Summary. Technology companies operate in markets with network effects, near-zero marginal costs, and winner-take-most dynamics, which produce high concentration through ordinary competitive success and also create powerful incentives to keep competitors out. Antitrust law tries to distinguish between the two, and it has been unusually active in the technology sector for the first time since the Microsoft case. This article explains the framework a technology lawyer needs: the elements of monopolization and attempted monopolization, why market definition decides most cases and why it is unusually hard for platforms after Ohio v. American Express, the conduct theories that recur (exclusive dealing, tying, self-preferencing, refusals to deal, predatory pricing), and the rule of reason under § 1. It covers the major recent cases, merger review, and the IP-antitrust interface including standard essential patents and FRAND, patent settlements after Actavis, and licensing restrictions. It closes with a compliance program, a worked example, an FAQ, and related reading.

Antitrust law asks a question that sounds simple and is not: did this company win because it was better, or because it prevented anyone else from competing?

In most industries that question is answerable from ordinary evidence about prices, output, and quality. In technology markets it is much harder, because the same features that make a platform valuable to users, network effects, data advantages, integration, and default placement, are also the features that make entry difficult. A messaging app is more useful when everyone is on it. An operating system is more useful with more applications. Those are genuine consumer benefits and genuine entry barriers at the same time.

This is why technology antitrust generates such passionate disagreement, and why the same conduct can be described accurately as "improving the product" and "foreclosing rivals."

The short answer

Sherman Act § 1, 15 U.S.C. § 1, prohibits contracts, combinations, and conspiracies in restraint of trade. It requires concerted action, meaning an agreement between two or more parties.

Sherman Act § 2, 15 U.S.C. § 2, prohibits monopolization, attempted monopolization, and conspiracy to monopolize. It reaches unilateral conduct.

Clayton Act § 7, 15 U.S.C. § 18, prohibits acquisitions whose effect "may be substantially to lessen competition, or to tend to create a monopoly."

FTC Act § 5, 15 U.S.C. § 45, prohibits "unfair methods of competition," which the Commission has asserted reaches beyond the Sherman Act's boundaries. The scope of that authority is contested.

Monopolization requires two elements: (1) possession of monopoly power in a relevant market, and (2) the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident. United States v. Grinnell Corp., 384 U.S. 563, 570-71 (1966).

Being a monopolist is not unlawful. The conduct is.

Remedies are substantial: treble damages and attorney's fees for private plaintiffs, 15 U.S.C. § 15; injunctive relief including structural remedies; and criminal penalties for hard-core § 1 violations, including individual imprisonment.

Part I: Section 2 monopolization

Market definition and market power

Nearly every § 2 case is decided at market definition, and defense counsel should start there.

The relevant market has a product dimension and a geographic one. The standard tool is the hypothetical monopolist test: would a hypothetical monopolist of the candidate product set profitably impose a small but significant and non-transitory increase in price? If not, the market is too narrow.

Monopoly power is "the power to control prices or exclude competition." It is usually inferred from a dominant market share plus barriers to entry, though direct evidence of the power to raise price or exclude is better where available. Courts have generally treated shares above roughly 70 percent as sufficient to support an inference, with shares below 50 percent rarely sufficient.

Why this is hard for platforms:

  • Zero-price products. If a service is free to users, the price-based hypothetical monopolist test does not work directly. Courts and agencies have adapted by considering quality, privacy, and advertising load as the competitive dimensions, which is analytically sound and evidentially messy.
  • Two-sided markets. Ohio v. American Express Co., 585 U.S. 529 (2018), held that for a two-sided transaction platform, the relevant market includes both sides, and a plaintiff must show a net anticompetitive effect across both. That holding was written for credit cards and has been argued, with mixed success, across the technology sector. Courts have distinguished platforms that are not "transaction platforms" in the Amex sense, and the Epic litigation addressed the question directly.
  • Aftermarkets. Eastman Kodak Co. v. Image Technical Services, Inc., 504 U.S. 451 (1992), held that a single brand's parts and service aftermarket can be a relevant market where information costs and switching costs lock in customers. This is the theory behind many app store, repair, and consumables disputes. See The Right to Repair Movement.

Exclusionary conduct

The second element asks whether the conduct is exclusionary rather than competition on the merits.

United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001) (en banc), is the foundational modern framework and still the best statement of the analysis. The court set out a burden-shifting structure:

  1. The plaintiff must show the conduct has an anticompetitive effect, meaning it harms the competitive process and thereby harms consumers, not merely that it harms a competitor.
  2. If so, the defendant may offer a procompetitive justification, a nonpretextual claim that the conduct is a form of competition on the merits.
  3. If the defendant does, the plaintiff must rebut the justification or show that the anticompetitive harm outweighs the procompetitive benefit.

Microsoft also held that the browser tying claim should be analyzed under the rule of reason rather than per se, because of the risk of condemning integration in "platform software products" that courts do not fully understand. That caution has aged into the central problem of technology antitrust.

The recurring conduct theories

Exclusive dealing. Agreements that foreclose rivals from a substantial share of distribution. The analysis considers the share foreclosed, the duration and terminability of the agreements, and the availability of alternative channels. Default placement agreements, preinstallation requirements, and revenue-share arrangements conditioned on exclusivity are the technology versions.

Tying. Conditioning the sale of one product on the purchase of another. Illinois Tool Works Inc. v. Independent Ink, Inc., 547 U.S. 28 (2006), overruled the presumption that a patent confers market power, so a tying plaintiff must prove market power in the tying product. In technology, the recurring question is whether integrated functionality is a "tie" at all, and Microsoft's rule of reason approach governs.

Self-preferencing. A platform ranking or featuring its own products ahead of rivals'. Whether this is exclusionary is genuinely contested in American law; the European Union has treated it as abuse of dominance, and the Digital Markets Act now prohibits it for designated gatekeepers.

Refusals to deal and interoperability. The general rule is that a firm has no duty to deal with rivals. Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398 (2004), described the exception recognized in Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585 (1985), as "at or near the outer boundary of § 2 liability." Aspen Skiing involved termination of a prior voluntary course of dealing that had been profitable, and a willingness to forgo short-run profits to achieve an anticompetitive end. Absent those facts, a refusal-to-deal claim is very difficult. Degrading interoperability, revoking API access, and terminating an existing partner are the technology fact patterns that come closest.

Predatory pricing. Requires pricing below an appropriate measure of cost and a dangerous probability of recouping the investment through later supracompetitive pricing. Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993). The recoupment requirement makes these claims very hard, and the analysis is awkward for products with near-zero marginal cost. Pacific Bell Telephone Co. v. linkLine Communications, Inc., 555 U.S. 438 (2009), rejected a price-squeeze claim where there was no antitrust duty to deal at the wholesale level and no predatory pricing at retail.

Acquisitions of nascent competitors. The theory that dominant firms buy potential rivals before they mature. This has driven both merger enforcement and monopolization claims, and it is evidentially difficult because it requires proving what the acquired firm would have become.

Part II: Section 1 agreements

The three levels of scrutiny

Per se unlawful: horizontal price fixing, bid rigging, market allocation, and group boycotts in their naked forms. These require no market analysis and are criminally prosecuted. No-poach and wage-fixing agreements among competitors for employees are treated as horizontal market allocation, and the Antitrust Division has prosecuted them criminally, with mixed trial results but continued enforcement.

Quick look: for restraints whose anticompetitive character is obvious but that warrant some justification analysis.

Rule of reason: everything else, including vertical restraints. Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877 (2007), overruled the per se rule against minimum resale price maintenance, applying rule of reason. Several states continue to treat RPM as per se unlawful under state law, which matters for distribution programs. See Gray Market Goods and the First Sale Doctrine.

NCAA v. Alston, 594 U.S. 69 (2021), is the Supreme Court's most recent extended treatment of the rule of reason, applying the burden-shifting framework and rejecting deferential treatment for the NCAA's restraints.

Proving agreement

Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007), requires a complaint to plead facts plausibly suggesting agreement; parallel conduct alone is insufficient because it is "just as much in line with a wide swath of rational and competitive business strategy." See Motions to Dismiss Under Rule 12.

Algorithmic pricing has produced a new version of this problem. Where competitors independently use a common pricing algorithm fed by their own confidential data, plaintiffs allege a hub-and-spoke conspiracy with the software vendor as the hub. Courts have divided on whether that states a claim, and both agencies have taken the position that using an algorithm does not immunize conduct that would be unlawful if done by people. This is an active and unsettled area, and any company deploying shared pricing software should get advice before, not after.

Part III: The current cases

The technology sector is in its most active antitrust period since the 1990s. The following summarizes the posture as of this writing; all of these matters have moved and will continue to move, and remedies and appeals remain in progress in several.

Search. A federal district court held in 2024 that Google violated § 2 by maintaining monopolies in general search services and general search text advertising, principally through exclusive default placement agreements with browser and device makers. The 2025 remedies decision declined to order divestiture of the Chrome browser and instead imposed behavioral remedies, including restrictions on exclusive default agreements and obligations to share certain search data with qualified competitors. Appeals followed.

Advertising technology. A separate federal action produced a 2025 holding that Google unlawfully monopolized markets for publisher ad servers and ad exchanges and unlawfully tied the two, with remedies proceedings following.

App stores. Epic Games, Inc. v. Apple Inc., 67 F.4th 946 (9th Cir. 2023), affirmed judgment for Apple on the federal antitrust claims while affirming an injunction under California's Unfair Competition Law barring Apple's anti-steering rules. Subsequent enforcement proceedings resulted in a 2025 order requiring Apple to permit developers to link to outside purchasing without commission, following a finding that Apple had not complied with the original injunction. In the parallel Google matter, a jury found for Epic in 2023, the district court entered a broad injunction addressing Play Store distribution and billing, and the Ninth Circuit affirmed in 2025.

Social media. The FTC's monopolization case against Meta, premised on the acquisitions of Instagram and WhatsApp, went to trial in 2025 and resulted in judgment for Meta, principally on market definition: the court concluded the FTC had not proved a "personal social networking services" market in which Meta held monopoly power, in light of competition from short-form video services.

Retail and cloud. The FTC's case against Amazon, and various private and state actions, remain pending.

Standard essential patents. FTC v. Qualcomm Inc., 969 F.3d 974 (9th Cir. 2020), reversed a district court judgment against Qualcomm, holding that its licensing practices, including refusing to license chip-level competitors and its "no license, no chips" policy, were not anticompetitive under § 2, and emphasizing Trinko's narrow duty-to-deal doctrine and the distinction between harm to rivals and harm to competition.

The pattern worth noticing: market definition is deciding these cases. Where the government has defined a market successfully, it has won on liability; where it has not, it has lost regardless of the conduct evidence.

Part IV: Mergers

The statute and the process

Clayton Act § 7 prohibits acquisitions that may substantially lessen competition. The Hart-Scott-Rodino Act, 15 U.S.C. § 18a, requires premerger notification and a waiting period for transactions above adjusted thresholds. HSR filing requirements were substantially expanded by rules effective in 2025, materially increasing the information required and the preparation time.

The 2023 Merger Guidelines

The Department of Justice and the Federal Trade Commission jointly issued new Merger Guidelines in December 2023, replacing the separate horizontal and vertical guidelines. They articulate a set of frameworks, including:

  • Concentration thresholds returning to more stringent levels than the 2010 guidelines.
  • Attention to potential and nascent competition, including acquisitions that eliminate a firm that might have become a competitor.
  • Serial acquisitions and roll-up strategies evaluated in the aggregate.
  • Multi-sided platforms treated as a distinct category, with attention to entrenchment of a dominant position.
  • Labor market effects as a cognizable harm.
  • Vertical theories including foreclosure and access to rivals' competitively sensitive information.

The guidelines are not binding on courts, and courts have applied them with varying degrees of receptiveness. They are, however, the operative statement of what the agencies will investigate, which makes them the working document for deal planning.

Practical deal considerations

  • Assess antitrust risk before signing, not during diligence.
  • Document creation discipline matters enormously. Ordinary-course documents describing a target as a "threat" or an acquisition as "eliminating a competitor" are the most damaging evidence in merger litigation, and they are almost always created by business people who have never seen a second request.
  • Allocate risk in the agreement: efforts covenants ("reasonable best efforts" versus "hell or high water"), divestiture obligations, timing outside dates, and reverse termination fees.
  • Gun-jumping is a real risk: coordinating pricing or exchanging competitively sensitive information before closing violates § 1 and HSR. Use clean teams.
  • Foreign filings may be required in dozens of jurisdictions with different thresholds and different theories.

Part V: The IP-antitrust interface

Intellectual property grants exclusive rights. Antitrust polices exclusion. The two are less in tension than the framing suggests, because the agencies' longstanding position, reflected in the DOJ and FTC Antitrust Guidelines for the Licensing of Intellectual Property, is that IP is comparable to other property for antitrust purposes, that IP does not necessarily create market power, and that licensing is generally procompetitive.

Three propositions follow, and each has a qualification.

1. A patent does not confer market power. Illinois Tool Works settled this for tying claims. The plaintiff must prove market power in the tying product with evidence.

2. A unilateral refusal to license is generally lawful. Patent law expressly provides that a patent owner shall not be denied relief or deemed guilty of misuse by reason of having "refused to license or use any rights to the patent," 35 U.S.C. § 271(d)(4). Combined with Trinko, this makes refusal-to-license claims very difficult. The qualification: a refusal that is part of a broader scheme, or that reverses a prior course of dealing on Aspen Skiing facts, may be actionable.

3. Licensing restrictions are analyzed under the rule of reason, considering whether the restraint is between competitors, whether it forecloses a substantial share, and whether it is reasonably necessary to a procompetitive collaboration. Field-of-use, territorial, and exclusivity restrictions are ordinarily lawful. Grantbacks, package licensing, and cross-licensing among competitors require more care. See How to License Your Patent.

Standard essential patents and FRAND

When a standard-setting organization adopts a technology, patents essential to practicing the standard acquire market power that comes from the standard's adoption rather than from the invention's merits. SSOs address this by requiring members to commit to license essential patents on fair, reasonable, and non-discriminatory terms.

Antitrust exposure arises from two behaviors:

  • Deception in standard setting. Failing to disclose essential patents while participating in the standard's development, then asserting them afterward, can support a monopolization claim. Broadcom Corp. v. Qualcomm Inc., 501 F.3d 297 (3d Cir. 2007), held that "in a consensus-oriented private standard-setting environment, a patent holder's intentionally false promise to license essential proprietary technology on FRAND terms, coupled with an SDO's reliance on that promise when including the technology in a standard, and the patent holder's subsequent breach of that promise, is actionable anticompetitive conduct."
  • Hold-up. Demanding supracompetitive royalties or injunctions after lock-in.

But the antitrust route is disfavored. FTC v. Qualcomm reflects considerable judicial skepticism about converting FRAND disputes into monopolization cases, emphasizing that breach of a FRAND commitment is primarily a contract matter, that harm to rivals is not harm to competition, and that Trinko forecloses most duty-to-deal theories.

The practical consequence: FRAND disputes are litigated as contract and patent cases, with damages determined through comparable-license analysis, and antitrust counterclaims are pleaded more often than they succeed. See Standard Essential Patents and FRAND Licensing in 5G and IoT.

Patent settlements

FTC v. Actavis, Inc., 570 U.S. 136 (2013), held that a reverse payment settlement, in which a patent holder pays an alleged infringer to stay out of the market, can violate the antitrust laws and is analyzed under the rule of reason rather than being immune because it falls within the patent's nominal scope. A large and unjustified payment can be evidence of the patentee's own doubts about the patent's validity.

Actavis arose in pharmaceuticals, but its logic reaches any settlement where a payment flows from the patentee to the accused infringer in exchange for delayed entry. Technology companies settling patent litigation with competitors should structure settlements to avoid the pattern, or be prepared to justify the payment as consideration for something other than delay.

Other interface issues

  • Patent pools are generally procompetitive where they include only essential, complementary patents, exclude substitutes, and do not facilitate coordination among members.
  • Patent misuse is a defense to infringement, not an antitrust claim, though the analyses overlap. Section 271(d) narrows it substantially.
  • Sham litigation is exempt from Noerr-Pennington immunity only where it is objectively baseless and subjectively motivated to interfere with a competitor's business.
  • Trade secret and non-compete practices intersect with labor market antitrust; see Non-Compete Agreements Under Siege.

Part VI: A compliance program

Antitrust compliance is unusual because the highest-risk conduct is often invisible to the legal department: a conversation at a trade association, a Slack message about a competitor's pricing, a slide describing an acquisition target as a threat.

Governance

  • Written antitrust policy, distributed and acknowledged.
  • Named compliance owner with a reporting line to the general counsel.
  • Annual risk assessment covering markets, market shares, distribution, and labor practices.

Training, targeted at the people who create risk

  • Sales and pricing teams: no discussion of price, output, customers, or territories with competitors, ever.
  • HR and recruiting: no agreements with other companies about hiring, wages, or benefits.
  • Business development: how to describe competitors and acquisitions in documents.
  • Product and engineering: interoperability decisions, API access terminations, and integration choices have antitrust dimensions.
  • Executives: the documents-in-litigation reality.

Document discipline

  • Guidance on writing about competition: describe your own product's merits, not your intent to exclude.
  • Prohibit language like "crush," "kill," "block them out," and "make it impossible to switch" in ordinary-course documents. This is not about hiding intent; it is about not creating misleading evidence of intent that does not exist.
  • Rules for market share estimates in board materials, which become the plaintiff's market definition evidence.

Trade associations

  • Counsel review of agendas before meetings.
  • Written protocol: leave the room and document leaving if a discussion turns to price, output, or customers.
  • No exchange of current or forward-looking competitively sensitive information, even through an intermediary or an aggregator, without counsel review.

Distribution and pricing

  • MAP policies drafted as unilateral policies, not agreements.
  • Exclusivity provisions reviewed for duration, terminability, and foreclosure share.
  • Most-favored-nation and parity clauses reviewed; these have drawn enforcement attention in platform contexts.
  • Bundled and loyalty discount structures analyzed before launch.

Transactions

  • Early antitrust assessment before signing.
  • Clean team protocols for diligence.
  • HSR analysis and document collection planning.
  • Efforts covenants and risk allocation negotiated deliberately.

Response

  • Protocol for civil investigative demands and second requests.
  • Preservation and litigation hold procedures. See Litigation Holds, Spoliation, and Rule 37(e).
  • Leniency analysis if criminal conduct is discovered; the Antitrust Division's Leniency Program provides substantial benefits to the first to report, and the timing advantage is decisive.

Part V-A: Private enforcement, standing, and the European overlay

Who may sue. Section 4 of the Clayton Act gives "any person ... injured in his business or property" a treble damages action. Two doctrines narrow it substantially.

Antitrust injury requires that the plaintiff's injury be "of the type the antitrust laws were intended to prevent and that flows from that which makes defendants' acts unlawful." Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977). A competitor injured because a rival competed harder has no claim; a competitor injured because a rival excluded it unlawfully does.

Indirect purchaser standing is barred federally. Illinois Brick Co. v. Illinois, 431 U.S. 720 (1977), holds that only the direct purchaser may sue for overcharges, and Hanover Shoe, Inc. v. United Shoe Machinery Corp., 392 U.S. 481 (1968), bars a defendant from arguing that the direct purchaser passed the overcharge downstream. Most states have enacted Illinois Brick repealer statutes allowing indirect purchaser claims under state law, which is why nearly every antitrust class action has parallel federal direct-purchaser and state indirect-purchaser tracks. Apple Inc. v. Pepper, 587 U.S. 273 (2019), held that consumers who bought apps through Apple's App Store were direct purchasers of Apple, which materially expanded platform exposure.

The limitations period is four years from accrual, 15 U.S.C. § 15b, with continuing violation and fraudulent concealment doctrines extending it in some cases. A government action tolls private claims under 15 U.S.C. § 16(i), which is why private suits routinely follow government complaints.

Section 5(a) of the Clayton Act, 15 U.S.C. § 16(a), makes a final government judgment prima facie evidence in a private action, which is a substantial advantage for follow-on plaintiffs and a reason government findings matter far beyond the government's own remedy.

The European layer. Any company of scale should assume parallel European exposure. Article 101 and Article 102 of the Treaty on the Functioning of the European Union are the analogues to Sherman Act §§ 1 and 2, and Article 102 abuse-of-dominance doctrine reaches self-preferencing and refusals to supply more readily than American monopolization law does. The Digital Markets Act goes further still, imposing ex ante obligations on designated gatekeepers, including prohibitions on self-preferencing, restrictions on combining personal data across services, and interoperability requirements, enforced with penalties up to 10 percent of worldwide turnover and 20 percent for repeat infringements. Compliance obligations under the DMA are not conditioned on proving market effects, which makes them operationally very different from an American case. The United Kingdom's digital markets regime, and comparable frameworks in other jurisdictions, add further layers.

The practical planning point: for a global platform, European regulation may drive product design before an American court reaches a judgment, and the design changes made for one jurisdiction become evidence in the other. Coordinate the analysis across jurisdictions rather than treating them as separate compliance projects. See Global Patent Litigation Strategies for the analogous coordination problem in IP enforcement.

A worked example

Northlight Systems, Inc. (fictional) operates a workflow platform with roughly 55 percent share of what it calls the "creative production workflow" category. It is considering four initiatives.

Initiative 1: exclusive integration agreements. Northlight wants three-year exclusive integration deals with the five largest asset-management vendors, prohibiting them from integrating with Northlight's competitors.

Analysis. Exclusive dealing by a firm with substantial share, foreclosing the primary distribution path for rivals. The relevant questions: what share of the market's access do those five vendors represent; can rivals reach customers another way; how long is the term; is it terminable. Three years with the five largest is aggressive. Recommendation: shorten to one year, make it terminable on notice, drop exclusivity in favor of preferred-placement and co-marketing commitments, and document the procompetitive rationale (joint engineering investment) contemporaneously.

Initiative 2: terminating a competitor's API access. A rival built a migration tool using Northlight's public API. Northlight wants to revoke its key.

Analysis. Refusal-to-deal territory. Under Trinko the general rule favors Northlight, but this has Aspen Skiing features: a prior course of dealing, termination directed at a specific rival, and a sacrifice of API revenue. Recommendation: if there is a genuine neutral basis (rate limits, security, terms violations), apply it to everyone through a published policy and enforce it consistently. If the only reason is that the rival is competing, do not do it.

Initiative 3: acquiring a two-year-old startup whose product could develop into a competitor.

Analysis. Squarely within the 2023 Merger Guidelines' nascent competition framework, and within the theory the agencies have pursued most aggressively. Recommendation: assess HSR reportability, assume the deal will be scrutinized regardless of size, and audit the document record now. If the deal team has written that the target is a "threat we need to take out," that document will define the case.

Initiative 4: a no-poach understanding with a partner company, agreed informally between two vice presidents at a conference.

Analysis. Stop immediately. Naked no-poach agreements among competitors for labor are treated as per se unlawful market allocation and have been prosecuted criminally. Legitimate ancillary restraints exist within genuine collaborations, but they must be narrow, in writing, and reasonably necessary to the collaboration. Recommendation: terminate the understanding, document the termination, assess leniency, and train the entire executive team the same week.

The through-line. Three of the four initiatives were fixable by changing how they were structured rather than by abandoning the business objective. The fourth was not fixable and had to stop. That ratio is typical, and it is why early antitrust review is cheap and late antitrust review is not.

Frequently asked questions

Is it illegal to have a monopoly? No. Monopoly power obtained through a superior product, business acumen, or historic accident is lawful. What is unlawful is acquiring or maintaining it through exclusionary conduct.

What share makes us a monopolist? There is no fixed number. Courts have generally found shares above roughly 70 percent sufficient to support an inference of monopoly power, along with entry barriers, and have rarely found monopoly power below 50 percent. For attempted monopolization, a dangerous probability of success is required, and courts have accepted lower shares with strong evidence of intent and entry barriers.

Can we refuse to do business with a competitor? Generally yes. The duty-to-deal exception is narrow and requires facts close to Aspen Skiing: termination of a profitable prior course of dealing, apparently to achieve an anticompetitive end. Terminating a long-standing partner because it began competing is the fact pattern to avoid.

Are exclusive contracts illegal? No, and they are common and often procompetitive. They become problematic when a firm with market power forecloses a substantial share of the market for a substantial period without adequate justification. Duration, terminability, and foreclosure share are the levers.

What about most-favored-nation clauses? They can be procompetitive, and they have also drawn enforcement attention in platform contexts where they prevent sellers from offering better prices elsewhere. Review platform-wide parity provisions before adopting them.

Can we talk to competitors at a trade association? About standards, safety, regulation, and industry advocacy, yes, with counsel involvement. About price, output, customers, territories, wages, or hiring, never. Have a protocol for leaving the room and documenting it.

Is using a pricing algorithm risky? It can be. Independently developed algorithms using your own data are ordinary business tools. Shared algorithms fed by competitors' confidential data, producing coordinated outcomes, have been alleged to constitute a hub-and-spoke conspiracy, and both agencies have said that delegating pricing to an algorithm does not immunize the result. Get advice before deploying.

Does our patent give us market power? Not automatically. Illinois Tool Works rejected that presumption. Market power must be proved with evidence about substitutes.

Do we have to license our SEPs? If you made a FRAND commitment to a standard-setting organization, you have contractual obligations, and breach is primarily a contract matter. Antitrust exposure exists mainly for deception during standard setting. Injunctions against willing licensees raise both contract and antitrust questions.

What happens in a government investigation? Typically a civil investigative demand or a second request, followed by document production, interviews or depositions, and economic analysis. Preservation obligations attach immediately. Engage specialist counsel at the first contact, not after the production.

Closing thought

The hardest thing about antitrust in technology markets is that the doctrine asks courts to distinguish between winning and excluding, in industries where the winning strategy often is to make the product so integrated, so networked, and so default that switching stops making sense.

Reasonable people disagree profoundly about where that line sits, and the disagreement is not going to be resolved by better legal writing. It will be resolved, slowly and unevenly, by cases that turn on market definition and on the specific documents companies wrote about their own intentions.

That last point is the practical one. In every major technology antitrust case of the last thirty years, from Microsoft forward, the most damaging evidence has been ordinary internal communication: an email describing a plan to cut off a rival's oxygen, a strategy deck framing an acquisition as eliminating a threat, a chat message about making it painful to leave.

None of those documents changed what the company did. They changed how a court understood it. Training the people who write them is the highest-return antitrust compliance activity there is, and it costs a fraction of the first day of a second request.

Related articles

This article is provided for general informational purposes and does not constitute legal advice. Antitrust enforcement priorities and major cases described here are in active litigation and have continued to develop; verify the current posture before relying on any outcome stated. Consult qualified antitrust counsel about any particular conduct, transaction, or investigation.