By: John W. Lettieri, President and Chief Executive Officer of the Economic Innovation Group (EIG)
Chairman Rubio, Ranking Member Cardin, and members of the committee, thank you for the opportunity to testify on how noncompete agreements impact American workers and the broader
Today, roughly 20 percent of the American workforce is not allowed to take a better job in the field of their choice – regardless of higher pay, better benefits, improved job satisfaction, or other factors. The reason: they are bound by a covenant not to compete, or “noncompete” agreement.
Noncompete agreements, much like occupational licensing laws and “no-poach” agreements, erect barriers to worker mobility and dampen the vitality of the U.S. economy. Once reserved for senior executives and those possessing valuable trade secrets, these provisions are now a common weapon in the arms race among employers looking to retain talent, hedge against competition, and keep their labor costs down.
Simply put, noncompete agreements are a tool of corporate protectionism – one that weakens the competitive forces essential to broad-based prosperity. At a time when policymakers are struggling to find ways to improve worker mobility, increase business formation, strengthen innovation, and boost wages, noncompete reform is an obvious place to start. Indeed, federal policy to limit the use of noncompete agreements would be among the most impactful and least expensive ways to jumpstart economic dynamism and improve the fortunes of American workers.
How Noncompetes Hurt Workers and Contribute to a Less Dynamic Economy
Healthy labor markets depend both upon vigorous competition for talent between firms and the ability of workers to freely market their skills to interested employers. Likewise, a dynamic economy depends upon the productivity-boosting exchanges that happen when individuals collaborate or apply their collection of ideas and experiences in new contexts. It’s the simple microeconomic transaction of a worker switching firms that facilitates innovation and helps know-how to proliferate throughout the economy. Recent research from economists at the University of Chicago finds that up to 70 percent of the decline in measures of economic dynamism can be attributed to a dramatic slowdown in the rate at which knowledge diffuses across the economy.2 Noncompete agreements directly undermine the economic processes integral to our continued prosperity.
An estimated one in five American workers are covered by a noncompete, and nearly twice as many have signed one at some point in the past. Their use among senior executives is commonplace, but a sizeable portion of lower-wage workers are covered by noncompetes as well. Much of the recent interest in noncompetes has been driven by a growing awareness that employers are requiring them of fast food workers, janitors, camp counselors, and other low-wage professions. Such examples are proof that noncompetes are often nothing more than a tool to suppress wages and limit the mobility of already vulnerable workers.
Source: Starr, Prescott, and Bishara, “Noncompetes in the US labor force.”
Why is limiting worker mobility potentially problematic? One key reason is that “job-hopping” – especially early in a worker’s career – is associated with stronger lifetime earnings. Research indeed finds that strict enforcement of noncompetes is associated with lower wages, as well as reduced job-to-job mobility and weaker rates of firm formation. Noncompetes also appear to exacerbate racial and gender wage gaps by exerting much larger wage effects on women and black employees than on white men.3 Additional research has identified noncompetes as one explanatory factor in the gender gap in entrepreneurship.4 It is also important to note that noncompetes do not simply impact the workers who sign such agreements; they bring negative externalities that have a chilling effect on the entire labor market, as my fellow witness Evan Starr and his colleagues have found.5
Noncompetes aren’t just bad for workers; they have negative effects for employers as well. By tamping down the amount of healthy churn throughout a labor market, they reduce the supply of available workers and can make it more difficult for businesses to grow. Such limitation could be especially harmful to smaller enterprises that already face disadvantages against larger incumbent businesses. The harmful effect of noncompetes on new businesses – both by stifling would-be entrepreneurs and limiting the pool of much-needed talent for startups – should be of utmost concern to this committee.
The vast majority of noncompete agreements are not subject to any negotiation between the employer and employee. In fact, surveys show a large share of these agreements are presented for signature only after the employee has already accepted the job offer – often on the first day of work.6 Employers frequently exploit workers’ lack of knowledge and resources when crafting noncompetes. For example, employers commonly request signature of noncompetes even in states where they are completely unenforceable – and workers sign them. Likewise, employers often craft extremely broad provisions knowing that employees generally lack an understanding of what is enforceable and therefore rarely challenge the terms in court. 7
Recent State Reforms
The growing evidence regarding how noncompetes impact workers and the economy has caught the attention of policymakers across the country from both sides of the aisle. While most states
have few (if any) rules limiting the use of noncompetes, several have enacted meaningful reforms in recent years that should serve as guideposts for federal policy efforts.
While the new laws include a wide variety of features, the two most common are 1) exemptions for lower-wage workers, and 2) transparency requirements to ensure workers are given prior notice before being asked to sign an agreement. Particularly worth noting are Maine and Washington, which enacted the most comprehensive policies in 2019.
- Maine, a state grappling with population loss and a scarcity of available workers, enacted a broad set of provisions primarily aimed at protecting lower-wage employees. Under the new law, employees making at or below 300 percent of the federal poverty level would be prohibited from signing a noncompete agreement. The law imposes several prior notice requirements, including requiring employers to disclose in job postings when a noncompete will be required. Maine’s law also enacts a ban on so-called “no poach” agreements between employers under which the parties agree not to recruit or hire each other’s employees or former employees. These agreements disproportionately impact low-wage workers and are particularly common within the franchise sector.
- Washington state’s new law sets a number of preconditions for the enforceability of noncompetes in the state. First, it exempts a much larger pool of employees than most other states, making noncompetes enforceable only for employees earning over $100,000 per year. (For independent contractors, the floor is set at $250,000 per year.) Employers are required to provide prior notice of the terms of the agreement by the time a job offer is made, and must compensate any terminated worker still subject to the terms of the agreement. Noncompetes in the state are now limited to 18 months in duration.
I want to briefly note three other state reforms that could inform potential federal policy efforts.
- Oregon was well ahead of its time in passing far-reaching noncompete reform in 2007. The state provided specific protections for low-wage workers alongside a range of broader measures to improve transparency and narrow the scope of what is enforceable under state law. The centerpiece of the law was a provision to void all new noncompetes for workers earning less than the median income for a household of four (it also covered hourly workers and employees in certain occupations). Other important elements include a “garden leave” provision requiring compensation of former employees covered by a noncompete, prior notice standards, and a rule limiting enforceable agreements to 18 months or less. Recent research finds the Oregon law had a significant positive effect on hourly wages and job-to-job mobility – exactly as intended.8
- Another noteworthy reform occurred in Massachusetts in 2018. Massachusetts is particularly interesting because many observers point to noncompetes as a key reason that Silicon Valley outpaced Boston as the country’s premier hub of innovation. (California is one of only three states that does not enforce noncompetes.) The Massachusetts statute, which applied only to agreements signed after October 1, 2018, included a one-year limitation on noncompetes, prior notice standards, and garden leave provisions requiring employers to pay 50 percent of the annualized base salary of a former employee on a pro rata basis while they are covered by the agreement. It banned noncompetes entirely for low-wage workers who are classified as nonexempt under the Fair Labor Standards Act, as well as employees terminated without cause and undergraduate or graduate students in an internship. The law notably maintains the employer-favorable standard of allowing courts to rewrite an overly broad agreement in order to make it enforceable. It also provides exceptions for the sale of a business or the dissolution of a partnership.
- While most states have targeted reforms to low-wage and moderate-wage workers, only one state has explicitly focused on entrepreneurship and innovation as the driving factors behind its policy efforts. In 2015, Hawaii banned noncompetes for workers in its technology industry in an effort to attract and retain highly skilled workers and compensate for its unique geographical limitations. Research finds meaningful increases in mobility and wages for new hires for Hawaii’s tech workers following the industry ban.9 Hawaii’s experience should inform federal efforts to create a more favorable policy environment for innovation and entrepreneurship nationwide.
How Should Federal Policymakers Respond?
The need for reform is urgent, but why should the federal government intervene in an issue that has until now been left to states? The answer is that noncompetes are clearly an issue of employment law, where the federal government commonly sets basic standards governing the relationship between employers and employees. In addition, even though employers exercise little self-restraint in how they deploy noncompete agreements, states have generally set few limitations of their own. But none of this would be of much consequence without the growing pile of research finding significant negative effects from noncompetes on wages, job-to-job mobility, entrepreneurship, the gender and racial income gaps, and more. In the absence of federal policy, we are left with a confusing patchwork of policy that fails to serve our national economic interests.
What form should federal policy take? While I believe the best answer would be a nearly universal restriction on noncompetes across all occupations and income levels, there are a wide variety of options from which Congress can choose.
Federal policy should be guided by the following core objectives.
- Require Transparency: Many of the negative effects of noncompetes can be reduced simply by ensuring greater transparency and improving workers’ awareness of their bargaining position. As noted above, employers exploit their informational advantage by requiring noncompetes in states where they are unenforceable, or by offering a noncompete on an employee’s first day of work when other options have been foreclosed. Rules governing noncompetes should be clear and easy to administer, and employees should be given adequate notice before being asked to sign away future job opportunities.
Examples: Employers should be required to notify job candidates of their intent to request signature as well as present noncompetes when the formal job offer is made – not after the employee has accepted the job. Employers should disclose their intent to require a noncompete in any job posting or advertisement for the position. Additionally, employers requesting a noncompete agreement should be required to fully inform their prospective employee of applicable state and federal laws and allow adequate time for the candidate to discuss the terms of the agreement before making a decision.
- Create Disincentives for Overuse: There are currently few disincentives for an employer to require noncompetes of its employees – even when agreements are written so broadly as to be unenforceable, and even when they cover employees who have no specialized skills or trade secrets. Federal law should seek to discourage overuse of noncompetes wherever possible by ensuring that employers carefully consider whether the benefits are worth the costs.
Examples: Employers should be required to compensate former employees while their noncompete is in effect. Noncompetes drafted in an overly broad manner should be rendered completely void – not rewritten by courts to make them enforceable. Federal law should also penalize employers who request signature in states where noncompetes are unenforceable.
- Limit the Pool of Eligible Workers: Most states currently have no restrictions on the kinds of workers that can be bound by a noncompete. Many options exist to narrow the eligible pool of workers by industry, by wage level, or by education attainment so that noncompetes are reserved only for senior executives and other top talent.
Example: Signature of a noncompete should be disallowed for any worker outside of the top five percent of the national income distribution.
- Limit the Scope of Agreements: Even when policymakers see a valid use for noncompetes under certain circumstances, most agree the scope of such agreements should be limited in various ways.
Examples: The duration of noncompetes should be limited to no more than one year, and any noncompete should be voided if an employee is terminated without cause or laid off.
I want to note my appreciation for two pieces of legislation introduced by members of this committee, which represent different ends of the broad spectrum of potential reforms. The first is the Freedom to Compete Act introduced by Chairman Marco Rubio (R-FL), which would prohibit noncompete agreements for any nonexempt worker under the Fair Labor Standards Act. Like many of the state laws noted above, this legislation would provide welcome relief to the lowest-wage American workers. At the other end of the spectrum is the Workforce Mobility Act, introduced by Senators Todd Young (R-IN) and Chris Murphy (D-CT). This legislation would limit the use of noncompetes in all but two cases: the sale of a business and the dissolution of a partnership – instances where the parties involved are on more or less equal footing.
With so many potential legislative options, policymakers must be mindful of why reform is necessary and what it can accomplish if properly crafted. There is perhaps a natural temptation to focus solely on the lowest-wage workers for whom the use of a noncompete seems most abusive and ridiculous. But Congress shouldn’t stop there. Achieving the full promise of noncompete reform requires enabling skilled workers to better deploy their talents and ideas throughout the economy, including by starting new firms and bringing innovations to market. Exemptions for low-wage workers alone will fall short of reaching this critical goal.
Workers should be free to seek better jobs and compete in the labor market without permission from their former employers. Employers should be rewarded for winning the competition for talent – not for holding workers hostage. And policymakers should be relentlessly focused on ensuring a policy environment that encourages competition, healthy risk-taking, and worker mobility. The pervasive use of noncompetes is proof that the United States has strayed from each of these fundamental principles. Fortunately, this is a solvable problem.
Noncompete reform is an issue whose time has clearly come. I urge Congress to act upon this rare opportunity to revive economic dynamism, improve worker mobility, and provide a long-overdue boost to the wages of American workers – all without spending a dollar of taxpayer money.
Thank you for holding this important hearing.
1 This testimony draws from a forthcoming paper on noncompete reform to be published by the American Enterprise Institute (2AEI). AEI is a nonpartisan, nonprofit, 501(c)(3) educational organization and does not take institutional positions on any issues. The views expressed in this testimony are those of the author.
2 Ufuk Akcigit, “What Happened to U.S. Business Dynamism.” Becker Friedman Institute Research Brief (2019).
3 Johnson, Matthew, Lavetti, Kurt and Lipsitz, Michael, “The Labor Market Effects of Legal Restrictions on Worker Mobility”, SSRN (2019)
4 Matt Marx, “Punctuated Entrepreneurship (Among Women),” Working Papers 18-26, Center for Economic Studies, U.S. Census Bureau (2019)
5 See Starr, Evan, James J. Prescott, and Norman Bishara, “Noncompetes in the US labor force.” U of Michigan Law & Econ Research Paper 18-013 (2019) and Starr, Evan, Frake Justin, and Agarwal, Rajshree “Mobility Constraint Externalities”, Organization Science (2019)
6 See Starr et al. (2019) at 5
7 Prescott, J. J., and Evan Starr, “Subjective Beliefs about Contract Enforceability” Working Paper (2019)
8 Lipsitz, Michael and Starr, Evan, “Low-Wage Workers and the Enforceability of Non-Compete Agreements” (September 10, 2019). Available at SSRN: https://ssrn.com/abstract=3452240 or http://dx.doi.org/10.2139/ssrn.3452240
9 Balasubramanian et al. 2018, “Locked In? The Enforceability of Covenants Not to Compete and the Careers of High-Tech Workers.” US Census Bureau Center for Economic Studies Paper No. CES-WP-17-09; Ross School of Business Paper No. 1339.