OPINION

STEPHEN J. MURPHY, III, District Judge.

Acting on a tip from a concerned citizen, the Wage and Hour Division of the U.S. Department of Labor commenced an investigation into potential child labor violations committed by Laurelbrook Sanitarium and School, Inc. ("Laurelbrook"). After concluding that Laurelbrook had violated the child labor provisions of the Fair Labor Standards Act of 1938, 29 U.S.C. §§ 201-219 ("FLSA" or "Act"), Department of Labor Secretary Hilda Solis ("Secretary") filed suit in federal court seeking prospective injunctive relief against future violations. After a seven-day bench trial, the district court denied the Secretary's request for a permanent injunction on the ground that Laurelbrook students are not "employees" for purposes of the FLSA, thus rendering the Act's prohibitions on child labor inapplicable to Laurelbrook's operations. We AFFIRM.

I

The district court made findings of fact, summarized below, to which we must defer unless clearly erroneous. Fed.R.Civ.P. 52(a)(6). Facts not specially found by the district court, but included below for completeness, are undisputed.

Founded in 1950 by a group of Seventh-Day Adventists, Laurelbrook is a non-profit corporation located in Dayton, Tennessee. Laurelbrook follows the philosophy and teachings of the Seventh-Day Adventist Church and its founder, Ellen G. White, which include the view that children are to receive an education with a practical training component.

In conformity with its beliefs, Laurelbrook operates a boarding school for students in grades nine through twelve, an elementary school for children of staff members, and a 50-bed intermediate-care nursing home that assists in the students' practical training (the Sanitarium). The school has been approved and accredited by the Tennessee Department of Education since the 1970s. The State of Tennessee accredits certain private schools through independent authorized accrediting agencies. The E.A. Sutherland Education Association (EASEA) is one such agency, whose purpose is to consider and adjudicate requests for accreditation from self-supporting (as opposed to denominational) schools, like Laurelbrook, which are operated by members of the Seventh-Day Adventist Church. Laurelbrook is currently accredited through EASEA.

Students in Laurelbrook's boarding school learn in both academic and practical settings, spending four hours of each school day in the classroom and four hours learning practical skills. The two aspects of the education are integrated, and all teachers instruct in both settings. Laurelbrook's stated mission is the "education of young people, providing a balanced program of spiritual, academic and vocational training to high school students (grades nine through twelve) with a goal of reproducing the character of God and preparing for His service." Students learn practical skills, in part, so they can later serve as missionaries in foreign lands. Boarding students keep busy with "wholesome activities" that teach them practical skills about "work, responsibility, [and] the dignity of manual labor," and that contribute to maintaining Laurelbrook's operations.

Laurelbrook offers the following vocational courses: Agriculture, Building Arts, Grounds Management, Mechanical Arts, Office Procedures, Plant Services, Water Services, Certified Nursing Assistant ("CNA"), Child Development, Environmental Services, and Food Service. The Sanitarium is an integral part of Laurelbrook's vocational training program. As part of their training, students are assigned to the Sanitarium's kitchen and housekeeping departments. Students sixteen and older may participate in the CNA program, which is approved by the State of Tennessee licensing authority. Students who receive their CNA certification may then be assigned to the Sanitarium to provide medical assistance to patients. Students assist patients in relation to the students' training and in line with Laurelbrook's guiding philosophy of education. Laurelbrook receives Medicaid funding for the care it provides at the Sanitarium.

The Sanitarium is staffed such that if the students were not training there, staff members could continue to provide the same patient services. But since the Sanitarium is integral to the education the school provides, Laurelbrook would not operate the Sanitarium if the school did not exist. In other words, the Sanitarium's sole purpose is to serve as a training vehicle for students. Therefore, the district court reasoned, students do not displace adult workers or other employees who might be willing to work at the Sanitarium.

Many of Laurelbrook's practical training courses (16 of 25) are approved by the Tennessee Department of Education for transfer credit. Several others are approved as "special courses," meaning a transferee school can accept the courses at its discretion. Students learn to use tools associated with specific trades, and the learning experience is similar to that received in vocational training courses at public schools. Adult staff members adequately supervise students and provide reasonable safeguards to protect them from hazardous activities. Laurelbrook has performed reasonably well in ensuring student safety.

Students do not receive wages for duties they perform. They are not entitled to a job with Laurelbrook upon graduation, and are expected to move on after graduation.

Laurelbrook provides "important tangible benefits and intangible training" and its students "reap great benefits" from the training and education provided. Any benefits Laurelbrook derives from its students are "secondary to its religious mission" of providing academic and practical training. Therefore, the district court reasoned, any such benefits are "much less" than those received by the students.

Laurelbrook does not compete for labor. Its vocational program compares favorably with programs operated by public high schools in the area. But because Laurelbrook operates a boarding school, it cannot compare perfectly with a public school where students are only in school for part of the day. Laurelbrook is responsible for its boarding students at all times, and its efforts to keep the students gainfully occupied are in keeping with its religious mission to teach students moral character.

In February 2007, the Secretary brought this action to enjoin future violations of 29 U.S.C. §§ 212(a), 212(c), and 215(a)(4), as well as Child Labor Regulation 3, 29 C.F.R. §§ 570.31-.37. The district court conducted a seven-day bench trial between August 19, 2008 and April 6, 2009, and subsequently denied the Secretary's request for a permanent injunction in a written order. The basis for the decision was that Laurelbrook students were not "employees" under the FLSA, thus rendering the Act inapplicable to Laurelbrook's entire operation. To reach this result, the district court considered the benefits each party received from the work performed, and concluded that the primary benefit of the work students performed at Laurelbrook ran to themselves, not Laurelbrook.

The Secretary timely appealed.

II

The issue before us is whether the district court erred in concluding that students at Laurelbrook are not employees under the FLSA. There is no settled test for determining whether a student is an employee for purposes of the FLSA. The district court applied what the parties characterize as a "primary benefit test" and considered which party (school or student) receives the primary benefit of the work the student performs. The Secretary claims this was error. She contends instead that the proper test is the one the Department of Labor's Wage and Hour Division ("WHD") has formulated for persons participating in employer-sponsored training programs. Under this test, the Secretary contends, Laurelbrook's students are employees. We must address two questions to decide this appeal: 1) whether the district court used the correct legal standard; and 2) whether the district court properly applied the correct standard. We answer both in the affirmative.

A

We begin with the applicable standards of review. "Whether a particular situation is an employment relationship is a question of law." Fegley v. Higgins, 19 F.3d 1126, 1132 (6th Cir.1994). Therefore, we review de novo the district court's ruling that Laurelbrook students are not employees. Id. The district court relied on various factual findings it made after a bench trial. We review these findings for clear error but review de novo the district court's application of the legal standard to them. Sec'y of Labor, U.S. Dep't of Labor v. 3Re.com, Inc., 317 F.3d 534, 537 (6th Cir.2003). "A district court's factual findings are clearly erroneous if, based on the entire record, we are left with the definite and firm conviction that a mistake has been committed." Shelby Cnty. Health Care Corp. v. Majestic Star Casino, 581 F.3d 355, 364-65 (6th Cir.2009) (citation and internal quotation marks omitted).

B

The FLSA prohibits the use of "oppressive child labor" in commerce or in the production of goods for commerce. 29 U.S.C. § 212(c). The prohibition is premised on an employment relationship. See id. § 203(l). While the Act does define the terms "employee," "employer," and "employ," see id. § 203(e)(1) ("[T]he term `employee' means any individual employed by an employer."); id. § 203(d) ("`Employer' includes any person acting directly or indirectly in the interest of an employer in relation to an employee. . . ."); id. § 203(g) ("`Employ' includes to suffer or permit to work."), the definitions are exceedingly broad and generally unhelpful, see Henthorn v. Dep't of Navy, 29 F.3d 682, 684 (D.C.Cir.1994) ("The Act provides generally unhelpful definitions of its critical terms."); cf. Nationwide Mut. Ins. Co. v. Darden, 503 U.S. 318, 323, 112 S.Ct. 1344, 117 L.Ed.2d 581 (1992) (noting that identical definition of employee under ERISA "is completely circular and explains nothing"). Despite the broad definitions, the concept of employment under the FLSA is not limitless. Tony & Susan Alamo Found. v. Sec'y of Labor, 471 U.S. 290, 295, 105 S.Ct. 1953, 85 L.Ed.2d 278 (1985). One such limitation is that "[a]n individual who, `without promise or expectation of compensation, but solely for his personal purpose or pleasure, work[s] in activities carried on by other persons either for their pleasure or profit,' is outside the sweep of the Act." Id. (quoting Walling v. Portland Terminal Co., 330 U.S. 148, 152, 67 S.Ct. 639, 91 L.Ed. 809 (1947)).

Whether an employment relationship exists under a given set of circumstances "is not fixed by labels that parties may attach to their relationship nor by common law categories nor by classifications under other statutes." Powell v. U.S. Cartridge Co., 339 U.S. 497, 528, 70 S.Ct. 755, 94 L.Ed. 1017 (1950). Rather, it is the "economic reality" of the relationship between parties that determines whether their relationship is one of employment or something else. Alamo, 471 U.S. at 301, 105 S.Ct. 1953 (quoting Goldberg v. Whitaker House Coop., Inc., 366 U.S. 28, 33, 81 S.Ct. 933, 6 L.Ed.2d 100 (1961)). "The issue of the employment relationship does not lend itself to a precise test, but is to be determined on a case-by-case basis upon the circumstances of the whole business activity." Donovan v. Brandel, 736 F.2d 1114, 1116 (6th Cir. 1984) (citing Rutherford Food Corp. v. McComb, 331 U.S. 722, 730, 67 S.Ct. 1473, 91 L.Ed. 1772 (1947)).

While the Secretary and Laurelbrook agree that economic realities govern here, there is considerable dispute between them regarding the more specific inquiry we should undertake in this instance. To state that economic realities govern is no more helpful than attempting to determine employment status by reference directly to the FLSA's definitions themselves. There must be some ultimate question to answer, factors to balance, or some combination of the two.

For example, in distinguishing between employees and independent contractors, courts have focused on "whether, as a matter of economic reality, the worker is economically dependent upon the alleged employer or is instead in business for himself." See, e.g., Hopkins v. Cornerstone Am., 545 F.3d 338, 343 (5th Cir.2008). To guide the inquiry, courts consider factors such as the degree of control exercised by the alleged employer, the extent of the relative investments of the worker and the alleged employer, the degree to which the worker's opportunity for profit or loss is determined by the alleged employer, the skill and initiative required in performing the job, and the permanency of the relationship. Id. No one factor is dispositive, nor can the "collective answers to all of the inquiries produce a resolution which submerges the dominant factor—economic dependence. . . . The five tests are aids— tools to be used to gauge the degree of dependence of alleged employees on the business with which they are connected. It is dependence that indicates employee status." Usery v. Pilgrim Equip. Co., Inc., 527 F.2d 1308, 1311-12 (5th Cir. 1976). Each factor must be applied with the ultimate notion of dependence in mind. Id. We have applied this general framework. See, e.g., Fegley, 19 F.3d at 1132 n. 6; Brandel, 736 F.2d at 1116-17, 1120; see also Imars v. Contractors Mfg. Servs., Inc., No. 97-3543, 1998 WL 598778, at *2-6 (6th Cir. Aug.24, 1998) (per curiam) (discussing our precedent on this issue); cf. Lilley v. BTM Corp., 958 F.2d 746, 750 (6th Cir.1992) (applying economic reality test in case involving Age Discrimination in Employment Act).

We must first determine the specific test to apply in assessing whether employment is present in a training or learning situation. Both sides propose different tests, which we now consider. Laurelbrook's proposed test is categorical, rather than fact-based. It contends that its status as an accredited vocational school should conclusively resolve this litigation in its favor because the Supreme Court has held that the FLSA does not apply to students attending vocational schools. In Walling v. Portland Terminal Co., 330 U.S. 148, 67 S.Ct. 639, 91 L.Ed. 809 (1947), the Supreme Court held that prospective railroad brakemen training for the position through the railroad's training program were not employees under the FLSA. Discussing the FLSA, the Court noted:

Id. at 152-53, 67 S.Ct. 639. Laurelbrook asserts that the Court's prime example of a non-employee is a vocational student. According to Laurelbrook, there is no need to analyze whether its students are employees because no vocational school students can be employees under the FLSA.

We disagree with the foregoing approach. For starters, we find the passage in Portland Terminal to be dicta because the Court was not considering a vocational school; it was considering an employer-based training program. Moreover, as a district court in this circuit has indicated, the passage was likely intended merely to "point out the absurdity of regarding as employment a student's regular school work performed primarily for its educational value." Marshall v. Baptist Hosp., Inc., 473 F.Supp. 465, 468 n. 3 (M.D.Tenn.1979), rev'd on other grounds, 668 F.2d 234 (6th Cir.1981). So, for example, "students required to perform a limited amount of varied cafeteria duty as an integrated part of a planned course of study in, for example, home economics should not be regarded as employees." Id.

More importantly, as noted above, determining employee status by reference to labels used by the parties is inappropriate. Powell, 339 U.S. at 528-530, 70 S.Ct. at 772. And concluding that students are not employees simply because they are students at a vocational school is precisely the type of labeling courts must resist. Such an approach bypasses any real consideration of the economic realities of the relationship and is antithetical to settled jurisprudence calling for consideration of the totality of the circumstances of each case. Indeed, courts have in the past determined that students in vocational training programs were nevertheless employees under the FLSA. See, e.g., Reich v. Shiloh True Light Church of Christ, 895 F.Supp. 799, 819 (W.D.N.C.1995) (holding that children enrolled in church-run vocational training program were employees), aff'd per curiam, 85 F.3d 616 (4th Cir.1996) (unpublished table decision); Baptist Hosp., 473 F.Supp. at 477 (holding that X-ray technicians-in-training enrolled in two-year, accredited college program were employees); see also id. at 468 n. 3 (providing example of where improper labeling likely resulted in arguably incorrect finding of non-employee status (citing Bobilin v. Bd. of Educ., Haw., 403 F.Supp. 1095, 1106 (D.Haw.1975))). The district court in this case likewise declined to apply such a test. We reject Laurelbrook's proposal that we resolve this appeal by concluding that the FLSA categorically does not apply to accredited vocational training programs. Laurelbrook's status as a vocational school cannot serve as a substitute for a full consideration of the realities surrounding its program.

The Secretary urges the application of a six-factor test created by the WHD to distinguish between employees and trainees. The test is laid out in the WHD's Field Operations Handbook and in a publication it issued for general public consideration. See Employment Relationship Under the Fair Labor Standards Act, WH Pub. 1297 (Rev. May 1980), available at http://www.osha.gov/pls/epub/wageindex. download?p_file=F11973/WH1297.pdf. The portion of the publication relating to trainees reads in its entirety as follows:

WH Pub. 1297, at 4-5.

The Secretary contends that both the test and her determination that it applies here are entitled to deference under Skidmore v. Swift & Co., 323 U.S. 134, 65 S.Ct. 161, 89 L.Ed. 124 (1944).1 Under so-called Skidmore deference, an agency's "policy statements, embodied in its compliance manual and internal directives," which interpret regulations and statutes the agency is charged with enforcing are entitled to a "measure of respect." Fed. Express Corp. v. Holowecki, 552 U.S. 389, 399, 128 S.Ct. 1147, 170 L.Ed.2d 10 (2008). "The weight [given an administrative] judgment in a particular case will depend upon the thoroughness evident in its consideration, the validity of its reasoning, its consistency with earlier and later pronouncements, and all those factors which give it power to persuade, if lacking power to control." Skidmore, 323 U.S. at 140, 65 S.Ct. 161.

Courts differ on whether the WHD's test is entitled to controlling weight in determining employee status in a training context. Some courts have said that the test is entitled to "substantial deference." See, e.g., Atkins v. Gen. Motors Corp., 701 F.2d 1124, 1128 (5th Cir.1983). Others have rejected it altogether. See, e.g., McLaughlin v. Ensley, 877 F.2d 1207, 1209-10 & n. 2 (4th Cir.1989). Still others strike a balance and consider the factors as relevant but not dispositive to the inquiry. See, e.g., Reich v. Parker Fire Prot. Dist., 992 F.2d 1023, 1027 (10th Cir.1993); see also Harris v. Vector Mktg. Corp., 753 F.Supp.2d 996, 1006-07 (C.D.Cal.2010).

We find the WHD's test to be a poor method for determining employee status in a training or educational setting. For starters, it is overly rigid and inconsistent with a totality-of-the-circumstances approach, where no one factor (or the absence of one factor) controls. See Rutherford Food, 331 U.S. at 730, 67 S.Ct. 1473 ("We think, however, that the determination of the relationship does not depend on such isolated factors but rather upon the circumstances of the whole activity."). Moreover, at least one court has found the test's all-or-nothing approach inconsistent with prior WHD interpretations and opinions endorsing a flexible approach, thereby diminishing any persuasive force the test might be entitled to under Skidmore. See Parker Fire, 992 F.2d at 1026. Furthermore, the test is inconsistent with Portland Terminal itself, which, as outlined below, suggests that the ultimate inquiry in a learning or training situation is whether the employee is the primary beneficiary of the work performed. While the Secretary's six factors may be helpful in guiding that inquiry, the Secretary's test on the whole is not.

Eschewing the positions taken by both parties, the district court focused on which party receives the primary benefit of the work performed by Laurelbrook students. This was the appropriate inquiry to make. The test finds solid grounding in the Supreme Court's decision in Portland Terminal, where, as noted above, the Court considered whether persons training to become railroad brakemen were employees of the railroad sponsoring the training program. 330 U.S. at 149, 67 S.Ct. 639. The Court began by setting out the guiding rationale for the decision: that the FLSA 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." Id. at 152, 67 S.Ct. 639. It considered various factors, including whether the men displaced the railroad's paid employees, whether the men impeded the railroad's business during training, whether the men were supervised, and whether the men received training relevant to the position of a railroad brakeman. After finding the training "most greatly benefit[ted]" the trainees and "accepting the unchallenged findings . . . that the railroads receive no `immediate advantage' from any work done by the trainees," the Court concluded that the men were not employees for purposes of the FLSA. Id. at 153, 67 S.Ct. 639.

Courts have read Portland Terminal as focusing principally on the relative benefits of the work performed by the purported employees. See, e.g., Isaacson v. Penn Cmty. Servs., Inc., 450 F.2d 1306, 1309 (4th Cir.1971) ("The rationale of Portland Terminal would seem to be that the railroad received no `immediate advantage' from the trainees' services. To state it otherwise, the principal purpose of the seemingly employment relationship was to benefit the person in the employee status."); Todaro v. Twp. of Union, 27 F.Supp.2d 517, 534 (D.N.J.1998) (characterizing Portland Terminal as "concluding that participation in railroad training program did not constitute employment under FLSA where primary benefit redounded to trainees"). While the Supreme Court found various other facets of the relationship significant, in our view, its decision rested upon whether the trainees received the primary benefit of the work they performed.2

Our reading of Portland Terminal finds support in our precedent. While we have not directly spoken to the issue, we have suggested that identifying the primary beneficiary of a relationship provides the appropriate framework for determining employee status in the educational context. See Baptist Hosp., 668 F.2d at 236. Baptist Hospital involved a two-year program for persons training to become X-ray technicians through Vanderbilt University Hospital, and later through a junior college. Graduates received an associate's degree and were eligible to sit for the Radiologic Technician ("RT") certification test. Id. at 235. Baptist Hospital was one of the program's member-hospitals that provided clinical training for students. The Secretary of Labor brought an action against the hospital seeking back wages for students and an injunction. After a bench trial, the district court concluded that the students were employees and that the hospital could not escape liability through good-faith reliance on an administrative interpretation relating to X-ray technician students. Id. at 236. We reversed on the issue of reliance, but expressly agreed with both the district court's test for employee status and its application of that test. Id. at 239.

Because we approved of the district court's decision regarding employee status without discussion, review of the district court's analysis is instructive. The analysis began with a recognition of the "evils" the FLSA targets: displacement of regular employees and exploitation of labor. In examining any training or educational situation for possible signs of these evils, the court thought it relevant to consider both the validity of the program as an educational experience and whether the primary benefit from the relationship flowed to the learner or to the alleged employer. Id. at 468. Though the court mentioned other factors as relevant, see. id. at 468 n. 4, it never applied them in its analysis.

With these principles in mind, the district court reviewed the details of Baptist Hospital's training program. First off, the training was deficient. Instead of placing students with employed RTs who would instruct students on proper X-ray procedures and supervise their performance, the program many times assigned students to areas of the hospital staffed only by other students. And even those students assigned to areas generally staffed by RTs often found themselves working alone or with other students, as the RTs were frequently reassigned to areas where they were needed most. Students staffed the chest X-ray rooms and frequently performed portable X-rays by themselves. Students often performed functions that relieved RTs and other employees from such duties. Id. at 472. Students also displaced RTs in more substantive ways. For instance, chest X-rays were the most common procedures at the hospital, at one time constituting nearly 40% of all procedures performed. Virtually all chest X-rays were performed by the students, whom the hospital did not pay. Additionally, students worked shifts assigned originally to employed RTs. The record was clear that were it not for the students' performance of essential tasks, the hospital would have had to hire more employees or require its regular employees to work overtime. Id. at 473.

After review of the facts, the district court found the students were employees under the FLSA. The determination turned on the court's finding that the hospital received the primary benefit of the training program:

Id. at 476-77 (footnote omitted).

The best reading of Baptist Hospital is that the district court ultimately was searching for the party who received the primary benefit of the relationship. But see Parker Fire, 992 F.2d at 1026 (characterizing Baptist Hospital as applying WHD's six-factor test). The court's consideration of both employee displacement and educational validity was made in furtherance of the ultimate determination of primary benefit. Both factors assisted in determining concretely which party primarily benefitted. Using students instead of paid employees for essential functions provided a substantial and immediate economic benefit to the hospital. The only benefits the students received from the arrangement were educational. But because the training was deficient, the educational benefits were slight. As the court noted, "[h]ad the training program been found to be educationally sound the court might nevertheless have concluded that the bulk of the benefit inured to the trainees, but because the trainees were shortchanged educationally the court finds that the hospital was the primary benefactor from the relationship between it and the trainees." Id. at 476-77. Given that the balance of benefits ran decidedly in favor the hospital, the court deemed the students employees.

Decisions from courts outside our circuit apply a primary benefit test to determine employment status. Some do so expressly. See McLaughlin, 877 F.2d at 1209 ("In sum, this court has concluded that the general test used to determine if an employee is entitled to the protections of the Act is whether the employee or the employer is the primary beneficiary of the trainees' labor."); Donovan v. Am. Airlines, Inc., 686 F.2d 267, 272 (5th Cir.1982) ("In this respect, the district court's balancing [of relative benefits] analysis appears to us to be more appropriate."); Isaacson, 450 F.2d at 1309; Wirtz v. Wardlaw, 339 F.2d 785, 788 (4th Cir.1964); Shiloh True Light Church, 895 F.Supp. at 817-18; Lane v. Carolina Beauty Sys., Inc., No. 6:90CV00108, 1992 WL 228868, at *6 (M.D.N.C. July 2, 1992) ("[B]ecause the benefits to Plaintiff derived from the teacher training program clearly outweighed the benefits to CBS, Plaintiff was not an `employee' under the FLSA. . . .").

Still others courts implicitly have applied a primary benefit test. See Blair v. Wills, 420 F.3d 823, 829 (8th Cir.2005) (holding that chores performed by teenager at boarding school he was ordered to attend as part of juvenile sentence were, as a matter of law, not "work"; though chores performed by students defrayed costs school would have incurred had it hired employees to perform tasks, chores were integral part of educational curriculum and were primarily for students' benefit); Woods v. Wills, 400 F.Supp.2d 1145, 1166 (E.D.Mo.2005), aff'd per curiam, 225 Fed. Appx. 420 (8th Cir.2007); Donovan v. Am. Airlines, Inc., 514 F.Supp. 526, 533-35 (N.D.Tex.1981) ("If the ultimate issue, as we believe it to be, is whether American has effectively assimilated these trainees into its work force in such a fashion that they are employees in economic reality, then an important analytical focus must be the relative benefits flowing to trainee and company during the training period."), aff'd, 686 F.2d 267 (5th Cir.1982); Bailey v. Pilots' Ass'n for the Bay & River Del., 406 F.Supp. 1302, 1307 (E.D.Pa.1976) (holding pilot apprentice was "employee" where his duties "were of immediate benefit" to pilot association, in that he substituted for paid employees and received no real training from association's apprenticeship program); Bobilin, 403 F.Supp. at 1108 (recognizing that mandatory cafeteria duty performed by students provided economic value to state, but finding that students were not employees given that duties served educational goals); see also Atkins, 701 F.2d at 1128 (approving of Secretary's six-factor test, yet focusing entirely on whether employer derived immediate benefit from trainees' activities).

We find that a primary benefit test provides a helpful framework for discerning employee status in learning or training situations. By focusing on the benefits flowing to each party, the test readily captures the distinction the FLSA attempts to make between trainees and employees. See Portland Terminal, 330 U.S. at 152, 67 S.Ct. 639 (stating that FLSA is not intended to reach persons "who, without any express or implied compensation agreement, might work for their own advantage on the premises of another."). The test is pitched at an appropriate level of generality to enable it to reach non-employer-based training relationships, such as those provided in schools, and can take into consideration the unique nature of the training situation presented here. It also helps to readily identify relationships Congress sought to stamp out by enacting the FLSA's child labor provisions: relationships that place adult employees in competition with minors. See Lenroot v. Interstate Bakeries Corp., 55 F.Supp. 234, 236 (W.D.Mo.1944), aff'd in part and rev'd in part on other grounds, 146 F.2d 325 (8th Cir.1945). If a purported employer receives the primary benefit from a working relationship with a child, it is likely that the child is in competition with adults, whom the employer cannot employ without complying with the FLSA's costly and burdensome requirements. If, however, a child receives the primary benefit of the work performed for a purported employer, and the child's presence does more harm to the purported employer's operations than good (or no good at all), it is unlikely that the child is competing with adults for the opportunity to hinder the employer's operations.

To conclude, we hold that the proper approach for determining whether an employment relationship exists in the context of a training or learning situation is to ascertain which party derives the primary benefit from the relationship. Factors such as whether the relationship displaces paid employees and whether there is educational value derived from the relationship are relevant considerations that can guide the inquiry. Additional factors that bear on the inquiry should also be considered insofar as they shed light on which party primarily benefits from the relationship. The district court applied such a test. It did not err by doing so.

C

Having determined that the district court applied the proper test for liability, we consider the second question: whether the district court properly applied the primary benefit test. After seven days of hearing testimony and receiving documentary evidence, the district court concluded that Laurelbrook students were the primary beneficiaries of the work performed: "[a]lthough there is benefit to the school and sanitarium from the students' activities, the totality of the circumstances shows that the primary benefit is to the students, who learn practical skills about work, responsibility, and the dignity of manual labor in a way consistent with the religious mission of their school." To guide its inquiry, the court considered many of the Secretary's six factors. We review the district courts factual findings for clear error and its application of the law to those facts de novo. 3Re.com, 317 F.3d at 537.

We note initially that the district court's factual findings are not as detailed as in some other cases. Nevertheless, we find that the findings do not prevent meaningful review. In Grover Hill Grain Co. v. Baughman-Oster, Inc., 728 F.2d 784 (6th Cir.1984), we addressed the specificity requirements of a district court's factual findings:

Id. at 792-93 (internal citations omitted). The district court's findings satisfy these requirements as they allow us to understand the basis for its ultimate conclusion.

In applying the primary benefit test, the district court recognized that students' activities at Laurelbrook contribute to Laurelbrook's maintenance, thereby benefitting Laurelbrook's operations. Laurelbrook receives payment for services it provides to patients at the Sanitarium; some of these services are performed by students at no cost to Laurelbrook.3 Hours worked by students in the Sanitarium also contribute to the Sanitarium's satisfaction of its licensing requirements. Laurelbrook sells flowers and produce grown at Laurelbrook with student help. The proceeds from these sales go directly to Laurelbrook's operations. As part of a course on collision repair, students assist in repairing cars for the public. Beneficiaries of these services pay Laurelbrook directly and the money is recycled back into school programs. Laurelbrook also earns revenue from the sale of wood pallets the students help build.

The value of these benefits to Laurelbrook, however, is offset in various ways. The district court found that Laurelbrook students do not displace compensated workers,4 and instructors must spend extra time supervising the students at the expense of performing productive work. Specifically, the court found that Laurelbrook is sufficiently staffed such that if the students did not perform work at the Sanitarium, the staff members could continue to provide the same services there without interruption. And while not specifically mentioned by the district court in its findings, there was evidence at trial that the same was also true of the work performed by students outside the Sanitarium. There was also testimony that, were it not for the instructors' supervisory responsibilities, instructors would be able to complete more productive tasks in less time. Moreover, as the district court found, Laurelbrook is not in competition with other institutions for labor, so Laurelbrook does not enjoy an unfair advantage over other institutions by reason of work performed by its students.

On the other side of the ledger are the tangible and intangible benefits that accrue to the students. The district court found that Laurelbrook provides it students with significant tangible benefits. Students are provided with hands-on training comparable to training provided in public school vocational courses, allowing them to be competitive in various vocations upon graduation. Students learn to operate tools normally used in the trades they are learning, while being supervised by instructors. Students engage in courses of study that have been considered and approved of by the state accrediting agency. In short, the educational aspect of the instruction at Laurelbrook is sound, in contrast to the training program at issue in Baptist Hospital, where the supervision was inadequate, the exposure to various aspects of the trade limited, and the overall value to the students nil. None of these educational shortcomings is present here. Indeed, the Tennessee Department of Education, through EASEA, has determined that Laurelbrook's vocational program provides benefits to the students sufficient to warrant accreditation.

Significant, too, are the intangible benefits students receive at Laurelbrook. As the district court found, receiving a well-rounded education—one that includes hands-on, practical training—is a tenet of the Seventh-Day Adventist Church. Laurelbrook provides students with the opportunity to obtain such an education in an environment consistent with their beliefs. The district court found that the vocational training portion of the education teaches students about responsibility and the dignity of manual labor. Thought not mentioned in the district court's opinion, there is ample evidentiary support for these findings. Parents testified to the benefits their children received from the program, stating that the students learn the importance of working hard and seeing a task through to completion. Some parents testified that their children have become more responsible and have taken on leadership roles since participating in Laurelbrook's program. Service in the Sanitarium engenders sensitivity and respect for the elderly and infirm. Laurelbrook alumni testified that the leadership skills and work ethic developed at Laurelbrook have proved highly valuable in their future endeavors. Employers also testified that Laurelbrook alumni have a strong work ethic, leadership skills, and other practical skills that graduates of other vocational programs lack.

The Secretary discounts the value of these intangible benefits, but we agree with the district court that they are of significant value. Courts that have addressed the value of such benefits have likewise concluded that they are significant enough to tip the scale of primary benefit in the students' favor even where the school receives tangible benefits from the students' activities. See, e.g., Blair, 420 F.3d at 829; Woods, 400 F.Supp.2d at 1166; Bobilin, 403 F.Supp. at 1108. The overall value of broad educational benefits should not be discounted simply because they are intangible.

After considering all of the evidence, the district court found that there is benefit to Laurelbrook's operations from the students' activities, but the primary benefit of the program runs to the students. We find no error in the district court's application of the primary benefit test.

III

The district court applied the proper test when it asked which party to the relationship received the primary benefit of the students' activities. We find the district court's findings of fact amply supported by the evidence adduced at trial, and agree with its application of the primary benefit test to conclude that the students at Laurelbrook are not employees for purposes of the FLSA. Accordingly, the judgment of the district court is AFFIRMED.