It is widely accepted that if you start a sole proprietorship and work for yourself, you are not your own employee.  The fact that you are solely in control of your work, and solely benefiting from the revenue it generates means that you are not vulnerable to the perverse incentives built into an employer/employee relationship.  Now how does that apply when two people own, manage, and work for their own business?   How about when 500 people own, manage, and work for their own business?   Business owners can certainly be considered employees of their own business, 1 so the question is: what are the qualities of a co-ownership relationship that create or do not create an employer/employee relationship?  How do you prove that there is no master-servant relationship between a group and its members?

In many cases, people will want to restructure relationships and activities such that they are, for employment law purposes, considered partners to an enterprise they are working for. 2  Consider, for example, the two cases studies described in the first section of this e-resource library – that of the cooperative grocery and the family farmers.   If the cooperative grocery were structured in such a way that everyone has meaningful control over the cooperative and their own realm of work, then they might succeed in arguing that an employment relationship does not exist.  In the case of extended family members that farm together, there would likely be no employer/employee relationship if the family members adopt an agreement that demonstrates that everyone participates meaningfully in ownership, management, profit, and losses of the farm.  In the case of friends helping friends harvest crops, there would likely be no employer/employee relationships if the farmers form a business partnership for the purposes of jointly harvesting and marketing the vegetables from multiple farms.

What is a Master-Servant Relationship?

Anyone who grapples with this issue must become familiar with the various factors courts consider in determining whether a master-servant relationship exists, and with how each factor has been examined.  Again, this varies from jurisdiction to jurisdiction, and from statute to statute within a single jurisdiction.  What follows is a relatively short examination of this question. The sharing economy movement could benefit from a much more detailed analysis of cases that have examined this question nationally.

Strong precedent was created in answer to the question when the U.S. Supreme Court decided the case of Clackamas Gastroenterology Associates, P.C.  v. Wells in 2003. 3  There, the Court was asked to decide whether the Americans with Disabilities Act applies to working shareholders of a small professional corporation.  The Court in Clackamas cited a prior case in explaining that “‘when Congress has used the term “employee” without defining it, we have concluded that Congress intended to describe the conventional master-servant relationship as understood by common-law agency doctrine.'” 4  The Court then adopted the Equal Employment Opportunity Commission’s (EEOC) guidelines for determining when a master-servant relationship exists (numbers added):

“We are persuaded by the EEOC’s focus on the common-law touchstone of control, see Skidmore v. Swift & Co., 323 U. S. 134, 140 (1944), and specifically by its submission that each of the following six factors is relevant to the inquiry whether a shareholder-director is an employee:

1)     Whether the organization can hire or fire the individual or set the rules and regulations of the individual’s work

2)     Whether and, if so, to what extent the organization supervises the individual’s work

3)     Whether the individual reports to someone higher in the organization

4)     Whether and, if so, to what extent the individual is able to influence the organization

5)     Whether the parties intended that the individual be an employee, as expressed in written agreements or contracts

6)     Whether the individual shares in the profits, losses, and liabilities of the organization.” 5

It remains to be fully seen whether Clackamas can or will always be relied upon to make similar arguments under state employment laws and under other types of employment statutes, such as FLSA.  Hundreds of cases have now cited or relied on Clackamas in making determinations of who is an employee, and many cases prior to Clackamas examine the status of partners in enterprises of all sizes.

The list of criteria weighed in the determination of partner status varies from case to case, and some courts place heavy weight on factors that other courts ignore.  Another case that summarized the factors courts consider is Simpson v. Ernst & Young, decided by the 6th Circuit in 1996.  The court looked at the application of ADEA and ERISA to a partner in large accounting firm, and named the following factors as relevant to the determination:

1)     the right and duty to participate in management

2)     the right and duty to act as an agent of other partners

3)     exposure to liability

4)     the fiduciary relationship among partners

5)     use of the term `co-owners’ to indicate each partner’s `power of ultimate control’

6)     participation in profits and losses

7)     investment in the firm

8)     partial ownership of firm assets

9)     voting rights

10) the aggrieved individual’s ability to control and operate the business

11) the extent to which the aggrieved individual’s compensation was calculated as a percentage of the firm’s profits

12) the extent of that individual’s employment security

13) and other similar indicia of ownership. 6

Does it Matter How You Remove a Partner?

The question of “whether the organization can hire or fire the individual,” named as a factor in Clackamas, is somewhat odd, if you think about it.  Almost no organization or partnership would tie its own hands in such a way to make it impossible to remove someone.  Thus, it appears that courts are not so much interested in “whether the organization can hire or fire the individual,” but how the organization can hire and fire the individual.  It seems that courts are primarily looking to see whether the organization can hire and fire people at will and whether there are barriers to simply firing someone.

For example, in a 7th Circuit case that examined a four-member partnership, the court thought it was significant that the “partnership agreement allowed for [the partner’s] involuntary termination only by a two-thirds vote of the general partners, meaning that the other three had to agree unanimously to remove him.” 7  Thus, the fact that a vote of a significant number of partners is required seems to help satisfy courts in examining this factor.  Most co-owned business entities are structured in such a way as to require a super-majority vote prior to removing any member, shareholder, or partner.  Typically, there is not one person empowered with the right to hire and fire co-owners, which lends to the argument that there is no employment relationship.

Does Size of the Group Make a Difference?

It appears that size of a group will matter to the extent that, under some courts’ analyses, the group size affects whether partners take part meaningfully in management.  However, a case that has acted as very strong precedent in the partnership question is Wheeler v Hurdman, 8 decided by the 10th Circuit in 1987.  There, the court held that a 502-member accounting firm was a partnership for employment law purposes. The court even acknowledged that size shouldn’t be a determining factor because “large partnerships may operate more democratically overall than small partnerships, which are frequently vulnerable to domination by a single partner or a small group of partners.” 9

How Much Control Must Each Partner Have?

The Wheeler court examined application of FLSA, Title VII, and ADEA to partner in the firm, and ultimately relied on these factors in making the decision that she was a bona fide partner:

1) participation in profits and losses

2) exposure to liability

3) investment in the firm

4) partial ownership of firm assets

5) voting rights

6) position under the partnership agreement

7) position under partnership laws. 10

Note that the Wheeler court placed very little emphasis on the factor of management and whether the partner is actually able to control the workings of the firm.  The court even noted that the practical needs of the business may result in partners giving up a certain amount of control over the day-to-day, and abdicating such control to managers, teams, or committees.  The court essentially recognized the following practical reality: Any time a group of people voluntarily works together, each individual gives up a certain amount of control to the group or to members of the group, and that this would be true in all cases but “sole proprietors and a limited number of dominant partners.” 11

In Fountain v. Metcalf, Zima & Co., the court focused on certain voting rights as the indicator of control, and not on the actual realities of management in the firm.  There the court ignored the argument that a managing partner was running the firm autocratically, and focused instead on the fact that the plaintiff partner had a right to vote his thirty-one percent ownership on member/shareholders’ amendments to the agreement, on admission of new member/shareholders, on termination of relationship with member/shareholders, on draws, and on distribution of profits and income.” 12

In contrast to Wheeler and Fountain, however, other courts have focused heavily on the issue of control, and found that factor to be a deal-breaker.  For example, the court in Caruso v. Peat, Marwick, Mitchell & Co. 13 examined the employment status of a partner in a 1350-member accounting firm, and the court looked at three primary factors:

1) the extent of ability to control and operate the business

2) the extent to which compensation is calculated as a percentage of the firm’s profits

3) the extent of employment security.

On the question of “ability to control and operate the business” in the Caruso case, it was significant that the firm was “managed by a board of directors separated from plaintiff by six levels of hierarchy” and that the plaintiff tended to seek approval of management-level partners in decisions about his own work. 14  The court held that the plaintiff was, indeed, an employee.

The Court in Clackamas also focused heavily on the question of control and common law definitions of the master-servant relationship:

“At common law the relevant factors defining the master-servant relationship focus on the master’s control over the servant. The general definition of the term “servant” in the Restatement (Second) of Agency § 2(2) (1957), for example, refers to a person whose work is “controlled or is subject to the right to control by the master.” See also id., § 220(1) (“A servant is a person employed to perform services in the affairs of another and who with respect to the physical conduct in the performance of the services is subject to the other’s control or right to control”).  […] We think that the common-law element of control is the principal guidepost that should be followed in this case.  […] A court should examine “whether shareholder-directors operate independently and manage the business or instead are subject to the firm’s control.” […] “[i]f the shareholder-directors operate independently and manage the business, they are proprietors and not employees; if they are subject to the firm’s control, they are employees.” 15

These cases beg the question: How does this apply in situations where neither the worker controls the business nor the business controls the worker?  What if a firm were divided into distributed and semi-autonomous circles of governance and operation, applying a governance model such as Holacracy 16 or dynamic governance? In such a case, each individual may have only a small amount of control over the entire firm, but may have a great deal of control and voice in the particular realm of that individual’s work.  In the example of a large consumer-owned grocery cooperative, such a governance model could lend strongly to the argument that each member does indeed have a large amount of control in management – not necessarily control over the cooperative as a whole, but over the member’s own work.

These cases also beg the question: How does this apply in situations where each member has a significant amount of control by virtue of a decision-making process that mandates the use of a consensus process or super-majority voting? In some worker collectives, decisions are made by unanimous consent, which means that every member has an enormous amount of control.  Even in the absence of a unanimity requirement, each member still has a large amount of control if the organization uses a decision-making process that requires that proposals be considered and modified until they are satisfactory to each member.  In such a process, everyone has the opportunity to be heard, and may even have the power to block proposals. Arguably, every member still has a large amount of power even if the organization can take a super-majority vote as a fall-back plan in the case of a deadlock.  Even without a consensus process, organizations that require major decisions be made by super-majorities, arguably, give every member a large amount of control over what happens, in that joining with a large enough minority can overturn a decision.

Does Choice of Entity Matter?

In some jurisdictions and under some statutes, courts have leaned heavily toward the assumption that shareholders of corporations are employees when they work for the corporation they co-own. 17  This assumption has actually caused a great deal of stress for California worker cooperatives, 18 which are generally formed as cooperative corporations.  If there is an assumption of an employer/employee relationship from the moment of incorporation, this means that cooperatives must have enough start-up capital to pay members minimum wage, obtain workers compensation, and so on.  The Catch-22 is that most workers cooperatives have great difficulty raising start-up capital, due to the reluctance of banks to loan or of individuals to invest.  This means that most worker cooperatives must bootstrap their business, and work for little or no pay in the beginning, like most partners to a start-up business do.  Some worker “cooperatives” choose to act as partnerships or limited liability companies (LLCs) 19 in order to avoid application of employment law to members.

The simple fact of incorporation, in the eyes of some courts, is enough to establish a presumption that the shareholder is separate from the entity, and therefore cannot be considered a partner.  The court in Wheeler seemed to imply that formation of an actual partnership under a state’s partnership statute was a significant factor.   That court also acknowledged that ERISA does not apply to partners in a business partnership, but indicated that it may apply to them if they form a corporation, thus indicating that the simple act of incorporating can make all the difference as far as determining an employment relationship.  The court explained:

“In United States v. Empey, 406 F.2d 157 (10th Cir.1969), this court found that a partner was not an employee for the purpose of certain tax statutes; therefore, it was necessary under the tax laws for professionals to incorporate in order to obtain retirement plan benefits available to corporate employees.” 20

However, other courts, including the Supreme Court in Clackamas, have rejected the de facto assumption that the formation of a corporation should determine employment status.  In Clackamas, the Court rejected the Ninth Circuit’s reliance on a Second Circuit decision that held that

“the use of any corporation, including a professional corporation, ‘precludes any examination designed to determine whether the entity is in fact a partnership.’ 271 F. 3d 903, 905 (2001) (quoting Hyland v. New Haven Radiology Associates, P. C., 794 F. 2d 793, 798 (CA2 1986)). It saw ‘no reason to permit a professional corporation to secure the `”best of both possible worlds”’ by allowing it both to assert its corporate status in order to reap the tax and civil liability advantages and to argue that it is like a partnership in order to avoid liability for unlawful employment discrimination.’” 271 F. 3d, at 905. 21

The Clackamas Court’s rejection of the presumption serves as powerful precedent, which may even serve to relieve California worker cooperatives of the automatic presumption of employment relationships.  Many lower courts have also found that a “partnership” relationship exists even when the entity is a corporation. 22  The court in Godoy v. Rest. Opportunity Ctr. of New York, Inc. also held that the members of a worker-owned cooperative were “partners,” in spite of the fact that they were working under a corporation. 23  Many courts may ultimately consider the incorporation status of an entity as relevant, but as only one factor among many in determining whether an employment relationship exists.

The lesson here is: If you and colleagues are planning to work for your own corporation without treating yourselves as employees, tread very carefully.  Look up the cases in your own jurisdiction to see if they have applied the presumption of employment to people that work for their own corporations.  Also look to whether cases in your jurisdiction have relied on Clackamas.  Thanks to Clackamas, we now have a very strong argument that the fact of incorporation should not matter, or, at the very least, that it is only one of many factors to be weighed.

Does it Matter Whether You Operate a Business?

Employment relationships can be found even when there is no enterprise involved.  For example, nannies and full-time housekeepers for individual households typically need to be treated as employees.  How would employment law apply if a group of people rents a house together and calculates each tenant’s share of the rent with consideration to the number of hours each tenant contributes to household chores, child care, maintenance work, and so on?  If one household member gets a large break on rent in exchange for taking on a large share of the work, should that household member be treated as an employee of the others?  This is, in fact, an incredibly important question relevant to student housing cooperatives and other shared housing arrangements all over the country. 24

Even when no entity exists and even when there is no actual enterprise involved, one could argue that the cases that examine partnerships are highly relevant to this question. The cases are essentially asking the question of whether a master-servant relationship exists, and the relationships of tenants in a housing cooperative can be examined with the same criteria. One might argue that if the tenants enter into a co-tenants’ agreement that provides for collective decision-making and so long as all tenants are involved in and mutually benefiting from the household work, then there is no master-servant relationship.


  1. See Goldberg v. Whitaker House Cooperative, Inc., 366 U.S. 28, 33 (1961) stating that “{t}here is nothing inherently inconsistent between the coexistence of a proprietary and an employment relationship. If members of a trade union bought stock in their corporate employer, they would not cease to be employees within the conception of this Act. For the corporation would ‘suffer or permit’ them to work whether or not they owned one share of stock or none or many.”
  2. See Godoy v. Rest. Opportunity Ctr. of New York, Inc., 615 F. Supp. 2d 186, 194 (S.D.N.Y. 2009), explaining that “status as a general partner carries important economic reality as well. Employees do not assume the risks of loss and liabilities of their employers; partners do…. Other common characteristics of partnerships are profit sharing; contributions to capital; part ownership of partnership assets, including a share of assets in dissolution of the enterprise; and the right to share in management subject to an agreement among the partners. These are economic realities, and no definition of ‘employee’ is co-extensive…. When individuals combine to carry on a business as partners all these factors introduce complexities and economic realities which are not consonant with employee status.”  
  3. Clackamas Gastroenterology Associates, PC v. Wells, 538 U.S. 440 (2003)
  4. Clackamas Gastroenterology Associates, PC v. Wells, 538 U.S. 440, 452 (2003), citing Nationwide Mut. Ins. Co. v. Darden, 503 U. S. 318, 322-323 (1992).
  5. Clackamas Gastroenterology Associates, PC v. Wells, 538 U.S. 440, 449-450 (2003), citing Equal Employment Opportunity Commission, Compliance Manual §§ 605:0008-605:00010, § 605:0009 (2000).
  6. Simpson v. Ernst & Young 100 F.3d 436, 443-444 (6th Cir. 1996)
  7. Solon v. Kaplan, 398 F. 3d 629, 633 (7th Cir. 2005)
  8. Wheeler v. Hurdman, 825 F. 2d 257, (10th Cir. 1987)
  9. Wheeler v. Hurdman, 825 F. 2d 257, 273, (10th Cir. 1987)
  10. Wheeler v. Hurdman, 825 F. 2d 257, 276, (10th Cir. 1987)
  11. Wheeler v. Hurdman, 825 F. 2d 257, 273, (10th Cir. 1987)
  12. Fountain v. Metcalf, Zima & Co., PA, 925 F. 2d 1398, 1401 (11th Cir. 1991)
  13. Caruso v. Peat, Marwick, Mitchell & Co., 664 F. Supp. 144, 149-150 (S.D.N.Y 1987).
  14. Caruso v. Peat, Marwick, Mitchell & Co., 664 F. Supp. 144, 150 (S.D.N.Y 1987).
  15. Clackamas Gastroenterology Associates, PC v. Wells, 538 US 440, 448  (2003), quoting the EEOC’s Amicus brief in the case.
  16. See “Organization Evolved:  Introducing Holacracy,” by Brian J. Robertson, August 6, 2009, available at
  17. For an examination of this issue, see McGinley, Ann C.  “Functionality or Formalism – Partners and Shareholders as Employees under the Anti-Discrimination Laws” 57 S.M.U. L. Rev. 3 (2004)
  18. See Neil A. Helfman’s piece “The Application of Labor Laws to Workers’ Cooperatives,” December 1992, available at  – Mr. Helfman writes: “As the law presently stands, it looks at form over function. The fact of incorporation may have more bearing upon the determination of an employment relationship, than the actual relationship between the parties. A worker in an incorporated cooperative who has managerial authority, for example, may be considered an employee just because the business is incorporated; while a junior partner of a thousand-partner accounting firm, who is under the control of others, is not. The rationale for this distinction is not entirely clear; one explanation is that in the former, worker members are providing services to an entity. For example, in the Matter of Construction Survey Cooperative (Case No. T-62-3) (1962)) before the California Unemployment Insurance Appeals Board, members of a workers’ cooperative were held not to be employees. In that case, member workers received compensation on the basis of their contributed labor. By common consent, the activities of the cooperative were directed by a manager, although ultimate authority for managerial decisions rested with the membership.  The appeals board found the workers to be principals of the cooperative, and held that under California law that it was incompatible for them to be employees of their own organization. When presented to the Employment Development Department, EDD representatives took the position that if the entity were incorporated, workers should be considered employees even if no other facts had changed.”  (As a source of this info, Mr. Helfman cites: Interview at California Employment Development Department (E.D.D.), November4, 1991, with David Johnson (Senior Tax Counsel), Terry Savage (Section Chief Auditor and attorney), and Noreen Vincent (tax auditor).
  19. Note that, for employment law purposes, LLCs and partnerships are generally treated the same.
  20. Wheeler v. Hurdman, 825 F.2d 257, 268 (10th Cir. 1987); see also Yates v Hendon 541 U.S. 1 (2004) holding that a doctor that was the sole shareholder of a professional corporation could qualify as participant/employee for purposes of ERISA, as the plan also covered one of his employees. 
  21. Clackamas at 443.
  22. See EEOC v. Dowd & Dowd, Ltd., 736 F.2d 1177 (7th Cir. 1984).
  23. Godoy v. Rest. Opportunity Ctr. of New York, Inc., 615 F. Supp. 2d 186, 195 (S.D.N.Y. 2009)
  24. By the way, there are other questions related to this example, including: Must the working tenant report the value of the rent discount as income?  Must the tenant group report the value of the labor as rental income?  Is the value of the work considered part of rent for the purposes of rent control ordinances?