both understand that the intern will not receive wages and that the intern is not entitled to a job upon completion of the internship, respectively.

Only Portland Terminal's reference to the railroad's receipt of "no `immediate advantage' from any work done by the trainees" is not accounted for by the Glatt factors. But the training in Portland Terminal was so different from a modern internship for academic, certification, and licensure purposes that we do not see how this particular consideration sheds light on the primary-beneficiary analysis here. In Portland Terminal, despite not receiving an "immediate advantage" from the training program, the railroad had a significant economic incentive to offer the training because it needed a ready pool of qualified brakemen from which it could hire. In the absence of the training, there may well not have been any. If the railroad had also obtained a direct and immediate financial or competitive advantage from providing a training program that it was going to have to offer for its own business reasons regardless of whether it received a direct advantage, that could have served as an indication that the railroad was taking unfair advantage of the situation.

But, as we have explained, the modern internship as a requirement for academic credit and professional certification and licensure is very different. For starters, the students seeking the internships — as opposed to a particular company's business requirements — drive the need for the internships to exist. Second, licensure and certification laws provide evidence that we as a society have decided that clinical internships are necessary and important. Third, we find it difficult to conceive that anesthesiology practices would be willing to take on the risks, costs, and detriments of teaching students in a clinical environment for extended periods (four semesters, for example) without receiving some benefit for their troubles. As we have further noted, though, the mere fact that an employer obtains a benefit from providing a clinical internship does not mean that the employer is the "primary beneficiary" of the relationship. Therefore, we cannot see how consideration of whether the employer gains an "immediate advantage" from an internship, in and of itself, brings us any closer to resolving who the primary beneficiary of the relationship is.

Instead, we focus on the Glatt factors. In order to allow the district court to apply these factors in the first instance and, if it desires, to permit the parties to supplement the record, we remand this case to the district court. But first we provide some guidance on applying some of the factors.11

The fourth factor focuses on the extent to which the internship accommodates the intern's academic commitment by corresponding to the academic calendar. In a case like this one, where the clinical training and the academic commitment are one and the same, this consideration must account for whether a legitimate reason exists for clinical training to occur on days when school is out of session.

As for the fifth factor — the extent to which the internship's duration is limited to the period in which the internship provides the intern with beneficial learning — this consideration must recognize the goals of the internship and determine whether the duration of the internship is necessary to accomplish them. In making this evaluation, the court should keep in mind that designing an internship is not an exact science. We cannot expect that the length of the internship will always match up perfectly with the skills to be taught and the experience to be gained through the program. An internship that is longer than absolutely necessary to accomplish the educational and experiential goals of the program does not necessarily weigh in favor of a determination that the intern is an "employee." Instead, the court should consider whether the duration of the internship is grossly excessive in comparison to the period of beneficial learning.

As part of this consideration, the court should also evaluate the extent to which the nature of the training requires the daily schedule that the intern must endure. In this case, graduation, certification, and licensure requirements all demanded that students participate in at least 550 cases involving a variety of procedures. Again, we imagine that it would be difficult, if not impossible, to plan the scheduling of SRNAs for precisely 550 different procedures over four semesters, particularly in view of the constantly changing nature of the medical schedule. Nor do we think that the law requires such precision. We also note that the SRNAs' clinical work was required to extend for four semesters, even if the Students finished 550 cases in a shorter period. As a result, it does not seem to us that the four-semester duration of the program would have been excessive, no matter how many cases the students completed during that time. But if the reason that the SRNAs completed well in excess of 550 cases during their four clinical semesters was because they were made to work grossly excessive hours, that would be an indication that the employer may have unfairly taken advantage of or otherwise abused the SRNAs and that they should be regarded as "employees" under the FLSA.

The sixth factor evaluates the extent to which the intern's work complements, rather than displaces, the work of paid employees while providing significant educational benefits to the intern. This case involves a unique consideration on this factor. The Students assert that CRNAs each worked fewer hours than they otherwise would have, in the absence of the SRNAs, meaning that the SRNAs displaced CRNA hours. For support, they point to the Revised Teaching Rule, which allowed Collier to be reimbursed by Medicare for providing anesthesia in two rooms while having to pay only a single CRNA — something that Collier could not have done if the SRNAs were not there.

We do not opine on whether, in fact, CRNAs worked fewer hours as a result of the SRNAs' presence, under the Revised Teaching Rule. But if they did, we do not think that such a fact, in and of itself, would resolve which party this factor favors. The analysis under this factor must also account for the existence of a Medicare rule that contemplates the use of two SRNAs to assist one CRNA in two rooms simultaneously. A Medicare rule obviously cannot inform whether a SRNA is an "employee" under the FLSA. Nevertheless, the rule's existence and endorsement of the staffing of two patient rooms with one CRNA and two SRNAs suggests that, at least from an anesthesia-administration point of view, there was nothing unsafe or wrong with Collier's scheduling of two SRNAs to be overseen by a single CRNA. Under these circumstances, therefore, it would not be appropriate to consider Collier's use of the Rule as evidence that Collier unfairly took advantage of the SRNAs when it scheduled two SRNAs to be supervised by a single CRNA. Of course, to the extent that CRNA hours may have been displaced by SRNA hours for reasons other than the Revised Teaching Rule, the court should evaluate those circumstances on their own merit.

In applying the factors to ascertain the primary beneficiary of an internship relationship, we caution that the proper resolution of a case may not necessarily be an all-or-nothing determination. That is, we can envision a scenario where a portion of the student's efforts constitute a bona fide internship that primarily benefits the student, but the employer also takes unfair advantage of the student's need to complete the internship by making continuation of the internship implicitly or explicitly contingent on the student's performance of tasks or his working of hours well beyond the bounds of what could fairly be expected to be a part of the internship.12 For example, in the context of an internship required for an academic degree and professional licensure and certification in a medical field, consider an employer who requires an intern to paint the employer's house in order for the student to complete an internship of which the student was otherwise the primary beneficiary. Under those circumstances, the student would not constitute an "employee" for work performed within the legitimate confines of the internship but could qualify as an "employee" for all hours expended in painting the house, a task so far beyond the pale of the contemplated internship that it clearly did not serve to further the goals of the internship.

Finally, we do not take a position at this time regarding whether the Students in this case were "employees" for purposes of the FLSA.

IV.

With these factors in mind, we vacate the district court's entry of summary judgment for Defendants and remand for further proceedings consistent with this opinion.

VACATED AND REMANDED.