Antitrust and Unfair Competition Law
Competition: SPRING 2023, Vol 33, No. 1
Content
- 133 Years Young: Sherman Act Section Two Keeps Up With Big Tech
- Antitrust Restoration From California Anchored By a New Monopolization Synthesis
- California Should Amend the Cartwright Act To Address Single-firm Monopolization
- Competition Beyond Rivalry: Adapting Antitrust Merger Review To Address Market Realties
- Dormant Commerce Clause: a Potential Brake On State Antitrust Legislation
- Executive Committee
- Message From the Advisors
- Message From the Editor
- Over-prescription Is Bad Medicine: the Case Against a Knee-jerk Revision of Antitrust Injury
- Should California Adopt Revisions Proposed By Congress and the New York State Legislature To Address Single-firm Conduct?
- Table of Contents
- Technological Monopolies, Innovation, and the Personal Freedom To Form Businesses: Like Oil and Water?
- The Adaptable Antitrust Laws
- The Risks of Requiring California-specific Merger Approvals
- Updating the Cartwright Act For the Twenty-first Century: Allowing Antitrust Claims For Unilateral Conduct
- Why Has California Waited So Long To Enact Its Own Merger Review Law?
- Restrictions On Worker Mobility and the Need For Stronger Policies On Anticompetitive Employment Contract Provisions
RESTRICTIONS ON WORKER MOBILITY AND THE NEED FOR STRONGER POLICIES ON ANTICOMPETITIVE EMPLOYMENT CONTRACT PROVISIONS
By Donald J. Polden1
I. INTRODUCTION
There is increasing concern in the United States about the difficulties that American workers are facing including concerns about limits on worker job mobility and how those limits affect workers’ wages and compensation, employee benefits, and finding their next job.2 Workers believe the challenges they face in changing jobs for better salary and benefits should be addressed so they have greater job mobility and opportunities for advancement. Workers are subject to restraints on job mobility across the spectrum of jobs, including those in executive-level positions as well as minimum-wage employees. Workers complain that they are not constrained in getting better jobs by their experience, training, or education but rather by their employers’ policies and employment practices, especially their growing use of employment contracts with restrictive practices that limit workers’ abilities to seek better work.3
This Article argues that California too should re-examine its laws concerning restrictive provisions in employment contracts and arrangements for their impacts on labor markets, job mobility, and wages. California is perhaps the most restrictive of states in prohibiting the enforcement of employer restraints such as non-competes. While current California statutory policy on non-compete provisions is seemingly clear, there is evidence that it is being evaded by employers who impose the noncompetition restrictions, even if unenforceable. California employers further restrain worker mobility through imposition of other restrictive provisions—such as no-shop clauses, no-hire provisions, confidentiality provisions—that are not explicitly covered by California’s statutory policy against restraints on mobility but nonetheless are widely used to hamper worker mobility.4
[Page 128]
This needed re-examination of the state’s policy to limit enforcement of such restrictions in the workplace coincides with a national awareness that such employment restrictions harm workers and employers and should be curbed or banned outright.
States have passed legislation to limit the constraints that employers impose on their workers that make job mobility more difficult or even impossible. For example, California, for more than a century, has legislatively proscribed enforcement of employer restraints on worker mobility, especially covenants not to compete.5 Similarly, Massachusetts and some other states have recently enacted legislation that curtails the use of non-competes except in narrowly specified circumstances, such as protection of trade secrets or commercial confidential materials.6Restrictive covenants are widely used in the United States and have been justified as necessary to protect employers’ interests but, as many studies discussed below conclude, they often have the effect of impeding the ability of workers to move to another employer in the same line of work. But there is growing evidence that these statutory limits on use of employee non-compete provisions are not completely effective in preventing employers from imposing non-compete restrictions, nor for that matter in preventing employers from utilizing many other types of restrictive contractual provisions on their workers. For example, while California and North Dakota refuse to enforce employer non-compete provisions, an empirical study showed that a high percentage (approximately 19%) of those states’ employers include non-compete provisions in their standard employment agreements.7Another study concluded that "[t]he use of non-competes is so pervasive that even volunteers in non-profit organizations, in states that do not even enforce them, are asked to sign away their post-employment freedom."8 Employment experts agree that there has been a dramatic increase in the use of non-compete clauses in employment contracts, including one research finding that those clauses are found in somewhere between 20% and 50% of all employment contracts, including those of minimum wage workers.9
Federal agencies also have been energetic in using the antitrust laws to prevent competitor-employers’ collusive conduct that constrains workers’ ability to move jobs, such as employer use of "no-poach" and "no-hire " agreements where they agree to not hire each other’s employees.10In 2016, the federal agencies, the Federal Trade Commission (FTC) and U.S. Department of Justice Antitrust Division (DOJ), issued a strong guidance for human resources officers, including threats of criminal enforcement, to avoid collusive conduct that suppresses worker wages and job mobility.11Recently, the FTC launched a rulemaking proceeding aimed at employer use of non-compete provisions in employment contracts.12 The agency asserted the need for federal involvement in this area by noting the national economic effect of non-compete and other restrictive provisions in employment contracts. According to the FTC, more than one in five American workers (totaling about 30 million workers) are bound by non-compete clauses the effect of which is to lower wages for workers subject to the non-competes and to workers not directly subject to them but who are harmed nonetheless.13 The agency also contends that these contractual provisions harm other employers who would otherwise hire workers that are handcuffed by the non-compete provisions imposed by their current employer.14
Employer conduct that constrains worker wages and mobility have traditionally been evaluated under two legal regimes—one involving employment contract provisions traditionally reviewed under state contract law and the other involving anti-competitive conduct prohibited by federal and state antitrust laws. While seemingly distinct forms of analyzing worker restraints, recent cases, agency actions, and scholarly works are re-examining economic and business foundations of these employment practices and contract provisions for their competitive effects both in markets for labor and in industries where such restrictive provisions are common.15 With better understanding of the economic effects of use of restrictive employment practices on worker mobility, courts and legislatures are beginning to re-examine their policies and laws
[Page 129]
that permit employers to impose restrictions that inhibit or eliminate workers’ job opportunities.
The first part of the Article discusses the current employment environment that permits employers to restrict the mobility of their workers through a broad range of restrictive employment conditions and terms frequently imposed by employers. The second part of the Article examines efforts at enforcement of restrictive provisions by employers and recent federal and state cases and law aimed at protecting worker mobility. The part also considers a revealing example of an employment situation in which an employer aggregated several restrictive contract clauses into one contract that stifled, and were clearly intended to stifle, workers’ ability to seek employment in the industry. The case demonstrates how, often, employment contracts contain a bundle of provisions that limit what workers can do during employment and after and that cumulatively inhibit their mobility to the next, better paying job. The example concerns an industry specific type of employment agreement that is apparently widely used in the entertainment industry, the so-called "Hollywood contract,"16and the Article examines those contracts with an aggregation of restrictive clauses to determine their effect on labor competition in the industry.
The third and final part of the Article examines the normative implications of efforts to constrain employee job mobility in California. It considers the broader reaches of constraining employee mobility and suggests a regime whereby employers are permitted only narrowly-drawn restrictions on their workers’ opportunities for job movement while also permitting employers to negotiate for greater workforce stability. The Article, following recent California Supreme Court holdings, recommends that California lawmakers provide clearer guardrails on restrictive covenants in employment relations and increase penalties for employer use of most restrictive provisions. It also recommends that lawmakers should strengthen courts’ ability to review employer use of restrictive covenants for the likelihood of competitive harms under the state antitrust statute.
II. RESTRICTIONS ON WORKER MOBILITY: USE OF NON-COMPETE AND OTHER RESTRICTIVE PROVISIONS
This part of the Article provides a background on the history and legal rules of restrictive provisions in employment contracts and then considers the effects of restrictive employment contract provisions on labor markets. Restrictive employment covenants are generally considered to be restraints of trade and analyzed under a common-law reasonableness standard that evaluates their beneficial effects mainly from the employers who wish to control their labor. Until recently they have not been examined for their competitive effects in markets for labor, for effects in the economy and for concerns that their use actually impedes entrepreneurship and innovation.17 However, as the Article describes, that is changing.
A. POLICY PERSPECTIVES ON RESTRICTIVE EMPLOYMENT COVENANTS
The use of covenants not to compete traces its history to English common law which recognized that some restrictions on competition were appropriate when there were business justifications for the new buyer or former employer to be concerned about subsequent competition from the purchaser of his business or a former employee.18 These concerns led to the recognition of noncompetition agreements at common law. The principal concerns were two-fold: First, non-compete clauses were justified because the former employee or former owner will use trade secrets, goodwill, and special knowledge related to the employer’s business and would wrongfully use the information following departure from employment.19 Second, a non-compete clause was justified as a legitimate method to permit an employer to gain back the investment it made in its employees’ training.
However, today there are growing concerns that national economic policy is being thwarted by widespread use of restrictive provisions in employment agreements and practices, business merger agreements, customer and client
[Page 130]
agreements, and others. While the reasons that employers seek to impose these restraints on their employees vary, they are almost always solely beneficial to employers; it is often difficult or impossible to discern how they are beneficial to employees. Yet, it is accurate that use of the restrictive provisions has become commonplace. For example, recent studies have shown that "bundles" or aggregations of restrictive provisions are seen in 80% of workers’ employment contracts.20 Further, research shows on a national scale, 18% of workers are bound by non-compete provisions but about 19% of workers in California and North Dakota, two states that refuse to enforce non-competes, have those provisions in their employment contracts.21The United States Treasury conducted an analysis of the use of non-compete provisions, acknowledging some of the beneficial aspects of such provisions for employers, but concluding that "a growing body of evidence suggests that non-compete agreements, as currently experienced by workers and enforced by states, are often deployed in ways that lack transparency and fairness."22
Restrictive covenants in employment agreements often contain promises or commitments not to compete with former employer (a non-compete provision), promises not to solicit or accept business from a former employer’s customers, promises not to recruit or hire away employees from a former employer (a non-solicitation provision), and/or promises not to use or disclose former employer’s commercially valuable information (a confidentiality or non-disclosure provision).23 An important justification for enforcement of non-compete and other restrictive covenants is the employers’ interest in protecting valuable trade secrets, methods of doing business, and other confidential information. Faced with strong competitive pressures in their market or industry, firms legitimately wish to protect their interests by ensuring that their employees do not take the information learned on the job, and take it elsewhere especially to competitors. Employers also are concerned about the business costs of competing for workers, for example by needing to pay higher salary or wages.24 This is a cost of doing business, not a legitimate justification for imposing employment terms that lock employees into service to one employer or curbing their ability to move to better jobs. Employers also benefit from investments in employee training, knowledge, and skills and wish to recoup those investments by keeping the employee on the job and preventing the loss of a well-trained worker. The courts have frequently been called up to balance the worker’s interest in moving to a better employment with the employer’s interest in recouping its investment in training the employee.25
A public policy evaluation of these restrictive provisions recognizes that employees should be able to pursue their occupation without hindrance and yet allow employers and employees to contract with who they please.26 However, these contractual provisions prevent a measure of competition and are therefore properly considered restraints on trade. Often state courts do not automatically reject these provisions but rather look to whether the burden imposed on the employee through these restrictions is reasonable and if it was truly designed to protect an employer’s legitimate interest in the restriction on the employee.27
There are different treatments of restrictive employment provisions amongst the states as to whether or not, and to what extent, to enforce these covenants. A few states, including California, specifically provide that covenants not to compete are not enforceable, with California going a step further and limiting the use of contractual provisions that bar former employees’ solicitation of former customers.28 California Business and Professional Code section 16600 ("Section 16600") states "every contract by which anyone is restrained from engaging in a lawful profession, trade, or business of any kind is to that extent void."29 California will still allow enforcement of some restrictive covenants in some limited exceptions articulated in Section 16600. The next section addresses the questions of enforceability of restrictive covenants.
[Page 131]
B. ENFORCEMENT OF RESTRICTIVE EMPLOYMENT CONTRACT PROVISIONS
Restrictive covenants have been regulated legislatively in most states.30 In many jurisdictions, state courts have articulated policies on their enforceability, usually applying the common-law reasonableness doctrine.31 The Restatement of Employment Law articulates a "rule of reason" standard for evaluating the reasonableness of non-compete provisions referring to the common law treatment.32 But several states are beginning to re-examine the competitive implications of restrictions on worker mobility by way of restrictive covenants. For example, some states that have routinely enforced employment contract covenants not to compete are amending their laws to curb employer restrictions most often by prohibiting or limiting the enforceability of covenants not to compete in employment arrangements.33 Further, some state attorney generals are aggressively pressuring the federal government to stop the use of non-compete agreements and provided an impetus for the FTC’s rulemaking proceeding on non-compete clauses.34The re-examination of these often longstanding state legislative and judicial policies reflects the concern that these restrictive covenants harm an important form of free market competition, namely that of competition for workers and employees.
There are signs of growing federal interest in examining the competitive and economic policies promoting the use of restrictive covenants and other forms of conduct restricting employees. The FTC has launched an investigation into the harmful effect of those restrictions on worker mobility, compensation, and long-term earnings, as well as the implications for the use of these employment devices on national economic policy.35 In furtherance of its investigation, the FTC has proposed a rule greatly limiting the broad application of non-compete provisions in employment relationships.36 The investigation and proposed rule are predicated on the agency’s finding that "[b]ecause non-compete clauses prevent workers from leaving jobs and decrease competition for workers, they lower wages for both workers who are subject to them as well as workers who are . . . [and] also prevent new businesses from forming, stifling entrepreneurship, and prevent novel innovation.37 These initiatives at the federal level are clearly advancing the White House’s launch of administrative and legislative efforts to curb policies that constrain worker mobility and inhibit competitive labor markets.38 Further, there has been an increase in legal scholarship on the harms of restraints on competition in labor markets39 and expert analysis of the restraints are advocating for limits.40 Moreover, the U.S. Department of Justice Antitrust Division has launched an energetic campaign against closely related employment restrictions—no-poach agreements and no-hire agreements.41 These agreements are usually between employers who agree not to hire each other’s workers and are considered to be illegal restraints of trade under the antitrust laws.42 Finally, there is a related and substantial increase in private class-action suits against employers utilizing these restrictive provisions that hamstring employees wishing to move to higher paying jobs which also harms their prospective employers as well as the employees.43
At the federal level, the greater emphasis on regulatory oversight of markets for labor stems from the recognition of the problem of monopsony in many markets for workers.44 The ability of employers to engage in collusive, anti competitive conduct, such as no-poach agreements, stems from the economic power that they have over workers in labor markets. While the problem of monopoly was the principal concern of Congress when it passed the Sherman Act in 1890, there has been very little recognition of the problem of buyer-side power in some markets, including workers. Restrictive contract provisions (such as non-compete clauses) can be useful to powerful buyers of worker services (take the example of home health care nurses and professionals) because they make it more difficult for workers to leave employment and begin working for another employer in the same or similar line of work. And the problem of buyer-side power in labor is compounded when the employer has power in the product market (e.g., home health care services) that expands the employer’s power to the labor
[Page 132]
market for home health care nurses. Section 2 of the Sherman Act, prohibiting monopolization of a market has an extensive history of enforcement, but that is not true with respect to monopsony markets and there are complexities with extending the reach of Section 2 to labor markets where buyers, such as employers, have market power.45
C. CALIFORNIA’S PROTECTION OF WORKER MOBILITY—BUSINESS CODE § 16600 AND FIXED-TERM EMPLOYMENT CONTRACTS
Common law authorizes, but limits, non-compete clauses and, to some degree, other restrictive terms in employment contracts. However, California has strong employment laws governing use of restrictive covenants in employment contracts and is an acknowledged leader among the states in forbidding the enforcement of non-competes.46California’s policy on other restrictive provisions-such as non-disclosure, non-solicitation, and no-hire provisions—is less clear and there is concern that the legislative and regulatory guardrails on the use of these covenants may not be enough to advance the positive, pro-worker mobility policies behind Section 16600. Since the 1800s, California labor and competition policy has limited the ability of employers to impose restrictions on their employees’ ability to move to other employment, especially a competitor of the employer. The California policy is embodied in the statutory provision in Section 16600, which provides that: "Except as provided in this chapter, every contract by which anyone is restrained from engaging in a lawful profession, trade, or business of any kind is to that extent void."47The provision, which was first enacted in 1872, provides a blunt statement of California’s policy that workers should be relatively unfettered in their decisions and efforts to seek new employment.48 In the leading case on the application of Section 16600 to employment contract restrictions, Edwards v. Arthur Anderson LLP, the California Supreme Court construed the statute expansively in the context of restraints on workers’ mobility, such as non-compete and non-solicitation clauses.49 Subsequently, the Supreme Court in Ixchel Pharma, LLC. v. Biogen, Inc.,50 a case involving an exception to Section 16600 for business transaction provisions, described the fundamental policy of Section 16600 and its decision in Edwards as follows:
Moreover, the rationale in Edwards focused on policy considerations specific to employment mobility and competition: The law protects Californians and ensures ‘that every citizen shall retain the right to pursue any lawful employment and enterprise of their choice.’ It protects the ‘important legal right of persons to engage in businesses and occupations of their choosing.’ the statute ‘evinces a settled legislative policy in favor of open competition and employee mobility."51
Further, the Ixchel Pharma court acknowledged that Section 16600’s prohibition is clear; the only exceptions to the ban on restraints are the express exemptions articulated in Sections 16601 (limited use of restrictions upon sale of goodwill of a business) and 16602 (limited use of restrictions upon dissolution of a partnership). Although the Ixchel Pharma case involved a joint venture agreement between two pharmaceutical companies rather than an employment agreement, the court concluded that the litigation involved competition-related issues and that the antitrust "rule of reason" should be applied to evaluation of the competitive effects of the arrangement.52 The court examined the agreement between the competitors as requiring one firm to not engage in business with any other entity in the development and use of a particular drug and then determine whether or not it violated Section 16600. Because the restriction on competition between the parties involved a restraint of trade (for example, concerted refusals to deal or exclusive dealing arrangements, in antitrust jargon), the court analyzed the competitive purposes and likely effects under the rule of reason was appropriate under cases interpreting the state’s antitrust law, the Cartwright Act.53 However, the Supreme Court did not perform the rule of reason analysis on the restraint, instead leaving it to the lower courts.
California and federal courts have given an expansive reading to the reach of Section 16600
[Page 133]
including its applicability to non-compete clauses. However, the settled law is less clear on the reasonableness of other restrictive covenants common in employment contracts, such as non-solicitation, confidentiality, and non-disclosure provisions in employment agreements.54 These forms of restraints on workers are common in California contracts (and most other jurisdictions) yet they evade the policy reasons favoring unfettered worker mobility. Courts tend to analyze them under the vague reasonableness standard for restrictive covenants and often fail to consider the competitive implications of the use of those types of restrictions. So summarizing, the chief problems are (1) the often vague justifications under contract law for these types of restrictive covenants, (2) what restrictions on employees in employment contracts present antitrust problems, and (3) the analytic prowess of the antitrust rule of reason in challenging restrictive covenants in employment contracts.
A recent California appellate court decision provides an example of the need for legislative and judicial action to clarify and strengthen Section 16600 and thereby provide answers to workers, courts, and employers. The case involved fixed-term contracts that included a bundle of restrictive covenants. These fixed-term provisions are also known as employment for a duration or for a term, which is simply an agreement that the employee will work for the employer, and the employer will permit the employee to work, for a stated period or duration.55Fixed-term agreements bind the employee to work for the employer for a set period of time, which benefits employers by stabilizing their workforces. In the abstract, fixed employment terms benefit employees by giving them some assurances of stability in their term of employment and salary, benefits, etc. But fixed-term agreements are frequently seen in some industries, including the Hollywood film and entertainment industry in which film and television companies retain their employees for contractually defined periods of time to prevent competitors from acquiring their talents. In the United States, these employment agreements for a stated duration are the exception to the usual, or "default," form of employer-employee relationship: employment at will.56
A problem with fixed-term employment contracts is created when the term of employment provision becomes an envelope that encases restrictive provisions that limit employees choices on new employment opportunities and constrains the labor market by preventing other employers from attracting new workers.57 A legal issue is, then, whether or not an otherwise ordinary contract term, such as a stated duration for the employment, can protect a contract containing other, perhaps multiple, restrictive provisions that together lock employees in their employment. Practically speaking, that would essentially create a "fixed term restraint of trade" or, as some scholars refer, "aggregation" or "bundle" of restricted clauses.58How does the strong public policy against restraints on employee mobility in Section 16600 apply to an agreement for a fixed-term when accompanied by multiple restraints on the employee’s mobility? A fresh look at the implications of fixed term employment contact bundled with other restrictive covenants would be helpful in creating clarity in the policy underlying Section 16600.
The recent decision was issued by a panel of the California Court of Appeal is 20th Century Fox Film Corp. v. Netflix, Inc,59 (hereinafter, 20th Century). The case involves two employees of 20th Century Fox ("Fox"), a major entertainment firm, who had signed multi-year, fixed-term employment contracts but then decided to leave Fox to work for Netflix, another major entertainment firm.60 Netflix was attempting to break into the Hollywood scene and was hiring talent to staff its new Hollywood production unit. Netflix offered both employees employment that included at least double the salary Fox was paying them. The evidence showed that the two Fox employees were paid considerably below market compensation for their job types. The employees were in breach of the Fox contracts when leaving for a new employer, but Fox did not sue the employees for breach of the term provisions. Instead, Fox sued its competitor, Netflix, for tortious interference and for violation of California law of
[Page 134]
unfair competition.61 In response, Netflix asserted affirmative defenses including that the employment contracts violated California public policy under Section 16600, that Netflix’s interference was justified, and that the contract provisions were unconscionable. Netflix also filed a cross-complaint asserting claims for unfair competition that Fox’s practices were unfair, unlawful, and fraudulent. The trial court ruled against Netflix on its complaint and on its defenses and imposed an injunction on Netflix against soliciting employees with "valid" contracts with Fox. A panel of the California court of appeals affirmed the lower court rulings.
For analytic purposes it is important to understand the terms, conditions, and requirements of the Fox contacts as they were found and examined in the 20th Century courts’ opinions. First, what provisions did the Fox contracts contain and what insights into Fox’s competitive purposes in requiring and enforcing such provisions against Netflix?
• First, the employment contracts were for two or more years duration but they could be unilaterally extended by Fox. Employees lacked the ability to extend the contract term of years on their own, only Fox could extend the term of the agreement. Fox routinely and unilaterally extended the duration of the agreements often before their expiration which raised the possibility that Fox was able to bind the employees for many years.62
• The employees’ compensation was set at sub-market levels, estimated to be below the 25th percentile of comparable or "market" salaries in the industry.63
• Many Fox employees at all levels of job types were required to sign the fixed term agreements. In fact, more than 125 Fox employees across its companies and subsidiaries signed the fixed term employment contracts.64
• The agreements contained "no shop" clauses which prohibited the employees from seeking or negotiating for new employment more than 90 days before the expiration of their current employment contract with Fox. The provision also included a "consent to injunction" provision in which employees were required to agree to accept a court injunction against them if they violated the "no shop" provision by discussing or seeking their next employment opportunity.65 Fox could agree to waive those provisions for non-specified reasons.
• The contacts included provisions prohibiting Fox employees from encouraging or soliciting other Fox employees to move to other employers for a period lasting for two years after leaving Fox employment.66
• The contracts also contained confidentiality provisions prohibiting employees from disclosing any information about Fox to others outside the company. However, it is not clear what of Fox’s assets deserved such confidentiality. The opinions of both the lower and appeals courts failed to indicate any trade secrets, commercially sensitive materials, or other Fox confidential materials that would warrant protection by a confidentiality covenant.67 Moreover, there was no indication in the record that Fox invested in training or advanced education for these employees which would suggest a legitimate purpose to recover its investment by prohibited disclosure.
• Fox executives agreed that these fixed term agreements were utilized to "lock in," "gain control of" " and maintain leverage" over its employees. Indeed, the record at the trial court failed to indicate any substantial corporate purpose for the imposition of the agreements across the workforce except to limit those employees’ ability to leave Fox for other employment.68
• Both the trial court and the appellate court emphasized that Netflix was enjoined from interfering with "valid" contracts between Fox and its employees. However, neither court defined or described would make a Fox fixed term employment contract "valid", or, conversely, what an "invalid" contract looked like.
[Page 135]
Neither the trial court nor the appellate court addressed Netflix’s argument that the employment contracts contained an aggregation of restrictive provisions that violated California’s statutory policy protecting worker mobility. Instead, the appellate court held, first, that fixed-term contracts were enforceable under California contract law.69 Then, however, the court refused or failed to consider Netflix’s arguments that collectively they were anticompetitive in purpose and, as seen in the case, effect. Netflix argued that the contractual provision were void as against public policy because the "series of interrelated contract provisions and practices to limit employment mobility . . .violated the established public policy protecting an employee’s right to move freely among jobs and employer’s right to compete for skilled employees."70 In this regard, Netflix claimed that the fixed term contracts with a constellation of restrictive provisions violated Section16600 and Labor Code section 2855, which prohibits personal services agreement for longer than seven years respectively.71 The Court of Appeals rejected Netflix’s public policy claims, finding, instead, that it was possible to sever any offending provisions, citing Armendariz v. Foundation Health Pyschcare Services, Inc.72 and Abramson v. Juniper Networks, Inc.73 Those cases were cited for the notion that contractual provisions that are considered invalid and unenforceable because they are illegal or contrary to public policy may be severed from the remainder of the contract if such severance is "in the interests of justice" and the unconscionability or illegality of the provisions does not "permeate[s] the whole contract . . ."74 Netflix claimed that at least five provisions in the Fox contracts with its employees contravened the policy of Section 16600 and that the contract as a whole, with the myriad restrictive employment provisions, was unconscionable. The court summarily dismissed concerns as to each of those provisions, with little or no analysis, and completely failed to comment at all about the most anticompetitive provision, the "no-shop" clause.75 The court’s ruling on the severability doctrine is a subject for the judgment of later courts’ analysis but it bears mentioning that if the five restrictive provisions were stricken from the fixed-term contracts there would be very little left of the contracts themselves.
The court found that the duration provision in the fixed-term agreements would benefit Fox, and ostensibly its employees, because the contracts would require periodic negotiations on employment terms when Fox unilaterally extended the duration of the agreements. According to the court, this process would establish rapport between employee and employer because of the negotiation that would take place every time Fox unilaterally extended the fix term contract.76 These fixed-term agreements would have terms set in them to last over a period of time, so there would be continuous back and forth negotiation between the employee and employer to ensure that the final terms over the fixed period of years would be favorable to both parties. According to the court theory, the frequent negotiation of contract terms would establish a trust between each other and this ensures better future relations and a more likely chance that the employee will want to continue to work for the employer.77 Factually, there was no evidence of the plausibility of this justification and no justifications were offered as to any legitimate purposes for the no-shop, consent to injunction, confidentiality, post-termination non-solicitation provisions (such as employees’ access to trade secrets or commercially confidential information, employer investment in training, etc.).
Netflix also challenged four other provisions in Fox’s fixed-term contracts According to the court, the unilateral extension of the duration of the agree at the sole option of Fox was consistent with public policy since it allowed Fox to "extend the stability and predictability of the parties’ economic relationship for a period of two years."78The court, however, failed to discuss how these restrictive provisions were beneficial to employees, if at all. In sweeping terms, the court held that Fox had "legitimate business reasons" for each of the restrictive provisions and therefore individually or collectively, they "did not rise to the level of a policy violation that rendered those agreements unenforceable."79 According to the court, those legitimate objectives of the contracts were to
[Page 136]
ensure, for Fox, "stability and predictability of its workforce.80 The court’s opinion did not analyze, consider, or even mention Netflix’s claims that the Fox contract restrictive provisions collectively constituted a violation of California public policy in Section 16600. The court also failed to consider the alternative, somewhat obvious explanations of Fox’s purpose in its widespread use of employment contracts containing an aggregation of provisions that, as its own executives observed, "locked in" Fox employees to "gain control" and "maintain leverage" over them. Fox’s desire for a "stable and predictable" workforce had obvious and predictable consequence for Fox’s employees’ compensation and for Fox’s competitors in the entertainment business who also seek to compete with Fox for talented and qualified employees in the Hollywood labor market.
D. COMPETITION NORMS FOR LABOR MARKETS IN CALIFORNIA—THE IXCHEL PHARMA CASE
The California Supreme Court in the 2020 Ixchel Pharma, LLC v. Biogen81 (Ixchel Pharma) case explicitly linked the policies of state competition law in the Cartwright Act to the policy foundations of Section 16600. The decision provides a look at the role of California state policy on business efforts to restrain competition through use of restrictive covenants. Ixchel Pharma involved an agreement between two bio-technology firms, Forward Pharma and Biogen, that resulted in Forward terminating its existing contract with Ixchel to jointly develop drugs for treatment of neurological diseases. The termination provision, according to Ixchel, was void and against California public policy under Section 16600 because there was a prospective business relationship that was thwarted by the agreement to terminate.
The Ixchel Pharma court emphasized the close relationship between the contractual provisions in Section 16600 and the state’s antitrust law:
Section 16600 appears alongside the Cartwright Act, which also employs broadly worded language to prohibit agreements in restraint of tradeSection 16600 should therefore be read in accordance with the Cartwright Act to incorporate the same rule of reason in such cases. . . . The similarities between the two statutes stretch beyond their language. They share a statutory purpose and doctrinal heritage in common law prohibitions on restraints of trade. They should therefore be interpreted together.82
The Ixchel Pharma case introduced an important element to the analysis of restrictive provisions by explicitly linking their impacts on competition in labor markets as they are affected by restrictive covenants. While the Ixchel Pharma case involved a business contractual provision rather than an employment-related provision, the court concluded that restraints in business transactions and in employment relationship are both considered "restraints of trade" under the statutory scheme of Section 16600. The Ixchel Pharma court concluded that the rule of reason, which is commonly applied to most antitrust restraints, should be applied in the case of restraints on business competition under Section 16600.83 As the court in Ixchel discussed, there are California cases that hold such agreements are per se, illegal and void rather than requiring a broader examination of their reasonableness but determined that inquiry into the reasonableness of the provisions under the antitrust rules was appropriate.
The decision in 20th Century case can be understood as reflecting 19th-century contract and tort-law principles extolling the notion of freedom to contact involving two executives who were capable of negotiating their employment contracts.84 But, the central failing of the court’s ruling is its rejection of the competition-law implications, especially that the obvious purpose of the contract was to restrict employees from seeking better employment with a competitor. The evidence showed that Fox refused to grant releases from the fix term contracts to employees "who joined competitors during the term of their agreement."85 The next section addresses this gap between the reach of the policy underlying Section 16600 and the overarching goals of the state and federal competition law and policy.
[Page 137]
III. A REAPPRAISAL OF CALIFORNIA POLICIES ON WORKER MOBILITY AND ANTICOMPETITIVE CONDUCT IN LABOR MARKETS
The Article makes the case for stronger, clearer policy-making regarding California’s stated goal of ensuring greater worker job mobility and enhancing greater competition in labor markets. Without question, California has led the country in articulating policies, and implementing them with laws and regulations, that seek to protect the ability of workers to seek better jobs and opportunities. That state policy is grounded in the research demonstrating that job mobility is essential for workers’ abilities to improve their earning potential, more vibrant (and competitive) markets for workers, and for increased entrepreneurship and innovation in product markets.86 But it also seems clear that there is more to be done, legislatively and judicially, in California to fully achieve those goals.
Non-compete and Other Restrictive Covenants in Employment Contracts. The Article presents evidence of the widespread use of non-compete provisions in California even though they are not enforceable. Almost 19% of employment contracts in the U.S. (and in California) include non-compete provisions and, nationally and in California, the percentage is almost four times higher for executive positions.87 Even unenforceable non-competes harm worker mobility.88 Further, the courts have not clearly extended the worker protections of Section 16600 to other restrictive employment terms such as no-shop, non-disclosure, confidentiality, non-solicitation and fixed term contract terms. Therefore, a re-evaluation of the scope of the policy prohibiting restraints on worker job mobility is necessary and appropriate to enhance the clarity of the state’s policy on restraining worker mobility. Legislative revisions to Section 16600 should prohibit non-compete clauses, not just make them unenforceable, and thereby increase the deterrence effect of 16600. Any revision should also clearly address the positive and negative labor and competition effects of other restrictive provisions, building on the existing policy on non-compete provisions. Legislative amendments should more precisely identify protectable employer interests for narrowly-drawn restrictive provisions but should also clearly state when such provisions fail to accommodate protectable interests of workers. This would be immeasurably helpful to California’s employers who want to protect their legitimate interests (such as protection of trade secrets that the employee used in their employment) but in a manner that doesn’t prevent their employees from improving their compensation, benefits, and access to better job opportunities.
A few states have adopted the new Uniform Restrictive Employment Agreement Act that clarifies the common-law treatment of restrictive employment provisions, including narrowing and articulating legitimate employer interests that support such restrictive contract terms.89 The uniform law goes beyond the common law (and that of Section 16600) remedy of voiding such agreements and provides that such illegal provisions are "prohibited" and subject to statutory damages (recommending $5000 per worker per violation; but noting that states can decide their own damage levels). A thoughtful re-examination of the policy and text of Section 16600 would benefit from a careful consideration of the national uniform law, perhaps adopting the provisions that fill gaps in the California law.
Further, statutory revisions are needed to bring policies on standards for enforcement of worker protection and competition norms in markets for labor into clearer focus with less indeterminacy. The opacity of California public policy favoring freedom of job mobility in Section 16600 that was thoughtfully captured by Judge O’Scannlain of the Ninth Circuit Court of Appeals in case involving the application of Section 16600 to settlement agreements:
The courts of California have not clearly indicated the boundaries of section 16600’s stark prohibition, but have nevertheless intimated that they extend to a considerable breadth. At the very least, we have no reason
[Page 138]
to believe that the State has drawn section 16600 simply to prohibit "covenants not to compete" and not also other contractual restraints on professional practice.90
More "clearly indicated boundaries" for the reach of the state policy served by Section 16600 would be beneficial for California workers and employers. Section 16600 should clearly state that it also applies to other forms of restraints on employees, such as non-disclosure, no-hire, and no-shop provisions, that serve, and often are intended to serve, to limit worker mobility away from the restraining employer to other job opportunities. It should recognize that some of those restrictive covenants can serve to protect legitimate employer interests such as protecting its intellectual property (i.e., trade secrets, confidential customer information, some goodwill with customers) and an appropriate "walking away" from a business following the sale of the business or the cessation of partnership interests. A revised (and perhaps revitalized) Section 16600 would explicitly recognize narrowly-drawn legitimate employer interests and craft them into an operational rule for defining appropriate (and restricting inappropriate) restrictive employment provisions.
Another area for further analysis is the use of aggregations of employee restrictive provisions in employment contracts and practices as was seen in the 20th Century case.91 As the appellate court’s opinion suggests, where there are aggregations of restrictive provision courts may simply examine each of the several restrictive provisions for their reasonableness but fail to evaluate the thicket of such restrictive provisions for their collective effect on employees. Perhaps obviously, the latter approach is most appropriate as aggregations of restrictive provisions in contracts exert an in terrorem effect on the affected employees.92As the Article and other commentary suggest, a comprehensive look at the employment and competition effects of such an aggregation may indicate an improper purpose and effect for the contract.
Competition Norms in Labor Markets. California should explicitly consider revisions to the Cartwright Act that would do two things: First, it should craft appropriate proscriptions on the use of restrictive covenants that restrain trade in labor markets and/or in product markets, and secondly, the legislature should address the problem of monopsony in labor markets. Buyer-side power is a significant economic problem that results in harms to workers.93However, California law does not explicitly prohibit monopolization, which may be problematic on a few levels, and therefore it does not prohibit monopsony. California law requires clear statement of state policy that the harmful effects of monopolization and monopsony harm citizens and violate state law and are prohibited by a revised Cartwright Act. The revision should identify what kinds of practices taken by a dominant firm in a seller market or a buyer market are proscribed. For example, in addition to traditionally proscribed conduct of monopolists, such as exclusive dealing or predatory pricing, the legislation could helpfully address conduct such as widespread use of non-competes or a thicket of restrictive provisions in a labor market. Further, it would be appropriate to consider the anticompetitive effects of restrictive provisions in particular industries or markets are prevalent or especially pernicious. Forexample, widespread use of non-competes in fast food or other minimum-wage labor markets, if taken by a firm with demonstrable power in the market, should constitute a violation of the Cartwright Act and likely are.94 However, a clear legislative prohibition on other restrictive covenants and provisions in employment relationships would reinforce California state policy seeking to permit unfettered job mobility and freedom.
Further, if there is no legislative appetite for re-visiting restrictions in labor markets, the California courts should address the question raised but unanswered in the Ixchel Pharma case: How should courts apply the rule of reason in cases involving restrictive employment contract terms? There are several considerations that inform this important answer. First, California antitrust law can draw meaning in its application from the federal Sherman Act,95 so it is appropriate to draw definitive concepts
[Page 139]
and processes from that statute in determining analytic tools like the rule of reason. Therefore, cases such as NCAA v. Alston,96 in which the United States Supreme Court applied the Sherman Act to conduct of the NCAA restraining economic benefits to student athletes, and the DOJ litigation involving no-poach agreements, would provide relevant guidance to California courts evaluating the legality of restraints in labor markets.
The Cartwright Act, like the Sherman Act, has been construed to prohibit restraints of trade under two forms of analysis: one treats restraints with no discernable procompetitive benefits as presumptively (or per se) illegal violations of law, and the second treats restraints with some arguable procompetitive attributes in a more comprehensively review to determine their reasonableness (i.e., the rule of reason).97 The proper analytic approach to restraints of trade under California and federal antitrust laws involves a characterization of a restraint by considering its purpose and the effects or likely effects of the restraint.98 If the characterization of the restraint leads to a conclusion that the purpose and likely effect in nearly all cases is to foreclose or prevent competition, then the restraint is considered as a presumptive, or per se, violation. If the analysis suggests that the restraint has some likely procompetitive benefits, then a fuller reasonableness analysis is conducted. It is important to point out that antitrust reasonableness analysis is markedly more rigorous and demanding on antitrust plaintiffs than the common-law reasonableness approach to restrictive covenants.
The U.S. Department of Justice Antitrust Division has been litigating cases against employers who collude with competitors to not hire each other’s employees ("no-poach" or "no-hire" agreements) as per se violations of federal antitrust law.99 That same classification of no-poach, no-hire, and no-shop agreements should be followed by California courts examining similar agreements under California law. No-poach and no-hire agreements are treated as horizontal restraints that most often lack any procompetitive attributes; they mainly eliminate competition for workers which suppresses worker wages. No-shop provisions should be treated as presumptively unlawful under California (and federal) antitrust law for the same reasons—there are no redeeming procompetitive benefits—as no-poach agreements. Even if no-shop provisions may be characterized as a vertical arrangement rather than a horizontal one, they should be found presumptively anticompetitive and illegal for the reason that they lack procompetitive justifications, almost always harm the targets-workers seeking job mobility-and, by definition, adversely affect competition for workers in the market or industry.100
Antitrust analysis of restrictive covenants would need to consider the purposes for their use in particular cases as well as the pro- and anticompetitive effects of the restrictive covenants involved. Consider non-compete provisions. Increasingly, covenants not to compete are commonly viewed in the same category as no-hire and no poach agreements because, although different in formation (non-competes are imposed by the employer and the no-poach are the product of agreement between competitors), they result in the same exclusionary outcomes for workers.101 They explicitly prevent workers from or severely constrain them in attempting to move to a better job. And, both non-competes and no-poach restraints have another anticompetitive effect: Other employers are prevented, or severely limited in their ability, to hire the restrained employees. More significantly, the employer justifications for both forms of restraints on employees are the same—to prevent other employers who may compete in the same product market (for example, media or entertainment) but do compete in the labor market from getting trained, experienced workers.102 The purpose and effect of the use of these employee restraints, in those factual circumstances, could be characterized by a fact-finder as a restraint of trade without significant procompetitive benefits and thus would violate the Cartwright or Sherman Act. However, non-compete agreements have not been treated as per se violations of the antitrust laws and DOJ is prosecuting some no-poach agreements as per se violations. Legislative guidance on proper treatment
[Page 140]
under the Cartwright Act for particularly pernicious restrictive covenants in employment arrangement would be very helpful.
The Ixchel Pharma court concluded that the appropriate analysis of anticompetitive aspects of a non-compete provision was the rule of reason in a case involving a non-compete provision in a business transaction. The same reasoning was applied by Pennsylvania Supreme Court, in Pittsburgh Logistics Systems, Inc. v. Beemac Trucking, LLC, a case involving a no-hire agreement between a logistics provider and a shipping company. The no-hire clause was considered ancillary to the services contract between the companies but found to be an unreasonable restraint of trade and therefore unenforceable.103 The court relied on similar holdings by other state courts to the effect that restraints on the mobility of workers should be analyzed and treated like agreements between competitors to not hire one another’s employees that exert a public harm therefore held illegal.104
Second, consider no-shop provisions. The no-shop clause in the Fox agreements prohibited Fox’s employees from seeking a new position until a three- or four-month period of time prior to the expiration of the employment contract. So, in a two-year fixed term agreement, Fox would have the contractual right to enjoin its employees from seeking or discussing new employment for 21 months. Moreover, any prospective suitor for that employee’s services could face a lawsuit for tortious interference with that contact if the prospective employer knows of the employee’s contractual bindings. There are few restrictive contract provisions that are facially as anticompetitive as conditioning when an employee can seek other employment. The use of "no shop" provisions have the clear, and arguably intended, purpose of preventing employees from testing their worth in the market by considering other opportunities and, reciprocally, prevents other employers from considering those workers for employment. The obvious target of these "no-shop" contract practices are the employer’s competitors in the market for qualified workers. Courts should evaluate these provisions under the per se rule because they lack any obvious redeeming value. Alternatively, no-shop provisions should be treated as restraints of trade and analyzed under a rule of reason approach like the no-hire provision in the Pittsburgh Logistics Systems, Inc. case.105
Consider treating aggregations of restrictive employment provisions as restraints of trade.
The 20th Century court failed to perform the kind of detailed consideration of the demonstrated purpose and effect of the aggregation of restrictive covenants in the Fox contracts. While it seemed clear that the underlying purpose of Fox’s imposition of an aggregate of restrictions was to prevent Fox’s competitors from gaining skilled, experienced employees, some of the restraints individually may have some off-setting procompetitive benefits. Confidentiality provisions are appropriate if an employer has intellectual property and trade secrets to protect from misappropriation by competitors. Non-solicitation provisions may be appropriate to prevent employees from soliciting customers to leave their current employer for a new venture. Fixed-term employment contracts may in many circumstances be appropriate and beneficial to both employees and employers. But the potential for anticompetitive effects from use of aggregations of restrictive clauses when there is no or very little justification is sufficient to treat them as restraints of trade and analyze them under the Cartwright Act as such.
IV. CONCLUSION
The Article encourages legislative and judicial consideration of the growing concerns about constraints on workers’ job mobility and on the state’s labor markets. The California state policy on these policy objectives is clearly stated but there is evidence that it is not being realized. The state policy of Section 16600 for unfettered worker mobility and freedom to seek new opportunities is still thwarted. This suggests, as the Article argues, that legislative re-examination of the California law, notably 16600 and the Cartwright Act would
[Page 141]
be beneficial. The California Supreme Court has indicated that antitrust law and policy should play a greater role in examining how labor and employment constraints may prevent the achievement of state policy favoring—greater opportunities for entrepreneurship and economic self-determination of workers.
——–
Notes:
1. Donald J. Polden is Dean Emeritus and Professor of Law, Santa Clara University School of Law. He expresses his appreciation to Michael Wang, a 2022 graduate of the Santa Clara Law for his very helpful research and editorial comments. In addition to teaching antitrust law courses, Professor Polden frequently writes in the area of federal antitrust law and policy and is a member of the California Bar Section on Antitrust and Unfair Competition. Previously, he tried antitrust cases in federal trial and appellate courts. He expresses his appreciation to the editors of the journal—Anu Reddy and Stephen Mclntyre—for the opportunity to submit this Article to be considered to efforts of the California Law Revision Commission’s study on state antitrust laws. Any errors and position statements are solely the responsibility of the author. The author was retained as an expert witness for Netflix in a case discussed in the article.
2. See Eric A. Posner, How Antitrust Failed Workers, 91-106 (2021)(describing the economic and legal concerns with widespread use of non-compete provisions in employment contracts). See also, for example, the President’s Executive Order on Promoting Competition in the American Economy, Exec. Order No. 14036, 86 Fed. Reg. 36987 (July 14, 2021), https://www.whitehouse.gov/briefing-room/presidential-actions/2021/07/09/executive-order-on-promoting-competition-in-the-american-economy/ (expressing concern with employment terms that hinder workers’ ability to seek new opportunities).
3. John M. McAdams, Non-Compete Agreements: A Review of the Literature (Fed. Trade Comm’n, 2019), SSRN: https://ssrn.com/abstract=3513639 ; Richard J. Pierce, Jr., The U.S. Federal Trade Commission Workshop on Non-Compete Clauses (Geo. Wash. Univ. Research Paper No. 2020-01, 2020). , https://ssrn.com/abstract=3520009. Princeton economist Alan Krueger summarized the new emphasis on protection of workers’ opportunities for job mobility: "New practices have emerged to facilitate employer collusion, such as non-compete clauses and no-raid pacts, but the basic insights are the same: employers often implicitly, and sometimes explicitly, act to prevent the forces of competition from enabling workers to earn what a competitive market would dictate, and from working where they would prefer to work." Alan B. Krueger, The Rigged Labor Market, Milkin Inst. Rev. (Apr. 28, 2017), https://www.milkenreview.org/articles/the-rigged-labor-market; see also Kenneth G. Dau-Schmidt & Jozie M. Barton, Non-Compete Covenants, in Oxford Handbook of the Law of Work (forthcoming 2023), https://ssrn.com/abstract=4342965.
4. See supra note 2 and accompanying text. On effects of restrictive provisions other than non-competes, see Camilla Alexandra Hrdy & Christopher B. Seaman, Beyond Trade Secrecy: Confidentiality Agreements That Act Like Noncompetes, 133 Yale L. J. (forthcoming 2023), https://ssrn.com/abstract=4384661; Heidi Lynne Kurter, 4 Things You Didn’t Know About Non-Disclosure Agreements, Forbes (Jan. 21, 2020 8:55 PM), https://www.forbes.com/sites/heidilynnekurter/2020/01/21/4-things-you-didnt-know-about-non-disclosure-agreements/?sh=46a3e76b7c12 (non-disclosure agreements)
5. See infra note 28 and accompanying text.
6. See infra note 32 and accompanying text.
7. J.J. Prescott et al., Understanding Noncompetition Agreements: The 2014 Noncompete Survey Project, 2016 Mich. St. L. Rev. 369, 461 (2016); see also Orly Lobel, Gentlemen Prefer Bonds: How Employers Fix the Talent Market, 59 Santa Clara L. Rev. 663, 698-99 (2020).
8. Evan Starr, The Use, Abuse, and Enforceability of Non-Compete and No-Poach Agreements: A Brief Review of the Theory, Evidence, and Recent Reform Efforts (2019), https://eig.org/wp-content/uploads/2019/02/Non-Competes-Brief.pdf.
9. Pierce, supra note 3, at 2. Arguably, the most pernicious use of non-competes occurs in saddling low-wage, fast food workers with non-competes that prevent them from working in other fast-food restaurants for two years after they depart. See Dau-Schmidt & Barton, supra note 3, at 5.
10. See, e.g., Final Judgment, United States v. eBay, Inc., No. 12-cv-5869 (N.D. Cal. Sept. 2, 2014) ECF No. 66. See generally Michael Murray, Antitrust Enforcement in Labor Markets: The Department of Justice’s Efforts, 59 Santa Clara L. Rev. 561 (2020) (Deputy Assistant Attorney General of Antitrust Division’s description of Justice Department policies and actions to prevent unlawful employer conduct that restrains employee compensation, benefits, and job mobility and violating the federal antitrust laws).
11. U.S. Dep’t of Just., Antitrust Div., & Fed. Trade Comm’n, Antitrust Guidance for HR Professionals (2016), https://www.justice.gov/atr/file/903511/download.
12. Non-Compete Clause Rule, 88 Fed. Reg. 3482 (proposed Jan. 19, 2023) (to be codified at 16 C.F.R pt. 910), https://www.federalregister.gov/documents/2023/01/19/2023-00414/non-compete-clause-rule.
13. Lina Khan, Noncompetes Depress Wages and Kill Innovation, N.Y. Times (Jan. 9, 2023), https://www.nytimes.com/2023/01/09/opinion/linakhan-ftc-noncompete.html (op-ed by the Chair of the Federal Trade Commission).
14. Non-Compete Clause Rule, 88 Fed. Reg. 3482.
15. See, e.g., Starr, supra note 7, at 3-12; Murray, supra note 9, at 572 & n.64.
16. See, Ashley Cullins, A New Era of Noncompetes May Shake Up Hollywood Contracts, Hollywood Rep. (Jan. 12, 2023 12:03 PM), https://www.hollywoodreporter.com/business/business-news/ftc-noncompetes-hollywood-contracts-1235296702/. Employment contracts used in the entertainment business often contain provisions or terms for a duration; i.e., the contract is for a stated term of time (referred to as "fixed term contracts"), often with rights of the employer to renew the duration of the contract. Because many of the contracts involve artistic performers, the contracts were considered personal service contracts for employees rendering "special, unique, unusual, extraordinary, or intellectual character." Because of abuses by the studio employers, California passed legislation that cap the total number of years to a period not exceeding seven years. Cal. Lab. Code § 2855; see Jonathan Blaufarb, The Seven-Year Itch: California Labor Code Section 2855, 6 Hasting Comm. & Ent. L.J. 653, 655-58 (1984).
17. See Posner, supra note 2, at 104-08 ("[A] noncompete may cause harm to (1) workers in the labor market to which the employees subject to the noncompete belong, (2) other labor markets that the employer draws from, and (3) the product market that the employer sells into . . .wages generally decline rather than increase in states that enforce or strictly enforce noncompetes . . . Noncompetes can further reduce labor market competition by deterring entry.").
18. For a description of the common law development of the law of non-competes in employment, see Harlan M. Blake, Employee Agreements Not to Compete, 73 Harv. L. Rev. 625, 629-47 (1960). For a comprehensive look at the contemporary law of restrictive employment practices, see Mark W. Bennett et al., Employment Relationships: Law & Practice ch. 11 (2018).
19. Posner, supra note 2, at 95-101.
20. Orly Lobel, Boilerplate Collusion: Clause Aggregation, Antitrust Law & Contract Governance, 106 Minn. L. Rev. 877, 878 (2021) (citing Natarajan Balasubramanian et al., Bundling Postemployment Restrictive Covenants: New Evidence from Firm and Worker Surveys (July 2021) (unpublished manuscript)).
21. McAdams, supra note 3, at 3.
22. Karen Dynan, The Economic Effects of Non-compete Agreements, U.S. Dept. Treasury: Treasury Notes Blog (Mar. 31, 2016), https://www.treasury.gov/connect/blog /Pages/The-Economic-Effects-of-Non-compete-Agreements-.aspx. The report of the Treasury Department can be found at U.S. Dept. Of the Treasury, Non-Compete Contracts: Economic Effects and Policy Implications (2016), https://home.treasury.gov/system/files/226/Non_Compete_Contracts_Econimic_Effects_and_PolicyJmplications_MAR2016.pdf.
23. Non-solicitation agreements include an employee agreement not to solicit other employees to leave the current employer, as well as agreeing not to start their own business to go compete against the employer. Confidentiality clauses, or non-disclosure agreements, limit what confidential information such as trade secrets, commercially sensitive materials, proprietary information and other intellectual property can be acquired and used by an employee during the term of employment and often for a certain period of years after the employment ends. Restrictive covenants are often embedded in employment contracts (usually with executive-level employees), in agreements that govern only the restrictive covenants or in workplace policies. See Bennett et al., supra note 18, at 43; see also Restrictive Covenants: What Are They and Where Can They Be Used, DLA Piper, https://www.dlapiperaccelerate.com/knowledge/2017/restrictive-covenants-what-are-they-and-when-and-where-can-they-be-used.html.
24. Posner, supra note 2, at 101-06 (describing the social and economic costs of non-compete provisions in worker contracts.).
25. See, e.g., Curtis 1000, Inc. v. Suess, 24 F.3d 941, 947 (7th Cir. 1994).
26. Stephen L. Brodsky, Restrictive Covenants in Employment and Related Contracts: Key Considerations You Should Know, A.B.A. (Feb. 8, 2019), https://www.americanbar.org/groups/litigation/committees/commercial-business/practice/2019/restrictive-covenants-employment-related-contracts/.
27. Blake, supra note 18 at 633-34 (history of use of workers’ covenants or agreements not to compete with his employer following termination of the employment relationship); Eric A. Posner, The Antitrust Challenge to Covenants Not to Compete in Employment Contracts, 83 Antitrust L.J. 165 (2020).
28. Edwards v. Arthur Anderson LLP, 44 Cal. 4th 937, 946 2008); AMN Healthcare, Inc. v. Aya Healthcare Services, Inc., 28 Cal. App. 5th 923 (2018).
29. Cal. Bus. & Prof. Code § 16600.
30. See, e.g., Mich. Comp. Laws § 445.774a(l) (restrictive covenants are enforceable if the agreement is reasonable); Tex. Bus & Com. Code. Ann. § 15.50(a) (non-compete covenants are enforceable if they are ancillary to, and the restraints do not impose a greater restriction than what is necessary to protect goodwill or a business interest) N.D. Cent. Code §§ 9-08-01 to 9-08-09 (general ban on any contracts that restrain an individual from entering into a profession, trade, or business of any kind); Okla. Stat. tit. 15, § 219A (prohibits non-compete contracts except for those written to protect the sale of goodwill of a business, dissolution of a partnership).
31. See Bennett et al., supra note 19, § 2-11, pp. 15-50 (surveying state court decisions on enforceability of restrictive covenants).
32. Restatement of Employment Law § 8.06 (Am. L. Inst. 2015). The Restatement provides that "a covenant . . . restricting the former employee’s working activities is enforceable only if it is reasonable tailored in scope, geography, and time to further a protectable interest of the employer." Id.
33. See, e.g., mass. Gen. Laws ch. 149, § 24L; D.C. Code §§ 32-581.01-.05.
34. Dallin Wilson & Erik Weibust, State Attorney General Keep Pressure on FTC to Regulate Non-Competes, Seyfarth: Trading Secrets (Dec. 13, 2019), https://www.tradesecretslaw.com/2019/12/articles/restrictive-covenants/state-attorneys-general-keep-pressure-on-ftc-to-regulate-non-competes/; Press Release, Cal. Dep’t of Just., Attorney General Becerra Renews Call for Nationwide Ban on Non-Compete Agreements, Reminds Businesses of Existing Prohibition in California (Mar. 12, 2020), https://oag.ca.gov/news/press-releases/attorney-general-becerra-renews-call-nationwide-ban-non-compete-agreements.
35. Press Release, Fed. Trade Comm’n, FTC Cracks Down on Companies That Impose Harmful Noncompete Restrictions on Thousands of Workers (Jan. 4, 2023), https://www.ftc.gov/news-events/news/press-releases/2023/01/ftc-cracks-down-companies-impose-harmful-noncompete-restrictions-thousands-workers.
36. Non-Compete Clause Rule, 88 Fed. Reg. 3482. According to the proposed rule, the FTC would prohibit employers from (i) entering into, or attempting to enter into a non-compete clause with an employee, and (ii) telling employees that they are subject to a non-compete clause. The proposal would also require employers to terminate any active non-compete clauses and inform those employees that any such contract provision is no longer valid. Id.
37. Non-Compete Clause Rulemaking, Fed. Trade Comm’n (Jan. 5, 2023), https://www.ftc.gov/legal-library/browse/federal-register-notices/non-compete-clause-rulemaking.
38. Press Release, The White House, FACT SHEET: Executive Order on Promoting Competition in the American Economy, (July 9, 2021), https://www.whitehouse.gov/briefing-room/statements-releases/2021/07/09/fact-sheet-executive-order-on-promoting-competition-in-the-american-economy/.
39. See, e.g., Lobel, supra note 20, at 87886; Hiba Hafiz, The Brand Defense, 43 Berkeley J. Emp. & lab. L. 1, 74-76 (2022); Gregory Day, Anicompeiive Employment, 57Am. Bus. L.J. 487, 520-25 (2020).
40. Chuck Loughlin, Hogan Lovells, Key Takeaways from the FTCs Noncompete Workshop (2020), https://www.hoganlovells.com/en/publications/key-takeaways-from-the-ftcs-noncompete-workshop.
41. No Poach Approach: Division Update Spring 2019, U.S. Dep’t of Just. (Sept. 30, 2019), https://www.justice.gov/atr/division-operations/division-update-spring-2019/no-poach-approach.
42. See, Donald J. Polden, Restraints on Workers’ Wages and Mobility: No-Poach Agreements and the Antitrust Laws, 59 Santa Clara l. Rev. 579, 588-609 (2020).
43. Lobel, supra note 6, at 694-96; see also Rachel Argenbright Rioux, Note, The Necessity for Employer Liability in Unenforceable Non-Compete Agreements, 86 UMKC L.R. 995 (2018).
44. Posner, supra note 2, at 71.
45. Id. at 74-75.
46. Lina Khan, supra note 13.
47. Cal. Bus & Prof. Code § 16600.
48. Ixchel Pharma, LLC v. Biogen, Inc, 9 Cal. 5th 1130 (2020) (providing extensive history of 16600).
49. Edwards v. Arthur Andersen LLP, 44 Cal. 4th 937 (2008) (striking down provision in employment agreement prohibiting employees for one year after termination of employment from soliciting former clients).
50. 9 Cal. 5th 1130.
51. Id. at 1158 (citing to Edwards, 44 Cal. 4th 937).
52. Id. 1160-62. Referring to prior appellate court decision refusing to apply antitrust norms to the interpretation of contract provisions under 16600, the court held: "The similarities between the two statutes stretch beyond their language. They share a statutory purpose and doctrinal heritage in common law prohibitions on restraints of trade. They should therefore be interpreted together." Id. at 1160.
53. Cal. Bus. & Prof. Code §§ 16700-16770. The Cartwright Act draws upon the language of the federal Sherman Antitrust Act to prohibit restraints of trade, In re Cipro Cases, 61 Cal. 3d 116 (2015).
54. See, e.g., AMN Healthcare, Inc. v. Aya Healthcare Servs., Inc., 28 Cal. App. 5th 923 (2018) (holding that confidentiality and non-disclosure provisions which prevented employees from soliciting other employees to leave employer for one year after termination of employment is void under policy of 16600); Dowell v. Biosense Webster, Inc., 179 Cal. App. 4th 564 (2019) (Eighteen-month noncompetition and non-solicitation agreements violate state restraint-of-trade statute by effectively preventing former employees from practicing chosen profession); Application Grp., Inc. v. Hunter Grp., Inc., 61 Cal. App. 4th 881 (1998) (refusing to enforce noncompetition agreements used by Maryland employer to prevent employees from seeking jobs with California-based competitor; agreements found to violate California statutory policy in 16600); Golden v. Cal. Emergency Med. Physicians., 782 F.3d 1083 (9th Cir. 2015) (settlement agreement that included provision barring physician’s employment by staffing consortium of medical facilities).
55. Restatement of Employment Law § 2.02 (Am. L. Inst. 2015) ("The employment relationship is not terminable at will by and employer if . . . (a) an agreement between the employer and the employee provides for (i) a definite term of employment . . ."). The author served as a member of the Consultative Group for the Restatement of Employment Law.
56. See Bennett et al., supra note 18, at ch. 2, pp. 3-8 (describing the widespread application of the employment at will doctrine in most U.S. jurisdictions; it is considered the "default rule").
57. Twentieth Century Fox Film Corp. v. Netflix, Inc., No. B304022, 2021 WL 5711822, at *1-4 (Cal. Ct. App. Dec. 2, 2021) (the court decided that the opinion was not to be considered "published").
58. Lobel, supra note 6, at 666-85.
59. Twentieth Century Fox, 2021 WL 5711822.
60. The court stated that Netflix approached multiple Fox employees and about a dozen similarly left Fox for Netflix. Id. at *1-2.
61. Id. at *5-6 (discussing California’s Unfair Competition Law set forth in Cal. Bus. & Prof. Code § 17200).
62. Id. at *1-3.
63. Id. at *2.
64. Id.
65. Id.
66. Id.
67. Id.
68. Id.
69. Id., at *8.
70. Id. at *6.
71. Section 2855 of the California Labor Code addresses the length of personal services contracts and prohibits such agreements that lock up workers for seven or more years. In the Twentieth Century Fox case, Netflix argued that due to Fox’s ability to, some contracts could extend beyond the limit in section 2855.
72. 24 Cal. 4th 83, 124-25 (2000)
73. 115 Cal. App. 4th 638, 658 (2004).
74. Twentieth Century Fox, 2021 WL 5711822, at *7.
75. See id. at *9-10 & n.10.
76. Id. at *8-9.
77. Id.
78. Id. at *9.
79. Id.
80. Id.
81. 9 Cal. 5th 1130 (2020).
82. Id. at 1159-60 (citations omitted).
83. Id. at 1160-62.
84. Twentieth Century Fox Film Corp. v. Netflix, Inc., No. B304022, 2021 WL 5711822, at *9 (Cal. Ct. App. Dec. 2, 2021) ("[T]he evidence showed that both [Fox employees] were sophisticated business executives who negotiated their fixed-term employment agreements with Fox at arm’s length."). Indeed, it is facially incongruent that so called "sophisticated business executives" would negotiate a substantially below market compensation arrangement that would include several restrictive provisions, both pre-termination and post-termination, without additional consideration for accepting those handcuffs on job mobility.
85. Id. at *2.
86. See Posner, supra note 2, at 101-06; McAdams, supra note 3, at 3-6, Pierce, supra note 3, at 2-4; see also Starr, supra note 7, at 7-12.
87. See Mark J. Garmaise, Ties That Truly Bind: Noncompetition Agreements, Executive Compensation, and Firm Investment, 27 J.L. econ. & Org. 376 (2011) (reporting that 70.2% of large national employers bound their employees with non-compete clauses); Jonathan M. Barnett & Ted Sichelman, The Case for Noncompetes, 87 U. Chi. L. Rev. 953, 980 (2020) (reporting on studies finding that 58-62% of firms headquartered in California have non-competes in their executive employment contracts).
88. See Dau-Schmidt & Barton, supra note 3, at 12 ("[E]ven unenforceable non-competes have a negative impact on employees’ job search and wages.").
89. Stewart J. Schwab, Regulating Noncompetes Beyond the Common Law: The Uniform Restrictive Employment Agreement Act, 98 Ind. L.J. 275 (2022).
90. Golden v. Cal. Emergency Med. Physicians., 782 F.3d 1083, 1093 (9th Cir. 2015).
91. See, Micha Mitch Danzig et al., California Court Deals Blow to Employee Mobility, MINTZ (Dec. 14, 2021), https://www.mintz.com/insights-center/viewpoints/2226/2021-12-14-california-court-deals-blow-employee-mobility (criticizing the appellate court’s treatment of the Netflix’s public policy arguments, including that "the contracts here seemed to create one-sided indentured servitude for as long as Fox wished to employ each individual." (emphasis added)).
92. Posner, supra note 2, at 102-05 (arguing that most restrictive clause are not negotiated with the employee); DauSchmidt & Barton, supra note 2, at 5 ("Even if a non-compete is unenforceable, its existence can raise barrier to job search because most employees don’t understand that many non-compete are not enforceable").
93. Posner, supra note 2, at 61-75.
94. See id. at 75.
95. See In re Cipro Cases I & II, 61 Cal. 4th 116, 137 (2015); Flagship Theaters of Palm Desert, LLC v. Century Theaters, Inc., 55 Cal.App.5th 381, 400(2020); see also John M. Landry & Kirk A. Hornbeck, One Hundred Years in the Making: The Cartwright Act in Broad Outline, 17 Competition: J. Antitrust & Unfair Competition L. Section St. Bar Cal. 7, 18-19 (2008).
96. Nat’l Collegiate Athletic Assn v. Alston, 141 S. Ct. 2141 (2021).
97. See, e.g., Ben-E-Lect v. Anthem Blue Cross Life & Health Ins. Co, 51 Cal. App. 5th 867 (2020).
98. See, e.g., Flagship Theaters, 55 Cal. App. 5th at 400.
99. See Murray, supra note 9, at 572-76.
100. It has been argued that restrictive covenants such as non-competes in employment contracts should be characterized as vertical restraints such as exclusive dealing and evaluated under the antitrust rule of reason. However, exclusive dealing occurs by agreement between parties in up-stream or down-stream relationships—buyers and seller. However, employees don’t fit this mold because they often don’t have the independence of vertically related businesses to simply move to another vendor (or customer). Further, widespread use of non-competes in markets (like fast food or home medical care) would exert the same effect as widespread no-poach agreements. Posner, supra, note 2, at 108-12. This is a reason that the FTC is considering a rule to ban non-competes as an "unfair method of competition" under section 5 of the Federal Commission Act. See Dau-Schmidt & Barton, supra note 3, at 10.
101. Posner, supra note 2, at 109.
102. See, e.g., Lydia DePillis, Noncompete Clauses Get Tighter, and TV Newsrooms Feel the Grip, N.Y. Times (Apr. 3, 2023), https://www.nytimes.com/2023/04/03/business/economy/noncompete-clauses-broadcast-news.html (describing the widespread industry use of non-compete clauses on many employees in the news business).
103. Pittsburgh Logistics Sys., Inc. v. Beemac Trucking, LLC, 249 A.3d 918, 936 (Pa. 2021) (the no hire agreement "creates a likelihood of harm to the public. . . impairs the employment opportunities and job mobility (of restricted former employees) [because] the effect of its enforcement. . . would have deprived its former employees of their current jobs and livelihoods [and therefore] undermines free competition in the labor market.").
104. See, e.g., Heyde Cos., Inc. v. Dove Healthcare LLC, 258 Wis.2d 28 (2002) (agreement between provider or rehabilitation services and a health care facility that hospital would not hire its physical therapists from the rehab services provider was found to be unreasonable); Common Tech. Sys. v. Densmore, 583 N.W.2d 125 (S.D. 1998) (South Dakota has a limitation on employee non-compete provisions similar to California’s Section 16600). But see Therapy Servs., Inc. v. Crystal City Nursing Ctr., Inc., 239 Va. 385 (1990).
105. See Henry Greco, On the Interpretations of No-Hire Provisions in Pennsylvania—The Case for Utilizing Federal Antitrust Law, 83 U. Pitt. L. Rev. 661, 667-71, 696-701 (2022).
[Page 142]