dissenting.
The practical question before us is whether inmates assigned to work at a plasma center operated by a private enterprise are covered by the Federal Labor Standards Act (FLSA). The majority finds that appellants are not in an employer/employee relationship under the FLSA with either the state or Cutter Biological. En route to this conclusion, it overlooks Supreme Court guidance and, I believe, misconstrues Congressional purpose. In my view, this court’s decision should reflect the central aims of federal legislation. Because I cannot subscribe to the majority’s niggardly interpretation of the FLSA, I must respectfully dissent.
The majority correctly outlines the factors to be considered in examining the “economic reality” of the employment situation for the purposes of the FLSA. This test looks to four factors: whether the alleged employer (1) had the power to hire and fire employees, (2) supervised and controlled employee work schedules or conditions of employment, (3) determined the rate and method of payment, and (4) maintained employment records. Bonnette v. California Health & Welfare Agency, 704 F.2d 1465, 1470 (9th Cir.1983). After discussing the guidelines, however, the majority then fails to apply them correctly to the facts of this case.
The reasons for this failure are hard to decipher. On the one hand, Judge Trott apparently holds that “the FLSA itself cannot be said to cover prison labor.” Supra at 1325. On the other hand, he goes on to apply the four-part economic realities test and finds that under the facts of this case, these appellants are neither Cutter’s nor the state’s employees. Id. at 1325-1326.1 While I am not certain which of these is my colleague’s position, I am confident that both are equally wrong. For her part, Judge Rymer chooses in her concurrence to avoid the first question in its entirety, focusing instead on the Bonnette factors. See supra at n. 1. However, the economic realities test cannot be considered in a vacuum. It must be conducted, throughout, with an eye to the overriding legislative purpose.
*1332I. Congressional Purpose
Context in this case means the reasons for the FLSA’s enactment by Congress. I agree with Judge Trott that the Act’s express purpose was to aid the unprotected and lowest paid of the working people. See supra at 1326. Also I understand, though I do not fully share, my colleague’s distaste at extending Congress’ economic protection to people he labels “convicted murderers, rapists, burglars, armed robbers, swindlers, thieves, and the likes.” See supra at 1324. But that is to miss the point. For to conclude that the FLSA applies in this case requires no particular sympathy for prison inmates.
In section 202(a) of the Act, Congress indicated that
the existence, in industries engaged in commerce or in the production of goods for commerce, of labor conditions detrimental to the maintenance of the minimum standard of living necessary for health, efficiency, and general well-being of workers (1) causes commerce and the channels and instrumentalities of commerce to be used to spread and perpetuate such labor conditions among the workers of the several States; ... [and] (3) constitutes an unfair method of competition in commerce.
29 U.S.C. § 202(a). Lest it had not been clear enough, the House also explained in a Report that, in the wake of the Act’s passage,
[n]o employer in any part of the United States in any industry affecting interstate commerce need fear that he will be required by law to observe wages ... higher than those applicable to his competitors. No employee ... need fear that the fair labor standards maintained by his employer will be jeopardized by oppressive labor standards maintained by those with whom his employer competes.
H.R.Rep. No. 2182, 75th Cong., 3d Sess. 6-7 (1938); see also International Ladies’ Garment Workers’ Union v. Donovan, 722 F.2d 795, 808 (D.C.Cir.1983).
The FLSA, in other words, is comprehensive legislation. It certainly is designed to directly assist exploited employees requesting its enforcement in a specific instance. But its sword is double-edged. It also purports to come to the indirect aid of compliant employers and their employees who might suffer from the ill effects of nonobservance. See Watson v. Graves, 909 F.2d 1549, 1554 (5th Cir.1990) (“The Act was drafted ... to eliminate unfair competition among employers competing for business in the market and among workers looking for jobs”) (citing Carter v. Dutchess Comm. College, 735 F.2d 8, 14 (2d Cir.1984)); International Ladies’ Garment Union, 722 F.2d at 808-09; see also Lerwill v. Inflight Services, Inc., 379 F.Supp. 690, 696 (N.D.Cal.1974) (enforcement of the Act “is also intended to protect the employers who comply with its terms”), aff'd sub nom. Lerwill v. Inflight Motion Pictures, Inc., 582 F.2d 507 (9th Cir.1978).
Put somewhat differently, the scope of the FLSA must be broad enough to prevent companies engaged in interstate commerce from taking advantage of employees’ substandard wages. See United States v. Darby, 312 U.S. 100, 109-10, 115, 61 S.Ct. 451, 454-55, 457, 85 L.Ed. 609 (1941). In general, whether it applies thus depends on whether application will further this goal.2
Neither one of my colleagues addresses the unfair competition issue. Indeed, Judge Rymer dismisses it in a footnote, on the ground that no relevant evidence has been presented. See supra at n. 4 (Rymer, J., concurring). It is true that appellants in this case are not Cutter’s competitors; their interest is in receiving a minimum wage, not in purifying the stream of commerce. But I remain convinced that Congress’ overall intent must determine the Act’s applicability to a specific case. See, e.g., Carter, 735 F.2d at 14.
In most cases, both concerns overlap: the persons seeking application of the FLSA are precisely those Congress wished to assist, and, in so doing, Congress protects compliant employers and their em*1333ployees as well. Here, however, we have a slightly different scenario: Congress, arguably, is indifferent to the economic well-being of prisoners. But it remains concerned, and so, by implication, must we, with the effect of their working conditions on other employers and their employees. In such cases, I believe, the principles underlying the FLSA must guide our determination regarding the Act’s coverage.
With this in mind, I now turn to the second question, namely whether, and in what circumstances, the FLSA extends to prison labor.
II. Application of the FLSA to Prison Inmates
If Judge Trott’s view is that prison labor can never be covered by the FLSA, it is at odds with precedent both from the Supreme Court and from this Circuit. First, such a narrow interpretation of an employment relationship disregards over forty years of Supreme Court case law requiring the courts to define “employer” and “employee” expansively under the Act. See, e.g., Tony & Susan Alamo Found. v. Secretary of Labor, 471 U.S. 290, 295-96, 105 S.Ct. 1953, 1958-59, 85 L.Ed.2d 278 (1985); Goldberg v. Whitaker House Coop., Inc., 366 U.S. 28, 32, 81 S.Ct. 933, 936, 6 L.Ed.2d 100 (1961); Rutherford Food Corp. v. McComb, 331 U.S. 722, 730, 67 S.Ct. 1473, 1476, 91 L.Ed. 1772 (1947); Walling v. Portland Terminal Co., 330 U.S. 148, 150-51, 67 S.Ct. 639, 640-41, 91 L.Ed. 809 (1947); United States v. Rosenwasser, 323 U.S. 360, 362-63, 65 S.Ct. 295, 296-97, 89 L.Ed. 301 (1945).3
Second, this position is hard to square with what we have said on this issue up to now. In Baker v. McNeil Island Corrections Center, 859 F.2d 124 (9th Cir.1988), a prison inmate brought a Title VII suit against the prison. The district court, finding that there was no employment relationship between a prison and its prisoners, dismissed the suit for failure to state a claim. We reversed, rejecting the claim that as a matter of law prisoners cannot be employees of the state.4 The Second Circuit, in Carter v. Dutchess Community College, 735 F.2d 8, 13-14 (2d Cir.1984), and the Fifth Circuit, in Watson v. Graves, 909 F.2d 1549, 1554 (5th Cir.1990) (agreeing “with the Carter court that status as an inmate does not foreclose inquiry into FLSA coverage”), reached the same conclusion.
I suspect that Judge Trott’s result derives in large part from his fragmentary *1334interpretation of the FLSA’s purpose. By-focusing exclusively on the fact that the FLSA was enacted to improve workers’ living conditions, he can then easily conclude that prisoners were not meant to fall within the statute’s protective reach. But in his haste to deny economic benefits to inmates, he has lost sight of the second major rationale behind the FLSA’s minimum wage provisions. As I discussed earlier, Congress intended the FLSA to have the widest possible impact in the national economy because one of its purposes was the establishment of minimum standards in the workplace in order to eliminate unfair competition among private companies and among workers looking for jobs. See Mitchell v. Robert De Mario Jewelry Inc., 361 U.S. 288, 292, 80 S.Ct. 332, 335, 4 L.Ed.2d 323 (1960); Carter 735 F.2d at 13. This national purpose is subverted when a court permits one company within an industry to avoid the strictures of the Act. By exempting appellants from FLSA coverage, the majority, without even discussing this issue, has given Cutter Biological an unfair cost advantage over its competitors, who are required to pay minimum wage. Quite simply, Cutter, with the state’s assistance, is making use of a cheap labor pool, bypassing the Act’s constraints, and thwarting its overriding purpose.
I cannot believe that Congress intended to create a loophole of such proportions, permitting employers to hire labor at substandard pay and then sending the output into the flow of commerce. Indeed there are, as Judge Trott remarks, “obvious policy considerations” to this issue, supra at 1324 — but they are triggered by nonen-forcement rather than by enforcement of the Act.
The majority’s recitation of Arizona’s law regarding prisoners, or of the purposes of correctional industries, supra at 1325, is irrelevant in this regard. What matters is that Cutter was offered a pool of workers whose wages were below the legal minimum; what matters is that one for-profit business thereby enjoyed an unfair advantage over its competitors. Moreover, the inmates worked alongside Cutter’s other, “regular” employees performing essentially the same job. The relevance of these factors becomes clearer when we consider the hypothetical case of a prison requiring its inmates to perform prison maintenance work or produce goods used solely by the state for less-than-minimum wages. In this situation, the FLSA’s concern about economic fairness in a competitive market does not come into play; in ours, it must.5
As an aside, it also is worth reminding that, under Ariz.Rev.Stat.Ann. § 41-1624, the Department can use part of the prisoners’ wages for purposes other than compensation. Possible uses might conceivably include payment for room and board, support for prisoners’ families, or various operating expenses. In short, the issue is not one of unjustified reward to a state prisoner, but of unwarranted benefit to a private enterprise.
Having concluded that, under the specific circumstances of this case, Arizona’s inmates fall under the FLSA’s umbrella, we arrive at the final hurdle: Are appellants employees of Cutter and the state?
III. The Bonnette Test
The minimum wage provisions apply only if appellees are “employers” of the inmates within the meaning of the FLSA. As the majority explains, the status is commonly *1335determined by the Bonnette test, described above. See supra at 1328 (Rymer, J., concurring). At the outset, however, it bears repeating that mere technical adherence to the test cannot suffice. For it is the economic realities of the situation we are interested in, and they must be analyzed “in light of the policies behind the FLSA.” Watson, 909 F.2d at 1555. My colleagues may well have scrupulously embarked on the four-part test, but they have left its substance behind.
A. The State’s Relationship With Appellants
Under the proper analysis of “economic reality” using the framework of Bonnette, the state qualifies as an employer. See Goldberg, 366 U.S. at 32-33, 81 S.Ct. at 936-937; Rutherford, 331 U.S. at 730, 67 5.Ct. at 1476; Bonnette, 704 F.2d at 1469-70; Carter, 735 F.2d at 12. The state Department of Corrections (DOC) determined which inmates could be employed and retained the power to remove an inmate from his job. Although not directly supervising during the employment day, the state maintained control over the inmates for security purposes. The method of payment was agreed to by the DOC in contract negotiations with Cutter Biological and the DOC determined the pay scale.6
Most disturbing to the majority, apparently, is the state’s degree of control. Specifically, its complete control over the prisoners would preclude it from being an employer because this control was not part of the “bargained for exchange of labor” that occurs in traditional employer/employee relationships. This analysis contradicts Ninth Circuit and Supreme Court case law.
In Baker, the court found that the most important factor in determining whether an employment relationship exists is “the extent of the employer’s right to control the means and manner of the worker’s performance.” Id. at 128. Contrary to the majority’s analysis, the court in Baker held that the prison’s complete control of its inmates made more certain the inmates’ status as employees of the institution. Id. The analysis in Baker precludes the majority from deciding that these prisoners cannot be employees of the state based on the state’s control.
Neither can the court depend on traditional common law notions of “bargained for exchange” to determine what is an employer/employee relationship for FLSA purposes. “The definition of ‘employer’ under the FLSA is not limited by the common law concept of ‘employer,’ and is to be given an expansive interpretation in order to effectuate the FLSA’s broad remedial purposes.” Bonnette, 704 F.2d at 1469; accord Rutherford, 331 U.S. at 729, 67 S.Ct. at 1476; Real v. Driscoll Strawberry Assoc., 603 F.2d 748, 754 (9th Cir.1979). The majority is required to determine whether an employment relationship exists based on the guidelines and analysis outlined in FLSA precedent; it errs in relying on a narrow traditional notion of the concept of employer that completely misinterprets the import of control in an employer/employee relationship.
In reality, there is nothing inconsistent in the coexistence of prisoner and employee status under the FLSA. If inmates are selected to work for private contractors participating in the market at large, they are in that capacity and given the FLSA’s purpose, “employees” of the state within the conception of the Act.
B. Cutter Biological’s Relationship with Appellants
In considering this relationship, the majority disregards facts and does not view the evidence in the light most favorable to appellants, as is required when reviewing a motion for summary judgment. See Tzung v. State Farm Fire & Casualty Co., 873 F.2d 1338, 1339-40 (9th Cir.1989). A close examination of the facts reveals that the inmates’ employment with Cutter Biological satisfies the Bonnette factors. I will outline briefly the facts that apply to each factor.
*1336(i) The power to hire and fire employees.
As the majority states, it is undisputed that the DOC determined whether a prisoner was eligible to work and could remove him from the job. However, the record indicates that an inmate had to be interviewed and selected by Cutter Biological and could be removed if Cutter Biological did not approve of the prisoner’s work. The number and qualifications of the workers also were decided by Cutter Biological, and once a prisoner had been hired, Cutter Biological could promote or demote him on the basis of his performance. Cutter Biological has never alleged that it was required to employ a prisoner it did not desire. It is true that prisoners discontinued their employment when Cutter Biological did not wish them to leave, but even private employers involuntarily lose personnel.
In sum, although Cutter Biological was limited to employing only those selected by the DOC, once an inmate was eligible, Cutter Biological had the power to hire and fire. It would “run counter to the breadth of the statute and to the Congressional intent to impose a qualification which permits an employer who exercises substantial control over work, but whose hiring decisions occasionally may be subjected to a third party’s veto, to escape compliance with the Act.” Carter, 735 F.2d at 12.
(ii) Supervision and control of employee work schedules and conditions.
Cutter Biological had nearly the same supervision and control over employee work schedules and conditions as a private employer. The prisoners were only available to work during certain hours and sometimes due to prison business they missed a day of work; this also is true of private employment situations. The details of the prisoners’ work assignments were jointly specified by Cutter Biological and the DOC. Cutter Biological completely determined the working conditions of the prisoners — what tasks each prisoner performed, how much the prisoner was paid within the DOC range (depending on which job he held), and whether the prisoner was promoted. As Judge Trott admitted, Cutter Biological “supervised the inmates in their daily activities.” This supervision qualifies Cutter Biological as an employer under the second Bonnette factor.
(iii) Rate and method of payment.
It appears that the pay scale was determined by the DOC. However, within that range Cutter Biological determined which prisoners were paid which rate. Prisoners were paid more if they had more responsibility and Cutter Biological had complete discretion over which prisoners held the more lucrative jobs. The method of payment was determined by contract between Cutter Biological and the DOC and by state law. , Cutter Biological, then, had some control over the rate and method of payment through its contract negotiations with the DOC.
(iv) Employment records.
The majority claims that Cutter Biological kept no employment records. It is unclear what they mean by employment records. The record contains Cutter Biological’s contemporaneous reports on how much each inmate worked and how much each was paid. I would consider those to be employment records.
In short, and after examining the “circumstance of the whole activity,” it is clear that the relationship between Cutter Biological and the prisoners could be nothing else but employer/employee. The inmates were not independent contractors; Cutter Biological had no rehabilitative or custodial relationship with them. It is undisputed that the prisoners performed precisely the same job that civilian employees performed for Cutter Biological. The prisoners worked for Cutter Biological and Cutter Biological permitted them to work because it profited from doing so. We are required to interpret the definition of employer expansively; under that mandate Cutter Biological must be an employer.
We are left, then, with the argument that responsibilities were divided between the state and Cutter Biological. In essence, the majority argues that the state’s *1337control over appellants precludes .Cutter from being an employer because it deprives the private company of one of the essential attributes of that status. The Labor Department has prevented this absurd consequence by adopting the concept of joint employment. Two parties are joint employers if they share or co-determine those matters governing the essential terms and conditions of employment. NLRB v. Greyhound Corp., 368 F.2d 778, 780 (5th Cir.1966), on remand from Boire v. Greyhound Corp., 376 U.S. 473, 84 S.Ct. 894, 11 L.Ed.2d 849 (1964). As we observed in Bonnette, 704 F.2d at 1470: “The fáct that the appellants delegated to the recipients various responsibilities does not alter [the economic reality]; it merely makes them joint employers.” Likewise, dividing the responsibilities of employment between the state and Cutter biological does not mean that neither is the employer; it means that they are joint employers. See Bonnette, 704 F.2d at 1469-70; NLRB v. Browning-Ferris Indus., 691 F.2d 1117, 1122-23 (3d Cir.1982); NLRB v. Checker Cab Co., 367 F.2d 692, 698 (6th Cir.1966)), cert. denied, 385 U.S. 1008, 87 S.Ct. 715, 17 L.Ed.2d 546 (1967).7
The majority’s error, in sum, is in finding that because Cutter Biological shared control over the worker it could not be the employer. In reality, the fact that Cutter Biological’s control was qualified does not place its employment relationship beyond the scope of the FLSA; it makes Cutter Biological a joint employer.
A useful analogy is provided by the independent contractor situation. In such cases, typically, a company will contract with someone who is responsible for finding workers and then supervising them. Although the company might not exercise control over basic components of its relationship with the workers, courts have not hesitated to find employment status. See, e.g., Castillo v. Givens, 704 F.2d 181, 190 (5th Cir.1983) (finding that such status could exist “even though the defendant-employer had no control over ... the right to set hours, hire and fire, or determine wages”); Usery v. Pilgrim Equipment Co., Inc., 527 F.2d 1308, 1312 (5th Cir.), cert. denied, 429 U.S. 826, 97 S.Ct. 82, 50 L.Ed.2d 89 (1976). Where, as here, the party sharing control also is an employer, they become joint employers. Cf. Castillo, 704 F.2d at 188.
IV. Conclusion
The crux of this case is simply this: Given the purposes of the FLSA, were appellants in an employer/employee relationship with the state and Cutter Biological? The majority starts off on the wrong foot by failing to discuss these purposes in full. It then errs by focusing on “isolated factors” rather than “upon the circumstances of the whole activity.” Rutherford, 331 U.S. at 730, 67 S.Ct. at 1477. In my view, however, this is the type of situation that the legislation was intended to govern, and these prisoners could be nothing else but employees.
The “economic reality” of the situation is that the inmates in this case were employed by a private company to produce a product for interstate commerce. The reality is that inmates “are not self-employed; nor are they independent, selling their products on the market for whatever price they can command. They are regimented under one organization, manufacturing what the organization desires, and receiving the compensation the organization dictates.” Goldberg, 366 U.S. at 32, 81 S.Ct. at 936 (footnote omitted). Likewise, there is no doubt that together the state and Cutter Biological have the power to hire- and fire, supervise and control work schedules and conditions, determine the rate and *1338method of payment and maintain employment records. The critical question is one that the majority neither asks nor answers: if this is not an employment relationship, what is it?
The majority’s cramped analysis under the FLSA allows a few select private companies within an industry to gain an unfair competitive advantage by circumventing minimum wage laws. Because I believe this was not Congress’ intention, I respectfully dissent.
. Adopting yet another alternative view, Judge Trott states that he finds "merit” to the argument that "the FLSA does not apply to prison inmates as a class,” id. at 1324, seemingly expressing uncertainty about a position which he then proceeds to adopt wholesale.
. Of course, there are certain categories of people expressly excluded from the FSLA. But, as Judge Trott acknowledges, prisoners are not. See supra at 1325.
. In Alamo, the Court held that we must construe the FLSA “liberally to apply to the furthest reaches consistent with congressional direction,” 471 U.S. at 296, 105 S.Ct. at 1959; in Rosenwasser, it stated;
A broader or more comprehensive coverage of employees ... would be difficult to frame. The use of the words "each” and "any” to modify "employee,” which in turn is defined to include "any” employed individual, leaves no doubt as to the Congressional intention to include all employees within the scope of the Act unless specifically excluded.
323 U.S. at 362-63, 65 S.Ct. at 296-97. Finally, in Walling, the Court explained that the "Act contains its own definitions, comprehensive enough to require its application to many persons and working relationships which, prior to the Act, were not deemed to fall within an employer-employee category.” 330 U.S. at 150-51, 67 S.Ct. at 640-41.
. Judge Trott’s hurried dismissal of Baker as "inapplicable” is unconvincing. Although Baker was interpreting the definition of employer under Title VII rather than the FLSA, “cases construing the definitional provisions of [the FLSA, Title VII or the ADEA] are persuasive authorities when interpreting the others.” Hyland v. New Haven Radiology Assocs., 794 F.2d 793, 796 (2d Cir.1986); accord Rutherford Food Corp. v. McComb, 331 U.S. 722, 723, 67 S.Ct. 1473, 1474, 91 L.Ed. 1772 (1947).
In Baker, as here, we were faced with the argument that the prison's “relationship with [the inmate is one] of prison/prisoner only, because the work positions are part of an inmate’s correctional programing.” Baker, 859 F.2d at 126; see also supra at 1325-1326. The district court in Baker, like Judge Trott in this case, invoked the irresistible authority of "common sense” to defeat a differing point of view. Baker at 128 (noting that the district court had found "common sense differences ... between inmate work assignments and ordinary employment relationships”); supra at 1324 (reaching its conclusion "with an eye guided by common sense and common intelligence”). Common sense notwithstanding, it is very difficult, in my view, to read Baker without concluding that, in certain circumstances at least, an employment relationship can exist between a prison and its prisoners.
. Although sharing my view that the FLSA is applicable to prison inmates, the Watson court distinguishes between inmates working outside the jail and inmates working for private companies "within the confines of the prison as a part of their sentence to hard labor.’’ 909 F.2d at 1553-55. It goes on to explain that in the latter situation:
there is no need to "protect the standard of living and general well-being of the worker in American industry." ... Neither is there fear of "upsetting the desired equilibrium in the work place,” because the "work place” was the prison itself.
Id. at 1555 (citations omitted).
I confess that the court’s logic escapes me. If the concern is with the impact of cheap labor on the economy, see id. at 1554-55, the pertinent distinction is between work performed, say, for prison maintenance and work performed for outside commerce. Whether the private company moves to the prison or the inmates migrate to the outside world is, from this perspective, irrelevant.
. The record does not indicate whether the state kept employment records.
. “Two or more employers may jointly employ someone for purposes of the FLSA.” Bonnette, 704 F.2d at 1469; accord Falk v. Brennan, 414 U.S. 190, 195, 94 S.Ct. 427, 431, 38 L.Ed.2d 406 (1973). An employee is employed jointly if "the employers are not completely disassociated with respect to the employment of a particular employee and may be deemed to share control of the employee, directly or indirectly_” 29 C.F.R. 791.2(b)(2) (1989). This determination of whether an entity possesses “sufficient indi-cia of control to be an ‘employer’ ” is essentially a factual inquiry. Boire v. Greyhound Corp., 376 U.S. 473, 481, 84 S.Ct. 894, 899, 11 L.Ed.2d 849 (1964).