Noncompetes, Human Capital Policy & Regional Competition Orly

Noncompetes, Human Capital Policy & Regional

Competition

Orly Lobel

Abstract

In the past few years, human capital law has become one of the most dynamic policy

fields in the United States. Multiple states have reformed their noncompete policies,

passing new legislation that limits their use. New bills that would similarly limit the

enforcement of noncompetes are currently before Congress. Nationwide, both the use of

noncompetes and litigation over their enforcement, are on the rise. As such, several state

attorneys general have taken up the issue by launching investigations into employers who

require their workforce to sign unenforceable noncompetes. An equally dazzling wealth of

studies, analysis, intellectual debates, and exchanges have emerged on the research side.

In particular, the past few years brought a significant number of empirical, experimental,

and theoretical studies offering more evidence and explanations about the key role that

human capital policy, including noncompete contracts, plays in industries and regions. In

this Article, written for a symposium honoring the scholarship of Professor Ronald Gilson,

I present the state of the scholarly field on human capital and economic competition and

develop three arguments about the future of noncompete research. First, in Part II, I

unpack the multiple dynamic effects that job mobility and noncompetes have on regions.

Beyond knowledge spillovers, it is important to recognize a range of distinct, though

interrelated effects. These include at least ten important aspects that are supported by job

mobility: behavioral, dynamic, firm-level, and regional-level effects. In particular, a

neglected aspect in the literature of noncompetes is the disproportionate harmful effect

noncompete clauses may have on women. Recent economic research on labor market

monopsonies and the relationship between mobility and wage growth allows us to see

connections between innovation policy and distributive justice. Second, I argue that while

the study of noncompetes has been invaluable to understanding talent flows, mobility

restrictions are far broader than merely formal covenants not to compete. Covenants that

restrict employee mobility appear in many shapes and forms. I introduce the range of

contractual restrictions that employers require in standard agreements and I argue that

these restrictions, too, should be understood and researched through the lens of labor

market competition and mobility. Third, I argue that the prevalence of practices that

subvert policy requirements, such as including unenforceable restrictions in employment

contracts underscores how we as scholars need to encompass market practices in the

empirical research, as well as recognize comparative advantages of proactive solutions

including antitrust and regulatory tools over contract doctrine.

Warren, Distinguished Professor, University of San Diego. For thoughtful comments and conversations,

I thank On Amir, Rachel Arnow-Richman, Peter Conti-Brown, Ronald Gilson, Michael Madison, and David

Zaring. For excellent research assistance, I thank Meghan Brown, Hannah Karraker, Austin Trickey. Sasha Nuñez

and Elizabeth Parker, who provided superb library support and smart suggestions.

932 The Journal of Corporation Law [Vol. 45:4

I. INTRODUCTION ............................................................................................................ 932

II. THE MULTIPLE DYNAMIC EFFECTS OF MOBILITY ...................................................... 935

III. MOBILITY PENALTIES: A NONCOMPETE BY ANY OTHER NAME ............................... 943

IV. BEYOND ENFORCEABILITY: PERVASIVE NONCOMPLIANCE WARRANTS A

REGULATORY APPROACH TO MOBILITY ................................................................ 947

V. CONCLUSION .............................................................................................................. 950

I. INTRODUCTION

Exactly two decades ago, in 1999, Ronald Gilson wrote a seminal article on human

capital law⎯before the field existed. No one was better equipped⎯neither then nor

now⎯to carve out this new policy field, which rests at the intersection of contract,

corporate, employment, intellectual property, and antitrust laws. In The Legal

Infrastructure of High Technology Industrial Districts: Silicon Valley, Route 128, and

Covenants Not to Compete, Gilson called attention to the legal infrastructure that either

allows or prohibits the use of noncompete clauses in employment relationships.

He argued

that the success of Silicon Valley’s high-tech industries should, in part, be linked to

California’s longstanding policy rendering noncompetes unenforceable.

Since then,

human capital law has become one of the most dynamic policy fields in the country.

the past five years, multiple states have reformed their noncompete policies, passing new

legislation that limits their use. New bills that would similarly limit the enforcement of

noncompetes are currently before Congress. Both the use of noncompetes, as well as

litigation over their enforcement, are on the rise nationwide, and several state attorneys

general have taken up the issue by launching investigations into employers who require

their workforce to sign unenforceable noncompetes.

In October 2016, the White House issued a Call for Action urging states to limit the

use of post-employment restrictions.

Also in 2016, the U.S. Treasury Department issued

a report on noncompetes warning that when noncompetes are enforced, “innovations

spread more slowly, possibly inhibiting the development of industrial clusters like Silicon

Valley.”

Currently, the Federal Trade Commission is contemplating a petition submitted

. Ronald J. Gilson, The Legal Infrastructure of High Technology Industrial Districts: Silicon Valley,

Route 128, and Covenants Not to Compete, 74 N.Y.U. L. REV. 575, 578 (1999).

. Id.

. See, e.g., COUNCIL OF ECON. ADVISORS, LABOR MARKET MONOPSONY: TRENDS,

CONSEQUENCES, AND POLICY RESPONSES 8 (Oct. 2016), https://obamawhitehouse.archives.gov/sites/de

fault/files/page/files/20161025_monopsony_labor_mrkt_cea.pdf [https://perma.cc/DW4W-WWFB].

. See id. (“[S]tate attorneys general have begun to be active in the field of employment restrictive

covenants . . . conducting proactive investigations into employers for requiring their workforce to sign

unenforceable contracts.”); see generally Illinois v. Check Into Cash, LLC, No. 2017-CH-14224 (Ill. Cir. Ct. Oct.

25, 2017); Illinois v. Jimmy John’s Enters., No. 2016-CH-07746 (Ill. Cir. Ct. June 8, 2016).

. Press Release, The White House, The Obama Administration Announces New Steps to Spur

Competition in the Labor Market and Accelerate Wage Growth (Oct. 25, 2016)

https://obamawhitehouse.archives.gov/the-press-office/2016/10/25/fact-sheet-obama-administration-announces-

new-steps-spur-competition [https://perma.cc/NH7G-LZDF]. I was part of the White House Working Group that

resulted in this Call for Action.

. U.S. DEP’T OF TREASURY, NON-COMPETE CONTRACTS: ECONOMIC EFFECTS AND POLICY

IMPLICATIONS 6 (2016), https://www.treasury.gov/resource-center/economic-policy/Documents/UST%20Non-

2020] Noncompetes, Human Capital Policy & Regional Competition 933

by over 40 law professors (including myself), senators, and non-profit organizations to

consider regulating noncompetes.

In the two decades since Gilson published the article,

an equally dazzling wealth of studies, analysis, intellectual debates, and exchanges have

emerged on the research side. In particular, the past few years brought a significant number

of empirical, experimental, and theoretical studies offering more evidence and explanations

about the key role that human capital policy, including noncompete contracts, plays in

industries and regions.

These articles attack the question of noncompetes from multiple

directions: longitudinal studies, comparative regional studies, patent network mapping,

surveys, behavioral lab experiments, ethnographies, simulations, and modeling.

On a personal note, Gilson’s 1999 article sparked my interest in the field, and in 2013,

I published Talent Wants to be Free: Why We Should Learn to Love Leaks, Raids, and

Free-Riding (Yale University Press), which developed a thick analysis of the multiple

effects of human capital law on competition, performance, innovation, and economic

growth.

Alongside over a dozen articles on human capital policy, I published a second

book in 2017, You Don’t Own Me: How Mattel v. MGA Entertainment Exposed Barbie’s

Dark Side (Norton), in which I demonstrated how other contractual restrictions such as

broad non-disclosure agreements and innovation assignment clauses effectively operate

like noncompetes to prevent employee mobility, slow down innovation, prevent new entry,

and concentrate industries.

Gilson’s contribution extends beyond inspiring new research and bringing

noncompete policy to the forefront. His article made other contributions that have received

less attention. Legal Infrastructure underscores the key role tacit knowledge—distinct

from codified knowledge—plays in innovative settings.

Before Gilson’s research, and

still to a significant extent today, the field of intellectual property focused on codified

knowledge. By requiring reduction to practice and concrete expressions, patent and

conceptual, or still in an ephemeral phase. In several recent articles, I examine these gaps

between intellectual property and contractual norms, building upon Gilson’s insights on

tacit knowledge.

Further, Gilson’s article was an important contribution to the canon of

regional comparison and experimentalism in state policy. The article sparked interest

around the world as an approach to decoding essential ingredients for economic prosperity

competes%20Report.pdf [https://perma.cc/LZJ8-6DZY]; Karen Dynan, The Economic Effects of Non-Compete

Agreements, U.S. DEP’T OF TREASURY (Mar. 31, 2016), https://www.treasury.gov/connect/blog/Pages/The-

Economic-Effects-of-Non-compete-Agreements-.aspx [https://perma.cc/5Z9V-2TTA].

. Petition to the Federal Trade Commission for Rulemaking to Prohibit Worker Non-Compete Clauses,

OPEN MKT. INST. (Mar. 20, 2019), https://openmarketsinstitute.org/petitions/here-is-a-petition/

[https://perma.cc/9X4T-LXKG].

. Orly Lobel, Gentlemen Prefer Bonds: How Employers Fix the Talent Market, 24 SANTA CLARA L. REV.

(forthcoming), http://ssrn.com/abstract=33731848 [https://perma.cc/8RJA-VHDX] (“[S]tate attorneys general

have begun to be active in the field of employment restrictive covenants”).

. ORLY LOBEL, TALENT WANTS TO BE FREE: WHY WE SHOULD LEARN TO LOVE LEAKS, RAIDS, AND

FREE RIDING (2013).

. ORLY LOBEL, YOU DON’T OWN ME: HOW MATTEL V. MGA ENTERTAINMENT EXPOSED BARBIE’S DARK

SIDE (2017).

. Gilson, supra note 2, at 582–86.

. See generally Rochelle Cooper Dreyfuss & Orly Lobel, Economic Espionage as Reality or Rhetoric:

Equating Trade Secrecy with National Security, 20 LEWIS & CLARK L. REV. 419 (2016) (examining Gilson’s take

on tacit knowledge).

934 The Journal of Corporation Law [Vol. 45:4

and growth. As I argue below, on each of these frontiers, Gilson’s work continues to shape

key bodies of research. Needless to say, Ronald Gilson has been incredibly influential for

me, as a leader and a mentor, and for so many other scholars.

Here, I develop three arguments about the future of noncompete research. First, in

Part II, I unpack the multiple dynamic effects that job mobility and noncompetes have on

regions. I argue that while Gilson focused on knowledge spillovers,

it is important to

recognize a range of distinct, though interrelated, effects. These include at least ten

important aspects that are supported by job mobility: behavioral, dynamic, firm-level, and

regional-level effects. In particular, a neglected aspect in the literature of noncompetes is

the disproportionate harmful effect noncompete clauses may have on women. Recent

economic research on labor market monopsonies and the relationship between mobility

and wage growth allows us to see connections between innovation policy and distributive

justice.

Second, I argue that while the study of noncompetes has been invaluable to

understanding talent flows, mobility restrictions are far broader than merely formal

covenants not to compete. Covenants that restrict employee mobility appear in many

shapes and forms. By the end of his article, Gilson points to the direction of human capital

policy study that looks beyond formal noncompetes by raising a caveat about the doctrine

of inevitable disclosure in trade secret law.

In Part III, I introduce the range of contractual

restrictions that employers require in standard agreements and I argue that these restrictions

too, should be understood and researched through the lens of labor market competition and

mobility.

Third, while Gilson focused on California’s statutory rule of non-enforcement, recent

empirical findings point to the importance of understanding mobility-in-action.

California’s Silicon Valley is indeed an example of an industry arc that has benefited from

its noncompete policy but also has sought to subvert it. In 2010, the Antitrust Division of

the U.S. Department of Justice filed a complaint against major tech corporations, including

Apple, Google, Intel and eBay, who colluded to refrain from hiring each other’s

employees.

More pervasively, recent findings show that California employers insert

noncompete clauses in the employment contracts at about the same rates as enforcing

states. In Part IV, I argue that the prevalence of such practices confirms how we, as

scholars, need to encompass market practices in the empirical research, as well as recognize

comparative advantages of proactive solutions including antitrust and regulatory tools over

contract doctrine. I conclude by suggesting that Gilson’s Legal Infrastructure continues to

provide an agenda for research in several different fields, including innovation, regional

development, entrepreneurship, and competition.

II. THE MULTIPLE DYNAMIC EFFECTS OF MOBILITY

“Hoping that similar names presage similar outcomes, regions christen themselves

. Gilson, supra note 2, at 584–86.

. Id. at 620–26.

. See Bill Baer, Assistant Att’y Gen., Remarks at the Conference Call Regarding the Justice Department’s

Settlement with eBay Inc. to End Anticompetitive “No Poach” Hiring Agreements (May 1, 2014),

https://www.justice.gov/opa/speech/assistant-attorney-general-bill-baer-speaks-conference-call-regarding-

justice-department [https://perma.cc/Y63E-JYTJ] (describing the DOJ’s settlement with eBay).

2020] Noncompetes, Human Capital Policy & Regional Competition 935

Silicon Mountain, Silicon Alley, Silicon Forest, or Silicon Glen” – Ron Gilson

Before Legal Infrastructure, debates about noncompetes were distorted through a lens

of labor versus business. The benefit of employee mobility was framed as the employee’s

right to pursue her profession, while the benefit of restriction was framed as a corporation’s

right to protect its investment in intangible property and training. Gilson shifted the

conversation to what is beneficial for industries and regions, understanding the rule against

noncompetes as solving a collective action problem.

Mobility benefits firms, but without

a mobility policy, firms will attempt to prevent their employees from moving to

competitors. In an optimal equilibrium of a competitive market, every firm should eschew

this kind of anti-competitive impulse in advance in order for everyone to benefit from a

continuous high-quality labor pool over time. Gilson’s article thus invited us to research

noncompete policy as a coordination mechanism that solves the arms race that employers

might face: the legal rule ensures that in the repeat game of frequent ongoing recruitment,

each employer is prohibited from attempting to gain a harmful one-shot advantage.

The

result is a collectively superior strategy of increased employee mobility and knowledge

flow.

The past two decades since Gilson published Legal Infrastructure have been

tremendously exciting and rewarding with the rise of numerous studies from multiple

academic disciplines. Together these studies provide robust support to the historical

intuition that regions and markets thrive on job mobility. As economist Evan Starr wrote

in 2018 when reviewing the most up-to-date empirical studies on noncompetes, “Despite

their different approaches, samples, and time frames, the results across these studies are

consistent: The enforceability of noncompetes is associated with the reduced movement of

workers and lower wages, both within and across jobs.”

As another 2018 article states,

“[P]olicymakers, economists, and legal scholars . . . overwhelmingly conclude that the

harms of noncompetes far outweigh their potential benefits.”

In my own research, I

develop an explanatory dynamic lens to understand the robust evidence from multiple

empirical studies.

The empirical exploration of noncompetes has allowed more

confidence, comparison, and nuance about the multiple effects of mobility restrictions in

. Gilson, supra note 2, at 576.

. Id. at 609.

. Id. at 596.

. Evan Starr, Are Noncompetes Holding Down Wages? HARV. L. SCH. 4 (June 13, 2018) (emphasis

omitted) https://lwp.law.harvard.edu/files/lwp/files/webpage_materials_papers_starr_june_13_2018.pdf

[https://perma.cc/YX9P-HNFP].

. Rebecca Morrow, Noncompetes as Tax Evasion, 96 WASH. U. L. REV. 265, 268 (2018).

. For a more in-depth look at the empirical evidence behind how noncompetes harm employers, see

generally LOBEL, supra note 9 (describing the harms of restrictive employment covenants); Diego Puga, The

Magnitude and Causes of Agglomeration Economies, 50 J. REGIONAL SCI. 203 (2010); Tomas Havranek &

Zuzana Irsova, Estimating Vertical Spillovers from FDI: Why Results Vary and What the True Effect Is, 85 J.

INT’L ECON. 234, 235 (2011); David B. Audretsch & Maryann P. Feldman, R&D Spillovers and the Geography

of Innovation and Production, 86 AM. ECON. REV. 630, 632 (1996); RICHARD FLORIDA, THE RISE OF THE

CREATIVE CLASS AND HOW IT’S TRANSFORMING WORK, LEISURE, COMMUNITY AND EVERYDAY LIFE 219 (2002)

(“Not only do people remain highly concentrated, but the economy itself—the high-tech, knowledge-based, and

creative-content industries that drive so much of economic growth—continues to concentrate in specific places

from Austin and Silicon Valley to New York City and Hollywood, just as the automotive industry once

concentrated in Detroit.”); Kenneth J. Arrow, The Economic Implications of Learning by Doing, 29 REV. ECON.

STUD. 155 (1962); KENNETH J. ARROW, COLLECTED PAPERS OF KENNETH J. ARROW: THE ECONOMICS OF

INFORMATION (1984).

936 The Journal of Corporation Law [Vol. 45:4

the labor market. Much of my goal in Talent Wants to be Free and subsequent research has

been to unpack the effects of mobility policies and show they are varied and multilayered.

The benefits begin when a company hires an employee and these benefits continue over

time. Of course, there are also costs to mobility and the tradeoffs between these costs and

benefits vary in relation to firm, industry, and region. The comparative regional studies are

aimed at understanding these tradeoffs. Gilson focused his inquiry on the comparison

between Silicon Valley on the San Francisco Peninsula and Route 128 outside of Boston.

The California Business and Professions Code voids “every contract by which anyone is

restrained from engaging in a lawful profession, trade, or business.”

Massachusetts’ law

is similar to most other states in enforcing “reasonable” noncompetes.

As Gilson

explained, California’s unique policy of voiding noncompetes dates back to 1872 as “a

serendipitous result of the historical coincidence between the codification movement in the

United States and the new state’s efforts at developing a coherent legal system out of its

conflicting inheritance of Spanish, Mexican, and English law.”

Indeed, initially this

section was California’s only antitrust statute and its language shares the same terms as the

federal antitrust law which prohibits restraints on trade. Gilson argued that Silicon Valley

experienced higher and more rapid growth than Boston’s Route 128 high technology area,

thanks in part to California’s voidance of contracts that restrain trade.

I argue that a policy which protects the ability of employees to move freely and

frequently between competitors supports at least ten distinct, albeit related, goals of

economic development:

1. Knowledge Spillovers: Gilson focused on the explanatory power of the benefits of

noncompetes’ non-enforcement on knowledge spillovers—the flow of tacit knowledge that

is not codified and is enriched by the movement of employees in a region. Mobility

increases knowledge exchange which in turn fuels innovation. Both sending and receiving

companies gain from this approach.

2. Dense Networks: Slightly different from knowledge spillovers is the densification

of talent pools and professional networks. When inventive employees move more often,

which Silicon Valley employees do,

they know, interact, and engage with each other

more and more frequently, creating denser inventor networks.

High density creative

communities enjoy a Medici Effect, the experience of people working within richer

professional networks and becoming better positioned to innovate.

Observing patenting

rates in relation to high mobility regions shows that not only is the number of co-authored

patents higher, but also the absolute overall number of patents is higher.

Higher job

. CAL. BUS. & PROF. CODE § 16600.

. MASS. GEN. LAWS ch. 149 §24L; Allied Adjustment Serv. v. Heney, 484 A.2d 1189, 1191 (N.H. 1984).

. Gilson, supra note 2, at 579.

. Id. at 607–09.

. Ulrich Kaiser et al., Does the Mobility of R&D Labor Increase Innovation?, 110 J. ECON. BEHAV. &

ORG. 91, 91 (2015).

. Bruce Fallick et al., Job-Hopping in Silicon Valley: Some Evidence Concerning the Microfoundations

of a High-Technology Cluster, 88 REV. ECON. & STAT. 472, 478 (2006).

. Lee Fleming & Matt Marx, Managing Creativity in Small Worlds, 48 CAL. MGMT. REV. 6, 15 (2006).

. FRANS JOHANSSON, THE MEDICI EFFECT (2004).

. Adam B. Jaffe et al., Geographic Localization of Knowledge Spillovers as Evidenced by Patent

Citations, 108 Q.J. ECON. 577, 595–96 (1993); Jasjit Singh, Collaboration Networks as Determinants of

Knowledge Diffusion Patterns, 51 MGMT. SCI. 756, 768 (2005); Ernest Miguélez & Rosina Moreno, Research

Networks and Inventors’ Mobility as Drivers of Innovation: Evidence from Europe, 47 REGIONAL STUD. 1668,

2020] Noncompetes, Human Capital Policy & Regional Competition 937

mobility indeed correlates with more collaboration, breakthroughs, and innovation.

3. Match Quality: When people are allowed to move, employees’ job search time is

shortened, and talent and skill are better employed in the job market. Employers can find

the best people for the job faster. Conversely, employees locked into single positions are

losses for search and match pools: “In blunt economic terms, the deadweight loss from

controls and restrictions over human capital is the person herself who is prevented from

using her talent, skill, and passion.”

Noncompetes not only reduce the overall mobility

of employees but they redirect employees to non-competition, a move away from one’s

expertise and industry.

4. Agglomeration Economies: In Legal Infrastructure, Gilson describes how increased

mobility and knowledge spillovers contribute to the formation and growth of clusters of

innovation.

Agglomeration economies benefit from economies of scale external to the

firm.

The availability of a rich talent pool means that firms locating to the region will

have an easier time finding skilled talent; at the same time, skilled employees are drawn to

a place that has a concentration of jobs.

Notably, while Legal Infrastructure focused on

agglomeration in Silicon Valley’s tech industry, Gilson mentions in a footnote the biotech

industry in San Diego—where I live and work—as another example of geographical

clustering.

In Southern California, the entertainment industry is another important

example of agglomeration.

Agglomeration economies result from knowledge spillovers

and create an environment that sustains such ongoing information exchanges.

5. Motivation & Behavior: An aspect that is not contemplated in Legal Infrastructure

is the behavioral effect of restricting human capital. When employees are asked to sign

away their future employment opportunities, their career trajectories are narrowed. In an

article I published together with On Amir, we find that employees who were asked to sign

restrictions, including merely partial restrictions on their employment opportunities,

performed worse on the tasks at hand than the non-restricted employees.

Noncompetes

may discourage employees from investing in their own human capital and work

performance.

6. Carrots & Sticks: Related to behavioral incentives are findings that carrots and

1677 (2013); Gerald A. Carlino et al., Urban Density and the Rate of Invention, 61 J. URB. ECON. 389, 397–401

(2006).

. Stefano Breschi & Francesco Lissoni, Mobility of Skilled Workers and Co-invention Networks: An

Anatomy of Localized Knowledge Flows, 9 J. ECON. GEOGRAPHY 439, 450 (2009); Morten T. Hansen, Knowledge

Networks: Explaining Effective Knowledge Sharing in Multiunit Companies, 13 J. ORG. SCI. 232, 232 (2002).

. Orly Lobel, The New Cognitive Property: Human Capital Law and the Reach of Intellectual Property,

93 TEX. L. REV. 789, 848 (2015).

. Evan Starr et al., Noncompetes and Employee Mobility, 2019 ACAD. MGMT. (2019).

. Gilson, supra note 2, at 608.

. Id. at 576 (citing ALFRED MARSHALL, PRINCIPLES OF ECONOMICS 222–30 (8th ed. 16th prtg. 1964)

(1890)).

. Id.

. Id. at 622 n.164 (citing David B. Audretsch & Paula E. Stephan, Company-Scientist Locational Links:

The Case of Biotechnology, 86 AM. ECON. REV. 641, 642–49 (1996)).

. For background on spatial tendencies of knowledge spillovers, see generally Puga, supra note 22;

Havranek & Irsova, supra note 22; Audretsch & Feldman, supra note 22; FLORIDA, supra note 22.

. On Amir & Orly Lobel, How Noncompetes Stifle Performance, HARV. BUS. REV. 26, 26 Jan–Feb. 2014.

See also generally On Amir & Orly Lobel, Driving Performance: A Growth Theory of Non-Compete Law 16

STAN. TECH. L. REV. 833 (2013) (describing the relationship between noncompete agreements and employee

performance).

938 The Journal of Corporation Law [Vol. 45:4

sticks—positive rewards and negative restrictions—alternate in practice as means to

encourage retention. Without the availability of noncompetes, California employers use

more positive retention incentives including performance-based pay, bonuses, and stock

options.

These compensation structures positively incentivize performances. Thus, the

gain is double: not only are employees not bound by restrictions that may diminish their

motivation, they are offered positive incentives that can boost their motivation.

7. Entrepreneurship: While Legal Infrastructure described Silicon Valley’s regional

advantage as eschewing vertical integration, the focus was on how independent start-ups

were founded and subsequently flourished because of the ease with which employees and

their knowledge flowed in the region. More recent noncompete research reveals that

noncompetes not only suppress mobility, but impose a particular harm on

entrepreneurship.

Employee behavior is patterned by the contracts they sign beyond the

question of whether they leave or stay with their employer.

Employees bound by

noncompetes not only leave their current employer less frequently, but when they do leave,

they are also more likely to go to larger incumbent competitors rather than risk the liability

of litigation on their own. As one study concludes, “if employees are finance-constrained

and hence unable to buy out their non-compete contracts, enforcement of these agreements

prevents startup of socially profitable spinoff firms.”

Here too, new empirical studies

exploiting exogenous changes in state policies support the theoretical predictions. For

example, a study looking at a change in Florida’s noncompete law finds that stronger

enforcement of noncompetes alters the size distribution of firms and their growth.

The

study shows that the law strengthening noncompete enforcement has favored and attracted

large incumbent firms.

Such firms were more likely to enter Florida after the change, and

conversely, start-up establishment has been hindered.

Following Florida’s law change,

. See generally Mark J. Garmaise, Ties That Truly Bind: Noncompetition Agreements, Executive

Compensation, and Firm Investment, 27 J. L. ECON. & ORG. 376 (2011) (describing retention incentives); Robert

M. Grant, Toward a Knowledge-Based Theory of the Firm, 17 STRATEGIC MGMT. J. 109 (1996) (exploring

integrated knowledge within firms); Paul Oyer & Scott Schaefer, Why Do Some Firms Give Stock Options to All

Employees?: An Empirical Examination of Alternative Theories, 76 J. FIN. ECON. 99 (2005) (analyzing stock

options as an employee incentive); Yuval Feldman & Orly Lobel, The Incentives Matrix: The Comparative

Effectiveness of Rewards, Liabilities, Duties and Protections for Reporting Illegality, 88 TEX. L. REV. 1151

(2010) (describing how different reporting mechanisms encourage the internal reporting of regulatory

infractions).

. Sampsa Samila & Olav Sorenson, Noncompete Covenants: Incentives to Innovate or Impediments to

Growth, 57 MGMT. SCI. 425, 436–37 (2011); Toby E. Stuart & Olav Sorenson, Liquidity Events and the

Geographic Distribution of Entrepreneurial Activity, 48 ADMIN. SCI. Q. 175, 183 (2003).

. See also Matt Marx & Lee Fleming, Non-compete Agreements: Barriers to Entry. . . and Exit?, in 12

INNOVATION POLICY AND THE ECONOMY 39, 45 (Lerner & Stern eds., 2012) (describing how noncompetes bind

employees to employers); Matt Marx, The Firm Strikes Back: Non-compete Agreements and the Mobility of

Technical Professionals, 76 AM. SOC. REV. 695, 698 (2011) (describing how noncompetes control employees).

. James E. Rauch, Dynastic Entrepreneurship, Entry, and Non-Compete Enforcement 2 (CESifo Working

Paper Series No. 5370, 2015); see also April M. Franco & Matthew F. Mitchell, Covenants Not to Compete,

Labor Mobility and Industry Dynamics, 17 J. ECON. & MGMT. STRATEGY 581, 602–03 (2008) (describing how

noncompetes reduce spin-outs and competition).

. Hyo Kang & Lee Fleming, Non-competes, Business Dynamism, and Concentration: Evidence from a

Florida Case Study, J. ECON. & MGM’T STRATEGY (Forthcoming 2020),

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3172477 [https://perma.cc/YZ9L-8D8D].

. Id. at 28.

. Id.

2020] Noncompetes, Human Capital Policy & Regional Competition 939

the market became more concentrated.

The researchers then expanded the analysis to find

that—consistent with Florida’s shift—states that strongly enforce noncompetes tend to

have high industry concentration.

8. Brain Gain: Legal Infrastructure encouraged interstate juxtaposition to understand

the comparative advantages of different state laws. Beyond the intrinsic advantage of a

certain policy, however, is the fact that a state is alone in its adoption. In other words,

California benefits not only from its policy in the absolute, but also from the uniqueness of

its policy compared to other states. California benefits twice: from the positive effects of

mobility within the region, and from talent flow from outside of the state. States that enforce

noncompetes over time will see a brain drain, as talent moves to places where it will be

more valued.

9. Monopsonies & Wages: Noncompetes decrease wages.

Employers calibrate

compensation largely based on competing external offers. When external offers are

reduced, employers face less pressure to increase wages. Search theory and research of

labor market concentration provide a helpful framework to understand the wage effects of

noncompetes. Even without noncompetes, the nature of job search and the inherently

burdensome switching costs for employees create job markets that are prone to employer

monopsonies. Alan Manning in Monopsony in Motion explains how workers must spend

time and effort to find jobs, allowing employers to reduce compensation—including wages,

benefits, and workplace amenities—or fail to increase compensation despite the worker’s

performance because the employer knows that the employee will incur high switching costs

finding an alternative job.

Burdett and Mortensen further show how search frictions

naturally lead to employer monopsony power.

In 2016, the Council of Economic

Advisors concluded that “employers may be better able to exercise monopsony power

today than they were in past decades” which can “undermine competition[,] tend to reduce

efficiency, and can lead to lower output, employment, and social welfare.”

The effects of

noncompetes on wages are now well-documented. In 2015, Hawaii passed a law banning

noncompete and non-solicitation clauses from employment contracts in the high tech

industry.

The Hawaii ban increased employee mobility in the sector by 11% and

. Id. at 26.

. Id. at 26, 33

. Matt Marx et al., Mobility Skills, and the Michigan Non-Compete Experiment, 55 MGMT. SCI. 875, 887

(2009).

. Evan Starr, Consider This: Training, Wages, and the Enforceability of Covenants Not to Compete, 72

INDUS. & LAB. REL. REV. 783, 783 (2019) (finding “an increase from non-enforcement of non-competes to mean

enforceability is associated with a 4% decrease in hourly wages” even in lower skilled jobs).

. See generally ALAN MANNING, MONOPSONY IN MOTION: IMPERFECT COMPETITION IN LABOR

MARKETS (2003); (explaining the costs employees bear associated in landscapes with enforced noncompete

agreements; Alan Manning, Imperfect Competition in the Labor Market, in 4 HANDBOOK OF LABOR ECONOMICS

973 (2011) (the same).

. Kenneth Burdett & Dale T. Mortensen, Wage Differentials, Employer Size, and Unemployment, 39

INT’L ECON. REV. 257, 272 (1998).

. COUNCIL OF ECON. ADVISERS, LABOR MARKET MONOPSONY: TRENDS, CONSEQUENCES, AND POLICY

RESPONSES 1, 10 (Oct. 2016), https://obamawhitehouse.archives.gov/sites/default/files/page/files/20161025_

monopsony_labor_mrkt_cea.pdf [https://perma.cc/S68S-VV5D].

. Orly Lobel, High Tech Hubs & Post-Employment Restrictions: Hawaii Joins California in its Ban of

Non-Competes, CASETEXT (Aug. 17, 2015), https://casetext.com/analysis/high-tech-hubs-post-employment-

restrictions [https://perma.cc/8H46-ZSYC].

940 The Journal of Corporation Law [Vol. 45:4

increased new-hire salaries by 4%.

Other studies demonstrate that noncompetes impose

wage externalities on employees who have not signed them: in a market that enforces

noncompetes, wages and mobility are lower even for those not bound by noncompetes.

A 2017 study looking at how noncompetes affect monopsony power via employee mobility

and wages uses a matched employer-employee dataset covering the universe of jobs in 30

states.

It finds that that higher enforceability of post-employment restrictions is

associated with “longer job spells” and a greater chance of employees relocating to a

different state.

Importantly, the study also finds persistent wage-suppressing effects that

last throughout a worker’s job and employment history.

The researchers conclude that

the results strongly suggest that noncompete enforceability lowers worker welfare, reduces

employees’ bargaining power relative to the firm, and locks them into their jobs; it further

creates monopsony power, leading to deviations from the “law of one wage” (that the same

talented employee will be valued equally and offered the same salaries from different

competitors in the market if the market is fully competitive), reduces the elasticity of labor

supply, and dampens labor market dynamism, further reducing both wage competition and

mobility.

10. Equality: An overlooked aspect in studies of employee mobility and noncompete

policy is the disproportionate negative effect noncompetes have on certain demographics.

In 1957, Gary Becker predicted that discrimination would be eliminated with competition

over employees.

Becker explained that even if some employers had a taste for

discrimination, underpaid and undervalued employees would be low-hanging fruit for

competitive recruitment.

Indeed, a recent en banc decision of the Ninth Circuit Court of

Appeals on the Equal Pay Act cites Talent Wants to be Free for the proposition that

employee mobility between competitors is key for eliminating the gender pay gap.

In a

forthcoming article, Gentlemen Prefer Bonds: How Employers Fix the Labor Market, I

argue that noncompetes harm equality in several ways. First, on average, women are more

likely to have geographic constraints based on family and spousal obligations. Under such

circumstances, a noncompete that restricts an employee’s ability to compete within a region

will likely mean a professional detour, a forced sabbatical out of the labor market, or

staying longer with one’s employer rather than relocating. Second, women and minorities

are more likely to have non-monetary preferences for a workplace that is free of

discrimination and hostility and that values diversity. For example, if a woman discovers

that her employer systematically allows harassment of its female employees, she will have

. Natarajan Balasubramanian et al., Locked In? The Enforceability of Covenants Not to Compete and the

Careers of High-Tech Workers (U.S. Census Bureau Ctr. for Econ. Studies, Working Paper No. CES-WP-17-09,

Ross Sch. of Bus., Working Paper No. 1339, 2017), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=

2905782 [https://perma.cc/EWD2-M9F6].

. Starr, supra note 20.

. Balasubramanian et al., supra note 56.

. Id. at 32.

. Id. at 36.

. Id. at 34.

. GARY S. BECKER, THE ECONOMICS OF DISCRIMINATION (1957).

. Id.

. Rizo v. Yovino, 887 F.3d 453, 471–72 (9th Cir. 2018) (McKeown, J., concurring) (citing ORLY LOBEL,

TALENT WANTS TO BE FREE 49–75 (Yale Univ. Press 2013)); see Orly Lobel, Knowledge Pays: Flipping

Transparency on Its Head and the Future of Pay Equity, 120 COLUM. L. REV. (forthcoming 2020) (describing

how employee mobility is essential for pay equity).

2020] Noncompetes, Human Capital Policy & Regional Competition 941

a strong interest in examining other opportunities in the market, yet a noncompete

restricting her mobility will prevent her from escaping the discriminatory workplace. Third,

without the opportunity to receive outside offers, historical pay gaps will persist or even

widen with internal raises.

Employees do not have full information about other job

opportunities and cannot dynamically discover their true value without external offers. The

more external offers available, the more equity norms and competitive pressures from

mobility drive employers to raise wages as retention efforts. When an employee discovers

information regarding her undervalued labor compensation by receiving an external offer

from a competitor employer, the employee can use that information to negotiate a higher

salary with her current employer. The process is iterative in that if her current employer

offers to match the higher salary to retain her services, the competitor employer can come

back with an even higher job offer.

Fourth, behavioral research consistently shows gender

differences in negotiations as well as risk aversion in the face of threatened costs. Women

on average negotiate less, ask for less, and—when they do negotiate—are penalized more

often than men for being perceived as strong negotiators.

These ten related but distinct effects of noncompetes help explain why, despite

Gilson’s correct analysis that California’s mobility policy helps solve a collective action

problem, not every firm will benefit from such a policy in the same way. It may well be

that some firms will rationally assess that the wage gains from less mobility outweigh the

benefits of agglomeration and richness of talent pools. Similarly, incumbent firms might

view entrepreneurship and competition as an immediate or long-term threat. When we

study the interplay of these ten effects of talent mobility, it is obvious that there will be

strong regional and industry variations in designing optimal human capital policies.

Clearly, there will be different optimum equilibria for different geographies and for

different points in an arc of an industry. As with debates about the right lines to draw in

any pillar of intellectual property (or tax or corporate law for that matter), a one-size-fits-

all approach may not be appropriate, or at least a uniform rule such as California’s absolute

prohibition on restraints on trade may vary in its success in supporting industries as wide-

ranging as software, hardware, pharma, entertainment, and apparel. Gilson emphasized that

different industries might have different development trajectories and needs:

[I]t may well be that a state concerned with regional development today should

not blindly seek to replicate the historical source of Silicon Valley’s success.

Given the opportunity to act by design rather than by historical accident, the

better approach may be to craft a legal infrastructure that has the flexibility to

accommodate the different balance between external economies and intellectual

property rights protection that may be optimal in different industries. In contrast,

for California, where the industrial distribution already reflects the long-term

presence of Business and Professions Code section 16600, the best course may

. Orly Lobel, Companies Compete But Won’t Let Their Workers Do the Same, N.Y. TIMES (May 4, 2017),

https://www.nytimes.com/2017/05/04/opinion/noncompete-agreements-workers.html [https://perma.cc/GFL3-

GM85].

. Id.

. See generally Lobel, Knowledge Pays, supra note 64 (describing how employee mobility is essential

for pay equity); LINDA BABCOCK & SARA LASCHEVER, WOMEN DON’T ASK: NEGOTIATION AND THE GENDER

DIVIDE (2003) (analyzing how negotiation affects the wage gap).

942 The Journal of Corporation Law [Vol. 45:4

simply be staying the course.

Still, in the twenty years since Legal Infrastructure, the research has leveraged

exogenous shifts in state policy, including changes in Michigan (1980), Vermont (2005),

Oregon (2008), South Carolina (2010), Georgia (2010), and Hawaii (2015).

Studies also

examine enforcement differences between states that do enforce noncompetes—taking

California out of the equation—finding correlation between weak enforcement, increased

mobility, and positive outcomes in those regions.

In both the longitudinal and

comparative regional studies, the findings support Gilson’s initial intuition: mobility is

good for regions.

The grounds are ripe for even further research. In 2018 and 2019, Massachusetts,

Washington, Maryland, and New Hampshire each passed a new law voiding most

noncompetes.

In Massachusetts, the reform was championed by the Alliance for Open

Competition, a national group of venture capitalists, executives, and entrepreneurs

dedicated to fostering innovation. According to the Alliance campaign, noncompetes were

stifling the emergence of start-up companies, “forcing innovative entrepreneurs to take on

tremendous legal and financial risks, and hampering the ability to meet our fullest

economic potential as a nation.”

At the national level, 2016 saw the White House’s Call

for Action, which urged the ban of most noncompete restrictions and promoted improved

transparency and fairness for noncompete agreements, as well as the U.S. Treasury

Department’s Report on Non-Competes finding that noncompetes reduce and harm

innovation.

Several federal bills have also been drafted that would limit the use of

noncompetes, either for low-wage workers or more broadly.

. Gilson, supra note 2, at 629.

. See generally, Marx et al., Mobility, Skills, and the Michigan Non-Compete Experiment, 55 MGMT. SCI.

875 (2009) (discussing empirical results of noncompete enforcement in Michigan); Garmaise, supra note 41

(analyzing the effects of disparate noncompete enforcement levels around the country); Sharon Belenzon & Mark

Schankerman, Spreading the Word: Geography, Policy, and Knowledge Spillovers, 95 REV. ECON. & STAT. 884

(2013); Kenneth A. Younge et al., How Anticipated Employee Mobility Affects Acquisition Likelihood: Evidence

from a Natural Experiment, 36 STRATEGIC MGMT. J. 686 (2015) (utilizing strategic factor market theory to

analyze the correlation between acquirers’ “expectations about employee departure on the firm post-acquisition”

upon acquirers’ decisions whether to bid on a firm); Starr, supra note 20, at 4 (discussing the impact in Hawai’i);

Michael Ewens & Matt Marx, Founder Replacement and Startup Performance, 31 REV. FIN. STUD. 1532 (2018)

(examining the effects of “14 states’ changes to noncompete laws from 1995 to 2016” upon venture capital

portfolio founder turnover).

. Garmaise, supra note 41, at 376.

. Gilson, supra note 2, at 577.

. Massachusetts Noncompetition Agreement Act, MASS. GEN. LAWS ch.149, § 24L (2018); Washington

Senate Votes to Restrict Non-Compete Contracts, U.S. NEWS (Mar. 5, 2019), https://www.usnews.com/news/best-

states/washington/articles/2019-03-05/washington-senate-votes-to-restrict-non-compete-contracts

[https://perma.cc/RFM7-UYNV]; Elisaveta Dolghih & Kenneth B. Walton, United States: New Hampshire Bans

Non-Compete Agreements For Low Wage Employees, MONDAQ (July 24, 2019),

https://www.mondaq.com/unitedstates/Employment-and-HR/829186/New-Hampshire-Bans-Non-Compete-

Agreements-For-Low-Wage-Employees [https://perma.cc/Y9PW-ME5Z].

. About Us, ALLIANCE FOR OPEN COMPETITION, http://opencompetition.wordpress.com/about

[https://perma.cc/52LQ-KRQW] (last visited Aug. 19, 2019).

. Ross Eisenbrey, White House Issues Call to Action on Non-Compete Clauses, EPI.ORG: WORKING

ECONOMICS BLOG (Oct. 25, 2016, 3:30 PM), http://epi.org/blog/white-house-issues-call-to-action-on-non-

compete-clauses/ [https://perma.cc/C8KD-QX55]; Dynan, supra note 7.

. Bill Would Ban Noncompetes for Low-Wage Workers—Proposed Legislation, Hum. Resources Mgmt.-

Compensation Guide (CCH) ¶ 34,005 (2015); The Workforce Mobility Act of 2018, S.2782, 115th Cong. (2018);

2020] Noncompetes, Human Capital Policy & Regional Competition 943

III. MOBILITY PENALTIES: A NONCOMPETE BY ANY OTHER NAME

As a matter of culture, high-tech entrepreneurs are the cowboys of our age. In the

United States, as Willie Nelson has told us, our heroes have always been cowboys.

Legal Infrastructure focused on non-compete clauses in employment agreements. As

with other areas of contract law, when policy closes on a practice, other practices arise that,

in essence, attempt to achieve the same negative result. In employment agreements,

restrictive covenants do not simply appear as a clause formally labeled “noncompete.”

Employment contracts regularly include non-solicitation of customers and co-workers, pre-

innovation assignment agreements, non-disclosure agreements, and a slew of other exit

penalties: “Through this web of extensively employed mechanisms, knowledge that has

traditionally been deemed part of the public domain becomes proprietary.”

Since Legal Infrastructure, the research, policy initiatives, and public debates on

employee mobility have also focused almost exclusively on the formal noncompete clause.

And yet, California’s Section 16600 is worded simply, stating that “every contract by

which anyone is restrained from engaging in a lawful profession, trade or business of any

kind is to that extent void.”

Interpreting 16600, the courts have found that both non-

solicitation clauses and other restrictive covenants imposing a penalty on an employee for

competing are invalid.

This broader interpretation of California’s policy suggests several

points. First, the regional advantage that Silicon Valley experienced from California’s

policy has relied not merely on invalidating noncompetes but also on an increased scrutiny

of other restrictions. Second, more research is needed to understand how other forms of

mobility restrictions affect regional development. There is still a dearth of scholarship of

how, for example, the enforcement of nondisclosure agreements or interpretation of trade

secrecy affects employee mobility and competition. Third, if restraints on trade are defined

as any contractual provision that increases an employee’s risk in accepting a job with a

competitor or founding their own new company, as well as any risk imposed on a

competitor when seeking to recruit an employee, then new types of contractual clauses can

be analyzed with similar tradeoffs in mind. In particular, much more research should be

devoted to the ways intellectual property—and tacit knowledge—are assigned, owned, and

policed in the job market.

Customer non-solicits are perhaps most readily understood as effectively

noncompetes because a business without clients is like a pool without water. The

externalities of non-competition should also be readily understood with regard to employee

non-solicitation agreements—which essentially reduce the job opportunities of every co-

worker that the former employee knew—regardless of whether that co-worker agreed to

be part of a restrictive regime. Co-worker non-solicit clauses are quite often drafted so

broadly that they bar not just hiring or recruiting but any “encouragement” to a former co-

Mobility and Opportunity for Vulnerable Employees (MOVE) Act, S. 1504, 114

Cong. (2015).

. Joseph Bankman & Ronald J. Gilson, Why Start-ups?, 51 STAN. L. REV. 289, 290 (1999) (citing WILLIE

NELSON, MY HEROES HAVE ALWAYS BEEN COWBOYS, ON THE ELECTRIC HORSEMAN ORIGINAL SOUNDTRACK

(Columbia Records 1979)).

. Orly Lobel, The New Cognitive Property: Human Capital Law and the Reach of Intellectual Property,

93 TEX. L. REV. 789, 792–93 (2015).

. Id. at 791.

. CAL. BUS. & PROF. CODE § 16600.

. Chamberlain v. Augustine, 156 P. 479, 480 (Cal. 1916).

944 The Journal of Corporation Law [Vol. 45:4

worker to quit and leave. Like noncompetes, a co-worker non-solicit functions as the

employer’s hedge against compensation increases.

And like with noncompetes, the harm

extends to the industry itself: the talent pool is diminished if employees leave the industry

or the country because of a mobility restraint. Relatedly, employers attempt to list within

nondisclosure agreements any information about not only customers but also co-workers,

including barring knowledge about the company’s pay scale to later offer a co-worker a

higher salary elsewhere. The knowledge that a departed employee would use to solicit a

former co-worker pertaining to a person’s skills, talent, personality, experience, and salary

is not and should not be the employer’s trade secret.

Only very recently have the

California courts recognized that employee non-solicitation clauses are unlawful restraints

on trade under Section 16600.

Another type of restrictive covenant that operates like a noncompete is the imposition

of monetary penalties for post-employment competition. In Muggill v. Reuben H.

Donnelley Corp., the California Supreme Court examined an employment clause that

terminated certain retirement benefits if a former employee engaged in competition with

his former employer.

The court concluded that the provision was unenforceable,

explaining that Section 16600 voids penalties and not only absolute noncompetes. The

Supreme Court of North Dakota (the only other state with a statute functionally identical

to California’s Section 16600) reached the same conclusion, stating that requiring an

employee to “purchase the freedom” to compete is a form of restraint on trade.

In a New

York case involving Merrill Lynch and two employees, the Court of Appeals held that a

forfeiture-for-competition clause was unenforceable, stating that “[a]n employer should not

be permitted to use offensively an anticompetition clause coupled with a forfeiture

provision to economically cripple a former employee and simultaneously deny other

potential employers his services.”

Still, such exit penalties are far less studied, and the

tradeoffs their use entails call for further research and analysis.

Trade secrets, the neglected stepchild of intellectual property research, is close kin of

noncompete policy. The majority of trade secret disputes are between employers and their

current or former employees.

In the final section of Legal Infrastructure, Gilson raised

the concern that the “[i]nevitable disclosure doctrine poses a serious threat to the

interemployer spillover of proprietary tacit knowledge that allows Silicon Valley to reset

its product cycle repeatedly.”

He describes “a disturbing line of recent cases developing

a doctrine of ‘inevitable disclosure’ that threatens to turn trade secret law into the

. I am grateful to Charles Tait Graves for thoughtful discussions about this issue in particular.

. Orly Lobel, Knowledge Pays: Reversing Information Flows & The Future of Pay Equity, 120 COLUM.

L. REV. (forthcoming 2020) (manuscript at 12022); see also, e.g., Metro Traffic Control v. Shadow Traffic

Network, 27 Cal. Rptr. 2d 573, 579 (Cal. Ct. App. 1994) (stating that hiring personnel does not convert the

employee into a “trade secret”).

. AMN Healthcare, Inc. v. Aya Healthcare Serv. Inc., 239 Cal. Rptr. 3d 577, 587 (Cal. Ct. App. 2018);

Barker v. Insight Glob., No. 16-cv-07186-BLF, 2019 U.S. Dist. LEXIS 6523, at *8–9 (N.D. Cal. Jan. 11, 2019);

WeRide Corp. v. Huang, 379 F. Supp. 3d 834, 851–52 (N.D. Cal. 2019).

. Muggill v. Reuben H. Donnelley Corp., 398 P.2d 147, 149 (Cal. 1965); Frame v. Merrill Lynch, 97 Cal.

Rptr. 811, 814 (Cal. Ct. App. 1971) (citing Muggill v. Reuben H. Donnelley Corp.).

. Werlinger v. Mut. Serv. Cas. Ins. Co., 496 N.W.2d 26, 30 (N.D. 1993).

. Post v. Merrill Lynch, 397 N.E.2d 84, 89 (N.Y. 1979).

. Dreyfuss & Lobel, supra note 13, at 467.

. Gilson, supra note 2, at 626.

2020] Noncompetes, Human Capital Policy & Regional Competition 945

equivalent of a judicially imposed covenant not to compete[.]”

The California courts have

rejected the doctrine of inevitable disclosure, understanding it, like Legal Infrastructure

did, as a de facto ex post implied noncompete.

A new study, which combines analyses of

state-level trade secret laws on the doctrine of inevitable disclosure with records of job

mobility from the Current Population Survey, confirms that states that have explicitly

rejected the doctrine of inevitable disclosure enjoy significantly increased job mobility as

well as higher returns to education compared to other states.

But the inevitable disclosure

is merely one narrow aspect of the ways trade secret law and litigation shape employee

mobility. In 2016, the Defend Trade Secrets Act (“DTSA”) created a federal civil cause of

action for trade secret misappropriation.

The DTSA expands the secrecy ecology in

several ways.

The DTSA includes very broad definitions of “trade secret” and

“misappropriation.”

The remedies under the DTSA are expansive as well—and

prosecutorial activity under the Economic Espionage Act, which the DTSA amended, has

also risen.

In August 2019, for example, prosecutors brought criminal charges against a

former Waymo employee who moved to Uber and triggered the now-settled Waymo-Uber

autonomous vehicle technology dispute.

The criminalization of trade secrecy disputes

and increased activity to prevent potential leaks of innovation has raised red flags and

voices of concern from top research leaders. In a 2018 New York Times Op Ed, MIT

President L. Rafael Reif wrote about a more farsighted vision in the race for innovation.

Reif wrote, “[S]topping intellectual property theft and unfair trade practices—even if fully

effective—would not allow the United States to relax back into a position of unquestioned

innovation leadership.”

Rather, we need to rely on a tradition of research, “rooted in a

national culture of opportunity and entrepreneurship, inspired by an atmosphere of

intellectual freedom, supported by the rule of law and, crucially, pushed to new creative

heights by uniting brilliant talent from every sector of our society and every corner of the

. Id. at 580.

. Whyte v. Schlage Lock Co., 101 Cal. App. 4th 1443, 1462 (2002).

. I. P. L. Png & Sampsa Samila, Trade Secrets, Non-Competes, and Mobility of Engineers and Scientists:

Empirical Evidence 25 (last revised Feb. 2015) http://ssrn.com/abstract=1986775 [https://perma.cc/6SBK-ALG3]

or http://dx.doi.org/10.2139/ssrn.1986775 [https://perma.cc/KH5A-K5VL[ (as of Dec. 2, 2013). See also Deqiu

Chen et al., Human Capital Driven Acquisition: Evidence from the Inevitable Disclosure Doctrine (Univ. of Int’l.

Bus. & Econ., 2020), https://ssrn.com/abstract=2713600 [https://perma.cc/6FWJ-V865]

or http://dx.doi.org/10.2139/ssrn.2713600 [https://perma.cc/5GZB-LWVS] (finding a significant increase in the

likelihood of being acquired for firms headquartered in states that recognize such a doctrine relative to firms

headquartered in states that do not).

. Defend Trade Secrets Act of 2016, Pub. L. No. 114-153, 130 Stat. 376.

. 18 U.S.C. § § 1836, 1839 (2016); Orly Lobel, The DTSA and the New Secrecy Ecology, 1 BUS.

ENTREPRENEURSHIP & TAX L. REV. 369 (2017).

. Bret A. Cohen et al., Explaining the Defend Trade Secrets Act, AM. BAR ASS’N (Sept. 20, 2016),

https://www.americanbar.org/groups/business_law/publications/blt/2016/09/03_cohen/ [https://perma.cc/SH5K-

BRZB].

. Dreyfuss & Lobel, supra note 13; Orly Lobel, NDAs Are Out of Control. Here’s What Needs to Change,

HARV. BUS. REV. (Jan. 30, 2018), https://hbr.org/2018/01/ndas-are-out-of-control-heres-what-needs-to-change

[https://perma.cc/LKT5-M6D2].

. Waymo LLC v. Uber Technologies, No. 3:17-cv-00939 (N.D. Cal. Feb. 6, 2020).

. L. Rafael Reif, China’s Challenge Is America’s Opportunity, N.Y. TIMES (Aug. 8, 2018),

https://www.nytimes.com/2018/08/08/opinion/china-technology-trade-united-states.html

[https://perma.cc/V2NC-SHQC].

. Id.

946 The Journal of Corporation Law [Vol. 45:4

world.”

The inevitable disclosure doctrine was debated in the enactment of the DTSA

and the language adopted in the DTSA is a compromise. The Act states that an injunction

issued under its authority may not “prevent a person from entering into an employment

relationship” and that any conditions placed on a former employee’s employment in an

injunction must be based on “evidence of threatened misappropriation and not merely on

the information the person knows.”

100

Needless to say, this new law introduces further

uncertainty for the ability of knowledge workers to move from one competitor to another.

Most broadly, bringing noncompete policy to the forefront, Gilson’s contribution was

in recognizing the key role that tacit knowledge plays, as distinct from codified knowledge.

Gilson concludes Legal Infrastructure “by offering a cautionary note” on “standard law

and economics prescription to protect fully intellectual property rights.”

101

Intellectual

property focuses on codified knowledge: knowledge that has been reduced to concrete

expressions that can be filed or documented. At the same time, contracts have extended

beyond these definitions to gain ownership in intangible tacit knowledge. In You Don’t

Own Me, which centers around the content and entertainment industries in Southern

California, I find that dominant actors in a concentrated industry combine intellectual

property claims, their employees’ contractual assignment clauses, and the threat of

litigation as a sledgehammer to prevent employees from pursuing their careers and also to

exclude new and smaller competitors from competing in the industry. In particular, I find

that in the toy industry, the use of such restrictive clauses in employment agreements

helped concentrate the industry, and that in turn, market concentration follows precisely

Kenneth Arrow’s predictions: the dominant actor (Mattel) had little incentive to innovate

and to “self-cannibalize” its own successful products (Barbie).

102

The same year Gilson wrote Legal Infrastructure, he also published an article with

Joseph Bankman titled Why Start-ups?.

103

The article begins with a puzzle about

entrepreneurship: Most start-ups are founded by employees who leave their company to

spinoff a new venture in the same field, so why aren’t more employers developing their

employees’ ideas? The puzzle is even greater than it first appears since, as Bankman and

Gilson point out, companies have information advantages and tax incentives that would

give them an edge over venture capitalists and new competitors.

104

So, they write with wit,

“Having demonstrated that elephants can’t fly, what accounts for Dumbo?”

105

otherwise put, why are there start-ups? Bankman and Gilson’s answer is that an employer’s

decision to bid for an employee’s idea can create an ex ante incentive for employees to

invest in entrepreneurial payoff, diverting energy away from the company’s core research

and development.

106

They suggest that because of this dynamic, employers must clarify

internal property rights to prevent employees from hoarding their information from other

. Id.

100

. 18 U.S.C. § 1836 (2016).

101

. Gilson, supra note 2, at 580.

102

. See Kenneth J. Arrow, Economic Welfare and the Allocation of Resources for Invention, in

THE RATE AND DIRECTION OF INVENTIVE ACTIVITY: ECONOMIC AND SOCIAL FACTORS 609, 620 (1962)

(describing the formula to decide market concentration and the incentive to innovate).

103

. Joseph Bankman & Ronald J. Gilson, Why Start-ups?, 51 STAN. L. REV. 289 (1999).

104

. Id. at 289.

105

. Id. at 299.

106

. Id. at 304.

2020] Noncompetes, Human Capital Policy & Regional Competition 947

team members and from the company.

107

But Why Start-ups? does not consider the

converse problem: that companies will develop such strongholds on employees’

knowledge, ideas, and potential for innovation that entrepreneurship will be severely

impeded. California courts have deemed “trailer clauses” or “holdover clauses” that assign

employee inventions post-employment to the employer to be the functional equivalent of

noncompetes and thereby void under Section 16600. In Applied Materials, Inc. v.

Advanced Micro-Fabrication Equipment, Inc., the Court held an employer’s assignment

clause invalid because it required assignment of inventions related to the employee’s work

with the employer within one year after termination of employment.

108

The Court held that

the assignment clause was void under Section 16600 as an unlawful restriction on

employee mobility because the assignment clause operated to restrict the employee’s job

opportunities after he left the employer.

109

In California, employers work around Applied

Materials by inserting a reporting clause, requiring that for one year after leaving the

company, the employee must report to the former employer any inventions they have made

in the field.

110

But again, more pervasive and impactful than trailer clauses is the use of

standard broad assignment agreements and standard IP claims to prevent employee

mobility and spinoffs. Today, standard employment contracts include far more than what

intellectual property statutes have defined as proprietary. Such clauses regularly define

“inventions” broadly, including items such as “all discoveries, improvements, processes,

developments, design, know-how, computer data programs, and formulae, whether

patentable or unpatentable.”

111

When employees bind themselves to a contract that pre-assigns their innovation

potential, the likelihood that they will leave and become an entrepreneur—coming from

the French word entreprendre, “one who undertakes innovations”—is severely decreased.

The issue then becomes not why start-ups, but rather, why not more start-ups? Here again,

evidence is emerging that women face higher costs in defending against litigation and in

returning to paid employment if they leave to found startups, further suggesting a

disproportionate harm to equality when human capital is constrained.

112

Two decades after Legal Infrastructure, the article can continue to set an agenda of

expanding the inquiry and lessons from noncompete scholarship to the research of human

capital policy and intellectual property law more broadly.

IV. BEYOND ENFORCEABILITY: PERVASIVE NONCOMPLIANCE WARRANTS A REGULATORY

APPROACH TO MOBILITY

As Gilson wrote in Legal Infrastructure, Silicon Valley has been the subject of

national and international envy. Gilson’s article compared Boston’s tendencies of vertical

107

. Id. at 305–07.

108

. Applied Materials, Inc. v. Advanced Micro-Fabrication (Shanghai) Equip., Inc., 630 F. Supp. 2d. 1084,

1091 (N.D. Cal. 2009).

109

. Id.

110

. Charles Tait Graves, Is the Copyright Act Inconsistent with the Law of Employee Invention Assignment

Contracts?, 8 N.Y.U. J. INTELL. PROP. & ENT. L. 1, 16 (2018).

111

. LOBEL, supra note 10.

112

. Matt Marx, Employee Non-compete Agreements, Gender, and the Timing of Entrepreneurship 3 SSRN

(2018), https://ssrn.com/abstract=3173831 [https://perma.cc/E5AA-7NEJ].

948 The Journal of Corporation Law [Vol. 45:4

integration with the benefits of a culture of entrepreneurship and dissemination of

knowledge across firms: “With this local industry structure, a single company need not be

a technological leader in every stage of a product’s manufacturing process.”

113

But when

we look today at the Valley, a puzzle persists. In many new, so-called disruptive fields, we

see clear winners, or “category kings.”

114

Vertical integration and dominance of companies

like Google, Amazon, and Uber are still very much a pattern.

While there are many reasons for market dominance, one explanation should be

considered relevant to our inquiry: that California’s policy of non-enforcement of

noncompetes is underenforced. If this is the case, even in California, mobility is overly

restricted and suboptimal. This explanation is supported by recent findings indicating that

California employers insert noncompetes into their employment contract at similar rates as

non-California employers.

115

Invalid noncompetes still chill the movement of employees.

The vast majority of invalid contracts will not be tested in court. Rather, they will have in

terrorem effect on employees. A recent study of 11,500 labor force participants finds that

signing a noncompete results in less mobility and redirects mobility away from competitors

to noncompetitors whether the noncompete was signed in a state the enforces noncompetes

or not.

116

Another new paper, by Sarath Sanga, examines executive contracts and finds that

in California, where noncompetes are void, firms pay severance in discretionary

installments to induce their executives to comply with the non-enforceable noncompete

agreements that still appear in these contracts.

117

By contrast, Sanga finds that outside

California, the same firms pay non-discretionary severance up front because the

noncompetes are enforceable.

118

Moreover, enforcement of such restrictions, including the

threat of enforcement of unenforceable restrictions, will vary systematically with different

types of companies. For example, larger companies with sufficient legal and financial

resources can be more aggressive in driving out competition even when their legal claims

are on weak grounds.

119

Incumbents may use litigation as a reputation strategy to deter

other employees from leaving.

120

In a footnote in Legal Infrastructure, Gilson considers “a Coasian response to section

16600.”

121

He suggests that Silicon Valley employers have likely not taken the route of

contracting around 16600 because of “the employers’ experience that the net effect of high

velocity employment was positive.”

122

But this positive calculation may well change once

113

. Gilson, supra note 2, at 591.

114

. Kenneth A. Bamberger & Orly Lobel, Platform Market Power, 32 BERKELEY TECH. L.J. 1051, 1085

(2017).

115

. Evan Starr et al., Noncompetes in the U.S. Labor Force 14 (U MICH. L. & ECON., Research Paper No.

18-013, 2019); Norman Bishara et al., An Empirical Analysis of Noncompetition Clauses and Other Restrictive

Postemployment Covenants, 68 VAND. L. REV. 1, 30 (2015).

116

. See generally id. (explaining how noncompetes redirect mobility).

117

. Sarath Sanga, Incomplete Contracts: An Empirical Approach, 34 J.L. ECON. & ORG. 650, 650 (2018).

118

. Id.

119

. See generally Patrick Bolton & David S. Scharfstein, A Theory of Predation Based on Agency Problems

in Financial Contracting, 80 AM. ECON. REV. 93 (1990) (showing how larger companies use lawsuits to suffocate

competition).

120

. Rajshree Agarwal et al., Reputations for Toughness in Patent Enforcement: Implications for Knowledge

Spillovers Via Inventor Mobility, 30 STRATEGIC MGMT. J. 1349, 1350 (2009); Jamal Shamsie, The Context of

Dominance: An Industry-Driven Framework for Exploiting Reputation, 24 STRATEGIC MGMT. J. 199, 199–206

(2003).

121

. Gilson, supra note 2, at 610 n.105.

122

. Id.

2020] Noncompetes, Human Capital Policy & Regional Competition 949

a region like Silicon Valley reaches a certain stage in its development and once certain

companies reach a dominant position in their industry. Then it may well be that the

tradeoffs between the different effects of job mobility are weighed in favor, from the

perspective of the employer, of non-mobility. In particular, the gains from depressed wages

may well outweigh the marginal gains of agglomeration, knowledge flows, and innovation.

Some of the largest high-tech companies in the Valley blatantly violated California’s

mobility policy, as well as federal antitrust law, in reaching gentlemen’s agreements not to

hire each other’s employees. In 2010, upon the discovery of these informal agreements, the

Antitrust Division of the U.S. Department of Justice filed a complaint against these

companies, including Apple, Google, Intel, eBay, and Pixar, deeming such do-not-hire

agreements to be collusive restraints on trade and competition.

123

Calling these practices

“blatant and egregious,” the DOJ concluded that these agreements were per se violations

of American antitrust law.

124

Following the settlement with the DOJ, in 2013 the Ninth

Circuit certified a private class of 64,000 former employees claiming that these agreements

depressed wages.

125

The lawsuit settled for over $400 million, and the DOJ announced that

in the future it will criminally investigate allegations that employers have agreed amongst

themselves on employee compensation or not to solicit or hire one another’s employees.

126

Again, these kinds of horizontal do-not-hire agreements are merely the tip of the

iceberg, given the pervasive practices of inserting unenforceable noncompetes and other

broad restrictions in employee contracts. When contracts involve unilateral terms and

unequal parties, retrospective defensive litigation—proving that clauses are

unenforceable—at the stage of a contract breach lawsuit is often inadequate. Several state

attorneys general have taken a more proactive approach. In 2016, N.Y. Attorney General

Eric Schneiderman announced settlements with three companies with overly broad

noncompetes and proposed a bill that would allow employees a private right of action for

damages if they are asked to sign an overly broad covenant.

127

A proactive approach could

also impose duties on the attorneys who draft clearly unenforceable provisions, deeming

such practice a violation of the Rules of Professional Conduct.

128

In 2019, the Center for

Public Interest Law (CPIL) at the University of San Diego School of Law sent a letter to

123

. Complaint, United States v. Adobe Systems, (D.D.C. Sept. 24, 2011) (No. 1:10-CV-01629) 2010 WL

11417874 .

124

. Bill Baer, Assistant Att’y Gen., Remarks at the Conference Call Regarding the Justice Dep’t’s

Settlement with eBay Inc. to End Anticompetitive “No Poach” Hiring Agreements (May 1, 2014),

https://www.justice.gov/opa/speech/assistant-attorney-general-bill-baer-speaks-conference-call-regarding-

justice-department [https://perma.cc/4J2Q-FZBD].

125

. In re High–Tech Employee Antitrust Litigation, 856 F.Supp.2d 1103 (N.D.Cal.2012); United States v.

eBay, Inc., 968 F.Supp.2d 1030 (N.D.Cal.2013); see also In re Animation Workers Antitrust Litigation, 123

F.Supp.3d 1175 (N.D. CA, 2015).

126

. Id.

127

. Press Release, N.Y. State Office of Att’y Gen., A.G. Schneiderman Proposes Nation’s Most

Comprehensive Bill to Curb Widespread Misuse of Non-Compete Agreements (Oct. 25, 2016),

https://ag.ny.gov/press-release/ag-schneiderman-proposes-nations-most-comprehensive-bill-curb-widespread-

misuse-non [https://perma.cc/894S-2GHR].

128

. Letter from Robert C. Fellmeth, Exec. Dir., Univ. of San Diego Ctr. for Pub. Interest Law, & Bridget

Fogarty Gramme, Admin. Dir., Univ. of San Diego Ctr. for Pub. Interest Law, to Andrew Tuft, Staff Counsel,

State Bar of Cal. (June 5, 2019), https://www.sandiego.edu/cpil/documents/2019_0605_COPRAC_Ethics_

Opinion_Noncompetes.pdf [https://perma.cc/5FFX-E99Q] (on file with author); Robert C. Fellmeth, Attorney

Ethics and Unenforceable Employment Contracts, L.A. & S.F. DAILY J. (June 11, 2019),

https://www.dailyjournal.com/articles/352891 [https://perma.cc/JF7M-HZ5Q].

950 The Journal of Corporation Law [Vol. 45:4

that effect to the California Bar asking for an ethical opinion: “We all know that it is

unethical for an attorney to file a clearly frivolous case, but the law as enforced is less clear

on the ethics of an attorney drafting a clearly unenforceable contract—especially as they

proliferate across the marketplace.”

129

The letter calls for a State Bar rule that allows

attorneys seeking compliance with the law to explain to clients that they are prohibited

from drafting such clauses.

130

At the federal level, the FTC and the Antitrust Division in the DOJ have only recently

begun to consider anti-competitive practices in the labor market within their scope of

regulating competition and unfair trade practices.

131

A petition to the FTC was filed in

2019 by the Open Markets Institute, the AFL-CIO, and over forty legal scholars, including

myself. The petition calls on the FTC to interpret Section 5 of the FTC Act’s prohibition

on “unfair methods of competition”

132

to promulgate a rule to ban noncompetes.

133

Similarly, the Workforce Mobility Act of 2018 is a bill that seeks to prohibit noncompetes.

The bill would fine employers who require such clauses from their employees. As noted

above, California’s Section 16600 and Section 1 of the federal Sherman Act share the

language of prohibiting contracts “in restraint of trade.”

134

Antitrust law is designed to

solve a collective action problem and as Gilson’s pioneering work taught us, this is the goal

of California’s prohibition on noncompetes. The time is ripe for an antitrust lens and

regulatory solutions to human capital policy.

V. CONCLUSION

Human capital law has become a dynamic field of both scholarly inquiry and policy

reform. A critical mass of recent studies suggests that job mobility and noncompetes have

significant effects on regional growth, competition, and innovation. These include

behavioral, dynamic, firm-level, and regional-level effects. While the harms of

noncompetes on employee wages and mobility are well-documented, these restrictions

have a disproportionately harmful effect on certain types of workers, primarily women and

racial minorities. Moreover, employers restrict mobility not merely through formal

noncompete clauses but through a myriad of human capital contractual restrictions and

corporate practices, including broad non-disclosure agreements and innovation assignment

clauses. Finally, beyond the shaping of formal human capital law on the books, scholars

and policymakers must be attentive to human capital policy in action. The prevalence of

unlawful practices, such as the inclusion of unenforceable restrictions in employment

contracts, calls for proactive policies, including antitrust and regulatory reforms to support

a competitive labor market.

129

. Fellmeth, supra note 128.

130

. See id. (asking the state of California to create an ethics rule allowing attorneys to explain to clients

they cannot file clearly frivolous cases).

131

. Antitrust Guidance for Human Resource Professionals, DEP’T OF JUSTICE & FED. TRADE COMM’N

(Oct. 2016), https://www.justice.gov/atr/file/903511/download [https://perma.cc/3W2P-HUB7].

132

. 15 U.S.C. § 45 (2006).

133

. OPEN MKT. INST., supra note 8.

134

. 15 U.S.C. § 1 (2004).