Responsible Public Affairs | Management & Human Resources | October 30, 2026

Standfirst. Noncompete clauses promise to protect relationships, training and confidential knowledge. New cross-country evidence and divergent reforms suggest a harder question for leaders: when a company can keep people from leaving, has it learned how to make them want to stay?

Executive Summary

• The OECD’s 2026 Employment Outlook estimates that roughly one-fifth to one-third of private-sector employees in 15 surveyed economies are bound by noncompete clauses. Its new analysis associates those restrictions with slower mobility, wage progression and productivity growth; the estimates and causal designs differ across findings. [1]
• Australia released draft legislation on September 7, 2026 to prohibit noncompetes for employees below its A$190,100 high-income threshold. The consultation closed October 2; the measure is a proposal, not enacted law. [2] [3]
• The United States has no operative nationwide FTC noncompete ban: the agency removed its vacated rule in February 2026. It nevertheless finalized a case-specific order in June barring Rollins from enforcing clauses against more than 18,000 workers. [4] [5]
• For employers, legitimate protection of trade secrets and customer relationships requires a narrower, evidence-led design. A blanket restraint can create legal exposure and hide weaknesses in pay, development and management.

Introduction: a contract clause is also a management decision

Most executives understand the departure risk: a sales director knows clients; a scientist understands a pipeline; a technician has learned a process that took years to refine. A noncompete seems to buy time and protect an investment. Yet a clause that prevents a former employee from joining a rival does not create loyalty, improve the job or make the successor ready.

The issue is newly consequential because governments are reassessing the balance between protection and mobility. The OECD’s 2026 analysis provides comparable evidence across 15 economies, while Australia is consulting on a ban for most earners and the U.S. Federal Trade Commission is pursuing individual cases after its broad rule collapsed. These are distinct legal paths, but they point to the same managerial test: can an employer identify the particular asset at risk, show why a narrower measure is inadequate and explain why the restriction is proportionate? [1] [2] [4]

The answer matters beyond legal compliance. Boards should ask whether restrictive covenants are protecting genuinely scarce information or substituting for succession planning, competitive compensation and a credible internal career path. Public-affairs leaders should expect the debate to cross labor, competition and innovation policy rather than sit within HR alone.

The theory: protection, bargaining power and the option to leave

A noncompete is a contractual restriction on working for or founding a competitor after employment ends. It differs from a nondisclosure agreement, which targets the use or disclosure of confidential information, and from a client non-solicitation clause, which limits approaches to defined customers. A no-poach agreement is different again: it is a restraint between employers, often without the worker’s participation. [3]

The economic case for a tailored restraint is real. Firms may invest more in training or share sensitive information when they can limit immediate appropriation by a competitor. The countervailing effect is also real: a worker with fewer outside options has less bargaining power and may remain in a poor match; rivals and new firms find it harder to recruit; knowledge circulates less readily. The OECD’s survey and firm-level analysis associates noncompetes with slower wage and productivity growth, while warning that parts of the evidence are preliminary. [1]

A useful distinction is asset protection versus labor-market control. The first begins with a demonstrable secret, relationship or investment and asks for the least restrictive protection. The second begins with a workforce category and applies a standard clause because departures are inconvenient. The legal status varies by jurisdiction, but the management difference is universal.

Question	Evidence of a defensible protection	Warning sign
What is at risk?	Identified confidential know-how, a defined client relationship or a documented investment	“Everyone knows something” is the only explanation
Who is covered?	Roles with actual access or influence, reviewed individually	Hourly and junior staff receive the same clause as executives
How long and where?	A period and scope tied to a realistic risk window and applicable local law	A broad radius or market definition copied across countries
What else works?	Confidentiality controls, targeted non-solicitation, access management, incentives and succession	The restraint is the first and only retention tool
What happens on exit?	Consistent review, written rationale and a proportionate release process	Threat letters are routine despite weak evidence of harm

Case One: Australia proposes a threshold, not a universal prohibition

Australia’s Treasury says an estimated one in five workers has a noncompete, including people in childcare, manual and clerical roles. Its September 2026 draft would ban such clauses for employees earning below A$190,100, the Fair Work Act high-income threshold cited by the government. It also canvasses restrictions on coworker non-solicitation, no-poach and wage-fixing arrangements and considers safeguards for legitimate business interests. The government’s stated aim is 2027 commencement following consultation and parliamentary passage. None of that should be described as current law. [2] [3]

The threshold is a policy compromise. It recognizes that many lower-paid workers have little access to trade secrets and limited power to negotiate, while leaving room to examine high-income roles differently. It also creates a design problem: pay is an imperfect proxy for access to sensitive information. A moderately paid specialist may know more than a highly paid manager. Employers should therefore avoid treating the threshold as a substitute for a role-level risk assessment.

Australia’s impact analysis compared the status quo, a partial ban and a full ban, preferring the partial approach. This is a proposal with an articulated rationale, not proof that its eventual effects on wages or innovation are known. [6]

Case Two: the U.S. ban failed, but the Rollins case did not disappear

In the United States, a federal court blocked the FTC’s 2024 nationwide noncompete rule. The agency abandoned its appeals in 2025 and removed the rule from the Code of Federal Regulations in February 2026. Employers therefore should not claim that a general federal ban is in force. State laws and case-specific federal enforcement remain relevant. [4]

The FTC’s Rollins action illustrates the narrower route. In April 2026, the agency alleged that the pest-control company used two-year noncompetes covering a wide range of employees, often within 75 miles of one of its more than 700 U.S. locations. The agency said the restrictions reached technicians and customer-service workers and described cease-and-desist letters and litigation against former staff. These are allegations in the complaint, not a court’s findings after trial. On June 22, the FTC finalized a consent order requiring Rollins to stop enforcing noncompetes against more than 18,000 current and former workers and notify them. A consent order resolves the matter and governs future conduct without establishing every allegation as an adjudicated fact. [5] [7]

The managerial lesson is not that all restrictions fail. It is that a uniform clause attached to roles with very different information access is difficult to justify as a precise protective instrument. An organization can spend heavily on enforcement while avoiding the harder question of why its people and customers would follow a departing worker.

Case Three: a no-hire clause can sit outside the employment contract

Adamas Amenity Services, a building-services contractor in New York and New Jersey, used provisions in agreements with client buildings that restricted the buildings from hiring Adamas workers directly without a substantial penalty, according to the FTC’s complaint. The FTC finalized a consent order on February 12, 2026 requiring Adamas and its affiliates to cease enforcing the no-hire provisions. The case shows why a review limited to employee contracts misses restrictions embedded in customer, supplier or staffing arrangements. [8]

Those arrangements can be presented as protecting a contractor’s recruitment investment and service continuity. They also restrict a worker’s route to a better offer from a client who already knows their work. A management team should distinguish a reasonable transition arrangement from a barrier to competition for labor. The FTC order establishes Adamas’s obligations; it should not be inflated into a general judicial rule for every service contract.

What the evidence supports—and what it cannot settle

The OECD reports that between one-fifth and one-third of employees in its 15-country survey sample are bound by noncompetes, and that 5% of private-sector employees said such a clause had stopped them moving jobs while 3% said it had stopped them starting a business. It also reports that workers subject to clauses move up the earnings distribution more slowly—about three percentile ranks over a 25-year career in its analysis. These are estimates from specified samples and models, not a universal effect for every occupation or country. [1]

A 2025 peer-reviewed study using changes in U.S. enforceability finds lower earnings and mobility where restrictions are stronger, especially among workers likely to sign them. That strengthens the case for a causal channel, but it does not eliminate differences in state law, industry or contract design. [9]

Evidence in favor of protection is harder to generalize because the asset being defended is often private and the alternative—what training or disclosure would have occurred without a restraint—is not directly observable. That uncertainty calls for narrower claims on both sides. A noncompete may have a legitimate use in a carefully defined role; its mere presence does not prove that value is being protected.

What Leaders Should Do Now

1. Inventory every restraint. Include employment contracts, incentive plans, equity awards, separation agreements, customer contracts, staffing agreements and franchise arrangements. Record jurisdiction, covered roles, duration, geography and enforcement history.
2. Start with the asset, not the template. Ask business owners to identify the information, relationship or investment that requires protection. Require a written reason why confidentiality, access controls or a narrower non-solicitation provision would not suffice.
3. Audit the employee experience. Compare the use of restraints with pay competitiveness, promotion rates, manager quality, regretted departures and the proportion of employees who cannot negotiate terms. A high retention rate produced by restricted exit is not the same as engagement.
4. Review exit decisions centrally. No manager should send a threat letter merely because a clause exists. A cross-functional review should examine local law, actual risk, evidence, proportionality and reputational cost.
5. Prepare for jurisdictional divergence. Australia’s measure remains proposed; the U.S. federal rule is defunct; state and national regimes vary. Maintain country-specific terms and date-stamped legal review rather than assuming a global clause can be saved with a choice-of-law sentence.
6. Test retention without a restraint. Run a controlled review in selected roles where risk is low: remove the clause, strengthen confidentiality and career development, and track retention, pay, recruitment, training and actual information incidents over time.

Conclusion

There are circumstances in which a company must protect secrets and hard-won customer relationships. That protection deserves precision. A clause applied to everyone can look like a retention strategy while doing little to improve the reasons people stay.

The new evidence should prompt boards to ask a simple, uncomfortable question: if the contractual door were open, what would we have to change to keep the people whose work matters most? The answer may be better pay, more responsibility, credible advancement, stronger managers or a clearer purpose. Those investments are harder than inserting a paragraph into an offer letter. They are also the foundation of a workforce that chooses to remain.

Key Evidence

• One-fifth to one-third: estimated share of private-sector employees bound by noncompetes in the OECD’s 15-economy survey evidence. [1]
• 5% and 3%: shares of surveyed private-sector employees reporting that a clause stopped a job move or business start, respectively. [1]
• A$190,100: income threshold in Australia’s September 7, 2026 draft ban; legislation remains pending. [2]
• More than 18,000: workers covered by the FTC’s June 22, 2026 final Rollins consent order. [7]
• February 2026: FTC removal of its vacated nationwide noncompete rule from the U.S. Code of Federal Regulations. [4]

Glossary

Noncompete: A term restricting work for, or the creation of, a competitor after departure. Non-solicitation: A restriction on approaching specified clients or coworkers. No-poach/no-hire: An arrangement between businesses restricting recruitment or direct hiring. NDA: A nondisclosure agreement protecting confidential information without necessarily restricting a subsequent job. Consent order: A binding agency resolution that does not necessarily adjudicate every factual allegation in a complaint.

References and Further Reading

1. Andrews, Dan, Andrea Garnero and Sara Holttinen, “Non-compete and related agreements: Hoarding talent, holding back growth?,” OECD Employment Outlook 2026, OECD, July 2026. Full chapter.
2. Australian Government, Ministers’ Media Centre, “Opening the door to better and higher paying jobs: banning unfair non-competes,” September 7, 2026. Release.
3. Australian Treasury, “Non-compete clauses and other restraints,” updated 2026. Policy page.
4. U.S. Federal Trade Commission, “Noncompete Rule,” status and litigation history, accessed October 30, 2026. Rule page.
5. U.S. Federal Trade Commission, “FTC Takes Action Against Noncompete Agreements, Securing Protections for Workers,” April 15, 2026. Release and allegations.
6. Australian Government, Office of Impact Analysis, “Non-compete clauses and other worker restraints,” September 21, 2026. Impact analysis.
7. U.S. Federal Trade Commission, “FTC Approves Final Consent Order in Pest-Control Noncompete Matter,” June 22, 2026. Final order release.
8. U.S. Federal Trade Commission, “Adamas,” case page, final consent order February 12, 2026. Case record.
9. Johnson, Matthew S., Kurt Lavetti and Michael Lipsitz, “The Labor Market Effects of Legal Restrictions on Worker Mobility,” Journal of Political Economy 133, no. 9 (2025): 2735–2793. DOI.
10. New York State Office of the Attorney General, “Noncompete agreements,” worker-rights guidance, accessed October 30, 2026. Guidance and cases.

Source and Methodology Note

Research cut-off: October 30, 2026. OECD survey prevalence covers 15 specified economies and private-sector employees; it must not be presented as a global workforce rate. The OECD’s wage and productivity relationships combine different empirical approaches and are qualified in its chapter. The Rollins and Adamas descriptions distinguish agency allegations from binding consent-order obligations. Australia’s draft has not passed Parliament and its intended 2027 start is conditional. U.S. state law differs materially; this article is management analysis, not jurisdiction-specific legal advice. The framework and recommendations are editorial analysis; no company-level causal claim about the effect of removing clauses is made.

Suggested Internal Links

• The Last Mile of Skills-Based Hiring
• The Burnout Blind Spot
• Green-Card Freeze Lands on the Org Chart

Publication Package

Meta title: Noncompetes and Retention: Keeping Talent Without Locking the Door

Meta description: New OECD evidence and 2026 decisions in Australia and the United States challenge noncompetes as a retention strategy. A practical framework for leaders.

Slug: keeping-talent-without-locking-the-door-noncompetes

Primary keyphrase: noncompete clauses and employee retention

Categories: Management; Human Resources

Tags: employee retention, noncompete clauses, labor mobility, talent strategy, OECD Employment Outlook, competition policy, workforce management

Excerpt: Noncompetes can protect genuine business assets, but broad clauses can mask weak retention strategies. New OECD evidence and 2026 policy decisions show why leaders should distinguish protection from control.

Hero image filename: Responsible Public Affairs | Management & Human Resources | October 30, 2026

Keeping Talent Without Locking the Door

Standfirst. Noncompete clauses promise to protect relationships, training and confidential knowledge. New cross-country evidence and divergent reforms suggest a harder question for leaders: when a company can keep people from leaving, has it learned how to make them want to stay?

Executive Summary

• The OECD’s 2026 Employment Outlook estimates that roughly one-fifth to one-third of private-sector employees in 15 surveyed economies are bound by noncompete clauses. Its new analysis associates those restrictions with slower mobility, wage progression and productivity growth; the estimates and causal designs differ across findings. [1]
• Australia released draft legislation on September 7, 2026 to prohibit noncompetes for employees below its A$190,100 high-income threshold. The consultation closed October 2; the measure is a proposal, not enacted law. [2] [3]
• The United States has no operative nationwide FTC noncompete ban: the agency removed its vacated rule in February 2026. It nevertheless finalized a case-specific order in June barring Rollins from enforcing clauses against more than 18,000 workers. [4] [5]
• For employers, legitimate protection of trade secrets and customer relationships requires a narrower, evidence-led design. A blanket restraint can create legal exposure and hide weaknesses in pay, development and management.

Introduction: a contract clause is also a management decision

Most executives understand the departure risk: a sales director knows clients; a scientist understands a pipeline; a technician has learned a process that took years to refine. A noncompete seems to buy time and protect an investment. Yet a clause that prevents a former employee from joining a rival does not create loyalty, improve the job or make the successor ready.

The issue is newly consequential because governments are reassessing the balance between protection and mobility. The OECD’s 2026 analysis provides comparable evidence across 15 economies, while Australia is consulting on a ban for most earners and the U.S. Federal Trade Commission is pursuing individual cases after its broad rule collapsed. These are distinct legal paths, but they point to the same managerial test: can an employer identify the particular asset at risk, show why a narrower measure is inadequate and explain why the restriction is proportionate? [1] [2] [4]

The answer matters beyond legal compliance. Boards should ask whether restrictive covenants are protecting genuinely scarce information or substituting for succession planning, competitive compensation and a credible internal career path. Public-affairs leaders should expect the debate to cross labor, competition and innovation policy rather than sit within HR alone.

The theory: protection, bargaining power and the option to leave

A noncompete is a contractual restriction on working for or founding a competitor after employment ends. It differs from a nondisclosure agreement, which targets the use or disclosure of confidential information, and from a client non-solicitation clause, which limits approaches to defined customers. A no-poach agreement is different again: it is a restraint between employers, often without the worker’s participation. [3]

The economic case for a tailored restraint is real. Firms may invest more in training or share sensitive information when they can limit immediate appropriation by a competitor. The countervailing effect is also real: a worker with fewer outside options has less bargaining power and may remain in a poor match; rivals and new firms find it harder to recruit; knowledge circulates less readily. The OECD’s survey and firm-level analysis associates noncompetes with slower wage and productivity growth, while warning that parts of the evidence are preliminary. [1]

A useful distinction is asset protection versus labor-market control. The first begins with a demonstrable secret, relationship or investment and asks for the least restrictive protection. The second begins with a workforce category and applies a standard clause because departures are inconvenient. The legal status varies by jurisdiction, but the management difference is universal.

Question	Evidence of a defensible protection	Warning sign
What is at risk?	Identified confidential know-how, a defined client relationship or a documented investment	“Everyone knows something” is the only explanation
Who is covered?	Roles with actual access or influence, reviewed individually	Hourly and junior staff receive the same clause as executives
How long and where?	A period and scope tied to a realistic risk window and applicable local law	A broad radius or market definition copied across countries
What else works?	Confidentiality controls, targeted non-solicitation, access management, incentives and succession	The restraint is the first and only retention tool
What happens on exit?	Consistent review, written rationale and a proportionate release process	Threat letters are routine despite weak evidence of harm

Case One: Australia proposes a threshold, not a universal prohibition

Australia’s Treasury says an estimated one in five workers has a noncompete, including people in childcare, manual and clerical roles. Its September 2026 draft would ban such clauses for employees earning below A$190,100, the Fair Work Act high-income threshold cited by the government. It also canvasses restrictions on coworker non-solicitation, no-poach and wage-fixing arrangements and considers safeguards for legitimate business interests. The government’s stated aim is 2027 commencement following consultation and parliamentary passage. None of that should be described as current law. [2] [3]

The threshold is a policy compromise. It recognizes that many lower-paid workers have little access to trade secrets and limited power to negotiate, while leaving room to examine high-income roles differently. It also creates a design problem: pay is an imperfect proxy for access to sensitive information. A moderately paid specialist may know more than a highly paid manager. Employers should therefore avoid treating the threshold as a substitute for a role-level risk assessment.

Australia’s impact analysis compared the status quo, a partial ban and a full ban, preferring the partial approach. This is a proposal with an articulated rationale, not proof that its eventual effects on wages or innovation are known. [6]

Case Two: the U.S. ban failed, but the Rollins case did not disappear

In the United States, a federal court blocked the FTC’s 2024 nationwide noncompete rule. The agency abandoned its appeals in 2025 and removed the rule from the Code of Federal Regulations in February 2026. Employers therefore should not claim that a general federal ban is in force. State laws and case-specific federal enforcement remain relevant. [4]

The FTC’s Rollins action illustrates the narrower route. In April 2026, the agency alleged that the pest-control company used two-year noncompetes covering a wide range of employees, often within 75 miles of one of its more than 700 U.S. locations. The agency said the restrictions reached technicians and customer-service workers and described cease-and-desist letters and litigation against former staff. These are allegations in the complaint, not a court’s findings after trial. On June 22, the FTC finalized a consent order requiring Rollins to stop enforcing noncompetes against more than 18,000 current and former workers and notify them. A consent order resolves the matter and governs future conduct without establishing every allegation as an adjudicated fact. [5] [7]

The managerial lesson is not that all restrictions fail. It is that a uniform clause attached to roles with very different information access is difficult to justify as a precise protective instrument. An organization can spend heavily on enforcement while avoiding the harder question of why its people and customers would follow a departing worker.

Case Three: a no-hire clause can sit outside the employment contract

Adamas Amenity Services, a building-services contractor in New York and New Jersey, used provisions in agreements with client buildings that restricted the buildings from hiring Adamas workers directly without a substantial penalty, according to the FTC’s complaint. The FTC finalized a consent order on February 12, 2026 requiring Adamas and its affiliates to cease enforcing the no-hire provisions. The case shows why a review limited to employee contracts misses restrictions embedded in customer, supplier or staffing arrangements. [8]

Those arrangements can be presented as protecting a contractor’s recruitment investment and service continuity. They also restrict a worker’s route to a better offer from a client who already knows their work. A management team should distinguish a reasonable transition arrangement from a barrier to competition for labor. The FTC order establishes Adamas’s obligations; it should not be inflated into a general judicial rule for every service contract.

What the evidence supports—and what it cannot settle

The OECD reports that between one-fifth and one-third of employees in its 15-country survey sample are bound by noncompetes, and that 5% of private-sector employees said such a clause had stopped them moving jobs while 3% said it had stopped them starting a business. It also reports that workers subject to clauses move up the earnings distribution more slowly—about three percentile ranks over a 25-year career in its analysis. These are estimates from specified samples and models, not a universal effect for every occupation or country. [1]

A 2025 peer-reviewed study using changes in U.S. enforceability finds lower earnings and mobility where restrictions are stronger, especially among workers likely to sign them. That strengthens the case for a causal channel, but it does not eliminate differences in state law, industry or contract design. [9]

Evidence in favor of protection is harder to generalize because the asset being defended is often private and the alternative—what training or disclosure would have occurred without a restraint—is not directly observable. That uncertainty calls for narrower claims on both sides. A noncompete may have a legitimate use in a carefully defined role; its mere presence does not prove that value is being protected.

What Leaders Should Do Now

1. Inventory every restraint. Include employment contracts, incentive plans, equity awards, separation agreements, customer contracts, staffing agreements and franchise arrangements. Record jurisdiction, covered roles, duration, geography and enforcement history.
2. Start with the asset, not the template. Ask business owners to identify the information, relationship or investment that requires protection. Require a written reason why confidentiality, access controls or a narrower non-solicitation provision would not suffice.
3. Audit the employee experience. Compare the use of restraints with pay competitiveness, promotion rates, manager quality, regretted departures and the proportion of employees who cannot negotiate terms. A high retention rate produced by restricted exit is not the same as engagement.
4. Review exit decisions centrally. No manager should send a threat letter merely because a clause exists. A cross-functional review should examine local law, actual risk, evidence, proportionality and reputational cost.
5. Prepare for jurisdictional divergence. Australia’s measure remains proposed; the U.S. federal rule is defunct; state and national regimes vary. Maintain country-specific terms and date-stamped legal review rather than assuming a global clause can be saved with a choice-of-law sentence.
6. Test retention without a restraint. Run a controlled review in selected roles where risk is low: remove the clause, strengthen confidentiality and career development, and track retention, pay, recruitment, training and actual information incidents over time.

Conclusion

There are circumstances in which a company must protect secrets and hard-won customer relationships. That protection deserves precision. A clause applied to everyone can look like a retention strategy while doing little to improve the reasons people stay.

The new evidence should prompt boards to ask a simple, uncomfortable question: if the contractual door were open, what would we have to change to keep the people whose work matters most? The answer may be better pay, more responsibility, credible advancement, stronger managers or a clearer purpose. Those investments are harder than inserting a paragraph into an offer letter. They are also the foundation of a workforce that chooses to remain.

Key Evidence

• One-fifth to one-third: estimated share of private-sector employees bound by noncompetes in the OECD’s 15-economy survey evidence. [1]
• 5% and 3%: shares of surveyed private-sector employees reporting that a clause stopped a job move or business start, respectively. [1]
• A$190,100: income threshold in Australia’s September 7, 2026 draft ban; legislation remains pending. [2]
• More than 18,000: workers covered by the FTC’s June 22, 2026 final Rollins consent order. [7]
• February 2026: FTC removal of its vacated nationwide noncompete rule from the U.S. Code of Federal Regulations. [4]

Glossary

Noncompete: A term restricting work for, or the creation of, a competitor after departure. Non-solicitation: A restriction on approaching specified clients or coworkers. No-poach/no-hire: An arrangement between businesses restricting recruitment or direct hiring. NDA: A nondisclosure agreement protecting confidential information without necessarily restricting a subsequent job. Consent order: A binding agency resolution that does not necessarily adjudicate every factual allegation in a complaint.

References and Further Reading

1. Andrews, Dan, Andrea Garnero and Sara Holttinen, “Non-compete and related agreements: Hoarding talent, holding back growth?,” OECD Employment Outlook 2026, OECD, July 2026. Full chapter.
2. Australian Government, Ministers’ Media Centre, “Opening the door to better and higher paying jobs: banning unfair non-competes,” September 7, 2026. Release.
3. Australian Treasury, “Non-compete clauses and other restraints,” updated 2026. Policy page.
4. U.S. Federal Trade Commission, “Noncompete Rule,” status and litigation history, accessed October 30, 2026. Rule page.
5. U.S. Federal Trade Commission, “FTC Takes Action Against Noncompete Agreements, Securing Protections for Workers,” April 15, 2026. Release and allegations.
6. Australian Government, Office of Impact Analysis, “Non-compete clauses and other worker restraints,” September 21, 2026. Impact analysis.
7. U.S. Federal Trade Commission, “FTC Approves Final Consent Order in Pest-Control Noncompete Matter,” June 22, 2026. Final order release.
8. U.S. Federal Trade Commission, “Adamas,” case page, final consent order February 12, 2026. Case record.
9. Johnson, Matthew S., Kurt Lavetti and Michael Lipsitz, “The Labor Market Effects of Legal Restrictions on Worker Mobility,” Journal of Political Economy 133, no. 9 (2025): 2735–2793. DOI.
10. New York State Office of the Attorney General, “Noncompete agreements,” worker-rights guidance, accessed October 30, 2026. Guidance and cases.

Source and Methodology Note

Research cut-off: October 30, 2026. OECD survey prevalence covers 15 specified economies and private-sector employees; it must not be presented as a global workforce rate. The OECD’s wage and productivity relationships combine different empirical approaches and are qualified in its chapter. The Rollins and Adamas descriptions distinguish agency allegations from binding consent-order obligations. Australia’s draft has not passed Parliament and its intended 2027 start is conditional. U.S. state law differs materially; this article is management analysis, not jurisdiction-specific legal advice. The framework and recommendations are editorial analysis; no company-level causal claim about the effect of removing clauses is made.

Suggested Internal Links

• The Last Mile of Skills-Based Hiring
• The Burnout Blind Spot
• Green-Card Freeze Lands on the Org Chart

#HumanResources #TalentStrategy #LaborMobility