Competition Law and Simulation-Based Market Power Analysis

Introduction

Simulation-based market power analysis refers to the use of economic models, computational simulations, and quantitative techniques to estimate how firms with different market positions are likely to behave under alternative competitive conditions. Competition authorities and courts can use such analysis to assess market definition, unilateral effects, coordinated effects, foreclosure, pricing power, merger effects, and competitive harm.

Simulation is particularly valuable where traditional indicators—such as market shares, concentration ratios, or the Herfindahl-Hirschman Index (HHI)—do not adequately capture the competitive dynamics of the market.

For example, a merger between two firms with relatively modest market shares may nevertheless create substantial unilateral market power if their products are close substitutes. A merger simulation can estimate the likely post-merger price increase by modelling consumer substitution and firms' pricing incentives.

1. Meaning of Simulation-Based Market Power Analysis

A competition-law simulation generally constructs a model of the relevant market and estimates what would happen under different competitive scenarios.

A simplified framework is:

Observed market data → Demand estimation → Competitive model → Counterfactual simulation → Predicted market outcome → Competition-law assessment

The simulation may compare:

• pre-merger and post-merger prices;
• output levels;
• market shares;
• margins;
• diversion ratios;
• consumer welfare;
• innovation incentives;
• entry or exit;
• foreclosure effects; and
• efficiencies.

The objective is not simply to calculate concentration but to estimate how firms would behave because of the competitive structure.

2. Legal Relevance

Simulation can be relevant to several major areas of competition law.

A. Merger control

Authorities can simulate whether a proposed merger is likely to:

• increase prices;
• reduce output;
• reduce quality;
• reduce innovation;
• eliminate an important competitive constraint; or
• facilitate coordinated conduct.

B. Abuse of dominance

Simulation may assist in examining:

• predatory pricing;
• excessive pricing;
• loyalty rebates;
• tying;
• bundling;
• refusal to supply;
• discriminatory access; and
• foreclosure.

C. Cartels and coordinated conduct

Models can examine whether market characteristics make coordination more sustainable.

Relevant variables can include:

• number of competitors;
• transparency;
• frequency of interaction;
• demand stability;
• switching costs;
• capacity constraints; and
• punishment mechanisms.

D. Market definition

Simulation can complement traditional tools such as the SSNIP test by estimating consumer substitution and diversion more directly.

3. Principal Types of Competition Simulations

3.1 Bertrand Price Competition

The Bertrand model assumes that firms compete primarily through prices.

A simplified profit function is:

πi=(Pi−MCi)Qi(P)\pi_i=(P_i-MC_i)Q_i(P)

where:

• PiP_i = price charged by firm ii;
• MCiMC_i = marginal cost;
• QiQ_i = quantity demanded.

A merger simulation can estimate the new equilibrium after two firms become a single entity.

3.2 Cournot Competition

The Cournot model assumes that firms compete through quantities rather than prices.

The simulation can examine whether a reduction in the number of independent firms produces:

• lower output;
• higher prices; or
• greater market power.

It can be useful in industries involving substantial production or capacity decisions.

3.3 Differentiated-Product Demand Models

Modern merger analysis frequently deals with differentiated products.

Consumers may regard:

• Product A and Product B as close substitutes;
• Product A and Product C as moderate substitutes; and
• Product A and Product D as distant substitutes.

Simulation therefore incorporates cross-price elasticities and diversion ratios.

4. Diversion Ratio and Market Power

The diversion ratio measures the proportion of customers lost by one product who switch to another product.

For example:

• Firm A loses 10% of its customers following a price increase.
• 40% of those customers switch to Firm B.
• 20% switch to Firm C.
• the remainder leave the market or switch elsewhere.

If Firm A and Firm B merge, the merged firm internalises the competitive effect between the two products.

This is particularly important for unilateral-effects analysis.

5. Upward Pricing Pressure

Simulation is closely related to the concept of Upward Pricing Pressure (UPP).

A simplified formulation is:

UPPA≈DAB(PB−MCB)−EAUPP_A \approx D_{AB}(P_B-MC_B)-E_A

where:

• DABD_{AB} = diversion from A to B;
• PB−MCBP_B-MC_B = margin on B;
• EAE_A = merger efficiencies affecting A.

A high diversion ratio combined with a high margin may indicate significant post-merger pricing incentives.

UPP is therefore more economically informative than simply asking whether the merging firms possess large market shares.

6. Simulation and the SSNIP Test

The SSNIP test asks whether a hypothetical monopolist could profitably impose a small but significant and non-transitory increase in price.

Traditionally, the test can be represented as:

Price increase → Customer substitution → Lost sales → Profitability

Simulation can improve this analysis by estimating the actual demand response.

It can therefore help identify:

• the relevant product market;
• geographic boundaries;
• substitution patterns; and
• competitive constraints.

However, simulation does not eliminate the need for legal judgment concerning market definition.

7. Data Required for Simulation

A reliable simulation normally requires some combination of:

1. transaction-level sales data;
2. prices;
3. quantities;
4. product characteristics;
5. customer switching information;
6. costs;
7. margins;
8. geographic information;
9. capacity information;
10. promotional data;
11. historical pricing; and
12. entry and exit information.

The quality of the conclusion depends heavily on the quality of the underlying data.

8. Simulation in Digital Markets

Simulation becomes particularly complex in digital markets.

Relevant factors include:

• zero-price services;
• multi-sided platforms;
• network effects;
• switching costs;
• interoperability;
• data advantages;
• algorithmic pricing;
• recommendation systems;
• self-preferencing;
• platform access; and
• economies of scale.

A conventional price-based model may be inadequate where consumers pay no monetary price.

The model may instead examine:

• quality;
• advertising exposure;
• privacy;
• search results;
• transaction costs;
• innovation;
• user engagement; and
• platform fees charged to another side of the market.

9. Simulation and Network Effects

Suppose platform A has:

• 60% of users;
• more sellers;
• more data;
• stronger network effects; and
• lower average transaction costs.

A simulation can examine whether those characteristics create a self-reinforcing competitive advantage.

The relevant question is not merely whether A has a large market share but whether:

additional users increase the platform's attractiveness sufficiently to make entry or expansion by competitors substantially more difficult.

10. Simulation-Based Analysis of Entry

Competition authorities may also simulate entry.

The model may examine:

• fixed entry costs;
• sunk costs;
• switching costs;
• economies of scale;
• network effects;
• capacity requirements;
• access to infrastructure; and
• expected profitability.

A theoretical entrant may appear profitable under a static model but unprofitable when realistic entry costs and customer-acquisition requirements are incorporated.

11. Simulation and Coordinated Effects

Simulation can also be used to examine whether a merger makes coordination easier.

Relevant characteristics include:

• market concentration;
• product homogeneity;
• price transparency;
• repeated interaction;
• capacity constraints;
• demand predictability;
• symmetry between competitors; and
• ability to detect deviations.

Simulation can compare outcomes under:

competitive equilibrium → coordinated equilibrium → post-merger equilibrium

The results must nevertheless be interpreted cautiously because coordination involves behavioural assumptions that may be difficult to establish empirically.

12. Simulation and Efficiencies

A merger may produce efficiencies such as:

• economies of scale;
• lower marginal costs;
• elimination of duplicated infrastructure;
• improved logistics;
• technological integration; or
• better production capacity.

Simulation can incorporate those efficiencies into the counterfactual.

For example:

Merger without efficiencies → predicted price increase

versus

Merger with verified cost reduction → smaller price effect

This allows authorities to examine whether claimed efficiencies could offset competitive harm.

13. Counterfactual Analysis

One of the most important aspects of simulation is construction of the counterfactual.

The authority may compare:

Scenario 1 — Existing competitive structure

A and B remain independent.

Scenario 2 — Proposed transaction

A and B operate as one firm.

Scenario 3 — Alternative counterfactual

A would have exited, expanded, or been acquired by another firm.

The legal significance of the predicted effect depends upon which counterfactual is legally and economically appropriate.

14. Limitations of Simulation

Simulation is not automatically conclusive.

14.1 Model specification risk

Different assumptions can produce different results.

14.2 Data limitations

Incomplete or inaccurate data can materially distort predictions.

14.3 Behavioural assumptions

The model may assume firms behave according to a particular competitive model that does not perfectly describe reality.

14.4 Dynamic competition

Static models may underestimate:

• innovation;
• entry;
• technological change;
• product repositioning; and
• changing consumer preferences.

14.5 Legal uncertainty

A numerical prediction does not itself establish that conduct violates competition law.

The authority must connect the economic evidence to the applicable statutory test.

15. Major Case Laws

1. United States v. Philadelphia National Bank, 374 U.S. 321 (1963)

The U.S. Supreme Court treated market concentration as an important indicator in merger analysis.

Although the decision predates sophisticated merger simulation, it established the importance of structural evidence in evaluating competitive effects.

Relevance

It provides an important foundation for understanding why quantitative measures of concentration can matter in merger cases.

Modern simulation analysis goes beyond this structural approach by attempting to estimate the economic consequences of a transaction.

2. United States v. General Dynamics Corp., 415 U.S. 486 (1974)

The Supreme Court recognised that historical market shares may sometimes provide an incomplete picture of competitive conditions.

The case involved evidence concerning the competitive significance of firms' future productive capacity.

Relevance to simulation

The decision illustrates an important principle underlying modern quantitative analysis:

current market shares may not adequately predict future competitive conditions.

Simulation models similarly attempt to incorporate economically relevant variables beyond simple market shares.

3. United States v. Baker Hughes Inc., 908 F.2d 981 (D.C. Cir. 1990)

The case concerned the evidentiary significance of market concentration in merger litigation.

The court emphasised that concentration statistics do not necessarily end the competitive inquiry.

Relevance

Simulation can provide additional evidence concerning whether concentration translates into actual competitive effects.

4. FTC v. Staples, Inc., 970 F. Supp. 1066 (D.D.C. 1997)

The proposed Staples–Office Depot merger became an important U.S. merger case concerning differentiated retail markets.

The government relied heavily upon evidence concerning competitive interaction between the merging firms.

Relevance

The case demonstrates why closeness of competition can be more important than simply looking at overall market shares.

Modern merger simulations formalise this concept through:

• diversion ratios;
• elasticities;
• margins; and
• predicted post-merger price effects.

5. FTC v. H.J. Heinz Co., 246 F.3d 708 (D.C. Cir. 2001)

The case concerned the proposed Heinz–Beech-Nut merger in the U.S. baby-food market.

The court considered market concentration and competitive effects in assessing the transaction.

Relevance

The case illustrates the importance of examining the competitive relationship between firms rather than treating market share as the sole indicator of market power.

6. FTC v. Whole Foods Market, Inc., 548 F.3d 1028 (D.C. Cir. 2008)

The case concerned Whole Foods' proposed acquisition of Wild Oats.

The dispute included issues concerning the appropriate market definition and the competitive relationship between the parties.

Relevance

It demonstrates the importance of determining whether the merging firms are particularly close competitors.

That question is central to simulation models using diversion ratios and differentiated-product demand.

7. FTC v. H&R Block, Inc., 833 F. Supp. 2d 36 (D.D.C. 2011)

The proposed H&R Block–TaxACT transaction involved digital tax-preparation services.

The court considered quantitative economic evidence concerning competition and the effects of the proposed merger.

Relevance

The case is particularly useful for studying how econometric evidence, market structure, and competitive effects can be combined in merger analysis.

8. FTC v. Sysco Corp., 113 F. Supp. 3d 1 (D.D.C. 2015)

The court rejected Sysco's proposed acquisition of US Foods.

The case involved extensive economic evidence concerning market definition and competitive effects in food-service distribution.

Relevance

It illustrates how quantitative evidence can be used to evaluate whether a merger would eliminate important competitive constraints.

Simulation-based analysis can build upon such evidence by estimating the magnitude of predicted effects.

16. European Competition-Law Context

The European Commission increasingly relies upon sophisticated economic evidence in merger and antitrust analysis.

Simulation can be relevant to:

• unilateral effects;
• coordinated effects;
• vertical foreclosure;
• portfolio effects;
• differentiated-product mergers;
• efficiencies; and
• innovation competition.

Important European cases illustrating the development of economic-effects analysis include:

Airtours v Commission

The case established important principles concerning coordinated effects and collective dominance.

Tetra Laval v Commission

The European courts required sufficiently convincing evidence when relying on complex economic theories of harm.

Impala v Commission

The case concerned the evidentiary assessment of coordinated effects in a concentrated market.

Intel v Commission

The litigation highlighted the importance of examining economic evidence concerning exclusionary effects, particularly in relation to rebates.

CK Telecoms UK Investments v Commission

The General Court's judgment concerning the Illumina/GRAIL transaction addressed the assessment of potential competitive harm and the evidentiary standard applicable to merger analysis.

These cases demonstrate why quantitative modelling must be connected to a clearly established theory of competitive harm.

17. Simulation Versus Traditional Market-Share Analysis

The two approaches are complementary rather than mutually exclusive.

18. Evidentiary Value

Simulation evidence should normally be assessed together with:

• documentary evidence;
• internal business documents;
• customer evidence;
• competitor evidence;
• pricing data;
• market shares;
• entry evidence;
• switching data;
• efficiencies; and
• industry characteristics.

An economic model should not become a substitute for establishing the underlying facts.

19. Competition-Law Issues Raised by Simulation

Simulation-based market-power analysis creates several important legal questions:

1. Transparency

Can opposing parties understand the assumptions and methodology?

2. Reproducibility

Can another economist reproduce the results?

3. Data integrity

Were the underlying datasets complete and reliable?

4. Model sensitivity

Do minor changes in assumptions substantially alter the outcome?

5. Counterfactual validity

Is the comparison scenario economically realistic?

6. Causation

Does the model establish that the challenged conduct caused the competitive effect?

7. Legal relevance

Does the predicted economic effect satisfy the statutory test?

20. Application to Emerging Markets

Simulation-based analysis is increasingly relevant to:

• artificial intelligence;
• cloud computing;
• app stores;
• digital advertising;
• online marketplaces;
• payment platforms;
• EV charging;
• battery-swapping networks;
• telecommunications;
• energy markets;
• logistics platforms;
• pharmaceutical markets;
• financial technology; and
• algorithmic pricing.

For example, an AI platform may have a competitive advantage because of:

more users → more data → better model → better service → more users

A simulation could examine whether that feedback loop materially increases market power or whether competitors can overcome the advantage through entry, interoperability, or access to alternative data.

21. Competition-Law Framework

A useful analytical framework is:

Step 1 — Define the relevant market

↓

Step 2 — Identify the competitive constraints

↓

Step 3 — Collect price, quantity, cost and substitution data

↓

Step 4 — Estimate demand and diversion

↓

Step 5 — Select an appropriate competitive model

↓

Step 6 — Construct the counterfactual

↓

Step 7 — Simulate the challenged conduct or transaction

↓

Step 8 — Test alternative assumptions

↓

Step 9 — Incorporate efficiencies and entry

↓

Step 10 — Compare predicted competitive outcomes

↓

Step 11 — Integrate quantitative evidence with legal evidence

↓

Step 12 — Apply the statutory competition-law test

Conclusion

Simulation-based market power analysis represents a movement from purely structural competition analysis toward economically informed effects analysis. Instead of asking only whether a firm has a large market share, simulation can examine how consumers substitute between products, how firms respond to competitive incentives, and how a merger or business practice may alter prices, output, quality, innovation, or access.

Its greatest value lies in merger analysis, differentiated-product markets, digital platforms, network industries, and other markets where simple concentration measures may provide an incomplete picture.