Indicator for Evaluating National Patent Performance: Comparative Analysis among the 30 OECD Countries

Transcription

1 Journal of Intellectual Property Rights Vol 17, March 2012, pp Indicator for Evaluating National Patent Performance: Comparative Analysis among the 30 OECD Countries Tae-Kyu Ryu and Yoo-Jin Han 509-ho, Centennial Hall, School of Global Service, Sookmyung Women s University, 52 Hyochangwon Road, Yongsan-gu, , Seoul, Korea Received 27 September 2011, revised 30 November 2011 In recent years, patenting has been considered one of the major drivers for enhancing national competitiveness and most of the advanced countries have been more actively enforcing patent protection. However, the indicators proposed in previous studies have not been able to deal adequately with the various dimensions of patenting, rather focusing on patent application counts. Therefore, in this research, a comprehensive patent performance indicator is proposed to provide a yardstick by which government policymakers can evaluate the whole process of converting patents into economic assets. The detailed steps of this research include: opting for the dimensions creation, administration, utilization and environment (also the most emphasized in the prior studies) in order to derive individual indicators; proposing a composite patent performance indicator by adopting the corresponding weights for each indicator calculated from the AHP method; and finally demonstrating the differences between the proposed indicator and the previous count-based patent indicator, and comparing the results for the 30 OECD countries. Keywords: Patent performance, indicator, OECD Nowadays, technological competence has become one of the most essential elements for enhancing a nation s competitiveness and many scholars have proposed diverse science and technology (S&T) indicators to measure a country s level of technological performance. In 1984, the OECD first introduced the S&T indicators 1, which were re-titled as the main S&T indicators in Another well-known example is the European report on the S&T indicators, which was first published by the European Commission (EC) in 1994 (ref. 2). Likewise, in order to grasp and model the status quo of one country s technological performance and formulate S&T policy accordingly, many countries have started to develop their own indicators. 3 Traditionally, the most frequently used proxy measures for technological performance were dichotomized into two components 1-3 : one related to R&D activities (R&D expenditures, R&D personnel, etc.); and the other related to patent activities (number of patent applications, number of triadic patent families 4, etc.). However, due to the limitations of R&D-related indicators that they only show the input side of innovation, more attention has been paid to the patent-related ones. Among others, seminal research was conducted, in which it was shown that multi-factor productivity can be better explained with patent stock than with R&D stock. 5 Later, several Corresponding author: yjhan@sm.ac.kr researchers tried to exhibit country-wise comparison results by employing their own patent-related metrics like comparing the number of patent applications of Israel to mainly the G7 countries 6 ; or by maintaining that the ranking results presented by the European Innovation Scoreboard (EIS) should be interpreted with caution since different orders are sometimes obtained depending on the methods adopted. 7 However, despite the shortcomings of the ranking method, many institutions such as the International Institute for Management and Development (IMD), the World Economic Forum (WEF) and the Economist Intelligence Unit (EIU) adopt this in order to grasp the status of particular countries in a specific economic sphere, thereby enabling policy-makers to establish country-specific goals. However, extant literature has only taken into account patent applications and/or grant counts 1-3, while not incorporating other aspects specific to patenting. Therefore, in this research, the authors aim to explore the relatively less studied areas of previous studies and propose a new patent performance indicator at the national level. Proposition of Indicator Individual Indicators First of all, three major dimensions that represent a country s patent performance creation, administration, and utilization were selected since they have been the

2 104 J INTELLEC PROP RIGHTS, MARCH 2012 most frequently emphasized fields in two earlier, representative studies. 8, 9 In particular, the study by Fu and Yang 9 elaborated the process whereby patents (i.e. innovation capacity) generate ultimate economic outcome (i.e. actual performance). Besides, other studies have emphasized the environment under which one country creates, administers, and utilizes patents, consequently generating economic rents. 10,11 Keeping this background in mind, in the first stage, a framework was developed by adopting Fu and Yang s point of view and the environment factor accentuated by the WEF and IMD. The individual indicators were drawn from this framework (Fig. 1). This was followed by the construction of the hierarchical anatomy for constituting a national patent performance indicator as shown in Fig. 2. Fig.1 Conceptual matching between Fu & Yang (2009) and this research Fig.2 1 st level hierarchical anatomy of national patent performance indicator The three dimensions creation, administration, and utilization were divided into quantity and quality. It is necessary to take quantity into account because in order to develop a new technology, the absolute amount of technologies has to reach a certain level. In other words, technology has a cumulative characteristic; therefore, in order to develop more sophisticated ones, technological stock needs to be taken. 12 On the other hand, quality also needs emphasis because core technologies can create much more benefits than peripheral ones. 13 In the case of environment, as in IMD s research, the dimension was divided into (i) domestic support, which represents the nationwide infrastructure that upholds the creation, administration, and utilization of patents, and (ii) international relationship, which links those activities to global patent regimes. Lastly, both absolute and relative aspects were taken into account since the absolute aspects show the extent of potential a country has in each domain, while the relative aspects demonstrate how competent a country is, in comparison to other countries. 10, 11 In addition, a main actor for each dimension was considered in order to show relative performance: in the case of creation, since inventors are the major subjects in producing patents 14, the absolute indicators were divided by the number of inventors; in case of administration, since examiners are the primary supervisors 15, by the number of examiners; and finally, in the case of utilization, since knowledge-based firms are the major players, by the number of knowledge-based firms. 16 Based on the aforementioned hierarchical structure and rationale, a finalized set of indicators was proposed as shown in Fig. 3 and Table 1. Fig. 3 Indicators for each 2 nd level

3 RYU & HAN: INDICATOR FOR EVALUATING NATIONAL PATENT PERFORMANCE 105 Table 1 Indicators and explanations 1 st level 2 nd level Indicator name Explanation Creation Quantity Quality (1) Number of domestic and foreign patent applications a Shows how many patents are created by a country s inventors (2) Number of domestic and Shows how more patents one country s inventors foreign patent applications/ 10,000 inventors a create than other countries inventors (3) Number of triadic patent Shows how many good patents are created by a applications country s inventors (4) Number of triadic patent Shows how many good patents one country s inventors applications/10,000 inventors a create in comparison to other countries inventors Administration Quantity Quality (5) Number of patent examinations a (6) Number of patent examinations/ a (7) Number of PCT international search reports b (8) Number of PCT international search reports/ Shows how many patents are administered by a country s Shows how many patents one country s patent examiners administer in comparison to other countries Shows international competitiveness of one country s Shows international competitiveness of one country s in comparison to other countries Utilization Environment Quantity Quality (9) Number of patents in force c Shows how many patents are currently utilized by a country s firms in knowledge-based industries (10) Number of patents in Shows the number of patents currently utilized by a force/number of firms in the country s firms in knowledge-based industries as knowledge-based industries c, d compared to those of other countries Domestic support International relationship (11) Value-added in the knowledge-based industries e (12) Value-added in the knowledge-based industries/ Number of firms in the knowledge-based industries d, e (13) Cost for patent application, registration, and maintenance f (14) Degree of IP informatization g Shows the extent of economic rents generated by a country s firms in knowledge-based industries Shows the extent of economic rents generated by a country s firms in knowledge-based industries as compared to those of other countries Shows how encouraging the conditions are in one country, where inventors can file, register, and maintain their patent rights. That is, the lower the cost, the more desirable the circumstance Shows how smooth the information infrastructure for patent application, registration, examination procedures is in a country. In an operational sense, the compound value of three aspects regarding IP administration informatizationis calculated: the spectrum of information (information on patents, utility models, designs, and trademarks), the possibility of on-line examination, and the possibility of tracking the examination process (15) WIPO contributions h Shows how actively one country allocates its budget so that it enlarges the base of international influence (16) Number of international Shows how positively one country participates in the patent-related treaties i IP-related activities in an international sense Sources: (a) Compendium of Patent Statistics (OECD, Paris) 2008; (b) The International Patent System: Performance Indicators (WIPO, Geneva), 2008; (c) World Competitiveness Yearbook (IMD, Lausanne), 2005; (d) Brouwer E and Kleinknecht A H, Innovative output, and a firm s propensity to patent: An exploration of CIS micro data, Research Policy, 28 (6) (1999) ; (e) Gambardella A, Guiri P and Luzzi A, The market for patents in Europe, Research Policy, 36 (8) (2007) ; (f) Archibugi D and Pianta M, Measuring technological change through patents and innovation surveys, Technovation, 16 (9) (1996) ; (g) Chung Y S, Computerization of industrial property administration in Korea, World Patent Information, 14 (2) (1992) 79-87; (h) Proposed Program & Budget for 2006/07 (WIPO, Geneva), 2007; and (i) Ganguli P, Towards TRIPs compliance in India: The Patents Amendment Act 1999 and implications, World Patent Information, 21 (4) (1999)

4 106 J INTELLEC PROP RIGHTS, MARCH 2012 One Composite Indicator The steps to be considered to obtain one composite indicator are shown in Table 2. In step 1, in case of absolute values (X i, namely, the indicators 1, 3, 5, 7, 9, 11, 13, and 15), square roots were used in order to 17, 18 narrow the gaps. In contrast, in case of relative values (X j, namely, the indicators 2, 4, 6, 8, 10, 12, 14 and 16), the values were used as such. Then, in step 2, to show the relative gaps among the 30 OECD countries more easily, the values obtained from step 1 (Y i, Y j ) were rescaled. In step 3, the rescaled values were multiplied with the weight (w i, w j ) for the respective indicators. Finally, in step 4, the multiplied values were summated, consequently obtaining the total score (Z). In terms of weighting, the Analytical Hierarchy Process (AHP), a widely used approach to scale the Table 2 Measurement method by step Data type Step 1 Step 2 Step 3 Step 4 value Sqrt(X i ) Y i =(sqrt(x i )-Min)/(Max-Min) Y i * w i value X j Y j =(X j -Min)/(Max-Min) Y j * w j X: real value; Y: normalized value; w: weights; Z: Total score; i: 1,2,3...,n; j: 1,2,3,m Table 3 Weighting value of each indicator Indicator* Weighting value Indicator* Weighting value Total 1.00 *See Table 1 for indicator name Z= 100*{Sum(Y i * w i ) +Sum(Y j * w j )} relative importance of each indicator, was applied. 19 Literature studies indicated that patent-related indicators have also been proposed by employing the AHP method. 20,21 To derive the relative weight of each indicator, 40 experts in the field of patents 15 professors and researchers, 10 patent managers in the firms, 10 patent attorneys, and 5 were surveyed. Amongst the surveyees, an inconsistency ratio of 0.1 was obtained 22, which allowed the use of the average of the whole reference group. The respective weights in each level are shown in Fig. 4. To obtain the final weight of each indicator, the weighting values were multiplied one by one. For example, in the case of inventors, the value at the third level (i.e., 0.30), the one at the second level (i.e., 0.5), and the one at the first level (i.e., 0.25) are multiplied. The same procedure was applied for each indicator, thereby obtaining the weighting values of each indicator as shown in Table 3. Comparative Analysis among 30 OECD Countries To compare the national patent performance indicators of the 30 OECD countries 23, data were obtained from various sources the secondary databases of the internationally well-known organizations and the survey of the 30 OECD countries patent offices. The data sources are presented in Table 4. Fig. 4 The weights at the three levels

5 RYU & HAN: INDICATOR FOR EVALUATING NATIONAL PATENT PERFORMANCE 107 The year 2004 was selected as the reference period because it was the most recent time span that covers most of the data. It should be noted that 2007 was the most recent available year for all the five types of data number of ; cost for patent application, registration, and maintenance; degree of IP informatization; WIPO contributions; and number After gathering data from the diverse sources and calculating the corresponding values, the patent performance indicators and the rankings of the OECD countries according to their scores (Table 6) were finally obtained. In addition, the results were compared with the number of domestic patent applications in respective countries since it is the most of international patent-related treaties because the widely used measure for investigating patent data collection 24 and survey were conducted in The descriptive statistics of each indicator are shown in Table 5. performances. Japan was ranked the first, leaving the other countries conspicuously behind; the United States, Netherlands, Switzerland, and Germany Table 4 Data sources and available years for each indicator Indicator Data source Year (1) No. of domestic & foreign patent applications WIPO IP statistics 2004 (2) No. of domestic & foreign patent applications/ WIPO IP statistics, OECD main science & technology indicators ,000 inventors* (3) No. of triadic patent applications OECD main science & technology indicators 2004 (4) No. of triadic patent applications/10,000 inventors * OECD main science & technology indicators 2004 (5) No. of patent examinations** WIPO IP statistics 2007 (6) No. of patent examinations//no. of WIPO IP statistics, Survey 2007 (7) No. of PCT international search reports # WIPO statistics on the PCT system 2004 (8) No. of PCT international search reports # /No. of WIPO statistics on the PCT system, Survey 2004 (9) No. of patents in force WIPO IP statistics 2004 (10) No. of patents in force/no. of firms in the WIPO IP statistics, OECD structural & demographic business 2004 knowledge-based industries statistics (11) Value-added in the knowledge-based industries OECD structural analysis database 2004 (12) Value-added in the knowledge-based industries/no. OECD structural analysis database, OECD structural & 2004 of firms in the knowledge-based industries demographic business statistics (13) Cost for patent application, registration and Survey 2007 maintenance (14) Degree of IP informatization Survey 2007 (15) WIPO contributions WIPO proposed program & budget for 2006/ (16) No. of international patent-related treaties WIPO website 2007 * Since data for the inventors of all 30 countries could not be obtained, the number of researchers was used as a substitute, ** Since exact number of patents requested for examination could not be determined, the number of applications was used normally, over 90 % of applicants request an examination, # Since the number of PCT international search reports is dependent upon the unit cost for the composition of the PCT international search report, the value was readjusted with the GDP per capita. Table 5 Descriptive statistics Indicator Mean S.D. Max Min (1) No. of domestic & foreign patent applications 26,846 76, , (2) No. of domestic & foreign patent applications/10,000 inventors (3) No. of triadic patent applications 840 2,209 9,286 0 (4) No. of triadic patent applications/10,000 inventors (5) No. of patent examinations 426 1,112 5,881 4 (6) No. of patent examinations /No. of (7) No. of PCT international search reports 662 2,225 11,571 0 (8) No. of PCT international search reports/no. of (9) No. of patents in force 165, ,599 1,633, (10) No. of patents in force/no. of firms in the knowledge-based industries (11) Value-added in the knowledge-based industries 533 1,176 6, (12) Value-added in the knowledge-based industries/no. of firms in the , knowledge-based industries (13) Cost for patent application, registration and maintenance (14) Degree of IP informatization (15) WIPO contributions 486, ,021 1,139,475 22,789 (16) No. of international patent-related treaties

6 108 J INTELLEC PROP RIGHTS, MARCH 2012 Table 6 The national patent performance indicator vs the number of domestic patent applications Rank Country National patent performance indicator Country No. of domestic patent applications Rank Country National patent performance indicator Country No. of domestic patent applications 1 Japan (0) 74.6 Japan 423, Italy 24.2 Hungary 2,657 2 United 54.1 United 356, Belgium (7 ) 24.0 Austria 2,514 States (0) States 3 Netherlands 52.2 Korea 140, Norway (6 ) 22.4 Finland 2,220 (12 ) 4 Switzerland 51.0 Germany 59, New Zealand 22.0 Switzerland 2,176 (15 ) (9 ) 5 Germany 45.9 Canada 38, Luxembourg 21.1 Denmark 2,015 (1 ) (9 ) 6 Korea (3 ) 38.3 United 29, Spain (7 ) 19.2 Czech 1,252 Kingdom Republic 7 United 36.4 Australia 22, Czech 16.9 Turkey 917 Kingdom (1 ) Republic(1 ) 8 Ireland 36.0 France 17, Iceland (2 ) 15.7 Ireland 845 (15 ) 9 France (1 ) 34.5 Mexico 13, Hungary (8 ) 15.2 Belgium Sweden 34.5 Poland 7, Greece (1 ) 14.6 Iceland 529 (3 ) 11 Austria (8 ) 30.7 New 6, Mexico (17 ) 12.9 Greece 514 Zealand 12 Denmark 30.4 Norway 5, Portugal (1 ) 12.7 Slovak 453 (8 ) Republic 13 Australia 26.5 Sweden 3, Slovak 12.1 Portugal 187 (6 ) Republic(1 ) 14 Canada (9 ) 25.9 Spain 3, Turkey (7 ) 10.5 Luxembourg Finland (3 ) 25.7 Netherlands 2, Poland (20 ) 9.0 Italy N/A Note: The number in the parenthesis indicates the change in the ranking and the arrow indicates rise or fall formed the second group; Korea, the United Kingdom, Ireland, France, Sweden, Austria, and Denmark the third group; and lastly Australia, Canada, Finland, Italy, Belgium, Norway, New Zealand, Luxembourg, Spain, Czech Republic, Iceland, Hungary, Greece, Mexico, Portugal, Slovak Republic, Turkey, and Poland the fourth group. The contrasting results, in comparison with previous studies, are shown in Table 6. The countries such as Australia, Canada, Norway, New Zealand, Spain, Hungary, Mexico, Turkey, and Poland were well ahead of the other countries in terms of the number of domestic patent applications; whereas, they ranked lower in terms of the composite indicator that encompasses the comprehensive aspects of patent performance. On the other hand, countries such as Netherlands, Switzerland, Ireland, Austria, Denmark, Belgium, and Luxembourg seemed uncompetitive with respect to the number of domestic patent applications; however, they ranked higher as far as the composite indicator was concerned. Finally, although the disparity between the two different measurement methods were not significantly visible for the rest, it was observed that their rankings vary slightly, ranging from 0 (Japan and United States) to 3 notches (Korea, Sweden and Finland). Conclusion In this research, a comprehensive patent performance indicator has been proposed by incorporating various activities in relation to patents. The aforementioned framework from which the individual indicators have been derived was mainly based on the prior literature; however, it also harbours practical implications. In particular, the Korean Intellectual Property Office (KIPO) has been putting emphasis on four areas, viz. creation, administration, utilization and environment, to formulate patent

7 RYU & HAN: INDICATOR FOR EVALUATING NATIONAL PATENT PERFORMANCE 109 policies. Therefore, the breakdown structure introduced in this research will help the policy-makers find the priority dimension for budget allocation. To sum up, the national patent performance indicator has been developed through three stages: in the first stage, 16 individual indicators were proposed according to the hierarchical structure. In this makeup, four dimensions creation, administration, utilization, and environment were mainly considered; next, each dimension was divided into the quantity and quality domains in order to derive indicators in a more accommodating way; and finally, a composite indicator proposed, by multiplying each indicator with the assigned weight. As an example, the cases of 30 OECD countries were compared in order to test the usefulness of the proposed indicator. As a result, Japan was ranked at the top; four nations including the United States were positioned in the second group; seven nations including Korea were distributed in the third group; and lastly, 18 economies lagged behind in the fourth group. There was a marked disparity when these results were compared with the case in which the number of domestic patent applications was used. However, since the results were dependent upon the 2004 dataset, it could not be ascertained that one country is invariably more innovative than the others. Moreover, since the patent indicators were mostly manufacturing-based measures, the knowledge service-oriented economies such as the United States were shown to get behind. Therefore, more diverse indicators representing knowledge service industries need to be incorporated in order to identify which country is more innovative than the others. This research can be extended further in two directions: firstly, since the framework proposed above is based on the notion of the actor-based innovation system, it would be possible to include more actors involved in the system. This concept was already introduced in 1997 (ref. 25). Thus the dynamics of actors behaviour within the innovation system can be simulated by adopting more actors specifically engaged in patent arena; and secondly, since parallel country-wise data is available, it would be possible to test whether and to what extent the value of the indicator is causally related to the economic growth. If the regression model is well-designed and yields a positive coefficient, it will, without a doubt, be good and atypical evidence to support the economic growth theory put forth in 1982 (ref. 26). Lastly, as mentioned above, this study is meaningful in that various aspects of patenting activities were accommodated in order to examine a nation s patent performance in a more integrated manner; however, because national-level data designed for international comparisons were insufficient, the dimensions for deriving indicators had to be delimited. Therefore, if large-scale comparable datasets become available from the international organizations such as the OECD and the WIPO, more dimensions could be accommodated in the future. That is, if more dimensions or sub-dimensions can be supplemented with a regularly accumulated dataset, the results and implications will be enriched. Acknowledgement This research was funded by the KIPO in References 1 Science and Technology Indicators (OECD, Paris), European Report on S&T Indicators (EC, Brussels), Grupp H and Mogee E, Indicators for national science and technology policy: How robust are composite indicators? Research Policy, 33 (9) (2004) A set of patents granted in three patent offices the United States Patent & Trademark Office, the European Patent Office and the Japan Patent Office - to protect a single invention. 5 Griliches Z, Patent statistics as economic indicators: A survey, Journal of Economic Literature, 28 (4) (1990) Grupp H and Schuber T, Review and new evidence on composite innovation indicators for evaluating national performance, Research Policy, 39 (1) (2010) Trajtenberg M, in Patents, Citations and Innovation: A Window on the Knowledge Economy, edited by A Jaffe and M Trajtenberg (MIT Press, Cambridge), 2002, p The International Patent System: Performance Indicators (WIPO, Geneva), Fu X and Yang Q G, Exploring the cross-country gap in patenting: A stochastic frontier approach, Research Policy, 38 (7) (2010) Global Competitiveness Report (WEF, Geneva), World Competitiveness Yearbook (IMD, Lausanne), Archibugi C and Coco A, Measuring technological capabilities at the country level: A survey and a menu for choice, Research Policy, 34 (2) (2005) Hirschey M and Richardson V J, Are scientific indicators of patent quality useful to investors? Journal of Empirical Finance, 11 (1) (2004) Lettl C, Rost K and Warburg L, Why are some independent investors heroes and others hobbyists? The moderating role of technological diversity and specialization, Research Policy, 38 (2) (2009) Criscuolo P and Verspagen B, Does it matter where patent citations come from? Inventor vs examiner citations in European patents, Research Policy, 37 (10) (2008) The Knowledge-Based Economy (OECD, Paris), 1996.

8 110 J INTELLEC PROP RIGHTS, MARCH In the case of economic size measures such as GDP and population, the square root of the values were used in order to narrow the gaps since absolute values do not represent the gaps per se, Atkinson A B, Income distribution in Europe and the United States, Oxford Review of Economic Policy, 12 (1) (1996) In the case of indicator 13, the real value was converted into rank because lower the value, the better. 19 Saaty T L, The Analytic Hierarchy Process (McGraw-Hill, New York), Razgaitis R, Early-Stage Technologies: Valuation and Pricing (Wiley, Brisbane), Chiu Y J and Chen Y W, Using AHP in patent valuation, Mathematical and Computer Modeling, 46 (7-8) (2007) If the value is under 0.1, the results are considered to be consistent; if between 0.1 and 0.2, the results are accepted based on the interpretation of the researchers; and if over 0.2 the survey needs to be re-conducted. 23 Australia, Austria, Belgium, Canada, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Japan, Korea, Luxembourg, Mexico, Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Spain, Sweden, Switzerland, Turkey, United Kingdom, and the United States. 24 At the time the research was conducted, the data for WIPO contributions and the number of international patent-related treaties were available only at the WIPO website. 25 Nelson R and Winter S, An Evolutionary Theory of Economic Change (Havard University Press, Cambridge), Edquist C and Johnson B, in Systems of Innovation: Technologies, Institutions and Organizations, edited by C Edquist (Pinter, London), 1997, p