Center for Advanced Study & Research on Innovation Policy


CASRIP Newsletter - Summer 2007, Volume 14, Issue 3

Patent Exhaustion and the Recycling Business in the United States and Japan

By Hirohito Nakada[*]


Recycle or reuse of goods and resources ever increasing promotion, support and acceptance and consumers are accepting recycled products, which often are less expensive than new or unused products.[1] One such example is in the resale of printing toner, a market which has expanded due to the ubiquity of personal computers.[2] Indeed, governments and companies are now considering the protection of the environment to such an extent that environmental protection can truly be classed as a global concern.[3] Given this social concern, the recycling of its own products and byproducts is becoming a responsibility of manufacturers.[4] In light of the importance of the emerging recycling businesses, manufacturers seek to profit either from the first sale of their products, the recycled product market, or prohibition of recycled products in order to promote the first sale of new products. The market strategy employed by printer manufactures involved reducing the price of the printer itself and gaining profit from the sale of toner or ink. Then, printer manufacturers tried to discourage the recycled products in order to benefit from the aforementioned market strategy.[5] Given antitrust regulations and the recent tendency of the tight application thereof, printer manufacturers and other manufacturers in different industries now intend to achieve shutout of recycled products by obtaining patents covering their products.[6]

Patented products are monopolized by the patentee. However, the patent can be exhausted at the first sale of the patented product under the “first sale doctrine”. This first sale doctrine is recognized in many countries. Based on this theory, consumers of the patented products can legitimately sell the consumed products to recyclers. The recyclers who own the consumed products are free to use, resell, or otherwise dispose of the consumed products. However, making a recycled product using the consumed products would be considered differently.

In Japan, on January 31, 2006, the Intellectual Property (“IP”) High Court rendered a decision regarding recycled ink tanks.[7] The IP High Court said that the patent does not exhaust under a certain situation and a recycled product manufacturer is prohibited to produce recycled product by using the consumed product. However, in the United States, such recycling is considered to be legitimate and we can find recycled ink tanks or ink tank refilling services in many retail stores.[8] [d1]

In this paper, I will discuss which policy is the better one, the intellectual property law that prevents the recycling of the patented products as in Japan, or the intellectual property law that does not hinder recycling as in the United States. Part II of this paper explains the policy of the United States courts regarding the recycling, and part III overviews that of Japanese courts. In part IV, I will compare these policies and weigh in favor of current American policy. Also, I will propose the policy which the American courts should follow. It should be noted that this paper discusses only domestic transactions, and does not include international exhaustion problems.


A. Basics of Patent Exhaustion Doctrine in the United States

Patent exhaustion theory in the United States, often referred to as the first sale doctrine, is explained as follows: “[a]n authorized sale of a patented product exhausts the patent monopoly as to that product. Thus, a purchaser of such a product from the patent owner or one licensed by the patent owner may use or resell the product free of control or conditions imposed by the patent owner.”[9] “The theory behind this rule is that in such a transaction, the patentee has bargained for, and received, an amount equal to the full value of the goods.”[10] The Supreme Court has recognized this patent exhaustion doctrine from the middle of the 19th century, for example, Adams v. Burke.[11] In addition to the patent exhaustion doctrine, the United States courts often mention the implied license. But the implied license is closely related to the patent exhaustion doctrine[12] and in certain situations, the patent exhaustion doctrine and the implied license are two alternative explanations for the same thing.[13]

With respect to recycling, the American courts often apply the repair-reconstruction dichotomy. This dichotomy derives from the patent exhaustion doctrine or the implied license.[14] As stated above, a purchaser of a patented product from an authorized dealer has the right to use the product; and this right includes a right to make any repairs on the product necessary for continued use.[15] On the other hand, the right of repair does not include a right to reproduce the product.[16] Such reproduction will constitute an infringement upon the patent owner’s right to exclude others from making the patent invention.[17] However, drawing a line between permissible “repair” and impermissible “reconstruction” is difficult and many cases address this issue.[18] Within the next paragraph, I will provide an overview of several important cases.

B. Cases Relating to Recycling

1. Conditional Sale

First, it should be noted that patent exhaustion does not apply to an expressly conditional sale in the United States.[19] For instance, there are some cases that analyzed the validity of the “single use only” limitation posed by the patentee.

The United States court basically permits a patentee to control the patent exhaustion by contract. In Mallinckrodt, Inc. v. Medipart Inc. (“Mallinckrodt”),[20] Mallinckrodt manufactured a patented device which delivered radioactive or therapeutic material in aerosol mist form to the lungs of a patient. The device fit into a lead-shielded container that was provided by Mallinckrodt to minimize exposure to radiation and for safe disposal after use. The device bore the inscription “Single Use Only.” In spite of the inscription, the hospitals shipped the used manifold/nebulizer assemblies to Medipart. After the exposure to gamma radiation, the “reconditioned” unites were shipped back to the hospitals from whence they came. The Federal Circuit allowed Mallinckrodt to condition the sale of the patented product by stating “[t]he appropriate criterion [for the judgment of validity of the restriction on reuse] [was] whether Mallinckrodt’s restriction [was] reasonably within the patent grant, or whether the patentee [had] ventured beyond the patent grant and into behavior having an anticompetitive effect not justifiable under the rule of reason”[21] and denied the patent exhaustion in the case of a conditional sale.[22] The Federal Circuit further stated that “[i]f the sale of the UltraVent was validly conditioned under the applicable law such as the law governing sales and licenses, and if the restriction on reuse was within the scope of the patent grant or otherwise justified, then violation of the restriction may be remedied by action for patent infringement.”[23] The Federal Circuit did not enter into the repair-reconstruction dichotomy, because “[the reconstruction-repair distinction] is dispositive if it is determined that the sale of the UltraVent device was accompanied by a valid restriction to single patient use, for ‘the traditional rule’ is that even repair of an unlicensed device constitutes infringement.”[24]

However, the court does not allow a patentee to restrain the patent exhaustion by unilateral intention of the patentee. In Jazz Photo Corp. v. International Trade Commission (“Jazz Photo”),[25] Fuji had a patent relating to “single-use” cameras and Jazz Photo imported a camera which it manufactured by recycling the body of Fuji’s camera. The Fuji camera was in a box and each of the boxes and outer cardboard covers of the cameras had statements instructing the purchaser to not remove the film and return the camera to the photo processor. The Federal Circuit, finding that “there was no express contractual undertaking by the purchaser,”[26] denied the validity of the restraint, because there was no showing of a “meeting of the minds.”[27]

2. Repair-Reconstruction Dichotomy

The Supreme Court first addressed the repair-reconstruction issue in Wilson v. Simpson (“Wilson”).[28] In Wilson, a patented machine used replaceable cutter-knives that lasted sixty to ninety days while the machine’s frame lasted many years.[29] The Court held that the defendant’s replacement of worn cutter did not constitute a reconstruction of the invention, because without such replacement “his purchase would have been useless to him, except for sixty or ninety days after a machine had been put in use.”[30]

One of the most important cases regarding the repair-reconstruction dichotomy is Aro Manufacturing Co. v. Convertible Top Replacement Co. (“Aro I”).[31] In Aro I, Aro owned a patent on a convertible, folding automobile top, which included a flexible fabric top, the support structures, and a sealing mechanism.[32] The useful life of the fabric portion was about three years, whereas the other components usually lasted the lifetime of the automobile. The defendant manufactured and sold replacement fabric tops for use exclusively with Aro’s patented combination. The Supreme Court held that “the replacement of the fabric involved in this case must be characterized as permissible ‘repair,’ not ‘reconstruction.’”[33] The Supreme Court also addressed that “reconstruction of a patented entity . . . is limited to such a true reconstruction of the entity as to ‘in fact make a new article,’ after the entity, viewed as a whole, has become spent.”[34] In addition, the Supreme Court rejected the factors that lower courts had used to distinguish repair from reconstruction, such as the relative life, cost and importance of the individual parts.[35] Also, the Court did not adopt Justice Brennan’s test which included the following factors: “the life of the part replaced in relation to the useful life of the whole combination, the importance of the replaced element to the inventive concept, the cost of the component relative to the cost of the combination, the common sense understanding and intention of the patent owner and the buyer of the combination as to its perishable components, whether the purchased component replaces a worn-out part or is brought for some other purpose, and other pertinent factors.”[36]

This Supreme Court decision in the Aro I case was later followed by lower courts decisions. Among them, Jazz Photo[37] handled the recycling of single-use cameras, as introduced above. The Federal Circuit concluded that the recycling is basically permissible repair, depending on how the actual recycling is done, stating “[a]s discussed in Aro Manufacturing, the replacement of unpatented parts, having a shorter life than is available from the combination as a whole, is characteristic of repair, not reconstruction.”[38] The court further stated that “[i]f the claimed component is not replaced, but simply is reused, this component is neither repaired nor reconstructed.”[39]

Another interesting case is Husky Injection Molding Systems Ltd. v. R & D Tool Engineering Co. (“Husky”),[40] in which the Federal Circuit stated three categories of repair-reconstruction situations. In Husky, Husky manufactured and sold injection molding systems. Customers changed the performance design on average after three to five years by buying substitute molds and corresponding carrier plates, which were not separately patented. R & D made molds and carrier plates, which were substituted for components of Husky’s injection molding system. The Federal Circuit identified three primary repair and reconstruction situations.[41] First, there is the situation in which the entire patented item is spent, and the alleged infringer reconstructs it to make it useable again. The Federal Circuit held that this would constitute a practical reconstruction.[42] Second, there is the situation in which a spent part is replaced. The Federal Circuit cited Wilson and Aro I and stated that this would constitute a permissible repair. Moreover, the Federal Circuit expanded this category by stating “[i]n Sage Products, we held that replacement was not limited to worn out articles, but also included articled that were effectively spent.”[43] Third, there is the situation in which a part is not spent but is replaced to enable the machine to perform a different function. The Federal Circuit said that this is a situation of “kin to repair.”[44] Then, the Federal Circuit found no infringement, because the carrier plates were readily replaceable, and the carrier plate is just one element of the patented combination and not separately patented.[45]

In the case that the parts are not readily replaceable,[46] the court found an infringement in Sandvik Aktiebolag v. E.J. Co. (“Aktiebolag”).[47] In Aktiebolag, Sandvik manufactured a commercial embodiment of the patented drill. The drill tip, which was not separately patented, dulled and required resharpening. Sandvik contends that it never intended for the drills to be retipped. E.J. retipped when the tip could not be sharpened because it had chipped, cracked or simply worn down after being resharpened several times. The court held that “[u]nder the totality of circumstances, we hold in this case that E.J.’s actions are a reconstruction.”[48] The court further held that “[t]here are a number of factors to consider in determining whether a defendant has made a new article, after the device has become spent, including the nature of the actions by the defendant, the nature of the device and how it is designed (namely, whether one of the components of the patented combination has a shorter useful life than the whole), whether a market has developed to manufacture or service the part at issue and objective evidence of the intent of the patentee.”[49]

C. Interim

Under the United States law, a patentee can control the patent exhaustion by contract, but cannot control the patent exhaustion by its unilateral intention.

In the case of unconditional sale, despite the significant number of cases dealing with repair-reconstruction issues, the distinction between “permissible repair” and “impermissible reconstruction” is unclear. Indeed, the courts abandoned all efforts to cabin the repair-reconstruction dichotomy within a rigid framework of rules.[50] However, given the cases above, it could be generally said that (i) if a part of a patented product is designed to be replaceable, (ii) if such a part itself is not an object of a patent, and (iii) if the part cannot be used because it is worn or broken, the customer who legitimately bought the patented product can replace the relevant part.[51] This rule makes a useful guidance, but it does not help to establish the line between permissible repair and impermissible reconstruction. Actually, the court admitted that “permissible ‘repair’ also includes replacement of parts that are neither broken nor worn.”[52]


A. Basics of Patent Exhaustion Doctrine in Japan

Patent exhaustion doctrine is explained in Japan as follows: “when the right holder legally distributes a relevant product, the patent has already achieved its purpose regarding that relevant product, the patent exhausts.”[53] The Supreme Court recognized this theory regarding domestic transactions in BBS Kraftfahrzeugtechnik AG v. K.K. Racimex Japan (“BBS”).[54] In BBS, the Supreme Court mentioned two reasons for recognition of patent exhaustion theory: (i) “if the approval of the patentee is required whenever the patented product is assigned or otherwise disposed of, the free distribution of products on the market is hindered,” and (ii) “it is not necessary to let the patentee gain double profit in the stream of commerce.”[55]

With regard to the recycling, there is no Supreme Court case decision. In light of the above, the legal doctrine which governs this area should be sought from lower court cases, which I will review in the following paragraph.

B. Cases Relating to Recycling

1. Conditional Sale

As opposed to the American law, a patentee cannot control patent exhaustion by contract under Japanese law. In Canon Inc. v. Recycle Assist Co. (“Canon”),[56] the IP High Court stated that “the patent exhaustion should not be hindered by intention of the patentee, because the patent exhaustion doctrine is recognized from the view point of the harmonization of the protection of invention under the Patent Act and the social public interest (reference is made to the BBS Supreme Court decision).”[57] Thus, the patent exhaustion is judged objectively from the view point of publicly accepted sense.

2. Cases before Canon Case

Before the Canon case, Japanese courts roughly took two approaches with regard to recycling: one was the “make approach,” the other was the “exhaustion approach.”[58] Under the make approach, courts considered whether alteration to a patented product constitutes permissible “repairing” or impermissible “making.” The reasoning behind the make approach is that the patentee’s right to “make” does not exhaust, because the patent exhaustion is recognized in a specific product that is assigned by the patentee or an authorized person.[59] Therefore, given that the assignee of the patented product can “repair” it as a part of “use,” if the assignee “repairs” it to the extent that the identity of the original patented product is lost, it should be considered “making” and the patentee’s enforcement of the patent should be permitted.[60] Under the exhaustion approach, courts consider whether alteration to a patented product is within the envisaged scope of use of the patented product at the time the patented product is placed in the market. The reasoning under the exhaustion approach is that in order to effectuate the gist of patent exhaustion doctrine, there is no necessity for the patent exhaustion to cover all uses of the patented product, but it is sufficient to cover the use foreseeable under the socially accepted sense. Hence, if the assignee of patented product uses the relevant product in an extraordinary way in light of the socially accepted sense, the effect of patent exhaustion should be denied and the patentee should be able to exercise its patent again.[61]

There are some cases which followed the “make” approach, and Fukuhiro v. Marusho Seiki K.K. (“Seisaki [sand manufacturing machine] Hammer)[62] is one such case. In Seisaki Hammer, Fukuhiro had a utility model related to hammers of sand manufacturing machines. The utility model facilitated the replacement of a wearable arm of a hammer by separating the arm and the hitting plate and by enabling the replacement of only the hitting plate. Marusho manufactured hitting plates which were replacement parts of utility modeled hammers and had no other use. The court found a contributory infringement, holding that “if the replacement of the part constitutes the use of the utility model over the extent that utility model holder is paid, it cannot be said mere repair, but it should constitute ‘make.’”[63]

Another “make” approach case is Nori Housou Souchi [packing machine for sheets of laver].[64] In this case, the plaintiff had patents for a packing machine for sheets of laver, and the alleged infringer sold the same kind of machines. Some of the accused products were manufactured by overhauling the plaintiff’s machine, i.e., by checking the whole machine, replacing and polishing some parts, painting the whole body of the machine, and thus making the condition of the machine akin to a new machine. The court held that either repair or conversion of the patented product constitutes exercising of the patent, i.e., “making”, should be established when the conduct of the alleged infringer is substantially deemed as the manufacture of the product covered by the invention.[65] The court then found an infringement stating that the overhaul by the alleged infringer is substantially deemed “making”.

The above two cases are district court decisions. Nevertheless, the Tokyo High Court also adopted the “make” approach in Welcome Foundation Ltd. v. Sawai Pharmaceutical Co. (“Acyclovir”).[66] In Acyclovir, the plaintiff had a patent for a chemical compound, including acyclovir. The defendant bought patented medicine which contained acyclovir, destroyed the tablets, abstracted the acyclovir and made medicine using that acyclovir. The court held that “[a]mongst the rights of the patentee, there is no exhaustion regarding the right to make. If the person who legally bought patented products or parts thereof on the market did an action which is judged as making of another patented product, such action of course constitutes an infringement.”[67] Ultimately, the court rejected the infringement claim, because defendant’s acyclovir is exactly the same acyclovir included in the plaintiff’s tablet and because it is impossible to deem that the defendant newly made the acyclovir. [d2]

Next, I will introduce several exhaustion approach cases. First, the Tokyo District Court issued a preliminary injunction in the Konica case.[68] In Konica, Konica had a utility model and design right regarding a single use camera. An undisclosed defendant sold single use cameras which were manufactured from used Konica’s plastic camera bodies and included new films and batteries. The court found an infringement, holding “if it is impossible to consider that the right holder unconditionally granted the right to freely use and resell the product, upon consideration of objective characteristic of the protected product, circumstances of the transaction, and method of use under the socially accepted sense, and if the defendant uses the intellectual property right over the extent of the condition, the right holder should exercise its utility model and/or design right.”[69]

Secondly, there is another case which recognized the exhaustion approach regarding the single use camera. In Fuji Photo Film Co. v. K & J K.K. et al (“Fuji Photo Film”),[70] Fuji had a patent, the utility models and design rights regarding a single use camera. Defendants imported or manufactured, and sold single use cameras which could be reused by replacing the film etc. of used Fuji’s single use cameras. The court applied practically the same reasoning in Canon, i.e., the introduction of Category I and Category II below. Then, the court found infringement, finding Category I was applicable regarding the patent and utility models, and Category I and Category II were applicable regarding the design rights.

3. Canon Case

The IP High Court handled the Canon case in a grand panel, which consisted of five judges, as opposed to the usual three judge panel. In Canon, Canon sold patented ink cartridges for Canon ink jet printers. Canon had a product patent and also a process patent. The recycler collected empty used Canon ink cartridges and, after cleaning, refilled the cartridges with ink in China and sold them in Japan. The IP High Court held that, in general, the patent is exhausted with respect to the patented products when sold by the patentee.[71] However, the IP High Court provided two categories of exceptions to patent exhaustion, because the patentee did not gain double profit and the free trade of goods was not hindered in these two categories.[72]

Category I is recycling products after the product life expires, and whether or not the life of the product has expired should be determined based on social and/or economic points of view.[73] For instance, the patent does not exhaust (i) if a product is unusable due to physical wear or change in chemical compound, (ii) if the shelf-life of a product has lapsed, as determined for health purposes, or (iii) if significant components are replaced or altered which extend the life of the whole product.[74] The life of the product does not expire when consumables or any other component whose life is shorter than the life of the whole product are replaced.[75] Category II is alteration or replacement of a component that constitutes an essential part of the patented invention. An essential part of the patented invention is the unique element of the particular patent which forms the core of the technical idea on which the solution is based.[76]

The IP High Court held that refilling the ink does not fall into Category I, but rather falls into Category II, and therefore, constitutes an infringement.

4. After the Canon Case

Quite recently, the Tokyo District Court rendered a judgment regarding recycling of single use cameras.[77] In this case, Fujifilm (f.k.a. Fuji Photo Film) alleged the infringement of two patents regarding single use cameras, one of which was the same patent as Fuji Photo Film. The other patent was issued after the Fuji Photo Film. One of the defendants imported the single use cameras which were made by replacing film cartridges etc. in the Fuji film products. The other defendant purchased refitted cameras from the defendant and sold them to consumers. The court found an infringement, stating that the conduct by the defendants falls into Category II in Canon. Although the court did not discuss the abovementioned Category I, it is clear that the Tokyo District Court followed the Canon standard.

C. Interim

Under the Japanese law, a patentee cannot control the patent exhaustion with the intention of the patentee.

Japanese courts traditionally took two approaches, i.e., the make approach and the exhaustion approach. But the Canon case explicitly denied the make approach, by stating that (i) it is difficult to determine whether the conduct constitutes repairing or making, if the patented product is not physically changed and also that (ii) the make approach is confusing, because it uses the word “make” in a different way than the word “make” in Item 1, Paragraph 3, Article 2 of Patent Act[78] . Because Canon adopted the exhaustion approach, Japanese courts should follow this approach in the future, subject to the survival of the Canon IP High Court decision under the Supreme Court’s review. Currently, the Canon case is under the review of the Supreme Court. It is the opinion of this author that the Supreme Court will affirm the Canon judgment, in view of the Japanese Supreme Court’s tendancy to affirm the high court decisions and the importance of the grand panel decision of the IP High Court.


A. Difference between United States cases and the Canon case

It seems that the United States courts and Japanese courts agree in that the patentee cannot control the patent exhaustion by its unilateral intention. However, the patentee can control the patent exhaustion by a valid contract in the United States[79] , as opposed to the Japanese courts which do not allow the patentee to control the patent exhaustion by contract, and also establish objective standards to determine whether a patent is exhausted.[80]

As stated above, the American courts apply the repair-reconstruction dichotomy to the recycling cases. The repair-reconstruction dichotomy is similar to the make approach in Japan, because both approaches try to distinguish permissible repairing and impermissible manufacturing. However, Japanese courts explicitly abandoned the make approach in the Canon case.[81] So, whether Japanese courts apply a different approach from the American courts should be analyzed.

Regarding Category I in Canon, the approach is generally the same as in the United States case law. In Aro I, the Supreme Court indicated that after the product has become spent as a whole, the revival of the product is deemed as reconstruction.[82] Also, in Aktiebolag, the Federal Circuit mentioned that the drill is “spent” when the tip can no longer be resharpened.[83] Then, the Federal Circuit stated that the conduct by E.J., the alleged infringer, is “effectively a re-creation of the patented invention after it is spent.”[84]

However, the lifetime of the product can be judged differently in the United States. For instance, in Mallinckrodt, the recycled product in question is for the disposal of radioactive or toxic materials. This product would fall into the second example of Category I, that is, “shelf-life of a product has lapsed, as determined for health purposes.” However, because the United States courts do not analyze the repair-reconstruction dichotomy in the case of a conditional sale, the shelf-life under the United States courts could not be found in this case. In addition, in Sage Products Inc, v. Devon Industries, Inc,[85] the Federal Circuit handled the case related to the biohazard container designed for the disposal of contaminated items, particularly those associated with sharp instruments commonly used in the medical field. Also, in Kendall Co. v. Progressive Medical Technology, Inc.,[86] the Federal Circuit dealt with the case relating to the medical kit which has a risk of contamination by a patient’s blood, body fluids and other excretions. In both cases, the Federal Circuit does not discuss the shelf-life of the product. Thus, it is possible to assume that the United States court takes a different position on shelf-life than the Japanese court. Actually, in Dana Corp. v. American Precision Co., Inc.,[87] the Federal Circuit ruled that the rebuild of patented clutches, using as many parts which are obtained by disassembling the worn clutches as possible, is permissible repair. The Federal Circuit also denied Dana’s argument that whether the patented product is “spent” should turn on the owner’s economic evaluation of the product, rather than on an examination of its physical characteristics.[88] Arguably, this reasoning is different from Canon’s holding that the expiration of the product’s life should be judged from a social or economical view point.[89]

Regarding Category II in Canon, it seems that the Canon’s approach could not be recognized under the United States common law, due to the fact that the United States case law does not recognize the “essential part” of the invention. In Aro I, the Supreme Court held that “this Court has made it clear in the two Mercoid cases that there is no legally recognizable or protected ‘essential’ element, ‘gist’ or ‘heart’ of the invention in a combination patent”[90] and denied the argument by the alleged infringer that the particular shape of the fabric was the ‘heart’ of the invention and therefore its patents should be held to grant it a monopoly on the fabric.[91] However, it would be important that Aro I case and the two Mercoid cases cited in it[92] coped with a “combination patent,” i.e., “a patent on the assembled or functioning whole, not on the separate parts.”[93] Arguably, it would be possible to assume that the United States Supreme Court might recognize the “essential part” of the invention, if the invention is not a “combination patent,” even though this would rarely happen.

In summary, if the detailed differences are ignored, both the American law and Japanese law are the same in the sense that a patentee can restrict the recycled product. In the United States, the patent generally exhausts at the time of the first sale, but a patentee can avoid the patent exhaustion by contract. Thus, in the case of recycling, the patent is not likely to exhaust if there is a contractual restraint, but the patent is likely to exhaust in the case of unconditional sale. On the other hand, in Japan, the patent generally exhausts at the time of the first sale, but exceptionally, the patent does not exhaust in the case of Category I and II in Canon. Thus, in the case of recycling, the patent is unlikely to exhaust.

B. Suggestion for Future Policy in the United States

1. Legitimacy of the Recycling – Big Picture

I will first discuss the big picture in the recycling issue – whether the court should allow recycling or not. This would depend on how the claim is written, but it would be a policy decision in some sense. Patent law is the adjustment of the incentive to inventors and the desire of availability of technology to the other persons. If these conflicting interests are balanced, the technical innovation is promoted best.[94] This is the reason why patent law limits the monopoly power of the patent holder. The patent exhaustion doctrine is also the adjustment of the monopoly power of the patentee and free trade of the patented products. The best balance between these two elements cannot be judged totally objectively, and thus the policy decision inevitably steps in.

Policy decision is generally made in consideration of various aspects, not only the incentive to the inventors and the desire of availability of technology, but also environment protection, antitrust regulations, consumer benefit etc. Arguably, when considering all these aspects, the better policy is to permit the recycling.

From the environment protection view point, permitting the recycling is of course preferable. The extinction of oil resource has been alarming for decades and ink cartridges and other recyclable components are often made of plastic, which is made from oil. In addition, disposal of plastic is problematic, because plastic does not decompose in the soil and burning plastic discharges toxic chemicals and carbon dioxide, which in turn causes global warming. There is strong necessity to recycle plastic products.

From the antitrust view point, the stronger the monopoly power of the patent is, the more anti-competitive it becomes. It is true that there is no case that identifies the restriction of recycling as antitrust violation either in the United States or in Japan, even though the antitrust claim or defense is often asserted. Moreover, in some situations, anti-competitive characteristics are strong. For instance, the pricing strategy of the printer manufacturer is a kind that ties ink cartridges to printers, by strategically obtaining patents to ink cartridges. Such characteristics do not directly result in violation of antimonopoly law, but they should be noted in the policy consideration.

From the customer benefit view point, it is better for customers to have a choice between genuine products by the patentee and recycled products by recyclers. Customers do not necessarily select the cheaper recycled products. Some customers choose new products because of the reliability or quality in spite of the high price. Thus, even from the manufacturers’ side, there is little or no necessity to expel the recycled products completely, because consistently both new products and recycled products are selected by customers. In addition, if recycling is permitted, the consumed cartridge has a market value. Then, customers can gain profit from the used cartridges, such as price discounts and free product.[95]

In consideration of the incentive to inventors, one can argue that enough incentives are being provided, due to the fact that the inventors/patentees can gain reasonable profit from the first sale. Because the United States allows the conditional sale of the patented product, the inventor/patentee can choose the marketing strategy, i.e., limiting the use of the cartridge for a single use, or permitting the recycled products. Such flexibility would enable the patentee to construct a market strategy from which the patentee can gain full value of the patented invention.[96] On the flip side of the coin, if the patentee sells patented products unconditioned, the patentee might face the risk that the recycler will reuse the patented product. Also, it should be noted that if the recycled product is permitted, the genuine products can have a certain amount of sale, because the consumed genuine product is the raw material of the recycled product. The patentee can gain profit at least from this sale. Moreover, a patentee obtains a patent for the whole invention, not for the parts modified or replaced. If a modification or replacement of the parts is an infringement, it would equal to a protection of the unpatented parts, as Aro I pointed out,[97] because the patent is infringed only if the all elements of the invention are satisfied by the accused devise.[98] That would give too much protection to the patentee over the framework of the patent law.

Regarding promotion of innovation, the free use of used patented products is helpful to innovation. It allows for improvements to the invention of the patented product, such as refillable ink cartridges[99] or inventions relating to efficient or high-quality recycling of the patented product. This would accelerate the competition in the innovation market, as well as the recycled product market.

In total, permitting recycling is the better policy, especially in jurisdictions where the contractual restraint on patent exhaustion is permitted, in the respect that it has many valuable aspects and it does not harm the incentive of the inventors.

2. Conditional Sale

I discussed above the big picture with regard to the premise that the patent does not exhaust in conditional sale, but it is arguable whether such contractual restraint should be permitted or not. In the common law countries, the possibility of fact specific decisions is preferable. Also, freedom of contract is favorable in the United States. Thus, in order to allow flexibility and freedom of contract, the contractual restraint on patent exhaustion should be permitted in the United States. Some critics say that the contractual restraint is problematic from the view point of free trade. However, if the restraint is marked on the package of the patented product, such problems would be solved. As a matter of fact, the restraint is usually printed on the package in mass consumption societies, and consumers can often know the restriction. Moreover, the contractual restraint is only applied on the parties of the contract under the basics of contract law, while the protection of the third party is often realized not by prohibiting the contractual restraint, but by applying other legal doctrines, e.g., protection of innocent third party. Therefore, the conditional sale should prevent the patent exhaustion in the United States.

3. Recognizing the “Essential Part” of the Invention

In Canon, the IP High Court recognized the “essential part” of the invention and used it in the criteria, but the American courts rejected it, as stated in Aro I and the two Mercoid cases. It is the opinion of this author that the American courts should continue to deny the recognition of the “essential part” or “heart” or “gist” of the invention in recycling cases, because there is no “essential part” in a combination patent. Since it is not appropriate to change the criteria between a combination patent and another kind of patent, such a concept should not be included in the criteria. Moreover, the recycler is not necessarily an expert in the industrial area in question and probably would not identify the “essential part” of the patent in question. In this case, the “essential part” standard would harm the predictability of the recyclers and other players in the market.[100] In addition, it is practically very difficult to decide the “essential part” of the claim, because “[e]ach element contained in a patent claim is deemed material to defining the scope of the patented invention.”[101]

4. Making Clear Criteria

In Canon, the IP High Court tried to make clear criteria for judgment of patent exhaustion in relation to the recycling, repair or reconstruction situation. It would be good to have such concrete guidance by the court in the civil law countries; nevertheless the American courts should not have such clear criteria. In common law countries, the court decision relies heavily on the specific fact pattern and having one clear criterion would be harmful to such a flexible legal system. Also, as noted above, the flexibility allows the manufacturers to choose various marketing strategies.

In addition, the United States does not always judge patent exhaustion objectively, because it allows the patentee to control the patent exhaustion by contract. This means the necessity of clear criteria is lower than in Japan where the patent exhaustion is always judged objectively.

5. Adopting Elements Test

Although the United States could not have concrete criteria like Canon decision, whether the American court should follow the elements test is another issue. Arguably the answer should be yes. The only applicable test in Aro I is too vague and conceptual.[102] Also, Aro I only converted the distinction between repair and reconstruction to whether the part modified or replaced was spent or not, but did not address any guiding criteria in order to asses whether the parts is “spent.” Given this problem, the lower courts would have difficulty implementing this test, and, in turn, elements test may lessen court confusion and provide a framework for consistent analysis.[103] As a matter of fact, Justice Brennan’s concurring opinion addressed that “there is no single test to which all must yield; rather the determination is to be based upon the consideration of a number of factors,” upon referring to older cases. Moreover, the Federal Circuit also conducted a kind of an elements test in the Aktiebolag case by analyzing various circumstances, although the Federal Circuit should follow the Aro I test.

In addition, an elements test may help the recyclers, i.e., the competitors of the patentee, to predict whether their business plan is lawful. The uncertainty in future business may cause a chilling effect to their business and technical innovation. This would be a great loss to society. Finally, an elements test does not harm the flexibility stated above. Then, the patentee would not be injured by the application of the elements test.

Factors should be considered in the elements test would be the same as those in Aktiebolag or in Justice Brennan’s opinion in Aro I.


Under the United States law, the patent does not exhaust if the first sale of the patented product is conditional, which means, the patentee can control the patent exhaustion by contract, but cannot control it by unilateral intention of the patentee. If the sale is unconditional, the American courts use the repair-reconstruction dichotomy when they discuss recycling cases. However, the line between permissible “repair” and impermissible “reconstruction” is a difficult one to draw.[104] In General, the replacement of the part of the patented product is a permissible repair, (i) if the part of patented product is designed to be replaceable, (ii) if such a part itself is not subject to a patent, and (iii) if the part cannot be used because it is worn or broken.[105] But permissible ‘repair’ is not limited to replacement of parts that are neither broken nor worn.[106] Supreme Court decisions tend to expand the scope of permissible repair in order to preclude extension of the monopoly on the patented machine or combination to unpatented parts.[107] The test introduced by the Supreme Court is that “reconstruction of a patented entity, comprised of unpatented elements is limited to such a true reconstruction of the entity as to ‘in fact make a new article,’ after the entity, viewed as a whole, has become spent.”[108]

In Japan, the patentee cannot control the patent exhaustion either by contract or unilateral intention of the patentee. Canon decision addressed objective criteria for the distinction whether a patent is exhausted or not. Canon explicitly rejected the traditional make approach,[109] and adopted the exhaustion approach. Currently, the Japanese test for the recycling case is that the patent basically exhausts by the first sale of the patented products. The exceptions for such result will be if the patented products are recycled after product life expires (Category I), or if a component of the patented product that constitutes an essential part of the patented invention is altered or replaced (Category II).

Category I above is generally true in the United States as well. However, shelf-life of the product could be differently understood in the United States and Japan. The American court seems to consider only the physical shelf-life of the products, while the Japanese court considers not only the physical shelf-life, but also the economic perspective.

Category II above has not been adopted in the United States, because the American court does not recognize the ‘heart’ or ‘gist’ of the invention.

The American court should maintain its policy under which the recycling is permitted, concerning various factors. It should permit the patentee to control the patent exhaustion by contract. Also, it should not adopt clear criteria for distinguishing permissible repair and prohibited reconstruction. Moreover, it should reject the test to determine the “essential part” of the invention. However, as opposed to the current practice, it should better adopt the elements test.

The American court recognizes the patent exhaustion broadly in the recycling case, but seeks the protection of a patentee by permitting the contractual restraint on the patent exhaustion. On the other hand, the Japanese court applies the patent exhaustion narrowly in the recycling case, but seeks the protection of free trade by denying the restraint on the patent exhaustion by the patentee’s intention. Each court tries to balance the protection of a patentee and the protection of free trade, and each court seems to find a balance which suits their own legal system. In other words, the United States, a common law country, chose the policy which allows flexible results while Japan, a civil law country, chose the policy which has more stable criteria.

Top of Page


  • [*] Hirohito Nakada is a 2007 graduate of the Intellectual Law and Policy LL.M. Program at the University of Washington School of Law.
  • [1] The Fifth Subcommitee of the Second Patent Committee, Risaikuruhinto Tokkyokentono Kankeino Kentou (Sono 1) [Consideration of the Relationship between Recycled Products and Patent (No. 1)], 52 Chizai Kanri 1281, 1281 (2002).
  • [2] Keiji Kondo, Risaikuruto Tokkyo Shingai [Recycling and Patent Infringement], 158 CIPIC Journal 15, 15 (2005).
  • [3] See Kyoto Protocol to the United Nations Framework Convention on Climate Change, December 11, 1997, 37 I.L.M. 22 (1998).
  • [4] The Fifth Subcommitee of the Second Patent Committee, supra at 1281.
  • [5] Keiji Kondo, supra at 15.
  • [6] Keiji Kondo, supra at 15.
  • [7] Canon Inc. v. Recycle Assist Co., 1922 Hanrei Jiho 30, (Intellectual Prop. High Ct., Jan.31, 2006).
  • [8] See id. at 46 (allegation by Recycle Assist).
  • [9] 5 Donald S. Chisum, Chisum on Patents § 16.03.
  • [10] 4 John Gladstone Mills III et al., Patent Law Fundamentals, § 20:40.50 (2d ed.).
  • [11] 84 U.S. (17 Wall.) 453 (1873).
  • [12] Chisum, supra § 16.03.
  • [13] Katsuya Tamai, America Tokkyohoni Okeru Kenri Shojinno Hori (1) [Patent Exhaustion Doctrine in American Patent law (1)], 54-10 Patent 19, 21(2001).
  • [14] See Jazz Photo Corp. v. Int’l Trade Comm’n, 264 F.3d 1094, 1105 (Fed. Cir. 2001).
  • [15] Chisum, supra § 16.03.
  • [16] Id.
  • [17] Id.
  • [18] Chisum, supra § 16.03.
  • [19] Mills, supra § 20:40.50.
  • [20] 976 F.2d 700 (Fed. Cir. 1992).
  • [21] Id. at 708.
  • [22] See id. at 706.
  • [23] Id. at 708.
  • [24] Id. at 709.
  • [25] Jazz Photo Corp., 264 F.3d 1094.
  • [26] Id. at 1108.
  • [27] Id. at 1108.
  • [28] 50 U.S. (9 How.) 109 (1850).
  • [29] Christina M. Sperry, Building a Mystery: Repair, Reconstruction, Implied License, and Hewlett-Packard Co. v. Repeat-O-Type Stencil Manufacturing Corp., 5 B.U.J. Sci. & Tech. L. 9, 11 (1999).
  • [30] Wilson, 50 U.S. at 125.
  • [31] 365 U.S. 336 (1961).
  • [32] Sperry, supra at 12.
  • [33] Aro Mfg. Co., 365 U.S. at 346.
  • [34] Id. at 346 (citation omitted).
  • [35] Sperry, supra at 12.
  • [36] Aro Mfg. Co., 365 U.S. at 363-364.
  • [37] Jazz Photo Corp., 264 F.3d 1094.
  • [38] Id. at 1107.
  • [39] Id. at 1107.
  • [40] 291 F.3d 780 (Fed.Cir. 2002).
  • [41] Id. at 785.
  • [42] Id. at 785.
  • [43] Id. at 785.
  • [44] Id. at 786.
  • [45] Id. at 789.
  • [46] Sandvik Aktiebolag v. E.J. Co., 121 F.3d 669, 674 (Fed. Cir. 1997).
  • [47] Sandvik Aktiebolag, 121 F.3d 669.
  • [48] Id. at 673.
  • [49] Id. at 673.
  • [50] Mark D. Janis, A Tale of the Apocryphal Axe: Repair, Reconstruction, and the Implied License in Intellectual Property Law, 58 Md. L. Rev. 423, 426 (1999).
  • [51] Katsuya Tamai, Nihon Kokunaini Okeru Tokkyokenno Shojin [Patent Exhaustion in Japan], in Shin-Saiban Jitsumu Taikei 4 Chiteki Zaisan Kanren Soshoho 233, 242 (2001).
  • [52] Surfco Hawaii v. Fin Control Systems Pty, Ltd., 264 F.3d 1062, 1065 (Fed.Cir 2001).
  • [53] Nobuhiro Nakayama, Chukai Tokkyoho Jo 667 (3d ed. 2000) (1983).
  • [54] BBS Kraftfahrzeugtechnik AG v. K.K. Racimex Japan, 1612 Hanrei Jiho 3, 7-8, (Sup. Ct., Jul. 1, 1997).
  • [55] Id. at 8.
  • [56] Canon, 1922 Hanrei Jiho 30.
  • [57] Id. at 48.
  • [58] Hisayoshi Yokoyama, Tokkyo Seihinni Taisuru Henkei Kouito Tokkyoken Shingai [Modification on the Patented Product and Patent Infringement], 170 Bessatsu Jurist 130, 130-131 (2004).
  • [59] Hisayoshi Yokoyama, Tsukaisute Shohinno Risaikuru Hanbaito Jitsuyoshinankenno Shojin – Konica Firumu Ittaigata Kamera Jiken [Sale of Recycled Single-use Product and the Exhaustion of the Utility Model – Konica Film Integrated Camera Case], 1201 Jurist 148, 149 (2001).
  • [60] Parties undisclosed, Hanrei Kogyo Shoyukenho (2ki han) 2399-263, (Osaka D. Ct., Jul. 23, 1992).
  • [61] Yokoyama, supra at131.
  • [62] Fukuhiro v. Marusho Seiki K.K., 1315 Hanrei Jiho 120 (Osaka D.Ct., Apr. 24, 1989).
  • [63] Id. at 128.
  • [64] Parties undisclosed, Hanrei Kogyo Shoyukenho (2ki han) 2399-263.
  • [65] Id. at 2399-282.
  • [66] Welcome Foundation Ltd. v. Sawai Pharmaceutical Co., 1779 Hanrei Jiho 89 (Tokyo High Ct., Nov. 29, 2001).
  • [67] Id. at 96.
  • [68] Konica Corp. v. Undisclosed defendant, 1712 Hanrei Jiho 175 (Tokyo D. Ct., Jun 6, 2000).
  • [69] Id. at 179.
  • [70] Fuji Photo Film Co. v. K & J K.K. et al., (Tokyo D. Ct. Aug. 31, 2000) available at
  • [71] Canon, 1922 Harei Jiho at 47.
  • [72] Id. at 47.
  • [73] Id. at 48.
  • [74] Id. at 48.
  • [75] Id. at 48.
  • [76] Id. at 48.
  • [77] Fujifilm Corp. v. K.K. Daito Boeki et al., (Tokyo D. Ct., April 24, 2007), available at
  • [78] Canon, 1922 Hanrei Jiho at 47-48.
  • [79] Mallinkrodt, 976 F.2d. at 706, 708.
  • [80] Canon, 1922 Hanrei Jiho at 48.
  • [81] Id. at 47.
  • [82] Aro Mfg. Co., 365 U.S. at 346.
  • [83] Sandvik Aktiebolag, 121 F.3d at 673.
  • [84] Id.
  • [85] 45 F.3d 1575 (Fed. Cir. 1995).
  • [86] 85 F.3d 1570 (Fed. Cir. 1996).
  • [87] 827 F.2d 755 (Fed. Cir. 1987).
  • [88] Id. at 760.
  • [89] Canon, 1922 Hanrei Jiho at 48.
  • [90] Aro Mfg. Co., 365 U.S. at 345.
  • [91] Id. at 344.
  • [92] Mercoid Corp. v. Mid-Continent Inv. Co., 320 U.S. 661 (1944), and Mercoid Corp. v. Minneapolis-Honeywell Regulator Co., 320 U.S. 680 (1944).
  • [93] Mercoid Corp., 320 U.S. at 684.
  • [94] See U.S. Const. art. I, § 8, cl. 8.
  • [95]  Personally, I obtained free recycled paper on which this thesis is printed by an empty printer cartridge.
  • [96] See David Pogue, Paying More for a Printer, but Less for Ink, N.Y. Times, May 17, 2007 at C1, C11 (introducing Kodak’s new marketing strategy for printers under which customers will pay more for a printer, but less for ink).
  • [97] Aro Mfg. Co., 365 U.S. at 345.
  • [98] See Warner-Jenkinson Co. v. Hilton Davis Chem. Co., 520 U.S. 17, 29 (1997) (addressing all elements rule regarding the doctrine of equivalents).
  • [99] See Hewlett-Packard Co. v. Recycle-O-Type Stencil Mfg. Corp., 123 F.3d 1445 (Fed. Cir. 1997).
  • [100] Tamai, supra at 248.
  • [101] Warner-Jenkinson Co., 520 U.S. at 29.
  • [102] See Sperry, supra at 27.
  • [103] Id. at 14.
  • [104] Chisum, supra § 16.03.
  • [105] Tamai, supra at 242.
  • [106] Surfco Hawaii, 264 F.3d at 1065.
  • [107] Chisum, supra § 16.03.
  • [108] Aro Mfg. Co., 365 U.S. at 346 (citation omitted).
  • [109] Canon, 1922 Hanrei Jiho, at 47-48.

Top of Page

Connect with us:

© Copyright 2015, All Rights Reserved University of Washington School of Law

4293 Memorial Way Northeast, Seattle, WA 98195