J Technol Transf (2013) 38:251–272

DOI 10.1007/s10961-012-9252-0

Licensing versus selling in transactions for exploiting

patented technological knowledge assets in the markets
for technology

Seongkyoon Jeong • Sungki Lee • Yeonbae Kim

Published online: 22 March 2012

 Springer Science+Business Media, LLC 2012

Abstract This paper examines the determinants of the types of technology transactions in
the markets for technology. On the basis of the relationship between the characteristics of a
firm’s patents and the firm’s decision on whether to license out or sell these patented
technologies, we empirically analyze the determinants of the decision. We employ inter-
locked patent data from the representative Korean market for technology, the National
Technology Bank, using a bivariate probit regression model in a theoretical framework that
includes the option and transaction cost perspectives. Overall, the results show that the
relationship between licensing and selling, the major alternatives in technology transac-
tions, is strongly substitutive. The major finding of this study is that firms in markets for
technology tend to prefer licensing their patents when uncertainty is low or transaction cost
is high, whereas they tend to prefer selling their patents under opposite conditions.

Keywords Technology transfer  Market for technology  Technology exploitation

strategy  Technology licensing

JEL Classification C35  D22  D81

S. Jeong
Department of R&D Policy, Korea Institute of Machinery & Materials,
104 Sinseong-Ro, Yuseong-Gu, Daejeon 305-343, Republic of Korea

S. Lee (&)
IP Policy Team, Korea Institute of Intellectual Property,
647-9 Yeoksam-Dong, Gangnam-Gu, Seoul 135-980, Republic of Korea
e-mail: [email protected]

Y. Kim
Technology Management, Economics and Policy Program (TEMEP),
College of Engineering, Seoul National University, San 56-1,
Shillim-Dong, Kwanak-Gu, Seoul 151-744, Republic of Korea

123
252 S. Jeong et al.

1 Introduction

In the past, most industrial firms used their technologies in their own products or services
in order to achieve a competitive advantage in their businesses (Elton et al. 2002;
Chesbrough 2003). External exploitation of technology, however, has become a growing
phenomenon in practice (Lichtenthaler 2008), and technological innovation through
transactions of intellectual property has increased substantially (Arora et al. 2001). This
phenomenon has resulted in the emergence of the open innovation paradigm (Chesbrough
2003), which emphasizes the strategic management of technological assets. Therefore,
many firms maximize the value of their knowledge assets through transactions such as
licensing out and selling their technologies.1
From the point of view of the technology supplier, it is critically important to decide the
specific form of a transaction for exploiting technological assets (Chiesa et al. 2008; Ford
and Ryan 1981). A firm can externally commercialize or exploit its technological
knowledge assets in several ways: selling technologies to a third party, forming a strategic
alliance, establishing a joint venture, licensing out, and initiating a new venture (Ford
1988; Granstrand 2000; Lichtenthaler 2005; Megantz 2002). Each specific type of tech-
nology transaction has different managerial and organizational implications (Chiesa et al.
2008) because of a correspondingly different degree of risk and potential for reward
(Megantz 2002).
However, thus far, there has been no empirical examination of the various types of
technology transactions. Most theoretical and empirical studies have considered only one
type of transaction—usually, licensing agreements (Arora 1996; Gambardella et al. 2007;
Motohashi 2008). A few scholars (e.g., Chiesa et al. 2008; Lichtenthaler 2005) have
suggested theoretical and conceptual approaches that take into account the types of
technology transactions and the perspective of technology owners; however, these
approaches have not fully developed the mechanism of choice with regard to types of
technology transactions. Furthermore, no study has empirically demonstrated the theo-
retical and conceptual approaches.
On the basis of the theoretical and conceptual classification of technology transactions,
this paper examines the determinants of technology suppliers’ choice of transaction types.
Under the framework arguing that a firm’s choice of technology licensing or selling is
determined by the degree of uncertainty and transaction cost, this study empirically
identifies a relationship between the characteristics of patents and a firm’s decision about
technology transaction types.
The distinctive contribution of this study can be described as follows. First, using the
existing theoretical framework, this paper offers a practical understanding of firms’
strategies in the markets for technology by comprehensively organizing relevant theories
on firms’ choice of technology transaction types. Second, this research empirically
examines the phenomenon of firms’ behavior with regard to the choice of technology
transaction using a noble data set. In general, although technology transaction types vary,
very limited standardized data can be obtained on the types of transactions and technology
readiness level, because firms would not voluntarily disclose such data. Finally, through
the suggested practical variables in our study, firms, especially potential technology
acquirers, can monitor technology suppliers’ strategies on the basis of the relationships

1
Elton et al. (2002) show that 10 % of patents from research-oriented firms have never been used, even for
a protective purpose, and that if the unused patents were to be licensed, the annual profits of patent-owning
firms would increase by 5 %.

123
Licensing versus selling in transactions 253

revealed between technology characteristics and technology transaction types in the actual
markets.
The remainder of the paper is organized as follows. The next section briefly discusses
the characteristics of technology licensing and selling and the implications of choosing
either transaction type. The article then introduces the theoretical framework for explaining
a firm’s choice of technology transaction and formulates hypotheses. The description of the
data and the empirical methodology are followed by a discussion of the results and overall
conclusions.

2 Theoretical background and hypotheses

2.1 Technology transaction types in the markets for technology

Unlike normal commodities, technology tends to include tacit knowledge, and it is not easy
to create or find substitute goods, i.e., the technology is irreplaceable. These characteristics
cause information asymmetries between technology suppliers, who have adequate know-
how relevant to the technology, and technology acquirers, who face difficulties in assessing
and absorbing the technology (Arora et al. 2001).
The information asymmetries enable technology suppliers to enjoy a dominant position
in transactions: they choose favorable sales policies and control prices, transaction types,
and follow-up services in the markets for technology (Park 2002). Because of their bar-
gaining power, technology suppliers choose the desired type of technology transaction as
part of a strategy to secure a competitive advantage, since each type has different mana-
gerial and legal implications (Chiesa et al. 2008; Megantz 2002).
In general, the two major types of technology transactions are licensing and selling,
depending on whether or not patent ownership is transferred (Chiesa et al. 2008; Megantz
2002). Selling involves the transfer of patent ownership from a technology supplier to a
buyer. After the transfer of the patent, the prior patent owner loses all rights on the patent,
including control of the use of technology (Irish 2005). Instead, the prior patent owner
receives a lump sum payment for the transfer of ownership (Chiesa et al. 2008). Under
licensing, a technology supplier grants a licensee the right of exploiting and commer-
cializing the technology protected by the patent; in return, the supplier receives a licensing
fee during the contract period. As a licensor, the technology supplier can control licensees
according to the codified legal terms of the licensing contract regarding confidentiality,
payment of licensing fee, liability, third-party claims, liquidated damages, exchange of
improvements, and expansion of fee.
When it comes to the revenue of technology suppliers, licensing provides a technology
supplier with various payment options. The technology supplier can receive a license fee in
lump sum or royalty proportional to the sales of the final product to which the transferred
technology contributed.2 Relating the sales of the licensee to the licensing fee enables the
licensee to lower the initial payment; however, this method increases the uncertainty
associated with the overall revenue of the technology supplier. This is because the total
amount of the licensing fee that the technology supplier will receive later cannot be fully
determined until the licensee finalizes her economic activities derived from the technology.

2
In reality, mixed licensing (i.e., lump sum plus running royalties) is a commonly chosen form of trans-
action (Chiesa et al. 2008). According to Thomson Financial’s Securities Data Company (SDC) Database,
66 % of all licensing contracts from 1997 to 2004 included royalties.

123
254 S. Jeong et al.

Therefore, in the case of licensing, the technology supplier is motivated to monitor the
licensees in order to maximize mutual revenue after the assignment of the licensing
contract. On the other hand, in the case of selling, the technology supplier faces no risk
regarding revenue from the technology acquirer, having surrendered all rights to the
technology.
As summarized in Table 1, the characteristics of selling and licensing differ in various
aspects: control of the sourcing firm, influence of the commercial outcome on the revenue
of the technology supplier, initial revenue of the technology supplier, and uncertainty of
the expected overall revenue of the technology supplier.

2.2 Determinants of type of technology transaction

According to classical decision theory, a trade-off between risk and expected return leads
firms to pursue the highest risk-adjusted return for available alternatives, wherein risk is
conceptualized as the variance of the probability of the gains and losses of a particular
alternative (Pratt 1964). However, it is not easy to conceptualize risk in the managerial
decision-making process in the real world. Hence, managers choose an alternative within
the decision context of bimodal distribution that consists of a threat of loss and opportu-
nities for gain (Shapira 1995; Steensma and Corley 2000). Decision making about the type
of technology transaction also entails a threat of loss and opportunities for gain, since the
characteristics of each transaction type have different implications for the source firm.
In this context, we review the backgrounds of the losses and gains for the types of
technology transactions and then develop hypotheses on firms’ strategies for the exploi-
tation of patented technological knowledge assets.

2.2.1 Option perspective

The primary perspective on the governance mode of technology exploitation is the option
perspective, which focuses mainly on the threat of commercial failure of technology (Folta
1998; Kogut 1991). The threat arises from the technology itself, its efficacy, and uncertain
response from the market (Steensma and Corley 2000; Tripsas et al. 1995). For a better
understanding of the uncertainty associated with technology, scholars classify it into
commercial uncertainty and dynamism of technology (Steensma and Corley 2000;
Galbraith and Merrill 1991).
Commercial uncertainty refers to the uncertainty associated with the commercial suc-
cess of a technology (Chakravarthy 1985). Even though technology that is difficult to
imitate leads to the development of products and processes that are more likely to be
superior, the actual value of the outcome and its sustainability in the market may remain

Table 1 Comparison of major technology transaction types

Characteristics Licensing Selling

Transfer of ownership No Yes

Control of the sourcing firm by the technology supplier Yes No
Influence of the commercial outcome on the revenue of the technology supplier Yes No
Initial revenue of the technology supplier (relative) Lower Higher
Risk involved in the expected overall revenue of the technology supplier (relative) Lower Higher
Modified from Chiesa et al. (2008)

123
Licensing versus selling in transactions 255

questionable. Until the final stage of development of products and processes, it is difficult
to predetermine whether the technology will work as designed and how the market will
respond to the products and processes it leads to. Even if the technology works as designed,
that feasibility does not guarantee the success of associated output in the target market
(McGrath et al. 1992). Furthermore, while the probability of commercial success is fairly
indeterminable (Steensma and Corley 2000), the time span from the completion of tech-
nology development at the commercialization stage to the finalization of profit activity in
the market is considerably long (Galbraith and Merrill 1991).
Dynamism of technology refers to the uncertainty in a dynamic environment charac-
terized by a Schumpeterian process of creative destruction. Significant technological
change leads to a decrease in production cost, and the change extirpates obsolescent
technologies while creating a new competitive environment (Robertson and Gatignon
1998; Steensma and Corley 2000). Furthermore, although technological change eventually
takes an identifiable path, the path is not precisely predictable; thus, the initial design and
planning may have to be adjusted at some point (Nelson and Winter 1982; Steensma and
Corley 2000); even a well-established technological paradigm eventually loses its com-
petency (Tushman and Anderson 1986). Accordingly, less obsolete technologies have a
higher probability of becoming commercially successful (Hicks and Hegde 2005).
Depending on the uncertainty associated, firms manage the level of commitment to
technology. When uncertainty is low, maximum benefits may be gained through high
commitment; conversely, when the uncertainty level is high, firms reduce exposure to risk
by exerting low commitment and maintaining a flexible position with respect to the
technology (Kogut 1991; McGrath 1997; Steensma and Corley 2000).
In the case of licensing, which entails a profit share, even though a licensor can con-
servatively estimate the total royalty, the actual amount of the royalty is obviously subject
to the risk of commercial failure, which always exists. Thus, the licensor is required to
extend relatively high commitment to the licensee in various ways, such as through
technical support and marketing, in order to maximize the royalties obtained from the
licensee. On the contrary, in the case of selling, a technology supplier will not have any
incentive to help its technology acquirer to increase sales.
Hence, technology suppliers would choose to license technology whose technological
uncertainty is low, whereas they would choose to sell technology whose technological
uncertainty is high.3
Hypothesis 1 There is a negative relationship between the degree of commercial
uncertainty and the willingness to license technology as opposed to selling it.
Hypothesis 2 There is a negative relationship between the degree of obsolescence and
the willingness to license technology as opposed to selling it.

2.2.2 Transaction cost perspective

Another major perspective related to the governance mode of technology exploitation is

that of the transaction cost. Transaction cost denotes the various costs incurred in obtaining
information and monitoring the fulfillment of a contract involved in a transaction (Robins
1987). On the basis of bounded rationality and opportunism of transaction participants at
3
Regardless of uncertainty, the cost of maintaining the ownership of patents increases every year, as does
the financial burden of ownership (Bessen 2008). Therefore, firms are motivated to sell obsolete technol-
ogies for financial reasons as well.

123
256 S. Jeong et al.

the firm level, the transaction cost perspective suggests that economic activities are per-
formed to minimize not only the production cost but also the associated transaction cost.
Not every firm in a contractual relationship always acts opportunistically; however, dis-
tinguishing opportunistic firms impedes the economic activities of their counterparts owing
to the bounded rationality of management (Williamson 1985). Thus, the threat of the
possibility of opportunistic behavior by the sourcing firm plays a significant role in
deciding the contractual relationship between firms (Steensma and Corley 2000).
When it comes to technology transfer, it is difficult for technology suppliers not only to
search for potential acquirers but also to control and redeem the use of sourced intellectual
property from the sourcing firm after the transaction (Clemons and Row 1992). Therefore,
the characteristics of technology transactions can induce the sourcing firm to generate the
rent dissipation effect, which lowers the source firm’s profit through competition and
makes the sourcing firm a potential rival of the source firm (Arora et al. 2001; Madhok and
Tallman 1998; Gambardella et al. 2007; Palomeras 2007). For this reason, firms are wary
of disclosing developed innovations and have a negative attitude toward external knowl-
edge commercialization (Boyens 1998), namely, the only used here (OUH) syndrome
(Kline 2003). Thus, firms are highly motivated to lower potential competition in the
markets for technology owing to a fear of negative returns on their competencies
(Lichtenthaler and Ernst 2006); in fact, firms selectively give potential technology acquirers
access to their innovations on the basis of the value of the innovations (Palomeras 2007).
The value of innovations is positively correlated with the value of patent rights (Hall
et al. 2005; Harhoff et al. 1999), which includes patent quality (Schankerman and Pakes
1986). On the basis of the premise that patent quality can be framed in the technological
and value dimensions of an innovation, several concepts have been thought to be subsets of
patent quality, such as importance, innovativeness, economic value, strength of protection,
and technological scope (Gambardella et al. 2007; Palomeras 2007). Among the subsets,
scope and importance of patent are highlighted as universal concepts because other con-
cepts are subject to patent citation activities that can be easily biased and distorted
(Gambardella et al. 2007).
Scope, often called broadness, refers to the technological space that the innovation
covers; the larger the scope, the greater the number of potential products that can be
derived (Merges and Nelson 1990; Palomeras 2007). Accordingly, a patent with a broader
scope may attract more potential technology acquirers who share the same technological
scope; thus, the technology affords the patent holder more opportunities to exploit it. By
taking advantage of the opportunities, technology suppliers can maximize their benefit by
retaining the ownership of patents with a broad scope and licensing them to multiple
acquirers. In addition, regardless of the expected profit from the markets for technology,
firms have incentives to hold patents with a broad scope from the perspective of strategic
management since they gain a higher valuation in the venture capital investment process
(Lerner 1994).
In addition, holding a patent with a broad scope may mitigate the possibility of legal
infringements within the technological scope as compared to holding a patent with a
narrow scope. The technology supplier may develop another innovation from the preceding
technological foundation that the integrated scope of the technology supplier’s own patent
portfolio circumscribes, either at present or in the future. Therefore, a technology acquirer
who obtains a patent with a broad scope has a greater likelihood of developing technol-
ogies within the domain of the technology supplier than a technology acquirer who obtains
a patent with a narrow scope. This phenomenon pertains not only to the original tech-
nology acquirer but also to any third parties who may receive the right to use or ownership,

123
Licensing versus selling in transactions 257

and the original technology supplier may thus be unable to control such third parties. Even
if the technology acquirer and the related party do not infringe the technological domain of
the technology supplier, the domain can be vulnerable to legal infringement by any other
firms irrelevant to the contract. For this reason, firms may desire to minimize the loss of
their integrated scope so that they can maintain their competency at creating innovations.
The other concept of patent quality, importance, refers to the resources devoted to the
research process involved in the invention. While many patents are notable and worthless
patents produce zero or even a negative stream of profits when exploited internally
(Gambardella et al. 2007), firms are interested in the commercial exploitation of an
important innovation (Palomeras 2007). Andrews (1979) and Lawani (1986) suggest that
investment in large-scale research projects leads to better and more innovative output.
Firms might therefore invest their resources in what they perceive as a strategically
important domain to improve their competency within the domain.
Correspondingly, firms would desire not to lose their ownership of important technol-
ogy in order to preserve their competency and lower the rent dissipation effect. Selling an
important patent entails a considerable transaction cost for the technology supplier owing
to the permanent loss of an important technological domain. Licensing the patent may also
result in a transaction cost since the licensee can learn about the licenser’s competency
through a contractual relationship. However, the legal terms of licensing grant the tech-
nology supplier greater authority over monitoring and controlling the licensee than the
conditions of selling; therefore, firms may wish to license important patents.
On the contrary, when the technology is relatively unimportant for the owner, firms may
desire to sell the patent in the hope of maximizing their revenue, because the transaction
cost would be relatively low owing to the low importance of the technology. On the other
hand, it may be highly appreciated by other potential users who consider it more important
than the technology supplier.
Hypothesis 3 There is a positive relationship between the degree of scope and will-
ingness to license the technology as opposed to selling it.
Hypothesis 4 There is a positive relationship between the degree of importance and
willingness to license the technology as opposed to selling it.

3 Data and variables

3.1 Data sources

In order to demonstrate firms’ patent exploitation strategies in practice, we employ data

from the National Technology Bank (NTB), the representative online market for tech-
nology in Korea. As of 2003, the total number of technologies to be traded in the entire
online the markets for technology in Korea was 13,630. NTB covered 11,399 or 84 % of
them (Chung and Park 2005).
NTB acts as an intermediary between potential technology traders by providing them an
Internet-based marketplace for technology. Thus, any form of patent and even uncodified
know-how developed in the private sector as well as the public sector can be registered on
the platform of NTB. In order to demonstrate firms’ patent exploitation strategies, we
exclude unpatented technologies and patents not registered by firms. A few foreign patents
are also excluded since they cannot be generalized with the domestic patents, viz., the
standards of approval and coverage of property protection vary by jurisdiction. We exclude

123
258 S. Jeong et al.

Table 2 Definition of technol-

Technology readiness level Value
ogy readiness level of NTB
Technology concept 1
Application formulated 2
Component and/or breadboard validation in a laboratory 3
environment
Component and/or breadboard validation in the relevant 4
environment
System prototype demonstration 5
Actual system completed through successful mission 6
operations
Advancing the market 7

technologies that have not yet been approved for registration by the Korean Intellectual
Property Office (KIPO), because the number of approved claims and the number of IPC
after the examination by KIPO can be different from those before the examination. Fur-
thermore, some of the patents can be ultimately rejected.
As of June 2008, 6,683 patents are registered in the NTB database. However, NTB does
not provide detailed information about the patents. Thus, we construct details about the
patents by matching the patent registration codes in the NTB database with those in the
Korea Intellectual Property Rights Information Service (KIPRIS) database: the number of
claims, number of inventors, application year, owners’ patent portfolio, and the techno-
logical field.

3.2 Variables

At the time of registering their patents on the platform of NTB, firms are required to choose
the desired transaction types from among the given multiple options. That is, firms can
choose not only licensing or selling but also both licensing and selling simultaneously as
their desired transaction type. After the registration process, the desired transaction types
of the patents are publicly displayed to potential technology acquirers through the NTB
platform. We construct dummy variables—i.e., License and Sell—that indicate the options
for licensing and selling respectively.
Technology readiness level is used as a proxy for the commercial uncertainty of patents.
A technology ready for commercialization would be more practicable than other tech-
nologies that need to be fully developed, since the former may generate a profit sooner than
the latter. For this reason, in terms of doubts about whether a technology will actually
function as designed, the former has lower technological uncertainty than the latter. That
is, the technology readiness level is correlated to risk and technological uncertainty
(Shenhar et al. 2005). The problem is that using general patent information does not
provide a clear background for assessing the technology readiness level. However, NTB
data enables us to measure the technology readiness level of each desired technology.
When technology suppliers register their patents, they are required to disclose the tech-
nology readiness level of their patents and choose one of the discrete levels specified by
NTB, as shown in Table 2.4 According to the specified levels, we code the technology
4
The Department of Defense (DoD) and the National Aeronautics and Space Administration (NASA) in the
United States also use a similar definition of technology readiness level.

123
Licensing versus selling in transactions 259

readiness level of NTB patents from 1 to 7 linearly, as a variable to indicate the degree of
commercial uncertainty of the patents.
Another uncertainty-related variable is Obsolescence, which corresponds to dynamism
of technology. This variable is determined for every year from the application year of the
patent to 2008. In this setting, an obsolescent technology is marked with a high value of
Obsolescence, indicating that it is far from the extant technological paradigm. One can
argue about why the application year is used rather than the registration year. This is
because the patent is protected by patent law after the application day owing to ‘‘the first-
to-file system’’ in Korea. Thus, from the application day, competitors or any technologi-
cally relevant group can recognize the patent as a new source of the owner’s innovation.
We use the number of claims as a proxy for Scope. A greater number of claims indicate
that a patent has a broader technological scope (Lanjouw and Schankerman 1999; Lerner
1994; Palomeras 2007). Firms carefully review information on claims when transacting
patents for acquisition or licensing (Gambardella et al. 2007; Lerner 1994) as the number
of claims is often used to assess economic value. However, the number of claims has to be
considered carefully, since some researchers argue that it could indicate the strength of a
patent (Gambardella et al. 2007). For a precise analysis, we use the number of claims
authorized by KIPO rather than the number of claims at the application stage.
The number of inventors is used as an indicator of Importance. It primarily suggests the
extent of resource investment because the number of inventors correlates to the resource
inputs for achieving expected outputs during the research and development process. The
number of inventors also implies the factors contributing to innovation. Palomeras (2007)
postulates that the number of inventors might be related to the characteristics of innovation
such as diversity of knowledge sources (Palomeras 2007) and complexity of the invention
(Reitzig 2002). These meanings of importance can be interpreted in diverse ways owing to
the broad use of terminology.5 In this study, Importance is regarded as the significance of
the patented invention or the scale of the research project.
As a control, we introduce each technological profile primarily in order to observe the
pure influence of the explanatory variables—e.g., Technology readiness level—on the
choice of technology transaction type, since the strategies of technology exploitation could
be subject to coherence with the owner’s technological profile (Chiesa et al. 2008). In
addition, we attempt to explain the relationship between the choice of technology trans-
action type and each technological profile.
Using Patel and Pavitt’s (1997) frame, we divide a firm’s technology profile into
quadrants such as Core, Background, Niche, and Marginal using the revealed technology
advantage (RTA) and patent share (PS). The RTA is defined as the firm’s share in total
patenting from 2004 to 2008 in each of the three-digit IPC technological classes divided by
the firm’s share of total patenting in all fields, as shown in Formula 1. This value reflects
the distinctive technical competencies of the firm.
PPij Number of patents of a patent holder in the specified technological field
Pij Total number of patents of a patent holder
RTA ¼ Pj ¼ Total number of patents in the specified technological field
ð1Þ
P
PP i ij
Total number of patents
i j
Pij

i the technological field, j the patent holder.

5
Gambardella et al. (2007) have a different concept of importance. They consider a patent whose owner’s
share in the technological field is high to be important.

123
260 S. Jeong et al.

PS is calculated as the share in total patenting for the same time span and in the same
technological classes with an RTA, as shown in Formula 2. This value reflects the tech-
nological commitment of the firm.
Pij Number of patents of a patent holder in the specified technological field
PS ¼ P ¼
j P ij Total number of patents of a patent holder
ð2Þ
i the technological field, j the patent holder.
Then, by using the standard of Gambardella et al. (2007), we define each technology
profile as follows6:
• Core: dummy equal to 1 if patent share [3 % and RTA [ 2
• Background: dummy equal to 1 if patent share [3 % and RTA \ 2
• Niche: dummy equal to 1 if patent share \3 % and RTA [ 2
• Marginal: dummy equal to 1 if patent share \3 % and RTA \ 2
Patentstock denotes the firm’s patenting activity in 2002–2007. A high value of pat-
entstock indicates greater experience and a profound understanding in dealing with patent
information and technology transactions, as well as the larger organizational size
(Palomeras 2007). According to the data from KIPRIS, the number of patents held by each
firm varies from 1 to more than 10,000; hence, the natural log transformed number of the
patentstock is used owing to possible statistical skewness.
We categorize the technological fields and use them as dummy variables on the basis of
the first-digit IPC classification established by World Intellectual Patent Organization
(WIPO). Scholars have suggested that some specific technological fields have distinct
transfer characteristics because of varying levels of codifiability (Arora and Forfuri 2003).
The definitions of variables are summarized in Table 3 with expected signs indicated in
the rows alongside the hypotheses.
Table 4 reports the means, standard deviations, and minimum and maximum values of
the independent and control variables. Interestingly, more than half of the patents in this
sample are categorized as core (61 %), since the possibility of exploiting core technology
is relatively low (Palomeras 2007). Table 5 shows that there is no critical correlation
between independent variables; thus, a serious multicollinearity problem would not occur.

4 Methodology

To account for the firms’ strategic behavior in terms of choice of technology transfer type,
we estimate the probability of a patent holder licensing and/or selling a given patented
innovation. In this model, firms are assumed to make decisions with the objective of own
utility maximization. We denote licensing as l and selling as s, where l, s = 1 for adoption
of technology transfer, and l, s = 0 for non-adoption of technology transfer. The utility
function that ranks the preference of the ith firm is then assumed to be a function of the
characteristics of the technology and the firm, ‘‘X’’—uncertainty, transaction cost, pat-
entstock, etc.—and a disturbance term having a zero mean:

6
Palomeras (2007) defines core patent differently from our study by judging a patent in the technological
field that accounts for more than 5 % of the patent holder’s entire patent portfolio to be a core patent.

123
Licensing versus selling in transactions 261

Table 3 Definition of dependent and independent variables

Variable name Description Proxy Expected sign

Dependent variables
License Dummy equal to 1 if the technology supplier
desires to license out the technology and
equal to 0 if not
Sell Dummy equal to 1 if a technology supplier
desires to sell the technology and equal to 0
if not
Explanatory variables License Sell
and controls
Technology readiness Technology readiness level from 1 to 7, Uncertainty ? -
level marked by the technology supplier
Obsolescence Time gap between the application year and Uncertainty - ?
2008
Scope Number of claims of the granted patent Patent Quality ? -
Importance Number of inventors of the granted patent Patent Quality ? -
Core Dummy equal to 1 if core technology in each Portfolio
IPC 3 digit and equal to 0 if not control
Background Dummy equal to 1 if background technology Portfolio
in each IPC 3 digit and equal to 0 if not control
Marginal Dummy equal to 1 if marginal technology in Portfolio
each IPC 3 digit and equal to 0 if not control
Niche Dummy equal to 1 if niche technology in each Portfolio
IPC 3 digit and equal to 0 if not control
Patentstock Natural logarithm numbers of the granted Firm control
patents of the firm listed in the patent
IPC_A Dummy equal to 1 if the first digit of IPC of Technological
the granted patent is A and equal to 0 if not control
IPC_B Dummy equal to 1 if the first digit of IPC of Technological
the granted patent is B and equal to 0 if not control
IPC_C Dummy equal to 1 if the first digit of IPC of Technological
the granted patent is C and equal to 0 if not control
IPC_D Dummy equal to 1 if the first digit of IPC of Technological
the granted patent is D and equal to 0 if not control
IPC_E Dummy equal to 1 if the first digit of IPC of Technological
the granted patent is E and equal to 0 if not control
IPC_F Dummy equal to 1 if the first digit of IPC of Technological
the granted patent is F and equal to 0 if not control
IPC_G Dummy equal to 1 if the first digit of IPC of Technological
the granted patent is G and equal to 0 if not control
IPC_H Dummy equal to 1 if the first digit of IPC of Technological
the granted patent is H and equal to 0 if not control

Ui1 ðXÞ ¼ b1 Xi þ ei1 for adoption and Ui0 ðXÞ ¼ b0 Xi þ ei0 for non-adoption ð3Þ
Only if Ui1 [ Ui0 will the ith firm adopt the alternative. Thus, for the ith firm, the prob-
ability of adoption is denoted as follows:
}ð1Þ ¼ }ðUi1 [ Ui0 Þ ð4Þ

123
262 S. Jeong et al.

Table 4 Descriptive statistics

Variable Obs Mean SD Min. Max.

Technology readiness level 6683 4.004489 0.364519 1 7

Obsolescence 6683 8.51444 3.233073 1 20
Scope 6683 5.84887 5.269927 1 71
Importance 6683 1.522969 1.102311 1 11
Core 6683 0.614694 0.486704 0 1
Background 6683 0.315876 0.464899 0 1
Marginal 6683 0.064492 0.245646 0 1
Niche 6683 0.004938 0.070102 0 1
Patentstock 6683 9.701034 1.691099 0 10.57513
IPC_A 6683 0.003142 0.055972 0 1
IPC_B 6683 0.005836 0.076174 0 1
IPC_C 6683 0.011223 0.105348 0 1
IPC_D 6683 0.001796 0.04234 0 1
IPC_E 6683 0.000898 0.029952 0 1
IPC_F 6683 0.005088 0.071151 0 1
IPC_G 6683 0.256921 0.436968 0 1
IPC_H 6683 0.715098 0.451402 0 1

}ð1Þ ¼ }ðb1 Xi þ ei1 [ b0 Xi þ ei0 Þ ð5Þ

}ð1Þ ¼ }ðei0  ei1 \b1 Xi  b0 Xi Þ ð6Þ
}ð1Þ ¼ }ðei \bXi Þ ð7Þ
}ð1Þ ¼ UðbXi Þ ð8Þ
where U is the cumulative distribution function for e. The functional form for U depends on
the assumptions about e. A probit model is based on the normal distribution of e; thus, the
probability of firm i licensing and selling its technology, respectively, is given by:
ZbXi  2
1 t
Ul ðbXi Þ ¼ pffiffiffiffiffiffi exp dt ð9Þ
2p 2
1

ZbXi  2
1 t
Us ðbXi Þ ¼ pffiffiffiffiffiffi exp dt: ð10Þ
2p 2
1

The two equations can be estimated individually by single equation probit methods.
However, such methods do not consider the correlation between the disturbances el and es
of the underlying stochastic utilities function (Greene 2003). The alternatives—i.e.,
licensing and selling—could be correlated since firms may choose licensing and/or selling
simultaneously in an actual transaction. As mentioned in Chap. 3, cases of such patents
have been identified among those registered on the NTB platform.
For this reason, in this paper, a bivariate probit model is employed to circumvent the
inadequacies of a single probit or logit model. This will enable us to examine the relative
preference for licensing and selling under the given conditions for the explanatory

123
 Table 5 Correlations of variables

(1) (2) (3) (4) (5) (6) (7) (8)

(1) Technology readiness level 1

(2) Obsolescence 0.0577* 1
(7) Scope 0.0127 -0.1217* 1
(8) Importance 0.0463* -0.0582* 0.1585* 1
(3) Core -0.0164 -0.0758* -0.0925* -0.0857* 1
(4) Background 0.0066 0.0608* 0.0968* 0.0832* -0.8583* 1
(5) Marginal 0.0202 0.0234 0.0108 0.0053 -0.3316* -0.1784* 1
Licensing versus selling in transactions

(6) Niche -0.0009 0.0410* -0.0373* 0.0247* -0.0890* -0.0479* -0.0185 1

(9) Patentstock -0.2056* -0.0184 0.0637* -0.1001* -0.1786* 0.1877* 0.0492* -0.1774*
(10) IPC_A 0.0947* -0.0428* 0.0026 0.0219 -0.0105 -0.0382* 0.0941* -0.004
(11) IPC_B 0.1931* -0.0177 -0.0213 0.0314* 0.0324* -0.0521* 0.0039 0.1067*
(12) IPC_C 0.0922* 0.0419* -0.0107 0.2252* 0.0464* -0.0724* 0.0299* 0.0533*
(13) IPC_D 0.0383* 0.0097 -0.0182 0.0632* 0.0263* -0.0288* -0.0111 0.0474*
(14) IPC_E 0.1093* -0.0218 0.0141 0.0447* 0.0237 -0.0204 -0.0079 -0.0021
(15)IPC_F 0.0395* -0.0023 -0.0279* -0.0168 0.0480* -0.0486* -0.0102 0.0250*
(16) IPC_G 0.0303* 0.012 0.1374* 0.0674* -0.4176* 0.4624* -0.0373* -0.0365*
(17) IPC_H -0.1123* -0.0123 -0.1220* -0.1321* 0.3776* -0.4055* 0.02 -0.0028

(9) (10) (11) (12) (13) (14) (15) (16) (17)

(1) Technology readiness level

(2) Obsolescence
(7) Scope
(8) Importance
(3) Core
(4) Background
263

123
 Table 5 continued
264

(9) (10) (11) (12) (13) (14) (15) (16) (17)

123
(5) Marginal
(6) Niche
(9) Patentstock 1
(10) IPC_A -0.1165* 1
(11) IPC_B -0.3196* -0.0043 1
(12) IPC_C -0.3092* -0.006 -0.0082 1
(13) IPC_D -0.0580* -0.0024 -0.0032 -0.0045 1
(14) IPC_E -0.1561* -0.0017 -0.0023 -0.0032 -0.0013 1
(15)IPC_F -0.1170* -0.004 -0.0055 -0.0076 -0.003 -0.0021 1
(16) IPC_G 0.1144* -0.0330* -0.0451* -0.0626* -0.0249* -0.0176 -0.0420* 1
(17) IPC_H 0.0640* -0.0889* -0.1214* -0.1688* -0.0672* -0.0475* -0.1133* -0.9316* 1
* Indicates significance at the 5 % level
S. Jeong et al.
 Table 6 Bivariate probit estimations of the determinants of desired technology transaction type
Model 1 Model 2 Model 3 Model 4

License Sell License Sell License Sell License Sell

Technology readiness level 0.625*** -0.520*** 0.615*** -0.553*** 0.626*** -0.520*** 0.626*** -0.533***
(0.054) (0.082) (0.052) (0.063) (0.054) (0.082) (0.062) (0.118)
Obsolescence -0.034*** 0.063*** -0.032*** 0.061*** -0.031*** 0.060*** -0.030*** 0.061***
(0.007) (0.022) (0.007) (0.022) (0.007) (0.022) (0.007) (0.023)
Scope 0.038*** -0.011 0.038*** -0.012 0.039*** -0.011 0.039*** -0.011
(0.008) (0.009) (0.008) (0.009) (0.008) (0.009) (0.008) (0.009)
Importance 0.104*** -0.106*** 0.105*** -0.109*** 0.114*** -0.111*** 0.114*** -0.111***
Licensing versus selling in transactions

(0.032) (0.037) (0.032) (0.037) (0.032) (0.037) (0.032) (0.037)

Core -0.346*** 0.197
(0.069) (0.152)
Background 0.287*** -0.069
(0.077) (0.158)
Marginal 0.444*** -0.371*
(0.151) (0.209)
Niche 0.007 6.386
(0.243) (3.7E?5)
Patentstock 0.235*** 0.155*** 0.237*** 0.144*** 0.244*** 0.150*** 0.247*** 0.144***
(0.013) (0.025) (0.013) (0.021) (0.013) (0.023) (0.013) (0.028)
IPC_A 0.081 8.174 0.187 9.394 0.066 8.099 0.097 8.235
(0.691) (3.0E?4) (0.702) (1.3E?6) (0.696) (2.6E?4) (0.689) (3.1E?4)
IPC_B -0.094 1.569** 0.018 1.678 -0.079 1.548** -0.070 1.595**
(0.574) (0.608) (0.595) (0.701) (0.575) (0.607) (0.575) (0.626)
IPC_C 0.350 0.628 0.435 0.740 0.324 0.656 0.334 0.688
(0.564) (0.544) (0.585) (0.631) (0.566) (0.544) (0.566) (0.561)
IPC_E -2.014** 7.863 -1.932** 8.851 -2.021** 7.848 -2.026*** 7.978
(0.780) (1.3E?6) (0.796) (7.0E?7) (0.780) (1.3E?6) (0.781) (1.8E?6)
265

123
 Table 6 continued
266

Model 1 Model 2 Model 3 Model 4

123
License Sell License Sell License Sell License Sell

IPC_F 0.131 7.565 0.212 8.867 0.130 7.542 0.126 7.700

(0.618) (1.3E?4) (0.637) (6.2E?5) (0.621) (1.3E?4) (0.622) (1.7E?4)
IPC_G -0.193 0.680 -0.097 0.764 -0.045 0.586 -0.065 0.668
(0.538) (0.532) (0.560) (0.618) (0.539) (0.530) (0.540) (0.552)
IPC_H -0.346 1.077** -0.277 1.198* -0.360 1.080** -0.377 1.140**
(0.535) (0.528) (0.556) (0.610) (0.536) (0.528) (0.537) (0.551)
_Cons -2.649*** 1.967*** -3.035*** 2.251*** -3.056*** 2.206*** -3.062*** 2.212***
(0.605) (0.757) (0.612) (0.727) (0.600) (0.714) (0.621) (0.849)
/artrho -1.498 -16.572 -1.504 -1.636
Rho -0.905 -1 -0.906 -0.927
Groups 26 26 26 26
Wald v2 test 904.06*** 971.41*** 881.48*** 891.95***
*, **, and *** indicate significance at the 10, 5, and 1 % level, respectively
S. Jeong et al.
Licensing versus selling in transactions 267

variables. The bivariate probit model is based on the joint distribution of two normally
distributed variables, as specified below (Brorsen et al. 1996; Greene 2003):
2 2
ðe þ es 2q el es Þ
1 l
f ðl; sÞ ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffi e 2ð1q2 Þ ð11Þ
2prl rs 1  q 2

e  ll
el ¼ ð12Þ
rl
e  ls
es ¼ ð13Þ
rs
where q is the correlation between l and s. ll, ls, rl, rs, and q denote the means and
standard deviations of the marginal distribution of l and s and the covariance. Following
Greene (1998), we assume that a bivariate normal distribution has the following
characteristics:
ll ¼ ls ¼ 0; rl ¼ rs ¼ 1; rls ¼ qrl rs : ð14Þ
The underlying algorithm for bivariate probit estimation is examined using the biprobit
option on STATA 10. In addition, as an exploratory study, we introduce each technological
profile—i.e., Core, Background, Niche, and Marginal—separately in four individual
models in order to observe the pure influence of the explanatory variables on the choice of
technology transaction and in order to analyze the relationship between the choice of
technology transaction and each technological profile.

5 Empirical results and discussion

Table 6 shows the empirical results. The coefficients of each variable are arranged
according to the alternatives, License and Sell, in four individual models that have different
technological profiles. The standard deviations of each coefficient are displayed in
parentheses below each coefficient. The models illustrate that the relationship between the
two alternatives is substitutive because the rho estimates are around -1 in every model.
As for the result of estimations, we find strong support for hypotheses 1 and 2. As the
Technology Readiness Level of the patent increases, the owner of the patent tends to prefer
to license the patent and is reluctant to sell it. Similarly, the owner of a patent with more
Obsolescence tends to prefer to sell the patent rather than licensing it out. Impressively, all
estimated coefficients related to technological uncertainty in the four models are strongly
significant at the 1 % level.
The first hypothesis on transaction cost, Hypothesis 4, is generally supported by the
estimation results. As for Importance, as the number of inventors increases, the willingness
to license out also increases, while the willingness to sell decreases in the same condition.
The other hypothesis on transaction cost, Hypothesis 3, is partially supported owing to the
exception of the willingness to selling on Scope. The relationship between Scope and
Licensing turns out to be positive as expected, and the coefficient is significant. On the
other hand, although the relationship between Scope and Sell was also expected to be
negative, the coefficient remains insignificant in every model. Thus, these results partially
support Hypothesis 3.
The above results might indicate that a patent with a broad scope may give the owner
more opportunities to license out, so that he can both license the patent (Irish 2005;

123
268 S. Jeong et al.

Halstead 1993) or prevent larger losses in the technology domain. However, at the same
time, the attractiveness of a broad scope to potential technology acquirers can overshadow
the owner’s threat of potential infringement, because selling can fortify the advantage from
the perspective of pure revenue. In fact, when entering the markets for technology, firms
tend to suggest that their technology has a broad scope in order to gain a greater benefit
than they could by presenting their technology as one with a narrow scope (Palomeras
2007). In their choice of a transaction type, firms are faced with both an advantage and the
threat of infringement by rivals. Thus, these mixed motives would result in an insignificant
relationship between willingness to sell and scope.
As for the control variables, we find that firms prefer to license Marginal and Back-
ground, while they prefer not to license out Core and not to sell Marginal. These results
enables us to conclude that firms prefer to license a patent with a low technological
advantage and not license a patent with a high technological advantage, as shown in Fig. 1.
Such a conclusion could be explained by the threat of technology leakage. Generally,
technological collaboration including licensing activity not only provides the sourcing
firms an opportunity to grasp the transferred knowledge but also causes unintended
knowledge leakage for the source firms (Hamel et al. 1989). The collaborators tend to try to
gain access to core competencies and value-creating disciplines of the partner by observing
the partner’s activity on how to manage and how to carry out innovation generated through
firm-specific skills (Hamel 1991). Although a source firm can obtain a reward from the
sourcing firm in the short term, failure can occur as a result of a collaborator learning about
competencies in the long term. Such failure can render the source firm to be first dependent
on the collaborator in the technological relationship, then redundant within the partnership,
and finally vulnerable outside the partnership (Prahalad and Hamel 1990).
In this context, technology suppliers might be led to refuse to grant the technology
acquirer access to core and niche technology, i.e., a patent with high RTA, because the
technological advantage is inherent not only in the technology but also in the organiza-
tional and technological assets related to the technology (Palomeras 2007; Teece 1986).
Contrarily, firms could be motivated to obtain access to partners’ technological competitive
advantages by associating only those technologies with their partners whose technological
competitive advantage is low, such as marginal and background technology. Thus, they
could learn about partners’ competencies while minimizing leakage.

Fig. 1 The relationship between

the willingness of licensing and
the revealed technology advance

123
Licensing versus selling in transactions 269

However, a few insignificant results do not match the aforementioned explanation. For
example, a demonstration on niche technology does not provide any statistical evidence,
perhaps because the threat induced by niche technology is not crucial owing to its minor
position in a firm’s technological portfolio; hence, the area of niche technology is rounded
with a dotted line in Fig. 1. As another example, the relationship between the willingness
to sell and core technology also turns out to be insignificant, although its sign matches the
explanation. This exception might be because core technology serves as an indication of
the reliability of an innovation to potential technology acquirers, thereby reducing asym-
metric information. In particular, technology acquirers are fairly motivated to obtain
ownership of core technology owing to the benefit of unlimited usage. Therefore, this
indication might increase the potential gain for technology suppliers when they choose
selling (Palomeras 2007); thus, the expected gain from selling core technology might
mitigate the threat of losing their competence.
Patentstock shows a positive sign for both alternatives at the 1 % significance level.
These coefficients for both alternatives show that firms with extensive knowledge assets
are more willing to exploit their technology externally than are those with limited
knowledge assets, regardless of the type of transaction. With regard to technological fields,
it may be preferable to sell patents in the fields of performing operations and transporting
(associated with IPC_B) and the field of electricity (associated with IPC_H), while patents
in fixed contractions field associated with IPC_E prefer to license out. This is probably
because the technological and commercial circumstances in the fields related to IPC_B and
ITC_H change rapidly (Seo et al. 1996). Thus, there is a greater possibility of patents in
this field becoming obsolescent in terms of the technological paradigm; therefore, it might
be preferable to sell these patents. In contrast, the field of fixed construction changes
relatively less than the others, and there is a lower possibility of patents in this field
becoming obsolescent in terms of the technological paradigm; thus, it might be preferable
to license out these patents.

6 Conclusions

In this study, we empirically demonstrate firms’ technology exploitation strategies asso-

ciated with the types of technology transactions by detailing related opportunities and
threats in the theoretical specification. Overall, our empirical results show that the major
types of technology transactions—i.e., licensing and selling—have a substitutive rela-
tionship. In addition, firms desire to retain ownership of technology with a low degree of
commercial uncertainty; therefore, they avoid selling technology in this situation. On the
other hand, firms have low willingness to license out technology with high uncertainty;
instead, they desire to sell it. Transaction cost also plays an important role in the choice—
firms desire to secure an opportunity to exploit important technology internally. Similarly,
they desire to license ownership of technology with a narrow scope in order to avoid the
threat of potential infringement from potential rivals, although the results do not support
that firms tend to avoid losing ownership of technology with a broad scope.
On the basis of the results, this study offers the following insights. First, it reveals that
refinement of the type of technology transaction should be highlighted. Previous empirical
research focuses only on the step prior to decision making about the type of transaction—
i.e., research questions on what to transfer or what to patent. However, given the theo-
retical taxonomy of technology transactions, our research shows that the relationship
between the two major types is substitutive, and that firms confront the threats and

123
270 S. Jeong et al.

opportunities in the markets for technology even at the moment of choosing the transaction
type. These findings pinpoint the importance of choosing the technology transaction type
for firms in the markets for technology.
Second, the empirical demonstration on strategic behaviors on the technology supply
side proves the previous theoretical concept (i.e., Chiesa et al. 2008). Our study clearly
shows that the choice of technology transaction type is determined by the degree of
uncertainty and transaction cost, as well as subsidiary factors such as technological
position in the owner’s patent portfolio, owner’s patenting activity, and technological
field.
Lastly, potential technology acquirers could build pragmatic and differentiating strat-
egies corresponding to technology suppliers’ strategies, which are practically detailed by
this study. Owing to the asymmetric information problem in the markets for technology,
technology acquirers might not be completely sure of the characteristics of technology
suppliers’ technology. By understanding technology suppliers’ choice of technology
transactions using pragmatic indicators such as the number of inventors, claims, applica-
tion year, and technology readiness level, a potential technology acquirer can postulate
technology suppliers’ motivation before a transaction. Thus, potential technology acquirers
can prepare for possible opportunistic behavior by technology suppliers, such as learning
about competencies and insincere support.
Our study has certain limitations. Our noble and unique data set, which enables this
study, presents a restricted view by dealing only with online market for technology and
not covering the markets for technology as a whole. Despite the NTB data being the
most representative, there are a number of unobservable intellectual asset transactions
outside online markets for technology. For this reason, the estimation we undertake
could be distorted by firms’ motives to join the online markets for technology. In
addition to the generalization problem, the regional characteristics could affect the
determinants and change the results of this study. Although Korea is the fourth largest
country in terms of the total number of patent applications and has strong industries
that are closely involved in intellectual assets (e.g., IT and manufacturing industries),
the development of the observed phenomenon into a universal one requires more
evidence from other regions.
This study can suggest certain directions for future research. First, the actual gap
between desired transactions and actual technology transactions can be studied. Not every
technology that a firm wishes to transact is traded in the markets for technology, although
technology suppliers generally have a greater bargaining power. The transaction will
consequently entail interaction with the buyer; thus, the characteristics of technology
acquirers may also prominently affect the actual choice of technology transfer type.
Second, in order to illustrate the relationship between technological profile and type of
technology transaction, comprehensive background theories must be developed with ample
evidence. Chiesa et al. (2008) also arrived at a similar conclusion through a case study in
their research; however, it was not convincing enough to be generally accepted. Therefore,
further study can offer a more thorough and applicable strategic framework to technology
traders. Lastly, as mentioned in the limitations of this research, a study with more generic
data in terms of regions and trading agents should be undertaken to fortify the generality of
our argument. In sum, a more thorough analysis of the aforementioned issues should
accompany a concrete, systemic approach and careful definition of strategies so that it can
provide a rich understanding on what governs choice of technology transaction from a
multi-theoretical perspective.

123
Licensing versus selling in transactions 271

References

Andrews, F. M. (1979). Scientific productivity: The effectiveness of research groups in six countries.
Cambridge: Cambridge University Press.
Arora, A. (1996). Contracting for tacit knowledge: The provision of technical service in technology
licensing contracts. Journal of Development Economics, 50, 233–256.
Arora, A., & Forfuri, A. (2003). Licensing the market for technology. Journal of Economic Behavior &
Organization, 52, 277–295.
Arora, A., Forfuri, A., & Gambardella, A. (2001). Markets for technology and their implications for
corporate strategy. Industrial and Corporate Change, 10, 419–451.
Bessen, J. (2008). The value of U.S. patents by owner and patent characteristics. Research Policy, 37,
932–945.
Boyens, K. (1998). Externe Verwertung von Technologischem Wissen. Wiesbaden: DUV, Dt. Univ.-Verl.
Brorsen, B. W., Dicks, M. R., & Just, W. B. (1996) Regional and farm structure effects of planting
flexibility. Review of Agricultural Economics, 18, 467–475.
Chakravarthy, B. S. (1985). Business-government partnership in emerging industries: Lessons from the
American synfuels experience. Advances in Strategic Management, 3, 257–275.
Chesbrough, H. W. (2003). Open innovation: The new imperative for creating and profiting from tech-
nology. Boston, MA: Harvard Business School Press.
Chiesa, V., Manzini, R., & Pizzurno, E. (2008). The market for technological intangibles: A conceptual
framework for commercial transactions. International Journal of Learning and Intellectual Capital,
5(2), 186–207.
Chung, K., & Park, S. (2005). A study of local large corporations technology marketing activating programs.
Journal of Korea Technology Innovation Society, 8(1), 1–28.
Clemons, E., & Row, M. (1992). Information technology and industrial cooperation: The changing eco-
nomics of coordination and ownership. Journal of Management Information Systems, 9, 9–28.
Elton, J., Shah, B., & Voyzey, J. (2002). Intellectual property: Partnering for profit. The Mckinsey Quarterly,
4, 59–67.
Folta, T. (1998). Governance and uncertainty: The tradeoff between administrative control and commitment.
Strategic Management Journal, 19, 1007–1028.
Ford, D. (1988). Develop your technology strategy. Long Range Planning, 21(5), 85–95.
Ford, D., & Ryan, C. (1981). Taking technology to market. Harvard Business Review, 59(2), 117–126.
Galbraith, C. S., & Merrill, G. B. (1991). The effect of compensation program and structure on SBU
competitive strategy: A study of technology intensive firms. Strategic Management Journal, 12,
353–370.
Gambardella, A., Giuri, P., & Luzzi, A. (2007). The market for patents in Europe. Research Policy, 36,
1163–1183.
Granstrand, O. (2000). The economics and management of intellectual property: Towards intellectual
capitalism. Northampton, MA: Edward Elgar Publishing.
Greene, W. (1998). Gender economics courses in liberal arts colleges: Further results. Journal of Economic
Education, 29(4), 291–300.
Greene, W. (2003). Econometric analysis. Upper Saddle River, NJ: Prentice Hall.
Hall, B. H., Jaffe, A., & Trajtenberg, M. (2005). Market value and patent citations. RAND Journal of
Economics, 36, 16–38.
Halstead, R. (1993). Protecting intellectual property: Understanding and using trademarks, patents, design,
and copyright in business. London: ICSA Publishing.
Hamel, G. (1991). Competition for competence and interpartner learning within international strategic
alliances. Strategic Management Journal, 12(1), 83–103.
Hamel, G., Doz, Y. L., & Prahalad, C. K. (1989). Collaborate with your competitors and win. Harvard
Business Review, 67(1), 133–139.
Harhoff, D., Narin, F., Scherer, F. M., & Vopel, K. (1999). Citation frequency and the value of patented
inventions. The Review of Economics and Statistics, 81, 511–515.
Hicks, D., & Hegde, D. (2005). Highly innovative small firms in the markets for technology. Research
Policy, 34, 703–716.
Irish, V. (2005). Intellectual property rights for engineers (2nd ed.). London: The Institute of Electrical
Engineers.
Kline, D. (2003). Sharing the corporate crown jewels. MIT Sloan Management Review, 44, 89–93.
Kogut, B. (1991). Joint ventures and the option to expand and acquire. Management Science, 37, 19–33.

123
272 S. Jeong et al.

Lanjouw, J., & Schankerman, M. (1999). The quality of ideas: Measuring innovation with multiple indi-
cators. NBER, WP 7345.
Lawani, S. M. (1986). Some bibliometric correlates of quality in scientific research. Scientometrics, 9,
13–25.
Lerner, J. (1994). The importance of patent scope: An empirical analysis. RAND Journal of Economics, 25,
319–333.
Lichtenthaler, U. (2005). External commercialization of knowledge: Review and research agenda. Inter-
national Journal of Management Reviews, 7, 231–255.
Lichtenthaler, U. (2008). Open innovation in practice: An analysis of strategic approaches to technology
transactions. Transactions on Engineering Management, 55(1), 148–157.
Lichtenthaler, U., & Ernst, H. (2006). Attitudes to externally organising knowledge management tasks: A
review, reconsideration and extension of the NIH syndrome. R&D Management, 36(4), 367–386.
Madhok, A., & Tallman, S. B. (1998). Stormy open marriages are better: Evidence from US, Japanese and
French cooperative ventures in commercial aircraft. Columbia Journal of World Business, 22, 87–95.
McGrath, R. (1997). A real options logic for initiating technology positioning investments. Academy of
Management Review, 22, 974–996.
McGrath, R., MacMillan, I., & Tushman, M. (1992). The role of executive team actions in shaping dominant
designs: Towards the strategic shaping of technological progress. Strategic Management Journal, 13,
137–161.
Megantz, R. C. (2002). Technology management: Developing and implementing effective licensing pro-
grams. New York: Wiley.
Merges, R. P., & Nelson, R. R. (1990). On the complex economics of patent scope. Columbia Law Review,
90, 839–916.
Motohashi, K. (2008). Licensing or not licensing? An empirical analysis of the strategic use of patents by
Japanese firms. Research Policy, 37(9), 1548–1555.
Nelson, R., & Winter, S. (1982). An Evolutionary Theory of Economic Change. Belknap: Cambridge.
Palomeras, N. (2007). An analysis of pure-revenue technology licensing. Journal of Economics & Man-
agement Strategy, 16, 971–994.
Park, H. (2002). A study on improved roles of cyber technology market in Korea. The Journal of Internet
Electronic Commerce Research, 2(2), 143–165.
Patel, P., & Pavitt, K. (1997). The technological competencies of the world’s largest firms: Complex and
path-dependent, but not much variety. Research Policy, 26, 141–156.
Prahalad, C. K., & Hamel, G. (1990). The core competence and the corporation. Harvard Business Review,
3, 71–91.
Pratt, J. W. (1964). Risk aversion in the small and in the large. Econometrics, 32, 132–139.
Reitzig, M. (2002). Improving patent valuation methods for management: Validating new indicators by
understanding patent strategies. LEFIC working paper 9.
Robertson, T. S., & Gatignon, H. (1998). Technology development mode: A transaction cost conceptuali-
zation. Strategic Management Journal, 19, 515–531.
Robins, J. A. (1987). Organizational economics: Notes on the use of transaction cost theory in the study of
organizations. Administrative Science Quarterly, 32, 68–86.
Schankerman, M., & Pakes, A. (1986). Estimates of the value of patent rights in European countries during
the post-1950 period. Economic Journal, 96, 1052–1076.
Seo, S., Jeong, K., Choi, S., Lee, G., Park, K., Hng, J., et al. (1996). A study on the establishment of
investment strategies of R&D projects in information and telecommunications area. Daejeon: ETRI.
Shapira, Z. (1995). Risk taking: A managerial perspective. New York: Russell Sage Foundation.
Shenhar, A. J., Dvir, D., Milosevic, D., Mulenberg, J., Patanakul, P., Reilly, R., et al. (2005). Toward a
NASA-specific project management framework. Engineering Management Journal, 17(4), 8–16.
Steensma, K. H., & Corley, K. G. (2000). On the performance of technology-sourcing partnerships: The
interaction between partner interdependence and technology attributes. Academy of Management
Journal, 43, 1045–1067.
Teece, D. J. (1986). Profiting from technological innovation: Implications for integration, collaboration,
licensing and public policy. Research Policy, 15, 285–305.
Tripsas, M., Schrader, S., & Sobrero, M. (1995). Discouraging opportunistic behavior in collaborative R&D:
A new role for government. Research Policy, 24, 367–389.
Tushman, M., & Anderson, P. (1986). Technological discontinuities and organizational environments.
Administrative Science Quarterly, 31, 439–465.
Williamson, O. E. (1985). The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting.
New York: Free Press.

123