Commercialization of Patents in the Czech Republic: What can we learn from linking R&D, patent and licensing datasets?

Transcription

1 Commercialization of Patents in the Czech Republic: What can we learn from linking R&D, patent and licensing datasets? Martin Mana Czech Statistical Office Karel Eliáš Czech Statistical Office Martin Srholec Centre for Technology, Innovation and Culture, University of Oslo 21 September 2007 Conference on Patent Statistics for Policy Decision Making 2-3 October 2007 Venice, Italy Abstract Market for technology has expanded over since the early 1990s. Although quality and availability of data on R&D, patents and overall innovation activity have increased enormously over the last decade, reliable and internationally comparable evidence on licensing (and licensees) remains rare. To fill in this gap, the Czech Statistical Office has launched annual surveys on licensing in Using micro data from this unique dataset, we bring evidence on licensing activity in the Czech Republic over the period In addition, we merge the data on licensing with information at the firm-level from the business register, annual R&D surveys and patent statistics. The concluding section outlines plans for econometric analysis of this micro data in the future. Keywords: Patent, license, commercialization, statistics, Czech Republic. 1

2 1. Introduction Market for technology has expanded over the 1990s (Arora et al. 2001) and marketbased international flows of technology have increased even more rapidly (Athreye and Cantwell 2007, Mendi 2007). Although some evidence on the market for technology exists, a systematic (and internationally harmonized) effort to measure commercialization of inventive activity is lacking. Most of the available data comes from secondary sources that are not tailored for this purpose or from surveys pursued by individual research projects, which remain fragmented and are rarely repeated over time. A lack of statistics on licensing (and licensees) is in a sharp contrast to the increasing quality and (availability) of data on research activity from the R&D surveys (OECD 2002) and the innovation process in general from the Community Innovation Surveys (OECD 2005). Only data on international payments for licensing fees - the socalled technology balance of payments are collected on a systematic basis in a large number of countries, but this information is designed for the purpose of the overall balance of payments by the IMF methodology (IMF 1992). Also this data is limited to the aggregated financial flows, while no details are available on characteristics of the organizations that receive (or pay) the licensing fees. More detailed information has been collected by a pilot survey of licensing organized by OECD (OECD 2004), but this project focused on commercialization of knowledge generated in the public sector and therefore very little evidence has been collected on licensing in the business sector. Similarly some aspects of inventive activity have been studied extensively, while for others we have surprisingly limited evidence. As pointed out by Giuri and Mariani (2006), we do not know much about the patent applicants themselves, nature of the inventive process that leads to the patent, and most importantly we have very limited information on whether (and why) are patents licensed to other organizations. Although the 2

3 PatVal survey organized by them and others helped to fill some of these gaps, there is a need for more evidence on the characteristics of patentees and the use of patents. Moreover, the analysis has been focused on evidence from developed countries. Without more evidence on the emerging markets for technology in countries behind the technology frontier, we cannot fully understand the process of technological catching-up and growth in the enlarged EU (and the world economy in general). The aim of this paper is to provide fresh evidence along these lines. Using data from the Czech Republic, we demonstrate what we can learn from merging information on inventive activity and the market for technology from various sources/surveys. Section 2 provides new evidence on characteristics of patentees by linking the patent database with the business register. Section 3 presents the Czech surveys of licensing and illustrates how the market for technology can be measured on a systematic basis by statistical offices. Section 4 provides new evidence on the nature of the inventive activity by linking information from R&D surveys, the patent database and the survey of licensing. Since this is the first time these various datasets have been merged (at least in the Czech Republic) and only descriptive analysis of the data have been conducted due to time (and other) constraints so far, the last section outlines agenda for econometric analysis of these data in the future. 2. Patent statistics A patent may be granted to a firm, an individual or a public body by a patent office. 1 Statistics of patents can be used to analyse national inventive activity. However, not all inventions are patented (e.g. a computer software) and not all patents have the same 1 A patent is an intellectual property right issued by authorized bodies to inventors to make use of, and exploit their inventions for a limited period of time (generally 20 years). The patent holder has the legal authority to exclude others from commercially exploiting the invention (for a limited time period). In return for the ownership rights, the applicant must disclose the invention for which protection is sought. (OECD 2006) 3

4 economical value. For example, patents are often used for assessing the performance of application-oriented types of R&D. Although patents are not perfect measures of inventive activity, they cover a considerable part of it and are therefore probably the most frequently used in the literature. 2.1 Patent data available from the national patent office (IPO CR) The Industrial Property Office of the Czech Republic (IPO CR) is responsible for the administration of industrial rights in the Czech Republic. Patent documents available from the IPO CR provide detailed information including the year of invention, technical classification, a country of the applicant, a country of the inventor etc. In addition, data about patents granted by the IPO CR (patents having effect in the Czech Republic) are available at the following breakdowns: By the type of patent registration (National patents or European patents 2 validated in the Czech Republic) By the patent holders country of residence By the International Patent Classification (IPC) 3 sections and classes By selected technical fields such as high-tech or biotechnology By region of the patent holders (CZ NUTS 3 regions) By cross combination of above categories 2 A European patent can be obtained for all the European Patent Convection (EPC) countries by filing a single application at the EPO in one of the three official languages (English, French or German). European patents granted by the EPO have the same legal rights and are subject to the same conditions as national patents (granted by the national patent office). It is important to note that a granted European patent is a bundle of national patents, which must be validated at the national patent office for it to be effective in member countries. The validation process could include submission of a translation of the specification, payment of fees and other formalities at the national patent office. This is because once a European patent is granted; competence is transferred to the national patent offices. (OECD 2006) 3 The International Patent Classification, which is commonly referred to as the IPC, is based on an international multilateral treaty administered by WIPO. The IPC is an internationally recognised patent classification system, which provides a common classification for patents according to technology groups. The IPC is a hierarchical system in which the whole area of technology is divided into a range of sections, classes, subclasses and groups. There are eight sections that are broken down into classes and subclasses. IPC is periodically revised in order to improve the system and to take account of technical development. The current (eighth) edition of the IPC entered into force on 1 Janua006. 4

5 The following tables show basic information available from the IPO CR sources for granted patents having effect in the Czech Republic from 2000 onwards (based on grant date as a reference period). Table 2.1: Patents having effect in the Czech Republic, Granted in the national way of which to the Czech applicants of which to the foreign applicants EP validated in CR Total Source: Industrial Property Office of the Czech Republic 2006 Annual report Over the past six years (except for the year 2004), the number of patents granted at the Czech national patent office (IPO CR) has continued to rise. The total number of new patents granted in the Czech Republic more than doubled between 2000 and 2006 and increased by 45% between 2005 and The continuing significant increase of granted European patents (EP) valid also for the Czech Republic was the essential factor. The number of patents granted in the national way declined (both patents granted to domestic and foreign applicants) between the year 2006 and In 2005 the Czech applicants accounted to 15% of total number of granted patents in the CR compare to only 8% in As already mentioned this decline was caused by accession of the Czech Republic to the European patent system. By contrast, the number of patents granted by the European Patent Office (EPO) with effect in the Czech Republic (validated the patent s owner delivered translation of the patent document to the Czech language and paid the administrative charges) has significantly increased. In 2006 the number of EP patents effective in the Czech Republic (validated the patent s owner delivered translation of the patent document to the Czech language and paid the administrative charges) exceeded the number of patents granted in the national way. 5

6 Figure 2.1 and table 2.2 provide detailed information about number of patents granted in the Czech Republic by holders residence for the years The number of patents annually granted at the IPO CR increased substantially since the year 2000 for many EU countries (e.g in Germany from 465 patent granted in 2000 to in 2006 or in Italy from 30 to 155). The rising interest of German applicants to protect their inventions (intellectual property) in the Czech Republic shows evidence of growing German investment in the Czech Republic. It also reflects to some extent our market attractiveness for foreign firms. It is expected that number of new patents granted to foreign residents will grow quickly. The main reason is ratification of the European Patent Treaty by the Czech Republic making easier for the foreign applicants to fill a patent in the Czech Republic. Other reason is continuing process of the internationalisation of our market. Figure 2.1: Share of main countries (regions) in total number of new patent granted in the Czech Republic, (%) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% Rest of the world (others) United States Other European countries Germany Czech Republic 0% Source: Czech Statistical Office 2007 based on database of the IPO of the CR. 6

7 Table 2.2: Patents granted in the Czech Republic by holder s country, Czech Republic (CR) Austria (AT) Belgium (BE) Denmark (DK) Finland (FI) France (FR) Germany (DE) China (CN) Ireland (IE) Italy (IT) Japan (JP) Korea (KR) Luxembourg (LU) Netherlands (NL) Russian Federation (RU) Slovakia (SK) Spain (ES) Sweden (SE) Switzerland (CH) United Kingdom (UK) United States (US) Other Total Source: Industrial Property Office of the Czech Republic 2006 Annual report Figure 2.2 shows distribution of valid patents in the Czech Republic by holders country as of In addition, table 2.3 provides information about type of the patent registration used by patentees from different countries. There was a total valid (active) patents in the Czech Republic at the end of The number valid patents granted to foreign applicants (11 371) is more than six time higher that to the Czech applicants (1 807). The entities from Germany had the highest innovative performance in the Czech Republic, measured by their share in total valid patents on the Czech territory (30 %). Nearly four thousands (3 910) German residents (mainly firms) were owners of an active patent in the Czech Republic as of It is even higher than the number of 7

8 patents owned by the Czech residents. The United States (13 %), followed by France (8 %) and Switzerland (7 %) are other countries with high shares. Figure 2.2: Distribution of valid patents in the CR as of by holders country Number (29,7%) % of total patents Germany Czech Republic United States France Switzerland Austria Italy UK Netherlands Japan Sweden Belgium Denmark Finland Spain Slovakia Luxembourg Korea Ireland Russia Portugal Greek China Other Source: Industrial Property Office of the Czech Republic 2006 Annual report 20% 18% 16% 14% 12% 10% 8% 6% 4% 2% 0% Table 2.3: Patents having effect in the CR by holders country as of DE CR US FR CH AT IT UK National patents European patents Total NL JP SE BE DK FI Other Total National patents European patents Total Source: Industrial Property Office of the Czech Republic 2006 Annual report The following three tables includes data relevant only to the Czech applicants and should allow getting more insight into important features of technical change based on IPC 8

9 classification (table 2.4 and table 2.5) and regional distribution of the invention activity (table 2.6). The data included in these tables were grouped by the Czech Statistical Office based on dataset from the IPO CR and can slightly differ from the data issued by this organisation (a different counting system was used). Table 2.4: Patents granted to Czech applicants by the IPC sections, IPC Section title Human necessities Performing operations, transporting Chemistry, metallurgy Textiles, paper Fixed constructions Mechanical engineering, weapons, Physics Electricity Total Note: Patent counts are recorded by the grant date and fractional counting is used. Source: Czech Statistical Office based on database of the IPO of the CR. Table 2.5: Patents granted to Czech applicants in selected technology areas, Hi-tech group, total 10,5 15,0 9,0 5,0 10,0 9,0 13,0 71,5 Aviation 1,0 2,0 3,0 0,0 0,0 3,0 0,0 9,0 Computer and automated business equipment 1,0 3,0 0,0 2,0 2,0 3,0 0,0 11,0 Communication technology 1,0 2,0 1,0 0,0 3,0 1,0 3,0 11,0 Lasers 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 Micro-organism and genetic engineering 7,5 8,0 4,0 3,0 5,0 2,0 10,0 39,5 Semiconductors 0,0 0,0 1,0 0,0 0,0 0,0 0,0 1,0 Biotechnology, total 5,5 8,0 4,0 3,0 4,0 2,0 8,0 34,5 Note: Patent counts are recorded by the grant date and fractional counting is used. Source: Czech Statistical Office 2007 based on database of the IPO of the CR. 9

10 Table 2.6: Patents granted to Czech applicants by regions, Hl. m. Praha 66,7 69,8 72,0 83,9 77,4 113,6 90,8 574,3 Jihomoravský 2,5 10,3 9,2 3,2 5,0 11,0 13,3 54,5 Moravskoslezský 26,5 43,0 28,5 33,2 33,5 26,5 30,3 221,5 Středočeský 1,0 1,0 0,0 1,0 0,0 4,2 5,0 12,2 Liberecký 8,5 5,0 3,4 12,0 22,0 17,0 10,0 77,9 Zlínský 26,8 17,0 18,3 24,3 18,0 24,0 25,7 154,2 Olomoucký 36,5 16,7 19,0 19,8 25,3 27,7 24,0 169,0 Pardubický 18,7 6,0 8,3 8,0 14,0 15,5 10,5 81,0 Královéhradecký 17,8 6,8 17,5 17,0 12,2 16,3 5,5 93,1 Ústecký 15,2 11,0 14,5 7,5 11,5 13,8 8,0 81,4 Plzeňský 17,5 29,0 12,8 22,7 40,0 51,3 18,3 191,6 Vysočina 8,8 12,5 10,4 7,8 13,5 10,4 9,5 73,0 Jihočeský 7,2 7,0 10,5 3,2 3,3 6,5 6,0 43,7 Karlovarský 14,8 6,0 14,0 14,5 15,7 8,3 8,0 81,3 Total ,6 Note: Patent counts are recorded by the grant date and the applicant s region of residence; fractional counting is used. Source: Czech Statistical Office 2007 based on database of the IPO of the CR. 2.2 New patent statistics (breakdowns) Since the patent database includes the name of the patentee, we were able to merge (link) the patent dataset with information from the Czech Business Register by identifying the ICO - identification number of the patentee. This allowed as to start merging patent data. Commencing by the one on patents granted4. However, only patentees registered in the Czech Republic can be identified in the business register, which implies that subjects with residence abroad cannot be included into this analysis. In addition, we are not able to identify the ICO of individual persons that are patentees, as they do not use this number. The following table provides an overview of the sample along this distinction. 4 The patent database (patent granted) has been connected with the business register only for the period over Additional work is under way on extending the dataset also to the patent applications and utility models and to the period before Note that utility models are very popular means of protection among the Czech inventors. For example, there were 938 applications for utility models, but only 641 applications for patents in

11 Table 2.7: Identification of Czech patentees, No of the Czech patentees, total Patentees with ICO Patentees without ICO % of patentees without ICO Source: Czech Statistical Office 2007 The business register provides us with additional information about the Czech applicants (patent holders) that enables us to breakdown the patent data as follows: By institutional sector (business sector of which foreign affiliates and national firms; government sector of which research institutions and higher education sectors/universities) By size category given by the number of employees (only for firms) By industry given by CZ-NACE Rev. 1.1 (only for firms) The following set of tables and figures illustrates the basic results from our work on merging patent data with the Czech Business Register. Information for numbers of Czech patentees with new patent granted and numbers of new patents assigned to the Czech entities over period are presented. Figure 2.3 shows the importance of different Czech entities (institutional sectors) in patenting activity. The business sector (national firms and foreign affiliates patentees) accounted for two third of the new patents granted in the Czech Republic in As a contrast, shares of public research institutions and universities were relatively low (4.9 %, respectively 5.9 % for universities). The same share for individuals reached in 2006 value of 15,2%. 11

12 100% Figure 2.3: Share of new Czech patent granted by institutional sector, (%) 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 60% 61% 62% 55% 51% 53% 48% National firms Foreign affiliaties Individuals Government insitutions Universities Source: Czech Statistical Office 2007 based on database of the IPO of the CR. Detailed numbers about Czech entities with new patent granted and number of patents registered to these entities by institutional sector are available in the next tables. Table 2.8a: Number of Czech entities with new patent by institutional sector, Business sector (firms) Foreign affiliates National firms Individuals Government sector Research institutions Universities Total (patentee) Table 2.8 b: Number of new patents by institutional sector, Business sector (firms) Foreign affiliates National firms Individuals Government sector Research institutions Universities Total (patents) Note: Patent counts are recorded by the grant date and fractional counting is used. Source: Czech Statistical Office 2007 based on database of the IPO of the CR. 12

13 Although the government research institutions accounted only 6.8 % of total patents granted to the Czech entities over period , they reached at the same time the highest patent activity, measured by an average number of new patents granted per a single patentee (2.4 new patent granted per government research institution patentee). In foreign affiliates this figure reached a value of 2.0, for the national firms 1.1; for universities 1.9 and individuals 0.7. The following set of tables refers only to patents assigned to the Czech firms and provides evidence on patent activity by firm size and industry. We distinguish the industries as very small (0-9), small (10-49), medium (50-249) and large (250 and more) firms (number of employees in brackets). Most patents are granted to large firms (figure 2.4 b). Somewhat surprisingly, however, more than a third (28.3%) of patents tend to be acquired by very small firms. In addition, if we concern only the number of patentee (firms) with new patent granted (figure 2.4a) than we can find out even a higher significant of very small firms. The share of very small firms accounted for more than half (55.0%) of the total new patents granted to the Czech firms over period Figure 2.4: Share of Czech firms with new patent and share of new patents in the CR by firm size category, (%) 60% 50% 40% 30% 20% a) Firm with patent granted b) Patents granted to firms very small small medium 60% 50% 40% 30% 20% 10% large 10% 0% % 13

14 Again, detailed numbers about Czech entities with new patent granted and number of patents registered to these entities by institutional sector are available in the following two tables. Table 2.9a: Number of Czech firms with new patent by size category, employees employees employees and more employees Total (firms) Table 2.9b: Number of new patents in the CR by size category, employees employees employees and more employees Total (firms) Note: Patent counts are recorded by the grant date and fractional counting is used. Source: Czech Statistical Office 2007 based on database of the IPO of the CR. Probably it is not surprise that the largest firms (patentees) had not only highest number of new patents granted (561) over period but also the highest patent activity (on average 1.5 new patent granted per a large firm/patentee). For the medium, small and very small firms were these values: 1.1; 0.9; and 0.7 respectively. A conventional view on structure of patenting activity used in the literature is based on the breakdown of patents by the IPC categories. Although the IPC classification provides detailed information on distribution of patents by scientific (or technology) fields, it is largely incompatible with the most widely used industrial classifications, which makes it difficult to compare structure of patents to other aspects of innovation (and other) 14

15 activities. Linking the patent database with the information from the business register offers us an unique opportunity to analyse the industrial composition of patenting activity in the Czech Republic. Table 2.10: Patents granted in the CR to Czech firms by industry over Industry Code Number of % of total patents patents AGRICULTURE, FORESTRY AND FISHING 01/05 9 0,6 MINING AND QUARRYING 10/ ,7 TOTAL MANUFACTURING 15/ ,8 Food, beverages, tobacco 15/ ,2 Textiles, apparel, fur, leather 17/ ,1 Wood, paper, printing, publishing 20/ ,2 Petroleum, nuclear fuel ,1 Chemical products ,7 of which: Pharmaceuticals ,4 Rubber, plastics ,4 Non-metallic mineral products ,8 Basic metals ,5 Fabricated metal products (excl. machinery) ,8 Non-electrical machinery n.e.c ,1 Office and computing machinery ,1 Electrical machinery ,7 Radio, TV and communication equipment ,9 Scientific and precision instruments ,5 Motor vehicles ,1 Other transport equipment ,2 Other manufacturing industries n.e.c. 36/ ,7 ELECTRICITY, GAS & WATER 40/ ,9 CONSTRUCTION ,0 COMMERCE, HOTELS, RESTAURANTS 50/ ,7 TRANSPORT AND COMMUNICATIONS 60/64 9 0,6 FINANCIAL INTERMEDIATION 65/67 1 0,1 BUSINESS (MARKET) SERVICES 70/ ,1 of which Research and Development ,4 OTHER SERVICES 75/ ,5 Total ,0 Source: Czech Statistical Office 2007 based on database of the IPO of the CR. Table 2.10 shows patents granted by industry (according to the NACE, rev. 1.1 classification). More than half of patents (59,8%) were granted to manufacturing firms, of which the chemical, non-electrical machinery n.e.c. and transport equipment industries are the most prevalent. It is interesting to note that this is broadly in line to what should be 15

16 expected from the OECD classification of industries by technology intensity, although the OECD taxonomy is based on (direct and indirect) R&D intensity of industries. Other important sectors are the R&D services and other business services (mainly in architectural, engineering and related consultancy services). Somewhat surprising is the relatively high share of retailing, wholesale and hotels category, which is most likely due to classification of firms into industries by the principal economic activity. Some firms classified in these sectors have a portfolio of activities that include R&D and other knowledge-intensive services. New patents granted per single Czech patentee ranged from an average some 1.3 in total business sector to almost 10 in pharmaceutical industry. Surprisingly, there were only 1.8 new patents granted per patentee in Research and development industry (CZ NACE 73). More information about the share of patents granted over period by main industries can be found in the next figure. Figure 2.5: Share of patents granted in the CR to Czech firms by main industry category, (%) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 58% 55% 52% 66% 63% 60% 64% 60% Manufacturing Services (without R&D - NACE 73) R&D industry (NACE 73) Source: Czech Statistical Office 2007 based on database of the IPO of the CR. 16

17 The last table in this chapter gives an overview on the top 10 Czech patentees over the period Most patents have been granted to the largest Czech pharmaceutical company Zentiva, a.s. Other firms on the list include the leading carmaker Skoda Auto, a.s., the crystal jewellery maker Preciosa, a.s. and Rieter Elitex, a.s. in the machinery industry and others. Also two universities and three government research institutes are on the list. A comparison of the total number of patents in which the patentees have participated to the number of patents by fractional assignments reveals that firms tend to invent alone, whereas universities and research institutes often develop new technologies in cooperation with other organizations. Table 2.11: Top 10 Czech patentees over Patentee Patents by fractional assignments The total number of patents Zentiva, a.s. (Pharmaceuticals) České vysoké učení technické v Praze (University) Škoda auto a..s. (Motor vehicles) Preciosa a.s. (Crystal fashion jewellery components) Vysoka skola chemicko-technologicka (University) Rieter Elitex a.s. (Machines and engineering components) Ústav makromolekulární chemie AV ČR (Research institute) Ústav organické chemie a biochemie AV ČR (Research institute) VÚTS Liberec a..s. (Research institute of textile machinery) Trustfin a.s.(knitting machines) Source: Czech Statistical Office 2007 based on database of the IPO of the CR. 17

18 3. Licensing statistics A licence 5 is one of the possibilities to use industrial rights and intellectual property on a commercial basis. Licensing transactions are a mean by which technology can be transferred from one party to another. Although the details of individual licensing agreements vary, they can include terms referring to technical support, training and other assistance to be provided by the licensor to the licensee. They can enable the licensee to acquire the right to use new technology without having to undertake costly R&D and to capitalise on the licensor s reputation and expertise. In exchange, the licensor derives fees and royalties. Better understanding of the economic use and impact of the intellectual property (mainly patents and utility models) requires complementary information to be gathered, not available in patent data bases. The Czech Statistical Office collected data on licence agreements by the annual statistical survey Lic 5-01 since the reference year The main focus of this survey is to monitor the number of valid licence agreements (purchased and sold) in order to protect industrial property, as well as, the value of licence fees (paid and received) in the Czech Republic in the respective year. Apart from the total number of valid licence agreements, the survey also inquires about newly concluded licence agreements, type of licence agreements and country of the contracting partner. The aforementioned survey Lic 5-01 provide monitors and classifies the concluded license agreements and the volume of license fees paid or received depending on the reporting unit (licenser or licensee) according to: 5 License is the means by which the owner of a patent gives permission to another person to carry out an action, which, without such permission, would infringe on the patent. A licence can thus allow another person to legitimately manufacture, use or sell an invention protected by a patent. In return, the patent owner will usually receive royalty payments. A license, which can be exclusive or non-exclusive, does not transfer the ownership of the invention to the licensee. 6 Over the period the Czech R&D Survey was used for collecting data on license agreements. 18

19 Industrial classification (ISIC), i.e. core economic activity of the entities that entered a licence agreement Size of enterprise defined with respect to the number of employees: 0, 1-9, 10-49, , , more than 500 employees Firm domicile defined by CZ-NUTS 3 (14 regions in the Czech Republic) Main high-tech and knowledge-intensive groups defined by ISIC Country of contracting partner, both, as licensor and licensee 3.1 Number of the Czech entities with valid license agreement, Reporting units of this survey represent all entities that sold or acquired a license (licensors and licensees) for any form of intellectual property (software licence agreements and trading license agreements for import and export goods and services are not included). Based on our national survey, it is estimated that there were some 143 active licensors in the Czech Republic in 2006 (entities that provided an active license/entities with licensing out activity). The number of licensees in the Czech Republic was the same year 456. Table 3.1 shows detailed numbers of the Czech licensors and licensees breakdown by institutional sector over period Number of reporting units that purchased rights to intellectual property in the Czech Republic increased by 30 % in 2006 compared to the year 2005, when 110 reporting units were recorded to do so. These numbers are estimations based on our annual survey. The results show superiority of entities from the business sector in licensing activities. In 2006 nearly all licensees (99%) were firms and also the share of firms in the total number of active licensors was very high 19

20 (90%). Nearly four out of ten licensors had a valid license agreement for using a patent or utility model 7. Table 3.1a: Number of Czech licensors by institutional sector, Business sector (firms) with license for using a patent or utility model Individuals with license for using a patent or utility model Government research institutions with license for using a patent or utility model Universities with license for using a patent or utility model Total with license for using a patent or utility model Table 3.1b: Number of Czech licensees by institutional sector, Business sector (firms) with license for using a patent or utility model Individuals with license for using a patent or utility model Government research institutions with license for using a patent or utility model Universities with license for using a patent or utility model Total with license for using a patent or utility model Source: Czech Statistical Office Number of valid licence agreements, In 2006, a total of 981 licences were effectively sold in the Czech Republic. Number of licences sold increased by 83 % in 2006 compared to the 2003, where 537 licences sold were recorded. The total of licences were effectively purchased in In 2006, the most important provider of licences in the Czech Republic was, like in the previous years, business sector, which contributed to the total sold licenses by 99.6 %. Figure 3.1 shows that while in 2003 the share of licences under patents or utility models 7 Not only patents but also utility models can be considered as a relatively objective indicator of the iinnovation activity level in the Czech Republic. 20

21 were 23 %, three year later (2006), this share increased to 28 %. Generally, the market for licensing in the Czech Republic is expanding in terms of number of deals. Figure 3.1: Number of valid licences agreements in the CR, Licenses purchased, total Licenses sold, total Licenses purchased on patents or utility models Licenses sold on patents or utility models Source: Czech Statistical Office 2007 Detailed numbers about valid licences sold (purchased) by institutional sector are available in the following two tables. Table 3.2a: Number of valid licences sold by Czech entities, Business sector (firms) Individuals Government research institutions Universities Total Table 3.2b: Number of valid licences purchased by Czech entities, Business sector (firms) Individuals Government research institutions Universities Total Source: Czech Statistical Office

22 3.3 Value of licences fees, Licence fees paid by licensee to licensor in the respective year (not starting from the validity of the licence agreement) are monitored. While the licence fees are declared on the basis of the respective year (only appear in the balance of the year, in which they were settled), the actual number of licence agreements is cumulatively monitored over the years (throughout the validity of the agreement). The same rule apply for licence fees paid to the licensor by the licensee. Figure 3.2 shows the value of licensee s fees received and paid by the Czech entities (mainly firms). License fees paid by the Czech entities are much higher than the revenue generated from licenses issued by the Czech entities. In 2006, the Czech entities paid twice more for license fees that they received. Over e time, we can also see that fees paid growing much quicker than fees received. While in 2004, the Czech entities paid together million CZK just in two year it was already million CZK, 3.3 times more. On contrary, the revenue from licenses fees increased from million CZK in 2004 to million CZK in 2006, 1.7 times more. Figure 3.2: Value of licences fees (received/paid) in the CR, Million CZK Fees paid, total Fees received, total Fees paid for licenses on patents or utility models Fees received from patents or utility models licences Source: Czech Statistical Office

23 Detailed information about the licence market (on which on patents or utility models) in the Czech Republic (licence fees received and paid) by institutional sector is provided in the following tables. Table 3.3a: Value of licence fees received by Czech entities, Million CZK Business sector (firms) of which on patents or utility models Government research institutions of which on patents or utility models Universities of which on patents or utility models Total of which on patents or utility models Table 3.3b: Value of licence fees paid by Czech entities, Million CZK Business sector (firms) of which on patents or utility models Government research institutions of which on patents or utility models Universities of which on patents or utility models Total of which on patents or utility models Source: Czech Statistical Office 2007 Data on licensing revenue shows that in 2006 the average revenue of the Czech licensor per license sold was 2 million CZK. For license under a patent it was even 2.9 million CZK. The same indicator for license fees paid by the Czech entities reached in 2006 value 5 times higher (10 million CZK). 23

24 Figure 3.3: Value of licence fees received/paid on patents by Czech entities by their country of the contractual partner, 2006 Million CZK Licence fees received Source: Czech Statistical Office 2007 To be further developed. US Germany Japan France CZ Netherlands UK Belgium Austria Italy Poland Canada China Belarusia Turkey Other Licence fees paid 24

25 3.4 Licensing under patents and utility models by Czech firms, The following set of tables and figures refers only to licences sold under patent and utility models and provides evidence on licensing-out by firm size and industry. Results will be further developed later. 1) By size categories given by the number of employees Table 3.4a: Number of Czech firms licensing-out patents and/or utility models by size category, Very small (0-9 employees) Of which with new one Small (10-19 employees) Of which with new one Medium (50-249employees) Of which with new one Large (250 and more employees) Of which with new one Total (All Czech firms) Of which with new one Table 3.4b Number of valid sold licences under patent or utility model by size category, Very small (0-9 employees) Of which newly sold Small (10-49 employees) Of which newly sold Medium ( employees) Of which newly sold Large (250 and more employees) Of which newly sold Total (All Czech firms) Of which newly sold Table 3.4c: Licence fees received under patent or utility model by Czech firms by size category, Very small (0-9 employees) Of which from new licences Small (10-49 employees) Of which from new licences Medium ( employees) Of which from new licences Large (250 and more employees) Of which from new licences Total (All Czech firms) Of which from new licences Source: Czech Statistical Office

26 Figure 3.4: Share of number sold licences and share of licences fees received under patent and/or utility model by Czech firms by size category, (%) a) Share of sold licences b) Share of licences fees 100% 100% 80% Very small 80% 60% Small 60% 40% 20% 0% Medium Large % 20% 0% Source: Czech Statistical Office ) By industry (CZ-NACE Rev. 1.1) Table 3.5a: Number of Czech firms licensing-out patents and/or utility models by industry, Manufacturing Of which with new one Services (without R&D - NACE 73) Of which with new one R&D industry (NACE 73) Of which with new one Other industries Of which with new one Total (All Czech firms) Of which with new one Table 3.5b Number of valid sold licences under patent or utility model by the industry (CZ- NACE), Manufacturing Of which newly sold Services (without R&D - NACE 73) Of which newly sold R&D industry (NACE 73) Of which newly sold Other industries Of which newly sold Total (All Czech firms) Of which newly sold

27 Table 3.5c: Licence fees received under patent or utility model by Czech firms by industry (CZ-NACE), Manufacturing Of which from new licences Services (without R&D - NACE 73) Of which from new licences R&D industry (NACE 73) Of which from new licences Other industries Of which from new licences Total (All Czech firms) Of which from new licences Source: Czech Statistical Office 2007 Figure 3.5: Share of number sold licences and share of licences fees received under patent and/or utility model by Czech firms by main industry groups, (%) a) Share of sold licences b) Share of licences fees 100% 80% 60% Services Manufacturing 100% 80% 60% 40% 20% 0% R&D industry 40% (NACE 73) 20% Other industries 0% Source: Czech Statistical Office 2007 To be further developed. 27

28 3.5 Licensing activities by type of license for the year 2006 The last set of tables in this chapter refers to both licensing-out and licensing-in activities breakdown in terms of license subject (patent licences; utility models licences; industrial designs licences; know-how licences; the trademark etc.). Also in this case results will be further developed later. Only numbers are provided here. Table 3.6a: Data on licensing out by type of license, 2006 New plant varieties, animal breeds Franchisi Trademarks ng Total Entities with valid sold license Of which with new one Number of valid sold licences Of which newly sold Licence fees received (Million CZK) Of which from new licences Table 3.6b: Data on licensing in by type of license, 2006 Patents Utility models Knowhow Industrial designs Entities with valid sold license Of which with new one Number of valid sold licences Of which newly sold Licence fees received (Million CZK) Of which from new licences Patents Utility models Knowhow Industrial designs Entities with valid purchased license Of which with new one Number of purchased valid licences Of which newly sold Licence fees paid (million CZK) Of which for new licences New plant varieties, animal breeds Franchisi Trademarks ng Total Entities with valid purchased license Of which with new one Number of purchased valid licences Of which newly sold Licence fees paid (million CZK) Of which for new licences Source: Czech Statistical Office 2007 To be further developed. 28

29 4. R&D, patents and licensing Since the ICO identification number is available for both of the patent and licensing datasets presented above as well as for the R&D statistics collected by the Czech Statistical Office, we can connect these datasets with each other. This provides us with unique evidence on interrelationships between the various stages of the inventive and licensing activity in the Czech Republic over A limited overlap between different surveys is often a serious problem for merging datasets. In order to keep as many observations in the analysis as possible, we first consider all firms that have been ever approached by the annual R&D surveys. Many of these firms answered to the questionnaire, but some did not (at least in most years). At this step of the analysis we take into account all of them, because we know some important information about them from the business register. Our sample is therefore a set of firms that (for various reasons) the Czech Statistical Office believes to be engaged in internal R&D activity. Since the patent database is exhaustive, we can observe how many of these firms have received a patent over the period. Table 4.1 shows the results. Altogether information about almost 6,000 organizations can be used for this exercise. On average every fifth of these firms have been granted a patent (based on fractional assignments) over the period. As can be expected, sharp difference in the patenting intensity came out by size categories. Large organizations are almost eight-times more likely to have a patent than their small counterparts. Sectoral differences are also according to the expectations. Firms in industry are more patent intensive than firms in market services and organizations in the sector of public services. However, not much difference has been detected between the two latter groups. Domestic publicly-owned organizations (such as universities) and foreign affiliates are more than 29

30 twice as much patent intensive compared to domestic-owned firms, which points to superior inventive capabilities in foreign- compared to domestic-owned firms. Table 4.1: Patenting intensity (number of patents per firm), Average St. dev. Observations Total ,987 Size Large (250 and more) ,051 Medium (50-249) ,811 Small (1-49) ,125 Sector Industry ,782 Market services ,884 Public services Ownership Domestic private ,533 Domestic public Foreign affiliates ,231 Note: Number of patents is based on fractional assignments. Table 4.2 is more restrictive. Here we take into account firms that have been granted a patent and at the same time answered at least once to the R&D survey over the period. Only 275 patentees fulfilled this criterion. We express the number of patents granted (fractional assignments) relative to total employment and to the number of researchers in these organizations. The number of patents per both employee and researcher tends to decrease with size. Much more patents appear to be produced per employee and researcher in business firms (in industry and market services) compared to organizations in public services, which may indicate low inventive productivity of the latter sector. As far as ownership is concerned, domestic-owned firms seem to be the most productive patentees relative to the overall size as well as to the number of researchers (closely followed by foreign affiliates in the latter indicator). Overall, we observe relatively clear differences along the size dimension and between business firms and other organizations. 30

31 Table 4.2: R&D and patents, Observations Patents per 100 employees Patents per 100 researchers Average Median St. dev. Average Median St. dev. Total Size Large (250 and more) Medium (50-249) Small (1-49) Sector Industry Market services Public services Ownership Domestic private Domestic public Foreign affiliates Note: Number of patents is based on fractional assignments. Number of researchers refers to full-time equivalent. Finally, Table 4.3 shows results if the patent database is merged with data from the licensing surveys, which yields 193 observations in both datasets. First, we analyse how many of the patentees received some income from licence fees (in other words, have sold at least one license regardless of the amount of money). Only about 10% of these patentees have generated income from licensing. Medium-sized, service and domestic publiclyowned firms seems to be the most successful, though we need to cautious in assessing this data, because the number of observations is relatively low (and therefore nonrepresentative to the total population). Second, we take into account the actual amount of money received from the licensing fees. Large foreign affiliates in industry and market services have the most valuable patents. It is obvious, however, that these results are highly influenced by a few outliers with relative high income from licence fees relatively to the others. 31