Empirical Research on Patent Compensation in China. Xiaodong Yuan

Empirical Research on Patent Compensation in China Xiaodong Yuan Abstract: The issues of patent compensation in China have attracted widespread attention of governments and public. What are primary elements affecting the amount of compensation decided by courts? From a comparative perspective, there are some special regulations on determining patent damages in China. Particularly, the approach of statutory compensation is a unique method of assessing compensation in the world. The paper undertakes a comprehensive empirical study on patent compensation decided by courts in China. The empirical research shows that some elements, which include the approach of statutory compensation, patent types and patent holder identity, can strongly affect the amount of compensation decided by courts in China. Key Words: patent compensation, statutory compensation, empirical study 1. Introduction The Chinese patent system has attracting widespread attention of governments and public. It focuses on relevant questions concerned with patentability requirements, longer pendency periods, and insufficient compensation, etc. The patent compensation practice and regulation of China have particularly attracted criticism in recent year because of insufficient compensation in patent lawsuit. Some foreign patent holders 1 complain that patent infringement compensations in China are generally low, causing them to believe that bringing suit is not worthwhile. 2 They emphasize that the rule of patent compensation of China is hard to protect patentee s benefits due to smaller awards than they expected. Meanwhile, some patentees in China also complain that suffering damages though having won lawsuit or damages are more than compensation. In general, patent holders had to bear the proof burden and long period, and take much cost in patent lawsuit. However, they This paper was supported by the National Nature Science Foundation of China (Project No. 71373088) Professor, Department of Intellectual Property, School of Management, Huazhong University of Science and Technology, Wuhan, China. 1 The United States International Trade Commission, China: Effects of Intellectual Property Infringement and Indigenous Innovation Policies on the U.S. Economy, Investigation No. 332-519, USITC Publication 4226, May 2011. 2 The United States International Trade Commission, China: Intellectual Property Infringement, Indigenous Innovation Policies, and Frameworks for Measuring the Effects on the U.S. Economy, Investigation No. 332-514, USITC Publication 4199.November 2010.P86. 1

can get a little damage even win a lawsuit. According to an investigation conducted by State Intellectual Property Office of China (SIPO), about 30% of patent holders in 30 Chinese provinces have suffering patent infringement, but only approximately 10% have filed lawsuits against infringer. 1 It has attracted a harsh criticism of insufficient compensation in China. Fundamentally, protecting benefits of patent holders and encouraging innovation are primary purposes of Patent Act. It will fail to fulfill such purposes if courts could not provide sufficient compensation for patent holder. Accordingly, damage awards are sufficient is crucial to patent system in China. In order to protect patent rights and facilitate technology innovation, China continues endeavoring to perfect patent systems. SIPO has prepared to fourth amend Patent Act focusing on improving patent protection and strengthening law enforcement since Feb. 2012. 2 However, how does the regulation of patent compensation of China be applied in practice? How to understand criticism of insufficient compensation in China? What are the fundamental elements that can affect amount of awarding compensation? From an empirical perspective, there is a little publication research on the understanding of Chinese patent compensation in detail. To close such gap, the paper endeavors to undertake a comprehensive empirical study on Chinese patent infringing cases and attempts to interpret the key factors affecting amount of awarding compensation in China. The paper firstly introduces the fundamental theory and provisions under the Patent Act of China, and then puts forward some hypotheses. To this end, constructs a theoretical framework and select research methodology. The empirical results of the study presents will be discussed in the following part. Finally, offers an analysis and explanation for results. The paper contributes new empirical evidence regarding impact elements which leads to amount of compensation decided by courts in China. 2. Literature review The justification of patent rights is to encourage creation and facilitate technology diffusion. Patent right can exclude others from making, using, selling, offering for sale or importing the invention. As a result, an infringer must be liable for damages arising from patent infringing activities. The damage award of a patent infringement lawsuit was deemed to be the legal value of the patent. 3 It is 1 SIPO: The direction for the draft of amend patent act (asking for comment draft) on 9 th Aug. 2012. 2 SIPO: The direction for the draft of amend patent act (asking for comment draft) on 9 th Aug. 2012. 3 Yi-Hsuan Lai, Hui-Chung Che, Evaluating patents using damage awards of infringement lawsuits: A case study, Journal of Engineering and Technology Management, 2009. 26, 167-180. 2

a fundamental mechanism of enforcing patent right if a patent holder filed to reach an ex ante licensing agreement. Therefore, a meaningful court remedy is an essential part of a functioning patent system. 1 The availability of remedies is basing on Patent Act. In China, the patent infringement is regarded as a special tort. The tort liability taken by the tortfeasor should be a significant rationale. The chief purpose of patent compensation is substantially and quickly compensating damages of the victim. To some extent, patent compensation is the evaluation of patent value and sometimes a patent value is discovered by court s decision. The patent compensation should recover the patent holder to the position he would have been if the infringement not taken place. Under such principle, the patent compensation should be firstly determined on the basis of actual losses suffered by the patent holder. It is not permit to determine the compensation depending on the gains obtained by infringer until it is hard to determine the actual losses. 2 However, according to Chinese traditional patent theory, it is hard to be applied in patent infringement because the results of patent infringement are impossible to restore patentee to the position he would have been if the infringement not taken place. 3 In such situation, patent compensation is firstly extended to the profits gained by infringer. Meanwhile, patent compensation will transfer the damages suffered by patent holder to the profits gained by infringer, namely loss shifting. From a comparative perspective, the principle of full compensation, which the compensation should be adequate to compensate for the infringement, is strictly abided by most countries in the world. In general, the Tort Act seeks to put the victim in the position he was in before the tort. Similarly, the Patent Act also endeavors to compensate the patent holder to the position he would have been if the infringement not taken place. Every case concerned with patent compensation should be demined in accordance with detailing situation. However, the damage is to an irreplaceable item of patent right, exact return to the status quo ante is impossible. It can never be accurately assessed by its nature so the aim is to come up with an amount which is fair, reasonable and just. Therefore, But For Test, which is a fundamental approach of determining the damages suffered by patent holders, is widely applied in common law. In order to adequately compensate for the infringement, the judge should endeavor to suppose the financial position but for patent infringement. The damages determined by court should be awarded to patentee after being minus actual profits from assuming profits. 1 John R. Allison, Mark A. Lemley, Extreme Value or Trolls on Top? The Characteristics of the Most-litigated Patents, University of Pennsylvania Law Review, 2010.158(1), 1-33 2 The Legal Affairs Department of SIPO, Guide to the Patent Act thirdly amended, Intellectual Property Press, 2009, P82. 3 Peter Meier-Beck, Damages for patent infringement according to German law Basic principles, assessment and enforcement, IIC, Volume 35, 2004.2, 113-124. 3

To accurate assessing damages, Patent Act in different country provides two or three kinds of approach to calculating compensation. For instance, the U.S Patent Act provides that compensation should be adequate to compensate for the infringement, but in no event less than a reasonable royalty for the use made of the invention by the infringer. 1 In practice, courts have established two primary approaches, which are lost profits and reasonable royalties, for awarding patent infringement compensation. The approach of lost profits focuses on the amount that the patentee has suffered by the infringement, 2 and a reasonable royalty can be calculated from an established royalty, the infringer s profit projections for infringing sales, or a hypothetical negotiation between the patentee and infringer. 3 Meanwhile, the U.K Patent Act allows a patent proprietor to freely elect approach between the damages suffered by patent proprietors and an account of the infringer s profits. 4 A large percentage of the patent proprietors would likely elect to claim for compensatory damages in practice and damages could be not precisely but objectively estimated by experienced judges. 5 Moreover, the German Patent Act provides three methods of assessing the damages, such as actual damages suffered by patentee, the profits made by the infringer, and reasonable royalty. 6 A patentee can claim a reasonable royalty from the infringer based on a hypothetical license agreement which is called as License Analogy. 7 Similar with German, Japanese Patent Act also provides three approaches to calculating compensation. There is without exception in China. The Patent Act of China has established a legal framework for determining patent infringement damages pursuant to different assessing approach. However, Patent Act of China has some distinct characteristics which are significantly different from other countries. Pursuant to Section 65 (1), Patent Act of China, The amount of compensation for a patent infringement shall be determined on the basis of the actual losses incurred to the patentee as a result of the infringement. If it is difficult to determine the actual losses, the actual losses may be determined on the basis of the gains which the infringer has obtained from infringement. If it is difficult to determine the losses incurred to the patentee or the gains obtained by the infringer, the amount shall be reasonably determined by reference to the multiple of the royalties for this patent. Moreover, there is a special approach statutory compensation besides such traditional approaches. Pursuant to Section 65 (2), Patent Act of China, court may decide a compensation in the sum of not less than RMB 10,000 Yuan but not more than RMB 1,000,000 1 Sec 284, 35 U.S.C. 2 Bandag, Inc. v. Gerrard Tire Co., 704 F.2d 1578, 1583 (Fed. Cir. 1983). 3 WordTech Sys., Inc. v. Integrated Network Solutions, Inc., 609 F.3d 1308, 1319 (Fed. Cir. 2010), (citing Georgia-Pacific Corp. v. US Plywood Corp. 318 F. Supp. 1116, 1120 (S.D.N.Y. 1970)). 4 Sec. 61, Patent Act of UK. 5 Gerber v Lectra [1995] RPC 383. 6 Sec 139, German Patent Act. 7 Marcus Schonknecht, Determination of patent damages in Germany, IIC, 2012 Vol 43, 3, 309-332. 4

Yuan, if it is difficult to determine the losses incurred to the patentee, the gains obtained by the infringer as well as the royalty obtained for the patent. 1 That is, the amount of damages can be determined by courts free discretion in accordance with the authority of Patent Act. Such method is called statutory compensation in China. From a comparative perspective, the approach of statutory compensation is a unique method of assessing patent infringement damages in the world. It is worth mentioned that the system of punitive compensation is not accepted by conventional patent theory of China. As well-known to us all, punitive damages in Patent Act originate from tort system, which have a long history, in the U.S. The Supreme Court of the U.S. explained that today punitive damages should be understood as quasi-criminal private fines designed to punish and deter the misconduct. 2 At present, punitive damages system of the U.S has a strong influence over Chinese patent theory. The draft of fourth amended Patent Act has added a provision relating to the punitive compensation for willful infringement. However, there is a conflict between the traditional compensatory damages with the punitive damages in China. The fundamental function of patent system is at compensation, but the aims of punitive damages are principally at retribution and deterring harmful conduct. 3 Such functions as retribution and deterrence are performed by other law or system. The Criminal Act fulfils the functions of retribution and punishment is initiated by the state. Moreover, the function of deterring infringing conduct is performed by the liability for stopping infringing action, which is equal the permanent injunction in the U.K and U.S, under Chinese Patent Act. From a traditional perspective, the system of patent compensation is aimed not at punishment but principally at compensation in China. Damage awards are one of the least-studied aspects of patent litigation. 4 At present, a majority of literature focuses on discussing whether damage awards are excessive or unpredictable in U.S. More importantly, there is a little literature in respect of empirical understanding as to damages. David addressed award value in a limited study of awards from 2002 to 2007 in U.S. 5 Michael investigate the the unpredictability of patent damages by analyzing the outcomes of 340 cases decided in US federal courts between 1995 and 2008. 6 To date, however, I have not see a paper of any empirical studies that attempt to explain damage awards in China. 1 Section 65, Patent Act of China. 2 Exxon Shipping Co. v. Baker, 128 S. Ct. 2605, 2621 (2008), U.S. 3 Exxon Shipping Co. v. Baker, 128 S. Ct. 2605, 2621 (2008), U.S. 4 Michael J. Mazzeoa, Jonathan Hillel c, Samantha Zyontzd, Explaining the unpredictable : An empirical analysis of U.S. patent infringement awards, International Review of Law and Economics 35 (2013) 58-72. 5 David W. Opderbeck, Patent Damages Reform and the Shape of Patent Law, 89 B.U.L.Rev. 127 (2009). 6 Michael J. Mazzeoa, Jonathan Hillel c, Samantha Zyontzd, Explaining the unpredictable : An empirical analysis of U.S. patent infringement awards, International Review of Law and Economics 35 (2013) 58-72. 5

3. Hypotheses Why does only China provide the approach of statutory compensation in Patent Act when assessing the amount of compensation? The traditional explanation is showed below. Sometimes patent holders are failed to take the burden of proving his actual losses or the gains obtained by infringer when filed litigation, because cases are very complex. If only depended on traditional approaches, courts were also hard to make a decision on determining the amount of compensation. As a result, an embarrassing position that the defendant has consisted of a patent infringement but the plaintiff cannot receive any compensation will be emerging. A set of puzzles arises from determining the amount of compensation in China. To avoid such embarrassing position, the court endeavors to elect an appropriate approach of determining damages. Furthermore, pursuant to Sec. 45 (2) of TRIPS, the judicial authorities may be authorized to order payment of pre-established damages when the infringer engaged in infringing activity. Finally, Chinese theory favor fair compensation to sufficient compensation and do not accept the approach of But For Test. The fair compensation refers to courts will decide the defendant compensating the plaintiff for suffered damages by carefully considering detail situation, such as the nature of infringement, financial situation of infringer, etc., under the principle of fairness, when the plaintiff did not prove his actual losses or the gains obtained by infringer. According to fairness principle, the courts should decide a reasonable amount of the compensation for the plaintiff by free discretion. Based on above mentioned reasons, China has gradually established the system of pre-established damage, which is called statutory compensation. The Supreme People s Court of the People s Republic of China issued Some regulation of applying law concerned with judging the cases of patent infringement disputes on 22th June 2001. According to such regulation, courts may decide compensation in sum of between RMB 5,000 Yuan and 300,000 Yuan but not more than RMB 500,000 Yuan, considering such facts as the type of patent, nature and particulars of the infringement, etc. As a policy, statutory compensation is regarded as an efficient and initial solution to determining the amount of compensation. It not only has effectively resolved above puzzles, but also enforced the theory of fair compensation. As a success patent policy, the approach of statutory compensation has been codified by Patent Act of China, which was thirdly amended on 27th October 2009 and become effective on 1st October 2009. Therefore, the approach of statutory compensation has applied by courts for about 13 years in China. However, what is the relationship between compensation and the approach of statutory compensation? In other words, what are primary elements affecting the amount of compensation decided by courts? Although Copyright Act usually provides statutory compensation for copyright 6

infringement in many countries, there are seldom countries providing statutory compensation for patent infringement in Patent Act except China. In general, there are several reasons contribute to the rule of statutory damages in Copyright Act. Firstly, the average cost of infringement detection is very high. With the development of digital technology, copying works without the author s authority is very cheap and easy. There are a lot of infringing activities conducted by copyright infringer. But it is hard to immediately detect such infringing activities. Secondly, it is hard to prove the actual damages suffered by author or the profits gained by infringer. Proving the amount of that harm nevertheless may be difficult, particularly in light of apportionment problems. 1 Finally, the losses caused by each infringing activity could not be a great magnitude of harm, though there are a lot of copyright infringements. Therefore, the traditional justification for statutory damages is that because actual damages are so often difficult to prove; only the promise of a statutory award will induce copyright owners to invest in and enforce their copyright sand only the threat of a statutory award will deter infringers by preventing their unjust enrichment. 2 However, the statutory damages usually not permitted in Patent Act, but entitled in Copyright Act and Trademark Act in many countries, such as the U.K, U.S and Germany, etc. In contrast, reversing grounds for not permitting the statutory damages in Patent Act might be raised in response to the patent infringement. Based on the business purpose, most patent infringers are entities though some activities of patent infringement can occur in private or individual person. Comparing the copyright infringement, the cost of detection may be lower and the magnitude of harm will be great for the patent holder. It would expect that a right to recover actual damages or profits might come closer to providing a sufficient incentive for detection and enforcement in the typical case involving patent. 3 Therefore, Patent Act is not permitted to lay down a rule of the statutory damages in many countries. Based on above theory, I putting forward a hypothesis that similar grounds for the statutory compensation are raised in Patent Act of China. Above all, the costs of detecting patent infringing activities and damages are very high in China. Because Chinese market is very large, many competitors favor imitating and infringing patents. Sometimes, a patent holder is hard to detect infringing activities conducted by competitors. Moreover, the capacity of bearing proof burden should be strong. Many patent holders in China are individual person or universities who are far away technology or product market. They usually do not have a basic capacity of proving actual losses and hardly bear the burden of proof. Finally, the value of patent-in-suit could not very high 1 Twin Peaks Prods., Inc. V. Publications Int'l, Ltd., 996 F.2d 1366, 1382 (2d Cir.1993), U.S. 2 Paul Goldstein, Copyright, 12.2, at 12:34 (2d ed. 1996),U.S. 3 Roger D. Blair, Thomas F. Cotter, An Economic Analysis of Damages Rules in Intellectual Property Law, William and Mary Law Review, Volume 39,1998.5,1585-1694. 7

and the losses could not be a great magnitude of harm for a patent holder. Hypothesis 1: The approach of statutory compensation applied by courts could achieve less compensation decided by courts than the other approach. The patents fall into three categories, which cover invention patent, utility model patent, and design patent, under Patent Act of China. Invention patent refers to any new technical solution relating to a product, a process or improvement thereof. In contrast, utility model patent is any new technical solution concerned with shape, structure, or their combination of a product and fit for practical use, which are called as petty patent. However, design patent refers to a new design of shape, pattern or their combination, or the combination of color with shape or pattern of a product that creates an aesthetic feeling and is fit for industrial application. There is an interesting phenomenon that most Chinese inventors favor filing design patents over invention patents. By contrast, foreign companies focused their patent filings in China on invention. 1 Meanwhile, a majority of patents in lawsuit are concerned with design patents in practice. Because the value of design patents is less than invention patent, the losses caused by each infringing activity could not be a great magnitude of harm. The amount of compensation is hard to trigger a sufficient incentive for patent holder to detection. Therefore, a majority of patent holders in lawsuit are failed to prove the amount of suffered losses or profits gained by the infringer, but the infringer consisting infringement. Hypothesis 2: Patents Types will strong affect the amount of compensation decided by courts. (1) The design patent will receive less compensation decided by courts than the other types. (2) The invention patent will achieve more compensation decided by courts than the other types. Fundamentally, the incentives to innovate that patents are intended to provide are predicated on a patent holder s ability to predictably defend his or her patent. 2 To some extent, the identity of patent holder represents a key capability to enforce and protect patent rights. Just as above mentioned, most foreign patentees not only focus invention patent filings, but also are companies in China. Companies usually have stronger capability to protect patent right than individual. However, a great proportion of patent holders are individual in China. The cost of infringement detection is very high for individual patentee. Furthermore, proving the amount of actual losses is very difficult due to individual patentee seldom exploring patent-in-suit, and it is hard to prove the profits gained by 1 The United States International Trade Commission, China: Intellectual Property Infringement, Indigenous Innovation Policies, and Frameworks for Measuring the Effects on the U.S. Economy, Investigation No. 332-514, USITC Publication 4199, November 2010. 2 Michael J. Mazzeoa, Jonathan Hillel c, Samantha Zyontzd, Explaining the unpredictable : An empirical analysis of U.S. patent infringement awards, International Review of Law and Economics 35 (2013) 58-72. 8

infringer for individual patentee. In contrast, a majority of invention patents are owned by companies and institutors in China. Because having stronger capability to protect patent right, companies would achieve more compensation than individual patentee. To establish a reasonable analysis, the paper constructed a theoretical framework below. Hypothesis 3: An individual patent holder could receive less compensation decided by courts than other patent holder who is a company. Invention Patent Value: Approach: Actual losses Infringer attain Royalty Statutory Companies Individual Utility model Design H1 Patent holders capability: H2 H3 Amount of compensation Figure 1: The theoretical framework 4. Methodology In the empirical analysis to follow, the goal of paper is to study impact elements relating to compensation. The paper will examine (a) the effect of statutory compensation approach on compensation decided by courts (Hypothesis 1), (b) whether the categories of patents will affect compensation decided by courts (Hypothesis 2) and (c) whether identity of patent holders play a role in determining the amount of compensation (Hypothesis 3). The sample, measures and methods are summarizes as follows. 4.1 Explanatory Variables To complement an empirical analysis, the paper established a series of explanatory variables that could potentially explain the compensation amount in each case. Above all, the dependant variable LA is the natural logarithm of compensation amount made by court in cases. Moreover, independent variable S reflects the approach applied by courts in decision. Although 9

there are four approaches of determining amount of compensation, the paper divided those approaches into two categories: statutory compensation and no statutory compensation. Independent variable S represents whether the decision made by court depended on the approach of statutory compensation. If the amount of compensation was determined by statutory compensation, it was noted 1. Otherwise, it was noted 0. Meanwhile, patent types can represent the value of patent-in-suit. According to Patent Act of China, there are three types of patents. To establish reasonable variables, the paper constructed two independent variables T1 and T2 to represent patent types. T1 refers to whether the patent-in-suit was a design patent. If a patent-in-suit was a design patent, it will note 1; otherwise note 0. Similar to T1, T2 represents whether the patent-in-suit was a utility model patent. If a patent-in-suit was a utility model design patent, it will note 1; otherwise note 0. Neither a design patent nor a utility model patent, it shall be an invention patent. Finally, the capability of a patent holder will be embodied by the identity of a patent holder. The independent variables I represents whether a patent holder is a company. If a patentee in lawsuit was a company, it will be noted 1; otherwise note 0. Therefore, the explanatory variables used are summarized below. Table 1: Variable definitions Category Title Code Variables description Dependant variable Independent variable Amount of compensation LA The natural logarithm of compensation account made by court in cases. Approach S Whether a statutory compensation was applied by courts in applied decision? If is, S=1; otherwise, S=0. Patent type T T1 Whether is a design patent? If is, T1=1; otherwise, T1=0. T2 Whether is an invention patent? If is, T2=1; otherwise, T2=0. Identity of a I Whether does a patent holder be a company or individual? If it patent holder is a company, I=1; otherwise, I=0. The paper established a multiple regression analysis to examine what are the primary elements affecting the amount of compensation. 10

4.2 Collecting Cases The paper endeavors to collective and examine the cases across mainland from the website of China Intellectual Property Right Judgments & Decisions 1. There are 5001 cases relating to patent lawsuit in the website of China Intellectual Property Right Judgments & Decisions until 31 December 2013. 2 If some cases proceeded with the first-instance and second-instance or even reexamine, it will be regarded as a case. About 2790 cases concerned with patent compensation was compiled after had deducted the cases with respect to the disputes of patent ownership and contract. There are 1004 cases that the plaintiff was supported by court, and 712 cases that were mediated by court, as well as 1080 cases that plaintiff and defendant were conciliated. According to Sec. 9 of Chinese Civil Procedure Act, courts should conduct mediation under the principle of voluntariness and lawfulness. If it is fail to conduct mediation, courts should render judgments without delay. The mediation refers to a mechanism of resolving disputes between plaintiff and defendant assisted by court. As a mediator, court will assist the parties to negotiate the patent settlement and produce a mediation agreement. However, the conciliation means plaintiff and defendant negotiate faced each other in the absence of judges and reach a conciliation agreement under the principle of voluntariness and mutual understanding. As an alternative dispute resolution (ADR) process, mediation or conciliation can successfully resolve disputes between patent holder and infringer. In contrast, there are 26 cases that the plaintiff not recovered damages in spite of the defendant infringing the plaintiff s patent. Meanwhile, the database added 7 cases disclosed in the other websites except the website of China Intellectual Property Right Judgments & Decisions. As a result, there are 2803 cases with respect to patent lawsuits in total, including 985 cases concerned with patent compensation decided by court. According to existing Patent Act of China, the plaintiff should follow a particular sequence when claim for compensation: actual losses, gains obtained by the infringer, referring to the multiple of the royalty, and statutory compensation. Such order is a compulsory regulation that should be abided by. However, it cannot be applied until 1st October 2009 in accordance with the theory of legal retroactivity. For the sake of a correctly application of amended Patent Act, the supreme 1 The website of China Intellectual Property Right Judgments & Decisions is operated by Intellectual Property Court of the supreme people s court of China on 2005. http://ipr.chinacourt.org/. 2 The number of patent civil cases has continued rapidly increasing for several years. Courts at various levels received 3196 first instance patent civil cases in 2006, and 4041 cases in 2007, and 4074 cases in 2008, and 4422 cases in 2009, and 5785 cases in 2010, as well as 7819 cases in 2011. Among them, there were a lot of cases involving patent infringement. Moreover, mediation of Intellectual Property lawsuits is the vast majority of the cases concluded and achieved great results. For example, the average withdrawing rate of first-instance Intellectual Property civil cases through mediation by local courts reached 61.08% in 2009, and 66.76% in 2010, as well as 71.32% in 2011. However, it is hard to get relevant decisions made by courts. Disclosed decisions in the website of China Intellectual Property Right Judgments & Decisions only occupy a small percentage of all cases. 11

people s court of China issued The notice with respect to learning and implementing amended Patent Act (It will be called Notice as following). According to the Notice, if the infringing action conducted before 1st October 2009, it will be applied to previous Patent Act; and the amended Patent Act will be applied if a suit of patent infringement had been filed after 1st October 2009. Moreover, it also should be abided by the amended Patent Act, if the infringing action, which conducted before 1st October 2009, continued to conduct until 1st October 2009 and was filed a suit by patent holder. After carefully examined the cases after 1st October 2009, I found that a majority of cases still applied by previous Patent Act. In other words, it is hard to evaluate or verify the implementation effects of the amended Patent Act in collective cases. Therefore, the analysis and discussion relevant to electing approaches primarily refer to previous Patent Act instead of the amended Patent Act. Table 2 compares the number of cases relevant to approaches applied by courts when determining the patent compensation as follows. Table 2: The Database of Approached applied by courts when determining compensation Actual losses Infringer Referring to Statutory Total gains royalties compensation Invention Patent 1 4 5 100 110 Utility Model Patent 0 1 4 233 238 Design Patent 1 1 3 632 637 Total 2 6 12 965 985 Table 3 describes the amount distribution of awarding compensation. Different type of patent would receive a distinctive amount of awarding compensation. It is obvious that the amount of awarding compensation concerned with invention patent is more than utility model patent and design patent. The maximum amount of awarding compensation is RMB 50,612,400 Yuan which was calculated by the approach of infringer s gains in a case of invention patent infringement. However, a majority of amount of awarding compensation is not more than RMB 200,000 Yuan. The minimum amount relating to design patent is RMB 3000 Yuan, and the minimum amount concerned with utility model patent is RMB 10,000 Yuan, as well as the minimum amount with respect to invention patent is RMB 15,000 Yuan. 12

Table 3: The amount of awarding compensation Minimum First Median Third Maximum Quartile Quartile Invention Patent 15,000 77,500 150,000 200,000 50,612,400 Utility Model Patent 10,000 50,000 80,000 150,000 1,760,000 Design Patent 3,000 25,000 50,000 80,000 500,000 4.3 Electing effective samples Table 2 clearly shows that decisions based on the approach of statutory compensation almost occupy 98.07% of all decisions made by courts. It is worth mentioned that only 20 cases are irrelevant to statutory compensation and in a small proportion of total cases. To select reasonable samples, the paper reserved such 20 cases irrelevant to statutory compensation and selected another stratified random samples, which consist of 20 cases relating to statutory compensation. The estimation samples are based on decision made by courts during the 2002-2009 periods. Consequently, 40 effective samples were extracted from 2803 patent infringement lawsuits retrieved from Chinese courts. As shown in Table 4, the mean amount of compensation is RMB 2,413,572.5 Yuan. Table 4: Sample Statistics (Number=40) Amount of compensation (RMB ) Patent Types Mean SD Media Min Max 2413572.50 1446851.27 116400 5000 50612400 Invention Utility model Design Company Individual 11 11 18 Identity 27 13 5. Results and Discussion To complete a regression analysis, the paper elects software SPSS 19.0 to study. It is supposed that amount of compensation determined by courts are different from statutory compensation approach and other approaches. To disclose such different, the paper firstly conducts a statistical t- 13

Test for natural logarithm of compensation account. Result of t-test is presented in Table 5. Table 5: The t-test of applying or not applying statutory compensation approach Applying Samples Not Applying Samples Mean difference SD t Sig. LA mean value LA mean value 10.7431 13.1034-2.36035.53718-4.394.000 Table 4 shows that LA mean value of applying statutory compensation approach is significantly different from not applying. LA mean value of applying, which is 10.7431, is quite distinct from LA mean value of not applying, because it is 13.1034. This presents that amount of compensation determined by the approach of statutory compensation is obviously less than decided by other approaches, such as actual losses, infringer gains and referring to royalties. To examine affecting degree of individual element, the paper endeavors to construct a hierarchical regression. As a unique approach, statutory compensation plays a significant role of determining amount of compensation. Model 1 firstly examines the correlation between compensation amount and approach. That is, the independent variable only includes S. Model 2 adds patent types as another independent variable, which includes T1 and T2. Because different category of patent embodies different value, the independent variable T can represent patent value affecting amount of compensation. Finally, Model 3 examines at least three elements which are S, T and I. The identity of a patent holder reflects a basic capacity of protecting patents. Model 3 endeavors to explain the correlation between dependent variable and independent variables. 14

Table 5: Hierarchical regression Model 1 Model 2 Model 3 Beta t p Beta t p Beta t p S -.580-4.394.000 S -.342-2.448.019 T1 T2 -.222.296-1.524 1.919.136.063 S -.355-2.875.007 T1 -.266-2.059.047 T2.197 1.414.166 I.360 3.343.002.337.490.613.337.153.123 F 19.307 5.398 11.176 P.000.009.002 Model 1 can explain the correlation between dependent variable LA and independent variable S due to, F=19.307, and p<0.001. Moreover, Beta is -.580. It means that correlation between LA and S is negative and highly significant. Just as expected in Hypothesis 1, the amount of compensation will be less than the amount decided by the other approach, if court chooses the approach of statutory compensation. In contrast, the amount of compensation determined by actual losses, or infringer gains or referring to royalties is more than compensation decided by statutory compensation. After independent variable of patent type T was putted into Model 2, the explanatory power become stronger because R2 has reached.490 and p is.009<0.05. The increment of explanatory power is 0.153 and. Such increment in Model 2 has a statistical significance. It means that patent types T including T1 and T2 have a significant explanatory power for amount of compensation LA. However, the independent variables of T1 and T2 do not have statistical significance because p of T1 is 0.136 and p of T2 is 0.063. The primary reason is the collinearity between T1 and T2, which contributes to the decrease of T explanatory power. Model 3 is added a new independent variable of patent identity I. Based on, Model 3 has a strong explanatory power which can explain the dependent variable LA 61.3%. Comparing with Model 2, Model 3 has increased.123 increments for explanatory power. Moreover, it has statistical 15

significance because p is.002<0.01. On the one hand, the coefficient of independent variable I is positive and highly significant, because Beta is.360 and p is.002<0.01. Just as expected in Hypothesis 3, it will achieve more compensation than an individual patentee if a patent holder is a company. On the other hand, the independent variable T1 has negative and significant coefficient with dependent variable LA due to Beta is -.266 and p is.047<0.05. It means that a design patent will receive less compensation than invention and utility model patent. Hypothesis 2 is tested. However, the independent variable T2 does not have statistical significance and explanatory power because of P=.166>0.05. To summarize, the independent variables S, T1 and I have outstanding to dependant variable LA. If an individual patentee owns a design patent and court determine compensation by approach of statutory compensation, the amount of compensation will be a little. On the contrary, the amount of compensation will be enormous, if an invention patent is owned by a company and compensation is decided by other approach instead of statutory compensation. 6. Conclusions This paper examines what kind of elements affect amount of compensation when courts made decision. In doing so, it helps bridge understanding patent compensation in practice. The primarily empirical findings of paper can be summarized as follows. Above all, the approach of statutory compensation can contribute to less compensation determined by courts than by other approaches. For instance, the actual losses suffered by patent holder refer to the legitimately expecting profits, which patent holder could expect to attain profits if without patent infringement action, subtract the actual profits attained by patent holders. Patent right is an exclusive right, which can exclude competitors from utilizing the invention without the consent of patent holder. As a consequence, the losses incurred to the patentee mainly reflect in future profit that the patent holder can expect to obtain without patent infringement, namely indirect losses. It is equal to lost profits in U.K or U.S. However, the Patent Act of China does not interpret what is mean of actual losses suffered by patent holder. The losses suffered by patentee may be calculated by the number of lost sales of patented products multiplying the reasonable profit of each patented product owing to the patent infringement. In general, lost profits include profits on lost sales as well as profits lost as result of price erosion. 1 However, the losses suffered by patentee only include profits on lost sales, not including profits lost as result of price erosion in practice. Moreover, the amount of profits obtained by the infringement may available by multiplying the reasonable profit of each infringement product by the quantities of the marketing products concerned with patent infringement. In general, 1 Michael G. Keeley, Estimating Damages in Patent Infringement Cases: an Economic Perspective, Cornerstone Research, 16

the profits gained by the infringer may be calculated according to operation profits. As for some infringers taking infringement as an occupation, the profits may be calculated on the basis of sales profit. Another approach is reference to the multiple royalties. Courts can determine a reasonable compensation from one time to triple royalty. Empirical research illustrates that these approach could lead to more compensation than statutory compensation. In practice, the approach of statutory compensation, which is extensively applied by Chinese courts, almost occupies 98.07% of all cases decided by courts. The results suggest that the criticism of insufficient compensation has closely related with an extensive application of statutory compensation approach. Moreover, amount of compensation can also be affected by patent types. Empirical study shows that only design patent has negative and significant influence on compensation due to collinearity. The results help to understand value different among different patent types and explain that why foreign companies favor filing invention patents, while most Chinese patent holders prefer design patents. Finally, individual patentee has negative and significant influence on compensation. In general, individual patent holders occupy a certain proportion for a long time in China. However, such situation is changing because the percentage of service invention is significant increasing. The results help to understand some patent holders cannot take the burden of proof because individual patentee is lack of capacity of patenting patent. While this paper deepens our understanding of what is primary elements contributing to compensation decided by courts, it is limited in ways that could be addressed in future studies. Firstly, the paper focuses on one approach of statutory compensation in isolation from the other approaches. Future research will comprehensive study the other approaches, such as reference to royalty. Another limitation of the paper is that it exams limited sample because cases have not be entirely published. Future study will collect more cases and select ampler and more reasonable samples. Finally, the samples cover period from 2001 to 2009 and not extend from 2010 to 2013. Because Patent Act of China amended in 2008 and has become effective in 2009, it is worth to further study in future. 17

Reference: David W. Opderbeck, Patent Damages Reform and the Shape of Patent Law, Boston University Law Review, 2009.89, 127-B.U.L.Rev. 127-187. Marcus Schonknecht, Determination of patent damages in Germany, IIC, 2012.43 (3), 309-332. Michael J. Mazzeoa, Jonathan Hillel c, Samantha Zyontzd, Explaining the unpredictable : An empirical analysis of U.S. patent infringement awards, International Review of Law and Economics 2013.35, 58-72. John R. Allison, Mark A. Lemley, Extreme Value or Trolls on Top? The Characteristics of the Most-litigated Patents, University of Pennsylvania Law Review, 2010.158(1), 1-33. Peter Meier-Beck, Damages for patent infringement according to German law Basic principles, assessment and enforcement, IIC, Volume 35, 2004.2, 113-124. Roger D. Blair, Thomas F. Cotter, An Economic Analysis of Damages Rules in Intellectual Property Law, William and Mary Law Review, Volume 39,1998.5,1585-1694. The Legal Affairs Department of SIPO, Guide to the Patent Act thirdly amended, Intellectual Property Press, 2009. The United States International Trade Commission, China: Intellectual Property Infringement, Indigenous Innovation Policies, and Frameworks for Measuring the Effects on the U.S. Economy, Investigation No. 332-514, USITC Publication 4199, November 2010. The United States International Trade Commission, China: Effects of Intellectual Property Infringement and Indigenous Innovation Policies on the U.S. Economy, Investigation No. 332-519, USITC Publication 4226, May 2011. Yi-Hsuan Lai, Hui-Chung Che, Evaluating patents using damage awards of infringement lawsuits: A case study, Journal of Engineering and Technology Management, 2009. 26, 167-180. 18