Market Definition and the Hypothetical Monopolist Test
A market is not a fact waiting to be discovered. It is a construct built for a particular purpose, and the purpose is to measure whether a firm or a merger could raise price without losing enough sales to make the attempt unprofitable.
The rule in short
Both the merger provision and the monopolization provision require a relevant market with a product dimension and a geographic dimension. Product markets are drawn by reasonable interchangeability of use and cross-elasticity of demand, tested by asking whether a hypothetical sole seller of a candidate group could profitably impose a small but significant non-transitory price increase. If it could not, the next-best substitute is added and the question is repeated.
Ask a merger lawyer which fact decides a case and the answer is usually the market. Share figures follow from boundaries, concentration measures follow from shares, and presumptions of harm follow from concentration. Draw the market one way and a transaction is unremarkable; draw it a notch narrower and the same transaction produces a share that shifts the burden onto the parties.
What the statutes ask for
The merger provision prohibits acquisitions whose effect may be substantially to lessen competition or tend to create a monopoly in any line of commerce or in any activity affecting commerce in any section of the country. Those two phrases are the product dimension and the geographic dimension, and courts have read them as requiring a market to be identified before effects can be assessed. The monopolization provision contains no comparable language, but power has to be measured against something, so the same requirement was read into it.
Outside those provisions, the requirement follows the mode of analysis rather than the statute. A restraint condemned per se needs no market at all, which is one reason characterizing a restraint is the first fight in most cases. Under the rule of reason a market is ordinarily required, because market power is the indirect route by which a plaintiff shows that a restraint could harm competition.
Substitutes, uses and the classic indicia
The product market includes commodities reasonably interchangeable by consumers for the same purposes, and the boundary is set by cross-elasticity of demand: how much sales of one product fall when the price of another rises. Around that core the Supreme Court identified a set of practical indicia for recognizing a narrower market within a broad one — industry or public recognition of a separate market, peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors.
Those indicia remain in the case law and continue to be argued, but they are descriptive rather than analytical. They tell a court what a separate market looks like without telling it how to decide a close case, and they can be assembled to support almost any boundary a party prefers. That gap is what the hypothetical monopolist test was designed to fill.
A test that asks a single question
The test posits a firm that is the only present and future seller of a candidate group of products in a candidate area, and asks whether that firm would find it profitable to impose at least a small but significant and non-transitory increase in price on at least one product in the group. The benchmark increase is conventionally five percent of the prevailing price, though the agencies apply it flexibly and use other terms of trade where price is not the competitive variable.
If the answer is no — because too many customers would switch to something outside the group, making the increase unprofitable — the candidate market is too narrow. The next-best substitute is added and the question is asked again. The process stops at the smallest group of products for which the answer is yes, which is why the test tends to produce narrow markets and why parties resisting a narrow market must show that the diverted sales would be large enough to defeat the increase.
Two quantitative tools implement the test in practice. The diversion ratio measures the share of sales lost by one product that would go to another, and it is estimated from win-loss records, switching data or customer surveys. Critical loss analysis compares the sales loss that would make a price increase unprofitable with the loss actually predicted. Both are contested in every case, and both depend on data produced during the second request stage of a merger investigation.
Courts sometimes treat a favorable market as a verdict. It is not. A narrow market establishes a high share, and a high share supports a presumption or an inference, but the parties may still show that entry is easy, that the shares overstate competitive significance, or that the merger produces verifiable efficiencies. A plaintiff who wins the market and does nothing else has won the first half of the case.
Where competition happens
The geographic dimension is drawn the same way. The question is where customers would turn if prices in one area rose, which depends on transport costs, perishability, regulation, licensing, and the willingness of buyers to travel. Markets may be local for a hospital or a ready-mix concrete plant, regional for a distribution business, and worldwide for a component sold by a handful of manufacturers.
Where sellers can identify individual buyers and charge them different prices, the agencies define markets around targeted customers rather than around geography, because a group of customers that cannot switch is a market even if it is scattered. That approach connects the exercise to the pricing questions taken up under the statute governing discrimination between purchasers, where the ability to sort customers is the premise rather than the conclusion.
Why prevailing price is the wrong benchmark against a monopolist
The test as stated uses prevailing prices. In a monopolization case that is a trap. A firm already charging a monopoly price has pushed price up to the point where customers are on the verge of switching to inferior substitutes, so observed substitution at that price makes the market look broad and the firm look constrained. The Supreme Court fell into this reasoning in a case about cellophane, and the error has carried its name ever since.
The correction is to run the test from the competitive price rather than the observed one, which requires estimating a price that does not exist in the record. Courts handle it inconsistently, and some avoid it by relying on direct evidence instead. The problem does not arise in merger cases, where the premise is that prices before the deal are competitive and the question is what the transaction would change.
The three routes, side by side
| Route | What it asks | Where it is used | Principal weakness |
|---|---|---|---|
| Practical indicia | Does the industry treat this as a separate market | Pleading and early motions | Descriptive; supports almost any boundary |
| Hypothetical monopolist test | Could a sole seller profitably raise price a small amount | Merger review and expert testimony | Data-hungry; sensitive to the diversion estimate |
| Direct evidence of effect | Did output fall or price rise because of the conduct | Conduct cases with a track record | Requires a period of actual conduct to measure |
| Targeted customer analysis | Can this group of buyers be isolated and charged more | Price discrimination and localized harm | Depends on the seller's ability to sort buyers |
The third route matters more than its rarity suggests. Where a plaintiff can show actual anticompetitive effects — reduced output, higher price, degraded quality traceable to the restraint — courts have accepted that proof in place of an elaborate market definition, on the reasoning that the market is only a proxy for the effect. That principle does the work in many exclusionary conduct cases and in disputes over exclusive arrangements, where the foreclosed share matters more than the outer boundary of the market.
The exercise is not unique to antitrust. Trade remedy proceedings run a comparable inquiry when they identify the domestic like product against which injury is measured, and scope rulings in those cases turn on the same kind of argument about where one product ends and another begins. In both settings the boundary is chosen for a purpose, and the party who frames the purpose usually frames the boundary.
Points to carry away
- The merger statute speaks of a line of commerce in a section of the country, which is the product and geographic dimension.
- Reasonable interchangeability of use and cross-elasticity of demand set the boundaries of a product market.
- The hypothetical monopolist test asks whether a sole seller could profitably raise price by a small but significant amount.
- Candidate markets are expanded until the test is satisfied, which produces the smallest market that works.
- Prevailing price is the wrong benchmark where the defendant already charges a monopoly price.
- Direct proof of anticompetitive effect can make an elaborate market definition unnecessary.
Questions readers ask
Can a single brand ever be a relevant market?
Yes, though it is unusual and the circumstances are narrow. The clearest recognized instance is an aftermarket for parts and service tied to durable equipment already purchased, where customers who chose the equipment on competitive terms are afterwards locked in by switching costs and cannot discipline the manufacturer's aftermarket pricing. Courts examining these claims look for information asymmetry at the time of the original purchase and for genuine lock-in rather than mere inconvenience, and they reject single-brand markets where buyers could have anticipated the lifecycle cost.
How are two-sided platforms treated?
The Supreme Court has held that a transaction platform serving two interdependent groups simultaneously — merchants and cardholders in the case it decided — should be analyzed as a single market rather than as two separate ones, because the product is the transaction itself and neither side can be served without the other. The consequence is that a plaintiff must show a net anticompetitive effect across both sides, not merely a price increase on one. Lower courts have confined the holding to platforms where the transaction is genuinely simultaneous.
Does the plaintiff have to propose only one market?
No. Plaintiffs commonly plead alternative markets, some narrow and some broad, and ask the court to accept whichever the evidence supports. The risk is not in pleading alternatives but in pleading a market that excludes obvious substitutes without explaining why, which invites dismissal for failure to plead a plausible market. Defendants respond by proposing a broader market in which their share is unremarkable, so the two proposals frame the case long before any expert testimony is heard.
Sources
- Cornell Legal Information Institute — 15 U.S.C. 18, Acquisition of Stock or AssetsThe line-of-commerce and section-of-the-country language that requires a defined market.
- Cornell Legal Information Institute — 15 U.S.C. 2, Monopolizing Trade a FelonyThe monopolization provision, whose power element is measured within a relevant market.
- Department of Justice and Federal Trade Commission — Merger GuidelinesThe agencies' statement of the hypothetical monopolist test and its application.
- Federal Trade Commission — MergersHow the agency describes product and geographic market analysis to non-specialists.
- Federal Trade Commission — Competitive EffectsConcentration measurement and the unilateral and coordinated theories markets are drawn to test.
- Federal Trade Commission — Monopolization DefinedThe agency's account of monopoly power and the share figures courts have treated as significant.
Justice Partners Journal is a publication, not a law firm. This article states general rules and cites its sources; it is not advice about any particular case, and the law differs by state and changes over time.
