Competition Law and Competition Implications of Forecasting Infrastructure Ownership

1. Introduction

Forecasting infrastructure ownership refers to the ownership or control of the technological, data, computational, analytical, and institutional infrastructure used to generate forecasts that influence markets and commercial decisions.

Forecasting infrastructure may include:

large datasets;

cloud and computing infrastructure;

artificial-intelligence and machine-learning systems;

demand-forecasting platforms;

pricing and inventory prediction systems;

logistics and supply-chain forecasting tools;

financial forecasting systems;

weather and climate forecasting databases;

market-intelligence platforms;

semiconductor and high-performance computing infrastructure;

APIs and data feeds;

proprietary prediction models; and

platforms that distribute forecasts to businesses and consumers.

The competition concern arises when one undertaking, or a small group of undertakings, controls an essential or strategically important part of this forecasting infrastructure.

Ownership itself is not automatically unlawful. Competition law becomes relevant when control over forecasting infrastructure is used to exclude competitors, restrict access to important inputs, discriminate between downstream users, facilitate coordination, strengthen an existing dominant position, or make market entry unnecessarily difficult.

2. Meaning of Forecasting Infrastructure Ownership

Forecasting infrastructure ownership can be understood as:

The ownership or control of the data, computing capacity, algorithms, platforms, databases, models, interfaces, or other infrastructure necessary to produce, distribute, or commercially exploit forecasts.

For example, imagine that a company owns:

the largest historical demand database;

the computing infrastructure required to process it;

the forecasting algorithm;

the API through which forecasts are supplied; and

a marketplace where businesses use those forecasts.

The company may therefore control not merely a forecasting product but an entire forecasting ecosystem.

This can create competition concerns because competitors may require access to one or more components of that ecosystem to compete effectively.

3. Why Forecasting Infrastructure Has Competition Significance

Forecasts increasingly influence:

prices;

production;

inventory;

transportation;

energy procurement;

financial decisions;

advertising;

staffing;

investment;

insurance;

agricultural decisions;

logistics;

consumer demand;

credit decisions; and

capacity planning.

Consequently, control over forecasting infrastructure can become a source of market power.

The competitive significance increases where:

the infrastructure is difficult to replicate;

the underlying data is unique;

economies of scale are substantial;

network effects exist;

switching costs are high;

interoperability is limited;

the infrastructure is vertically integrated;

access is technically restricted;

competitors depend upon the infrastructure; or

the owner also competes in downstream markets.

4. Forecasting Infrastructure as a Potential Bottleneck

A forecasting infrastructure can operate as a bottleneck.

A bottleneck occurs where competitors cannot reasonably reproduce or replace an important input.

For example:

Historical data → Computing infrastructure → Forecasting model → Forecast → Downstream market

If a dominant undertaking controls the first three stages, competitors may face significant disadvantages.

However, competition law does not normally require every proprietary asset to be shared.

The important questions are:

Is the infrastructure commercially important?

Does the owner possess substantial market power?

Can competitors reasonably reproduce it?

Is access indispensable or merely useful?

Is access technically or contractually restricted?

Does the owner discriminate between users?

Is the infrastructure used to disadvantage downstream competitors?

Does the conduct produce exclusionary effects?

5. Relevant Markets

Forecasting infrastructure can create several related markets.

A. Data Market

There may be a market for:

historical datasets;

real-time data;

consumer data;

market data;

weather data;

logistics data; or

industry-specific datasets.

B. Forecasting Technology Market

This may include:

forecasting software;

AI prediction systems;

analytics platforms;

demand-planning systems; and

predictive APIs.

C. Computing Infrastructure Market

This may include:

cloud computing;

GPUs;

high-performance computing;

storage;

model-training infrastructure.

D. Downstream Market

The forecasting infrastructure may support a separate downstream market such as:

retail;

financial services;

transportation;

energy;

advertising;

insurance;

e-commerce.

Competition authorities may therefore need to examine both the infrastructure market and the downstream market.

6. Data as a Source of Forecasting Power

Forecasting quality often depends heavily on data.

A company possessing a large volume of high-quality historical data may produce forecasts that competitors cannot easily reproduce.

This creates several possible competitive advantages:

better prediction accuracy;

faster prediction;

lower forecasting costs;

improved personalization;

better inventory management;

superior pricing decisions; and

stronger customer retention.

The existence of a large dataset, however, does not automatically establish dominance.

The relevant questions include whether the data is:

unique;

difficult to obtain;

commercially important;

timely;

accurate;

scalable; and

capable of being substituted.

7. Network Effects

Forecasting infrastructure may benefit from network effects.

For example:

More users → more data → better forecasts → more users → more data.

This can produce a reinforcing cycle.

A leading forecasting platform may therefore become increasingly attractive as its user base expands.

Competitors may find it difficult to break into the market because they start with:

less data;

fewer users;

fewer observations;

weaker predictive accuracy; and

fewer commercial relationships.

This can create data-driven entry barriers.

8. Economies of Scale

Forecasting infrastructure frequently involves high fixed costs.

Developing:

computing infrastructure;

databases;

AI models;

data pipelines;

cybersecurity;

APIs; and

technical personnel

can require substantial investment.

Once the infrastructure has been created, however, serving additional customers may be relatively inexpensive.

This creates economies of scale.

A large incumbent may therefore have a cost advantage that smaller competitors cannot easily reproduce.

Economies of scale are not unlawful themselves, but they can become relevant when combined with exclusionary conduct.

9. Vertical Integration

A particularly important competition issue arises when the owner of forecasting infrastructure also operates in a downstream market.

For example:

Company A owns forecasting infrastructure

and simultaneously

Company A competes with users of that infrastructure.

The company may potentially have an incentive to:

deny access;

increase access prices;

degrade access;

delay API access;

provide inferior data;

offer preferential forecasts to its own subsidiary;

use customer data against competitors; or

manipulate interoperability.

This creates a classic vertical foreclosure concern.

10. Self-Preferencing

Suppose a forecasting platform supplies forecasts to competing retailers while operating its own retail business.

It might theoretically provide:

superior forecasts to itself;

faster access to forecasts;

more detailed information;

earlier access to demand signals; or

better API functionality.

Such conduct may create a competitive advantage unrelated to the merits of the downstream business.

Competition authorities would examine whether the conduct amounts to exclusionary self-preferencing or another abuse.

11. Refusal to Supply

A dominant forecasting infrastructure owner may refuse access to competitors.

The legal analysis depends heavily on the circumstances.

A refusal to deal is not automatically unlawful.

Competition law generally distinguishes between:

legitimate proprietary control; and

exclusionary refusal that prevents effective competition.

Relevant factors may include:

indispensability;

feasibility of replication;

availability of substitutes;

prior supply relationships;

objective justification;

effect on competition; and

effect on consumers.

12. Essential-Facility-Type Concerns

Forecasting infrastructure may sometimes be argued to constitute an essential facility.

For example, a market participant may claim that:

Without access to the dominant forecasting database, it cannot compete effectively.

The essential-facilities concept is applied cautiously.

Mere usefulness is normally insufficient.

The infrastructure must generally be exceptionally difficult or impossible to reproduce, and the refusal must create significant competitive harm.

13. Discriminatory Access

A forecasting infrastructure owner may offer access to multiple customers but apply different conditions.

Examples include:

different prices;

different data quality;

different API limits;

different latency;

different update frequency;

different forecasting accuracy;

different contractual terms.

If a dominant undertaking applies discriminatory conditions that disadvantage competitors, competition-law concerns may arise.

14. Excessive Pricing

Another possible issue is excessive pricing.

If a dominant firm controls a uniquely important forecasting infrastructure, it might charge extremely high access fees.

However, excessive-pricing cases are difficult because high prices can also reward innovation and investment.

Authorities therefore generally need to distinguish:

legitimate returns on investment

from

exploitative pricing by a dominant undertaking.

15. Margin Squeeze

A vertically integrated forecasting provider could potentially engage in a margin squeeze.

For example:

it charges competitors a high wholesale price for forecasting infrastructure;

simultaneously it competes with them downstream;

its downstream affiliate receives the input at an economically advantageous internal price.

Competitors may then be unable to compete profitably.

16. Tying and Bundling

Forecasting infrastructure may also be tied to another product.

For example:

Access to the forecasting database is available only if the customer also purchases the company's cloud services.

Or:

Customers purchasing forecasting software must also use the company's proprietary data-storage system.

If the undertaking is dominant and the arrangement forecloses competitors, tying or bundling concerns may arise.

17. Exclusivity

Forecasting infrastructure providers may enter exclusive agreements.

Examples include:

exclusive access to retailer data;

exclusive forecasting contracts;

exclusive cloud arrangements;

exclusive API arrangements;

exclusive distribution of forecasts.

Exclusivity may reduce the amount of data available to rivals and thereby reinforce the incumbent's advantage.

The competitive analysis depends upon factors such as:

duration;

market coverage;

market power;

foreclosure;

availability of alternatives; and

legitimate business justification.

18. Algorithmic Coordination

Forecasting infrastructure can create another important concern: algorithmic coordination.

Suppose several competitors use the same forecasting or pricing infrastructure.

The infrastructure may generate:

demand forecasts;

price recommendations;

inventory recommendations; or

capacity recommendations.

If competitors independently use the same system, prices or strategies could become more similar.

The mere use of similar software does not automatically constitute a cartel.

However, competition law becomes more concerned where there is:

communication of competitively sensitive information;

agreement to follow recommendations;

coordination through a common intermediary; or

conscious alignment of competitive strategies.

19. Common Ownership and Information Exchange

Ownership structures can also matter.

If the same company controls forecasting infrastructure used by several competitors, it may have access to sensitive information such as:

future demand;

expected prices;

inventory;

production levels;

capacity;

customer behavior.

This can raise concerns about information exchange and competitive sensitivity.

The key issue is whether the infrastructure allows competitively sensitive information to flow between independent competitors in a way that reduces competition.

20. Merger and Acquisition Concerns

Forecasting infrastructure ownership has important merger implications.

A large technology company may acquire:

a forecasting startup;

a specialized data provider;

a cloud platform;

an AI model developer;

a market-data company.

The acquisition may combine:

data + computing + forecasting + distribution.

Competition authorities may investigate whether the transaction:

eliminates an emerging competitor;

combines complementary sources of market power;

increases entry barriers;

forecloses rivals;

enables data accumulation;

creates interoperability problems; or

strengthens vertical integration.

21. Killer Acquisitions

A dominant forecasting platform may acquire a small forecasting startup before the startup becomes a significant competitor.

The transaction may be commercially small in revenue terms but strategically significant.

Competition authorities may therefore examine:

innovation potential;

future competitive significance;

proprietary technology;

unique datasets;

talent;

patents;

customer relationships.

22. Innovation Competition

Competition in forecasting infrastructure is not limited to price.

Competition may occur through:

accuracy;

speed;

transparency;

privacy;

explainability;

model quality;

interoperability;

reliability;

customization.

A dominant infrastructure owner could potentially reduce innovation by making it difficult for alternative forecasting systems to access important inputs.

23. Consumer Welfare

Forecasting infrastructure affects consumers indirectly.

Better competition may result in:

lower prices;

better product availability;

improved logistics;

more accurate services;

greater innovation.

Conversely, exclusionary control may produce:

higher prices;

reduced choice;

weaker innovation;

lower service quality;

reduced privacy;

increased dependency on a single provider.

24. Major Case Laws

1. United States v. Microsoft Corp. (2001)

Facts

Microsoft possessed a dominant position in PC operating systems and engaged in various practices involving browser distribution and relationships with computer manufacturers and software developers.

Competition principle

The case demonstrated how a dominant technology platform can use control over an important technological ecosystem to protect or extend its market position.

Relevance to forecasting infrastructure

Forecasting infrastructure can similarly become strategically important where its owner controls an upstream technological layer and uses that control to disadvantage competing products.

The case is particularly relevant to:

platform power;

technological leverage;

exclusionary conduct;

interoperability;

tying-related concerns.

2. Bronner v. Mediaprint (1998)

Facts

The case concerned access to a newspaper home-delivery system.

The claimant argued that access to the infrastructure was necessary to compete.

Principle

The European Court of Justice applied a demanding test to refusal-to-supply claims, emphasizing the importance of indispensability and the absence of realistic alternatives.

Relevance

The case provides an important framework for determining whether forecasting infrastructure can legitimately be treated as an indispensable facility.

A forecasting database being commercially valuable is not necessarily enough.

3. United Brands v. Commission (1978)

Facts

United Brands was found to hold a dominant position in the relevant banana market and was examined for abusive conduct.

Principle

The case remains an important authority on:

dominance;

relevant market;

abuse of dominant position; and

the special responsibilities of dominant firms.

Relevance

An owner of forecasting infrastructure that achieves substantial market power may acquire special responsibilities not to use that position to eliminate competition.

4. Intel v. Commission

Facts

Intel was investigated for practices involving rebates to customers and efforts to preserve its position in the market for x86 central processing units.

Principle

The litigation developed important principles concerning exclusionary rebates by dominant firms and the need to examine their actual or potential effects.

Relevance

Forecasting infrastructure providers may similarly use discounts or commercial incentives to lock customers into their infrastructure.

For example:

“Use our forecasting platform exclusively and receive substantially discounted access.”

The competitive effect would need to be examined rather than assuming that every discount is unlawful.

5. Google Shopping

Facts

The European Commission examined Google's treatment of competing comparison-shopping services within its search results.

Principle

The case concerned the use of dominance in one digital layer to favor the undertaking's own service in another layer.

Relevance

The analogy is significant for forecasting infrastructure.

A forecasting platform that also operates downstream could potentially favor its own downstream service through:

preferential data access;

better forecasting outputs;

priority APIs;

superior model functionality.

The legal question would be whether such conduct constitutes exclusionary abuse.

6. Google Android

Facts

The European Commission examined contractual practices concerning Google's Android ecosystem, including arrangements involving search, browsers, and application distribution.

Principle

The case illustrates how control over one technological ecosystem can be leveraged through contractual arrangements involving complementary products.

Relevance

Forecasting infrastructure could similarly be bundled with:

cloud services;

AI models;

data storage;

analytics;

software;

APIs.

The case therefore provides useful principles for analysing tying, bundling and ecosystem leverage.

7. Qualcomm

Competition principle

Competition authorities and courts in different jurisdictions have examined Qualcomm's conduct involving chipsets, licensing arrangements and commercial relationships.

Relevance

The Qualcomm litigation demonstrates the importance of analysing technology markets where a firm possesses control over an important upstream technological input.

Forecasting infrastructure may present similar issues where competitors depend on a proprietary technological layer.

8. Amazon Marketplace Investigations

Competition authorities have examined Amazon's use of information generated through its marketplace relationships with sellers.

Competition concern

A platform may obtain information from businesses that use its infrastructure and subsequently compete with those businesses.

Relevance to forecasting infrastructure

The same issue can arise when the infrastructure owner receives:

demand forecasts;

inventory information;

sales projections;

customer trends;

future production information.

The owner could potentially gain a competitive advantage from information generated by infrastructure users.

25. Indian Competition Law Framework

In India, forecasting infrastructure ownership is not an independent statutory offence.

The principal framework is the Competition Act, 2002.

Several provisions may become relevant.

Section 3 – Anti-Competitive Agreements

Section 3 can become relevant where forecasting infrastructure is used to facilitate:

price fixing;

market allocation;

bid coordination;

information exchange;

restrictive agreements;

exclusionary arrangements.

For example, competitors using a common forecasting platform could raise concerns if the arrangement facilitates coordinated conduct.

26. Section 4 – Abuse of Dominant Position

Section 4 is especially relevant where a forecasting infrastructure owner is dominant.

Potential forms of abuse may include:

A. Unfair or discriminatory conditions

A dominant infrastructure provider may impose discriminatory terms.

B. Unfair or discriminatory pricing

Different competitors may be charged different prices without objective justification.

C. Limiting technical development

The dominant provider may deliberately restrict interoperability or innovation.

D. Denial of market access

The infrastructure may be used to prevent competitors from entering or operating in a downstream market.

E. Leveraging dominance

The owner may use dominance in forecasting infrastructure to strengthen its position in another market.

27. Section 5 – Combinations

Where ownership changes occur through:

mergers;

acquisitions;

amalgamations;

Section 5 and the combination-control framework may become relevant.

The CCI may consider whether the transaction:

strengthens data concentration;

increases entry barriers;

eliminates a potential competitor;

enables foreclosure;

combines upstream and downstream power.

28. Section 6 – Regulation of Combinations

A transaction involving forecasting infrastructure may require competition scrutiny where it is capable of causing an appreciable adverse effect on competition.

Important considerations may include:

market concentration;

countervailing buyer power;

entry barriers;

substitutes;

innovation;

vertical integration;

data concentration;

network effects.

29. CCI v. SAIL

Competition Commission of India v. Steel Authority of India Ltd.

Importance

The Supreme Court's decision is an important Indian competition-law authority concerning the functioning and jurisdiction of the CCI.

Relevance

For forecasting infrastructure, it reinforces the broader principle that competition-law analysis must be conducted within the statutory framework rather than treating every commercial dispute as a competition violation.

30. CCI v. Bharti Airtel Ltd.

Importance

The Supreme Court considered the relationship between sectoral regulation and competition law.

Relevance

Forecasting infrastructure may operate in regulated industries such as:

telecommunications;

banking;

energy;

financial markets;

aviation.

Competition questions may therefore overlap with sector-specific regulation.

31. Excel Crop Care Ltd. v. CCI

Importance

The Supreme Court examined competition-law principles in the context of cartelisation and penalty assessment.

Relevance

Although not a forecasting-infrastructure case, it is useful when analysing whether common forecasting systems are being used to facilitate coordinated conduct among competitors.

The important distinction is between:

independent use of a common technology

and

technology-assisted coordination pursuant to an agreement or concerted practice.

32. Competition Concerns Across the Forecasting Value Chain

33. Legitimate Business Reasons

Ownership of forecasting infrastructure can generate genuine efficiencies.

A company may legitimately:

invest heavily in infrastructure;

protect proprietary technology;

charge for access;

limit access for cybersecurity reasons;

protect confidential data;

restrict misuse;

maintain quality standards;

develop proprietary forecasting models.

Competition law should therefore distinguish legitimate commercial conduct from exclusionary conduct.

34. Key Test for Competition Authorities

A useful analytical sequence is:

Step 1 – Identify the infrastructure

What exactly is controlled?

Step 2 – Define the relevant market

Is the market:

data;

forecasting software;

computing;

APIs;

downstream services?

Step 3 – Determine market power

Does the owner possess substantial market power?

Step 4 – Examine entry barriers

Can competitors reproduce the infrastructure?

Step 5 – Examine conduct

Has the owner:

refused access;

discriminated;

tied products;

bundled services;

imposed exclusivity;

degraded interoperability;

self-preferenced?

Step 6 – Examine effects

Does the conduct substantially reduce competition?

Step 7 – Examine efficiencies

Are there legitimate technical or commercial justifications?

35. Important Distinction: Ownership vs Abuse

The most important principle is:

Ownership of forecasting infrastructure is not itself an antitrust violation.

A company can lawfully own:

databases;

AI systems;

cloud infrastructure;

forecasting software;

computing facilities.

The competition issue arises when market power and exclusionary conduct combine to harm the competitive process.

36. Forecasting Infrastructure and Digital Monopolisation

Forecasting infrastructure can contribute to a broader process of digital concentration.

For example:

More users

↓

More data

↓

Better forecasts

↓

More customers

↓

More data

↓

Higher entry barriers

This feedback loop can produce durable market power.

Competition authorities therefore increasingly need to consider not only current market shares but also:

data advantages;

ecosystem effects;

switching costs;

innovation;

interoperability;

potential competition.

37. Possible Remedies

Where competition problems are established, possible remedies may include:

Structural remedies

divestiture;

separation of business units.

Behavioural remedies

non-discriminatory access;

interoperability;

data portability;

transparent access terms;

restrictions on self-preferencing;

restrictions on exclusivity.

Data-related remedies

controlled data access;

privacy-preserving data sharing;

information firewalls.

Merger remedies

divestiture of overlapping assets;

licensing;

access commitments;

interoperability commitments.

38. Challenges for Competition Authorities

Forecasting infrastructure creates several difficult enforcement questions.

A. Accuracy

How should authorities determine whether one forecasting system is genuinely superior?

B. Data Replicability

Can competitors realistically recreate the database?

C. Dynamic Competition

A small company today may become a major competitor tomorrow.

D. Algorithmic Complexity

It may be difficult to determine how a forecasting model produces its results.

E. Rapid Technological Change

Infrastructure that is indispensable today may become replaceable tomorrow.

39. Six Core Competition Risks

For examination purposes, remember these six major risks:

Data concentration

Infrastructure bottlenecks

Vertical foreclosure

Self-preferencing

Exclusivity and switching costs

Algorithmic coordination

Other important risks include tying, bundling, discriminatory access, refusal to supply and anti-competitive acquisitions.

40. Short Case-Law Revision Table

41. Conclusion

Forecasting infrastructure ownership is not, by itself, anti-competitive. It can encourage investment, innovation, efficiency and better prediction.

However, competition concerns can arise when a firm controls an important forecasting infrastructure and uses that control to:

exclude competitors;

deny indispensable access;

discriminate between users;

impose exclusivity;

tie or bundle services;

self-preference downstream businesses;

exploit commercially sensitive information;

facilitate coordination;

foreclose innovation; or

strengthen an already dominant position.

The central competition-law question is therefore not simply:

“Who owns the forecasting infrastructure?”

It is:

“Does control over that infrastructure give the owner market power that is being used in a manner that restricts effective competition?”

Under Indian competition law, the principal analytical provisions are Sections 3 and 4 of the Competition Act, 2002, together with the merger-control provisions in Sections 5 and 6 where ownership changes occur.

The most important conceptual distinction for examination purposes is:

Infrastructure ownership → may create market power → market power + exclusionary conduct → potential competition-law problem.

Thus, forecasting infrastructure should be analysed as a potential data bottleneck, technological bottleneck, platform asset and source of vertical leverage, rather than as an automatically prohibited form of ownership.