Valuing academic patents and intellectual properties: Different perspectives of willingness to pay and sell

Technovation 33 (2013) 13–24

Contents lists available at SciVerse ScienceDirect

Technovation journal homepage: www.elsevier.com/locate/technovation

Valuing academic patents and intellectual properties: Different perspectives of willingness to pay and sell So Young Sohn n, Won Sang Lee, Yong Han Ju Department of Information and Industrial Engineering, Yonsei University, Seoul, Republic of Korea

a r t i c l e i n f o

abstract

Available online 10 November 2012

Academic inventors tend to lack the ability of valuing technologies in their areas. We apply classification tree analysis to discover different perspectives of Willingness to Pay (WTP) and Sell (WTS) of academic inventors when valuing their patents and technologies. Predictor factors considered are development environment, technology characteristics, ownership and patenting policy, and technology transfer characteristics. According to the result of Korean student data, WTS and WTP are differently perceived for the same technology: WTP is higher than WTS for the low valued technologies. The ownership policy, scalability and degree of innovation of technology, among the discovery of significant factors on WTS and WTP, are mainly considered as the important factors on WTS and WTP. From the finding of this research, we provide the policy implication on academic patenting and its ownership for further development of academic patents. & 2012 Elsevier Ltd. All rights reserved.

Keywords: Intellectual property value Decision tree Willingness to sell Willingness to pay

1. Introduction Academic inventions are an important source of corporate innovations (Geuna and Nesta, 2006). Various studies have been conducted in terms of management, legal aspects, and technology transfers of academic patents (Duderstadt, 2001; Jensen and Thursby, 2001; Mok et al., 2010; Sohn and Lee, 2012). Shane (2002) suggested a conceptual framework to verify the influence of patent effectiveness on the licensing, commercialization, royalty generation based on MIT inventions using various statistical models. Agrawal and Henderson (2002) investigated the degree to which patents are representative of the magnitude, direction, and impact of the knowledge spilling out of the university by focusing on the case of MIT using descriptive statistics and regression analysis. In terms of law, Colyvas et al. (2002) showed, using the case studies, how the intellectual property right can affect the commercialization of university inventions after Bayh–Dole act in 1980. They suggested that the IPR could be important for embryonic inventions and the marketing effort of university institution for technology transfer was important for university inventions. The authors also mentioned that the ability to issue exclusive licenses was most important for embryonic inventions while the dangers of exclusivity were greatest for these types of inventions. The increasing interest in academic inventions and patent made many researchers to focus on recognizing and exploiting the commercial opportunities; and promoting the community of

n

Corresponding author. Tel.: þ82 2 2123 4014; fax: þ82 2 364 7807. E-mail address: [email protected] (S.Y. Sohn).

0166-4972/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.technovation.2012.10.003

practice between different stakeholders (D’Este et al., 2012; Theodorakopoulos et al., 2012). Particularly, estimating value of academic inventions and understanding what factors affect it have emerged as the essential tasks to boost technology transfer activities from academia to commercial use. One of the largest studies regarding estimating patent value is the survey-based PatVal-EU project (Final Report of the PatVal-EU Project, 2005). But this project examined the value of retained patents that are not necessarily academic (Giuri et al., 2007; Crespi et al., 2007). Mowery and Ziedonis (2002) found that governmental policy can affect academic patent quality and quantity in the United States. Sapsalis et al. (2006) studied the distribution and determinants of patent value by comparing academic patents to corporate patents in Belgium. These studies of academic patent and its value estimation, however, do not take into account the perspectives of engineering students who were highly involved in the development of the technologies themselves. Because engineering graduate students have the high possibility of continuously working in research and development (R&D) or related areas, it becomes more necessary for students to be educated about what the estimated value of developed technology can be and how the value can be related with factors, such as ownership policy and environment of development, technology characteristics, and patenting and technology transfer characteristics (Mok et al., 2010). Therefore, this paper investigates the estimated value of technology and related factors in the perspective of engineering students in Korea, one of the most R&D intensified and engineering education oriented countries. Particularly, the ownership policy of developed technology among the related factors can be the important issue to

14

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

students. Generally, if students develop technology on their own, they are entitled to ownership of intellectual property (IP) rights. On the other hand, if students develop technology as part of a research team that is sponsored by or under contract with an industry or government, the ownership of IP rights does not belong to the inventors. While the view of researchers, including students, is not sufficiently considered in Korea, we can find the exceptional case that the teacher’s exception policy grants exclusive rights to academic inventors in Sweden, one of the R&D intensified countries (SOU 2005:95, 2005). While the teacher’s exception policy is currently effective in Sweden, most countries including Korea do not currently employ the teacher’s exception policy. Therefore, we assume that an ownership policy can be related with estimated value of patents or technology by students involved in corresponding R&D. To estimate the value of patents or technologies, several approaches have been used in terms of future cash flow projections generated by the patents: regression models of patent indicators, net present value, and real option pricing with Monte Carlo simulation (Gambardella et al., 2005; Giuri et al., 2007; Hall et al., 2007; Meng, 2008; Wartburg and Teichert, 2008; Ernst et al., 2010). However, it would be inappropriate to expect engineering graduate students to estimate the values of academic patents using these approaches that depend on many assumptions. In this paper, we estimate the value of technology in terms of Willingness to Sell (WTS) and Willingness to Pay (WTP) based on the results of a survey administered to engineering graduate students in Korea involved in technology development. WTP measures the benefit received by individuals (Johannesson, 1996; Coate and Morris, 1999; Jeon et al., 2010), and WTS represents the expected selling price for individuals (Hanemann, 1985). The measures of WTP and WTS, along with factors that influence them, are widely used to estimate the values of intangible goods (Shapiro, 1985; Johannesson, 1996; Noy et al., 2006; LeVert et al., 2009). We consider that these measures can be especially suitable for estimating value of intangible assets by students. As WTS and WTP are different measures of valuing the same goods, we consider both for an overall, balanced understanding of the value of technology in academia. When purchasing a patent, customers consider a specific patent in particular. On the other hand, when inventors sell a patent, they can proceed with licensing the patent to several customers simultaneously. The estimated value of the same patent can vary according to inherent differences in patent buying and selling situations. This is why we consider both WTS and WTP to estimate the value of technology. Then we apply a decision tree (DT) to identify variables that influence WTS and WTP. Among data mining methods, DT is one of the most frequently used methods for knowledge discovery. Decision tree is easy to interpret, and it is robust to input noise (Gayatri et al., 2010; Szepannek et al., 2005; Doctor et al., 2001). By analyzing both WTS and WTP using a decision tree, we expect to understand the relationship between related factors to the estimated value of technology evaluated by students. The structure of this paper is as follows. Section 2 summarizes the related literature. Section 3 presents our research design, and in Section 4, we analyze WTP and WTS and the associated factors using a decision tree. Lastly, in Section 5, we conclude the study and suggest directions for future research.

2. Literature review One of the most important studies on the estimation of patent value is the PatVal-EU project (Final Report of the PatVal-EU Project, 2005). The project is a European survey that was carried

out in France, Germany, Italy, the Netherlands, Spain and the United Kingdom from May 2003 to January 2004. The PatVal-EU project surveyed approximately 10,000 inventors for their best estimates of the values of their inventions, focusing on determinants for innovative performances of European countries and their potential contributions to economic growth. Using PatValEU project data, researchers estimated patent value and found that the distribution of patent value was skewed, with only a small share of highly valued patents: 16.81% of patents were worth more than three million Euros (Final Report of the PatValEU Project, 2005). Using the PatVal-EU project data set, other studies have been conducted for estimating the values of patents, which consider the value distribution of patents together with environmental characteristics of patents. Giuri et al. (2007) studied the characteristics of European inventors, the sources of their knowledge, the importance of formal and informal collaborations, their motivations, and the actual uses and values of patents using frequency analysis. Gambardella et al. (2007) studied the European market for patents and discussed the determinants of patent licensing using a probit regression. They considered the economic value of patents as an important factor in licensing, as well as in patent protection. Deng (2007) investigated the private value of European patents using the modified patent filing model and the patent renewal model. This study found that estimates of the private value of European patent rights vary according to different nationality, technology field, and cohort group. In particular this study pointed out that the value distribution of patents was quite skewed and even more skewed for EPO patent families. Harhoff and Hoisl (2007) discussed some specific differences in national legal provisions dealing with inventor compensation. Using ordered probit analysis, they also demonstrated that the number of inventors, technology field, size of patent family, and cites receiving the developed technology within five years of patenting are significantly associated factors on monetary patent value. Beyond the PatVal-EU project, other studies about patent value and commercialization of academic patents have been conducted, and are mainly concerned with patent value in light of technological and environmental characteristics, corporate patenting activity, effective management of governmental support, national differences, and other related issues. Goldenberg and Linton (2012) estimated the value of patents by considering patents as the compound options, so that it can be utilized by patent policy makers, inventors and patent attorneys. Gronqvist (2009) estimated private values of Finnish patents using renewal ratio and cost and found that private values of patents were determined by the degree of utility that the patent owner could obtain with the patent, and showed that patent value has a significant relationship to characteristics of technology such as the developer, the environment of the sponsor, assignee type, entity size, and technology category. Bessen (2008) estimated values of U.S. patents using regression and Monte Carlo simulation, and observed that they were substantially higher than those of European patents. Gallini (2002) studied the effect of strong patents due to the U.S. Patent Reform on patenting activity, specifically patent value. Sneed and Johnson (2009) investigated how specific attributes of patents affect patent value by analyzing unique patent auction data using Hexkman’s two-step model. Hausman and Leonard (2007) mentioned that a patent’s owner must receive a royalty that at least compensates for lost profit. Chiu and Chen (2007) proposed an analytical hierarchy process (AHP) scoring system for intellectual property with respect to the licensor. Boardman and Ponomariov (2009) pointed out that there is too little systematic assessment of university scientists who worked with private companies, despite the increased importance of university-industry interaction. Pries and Guild (2011) found

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

that the business model of academic research can be affected by legal protection, commercial uncertainty, specialized complementary area, and technological dynamism. Bathelt et al. (2010) developed conceptualization of spin-off processes in terms of technology transfer, and applied this concept to a current regional case study. The author’s suggested typology of university spin-off/ start-up firms is based on type of university sponsorship, university involvement in firm formation, the character of knowledge applied These previous studies, however, tend to lack perspective of the researcher in estimating the value of technology in academia. In this paper, we use WTS and WTP analysis of graduate student researchers to estimate the value of their technology. With regard to WTP and WTS, Shapiro’s (1985) study on patent licensing and R&D rivalry suggested that any firm’s WTP for a license depends upon the set of other firms that are also purchasing licenses. In addition, the PatVal-EU project (Final Report of the PatVal-EU Project, 2005) used the WTS concept to measure the inventor’s perceived value of a patent. As revealed in previous studies, patent value can be related with various factors, including: inventor characteristics, related policy, technological characteristics, and the environment associated with technology development and patenting activity. As this paper examines the integrated aspects of technology development in academia that impact engineering students’ WTP and WTS, we cover the entire process of academic patenting. In order to identify factors that are associated with the value of a patent or the technology in which engineering students are involved, we categorize most of the research variables reviewed in previous studies into three groups: technology ownership policy (Sapsalis et al., 2006; Hausman and Leonard, 2007) and environment in which technology is developed, technology characteristics, and patenting and technology transfer characteristics (Gambardella et al., 2007; Gronqvist, 2009) as shown in Table 1. The category of technology ownership policy and environment encompasses all of the variables that explain research conditions, ownership policy of research results, duration of research, participants involved in research, and any policy that can be related with development of technology. Technology characteristics include variables about the degree of sophistication and uniqueness of developed technology, the main area of developed technology, and its potential market. Patenting and technology transfer characteristics cover the appropriate range of patented technology, its expected life, the use of patent attorneys during the patenting process, methods of commercializing the developed technology (or patent), and the size and type of firms that buy the developed technology (or patent). Each of these three categories has been partially explored in previous studies, but no study has covered them comprehensively. We here consider all three categories together for WTS and WTP to provide the integrated view by considering its development to commercialization. Based on this categorization, we establish our research hypotheses as follows. Hypothesis 1. Engineering graduate student’s WTS and WTP on their technology or patent are related to the ownership policy & the development environment. Hypothesis 2. Engineering graduate student’s WTS and WTP on their technology or patent are related to the technology characteristics. Hypothesis 3. Engineering graduate student’s WTS and WTP on their technology or patent are related to the patenting & technology transferring characteristics. These hypotheses are designed to examine which factors and policy are associated with the estimated value of developed technology when considering both WTP and WTS. To test our

15

hypotheses, we define our research variable and conduct the survey to engineering graduate students.

3. Research variables and data 3.1. Variables As presented in Section 2, many different approaches have been applied to estimate patent value. These approaches have mainly considered technology that has already been patented, and have not focused on academic patents in depth. The current study considers the value of both patented technology and technology with the potential for academic patents. It takes a considerable amount of time for certain technology to be developed and ultimately patented. It can be argued that patent cannot completely protect the developed technology because patent application opens patented technology to public. In spite of such potential risk, patent can provide the legal protection to inventor for technology leakage, technology dispute, and its commercial use. Engineering graduate students who are involved in R&D may not stay in school long enough to see the patenting process to completion, as they often graduate before this process is completed. This is why we define the scope of academic patents to include both patented technology and technology that is still under development but has patent potential. Therefore, based on the assumption that the value of a patent (or technology) can be represented by WTP and WTS, we surveyed engineering graduate students regarding both their WTS and WTP for technology that they invented (or technology that they are developing). In this paper, because we target both patented technology and technology that is under development, the environment in which the technology is developed must be considered. The current study emphasizes both ownership policies and patenting and technology transfer characteristics that students follow in applying for patents on their inventions because of our focus on the researcher’s perspective. In order to predict the value of an academic patent, we consider three categories of variables, as listed in Table 2, for our survey: technology ownership policy and environment of development, technology characteristics, and patenting and technology transfer characteristics. This categorization of variables is meant to explain various aspects of WTS and WTP. In addition to WTS and WTP, Table 2 has 29 explanatory variables. The variable, ‘Degree of conflict of patent ownership in willingness to do research’ in the category of technology ownership policy and environment of development explains how conflicts over patent ownership can be related with the willingness of researchers to do their research. ‘Usefulness of teacher’s exception policy’ measures how useful researchers consider the teacher’s exception policy to be for their research. Based on the assumption that a competitive laboratory can be a useful environment for research, ‘Laboratory competitiveness’ rates the domestic competitiveness of laboratories wherein researchers work. ‘Technology level’ in the technology characteristics category examines how innovative the developed technology is. Here, ‘imitation level’ means that own technology does not have any innovation point. ‘Absorption level’ means that the developed technology partially or fully utilizes the other existing technology. ‘Technology category’ explains the area of technology. Among categories, culture technology represents convergence of science, technology and arts which can be applied to create or produce contents like movies, games, and animations. ‘Level of utilization of existing information’ means the degree to which researchers utilize the results of previous research in the development of their technology. ‘Necessity of complementary technologies in commercialization’ considers whether the developed technology

16

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

Table 1 Variables used to estimate patent value. Project

Variables

Categorya References

WTP and WTS

Patent licensing and R&D rivalry

PatVal-EU project

Estimation of patent licensing value

Technology field (Electrical engineering, instruments, chemicals and pharm, process engineering, and mechanical engineering) Inventor’s employers Gender Age Education Importance of inventor’s rewards Inventor’s share of monetary compensation Existence of research collaborations in the invention process Inventor’s nationality Technology field Patent age Share of salary received as inventor compensation for the surveyed patent, Inventor’s education Inventor’s age Inventor productivity Main technology field Number of employees Originality Generality Oppositions received PCT application filed Citation Size of patent family Job mobility Environment of the invention Invention process Brand

Value of patent T

Using AHP in patent valuation

Geographic area Technology essence

OE T

Cost dimension Product market Technology market Assignee type PTO (Patent and trademark office) entity status Technology category Renewal

OE T T P P T OE

Technology field Breadth Average patenting rate

T OE P

Number of employees, expenditure of R&D, Capital intensity Firm characteristics (size, type, age) Patent protection

OE OE P

Disclosure Technology transfer efficiency Public owner

P OE OE

Citations (forward & backward) Family size Scope Claims Lag Elapsed time Third party observation

P P T P P P OE

Number of designate countries University sponsorship

P OE

Character of university knowledge applied Employees Share of sales Type of invention Ownership

OE OE P P OE

Private value of European patents

Patent value and the German Employees’ Inventions Act

The value of U.S. patents by owner and patent characteristics

The private values of patents by patent characteristics

The patent paradox revisited: an empirical study of patenting in the U.S. semiconductor industry, 1979–1995

The economics of patents: lessons from recent U.S. patent reform

Selling ideas: the determinants of patent value in an auction environment

The market for patents in Europe

A knowledge-based typology of university spin-offs in the context of regional economic development

a

Shapiro (1985) Giuri, et al. (2007)

OE

OE OE P OE T P OE OE OE OE T OE P P P P P P OE OE P T

Deng (2007)

Harhoff and Hoisl (2007)

Hausman and Leonard (2007) Chiu and Chen (2007)

Bessen (2008)

Gronqvist (2009)

Hall and Zeidonis (2001)

Gallini (2002)

Sneed and Johnson (2009)

Gambardella et al. (2007) Bathelt et al. (2010)

OE: technology ownership policy and the development environment, T: technology characteristics, P: patenting and technology transfer characteristics.

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

17

Table 2 Research variables. Category

Description

Variable

Scale

References

Target variable

Value of academic patent

WTS

(Unit: Korean Won) 1 ¼Below ten million won 2 ¼Below 50 million won and over ten million won 3 ¼Below 100 million won and over 50 million won 4 ¼Over 100 million won

Shapiro (1985), Johannesson (1996), Hanemann, (1985), Gambardella et al. (2005), Hall et al. (2007)

WTP

Type of patent ownership B9 Technology ownership policy and environment of development B12 Degree of conflict of patent ownership in willingness to do research Preferred ownership policy B13

Technology characteristics

B17

Usefulness of teacher’s exception policy Average duration of development Number of participating researchers in R&D Laboratory competitiveness in domestic Writing research note

E4

Technology level

F1

Research experience in major technology area Level of utilization of existing information when technology is developed

F2

D3 E1 E3

F3–1 (research report) F3–2 (paper) F3–3 (patent) F4

Previous experiences of patent development Laboratory has patent that has F5 not been commercialized Market of developed G1 technology Technology category G2

Uniqueness of technology

G3

Position in the technological life cycle Necessity of complementary technologies in commercialization Ripple effect (outcome)

G4

Possibilities of expansion to various technology sectors

G5

G6

G7

1 ¼developer 2 ¼universities 3¼ sponsor 4¼ joint Reitzig (2004), (sponsor and universities) 5¼ joint (sponsor and developer) 6¼ etc 5 Likert scale (1¼ very high 2 ¼high 3¼ normal Reitzig (2004) 4 ¼low 5¼ very low) 1 ¼universities 2 ¼developer 3¼ country 4 ¼sponsor 5 Likert scale (1¼ very low 2 ¼low 3¼ normal 4 ¼high 5¼ very high) 1 ¼less than six months 2¼ 6–12 months 3¼ 1–2 years 4¼ 2–4 years 5 ¼more than four years Number of people

Argyres and Liebeskind (1998) Sohn and Lee (2012) SOU 2005:95 (2005)

1 ¼imitation 2¼ absorption 3¼ improvement 4 ¼innovation Years

Chiu and Chen (2007)

1 ¼very low 2 ¼low 3¼ normal 4 ¼high 5¼ very high

Chakrabarti et al. (2006)

Schwartz (2004)

Harhoff and Hoisl (2007); Reitzig(2004); Hall and Zeidonis (2001) 1 ¼top 10% 2 ¼top 1–25% 3¼ top 25 ¼50% 4¼ less Jaffe and Lerner (2001) than top 50% 1 ¼yes 2¼ no Chamas (2008)

Lynskey (2006)

Harhoff et al. (2003)

1 ¼yes 2¼ no

Lynskey (2006)

1 ¼yes 2¼ no

Svensson (2007)

1 ¼existing business 2 ¼newly created business 3 ¼converged business 1 ¼IT (Information Technology) 2 ¼BT (Biotechnology) 3 ¼NT (Nano Technology) 4 ¼CT (Culture Technology) 5 ¼ET (Environment Technology) 6 ¼ST (Space Technology) 7 ¼etc 1 ¼Very low 2 ¼Low originality, but efficiency and application exist 3 ¼High originality, but low efficiency and application 4 ¼High differentiation, but low possibility of developing new technology 5 ¼High differentiation and high possibility of developing new technology 1 ¼declining 2¼ introductory 3¼ mature 4 ¼growing 1 ¼yes 2¼ no

Nerkar and Roberts (2004)

Case 1: Has an influence on other industries Case 2: Has a social ripple effect, such as an increase in hiring Case 3: Has an import substitution effect Case 4: Has an export increasing effect 1 ¼none 2¼ selection of one case 3¼ selection of two cases 4 ¼selection of three cases 5 ¼selection of four cases 1 ¼There is no possibility of expansion. 2 ¼There is little expansion possibility. 3 ¼There is expansion possibility

Deng (2007); Harhoff and Hoisl (2007)

Baark (1988)

Haupt et al. (2007) Feshtman and Kamien (1992)

Motohashi (2005);

Hong et al. (2010)

Long (1989)

18

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

Table 2 (continued ) Category

Description

Variable

Scale

References

4 ¼There is a larger degree of expansion possibility 5 ¼There is the largest degree of expansion possibility Patenting & technology transfer

Has experienced patent rejection Has experienced objection during patent application Domestic or international patent (domestic/international) Expected life of patent Employment of patent attorney Method of patent commercialization

H1 H2 H3 H4 H5 H6

Same business area between seller and buyer of patent Type of buyer

H7 H8

Necessity of patent renewal

H9

1 ¼yes 2 ¼ no 3 ¼no experience with patent registration 1 ¼yes 2 ¼ no 3 ¼no experience with patent application 1 ¼domestic 2¼ international 3 ¼both Number of years 1 ¼yes 2 ¼ no 1 ¼Technology transfer by selling the developed technology 2 ¼Spin-off (creating new business) 3 ¼Patent licensing 4 ¼No experience with commercialization of patent 5 ¼Other 1 ¼yes 2 ¼ no 3 ¼no experience with patent registration 1 ¼Large companies (more than 250 employees) 2 ¼Mid-sized enterprises (between 100 and 250 employees) 3 ¼Small companies (less than 100 employees) 4 ¼Private research institute 5 ¼Government-funded institute 6 ¼University or educational institute 7 ¼Governmental institute 8 ¼Other 9 ¼No experience with commercialization 1 ¼yes 2 ¼ no

requires other technologies that complement it for commercialization. ‘Ripple effect’ refers to the expected influence that the developed technology will have on any industry. ‘Has experienced patent rejection’ and ‘Has experienced objection during patent application’ in the category of patenting and technology transfer characteristics describe such obstacles in the general experience of researchers during any previous patent applications. ‘Same business area between seller and buyer of patent’ considers whether the researchers (will) sell their developed technology to buyers in the same area of business as the developed technology, such as IT, NT, and BT. Because respondents might find it difficult to estimate the value of their own technology without any guidelines, students are asked to state their WTS and WTP for technology by selecting one option from the following ranges of money: (1) less than 10 million Korean won, (2) between 10 and 50 million Korean won, (3) between 50 and 100 million Korean won, and (4) over 100 million Korean won (USD $1 is equal to KRW 1,336.39 in April, 2009). In this case, traditional regression analysis is not proper because the value of technology is observed in a continuous scale. Therefore we apply a decision tree analysis that can be easily conveyed as having a simple top-down tree structure where decisions are made at each node by considering ownership policy of their technology and environmental factors of R&D, characteristics of technology, and conditions of the patenting process as explanatory variables. In the following section, we empirically investigate the relationship between explanatory variables in Table 2 to WTS and WTP using two decision trees, respectively. 3.2. Data The survey was administered in April 2009 to engineering graduate students with R&D experience from seven major, research-oriented universities in Seoul, Korea. Because those universities mostly hold the outstanding technology and actively

Rodriguez (2008) Adams (2005) Mengistie (1995) Deng (2007); Reitzig (2004) Macdonald and Lefang (1998) Kollner and Dowing (2004)

Cabral et al. (1999) Cabral et al. (1999)

Gronqvist (2009)

commercialize those technologies, the survey can have the representative of the population in Korea. A total of 219 students responded, of which 80% were male (195 respondents) and 15% were female (37 respondents), with no response from 5% (13 respondents). Respondents were enrolled in either a Master’s level program (43%; 105 respondents), a PhD program (40%; 99 respondents), or a post-doctoral program (6%; 14 respondents), with no response from 11% (27 respondents). The academic majors of the respondents included chemical engineering (7%; 16 respondents), mechanical engineering (15%; 38 respondents), industrial engineering (5%; 13 respondents), electrical and electronic engineering (23%; 56 respondents), material science and engineering (14%; 35 respondents), bioengineering (7%; 16 respondents), and computer engineering (7%; 17 respondents), with no response from 22% (54 respondents). Among the respondents, 20% were inventors of technology with a patent (48 respondents) and 67% were not (164 respondents). Thirteen percent of participants did not respond to this question (33 respondents). In the case of respondents with no patent, the survey was based on their technology in development. The survey results for WTS and WTP are displayed in Fig. 1. Table 3 shows the frequency of WTS and WTP level by academic major. It shows that most students estimate their technology to be worth 10,000,000 to 50,000,000 KRW in terms of WTP and WTS, with the exception of material science and engineering, and computer engineering majors. In addition, WTS tends to be higher than WTP in chemical engineering, industrial engineering, bioengineering, and computer engineering, while WTS tends to be equal to WTP in mechanical engineering, electrical and electronic engineering, and material science and engineering. In Fig. 1, out of all the respondents, 12% responded that WTP for their technology is higher than WTS while 37% responded that WTP is lower than WTS. Detailed information of WTS/WTP is

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

19

Fig. 1. Distribution of WTS/WTP on patent or technology in which engineering graduate students are involved. (Unit: persons). Table 3 Frequency distribution of WTS/WTP by academic major. Major

Overall Chemical engineering Mechanical engineering Industrial engineering Electrical and electronic engineering Material science and engineering Bioengineering Computer engineering No response

WTS

WTP

Less than KRW 10,000,000 (%)

KRW 10,000,000–KRW 50,000,000 (%)

KRW 50,000,000– More than KRW 100,000,000 KRW100,000,000 (%) (%)

Less than KRW 10,000,000 (%)

KRW 10,000,000–KRW 50,000,000 (%)

KRW 50,000,000– More than KRW 100,000,000 KRW100,000,000 (%) (%)

17.47 16.67

35.54 25.00

27.11 41.67

19.88 16.66

25.93 16.67

42.59 50.00

17.28 25.00

14.20 8.33

10.00

53.33

16.67

20.00

26.67

46.67

16.66

10.00

0.00

30.00

60.00

10.00

0.00

80.00

20.00

0.00

22.45

36.73

26.53

14.29

31.91

46.81

10.64

10.64

18.75

21.88

18.75

40.63

16.67

30.00

20.00

33.33

26.67 12.50

20.00 56.25

53.33 6.25

0.00 25.00

20.00 56.25

40.00 18.75

33.33 6.25

6.67 18.75

9.38

43.75

37.50

9.38

29.03

38.71

29.03

3.23

Table 4 Gap distribution in WTS/WTP. WTS/WTP

Less than KRW 10,000,000 KRW 10,000,000–KRW 50,000,000 KRW 50,000,000–KRW 100,000,000 More than KRW100,000,000 Total

Less than KRW 10,000,000 (%)

KRW 10,000,000–KRW 50,000,000 KRW 50,000,000–KRW 100,000,000 More than KRW100,000,000 (%) (%) (%)

51.7 41.4

37.5 56.9

10.7 42.9

5.3 10.5

6.9

5.6

39.3

24.6

0.0 100

0.0 100

7.1 100

59.6 100

shown in Table 4 where we can compare the difference between WTS and WTP. There is a perceived gap in value between selling and paying for the same developed technology. The predicted pattern is higher WTS than WTP, but interestingly, we discover that WTP is higher than WTS for patents with an estimated value of less than 10,000,000 Korean won. Specifically, 48.3% of researchers who evaluate their WTS at less than 10,000,000

Korean won believe that their WTP is higher than WTS, as opposed to the predicted pattern. This means that researchers have high WTP for low-priced inventions in particular. This result suggests that inventors prefer to buy technology rather than to develop it when the value of the technology is assessed by them as having low WTS. This gap in perception leads us to analyze WTS and WTP cases using two decision trees, respectively, in the

20

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

Table 5 Value of WTS/WTP and gap based on academic major. Value/Major

WTS

Overall Chemical engineering Mechanical engineering Industrial engineering Electrical and electronic engineering Material science and engineering Bioengineering Computer engineering No response

KRW KRW KRW KRW KRW KRW KRW KRW KRW

WTP 51,748,000 56,246,000 49,001,500 64,000,000 46,329,000 62,194,000 47,331,000 47,187,500 51,099,000

following section in order to examine what factors are related with the value of academic patents.

4. Decision tree analysis Based on observed variables in Table 2, we construct two decision trees for WTP and WTS, respectively. A decision tree is frequently used for knowledge discovery due to its simple topdown structure. Decision trees have also been applied to technology management (Sohn and Moon, 2004; Moon and Sohn, 2005; Padmanabhan and Tuzhilin, 2003). In this paper, the decision tree analysis is conducted using the Classification and Regression Trees (CART) algorithm. We first identify important variables by concurrently constructing decision trees for three categories of relatively small data size compared to the number of variables in Table 2. After the important variables from each of these three categories are recognized, we construct a final decision tree with a hierarchical structure, using all the variables selected. This final decision tree facilitates the examination of interaction effects between important factors from all three categories. 4.1. Decision tree for WTS/WTP For the sake of empirical analysis, the data set is divided into a training group and a validation group, in a 7:3 ratio. Each leaf of the decision tree (DT) represents a WTS or WTP in four intervals. The left column of leaves represents the results of the training group, and the right column represents the results of the validation group. We are only concerned with rules in which results are consistent between the training and validation groups. The relative frequency for an interval of WTS and WTP in leaf nodes is interpreted as follows: if the relative frequencies of interval 1 (less than 10 million won) and interval 2 (between 10 and 50 million won) are greater than 50%, we classify the WTS or WTP as low. If the relative frequencies of interval 3 (between 50 and 100 million won) and interval 4 (over 100 million won) are greater than 50%, we classify the WTS or WTP as high. If the relative frequencies of interval 1, interval 2 and interval 3 are greater than 80%, we classify the WTS or WTP as less than 100 million won at most. If the relative frequencies of interval 2, interval 3 and interval 4 are greater than 80%, we classify the WTS or WTP as over 10 million won at least. In terms of quantity, we estimate patent (technology) values by summing the multiplications of the relative frequency and the median values for each interval. The median values for each interval are 5,000,000 Korean won for interval 1, 30,000,000 Korean won for interval 2, 75,000,000 Korean won for interval 3, and 100,000,000 Korean won for interval 4. Because the fourth interval does not have an upper bound, we cannot use its median value, and instead we use

KRW KRW KRW KRW KRW KRW KRW KRW KRW

Gap (WTS-WTP) 41,233,500 42,913,500 37,829,500 39,000,000 34,258,500 58,163,500 44,667,500 31,875,000 38,067,000

KRW KRW KRW KRW KRW KRW KRW KRW KRW

10,514,500 13,332,500 11,172,000 25,000,000 12,070,500 4,030,500 2,663,500 15,312,500 13,032,000

the minimum value of the interval. If the leaf of DT does not have a frequency, we classify that case as an unavailable estimation. Table 5 shows the value of WTS and WTP by academic major of graduate students. In Table 5, academic majors are arranged in descending order in terms of WTS as follows: industrial engineering, material science and engineering, chemical engineering, mechanical engineering, bioengineering, computer engineering, and electrical and electronic engineering. Academic majors arranged in descending order in terms of WTP are as follows: material science and engineering, bioengineering, chemical engineering, industrial engineering, mechanical engineering, electrical and electronic engineering, and computer engineering. In addition, the gap between WTS and WTP is arranged in descending order as follows: industrial engineering, computer engineering, chemical engineering, electrical and electronic engineering, mechanical engineering, material science and engineering, and bioengineering. Graduate students with computer engineering majors estimate the lowest levels of WTS and WTP. Interpretation of this result is that the inventions of computer engineers – mostly software – are difficult to patent and are not well protected by copyright either. The gap between WTS and WTP is small in material science and engineering as well as in bioengineering. This result can be interpreted that because most original technologies are developed in material science & engineering, gap between WTS and WTP is smaller than other majors. In addition, there is small gap between WTS and WTP in technology of bioengineering because this area should have many advanced tests before completing valuation. Table 6 summarizes selected variables from individual decision trees obtained from the three categories of explanatory variables. In Table 6, we see that more than three variables are selected in each category of explanatory variables. Interestingly, significantly associated variables are different for WTP and WTS in the technology ownership policy and environment of development category. Respondents consider the teacher’s exception policy and competitiveness of laboratory to be influential in terms of WTS, while they consider the average duration of development, ownership type and degree of conflict of patent ownership to be significantly associated variables for WTP. Table 6 shows important variables for each category. It does not explain the interaction effects between these categories, however, so we construct a DT using important variables selected from the three categories to analyze the interaction effects. As shown in Fig. 2, usefulness of the teacher’s exception policy, possibilities of expansion to various technology sectors, technology level, type of buyer, and uniqueness of technology are selected for classification of WTS. In particular, the teacher’s exception policy turns out to be the most significantly associated variable in classifying WTS. Teacher’s exception policy is not enforced in Korea. This result reflects that engineering graduate students expect that this policy can increase the value of their technology or patent. In addition, if engineering graduate students think that technology has high expansion possibilities to

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

21

Table 6 Comparison of selected variables for WTS and WTP from three categories of explanatory variables. Category

Selected variables in WTS

Selected variables in WTP

Technology ownership Usefulness of teacher’s exception policyCompetitiveness of policy and environment of laboratoryParticipants in R&D development Technology levelLevel of utilization of patent Technology characteristics informationPossibility of expansion to various technology sectorsUniqueness of technology Patenting and technology Type of buyerExpected life of patentPrior experience with patent transfer characteristics application and rejection

Average duration of developmentOwnership typeDegree of conflict of patent ownership on willingness to conduct researchParticipants in R&D Technology categoryLevel of utilization of research report informationPossibility of expansion to various technology sectorsUniqueness of technology Expected life of patentPrior experience with patent application and rejection

experienced commercialization, and the technology has high originality but low efficiency and application, then WTS is as high as 56,818,000 Korean won. Rule 1–3. If graduate student researchers believe that the teacher’s exception policy is useful, expansion possibilities of relevant technology to various technology sectors is high, the level of technology is below the innovation stage, and yet graduate researchers have experience with commercialization, then WTS is as low as 48,750,000 Korean won. Rule 1–4. If graduate student researchers believe that the teacher’s exception policy is useless, then WTS is as low as 32,857,000 Korean won. Rule 1–5. If graduate student researchers believe that the teacher’s exception policy is useful, and expansion possibilities of relevant technology to various technology sectors is low, then WTS is as low as 41,923,000 Korean won. In contrast to the decision tree for WTS, Fig. 3 shows different patterns between explanatory variables and WTP. As shown in Fig. 3, the possibility of expansion to various technology sectors, expected life of patent, type of patent ownership, and number of participating researchers in R&D are selected for the classification of WTP. In case of WTP, the possibility of expansion to various technology sectors is the most important variable because this is first selected for a tree. This may reflect that opportunity of commercialization is considered to be most important factor. In contrast to teacher’s exception policy in WTS, type of patent ownership is selected important variable for high degree of WTP. If type of patent ownership is developer or university, then students respond for high degree of WTP. If developer or university can have patent ownership, they get a strong attachment to their technology. Based on the results of the decision tree for WTP, we outlined four rules as follows:

Fig. 2. Decision tree for WTS.

various technology sectors, then their WTS is high. Overall, we obtain five rules regarding WTS, as follows: Rule 1–1. If graduate student researchers believe that the teacher’s exception policy is useful, expansion possibilities of relevant technology to various technology sectors is high, and technology is at the innovation stage, then WTS is as high as 90,000,000 Korean won. Rule 1–2. If graduate student researchers believe that the teacher’s exception policy is useful, expansion possibilities of relevant technology to various technology sectors is high, the level of technology is below the innovation stage, the graduate student researchers have not

Rule 2–1. If expansion possibilities of relevant technology to various technology sectors are low, and the expected life of the patent is less than 3.5 years, then graduate student researchers’ WTP is as low as 38,000,000 Korean won. Rule 2–2. If expansion possibilities of relevant technology to various technology sectors are low, and the expected life of the patent is longer than 3.5 years, then graduate student researchers’ WTP is as low as 34,410,000 Korean won. Rule 2–3. If expansion possibilities of relevant technology to various technology sectors are high, and the type of patent ownership is sponsor or co-ownership (sponsor and university/developer), then graduate student researchers’ WTP is as low as 37,585,000 Korean won. Rule 2–4. If expansion possibilities of relevant technology to various technology sectors are high, the type of patent ownership is developer or university, and number of participating

22

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

Fig. 3. Decision tree for WTP.

researchers in R&D is less than 9, then graduate student researchers’ WTP is as high as 62,000,000 Korean won. Both WTS and WTP cases show that there can be interaction effects among technology ownership policy and environment of development, technology characteristics, and patenting and technology transfer characteristics, and our research hypotheses can be supported from these findings. Moreover, we found a tendency for graduate students to strongly consider the teacher’s exception policy, possibilities of expansion to various technology sectors, technology level, type of buyer, and uniqueness of technology in estimating WTS. In estimating WTP, type of patent ownership, possibilities of expansion to various technology sectors, and number of participating researchers in R&D are selected as important factors. Based on the results of DT analysis, we could identify important information in terms of WTS and WTP. In the case of WTS, the teacher’s exception policy is a very important factor; however, expansion possibility for the relevant technology is an important factor in WTP. It follows that graduate student researchers who primarily consider selling their technology should consider factors that are related with the selling price. It is particularly beneficial for students to focus on developing technology with many possibilities for expansion to other areas of technology and business.

5. Conclusions In this paper, we surveyed both the WTP and WTS of engineering graduate students in Korea for academic patents (technology), along with related variables. As a result, we discovered that WTS and WTP are perceived differently for the same technology. One interesting point is that WTP is higher than WTS when the estimated value of the developed technology is low. This result is not consistent with the general understanding that WTS is usually higher than WTP (Bockstael and McConnell, 1980). We discovered that values of WTS and WTP vary according to different academic majors. The value of WTS was highest in industrial engineering while the value of WTP was highest in material science and

engineering. Bioengineering and material science and engineering have a smaller gap between WTS and WTP than other majors. Computer engineering has the lowest WTS and WTP. We found well-informed estimations of the market price of patents in material science and engineering and bioengineering, represented by the significantly small gap between WTS and WTP in these academic fields. We also examined if both WTS and WTP are associated with categories of ownership policy and environment of development, technology characteristics, and patenting and technology transfer characteristics. In terms of the technology ownership and environment category, perception of the usefulness of the teacher’s exception policy was positively related with WTS. The high possibility for expansion and a high uniqueness of technology positively influence WTS. In addition, the innovation level of technology is highly related to both WTS and WTP. In the case of patenting and technology transfer characteristics, the expected life of patent, type of ownership, and number of participating researchers in R&D have an effect on WTP. The results of DT analysis show that perceptions of the usefulness of the teacher’s exception policy are considered the single most important variable with regard to WTS, while possibilities of expansion to various technology sectors is the most important variable for WTP. In particular, rules 1–1 and 1–2 of the DT for WTS suggest that (1) if the teacher’s exception policy is considered useful, (2) if the possibility of expansion to various technology sectors is high, and (3) if the developed technology is at the innovation stage, then WTS is high. This result implies that the teacher’s exception policy should be introduced as an option for graduate student researchers at the universities surveyed for this study, and arguably to engineering graduate schools in general. Inventors are also required to develop a highly expandable technology in order to enhance the patent value. Decision tree analysis for WTP explains that if the possibility for expansion of relevant technology to various technology sectors is low, then WTP is low. Rule 2–4 of the DT for WTP also shows that (1) if the expansion possibility of relevant technology is high, and (2) patent ownership is granted to the developer or university, then WTP is high. Researchers should therefore consider scalability and ownership in the development of their technology.

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

From the findings of this research, we suggest important areas for further investigation for academic patenting. First, it is necessary to further analyze the gap between WTS and WTP in terms of paired comparison. Based on the existence of interevaluator variability (Sohn et al., 2012), in this way, we expect to provide useful information in understanding the value of academic patents. Further analysis should be focused on three cases in particular: (1) WTS being higher than WTP, (2) WTS and WTP being of equal value, and (3) WTS being lower than WTP. In this study we found examples of WTS being lower than WTP when the estimated value of technological inventions is low, or in other words, when the market value is cheap. Second, it would be worthwhile to conduct a similar study in the corporate realm. It is critically important to consider the perspectives of corporations, which are the biggest purchasers of technology, when estimating the value of academic patents (technology) objectively. Third, we need to concentrate on the relationship between academic patent value and academic patenting policy, especially ownership policy that benefits the academic inventor. Fourth, it is necessary to conduct a cross-country survey to analyze the relationship between academic patent value and ownership policy by country. This would help us understand how ownership policy is related to patent value. Above all, creativity education for the researchers is needed to improve and induce innovative ideas for invention which can increase both WTS and WTP (Sohn and Jung, 2010; Sohn and Ju, 2011). These areas are left for further research that can add valuable insight to the implications for academic patenting revealed in our study.

References Adams, S., 2005. Electronic-text material in patent applications-some questions for patent offices, applications and searchers. World Patent Information 27 (2), 99–103. Agrawal, A., Henderson, R., 2002. Putting patents in context: exploring knowledge transfer from MIT. Management Science 48 (1), 44–60. Argyres, N.S., Liebeskind, J.P., 1998. Privatizing the intellectual commons: universities and the commercialization of biotechnology. Journal of Economic Behavior and Organization 35 (3), 427–454. Baark, E., 1988. The value of technology: a survey of the Chinese theoretical debate and its policy implications. Research Policy 17 (5), 269–282. Bathelt, H., Kogler, D.F., Munro, A.K., 2010. A knowledge-based typology of university spin-offs in the context of regional economic development. Technology 30 (9–10), 519–532. Bessen, J., 2008. The value of U.S. patents by owner and patent characteristics. Research Policy 37 (5), 932–945. Bockstael, E., McConnell, E., 1980. Calculating equivalent and compensating variation for natural resource facilities. Land Economics 56 (1), 56–63. Boardman, P.C., Ponomariov, B.L., 2009. University researchers working with private companies. Technovation 29 (2), 142–153. Cabral, L.B., Salant, D.J., Woroch, G.A., 1999. Monopoly pricing with network externalities. International Journal of Industrial Organization 17 (2), 199–214. Chakrabarti, S., Amba, S., Ramasami, T., 2006. Study of landscape of global leather patents and analysis of technology linkage to trade. World Patent Information 28 (3), 226–234. Chamas, C.I., 2008. Nanotechnology intellectual property in Brazil: preliminary research note. World Patent Information 30 (2), 146–149. Chiu, Y.J., Chen, Y.W., 2007. Using AHP in patent valuation. Mathematical and Computer Modelling 46 (7/8), 1054–1062. Coate, S., Morris, S., 1999. Policy persistence. The American Economic Review 89 (5), 1327–1336. Colyvas, J., Crow, M., Gelijns, A., Mazzoleni, R., Nelson, R., Rosenberg, N., Sampat, B.N., 2002. How do university inventions get into practice? Management Science 48 (1), 61–72. Crespi, G.A., Geuna, A., Nesta, L., 2007. The mobility of university inventors in Europe. The Journal of Technology Transfer 32 (3), 5–215. Deng, Y., 2007. Private value of European patents. European Economic Review 51 (7), 1785–1812. D’Este, P., Mahdi, S., Neely, A., Rentocchini, F., 2012. Inventors and entrepreneurs in academia: what types of skills and experience matter? Technovation 32 (9– 10), 293–303. Doctor, R.N., Newton, D.P., Perason, A., 2001. Managing uncertainty in research and development. Technovation 21 (2), 79–80. Duderstadt, J.J., 2001. The future of the university in the digital age. American Philosophical Society 145 (1), 54–72.

23

Ernst, H., Legler, S., Lichtenthaler, U., 2010. Determinants of patent value: insight from a simulation analysis. Technological Forecasting & Social Change 77 (1), 1–19. Feshtman, C., Kamien, M.I., 1992. Cross licensing of complementary technologies. International Journal of Industrial Organization 10 (3), 329–348. Gambardella, A., Giuri, P., Luzzi, A., 2007. The market for patents in Europe. Research Policy 36 (8), 1163–1183. Gambardella, A., Harhoff, D., Verspagen, B., 2005. The Value of Patens. Universita Bocconi, Ludwig-Maximiliens Universitaet, and Eindhoven University, Working Paper. Gallini, N.T., 2002. The economics of patents: lessons from recent U.S. patent reform. Journal of Economic Perspectives 16 (2), 131–154. Gayatri, N., Nickolas, S., Reddy, A.V., 2010. Feature selection using decision tree induction in class level metrics dataset for software defect predictions. Proceedings of the World Congress on Engineering and Computer Science Vol. I, 124–129. Geuna, A., Nesta, L.J.J., 2006. University patenting and its effects on academic research: the emerging European evidence. Research Policy 35 (6), 790–807. Giuri, P., Mariani, M., Brusoni, S., Crespi, G., Francoz, D., Gambardella, A., GarciaFontes, W., Geuna, A., Gonzales, R., Harhoff, D., Hoisl, K., Le Bas, C., Luzzi, A., Magazzini, L., Nesta, L., Nomaler, O., Palomeras, N., Patel, P., Romanelli, M., Verspagen, B., 2007. Inventors and invention processes in Europe: results from the PatVal-EU survey. Research Policy 36 (8), 1107–1127. Goldenberg, D.H., Linton, J.D., 2012. The patent paradox—new insights through decision support using compound options. Technological Forecasting & Social Change 79 (1), 180–185. Gronqvist, C., 2009. The private value of patents by patent characteristics: evidence from Finland. Journal of Technology Transfer 34 (2), 159–168. Hall, B.H., Zeidonis, R.H., 2001. The patent paradox revisited: an empirical study of patenting in the U.S. semiconductor industry, 1979–1995. RAND Journal of Economics 32 (1), 101–128. Hall, B.H., Thoma, G., Torrisi, S. 2007. The Market Value of Patents and R&D: Evidence from European Firms, NBER Working Paper, No. 13426. Hanemann, M.W., 1985. Some issues in continuous and discrete response contingent valuation studies. North-eastern Journal of Agricultural and Resource Economics 14, 5–13. Harhoff, D., Hoisl, K., 2007. Institutionalized incentives for ingenuity-patent value and the German Employees’ Invention Act. Research Policy 36 (8), 1143–1162. Harhoff, D., Scherer, F.M., Vopel, K., 2003. Citations, family size, opposition and the value of patent rights. Research Policy 32 (8), 1343–1363. Haupt, R., Kloyer, M., Lange, M., 2007. Patent indicator for the technology life cycle development. Research Policy 36 (3), 387–398. Hausman, J., Leonard, G.K., 2007. Estimation of patent licensing value using a flexible demand specification. Journal of Econometrics 139 (2), 242–258. Hong, S.J., Seo, J.W., Kim, Y.S., Kang, S.H., 2010. Construction technology valuation for patent transaction. KSCE Journal of Civil Engineering 14 (2), 111–122. Jaffe, A.B., Lerner, J., 2001. Reinventing public R&D: patent policy and the commercialization of national laboratory technologies. RAND Journal of Economics 32 (1), 167–198. Jeon, H.J., Kim, M.S., Sohn, S.Y., 2010. Conjoint and WTP analysis of future mobile phones for digital convergence. Technological Forecasting & Social Change 77 (3), 457–465. Jensen, R., Thursby, M., 2001. Proofs and prototypes for sale: the licensing of university inventions. The American Economic Review 91 (1), 240–259. Johannesson, M., 1996. A note on the relationship between ex ante and expected willingness to pay for health care. Social Science & Medicines 42 (3), 305–311. Kollner, H., Dowing, M., 2004. Licensing as a commercialization strategy for new technology-based firms. Research Policy 33 (8), 1141–1151. LeVert, M., Stevens, T., Kittredge, D., 2009. Willingness-to-sell conservation easements: a case study. Journal of Forest Economics 15 (4), 261–275. Long, F., 1989. Technology. Trade and international economic policy. Technology in Society 11 (2), 261–266. Lynskey, M.J., 2006. Transformative technology and institutional transformation: coevolution of biotechnology venture firms and the institutional framework in Japan. Research Policy 35 (9), 1389–1422. Macdonald, S., Lefang, B., 1998. The patent attorney as an indicator or innovation. Computer Law and Security Report 14 (1), 8–13. Meng, R., 2008. A patent race in real options setting: investment strategy, valuation, CAPM beta, and return volatility. Journal of Economic Dynamics & Control 32 (10), 3192–3217. Mengistie, G., 1995. The patent system in Ethiopia. World Patent Information 17 (1), 29–34. Mok, M.S., Sohn, S.Y., Ju, Y.H., 2010. Conjoint analysis for intellectual property education. World Patent Information 32 (2), 129–134. Moon, T.H., Sohn, S.Y., 2005. Intelligent approach for effective management of governmental funds for small and medium enterprises. Expert Systems with Applications 29 (3), 566–572. Motohashi, K., 2005. University-industry collaborations in Japan: the role of new technology-based firms in transforming the National Innovation System. Research Policy 34 (5), 583–594. Mowery, D.C., Ziedonis, A.A., 2002. Academic patent quality and quantity before and after the Bayh–Dole act in the United States. Research Policy 31 (3), 399–418. Nerkar, A., Roberts, P.W., 2004. Technological and product-market experience and the success of new product introductions in the pharmaceutical industry. Strategic Management Journal 25 (89), 779–799.

24

S.Y. Sohn et al. / Technovation 33 (2013) 13–24

Noy, A., Raban, D.R., Ravid, G., 2006. Testing social theories in computer-mediated communication through gaming and simulation. Simulation Gaming 37 (2), 174–194. ¨ ¨ Nyttiggorande av hogskoleuppfinningar SOU 2005:95, available at /http://www. regeringen.se/sb/d/108/a/53408S. Padmanabhan, B., Tuzhilin, A., 2003. On the use of optimization for data mining: theoretical interactions and eCRM opportunities. Management Science 49 (10), 1327–1343. Pries, F., Guild, P., 2011. Commericializing inventions resulting from university research: analyzing the impact of technology characteristics on subsequent business models. Technovation 31 (4), 151–160. Reitzig, M., 2004. Improving patent valuations for management purposesvalidating new indicators by analyzing application rationales. Research Policy 33 (6–7), 939–957. Rodriguez, V., 2008. Governance of material transfer agreements. Technology in Society 30 (2), 122–128. Sapsalis, E., Potterie, B.P., Navon, R., 2006. Academic versus industry patenting: an in-depth analysis of what determines patent value. Research Policy 35 (10), 1631–1645. Schwartz, E.S., 2004. Patents and R&D as real options. Economic Notes by Banca Monte dei Pashi di Siena SpA 33 (1), 23–54. Shapiro, C., 1985. Patent licensing and R&D rivalry. The American Economic Review 75 (2), 25–30. Shane, S., 2002. Selling university technology: patterns from MIT. Management Science 48 (1), 122–137. Sneed, K.A., Johnson, D.K.N., 2009. Selling ideas: the determinants of patent value in an auction environment. R&D Management 39 (1), 87–94.

Sohn, S.Y., Moon, T.H., 2004. Decision tree based on data envelopment analysis for effective technology commercialization. Expert Systems with Applications 26 (2), 279–284. Sohn, S.Y., Jung, C.S., 2010. Effect of creativity on innovation: do creativity initiatives have significant impacts on innovative performance in Korean firms? Creativity Research Journal 22 (3), 320–328. Sohn, S.Y., Ju, Y.H., 2011. Design and implementation of youth engineering adventure program in Korea. International Journal of Engineering Education 27 (5), 1107–1116. Sohn, S.Y., Lee, M.Y., 2012. Conjoint analysis of R&D contract agreements for industry-funded university research. The Journal of Technology Transfer 37 (4), 532–549. Sohn, S.Y., Doo, M.K., Ju, Y.H., 2012. Pattern recognition for evaluator errors in a credit scoring model for technology-based SMEs. Journal of the Operational Research Society 63 (8), 1051–1064. Svensson, R., 2007. Commercialization of patents and external financing during the R&D phase. Research Policy 36 (7), 1052–1069. Szepannek, G., Luebke, K., Weihs, C., 2005. Understanding pattern with different subspace classification. Lecture Notes in Computer Science 3597, 110–119. Theodorakopoulos, N., Preciado, D.J.S., Bennett, D., 2012. Transferring technology from university to rural industry within a developing economy context: the case for nurturing communities of practice. Technovation 32 (9–10), 550–559. The Value of European Patents: Evidence from a Survey of European Inventors, Final Report or the PatVal EU Project, 2005. Wartburg, I., Teichert, T., 2008. Valuing patents and licenses from a business strategy perspective—extending valuation considerations using the case of nanotechnology. World Patent Information 30 (2), 106–114.