Last updated 3 months ago. Our resources are updated regularly but please keep in mind that links, programs, policies, and contact information do change.
The patent system in the United States represents a fundamental economic contract between the government and the inventor. Rooted in the Constitution’s Article I, Section 8, Clause 8, the system grants inventors the exclusive right to their discoveries for “limited Times.”
This phrase—”limited Times”—is the fulcrum upon which American innovation balances. It creates a temporary monopoly, allowing inventors to charge premium prices and exclude competitors, theoretically to recoup research and development costs. In exchange, the inventor must fully disclose the invention’s details to the public, ensuring that once the patent expires, the knowledge enters the public domain.
However, the definition of an appropriate “limited time” has become fiercely debated. Since 1995, the standard term for a U.S. utility patent has been 20 years from the date of filing. This uniform duration applies regardless of the technology involved—whether it’s a pharmaceutical compound requiring a decade of clinical trials and billions of dollars to develop, or a software algorithm written in a few weeks that may become obsolete just as quickly.
As technological change accelerates and the economic realities of different industries diverge, a growing chorus of economists, legal scholars, and industry stakeholders argues that the “one-size-fits-all” model is broken.
On one side, the pharmaceutical and biotechnology sectors contend that regulatory hurdles eat away at their effective patent life, leaving them insufficient time to earn a return on high-risk investments. On the other, the software and electronics industries argue that 20 years is an eternity in digital time, stifling competition and fueling a litigation industry driven by “patent trolls” who exploit old patents to extract rents from modern innovators.
How Patent Duration Actually Works
To understand the arguments for lengthening or shortening patent terms, you first need to understand the complex machinery that determines how long a patent actually lasts. It’s rarely as simple as “20 years.” The effective life of a patent is shaped by filing dates, administrative delays, maintenance fees, and specific regulatory adjustments.
From 17 to 20 Years
Prior to 1995, the term of a U.S. patent was 17 years from the date the patent was issued by the United States Patent and Trademark Office (USPTO). This system, established in 1861, created a perverse incentive known as the “submarine patent.”
Unscrupulous inventors could file an application and then intentionally delay the examination process—sometimes for decades—by filing endless continuations and amendments. These patents would remain “submerged” in secrecy while an industry developed and matured. Once the technology was widely adopted, the inventor would allow the patent to issue, suddenly springing a fresh 17-year monopoly on a mature market.
The Uruguay Round Agreements Act of 1994, which implemented the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS), fundamentally changed this calculation. Effective June 8, 1995, the U.S. aligned its laws with global standards, changing the patent term to 20 years from the date of filing.
This shift had two primary effects: It eliminated submarine patents (because the clock starts ticking at the filing date, any delay in the examination process eats into the patent holder’s term), and it brought the U.S. into conformity with major trading partners like Europe and Japan.
Different Patents, Different Terms
While the 20-year term applies to utility patents (which cover processes, machines, articles of manufacture, and compositions of matter), other types of patents have different durations:
| Patent Type | Term Duration | Basis of Calculation | Maintenance Fees Required? |
|---|---|---|---|
| Utility Patent | 20 Years | From earliest filing date | Yes (3.5, 7.5, 11.5 years) |
| Design Patent | 15 Years* | From date of issuance | No |
| Plant Patent | 20 Years | From earliest filing date | No |
*For design patents filed before May 13, 2015, the term is 14 years.
Patent Term Adjustment
Because the 20-year clock begins at filing, bureaucratic delays at the USPTO can significantly reduce the “effective” patent life. If the USPTO takes five years to examine and grant a patent, the inventor is left with only 15 years of enforceability.
To address this, Congress enacted provisions for Patent Term Adjustment (PTA). Under 35 U.S.C. § 154(b), the USPTO must guarantee certain timeframes. For instance, if the office fails to issue a first office action within 14 months of filing, or if the overall pendency of the application exceeds three years, the patent term is extended day-for-day to compensate for the delay.
This ensures that the administrative inefficiency of the government doesn’t penalize the inventor. However, this adjustment is reduced if the applicant engages in delays, such as requesting extensions of time to respond to examiner queries.
Maintenance Fees
A utility patent doesn’t automatically last for its full 20-year statutory term. It must be actively maintained through progressively increasing fees. These maintenance fees are due at 3.5, 7.5, and 11.5 years after the patent is granted.
The policy rationale is to clear “dead wood” from the patent system. If an invention is not commercially successful or has become obsolete, the patent holder will likely choose not to pay the fee, allowing the patent to expire early and the technology to enter the public domain.
As of the 2025 USPTO Fee Schedule, these fees have been adjusted to reflect inflation and agency funding needs:
| Payment Interval | Due Date (Post-Grant) | 2025 Fee (Large Entity) | 2025 Fee (Small Entity) |
|---|---|---|---|
| Stage 1 | 3.5 Years | ~$2,150 | ~$860 |
| Stage 2 | 7.5 Years | ~$4,040 | ~$1,616 |
| Stage 3 | 11.5 Years | ~$8,280 | ~$3,312 |
Statistics indicate that roughly 50% of patents lapse before their full term because owners decline to pay these fees. This suggests that for a vast number of inventions, the “optimal” patent term determined by the market is significantly shorter than the 20 years granted by statute.
The Pharmaceutical Argument
The strongest arguments for extending patent duration—or at least modifying how it’s calculated—come from the life sciences sector. In the pharmaceutical and biotechnology industries, the 20-year term is often viewed as illusory.
Unlike a software app that can be launched weeks after coding, a new drug faces a regulatory marathon that consumes a massive portion of its patent life before a single unit is sold.
The Investment and Risk
Developing a new pharmaceutical is an undertaking of immense scale and risk. Industry estimates suggest that the average cost to bring a new drug to market, including the cost of failures, ranges from $1 billion to $2.8 billion. This figure accounts for the high attrition rate in drug discovery. Thousands of compounds are screened and tested for every one that successfully navigates Phase I, II, and III clinical trials to receive FDA approval.
The economic logic of the patent system is “dynamic efficiency”: society tolerates high monopoly prices today to incentivize the investment required for future breakthroughs. Without a guaranteed period of exclusivity to recoup these billions, rational investors would likely shift capital to lower-risk industries, potentially stalling progress on treatments for complex diseases like Alzheimer’s, cancer, and ALS.
The Effective Patent Life Gap
The core problem for pharma is the discrepancy between Nominal Patent Life (20 years) and Effective Patent Life (the time the product is actually on the market).
A pharmaceutical company typically files for a patent early in the research phase to secure priority over competitors. However, the subsequent clinical trials and FDA review process can take 10 to 12 years.
By the time a drug is approved for sale, more than half of its patent term may have already evaporated. While the statutory term is 20 years, the effective exclusivity period for many drugs is often closer to 7 to 11 years. Industry advocates argue that this shortened window forces companies to charge higher prices to recoup costs quickly, or creates a bias against developing drugs that require longer, more complex clinical trials.
The Hatch-Waxman Compromise
To address this imbalance, Congress passed the Drug Price Competition and Patent Term Restoration Act of 1984, commonly known as Hatch-Waxman. This landmark legislation attempted a “grand bargain” between brand-name innovators and generic manufacturers.
Hatch-Waxman allows brand companies to apply for an extension of their patent term to compensate for the time lost during FDA regulatory review. However, this extension is strictly capped: The extension cannot exceed 5 years, the total effective patent life (remaining term + extension) cannot exceed 14 years from the date of FDA approval, and only one patent per drug product can be extended.
Distinct from patents, the FDA grants “regulatory exclusivity” periods. For a New Chemical Entity (NCE), the FDA grants 5 years of data exclusivity. During this time, generic competitors cannot rely on the innovator’s safety and efficacy data to file their own Abbreviated New Drug Application (ANDA). For biologics (large-molecule drugs), the Affordable Care Act established a longer 12-year exclusivity period, reflecting the higher complexity of these products.
Calls for Further Extension
Despite these provisions, some economists and industry groups argue that the current caps are outdated.
As science advances, the “low-hanging fruit” of drug discovery has been picked. Remaining medical challenges are more complex, requiring longer and more expensive trials. This extends the development timeline, further eroding the effective patent term.
There’s a specific push for “ultra-long” patent terms or “transferable exclusivity vouchers” for novel antibiotics. Because new antibiotics must be used sparingly (stewardship) to prevent resistance, sales volumes are low. A standard 20-year term (with only ~10 years of sales) is insufficient to incentivize development of drugs that are meant to be kept on the shelf as a last resort.
Some scholars have proposed extending patent terms in exchange for regulated price caps. A “variable patent term” could offer a longer monopoly duration if the drug is sold at a lower price, theoretically balancing the innovator’s total revenue needs with public access.
The Evergreening Problem
While the pharmaceutical industry argues that patent terms are effectively too short, critics contend that through legal maneuvering, these companies artificially extend their monopolies far beyond the statutory intent. This phenomenon, known as “evergreening,” involves filing dozens or even hundreds of secondary patents on trivial modifications of a drug to block generic entry long after the primary patent has expired.
Patent Thickets
A “patent thicket” is a dense web of overlapping intellectual property rights that a competitor must “hack” through to commercialize a new product. In the pharmaceutical context, this involves filing patents not just on the active ingredient, but on formulations, methods of use, manufacturing processes, and delivery devices.
Because generic companies must certify that they don’t infringe any valid patent listed in the FDA’s “Orange Book,” a thicket of 100 patents creates a formidable barrier. Even if 90 of those patents are weak or invalid, the cost and time required to litigate them (or the risk of launching “at risk”) effectively deter competition.
The Humira Case
The most prominent example of this strategy is AbbVie’s blockbuster biologic Humira (adalimumab). The primary patent on the active antibody expired in 2016. However, AbbVie filed over 247 patent applications related to Humira and obtained over 132 patents.
These secondary patents covered formulation changes, manufacturing steps, and dosing regimens. When biosimilar competitors attempted to enter the market in 2016, they faced this wall of litigation. AbbVie successfully settled with these competitors, delaying their entry into the U.S. market until 2023—seven years after the primary patent expired.
During those extra seven years of monopoly, Humira generated billions in revenue. Critics argue that this cost the U.S. healthcare system significantly, estimating the cost of such thickets to be in the billions annually.
AbbVie and supporters argue that these secondary patents represent genuine innovation in how the drug is made and used, benefiting patients through less painful injections or more stable formulations. They contend that the patent system is designed to protect all inventions, not just the first molecule discovery.
Product Hopping and Pay-for-Delay
Product Hopping occurs when a brand-name company makes a minor modification to a drug (like switching from a tablet to a film strip) shortly before the generic version of the original is set to launch. The company then aggressively switches patients to the new “patent-protected” version and withdraws the old one from the market.
Because pharmacists generally cannot substitute a generic tablet for a brand-name film strip, the generic market is decimated.
Pay-for-Delay settlements involve a brand company paying a generic challenger to not launch their product for a certain number of years. While the Supreme Court has scrutinized these deals, they remain a contentious method of effectively extending patent duration by private contract rather than public law.
Generic Manufacturers Push Back
The Association for Accessible Medicines (AAM) and other generic advocates argue that these practices undermine the Hatch-Waxman balance. They support legislative reforms that would limit patent assertions, restrict “skinny label” attacks, and reform terminal disclaimers.
The USPTO is currently considering rules where if a patent is tied to another via a terminal disclaimer (often used in thickets), invalidating the parent would invalidate the whole family. This is fiercely opposed by pharma but supported by generics.
The Tech Sector’s View
In stark contrast to the pharmaceutical industry, the technology sector—comprising software, electronics, and internet services—often views the 20-year patent term as excessively long. The pace of innovation in these fields is governed not by clinical trials, but by Moore’s Law and agile development cycles.
Innovation Cycles vs. Patent Terms
In the software industry, a product lifecycle may be as short as 18 months. A patent granted today will remain in force until 2045. By that time, the computing landscape will likely be unrecognizable.
Critics argue that granting a 20-year monopoly on a software method is economically inefficient because the “invention” becomes a basic building block for future development almost immediately.
Software development is highly cumulative. New code is built on top of old code. When basic functions—like “slide-to-unlock” or “one-click checkout”—are patented for two decades, it creates a “royalty stack” or “anticommons” where new developers must navigate a minefield of licenses to create even simple applications.
Amazon’s 1-Click Patent, granted in 1999, gave Amazon exclusive rights to the idea of storing customer credentials for single-action purchasing. It prevented competitors from streamlining their checkout processes until 2017. Critics argue this retarded the growth of the entire e-commerce sector for the benefit of one player, protecting a business method that arguably would have been “obvious” to others shortly thereafter.
Patent Trolls
The long tail of the 20-year term is the primary habitat of Non-Practicing Entities (NPEs), or “patent trolls.” These entities acquire patents—often from bankrupt companies—not to produce products, but to sue operating companies for infringement.
Empirical studies show a distinct pattern: operating companies (like Apple or IBM) tend to litigate their patents early in the term to protect market share. In contrast, NPEs overwhelmingly litigate in the last few years of the patent term.
Why wait? Trolls wait for a technology to become ubiquitous and for companies to become “locked in.” If a patent covers a widely used standard (like WiFi or JPEG), asserting it in year 18 allows the troll to demand royalties from the entire industry.
Startups are particularly vulnerable, as they lack the legal war chests to defend against infringement claims. Settlements drain capital that would otherwise go to hiring or R&D.
If patent terms for software were reduced to, say, 5 or 7 years, the business model of the patent troll would be severely undermined. The window for asserting vague, old patents against modern technology would close before the technology becomes an industry standard.
The Smartphone Wars
The “smartphone wars” of the early 2010s illustrated the chaos of long patent terms in tech. Apple and Samsung engaged in a global legal battle over patents covering features like “slide-to-unlock,” “pinch-to-zoom,” and the rectangular shape of the device.
These battles cost hundreds of millions in legal fees. While they eventually settled, they highlighted how patents on interface designs could be weaponized to block competitors.
The term for design patents (covering the look of a device) was actually extended from 14 to 15 years in 2015. Tech critics argue this moves in the wrong direction, as consumer aesthetic trends cycle far faster than 15 years.
Economic Theory
The debate over patent duration is not just a clash of industries. It’s a fundamental question of economic theory. Economists define the “optimal” patent life as the point where the marginal benefit of incentivizing innovation equals the marginal cost of monopoly deadweight loss.
Static vs. Dynamic Efficiency
Dynamic Efficiency (Incentive to Create) favors longer terms. The promise of future monopoly profits encourages firms to invest in R&D today. If the term is too short, firms won’t invent.
Static Efficiency (Access to Goods) favors shorter terms. Once an invention exists, social welfare is maximized by distributing it at the marginal cost of production (which is low for drugs and near-zero for software). Patents artificially inflate prices, preventing some consumers from accessing the good (deadweight loss).
Economist William Nordhaus modeled this tradeoff in the 1960s. He concluded that the optimal patent life depends on the elasticity of demand and the elasticity of innovation with respect to R&D spending.
If a small increase in patent life yields a huge burst of innovation (high innovation elasticity), terms should be long. If innovation is largely driven by other factors (first-mover advantage, academic curiosity) and demand is sensitive to price, terms should be short.
The Uniformity Problem
The fundamental economic flaw in the U.S. system is that it applies a single duration (20 years) to all industries. Nordhaus’s model suggests this is inevitably inefficient.
Pharmaceuticals: High R&D costs and high risk suggest a longer optimal term (perhaps 20-30 years).
Software: Low reproduction costs and cumulative innovation suggest a much shorter optimal term (perhaps 3-5 years).
Because the TRIPS agreement mandates a minimum 20-year term for all fields of technology, the U.S. cannot easily adopt variable statutory terms without violating international law. This forces policymakers to look for other levers—like maintenance fees or regulatory exclusivity—to mimic the effects of variable terms.
2025 Legislative Outlook
As of 2025, the debate over patent duration and quality has coalesced into several major legislative and regulatory initiatives. These proposals don’t explicitly change the “20 year” number but aim to alter the effective strength and duration of patents.
PERA and PREVAIL Acts
Two major bills are currently winding through Congress:
Patent Eligibility Restoration Act (PERA): This bill aims to expand the types of inventions eligible for patenting, particularly in diagnostics and software, effectively overturning Supreme Court decisions that restricted them. Proponents argue this is necessary to secure U.S. leadership in AI and biotech. Critics argue it will unleash a new wave of patent trolls and thickets by allowing abstract ideas to be patented again.
PREVAIL Act: This bill focuses on the Patent Trial and Appeal Board (PTAB). It seeks to limit the ability of third parties to challenge the validity of patents at the USPTO. By raising the standard of proof required to invalidate a patent and limiting standing, PREVAIL would make issued patents harder to kill.
Supporters (Pharma/Inventors) argue that a “quiet title” is necessary for investment security. If a patent can be easily invalidated at the PTAB, the 20-year promise is worthless.
Opponents (Tech/Consumer Groups) argue that the PTAB is a vital tool for cleaning up “bad patents” that shouldn’t have been granted. They fear PREVAIL will cement patent thickets and keep drug prices high by protecting weak secondary patents.
Terminal Disclaimer Reform
The USPTO has proposed a significant rule change regarding Terminal Disclaimers. Currently, companies file “continuation” patents that are slight variations of a parent patent. To get around the rule against “double patenting,” they file a terminal disclaimer, tying the expiration of the new patent to the old one.
The new rule would state that if the parent patent is invalidated, all patents tied to it by terminal disclaimer fall with it. This would destroy the “hydra” nature of patent thickets. A challenger would only need to invalidate one key patent to topple the entire thicket. This effectively weakens the “evergreening” strategy without changing the statutory term.
Maintenance Fees as Policy
With the 2025 Fee Schedule increasing maintenance fees, there’s growing discussion about using these fees more aggressively. Some economists propose drastically hiking fees in the later years (like Year 15 or 19) to force patent holders to release low-value patents.
A “Year 19” maintenance fee of $50,000 would make a pharma company with a blockbuster drug pay instantly, while a patent troll holding a vague software patent with uncertain litigation value might let it expire. This creates a de facto variable term system based on private market valuation.
International Models
Some reformers point to the “Utility Model” system used in China and Germany as a template. These are “mini-patents” with a shorter term (10 years) and no substantive examination.
China uses these aggressively for incremental innovations. Adopting a similar tier for software or minor improvements could allow for protection that matches the speed of the industry without locking up ideas for 20 years. However, this would require a major overhaul of the U.S. Patent Act.
The Split Answer
The question “Should patent duration be longer or shorter?” ultimately yields a split answer: It depends on the industry.
For the pharmaceutical sector, the effective patent life is arguably squeezed by regulation, threatening the investment model for the hardest diseases. However, the industry’s response—building thickets and evergreening—has swung the pendulum too far, creating indefinite monopolies that harm public access.
The solution likely lies not in extending the 20-year term, but in reforming the quality of secondary patents and perhaps extending regulatory exclusivity for truly novel drugs (like antibiotics) while cracking down on thickets for older ones.
For the technology sector, the 20-year term is almost certainly too long. It’s a relic of the industrial age applied to the information age. While international treaties make shortening the term difficult, the U.S. is moving toward reforms that reduce the enforceability of these patents (via the PTAB or fee structures) rather than the statutory number.
As 2025 progresses, the fate of the PERA and PREVAIL Acts, along with USPTO rule changes, will decide whether the U.S. patent system tilts toward stronger, longer-lasting property rights or toward a more fluid, competitive commons. The “limited times” envisioned by the Founders remain in the text, but the struggle to define those limits in a digital and biological world continues.
Our articles make government information more accessible. Please consult a qualified professional for financial, legal, or health advice specific to your circumstances.