Competition Law and Machine Labour Market Power

1. Introduction

Machine labour market power refers to the competitive and economic power that may arise when machines, robots, artificial intelligence systems, automated platforms, or algorithmic technologies increasingly perform tasks traditionally performed by human workers.

The concept has two connected dimensions:

1. Employers' market power over human labour, where technology changes employers' ability to hire, monitor, compensate, or replace workers; and
2. Market power arising from machine labour itself, where control over robots, AI systems, automation infrastructure, or algorithmic labour platforms gives an undertaking an advantage over competing firms.

Competition law is increasingly relevant because labour markets are themselves markets in which firms compete to obtain workers. Conduct that restricts worker mobility or suppresses competition for labour can therefore have competition-law implications.

The central question is:

Does control or use of machine labour merely improve productive efficiency, or does it create or reinforce market power that restricts competition in labour or product markets?

2. Meaning of Machine Labour

Machine labour includes work performed or substantially assisted by:

• industrial robots;
• autonomous vehicles;
• AI systems;
• generative AI;
• algorithmic decision systems;
• automated warehouses;
• robotic manufacturing;
• AI-powered customer service;
• autonomous logistics;
• software agents;
• automated coding systems;
• algorithmic management platforms.

Machine labour can substitute for human labour, complement human labour, or create entirely new forms of work.

3. Machine Labour and Competition Law

Traditional competition law focuses heavily on product markets.

Modern competition policy increasingly recognizes that labour markets are also competitive markets.

A firm may possess buyer-side market power, commonly called monopsony power, when it can obtain labour on terms less favorable to workers than would prevail in a competitive labour market.

Machine labour can affect this power in several ways.

Example

Suppose a factory employs 10,000 workers.

It introduces robots capable of performing many of their tasks.

The factory may now have greater bargaining power over the remaining workers because:

• fewer workers are required;
• alternative employment may be limited;
• replacement becomes easier;
• wage bargaining may weaken.

Automation itself is not necessarily anti-competitive. The competition-law issue arises when market power is created or exercised in an unlawful manner.

4. Monopsony and Labour Market Power

Monopoly

One powerful seller.

Monopsony

One powerful buyer.

In labour markets:

Employer = buyer of labour

Worker = seller of labour

A dominant employer may therefore possess monopsony power.

Potential indicators include:

• concentration of employers;
• high switching costs;
• geographic limitations;
• specialized skills;
• non-compete agreements;
• no-poach arrangements;
• information asymmetry;
• algorithmic wage-setting;
• worker dependence on a platform.

5. Machine Labour Can Increase Employer Bargaining Power

Automation may reduce the employer's dependence on workers.

For example:

Before automation:

10,000 workers → production

After automation:

4,000 workers + 2,000 robots → similar production

The employer's demand for labour may fall.

This can produce legitimate productivity benefits, but it can also alter the competitive structure of the labour market.

Competition authorities therefore need to distinguish:

technological efficiency

from

anti-competitive exploitation or exclusion.

6. Machine Labour and Labour-Market Concentration

Suppose five companies employ almost all skilled workers in a particular region.

If one company introduces highly advanced automation and becomes significantly more productive, competitors may have difficulty matching it.

The company could potentially gain power in:

• product markets;
• labour markets;
• technology markets.

This creates a relationship between:

Labour concentration + automation + technological advantage + market power.

7. Case Law 1: United States v. Topco Associates

United States v. Topco Associates, 405 U.S. 596 (1972)

Topco involved territorial restrictions among grocery suppliers.

Although not a labour case, it is a useful competition-law authority concerning agreements that allocate markets.

Principle

Agreements that divide competitive opportunities can seriously restrict competition.

Relevance to machine labour

The analogous concern can arise where employers coordinate to divide labour markets or prevent workers from moving between competing employers.

For example:

Company A will recruit workers only from one region, while Company B agrees to recruit from another.

Such arrangements may reduce competition for labour.

8. Case Law 2: United States v. Apple Inc.

United States v. Apple Inc.

The Apple litigation concerning employment-related coordination among technology companies is an important example of competition law being applied to labour-market conduct.

Historically, the U.S. Department of Justice prosecuted arrangements involving restrictions on recruiting employees among technology companies.

Principle

Agreements among employers that restrict employee mobility can raise antitrust concerns because they reduce competition among employers for labour.

Relevance

Machine labour makes this issue more significant because technology companies may compete simultaneously for:

• AI researchers;
• engineers;
• robotics specialists;
• data scientists;
• machine-learning experts.

Restricting their movement can strengthen employer bargaining power.

9. Case Law 3: Aya Healthcare Services, Inc. v. AMN Healthcare, Inc.

Aya Healthcare Services, Inc. v. AMN Healthcare, Inc.

This litigation concerned labour-market restrictions and competition among healthcare staffing firms.

The dispute involved contractual restrictions affecting workers' ability to move between employers.

Principle

Employment restrictions can have competition implications where they substantially restrict worker mobility and employer competition.

Relevance

In machine-intensive sectors, similar concerns could arise for workers with highly specialized technological skills.

10. Case Law 4: FTC v. Surescripts

FTC v. Surescripts, LLC

This case concerned market power and exclusionary conduct in healthcare-related technology markets.

Although it was not fundamentally a machine-labour case, it illustrates how control over technological infrastructure can reinforce market power.

Relevance

The same reasoning can be applied conceptually to automated labour infrastructure where one platform controls:

• worker access;
• employer access;
• algorithms;
• job allocation;
• performance evaluation.

11. Case Law 5: Ohio v. American Express Co.

Ohio v. American Express Co., 585 U.S. 529 (2018)

The Supreme Court analyzed a two-sided transaction platform.

The importance of the case extends beyond payment systems because many modern digital labour platforms are also two-sided markets.

For example:

Workers ↔ Platform ↔ Employers/Customers

A platform may simultaneously serve:

• workers;
• employers;
• consumers.

Relevance

Machine-labour platforms may need to be analyzed across both sides of the platform.

A restriction benefiting one side can affect competition on the other side.

12. Case Law 6: United States v. Microsoft Corp.

United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001)

Microsoft involved exclusionary conduct in a technology platform environment.

Principle

A powerful technology platform can unlawfully use its position to restrict competitive opportunities in related markets.

Relevance to machine labour

A dominant AI or automation platform might control:

• worker-management software;
• robotics systems;
• AI agents;
• operating infrastructure;
• employment marketplaces.

If it uses control over one layer to disadvantage competing systems, Microsoft provides an important analytical analogy.

13. Case Law 7: Commercial Solvents

Commercial Solvents Corporation v Commission

Joined Cases 6/73 and 7/73 (1974)

Commercial Solvents concerned upstream market power and downstream competition.

Principle

A dominant undertaking controlling an important input cannot necessarily use that position to eliminate downstream competitors.

Relevance to machine labour

Consider:

Robotics manufacturer → automated manufacturing firms

If a dominant robotics supplier restricts access to critical technology to disadvantage competing manufacturers, the upstream/downstream foreclosure principles become relevant.

14. Case Law 8: Bronner v Mediaprint

Oscar Bronner GmbH & Co. KG v Mediaprint

Case C-7/97 (1998)

Bronner established important principles regarding refusal of access to infrastructure.

Relevance

Machine labour may depend upon:

• robotic infrastructure;
• AI compute;
• automated logistics;
• labour-management platforms;
• specialized machine interfaces.

A firm controlling such infrastructure does not automatically have a legal obligation to provide access. The stringent conditions associated with the essential-facilities doctrine remain relevant.

15. Labour Market as a Relevant Market

Competition authorities may define labour markets according to:

Skill

Example:

AI researchers.

Occupation

Example:

Software engineers.

Geography

Example:

AI engineers in a particular city.

Industry

Example:

Robotics engineers employed in automotive manufacturing.

Platform

Example:

Workers dependent upon one digital labour platform.

The relevant labour market can therefore be narrower than the general employment market.

16. Machine Labour and Monopsony

A machine-intensive company may become a powerful labour buyer when:

• it employs a large proportion of specialized workers;
• competitors cannot easily recruit;
• workers have significant switching costs;
• geographic mobility is limited;
• skills are highly specialized;
• employment is concentrated among a few firms.

Automation can reinforce this position if machines reduce the number of alternative employers or jobs.

17. No-Poach Agreements

A no-poach agreement occurs when competing employers agree not to recruit one another's workers.

Example:

Company A and Company B agree that neither will hire the other's AI engineers.

This can reduce:

• worker mobility;
• wage competition;
• employer competition;
• innovation;
• recruitment opportunities.

Such agreements can raise serious competition-law concerns independently of whether machine labour is involved.

18. Wage-Fixing Agreements

An even more direct restriction occurs where competing employers agree on wages or employment terms.

Example:

Five robotics companies agree that machine-learning engineers will receive no more than ₹20 lakh annually.

Such coordination can suppress competition for labour.

The introduction of automation does not make such coordination legitimate.

19. Algorithmic Wage Setting

Machine labour creates a new concern: algorithmic wage coordination.

Suppose several employers use the same algorithm to determine:

• wages;
• bonuses;
• staffing levels;
• scheduling;
• recruitment.

If the system enables competitors to coordinate their labour-market behavior, competition concerns may arise.

The important question is not simply:

“Was an AI system used?”

It is:

“Did the technology facilitate or implement conduct that restricts competition?”

20. Algorithmic Management

Platforms increasingly use algorithms to determine:

• worker assignments;
• pay;
• schedules;
• performance;
• incentives;
• access to jobs;
• termination.

This can improve efficiency.

But it can also create market power if workers have no meaningful alternative platform.

21. Machine Labour and Platform Dominance

Consider a delivery platform controlling:

• customer demand;
• automated worker allocation;
• pricing;
• worker ratings;
• payment.

Workers may become dependent on the platform.

If competitors cannot attract sufficient workers or customers, network effects can strengthen the platform's position.

This creates a feedback loop:

More workers → better service → more customers → more jobs → more workers.

22. Switching Costs

Machine-labour platforms can impose substantial switching costs.

Workers may lose:

• reputation scores;
• accumulated ratings;
• customer relationships;
• platform-specific training;
• bonuses;
• preferred status.

If these benefits cannot be transferred to competitors, worker mobility may decline.

23. Data as a Source of Labour-Market Power

Automated labour platforms can collect:

• worker productivity;
• acceptance rates;
• performance;
• availability;
• location;
• customer ratings;
• compensation history.

The platform may use this data to optimize labour allocation.

Competitors lacking comparable data may face entry barriers.

Thus:

Data → better algorithm → more workers/customers → more data

can become a self-reinforcing cycle.

24. Machine Labour and Vertical Integration

Suppose a company owns:

1. robots;
2. robotics software;
3. manufacturing facilities;
4. logistics systems.

It may become vertically integrated across the entire production chain.

Vertical integration can produce efficiencies.

But competition concerns may arise if the company uses control at one level to exclude competitors at another.

25. Automation and Entry Barriers

Advanced automation can increase capital requirements.

A new competitor may need:

• expensive robots;
• AI systems;
• cloud infrastructure;
• specialized software;
• skilled technicians;
• large-scale data.

These costs may create barriers to entry.

However, high investment requirements are not automatically anti-competitive.

The question is whether barriers arise from legitimate economies of scale or from exclusionary conduct.

26. Machine Labour and Product-Market Dominance

Machine labour may create significant cost advantages.

For example:

Company A: 1,000 human workers.

Company B: 200 workers + advanced automation.

Company B may have:

• lower costs;
• faster production;
• greater capacity;
• more consistent quality.

Such advantages can represent competition on the merits.

Competition law should not penalize firms merely because they successfully automate.

27. When Automation Becomes a Competition Concern

Automation may become relevant to competition law where it is combined with conduct such as:

• exclusionary contracts;
• discriminatory access;
• no-poach agreements;
• wage fixing;
• predatory conduct;
• refusal to deal;
• tying;
• exclusive arrangements;
• acquisition of emerging competitors;
• algorithmic coordination.

28. Machine Labour and Merger Control

A dominant company may acquire a robotics startup.

The startup might have:

• little current revenue;
• few customers;
• innovative technology;
• potential to reduce labour costs dramatically.

Competition authorities may consider whether the acquisition removes an important potential competitor.

This is especially relevant where traditional turnover-based thresholds might underestimate the target's competitive significance.

29. Killer Acquisitions in Automation

A large automation company could acquire a startup developing:

• autonomous robots;
• AI workers;
• automated coding;
• warehouse robotics;
• autonomous delivery.

The acquisition could eliminate a future technological competitor.

Competition authorities may therefore examine:

• innovation pipelines;
• patents;
• research capabilities;
• future market entry;
• technological substitutability.

30. Machine Labour and Innovation

Competition in machine labour is not limited to prices.

It may involve:

• productivity;
• safety;
• robotics design;
• AI capabilities;
• worker augmentation;
• automation quality;
• energy efficiency.

An overly concentrated market could reduce incentives to develop better machine-labour technologies.

31. Positive Effects of Machine Labour

Competition law should recognize legitimate efficiencies.

Machine labour may:

1. Increase productivity

Machines can perform repetitive tasks efficiently.

2. Reduce costs

Automation may reduce production expenses.

3. Improve safety

Dangerous tasks can be automated.

4. Increase output

Companies can produce more with fewer resources.

5. Encourage innovation

Automation can create new products and services.

6. Complement human labour

AI can increase human productivity rather than simply replacing workers.

32. Potential Anti-Competitive Effects

Potential concerns include:

A. Labour displacement

Large-scale automation can reduce demand for certain labour categories.

B. Monopsony

Remaining workers may have fewer alternative employers.

C. Platform dependence

Workers may depend on one algorithmic platform.

D. Data concentration

A dominant platform may possess superior labour-market data.

E. Algorithmic coordination

Common algorithms may facilitate coordinated conduct.

F. Entry barriers

Automation infrastructure can be expensive.

G. Vertical foreclosure

Infrastructure owners may disadvantage competing employers.

33. Human Labour and Machine Labour as Complements

Machine labour does not always replace humans.

Consider:

AI + doctor

Robot + technician

AI + lawyer

Autonomous vehicle + safety operator

Competition analysis should therefore consider whether machines:

• replace workers;
• complement workers;
• create new occupations;
• alter skill requirements.

34. Competition Effects Across Time

Machine labour can change market power dynamically.

Stage 1

Human labour dominates.

Stage 2

Automation supplements workers.

Stage 3

Automation becomes essential.

Stage 4

A few firms control automation infrastructure.

Stage 5

Infrastructure providers may influence both product and labour markets.

Therefore, competition authorities should consider long-term competitive effects, not merely current market shares.

35. Regulatory Framework

A useful competition-law framework is:

Step 1 — Identify the market

Is the relevant market:

• labour;
• robotics;
• AI infrastructure;
• software;
• manufacturing?

Step 2 — Determine market power

Consider:

• market share;
• alternatives;
• switching costs;
• network effects;
• entry barriers.

Step 3 — Identify conduct

Look for:

• no-poach;
• wage fixing;
• exclusivity;
• algorithmic coordination;
• discriminatory access;
• tying;
• foreclosure.

Step 4 — Analyze effects

Examine:

• wages;
• worker mobility;
• entry;
• innovation;
• productivity;
• consumer prices.

Step 5 — Evaluate efficiencies

Consider:

• automation;
• investment;
• productivity;
• safety;
• innovation.

Step 6 — Determine remedy

Potential remedies may include:

• ending restrictive agreements;
• access obligations;
• interoperability;
• non-discrimination;
• merger remedies;
• monetary penalties.

36. Important Distinction: Automation vs Anti-Competitive Conduct

37. Practical Example

Suppose AutoTech Ltd. operates 70% of the robotics market in a region.

It also operates an employment platform used by most robotics engineers.

AutoTech introduces an AI system that:

• determines salary ranges;
• recommends hiring decisions;
• controls worker access;
• predicts employee departure;
• provides discounts to companies that agree to use AutoTech's employment platform exclusively.

The competition analysis could examine:

1. AutoTech's position in robotics;
2. its position in labour-market infrastructure;
3. whether the employment platform constitutes a separate market;
4. whether exclusivity forecloses rival platforms;
5. whether algorithmic wage-setting facilitates coordination;
6. whether workers can easily switch;
7. whether the conduct produces efficiency benefits;
8. whether competing employers can access labour on reasonable terms.

38. Indian Competition-Law Perspective

Under the Competition Act, 2002, machine-labour issues can potentially intersect with:

Section 3

Agreements between competing enterprises that restrict competition, including arrangements concerning wages or recruitment where the statutory conditions are satisfied.

Section 4

Abuse of dominant position, including potentially:

• unfair or discriminatory conditions;
• exclusionary conduct;
• denial of market access;
• leveraging dominance into related markets.

Merger control

Acquisitions involving robotics, AI, labour platforms and automation technologies may require examination where the statutory thresholds and other requirements apply.

39. Key Case-Law Lessons

40. Limitations of the Case Law

It is important to recognize that machine labour is a relatively new competition-law subject.

Most established cases concern:

• traditional labour markets;
• technology platforms;
• infrastructure;
• vertical foreclosure;
• dominance;
• exclusive dealing.

They are therefore analogical authorities, rather than cases directly deciding whether robots or AI themselves possess “market power.”

The legal principles can nevertheless be applied to machine-labour markets where the underlying competitive issue is similar.

41. Future Competition-Law Issues

Future disputes may concern:

1. AI workers

Whether autonomous AI agents compete with human labour.

2. Robot labour markets

Whether access to robotic capacity becomes concentrated.

3. Autonomous employment platforms

Whether algorithms control access to work.

4. AI wage setting

Whether common algorithms facilitate wage coordination.

5. Machine-to-machine bargaining

Whether autonomous systems negotiate prices or labour contracts.

6. Robotics infrastructure

Whether dominant manufacturers restrict access to essential components.

7. AI recruitment

Whether dominant platforms control recruitment data.

8. Human-machine substitution

Whether automation dramatically changes relevant labour markets.

42. Conclusion

Machine labour market power sits at the intersection of competition law, labour economics, automation and digital-platform regulation.

The central distinction is between:

legitimate technological efficiency

and

unlawful use of technological or labour-market power to restrict competition.

Competition law should not prevent firms from adopting robots, AI or automation merely because automation reduces labour requirements.

However, competition concerns can arise where firms:

• coordinate wages;
• agree not to recruit workers;
• restrict worker mobility;
• use algorithms to facilitate anti-competitive coordination;
• control essential automation infrastructure;
• exclude rival employers or platforms;
• impose exclusionary agreements;
• leverage dominance from machine infrastructure into labour or product markets.

The key authorities—Microsoft, Commercial Solvents, Bronner, Intel, Ohio v. American Express, and labour-mobility cases—provide a framework for analysing these issues even though most predate advanced AI and autonomous machine labour.

Quick Revision Formula

Machine Labour Market Power =

Automation + Labour Market + Employer Power + Monopsony + AI + Robotics + Data + Algorithmic Management + Worker Mobility + Switching Costs + Infrastructure Control + Exclusivity + Innovation + Foreclosure + Proportionate Remedies

One-line exam definition

Machine labour market power is the ability arising from control or deployment of AI, robotics, automation or algorithmic labour infrastructure to influence competitive conditions in labour or related product markets, particularly where such power can restrict worker mobility, exclude competing firms, increase entry barriers or reinforce technological dominance.