Skip to main content
v2026.11,610 entries · CC-BY 4.0
Dictionary termTrack DProposedv2026.1

Non-Practicing Entity (NPE)

An entity that owns one or more issued or pending patents but does not manufacture, sell, or otherwise commercially practice the invention itself, and instead generates value from the patent through licensing, litigation/demand-letter assertion, or both. "Non-practicing" describes the entity's relationship to the patent, not the patent's validity -- individual inventors, universities and research institutions, pre-commercialization startups, failed operating companies holding residual portfolios, and dedicated patent-acquisition/licensing firms are all NPEs in this technical sense, even though only a narrower subset of them (the FTC's "patent assertion entity," or PAE, category, and the pejorative "patent troll" label) draw the negative connotation the term is often assumed to carry.

ByCASRAI Editorial Board
· Last updated 22 Aug 2026

Ask about Non-Practicing Entity (NPE)

Answers are drawn from this dictionary entry and the rest of the CASRAI corpus, with a link to every source.

Answers are AI-generated from CASRAI’s own published pages and can be wrong, so check the linked sources before relying on one; your question is logged without personal data — never sold, never used to train a third-party model — to show us what CASRAI is missing, so please do not type personal or confidential details. How we use this

Examples

Worked examples

  • Is an instance

    A university technology-transfer office holds an issued patent on a diagnostic assay invented by a faculty researcher. The university does not manufacture diagnostic kits itself; it out-licenses the patent to an existing diagnostics company, or to a faculty spinout, which will make and sell the product. In the neutral, technical sense, the university is an NPE -- but because the patent originated from the university's own research and the licensing goal is commercialization rather than litigation revenue, it is not the kind of entity the "patent troll" label was coined to describe.

  • Is an instance

    A firm purchases a portfolio of patents out of a bankrupt electronics company's asset sale, having done none of the underlying research itself, then sends infringement demand letters to hundreds of small businesses that use a common product feature the portfolio covers, offering to settle for an amount set just below the cost of hiring a defense attorney. This acquired-to-assert pattern is the model the FTC's "patent assertion entity" (PAE) category and the pejorative "patent troll" label both describe.

Counter-examples

Looks similar, but isn't

  • Not an instance

    A semiconductor manufacturer designs, manufactures, and sells its own chips, and also holds a large patent portfolio that it licenses to competitors under cross-licensing agreements. Because it practices its own patents by selling products that use them -- licensing to others is a secondary activity, not its sole business model -- it is not an NPE, regardless of how large or actively licensed its portfolio is.

Editorial commentary

A non-practicing entity (NPE) is a patent holder that does not manufacture, sell, or otherwise commercially use the invention covered by its patent. Instead of practicing the patent itself, an NPE generates value from it in one of two ways — licensing the patent to a company that will make or sell the product, asserting the patent through infringement litigation or demand letters against companies that allegedly do, or some mix of both. “Non-practicing” describes the entity’s relationship to the patent (it doesn’t use the invention commercially), not the validity or value of the patent itself — an NPE’s patents can be entirely legitimate.

NPE is frequently used as a synonym for “patent troll,” but the two terms are not the same thing, and conflating them causes real confusion in technology-transfer contexts — because under the plain, technical definition above, a university that patents faculty inventions and licenses them out is also an NPE. Getting this distinction right matters for how a TTO talks about its own patent portfolio and how it evaluates entities that approach it.

NPE, PAE, and “patent troll”: three different terms, one common source of confusion

These three terms are often used interchangeably in casual conversation, but they sit at different levels of precision:

  • Non-practicing entity (NPE) — the broadest, most neutral term. Any patent owner that doesn’t practice its own patent falls under it: individual inventors, universities, national labs, failed operating companies holding leftover patents, and dedicated patent-monetization firms alike.
  • Patent assertion entity (PAE) — a narrower analytical category used by the U.S. Federal Trade Commission in its 2016 study of patent monetization, Patent Assertion Entity Activity: An FTC Study. The FTC defines a PAE as “a business that acquires patents from third parties and seeks to generate revenue by licensing those patents to or asserting them against alleged infringers.” Critically, the FTC’s PAE definition explicitly excludes most individual inventors, research labs, universities, and companies that develop their own patent portfolios — because those entities typically patent inventions they created themselves, rather than acquiring someone else’s patents purely to monetize them.
  • “Patent troll” — a pejorative, not a technical term. It is commonly attributed to Peter Detkin, an Intel assistant general counsel, who coined it around 1999 to describe entities that, in his words, try “to make a lot of money from a patent that they are not practicing, have no intention of practicing, and have not made any investment in.” The label carries a value judgment — low-quality or overly broad patents, acquisition purely for litigation leverage, demand letters sent in bulk to end users rather than manufacturers, and settlement amounts calibrated just below the cost of a legal defense rather than reflecting the invention’s actual value. Because it’s a value judgment and not a defined legal category, “none of [its definitions] are considered fully satisfactory” even among people who study the phenomenon, and it should be attributed as an opinion (“critics call X a patent troll”) rather than asserted as neutral, settled terminology.

The practical takeaway: every PAE and every colloquial “patent troll” is an NPE, but the reverse is not true — most NPEs, including essentially every research university, are neither.

Why universities are technically NPEs — and why that’s not the same as being a “patent troll”

A university does not manufacture or sell products. When a university technology-transfer office (TTO) patents an invention disclosed by a faculty researcher, the university becomes the patent’s owner but, by definition, someone else — a licensee, whether an established company or a faculty spinout — has to actually build and sell whatever the patent covers. That makes the university an NPE in the plain, technical sense used above.

Universities hold this position at scale specifically because of the Bayh-Dole Act (1980), which lets universities and other federal-funding recipients elect to retain title to inventions made with federal research funding, then patent and license them rather than assigning that ownership to the government. Bayh-Dole is the statutory reason the modern university TTO — and, with it, the university-as-patent-owner — exists at all.

But being an NPE in the technical sense is not the same as being a “patent troll” in the pejorative sense, and the distinction the field draws is a pattern-of-conduct distinction, not a legal one:

  • Patent origin. University patents typically cover inventions that originated in the university’s own labs. PAEs and colloquial “trolls” typically acquire patents from third parties (often out of bankrupt companies’ asset sales) specifically to assert them — they didn’t do the underlying research.
  • Licensing intent. A TTO’s stated mission is to move an invention into commercial use — the license itself is meant to result in a product reaching the market, with royalties tied to actual sales. A litigation-driven NPE’s business model is the assertion or settlement itself, independent of whether the technology ever reaches a product.
  • Enforcement pattern. Universities rarely file first-strike infringement suits against end users of a product; when university-owned patents do end up in litigation, it is far more often to enforce a license already in place or respond to a licensee dispute than as a standalone revenue strategy.

This is exactly the distinction the FTC’s PAE study draws when it excludes universities from its PAE definition, and it’s reflected in the broader patent-policy literature: entities such as universities and national laboratories that don’t practice their asserted patents are generally not described as “patent trolls” when they license their technology on reasonable terms established in advance, rather than acquiring patents opportunistically to extract settlements.

That said, this is a description of typical conduct, not a blanket exemption. A university licensing office that starts behaving like a PAE — buying in outside patents to assert, or sending broad demand letters to end users rather than negotiating licenses with makers of a product — would draw exactly the same scrutiny and the same “troll” comparison that any other entity engaged in that conduct would. The label tracks behavior, not institutional identity.

Why this distinction matters operationally for a tech-transfer office

  • Framing your own program. Being precise about NPE vs. PAE vs. “patent troll” matters when a TTO talks to faculty inventors, licensees, journalists, or state legislators about the university’s own patenting and licensing activity — being an NPE is a factual, defensible description; being called a “patent troll” is a reputational hit worth actively managing against, not a label a research university wants attached to its licensing program.
  • Evaluating counterparties. TTOs are sometimes approached by patent aggregators or invention-monetization firms interested in acquiring or co-licensing university-owned patents specifically for assertion purposes. AUTM’s widely adopted 2007 statement, In the Public Interest: Nine Points to Consider in Licensing University Technology — originally signed by eleven major research universities and the Association of American Medical Colleges — includes a principle urging licensing offices to be mindful of the implications of dealing with patent aggregators, precisely because doing so can pull a university’s licensing program into PAE-style conduct.
  • Responding to a demand letter or infringement claim. Understanding the NPE/PAE taxonomy helps a TTO or general counsel triage quickly whether an incoming claim is from an operating competitor asserting patents on products it actually sells, or from an entity that holds patents purely for assertion — which affects both the litigation risk calculus and settlement strategy.
  • Explaining licensing structure to spinouts. A faculty spinout licensing a patent from its own university needs to understand that the university, as licensor, is inherently non-practicing by design — the spinout itself is the practicing party that has to commercialize the invention, and the license terms (exclusivity, field-of-use, milestones) exist to make sure that actually happens.

Litigation dynamics: venue, injunctive relief, and validity challenges

Three developments in U.S. patent litigation over the past two decades specifically shaped how NPE assertion works in practice, and they matter for how a university TTO thinks about its own leverage and its own exposure:

  • Injunctive relief is generally unavailable to non-practicing patent owners. In eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006), the Supreme Court held that a patent owner must satisfy the traditional four-factor equitable test — including irreparable harm — before a court will grant a permanent injunction, rather than an injunction issuing automatically upon a finding of infringement. Because an entity that doesn’t practice its own patent generally cannot show it is losing sales or market share to the infringer, NPEs (including universities) are typically limited to monetary relief — damages or an ongoing royalty — rather than the ability to force a defendant to stop selling a product. This shifted NPE assertion economics firmly toward settlement value rather than injunctive leverage, and it is one reason a TTO’s own licensing strategy generally centers on negotiated royalty terms rather than the threat of shutting a licensee’s product down.
  • Venue was narrowed. TC Heartland LLC v. Kraft Foods Group Brands LLC (2017) held that, for patent infringement suits, proper venue under the patent venue statute is limited to where the defendant is incorporated or where it has committed acts of infringement and has a regular and established place of business — reversing decades of looser appellate practice that had let plaintiffs, including many NPEs, file the large majority of U.S. patent suits in a small number of plaintiff-favorable districts. The decision measurably reduced that concentration and gave defendants a more predictable venue picture.
  • Patent validity can be challenged administratively, not just in district court. The Leahy-Smith America Invents Act (2011) created inter partes review (IPR) at the Patent Trial and Appeal Board (PTAB), letting an accused infringer petition to have a patent’s validity reviewed by the USPTO rather than only litigating validity as a defense in district court. IPR is faster and cheaper than district-court litigation and has become a standard defensive tool against an asserted patent, NPE-held or otherwise — a company facing an NPE demand letter will often evaluate an IPR petition alongside, or instead of, a district-court defense.

For a TTO, the practical read-through is that the same legal environment that curbs aggressive NPE litigation strategies also curbs a university’s own leverage if a licensing negotiation ever broke down into litigation: a university asserting its own unlicensed patent is unlikely to obtain an injunction against an infringer under eBay, which is one more reason the field’s working assumption is that university patent strategy runs through negotiated licensing — not litigation threat — as the primary route to both revenue and technology commercialization.

Common types of non-practicing entities

Type Typical patent origin Typical monetization approach
Universities & research institutions Own faculty/staff research, retained under Bayh-Dole or equivalent Out-licensing (exclusive or non-exclusive) to existing companies or faculty spinouts, generally toward commercialization
Individual inventors Own invention, no manufacturing capacity Licensing to a company able to bring the product to market
Pre-commercialization startups Own R&D, product not yet launched Temporary state — patents held prior to the company itself becoming a practicing entity
Patent aggregators / licensing firms (PAEs) Acquired from third parties, often via bankruptcy asset sales Portfolio licensing and/or litigation against alleged infringers, independent of any product
Failed operating companies Own former R&D, company no longer manufactures Residual patent portfolio sold or asserted post-failure, sometimes to a PAE

Frequently confused questions

Is a university a patent troll?

Not in the way the term is generally used. A university is an NPE in the technical sense — it doesn’t manufacture products — but the FTC’s own PAE study specifically excludes universities from its patent-assertion-entity definition, and the broader literature distinguishes universities from “patent trolls” based on where the patents originate (a university’s own research, versus patents acquired from third parties) and what the licensing is for (commercialization, versus litigation revenue).

Is being an NPE illegal or improper?

No. Patents are property, and licensing or asserting a patent you hold — including one you don’t practice yourself — is a legitimate use of a patent right on its own. What draws the “patent troll” label and the associated policy criticism is a specific pattern of conduct (opportunistic acquisition, low-quality or overly broad claims, mass demand letters, settlement amounts set just below litigation-defense cost), not the mere fact of not manufacturing anything.

What’s the difference between an NPE and a PAE?

NPE is the broad, purely descriptive category: any patent owner that doesn’t commercially practice its own patent. PAE is the FTC’s narrower analytical subset of NPEs — entities that specifically acquire patents from third parties in order to license or assert them — and it is defined to exclude universities, individual inventors, research labs, and companies that developed their own patent portfolios.

Should a TTO avoid ever working with a patent aggregator?

That’s an institutional-policy question, not a legal one, and AUTM’s Nine Points statement frames it as something to weigh carefully rather than an outright prohibition — but doing so does carry the risk of aligning a university’s licensing program with the conduct that draws the “patent troll” comparison, which is worth factoring into any such decision.

Related CASRAI resources

Machine-readable encodings

Use in your systems

JATS XML <role> element
xml
<role vocab="credit"
      vocab-identifier="https://casrai.org/dictionary/"
      vocab-term="Non-Practicing Entity (NPE)"
      vocab-term-identifier="https://casrai.org/dictionary/term/non-practicing-entity-npe" />
Schema.org DefinedTerm (JSON-LD)
json
{
  "@context": "https://schema.org",
  "@type": "DefinedTerm",
  "@id": "https://casrai.org/dictionary/term/non-practicing-entity-npe",
  "name": "Non-Practicing Entity (NPE)",
  "identifier": "https://casrai.org/dictionary/term/non-practicing-entity-npe",
  "description": "An entity that owns one or more issued or pending patents but does not manufacture, sell, or otherwise commercially practice the invention itself, and instead generates value from the patent through licensing, litigation/demand-letter assertion, or both. \"Non-practicing\" describes the entity's relationship to the patent, not the patent's validity -- individual inventors, universities and research institutions, pre-commercialization startups, failed operating companies holding residual portfolios, and dedicated patent-acquisition/licensing firms are all NPEs in this technical sense, even though only a narrower subset of them (the FTC's \"patent assertion entity,\" or PAE, category, and the pejorative \"patent troll\" label) draw the negative connotation the term is often assumed to carry.",
  "inDefinedTermSet": "https://casrai.org/dictionary/domain/compliance-regulatory#set",
  "url": "https://casrai.org/dictionary/term/non-practicing-entity-npe",
  "sameAs": [],
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "publisher": {
    "@id": "https://casrai.org/#organization"
  },
  "dateModified": "2026-08-22T11:57:41",
  "inLanguage": "en"
}

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →