chevron-down Created with Sketch Beta.
January 11, 2023 Feature

The Right to Repair, Competition, and Intellectual Property

Michael A. Carrier

©2022. Published in Landslide, Vol. 15, No. 2, December/January 2023, by the American Bar Association. Reproduced with permission. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association or the copyright holder.

The right to repair products is critical. But in recent years, manufacturers have made it extremely difficult to do so. They have made parts unavailable, prevented products from working, restricted access to service manuals, and relied on intellectual property (IP), software restrictions, and licenses.

As discussed elsewhere, repair restrictions have affected medical ventilators, military equipment, technological devices, and even wheelchairs. This article focuses on agricultural equipment, in particular, tractors.

A Case Study: Tractors

The casual observer would be surprised by the role software plays in farm equipment. Today’s John Deere tractors “cost as much as $800,000” and “depend on multiple electronic control units (ECUs) to operate everything from the engine to the power seat.” The problem is that when any of these parts fail, the machine goes into “limp mode,” which “disables most of the equipment’s functionality” until “it is repaired and the error codes are cleared.”

Software erects a hurdle in the form of copyright law. The Digital Millennium Copyright Act (DMCA) punishes conduct that circumvents technological protection measures (TPMs) protecting copyrighted works, which include the software in tractors. The DMCA created a “triennial exemption process” that allows the Librarian of Copyrights to grant exemptions every three years for certain classes of works. Since 2015, there has been an exemption for motorized land vehicles, which includes tractors.

While this exemption is helpful for users able to circumvent the TPMs, it does not cover the trafficking of such tools, which prevents those not handy enough to fix the products themselves from benefiting from the exemption. In addition, just after the exemption was first granted in 2015, John Deere “started requiring farmers to sign licensing agreements.” These licenses compel users to agree that the licensed material is “protected under copyright law, trade secret law, and laws governing confidential information” and that they will not “‘modify,’ ‘reverse engineer,’ or ‘reproduce’ the covered information” even though these are “necessary steps to understanding, repairing, and improving [the] equipment.”

These restrictions’ significant effects are exacerbated given the sensitivity of timing. The finite duration of harvesting seasons and idiosyncrasies of weather raise the stakes for each day farmers are not able to use their equipment. Just to give one example, a Nebraska farmer “lost half a day of harvesting corn while waiting for mechanics to drive 65 miles to his farm,” with the wait “contribut[ing] to a loss of at least 15% of the crop.”

The role of software and importance of timing increase the leverage of manufacturers, in particular market leader John Deere. In the market for large farm tractors in 2018, for example, John Deere had 53% market share, with most of the rest of the market taken by CNH Industrial (35%) and AGCO (7%). Deere’s power is buttressed by competitors using similar restraints, which apply “across the board.” This power is strengthened even more by dealer consolidation. Of Deere’s dealerships, 82% “are a part of a large chain with seven or more locations.” As a result, “some farmers only have one dealership choice near them,” which can “force them to travel long distances and cross state lines to get another quote from a dealer they might trust more.”

In short, the combination of ubiquitous software, time-sensitive applications, competitors’ use of similar restraints, and dealer consolidation suggests that John Deere effectively has market power.

Competition Law

Because of the competitive effects of repair restrictions like those imposed by Deere, competition law would seem to offer a natural antidote. But it has not played the robust role it could. The reason is that the primary instrument for effectuating competition—antitrust law—has erected nearly impossible hurdles in front of plaintiffs making right-to-repair claims.

First, an essential facility claim requires a monopolist to share facilities necessary to compete in a market. But in part because these claims could be interpreted to cover a wide array of products, courts almost never allow them to proceed.

Second, courts have narrowly construed refusals to license, especially when they involve IP. For example, in In re Independent Service Organizations Antitrust Litigation (Xerox), the Federal Circuit held that a refusal to sell patented parts did not exceed “the scope of the patent grant” and thus could not violate antitrust law.

Third, tying claims require a plaintiff to show several factors: (1) two separate products, (2) coercion, (3) market power in the market for the “tying product” (the one the consumer desires), and (4) a not insubstantial amount of commerce in the market for the “tied product” (the one the consumer is forced to take). In particular, courts have made it difficult for plaintiffs to demonstrate coercion.

Finally, courts have hamstrung the most relevant doctrine. In Eastman Kodak Co. v. Image Technical Services, the U.S. Supreme Court recognized that a manufacturer of equipment in a competitive “primary” market could have monopoly power in an “aftermarket” for service and parts. (Deere’s 53% market share in the primary market likely would not reach the 70% required for monopoly power. ) Aftermarket power could result from customers being “locked in” to the manufacturer’s product because of high switching costs, thereby weakening the significance of the primary market. This theory could be a natural fit for claims challenging a manufacturer’s repair restrictions.

But courts soon “severely limited Kodak’s scope so that it is no longer a viable weapon for antitrust plaintiffs” and “bent over backwards to construe Kodak as narrowly as possible.” Most notably, several courts required plaintiffs to show that defendants changed their service policies “to exploit the installed base of consumers.” For example, the Sixth Circuit stated that “the change in policy in Kodak was the crucial factor in the Court’s decision.”

FTC Act Section 5

Even though courts have limited antitrust law’s ability to address right-to-repair claims, there should be an available competition-based claim. Into this breach steps Section 5 of the Federal Trade Commission (FTC) Act, which allows the FTC to challenge “unfair methods of competition.” Section 5 can reach expansively to target conduct lying outside antitrust’s scope. While such an approach could be applied in a way that has few limits, the framework (summarized here) is designed to modestly expand Kodak in a predictable manner that is consistent with the decision’s rationale.

In particular, the application of Section 5 discussed here builds on the “gap filler” rationale that Susan Creighton and Thomas Krattenmaker proposed for settings in which one of the elements of an antitrust claim is not satisfied. The element targeted here is market power, as courts’ unwillingness to find this factor satisfied could be addressed by practical indicators of market power based on consumers’ lack of choice. Given developments in the past several years, the rationales underlying courts’ and commentators’ skepticism of market power in aftermarkets immediately after Kodak is less relevant today.

Market Power

This Section 5 gap filler applies to five scenarios that functionally prevent consumers from having a choice.

First, multiple manufacturers could impose similarly restrictive terms. As a result, even if the manufacturer with whom the customer is dealing does not have significant market power, the use by other firms of analogous terms could effectively prevent choice.

Second, a manufacturer could have control over a separate level of the distribution chain needed to service the product. Farmers using John Deere tractors need to obtain service at certified Deere dealers. And such dealers may be the only repair facilities in the vicinity. For example, in Montana’s “58 million acres of farmland,” there are “only three large John Deere chains with a combined 19 locations serving Montana farms.”

Third, today’s range of more intrusive restrictions leads to users lacking knowledge. One of the main strands underlying hostility to Kodak aftermarket claims is a purchaser’s ability to consider the policies of a single manufacturer, engage in “lifecycle pricing” that considers these costs, and make comparisons with rivals. But today’s use of not only simple parts and servicing policies but also a range of more hidden restrictions makes this virtually impossible.

Fourth is revealed market power over time. Despite the presence of suffocating policies threatening lives and livelihoods, multiple generations of purchasers are still buying the products. The inability to switch away from such restrictive policies provides an indication of equipment owners’ sustained market power.

Fifth is the importance of time-sensitive uses. In the agricultural setting discussed above, delay has dramatic consequences. Farmers suffering equipment breakdowns when a tractor is needed to harvest crops face timing constraints that increase the manufacturer’s market power as users do not have the luxury of deliberately looking to other manufacturers in a theoretically competitive primary market for alternatives.

Anticompetitive Effects

Once market power—or its functional equivalent—is shown, the analysis should consider anticompetitive and procompetitive effects. First, there are uniquely severe anticompetitive effects. Antitrust’s typical anticompetitive effects have been higher prices and lower output, and to a lesser extent, reduced innovation and quality. The first two effects are relatively easy to measure, and (where significant enough) make consumers’ lives worse. But as concerning as these effects are, there’s a whole level of harm higher than that—in fact, significantly higher. Consumers are not “just” experiencing an anticompetitive market with higher prices, but also suffering direct effects on their lives and livelihoods. For example, farmers are not able to fix machines that are their literal livelihood. If they cannot use their tractor, they will not be able to harvest their crops. And as explored elsewhere, the ability to repair broken medical ventilators, military equipment stranded on a battlefield, and wheelchairs needed for mobility affects lives. These effects present a compelling argument for the FTC to use Section 5.

Intellectual Property

Manufacturers have claimed procompetitive justifications for their repair restrictions. Of several rationales they have offered, the focus here is IP. This section briefly discusses design patents, trademarks, and trade secrets before exploring in detail manufacturers’ primary justification: copyrights.

Design Patents

Design patents protect “new, original and ornamental design[s].” Manufacturers have frequently obtained design patents in the automobile industry. Ford, for example, has claimed that its “designers create the appearance of headlamps, hoods and other parts to appeal aesthetically to customers”; that “[g]iven the importance of vehicle design, [the company] invests heavily in design and protects some of its artistic products through design patents”; and that a “knock-off business model free-rides off [its] investment and creativity.” Design patent protection makes sense where “the design . . . make[s] some type of material aesthetic contribution to the art,” having “some visual content that actually matters to consumers of the relevant product.” But it does not seem appropriate for the “internal parts of a product, which no one buys for their appearance,” implicated in repair settings. As Sarah Burstein has explained: “internal, mechanical parts are going to be created regardless of whether design patent protection is available,” “[t]he public gains nothing by protecting them,” “patenting such designs raises serious concerns related to circumvention of the utility patent system,” and “[p]roviding separate protection for spare parts . . . provides a windfall to the . . . manufacturer.”


Manufacturers have used trademarks to block “the importation of replacement parts.” To do this, they have placed trademarks on “internal parts like batteries, processors, and cables” that users never see and “logos . . . no bigger than a grain of rice.” The setting in which these issues arise—counterfeiting and importation of purportedly trademark-protected goods—is one where defenses are less likely to be fully considered. But such use is at odds with trademark law and policy in several ways.

First, it is not consistent with trademark law’s purpose, which is to prevent customer confusion: consumers will be aware of the fact that an independent servicer—which it chose—repaired their products. Second, under trademark law, repairers are allowed to refurbish parts as long as they do not “deceive the public.” Third, manufacturers have targeted independent repairers’ use of the manufacturers’ trademarks even though—in referring to the plaintiff’s product itself—repairers’ use of the trademark will typically be justified based on the doctrine of “nominative use.” Finally, the functionality defense would prohibit enforcement of a trademark embedded in a system necessary for the device to work.

Trade Secrets

Manufacturers also have sought to block independent repair organizations by using trade secrets. They have claimed that allowing these groups to service their products would “increase[] the likelihood of trade secrets becoming public knowledge” and “place [original equipment manufacturers], suppliers, [and] distributor and repair networks at risk.” Despite these claims, three doctrines should prevent manufacturers from being successful. First, unlike information that gives an advantage over competitors, repair information does not derive independent economic value from being secret. Second, protection does not apply when independent repairers can discover the information legally through reverse engineering.

And third, when information “is readily shared with authorized dealers (and their repair personnel) all over the country,” the owner may not have engaged in reasonable efforts to maintain secrecy. This could be the case “even where manufacturers have entered into confidentiality agreements with their authorized dealers” because the dealers’ repair personnel may not have “entered into similar agreements with their employers.” In fact, the repair information could be so widely available and generally known that it is not considered a secret at all.


In the realm of copyright law, manufacturers of video games and gaming consoles, to pick one example, have asserted that “repair restrictions in the form of [TPMs] are needed to protect video games from being pirated.” In particular, they contend that “‘some game console repairs may require replacing hardware components or parts of components, and some of these hardware fixes may require’ circumvention of a console’s anti-piracy TPMs.”

For several reasons, the copyright justification is not compelling in this setting. First, any reasonable assessment of the relationship between IP and competition law makes clear that IP rights are not absolute. The Supreme Court in FTC v. Actavis confirmed its decades-long approach of applying antitrust scrutiny to IP-based conduct. The Court held that it “would be incongruous to determine antitrust legality by measuring [a] settlement’s anticompetitive effects solely against patent law policy, rather than by measuring them against procompetitive antitrust policies as well.” Citing cases going back to 1926, the Court explained that it “has struck down overly restrictive patent licensing agreements—irrespective of whether those agreements produced supra-patent-permitted revenues.”

To similar effect, the D.C. Circuit in United States v. Microsoft rejected Microsoft’s assertion that the license restrictions it imposed on original equipment manufacturers were justified as the “exercis[e of] its rights as the holder of valid copyrights.” The court explained that Microsoft “claims an absolute and unfettered right to use its intellectual property as it wishes,” but “[t]hat is no more correct than the proposition that use of one’s personal property, such as a baseball bat, cannot give rise to tort liability.” The court instead cited the long-standing proposition that “[i]ntellectual property rights do not confer a privilege to violate the antitrust laws.”

Second, Section 5 provides leeway to consider policy considerations not as directly relevant in the case law. The most fundamental question in copyright law is how to assess the trade-off between incentives and access. In the right-to-repair setting, access concerns should be paramount. For starters, manufacturers have never shown that they need to control the market for service and parts to incentivize the creation of products with protectable expression. Nor would they need to do so to be motivated to service or provide parts for faulty products because—in addition to the lack of connection with protecting expression—the need to fix the products is reason enough. This favoring of access gains support from courts’ treatment of reverse engineering as fair use. Because courts have found that using reverse engineering to create a competing system is fair use, the lesser step of repairing the system itself also would be. Finally, users expect the right to repair their product, and innovation relies in significant part on users’ contributions—the “user innovation” highlighted by Eric von Hippel.

Third, copyright doctrine supports the right to repair. The “first sale” doctrine in § 109 of the Copyright Act “allows those who have acquired products . . . considerable freedom to use, modify, and resell those products as they wish, even if the products are protected . . . by IP rights.” As Pamela Samuelson has explained, this right “serves many positive functions,” including “promot[ing] broader public access to products,” “enabl[ing] preservation of products,” “protect[ing] privacy and autonomy,” and fostering “more innovation.”

Additionally, § 117(c) allows the copying of computer programs for “maintenance or repair” of a machine. The drafters explained that “[w]hen a computer is activated, certain software or parts thereof is automatically copied into the machine’s random access memory, or ‘RAM.’” Court holdings that such copying is a “reproduction” reserved to the copyright owner “call[] into question the right of an independent service provider . . . to even activate” a computer “for the purpose of servicing the hardware components.” Section 117(c) serves a purpose that the drafters believed was “important” and supports repair.

Moreover, even though copyright law—in particular, the DMCA—makes unlawful the circumvention of TPMs that prevent access to copyrighted works, the legislative history and case law suggest that it was not intended to be applied in right-to-repair settings. The Register of Copyrights found that the exemption for tractors was warranted because “facilitating diagnosis, repair, and modification of vehicles may constitute a noninfringing activity as a matter of fair use,” § 117, or both.

In enacting the DMCA, Congress was concerned that copyright owners would not “make their works . . . available on the Internet” because of the “massive piracy” resulting from the “ease with which digital works [could] be copied and distributed worldwide virtually instantaneously.” The legislation was designed to encourage the availability of the “movies, music, software, and literary works that are the fruit of American creative genius.”

The legislative history makes clear that the software in functional devices was not the intended target. Worried about potential abuse, the drafters crafted an interoperability exemption so that computer programs could exchange information. They explained that “manufacturers, consumers, retailers, and professional servicers should not be prevented from correcting an interoperability problem . . . resulting from a technological measure.”

Manufacturers have used copyright to target the use of not only software but also service manuals. Just to pick one example, Toshiba sent a cease and desist letter to an individual who “distribut[ed], by download, copyright[ed] repair manuals . . . that are proprietary.” Copyright’s originality standard is low, and courts have found that “manuals can possess sufficient originality to allow copyright protection, thin as it may be.”

But a service manual, which “contain[s] useful information for diagnosing and repairing . . . common failures,” is largely factual in nature. No one is interested in the manual because they are looking for flowery prose or creative expression. Absent access to the manual, the machine cannot be fixed. This seems to violate the foundational idea-expression dichotomy. And the manual’s factual, functional, non-market-displacing use seems to present a quintessential example of fair use. Nor are incentives needed to create service manuals, which manufacturers must offer for their products.


In short, restrictions on the right to repair can present severe anticompetitive effects. The policy considerations relevant to Section 5 raise questions about whether this conduct can be justified by IP. It is time for competition law to address the right to repair.

    The material in all ABA publications is copyrighted and may be reprinted by permission only. Request reprint permission here.

    Michael A. Carrier

    Rutgers Law School

    Michael A. Carrier is a Distinguished Professor at Rutgers Law School whose scholarship focuses on the intersection of IP and antitrust law. This article is adapted from the author’s article “How the Federal Trade Commission Can Use Section 5 to Strengthen the Right to Repair,” 36 Berkeley Tech. L.J. (forthcoming 2023).