evidoria

← Back to browse

Good practice Imported

California's Non-Compete Ban (§16600) — and the Michigan Natural Experiment That Measured Its Effect

United States of America · San Jose · See the United States of America profile

Evidence: Quasi-experimental Top 50% 67/100 · Ask Evidence Copilot about this practice

California voids virtually all employee non-compete contracts (§16600). A natural experiment when Michigan accidentally reversed its own ban in 1985 showed enforcement cuts inventor job mobility by 8.1%, rising to 16.2% for narrow technical specialists.

-8.1% %
Michigan inventor job mobility decline after non-compete re-enforcement (1985-2006)
-15.4% %
Mobility decline for inventors with firm-specific skills (1985-2006)
-16.2% %
Mobility decline for narrow technical specialists (1985-2006)

Details

Maturity
Established
Promoter
State of California — Business and Professions Code §16600
Period
1872–present
Keywords
legal policy, labour mobility, innovation ecosystems, talent retention

Context

California's Business and Professions Code Section 16600 voids virtually any contract restraining a person from engaging in a lawful profession, trade, or business. Legal scholar Ronald Gilson (1999) traced the rule to an 1872 codification, reaffirmed by the California Supreme Court in Edwards v. Arthur Andersen (2008), and argued it is a causal antecedent of Silicon Valley's high-velocity, cross-firm learning labor market, in contrast to Route 128 (Massachusetts), where non-competes are enforced.

Objectives

Test Gilson's hypothesis that a state's legal treatment of non-compete contracts shapes talent mobility and regional innovation ecosystems, using a credible natural experiment rather than cross-sectional comparison alone.

Activities

Michigan supplied an inadvertent natural experiment: the Michigan Antitrust Reform Act (MARA), passed March 27, 1985, repealed a 1905 statute that had voided non-competes in the state, without legislators ever discussing the change. Marx, Strumsky and Fleming (Management Science, 2009) used U.S. patent records for 98,468 inventors (372,908 patents, 1963-2006) and a difference-in-differences design against other non-enforcing states to isolate the effect.

Results

After MARA, Michigan inventors' job mobility fell 8.1% relative to inventors in states that kept non-competes unenforceable; the effect reached 15.4% for inventors with firm-specific skills and 16.2% for narrow technical specialists. State-level patenting rates were unaffected, indicating the mechanism is mobility, not innovation output directly.

Conclusions

The study cannot separate voluntary from involuntary job changes and relies on patent data as a proxy for the wider workforce, so results speak most directly to inventors, not all employees; Michigan's own courts also moderated enforcement over time through a 'reasonableness' doctrine. It nonetheless remains the most-cited causal test of Gilson's hypothesis linking non-compete law to a region's talent mobility and innovation ecosystem.

Implementation
Indicative cost
Low (< €50k) — A statutory default rule; no direct public expenditure required to operate.
Time to results
Long (> 3 years) — In continuous force since an 1872 codification, with measured effects only established via a 1985-2006 natural experiment analysis published in 2009.
Staffing & skills
California state legislature and courts (interpretation via case law, e.g. Edwards v. Arthur Andersen 2008)

Conditions for success

  • Blanket statutory unenforceability rather than a case-by-case 'reasonableness' test, which keeps the labor market fluid (per Gilson's hypothesis)

Common failure modes

  • Courts can reintroduce enforcement gradually via judicial doctrine, as seen when Michigan's own courts moderated enforcement of the reinstated law through a 'reasonableness' standard

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Replication kit

Reusable artefacts from this practice — as published by their sources.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful