No Poaching Allowed: Antitrust Issues in Labor Markets

The article below addresses another aspect of employees moving from one company to another — agreements between two or more competitive corporations not to solicit, or hire, each other’s employees.  The article below first appeared in Antitrust, Vol. 26, No. 3, Summer 2012. © 2012 by the American Bar Association. Reproduced with permission. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association.

Michael Lindsay

Michael Lindsay

Kate Santon

Kate Santon

The article below first appeared in the ABA’s Magazine, Antitrust, Vol. 26, No. 3.  The required ABA admonition follows.

Finally, while we generally try to stay away from footnoted articles on this Blog, and attempt to write articles that are easily understood by those who have, or do not have, a formal legal education,  here we have reproduced the article in its orginal form.  Portions of it may sound a bit like ‘inside baseball’ for the non-antitrust practitioner.  But, we are confident that you will understand the key points and the practical suggestions at the article’s conclusion.

We hope you find the article helpful.

 

Dorsey

Employees can be a company’s most important asset, but unlike other assets, employees can walk. When an employee walks out the door, the employee takes with him the human capital that results from the joint investment of the employee (through his time and effort) and the employer (through its formal and informal training programs––and through the compensation paid to the employee). The employee might also take business relationships and confidential information. With employee mobility at an all-time high (n.1), employers might ask how far they can go by agreeing among themselves to limit the kinds of competition they will engage in for employment services.

The (unsurprising) answer is “not very far.” In 2009, the U.S. Department of Justice investigated allegations of such agreements among a number of high-tech companies that restricted their ability to hire each other’s employees. (n.2) The investigation resulted in civil complaints against a number of high-tech companies, (n.3), followed by settlement agreements (n.4),—and a federal class action. (n.5)  The popular press certainly saw no difference between hiring-restriction agreements and the more familiar (and illegal) practice of price fixing. One headline referred to it as “Tech-Recruiting Collusion,” (n.6), and many others referred to the underlying agreements as “anti-poaching.” (n.7) Perhaps “poaching” was already used to describe competition for employees in the high-tech world, but if not, it came to have that meaning. (n.8)  The resulting DOJ case, along with a follow-on federal class action have highlighted the antitrust aspects of employer competition for employees. So what are the antitrust guides that employers should follow in the non-unionized economy? (n.9)

Traditional Post-Employment Covenants

Employers traditionally have used both incentives and disincentives to keep valued employees from leaving. “Incentives” include compensation (salary, bonuses, benefits packages, and employee-specific perks) and working conditions. “Disincentives” include post-employment conditions in an employment agreement that make leaving relatively less attractive. For example, an employment agreement might include a non-compete covenant (employee agrees not to pursue a similar profession or trade in competition against the employer within a stated geography for a specified time), a customer non-solicitation covenant (employee agrees not to solicit employer’s customers, but is otherwise free to compete, and is free to make unsolicited sales to employer’s customers), a co-employee non-solicitation covenant (employee agrees not to solicit former co-employees to leave the company), and confidentiality agreements (employee agrees not to take, use, or disclose employer’s confidential information).

Each of these “disincentives” is an agreement between the employer and employee, or in antitrust-speak, between the buyer and seller. As vertical agreements, these employment covenants are unlikely to have any antitrust implications. (n.10)

Employment agreements, however, are not a complete or perfect solution to the problem of protecting an investment in human capital. State law regulates the enforceability of post-employment restrictive covenants, and some states (notably California) severely limit the enforceability of noncompete and non-solicitation agreements. Especially in a state that limits enforceability of vertical agreements (that is, of post-employment covenants with the employee), employers might look to other means to protect their interests.

“No-Switching” Agreements Between Competitors

An employer will be particularly concerned if its employee joins or forms a competitor of the employer—that is, “switches” to a competing employer. After all, that is where the employee’s use of knowledge and relationships can hurt the former employer most. Consequently, an employer might be tempted to reach an agreement with its competitors not to hire each other’s employees. After all, the employer thinks, this reaches the same result as a perfectly lawful (in most states) noncompete agreement with the employee, and it avoids the messy and fast-paced lawsuit for a temporary restraining order or preliminary injunction for breach of employment covenants. And doesn’t Section 6 of the Clayton Act make clear that antitrust laws do not apply to markets for human labor? (n.11)

Agreements Between Competitors. Some employers have tried this approach––and failed. Section 6 clearly does not provide an antitrust exemption for a “buyer cartel” for labor services. For example, in Nichols v. Spencer International Press, Inc., (n.12) two competitors in the sale of encyclopedias and other reference books had a “no switching” agreement.  According to the complaint, under their agreement, these competing firms “refused to permit employees or other sales personnel to go to work for competitors . . . and . . . refused to hire employees and sales personnel with any other defendant or of any other competitor . . . .” (n.13)  The court rejected the argument that Section 6 immunized this agreement, but it did so primarily on the grounds of effects in the downstream market—that is, the market for encyclopedias. According to the court, “agreements among supposed competitors not to employ each other’s employees not only restrict freedom to enter into employment relationships, but may also, depending upon the circumstances, impair full and free competition in the supply of a service or commodity to the public.” (n.14)  The court held that it could not determine as a matter of law “that the effect of the ‘no switching’ agreement . . . upon the business of supplying encyclopedias and reference books is so negligible that the agreement is not a restraint of trade in such products.” (n.15) Consequently, the court remanded for determination of the reasonableness of the agreement—but apparently with respect to its effects in the downstream market. (n.16)

Vertical Agreements and Copperweld Defenses.  An agreement might have horizontal aspects but be justifiable on other grounds. For example, in Williams v. I.B. Fischer Nevada, (n.17) the court considered an agreement under which management employees of a fast-food franchise could not move from one Jack-In-The-Box franchisee to a different franchisee without the first franchisee’s consent. The court said that the purpose of this no-switching agreement was “to prevent the franchises from ‘raiding’ one another’s management employees after time and expense have been incurred in training them.” (n.18)

Two factors distinguish the Jack-in-the-Box agreement from the Nichols agreement. First, although the Jack-in-the-Box agreement restricted competition between franchisees (that is, horizontal competition for skilled labor), the agreement was actually vertical—between the franchisor and each franchisee. The district court did not make much of this fact in and of itself. Instead, it focused on a second factor—that the agreement was within a single enterprise, the Jack-in-the-Box franchise system. (The agreement did not prohibit management employees from seeking management positions outside the Jack-in-the-Box system.)

Both the district court and the Ninth Circuit used a (n.19) analysis to determine that the franchisor and franchisees were incapable of conspiring with one another. (n.20)  The Ninth Circuit concluded that the franchisor and franchisees had a sufficient unity of purpose to qualify as a single entity under Copperweld. The district court explained in more detail that “the franchisor does everything in its power to minimize competition and promote uniformity between franchises” and does so to help both the franchisor itself and its franchisees. (n.21) The franchisor and franchisees all benefit from “uniformity of quality food and service” that results in an “enhanced reputation” and increased business. The larger sales volume helps the franchisees directly and helps the franchisor sell more franchises at higher prices, as well as earning more in franchise fees that provide a percentage of the restaurants’ gross sales.

Whether the I.B. Fischer Nevada case’s Copperweld analysis survives American Needle (n.22) is debatable. (n.23)  The factors that the court described to support its Copperweld conclusions, however, would also serve in a rule of reason analysis (possibly even sustaining the agreement under a “quick look”). (n.24)

Market Definition Upstream and Downstream. Market definition will obviously be important in any rule of reason case, and the I.B. Fischer Nevada district court opinion is also illuminating for its discussion of market definition. The plaintiff argued for a relevant market that was “a specialized labor market for Jack-in-the-Box management employees,” in which the no-switching agreement “limits the quantity of the labor supply,” which in turn “affects the price of labor and drives up the price of the products offered to the public” (n.25) in plaintiff’s proposed downstream market for “Jack-in-the-Box products.” By contrast, the defendants argued for a relevant market of the labor market for restaurant managers in general” and a downstream market for fast-food products. The court correctly expressed doubts about market definitions designed around a single firm’s inputs and outputs. (n.26)

Other Ancillary Agreements. A no-switching agreement might also be justified when it is adopted as part of a corporate spin-off. For example, in Eichorn,(n.27) the Third Circuit addressed agreements entered between and among a parent company and its affiliates regarding their conduct following a proposed a spin-off. Following the transaction, the companies would not hire each other’s employees who made more than $50,000 annually. Employees of one of the former affiliates were disadvantaged (relative to their pre-spin position), because the no-switching agreement prevented them from taking advantage of certain pension-eligibility rules at their former employer. The Third Circuit recognized this as a rule of reason case and found it relatively easy to justify the restraint as ancillary to the sale of a business.

Although the Eichorn opinion does not clearly say so, the post-spin former affiliates were probably not significant competitors of each other, at least not as of immediately after the spin-off. Nevertheless, there was an efficiency-based justification for the restraint—namely, making the sale transaction more feasible. Perhaps some lesser restraints might have served the purpose (for example, a limitation on the pension benefits rules that made a return to the former parent or other affiliate more attractive relative to joining an entirely unrelated company). The rule of reason can certainly accommodate that fact-intensive inquiry, but only because there is a main (and procompetitive) agreement to which the restraint is ancillary. What if there is no such agreement?

“No-Poaching” Agreements Between Firms

In September 2010, the DOJ filed a complaint against several high-tech companies alleging that their agreements relating to hiring practices violated the antitrust laws. (n.28)  These agreements were clearly not pure no-switching agreements. Rather, they restricted use of a particular recruiting tool: cold-calling each other’s employees. (The agreements apparently did not apply to employee-initiated contacts.) In other words, they were closer to non-solicitation agreements than to no-hire agreements. The DOJ filed a second complaint in December 2010 against two high-tech companies, however, and this complaint alleged not only a ban on cold-calls, but two other provisions that went even further: an agreement to notify the other firm when making an offer to an employee of that other firm, and an agreement not to counteroffer above the first offer. (Apparently the current employer could match the would-be employer’s offer but could not exceed it.) (n.29)

Per Se Violations. In both cases the DOJ challenged the agreements as per se violations, and it is easy to see why. If the facts are recast as agreements among competing sellers—that is, agreements not to make sales calls on each other’s customers, to give notice when offering to sell a product to each other’s customers, and not to offer a lower price than what the new seller was offering—then no one would have any doubt that the allegations described a per se violation. (n.30) As the Adobe Competitive Impact Statement put it, “There is no basis for distinguishing allocation agreements based on whether they involve input or output markets. Anticompetitive agreements in both input and output markets create allocative inefficiencies.” (n.31)

The Lucasfilm Competitive Impact Statement was even stronger, because the alleged agreement was both more extensive and, in the DOJ’s words, more “pernicious.” The DOJ restated the same principle: “Antitrust analysis of downstream customer-related restraints applies equally to upstream monopsony restraints on employment opportunities.” (n.32)  In denying a motion to dismiss the complaint in the follow-on private class action, the district court declined to determine whether to apply the per se rule or the rule of reason. But the court did hold that “it is plausible to infer that even a single bilateral agreement would have the ripple effect of depressing the mobility and compensation of employees of companies that are not direct parties to the agreement.” (n.33)

Direct Restraint of Labor Markets. The DOJ correctly focused on the direct restraint that the agreements imposed on the labor markets, rather than any effects in downstream markets. Although the firms were to some extent competitors with each other, the real vice was the effect on the labor market. By reducing competition for the skill-set of highly trained technical employees, the agreements “diminish[ed] potential employment opportunities for those same employees.” (n.34)  Employees were “deprived of information and access to better job opportunities.” (n.35)

Ancillary Restraints. The DOJ acknowledged that the companies had “legitimate collaborative projects” and “extensive business relationships,” (n.36), which creates the possibility that some form or amount of recruiting restraints might be justified. But as the DOJ observed, the rule of reason requires that the restraint be designed to protect the legitimate business interest (n.37)––and here the restraints were not limited in some way related to any specific collaboration, such as being limited to a product group involved in a particular collaboration. (n.38) Although the DOJ does not suggest this, the no-cold-call restraint may also have been too narrow as well as too broad. If a collaboration gives one company an opportunity to get to know—and potentially hire away—a critical employee, then a ban on initiating a hiring dialogue with that employee may be insufficient to protect the employer’s interest (and thus induce the employer to participate in the collaboration in the first place).

Practical Implications

A counselor can give the following practical advice to an employer facing the prospect of employee departures:

Where possible, rely on employment covenants. A noncompete agreement—where enforceable, and subject to geographic and time reasonableness limitations—can keep employees from leaving for a competitor. The employer should identify employees (by individual or by group) for whom noncompete agreements are particularly important. In most states, the noncompete agreement either must be signed when the employee is offered the position or must be supported by independent consideration.

Consider unilateral practices—or at least the effects of one’s own overly aggressive recruiting. An employer should consider whether its own recruiting practices might start a bidding war, either in the employer’s own industry or in some broader labor market. An employer might, for example, have its own unilateral “do not call” list and impose a similar restriction on any recruiting firms that it engages. The employer should document that this is in fact a unilateral policy, adopted and enforced without communications with any other company that competes in the employer’s upstream or downstream markets. Additionally, the employer should periodically confirm that the ban on communications with other firms has been honored.

Consider agreements that provide for efficient dispute resolution. There is a difference between an agreement not to solicit or hire each other’s employees and an agreement to streamline resolution of claims that an employee’s hiring violates a noncompete (for example, an agreement to arbitrate any such disputes). This will have the greatest value only in industries with relatively extensive noncompetes and relatively few industry players.

Adopt a payback-for-training requirement. Before an employee begins a new job, the employer may ask the employee to sign a contract in which the employee agrees to repay training costs up to a set amount if the employee resigns or is fired within a certain time period following the employee’s start date. To ensure enforceability, an employer should draft terms that are reasonable regarding the repayment amount and the length of time an employee must remain at the company to escape the repayment obligation. (n.39)

Consider a term-of-years agreement or employee retention agreement. With valid consideration, such as a sign-on bonus, an employee retention agreement can be a very useful retention tactic. If the employee leaves before the end of the agreed-upon length of employment, a portion (or all) of the sign-on bonus must be repaid. With sufficient consideration, and as long as the sign-on bonus and time period are reasonable in amount, the contract would likely be enforceable.

Think about employees whose retention might be jeopardized through a joint venture or other collaboration. Before entering any kind of collaboration––whether with a collaborator or not, an employer should ask itself whether the collaborator might use the collaboration for recruiting—and how the employer might protect itself.

Conclusion

Retaining key employees and minimizing undesired employee turnover is increasingly challenging. Traditional mechanisms are certainly imperfect tools, but illegal agreements with other firms are not an improvement. Nevertheless, an employer that takes a careful approach—and takes the time to consider specific vulnerabilities and to craft narrow solutions—stands a better chance of lawfully retaining valuable employees.

(Footnotes)

1See U.S. Dep’t of Labor Bureau of Labor Statistics, News Release, Employee Tenure in 2010, available at http://www.bls.gov/news.release/pdf/tenure.pdf (in the professional and technical services industry, the average employee tenure has only increased slightly over the last few years from 3.6 years in 2004 to 4.0 years in 2010).  The problem of employee mobility is particularly acute in high tech industries.  The average annual employee turnover rate for the professional and technical services industry in 2011 was 27% while the average among all industries was 16%.  See U.S. Dep’t of Labor Bureau of Labor Statistics, Economic News Release:  Job Openings and Labor Turnover Survey News Release, available at http://www.bls.gov/news.release/archives/jolts_03132012.htm.

2 Although the DOJ did not file its complaint until 2010, the investigation reportedly began in 2009.  See Cecilia Kang, Federal Antitrust Probe Targets Tech Giants, Sources Say, WASH. POST, June 3, 2009, http://www.washingtonpost.com/wp-dyn/content/article/2009/06/02/AR2009060203412.html.

3 United States v. Adobe Sys., Inc., Case No. 1:10-cv-01629 (D.D.C. filed Sept. 24, 2010), available at http://www.justice.gov/atr/cases/f262600/262654.htm; United States v. Lucasfilm Ltd., Case:  1:10-cv-02220 (D.D.C. filed Dec. 21, 2010), available at http://www.justice/gov/atr/cases/f265300/265395.htm.

4 See U.S. Dep’t of Justice, Justice Department Requires Six High Tech Companies to Stop Entering into Anticompetitive Employee Solicitation Agreements (Sept. 24, 2010), available at http://www.justice.gov/opa/pr/2010/September/10-at-1076.html.

5 Consolidated Amended Complaint, In re High-Tech Employee Antitrust Litigation, Docket No. 11-CV-2509-LHK (N.D. Cal.).

6 Dawn Kawamoto, Justice Department Wraps Up Tech-Recruiting Collusion Probe, DAILY FIN., Sept. 7, 2010, http://www.dailyfinance.com/2010/09/17/justice-department-wraps-up-tech-recruiting-collusion-probe/.

7 See, e.g., Chloe Albanesius, Judge Refuses to Toss Anti-Poaching Case Against Apple, Google, PCMag, Apr. 20, 2012, http://www.pcmag.com/article2/0,2817,2403308,00.asp.  The same “no poaching” language, of course, can be used to describe market allocation schemes among sellers.  See, e.g., Brief for Appellee United States of America, United States v. Rose, No. 05-10447, at 9 (5th Cir. 2005) (“The three companies allocated customers by agreeing that each company would have “protected accounts”—choline chloride buyers/users secretly assigned to only one manufacturer.  The two companies not assigned to that customer would not attempt to poach its business by offering lower prices.)”(emphasis added), available at http://www.justice.gov/atr/cases/f211800/211870.pdf.

8 See, e.g., Jay Yarrow, Hulu, Facebook Open Offices in Seattle—Now in Poaching Distance of Microsoft, Amazon, S.F. CHRON., Sept. 16, 2010 (“Microsoft and Amazon are about to get some competition for the best and brightest [workers] the Northwest has to offer.”), http://www.sfgate.com/cgi-bin/article.cgi?f=/g/a/2010/09/16/businessinsider-suddenly-every one-wants-to-be-in-seattle-2010-9.DTL.

9 The non-statutory labor exemption permits employers to take collective actions that would not otherwise be permitted.  See, e.g., California ex rel. Harris v. Safeway, Inc., No. 08-55671, 2011 WL 2684942 (9th Cir. July 12, 2011) (revenue-sharing provision in an agreement among competing grocers to provide strategic support to one another if employees went on strike at or picketed any one of the chains did not violate Section 1 of the Sherman Act under a “per se-plus” or a “quick look-minus” analysis.) We do not suggest that no-poaching agreements would be lawful in the unionized part of the private-sector economy—only that this issue is beyond the scope of this article.

10 See, e.g., Marsh USA Inc. v. Cook, 354 S.W.3d 764 (Tex. 2011).

11 15 U.S.C. § 15 (“The labor of a human being is not a commodity or article of commerce.”).

12 371 F.2d 332 (7th Cir. 1967).

13 Id. at 333.  The duration of the “non-hire” period was six months.  Id. at 333–34.

14 Id. at 336.

15 Id. at 337.

16 In fairness to the court, the plaintiff seems to have focused its case on the effects in the downstream market, rather than arguing that the labor restraint was a per se violation.

17 794 F. Supp. 1026 (D. Nev. 1992), aff’d, 999 F.2d 445 (9th Cir. 1993).

18 794 F. Supp. at 1029.

19 Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752 (1984).

20 794 F. Supp. at 1030–31; Williams v. I.B. Fischer Nevada, 999 F.2d 445, 447–48 (9th Cir. 1993).

21 794 F. Supp. at 1032.

22 Am. Needle, Inc. v. Nat’l Football League, 130 S. Ct. 2201 (2010) (holding that a joint venture of the teams in the NFL and the NFL itself were not immune from the Sherman’s Act’s reach because the teams in the NFL and the NFL did not possess either the unitary decision-making quality or the single aggregation of economic power characteristic of a single entity).

23 See, e.g., Timothy Cornell, The Majority-Owned Venture:  Does Copperweld Provide Immunity After American Needle?, METROPOLITAN CORP. COUNSEL, Nov. 2, 2010, available at http://www.metrocorpcounsel.com/articles/13199/majority-owned-venture-does-cooperweld-provide-immunity-after-american-needle; See also In re Florida Cement and Concrete Antitrust Litig., 746 F. Supp. 2d 1291 (S.D. Fla. 2010).

24 James A. Keyte, American Needle:  A New Quick Look for Joint Ventures, ANTITRUST, Fall 2010, at 48.

25 794 F. Supp. at 1033.

26 Id. at 1034.

27 Eichorn v. AT&T Corp., 248 F.3d 131 (3d Cir. 2001).

28 United States v. Adobe Sys., Inc., Case No. 1:10-cv-01629 (D.D.C. filed Sept. 24, 2010), available at http://www.justice.gov/atr/cases/f262600/262654.htm.

29 United States v. Lucasfilm Ltd., No. Case:  1:10-cv-02220 (D.D.C. filed Dec. 21, 2010), available at http://www.justice.gov/atr/cases/f265300/265395.htm.

30 The allegations are only that—allegations.  The DOJ cases were settled without admissions, and the federal class action has not proceeded significantly past the motion to dismiss stage.  See Order Granting in Part and Denying in Part Defendants’ Joint Motion to Dismiss; Denying Lucasfilm Ltd.’s Motion to Dismiss, In re High-Tech Employee Antitrust Litigation, No. 11-CV-02509-LHK (N.D. Cal. Apr. 18, 2012) [hereinafter High-Tech Rule 12 Order].

31 Competitive Impact Statement, available at http://www.justice.gov/atr/cases/f262600/262650.htm [hereinafter Adobe CIS].

32 Competitive Impact Statement, available at http://www.justice.gov/atr/cases/f265300/265397.htm [hereinafter Lucasfilm CIS].  The DOJ’s citation to United States v. Brown, 936 F.2d 1042 (9th Cir. 1991) as “limited to an input market (the procurement of billboard leases),” however, is inapt.  Although billboards are certainly an input in the overall marketing process, they are also a direct means of competition for sales.  Thus, an agreement to limit competition for billboards is more closely related to competition in the downstream market than would be an agreement to limit some other input.

33 High-Tech Rule 12 Order, supra note 30, at 20–21.

34 Adobe CIS, supra note 31, at 2.

35 Lucasfilm CIS, supra note 32, at 3.

36 Adobe CIS, supra note 31, at 9.

37 Id. at 9 (“Restraints that are broader than reasonably necessary to achieve the efficiencies from a business collaboration are not ancillary and are properly treated as per se unlawful.”).

38 Id. (“The agreements were not limited by geography, job function, product group, or time period.  This overbreadth and other evidence demonstrated that the no cold calling agreements were not reasonably necessary for any collaboration and, hence, not ancillary.  The lack of reasonable necessity for these broad agreements is demonstrated also by the fact that Defendants successfully collaborated with other companies without similar agreements, or with agreements containing more narrowly focused hiring restrictions.”).

39 Loraine Lawson, Indentured ITs:  Should Your Employees Sign Training Contracts?, TECHREPUBLIC, June 5, 2000, http://www.techrepublic.com/article/indentured-its-should-your-employees-sign-training-contracts/1061380.

Dorsey & Whitney

Dorsey is a business law firm, applying a business perspective to clients' needs. We make it our first priority to know the context in which you do business - your market, your competitors, your industry.

You may also like...