To avoid any suggestion to the contrary: This is not a political commentary, nor is it intended to express an opinion about whether noncompetes are good or bad, or when (if ever) they should be used. Similarly, I will not elaborate on my oft-expressed view that sweeping political decisions with potentially significant ramifications for people, companies, and the economy should be based on more than political expediency, soundbites, and preliminary, nuanced, inconclusive, and sometimes inconsistent studies.
Instead, I’m going to discuss President-elect Biden’s announcement about his policy toward noncompetes and suggest the steps companies should be taking now to protect themselves. So, feel free to skip down to “What to do about it” if you just want to see my recommendations.
With that out of the way…
The looming problem: a possible federal ban on (most) noncompetes.
President-elect Biden’s website announced the following plan for noncompetes:
Eliminate non-compete clauses and no-poaching agreements that hinder the ability of employees to seek higher wages, better benefits, and working conditions by changing employers. In the American economy, companies compete. Workers should be able to compete, too. But at some point in their careers, 40% of American workers have been subject to non-compete clauses. If workers had the freedom to move to another job, they could expect to earn 5% to 10% more – that’s an additional $2,000 to $4,000 for a worker earning $40,000 each year. These employer-driven barriers to competition are even imposed within the same company’s franchisee networks. For example, large franchisors like Jiffy Lube have no-poaching policies preventing any of their franchisees from hiring workers from another franchisee. As president, Biden will work with Congress to eliminate all non-compete agreements, except the very few that are absolutely necessary to protect a narrowly defined category of trade secrets, and outright ban all no-poaching agreements.
There are a lot of issues to unpack. I’m going to discuss eight of them. (There are more; but these eight are the most important.)
First, lumping noncompetes and no-poach agreements together is mixing apples and oranges. While noncompetes have some impact on a person’s job mobility, they reflect a balance between the impact on mobility and allowing companies to freely and efficiently share information and customer goodwill with their employees (and secure the benefits that flow from that) without fear that the information or goodwill will be taken to and used at a competitor who did not have to invest the resources to develop it. In contrast, in most instances, no-poach agreements (which are not agreed upon by the parties directly affected by them) just hinder movement of employees without protecting corresponding legitimate business (and societal) interests.
Further, noncompetes do not typically prevent employees from joining a competitor; rather, they restrict the role that the employee can undertake for the competitor. See Cleaning up overly broad noncompetes: the “Janitor Rule.” In contrast, no-poach agreements often effectively prevent many employees from working for the other contracting companies in any capacity.
Conflating these two types of agreements also obscures that the research on noncompetes is actually quite nuanced about whether and how noncompetes hinder (predominantly low-wage) employees from seeking higher wages or obtaining better benefits or working conditions (more on that below) and that there is a dearth of research concerning no-poach agreements.
Given that the two types of agreements are fundamentally different and that no-poach agreements are rarely justified, I focus only on the noncompete portion of the announcement. (To be clear, I am not suggesting that no-poach agreements are never justified. They may or may not be in certain limited contexts, such as the sale of a business or in the franchise context. I offer no opinion on that here.)
Second, the statement that “40% of American workers have been subject to non-compete clauses” comes from a well-regarded study (“Noncompete Agreements in the U.S. Labor Force” (2015)) by some of the top researchers in the field, Professors Evan Starr, J.J. Prescott, and Norm Bishara. The takeaway is that a lot of people have (at some point in their career, not necessarily at the time of the survey) signed a noncompete. But, how does that translate into actual impact on them? There are many ways to answer that, and they run the gamut from beneficial impacts (see below), to no impact on career plans, to restrictions on permissible conduct at a new job, to taking a job that is different from a preferred job, to taking a temporary career detour. In my experience, most matters fall closer to the “restrictions on permissible conduct at a new job” side of the spectrum and are resolved (relatively) amicably.
Third, the assumption that “[i]f workers had the freedom to move to another job, they could expect to earn 5% to 10% more – that’s an additional $2,000 to $4,000 for a worker earning $40,000 each year” is based on the U.S. Department of the Treasury March 2016 report, “Non-compete Contracts: Economic Effects and Policy Implications” (the “Treasury Report”). However, while the Treasury Report relies in part on objective, verifiable resources (including my 50-State Noncompete Chart), it also relies on “[r]esearch on non-competes [that] is still at an early stage.” Treasury Report, at p. 11. Indeed, as the Treasury Report acknowledges, “The literature on the effect of non-competes on wages is small, consisting largely of case studies, surveys of specific professions (e.g., electrical engineers), theoretical papers, and a recent analysis based on a broad online survey.” Treasury Report, at p. 18.
Of course, if noncompetes were in fact the root cause of comparatively depressed wages, one would think that California – the state trotted out as the poster child for banning noncompetes – would have the highest median income (all things being equal). But it doesn’t. It has the 10th. The states (and D.C.) with the top ten median incomes in 2020, in order, are:
- Washington, DC ($82,604)
- Maryland ($81,868)
- New Jersey ($79,363)
- Hawaii ($78,084)
- Massachusetts ($77,378)
- Alaska ($76,715)
- Connecticut ($76,106)
- New Hampshire ($74,057)
- Virginia ($71,564)
- California ($71,228)
Instructively, every other state (and D.C.) on the list above California (i.e., with a higher median income) enforces noncompetes.
In fact, Massachusetts – which is usually trotted out as the whipping boy to California’s success in Silicon Valley – has a median income that is $7,150 higher than California (i.e., roughly 10% higher). And, Massachusetts’ cost of living index (131.6) is 20 percentage points lower than California’s (151.7).
Nevertheless, people still reflexively point to the rise of Silicon Valley and the (somewhat exaggerated) fall of Massachusetts’ Route 128 as reflecting the different noncompete enforcement regimes in the states. (California bans noncompetes, whereas Massachusetts enforces them.) Although the fallacy of this assumption is manifold, it’s beyond the scope of this article. (Anyone interested in the topic can read “Misconceptions In The Debate About Noncompetes” (July 8, 2019) (a free version of which is available here: “Correlation Does Not Imply Causation: The False Comparison of Silicon Valley and Boston’s Route 128” (July 9, 2019)) and “The Case for Noncompetes” (July 2020) by Professors Jonathan Barnett and Ted Sichelman.)
Fourth, the President-elect’s announcement groups all workers together, disregarding that the data (preliminary as it is) shows that while some workers (low-wage workers in particular) may suffer adverse wage impacts from noncompetes, other workers actually benefit from noncompetes. For example, noncompetes appear to systematically increase earnings for CEOs and physicians. See Final Transcript of the FTC January 9, 2020 Workshop (“Workshop Tr.”), p. 179. (If you’re interested in watching the FTC workshop, here are the links: Non-Competes in the Workplace Workshop Part 1 and Part 2).)
Fifth, the announcement ignores additional benefits of noncompetes. As the Treasury Report observed, some of the identified additional benefits include:
- Increased training: “Non-compete enforcement is associated with more worker training.” (P. 12.) “By reducing the probability of worker exit, non-competes may increase employers’ incentives to provide costly training.” (P. 3.) And, “[w]hile non-compete enforcement is associated with increased training for some workers, the details of this enforcement are important: strong ‘consideration’ requirements can support training and wage growth while diminishing the likelihood that non-compete contracts result purely from inadequate worker knowledge.” See also “Training the Enemy? Firm-Sponsored Training and the Enforcement of Covenants Not to Compete,” by Evan Starr (Jan. 25, 2015); “Noncompete Agreements in the U.S. Labor Force,” by Evan Starr, J.J. Prescott, and Norman Bishara (Oct. 12, 2020) (finding that employees who receive their noncompete before accepting their job receive 11% more training than those not bound by noncompetes).
- Increased innovation: “Non-competes are sometimes used to protect trade secrets, which can promote innovation.” (P. 3.)
- Better matching of employees with employers: “Employers with especially high turnover costs could use non-competes to match with workers who have a low desire to switch jobs in the future.” (P. 3.)
Indeed, although the Treasury Report observed that “[w]orker bargaining power is reduced after a non-compete is signed, possibly leading to lower wages,” (p. 3 (emphasis added)), the opposite (according to an October 12, 2020 paper by Professors Evan Starr, J.J. Prescott, and Norman Bishara) is also true: Employees “presented with a noncompete before accepting the associated job offer earn 9.7% higher wages . . . than those not bound by noncompetes.” (Emphasis added.) As Professor Matt Marx has observed, “If it were the case that workers made fully informed decisions about signing a non-compete and could negotiate higher compensation in exchange for doing so, these agreements could be valuable for both workers and firms.” (Emphasis added.)
It also bears noting that while increased wages supposedly derive from job churn (i.e., changing jobs more frequently), which is allegedly inhibited by noncompetes, not only is that purported wage effect offset when the noncompete is presented before the employee accepts the job, but providing such advance notice also leads to 11% more training and 6.6% more job satisfaction. Instructively, these benefits may in turn explain why some employees with noncompetes stay in their jobs longer. Accordingly, a simple solution to the (presumed) wage issue (which would concomitantly increase training and job satisfaction) may be to require advance notice of the noncompete, rather than wrest the decision to forgo increased job satisfaction and training out of the hands of the employee.
Of course, there may be additional benefits that will become apparent as additional research is done. For example, as Professors Evan Starr, Natarajan Balasubramanian, and Mariko Sakakibara explained in “Enforcing Covenants Not to Compete: The Life-Cycle Impact of New Firms,” “States that permit stronger enforcement of noncompete agreements tend to have fewer – but better (higher-quality ideas and more likely to survive) – startups.” And, financial services industry advisors subject to noncompetes “tend to be more productive, take fewer risks and align their behaviors with the goals of their employer” (at least in the mutual fund industry).
Sixth, the Treasury Report (upon which the President-elect’s announcement relies) itself relies in part on employee-reported data, the trustworthiness of which is at best dubious. For example, the Treasury Report notes that “only 24 percent of workers report that they possess trade secrets. Moreover, less than half of workers who have non-competes also report possessing trade secrets, suggesting that trade secrets cannot explain the majority of non-compete activity.” (P.4 (emphasis added).) The Treasury Report reinforces that point, noting, “Many workers who do not report possessing trade secrets are nonetheless covered by non-competes.” (P.24 (emphasis added).) While my experience confirms that employees report these things, my experience also reflects that employees frequently do not fully appreciate what a trade secret is. For example, employees oftentimes believe that something they worked on is theirs, and therefore not a trade secret. They are wrong. And the drafters of the Treasury Report were aware of this: “It is worth noting, however, that this is based on worker self-reports; employers may disagree.” (P. 24, n.41.)
There are additional statements in the Treasury Report that should be fully explored before they are relied upon to justify a political position. (One such example is the Report’s assertion that “it is not clear that relationships with clients constitute a socially valuable investment analogous to trade secrets” (p.7).) These other statements are not critical to the present discussion, and therefore are not explored here.
Seventh, the statement that “Biden will work with Congress to eliminate all non-compete agreements, except the very few that are absolutely necessary to protect a narrowly defined category of trade secrets, and outright ban all no-poaching agreements” is a bit unclear.
As a threshold matter, in 2016, Congress (virtually unanimously) passed the Defend Trade Secrets Act (“DTSA”), stating that it was the “sense of Congress that . . . trade secret theft, wherever it occurs, harms the companies that own the trade secrets and the employees of the companies” and that the Economic Espionage Act (which was amended by the DTSA to provide a federal private right of action to trade secret owners) “applies broadly to protect trade secrets from theft . . . .” Accordingly, trade secrets are anything but a “narrowly defined category.” And, as Congress and President Obama agreed, they are so vital to the economy that it was necessary to add a federal private right of action – on top of noncompetes – to help protect them.
Further, despite the announcement’s reliance on the Treasury Report, this statement may suggest that the Biden Administration plans to ignore the Treasury Report’s ultimate conclusions and recommendations – as well as those of the Obama Administration’s follow-up report and resulting Call to Action – all of which called for a rational response, as opposed to a throw-the-baby-out-with-the-bathwater response. Specifically, the Call to Action’s suggestions (which flesh out the Treasury Department’s “Directions for Reform” (pp.24-25)) were stated as follows:
- Ban non-compete clauses for categories of workers, such as workers under a certain wage threshold; workers in certain occupations that promote public health and safety; workers who are unlikely to possess trade secrets; or those who may suffer undue adverse impacts from non-competes, such as workers laid off or terminated without cause.
- Improve transparency and fairness of non-compete agreements by, for example, disallowing non-competes unless they are proposed before a job offer or significant promotion has been accepted (because an applicant who has accepted an offer and declined other positions may have less bargaining power); providing consideration over and above continued employment for workers who sign non-compete agreements; or encouraging employers to better inform workers about the law in their state and the existence of non-competes in contracts and how they work.
- Incentivize employers to write enforceable contracts, and encourage the elimination of unenforceable provisions by, for example, promoting the use of the “red pencil doctrine,” which renders contracts with unenforceable provisions void in their entirety.
Whether one agrees with all of the White House’s suggestions or not, the Call to Action represents a careful balance of myriad competing interests. It is no easy task to satisfy everyone, and states have varied on which of these recommendations to choose. In that vein, there are many additional methods to rein in noncompete abuse (I remain a fan of springing (or “time out”) noncompetes (see below) and the so-called “purple pencil” approach to overly broad noncompetes), but recommendations for appropriate regulations (assuming they even belong at the federal level) are beyond the scope of this post.
Eighth, the announcement ignores the potential unintended, unstudied consequences of a ban, including the impacts on small companies that have limited resources to protect themselves from larger competitors who hire their employees to unfairly compete with them. Indeed, another part of the equation is omitted entirely: the impact on the remaining employees, whose interests are aligned with the company, its survival, and its success.
Is President-elect Biden’s (apparent) approach the right approach?
No. But, as I said, I’m not going to delve into that question. The point of this post is not to debate the wisdom of a national (near-complete) ban of a tool that 47 states and D.C. have all made the affirmative decision to allow (including the over 30 states that have recently reviewed their noncompete laws). Want to read about that issue? See my testimony to the United States Senate Committee on Small Business & Entrepreneurship (November 28, 2019); the joint submission of me and 21 other lawyers and my paralegal (Erika Hahn) to the FTC (March 11, 2020).
To be clear, though it may occasionally appear otherwise, I actually don’t think that noncompetes are warranted nearly as often as I see them used. But, I recognize that they are sometimes necessary. That’s because every day (quite literally), I see employees and new employers engage in conduct that underscores the need for noncompetes. Conversely, I also regularly see conduct by employees and employers that is beyond reproach. But, it’s often not clear which of these paradigms will apply when an employee leaves, and noncompetes should be available when circumstances require them.
What to do about it.
Protecting trade secrets, confidential business information, goodwill, and other recognized legitimate business interests does not happen by accident. Companies need to plan. More on that here: “A primer and checklist for protecting trade secrets and other legitimate business interests before, during, and after lockdown and stay-at-home orders.” (It’s long, but worth a read, and includes a step-by-step guide to evaluating and establishing a proper plan.)
But, when one of the key tools to protect a company’s legitimate business interests (i.e., noncompetes) is in the crosshairs, companies need to focus more closely on the remaining options to ensure they have the protections they need and that fit their circumstances. These options each have their pros and cons, but can work together to form a workable back-up plan when noncompetes are not available – or even when they are simply not necessary or enforceable in the particular circumstances. Accordingly, they should be part of an overall legitimate business interest protection strategy.
The key tools beyond noncompetes.
Key among the tools to protect a company’s legitimate business interests are the following:
- Nondisclosure agreements. Also known as an “NDA” or “confidentiality agreement,” these agreements restrict a person’s use of confidential information and trade secrets. They serve multiple important purposes and are a fundamental building block to protecting trade secrets. For example, they put employees on notice that the company has information that may be confidential in general and identifying for the employee particular types of information that the company, in fact, considers confidential. Most courts view them as the price of admission before they will step in to help protect the company’s trade secrets. But, they are also an important tool in the protection of other confidential information and even customer goodwill, insofar as the goodwill and information about the client typically go hand in glove.
While in the past nondisclosure agreements were generally considered enforceable without much scrutiny, a recent First Circuit decision (TLS Mgmt. & Mktg. Servs., LLC v. Rodriguez-Toledo, 966 F.3d 46, 57 (1st Cir. 2020)) potentially upends that paradigm – at least in the First Circuit. After this case, if an NDA is significantly over-inclusive, it could be at risk of being unenforceable in its entirety. See TLS Mgmt. & Mktg. Servs., LLC, 966 F.3d at 57 (refusing to narrow the “astounding [over]breadth”).
Companies should now be reviewing the language they use for their nondisclosure agreements and make sure it’s tied to the information that the company in fact treats as confidential. See id. at 57, 59 (invalidating an agreement that prohibited the use and disclosure of information that (1) included general knowledge, (2) was otherwise publicly available, and (3) was received from third parties (such as the plaintiff’s former clients)). And, while you’re at it, don’t forgot to comply (or make an informed decision to not comply) with the Defend Trade Secret Act’s whistleblower notification requirements. 18 U.S.C. § 1833(b).
Of course, even if the NDA is enforceable, depending on the jurisdiction, it likely will not prevent an employee from taking a job where they will be in a position to use or disclose the company’s confidential information. As a consequence, oftentimes violations occur without the former employer’s knowledge – and by the time it’s discovered, it may be too late.
- Nonsolicitation agreements. These agreements typically prohibit the solicitation of a company’s customers (and sometimes referral sources, suppliers, and others). And they do so without restricting the nature of the former employee’s employment with a competitor. That said, sometimes nonsolicitation agreements are drafted so broadly that courts will consider them to effectively constitute a noncompete, and refuse to enforce them on that basis.
When customer goodwill (and potentially customer confidential information) is at issue, a properly drafted nonsolicitation agreement (if complied with) offers meaningful protection. But like an NDA, nonsolicitation agreements are not a prophylactic, and oftentimes violations occur without the former employer’s knowledge.
- A variation on the theme of nonsolicitation agreements are agreements that prohibit the former employee from causing a customer or other party to terminate or reduce its relationship with the former employer, or that expressly prohibit the former employee from even servicing the customer. They are known as noninterference and no-service agreements, respectively. While they provide more protection than a nonsolicitation agreements, they are harder to enforce.
- No-recruit / no-raid agreements. These agreements, which prohibit the solicitation of a company’s employees, are sometimes referred to as nonsolicitation agreements, but more properly are known as no-raid, anti-raiding, antipiracy, nonrecruit, or similarly named agreements. In the past, companies often did not enforce these restrictions absent a contemporaneous breach of a noncompete or other agreement. However, enforcement actions involving these agreements seem to have become more commonplace, and the law is developing to hold these agreements to a standard similar to those of nonsolicitation agreements.
- A variant of no-raid agreements that are an absolute ban on the hiring – as opposed to just a bar to the solicitation – of a company’s employees are called no-hire agreements. While they provide more protection than no-recruit agreements, they are harder to enforce.
- Alternative forms of noncompetes. If noncompetes are banned, similar (but different) types of agreements may remain permitted. At the moment, this issue varies by state (with some states viewing them as noncompetes and others not). How a federal overlay would apply is, at this point, unknown. However, if permitted, the impact of a noncompete is ameliorated to some extent because the employee is paid during the restricted period or permitted to compete.
- Garden leave/notice covenants. Generally, these clauses call for compensating an employee (or former employee) while they are restricted from working for a competitor. There are essentially two types, each with its own variations: (1) the traditional version, in which the employee is required to provide notice of resignation for a specified (typically, lengthy) period, during which they remain employed by their soon-to-be-former employer (but do no work) and, as a consequence, continue to be bound by their fiduciary duties (including to not compete), and (2) a model in which the employee is paid some amount of compensation during the restricted period of a traditional noncompete.
- Forfeiture-for-competition agreements and compensation-for-competition agreements. These are agreements by which an employee either forfeits certain benefits or pays some amount of money (often a percentage of revenues) if they engage in activities that are competitive with their former employer.
- Invention assignment agreements. These agreements require employees to assign to the company any improvements to the company’s intellectual property. Such agreements both protect the company’s investment in its intellectual property and preserve to the company the benefits of new developments that arise as a consequence of work being done for the company. They tend to be enforceable, assuming they follow applicable state limitations.
- Springing/“time out” noncompetes. In Massachusetts, a court is expressly permitted to, in effect, create a noncompete for someone who has engaged in unlawful conduct, such as breaching their nondisclosure agreement or nonsolicitation agreement, misappropriating trade secrets, or breaching their fiduciary duties to their employer. As such, while the employee is permitted to compete in the first instance, they essentially forfeit that right if they engage in otherwise unlawful behavior. Much like the photograph above, when people don’t follow the rules (whether it’s the instruction to keep off the grass or to not use the company’s trade secrets or solicit the company’s customers), a fence (real or metaphorical) may be necessary. Accordingly, this is not a separate agreement; it’s a remedy. Nevertheless, it can be identified as a remedy in an agreement. There is no case law about its enforceability at this point, but I have had clients use it for years before it was incorporated in the Massachusetts Noncompetition Agreement Act, and, on a positive (albeit unhelpful) note, it’s never needed to be tested.
To understand the relative enforceability (i.e., how likely a court is to enforce the particular agreement) and strength of protection offered by these agreements, I have plotted them on the graph below. The farther up the X axis of the graph, the greater the protection offered by the type of agreement. The farther right on the Y axis, the more likely a court is to enforce the agreement. (The graph is not drawn to scale; it’s intended to give just a general sense of how the different types of agreements compare.)
These agreements, for the most part, are not mutually exclusive. Which agreements make sense will depend on the particular circumstances of the company, the industry, the business interests, and the employee at issue. Careful consideration should be given – now – to make sure that the appropriate protections are in place, regardless of what happens to noncompetes moving forward.
Any agreements you use should be supplemented with all appropriate policies, including in particular, policies governing the proper use of company owned equipment and technology and personal devices (BYOD policies), trade secrets and confidential business information policies, and codes of conduct.
Training, training, training.
What are the three most important steps to protecting a company’s trade secrets, other confidential information, and goodwill? Okay, it’s not really training, training, training – but training is crucial to making the protections work.
Even with the best of agreements in place, protecting legitimate business interests comes down to training – starting before a new employee walks in the door, continuing during the employment cycle, and repeating at the end (where it starts for the new employer). Preventing information from entering the company and contaminating the company’s existing information and research is critical. (This happens often not just at the start of employment, but later as well, when issues arise that an employee worked on at their former employer.) Relatedly, making sure that company information doesn’t leave the company (during or at the end of the employment relationship) is also critically important.
To help train employees about how to avoid these missteps, we have created some short “Ten Minute Training” videos for each of those scenarios. (Our trade secret training video for remote workers — Protecting Trade Secrets – Working at Home — is free and just requires registration; our other training videos, including for employees transitioning from one company to another, are available for free for our corporate and individual clients and can be licensed by other companies.) Please feel free to contact us if you would like access.
Now is the time to review and update your overall legitimate business interest protection strategy, including (1) your agreements, to ensure that they are sufficiently protective, without going too far; (2) all applicable policies and codes of conduct; and (3) your training program for preventing the infiltration and exfiltration of your information and goodwill (contact us if you want any of the training videos).
Taking these steps should help to both reduce the need for enforcement actions and increase the likelihood of success if and when you do need to enforce your rights or defend your conduct.
 The “purple pencil” is an equitable, middle-ground to handling overly broad restrictions (which a Massachusetts state senator coined the “purple pencil” approach). It’s a hybrid of the reformation and red pencil approaches, requiring courts to strike the noncompete in its entirety unless the language reflects a good faith effort to draft a narrow restriction, in which case the court may reform it.
 The concept of garden leave, sometimes called “gardening leave,” originated in the United Kingdom, see Norman D. Bishara & Michelle Westermann-Behaylo, The Law and Ethics of Restrictions on an Employee’s Post-Employment Mobility, 49 American Business Law Journal, Issue 1, 1, at 25 (Spring 2012), and takes its name from the idea that the employee is placed on leave, and therefore having no work to do, can stay home and tend their English garden.