Algorithmic Pricing and Antitrust

Short Answer

Pricing software does not by itself establish an antitrust violation. Risk increases when competitors exchange nonpublic pricing information through a common provider or agree to follow common recommendations. The DOJ’s RealPage complaint and November 24, 2025 proposed final judgment illustrate that theory. A proposal is not proof of the complaint’s allegations or, without the entered judgment, a statement of current court obligations. Texas Business and Commerce Code § 15.05 and federal Sherman Act Section 1 apply according to their elements. Review data sharing, independence and the actual use of recommendations before adopting a shared tool.

Which Laws Apply

Texas AI-specific: Business and Commerce Code § 552.003 (preemption of local AI regulation).

Generally applicable Texas law: Texas Free Enterprise and Antitrust Act, Business and Commerce Code § 15.05; DTPA for pricing representations.

Federal: Sherman Act sec. 1, 15 U.S.C. § 1; FTC Act sec. 5.

The Theory

Sherman Act Section 1 and Business and Commerce Code § 15.05(a) address agreements restraining trade. A common vendor is relevant to the inquiry but is not itself proof of an agreement. The DOJ’s RealPage complaint alleges pooling nonpublic, competitively sensitive information and features supporting coordinated rents. Buyers should identify what competitor information enters the system, whether independent decisions remain possible and how departures from recommendations are treated.

The Proposed RealPage Judgment

On November 24, 2025, the Justice Department filed a proposed final judgment in its antitrust case against RealPage in the U.S. District Court for the Middle District of North Carolina (No. 1:24-cv-00710-WO-JLW). The filed proposal would bar specified uses of nonpublic competitor information, restrict model training to data at least 12 months old, change particular recommendation features and provide monitoring. The proposal remains subject to court approval: under the Tunney Act, the court may enter it only after public comment and a finding that entry is in the public interest. These are terms in the dated proposal, not a finding that the complaint’s allegations are true. Later case relief must be read in the entered order before it is treated as a current obligation.

Other State Measures

TRAIGA Business and Commerce Code § 552.003 preempts political-subdivision regulation regarding the use of AI systems. Whether a pricing ordinance falls within that language depends on its text and the software’s function. Statewide antitrust law and an independently supported antitrust claim require their own analysis.

Practical Steps

Ask what data the tool uses and whether competitors’ nonpublic data feeds recommendations.

Keep pricing decisions independent and documented; do not commit to follow recommendations.

Avoid features that discourage price decreases or align prices with competitors.

Review vendor marketing that promises above-market results through shared data.

Illustrative Example (Hypothetical)

Hypothetical: a Texas apartment owner adopts a pricing tool using public listings and its own leasing data. Managers retain independent authority. Those facts differ from pooling competitors’ nonpublic current lease data and enforcing common recommendations. Ask what information is exchanged, whether an agreement exists and how the recommendations are used; the label AI does not answer the antitrust question.

What Is Unsettled

Agreement, information exchange and competitive effects remain fact-dependent. A complaint, a proposed settlement and an entered judgment have different legal effects. Preserve the data-source and recommendation-use records needed to test the relevant elements.